DeLee and Drez's Orthopaedic Sports Medicine, Third Edition: Expert Consult - Online and Print, 2-Volume Set

C o n t r i b u t o r s

Karim Abdollahi, MD Clinical Professor of Orthopedic Surgery, Loma Linda University Medical Center, Loma Linda; Orthopedic Surgeon, South Coast Medical Center, Laguna Beach, California Thoracic Outlet Syndrome

Shafeeq Ahmed, MD Hospitalist, Cardiovascular Services, Hartford Hospital, Hartford, Connecticut Management of Hypertension in Athletes

Mir Haroon Ali, MD, PhD Resident, Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota Heterotopic Bone Around the Elbow

David B. Allen, MD Professor of Pediatrics, University of Wisconsin School of Medicine and Public Health; Director of Endocrinology and Endocrine Fellowship Training, American Family Children’s Hospital, Madison, Wisconsin Diabetes Mellitus

Louis C. Almekinders, MD Attending Surgeon and Cofounder, North Carolina Orthopaedic Clinic, Durham, North Carolina Physiology of Injury to Musculoskeletal Structures

Annunziato Amendola, MD Professor and Director of the University of Iowa Sports Medicine Center, University of Iowa Hospitals and Clinics, Iowa City, Iowa Leg Pain and Exertional Compartment Syndromes; Stress Fractures of the Leg

James R. Andrews, MD Orthopedic Surgeon, Alabama Sports Medicine Center, Birmingham, Alabama Throwing Injuries in the Adult

Robert A. Arciero, MD Professor of Orthopaedic Surgery, University of Connecticut Health Center, Framingham, Connecticut Sports Medicine Terminology

April Armstrong, MD, MSc, FRCSC Assistant Professor, Department of Orthopaedics, Milton S. Hershey Medical Center, Hershey, Pennsylvania The Female Athlete

Robert E. Atkinson, MD Associate Professor and Division Chief, Department of Orthopedic Surgery, University of Hawaii John A. Burns School of Medicine, and Program Director, University of Hawaii Residency Program; Associate, The Queen’s Medical Center, Honolulu, Hawaii Athletic Injuries of the Adult Hand

Geoffrey S. Baer, MD, PhD Assistant Professor of Orthopedic Surgery, Division of Sports Medicine, Department of Orthopedics and Rehabilitation, University of Wisconsin Medical School, Madison, Wisconsin Tendon Injuries of the Foot and Ankle

Roald Bahr, MD, PhD Chair, Oslo Sports Trauma Research Centre, Oslo, Norway Preventing Hamstring Strains

Sue D. Barber-Westin, BS Director, Clinical and Applied Research, Cincinnati Sportsmedicine Research and Education Foundation, Cincinnati, Ohio High Tibial Osteotomy in the Anterior Cruciate Ligament–Deficient Knee with Varus Angulation

Bryce Bederka, MD The Bone and Joint Clinic, Portland, Oregon Leg Pain and Exertional Compartment Syndromes; Stress Fractures of the Leg

J. Michael Bennett, MD Clinical Instructor, Department of Orthopedic Surgery, Sports Medicine, and Arthoscopy, University of Texas at Houston; Orthopedic Surgeon, Fondren Orthopedic Group; Orthopedic Surgeon, Texas Orthopedic Hospital, Houston, Texas Vascular Problems of the Shoulder

Thomas M. Best, MD, PhD Professor and Pomerene Chair in Family Medicine; Chief, Division of Sports Medicine; Director, Primary Care Sports Medicine Fellowship, The OSU Sports Medicine Center, the Ohio State University, Columbus, Ohio Physiology of Injury to Musculoskeletal Structures; Sudden Death in Athletes: Causes, Screening Strategies, Use of Participation Guidelines, and Treatment of Episodes

Bruce D. Beynnon, MS, PhD Professor of Orthopaedics and Rehabilitation, University of Vermont College of Medicine, Burlington, Vermont Relevant Biomechanics of the Knee

vii

viii

Contributors

Mark J. Billante, MD Orthopaedic Surgeon, Greater Austin Orthopaedics, Austin, Texas Knee Replacement in Aging Athletes

Leslie Bonci, MPH, RD Director of Sports Medicine Nutrition, Center for Sports Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Nutrition for Sports

Andrew H. Borom, MD Tallahassee Orthopedic Clinic, Tallahassee, Florida Sports Shoes and Orthoses

James P. Bradley, MD Clinical Professor of Orthopedics, University of Pittsburgh Medical Center; Director, Sports Medicine Department, University of Pittsburgh Medical Center Shadyside Hospital; Director, Sports Medicine Department, University of Pittsburgh Medical Center St. Margaret Hospital, Pittsburgh, Pennsylvania Elbow Injuries in Children and Adolescents; Pediatric Elbow Fractures and Dislocations

Barton R. Branam, MD Physician, Ohio Valley Orthopaedics and Sports Medicine, Cincinnati, Ohio Allograft Tissues

Mark R. Brinker, MD Clinical Professor, Department of Orthopedic Surgery, Baylor College of Medicine; Director, Acute and Reconstructive Trauma Department, Texas Orthopedic Hospital, Fondren Orthopedic Group, Houston, Texas Physiology of Injury to Musculoskeletal Structures

Stephen F. Brockmeier, MD Orthopedic Surgeon, Perry Orthopedics and Sports Medicine, Charlotte, North Carolina Meniscal Injuries

James W. Brodsky, MD Clinical Professor of Orthopaedic Surgery, University of Texas Southwestern Medical School; Fellowship Director, Baylor University Medical Center, Dallas, Texas Stress Fractures of the Foot and Ankle

David E. Brown, MD Clinical Associate Professor of Orthopedic Surgery, University of Nebraska Medical Center; Surgeon, OrthoWest, PC, Omaha, Nebraska Sports Pharmacology: Ergogenic Drugs in Sports

Lauren Brown, BA Research Assistant, Department of Orthopaedics and Rehabilitation, University of Vermont, Burlington, Vermont Relevant Biomechanics of the Knee

Thomas E. Brown, MD Associate Professor, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, Virginia Physical Activity and Sports Participation after Total Hip Arthroplasty

Shawn M. Brubaker, DO Staff, Shasta Orthopaedics and Sports Center, Redding, California Physical Activity and Sports Participation after Total Hip Arthroplasty

Nathan Bruck, MD Instructor, Tel Aviv University Medical School, Tel Aviv; Attending Physician, Sheba Medical Center, Tel-Hashomer, Israel Stress Fractures of the Foot and Ankle

Joseph A. Buckwalter, MD Professor and Head, Department of Orthopaedics and Rehabilitation, University of Iowa, Iowa City, Iowa Physiology of Injury to Musculoskeletal Structures

Wayne Z. Burkhead, MD Orthopedic Surgeon, Private Practice, Dallas, Texas Impingement Lesions in Adult and Adolescent Athletes

Brian Busconi, MD Associate Professor of Orthopaedics and Physical Rehabilitation, University of Massachusetts Medical School, Worcester, Massachusetts Hip and Pelvis

Kenneth P. Butters, MD Orthopedic Surgeon, Slocum Center for Orthopedics and Sports Medicine, Eugene, Oregon Nerve Lesions of the Shoulder

S. Terry Canale, MD Professor and Department Chair, University of Tennessee-Campbell Clinic Department of Orthopaedic Surgery; Staff Physician, Campbell Clinic, Memphis, Tennessee Osteochondroses and Related Problems of the Foot and Ankle

Robert C. Cantu, MD, FACS Medical Director, National Center for Catastrophic Sports Injury Research, University of North ­ Carolina Medical Center, Chapel Hill, North Carolina; Chief of Neurosurgery, Emerson Hospital, Concord, ­Massachusetts Head Injuries

Robert V. Cantu, MD Assistant Professor of Orthopaedic Surgery, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire Head Injuries

Chang-Hyuk Choi, MD Assistant Professor, Department of Orthopaedic Surgery, Hanyang University Medical School, Hanyang University Hospital, Seoul, Korea Injuries of the Proximal Humerus in Adults

Contributors

Luke Choi, MD Orthopaedic Resident, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, Virginia Overuse Injuries

Thomas O. Clanton, MD Professor and Chairman, Department of Orthopaedics, University of Texas–Houston Medical School, Houston, Texas Sport Shoes and Orthoses; Etiology of Injury to the Foot and Ankle

Jack Clement, MD, PhD Physician, South Texas Radiology Imaging Centers, San Antonio, Texas Basic Imaging Techniques in the Adult; Imaging Considerations in the Skeletally Immature Patient

Brian J. Cole, MD, MBA Professor of Orthopaedic Surgery, Rush University; Director, Cartilage Restoration Center, Rush University Medical Center, Chicago, IIIinois Articular Cartilage Lesion

Fred G. Corley, Jr., MD Professor, Department of Orthopaedics, University of Texas Health Science Center at San Antonio; Professor of Orthopaedic Surgery, Department of Orthopaedic Traumatology, University Health System; Professor of Orthopaedic Surgery, Department of Orthopaedics, University of Texas Medicine Clinic, University of Texas Health Science Center, San Antonio, Texas Arm

Jason A. Craft, MD Assistant Professor, Department of Orthopaedic Surgery, University of Mississippi School of Medicine; Assistant Professor, University Hospital and Clinic, University of Mississipi Medical Center, Jackson, Mississippi Fractures of the Coracoid in Adults and Children; Glenoid and Scapula Fractures in Adults and Children

Ralph J. Curtis, Jr., MD Orthopedic Surgeon, Orthopedic Surgery Association, San Antonio, Texas Anatomy, Biomechanics, and Kinesiology of the Child’s Shoulder; Glenohumeral Instability in the Child

Frances Cuomo, MD Assistant Professor, Department of Orthopaedic Surgery, New York University School of Medicine; Chief, Shoulder and Elbow Service, New York University Hospital for Joint Diseases, New York, New York Injuries of the Proximal Humerus in Adults

ix

Thomas M. DeBerardino, MD Associate Professor, Department of Surgery, F. E. Edward Hébert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Director, John A. Feagin, Jr. Sports Medicine Fellowship at West Point; Head Team Physician, United States Military Academy at West Point; Physician, Keller Army Community Hospital, West Point, New York The Team Physician: Preparticipation Examination, On-Field Emergencies, and Ethical and Legal Issues

Richard E. Debski, PhD Associate Professor, Department of Bioengineering, University of Pittsburgh, Pittsburgh, Pennsylvania Fundamentals of Biomechanics; Functional Anatomy and Biomechanics of the Adult Shoulder

Marc M. DeHart, MD Surgeon, Texas Orthopedics, Sports, and Rehabilitation Associates, Austin, Texas Deep Venous Thrombosis and Pulmonary Embolism

Marlene DeMaio, MD, CAPT, USNR Surgeon, Department of Orthopaedic Surgery, Bone and Joint/Sports Medicine Institute, Naval Medical Center, Portsmouth, Virginia The Female Athlete

Allen Deutsch, MD Clinical Assistant Professor, Department of Orthopaedic Surgery, Baylor College of Medicine; Physician, KelseySeybold Clinic; Physician, St. Luke’s Episcopal Hospital, Houston, Texas Fractures of the Coracoid in Adults and Children; Glenoid and Scapula Fractures in Adults and Children

William W. Dexter, MD Professor, University of Vermont College of Medicine; Director, Sports Medicine Program, Maine Medical Center, Portland, Maine Dermatologic Disorders

David R. Diduch, MD Professor of Orthopaedic Surgery, Head Orthopaedic Team Physician, and Fellowship Director of Sports Medicine, University of Virginia, Charlottesville, Virginia Knee Replacement in Aging Athletes

William Dillin, MD Orthopaedic Surgeon, Kerlan-Jobe Orthopaedic Clinic, Los Angeles and Orange County, California Thoracolumbar Spine Injuries in the Adult

Jeffrey R. Dugas, MD Affiliate Professor, College of Human Services, Troy University, Troy; Fellowship Director, American Sports Medicine Institute, Birmingham, Alabama Throwing Injuries in the Adult



Contributors

Craig J. Edson, MHS, PT Physical Therapist, Geisinger/HealthSouth Rehabilitation Hospital, Danville, Pennsylvania Multiple Ligament Knee Injuries

T. Bradley Edwards, MD Clinical Instructor, Department of Orthopaedic Surgery, University of Texas at Houston; Clinical Assistant Professor, Department of Orthopaedic Surgery, Baylor University; Orthopaedic Surgeon, Fondren Orthopedic Group, Houston, Texas Development of Skills for Shoulder Surgery; Glenohumeral Arthritis in the Athlete

Marsha Eifert-Mangine, EdD, PT, ATC Assistant Professor, Department of Health Sciences, Physical Therapy Program, College of Mount Saint Joseph; Physical Therapist, NovaCare Rehabilitation, Cincinnati, Ohio Use of Modalities in Sports

Frank J. Eismont, MD Leonard M. Miller Professor and Chairman, Department of Orthopaedics; Director, Residency and Fellow Education, University of Miami, Miami; Orthopaedic Surgeon, University of Miami Medical Group, Miami, Florida Thoracolumbar Spine Injuries in the Adult

Hussein Elkousy, MD Volunteer Faculty, Department of Orthopaedic Surgery, University of Texas Health Science Center at Houston; Volunteer Faculty, University of Texas Medical Branch, Galveston; Volunteer Faculty, Baylor College of Medicine, Houston; Staff Surgeon, Texas Orthopedic Hospital, Houston, Texas Development of Skills for Shoulder Surgery; Muscle Ruptures Other Than the Rotator Cuff

Gregory C. Fanelli, MD Chief Emeritus, Sports Medicine and Arthroscopic Surgery, Geisinger Medical Center, Danville, Pennsylvania Multiple Ligament Knee Injuries

Mario Ferretti, MD Researcher, Department of Orthopaedic Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania Anterior Cruciate Ligament Injuries in the Adult

Gary B. Fetzer, MD Orthopaedic Surgeon,TRIA Orthopaedic Center, Bloomington, Minnesota Lateral and Posterolateral Injuries of the Knee

Larry D. Field, MD Clinical Instructor, Department of Orthopaedic Surgery, University of Mississippi School of Medicine; Director, Upper Extremity Service, Mississippi Sports Medicine and Orthopaedic Center, Jackson, Mississippi Osteochondritis Dissecans of the Elbow; Olecranon Bursitis; Elbow Dislocations in the Adult Athlete and Pediatric Patient

Daniel C. Fitzpatrick, MD, MS Orthopedic Surgeon, Slocum Center for Orthopedics and Sports Medicine, Eugene, Oregon Nerve Lesions of the Shoulder

Kevin R. Ford, MS Research Biomechanist, Cincinnati Children’s Hospital, Cincinnati, Ohio Return-to-Sport Plyometric Training in the Rehabilitation of Athletes following Anterior Cruciate Ligament Reconstruction

Donald E. Fowler Medical Student, University of Virginia, Charlottesville, Virginia Exercise Physiology

W. Anthony Frisella, MD Orthopaedic Surgeon, St. Peters Bone and Joint Surgery, St. Peters, Missouri Injuries of the Proximal Humerus in Adults

Freddie H. Fu, MD, DSc Chairman, Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Anterior Cruciate Ligament Injuries in the Adult; Anterior Cruciate Ligament Injuries in the Child

Lorenzo Gamez, MD Assistant Professor of Orthopaedics and Physical Rehabilitation, Department of Orthopaedics, University of Massachusetts Medical School, Worcester, Massachusetts Entrapment Neuropathies of the Foot

Seth C. Gamradt, MD Assistant Professor, Department of Orthopaedic Surgery and Sports Medicine, University of California, Los Angeles, Los Angeles, California Glenohumeral Instability in Adults

William E. Garrett, Jr., MD, PhD Professor, Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina Physiology of Injury to Musculoskeletal Structures; Sports Medicine Terminology

Gary M. Gartsman, MD Clinical Professor, Department of Orthopaedic Surgery, University of Texas Health Sciences Center at Houston Medical School; Surgeon, Fondren Orthopedic Group, Houston, Texas Adhesive Capsulitis

Contributors

Christian Gerber MD, FRCSEd (Hon) Professor of Orthopaedic Surgery, University of Zürich; Chairman, Department of Orthopaedics, Uniklinik Balgrist, Zürich, Switzerland Suture Materials

Eric Giza, MD Orthopaedic Surgeon, Orthopaedic Sports Medicine Service, University of California, Davis Health System, Sacramento, California Ankle Instability Prevention; Principles of Injury Prevention; Spine-Related Injury Prevention in the Athlete: Trunk Stabilization

S. Raymond Golish, MD, PhD Resident, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, Virginia Design and Statistics in Sports Medicine

Jorge E. Gómez, MD Clinical Professor, Sports Medicine and Pediatrics, University of Texas Health Science Center; Team Physician, University of Texas, San Antonio, Texas Heat Illness; Cold Injury; Altitude

Andreas H. Gomoll, MD Assistant Professor of Orthopaedic Surgery, Harvard Medical School; Orthopaedic Surgeon, Brigham and Women’s Hospital, Boston, Massachusetts Articular Cartilage Lesion

Letha Y. Griffin, MD, PhD Physician, Peachtree Orthopaedic Clinic, Atlanta, Georgia The Female Athlete

Philippe P. Grondin, MD Fellow, Department of Orthopaedics, University of British Columbia, Vancouver, British Columbia, Canada Tendinopathies around the Arm

Andrew J. Grove, MD Adjunct Assistant Professor, Department of Pediatrics, Division of Adolescent Medicine, Medical College of Wisconsin; Primary Care/Sports Medicine Physician, Student Health Service, Marquette University, Milwaukee, Wisconsin Heat Illness; Cold Injury

Dan Guttman, MD Clinical Instructor, Department of Orthopaedic Surgery, University of New Mexico, Albuquerque; Chief of Upper Extremity Surgery and the Hip Arthritis Service, Taos Orthopaedic Institute, Taos, New Mexico Injuries of the Proximal Humerus in Adults

Gregory P. Guyton, MD Attending, Department of Orthopaedics, Union Memorial Hospital, Baltimore, Maryland Entrapment Neuropathies of the Proximal Humerus

xi

Christopher D. Harner, MD Professor, Department of Orthopaedic Surgery, University of Pittsburgh School of Medicine; Chief, Division of Sports Medicine; Fellowship Director and Medical Director, Center for Sports Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania

Posterior Cruciate Ligament Injuries in the Adult; Posterior Cruciate Ligament Injuries in the Child

Justin D. Harris, MD Orthopaedic Surgeon, Nebraska Orthopaedic and Sports Medicine, Lincoln, Nebraska Multiple Ligament Knee Injuries

Jennifer A. Hart, MPAS, PA-C Physician Assistant, Department of Orthopaedic Surgery, Division of Sports Medicine, University of Virginia, Charlottesville, Virginia Basic Arthroscopic Principles; Infection: Prevention, Control, and Treatment

Joseph M. Hart, PhD, ATC Assistant Professor of Research, University of Virginia, Charlottesville, Virginia Exercise Physiology

Andrew Haskell, MD Assistant Clinical Professor, University of California, San Francisco, San Francisco; Attending, Palo Alto Foundation Medical Group, Palo Alto, California Biomechanics

Martin J. Herman, MD Associate Professor, Departments of Orthopaedic Surgery and Pediatrics, Drexel University College of Medicine; Attending, St. Christopher’s Hospital for Children and Hahnemann University Hospital, Philadelphia, Pennsylvania Cervical Spine Injuries in the Child

Timothy E. Hewett, PhD, FACSM Associate Professor, University of Cincinnati College of Medicine; Director, Sports Medicine Biodynamics Center, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio Return-to-Sport Plyometric Training in the Rehabilitation of Athletes following Anterior Cruciate Ligament Reconstruction

Christopher B. Hirose, MD Orthopaedic Surgeon, Panorama Orthopaedics, Golden, Colorado Etiology of Injury to the Foot and Ankle

xii

Contributors

Nicholas J. Honkamp, MD Staff Physician, Iowa Methodist Medical Center; Private Practice, Des Moines Orthopaedic Surgeons, Des Moines, Iowa

Anterior Cruciate Ligament Injuries in the Adult; Anterior Cruciate Ligament Injuries in the Child; Posterior Cruciate Ligament Injuries in the Child; Posterior Cruciate Ligament in the Adult

Florian G. Huber, MD Orthopaedic Surgeon, Peninsula Orthopaedic Associates; Orthopaedic Trauma Surgeon, Division of Orthopaedic Trauma, Peninsula Regional Medical Center, Salisbury, Maryland Arm

Jack V. Ingari, MD Assistant Professor of Surgery, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Associate Professor of Surgery, University of Texas Health Sciences Center at San Antonio; Staff, Attending Hand Surgeon, The Hand Center of San Antonio, San Antonio, Texas The Adult Wrist

Darren L. Johnson, MD Professor and Chairman, Department of Orthopaedic Surgery, University of Kentucky School of Medicine; Chief of Orthopaedic Surgery, The Kentucky Clinic, University of Kentucky, Lexington, Kentucky Design and Statistics in Sports Medicine; Allograft Tissues; Medial Collateral Ligament Injuries in Adults; Pediatric Medial Knee Injuries

Rob Johnson, MD Associate Professor, Department of Family Medicine and Community Health, University of Minnesota; Director, Primary Care Sports Medicine, Department of Family Medicine, Hennepin County Medical Center, Minneapolis, Minnesota Infectious Disease and Sports

Robert J. Johnson, MD Professor Emeritus, Department of Orthopaedics, The University of Vermont College of Medicine, University of Vermont, Burlington, Vermont Relevant Biomechanics of the Knee

Ron M. Johnson, PT, MPT, ATC, LAT, ATC, CSCS Facility Director, Excel Sports Therapy, Gulf Coast Rehabilitation, PC, Shiner, Texas Therapeutic Exercise Prescription

James S. Keene, MD Professor of Orthopedic Surgery, University of Wisconsin; Chairman, Division of Sports Medicine, Department of Orthopedic Surgery and Rehabilitation, University of Wisconsin Medical School, Madison, Wisconsin Tendon Injuries of the Foot and Ankle

Sami O. Khan, MD Orthopaedic Surgeon, Resurgens Orthopaedics, Decatur, Georgia Elbow Dislocations in the Adult Athlete and Pediatric Patient

Richard Y. Kim, MD Orthopaedic Surgeon, Hackensack University Medical Center, Hackensack, New Jersey Entrapment Neuropathies around the Elbow

Donald T. Kirkendall, PhD Adjunct Assistant Professor, Department of Exercise and Sport Sciences, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina Physiology of Injury to Musculoskeletal Structures; Sports Medicine Terminology

Scott H. Kitchel, MD Athletic Medicine Staff, University of Oregon, Eugene, Oregon Thoracolumbar Spine Injuries in the Adult

Sandra E. Klein, MD Assistant Professor, Department of Orthopaedic Surgery, Washington University School of Medicine; Attending, Barnes-Jewish Hospital at Washington University School of Medicine, St. Louis, Missouri Conditions of the Forefoot

William Knopp, MD Assistant Professor, Department of Family and Community Medicine, University of Minnesota School of Medicine; Faculty, Family Medicine and Primary Care Sports Medicine, Methodist Hospital/University of Minnesota Family Medicine Residency Program, Minneapolis, Minnesota Infectious Disease and Sports

Mininder Kocher, MD, MPH Associate Professor of Orthopaedic Surgery, Harvard Medical School; Associate Director, Division of Sports Medicine, Children’s Hospital Boston, Boston, Massachussetts The Young Athlete

Melissa D. Koenig, MD Orthopaedic Surgeon, Kaiser Permanente Medical Group, Houston, Texas Ligament Injuries of the Foot and Ankle in Adult Athletes

Sumant G. Krishnan, MD Assistant Clinical Professor, Department of Orthopaedic Surgery, University of Texas Southwestern Medical Center at Dallas Southwestern Medical School; Staff, Shoulder Service, W.B. Carrell Memorial Clinic, Dallas, Texas Impingement Lesions in Adult and Adolescent Athletes

Contributors

Irving L. Kron, MD Professor and Chairman, Department of Surgery, Division of Thoracic and Cardiovascular Surgery, University of Virginia Health System, Charlottesville, Virginia Vascular Problems—Popliteal Artery Entrapment

John E. Kuhn, MD Associate Professor, Department of Orthopaedics and Rehabilitation, Division of Sports Medicine, Vanderbilt University Medical School; Chief of Shoulder Surgery, Vanderbilt University Medical Center, Nashville, ­Tennessee Scapulothoracic Disorders in Athletes

Robert F. LaPrade, MD, PhD Professor, Sports Medicine and Shoulder Surgery, Department of Orthopaedic Surgery, University of Minnesota, Minneapolis, Minnesota Lateral and Posterolateral Injuries of the Knee

William C. Lauerman, MD Professor of Orthopaedic Surgery, Georgetown University School of Medicine; Chief, Division of Spine Surgery, Georgetown University Hospital, Washington, DC Thoracolumbar Spine Injuries in the Child

Christine Lawless, MD, MBA President, Sports Cardiology Consultants LLC, Columbus, Ohio Sudden Death in Athletes: Causes, Screening Strategies, Use of Participation Guidelines, and Treatment of Episodes

James Lebolt, DO Orthopedic Surgeon, Virginia Sports Medicine and Orthopedic Institute, Christiansburg; Orthopedic Surgeon, Montgomery Regional Hospital, Blacksburg, Virginia Throwing Injuries in the Adult

Igor Cezar da Silva Leitao, MD Assistant Professor, Department of Orthopaedics, Santa Casa de Misericordia de Juiz de Fora, Juiz de Fora/MG, Brazil; Visiting Fellow from Hospital Felicio Rocho, University of Texas Health Science Center at San Antonio, San Antonio, Texas Injuries to the Sternoclavicular Joint in the Adult and Child

Kenneth C. Lin, MD Orthopedic Surgeon, Evergreen Orthopedics, Monroe, Washington Impingement Lesions in Adult and Adolescent Athletes

Thomas N. Lindenfeld, MD Associate Director, Cincinnati Sports Medicine and Orthopaedic Center, Cincinnati, Ohio Complex Regional Pain Syndromes Including Reflex Sympathetic Dystrophy and Causalgia

xiii

Turner C. Lisle, MD Resident, Department of Surgery, University of Virginia, Charlottesville, Virginia Vascular Problems—Popliteal Artery Entrapment

Walter R. Lowe, MD Associate Professor, Department of Orthopedic Surgery, Baylor College of Medicine; Team Physician, Houston Texans, Houston, Texas Superior Labral Injuries

David J. Lunardini, BS Medical Student, University of Virginia, Charlottesville, Virginia Exercise Physiology

Mark W. Maffet, MD Assistant Professor, Department of Orthopedic Surgery, Baylor College of Medicine; Team Physician, Houston Comets; Team Physcian, Houston Baptist University, Houston, Texas Superior Labral Injuries

Jeffrey J. Mair, DO Attending, Riverview Medical Center and Twin Cities Orthopedics, Waconia, Minnesota Lateral and Posterolateral Injuries of the Knee

Robert Mangine, MEd, PT, ATC Adjunct Instructor, College of Mount Saint Joseph; Adjunct Clinical Instructor, Department of Ortho­ pedics, and Head Football Trainer, University of Cincinnati; Director of Clinical Residency, NovaCare Rehabilitation, Cincinnati, Ohio Use of Modalities in Sports

Roger A. Mann, MD Associate Clinical Professor, Department of Orthopaedic Surgery, University of California, San Francisco, School of Medicine, San Francisco; Director, Foot Fellowship Program, Oakland, California Biomechanics; Entrapment Neuropathies of the Foot

John G. Mastronarde, MD, MSc Associate Professor, Ohio State University; Director, OSU Asthma Center, The Ohio State University Medical Center, Columbus, Ohio Exercise-Induced Bronchospasm

Carl G. Mattacola, PhD Associate Professor and Director, Rehabilitation Sciences Doctoral Program, Division of Athletic Training, University of Kentucky, College of Health Sciences, Lexington, Kentucky Design and Statistics in Sports Medicine

xiv

Contributors

Augustus D. Mazzocca, MD Associate Professor of Orthopaedic Surgery, University of Connecticut, Farmington, Connecticut Injuries to the Acromioclavicular Joint in Adults and Children; Sternum and Rib Fractures in Adults and Children

Wendy McBride, MD Pediatric Emergency Physician, CarePoint PC, Physician, Sky Ridge Medical Center, Denver, Colorado Heat Illness; Altitude

Kendra McCamey, MD Assistant Clinical Professor, Department of Family Medicine, The Ohio State University Medical Center; Team Physician, Department of Athletics, The Ohio State University, Columbus, Ohio Exercise-Induced Bronchospasm

Michael P. McClincy, BS Medical Student, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Forearm Fractures: Pediatric Elbow Fractures and Dislocations

Edward R. McDevitt, MD Assistant Professor of Surgery, Uniformed Services University of the Health Sciences, Bethesda; Orthopaedic Surgeon, Anne Arundel Medical Center, Annapolis, Maryland Sports Pharmacology: Ergogenic Drugs in Sports; Sports Pharmacology: Recreational Drug Use

Patrick J. McMahon, MD Adjunct Associate Professor, Department of Bio­ engineering, University of Pittsburgh; Founder, McMahon Orthopedics and Rehabilitation, Pittsburgh, Pennsylvania Functional Anatomy and Biomechanics of the Adult Shoulder

Jennifer J. F. McVean, MD Fellow, University of Wisconsin; Fellow, American Family Children’s Hospital, Madison, Wisconsin Diabetes Mellitus

W. Andrew Middendorf, MPT, ATC Physical Therapist, NovaCare Rehabilitation/University of Cincinnati, Cincinatti, Ohio Use of Modalities in Sports

Chealon D. Miller, MD Resident, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, Virginia Infection: Prevention, Control, and Treatment

Mark D. Miller, MD S. Ward Casscells Professor of Orthopaedic Surgery and Head, Division of Sports Medicine, University of Virginia, Charlottesville; Adjunctive Clinical Professor and Team Physician, James Madison University, Harrisonburg, Virginia Design and Statistics in Sports Medicine; Basic Arthroscopic Principles; Infection: Prevention, Control, and Treatment

Bernard F. Morrey, MD Professor of Orthopedics, College of Medicine, Mayo Clinic; Consultant, Division of Adult Reconstruction, Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota Biomechanics of the Elbow and Forearm; Tendinopathies around the Elbow

Vasilios Moutzouros, MD Clinical Instructor, Wayne State School of Medicine; Senior Staff, Department of Orthopaedics, Division of Sports Medicine, Henry Ford Center, Detroit, Michigan Osteochondroses

Van C. Mow, PhD Stanley Dicker Professor and Chairman, Department of Biomedical Engineering; Director, Shelley Liu Ping Laboratory for Functional Tissue Engineering Research, Fu Foundation School of Engineering and Applied Science, Columbia University, New York, New York Physiology of Injury to Musculoskeletal Structures

Gregory D. Myer, MS Sports Biomechanist, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio Return-to-Sport Plyometric Training in the Rehabilitation of Athletes following Anterior Cruciate Ligament Reconstruction

Frank R. Noyes, MD Clinical Professor, Department of Orthopaedic Surgery, University of Cincinnati; Chairman, Medical Director, and Chief Operating Officer, Cincinnati Sports Medicine and Orthopaedic Center, Cincinnati, Ohio High Tibial Osteotomy in the Anterior Cruciate Ligament-Deficient Knee with Varus Angulation

Eugene T. O’Brien Orthopaedic Surgeon, Churchill Evaluation Centers, San Antonio, Texas Wrist Injuries in the Child

Agbecko Ocloo, MD Staff Surgeon, Korle Bu Teaching Hospital, Korle Bu, Accra, Ghana Patellar Fractures

Daniel P. O’Connor, PhD Director, Joe W. King Orthopedic Institute, Houston, Texas Physiology of Injury to Musculoskeletal Structures

Nnamdi Okeke, BS Medical Student, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Anterior Cruciate Ligament Injuries in the Adult

Contributors

Brett D. Owens, MD Assistant Professor, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Assistant Professor, Texas Tech University Health Science Center; Chief, Sports Medicine and Shoulder Service, William Beaumont Army Medical Center, El Paso, Texas The Team Physician: Preparticipation Examination, On-Field Emergencies, and Ethical and Legal Issues

Russ Paine, BS, PT Director of Sports Medicine Research and Rehabilitation, Memorial Hermann Sports Medicine Institute; Team Physical Therapist for Houston Rockets, Houston, ­Texas Proprioception and Joint Dysfunction; Language of Excercise and Rehabilitation

Selene G. Parekh, MD, MBA Clinical Associate Professor, Duke University; Attending, Division of Orthopaedic Surgery, Duke University Medical Center; Adjunct Faculty, Duke University School of Business, Durham, North Carolina Heel Pain

William D. Parham, PhD, ABPP Dean of Graduate School of Professional Psychology, John F. Kennedy University, Pleasant Hill, California Psychological Adjustment to Athletic Injury

Richard D. Parker, MD Professor of Surgery, Cleveland Clinic Lerner College of Medicine; Chairman, Department of Orthopaedic Surgery, Orthopaedic Rheumatologic Institute, Cleveland Clinic Foundation, Cleveland, Ohio Patellar and Quadriceps Tendinopathies and Ruptures; Osteochondroses; Patellofemoral Instability: Recurrent Dislocation of the Patella; Acute Dislocation of the Patella; Chronic Dislocation of the Patella; Patellar Fractures

Johnathan P. Parsons, MD Assistant Professor of Internal Medicine, The Ohio State University; Associate Director, The Ohio State University Asthma Center, Columbus, Ohio Exercise-Induced Bronchospasm

Jayesh K. Patel, MD Chief Resident, Department of Orthopaedic Surgery, University of Kentucky, Lexington, Kentucky Medial Collateral Ligament Injuries in Adults; Pediatric Medial Knee Injuries

Mark V. Paterno, MS, PT, MBA, SCS, ATL Assistant Professor, Department of Pediatrics, Division of Sports Medicine, University of Cincinnati College of Medicine; Coordinator of Orthopaedic and Sports Physical Therapy, Division of Occupational Therapy and Physical Therapy, Cincinnati Children’s Medical Center, Cincinnati, Ohio Return-to-Sport Plyometric Training in the Rehabilitation of Athletes following Anterior Cruciate Ligament Reconstruction

xv

Russell S. Petrie, MD Orthopaedic Surgeon, Newport Orthopaedic Institute, Newport Beach, California Elbow Injuries in Children and Adolescents

Peter D. Pizzutillo, MD Professor, Departments of Orthopedic Surgery and Pedia­trics, Drexel University School of Medicine; Director, St. Christopher’s Hospital for Children; Physician, Hahnemann University Hospital, Philadelphia, Pen­nsylvania Cervical Spine Injuries in the Child

Michael D. Pleacher, MD Assistant Professor of Pediatrics, University of New Mexico, Albuquerque, New Mexico Dermatologic Disorders

Teodor T. Postolache, MD Associate Professor and Director, Mood and Anxiety Program, University of Maryland School of Medicine, Baltimore, Maryland; Director, Institute for Sports Chronobiology, Washington, DC Sleep and Chronobiology in Sports

Matthew T. Provencher, MD Director, Orthopaedic Shoulder, Knee, and Sports Surgery, Department of Orthopaedic Surgery, Naval Medical Center, San Diego, San Diego, California Injuries to the Acromioclavicular Joint in Adults and Children; Sternum and Rib Fractures in Adults and Children

Anil S. Ranawat, MD Orthopaedic Surgeon, Hospital for Special Surgery, New York, New York Forearm Fractures; Pediatric Elbow Fractures and Dislocations; Posterior Cruciate Ligament Injuries in the Adult; Posterior Cruciate Ligament Injuries in the Child

Michael A. Rauh, MD Clinical Assistant Professor of Orthopaedic Surgery, Department of Orthopaedic Surgery, University Sports Medicine, State University of New York at Buffalo, Buffalo, New York Patellar and Quadriceps Tendinopathies and Ruptures

William D. Regan, MD Assistant Professor of Orthopaedic Surgery, University of British Columbia Faculty of Medicine, Vancouver, British Columbia, Canada Tendinopathies around the Elbow

Scott B. Reynolds, MD Orthopaedic Surgeon, Nebraska Orthopaedic Associates, Omaha, Nebraska Throwing Injuries in the Adult

xvi

Contributors

David R. Richardson, MD Assistant Professor and Director of Residency Program, University of Tennessee-Campbell Clinic Department of Orthopaedic Surgery, University of Tennessee; Staff Physician, Campbell Clinic, Memphis, Tennessee Osteochondroses and Related Problems of the Foot and Ankle

John T. Riehl, MD Resident, Geisinger Pennsylvania

Medical

Center,

Danville,

Multiple Ligament Knee Injuries

David Ring, MD, PhD Associate Professor of Orthopaedic Surgery, Harvard Medical School; Medical Director and Director of Research, Orthopaedic Hand and Upper Extremity Service, Massachusetts General Hospital, Boston, Massachusetts Fractures of the Elbow in the Adult

Kristin N. Rinheimer, MS, PAC Physician’s Assistant, Geisinger Medical Center, Danville, Pennsylvania Multiple Ligament Knee Injuries

Samuel P. Robinson, MD Orthopaedic Surgeon, Jordan-Young Institute, Virginia Beach, Virginia Pediatric Elbow Fractures and Dislocations

Charles A. Rockwood, Jr., MD Professor and Chairman Emeritus, Department of Orthopaedics, University of Texas Health Science Center at San Antonio, San Antonio, Texas Injuries to the Sternoclavicular Joint in the Adult and Child

Scott A. Rodeo, MD Co-Chief, Sports Medicine and Shoulder Service, Hospital for Special Surgery, New York, New York Meniscal Injuries

Anthony A. Romeo, MD Director, Shoulder Service, Department of Orthopaedic Surgery, Rush University, Chicago, Illinois Injuries to the Acromioclavicular Joint in Adults and Children; Sternum and Rib Fractures in Adults and Children

Melvin P. Rosenwasser, MD Robert E. Carroll Professor of Orthopaedic Surgery, Columbia University College of Physicans and Surgeons, New York; Director, Orthopaedic Hand and Trauma Service, New York/Presbyterian Hospital, New York, New York Entrapment Neuropathies around the Elbow

Charles E. Rosipal, MD Physician, GIKK Ortho Specialists, Omaha, Nebraska Injuries to the Sternoclavicular Joint in the Adult and Child

Timothy G. Sanders, MD Visiting Professor, University of Kentucky College of Medicine, Lexington, Kentucky; Director of Education and Research, National Musculoskeletal Imaging, Weston, Florida Imaging of the Glenohumeral Joint

Felix H. Savoie, MD Lee C. Schlesinger Professor and Vice Chairman, Department of Orthopaedic Surgery, Tulane University School of Medicine; Director, Tulane Institute of Sports Medicine, Tulane University Medical Center, New Orleans, Louisiana Osteochondritis Dissecans of the Elbow

David L. Saxton, MD Clinical Faculty, Oklahoma University Medical Center, Oklahoma City, Oklahoma Complex Regional Pain Syndromes Including Reflex Sympathetic Dystrophy and Causalgia

Andrew J. Schorfhaar, DO Assistant Professor and Team Orthopaedic Surgeon, Department of Radiology, Division of Sports Medicine, Michigan State University, East Lansing, Michigan Lateral and Posterolateral Injuries of the Knee

Jon K. Sekiya, MD Associate Professor, Department of Orthopaedic Surgery, University of Michigan, Ann Arbor, Michigan Fundamentals of Biomechanics

Agam Shah, MD Othopaedic Surgeon, Newton-Wellesley Newton, Massachusetts

Hospital,

Hip and Pelvis

Wei Shen MD Resident, Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Anterior Cruciate Ligament Injuries in the Adult; Anterior Cruciate Ligament Injuries in the Child

Thomas Shepler, MD Associate Professor of Surgery, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Staff Physician, Reston HCA Hospital, Reston, Virginia The Pediatric Hand

Holly J. Silvers, MPT ACL Prevention Project Coordinator, Santa Monica Orthopaedic and Sports Medicine Group, Santa Monica, California Anterior Cruciate Ligament Tear Prevention in the Female Athlete; Ankle Instability Prevention

Contributors

Manuj Singhal, MD Physician, Orthopaedic Associates of Lewisville, Lewisville, Texas Medial Collateral Ligament Injuries in Adults; Pediatric Medial Knee Injuries

Timothy Steiner, MD Staff Orthopaedic Surgeon, Orthopaedics of Southern Indiana, Bloomington, Indiana

Patellofemoral Instability: Acute Dislocation of the Patella; Patellofemoral Instability: Recurrent Dislocation of the Patella

Scott P. Steinmann, MD Professor of Orthopedics, Mayo Clinic College of Medicine, Rochester, Minnesota Heterotopic Bone around the Elbow

Ligament Injuries of the Foot and Ankle in the Pediatric Athlete

Dean C. Taylor, MD Professor of Surgery and Director, Duke Sports Medicine Fellowship, Duke University, Durham, North Carolina Sports Medicine Terminology

Samir G. Tejwani, MD Orthopedic Surgeon, Department of Orthopedic Surgery, Division of Sports Medicine, Kaiser Permanente, Fontana, California Elbow Injuries in Children and Adolescents

Richard J. Thomas, MD Attending, Sports Medicine and Orthopaedic Surgery, OrthoGeorgia, Macon, Georgia Olecranon Bursitis

Paul D. Thompson, MD Director of Cardiology and Director of the Athlete’s Heart Program, Hartford Hospital, Hartford Connecticut Management of Hypertension in Athletes

Center,

Danville,

Multiple Ligament Knee Injuries

Steven M. Topper, MD Clinical Assistant Professor, Uniformed Services University of the Health Sciences, Bethesda, Maryland; Attending, The Colorado Hand Center, Colorado Springs, Colorado Wrist Arthroscopy

Joseph S. Torg, MD Professor of Orthopaedic Surgery, Temple University School of Medicine; Attending Staff, Temple University Hospital, Philadelphia, Pennsylvania Cervical Spine Injuries in the Adult

C. Thomas Vangsness, Jr., MD Professor of Orthopaedic Surgery, Department of Orthopaedic Surgery, University of Southern California, Los Angeles, California Osteochondritis Dissecans

Marius von Knoch, MD Associate Professor of Orthopaedic Surgery, University of Duisburg-Essen, Essen, Germany Suture Materials

J. Andy Sullivan, MD Clinical Professor of Pediatric Orthopedics, Department of Orthopedic Surgery, University of Oklahoma, College of Medicine; Attending, Oklahoma University Medical Center, Okalahoma City, Oklahoma

Daniel J. Tomaszewski, MD Attending, Geisinger Medical Pennsylvania

xvii

Keith L. Wapner, MD Clinical Professor of Orthopedic Surgery and Director of Foot and Ankle Orthopedic Fellowship, University of Pennsylvania; Adjunct Professor of Orthopedic Surgery, Drexel College of Medicine, Philadelphia, Pennsylvania Heel Pain

Russell F. Warren, MD Professor, Department of Orthopaedic Surgery, Weill Medical College of Cornell University; Professor of Orthopaedic Surgery and Attending Orthopaedic Surgeon, Hospital for Special Surgery, New York, New York Glenohumeral Instability in Adults

Scott Waterman, MD Sports Medicine Fellow, Keller Army Community Hospital, West Point, New York The Thigh

Robert G. Watkins IV, MD Co-Director, Marina Spine Center, The Marina Hospital, Marina del Rey, California Spine-Related Injury Prevention in the Athlete: Trunk Stabilization

Adam Nelson Whatley, MD Chief Resident, Department of Orthopaedic Surgery, Louisiana State University Health Sciences Center, New Orleans, Louisiana Parsonage-Turner Syndrome

Alexis C. Wickwire, MS Department of Bioengineering, University of Pittsburgh, Pittsburgh, Pennsylvania Fundamentals of Biomechanics

xviii

Contributors

Kaye E. Wilkins, DVM, MD Professor of Orthopaedics and Pediatrics, University of Texas Health Science Center at San Antonio; Staff Orthopaedic Surgeon, University Hospital, San Antonio; Staff Orthopaedic Surgeon, Christus Children’s Hospital, San Antonio, Texas Injuries of the Proximal Humerus in the Skeletally Immature Athlete

Gerald R. Williams, Jr., MD Professor of Orthopaedic Surgery, Jefferson Medical College; Chief, Shoulder and Elbow Service, The Rothman Institute, Philadelphia, Pennsylvania Glenoid and Scapula Fractures in Adults and Children; Fractures of the Coracoid in Adults and Children

Matthew D. Williams, MD Assistant Professor of Orthopaedic Surgery, Louisiana State University Health Sciences Center, New Orleans; Orthopaedic Surgeon, Acadiana Orthopaedic Group, Lafayette, Louisiana Adhesive Capsulitis; Glenohumeral Arthritis in the Athlete

Michael A. Wirth, MD Professor and Charles A. Rockwood, Jr., MD Chair, Department of Orthopaedics, University of Texas Health Science Center at San Antonio, San Antonio, Texas Injuries to the Sternoclavicular Joint in the Adult and Child

Valerie M. Wolfe, MD Fellow, Department of Orthopaedic Surgery, New York Presbyterian/Columbia Hospital, New York, New York Entrapment Neuropathies around the Elbow

Brett W. Wolters, MD, MS Attending Orthopaedic Surgeon, Memorial Medical Center, Springfield, Illinois Lateral and Posterolateral Injuries of the Knee

Savio L.-Y. Woo, PhD, DSc Whiteford Professor and Director, Musculoskeletal Research Center, Department of Bioengineering, University of Pittsburgh, Pittsburgh, Pennsylvania Physiology of Injury to Musculoskeletal Structures

Robert M. Wood, MD, FRCS Orthopedic Surgeon, Sports Medicine North, Lynnfield, and North Shore Medical Center, Salem, Massachusetts Etiology of Injury to the Foot and Ankle

Virchel E. Wood, MD Professor, Department of Orthopaedic Surgery, Loma Linda University School of Medicine; Consulting Chief, Hand Surgery Service, Loma Linda University Medical Center, Loma Linda, California Thoracic Outlet Syndrome

Adam Yanke Medical Student, Department of Orthopaedic Surgery, Rush University, Chicago, Illinois Articular Cartilage Lesion

John A. Zavala, MD Chief Resident, Georgetown Washington, DC

University

Hospital,

Thoracolumbar Spine Injuries in the Child

Mary L. Zupanc, MD Professor, Department of Neurology and Pediatrics; Chief, Division of Pediatric Neurology; Heidi Marie Bauman Chair of Epilepsy; Director, Pediatric Comprehensive Epilepsy Program, Medical College of Wisconsin, Milwaukee, Wisconsin Sports and Epilepsy

p r e f a c e

It is an honor and a pleasure to introduce the third edition of the popular textbook DeLee & Drez’s Orthopaedic Sports Medicine: Principles and Practice. In the preface to the previous edition, we emphasized that the orthopaedic sports specialist must be soundly schooled in a variety of conditions that are not commonly encountered in the dayto-day practice of orthopaedics. We would suggest that with the advent of subspecialty certification in orthopaedic sports medicine, this is even more important today. This textbook, unlike anything else in your library, can help you become properly schooled! So, what’s new in this edition? We have all new chapters, written by experts and leaders in their subspecialty. The contributors have done a fabulous job of reviewing and sharing their experience and the current state of knowledge in their respective fields. For this, we profoundly thank them! We introduce a new 4-color design and format, including full-color intraoperative and clinical images, to this edition. We offer expanded coverage of key topics, including new chapters on arthroscopic principles, allograft tissue, complications in athletes, nutrition, pharmacology, and psychology. We have also asked all ­ contributors to completely update all surgical

procedures from ACL reconstruction to cartilage transplantation to the latest arthroscopic shoulder techniques, including labral and rotator cuff repairs. With this edition we have also created a new website that includes the full content of the text, a self-assessment module, links to PubMed, an image library, content updates, and “bonus” video. We would like to recognize and thank our new web editor, Dr. Scott Montgomery, who is already working hard on developing this website. You can access it at www.expertconsult.com So, this is an exciting new edition. The contributors have done a masterful job. With our new web editor, it will always remain an up-to-date “work in progress.” We are confident that this two-volume treatise will remain the gold standard resource for sports medicine providers. Finally, we offer our sincere appreciation to Kim Murphy, Cathy Carroll, Dan Pepper, Jodi Kaye and all of the staff at Elsevier Science for their unfailing support in producing this third edition of DeLee and Drez! Jesse Delee David Drez Mark Miller

ixx

C H A P T E R  

 1

Basic Science and Injury of Muscle, Tendon, and Ligament S ecti o n

A

Physiology of Injury to Musculoskeletal Structures 1. Muscle and Tendon Injury Mark R. Brinker, Daniel P. O’Connor, Louis C. Almekinders, Thomas M. Best, Joseph A. Buckwalter, William E. Garrett, Jr., Donald T. Kirkendall, Van C. Mow, and Savio L.-Y. Woo

SKELETAL MUSCLE Structure The primary function of skeletal muscle is to generate force, producing joint and limb locomotion and movement. Muscle also maintains posture, assistsz with joint stability, and generates heat. The muscle moment arm affects the ability to generate joint torques; a larger moment arm requires less muscle force to resist a given externally applied moment. Muscle ­tissue characteristics include excitability, contractility, elasticity, and extensibility. Muscle originates from bone or dense connective tissue, either directly or from a tendon of origin. The muscle fibers pass distally, usually to a tendon of insertion, and connect with bone. This structural framework ­supports the musculotendinous unit against injury and organizes the individual units into tissues and organs (Fig. 1A1-1). The muscle-tendon unit crosses one or more joints. Generally, muscles that cross one joint are located close to bone and frequently are involved more in postural or tonic activity (e.g., soleus). Morphologically, one-joint muscles are broad and flat and have slower contraction and increased strength (force output) relative to two-joint ­muscles. Twojoint or phasic muscles typically lie more superficially (e.g., gastrocnemius, rectus femoris). Compared with onejoint muscles, two-joint muscles are capable of quicker contraction and greater length change but are less effective in ­producing tension over the full range of motion.

Architecture The muscle fiber is the basic structural unit of skeletal muscle, and the sarcomere is the smallest contractile unit of the fiber. Fibers are grouped into bundles known as ­fascicles, which are usually oriented obliquely to the longitudinal

axis. Fiber arrangement within the muscle is variable and includes fusiform, parallel, unipennate, bipennate, and multipennate arrangements (Fig. 1A1-2). Fiber architecture plays a major role in muscle function. In general, fusiform muscles permit greater range of motion. Pennate muscles usually produce more force than parallel-fibered muscles of the same weight, but their maximal velocity of shortening is slower, and the work performed can be ­considerably less.1 A muscle’s force production is proportional to crosssectional area and fiber orientation. Cross-sectional area of muscle is a difficult property to define because there is no location within the muscle belly that is crossed by all fibers. Force production is independent of fiber arrangement when differences in cross-sectional area are considered (Fig. 1A1-3). The fibrous connective tissue network within the muscle is also important. Connective tissue surrounds the whole muscle (epimysium), each bundle of fibers (perimysium), and the individual fibers (endomysium). The connective tissue framework is continuous within the muscle and attaches to the tendon of insertion to produce an efficient means for movement. Tendons provide a wide area for the attachment of muscle fibers.

Myofibrillar Proteins Proteins compose about 12% of the total weight of vertebrate striated muscle. Muscle proteins include myosin, actin, tropomyosin, troponin, and others (Table 1A1-1). Myosin (Fig. 1A1-4) is a hexameric molecule composed of two high-molecular-weight (200,000 Da) heavy chains and four low-molecular-weight light chain subunits (A-1, A-2, and two DTNB units). The myosin molecule can be cleaved by trypsin to yield two fragments, heavy ­ meromyosin and light meromyosin. Papain further cleaves the light 



DeLee & Drez’s Orthopaedic Sports Medicine

Fascicle

Fiber Z Band

Z

Myofiber M

Z

H

I Band

I A

M Line

H Zone

A Band

I

I Band Z Band Figure 1A1-1   Schematic drawing of the structure of striated muscle, showing the organizational framework necessary for effective function. (See text for further explanation of structures.)

­meromyosin fragment into a globular protein, S-1, and a helical protein, S-2. Myosin’s adenosine triphosphatase (ATPase) activity and actin-combining property are associated with the heavy meromyosin component. Myosin’s solubility properties are associated with the light meromyosin fraction. Functionally, heavy chain meromyosin possesses ATPase activity; light chain meromyosin appears to regulate this action but is not essential for ATPase activity. Actin, tropomyosin, and troponin are incorporated into the thin filaments. Of these, actin is present in the largest amount. Actin molecules are small, spherical structures arranged in the thin filaments as if to form a twisted strand of beads. The polarity of actin and myosin molecules is essential to muscular contraction. Tropomyosin and troponin constitute a protein complex that enables calcium to regulate the contraction-relaxation cycle of actomyosin. Tropomyosin molecules (65,000 Da) are long, thin proteins that attach end to end, forming a

A

B

C

D

Figure 1A1-2   Muscle fiber architecture. A, Parallel; B, unipennate; C, bipennate; D, fusiform.

thin filament. Each actin strand carries its own tropomyosin filament, which lies on the surface near the groove between paired actin strands. Together with the troponin myofibrillar protein, they collectively form native tropomyosin, with two main subunits (α and β polypeptide chains, 34,000 and 35,000 Da, respectively) that differ in cysteine content and electrophoretic mobility. The ratio of these two subunits varies among fiber types.2,3 Native tropomyosin makes actomyosin highly sensitive to calcium concentration. At low calcium concentrations, the tropomyosin threads move out of their actin groove and cover the actin region where the myosin cross-bridges attach. When calcium concentrations approach 10−5 M, tropomyosin binds to its target protein, troponin T, resulting in allosteric conformational changes of the troponin-­tropomyosin complex and movement of the tropomyosin further back into the grooves.4 This movement permits actin and myosin to bind, leading to adenosine triphosphate (ATP) hydrolysis and initiation of contraction. Troponin has a globular shape and is located adjacent to the tropomyosin molecule. It is a noncovalent complex of three subunits, troponin T, C, and I, each of which has a distinct physiologic function in muscle. Thin filaments are commonly 1 μm in length, and each troponin-tropomyosin complex is associated with seven actin monomers. C protein is located in the cross-bridge–bearing region of the thick filament. M protein is associated with the enzyme creatine phosphokinase and is localized to the M line in the middle of the thick filament. Titin is an elastic, sarcomeric protein of about 3 million Da that spans the gap between the Z band and the M line in

Basic Science and Injury of Muscle, Tendon, and Ligament

A



C

B Figure 1A1-3   The role of muscle architecture in force development and length change. Length of A is twice that of B;   cross-sectional area of A equals that of B, Maximal force of A is one half that of B, whereas maximal length change of A is twice   that of B. In C, the force is diminished by only a small factor when fibers are arranged in pennate fashion.

the sarcomere. Titin is thought to account for a significant portion of a muscle’s resistance to stretch when extended under relaxed conditions and may play a central role in protecting the muscle against overstretch.5

Ultrastructure Skeletal muscle has dark and light bands (Fig. 1A1-5). The dark bands are composed of thick filaments with small projections, or cross-bridges, extending from the filament. The primary thick filament protein is myosin. Light bands are composed of thin filaments. The primary thin filament protein is actin. The basic functional unit of skeletal muscle is the ­sarcomere, which extends from one Z band to the next and is divided into I bands and A bands. The Z band is composed of at least four proteins: α-actinin, desmin, filamin, and zeugmatin. The I band contains actin, tropomyosin, and troponin. The A band consists of myosin and the actintropomyosin-troponin complex. The light bands narrow with contraction and muscle shortening, whereas the dark bands do not change in length. The H zone, which includes the M line, is a region near the center of the A band. At the cellular level, skeletal muscle is postmitotic and multinucleated, with several hundreds to thousands of nuclei per centimeter of fiber length. Each fiber is divided into an array of overlapping domains. Within each domain, the nucleus controls the structural proteins. Any increase in fiber size must be accompanied by an increase in nuclei. Inhibiting the increase of nuclei inhibits the growth of the fiber.6 Nuclei arise from undifferentiated myogenic cells called satellite cells that lie underneath the basal lamina of

  Table 1A1-1  Relative Proportions of Myofibrillar Proteins in Rabbit Skeletal Muscle Protein Myosin Actin Tropomyosin Troponin C protein M proteins α-Actinin β-Actinin

Percentage of Total Structural Protein 55 20 7 2 2 <2 10 2

Adapted from Carlson FD, Wilkie DR: Muscle Physiology. Englewood Cliffs, NJ, Prentice-Hall, 1974.

muscle fibers. The daughter cell resulting from mitosis of a satellite cell becomes the newest nucleus of the muscle fiber.7 Inhibition of this division appears to halt muscle growth. The density of nuclei (nuclei/fiber length) is related to the fiber type. Type I (slow) fibers have a higher density of nuclei than type II (fast) muscle fibers. When computed per unit of volume rather than length, the larger type II (fast) muscle fibers contain more nuclei than the type I fibers.

Sliding Filament Model The sliding filament model of muscle contraction states that an arrangement of rods or filaments is stacked parallel to the long axis of the muscle. These rods are composed of thick filaments of myosin. Another set of filaments, the thin filaments, extends from the Z band through the I band and part way into the A band, stopping short of the H zone. The I band decreases with sarcomere shortening during contraction, but the A band length and the actin and myosin filaments do not change. These changes in sarcomere pattern conform with the expected mechanical behavior of two sets of filaments sliding past one another.

Activation of Contraction: Sarcoplasmic Reticulum Regulation of skeletal muscle contraction is accomplished primarily by release of calcium that is stored in the sarcoplasmic reticulum (SR). The SR has two distinct units, the longitudinal tubules, which extend outward to form two large lateral sacs in the region of the Z band, and the transverse tubules, which bisect the longitudinal tubules. Two lateral sacs and their corresponding transverse tubule are referred to as a triad. Mammalian skeletal muscle has two triads per sarcomere (Fig. 1A1-6). An action potential passes over the sarcolemma into the transverse tubular system and longitudinally into the sacs, resulting in an increase in permeability and a release of calcium ions into the sarcoplasm. This free calcium binds to troponin C, causing a change in troponin I that permits interaction of the thick and thin filaments. Cross-bridge cycling continues as long as the free calcium concentration is maintained. After completion of the electrical event, relaxation of the muscle occurs by active transport of calcium into the longitudinal tubules of the SR, forcing calcium to dissociate from troponin and allowing tropomyosin to move back into the groove between actin strands, where it prevents any further cross-bridge attachment.

DeLee & Drez’s Orthopaedic Sports Medicine



Figure 1A1-4   Schematic drawing of   myosin molecule showing subunit structure. The heavy chain component possesses the adenosine triphosphatase activity, and the light chain confers the solubility properties of the molecule. DTNB, 5’-dithiobis-(2-nitrobenzoic) acid; HMM, heavy meromyosin; LMM, light meromyosin.   (Adapted from Carlson FD, Wilkie DR: Muscle Physiology. Englewood Cliffs, NJ, Prentice-Hall, 1974.)

HMM HMM S-2

HMM S-1

LMM DTNB light chains

Light chains α1 and α2

Site of trypsin splitting

Fiber-Type Differences Different striated muscles exhibit significant variations in innervation, physiology, biochemistry, and circulation. Currently, two major classes of distinct fiber types are recognized: type I and type II (Table 1A1-2).8 The type I, or

A

M Line H Band

Z Band

A I Band Band Figure 1A1-5   Electron micrograph (A) and schematic drawing (B) of skeletal muscle.

B

slow-twitch oxidative, fibers have the slowest ­contraction time and the lowest content of glycogen and glycolytic enzymes. They are rich in mitochondria and myoglobin and are fatigue resistant. Morphologically, their sarcomeres contain a wide Z band. The type II fibers have two subclasses. The type IIa, or fast-twitch oxidative glycolytic, fibers have a faster contraction time but less resistance to fatigue than type I fibers. Type IIa fibers have a higher content of mitochondria and myoglobin and are more fatigue resistant than the type IIb, or fast-twitch glycolytic, fibers. Type IIa fibers have high levels of myosin ATPase, oxidative enzymes, and glycogen. Morphologically, the type IIa Z bands are slightly more narrow than those in type I fibers. The type IIa fiber is termed an intermediate fiber because it possesses both forms of myosin that are present in the type I and IIb fibers. Type IIb fibers have high levels of glycogen and glycolytic enzymes and contract the fastest but are the least fatigue resistant. Most mammalian carnivores, including humans, possess a third subclass, the IIc, or superfast, fiber, most prominently in the jaw muscles. Type IIc fibers possess a unique different form of myosin from those characteristic of type I and type II fibers. At birth, 10% of muscle fibers may be classified as type IIc; this declines to about 2% after the first year of life. During physical training, 10% of these fibers may be present in some muscles of endurance athletes.9 Their presence has yet to be explained, although it is believed by some that this fiber type is an undifferentiated, transitional form between type I and type IIa fibers.10 Biochemically, the different fiber types have distinct isomers of myosin, tropomyosin, and troponin in the sarcomere,11,12 with considerably more heterogeneity than had been expected previously. Rather than three distinct fiber types, there appears to be a discrete number of basic sets of structural protein isomers for fast and slow muscles. The difference in the sensitivity of myosin to ­ retaining or losing ATPase after exposure to either high or low pH is a reliable method for classifying muscle fibers histochemically.13,14 All human muscles are composed of a mixture of fiber types; those muscles with similar function have similar fiber populations.9,15-19 The mean fiber composition in most human muscles is 50% slow-twitch and 50% fast-twitch fibers, although some muscles have a predominance of either slow-twitch or fast-twitch fibers.

Basic Science and Injury of Muscle, Tendon, and Ligament

Myofibrils

Transverse tubule



Figure 1A1-6   Details of the sarcoplasmic reticulum, the system of membranes responsible for transmission of the electrical signal from one muscle cell to the next. Each electrical signal passes inward along the transverse tubule, causing release of calcium from the lateral sacs.

Lateral sacs Cisternae

Triad

High-performance athletes tend to have fiber compositions that would appear to be advantageous for their particular event. Endurance athletes tend to have high percentages of type I fibers, whereas those in nonendurance events (e.g., sprinters) tend to have higher percentages of fast-twitch fibers.9,20,21 Subjects with a higher percentage of type II fibers in the quadriceps have been shown to generate more knee extensor force and torque at higher velocities than subjects with fewer type II fibers.22 Considerable variation exists in the fiber composition of skeletal muscle within subgroups of athletic ability, ­ suggesting that factors other than fiber-type composition contribute to performance.

Physiology Central Nervous System Control of Force Production Two basic strategies allow the central nervous system to control force production: either the rate of discharge of ­ firing neurons can be increased, a strategy known as ­temporal summation, or the number of active motor units

can be increased, a strategy known as spatial summation or recruitment.

Mechanical Events: Twitch and Tetanus Electrical stimulation of a muscle is not followed by the immediate development of force but rather by a latency period lasting about 15 msec during which the muscle, maintained at a fixed length, produces no force. A short-lived fall in tension may occur before positive tension is developed.23 A single stimulus produces a single transient rise in tension, known as a twitch. Two twitches separated by an appropriate time interval result in two identical force recordings. Increasing the frequency of stimulation produces a summation of the force recordings known as unfused tetanus (Fig. 1A1-7). Tetanic fusion, at which an increase in stimulus frequency results in no further increase in force, occurs at a frequency of 50 to 60 Hz in mammalian skeletal muscle at body temperature. Classic experiments in muscle physiology demonstrated the relationship between muscle length and tension.24 The passive tension curve was measured on the resting muscle at a series of different lengths, whereas the active

  Table 1A1-2  Histochemical Reactions of Human Skeletal Muscle Muscle Fiber Type I

IIa

IIb

IIc

Fatigue Resistant

Fatiguing

Superfast Twitch

3+ 0 0 2+ 2+ 2+ 3+ 3+

3+ 3+ 0 1+ 1+ 2+ 2+ 3+

3+ 3+ 2+ 2+ 2+ 1+ 2+ 3+

Fast Twitch Example Routine ATPase ATPase preincubated pH 4.6 ATPase preincubated pH 4.3 NADH-TR SDH Glycerophosphate-menadione-linked PAS Phosphorylase

Slow Twitch 1+ 3+ 3+ 3+ 3+ 0 1+/2+ 1+/0

ATP, adenosine triphosphatase; NADH-TR, nicotinamide adenine dinucleotide trireductase; PAS, periodic acid–Schiff stain; SDH, succinate dehydrogenase. From Dubowitz V, Brooke MH: Muscle Biopsy: A Modern Approach. London, WB Saunders, 1973.

DeLee & Drez’s Orthopaedic Sports Medicine



TWITCH

4

PAIRED TWITCH

3

s2

s1

s1

UNFUSED TETANUS

TENSION

Total s2

2 Active Passive

FUSED TETANUS

1 Lo 0

s1

s2

s3

s4

s5

s1

s2

s3

s4

s5

Figure 1A1-7   Twitch and tetanus. As the frequency of stimulation is increased, muscle force rises to an eventual plateau level known as fused tetanus.

(tetanized) curve was measured as the muscle was maintained in isometric contraction at a series of constant lengths. Figure 1A1-8 shows that the tension is greater when the muscle is tetanized than when passively stretched to the same length.

Motor Unit The second mechanism for controlling force production involves motor unit recruitment. The motor unit ­classically is defined as an α-motoneuron and the specific muscle fibers it innervates.25-27 A single α-motoneuron can innervate between 10 and 2000 muscle fibers.28 All fibers within a motor unit are homogeneous with regard to histochemically identifiable contractile and metabolic properties.29-31 The fibers in a motor unit generally are distributed throughout the muscle.32 Fibers within a motor unit rarely are located adjacent to each other. The motor end plate has a characteristic location on the muscle fiber at about the midpoint of the length of the fiber, and fiber density decreases as a function of the distance from the motor end plate.33,34 One of the important areas of research in skeletal muscle physiology involves the systematic response of the different motor units to distinct physiologic tasks. The existence of an orderly procedure for motor unit recruitment by the central nervous system is based on the size principle.35-39 Basically, there appears to be a continuum of thresholds for motor unit activation based on motor unit size. Slowtwitch motor units are innervated by small, low-threshold α-motoneurons, whereas fast-twitch motor units are innervated by larger, higher threshold α-motoneurons. This size principle helps to account for the efficiency of skeletal muscle movement and locomotion. The relative importance of increased firing rate (temporal summation) and recruitment (spatial summation) as mechanisms for increasing the force of voluntary contraction has been investigated. In humans, only at low levels of force output is recruitment the primary mechanism for increased force production.40 An increase in firing rate becomes more important at intermediate (>50% maximal force) force levels

0

1 2 LENGTH (fraction of optimum) Figure 1A1-8   Tension-length curve of skeletal muscle.   L0, rest length.

and thereafter accounts for most of increased force production (Fig. 1A1-9). The common drive principle accounts for the behavior of the firing rates of motor units and provides a simple explanation for the control of motor unit activation.41 Common drive explains that the nervous system does not control the firing rates of motor units individually; rather, it acts on the pool of motoneurons in a uniform fashion.

Contractile Properties The basic feature that differentiates motor units is their contractile properties: time to peak tension in a twitch and the one-half relaxation time. A slow-twitch motor unit possesses a relatively long time to peak tension, whereas a short time to peak tension is characteristic of the fasttwitch motor unit. A prime determinant of a muscle’s time to peak tension is the rate at which ­myosin splits ATP into adenosine diphosphate (ADP) + inorganic phosphate (Pi). The respective enzyme is referred to as actin-­activated myosin ATPase or, more commonly, ATPase.42-45 Under normal physiologic conditions of high concentrations of magnesium, the enzymatic activity of myosin is reduced greatly. High rates of ATP hydrolysis during muscle contraction are due to interactions that remove the inhibitory effects of the divalent cations magnesium and calcium. Differences in the specific ATPase activities of myosin are due to the existence of several isoforms of the protein.14,46,47 Traditionally, myosin isoenzymes have been identified based on their susceptibility to loss of ATPase activity in response to an alteration in pH.46,47 Myosin from slow-twitch muscles is acid stable but alkaline labile, whereas the opposite is true for fast-twitch muscles. The sliding filament theory of skeletal muscle contraction argues that myosin is the most important contractile protein because it hydrolyzes ATP to yield energy for formation of the actomyosin complex. The activity of the ATPase of myosin correlates closely with the intrinsic speed of muscle shortening, thus demonstrating a link between the contractile and biochemical properties of skeletal muscle (Fig. 1A1-10).42 Because the rate-limiting step of muscle contraction appears to be the rate of energy delivery from

30% 0

100

100

0

50

50 33%

0

100 0

100

50

50 39% 0

1 VOLUNTARY FORCES (kg)

100 2

TOTAL PERCENTAGE EXTRA FORCE ACCOUNTED FOR

50

50

0

A

0

100

PERCENTAGE EXTRA FORCE FROM INCREASED RATE

PERCENTAGE EXTRA FORCE FROM RECRUITMENT

Basic Science and Injury of Muscle, Tendon, and Ligament

20 10 16% 0 20 10 14% 0 20 10 10% 0

B



0

1 2 VOLUNTARY FORCES (kg)

Figure 1A1-9   A, Calculated percentage of increase in force due to recruitment and increased firing rate for three subjects. B, Calculated total percentages of force accounted for by the units studied at various force levels.   (From Milner-Brown HS, Stein RB, Yemmon R: Changes in firing rate of human motor units during linearly changing voluntary contractions. J Physiol [Lond] 230:371-390, 1973.)

ATP hydrolysis, one can observe changes in the ATPase of myosin to study alterations in contractile characteristics. Much of skeletal muscle regulation and physiologic adaptation to a given stimulus involves the myosin molecule.48

Adaptability of Mammalian Skeletal Muscle

Actin Activated ATPase - µm P(g × sec)−1

Physiologic overload of skeletal muscle can result in ­adaptation of all components of the motor unit, including the muscle tissue, the neuromuscular junction, and the

30

Cross-Reinnervation and Electrical Stimulation

25 20 15 10 5 0

corresponding α-motoneuron.26 The plasticity of ­skeletal muscle has been studied under different conditions to show that the different fiber types within a muscle adapt to ­various forms of overload in several ways.49-59 Each of these adaptations results in a logical alteration of the ­morphology and the function of the muscle fiber. Muscle is thought to adapt to the function it performs, which in turn implies that specificity becomes an important factor in any type of functional overload, including exercise training. A variety of stimuli, including cross-reinnervation, electrical stimulation, hypergravity stress, thyrotoxicosis, compensatory hypertrophy, and exercise, have been used to study and elucidate mechanisms for this adaptive response.

0 5 10 15 20 25 30 Contractile Speed - Muscle Lengths/sec

Figure 1A1-10   Relationship between maximal speed of shortening and actin-activated myosin adenosine triphosphatase (ATPase) from a variety of animal species.   (From Barany M:   ATPase activity of myosin correlated with speed of muscle shortening. J Gen Physiol 50:197-218, 1967, by permission of the Rockefeller University Press.)

The pattern of muscle stimulation plays an important role in determining the functional properties of skeletal muscle.54 Cross-reinnervation experiments have shown that the isometric twitch speed of cat skeletal muscle is determined largely by the motor innervation it receives.50 Motoneurons innervating slow-twitch skeletal muscle generate a sustained, low-frequency pattern of activity; motoneurons innervating fast-twitch muscles generate intermittent bursts of more intense activity.60 Transformation of a fasttwitch muscle into a slow-twitch muscle can be brought about with unaltered innervation by transmitting to the nerve (through implanted electrodes) a frequency pattern that normally is delivered to a slow-twitch muscle.61 Neural influences are also largely responsible for the reciprocal changes that occur in functional59 and biochemical62 properties. Without a change in the frequency pattern of electrical impulses reaching a muscle, crossreinnervation has no significant effect on force-velocity

10

DeLee & Drez’s Orthopaedic Sports Medicine

properties of skeletal muscle. These observations discount the theory that a chemotrophic substance affects the physiologic and biochemical differences between fast-twitch and slow-twitch muscle.

Compensatory Hypertrophy Several researchers have induced functional overload in skeletal muscle by ablating synergistic muscles and studying the remaining intact system. This compensatory hypertrophy model has been used to stimulate histochemical and biochemical changes in skeletal muscle fibers. Experiments involving compensatory hypertrophy of the fast-twitch rat plantaris muscle show a decrease in calcium-activated ATPase activity, an increase in the number of histochemically determined alkaline-labile fibers, and alteration of the myosin light chain pattern.49,52,56,57 Changes induced by the compensatory hypertrophy model are not as complete as the changes observed with cross-reinnervation and electrical stimulation. Because compensatory hypertrophy does not directly affect the intact nerve, it more closely represents an ideal physiologic situation. The factors operating in the compensatory hypertrophy model to effect fiber-type transformations are likely the same as those associated with hypergravity stress because neural input is not ­manipulated in either model. Changes in the soleus and plantaris muscles of rats maintained for 6 months under hypergravity stress reinforce Lomo’s54 conjecture that the pattern of electrical stimulation determines the contractile properties of skeletal muscle. Other models have shown that muscle is capable of responding to a hormonal stimulus.51,53

Exercise Training programs are based on the principles of overload, specificity, and reversibility.63 Overload means that a certain level of stimulus is necessary for adaptation to occur. The specificity concept implies that specific stimuli result in specific responses, that specific training leads to specific adaptations, and that fatigue is specific to the type of exercise performed. Reversibility implies that the effects of training can be reversed with a change in the training stimulus. The adaptive changes of skeletal muscle to endurance exercise are well studied.64,65 Endurance exercise is characterized by activation of large muscle groups that generate high metabolic loads, resulting in adaptation of the respiratory and circulatory transport systems as well as the enzymatic capacity of the muscle. Skeletal muscle has a tremendous potential for adaptation in oxidative potential with endurance training. In certain circumstances, this type of training can double the oxidative capacity of skeletal muscle. For example, a 5-month bicycle ergometer program of 1 hour per day 4 days a week at a load requiring about 75% of maximal oxygen power doubled succinyl dehydrogenase (SDH) and phosphofructokinase (PFK) enzymatic activities.65 By contrast, anaerobic capacity, as measured histochemically by α-glycerophosphate dehydrogenase activity, was increased in the fast-twitch fibers only.65 Mean oxygen uptake increased 13%, and muscle glycogen was 2.5 times higher than before training. The percentages of slow-twitch or fast-twitch fibers, as identified from myosin ATPase activity, were unaffected.

High-force, low-repetition training results in an increase in muscle strength that is proportional to the tissue’s crosssectional area. Some debate exists with regard to whether this increase in cross-sectional area is due to muscle hypertrophy (an increase in the size of the muscle fibers) or to muscle hyperplasia (an increase in the number of muscle fibers). At present, it appears likely that most of the change is due to hypertrophy. Along with an increase in the size of the fibers, an increase in the amount of contractile proteins, particularly myosin, occurs. The contributions of ­ hyperplasia (i.e., fiber splitting, fiber branching, and fiber fusion) to the increase in muscle mass are yet to be determined. There is also a strong neurologic component in strength training. Initial responses to training include an alteration in central nervous system firing of motor neurons to produce a more synchronized, and thus more ­effective, recruitment of muscle neurons. Prolonged endurance exercise in athletes increases skeletal muscle capillary density.66 This increase in capillary density is highly correlated with the improvement in whole-body maximal oxygen consumption. Another predominant effect of endurance training on skeletal muscle fibers is a marked increase in volume and density of the mitochondria. Prolonged resistance exercise, such as long-term weightlifting and powerlifting, appears to produce fasttwitch fiber hypertrophy, which results in reduced capillary density.67 The percentages of fiber-type composition in human skeletal muscle vary considerably among muscles and among individuals. Certain compositions of fiber types would logically appear advantageous for particular athletic events. Saltin and others9 were the first to show the conversion of type IIb to type IIa fibers; however, the ratio of type I to type II fibers remained constant. Studies reporting increases in the percentage of red compared with white fibers employed oxidative capacity, as measured by SDH or DPNH-diaphorase activity, to classify fibers. In no instance in humans has a change been shown in fiber characteristics as determined histochemically by myosin ATPase. A transition from fast-twitch to slow-twitch fibers can be brought about by increased contractile activity in animal models.68,69 Similar to long-term nerve stimulation, high-intensity endurance running leads to transition of fast-twitch to slow-twitch fibers in the sarcoplasmic reticulum, a decrease in fast-type myosin light chains, and an increase in slow-type myosin light chains. Increased contractile activity may induce changes that are qualitatively similar to changes seen in long-term nerve stimulation. Furthermore, endurance training not only affects the metabolic properties of the muscle fiber but also produces fast to slow transitions in the Ca2+-handling system. As judged by conventional histochemical techniques, type IIb fibers decreased, whereas type IIa and I fibers increased in the plantaris, extensor digitorum longus, and vastus lateralis muscles. Increased contractile activity, brought about in a physiologic manner, is capable of inducing fiber-type transitions in certain instances.68 Similar transformation of type IIb into type IIa fibers and of type IIa into type I fibers have been identified in sedentary human subjects.70 These fiber-type changes were depicted by histochemical staining of myofibrillar ATPase and were accompanied by

Basic Science and Injury of Muscle, Tendon, and Ligament

an enhancement of the oxidative capacity in all fiber types. Alteration in fiber types with exercise has yet to be shown in highly trained athletes. Skeletal muscle fibers possess the potential for synthesis of all types of myofibrillar proteins. Skeletal muscle fiber transformations appear to occur only when a departure from normal function is severe and sustained. The exact duration, frequency, and magnitude of stimulus required for fibertype transformations are uncertain. The forms of overload discussed here represent nonphysiologic ­conditions and are not characteristic of physical training. For well-trained athletes, exercise may be unable to meet the necessary requirements to bring about a change in fiber type.

Muscle Injury and Repair Muscle injury results from several mechanisms that are governed by separate pathologic processes. Not enough basic research regarding muscle injury has been done to elucidate the precise changes that occur with injury, and experimental and scientific data regarding prevention and rehabilitation are lacking. A predictable set of events occurs in response to muscle injury, although the molecular mechanisms that regulate and control these events are not well understood. Muscle fiber regeneration begins with the satellite cell, a quiescent cell that is activated with inflammation and repair. These cells are located between the basal lamina and the plasma membrane of individual myofibers. During muscle regeneration, trophic substances released by the injured muscle presumably activate the satellite cells.71 A host of growth factors and cytokines has been shown to cause proliferation of satellite cells and their transformation into myotubes and muscle fibers.72 This process depends in part on the prostaglandin-mediated cyclooxygenase-2 pathway; for instance, cellular muscle repair mechanisms are compromised in cyclooxygenase–2–deficient mice.73 The importance of the cyclooxygenase-2 pathway to muscle repair should be considered when deciding whether to use nonsteroidal anti-inflammatory medications following muscle injury. The postinjury response of the satellite cell appears similar to the process of fetal development, suggesting that the expression of myosin heavy chain isoforms provides a useful marker for regeneration. The time between injury and the initiation of proliferation is affected by several factors, including the type of injury and the metabolic state of the muscle.74 In addition to the regeneration of damaged fibers, successful repair includes the synthesis of collagen. Some connective tissue production is necessary for restoration of the tissue’s tensile strength and architecture; however, animal models have shown that certain conditions, such as immobilization, lead to fibrosis and scarring of the muscle.75 As more is learned about what factors control scar formation, treatment practices may be changed.

Muscle Laceration Laceration of muscle is much more common in trauma than in athletics. After laceration and repair, the two segments of the muscle heal by dense scar formation.76 Muscle does not regenerate across the scar, and functional continuity is

11

Transected Isolated Proximal nerve

Figure 1A1-11   Schematic drawing of lacerated muscle. The laceration leaves fibers intact proximally and distally while dividing the central fibers. Scar tissue isolates the distal segment from its nerve supply.   (Redrawn from Garrett WE Jr, Seaber AV, Bokswich J, et al: Recovery of skeletal muscle following laceration and repair. J Hand Surg [Am] 9[5]:683-692, 1984.)

not restored. The muscle segment isolated from the motor point loses its innervation. After healing, the segment isolated from the motor point develops the histologic picture of denervated muscle. Electrical mapping studies show that muscle activation does not cross the scar.76 Consequently, the muscle loses a significant proportion of its ability to produce tension. Partial lacerations decrease the ability of the muscle to generate tension, but to a lesser degree than muscles that are transected completely. The isolated segment may be able to transmit force and to shorten, but the active contractile function of the muscle remains only in the portion with an intact nerve supply (Fig. 1A1-11). Treatment should stress the repair or reconstruction of a muscle using its long tendons of origin and insertion as well as the epimysial connective tissue to anchor the repair; muscle tissue alone is inadequate for suture repair.

Muscle Cramps Ordinary muscle cramps are common during and after athletic exercise and are frequent in young healthy people not involved in athletics. Cramps occur most frequently in the gastrocnemius complex and can arise during exercise, at rest, or while asleep. The cause of muscle cramps is uncertain. Their onset frequently follows contraction of shortened muscles. The cramp often originates as fasciculations from a single focus or several distinct foci within the muscle and then spreads throughout the muscle in an irregular pattern. Electromyographic studies reveal fascicular twitching in a single focus, followed by high-frequency discharges within the muscle fibers.77 The entire motor unit is involved, and the initiating source is within the motor nerve fiber rather than within the individual muscle fibers. Specifically, the focus is thought to be located in the terminal arborizations of the motor nerve fibers. Layzer78 supported these findings on peripheral motor nerve involvement and suggested that the disturbance could arise from hyperexcitable motor neurons in the spinal cord. Despite their common occurrence, the cause of muscle cramps during exercise remains poorly understood. Many of the studies to date have been conducted in ultraendurance athletes to investigate the proposed mechanisms, which include dehydration, electrolyte disturbances, and muscle fatigue. Maughan79 followed 90 competitors at the 1982 Aberdeen marathon and found no correlation between hydration status and electrolyte balance and the incidence of muscle cramps.

12

DeLee & Drez’s Orthopaedic Sports Medicine

Ordinary muscle cramps are also associated with a variety of conditions unrelated to exercise. Excessive sweating or diuresis can cause saline loss and may produce cramps. Patients in renal failure who are on long-term hemodialysis often have muscle cramps. These conditions may be related to an alteration in sodium concentration, and administration of a saline solution sometimes is helpful.80 Low levels of serum calcium or magnesium have been implicated.77 Neither of these ionic disturbances is necessarily present in muscle cramps after exercise, however. The cramp sometimes can be interrupted by forceful stretching of the involved muscle or activation of the antagonistic muscle. After resolution of the knotted and painful contraction, the muscle shows evidence of altered excitability and fasciculations for many minutes after the cramp. The muscle may be painful for several days after the event. Electrolyte and hydration balance is thought to be helpful in preventing cramps, although the value of this treatment has not been proved. Drugs have been more helpful in treating cramps occurring in nonathletic individuals. Quinine sulfate and chloroquine phosphate have been beneficial, particularly for night cramps.81,82 The use of these medications to prevent or control exercise-induced muscle cramps is questionable at this time.

Delayed-Onset Muscle Soreness Muscle pain after unaccustomed vigorous exercise is a ­common phenomenon in athletes. Muscle soreness is especially marked after the initiation or resumption of training after a period without training. This pain should be distinguished from discomfort occurring during exercise, which is often associated with muscle fatigue. Typically, delayed-onset muscle soreness begins hours after exercise and is prominent on the first and second days after activity. The painful areas are typically located along the tendon or fascial connections within the muscle. Several different pathologic mechanisms have been proposed to explain delayed-onset muscle soreness. Hough83 reported that delayed pain did not necessarily follow more fatiguing work. He found that exercise routines that produced considerable fatigue did not produce as much delayed-onset muscle pain as high-intensity, rhythmic contractions marked by relatively little fatigue or exercises associated with a sudden contraction or jerk. Hough83 proposed that the soreness and reduction in the ability of the muscle to produce tension could be explained by small ruptures within the muscle. Asmussen84 found that negative (eccentric) work produced more delayed-onset muscle soreness than positive (concentric) work, despite the greater fatigue induced by positive work. He concluded that the pain was due primarily to mechanical stress rather than fatigue or metabolic waste products and believed that the location of the injury was the connective tissue within the muscle rather than the muscle fibers.84 Abraham85 investigated the hypothesis that connective tissue breakdown might be associated with delayed-onset muscle soreness by monitoring hydroxyproline, a modified amino acid found almost exclusively in collagen. After a weightlifting ­ program, a significant increase in urinary hydroxyproline occurred in subjects experiencing delayed muscle

s­ oreness.86 Elevated levels of myoglobin excretion were also noted in subjects who developed pain as well as in subjects who did not develop pain. A correlation thus exists between muscle soreness and collagen breakdown but not between ­soreness and muscle breakdown. Serum ­lactic acid concentration is not related to exercise-induced delayed muscle soreness. An alternative theory of muscle soreness implicates muscle spasm and electrical activity as the cause of pain rather than breakdown of connective tissue or muscle fibers.87,88 DeVries87 proposed that exercise produces ischemia that subsequently causes pain, which initiates reflex tonic muscle contraction, which prolongs the ischemia. Using quantitative electromyography, DeVries87 showed that muscular activity can be present when pain is present. Stretching of the muscle diminished the pain and the electromyographic activity. Abraham85 reinvestigated the electromyographic data and was unable to show significant differences in subjects with and without muscle soreness. The weight of the evidence seems to be with the tissue injury theory of muscle damage as a cause of delayed muscle soreness. Electromyographic changes may accompany the tissue injury, so treatment that alters the muscle spasm or electromyographic manifestations may be of benefit for delayed-onset muscle soreness. Studies of delayed-onset muscle soreness have ­investigated changes that occur on an ultrastructural level (Fig. 1A1-12). Electron microscopy of muscle in subjects with pain in the vastus lateralis after cycling showed ­significant alterations in the sarcomere and the cross­striations.89 Three days after heavy exercise, 50% of the muscle fibers displayed disorganization of the ­myofibrillar material. Armstrong90 disputed these findings, showing histologic evidence of injury to less than 5% of muscle fibers active during exercise. It is known that a loss of desmin labeling occurs rapidly after eccentric exercise.91 Z-band streaming, A-band disruption, and myofibril disorganization can be seen within 10 minutes of exercise.92 There appears to be general consensus that the initial injury causing delayed-onset muscle soreness results from mechanical damage or overstretch of the contractile apparatus.93 An inflammatory response quickly follows that is characterized by invasion of neutrophils and the release of cytokines that attract additional inflammatory cells. An unexplored and especially important aspect of inflammation following injury is the role of inflammatory cells in extending injury and possibly directing repair. Neutrophils may promote further damage through the release of oxygen free radicals and lysosomal proteases. Macrophages subsequently invade damaged fibers to phagocytose cellular debris. It is becoming increasingly evident that neutrophils and macrophages play a role in muscle repair, at least in laboratory models.94,95 The clinical significance of these findings awaits further investigation. The pain with delayed-onset muscle soreness ­generally is described as a dull ache that may be localized to the musculotendinous junction or experienced throughout the muscle. Other clinical findings may include muscle and joint stiffness, swelling, and decreased joint range of motion. A marked reduction in maximal muscle force and power production is typical and may be due to inhibition of voluntary effort secondary to pain as well as to a decline

Basic Science and Injury of Muscle, Tendon, and Ligament

13

Figure 1A1-12   Transmission electron micrographic (TEM) and light microscopic (LM, inset) views of injured rat soleus muscle fibers immediately after downhill walking. Note the A band disruption and the Z band damage.   (TEM from Dr. R. W. Ogilvie. LM from Armstrong RB, Marum P, Tullson P, et al: Acute hypertrophic response of skeletal muscle to the removal of synergists. J Appl Physiol 46:835-842, 1975.)

in the intrinsic force-generating capacity of the muscle as a result of myofiber damage. In some animal models, the decrease in force production follows a biphasic pattern with a secondary loss of muscle contractile force at 48 hours after injury.96 Delayed-onset muscle soreness typically occurs after any unaccustomed physical activity with a strong eccentric component. Prior training with eccentric exercise has been noted to protect against similar injury in humans,97 rats,98 and mice.99 Aging has important effects because muscles from older rats show more injury and recover more slowly than muscles from younger rats.100

exist to determine which pathologic processes are involved. The treatment regimen ­generally includes rest and ice with an early return to gentle motion.89 Prolonged immobilization has been associated with longer periods of disability than shorter delays in restoration of motion. Active and passive motion should be emphasized, and care is necessary in therapy to avoid reinjury. A study conducted at the U.S. Naval Academy reported that using early immobilization in 120 degrees of knee flexion for the first 24 hours after the injury allowed return to full athletic activity in an average of 3.5 days.104 The general efficacy and effectiveness of such treatment remains to be ­evaluated.

Muscle Contusions

Myositis Ossificans

Direct trauma to muscle is a common athletic injury, ­particularly in contact sports. Damage and partial disruption of muscle fibers occur, and intramuscular hematoma frequently results. Direct trauma may affect any muscle, but contusion of the quadriceps and gastrocnemius muscles are most prevalent. The injuries are characterized by tenderness, diffuse swelling or a palpable hematoma, and limitation of motion and strength.

A complication of muscle contusions is the occurrence of myositis ossificans tissue calcification or ossification at the site of injury. Heterotopic bone may form in 20% of patients with a quadriceps hematoma.105 The pathogenesis of heterotopic bone formation is poorly understood. A major risk factor is reinjury during the early stages of recovery.106 Myositis ossificans usually becomes radiologically ­evident 2 to 4 weeks after injury and often extends to the underlying bone.101 The mass may enlarge or may be symptomatic for several months before stabilizing. History of previous contusion is important because the mass and its radiographic appearance can mimic osteogenic ­sarcoma (Fig. 1A1-13). The histologic features may be similar if a biopsy is performed early in the course of myositis ­ossificans. Heterotopic bone often resorbs with time. Normal function is often possible before complete resorption, but the recovery period is longer than that of an uncomplicated contusion. No specific treatment is recommended in

Quadriceps Contusions Adequate acute treatment of muscle injuries is important to limit hematoma formation and inflammation. Jackson and Feagin101 reviewed quadriceps injuries occurring in military cadets and found them to be a significant cause of ­athletic and occupational disability. The initial injury severity, based on range of motion, correlates well with the severity and the duration of disability.102 The possible pathologic mechanisms were described by Ryan,103 but few scientific data

14

DeLee & Drez’s Orthopaedic Sports Medicine

of diastolic blood pressure.108 These objective measures are used to increase the reliability of the diagnosis.

Acute Compartment Syndromes

Figure 1A1-13   Radiograph of myositis ossificans of the rectus femoris. This 22-year-old patient suffered a quadriceps contusion that resulted in the condition shown. The resultant heterotopic bone gradually resorbed over time.

addition to the treatment for the contusion. Early surgery may exacerbate heterotopic bone formation and prolong disability. Surgery may be considered later in the disease course to remove the heterotopic bone only if it is causing symptoms and has matured, and no clinical or radiologic improvement can be noted.107

Compartment Syndromes Compartment syndrome is a pathologic condition of skeletal muscle characterized by a rise in intracompartmental pressure above capillary pressure, thus restricting capillary blood flow. Many factors can cause a rise in intracompartmental pressure. The most common cause is muscle edema following transient muscle ischemia. Hemorrhage or direct trauma to muscle can also result in a compartment syndrome. If recognized early, the elevated pressure can be relieved by incising the investing fascia, restoring the circulation and function of the compartmental muscles and neurovascular components. The pathophysiology of compartment syndromes involves increased intracompartmental fluid from hemorrhage or intracellular or extracellular edema. The fluid increases intracompartmental pressure, compromising capillary perfusion and subjecting the muscle within the compartment to ischemic injury. The level of pressure that interferes with capillary circulation is less than that of the major vessels within a compartment. The presence of a pulse distal to the compartment thus does not rule out a compartment syndrome. Clinical evaluation relies on the presence of pain, ­particularly with passive stretch of the muscles in the involved compartment, increased compartment pressure noted with palpation, and altered nervous function as noted by ­paresthesias in the sensory distribution of nerves within the compartment. Compartment pressures within the compartment can be measured by needle manometer,108 wick catheter,109 solid-state transducer,110 and noninvasive ­auscultation. Significant muscle damage can occur with pressures above 30 to 40 mm Hg20 or within 10 to 30 mm Hg

Acute compartment syndromes have been associated with direct trauma to bone or soft tissue. Tibial shaft fractures comprise a large proportion of fractures with compartment syndromes. Direct soft tissue injury and muscle trauma also can result in elevated pressure that compromises tissue perfusion. Indirect injury secondary to exertion is another cause of compartment syndrome. Indirect injuries can be acute or chronic. The acute syndromes are not well understood, but several factors may contribute. Intense muscular activity causes a large rise in interstitial pressure; intermittent pressure levels of greater than 100 mm Hg are common during some forms of exercise.110 Muscle perfusion during such exercise is possible only intermittently between muscular contractions. Increasing exercise also causes a muscle volume increase up to 20% as a result of increased blood content and intracompartmental fluid accumulation. Acute exertional compartment syndromes are uncommon in sports.109 They are typically associated with intense muscular contractions in individuals unaccustomed to such activity, such as military recruits in basic training. Direct measurement of compartment pressures is preferred, although if unavailable, surgical decompression of the muscle and neurovascular components by fascial release probably should be undertaken.

Chronic Compartment Syndromes Chronic exertional compartment syndromes occur more frequently than acute forms. The presenting complaints are usually diffuse pain or a deep ache over the anterior or lateral compartment of the leg, usually after a relatively long exercise period. The pain usually is severe enough to interrupt the activity or reduce the intensity of the exercise. The symptoms are often bilateral, and sensory changes may be present. Occasionally, muscle hernias may be present near the fascial opening through which the distal branch of the superficial peroneal nerve passes (Fig. 1A1-14). Chronic compartment syndrome is difficult to diagnose clinically. Corroboration with objective pressure measurements is the gold standard for diagnosis. Resting pressure values may be slightly higher, but the primary characteristic is pressure elevation above normal during exercise and a slower return to resting value at the end of exercise. A resting pressure of greater than 12 mm Hg and 1-minute recovery pressures of greater than 30 mm Hg or 5-minute postexercise pressure of greater than 20 mm Hg is diagnostic of chronic exertional compartment syndrome.111 For most cases, treatment should begin with relative rest and cross-training to avoid the exacerbating activity. Other treatment options include physical therapy and biomechanical correction. If unsuccessful, elective fasciotomy should be considered.109,111 Subjective improvement and normalization of compartment pressures have been reported. Fascial release adversely affects muscle strength, and these procedures should not be advocated without

Basic Science and Injury of Muscle, Tendon, and Ligament

Anterior compartment Lateral compartment Superficial peroneal nerve Fascial defect

Medial dorsal cutaneous nerve Intermediate dorsal cutaneous nerve

Figure 1A1-14   Schematic drawing of the relationship of the branches of the superficial peroneal nerve to the fascial defect.   (Adapted from Garfin SR, Mubarak SJ, Owen CA: Exertional anterolateral-compartment syndrome. Case report with fascial defect, muscle herniation, and superficial peritoneal-nerve entrapment. J Bone Joint Surg Am 59:404-405, 1977.)

accurate diagnosis and counseling.112 Vigorous exercise should be avoided for 6 to 8 weeks postoperatively.

Medial Tibial Syndrome Medial tibial syndrome, or shin splints, is a syndrome of exercise-related pain localized to the medial aspect of the distal third of the tibia, coursing across the junction of muscle and tendon to bone. This syndrome previously had been ascribed to recurrent deep posterior compartment syndrome.113,114 Objective measurement of pressure within the anterior and posterior compartments, however, has shown no pressure elevation.29 This condition most likely is a stress reaction of the bone or muscle origin in response to repetitive exercise, such as running long distances on hard surfaces. Medial tibial syndrome is more common in athletes with significant hindfoot valgus and midfoot pronation, often called flatfoot.

Muscle Strain Injuries Clinical Studies of Muscle Strain Injury Mechanism of Injury

Muscle strain injuries represent about half of all athletic injuries.115 Epidemiologic studies show that muscle strain injuries occur most often in athletes involved in sports

15

requiring bursts of speed or acceleration, such as track and field, football, basketball, rugby, and soccer.116 Strains are injuries caused by stretching or muscle activation during a lengthening (eccentric) contraction.117-119 Muscle may be more prone to injury during eccentric contraction because the passive or connective tissue element of muscle allows significantly higher active muscle force ­production when muscle is stretched than when it is at static length or allowed to shorten. Certain muscles are more prone to strains than ­others. The two-joint muscles are at highest risk for injury.120 With these muscles, physiologic joint motion can place the muscles in positions of increased passive tension. For example, the hamstrings increase passive tension as the hip flexes and the knee extends. In addition, twojoint muscles often function in an eccentric manner. With eccentric contractions, the muscle can be considered to be controlling or regulating motion as a function of energy absorption. Much of the muscle action involved with running or sprinting is eccentric.121-124 For instance, during running, the hamstrings act not so much to flex the knee as to ­ decelerate knee extension before foot strike. The muscles most likely to be injured have a relatively high percentage of type II or fast-twitch muscle fibers.125 For unknown reasons, certain muscles within a particular group are more susceptible to strain injury,126 including the adductor longus in the adductors, the biceps femoris in the hamstrings, and the rectus femoris in the quadriceps. The rate of muscle stretch may also affect the site of injury.127 Recent clinical studies have begun to elucidate the relative contribution of the muscle and the tendon to change in overall muscle length during running.128,129 Late swing and early stance have been suggested as potentially injurious phases of the gait cycle for the hamstring muscle group. Studies combining kinematic with electromyogram analysis indicate that the lengthening phase of the hamstring ­muscles actually starts at about 45% of the gait cycle, with the largest rates of lengthening occurring shortly ­thereafter.128 Lengthening terminates at about 90% of the gait cycle, that is, in swing phase before foot contact. These observations call into question the hypothesis that hamstring injuries occur after foot strike and raise the ­possibility that the greatest period of vulnerability is much earlier in the swing phase than previously thought. Structural Changes with Muscle Strain Injury

A muscle strain injury may be partial or complete depending on whether the muscle-tendon unit is grossly disrupted.117 Complete tears produce muscle asymmetry at rest compared with the contralateral contour and a bulge with ­ voluntary contraction on the side of the muscle­tendon unit that still is attached to bone. Muscle strain injuries can be distinguished clinically from exercise-induced muscle soreness. Both conditions are more prone to occur with eccentric exercise.84,130-132 In both injuries, passive stretching and active contraction of the affected region produce discomfort. A strain injury, however, is an acute and usually painful event that is recognized by the patient at the time of injury, whereas muscle soreness is characterized by focal muscle pain and swelling 24 to 48 hours after exercise.90,133

16

DeLee & Drez’s Orthopaedic Sports Medicine

Direct muscle injury or contusion causes injury to muscle at the place of contact. Imaging studies have confirmed that the locus of tissue damage in an indirect strain injury occurs at or near the musculotendinous junction.126,134 Reports of surgical exploration of muscle injuries confirm the existence of tears near the muscle-tendon junction in (1) the gastrocnemius medial head (often incorrectly called a plantaris rupture),18,135 (2) the rectus femoris muscle,136 (3) the triceps brachii muscle,137 (4) the adductor longus muscle,138 (5) the pectoralis major muscle,139 and (6) the semimembranosus muscle.3 High-resolution imaging studies have also localized acute hamstring injuries to the muscle-tendon junction (Fig. 1A1-15). Bleeding often occurs after muscle injury, but it often takes a day or more to detect subcutaneous ecchymosis as the blood escapes through the perimysium and fascia to the subcutaneous space.134 Computed tomography has shown an inflammatory or edematous response within the muscle tissue itself.140 In certain instances, a hematoma can form between the muscle tissue and the surrounding fascial compartment, as shown by ultrasonography.141 There has been some interest in using magnetic resonance imaging for prognosis following hamstring strains.142,143 Both T1-weighted and inversion ­ recovery T2-weighted, spin-echo imaging in axial and sagittal planes can be used to determine volume of muscle affected and correlate with length of rehabilitation following a first-time injury.142,143 These studies suggest that ­magnetic resonance may have some value in predicting return to sport. However, as recently pointed out, there are no consensus guidelines or agreed-on criteria for safe return that completely eliminate the high risk for recurrence.144 Prevention and Treatment

It is difficult to find scientific evidence of the best methods of preventing or treating these often debilitating problems. Most athletes routinely practice stretching largely because it is believed to decrease the risk for muscle injury.145-147 Adequate warm-up also is cited as a way of preventing muscle injury.117,147 Fatigue is thought to predispose muscle to injury,117 as is a prior incomplete injury.116 There have been few clinical or laboratory studies to support any of these hypotheses.148-150 The most commonly applied treatments for muscle injuries (strains, contusions, lacerations) have been rest, ice, compression, elevation, anti-inflammatory drugs, and mobilization. The natural regeneration process can be slow and may produce incomplete healing. The isolation and use of growth factors has opened up a new area for investigation and treatment. The number and diameter of regenerating myofibers and recovery of muscle strength following a contusion injury in mice can be accelerated with the use of basic fibroblast growth factor (b-FGF), insulin-like growth factor (IGF-I), and nerve growth factor (NGF). These growth factors are known to enhance fibroblast proliferation and differentiation into muscle cells in vitro.151 Not all growth factors are beneficial to muscle healing, however. Administration of NS-398, a cyclooxygenase2–specific inhibitor, following muscle laceration in mice resulted in higher expression of transforming growth

Authors’ Preferred Method of Treatment of Muscle Strain Injuries There is no consensus for the treatment of muscle strain ­injuries.116,117 Various treatment regimens have been adapted empirically from clinical practice. Few studies have been performed to compare the effects of these different treatment strategies. Suggested modalities at various ­phases of injury include rest, ice, compression, physical therapy to improve joint range of motion and function, bandaging, and medications (topical anesthetics, analgesics, ­ muscle relaxants, and anti-inflammatory agents). Occasionally, surgery has been advocated for persons with complete dissociation of the muscle-tendon unit.154,155 Based on available laboratory and clinical studies, we have devised the following treatment regimen for ­ muscle strain injuries. We avoid immobilization and prefer to begin active stretching and muscle activation as soon as these exercises can be performed without great discomfort. ­After the initial injury, the tensile strength of the muscle-­tendon unit is weaker than normal, and large forces should be ­avoided. Forces large enough to disrupt the muscle are unlikely to occur in a controlled rehabilitation setting. We stress full recovery of muscle length and joint range of motion. Strengthening exercises are resumed early, and progressive resistance is emphasized. A recent prospective randomized trial demonstrated that a program emphasizing progressive agility and trunk stabilization was superior to a strategy ­employing static stretching and isolated progressive resistance exercises in individuals with a hamstring strain.156 Although time to return to sport was similar, the risk for reinjury was significantly reduced in the group treated with progressive agility and trunk stabilization ­exercises. We apply ice during the acute phase of the injury and heat before performing stretching exercises after the acute phase. Therapeutic exercise should be of sufficiently high intensity to impart a strengthening effect. The accommodating resistance of isokinetic devices allows the injured athlete to work at a comfortable level through a full range of motion.

factor-β1 (TGF-β1), which causes fibroblasts to differentiate into fibrotic cells rather than muscle cells, leading to an increase in fibrosis.152 By contrast, inhibition of TGF-β1 with injection of suramin following muscle strain in mice inhibits the production of fibrous scar formation by fibroblasts.153 Further work in this direction may lead to new treatment techniques for muscle strains.

Laboratory Studies of Muscle Strain Injury McMaster157 showed in 1933 that normal tendon did not rupture when the gastrocnemius muscle-tendon unit of rabbits was pulled to failure. Failure occurred at the bonetendon junction, the myotendinous junction, or within the muscle. A series of experiments in a rabbit model showed that activation of normal muscle by nerve stimulation alone

Basic Science and Injury of Muscle, Tendon, and Ligament

17

Energy Absorbed to Failure

Energy Absorbed (Joules)

Group 1

Figure 1A1-15   Acute left biceps femoris muscle strain and chronic right hamstring injury. A prone axial computed tomographic image of the proximal thighs in this patient demonstrates an area of low density in the region of the long head of the left biceps femoris muscle (left arrow), typical of an acute muscle strain. Calcifications are noted in the comparable muscle group on the right side (right arrow), probably due to an old injury in this patient.   (From Garret WE Jr, Rich FR, Nikolaou P, Vogler JB: Computed tomography of hamstring muscle strains. Med Sci Sports Exerc 21:506-514, 1989. © The American College of Sports Medicine, 1989.)

produced no disruption of the muscle-tendon unit.158 Gross or microscopic muscle injury required stretch of the muscle. The forces produced at the time of muscle failure were several times the maximal isometric force produced by the activated muscles.158 Passive Stretch

In a rabbit model, muscles stretched from the proximal or distal tendon without preconditioning or muscle ­activation consistently showed injury near the muscletendon junction (usually distally). A small amount of muscle fiber was left attached to the tendon, usually 0.1 to 1.0 mm in length. The disruption occurred predictably near the muscle-tendon junction within the strain rates tested and for all muscles tested regardless of architectural features or direction of strain. A more recent study showed that at higher rates of stretch, certain muscles tear in the distal muscle belly.127 Injury can be predicted in an animal model to occur at the location of maximal strain.127 Active Stretch

Experiments have been performed to measure the amount of force needed to produce failure, energy absorption before failure, and muscle length before failure in passive and active muscles under conditions simulating powerful eccentric contractions.136 The total amount of strain before failure did not differ among muscles stretched to the point of failure under three conditions of motor nerve activation: (1) tetanically stimulated, (2) submaximally stimulated, and (3) unstimulated. The force generated at failure was only about 15% higher in stimulated muscles. The location of failure, near the myotendinous junction, did not change. The energy absorbed was about 100% higher, however, in muscles stretched to failure while activated (Fig. 1A1-16). These data confirm the importance of considering muscles as energy absorbers. The passive components of stretched muscle have the ­ability to absorb energy, but the potential

Group 2

Group 3

232 ±47

454 ±67

n=8

258 ±36

516 ±109

0 Hz 64 Hz P < .0003

437 ±111

0 Hz 16 Hz P < .0002

534 ±79

16 Hz 64 Hz P < .01

Figure 1A1-16   Average relative energy absorbed by the muscle-tendon unit before failure in groups 1 through 3. All values are plus or minus the standard deviation. 0 Hz, no stimulation; 16 Hz, wave-summated stimulation; 64 Hz, tetanic stimulation.   (From Garrett WE Jr, Safran MR, Seaber AV, et al: Biomechanical comparison of stimulated and nonstimulated skeletal muscle pulled to failure. Am J Sports Med 15:448-454, 1987.)

to absorb energy is increased greatly by concomitant active contraction of the muscle. This concept helps to explain the ability of muscles to prevent injury to themselves as well as the supporting joint structures. Muscles can be injured when they are incapable of withstanding a certain force or strain. The ability of a muscle to withstand force and strain is a measure of energy absorption. In engineering terms, strain energy is the area under the curve relating stress to strain. Muscle can be ­considered to have passive and active components to absorb energy. The passive component does not depend on muscle activation and is a function of the muscle’s connective tissues, including the muscle fibers themselves and the connective tissue associated with the cell surface and between fibers. The active contractile mechanism of the muscle can double the ability of muscle to absorb energy. Conditions that diminish the muscle’s contractile ability also might diminish the ability of muscle to absorb energy. For example, muscle fatigue and muscle weakness, which suggest that the tissue’s ability to absorb energy is diminished, often are considered as factors predisposing muscle to injury. At low levels of strain, most energy absorption is due to the active rather than the passive elements. Because most physiologic activity in eccentrically contracting muscle occurs at relatively low levels of muscle strain, energy absorption during eccentric contraction is due more to active than to passive force in the muscle. The ability of the muscle to absorb energy not only can protect a muscle but also can protect associated bones and joints.118 Nondisruptive Injury

In nondisruptive stretch-induced injury, muscle fibers do not tear at the junction of the muscle fiber and the tendon but instead tear within those fibers that are a short distance from the tendon.159 In the acute phase, these injuries are marked by fiber disruption and hemorrhage within the muscle (Fig. 1A1-17). By 24 to 48 hours after injury, an inflammatory reaction, including inflammatory

18

DeLee & Drez’s Orthopaedic Sports Medicine

A

B

Figure 1A1-17   A, Gross appearance of tibialis anterior muscle following controlled passive strain injury. A small hemorrhage (H) is visible at the distal tip of injured muscle at 24 hours. I, injured; C, control. B, Histologic appearance of muscle immediately after passive strain injury. Note the rupture of fibers at the distal muscle-tendon junction, along with hemorrhage. T, tendon; M, intact muscle fibers (Masson stain ×100).�   �������������������������������������������������������������������������������������������������������� (From Nikolaou PK, Macdonald BL, Glisson RR, et al: Biomechanical and histological evaluation of muscle after controlled strain injury. Am J Sports Med 15:9-14, 1987.)

Viscoelastic Behavior of Muscle Among the factors believed to be important in the prevention of injury are innate flexibility, warm-up, and stretching before exercise. Muscle response to stretching classically has been explained on a neurophysiologic basis with reference to stretch reflexes,161,162 although the muscle’s ­ viscoelastic properties probably account for the adaptation to acute stretch. When a ligament or tendon is stretched and held at a constant length, the tension at that length gradually decreases over time,2,163,164 a property known as stress relaxation (Fig. 1A1-19). Additionally, cyclic stretching of ligaments and tendons to the same length results in a decrease in tension with each stretch. Most of the studies pertaining to the viscoelastic behavior of muscle have focused on active force production. Much

less is known about the muscle’s viscoelastic behavior ­during stretching in a manner relevant to current athletic and rehabilitation regimens. Laboratory studies have confirmed and described mathematically muscle’s viscoelastic behavior127; the relevance of these studies to preventing injury is unknown.

FORCE GENERATION (% Control)

cells and edema, becomes pronounced. By the seventh day, the inflammatory reaction begins to be replaced by fibrous ­tissue near the site of injury. Although some muscle fibers regenerate, normal histology is not restored, and scar tissue persists.75 Immediately after a nondisruptive injury, muscle can produce about 70% of normal force, but force production decreases within 24 hours to only 50% of normal. Recovery of force production by 7 days is 90% complete (Fig. 1A1-18). The recovery of contractile ability is relatively rapid. The initial loss of function may be a result of the hemorrhage and edema at the site of injury. Another ­possibility is that oxygen free radicals produced after injury are responsible for the decline in muscle function observed in the first 24 hours after injury.160

100 90 80 70 60 50 40 30 20 10 0

immediate 24 hours

48 hours

7 days

TIME AFTER INJURY Figure 1A1-18   Percentage of control force generation over a range of frequencies versus time after controlled passive injury. Immediately after injury, N = 30; at 24 hours, N = 7; at   48 hours, N = 8; and after 7 days, N = 8. All values are plus or minus the standard error of the mean.   (From Nikolaou PK, Macdonald BL, Glisson RR, et al: Biomechanical and histologic evaluation of muscle after controlled strain injury. Am J Sports Med 15:9-14, 1987.)

Basic Science and Injury of Muscle, Tendon, and Ligament 80

Relaxation Sequence

N = 12

Tension (N)

75

5-10 4 3 2 1

70 65 60 0

5

10

15 20 25 30 35 Time (sec) Figure 1A1-19   Relaxation curves for extensor digitorum longus muscle-tendon units subjected to repeated stretch to the same tension. There was a statistically significant (P < .05) difference between the first relaxation curve and the subsequent nine curves. The second relaxation curve also showed a statistically significant difference from the other nine curves   (P < .05). There were no differences among curves 3 through 10.   (From Taylor DC, Dalton JD, Seaber AV, Garrett WE: Viscoelastic properties of muscle-tendon units. The biomechanical effects of stretching. Am J Sports Med 18:300-309, 1990.)

Effect of Repetitive Stretching on Failure Properties The viscoelastic behavior of muscle-tendon units is separate from reflex effects in the muscle.163 Reflex effects can have great significance in flexibility and performance in sports and exercise, but it is not clear whether these effects have any bearing on the prevention of muscle injury. Stretching often has been advocated to prevent injury. This hypothesis has been tested in a rabbit model.148 Muscles were stretched cyclically using 50% or 70% of the force needed to produce failure in the contralateral leg. Ten cycles at 50% of maximal force resulted in a significant increase in the length of the muscle at failure without affecting force at failure or energy absorption. Many of the muscles that were stretched cyclically to 70% of maximal force, however, showed macroscopic evidence of disruption before completing the 10 cycles; length and force to failure were unaffected in the specimens that showed no disruption. Cyclic stretching induces a protective effect in some cases, but muscles subjected to stretch without preconditioning are actually more subject to injury. Stretch that produces greater than 70% of maximal sustainable force in the muscle can make the muscle more likely to be injured. In summary, a cyclic stretching routine may make muscle less likely to be injured because it increases the length to which a muscle can stretch before failure occurs. Stretching produces significant effects on muscle at physiologic lengths, which produce stress relaxation through viscoelastic effects, and at highly stretched lengths, which affect the mechanical failure properties of muscle.

Effects of Warm-up Warm-up by a short period of muscle activity is emphasized in the prevention of muscle injuries. Viscoelastic materials are sensitive to temperature, which changes within the muscle with activation. Muscle held isometrically and stimulated for one single tetanic contraction lasting 10 to 15 seconds produces a temperature rise of about 1° C within the muscle.136,150 Following this single contraction, more

19

stretch can occur before failure and more force production is possible. These changes may be due in part to the small temperature change within the muscle. These changes, however, also may be due to the effects of stretch on viscoelasticity. Although the total length of the muscle-tendon unit does not change, the region most susceptible to injury probably undergoes some degree of stretch during an isometric contraction as the muscle belly and fibers shorten slightly. The resulting change in failure properties may be related to these stretch effects rather than a temperature change induced by muscle activity.

Summary of Basic Studies on Muscle Injury The cause of most muscle injuries involves powerful eccentric contractions. Disruptive and nondisruptive injuries show pathologic changes near the muscle-tendon junction. Active muscular contraction has an important role in the ability of a muscle to absorb energy. The separate effects of stretch, muscle activation, and temperature are being evaluated with respect to improvement of performance or prevention of injury. Stretch is required to injure normal muscle; strong active contractions involving shortening of the muscle do not appear to create injury. The ability of muscle to withstand stretch may be important in preventing injury. Stretching may increase tissue extensibility because of viscoelastic changes in the muscle. Several stretches that are each held for a specific period seem to be more beneficial than a single stretch. Ballistic stretching should be avoided because high velocities increase stretch forces, and quick shortening does not allow time-dependent or viscous changes to occur in the muscle. Empirically, three to four stretches held for 5 to 10 seconds or more may be optimal. Warm-up increases muscle extensibility. The usual ­athletic warm-up involves increasing muscle temperature by metabolic activity and stretching the muscles and tendons by active muscle force production. Because extensibility of connective tissue increases with temperature, a warm-up period before a stretching routine may be ­effective. Longer term adaptations in muscle are important in the prevention of acute injury. Often the physiologic requirement of a muscle in sports is the control or deceleration of a joint or limb. The muscle is required to absorb the kinetic energy of the limb. The ability of the muscle to absorb energy can protect it from injury. Strong muscles can absorb more energy than weak muscles; strong muscles undergo less deformation or stretch than weak muscles. Strengthening may thus help to prevent strain injury. Conditioning has a similar effect. Fatigue is a situation in which the ability of muscle to generate active force is declining. Fatigued muscles can absorb less energy than nonfatigued muscles. Muscle strength and conditioning appear to be valuable components of an injury prevention program, particularly in individuals in whom muscle injury is most likely.

Conclusions More clinical and basic laboratory studies of muscle injuries have become available. Clinical imaging studies ­provide information about the location and nature of the

20

DeLee & Drez’s Orthopaedic Sports Medicine

initial muscle injury and the clinical course. Few studies exist that evaluate or compare means of preventing injury. Such clinical studies would require large numbers of subjects to obtain reliable data. Basic laboratory studies can be helpful in the practical management and prevention of muscle strain injuries. These studies have shown the pathophysiology of muscle stretch injury, and the findings are consistent with clinical observations. As the events of injury and repair are�������� better understood, more emphasis is being placed on the treatment and prevention of these injuries and on methods of improving performance.

TENDON Tendons are fibrous connective tissues designed to transmit the force of muscle contraction to bone to effect limb movement. Tendons have a complex architecture: highly aligned matrix containing 90% type I collagen to provide tensile strength; elastin to provide compliance and elasticity; proteoglycans serving as pulse dampeners; and lipids, which may reduce shear stress-induced friction.165-169 On microscopic examination, tendons consist of a network of interlacing fibers with variously shaped cells and ground substance. Eighty-five percent of the dry weight of this structure is collagen, and the mechanical and physiologic behavior of collagen is the most important factor in determining tendon properties.170 Two cell populations are present in the major compartments of tendon. The surface epitenon contains large, polygonal cells (tendon surface cells) in syncytia embedded in a lipid- and proteoglycanrich matrix containing 25% collagen. The internal portion of tendon contains fibroblasts (tendon internal fibroblasts) in syncytial layers amidst linear and branching collagen fascicles and bundles.171-175 The matrix of tendon surface cells contain collagen types I and III, fibronectin, TGF-β, and positive IGF-I and negative modulators of cell division (IGF-I-binding proteins).168,171,172,176 Tendon surface cells are most active in migrating into and populating a wound bed in tendon after injury.165,177-182 Tendon internal fibroblasts migrate and divide less in response to injury.179 Tendon internal fibroblasts and tendon surface cells in mature tendon are present as syncytia situated in layered longitudinal sheets that are intimately connected to each other.183 Within a syncytium, cells are connected by connexin 43 (cx43) and cx32, but between syncytia they are connected only by cx43.183 Cells within the epitenon communicate with cells in the internal compartment by cx43 gap junctions. Tendon internal fibroblasts express IGF-I messenger RNA (mRNA).184 Tendon internal fibroblasts and tendon surface cells cache stores of IGF-I in the tendon epitenon and internal compartment that appear to be used in response to trauma. The epitenon and internal compartment of rat and avian tendon express mRNA for IGFbinding protein (IGF-BP5), particularly in the epitenon tendon surface cells. Rabbit flexor digitorum profundus tendons respond to human recombinant IGF-I by synthesizing DNA and matrix.185 Tendons must be capable of resisting large tensile stresses to perform their primary function, which is to transmit forces from muscle to bone. Tendon also

maintains the length of the moment arm during muscle contraction to optimize force production. In addition to this load-transmitting role, tendons satisfy kinematic requirements (they must be flexible enough to bend at joints) and damping requirements (they must absorb sudden shock to limit damage to muscle).

Structure Tendons and ligaments are both dense, regularly arranged connective tissues, but there are significant differences between them with respect to structure and histologic and biochemical properties.186 The collagen fibers in a tendon are more parallel to the longitudinal axis than is the case in a ligament. The collagen fibers, composed of thinner fibrils, extend the entire length of the tendon. Fibroblasts, which are few in number, are located more centrally in the tendon, between the collagen bundles or fibrils. Present knowledge of tendon morphology is outlined in Figure 1A1-20. The surface of the tendon is enveloped in a white, glistening, synovial-like membrane, called the epitenon. The epitenon is continuous on its inner surface with the endotenon, a thin layer of connective tissue that binds collagen fibers and contains lymphatics, blood vessels, and nerves.187 In some tendons, the epitenon is surrounded by a loose areolar tissue called the paratenon. Typically, the paratenon surrounds tendons that move in a straight line and are capable of great elongation owing to the presence of elastic fibers. This paratenon functions as an elastic sheath permitting free movement of the tendon against the surrounding tissue. Together the epitenon and the paratenon compose the peritendon (Fig. 1A1-21). In some tendons, the paratenon is replaced by a true synovial sheath or bursa consisting of two layers lined by synovial cells. This double-layered sheath, which is lined by synovium, is referred to as a tenosynovium. Within this synovial sheath, the mesotendon carries important blood vessels to the tendon.188 The flexor tendons of the forearm and hand and the Achilles tendon are surrounded by this well-defined sheath lined with synovial cells. In the absence of a synovial lining, the paratenon often is called a tenovagina. The perimysium becomes continuous with the endotenon at the musculotendinous junction. The tendon-bone interface marks the site where collagen fibers enter bone as Sharpey’s fibers, and the endotenon becomes continuous with the periosteum. The insertion of tendon into bone generally is classified into two types. The simpler type, termed direct insertion, occurs when the tendon fibrils pass directly into bone through zones of fibrocartilage with little interdigitation into the surrounding periosteum. As described by Cooper and Misol,189 the tendon inserts into a zone of fibrocartilage, then into a layer of mineralized fibrocartilage, and finally into bone. The periosteum is continuous with the endotendon. Dissipation of force is achieved effectively through this gradual transition from tendon to fibrocartilage to bone. The second type of insertion is more complex and involves the periosteum as well as the underlying bone; the superficial fibrils insert into the periosteum, whereas the deeper fibrils fan out into bone directly. When tendons insert at an angle into the bone, a larger area of fibrocartilage can be found on one

Basic Science and Injury of Muscle, Tendon, and Ligament

21

Figure 1A1-20   Basic tendon morphology.   (Adapted from Kastelic J, Galeski A, Baer E: The multicomposite structure of tendon. Connect Tissue Res 6:11-23, 1978.)

MICROFIBRIL

TENDON

FIBRIL FASCICLE

TROPOCOLLAGEN

Fibroblasts

Waveform or crimp structure

side of the insertion.190 This is thought to be an adaptation to the compressive forces experienced by the tendon on that side. Tendon is well-vascularized tissue, although less so than muscle. The blood supply to tendon has several sources, including the perimysium, periosteal attachments, and surrounding tissues. Blood supplied through the surrounding tissues reaches the tendon through the paratenon, mesotenon, or vincula. A distinction between vascular and the so-called avascular tendons has been made to denote ­ differences in blood supply. Vascular ­ tendons

Fascicular membrane

are surrounded by a paratenon and receive vessels along their borders; these vessels then coalesce within the tendon. The relatively avascular tendons are contained within tendinous sheaths, and the mesotenons within these sheaths function as vascularized conduits called vincula. The muscle-tendon and tendon-bone junctions along with the mesotenon are the three types of vascular supply to the tendon inside the sheath. Other sources of nutrition191,192 include diffusional pathways from the synovial fluid, which provide an important supply of nutrients for the flexor tendons of the hand.

Tendon

Paratenon Peritendon Epitenon Endotenon Fibroblast Primary bundle Fibril Microfibril Collagen fibril Tropocollagen Figure 1A1-21   Structural organization of tendon.

22

DeLee & Drez’s Orthopaedic Sports Medicine

The nervous supply to a tendon is sensory in nature. The proprioceptive information supplied to the central nervous system by these nerves usually is picked up through mechanoreceptors located near the musculotendinous junction.

Biochemistry The cellular component of tendon is the tenocyte, which is responsible for the production of collagen and the matrix proteoglycans. Similar to all types of connective tissue, tendons consist of relatively few cells (fibroblasts) and an abundant extracellular matrix. In tendons and ligaments, the main constituent is collagen, along with small amounts of elastin, ground substance, and water. Collagen constitutes about one third of the total protein in the body and is present in large amounts in specialized connective tissues, such as tendon, ligament, skin, joint capsule, and cartilage.

Collagen Twelve different but homologous collagen types are recognized.193 It is convenient to think of two major classes of collagen—those that are fiber forming and those that are not. Collagen types I, II, and III are known as fibril-forming collagens. After being secreted into the extracellular space, these collagens assemble into collagen fibrils. Collagen types I and III are the main forms comprising normal connective tissue. Type I is more common, constituting 90% of the collagen in the body. The remaining collagen types constitute the second major group—the non–fiber-forming group; types IV and V are the basement membrane collagens. The fibroblast is a spindle-shaped, contractile cell that synthesizes connective tissue matrix precursors, including collagen, elastin, and proteoglycans.194,195 Collagen is produced within the fibroblast as a large precursor molecule (procollagen), which is secreted and cleaved extracellularly to form tropocollagen. Soluble tropocollagen molecules form noncovalent cross-links, resulting in insoluble collagen molecules that aggregate to form collagen fibrils. After collagen fibrils have been synthesized in the extracellular space, they are strengthened greatly by the formation of covalent cross-links within and between the constituent collagen molecules. In its normal state, mature collagen can be degraded only by collagenase, whereas ruptured collagen fibrils are susceptible to digestion by trypsin. When isolated collagen fibrils are viewed in an electron microscope (Fig. 1A1-22), they exhibit cross-­striations every 64 to 68 nm. This pattern reflects the packing arrangement of the individual collagen ­molecules in the fibril. Collagen is the strongest fibrous protein in the body. The arrangement of fibers in parallel to their longitudinal axis results in tendon having one of the highest tensile strengths of all soft tissues. All types of collagen have in common a triple helical domain, which is combined ­differently with globular and nonhelical structural ­elements (Fig. 1A1-23). The triple helical collagen molecule is stiff compared with a single polypeptide chain. The most common collagen molecule, type I collagen (also found in skin and bone), is composed of three α-peptide chains, each with about 1000 amino acids, resulting in a total

Figure 1A1-22   Electron micrographs of collagen demonstrating the periodicity and the regularity of the molecule. Precipitated from collagen solution by dialysis against 1% sodium chloride.   (From Bloom W, Fawcett DW: A Textbook of Histology, 8th ed. Philadelphia, WB Saunders, 1962, p 105. Original investigators: J. Gross, F. O. Schmitt, and J. H. Highberger.)

molecular weight of about 340,000 Da.196,197 The α-chains exist in several different isomeric forms (Table 1A1-3). Type I collagen contains two α1-chains and one α2-chain.198 These three α-chains are wound around each other in a regular helix to generate a rod-like collagen molecule about 300 nm long and 1.5 nm in diameter. Normal human adult flexor tendons are composed largely of type I collagen (>95%); the remaining 5% consists of type III and type IV collagen.199,200 The amino acid sequence of the collagen molecule has been studied extensively to understand the cross-linking mechanism of these structures. They are arranged in a characteristic triple helical pattern that gives the molecule its rod-like form and its rigid properties. Every third amino acid in the α-chain is glycine; other amino acids commonly present are proline (15%) and hydroxyproline (15%).196 Consequently, nearly two thirds of the collagen molecule consists of these three amino acids. Hydroxyproline is derived from proline and is almost unique to collagen, and another amino acid, hydroxylysine, is unique to collagen.

Physical Properties of Collagen The mechanical properties of soft collagenous tissues are highly dependent on their structural integrity, which is determined primarily by the architecture and properties

Basic Science and Injury of Muscle, Tendon, and Ligament

rupture, abnormal curvature of the spine, and problems with skin breakdown and wound dehiscence.

64 nm Fibril

Elastin

Overlap Zone Microfibrils Hole Zone Packing of molecules

Collagen molecule

23

280 nm

α2 Triple α 1 helix α1

1.5 nm

Elastin is a protein found in connective tissues that permits these structures to undergo changes in length without incurring any permanent change in structure, while expending little energy in the process. Elastin is responsible for the wavy pattern of the tendon when viewed by a light microscope. Tendons of the extremities possess small amounts of this structural protein, whereas elastic ligaments, such as the ligamentum flavum and ligamentum nuchae, have greater proportions of elastin. Elastin in most tendons is found primarily at the fascicle surface201; it usually comprises less than 1% by dry weight. The elastin content of the aorta can be 30% to 60% of dry weight. Elastin, similar to collagen, has lysine-derived cross-links. The amino acids desmosine and isodesmosine are unique to elastin. Their formation depends on the presence of copper. The elastic potential of elastin is due primarily to the cross-linking of lysine residues through desmosine, isodesmosine, and lysinonorleucine.

Ground Substance Collagen α-chain (purity to molecules) Figure 1A1-23   Microstructure of collagen showing the three α-chains of the triple helix. Three separate α-chains are wrapped around each other to form a rope-like, triple-stranded, helical rod. Every third amino acid of the native molecule is glycine.

of the collagen fibers as well as by the amount of elastin present in the tissue. The physical properties of collagen rely on covalent cross-links within and between the molecules. The triple helix conformation of collagen is stabilized mainly by hydrogen bonds. The three chains are held together strongly by hydrogen bonds between glycine residues and between hydroxyl groups of hydroxyproline. This helical conformation is reinforced by hydroxyproline-forming and proline-forming hydrogen bonds to the other two chains. The degree of cross-linking is a key to the tensile strength of collagen and its resistance to enzymatic and chemical breakdown. This conclusion is evident clinically in the condition of lathyrism, in which an absence of cross-linking produces a collagen that is significantly weakened, resulting in an increased incidence of aortic

About 1% of the total dry weight of tendon is composed of ground substance, which consists of proteoglycans, ­glycosaminoglycans, structural glycoproteins, plasma proteins, and a variety of small molecules. The waterbinding capacity of these structures helps to account for the viscoelastic properties of tendinous materials. Proteoglycans and glycosaminoglycans are thought to be important for stabilizing the collagenous skeleton of connective tissue. Proteoglycans are high-molecularweight macromolecules consisting of a protein core to which glycosaminoglycan side chains are attached. Glycosaminoglycans are macromolecules containing repeated disaccharides composed of a hexosamine residue and a uronic acid residue. Glycosaminoglycans that are abundant in mammalian tissues include hyaluronic acid, chondroitin sulfate, dermatan sulfate, keratan sulfate, and heparin-heparan sulfate. Except for hyaluronic acid, the glycosaminoglycans are negatively charged owing to the presence of sulfate or carboxyl groups, and this confers predictable mechanical and chemical properties on the connective tissue. Regions of tendon that ­experience primarily tensile forces have a lower proteoglycan content

  Table 1A1-3  Principal Collagen Types and Their Properties Type

Molecular Formula

Polymerized Form

Distinctive Features

Tissue Distribution

I

[α-1 (I)]2 α-2(I)

Fibril

Skin, tendon, bone, ligaments, cornea

II

[α-1 (II)]3

Fibril

III

[α-1 (III)]3

Fibril

IV

[α-1 (IV)2 α-2 (IV)]

Basal lamina

Low hydroxylysine Low carbohydrate High hydroxylysine High carbohydrate High hydroxyproline Low hydroxylysine Low carbohydrate Very high hydroxylysine High carbohydrate

Articular cartilage, intervertebral disk, notochord, vitreous body of eye, fetal collagen Skin, blood vessels, internal organs Basal laminae

24

DeLee & Drez’s Orthopaedic Sports Medicine

and higher rates of collagen synthesis than areas that ­experience frictional and compressive forces in addition to tensile forces.202

and development. Gap junction regeneration and function are most likely essential for an organized wound healing response in tendon.

Cellular Interaction

Mechanical Properties of Tendon

Results of light and electron microscopy studies have shown that cells in the epitenon and internal compartment of whole tendon are connected physically to each other.172,183 Epitenon cells and internal fibroblasts in vivo are layered in longitudinal syncytia that appear optimal for rapid, repeated chemical and electrical coupling, similar to osteocytes in bone.203 Cells within tendon and in vitro are coupled and respond to a mechanical perturbation (e.g., a micropipet mechanical stimulation of a target cell plasma membrane) by releasing intracellular calcium stores and propagating a calcium wave to adjacent cells for four to seven cell diameters.204-206 In vivo, tendon fixed with glutaraldehyde under tension contained cells that were indented dramatically, similar to marshmallows squeezed between rods.166 Avian tendon cells express gap junctions and have at least three forms of cx43: a 42-kDa nonphosphorylated form and two intermediate forms of 44 to 47 kDa that are phosphorylated at serine.207-209 Preliminary data indicate that quiescent tendon surface cells have predominantly the nonphosphorylated form of cx43 but have phosphorylated forms during log phase.207 Tendon surface cells and tendon internal fibroblasts express mRNA for cx42, cx43, cx45, and cx45.5 by polymerase chain reaction detection and cloning, but cx43 is the only form detected by Northern analysis.184 The Western blot technique detects cx26, cx32, and cx43, whereas cx32 and cx43 have been visualized by scanning confocal microscopy183 with cx43 appearing to predominate. Quiescent tendon internal fibroblasts have about half each of the 43-kDa nonphosphorylated form and 45-kDa phosphorylated form of cx43 and have all forms during log phase growth. Ingber210 postulated in his tensegrity model that cells are connected and signal through direct mechanical linkage from the matrix through integrins and through the cytoskeletal system to the nucleus.210 Gap junction proteins may not have cytoskeletal ­ connections but are known to pass signaling molecules intercellularly.211-213 A direct response in a cell by perturbing matrix and signaling through an integrin has not been demonstrated, although endothelial cells subjected to shear stress release [Ca2+]ic and cluster integrins at the cell front.214,215 Direct evidence that connexins are involved in signaling and growth control derives from experiments with the c6 glioma cell line, which is poorly electrically coupled.216 If the cells are transfected with cx43 complementary DNA and express the protein, they gain the ability to communicate with neighbors electrically and with a propagated calcium wave in response to a mechanical stimulation.216,217 Loss of cx43 and gap junctions leads to loss of regulation of DNA synthesis and cell division.217 Although cx43 gap junction expression can be up-regulated by mechanical load in tendon cells,165 cx43 is up-regulated by load in osteoblasts and vascular smooth muscle cells as well.218 The connexins may be involved intimately in regulating embryogenesis

The primary function of tendon is to transmit muscle forces to the skeletal system to provide joint and limb locomotion and movement. To do this effectively, tendons must be capable of resisting high tensile forces with limited elongation. Tensile strength of 98 N/mm2 has been reported.219 The densely packed collagen fiber bundles arranged in parallel along the length of the tissue provide efficient resistance to tensile loading; however, tendons have weak resistance to shear and compression forces. Thus, from a functional point of view, tendons are designed to transmit tensile loads with minimal energy loss and deformation. Biomechanical studies of tendon have revealed that the stress-strain relationship is similar to other parallelfibered collagenous tissues, such as ligaments, but not the same as skin, where the collagen fibers are more randomly organized. As with most biologic tissues, tendons show complex time-dependent and history-dependent nonlinear viscoelastic properties.220-223 These properties include stress relaxation (decreased stress with time under constant ­elongation) and creep (increased elongation with time under constant load) (Fig. 1A1-24). In addition, the shape of the load-elongation curve depends on the previous loading history. Clinically, we recognize these time-­dependent and history-dependent characteristics; for example, increased tendinous and ligamentous laxity occurs after exercise. Several models have been developed to describe and predict the mechanical and viscoelastic behavior of tendons and other biologic tissues.222,224,225 Figure 1A1-25 represents a typical load-elongation curve for tendon. Abrahams220 described three distinct regions of such a curve before rupture of the structure. Under tension, the fibers straighten, and the system becomes stiffer. Different components of the structure take up loads at different levels, resulting in a nonlinear, concave, upward load-elongation curve. The initial toe region represents the alignment of fibers in the direction of stress as well as fiber recruitment. In this region, little force is required initially to elongate the tissue, which may protect the tendon from large loads during normal joint motion. The toe region is followed by a steep linear portion of the loadelongation curve, in which most fibers are aligned in parallel to the longitudinal axis of tension and elongation of the helical structure on the fibrillar and macromolecular levels occurs. The slope of the curve in this linear region often is referred to as the elastic stiffness of the tendon. With further loading, small reductions in stiffness sometimes can be observed, which can be attributed to failure of a few fiber bundles. Eventually, major failure of the tendon will occur as the fibers recoil and blossom into a tangled bud at the ruptured end. At strains of 4% to 8%, collagen fibers begin to slide past one another, resulting in disruption of their cross-linked structure. Normal physiologic forces or loads are reported to cause strain of less than 4%,226,227 but ­certain activities, such as sports, occasionally induce stronger loads. There is a large safety factor because the maximal isometric contractile force of a muscle is ­usually about

Basic Science and Injury of Muscle, Tendon, and Ligament

25

X Ultimate failure Load

creep 3

Deformation time

LOAD 2

Deformation 1 Stress

stress relaxation time

Figure 1A1-24   Biomechanical properties of collagen. Under a constant load, the tendon will undergo time-dependent relaxation (creep), whereas under a constant deformation, the structure will undergo stress relaxation (i.e., reduction in load over time).

one third of the maximal load of the tendon, although repetitive loading at submaximal failure loads can result in fatigue and eventual failure of the tendon. The tendon fibroblasts themselves, however, appear to tolerate repetitive tensile loads well. In vitro studies have shown no negative effects on tendon fibroblasts with repetitive stretching up to 25% strain.228 Large variations in mechanical properties of tendons usually are attributable to differences in species, type, and age, as well as testing conditions, such as temperature and humidity. These variables are important to note when comparing the results of different studies. Preconditioning of the specimen through cyclic stretching at low levels of elongation before testing helps to eliminate some of this variation. In addition, one must distinguish between the structural properties of the tendon-bone complex and the mechanical properties of the tendon itself. Structural properties (i.e., linear stiffness, ultimate load, ultimate elongation, and energy absorbed at failure) describe the tensile properties of the tendon-bone complex and are obtained directly from the load-elongation curve. Mechanical properties (i.e., elastic modulus, ultimate tensile strength, ­ultimate strain, and strain energy density) are represented by the stress-strain relationship and are properties of the tendon itself (Fig. 1A1-26).

Adaptability of Collagen Aging  After collagen maturation, the mechanical properties of tendon reach a plateau, followed shortly thereafter by a decrease in tensile strength. This decrease in tensile strength correlates with decreases in both the amount of insoluble collagen and the total collagen present.229 There is a concomitant increase in stiffness,230 which likely is due to a marked increase in collagen cross-linking.231,232 Other

ELONGATION Figure 1A1-25   Typical load-elongation curve to failure showing primary or “toe” region (1), secondary or “linear” region (2), and end of secondary region (3).

extracellular changes include a decrease in the content of mucopolysaccharides and water.

Training Studies of exercise-related changes in tendon properties are inconclusive. Most studies have shown that training results in increased maximal tensile failure load,233-235 which is a structural property of the tendon-bone interface.233,235 The effect of exercise on mechanical properties, cross-sectional area, and collagen content are less clear. These properties in flexor tendons of swine after 1 year of moderate exercise showed no difference from control animals in one study,236 although a similar study observed increases in strength, size, and collagen content in extensor tendons.235 A third study in rabbits234 found that the ultimate load was higher for trained than for nontrained animals, but the weight, water, and collagen contents of the tendons were no different. Flexor and extensor tendons may respond differently to exercise, with flexor tendons having a limited capacity for adaptability and extensor tendons having a greater training potential, although these differences need to be investigated more thoroughly. Ultrastructural investigations have shown that exercise leads to an increased number of collagen fibrils that are thinner in diameter compared with controls.237,238 Studies of exercised rabbits revealed that the collagen fiber crimp angle is increased, whereas crimp length and elastic modulus are lowered.239 Anabolic steroids accentuate these changes and can lead to increased collagen dysplasia.239,240 Further research is needed in this area.

Immobilization Several studies have shown decreased tensile strength241,242 and increased collagen turnover241 in tendon after immobilization. Similar to ligament, tendon shows a decrease in stiffness with immobilization. Many of the differences in these studies may be attributable to differences in age of the animals studied as well as to duration of ­immobilization.

DeLee & Drez’s Orthopaedic Sports Medicine

26

Decrease stress

STRUCTURAL PROPERTIES OF BONE-TENDON COMPLEX (LOAD-DEFORMATION CURVE)

A

Linear slope

Immobilization Energy absorbed Ultimate deformation

Deformation (mm)

Physiologic activities

Exercise

MECHANICAL PROPERTIES MASS

LOAD (N)

Failure load

Increase stress

MECHANICAL PROPERTIES OF TENDON SUBSTANCE (STRESS-STRAIN CURVE)

STRESS (N/m2)

Ultimate stress STRESS AND STRAIN DURATION

Elastic modulus

B

Ultimate strain

Strain (%)

Figure 1A1-26   Representative plots of tensile testing to failure. The structural properties of the tendon-bone complex are obtained from the load-elongation curve (A), and the mechanical properties of the tendon substance are obtained from the stressstrain curve (B).

Based on the available information, Woo and associates243 developed a hypothetical curve that predicts the mechanical response of tendons and ligaments to various periods of exercise and immobilization (Fig. 1A1-27). This diagram suggests that for tendons and ligaments within the normal range of physiologic activity, immobilization results in profound shifts in deformation properties when subjected to increasing forces. Short-term training has little or no observable effect on these properties, and long-term training has a minimal effect. The clinical significance of these animal results suggests that connective tissue is more responsive to a decrease in mechanical stimuli than to progressively increasing loads.

Medication and Tendons Corticosteroid Treatment

Figure 1A1-27   A hypothetical curve showing the nonlinear properties of collagenous tissues and the effects of stress and motion on the equilibrium responses of soft connective tissues.    (From Woo SL: Mechanical properties of tendons and ligaments. I. Quasi-static and nonlinear viscoelastic properties. Biorheology 19:385-396, 1982. With permission of Pergamon Press Ltd.)

No area of tendon research has received as much attention in the orthopaedic and sports medicine literature as that of corticosteroid injections into or around tendinous tissue. Glucocorticoids are often used in the treatment of athletic injuries for their marked anti-inflammatory effect. The biosynthesis of collagen is inhibited by glucocorticoids.244 There are case reports of local injections around the Achilles tendon245 and patellar tendon246 resulting in

rupture of the tendinous tissue. The effects of corticosteroids can be systemic as well as local, as shown by reports of bilateral rupture of the Achilles tendon in patients receiving oral glucocorticoid therapy.247,248 Laboratory studies of the effects of corticosteroid injections have produced confusing results. Important variables include the duration of the study, amount of steroid injected, and site of injection. Oxlund249 showed that local ­administration of hydrocortisone acetate, 20 mg/kg every third day for 24 days, around the peroneal tendons of rats increased the tensile strength and stiffness of muscle ­tendons with no change in collagen content. In the same study, another group of animals received injections into both knee joint cavities that decreased tensile strength of the ­ posterior cruciate ligament-bone interface. Systemic effects of this local cortisol treatment included decreased thickness and fat content of the skin. A similar study showed that daily intramuscular injections of prednisolone (2 mg/kg/body weight for 14 days) increased the maximal load, maximal stress, and energy absorption for the muscle tendons in rabbits.250 Elastic stiffness, measured after exhaustion of the viscous properties of the tendons, also increased, which was thought to be due to an increased stabilization of the collagen cross-linking pattern. To evaluate the long-term effects of corticosteroid injections, Oxlund251 injected 10 mg/kg of cortisone around the peroneal tendons every third day for 55 days. He found that although the mechanical properties of the tendons were not altered, their dry weight and hydroxyproline content were reduced. The thickness and collagen content of skin remote from the injection site were reduced, although the strength of skin specimens was increased.

Basic Science and Injury of Muscle, Tendon, and Ligament

As a result of these studies and others, it is currently believed that corticosteroids act on collagenous tissues in two ways.251 Initially, during the first 1 to 2 weeks, corticosteroids induce a relatively fast increase in the mechanical and structural stability of the injected tissues. This increase is believed to be due to a change in the cross-linking pattern of the collagen. With continued treatment, progressive thinning and a reduction in collagen occur as a result of inhibited collagen synthesis, ultimately leading to reduced collagen content. Nonsteroidal Anti-Inflammatory Drugs

Vogel252 was the first investigator to show that indomethacin treatment resulted in increased tensile strength, proportion of insoluble collagen, and total collagen content in rat tail tendons. Carlstedt and coworkers253 examined the influence of indomethacin on the biomechanical and biochemical properties of rabbit tendons and found increased strength in tendon repair after indomethacin treatment. They noted a slight decrease in the amount of soluble collagen, which may have been due to increased cross-linking after indomethacin treatment, and concluded that the increased tendon strength resulted from the increased cross-linkage. The effects of nonsteroidal anti-inflammatory drugs on tendon fibroblasts have been studied in in vitro models. The results suggest that nonsteroidal anti-inflammatory drugs can depress DNA synthesis and stimulate protein synthesis through direct effects on the fibroblasts.254

Tendon Healing Mechanisms of Tendon Injury Injury to tendons can result from acute trauma (e.g., laceration) or repetitive loading (e.g., overuse injury). The former is discussed first with respect to flexor tendon injuries of the hand, and overuse tendinous injuries incurred in sports are covered last. Considerable scientific data are available regarding the acutely traumatized tendon. Much of this work has been done on tendons of the hand owing to their propensity for injury. Injury to tendon can occur in numerous ways; in the hand, avulsion directly from bone and mid-substance transection of the tendon itself are the two major mechanisms. These injuries often occur after crush injuries to the hand, of which 75% involve ­associated injury to the surrounding soft and hard tissues. The tendons of the hand are also subject to compressive loads when these tendons wrap around articular surfaces and in oblique tendon insertions into bone. During active flexion, the pressure between the pulley and the flexor tendon may be 700 mm Hg.255 These compressive loads are capable of altering the histologic structure of the ­tendon.256

Primary Tendon Healing As in other areas in the body, tendon healing proceeds in three phases: (1) an inflammatory stage, (2) a reparative or collagen-producing stage, and (3) a remodeling phase. There are two different theories of primary tendon healing, or the healing of two divided tendon ends brought into apposition by sutures.

27

One theory suggests that healing depends on the surrounding tissues and that the tendon itself plays no significant role.198,257-262 This theory holds that the tendon is an inert, almost avascular structure whose cells are incapable of contributing to the healing process. Using a canine model, Potenza259,261 showed that the tendon is invaded by fibrovascular tissue at the location of suture placement. At 28 days, the collagen produced by these fibroblasts is immature, but by 128 days, it is indistinguishable from that of normal tendon. By contrast, several studies191,192,263-268 have suggested that the inflammatory response is not essential to the healing process and that tendons possess an intrinsic capacity for repair. A study of rabbit flexor tendons showed an intrinsic tendon repair response consisting of proliferation of tendon cells and production of mature collagen. Efforts to show the intrinsic capacity of tendon healing previously failed owing to an inability experimentally to isolate the tendon from the inflammatory response. Lindsay and Thomson265 were the first to show (in chickens) that an experimental tendon suture zone can be isolated from the perisheath tissues, and that healing progressed at the same rate as when the perisheath tissues were intact. Later, in isolated segments of profundus tendon in rabbits, these researchers showed that an active metabolic process existed in the experimentally free tendons by the presence of anabolic and catabolic enzymes.266 In addition, sutured free tendon grafts of rabbit flexor tendons healed without adhesions within a vascular synovial environment of the suprapatellar bursa. Consequently, it is now accepted that tendons may possess intrinsic and extrinsic capabilities for healing, and the contribution of each of these two mechanisms probably depends on the location, extent, and mechanism of injury and rehabilitation program used after the injury. Tendon healing begins with the formation of a blood clot and an inflammatory reaction that includes an outpouring of fibrin and inflammatory cells. The degree of inflammation is related to the size of the wound and the amount and type of trauma that has occurred. The presence of nitric oxide also appears to limit the duration and intensity of the inflammatory response following tendon injury.269,270 A clot forms between the two tendon ends and is invaded by cells resembling fibroblasts and migratory capillary buds. This process occurs during the first 3 days after injury. The fibroblasts are believed to arise from the endotenon and epitenon.271 Fibroblasts residing in the endotenon differ from those in the epitendinous tissues.202 The epitendinous fibroblasts resemble synovial cells and are particularly active in response to injury. Mesenchymal cells, which are capable of differentiating into fibroblasts, also appear in the area. A study of patellar tendon injury in a rat model showed that in the first few days of acute injury and inflammation, the mesenchymal cells are primarily from the circulation.272 In the subsequent days and weeks, the number of circulation-derived mesenchymal cells decreases and that of locally derived mesenchymal cells from the tendon tissue itself increases.272 This inflammatory phase is evident until the 8th to 10th day after injury. Collagen synthesis begins within the first week and reaches its maximal level after about 4 weeks. At 3 months, collagen synthesis still proceeds at a rate

28

DeLee & Drez’s Orthopaedic Sports Medicine

3 to 4 times normal. Type I collagen is synthesized and extruded into the extracellular space as procollagen, which is converted to type I collagen by the enzyme procollagenase. The expression of procollagen mRNA, and thus the production of procollagen, depends on the presence of TGF-β1273; other growth factors are also expressed throughout the acute inflammatory phase.274 Initially, the collagen fibrils are oriented perpendicular to the long axis of the tendon, but by 2 months these fibrils usually are oriented parallel to the axis of tensile loading. Restoration of the gliding function of the tendon depends on the dissolution and reformation of the collagen fibers during the scar remodeling phase. This phase starts at about the 15th day, and by 28 days, most of the fibroblasts and collagen between the tendon stumps are oriented longitudinally. Collagenase is present in the wound on the second day after injury. Between 4 and 6 weeks, collagen synthesis and collagen degradation reach equilibrium. Collagen maturation and remodeling begin in the third week and can continue for 1 year after injury.275,276 The strength of the tendon repair results from the organization of collagen fibrils at the bone site; these fibrils cross-link with each other and with those of the tendon on each side of the wound.

Biomechanics of Tendon Healing A classic experimental study using the extensor carpi radialis and flexor carpi ulnaris tendons in dogs showed that tensile strength progressed through three phases parallel with the phases of healing: (1) rapid decrease in tensile strength as a result of wound edema, which lasts about 5 days, ­during which tensile strength depends primarily on the suture; (2) increase in tensile strength, reaching a plateau on about the 16th day; and (3) second increase in tensile strength, beginning between 19 and 21 days and continuing for an undetermined period (length of study was 72 days). The mechanical strength of the healing ­ tendon was related closely to the three histologic phases of the ­healing process: (1) exudation and fibrous union; (2) fibroplasia; and (3) maturation, organization, and differentiation. Function and motion during the first two phases of healing appeared to increase cellular reaction and separation at the suture lines. Active, unprotected use even after 3 weeks of ­immobilization may be associated with stretching of the suture line and always leads to an increased ­cellular response. On the basis of this work, clinicians for the next 40 years immobilized patients with injured tendons until the third phase of the healing process, when range of motion was encouraged to stimulate increased tendon strength and gliding. The strength of an injured tendon that has been sutured properly increases rapidly during the fibroplastic phase, when granulation tissue is produced to repair the defect. Quantitative changes in acid mucopolysaccharides (hydroxyproline and hexosamine) accompany collagen production, and the ratio of wound collagen to mucopolysaccharide content is a direct measure of increasing tensile strength. The strength of the healing tendon increases as the collagen becomes stabilized by cross-links and the fibrils assemble into fibers. A study of rotator cuff repair in sheep showed that the stiffness of the repair construct

­ uring this stage can be improved by augmenting the d repair with a collagen patch (swine small intestine submucosa), although the patch had no effect on load to failure277 (the long-term effect of xenograft collagen patches on the rate and quality of healing in rotator cuff repair remains uncertain278). During the maturation phase, the mechanical strength of healing tendon increases owing to remodeling and reorganization of the fiber architecture. A gradual shift of collagen production from type III to type I may contribute to increased mechanical strength. Bioscaffolds, in particular porcine small intestinal submucosa (SIS), have also been applied to enhance patellar tendon (PT) healing in rabbits following the harvest of the central third of the PT for anterior cruciate ligament reconstruction.279 A layer of SIS was applied both anterior and posterior to the defect to act as a biologic agent to enhance healing, not as a structural replacement. After 12 weeks of healing, the PT defect filled with more healing neo-PT tissue following SIS treatment than those without treatment, as the cross-sectional area was 68% greater. SIS treatment also resulted in a 57% higher stiffness and a 70% greater ultimate load of the healing central bonepatella tendon-bone (������������������������������������ BPTB) ������������������������������ complex compared to without treatment. These results demonstrated the potential of SIS treatment to increase the quantity of healing PT tissue and structural properties of the healing central BPTB complex.

Factors Affecting Healing Active Mobilization Active mobilization in the immediate postoperative period may have a deleterious effect on tendon healing. Early active mobilization (<3 weeks after surgery) increases tension across the suture line and can lead to gap formation and tendon ischemia. Every study on primary tendon healing has shown that early active mobilization by contraction of the attached muscle is contraindicated to prevent suture problems and wound insufficiency. Poor results are probably due to increased tension on the suture line with resultant ischemia, tenomalacia, and ­possible tendon rupture or gap formation between the tendon ends. After 3 weeks of immobilization, however, Mason and Allen276 showed that a tendon’s tensile strength triples following 2 weeks of active mobilization, suggesting that mobilization may stimulate the healing rate and that the repaired flexor tendons’ once fibrous union has strengthened. Tendons undergoing minimal tension at the site of repair have also been shown to be weaker at 3 weeks than those with significant tension, but this strength increases rapidly during the following 3 weeks.280

Stress Mechanical stress promotes orientation of the collagen fibrils.281 Remodeling of collagen scar tissue into mature tendon tissue depends on the presence of tensile forces.259,282 Tendons under minimal tension at the site of repair are weaker at 3 weeks than those under significant tension. A study of rats with collagenase-induced Achilles tendon injuries found that exercising immediately after injury increased and prolonged the presence

Basic Science and Injury of Muscle, Tendon, and Ligament

of ­inflammatory cells (neutrophils and macrophages) and decreased the stiffness and ultimate force.283 The optimal amount of tension necessary to promote an acceptable clinical response is currently unknown.

Controlled Passive Mobilization The concept of immediate passive mobilization was introduced by Kleinert and coworkers,284 who showed that during limited active extension there is reciprocal relaxation of the flexor tendons, allowing passive extension of the repaired tendon. This controlled passive motion was found to be effective experimentally and clinically in decreasing the tethering effect of adhesions. Rates of tendon repair and gliding function improve significantly with early passive mobilization. Tendons at 12 weeks following repair regained more than one third the strength of intact control tendons and maintained good gliding function within the sheath during the repair process. Tensile strength during tendon healing increases faster with controlled passive mobilization than with immobilization. From a biomechanical point of view, the optimal procedure appears to be to use a strong suture repair to reduce gap formation and scar tissue in the early phase of healing. After the initial healing phase, tensile stress placed on the tendon through controlled passive mobilization appears to promote earlier reorganization and remodeling of the collagen, leading to achievement of higher ­tensile strength.

Corticosteroids and Nonsteroidal Anti-Inflammatory Drugs Large doses of corticosteroids inhibit wound healing.285,286 Studies on the effects of small and moderate doses are conflicting. Experimental data on the healing of injured tendons are lacking. Studies done on other fibroblastic structures have shown that corticosteroids suppress the formation of adhesions287,288 but may lower the tensile strength of sutured tendons, leading to subsequent spontaneous rupture.287,289 For example, rats injected with cortisone have increased stiffness in 10-day wounds but decreased energy absorption before tendon failure. A study using a rat model found that the administration of either indomethacin, a nonselective anti-inflammatory drug, or celecoxib, a cyclooxygenase-2–specific anti-­inflammatory drug, inhibited tendon-to-bone healing.290 The authors concluded that interfering with cyclooxygenase-2 activity in the early inflammatory cascade has an adverse effect on tendon-to-bone healing.

Overuse Tendon Injuries In contrast to traumatic tendinous injury, sport-related injuries more often involve repetitive submaximal loading of the tissues, resulting in what commonly is referred to as overuse injury. An estimated 50% of all sports injuries are due to overuse, and the tissue most often affected is the musculotendinous unit.291 In response to cyclic loading, dysfunction may result from changes caused by fatigue within tendons or inflammatory changes in the surrounding tissue.

29

The most frequent problem with overuse in the upper extremity involves the supraspinatus tendon with or without associated biceps tendinitis in swimmers, throwers, and weightlifters. Tennis players frequently have problems in the proximal wrist extensor tendons (i.e., tennis elbow). Other upper extremity overuse syndromes include golfer’s elbow, or tendinitis of the proximal flexor or pronator tendons, and crossover tendinitis of the adductor pollicis longus and extensor pollicis brevis tendons in rowers. Overuse problems in lower extremity tendons frequently are seen in runners and often involve the Achilles and posterior tibial tendons. Ballet dancers typically present with flexor hallucis longus tendinitis, and sports involving jumping (e.g., basketball, volleyball) often cause symptoms in the patellar tendon.

Classification of Overuse Injuries Controversy exists in the literature about a universal ­classification of overuse tendon injuries and the pathologic entities responsible for them. A classification of Achilles tendon disorders292 provides a guide to the structural ­manifestations of overuse injury as follows: (1) peritendinitis, or inflammation of the peritendon; (2) tendinosis with peritendinitis; (3) tendinosis without peritendinitis; (4) partial rupture; and (5) total rupture. Others have added a sixth category, tendinitis, in which the primary site of injury is the tendon, and there is an associated reactive peritendinitis.293 This classification is not universal because some tendons lack a paratenon and instead have synovial sheaths. No firm scientific evidence shows that certain histopathologic conditions are actually separate entities. For instance, human biopsy studies have been unable to show histologic evidence of inflammation within the tendon substance.294 Because of uncertainty regarding the histologic features of these conditions, several authors295,296 have suggested use of the term tendinopathy rather than tendinitis. With adjectives such as insertional or mid-substance, or the prefix peri, it is clear where the tendon problem is located. Spontaneous tendon rupture during sporting activities occurs despite laboratory studies showing that under normal circumstances healthy tendon is not the weak link of the musculotendinous unit. Healthy tendon is stronger than its muscle or muscle-tendon junction. Biopsies have shown that preexisting degenerative changes are present in most tendons that spontaneously rupture.297 Histologic tendon degeneration can occur without clinical symptoms.293 Thus, there appears to be preexisting but generally asymptomatic pathology that precedes spontaneous tendon rupture during sport-related activity. Studies have shown that in cases of chronic tendon pain, the pathologic lesion is typical of a degenerative process rather than an inflammatory one, and that this degeneration occurs in areas of diminished blood flow. Several authors have documented the existence of areas of marked degeneration without acute or chronic inflammation in most of these cases.214,299-302 These changes are separate and distinct from the site of rupture. A review of patients with chronic tendinitis syndrome revealed similar findings of tendon degeneration.292,303 Nirschl303 described the pathology of chronic tendinitis

30

DeLee & Drez’s Orthopaedic Sports Medicine

as angiofibroblastic hyperplasia. A characteristic pattern of fibroblasts and vascular, atypical, granulation-like tissue can be seen microscopically.303,304 Cells characteristic of acute inflammation are virtually absent. These observations suggest that factors other than mechanical overuse play an important role in the pathogenesis of these tendon lesions. Age appears to be an important factor. Several studies have identified a correlation between the incidence of chronic tendon problems and age.305,306 In vitro studies have shown decreased proliferative and metabolic responses of aging tendon tissue.307 Other causative ­factors include the lack of blood flow in certain areas, which may predispose a tendon to rupture or may result in chronic tendinopathy. Areas of decreased vascular supply have been shown in the supraspinatus and Achilles tendons.308,309 Rupture of these tendons occurs regularly in the areas of diminished blood supply.293 Biopsy specimens of young patients with chronic tendinitis have revealed a change in the morphology of tenocytes adjacent to areas of collagen degeneration.293 Tendon compression may also have a role in insertional tendinopathy. For example, compression of the tendon at the origin on the inferior pole of the patella may play a role in patellar tendinopathy, or jumper’s knee.310 Finally, fluoroquinolones have been associated with the development of tendinopathy or even tendon ruptures. Several cases of severe tendon problems as a result of drugs such as ciprofloxacin have also been reported.311

Clinical Considerations Diagnosis There are many common clinical presentations of acute and chronic pathologic processes involving tendons. These entities are common in the general population and among athletes in particular.

Trauma-Induced Conditions Some conditions involve trauma-induced or chronic overload changes within the tendon substance. These conditions often involve degenerative changes without significant accompanying inflammatory changes and may be considered a form of tendinosis or degenerative tendinopathy. These conditions cause symptoms of localized pain, especially when the tendon is placed under tension by muscle action or stretch. Often the degeneration occurs near the bony origin or insertion of the tendon as an insertional tendinopathy. The following tendons are commonly involved: 1. Wrist extensors. The common wrist extensor tendon is involved near its origin on the lateral epicondyle. The condition frequently is termed lateral epicondylitis or tennis elbow. 2. Patellar tendon. Pathologic changes occur near the attach­ ment of this tendon on the patella, which lead to a painful condition called jumper’s knee, or infrapatellar tendinitis. It is particularly painful during periods of high tension in the tendon, such as in jumping or landing from a jump.

3. Posterior tibial tendon. Injury and complete rupture may occur near the tarsal attachment. This injury is common in runners and dancers. 4. Supraspinatus tendon. This frequent tendinopathy occurs near the junction of the rotator cuff and the greater tuberosity. Changes can occur within the tendon, causing pain in response to eccentric loading. This injury is common in eccentric loading during exercises typically performed by throwing athletes.

Peritendinitis Conditions characterized by reactive and often inflammatory changes in the peritendinous structures were identified previously as peritendinitis. Peritendinitis may occur with or without tendinosis. The following tendons are examples of those that may be associated with peritendinitis: 1. Achilles tendon. The peritendinous tissue around the tendon may show significant fibrosis and inflammatory changes. Significant tendon degeneration in the presence of inflammation of the peritendinous tissue (i.e., the paratenon or tendon sheath) is referred to as tendinosis with peritendinitis. This condition has been documented in competitive runners.300 2. Rotator cuff tendons. In addition to the tendinosis noted earlier, significant involvement of the peritendinous tissue may occur with or without degenerative changes in the rotator cuff tendon. The subacromial bursa may be considered part of the peritendon of the rotator cuff, and its involvement is common in the well-known impingement syndrome. The symptoms often emanate from inflamed peritendinous tissue rather than from traction or tension within the tendon. The shoulder pain experienced in impingement syndrome often occurs during complete rest and at night rather than in response to tendon loading. 3. Flexor hallucis longus. A common injury in dancers, peritendinitis of this tendon is characterized by pain from the peritendon at a site of injury near the tendon sheath posterior to the medial malleolus. 4. Abductor pollicis longus. Inflammation within the peritendon here is caused by a tight sheath in the first extensor compartment of the wrist. These examples show that changes can occur within the tendon, the peritendon, or both. Usually these conditions are characterized by pain. They may lead to complete failure of a tendon with resulting loss of function. Complete failure of the Achilles tendon and the rotator cuff are perhaps the best-known conditions of tendon failure. As discussed earlier, tendon failure usually involves a tendon that has a preexisting degenerative change or a tenuous area of vascular supply.293,308,309

Treatment of Problems of the Tendon and Peritendon Most pathologic conditions involving tendinous and peritendinous structures are mild and resolve spontaneously. Time and changes in the routine of physical activity are often successful in alleviating symptoms. Restriction of physical activity can include anything from complete rest to

Basic Science and Injury of Muscle, Tendon, and Ligament

maintenance of vigorous exercise with avoidance of motions that cause pain. Nonsteroidal anti-­inflammatory drugs are useful for their analgesic effects. The anti-inflammatory effects of nonsteroidal anti-inflammatory drugs may not affect these problems directly because acute inflammatory changes do not appear to be a prominent feature in most of these chronic tendon problems. In several controlled studies, the efficacy of these drugs appears similar to analgesics without known anti-inflammatory effects or even to placebo.312,313 Therapeutic exercises are prescribed frequently as ­treatment for painful conditions of tendons. The added stimulus of training may help to strengthen the injured area or the surrounding normal tendon. Eccentric exercises have been used widely with promising results.314 These exercises ­ usually are designed to increase the strength of the tendon and are performed at a highly intense level. ­Traction injuries to the rotator cuff often are treated by vigorous strengthening of the shoulder abductors and external rotators. Athletes using overhead motions, such as baseball pitchers and tennis players, may benefit from a program of preventive exercise. Corticosteroid injections may be used in the management of tendon problems to treat inflammatory involvement of the peritendinous structures. Direct injection into the tendons usually is avoided because glucocorticoids inhibit collagen biosynthesis and because the nature of the intratendinous pathology often involves no inflammation. When inflamed peritendinous structures are responsible for the pain, injections may be quite helpful. Injections of the subacromial bursa, the retrocalcaneal bursa, and the first dorsal compartment of the wrist (for de Quervain’s tenosynovitis) can be useful clinically. Complete rupture of tendons is frequent. In injuries occurring in athletes, these conditions usually are best treated by surgical repair. Rotator cuff tears are best treated acutely. Achilles tendon ruptures can be treated conservatively227,315,316; however, strength recovery may be better with surgery, and the risk for reinjury is lower with surgery.317-319 Good surgical repair diminishes the need for immobilization and restricted activity. Certain conditions involving the tendons create significant pain and may be treated surgically when they do not respond to conservative management. As discussed previously, cases of failed response to treatment of chronic overuse injuries usually involve pathology of the tendon. This pathology appears to involve a chronic degenerative process secondary to diminished vascular supply rather than acute inflammation. These conditions may involve pathology within the tendons; examples are tennis elbow and jumper’s knee. Tennis elbow can be treated successfully by surgical removal of the abnormal portion of the tendon of origin of the wrist extensors. Similarly, jumper’s knee may respond to excision of the abnormal portion of the patellar tendon near its attachment to the patella. Growth factors offer a potentially promising advancement in the treatment and reconstruction of tendon injuries. Platelet-­derived growth factor-BB, IGF-I, and b-FGF have been shown to induce proliferation of tendon cells in cultured rabbit flexor tendon cells.320 Inhibition of TGF-β1 in rabbit flexor tendons in vitro reduced the production

C l Muscle

r i t i c a l

P

31

o i n t s

strains account for about half of all athletic injuries.

l Most muscle injuries involve powerful eccentric contractions

and pathologic changes at the muscle-tendon junction. in muscle that are induced by strength and conditioning exercise are important in the prevention of acute injury; stronger, well-conditioned muscles are less likely to be injured during physical activity. l Healthy tendon rarely ruptures because it is stronger than its corresponding muscle and muscle-tendon junction. l Tendon ruptures are often attributable to preexisting, noninflammatory tendon degeneration and pathology ­resulting from various factors including repetitive trauma (overuse), certain medications (e.g., fluoroquinolones), age, and compression. l Treatment of muscle and tendon injuries should be specific to the type of tissue damage and the underlying pathophysiology. l Adaptations

of collagen I, which may be useful in modulating fibrous scar formation and adhesions.217 Ongoing studies are investigating the responses of specific tendon structures (e.g., tendon vs. sheath) to various growth factors under different dosing regimens.

SUMMARY Pathologic involvement of tendons and peritendinous tissues is common in sports medicine. Certain tendons and certain locations are involved frequently. An attempt should be made to understand the pathologic processes, at least insofar as which tissues are involved and how. This understanding enables a rational approach to treatment and prevention.

S U G G E S T E D

R E A D I N G S

Eming SA, Krieg T, Davidson JM: Gene transfer in tissue repair: Status, challenges and future directions. Expert Opin Biol Ther 4:1373-1386, 2004. Hildebrand KA, Frank CB, Hart DA: Gene intervention in ligament and tendon: Current status, challenges, future directions. Gene Ther 11:368-378, 2004. Hsu C, Chang J: Clinical implications of growth factors in flexor tendon wound healing. J Hand Surg [Am] 29:551-563, 2004. Huang D, Balian G, Chhabra AB: Tendon tissue engineering and gene transfer: The future of surgical treatment. J Hand Surg [Am] 31:693-704, 2006. Jarvinen TA, Jarvinen TL, Kaariainen M, et al: Muscle injuries: Biology and ­treatment. Am J Sports Med 33:745-764, 2005. Lin TW, Cardenas L, Soslowsky LJ: Biomechanics of tendon injury and repair. J Biomech 37:865-877, 2004. Maganaris CN, Narici MV, Almekinders LC, Maffulli N: Biomechanics and ­pathophysiology of overuse tendon injuries: Ideas on insertional tendinopathy. Sports Med 34:1005-1017, 2004. Mehta V, Mass D: The use of growth factors on tendon injuries. J Hand Ther 18:87-92, 2005. Tidball JG: Inflammatory processes in muscle injury and repair. Am J Physiol Regul Integr Comp Physiol 288:R345-R353, 2005. Woo SLY, Renstrom P, Arnocsky SP (eds): Tendinopathy in Athletes. IOC ­Encyclopaedia of Sports Medicine, vol XII. Oxford, UK, Blackwell ­ Publishing Ltd, 2007.

R eferences Please see www.expertconsult.com

32

DeLee & Drez’s Orthopaedic Sports Medicine

S ecti o n

A

Physiology of Injury to Musculoskeletal Structures 2. Ligamentous Injury Mark R. Brinker, Daniel P. O’Connor, Louis C. Almekinders, Thomas M. Best, Joseph A. Buckwalter, William E. Garrett, Jr., Donald T. Kirkendall, Van C. Mow, and Savio L.-Y. Woo

The specialized dense fibrous tissue structures—ligaments, tendons, and joint capsules—form a major component of the musculoskeletal system.1 These tissues consist primarily of highly oriented, tightly packed collagen fibrils that give them flexibility and great tensile strength. Despite their similarities, ligaments and tendons differ in structure and function. Skeletal ligaments and joint capsules have much lower length-to-width ratios than tendons. Unlike tendons, they connect adjacent bones to each other rather than connecting muscle to bone, and they more often form layered sheets or lamellae of collagen fibrils rather than cords. Skeletal ligaments and joint capsules (to a lesser degree) help stabilize synovial and nonsynovial joints by guiding normal joint motion and preventing abnormal motion. They also may have sensory functions in providing joint proprioception and initiating protective reflexes.1,2 Sports activities frequently result in ligament sprains or tears that compromise ligament function and thereby decrease joint stability. Significant joint instability due to loss of ligament function eliminates participation in competitive sports and may lead to degeneration of the affected joint.3,4 This section first outlines the types of ligaments and their structure and composition. Next, the response of ligaments to injury is reviewed, including inflammation, repair, and remodeling. Finally, ligament autografts, allografts, xenografts, and grafts formed from reconstituted collagen are reviewed.

intracapsular ligaments, such as the anterior cruciate ligament of the knee and the ligamentum teres of the hip. The areolar connective tissue that surrounds the joints covers capsular ligaments, including the capsular ligaments of the hip and the glenohumeral joint. Fat and areolar connective tissue separate extracapsular ligaments—such as the coracoacromial ligament, the coracoclavicular ligaments, and the costoclavicular ligaments—from the underlying joint capsule and capsular ligaments. Despite these differences, the presumed primary functions of intracapsular, capsular, and extracapsular ligaments remain the same: stabilizing the relationship between adjacent bones and stabilizing and guiding joint motion.

TYPES OF LIGAMENTS

SUBSTANCE OF LIGAMENTS

Surgeons and anatomists have identified hundreds of skeletal ligaments. Generally, they have named ligaments by their location and bony attachments (e.g., anterior glenohumeral ligament, anterior talofibular ligament) or by their relationship to other ligaments (e.g., medial collateral ligament of the knee, posterior cruciate ligament of the knee). Many ligaments, like the anterior cruciate ligament, form discrete, easily identifiable structures with distinct mechanical functions, but others, like the hip joint capsular ligaments, blend with the joint capsule, making it difficult to define their exact structure and function. The anatomic relationships of ligaments to the joints they help stabilize vary. They may be intracapsular, capsular, or extracapsular. A thin layer of synovium covers

Bundles of collagen fibrils form the bulk of the ligament substance.1,5-7 Some ligaments, including the anterior cruciate ligament of the knee, consist of more than one band of collagen fibril bundles; as the joint moves, different bands become taut.8 The alignment of collagen fiber bundles within the ligament substance generally follows the lines of tension applied to the ligament during normal activities. In addition, light microscopic examination has shown that the collagen bundles have a wave or crimp pattern. The crimp pattern of matrix organization may allow slight elongation of the ligament without incurring damage to the tissue.7 In some regions, the ligament cells align themselves in rows between collagen fiber bundles, but in other regions, the cells lack apparent orientation relative to the alignment of the matrix collagen fibers. Scattered

STRUCTURE OF LIGAMENTS Skeletal ligaments vary in length, shape, and thickness. They cross joints with wide ranges of motion (e.g., the hip and the glenohumeral joint) as well as those with little or no normal motion (e.g., the sacroiliac joint and the proximal tibiofibular joint). Grossly, they appear as firm, white fibrous bands, sheets, or thickened strips of joint capsule securely anchored to bone. They consist of a proximal bone insertion, the substance of the ligament or the capsule, and a distal bone insertion. Because most insertions are no more than 1 mm thick, they contribute only a small amount to the volume and the length of the ligament.

Basic Science and Injury of Muscle, Tendon, and Ligament

blood vessels penetrate the ligament substance, forming small-diameter, longitudinal vascular channels that lie parallel to the collagen bundles. Nerve fibers lie next to some vessels, and nerve endings with the structure of mechanoreceptors have been found in some ligaments.1,2,5

INSERTIONS Insertions of ligaments attach the flexible ligament substance to the rigid bone and allow motion between the bone and the ligament without damage to the ligament. Despite their small size, ligament insertions have a more variable and elaborate structure than the ligament substance,9-13 and they may have different mechanical properties. ­Measurement of ligament mechanical properties shows that during activity, insertions or the ligament regions near insertions deform more than the ligament substance.14 Ligament insertions vary in size, strength, angle of the ligament collagen fiber bundles relative to the bone, and proportion of ligament collagen fibers that penetrate directly into bone.1,11,12 Based on the angle between the collagen fibrils and the bone and the proportion of the collagen fibers that penetrate directly into bone, investigators group ligament insertions into two types: direct and ­indirect.

Direct Insertions Direct ligament insertions into bone, like the femoral insertion of the medial collateral ligament of the knee or the tibial insertion of the anterior cruciate ligament of the knee, consist of sharply defined regions where the ligament appears to pass directly into the cortex of the bone.9-11,13,15 Attempts to separate a ligament with a direct insertion from the bone usually fail unless the surgeon cuts the substance of the ligament through the region next to the insertion. The thin layer of superficial ligament collagen fibers of direct insertions joins the fibrous layer of the periosteum. Most of the ligament insertion consists of deeper fibers that directly penetrate the cortex, often at a right angle to the bone surface. In ligaments that approach the bone surface at a right angle, the ligament collagen fibers follow a straight line into the bone, but in ligaments that approach the bone surface at an oblique angle, the ligament collagen fibers make a sharp turn to enter the insertion. The angle of ligament collagen fibril insertion may not be apparent from gross examination. For example, the medial collateral ligament of the femur approaches the surface of the femur obliquely; grossly, it appears to insert at a 50- to 70-degree angle, but microscopic study shows that the collagen fibers of the ligament enter the bone at about a 90-degree angle. The deeper collagen fibers pass through four zones with increasing stiffness: ligament substance, fibrocartilage, mineralized fibrocartilage, and bone.9-11,13 In the fibrocartilage zone, the cells become larger and more spherical than the cells in most regions of the ligament substance. A sharp border of mineralized and unmineralized matrix separates the fibrocartilage zone from the mineralized fibrocartilage zone. From this latter zone, the ligament collagen fibers pass into the bone and blend with the bone collagen fibers.

33

Indirect Insertions Indirect or oblique ligament insertions into bone,9-11,13,15,16 such as the tibial insertion of the medial collateral ligament of knee or the femoral insertion of the lateral collateral ligament, are less common than direct insertions. They usually cover more bone surface area than direct insertions, and their boundaries cannot be easily defined because the ligament passes obliquely along the bone surface rather than directly into the cortex. Ligaments with indirect insertions into bone can often be elevated from the bone without ­cutting the ligament substance and may not have a zone of fibrocartilage. Like direct insertions, indirect insertions have superficial and deep collagen fibers, but the superficial ligament collagen fibers passing into the fibrous layer of the periosteum form most of the substance of indirect insertions. The deeper collagen fibers of indirect insertions approach the bone cortex at oblique angles and do not pass through well-defined fibrocartilage zones. The structure of indirect insertions, particularly the distribution of ligament fibers between bone and periosteum, may change with skeletal development and alter the mechanical properties of the insertions. A study of the maturation of the rabbit medial collateral ligament insertion into the tibia showed that with skeletal growth, more of the ligament collagen fibers entered the bone, decreasing the contribution of the ­periosteal component to the insertion.16 Changes ­occurring with age in the periosteum may also alter the structure and the mechanical properties of the insertions.13

COMPOSITION OF LIGAMENTS Different ligaments and different regions of the same ligament may vary slightly in matrix composition and in cell shape and density.6,17,18 They all consist, however, of fibroblasts surrounded by an extracellular matrix formed by a highly ordered arrangement of macromolecules, primarily type I collagen, and water that fills the macromolecular framework.1,6 The composition of the matrix, the organization of the matrix macromolecules, and the interaction between the matrix macromolecules and the tissue water determine the mechanical properties of the tissue. The biochemical composition of ligament insertions has not been studied extensively because of the difficulties of isolating and analyzing such small volumes of tissue, but, like ligament substance, the insertions consist of a small number of cells surrounded by an abundant extracellular matrix formed mainly by type I collagen fibrils.

Cells Fibroblasts are the dominant cells of ligaments, although endothelial cells of small vessels and nerve cell processes are also present.1,6,10-12,19 Fibroblasts form and maintain the extracellular matrix. They vary in shape, activity, and density among ligaments, among regions of the same ligament, and with the age of the tissue.1,6,11,12,19 Many fibroblasts are spindle shaped and extend between the collagen fibrils.11 Younger ligaments have a higher concentration of cells that synthesize new matrix. These cells frequently have a large volume of cytoplasm containing substantial

34

DeLee & Drez’s Orthopaedic Sports Medicine

amounts of endoplasmic reticulum. With increasing age, the cells become less active, but at any age, they synthesize the matrix macromolecules necessary to maintain the structure and the function of the tissue.

Tissue fluid contributes 60% or more of the wet weight of most ligaments. Because many ligament cells lie at some distance from vessels, these cells must depend on diffusion of nutrients and metabolites through the tissue fluid. The interaction of the tissue fluid and the matrix macromolecules influences the mechanical properties of the tissue.

articular cartilage-type proteoglycans contain long, negatively charged chains of chondroitin and keratan sulfate. Smaller proteoglycans contain dermatan sulfate. Because of their long chains of negative charges, the large articular cartilage-type proteoglycans tend to expand to their maximal domain in solution until restrained by the collagen fibril network. As a result, they maintain water within the tissue and exert a swelling pressure, thereby contributing to the biomechanical properties of the tissue and filling the regions between the collagen fibrils. The small dermatan sulfate proteoglycans usually lie directly on the surface of collagen fibrils and appear to affect formation, organization, and stability of the extracellular matrix, including collagen fibril formation and diameter, rather than having a direct mechanical role.25,26

Matrix Macromolecules

Noncollagenous Proteins

Four types of molecules (collagen, elastin, proteoglycans, and noncollagenous proteins) form the molecular framework of the ligament matrix and constitute about 40% of the wet weight of most ligaments.1,6

Although noncollagenous proteins contribute only a small percentage of the dry weight of dense fibrous tissues, they appear to help organize and maintain the macromolecular framework of the collagen matrix, aid in the adherence of cells to the framework, and possibly influence cell ­function. One noncollagenous protein, fibronectin, has been identified in the extracellular matrix of ligaments and may be associated with several matrix component molecules as well as with blood vessels. Other noncollagenous proteins undoubtedly exist within the ligament matrix, but their identity and their functions have not yet been defined. Many of the noncollagenous proteins also contain a few monosaccharides and oligosaccharides.1,6,10

Matrix Water

Collagen Fibrillar collagen has the form of cylindrical cross-banded fibrils when examined with electron microscopy. These fibrils give ligaments their form and their tensile strength and constitute 70% to 80% of the dry weight of the ligament. Type I collagen is the major component of the molecular framework, composing more than 90% of the collagen content of ligaments. Type III collagen constitutes about 10% of the collagen, and small amounts of other collagen types may be present as well. Ligaments have more type III collagen than do tendons.17

Elastin Most ligaments have little elastin (usually less than 5%), but a few, such as the nuchal ligament and the ligamentum flavum, have high concentrations (up to 75%). Elastin forms protein fibrils or sheets, but elastin fibrils lack the cross-banding pattern of fibrillar collagen and differ in amino acid composition, including two amino acids not found in collagen (desmosine and isodesmosine). Also unlike collagen, elastin amino acid chains form random coils when the molecules are unloaded. This conformation of the amino acid chains makes it possible for elastin to undergo some deformation without rupturing or tearing and then, when the load is removed, to return to its original size and shape.

Proteoglycans Proteoglycans form only a small portion of the macromolecular framework of the ligament, usually less than 1% of the dry weight,6 but may have important roles in organizing the extracellular matrix and interacting with the tissue fluid.1,10,20-25 Most ligaments have a higher concentration of glycosaminoglycans than do tendons.17 Like tendon, meniscus, and articular cartilage, ligaments contain two known classes of proteoglycans. Larger,

HEALING OF LIGAMENTS Ligament healing (i.e., restoration of structural integrity following injury) depends on the response of the tissue to injury.27,28 Ligament strains and tears disrupt the matrix, damage blood vessels, and injure or kill cells. Damage to cells, matrices, and blood vessels and the resulting hemorrhage start a response that includes inflammation, repair, and remodeling.28,29 These events form a continuous sequence of cell, matrix, and vascular changes that begins with the release of inflammatory mediators and ends when remodeling ceases.28

Inflammation Inflammation is the cellular and vascular response to injury that includes release of inflammatory mediators, vasodilation, increased blood flow, exudation of plasma, and migration of inflammatory cells.28 These tissue events cause swelling, erythema, increased temperature, pain, and impaired function. Acute inflammation lasts 48 to 72 hours after most ligament injuries and then gradually resolves as repair progresses. Some of the events that occur during inflammation, including the release of cytokines or growth factors, may help stimulate tissue repair.28 Inflammation begins immediately after injury as inflammatory mediators are released from damaged cells. These mediators promote vascular dilation and increase vascular permeability, leading to exudation of fluid from vessels in the injured region, which causes tissue edema. Ligament

Basic Science and Injury of Muscle, Tendon, and Ligament

tissue immediately surrounding the injury becomes swollen and increasingly friable. Uninjured ligament tissue at a distance from the injury also swells as a result of exudation of fluid from the dilated vessels. Blood escaping from the damaged vessels forms a hematoma that temporarily fills the injured site. Fibrin accumulates within the hematoma, and platelets bind to fibrillar collagen, thereby achieving hemostasis and forming a clot consisting of fibrin, platelets, red cells, and cell and matrix debris. The clot provides a framework for vascular and fibroblast cell invasion. As they participate in clot formation, platelets release vasoactive mediators and the cytokines or growth factors (transforming growth factor-β and platelet-derived growth factor). Within hours of the injury, polymorphonuclear leukocytes appear in the damaged tissue and the clot. Shortly thereafter, monocytes arrive and increase in number until they become the predominant cell type. Enzymes released from the inflammatory cells help digest necrotic tissue, and monocytes phagocytose small particles of necrotic tissue and cell debris. Endothelial cells near the injury site begin to proliferate, creating new capillaries that grow toward the region of tissue damage. Release of chemotactic factors and cytokines from endothelial cells, monocytes, and other inflammatory cells helps to stimulate migration and proliferation of the fibroblasts that begin the repair process.28

Repair Ligament repair involves the replacement of necrotic or damaged tissue by cell proliferation and synthesis of new matrix.28 Repair depends on the fibroblasts that migrate into the injured tissue and clot. Within 2 to 3 days of the injury, fibroblasts within the wound begin to proliferate rapidly and synthesize new matrix. They replace the clot and the necrotic tissue with a soft, loose fibrous matrix ­containing high concentrations of water, glycosaminoglycans, and type III collagen. Inflammatory cells and fibroblasts fill this initial repair tissue. Within 3 to 4 days, vascular buds from the surrounding tissue grow into the repair tissue and then canalize to allow blood flow to the injured tissue and across small tissue defects. This vascular granulation tissue fills the tissue defect and extends for a short distance into the surrounding tissue but has little tensile strength. During the next several weeks, as repair progresses, the composition of the granulation tissue changes. Water, glycosaminoglycan, and type III collagen concentrations decline, the inflammatory cells disappear, and the concentration of type I collagen increases. Newly synthesized collagen fibrils increase in size and begin to form tightly packed bundles, and the density of fibroblasts decreases. Matrix organization increases30-33 as the fibrils begin to align along the lines of stress; the number of blood vessels decreases; and small amounts of elastin may appear within the site of injury. The tensile strength of the repair tissue increases as the collagen concentration increases.

Remodeling Repair of many ligament injuries results in an excessive volume of highly cellular tissue with limited mechanical properties and a poorly organized matrix. Remodeling reshapes and strengthens this tissue by removing, reorganizing,

35

and replacing cells and matrix.28 In most ligament injuries, ­evidence of remodeling appears within several weeks of injury as fibroblasts and macrophages decrease, fibroblast synthetic activity decreases, and fibroblasts and collagen fibrils assume a more organized appearance. As these changes occur in the repair tissue, collagen fibrils grow in diameter, the concentration of collagen and the ratio of type I to type III collagen increase, and the water and proteoglycan concentrations decline. Unfortunately, the collagen fiber diameters do not increase to normal because of elevated levels of collagen V. In the months following injury, the matrix orientation continues to align, presumably in response to loads applied to the repair tissue. Although the matrix orientation improves, this is not reflected in the healing tissue’s mechanical properties29,34-36; for example, it has 50% to 70% of normal tensile strength. Generally, this decrease in tissue quality does not cause clinically apparent disturbances of joint function because the volume (cross-­sectional area) of the repaired tissue remains greater than the ­volume of uninjured ligament. The most apparent signs of remodeling disappear within 4 to 6 months of injury: The cell density and the number of small blood vessels decline to near-normal levels, and the collagen concentration increases nearly to normal. Removal, replacement, and reorganization of repair tissue, however, continue to some extent for years.32 Although mature repair tissue restores the structural integrity of the ligament, it can usually be distinguished from normal tissue and may differ in ­composition. Most important, as already noted, mature ligament repair ­tissue lacks the tensile strength of normal ligament tissue.29,34-36

Variables That Influence Ligament Healing Many variables influence ligament healing.28 Among the most important variables are the type of ligament, the size of the tissue defect, and the amount of loading applied to the ligament repair tissue. Injuries to capsular and extracapsular ligaments stimulate production of repair tissue that will fill most defects, but injuries to intracapsular ­ligaments, such as the anterior cruciate ligament, often fail to produce a successful repair response. This may be due to the synovial environment, limited vascular ingrowth, and fibroblast migration from surrounding tissues or other ­factors. The extent of the injury and the type of treatment can influence the volume of necrotic tissue and the size of the tissue defect. Treatments that achieve or maintain apposition of torn ligament tissue and that stabilize the injury site decrease the volume of repair tissue necessary to heal the injury. They may also minimize scarring and help provide near-normal tissue length. For these reasons, avoidance of wide separation of ruptured ligament ends and selection of treatments that maintain some stability of the injured site during the initial stages of repair are generally desirable. Early controlled loading of ligament repair tissue can promote healing, but excessive loading will disrupt repair tissue and delay or prevent healing.1,28,34,37-41

LIGAMENT GRAFTS Most ligament sprains and tears heal satisfactorily and restore the structural integrity of the ligament. By contrast, ruptures of some ligaments, such as the anterior cruciate

36

DeLee & Drez’s Orthopaedic Sports Medicine

ligament, or severe or untreated injuries of other ligaments may fail to heal.28 Failure to heal or loss of ligamentous tissue can leave the patient with an unstable joint. For this reason, surgeons have used autografts and allografts to reconstruct ligaments that have failed to heal or are likely to fail to heal. Currently, autografts and allografts are most commonly used in reconstruction of a deficient anterior cruciate ligament, although other ligaments have also been reconstructed with dense fibrous tissue grafts. Xenografts have not been widely used for ligament reconstruction, and artificial ligaments composed of reconstituted collagen remain experimental. After transplantation, dense fibrous tissue autografts and allografts undergo remodeling, which includes revascularization, repopulation with cells from the recipient site, and synthesis of new matrix by these cells.42-46 When placed in a recipient site, grafts also undergo a change in biomechanical properties that adversely affects their ability to withstand large tensile loads.41,44 For example, the central and medial portions of the patellar tendon have ultimate loads higher than those of the anterior cruciate ligament,47 but transplantation of these portions of the patellar tendon to reconstruct the anterior cruciate ligament reduces the ultimate load to 10% to 30% of the femur–anterior cruciate ligament–tibial complex.8,45,48 As the graft heals to the recipient tissues and undergoes remodeling, its mechanical properties improve, but it never approaches the stiffness and the strength of the graft tissue before transplantation. Despite the failure of grafts to restore the biomechanical properties of ligaments to normal in experimental studies, clinical evaluations show that they can significantly improve the stability of joints with ligamentous insufficiency.49-53 Clinical testing may fail to show the weakness of grafts demonstrated in experimental studies because clinical tests of ligament function use small loads and do not attempt to test the ultimate strength of the reconstruction. Healing of the site of the graft attachment may be responsible for most of the increase in graft strength observed after transplantation. One study of rabbit patellar tendon autografts showed that immediately after transplantation, most grafts failed at the femoral fixation site; 6 weeks later, the grafts ruptured at the substance within the tibial tunnel; and 30 weeks after transplantation, five of six grafts ruptured in the intra-articular part of the graft, and one failed within the femoral bone tunnel. Fifty-two weeks after transplantation, all grafts failed in the intra-articular part of the graft substance. These results show that initially the weakest regions of the reconstruction are the bone attachment sites and the parts of the graft within the bone tunnels. Once these regions incorporate into the bone, presumably increasing their strength, the intra-articular parts of the graft become the weakest regions.

Autografts Autografts can be transplanted either with a vascular pedicle that maintains their blood supply or as free tissue. Tissues used for ligament autografts include fascia lata, portions of the patellar tendon, other tendons, and meniscus.49 Clinically, autografts heal and can be used in reconstructive surgery, although they may differ in structure and mechanical properties from the original ligament.

Nonvascularized Autografts Harvesting an autograft leaves the matrix intact, but disruption of the blood supply causes ischemic necrosis or injury to cells in the graft.54 Graft cells can survive transplantation,55 but the proportion of cells that survive has not been determined. Soon after transplantation, inflammatory cells, fibroblasts, and vessels at the recipient site invade the graft and increase cell density.54,56,57 As the new cells and vessels penetrate the substance of the matrix, the graft usually swells. The increased water content decreases graft stiffness and strength.58 The remodeling process will begin to take place once the new cells establish themselves within the graft. The ­collagen fibril alignment within the graft initially appears disorganized, but progressive realignment occurs, and edema and cell density decrease until the histologic appearance of the graft approaches that of the normal tissue.56 Changes in the biochemical composition of the graft matrix accompany the changes in histologic appearance. Patellar tendon normally lacks type III collagen and has a low concentration of glycosaminoglycans. After a patellar tendon transplantation to reconstruct the anterior cruciate ligament, type III collagen and glycosaminoglycan concentrations in the graft increase until they reach levels normally found in the anterior cruciate ligament, and the pattern of collagen cross-links changes to resemble that found in the anterior cruciate ligament.17,59 Despite their normal appearance and the changes in their matrix composition, autografts generally have significantly less stiffness and strength than normal ligaments. Most studies show that they reach only about 20% to 40% of normal values 6 months to a year after implantation.56,58,60 Loading of the grafts appears to influence graft healing and remodeling. Excessive or insufficient tension may adversely affect revascularization and remodeling of the graft. A series of studies on anterior cruciate ligament reconstruction using patellar tendon autograft in dogs showed that grafts fixed under high tension (39 newtons [N]) had poor revascularization and myxoid degeneration 3 months after surgery.61 Grafts fixed under low tension (1 N) showed more extensive revascularization, no degeneration, and greater strength and stiffness. Augmentation of the autograft with a Dacron prosthesis increased the stiffness and ultimate load (strength) of the reconstruction in the immediate postoperative period, presumably because of the better initial fixation of the prosthetic material.62 By 3 months after surgery, the strength of the augmented grafts had decreased, whereas the strength of the grafts without Dacron augmentation had increased. Dacron augmentation apparently delayed revascularization and remodeling of the grafts such that the declining strength of the augmented grafts resulted from the delay in remodeling combined with degradation of the Dacron material. The inhibition of graft remodeling may have resulted from decreased loading of the graft, inasmuch as no delay of revascularization and remodeling was observed when the synthetic graft was not placed parallel to the autograft.63 The extent to which a graft restores the normal anatomy of a ligament and is subjected to normal loading, as well as the presence of well-vascularized surrounding tissues that can supply cells and vessels to invade the graft, may

Basic Science and Injury of Muscle, Tendon, and Ligament

influence the results of ligament autograft procedures. Under ideal conditions, fresh medial collateral ligament autografts in rabbits can achieve near-normal strength.64 The bone–medial collateral ligament complexes were removed completely and then replaced using internal fixation. The autograft ligament substance and insertions showed early weakening, but graft strength gradually increased over time. The graft complexes showed their lowest failure load (65% of control values) at 24 weeks. By 48 weeks, the failure load had increased to about 90% of control values, and the structural properties of the bone-graft-bone complex could not be distinguished from those of control samples. Immediate anatomic replacement of a ligament with rapid cell and vascular invasion of the graft from surrounding soft tissues may represent the best circumstance for optimal mechanical recovery. Even under these optimizing circumstances, healing and incorporation of the autograft proceed slowly; whether an autograft transplanted under ideal biologic and mechanical conditions can eventually recover the stiffness and strength of the normal ligament remains uncertain.

Vascularized Autografts Methods of maintaining the vascularity of autografts have been developed in an effort to decrease ischemic necrosis and avoid the need for recipient site cells to revascularize and repopulate the graft.8,65-69 Unfortunately, studies of the healing and biomechanical properties of vascularized and nonvascularized patellar tendon graft reconstructions of the anterior cruciate ligament in dogs and primates do not show significant advantages for the vascularized autografts.65,66 The vascularized grafts initially had a better blood supply and matrix organization than the nonvascularized grafts,66 but 12 weeks after transplantation, little or no difference in matrix organization, cell population, blood supply, or graft strength was evident between the two graft types.66 Similar findings have been noted when comparing vascularized and nonvascularized anterior cruciate ­ligament autografts in cynomolgus monkeys.65

Specific Tissues Used for Autografts Iliotibial Band and Fascia Lata The ease and low morbidity of harvesting the distal portion of the fascia lata, the iliotibial band, make it an attractive source of tissue for ligament autografts,70 but experimental studies show that these grafts fail to restore the normal biomechanical properties of the anterior cruciate ligament complex.57,71-73 For example, 4 years after anterior cruciate ligament reconstruction using the iliotibial tract in dogs, the grafts remained considerably weaker than normal anterior cruciate ligament complexes, and the treated knees showed evidence of instability and arthritic change. The mean ultimate load to failure of iliotibial tract grafts used to reconstruct anterior cruciate ligaments in dogs has been shown to be about 40% of control values.57,73 Fascia lata grafts used to replace the anterior cruciate ligament in goats produced similar results: 2 months after surgery, mean graft stiffness was no more than 10% of control value

37

for the anterior cruciate ligament, and mean graft ultimate load strength was no more than 15% of control value.71 Eight weeks after reconstruction, the anterior-posterior translation of the treated knees was about 7 times greater than that of control knees.

Patellar Tendon Because of their size, their strength, and their availability, autografts consisting of the central or medial third of patellar tendon, together with the corresponding bony insertion sites, have been widely used for reconstruction of the anterior cruciate ligament.68,69,74,75 Experimental studies, however, show that despite the great strength of these grafts before transplantation, the structural properties of the grafts fail to reach those of the normal anterior cruciate ligament complex after transplantation.76 In most studies, patellar tendon grafts had about one third of the ultimate load of controls 24 months after reconstruction; in some studies, the reconstructed joints showed increased anterior-posterior translation and evidence of joint ­degeneration. Experimental investigations in dogs have confirmed that patellar tendon autografts fail to achieve the normal stiffness and strength of the anterior cruciate ligament.61,62 One group of investigators found that the ultimate load values of patellar tendon autografts in dogs were about 72% of normal anterior cruciate ligament values, with equivalent stiffness.61,62 After implantation, the ultimate load values of the patellar tendon graft complex declined to about 10% of control values, and stiffness declined to about 11% of control values. Three months after implantation, the ultimate load value had increased to 20% of control values, and graft stiffness had increased to 22% of control values, but the grafted knees had nearly 3 times the anterior-posterior instability as the control knees. Twenty months after surgery, the translation of the grafted knees was within 1 mm of the control knees, but the ultimate load values had reached only 32% of control values. Four to 6 months after reconstruction, stiffness, ultimate load to failure, and energy absorbed to failure of patellar tendon grafts are commonly less than half normal values, and joints with patellar tendon grafts have increased laxity.76-78 Studies of goats, rabbits, and rhesus monkeys also showed that patellar tendon autografts used in reconstructions of anterior cruciate ligaments did not develop the properties of the normal ligaments.56,77,79 In general, ultimate load and stiffness both recover to less than half of the normal ligament values, and the reconstructed knees are more lax and more likely to show degenerative changes over a period of 1 to 2 years after implantation.

Semitendinosus Tendon Although less extensively studied than patellar tendon autografts, the semitendinosus tendon provides a clinically useful alternative tissue graft for anterior cruciate ligament reconstruction or augmentation of anterior cruciate ligament repairs.49,52,80-82 The ultimate load of semitendinosus tendon anterior cruciate ligament reconstructions in rabbits was less than 30% of controls; 26 weeks after reconstruction, the grafts failed at loads of less than 15% of

38

DeLee & Drez’s Orthopaedic Sports Medicine

control values, and the reconstructed knees showed degenerative changes.83

Allografts Dense fibrous tissue allografts acquired from carefully selected donors and processed and stored according to established standards and guidelines, such as those of the American Association of Tissue Banks, offer an acceptable alternative to autografts.84-86 Allografts in humans have not shown evidence of clinically detectable rejection or a higher rate of infection than would be expected in autografts.84-86 Because allograft reconstruction requires no harvesting of an autograft, surgical time is decreased and complications at the donor site are eliminated, which is why many surgeons use allografts to reconstruct the anterior cruciate ligament.50,53,85 Less frequently, allografts are used to reconstruct other ligaments, including the medial collateral ligament of the knee. Allografts may consist of the substance of donor fascia, tendon, or ligament, or they may consist of tendon or ligament and their bone insertions. Allografts that include bone usually allow better immediate fixation to the ­recipient site bone and retain the structure of the insertion. Dense fibrous tissue allografts that lack bone insertions heal to the recipient site,85 but they may have less secure initial fixation and possibly greater laxity after healing.84 Because they do not include allograft bone, allografts consisting only of dense fibrous tissue have a lower potential of host immunologic reactions. Examination of allografts at intervals after transplantation shows invasion of host cells and vessels and healing to the host tissue.43,46,73,85 Although some allografts may remodel more slowly than autografts,45 most studies show that incorporation and remodeling of allografts closely resemble the characteristics of autografts.48,78,85,87,88 After healing and remodeling, comparable types of allografts and autografts show similar strength and stiffness.48,78,87,88 Fresh allografts can incite a prominent inflammatory reaction, including lymphocytic infiltration and synovitis.43 Freezing the grafts decreases this reaction; in animal experiments, immunologic reactions from frozen and freeze-dried allografts have not led to destruction of the grafts.43,48,78,87,89 Consequently, most ligament allografts are currently preserved by deep freezing or freeze-drying. Animal and human studies show that frozen and freezedried grafts may elicit humoral and cellular immune responses48,73,90 but the available evidence suggests that these methods do not adversely alter the biomechanical properties of the grafts.87 Allograft sterilization, however, may adversely affect graft mechanical properties. Donor age may also adversely affect the mechanical properties of allografts. With increasing age, the stiffness and the ultimate load of human anterior cruciate ligaments and other dense fibrous tissues decline.91 Therefore, older donors may not provide optimal allografts for ligament reconstruction.

Anterior Cruciate Ligament Allografts Several groups of investigators have examined the results of bone–anterior cruciate ligament–bone allograft reconstructions of the anterior cruciate ligament in animals.48,78,87-89,92-94 These grafts have been found to recover less than one third

of the ultimate load and stiffness values and to have greater instability compared with the unoperated control knee. Other animal experiments have measured the strength of dense fibrous tissue allografts that do not include bone insertions. Although most showed that ultimate load strength was less than one third of the normal ligament up to 30 weeks after reconstruction,78,88,94 one study reported an ultimate load of 67% of those for the control values at 24 weeks.92

Medial Collateral Ligament Allografts Because vascularized soft tissues surround medial collateral ligament allografts, cells and vessels can potentially invade these grafts more rapidly than is possible with anterior cruciate ligament allografts.95 Medial collateral ligament allografts may remodel more rapidly and achieve greater average strength relative to control ligaments than anterior cruciate ligament allografts.95 The reported experiments in dogs and rabbits, however, suggest that as with allograft reconstruction of the anterior cruciate ligament, allograft reconstruction of the medial collateral ligament restores only part of the normal strength of the ligament.45,95,96 Although the biomechanical properties of allografts slowly improve, they may never achieve normal values. When allografts have healed to the recipient site tissues and have been revascularized by host blood vessels and repopulated with new cells, they appear to reach a state of equilibrium in which their material properties neither improve nor deteriorate. This equilibrium state may provide sufficient mechanical strength to withstand the loads associated with normal activities but may not be sufficient to withstand unusually high loads.

Xenografts The ready availability of xenografts (animal tissue) makes them a potentially attractive tissue for reconstruction of ligaments. The small number of appropriate human donors and the technical difficulties and expense of acquiring ideal human allografts limit their availability. Bovine xenografts from young animals could be acquired, prepared, and stored in large quantities, thereby reducing the cost relative to allografts. Experimental studies suggest that xenograft ligament reconstructions partially restore ligament function and that the host tissues invade and begin to incorporate and replace the xenograft collagen.97-99 Bovine xenograft reconstructions of dog medial collateral ligaments had near-normal loads to failure 4 months after operation, and the host fibroblasts and vessels progressively invaded the grafts.33 The host cells appeared to synthesize new matrix within the grafts. Early results of ligament reconstruction with xenografts in humans showed that the grafts could restore anterior cruciate ligament function. One clinical follow-up study of 40 knees that had been treated for anterior cruciate ­ deficiency with bovine tendon xenografts preserved in glutaraldehyde, however, showed a high incidence of graft failure and synovitis.100 About half the grafts ruptured between 12 and 20 months after the operation. Six of the first 30 knees developed severe synovitis within 8 months of surgery, requiring total synovectomy and graft removal. The authors subsequently found no evidence of synovitis in 10 knees following a change in their graft-rinsing ­procedure.

Basic Science and Injury of Muscle, Tendon, and Ligament

Reconstituted Collagen Implants Ligament prostheses formed from reconstituted collagen offer a possible alternative to xenografts.34,101,102 Preliminary experiments show that host cells do invade these implants and that they can develop strength similar to autogenous grafts. Future developments may allow these biologic implants to be created in various sizes and shapes with different concentrations and orientations of collagen fibrils and to incorporate cytokines or other growth factors to stimulate cell migration, proliferation, and differentiation.34

NEW TREATMENT STRATEGIES IN LIGAMENT HEALING Novel treatment strategies based on functional tissue engineering have shown some promise in restoring the normal function of injured ligaments and tendons.103 These approaches include innovative biologic and bioengineering techniques by means of using growth factors, gene transfer therapy, cell therapy, scaffolding materials, and mechanical stimuli. In particular, the use of biologic scaffolds, namely, the porcine small intestine submucosa, offers distinct promise in accelerating ligament and tendon healing and regeneration.104 Furthermore, the small intestine submucosa can be modified in vitro by seeding marrow stromal cells on the scaffold and applying cyclic stretching to increase the alignment of cells, as well as to improve the production and orientation of collagen. Hence, when applied in vivo, the tissue-engineered scaffold could serve to accelerate the initiation of the healing process, ultimately helping to make a better neo-ligament or tendon. Biotechnology in general has seen many recent exciting developments, such as the sequencing of the human genome, stem cell–based therapies, and the promise of tissue engineering. Still, these opportunities present many challenges as the knowledge gained about a particular gene, protein, or cell is eventually translated to a clinical application. Solving such complicated issues will require that experts from different disciplines work together. The role of biomedical engineers within this framework can help link interactions at various levels of scale: molecules to cells, cells to tissues, tissues to organs, and organs to body function. Such a seamless interaction of talented biologists, biomedical engineers, and clinicians, as well as experts from many other disciplines, will lead to therapies that allow injured ligaments and tendons to heal closer to normal. With the help of funding agencies made aware of the need for this research, the efforts of a team-based approach, and the new developments in functional tissue engineering, the future appears bright.

C

r i t i c a l

P

39

o i n t s

l Ligaments

are composed primarily of collagen fiber bundles that are aligned in parallel to normal lines of tension. l Ligament insertions may be direct, in which most of the collagen fibers pass into the cortex of the bone, or indirect, in which most of the collagen fibers pass obliquely into the fibrous layer of the periosteum. l Ligament healing follows the basic injury response of inflammation, repair, and remodeling. Inflammation ­ involves release of inflammatory mediators, increased blood flow, exudation of plasma, and migration of ­inflammatory cells. Repair entails cell proliferation and synthesis of new matrix, which depends on proliferation of fibroblasts. Remodeling reshapes and strengthens the repair tissue by removing, reorganizing, and replacing cells and matrix to align the matrix orientation along lines of mechanical stress. l Ligament grafts currently in use include vascularized autograft, nonvascularized autograft, and allograft. Xenografts and reconstituted collagen grafts are currently under experimental investigation. l Functional tissue engineering approaches, ���������� including growth factors, gene transfer therapy, cell therapy, scaffolding materials, and mechanical stimuli,������������ have shown some potential to restore the normal ligament function.

S U G G E S T E D

R E A D I N G s

Dahners LE, Mullis BH: Effects of nonsteroidal anti-inflammatory drugs on bone formation and soft-tissue healing. J Am Acad Orthop Surg 12:139-143, 2004. Eming SA, Krieg T, Davidson JM: Gene transfer in tissue repair: Status, challenges and future directions. Expert Opin Biol Ther 4:1373-1386, 2004. Frank C, Amiel D, Woo SLY, Akeson W: Normal ligament properties and ligament healing. Clin Orthop 196:15-25, 1985. Frank C, Woo S, Andriacchi T, et al: Normal ligament structure, function, and composition. In Woo SLY, Buckwalter JH (eds): Injury and Repair of the Musculoskeletal Soft Tissues. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1988. Hildebrand KA, Frank CB, Hart DA: Gene intervention in ligament and tendon: Current status, challenges, future directions. Gene Ther 11:368-378, 2004. Petrigliano FA, McAllister DR, Wu BM: Tissue engineering for anterior cruciate ligament reconstruction: A review of current strategies. Arthroscopy 22:441-451, 2006. Rosenberg L, Choi HU, Neame PJ, et al: Proteoglycans of soft connective tissue. In Leadbetter WB, Buckwalter JA, Gordon SL (eds): Sports Induced Inflammation—Basic Science and Clinical Concepts. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1990. Woo SL, Abramowitch SD, Kilger R, Liang R: Biomechanics of knee ligaments: Injury, healing, and repair. J Biomech 39:1-20, 2006. Woo SL, Jia F, Zou L, Gabriel MT: Functional tissue engineering for ligament healing: Potential of antisense gene therapy. Ann Biomed Eng 32:342-351, 2004. Woo SLY, Buckwalter JA: Ligament and tendon autografts and allografts. In ­Friedlander GE, Goldberg VM (eds): Biologic Restoration of Bone and Articular Surfaces. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1991.

R eferences Please see www.expertconsult.com

40

DeLee & Drez’s Orthopaedic Sports Medicine

S ecti o n

A

Physiology of Injury to Musculoskeletal Structures 3. Articular Cartilage Injury Mark R. Brinker, Daniel P. O’Connor, Louis C. Almekinders, Thomas M. Best, Joseph A. Buckwalter, William E. Garrett, Jr., Donald T. Kirkendall, Van C. Mow, and Savio L.-Y. Woo Synovial joints allow the rapid controlled movements ­necessary for sports. Normal function of these complex diarthrodial structures depends on the structural integrity and macromolecular composition of articular cartilage. Sports-related traumatic disruptions of cartilage structure or alterations in the macromolecular composition or organization change the biomechanical properties of the tissue and compromise joint function. These changes can lead to progressive pain and disability. Sports injuries to ­articular cartilage present more difficult diagnostic and treatment problems than injuries to ligament, tendon, or bone because less is understood about these injuries and because of the unique structure and function of articular cartilage. The specialized composition and organization of articular cartilage1,2 make the diagnosis of many injuries ­difficult, but these characteristics also provide the unique biomechanical properties that permit normal synovial joint function. When compressed, articular cartilage is soft and yields its interstitial water easily,3-6 yet it is stiff in tension along planes parallel to the articular surface. Intact cartilage ­ provides a smooth, lubricated gliding surface with a coefficient of friction lower than most fabricated bearing materials.7-10 In the joint, cartilage distributes the loads of articulation, thereby minimizing peak stresses acting on the subchondral bone. The tensile strength of the tissue provides its structural integrity under such loads. These ­biomechanical properties make the tissue remarkably durable and wear resistant, enabling it to last many decades, even under high and repetitive stresses. Alterations in the mechanical properties of cartilage due to injury, disease, or increasing age have not been well defined, but the available information shows that these properties change with age and loss of structural integrity. Cartilage from skeletally immature joints (open growth plates) is much stiffer than cartilage from skeletally mature joints (closed growth plates).11 Older cartilage and fibrillated cartilage have much lower tensile stiffness and strength.12,13 Declines in stiffness and strength may increase the probability of injury to cartilage. Participation in sports often subjects the articular cartilage to intense repetitive compressive forces that can cause injury and deterioration of the tissue. Falls or other high-energy impact can damage the articular cartilage without disrupting the articular surface. These abnormally large forces generate high shear stresses at the cartilage­subchondral bone junction, causing matrix lesions14,15 that

may lead to clinically significant cartilage deterioration and joint dysfunction.16-18 Furthermore, repetitive trauma that leaves the articular surface intact can cause other injuries, including subtle damage to the matrix macromolecular framework and cartilage cells (chondrocytes).15,18-20 These injuries disrupt the well-organized macromolecular fabric of the matrix, alter cell function, and disturb normal cell-matrix interactions, thereby adversely altering the cell activities required to maintain cartilage mechanical properties.21-28 Because cartilage lacks nerves and blood vessels, damage limited to cartilage is not likely to be detected at the time of initial injury. Cartilage with this type of damage may become fatigued and fail more easily when subjected to subsequent acute or repetitive trauma. Less frequent but more severe sports injuries can acutely disrupt the articular surface by fracturing both cartilage and the underlying bone.29,30 Therefore, both acute traumatic injuries and repetitive excessive loading can lead to cartilage deterioration and impairment of synovial joint function. Sports injuries to articular cartilage remain poorly understood. Because injuries limited to cartilage do not cause pain or inflammation, patients and physicians rarely suspect cartilage damage following excessive acute or repetitive joint loading. Even when the physician suspects that cartilage injury exists, making a precise diagnosis of many types of cartilage injuries is difficult. The natural history of many types of cartilage injury remains unknown. Some of the deterioration of articular surfaces currently attributed to sports-related ligamentous or meniscal injuries may actually be due to undetected cartilage damage. This chapter first reviews the important aspects of articular cartilage composition, organization, and biomechanical properties that make possible normal synovial joint function. The next sections summarize the response of cartilage to different types of injury and the results of cartilage repair, cartilage shaving, and abrasion of subchondral bone. The last section describes the use of cartilage grafts to replace lost or damaged articular surfaces.

COMPOSITION OF ARTICULAR CARTILAGE Like the dense fibrous tissues and meniscus, articular cartilage consists of cells, matrix water, and a matrix macromolecular framework.1,2,31-37 Unlike the most dense

Basic Science and Injury of Muscle, Tendon, and Ligament

41

Articular surface STZ (10%-20%)

Middle zone (40%-60%)

Deep zone (30%) Calcified zone

Subchondral bone Tide mark Chondrocyte Figure 1A3-1  Normal articular cartilage structure. Histologic (A) and schematic (B) views of a section of normal articular cartilage. The tissue consists of four zones: the superficial tangential zone (STZ), the middle zone, the deep zone, and the calcified zone. Notice the differences in cell alignment among zones. The cells of the superficial zone have an ellipsoidal shape and lie with their long axes parallel to the articular surface. The cells of the other zones have a more spheroidal shape. In the deep zone, they tend to align themselves in columns perpendicular to the joint surface.   (From Nordin M, Frankel VH: Basic Biomechanics of the Musculoskeletal System, 2nd ed. Philadelphia, Lea & Febiger, 1989, pp 31-57. Used with permission.)

A

B

fibrous tissues, cartilage lacks nerves, blood vessels, and a lymphatic system. The composition of articular cartilage is responsible for its unusual physiologic requirements, cell behavior, and responses to injury.

Chondrocytes The only type of cell in normal cartilage is the highly specialized chondrocyte. Chondrocytes contribute relatively little to the total volume of mature human articular cartilage, usually 5% or less. Like other mesenchymal cells,33,38 chondrocytes surround themselves with their extracellular matrix and rarely form cell-to-cell contacts. In normal cartilage, they are isolated in the extracellular matrix. Because the tissue lacks blood vessels, the cells depend on diffusion through the matrix for their nutrition and rely primarily on anaerobic metabolism. Figure 1A3-1A and B shows a histologic section of normal adult articular cartilage with the chondrocytes embedded in the matrix and a schematic representation of chondrocyte morphology. Three distinct zones of chondrocytes are seen. The superficial tangential zone contains ellipsoidal cells with their long axes aligned parallel to the surface. The middle zone contains spherical cells randomly distributed throughout the region. The deep zone contains ­similar spherical cells forming columns aligned perpendicular to the tidemark and the calcified zone. Articular cartilage chondrocytes contain the organelles necessary for matrix synthesis, including endoplasmic reticulum and Golgi membranes. Also, they frequently contain intracytoplasmic filaments and glycogen, and some chondrocytes have a cilium that extends from the cell into the extracellular collagen-proteoglycan matrix. These structures may sense mechanical changes in the matrix. After completion of skeletal growth, chondrocytes rarely divide, but throughout life they synthesize and maintain the

extracellular matrix that gives cartilage its essential material properties. Synthesis and turnover of proteoglycans are relatively fast, whereas collagen synthesis and turnover are very slow.37,39,40 (For more details concerning turnover of cartilage matrix macromolecules, see the excellent review by Lohmander.41)

Extracellular Matrix Tissue Fluid Water contributes up to 80% of the wet weight of articular cartilage. The interaction of water with the matrix macromolecules significantly influences the material properties of the tissue.5,26,37,42-48 This tissue fluid contains gases, small proteins, metabolites, and a high concentration of cations to balance the negatively charged proteoglycans.46,47,49,50 The volume, the concentration, and the behavior of the tissue water depend on its interaction with the structural macromolecules. In particular, large aggregating proteoglycans organize the tissue water and impede its flow through the matrix. The large proteoglycans also help maintain the fluid within the matrix and the fluid electrolyte concentrations. These matrix macromolecules have large numbers of negative charges, which attract positively charged ions and repel negatively charged ions. This increases the concentration of positive ions (e.g., sodium) and decreases the concentration of negative ions (e.g., chloride). This increase in the total inorganic ion concentration increases the tissue osmolarity, attracting water into the matrix; that is, it creates a Donnan effect. The collagen network resists this Donnan osmotic pressure caused by the inorganic ions that are associated with proteoglycans.46,47,51 This interaction between proteoglycans and tissue fluid significantly influences the compressive stiffness and resilience of articular cartilage.5,47,52-54

42

DeLee & Drez’s Orthopaedic Sports Medicine

Articular surface

Zones

STZ

Superficial tangential (10%-20%)

Middle (40%-60%)

Deep (30%)

Tide mark

Middle zone

Subchondral bone Calcified cartilage

Cancellous bone

A B Deep zone Figure 1A3-2  Schematic representation (A) and scanning electron micrographs (B) of the interterritorial matrix collagen fibril orientation and organization in normal articular cartilage. In the superficial tangential zone (STZ), the fibrils lie roughly parallel to the articular surface. In the middle zone, they assume a more random alignment, and in the deep zone, they lie roughly perpendicular to the articular surface.   (From Nordin M, Frankel VH: Basic Biomechanics of the Musculoskeletal System, 2nd ed. Philadelphia, Lea & Febiger, 1989, pp 31-57. Used with permission.)

Structural Macromolecules The structural macromolecules that provide 20% to 40% of the wet weight of cartilage include collagens, proteoglycans, and noncollagenous proteins.34,37,41 Chondrocytes synthesize all three types of molecules from amino acids and sugars, but differences in the types and the organization of amino acids and sugars give each type of molecule a different form and function.33,34,37,41 Abnormalities in these molecules or in their organization can adversely affect the durability and the mechanical properties of the cartilage and may lead to deterioration of the articular surface.5,8,9,12,37,47,55-58 Collagens contribute about 60% of the dry weight of cartilage, proteoglycans contribute 25% to 35%, and the noncollagenous proteins and glycoproteins contribute 15% to 20%. Collagens are distributed relatively uniformly throughout the depth of the cartilage except in the collagenrich region near the surface.45 The collagen fibrillar meshwork and cross-linking give cartilage its form and tensile strength.11,13,59,60 Figure 1A3-2A shows a ­schematic representation of the fibrillar collagen ultrastructure throughout the depth of the tissue, along with three scanning electron micrographs that show the appearance of the fibrillar collagen network in the three zones of uncalcified cartilage (see Fig. 1A3-2B).61 Proteoglycans and noncollagenous proteins bind to the collagenous meshwork or become mechanically entrapped within it, and water fills this molecular framework. Proteoglycans give cartilage its stiffness in compression and its resilience. Figure 1A3-3A shows a proteoglycan monomer attached to a hyaluronate chain and a linking protein, a proteoglycan aggregate (��������������� see Fig. 1A3-3�B), and an electron micrograph of a proteoglycan aggregate (��������������� see Fig. 1A3-3�C).

Some noncollagenous proteins organize and stabilize the matrix macromolecular framework, whereas others bind chondrocytes to the macromolecules of the matrix. More detailed reviews of cartilage molecular organization are available.1,2,34,37,41,62

COLLAGENS Collagens, a family of 13 or more protein molecules produced by more than 20 distinct genes, form a critical part of every tissue.40,63 All collagen molecules have a region consisting of three amino acid chains wound into a triple helix that provides some of the distinctive structural and mechanical properties of collagens. By virtue of their tensile stiffness and strength, collagens contribute to the structure of the articular cartilage matrix. Articular cartilage contains collagen types II, VI, IX, X, and XI. Collagen types II, IX, and XI form the cross-banded fibrils seen on electron microscopy. These fibrils organize into a tight meshwork that extends throughout the tissue and provides the tensile stiffness and strength of articular cartilage.11-13,60,64 This meshwork also contributes to the cohesiveness of the tissue by mechanically entrapping the large proteoglycans. Type II collagen accounts for 90% to 95% of the cartilage collagen and forms the primary component of the cross-banded fibrils. Type IX collagen contains both collagenous and noncollagenous regions and has one or possibly two chondroitin sulfate chains.13,65 Type IX collagen molecules bind covalently to the superficial layers of the cross-banded fibrils and project into the matrix. Type XI collagen molecules also bind covalently to type II collagen molecules and probably form part of the interior

ion

(G 3)

re g

om ain

–r i ch

Te rm ina

ld

ulf ate ro i tin s

C-

nd Ch o

Ke r re rat gio an n( su G lfa 2) te– ric hr eg ion

nd co Se

Link protein

HA b

ind

ing

glo

re g

Hyaluronic acid (HA)

bu la

ion

(G 1)

Basic Science and Injury of Muscle, Tendon, and Ligament

43

Figure 1A3-3  Proteoglycan structure. A, Details of proteoglycan monomer structure, showing chondroitin sulfate and keratan sulfate chains and the interaction of the monomer with hyaluronate chain and link protein. B, Molecular conformation of a typical proteoglycan aggregate, showing size of the molecule. C, An electron micrograph of a proteoglycan aggregate.

Protein core

Keratan sulfate chain Chondroitin sulfate chain

A

Link protein

200-400 nm

Hyaluronic acid

B

1200 nm

C structure of the cross-banded fibrils. The functions of types IX and XI collagens remain uncertain, but presumably they act together to help form and stabilize the primarily type II collagen fibrils. The position of type IX collagen molecules on the surface of the fibrils suggests that they may influence the diameter and the stability of fibrils and may interact with other matrix macromolecules. The position of type XI collagen molecules within the fibrils suggests that they may contribute to the formation and may influence the diameter of fibrils.

The functions of types VI and X collagen remain unknown. Type VI collagen is often described as an adhesion protein and in some regions reaches its highest concentration in the immediate vicinity of the chondrocytes. Type X collagen appears near the cells of the calcified cartilage zone of articular cartilage and the hypertrophic zone of the growth plate, where the longitudinal cartilage septa begin to mineralize. This pattern of distribution suggests that it contributes to mineralization of cartilage.

44

DeLee & Drez’s Orthopaedic Sports Medicine

PROTEOGLYCANS Proteoglycans consist of a protein core and one or more glycosaminoglycan chains (long, unbranched polysaccharide chains consisting of repeating disaccharides that ­contain an amino sugar). Each disaccharide unit has at least one negatively charged carboxylate or sulfate group so that the glycosaminoglycans form long strings of negative charges that repel other negatively charged molecules and attract cations. The sugars included in the polysaccharide chains of these molecules vary, but all the molecules consist of repeating disaccharide units containing a ­derivative of either glucosamine or galactosamine.41,66 Glycosaminoglycans found in cartilage include hyaluronic acid, chondroitin sulfate, keratan sulfate, and dermatan sulfate. Articular cartilage contains at least three types of proteoglycans: a large aggregating proteoglycan called aggrecan, which contains large numbers of chondroitin sulfate and keratan sulfate chains, and two small dermatan ­sulfate-­containing, nonaggregating proteoglycans called biglycan and decorin. Other matrix molecules that have glycosaminoglycan chains and therefore could be classified as proteoglycans include type IX collagen and fibromodulin (discussed later). The tissue probably also contains other small proteoglycans and some large nonaggregating ­proteoglycans. The aggrecan molecules fill most of the interfibrillar space of the cartilage matrix1,2,34,37,41,67 and contribute about 90% of the total cartilage matrix proteoglycan. The large nonaggregating proteoglycans contribute less than 10%, and small nonaggregating proteoglycans contribute about 3%. The large nonaggregating proteoglycans resemble the large aggregating proteoglycans in structure and composition41 and may represent degraded aggregating proteoglycans. The large aggregating proteoglycans consist of protein core filaments with many covalently bound chondroitin sulfate and keratan sulfate chains (see Fig. 1A3-3).37,41,62,66 Chondroitin sulfate and keratan sulfate form about 95% of the molecule, and protein forms about 5%. The protein cores consist of five domains: three globular domains and two extended domains (see Fig. 1A3-3). A short extended region of the protein core separates the G1 region (first globular domain) from the G2 region (second globular domain). A longer extended region of the protein core contains covalently bound keratan sulfate and chondroitin sulfate chains and separates G2 and G3 (the third globular domain). Keratan sulfate chains cluster together near the G2 domain (the keratan sulfate-rich region), and the chondroitin sulfate chains cluster together between the keratan sulfate-rich regions and the G3 domain (the chondroitin sulfate-rich region). The G1 domain binds noncovalently to hyaluronic acid filaments and small proteins called link proteins. The functions of the G2 and G3 domains remain unknown. Because each keratan sulfate and chondroitin sulfate chain contains many negative charges, adjacent chains repel each other and tend to maintain aggrecan molecules in an expanded form. This conformation promotes the trapping of proteoglycans within the fine collagen meshwork.37,67 In the articular cartilage matrix, most aggregating proteoglycan monomers noncovalently associate with hyaluronic acid filaments and link proteins to form proteoglycan aggregates (see Fig. 1A3-3B). These large ­molecules

have a central hyaluronic acid backbone that can vary in length from several hundred nanometers to more than 10,000 nm.37,41,55,62 Large aggregates may have more than 300 associated monomers. Link proteins and other small noncollagenous proteins stabilize the association between monomers and hyaluronic acid. These proteins maintain the stability of the matrix and provide added mechanical strength and appear to play a role in directing the assembly of aggregates.58,62 Aggregate formation anchors proteoglycans within the matrix, preventing their displacement during deformation of the tissue, and also helps organize and stabilize the relationship between proteoglycans and collagen fibrils. Proteoglycans at physiologic concentrations form elastic networks capable of storing energy.56,57 Link proteins greatly increases the stiffness and the strength of these proteoglycan networks.58 The interaction between the collagen and the proteoglycan networks provides the strength and the cohesiveness of the articular cartilage extracellular matrix.5,28 The interaction between large proteoglycans and the tissue fluid contributes significantly to the compressive stiffness and the resilience of cartilage. These proteoglycans have a structure that fills a large volume with negatively charged glycosaminoglycan chains (see Fig. 1A3-3) that interact with water and cations. By repelling each other, these charged chains hold the monomers stiffly extended, thereby inflating the collagen fibril meshwork with water. Compression of the intact matrix drives the glycosaminoglycan chains closer together, increasing resistance to further compression and forcing water out of the molecular domain. Release of compression allows the molecules to re-expand and imbibe the lost fluid. Comparison of the maximal volume that can be occupied by proteoglycans in solution with their concentration in articular cartilage matrix shows that if the cartilage matrix proteoglycans expanded fully, they would fill a volume many times larger than the tissue that contains them.62 In the matrix, their domains must overlap or be collapsed; the repulsive forces from charges on these ­molecules and the osmotic pressure generated by counter-ions associated with these charges exert a constant pressure to expand. Only the collagen fibril meshwork restrains the expansion of the proteoglycans.46,47,49,51-54,68 Disruption of this collagen meshwork releases the matrix proteoglycans to extend their protein cores and glycosaminoglycan chains, thereby increasing the concentration of water and decreasing the proteoglycan concentration within the extracellular matrix. These molecular changes increase the permeability and decrease the stiffness of the matrix. Thus, either loss of proteoglycans or collagen network disruption due to acute or repetitive trauma will have significant detrimental effects on the mechanical properties of articular ­cartilage.3,5,12,13,28,47 Small nonaggregating proteoglycans have shorter protein cores than aggrecan molecules, and two of them—biglycan and decorin—contain a different type of glycosaminoglycan (dermatan sulfate) as well as other glycosaminoglycans.66 In contrast to the many glycosaminoglycan chains in aggrecan molecules, biglycan has only two glycosaminoglycan chains, and decorin has only one. In adult articular cartilage, at least some of these smaller proteoglycans form close associations with collagen fibrils.

Basic Science and Injury of Muscle, Tendon, and Ligament

45

Unlike the large aggregating molecules, they do not fill a large volume of the tissue or contribute directly to the mechanical behavior of the tissue. Instead, they bind to other macromolecules and probably influence cell function, such as inhibiting cartilage repair.

NONCOLLAGENOUS PROTEINS AND GLYCOPROTEINS The noncollagenous proteins and glycoproteins are not as well understood as collagens and proteoglycans. They consist primarily of protein and have a few attached monosaccharides and oligosaccharides. Some of these molecules appear to help organize and maintain the macromolecular structure of the matrix, whereas others may help stabilize the relationship between chondrocytes and other matrix macromolecules. Link proteins help organize and stabilize the matrix through their effects on proteoglycan aggregation.37,41,58 Other noncollagenous proteins found in articular cartilage may help mediate the adhesion of chondrocytes to the matrix, possibly stabilizing the relationship between the chondrocytes and the matrix.41 Fibromodulin, a glycoprotein that contains keratan sulfate, may be considered a form of proteoglycan as well as a glycoprotein. It appears to be associated with cartilage collagen fibrils and may influence collagen turnover.

Cell-Matrix Interactions Maintenance of cartilage depends on continual complex interactions between chondrocytes and the matrix they synthesize. Normal degradation of matrix macromolecules, especially proteoglycans, requires that the chondrocytes continually synthesize new molecules.37,41,66 The cells sense the content of the matrix and respond appropriately to maintain the tissue’s biomechanical properties. For example, experimental depletion of matrix proteoglycan with papain stimulates proteoglycan synthesis.41 If the cells did not replace the lost proteoglycans, the tissue would deteriorate. Mechanical loading also affects cartilage homeostasis.69-73 These interactions between the cells and their matrix have considerable importance for sports injuries to articular cartilage. Either a mechanical injury that interferes with the ability of chondrocytes to replace matrix macromolecules or a lack of appropriate mechanical stimulation will lead to deterioration of the tissue. At present, the physiologic mechanisms through which chondrocyte synthesis of matrix macromolecules is stimulated or suppressed remain unknown. Chondrocytes respond to changes in patterns of matrix deformation due to persistent changes in joint use, although the complete mechanisms of chondrocyte control and modulation are unknown. Both mechanical and physicochemical events during matrix deformation likely play significant roles in stimulating chondrocytes. When cartilage is loaded, it deforms. Figure 1A3-4 shows a chondrocyte embedded in the charged extracellular matrix in an undeformed state and in a deformed state. Deformation during compression alters the charge density around the cells and induces a streaming potential throughout the tissue.

Figure 1A3-4  Confocal laser microscopic view of a chondrocyte from the middle zone of articular cartilage embedded in the extracellular matrix in the unloaded state   (t = 0) and the compressed state at equilibrium (Equil.).

These physicochemical effects vary according to proteoglycan concentration relative to depth from the surface in the different zones of the charged collagen-proteoglycan extracellular matrix.34,37,41,47,66 Recent studies have suggested that these physicochemical events are important in modulating chondrocyte proteoglycan biosynthesis.74,75 The increase in charge density within the extracellular matrix increases the interstitial Donnan osmotic pressure, increases the osmotic pressure gradients, and produces streaming polarization and electro-osmosis effects around the chondrocytes.46,53 These electrical events may be important in stimulating chondrocyte biosynthetic activities. The stress-strain environment and the strain-energy density around the cell may also play important roles in stimulating chondrocytes.76,77 In addition to these mechanical, electrical, and physicochemical events, biochemical agents such as growth factors,78 cytokines, and enzymes are also potent stimulators of chondrocytes. Mechanical, electrical, physicochemical, and biochemical effects may each play a role in modulating chondrocyte activities, and these effects may be synergistic. Studies addressing these important questions offer great challenges for the future.

ORGANIZATION OF ARTICULAR CARTILAGE To form articular cartilage, chondrocytes organize collagens, proteoglycans, and noncollagenous proteins into a unique, highly ordered structure (see Figs. 1A3-1 and 1A3-2). The composition, organization, and mechanical properties of the matrix as well as cell morphology and function vary according to the depth from the articular surface.1,34,37,41,47,66 Matrix composition, organization, and function also vary with distance from the cell.1,2,79

46

DeLee & Drez’s Orthopaedic Sports Medicine

Zones of Articular Cartilage Morphologic changes in articular cartilage cells and matrix from the articular surface to the subchondral bone make it possible to identify four zones or layers of articular ­cartilage: the superficial tangential zone, the middle or transitional zone, the deep or radial zone, and the zone of calcified cartilage (see Figs. 1A3-1 and 1A3-2). Although each zone has distinct morphologic features, the boundaries between zones cannot be sharply defined. Recent ­biologic and mechanical studies, however, have shown that this morphologic zonal organization has functional significance. Cells in each zone differ in shape, size, and orientation relative to the articular surface and appear to differ in synthetic activity. They may also respond differently to mechanical loading, suggesting that development and maintenance of normal articular cartilage depend in part on differentiation of phenotypically distinct populations of chondrocytes. The heterogeneity of chondrocytes presumably is responsible for the differences in matrix composition and organization that result in different mechanical properties in each zone.

Superficial Zone The thinnest zone, the superficial tangential zone, has two layers. A sheet of fine fibrils with little polysaccharide and no cells covers the joint surface (see Fig. 1A3-1). This layer presumably corresponds to the clear film that can be stripped from the articular surface in some regions. On phase-contrast microscopy, it appears as a narrow bright line, the “lamina splendens.” In the next layer of the superficial zone, flattened ellipsoid chondrocytes are arranged so that their major axes are parallel to the articular surface (see Fig. 1A3-1). They synthesize a matrix that has a high collagen concentration and a low proteoglycan concentration relative to the other cartilage zones. Water content is the highest in this zone, averaging 80%.37,45,47

Transitional Zone The transitional, or middle, zone has several times the volume of the superficial zone (see Fig. 1A3-1). The cells of this zone have a higher concentration of synthetic organelles, endoplasmic reticulum, and Golgi membranes than do the cells of the superficial zone. They assume a spheroidal shape and synthesize a matrix with collagen fibrils of a larger diameter and a higher concentration of proteoglycans than is found in the superficial zone. In this zone, the proteoglycan concentration is higher than in the superficial zone, but the water and the collagen concentrations are lower.45,47,48

Deep Zone The chondrocytes in the deep zone resemble those of the middle zone, but they tend to align in columns perpendicular to the joint surface (see Fig. 1A3-1). This zone contains the collagen fibrils with the largest diameter, the highest concentration of proteoglycans, and the lowest concentration of water. The collagen fibers of this zone pass through the tidemark (a thin basophilic line seen on light microscopic sections of decalcified articular cartilage that marks the boundary between calcified and uncalcified

cartilage)16,80 into the calcified zone, anchoring adult articular cartilage to the subchondral bone.80

Zone of Calcified Cartilage A zone of calcified cartilage lies between the deep zone of uncalcified cartilage and the subchondral bone (see Fig. 1A3-2). The cells of the calcified cartilage zone have a smaller volume per cell than the cells of the deep zone and contain only small amounts of endoplasmic reticulum and Golgi membranes. During aging, the tidemark advances, causing thinning of the uncalcified cartilage.16,81 This remodeling process may be due to repetitive microtrauma in the deep zone of cartilage.14,15,18 Results of stress-strain analysis suggest that this cartilage thinning is detrimental to the tissue.18,82

Matrix Regions Variations in the matrix within zones have been described by dividing the matrix into regions or compartments called the pericellular region, the territorial region, and the interterritorial region (Figs. 1A3-5 and 1A3-6).1,34,79,83 The pericellular and the territorial regions appear to serve the needs of chondrocytes; they bind the cell membranes to the matrix macromolecules and protect the cells from damage during loading and deformation.77 They may also help transmit mechanical signals to the chondrocytes. The primary function of the interterritorial matrix is to provide the mechanical properties of the tissue.5,28,47

Pericellular Matrix Chondrocyte cell membranes appear to attach to the thin rim of the pericellular matrix that covers the cell surface (see Figs. 1A3-5 and 1A3-6). This matrix region probably contains noncollagenous proteins and possibly nonfibrillar collagens and is rich in proteoglycans.34,79 In addition, this matrix region contains little or no fibrillar collagen.

Territorial Matrix An envelope of territorial matrix surrounds the pericellular matrix of individual chondrocytes and, in some locations, pairs or clusters of chondrocytes and their pericellular matrices (see Figs. 1A3-5 and 1A3-6). In the deep zone, a territorial matrix surrounds each chondrocyte column. The thin collagen fibrils of the territorial matrix nearest the cell appear to adhere to the pericellular matrix. At a distance from the cell, they decussate and intersect at various angles, forming a fibrillar basket around the cells.83 This collagenous basket may provide mechanical protection for the chondrocytes during loading and deformation of the tissue. The results of a recent stress-strain analysis support this concept; the matrix regions surrounding the cells functioned as a protective buffer against the development of high stresses and strains in the chondrocyte when the tissue was loaded.77 The boundary between the territorial and the interterritorial matrices is marked by an abrupt increase in collagen fibril diameter and a transition from the basket-­like orientation of the collagen fibrils to a more parallel arrangement. Many collagen fibrils connect these two regions, making it difficult to identify the precise division between these two regions.34,84

Basic Science and Injury of Muscle, Tendon, and Ligament

A

47

B

Figure 1A3-5  A and B, Electron micrographs showing chondrocytes and the regions of articular cartilage matrix. In both electron micrographs, short cell processes protrude from the chondrocytes through the pericellular matrix (arrowheads) to the border between the pericellular matrix and the territorial matrix (*). The interterritorial matrix (**) contains larger collagen fibrils and surrounds the territorial matrices, the pericellular matrices, and the cells.   (Reprinted with permission from Buckwalter JA, Hunziker EB, Rosenberg RC,   et al: Articular cartilage: Composition and structure. In Woo SL, Buckwalter JA [eds]: Injury and Repair of the Musculoskeletal Soft Tissues. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1988, pp 405-426.)

B

A

C

Figure 1A3-6  Electron micrographs showing the matrix compartments of articular cartilage. A, The pericellular matrix (arrowheads) consists of a narrow, dense coat of proteoglycans and possibly glycoproteins as well as other nonfibrillar molecules. The territorial matrix (*) surrounds the pericellular matrix. Notice the intimate contact between the pericellular matrix and the cell membrane. B, The territorial matrix consists of a dense network of fine collagen that may extend around one cell or a group of   cells (a territorium or a chondron). Notice the absence of a sharp border between the pericellular and the territorial matrices.   C, The interterritorial matrix consists of parallel large-diameter collagen fibrils and fibers with matrix granules (arrows) interspersed between fibrils. These granules consist of proteoglycans precipitated with ruthenium hexamine trichloride.���   (Reprinted with permission from Buckwalter JA, Hunziker EB, Rosenberg RC, et al: Articular cartilage: Composition and structure. In Woo SL, Buckwalter JA [eds]: Injury and Repair of the Musculoskeletal Soft Tissues. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1988, pp 405-426.)

48

DeLee & Drez’s Orthopaedic Sports Medicine

Interterritorial Matrix The interterritorial matrix (see Figs. 1A3-5 and 1A3-6) makes up most of the volume of mature articular cartilage. It contains the largest-diameter collagen fibrils. Unlike the collagen fibrils of the territorial matrix, these fibrils are not organized to surround the chondrocytes, and they change their orientation relative to the joint surface by 90 degrees from the ­superficial zone to the deep zone (see Fig. 1A3-2).20,85-87 In the superficial zone, the fibril diameters are relatively small, and the fibrils generally lie parallel to the articular surface. Creation of pinhole defects in some articular surfaces produces split lines that show a tendency to extend parallel to the plane of joint motion,88,89 suggesting that collagen fibrils in the interterritorial matrix of the superficial zone may also be oriented relative to the motion of the joint. In the middle zone, interterritorial matrix ­ collagen fibrils assume more oblique angles relative to the articular surface, and in the deep zone, they generally lie perpendicular to the joint surface. Because the interterritorial matrix forms most of the volume of cartilage, and because collagen provides the tensile stiffness and strength of articular cartilage, these biomechanical properties should vary with changes in collagen fibril orientation and organization in the interterritorial matrix (see Fig. 1A3-2). Cartilage tensile stiffness and strength vary among cartilage zones.11,13,60

MECHANICAL PROPERTIES OF CARTILAGE The behavior of cartilage when subjected to compression, tension, or shear depends on the concentration, properties, and organization of the matrix macromolecules, the water content, and the physical and electrical interactions between the water and the macromolecular framework.10,28,64 Because cartilage has a solid phase (the macromolecular framework of collagens, proteoglycans, and noncollagenous proteins) and a fluid phase (the tissue water), it behaves as a biphasic (two-phase) viscoelastic material, that is, its response to loading combines viscosity, a characteristic of fluids, with elasticity, a characteristic of solids. When a viscoelastic material is subjected to a constant load or a constant deformation, its response varies with time. A viscoelastic material subjected to a constant load responds with rapid initial deformation followed by further slow, progressive deformations until it reaches an equilibrium state, a behavior called creep. A viscoelastic material subjected to constant deformation responds with high initial stress followed by a slowly progressive decrease in the stress required to maintain the deformation, a behavior called stress relaxation. Because articular cartilage is a porous-permeable hydrated soft tissue, creep and stress relaxation in compression are predominantly caused by fluid flow through the matrix.4,5,28 In shear, when no interstitial fluid flow occurs, creep and stress relaxation occur because the macromolecular framework is altered.5,90-93 Articular cartilage can undergo deformation and then revert to its original form by reversing fluid flow. This feature is important in joint lubrication and tissue fluid circulation, particularly because of the load-bearing function of cartilage and because of its lack of a vascular supply and lymphatic system.9,10

Changes in the composition or the organization of the matrix macromolecular framework can cause deterioration of these essential mechanical properties. Collagen fibrils provide tensile strength but little resistance to compression. Interaction of the proteoglycans with water provides resistance to compression, swelling pressure, and resilience but little tensile strength. Disruption of the collagen fibril meshwork allows the proteoglycans to expand, increases the water concentration, decreases the proteoglycan concentration,28,49,51,53,54 and results in a decrease in cartilage stiffness and an increase in matrix permeability.3,5,28 These changes make the tissue less capable of supporting loads and more vulnerable to further injury, which may cause progressive mechanical failure of the matrix and clinical osteoarthritis.22,28,94

CARTILAGE INJURY AND REPAIR The response of cartilage to trauma and the potential for repair of cartilage depend to a large extent on the type of injury sustained and whether the injury involves the subchondral bone. Sports-related trauma can disrupt the cartilage matrix, causing visible splits in the articular surface, or damage the macromolecular framework and alter cell function without disrupting the tissue surface.1,14,15,20,28,95-100 Because cartilage lacks blood vessels, damage to the cartilage alone does not cause inflammation. By contrast, if an injury disrupts both the cartilage and the subchondral bone, the blood vessels in the bone participate in the inflammation that initiates the fracture healing.22 The clot and the repair tissue from bone can then fill the articular cartilage defect and follow the sequence of inflammation, repair, and remodeling, as in the repair of other tissues such as ligament.22 Unlike ligament repair, the tissue that repairs cartilage defects differentiates initially toward articular cartilage rather than toward dense fibrous tissue.1,22,94,101 Differences in the potential for cartilage repair separate acute injuries of articular cartilage into three general types1,21,22,94: (1) loss of matrix macromolecules or disruption of the macromolecular framework without visible tissue disruption; (2) mechanical disruption of articular cartilage alone; and (3) mechanical disruption of articular cartilage and subchondral bone. These categories overlap. Progressive loss of matrix macromolecules or disruption of the organization of the matrix macromolecular framework eventually results in mechanical disruption of the tissue; mechanical disruption of the cartilage may release tissue factors that stimulate matrix degradation and loss of matrix macromolecules. Each of these types of cartilage damage presents a different problem for repair Table 1A3-1 summarizes the current understanding of the clinical presentation, the tissue response, and the potential for healing of the three types of acute sports-related injuries to cartilage. The exact mechanisms of injury and the natural history of these injuries remain poorly understood. Injuries limited to the articular cartilage do not cause pain or inflammation, but these injuries commonly occur with injuries to other tissues that have nerves and blood vessels, such as synovium, ligament, joint capsule, meniscus, and bone. Therefore, people with injuries of the articular cartilage may have symptoms and signs from the damage to the innervated vascularized tissues.

Basic Science and Injury of Muscle, Tendon, and Ligament

49

  Table 1A3-1 Acute Sports Injuries to Articular Cartilage Injury Type

Clinical Presentation

Tissue Response

Potential for Healing

Damage to matrix or cells without visible disruption of articular surface

Synthesis of new matrix macromolecules Cell proliferation?

Cartilage disruption (cartilage fractures or ruptures)

No known symptoms Direct inspection of the articular surface and current clinical imaging methods cannot detect this type of injury May cause mechanical symptoms, synovitis, and joint effusions

If basic matrix structure remains intact and a sufficient number of viable cells remain, the cells can restore the normal tissue composition If the matrix or the cell population sustains significant damage, or if the tissue sustains further damage, the lesion may progress Depending on the location and the size of the lesion and the structural integrity, stability, and alignment of the joint, the lesion may or may not progress

Cartilage and bone disruption (osteochondral fractures)

May cause mechanical symptoms, synovitis, and joint effusions

No fibrin clot formation or inflammation Synthesis of new matrix macromolecules and cell proliferation, but new tissue does not fill cartilage defect Formation of a fibrin clot, inflammation, invasion of new cells, and production of new tissue

Matrix Damage without Visible Tissue Disruption Acute or repetitive direct blunt trauma and acute or repetitive high-energy joint loading can cause cartilage damage without visible tissue disruption (see Table 1A3-1). These injuries are often associated with other joint injuries such as ligament and meniscal tears, but lack of an inflammatory response, lack of nerves, and lack of visible disruption of the articular surface prevent detection of these injuries even on direct examination of the articular surface or with magnetic resonance imaging. The intensity and the type of cartilage loading that can cause matrix damage without gross tissue disruption have not been well defined.14,28,80 Physiologic levels of joint loading do not appear to cause cartilage injury, but impact loading above that associated with normal activities and less than that producing cartilage disruption can cause tissue alterations. Other causes of this type of cartilage injury include traumatic or surgical disruption of the synovial membrane, prolonged joint immobilization, some medications, joint irrigation, and synovial inflammation.1,14,15,21,22,94 Loss of proteoglycans or disruption of their organization appears to occur before other signs of tissue injury.102-104 The loss of proteoglycans and the alteration of their molecular structure may be due to either increased degradation or altered synthesis of the molecules.1,21,22,104 Loss of matrix proteoglycans decreases cartilage stiffness and increases its hydraulic permeability.3,5,28 These alterations in mechanical properties may cause greater loading of the remaining macromolecular framework, including the collagen fibrils, thus increasing the vulnerability of the tissue to damage from further impact loading. These types of injuries may cause matrix abnormalities other than loss of proteoglycans (e.g., rupture or distortion of the collagen fibril meshwork or disruption of the collagen fibril-proteoglycan relationships), and they may alter chondrocyte function or even damage the chondrocytes. For example, impact loading of dog articular cartilage causes cartilage swelling, collagen fibril swelling, and disturbances in the relationships between collagen fibrils and proteoglycans.15 Mildly fibrillated cartilage from

Depending on the location and the size of the lesion and the structural integrity, stability, and alignment of the joint, the lesion may or may not progress

human femoral condyles shows a significant decrease in tensile stiffness and an increase in swelling; cartilage specimens obtained adjacent to focal lesions show a dramatic loss of tensile properties and an even greater increase in swelling.12,105 Lack of a reliable clinical method of detecting cartilage injuries that have no visible disruption of the matrix has made it impossible to define the natural history of this type of damage. Probing the intact articular surface may reveal soft regions, but the significance of such softening remains uncertain. Methods that may eventually help identify these injuries in humans include imaging techniques that show the state of water within the cartilage and devices that directly measure cartilage stiffness. Chondrocytes can sense changes in matrix composition and synthesize new molecules, which allow the cells to repair damage to the macromolecular framework. The available evidence indicates that following a loss of proteoglycans, the cells increase synthesis of these macromolecules and the matrix concentration of proteoglycans begins to normalize. As a result, the material properties of the matrix return toward normal. Following significant depletion of proteoglycans, repair of the matrix may require many weeks or possibly months.22 If the cells fail to repair significant matrix macromolecular abnormalities, or if the loss of matrix molecules progresses, the tissue will ­deteriorate.94 The threshold for this injury becoming irreversible and leading to progressive loss of articular cartilage is unknown. Presumably, if the basic collagen meshwork remains intact and enough chondrocytes remain viable, the chondrocytes can restore the matrix as long as the loss of matrix proteoglycan does not exceed the production capability of the cells.1,21,94 When these conditions are not met, the cells will fail to restore the matrix, the chondrocytes will be exposed to excessive loads, and the tissue will degenerate.94 For these reasons, insults causing this type of articular cartilage injury—including immobilization, exposure of articular cartilage, and inflammation—should be minimized. Because this type of matrix macromolecular injury may temporarily increase the vulnerability of cartilage to mechanical injury, minimizing impact loading of cartilage

50

DeLee & Drez’s Orthopaedic Sports Medicine

following severe blunt trauma, prolonged immobilization, or inflammation is recommended.

Articular Cartilage Injuries That Disrupt the Tissue Severe blunt trauma, penetrating injuries, and fractures can cause visible disruption of the articular cartilage matrix. These injuries rupture, lacerate, or fracture the matrix macromolecular framework and kill chondrocytes at the site of injury without directly damaging the subchondral bone. Penetrating joint injuries that lacerate cartilage rarely occur as a result of sports, but study of experimental lacerations has provided most of our current understanding of the potential for healing of injuries limited to cartilage. The local response to these injuries depends entirely on chondrocytes. Because cartilage lacks blood vessels, no hemorrhage, fibrin clot formation, or inflammation occurs. Undifferentiated mesenchymal cells cannot migrate from blood vessels to the site of injury, proliferate, differentiate, and synthesize a new matrix. The chondrocytes are tightly encased in the collagen-proteoglycan matrix and cannot migrate to the site of injury. The chondrocytes near the injury site respond to tissue injury by proliferating and increasing the synthesis of matrix macromolecules1,21,22,24,25,27,94; however, the newly synthesized matrix and the proliferating chondrocytes are unable to fill the tissue defect, and soon after injury, the increased proliferative and synthetic activity ceases. When cartilage injury is associated with damage to the synovial membrane, blood may fill the joint, but fibrin clots do not form in the cartilage injury, and cells from the synovium and blood vessels do not migrate into the cartilage defect. The failure of clot formation and of cell migration and adhesion may be due to the proteoglycans, specifically inhibition of cell adhesion to the cartilage matrix by dermatan sulfate proteoglycans.

Cartilage Lacerations Penetrating injuries of synovial joints or surgical instruments passing across an articular surface can cut or abrade articular cartilage without damaging subchondral bone. This type of injury and the respective tissue response has been studied more extensively than blunt trauma to cartilage.1,21,22,24,25,27 Lacerations perpendicular to the articular surface kill chondrocytes at the site of injury and create matrix defects.1,21,106 These lesions cannot cause hemorrhage or initiate an inflammatory response, platelets do not bind to the damaged cartilage, and a fibrin clot does not appear. Inflammatory cells, capillaries, and undifferentiated mesenchymal cells do not migrate to the site of injury. Chondrocytes near the site of injury proliferate and form clusters, or clones, and synthesize new matrix but do not migrate to the site of the lesion. The new matrix they produce remains near the cells and therefore does not repair the damage. Shortly after the injury, these cells proliferate, and synthetic activities cease. Unlike morphologically similar osteoarthritic lesions, experimental lacerations of articular cartilage show no evidence of progression.

Superficial lacerations or abrasions of cartilage that are tangential or parallel to the articular surface also do not stimulate a successful repair response. Some cells next to the site of injury die, whereas others show evidence of increased proliferation and matrix synthesis. A thin layer of new, acellular matrix may form over the injury surface. The available evidence shows that, similar to perpendicular cartilage defects, the remaining normal tissue does not deteriorate.

Blunt Trauma that Disrupts Tissue Cartilage Alone During sports activities, impact loading, twisting, and direct blows to synovial joints occur frequently. The resulting compression of an articular surface can rupture the cartilage matrix, producing chondral fissures, flaps, or fractures without bone injury. If the compressive force is sufficiently high, the uncalcified cartilage may shear off the calcified cartilage.14 The mechanisms of these injuries have not been studied extensively, but the available evidence shows that impact loading of the articular surface can rupture the matrix.80 Disruption of normal articular cartilage by a single impact requires substantial force. A study80 of the response of human articular cartilage to blunt trauma showed that impact loads exceeding 25 newtons per square millimeter (25 MPa) caused chondrocyte death and cartilage fissures. The authors suggested that a stress level that could cause acute cartilage disruption required a force greater than that necessary to fracture the femur. Another study107 measured the pressure on human patellofemoral articular cartilage during impact loading and found that impact loads less than those necessary to fracture bone can cause stresses greater than 25 MPa in some regions of the articular surface. With the knee flexed 90 degrees, 50% of the load necessary to cause a bone fracture produced joint pressures greater than 25 MPa over nearly 20% of the contact area of the patellofemoral joint. At 70% of the bone fracture load, nearly 35% of the contact area of the patellofemoral joint pressures exceeded 25 MPa; at 100% of the bone fracture load, 60% of the patellofemoral joint pressures exceeded 25 MPa. These results suggest that impact loads can disrupt cartilage without fracturing bone. Other experimental investigations show that repetitive impact loads can split the articular cartilage matrix and can initiate progressive cartilage degeneration.96,108 Cyclic loading of human cartilage samples in vitro causes surface fibrillation.108 Periodic impact loading of bovine metacarpal phalangeal joints in vitro combined with joint motion causes rapid degeneration of the articular cartilage.109 Repeated overuse combined with peak overloading of rabbit joints in vivo causes articular cartilage damage, including formation of chondrocyte clusters, fibrillation of the matrix, thickening of subchondral bone, and penetration of subchondral capillaries into the calcified zone of the articular cartilage.96,110 The extent of cartilage damage appears to increase with longer periods of repetitive overloading, and deterioration of the cartilage continues following cessation of excessive loading. An investigation of cartilage plugs in vitro also showed that repetitive loading disrupts the tissue and the severity of the damage increases as the load and the number of loading cycles increase.100 At a compression load of

Basic Science and Injury of Muscle, Tendon, and Ligament

1000 pounds per square inch, 250 cycles caused surface abrasions, 500 cycles produced primary fissures penetrating to calcified cartilage, and 1000 cycles produced secondary fissures extending from the primary fissures. After 8000 cycles, the fissures coalesced and undermined the cartilage fragments. Higher loads caused similar changes with fewer cycles. These experiments suggest that repetitive loading can cause propagation of vertical cartilage fissures from the joint surface to calcified cartilage and extension of oblique fissures into areas of intact cartilage, extending the damage and creating cartilage flaps and free fragments. Clinical studies have identified articular cartilage fissures, flaps, and free fragments similar to those produced experimentally by single and repetitive impact loads.95,97-99 In at least some patients, acute impact loading of the articular surface or twisting movements of the joint apparently caused these injuries. In other patients, cartilage damage may have resulted from repetitive loading. Other joint injuries, including rupture of the anterior cruciate ligament and meniscal tears, frequently occur in association with cartilage damage. Taken together, the clinical and the experimental studies suggest that closed injuries to synovial joints, including direct blows and loading combined with torsion, can split articular cartilage matrix without causing bone fractures. These injuries disrupt the cartilage matrix macromolecular framework and kill chondrocytes near the injury. Because chondrocytes cannot repair these matrix injuries, the fissures either remain unchanged or progress. The experimental studies suggest that excessive repetitive loading weakens the cartilage macromolecular framework before visible matrix disruption occurs. Presumably, the chondrocytes could repair at least some of this molecular damage before cartilage fissures developed if the tissue were protected from further injury.

Osteochondral Fractures and Osteochondral Defects Sports injuries can cause fractures that extend through cartilage into the subchondral bone. Severe osteochondral fractures may result in loss of part of the articular surface. Unlike injuries limited to cartilage, fractures that extend into the vascular subchondral bone cause pain, hemorrhage, and fibrin clot formation and activate the inflammatory response (see Table 1A3-1).1,21,24,25,27,94,101,111-118 Undifferentiated mesenchymal cells migrate into the region of the fibrin clot, proliferate, and synthesize a new matrix, so most osteochondral defects fill with new cells and matrix. Soon after an osteochondral injury, blood escaping from the blood vessels in damaged bone forms a hematoma that temporarily fills the injury site. Fibrin forms within the hematoma, and platelets bind to fibrillar collagen and establish hemostasis. A continuous fibrin clot fills the bone defect and extends for a variable distance into the cartilage defect. Platelets within the clot release potent vasoactive mediators (including serotonin, histamine, and thromboxane A2) and growth factors or cytokines (small proteins that influence multiple cell functions, such as migration, proliferation, differentiation, and matrix synthesis, including transforming growth factor-β and platelet-derived

51

growth factor).22 Bone matrix also contains growth factors, including transforming growth factor-β, bone morphogenetic protein, platelet-derived growth factor, insulin-like growth factor I, insulin-like growth factor II, and possibly others. Release of these growth factors after injury may stimulate vascular invasion and migration of undifferentiated cells into the clot and influence their proliferative and synthetic activities. Shortly after entering the tissue defect, the undifferentiated mesenchymal cells proliferate and synthesize a new matrix. Within 2 weeks of injury, some of the mesenchymal cells assume the rounded form of chondrocytes and begin to synthesize a matrix that contains type II collagen and a relatively high concentration of proteoglycans.1,21,22,94 These cells produce regions of hyaline-like cartilage in the chondral and bone portions of the defect. In many osteochondral defects, the regions of hyaline-like cartilage first appear next to the exposed bone matrix, leaving the central region of the defect filled with more fibrous tissue.101 Six to 8 weeks after injury, the repair tissue within the chondral region of most defects contains many chondrocyte-like cells in a matrix consisting of type II collagen, proteoglycans, some type I collagen, and noncollagenous proteins.101 By contrast, the repair cells in the bone portion of the defect produce immature bone, fibrous tissue, and hyaline-like cartilage21,94,101 and soon restore the original level of subchondral bone. Capillaries that had approached or entered the chondral portion of the defect recede. Six months after injury, the mesenchymal cells have repaired the bone defect with a tissue consisting primarily of bone with some regions of fibrous tissue, small blood vessels, and hyaline cartilage.101 By contrast, the chondral portions of large osteochondral defects rarely fill completely with repair tissue.21,94,101 In animal experiments, repair tissue fills about two thirds of the total volume of the chondral portion of large osteochondral defects but fills more than 95% of the volume of the bone portion.1,22 In addition, the tissue in the chondral and bone ­portions of the defect differs significantly in composition.101 The chondral repair tissue does not contain bone or blood vessels and has a significantly higher proportion of hyalinelike cartilage. In most regions of the chondral defects, the repair tissue has a composition and a structure intermediate between hyaline cartilage and fibrocartilage and rarely replicates the elaborate structure of normal articular cartilage. The differences in the differentiation of the repair tissue in the chondral and the bony portions of the same defect show that the environment in the two regions causes the repair cells to produce different tissues. The important differences in environment may be mechanical, biologic, electrical, or other unknown factors. The cartilage repair tissue occasionally persists unchanged or progressively remodels to form a functional joint surface, but in most large osteochondral injuries, the chondral repair tissue begins to show depletion of matrix proteoglycans, fragmentation, and fibrillation; increasing collagen content; and loss of cells that have the appearance of chondrocytes within 1 year or less.1,21,22,94,116,118 The remaining cells often assume the appearance of fibroblasts as the surrounding matrix becomes primarily densely packed collagen fibrils. This fibrous tissue usually fragments and often disintegrates, exposing areas of bone.1,21,94,118

52

DeLee & Drez’s Orthopaedic Sports Medicine

The inferior mechanical properties of cartilage repair tissue may be responsible for its frequent deterioration.94 Several experimental studies show that even repair tissue that successfully fills osteochondral defects lacks the stiffness of normal articular cartilage.119-121 Cartilage repair tissue formed in rabbit metatarsophalangeal joint arthroplasty sites deforms more easily and takes longer to recover from deformation than does normal articular cartilage.120 Repair cartilage formed in pig joints swells in Ringer’s solution more than normal cartilage and has greater permeability and less stiffness on compression.121 Chondral repair cartilage in primate osteochondral defects is more permeable and less stiff on compression than normal articular cartilage.119,122,123 Differences in matrix composition and organization may explain the differences between the mechanical properties of repair cartilage and normal cartilage.94 The increased swelling of repair cartilage indicates a lack of organization or a weakness of the collagen fibril meshwork. Microscopic studies of repair cartilage show that the orientation of the collagen fibrils in even the most hyaline-like cartilage repair tissue does not follow the pattern seen in normal articular cartilage.101 In addition, the repair tissue cells may fail to establish the normal relationships between matrix macromolecules, particularly between cartilage proteoglycans and the collagen fibril network. This failure may occur because of lack of organization of the macromolecules, insufficient concentrations of some macromolecules, or the presence of molecules that interfere with the assembly of a normal cartilage matrix. For example, type I collagen or high concentrations of dermatan sulfate proteoglycans might interfere with the establishment of normal collagen proteoglycan relationships. The decreased stiffness and increased permeability of repair cartilage matrix subjects the macromolecular framework to increased strain fields during joint use, resulting in progressive structural damage to the matrix collagen and proteoglycans.94 Mechanical failure of the matrix may expose the repair chondrocytes to excessive loads, further compromising their ability to restore the matrix. Thus, although cartilage repair tissue may initially have a composition and a structure that closely resemble normal articular cartilage, defects in organization of the matrix macromolecular framework could compromise the function and the durability of the repair tissue. The success of chondral repair in osteochondral injuries may depend on the severity of the injury, as measured by the volume of tissue injured, the surface area of cartilage injured, the stability of the injury site, and the age of the individual.22 Smaller osteochondral defects heal more predictably and more successfully than larger defects.1,22,101,114,124 A study of intra-articular fractures of the distal femur in rabbits showed that anatomically reduced cartilage fractures that were stabilized by compression fixation healed with apparently normal articular cartilage.125 Inadequately and adequately reduced fractures that were not stabilized by compression fixation healed with fibrocartilage. Age may also influence the success of osteochondral repair, although evidence is lacking. Bone is known to heal more rapidly in children than in adults, and the articular cartilage chondrocytes in skeletally immature animals show

a better proliferative response to injury and synthesize larger proteoglycan molecules than do those from mature animals.55,62,126,127

SURGICAL TREATMENT OF ARTICULAR CARTILAGE DAMAGE Localized cartilage disruption can compromise joint function and may lead to progressive cartilage deterioration. ­Consequently, surgeons have sought effective methods for treating localized cartilage damage. Currently, shaving damaged cartilage and abrading exposed subchondral bone are the most common surgical treatments for damaged cartilage.

Shaving Fibrillated Articular Cartilage Many surgeons shave fibrillated cartilage to decrease joint symptoms and achieve a smoother articular surface. ­Shaving degenerating articular surfaces can remove frayed and fibrillated superficial cartilage, but the efficacy of this procedure in decreasing pain, improving joint function, or stimulating restoration of the normal articular surface has not been established.94,128 Several reports describe a decrease in symptoms following arthroscopic débridement of loose cartilage fragments and flaps, torn menisci, osteophytes, and proliferative synovium in osteoarthritic knee joints,129,130 but experimental and clinical studies have not shown a clear benefit to shaving fibrillated cartilage. ­ Shaving normal rabbit patellar cartilage did not stimulate a ­significant repair response but also did not cause progressive deterioration.131 In one series of patients, only 25% of patients had satisfactory results, and the investigators concluded that the procedure “is ­disappointing and ineffective.” 84,132,133 A study of human femoral articular cartilage after arthroscopic ­ shaving for treatment of cartilage damage found no evidence of restoration of a smooth articular surface; to the contrary, shaving may have actually increased fibrillation and chondrocyte necrosis in and adjacent to the region of the original defect.134 Removal of fragments of degenerating cartilage with joint irrigation may decrease symptoms of mechanical catching and pain in some patients. Experimental injection of cartilage fragments into rabbit knees produced inflammatory arthritis with joint effusions, increased levels of synovial enzymes, and articular cartilage friability, pitting, and discoloration,135 suggesting that cartilage fragments can contribute to synovitis. Presumably, débridement of fibrillated cartilage and the ­ associated joint irrigation would temporarily decrease this synovial inflammation by removing cartilage particles as well as any degradative enzymes and inflammatory mediators.

Abrasion of Subchondral Bone Arthroscopic abrasion offers a potentially attractive treatment for articular surfaces that have small regions of full-thickness cartilage loss.128 Arthroscopic abrasion stimulates formation of cartilage repair tissue and can

Basic Science and Injury of Muscle, Tendon, and Ligament

provide temporary symptomatic improvement in selected patients.94,128 The surgeon removes the most superficial layers of subchondral bone, usually 1 to 3 mm, to disrupt intraosseous vessels.136-138 The resulting hemorrhagic exudate forms a fibrin clot, and undifferentiated cells invade the clot, forming repair tissue over the abraded bone. Protection of the joint from excessive loading during the healing response allows the repair tissue to remodel and form a new articular surface. The repair tissue has a fibrocartilaginous appearance and contains variable concentrations of type II collagen,137,138 similar to the repair tissue formed in the chondral regions of experimental osteochondral defects,1,21,101 but no regeneration of normal articular cartilage occurs. Several authors have reported that arthroscopic abrasion of the knee can decrease pain in 60% or more of patients.136-139 Abrasion of exposed subchondral bone results in formation of a fibrocartilaginous repair tissue, with variable concentrations of type II collagen, that in some patients persists for years.136-138 Despite the ability of arthroscopic abrasion to stimulate formation of cartilage repair tissue and the encouraging reports of symptomatic improvement, it is difficult to assess the value of this procedure. The available evidence suggests that it can provide temporary symptomatic improvement in selected patients.94,128

CARTILAGE GRAFTS The mechanical properties of most naturally occurring cartilage repair tissue following osteochondral injury or surgical treatment is poor. Consequently, various grafts, including osteochondral autografts, osteochondral allografts, periosteal grafts, and perichondrial grafts, have been proposed as alternatives to replace regions of damaged or lost articular surface. In addition, chondrocytes or undifferentiated mesenchymal cells can be isolated and grown in culture to be implanted in a gel or other artificial matrix to replace articular cartilage.

Cartilage Autografts Animal experiments show that articular cartilage autografts transplanted with a thin shell of bone heal to the recipient site tissue.140-144 These studies also show that chondrocytes in adequately stabilized autografts remain viable, and the matrix remains intact for 1 year or more. Surgeons rarely use cartilage autografts in humans because of the lack of donor sites. Possible cartilage autograft donor sites include the proximal tibiofibular joint, the sternum, and the patella. Animal experiments show that sternal osteochondral autografts can replace segments of articular cartilage.145 In humans, osteochondral patellar grafts have been used to replace severely damaged portions of the tibial articular surface,146 and osteochondral grafts from the intercondylar notch and femoral trochlea have been used to replace full-thickness defects of the femoral condyles.147,148 Radiographs show that the graft bone heals with the recipient site bone, and clinical evaluation shows satisfactory joint function without knee effusions or evidence of degeneration for 18 years or more. A randomized trial comparing osteochondral autografting, abrasion chondroplasty, and microfracture of the subchondral bone in the treatment

53

of recalcitrant osteochondral lesions of the talus found no significant differences in outcome among these procedures.149

Periosteal and Perichondrial Autografts Periosteum and perichondrium provide other sources of tissue for repair of articular cartilage defects. Cells from both tissues can synthesize the necessary matrix macromolecules to form hyaline cartilage and survive transplantation without a vascular pedicle.150-159 In experimental and clinical studies, most grafts have been harvested from the recipient, but allografts also can restore a joint surface.160 Experimental work with rabbit articular surfaces shows that periosteal grafts will fill large defects with hyaline-like cartilage; this tissue remains intact for as long as 1 year after transplantation.155-158,161,162 Perichondrial grafts have produced similar results in rabbits and dogs.150,151,154,163-169 The graft cells produce a matrix that contains type II collagen,150,156,166 and the concentration of type II collagen may increase with time.151 The mechanical properties of these grafts have not been thoroughly examined, but the viscoelastic properties of perichondrial grafts improved after surgery and were similar to normal cartilage by 26 weeks after surgery in one published study.169 These observations suggest that these grafts have some capacity to remodel. Although periosteal and perichondrial cells can ­survive transplantation and form a new articular surface, the results vary among animals and possibly among joint regions. Investigations of rabbit perichondrial grafts found that 38% to 50% produced unacceptable results such as fractures, failures of graft attachment, or infection.150,151 A study of periosteal grafts in rabbits showed frequent ­success in the femoral condyles and patellar grooves but a high rate of failure in the patella.170 Although the composition and the mechanical ­properties of grafts can improve after transplantation,151,169 several studies show that the grafts may deteriorate. In one set of experiments, dog perichondrial grafts formed new cartilage with smooth articular surfaces by 2 to 8 months after surgery, but by 12 to 17 months, the grafts degenerated, leaving exposed bone in some regions and fragments of graft tissue in others.163,164 A study of periosteal grafts in sheep also found graft deterioration following use of the joint; moderately well-differentiated fibrocartilage covered the previously exposed bone 1 year after periosteal grafting, but by 2 years after transplantation, the graft tissue had degenerated. The age of the graft may influence the results, although the evidence is sparse. With increasing age, periosteum and perichondrium become thinner and less cellular, and the potential of the cells to proliferate and to synthesize new matrix decreases.22 Cryopreserved periosteal allografts from young rabbits produce better results than grafts from older animals.160 Early motion of the joint following grafting may influence the results. It is not clear how motion affects cell function in cartilage repair tissue. Motion stimulates fibrocartilage formation following joint resection,171 and passive motion may promote cartilage formation by periosteal grafts.155,156,159,161,162,172 Grafts placed in immobilized joints form less cartilage than grafts treated with passive motion.172 The long-term benefits of passive motion treatment, however, remain uncertain.22 The apparent beneficial

54

DeLee & Drez’s Orthopaedic Sports Medicine

effect of early motion treatment of rabbit perichondrial grafts did not persist in one series of studies.166,169,173 ­Passive joint movement appeared to improve the quality of the grafts initially, but 1 year after transplantation, no benefits were apparent. Thus, passive joint motion soon after graft transplantation may affect the initial behavior of the graft cells, but the mechanism, the optimal timing and duration of motion, and the long-term benefits have not yet been established. Surgeons have used perichondrial grafts to replace damaged or lost articular cartilage in human osteoarthritic and rheumatoid joints. Degenerated or damaged cartilage is removed, and a rib perichondrial autograft is placed in the defect.165,168,174-176 Most reports of this procedure have described results for joints of the upper extremity. Some patients reported improved range of motion and decreased pain, but the results have been unpredictable.165 One clinical series showed that the results of rib perichondrial allograft arthroplasty in metacarpophalangeal and proximal interphalangeal joints depended to a large extent on the age of the patient.177 All patients in their 20s had good results with metacarpophalangeal joint arthroplasty, whereas only 75% of the patients in their 30s had good results. Seventy-five percent of teenaged patients had good results with proximal interphalangeal joint arthroplasty, compared with 66% of patients in their 20s and none of the patients older than 40 years. These results support the concept that the production of new cells and matrix in perichondrium and periosteum declines with age. The best clinical results with these grafts may be expected in skeletally immature patients and in those who have recently reached skeletal maturity.

Cartilage Allografts Cartilage allografts have the advantage of providing osteochondral segments of any size or shape without donor site morbidity. Biopsies of allograft cartilage show that many chondrocytes remain viable years after transplantation.178 Surgeons have used allografts to replace segments of articular surfaces, entire articular surfaces, and entire synovial joints. Large grafts have been used for joint reconstruction following tumor resection or major trauma.179,180 Smaller grafts, usually consisting of articular cartilage and a thin shell of subchondral bone, have been used to replace damaged regions of articular cartilage in young, physically active patients. Generally, larger grafts have more frequent and more severe surgical and postoperative complications, including infection and mechanical failure.180-182 Fresh and cryopreserved grafts have been used experimentally and clinically. Fresh grafts presumably have the advantage of maintaining the maximal viability of the chondrocytes178; one study of osteochondral allografts from large dogs showed that fresh grafts produced better results than frozen grafts.183 Use of preserved grafts makes it possible to accumulate a bank of grafts of different sizes and shapes, and freezing has the advantage of decreasing graft immunogenicity.183,184 Cryopreservation and storage for up to 28 days do not alter the mechanical properties or the structure of cartilage.185,186 Chondrocytes can survive freezing and thawing,187,188 but in one study, only a few chondrocytes survived freezing.183 The frozen grafts also had more evidence of structural deterioration and

lower concentrations of glycosaminoglycans than the fresh grafts.183 Cartilage allografts can survive transplantation and heal to the recipient-site tissue.67,143 A study of replacement of the tibial articular surface in skeletally mature rabbits showed that articular cartilage, growth plate, and cultured chondrocyte allografts, when correctly positioned and secured, all resulted in significantly better repair than did the natural repair response.39 Other studies have shown that some allograft cartilage degenerates when subjected to loading,140,183,189 and one study of dog osteochondral allografts showed that the allograft cartilage became thin, dull, and rough over time.183 The size of the graft (including the amount of bone transplanted with the cartilage), freezing of the graft, and antigen matching may affect the host response to osteochondral allografts. Small cartilage allografts do not cause an apparent inflammatory reaction,67,143 but large osteochondral allografts can cause synovial inflammation.183,190 This difference may be an effect of the amount of bone in the graft. Large antigen-mismatched osteochondral allografts stimulate systemic humoral, cell-mediated, and antibody-dependent cell-mediated and local immune responses,183,190 although even large antigen-matched osteochondral grafts can cause synovitis.183 Host immune responses adversely affect cartilage grafts. One study compared the results of leukocyte antigen­mismatched frozen allografts, leukocyte antigen-mismatched fresh allografts, leukocyte antigen-matched fresh allografts, and leukocyte antigen-matched frozen allografts in dogs.183 Leukocyte antigen-mismatched fresh allografts stimulated the most severe inflammatory response. Invasive pannus appeared more frequently in the joints with fresh grafts, especially those joints with leukocyte antigen-mismatched grafts. In some of these joints, the pannus eroded the cartilage down to the subchondral bone. Antigen mismatching increased cartilage deterioration and exacerbated the damage due to freezing. Fresh antigen-matched grafts produced results similar to those seen with autogenous grafts. This study shows that for large segmental osteochondral allografts, fresh tissue-matched grafts produce the best results. Fresh osteochondral allografts can replace localized regions of damaged articular cartilage in humans.182,191,192 Fresh osteochondral grafts used to replace portions of damaged tibial plateaus decreased pain and improved function in 10 of 12 patients who were followed for more than 2 years.191 Evaluation of 40 knees 2 to 10 years after transplantation of fresh osteochondral allografts for localized degeneration of the articular surface showed that 31 of the grafts had healed and 9 had failed.182 Of the 31 successful transplants, 13 had an excellent result, 14 had a good result, and 4 had a fair result. Fresh osteochondral shell allografts were successful in most cases of post-traumatic degenerative arthritis of the patella, for post-traumatic arthritis and traumatic defects of the tibial plateau, and for traumatic defects, osteochondritis dissecans, and avascular necrosis of the femoral condyle. Only 3 of the 10 grafts for unicompartmental degenerative arthritis of the knee involving both the femur and the tibia succeeded. This evidence suggests that fresh allografts can provide improvement for selected patients with disabling symptoms due to isolated regions of degenerated or damaged cartilage.

Basic Science and Injury of Muscle, Tendon, and Ligament

Chondrocyte and Artificial Matrix Grafts Use of synthetic matrix grafts, with or without cells and growth factors that stimulate cartilage formation, is another method of replacing regions of damaged or lost articular cartilage.1,21,94,193-204 Synthetic matrices can be created in various sizes and shapes to fill any chondral defect precisely. A synthetic matrix provides a framework for cell migration and attachment and may protect the cells from excessive loading. The cells included in these grafts may be autografts or allografts. Chondrocytes or ­mesenchymal cells harvested from the intended recipient or from another individual can be grown and maintained in culture and then reimplanted. Most synthetic matrices used to replace articular ­cartilage and to implant growth factors or cells consist of reconstituted collagen, but one group of investigators has reported improved cartilage repair with carbon fiber pads.205 The matrix composition and organization can influence cell migration, proliferation, and differentiation.206,207 One in vitro study showed that a collagenous matrix promoted formation of cartilage by mesenchymal cells.207 Several experimental studies have shown the feasibility of implanting chondrocytes or mesenchymal cells in cartilage defects.1,94 The implanted cells survive and synthesize a collagenous matrix that often resembles that of normal articular cartilage.201-204 This tissue resembles hyaline cartilage more closely than the ­ tissue that forms in defects not treated with ­chondrocyte ­ artificial matrix and synthetic matrix transplants. In one series of experiments, the investigators created 4-mm diameter osteochondral defects in rabbit articular surfaces and then placed collagen gels containing allograft chondrocytes in the defects.203,204 Eighty percent of the treated defects showed successful healing 24 weeks later. Other groups have also reported improved cartilage healing using similar methods in animal models.196,198,199,201,202,208 Thus far, investigators have not identified the optimal type of synthetic matrix, defined the benefits of implanting various cell types and growth factors, or shown that implantation of synthetic matrices containing chondrocytes or undifferentiated mesenchymal cells or growth factors can predictably restore a durable articular surface in large cartilage defects. Yet, the existing studies show that this approach has the potential to improve repair of limited osteochondral defects.

CONCLUSIONS The specialized composition and organization of articular cartilage provide the unique biomechanical properties that make possible normal synovial joint function. Blunt trauma, high-energy joint loading, and forceful joint loading and twisting can damage cartilage without causing visible disruption of the matrix, can fracture cartilage without damaging the subchondral bone, or can fracture both cartilage and subchondral bone. Injuries that alter the biomechanical properties of cartilage compromise smooth, pain-free joint motion and may lead to progressive deterioration of the articular surface. Damage to cartilage that does not cause visible matrix disruption has been studied

C

r i t i c a l

P

55

o i n t s

l The

unique viscoelasticity and biomechanical properties of articular cartilage minimizes stress to subchondral bone while maintaining its own structural integrity, providing remarkable durability. l High-energy forces generate high shear stresses at the cartilage–subchondral bone junction that can cause matrix lesions, leading to cartilage deterioration and joint dysfunction. By contrast, repetitive trauma can cause subtle damage to the matrix macromolecular framework and chondrocytes, ultimately altering the chondrocyte-matrix functions that maintain cartilage mechanical properties. l After maturity, chondrocytes rarely divide but continue to synthesize and maintain the extracellular matrix (collagens, proteoglycans, and noncollagenous proteins and glycoproteins); turnover of proteoglycans is relatively fast, whereas collagen turnover is very slow. l Morphology of articular cartilage cells and matrix delineates four zones or layers of articular cartilage: the superficial tangential zone (large water content), the middle or transitional zone (metabolically active cells with less water than the superficial layer), the deep or radial zone (high concentration of proteoglycans and the least water), and the zone of calcified cartilage. l Three general types of articular cartilage injury exist: (1) disruption of the macromolecular framework; (2) mechanical disruption of articular cartilage; and (3) mechanical disruption of articular cartilage and subchondral bone. Tissue response and potential for repair depend on the type of injury, but in general, the potential for restoring normal tissue is limited. l Surgical treatment of articular cartilage injury includes shaving fibrillated cartilage, subchondral abrasion, and ­various autograft, allograft, and chondrocyte–artificial ­matrix graft procedures.

in animals but is difficult to detect in humans. Improved imaging methods or devices that measure in vivo cartilage mechanical properties may solve this problem. Currently, the natural history of such injuries remains unknown, but the experimental evidence shows that the chondrocytes can repair damage if the basic structure of matrix remains intact and the cartilage is spared further injury. Chondrocytes cannot heal injuries that disrupt cartilage alone, but injuries that disrupt both cartilage and bone stimulate an inflammatory response and migration of undifferentiated cells into the injury site. These cells proliferate, differentiate into chondrocyte-like cells, and synthesize a new matrix, but the tissue they produce usually fails to restore the normal volume of articular cartilage. Furthermore, this repair tissue lacks the biomechanical properties of articular cartilage. Because cartilage has a limited capacity for restoring normal tissue after significant injury, reliable methods of repairing or replacing damaged articular surfaces have been sought. Thus far, all of these methods have substantial limitations, but the knowledge gained from studying them provides a basis for developing better methods of treating articular cartilage injuries and preventing progression of cartilage damage.

56

DeLee & Drez’s Orthopaedic Sports Medicine

S U G G E S T E D

R E A D I N G s

Athanasiou KA, Spilker RL, Buckwalter JA, et al: Finite element biphasic modeling of repair articular cartilage. Adv Bioeng 15: 95-96, 1989. Beris AE, Lykissas MG, Papageorgiou CD, Georgoulis AD: Advances in articular cartilage repair. Injury 36(Suppl 4):S14-23, 2005. Buckwalter JA, Rosenberg LC, Coutts R, et al: Articular cartilage: Injury and repair. In Woo SL, Buckwalter JA (eds): Injury and Repair of the Musculoskeletal Soft Tissues. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1988, pp 465-482. Chiang H, Kuo TF, Tsai CC, et al: Repair of porcine articular cartilage defect with autologous chondrocyte transplantation. J Orthop Res 23:584-593, 2005. Chow JC, Hantes ME, Houle JB, Zalavras CG: Arthroscopic autogenous osteochondral transplantation for treating knee cartilage defects: A 2- to 5-year follow-up study. Arthroscopy 20:681-690, 2004. Dorotka R, Bindreiter U, Macfelda K, et al: Marrow stimulation and chondrocyte transplantation using a collagen matrix for cartilage repair. Osteoarthritis Cartilage 13:655-664, 2005.

Gobbi A, Francisco RA, Lubowitz JH, et al: Osteochondral lesions of the talus: Randomized controlled trial comparing chondroplasty, microfracture, and osteochondral autograft transplantation. Arthroscopy 22:1085-1092, 2006. Guilak F, Fermor B, Keefe FJ, et al: The role of biomechanics and inflammation in cartilage injury and repair. Clin Orthop 423:17-26, 2004. Kelly DJ, Prendergast PJ: Mechano-regulation of stem cell differentiation and tissue regeneration in osteochondral defects. J Biomech 38:1413-1422, 2005. Moore EE, Bendele AM, Thompson DL, et al: Fibroblast growth factor-18 stimulates chondrogenesis and cartilage repair in a rat model of injury-induced osteoarthritis. Osteoarthritis Cartilage 13:623-631, 2005.

R eferences Please see www.expertconsult.com

S ecti o n

A

Physiology of Injury to Musculoskeletal Structures 4. Meniscus Injury Mark R. Brinker, Daniel P. O’Connor, Louis C. Almekinders, Thomas M. Best, Joseph A. Buckwalter, William E. Garrett, Jr., Donald T. Kirkendall, Van C. Mow, and Savio L.-Y. Woo

As recently as the 1970s, some surgeons assumed that knee menisci were vestigial structures that had little effect on joint function, and they treated small meniscal tears with total meniscectomy. Since then, understanding of meniscal function and treatment of meniscal injuries have advanced considerably. Research has included comparisons of meniscal structure and function among species, investigations of the relations between the biomechanical function of menisci and their composition and organization, and long-term studies showing that total meniscectomy adversely affects joint function and increases the probability of joint degeneration. The adverse effects of loss of menisci on joint function have encouraged surgeons and other investigators to seek methods of promoting meniscal healing and replacement. The menisci of humans are often described as semilunar in shape.1-4 Menisci exist in many animals other than humans.1,5 Monkeys, some types of bats, crocodiles, and bullfrogs have menisci (in the latter two, the menisci are discoid). In birds, the menisci are C-shaped, similar to the knee joints of humans and the larger bovine species. Bovine knee menisci are closer to discoid in shape than are menisci of human knees. There is also a disk-shaped fibrocartilaginous disk in the temporomandibular joint of humans and rodents that is very similar to knee meniscus in terms of composition and structure. A meniscus is present in the ankle joint of kangaroos, which is subjected to large forces

during jumping.5,6 From these and other comparative ­anatomic studies, investigators have recognized that a firm intra-articular fibrocartilaginous structure with great tensile strength is required in joints across which high forces are transmitted as well as in joints in which rotation and translation occur, which requires excellent lubrication and probably additional mechanical stability.5,7-9 Recent investigations using microscopy, biochemistry, and bioengineering have significantly advanced the understanding of the relationships between meniscal function and meniscal composition and organization. Like articular cartilage, knee meniscal tissue performs important mechanical functions,1,10-12 including load bearing,13-25 shock absorption,4,26,27 joint stabilization,26,28-31 and possibly joint lubrication.4,5,32-34 These functions depend on a highly organized extracellular matrix consisting of fluid and a macromolecular framework formed of collagen (types I, II, III, V, and VI), proteoglycans, elastin, and noncollagenous proteins. Unlike articular cartilage, the peripheral 25% to 30% of the lateral meniscus and the peripheral 30% of the medial meniscus1,2,35-38 have a blood supply, and the peripheral regions of the meniscus, especially the meniscal horns,1,39-41 have a nerve supply. Loss of menisci alters the loading of articular cartilage in ways that may increase the probability and the severity of degenerative joint disease.1,13,14,16,26,32,42-45 As a result,

Basic Science and Injury of Muscle, Tendon, and Ligament

surgeons and investigators have studied the response of menisci to injury and have sought methods of preserving, repairing, and replacing menisci.46-53 Tears through the vascular regions of the meniscus can heal, but tears through the avascular regions do not undergo a repair process that can heal a significant tissue defect.1,2,36,54 A number of reports suggest that methods of stimulating healing of tears in avascular meniscal tissue and of replacing meniscal tissue exist and may help to maintain or restore meniscal function.1,52-62 This chapter first summarizes the current understanding of meniscal composition, structure, mechanical properties, blood supply, and nerve supply. Subsequent sections review the response of menisci to injury and methods of stimulating meniscal healing. The final section discusses meniscal grafts.

COMPOSITION Like ligament, cartilage, and bone, meniscus consists of scattered cells surrounded by an abundant extracellular matrix.1,11,12,38,47,59,63 Structural integrity and function of the tissue depend on interactions between the cells and their surrounding matrix. The material properties of the tissue result from the composition and the organization of the matrix macromolecules and the interactions between these solid components of the matrix and the tissue fluid.

Cells Based on morphologic characteristics, there are two major types of meniscal cells.38,64 Near the surface, the cells have flattened ellipsoid or fusiform shapes; in the deep zone, the cells are spherical or polygonal. These differences in cell shape and size between the superficial and the deep regions of the tissue resemble the changes in cell morphology seen between the superficial and the deep regions of articular cartilage. Like the cells from the superficial and the deep zones of articular cartilage, the superficial and the deep meniscal cells appear to have different metabolic functions or perhaps different responses to loading.60 These cells produce and maintain the macromolecular framework of the meniscal tissue. The cells contain the organelles, endoplasmic reticulum, and Golgi membranes that are necessary to accomplish their primary function of synthesizing matrix macromolecules. Like most other mesenchymal cells, they lack cell-to-cell contacts.1,11 Because most of them lie at a distance from blood vessels, they rely on diffusion through the matrix for transport of nutrients and metabolites. The membranes of meniscal cells attach to matrix macromolecules through adhesion proteins (fibronectin, thrombospondin, type VI collagen38,65-67). The matrix, particularly the pericellular region, protects the cells from damage due to physiologic loading of the tissue.68,69 Deformation of the macromolecular framework of the matrix causes fluid flow through the matrix4,33,70,71 and influences meniscal cell function.68,69 In all these features, the meniscal cells resemble the articular cartilage chondrocytes. Because meniscal tissue is more fibrous than hyaline cartilage, some authors have proposed that meniscal cells be called fibrochondrocytes.38,61

57

Extracellular Matrix Water Water contributes 65% to 75% of the total weight of meniscus.4,70,72,73 Table 1A4-1 summarizes the water, sulfated glycosaminoglycan, and hydroxyproline content of meniscus.70 Some portion of the water may reside within the intrafibrillar space of the collagen fibers.74-76 Most of the water is retained within the tissue in the solvent domains of the proteoglycans by means of both their strong hydrophilic tendencies and the Donnan osmotic pressure exerted by the counter-ions associated with the negative charge groups on the proteoglycans.4,77-81 Because the pore size of the tissue is extremely small (<60 nm), very large hydraulic pressures are required to overcome the drag of frictional resistance when forcing fluid flow through the tissue. These interactions between water and the macromolecular framework of the matrix significantly influence the viscoelastic properties of the tissue.

Matrix Macromolecules Collagens The macromolecular framework of the meniscal matrix consists primarily of collagens, which may contribute up to 95% of the dry weight of the tissue (see Table 1A4-1). Most of this collagen is type I.1,38,82 Types II, III, V, and VI each may contribute 1% to 2% of the total amount of tissue collagen (strictly speaking, type VI may be classified as a matrix glycoprotein).38,66,83 The large-diameter type I collagen fibrils that lie mostly in the outer radial two thirds of the meniscus give this tissue its ultrastructural arrangement and tensile stiffness and strength.4,33,64,70,82,84-86 The type II collagen fibrils have smaller diameters and are located in the inner one third, avascular region of the meniscus near the surface of the tissue.87 This inner region is also rich in proteoglycans and has a hyaline appearance.1,88 Little information exists about types III and V collagen in the meniscus. Type VI collagen may play a role in stabilizing the type I and II collagen framework of the meniscus and in maintaining fibrochondrocyte adhesion to the matrix.1,38,83

  TABLE 1A4-1  Composition of Meniscus by Region Sulfated Glycosaminoglycan Water Region Number (% Dry Weight) (%) Content (%) LA LC LP MA MC MP

18 18 18 12 14 18

1.80 + 0.50 1.68 + 0.56 1.75 + 0.45 2.20 + 1.01 2.06 + 0.68 1.94 + 0.83

75.02 + 2.14 72.99 + 2.40 73.39 + 2.44 72.12 + 9.73 76.77 + 2.68 74.88 + 7.32

Hydroxyproline (% Dry Weight) 14.3 + 3.7 13.2 + 2.0 15.2 + 3.1 13.2 + 3.6 13.9 + 3.4 13.9 + 3.6

LA, lateral anterior; LC, lateral central; LP, lateral posterior; MA, medial anterior; MC, medial central; MP, medial posterior. From Fithian DC, Kelly MA, Mow VC: Material properties and structure-function relationships in the menisci. Clin Orthop 252:19-31, 1990.

58

DeLee & Drez’s Orthopaedic Sports Medicine

Proteoglycans

Elastin

Some meniscal regions have a proteoglycan concentration of up to 3% of their dry weight.1,38,63,70,72,89 Like proteoglycans from other dense fibrous tissues, meniscus proteoglycans can be divided into two general types. The large, aggregating proteoglycans expand to fill large volumes of matrix and contribute to tissue hydration and the mechanical properties of the tissue. The smaller, nonaggregating proteoglycans usually have a close relationship with fibrillar collagen.38,90,91 The large aggregating proteoglycans from meniscus have the same structure as the large aggregating proteoglycans from articular cartilage.63,91 The concentration of large aggregating proteoglycans suggests that they probably contribute less to the properties of meniscus than to the properties of articular cartilage.4,70,79-81 As with the quantitatively minor collagens, the smaller nonaggregating meniscal proteoglycans may help organize and stabilize the matrix, but at present, their function remains unknown.

Elastin contributes less than 1% of the dry weight of the meniscus.1,11,12,38 The contribution of elastin to the mechanical properties of meniscal tissue is uncertain. The sparsely distributed elastic fibers are unlikely to play a significant role in the organization of the matrix or in determining the mechanical properties of the tissue.

Noncollagenous Proteins Noncollagenous proteins also form part of the macromolecular framework of meniscus and may contribute as much as 10% of the dry weight of the tissue in some regions.38,88 Two specific noncollagenous proteins, link protein and fibronectin, have been identified in meniscus.38 Link protein is required for the formation of the stable proteoglycan aggregates that are capable of forming strong networks.92,93 Fibronectin serves as an attachment protein for cells in the extracellular matrix.65,67 Other noncollagenous proteins such as thrombospondin94 may serve as adhesive proteins in the tissue, thus contributing to the structure and the mechanical strength of the matrix. The exact details of their composition and function in the meniscus remain largely unknown.

A

A

STRUCTURE Within the meniscus, the diameter and the orientation of the collagen fibrils and cell morphology vary from the surface to the deeper regions.1,4,33,38,64,78,84,85,95 The highly ordered arrangement of collagen fibrils within the tissue correlates closely with the biomechanical properties of meniscus. As in articular cartilage, the mechanisms responsible for organizing the collagen fibrils within meniscus remain unknown, but a study of meniscal development suggests that weight-bearing may influence meniscal collagen fibril organization.37 The meniscal surface is a thin layer that is rich in type II collagen and consists of a randomly woven mesh of fine collagen fibrils that lie parallel to the surface. Below this surface layer, large, circumferentially arranged collagen fiber bundles (mostly type I) course throughout the body of each meniscus.1,4,70,84,85 These circumferential collagen bundles give meniscus great tensile stiffness and strength parallel to their orientation.4,70,85,95,96 They insert into the anterior and the posterior meniscal attachment sites on the tibial plateau, and large forces are transmitted through these attachment sites. Figure 1A4-1A illustrates these large fiber bundles and the thin superficial surface layer. Figure 1A4-1B is a photograph of a bovine medial meniscus with the surface layer removed, showing the large collagen bundles of the deep zone.

B

B

Figure 1A4-1  A, Diagram of collagen fiber architecture throughout the meniscus. Collagen fibers of the thin superficial sheet are randomly distributed in the plane of the surface and are predominantly arranged in a circumferential fashion deep in the substance of the tissue. A indicates the anterior insertion of the meniscus into the tibia, and B indicates the posterior insertion. B, Macrophotograph of bovine medial meniscus with the surface layer removed, showing the large circumferentially arranged collagen bundles of the deep zone.   (A, Redrawn from Bullough PG, Munuera L, Murphy J, et al: The strength of the menisci of the knee as it relates to their fine structure. J Bone Joint Surg Br 52:564-570, 1970; B, from Proctor CS, Schmidt MB, Whipple RR, et al: Material properties of the normal medial bovine meniscus. J Orthop Res 7:771-782, 1989.)

Basic Science and Injury of Muscle, Tendon, and Ligament

59

anterior 30º

30º

60º

60º

90º 90º 120º

120º

150º

150º posterior Figure 1A4-2  Radial collagen fiber bundles of the meniscus. Radial tie fibers consisting of branching bundles of collagen fibrils extend from the periphery of the meniscus to the inner rim in every radial section throughout the meniscus. They are more abundant in the posterior sections and gradually diminish in number as the sections progress toward the anterior region of the meniscus.   (Redrawn from Kelly MA, Fithian DC, Chern KY, Mow VC: Structure and function of meniscus: Basic and clinical implications.   In Mow VC, Ratcliffe A, Woo SL [eds]: Biomechanics of Diarthrodial Joints, vol 1. New York, Springer-Verlag, 1990, pp 191-211.)

Radial sections of meniscus (Fig. 1A4-2) show radially oriented bundles of collagen fibrils, or “radial tie fibers,” among the circumferential collagen fibril bundles, weaving from the periphery of the meniscus to the inner region.78,85,97 Their prevalence and the tensile properties of specimens prepared from radial sections argue that these radial tie fibers form sheaths that secure the loosely arranged large circumferential collagen bundles.78,97 Presumably, they help to increase the stiffness and the strength of the tissue in a radial direction,97 thereby resisting longitudinal splitting of the collagen framework. Finite-element analyses of meniscal function show that large radial stresses and strains develop in the central region of the tissue that may be capable of producing longitudinal splits similar to bucket handle tears.98-100 In cross section, these radial tie fibers appear to be more abundant in the middle and the posterior sections than in the anterior sections of the meniscus. This arrangement may be related to meniscal function because most loads of tibiofemoral articulation are transmitted through the middle and the posterior portions of the meniscus.

MECHANICAL PROPERTIES Mechanical functions of the meniscus include distributing loads over a broad area of articular cartilage, absorbing shock during dynamic loading, improving joint stability, and possibly participating in joint lubrication.* These functions depend on the solid phase of the meniscal matrix (primarily the two major matrix macromolecules, collagens and proteoglycans) and on the tissue fluid. In biomechanical terms, the collagen network and the proteoglycans form a cohesive porous-permeable solid matrix.4,81 The interaction between the tissue fluid and the macromolecular solid matrix and, in particular, the flow of water through the framework make important contributions to the mechanical properties of the tissue. *See

references 4,5,13-31,33,34,38,43,44,50,66,75,80,81,89,101-104

Figure 1A4-3 shows the load-carrying mechanisms for tendons and ligaments, articular cartilage, and meniscus. These three tissues have strong viscoelastic tendencies, manifested by creep and stress relaxation behavior in response to loading and deformation. Creep is the increasing deformation that occurs over time when a viscoelastic material or structure is subjected to a constant load, whereas stress relaxation is the decreasing stress that occurs over time when a viscoelastic material or structure is subjected to a constant deformation. Creep and stress relaxation depend on the viscoelastic properties of the macromolecules and their organization within the individual tissues. Tendons and ligaments are linear fibrous structures that transmit loads along the length of their fibers. Type I collagen fibers predominate in these materials, and they form large undulating fiber bundles with high levels of tensile stiffness and strength.105-107 The viscoelastic behavior of tendons and ligaments results from the uncoiling of the collagen fibrils within the substance of these tissues106,107 and has significant implications for warm-up and stretching exercises before participation in sports. When tendons and ligaments are loaded in tension, they creep, thus loosening the joint; when tendons and ligaments are held at a constant stretch, less load is required to maintain the stretched position with time, resulting in a less stiff joint. The loading pattern on articular cartilage is complex, both temporally and spatially. At the articular surface, large tensile stresses are produced; hence, there is a strong collagen-rich surface zone.90 In the middle zone, tensile stresses may be produced in any direction, depending on the loading pattern and motion; hence, a randomly distributed collagen network in this zone would seem to function best. In the deep zone,88 tensile and shear stresses are produced; hence, theoretically, a radial fiber orientation would be optimal.50,108 High compressive stresses are also developed within articular cartilage owing to joint loading. They are resisted by both physicochemical and mechanical forces. Articular cartilage also exhibits pronounced viscoelastic effects. This viscoelasticity is mainly due to the high frictional drag exerted on the microporous matrix by

60

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 1A4-3  Direction of primary loading for three major types of connective tissue (tendon and ligament, articular cartilage, and meniscus).   (Redrawn from Mow VC, Fithian DC, Kelly MA: Fundamentals of articular cartilage and meniscus biomechanics. In Ewing JW [ed]: Articular Cartilage and Knee Joint Function:   Basic Science and Arthroscopy. New York,   Raven Press, 1990, pp 1-78.)

Compression

Meniscus

Compression

Tension

Tendon and ligament Tension

interstitial fluid flow.50,81 Very high interstitial pressures are required to force fluid to flow through the permeable solid matrix. When cartilage is suddenly loaded, the interstitial hydraulic pressure is largely responsible for its initial compressive stiffness. With time, creep occurs as the interstitial fluid flows away from the high-pressure regions and exudes from the tissue. This process slowly relieves the high hydraulic pressure, and the load acting on the tissue is slowly transferred to the solid matrix. Load sharing (ratio of loads carried between the fluid pressure and the solid matrix stress) during normal physiologic function is estimated to be 22:1. Thus, in articular cartilage, viscoelasticity is mainly due to interstitial fluid flow, whereas in tendons and ligaments, viscoelasticity is mainly due to movement of matrix macromolecules, primarily fibrillar collagen. The material properties of meniscal tissue differ from those of tendons and ligaments and from those of articular cartilage.70,81,108 The meniscus is loaded in compression in a direction perpendicular to the predominant collagen fiber direction (see Fig. 1A4-3). Because of the triangular shape of the meniscal tissue and its location along the periphery of the tibiofemoral articulation, the compressive force tends to extrude the meniscus outward toward the joint margins. A high tensile stress, often referred to as a hoop stress (a term derived from the hoops of a barrel), must be developed in the circumferential collagen fibers of the tissue to resist this extrusion effect.24 Thus, the geometric configuration of the meniscus and its nearly frictionless articulations with the femoral and the tibial surfaces provide an efficient mechanism for converting the compressive loadings in the knee into tensile loads running parallel to the circumferentially arranged, strong collagen fibers. In this sense, the meniscus behaves as ligament. Recent studies of the tensile stiffness and the strength of bovine and human menisci have shown that meniscal tissue

Articular cartilage

is anisotropic and inhomogeneous.4,70,95,96 Figure 1A4-4A shows that for the bovine medial meniscus, the posterior specimens are significantly stiffer in tension than the anterior specimens, except at the surface. Figure 1A4-4B shows that for both posterior and anterior specimens, the circumferential specimens are stiffer in tension than the radial specimens, except at the surface.4,95,96 The surface of meniscal tissue is more than 5 times stiffer than the surface zone of articular cartilage.70 In general, the tensile stiffness of meniscal specimens harvested from the circumferential direction may be as much as 100 times greater than the stiffness of specimens obtained from the radial direction. The central and the posterior parts of human medial meniscus have less tensile stiffness than the anterior part of the medial meniscus and all parts of the lateral meniscus (Fig. 1A4-5).70 The posterior and circumferential meniscal deep-zone tissues are both more than 20 times stiffer than articular cartilage.70,81,108 This high tensile stiffness and strength is consistent with the ligament-like function of the meniscus. Along with the large hoop stresses that can develop along the circumferential fibers as a result of extrusive forces on the meniscus, finite-element analyses have predicted large radial stresses and strains deep within the midsection of the meniscus.99,100 Such forces can be resisted only by the relatively weak radial tie fibers.4,97 These analyses suggest that longitudinal lesions similar to bucket handle tears of the meniscus are caused by these high-magnitude radial stresses and strains. In compression, meniscus closely resembles articular cartilage, except that the high Donnan osmotic swelling pressure component is not present owing to the low concentration of proteoglycans. The collagen-proteoglycan framework does form a porous-permeable solid matrix, and, like articular cartilage, meniscus exhibits pronounced biphasic viscoelastic effects in compression, mainly due to the high frictional drag exerted on the microporous

Basic Science and Injury of Muscle, Tendon, and Ligament

200

300

posterior

Modulus (MPa)

Modulus (MPa)

300

200

circumferential

100

100

radial

anterior 0

61

1 (surface)

A

3

0

5 (deep)

B

Slice

1 (surface)

3

5 (deep)

Slice

Figure 1A4-4  Demonstration that menisci exhibit inhomogeneous and anisotropic tensile stiffness. A, Tensile stiffness of bovine medial meniscus varies between anatomic regions (posterior and anterior) and among zones (depth). This type of variation indicates that the meniscus has inhomogeneous material properties. B, Tensile stiffness of bovine medial meniscus varies with direction (circumferential or radial). This type of variation indicates that the meniscus has anisotropic material properties.   (A and B, From   Proctor CS, Schmidt MB, Whipple RR, et al: Material properties of the normal medial bovine meniscus. J Orthop Res 7:771-782, 1989.)

collagen-proteoglycan matrix by interstitial fluid flow.4,81 In fact, even higher interstitial pressures are required; the permeability of the porous-permeable meniscal solid matrix is one sixth that of articular cartilage,4,81,108 which means that the frictional drag of fluid flow through meniscus is 6 times higher than that of articular cartilage. Meniscal tissue also has a compressive stiffness only one half that of articular cartilage,4,81,108 so when the soft meniscal tissue is loaded in compression, it can be deformed easily, forcing interstitial fluid flow. Because of the high frictional drag, a large amount of energy per unit volume of tissue is dissipated when the tissue is compressed. By virtue of its relative mass, the meniscus provides an excellent energy-absorption mechanism for the knee, damping out shocks experienced

Anterior

159.58 ±26.2 n=7

93.18 ±52.4 n=8

L

M

110.23 ±40.7 n=7

159.07 ±47.4 n = 10

228.79 ±51.4 n = 11

294.14 ±90.4 n=7 Figure 1A4-5  Variation in circumferential tensile stiffness of human medial (M) and lateral (L) menisci. Note that the central and posterior parts of the medial meniscus are weakest, as measured by tensile stiffness.   (From Fithian DC, Kelly MA, Mow VC: Material properties and structure-function relationships in the menisci. Clin Orthop 252:19-31, 1990.) Posterior

during normal daily activities and sports.27 These properties may help protect the cartilage and subchondral bone from being damaged by excessive physiologic loads. The ultrastructure of the meniscus also provides unique shear properties. The large type I collagen fiber bundles are held together by the radial tie sheaths very much like timber stacks in lumberyards. When the meniscus is sheared in planes containing the circumferential collagen fibers (Fig. 1A4-6, inset), resistance to shear is provided only by the sparse radial ties (as shown). Consequently, the shear modulus of the meniscus is very low. In addition, the shear modulus decreases with increasing shear strain (see Fig. 1A4-6), in contrast to other connective tissues. This characteristic means that meniscal tissues can easily conform to the anatomic forms of the mating femoral and tibial articulating surfaces in the axial plane, which may be an important consideration in selecting meniscal allografts. Meniscal shear properties span a broad range and occupy a unique place in the spectrum of connective tissue mechanical properties. The function and the tensile properties of meniscus are very similar to those of tendons and ligaments, and the function and the material properties of meniscus in compression are very similar to those of articular cartilage. Figure 1A4-7 provides a comparison of the mechanical properties of articular cartilage, meniscus, and ligament in tension, ­compression, and shear.

BLOOD SUPPLY Branches from the geniculate arteries form a capillary plexus along the peripheral borders of the menisci.1,35,36,38 Small radial branches project from these circumferential parameniscal vessels into the meniscal substance.37 During the fetal period, blood vessels have been observed throughout the meniscus, with the greatest number occurring in

62

DeLee & Drez’s Orthopaedic Sports Medicine

0.10 |G*| (MPa)

Ligament tensile modulus

Meniscus

0.12 100

0.08

13%

0.06 10% 0.04 7% 0.02 0.00

A.C. shear modulus

Proteoglycan shear modulus

0.14

0.01

0.02

0.03

0.04

0.05

101

104

105

106

107

108

A.C. and meniscus compressive modulus meniscus A.C. tensile moduli Figure 1A4-7  Comparison of the material properties of articular cartilage (A.C.), meniscus, and ligament in terms of tension, compression, and shear. Note that meniscus differs from both articular cartilage and ligament.

0.06

Shear Strain Figure 1A4-6  The shear stiffness of meniscal specimens containing circumferential collagen fibers (inset) decreases with increasing shear strain. This is known as a shear-softening effect; it enhances the ability of the tissue to conform to its mating articulating surfaces. Note that shear stiffness increases with increasing compressive strain (7%, 10%, and 13%). This behavior promotes knee joint stability.

the peripheral one third of the tissue. From birth through adolescence, the densities of meniscal cells and blood vessels decrease. In the adult, the penetrating vessels extend from the periphery into only 10% to 30% of the medial meniscus and 10% to 25% of the lateral meniscus, leaving the more central regions without blood vessels.

NERVE SUPPLY Nerves enter the joint capsule, the knee ligaments, and the periphery of the menisci and the meniscal horns but do not enter the central regions of the menisci.1,39-41 As with other intra-articular connective tissues, the functions of nerve endings in these tissues have not been clearly defined, but meniscal nerves may contribute to joint proprioception.39,41

INJURY AND REPAIR Traumatic meniscal tears occur most frequently in young, active people. A sudden change in direction while running, forceful squatting, twisting the knee, or application of external forces to the knee (e.g., rotation, varus, valgus, or hyperextension) subjects the meniscus to tension, compression, and shear. Tension, compression, or shear forces that exceed the strength of the meniscal matrix in any direction (i.e., circumferential or radial) tear the tissue. Acute traumatic injuries of normal meniscal substance usually produce longitudinal or transverse tears, although the morphology of tears can be quite complex.98 The configuration of tears due to overloading of normal meniscal tissue depends strongly on the direction and the rate of stretch.4 Unlike acute traumatic tears through apparently normal meniscal tissue, degenerative meniscal tears occur in association with age-related changes in the tissue. These degenerative tears are most common in individuals older than 40 years. Often, these individuals do not recall a specific injury, or they recall only a minor load applied to the knee.

Degenerative tears often have complex shapes or may appear as horizontal clefts or flaps, as though they were produced by shear failure (see Fig. 1A4-6, inset). Multiple degenerative tears often occur within the same meniscus. These features of degenerative meniscal tears suggest that they result more from age-related changes in the collagenproteoglycan solid matrix than from specific acute trauma. The response of meniscal tissue to tears depends on whether the tear occurs through a vascular or an avascular portion of the meniscus.1,36 The vascular regions respond to injury as other vascularized dense fibrous tissues do. The tissue damage initiates a sequence of cellular and vascular events recognized as inflammation, repair, and remodeling109 that can result in healing and restoration of tissue structure and function. The avascular regions of meniscal tissue, like articular cartilage, cannot repair significant tissue defects.

Repair in Vascular Regions of the Meniscus A tear in the vascular regions of the meniscus damages blood vessels, causing hemorrhage and fibrin clot formation. Injury to meniscal cells and clot formation activate inflammation. Platelets within the clot and inflammatory cells release mediators that stimulate cell migration, proliferation, and differentiation.109 The response of meniscal cells to injury and their contribution to repair of vascularized regions of the meniscus are uncertain. As in other vascularized dense fibrous tissues (e.g., tendon and ligament), mesenchymal cells and vascular buds invade the fibrin clot. If the injury site is sufficiently stable to allow repair, granulation tissue replaces the fibrin clot, and vessels soon cross the site of the defect. Repeated loading and motion early during repair may disrupt the immature repair tissue and prevent healing. Successful repair replaces damaged tissue with new tissue consisting of a high concentration of fibroblasts and small blood vessels surrounded by a poorly organized matrix. Following successful repair, the newly formed tissue begins to remodel.109 Cell density and vascularity decline, and excess tissue is resorbed. The collagen fibrils at the injury site assume a higher degree of orientation. Loading of the meniscus presumably influences remodeling of meniscal repair tissue, in the same manner that loading appears to affect collagen fibril organization in immature meniscal

Basic Science and Injury of Muscle, Tendon, and Ligament

tissue.37 Repair and subsequent remodeling of the repair tissue in the vascular regions of the meniscus can restore the structural integrity of the tissue and presumably many of its material properties. Unfortunately, very little basic science information exists on the composition, remodeling, structure, and material properties of meniscal repair tissue.

Meniscal Regeneration Partial meniscal resection through the peripheral vascularized region or complete meniscal resection initiates production of repair tissue that extends from the remaining peripheral tissue into the joint.3,102-104,110-113 Although the repair cells usually fail to replicate normal meniscal tissue, many authors have referred to this phenomenon as meniscal regeneration.3,110,112 Some repaired menisci grossly resemble normal menisci, but the functional capabilities and mechanical properties of this “regenerated” meniscal tissue have not been studied. Surgeons have reported meniscal regeneration in many clinical situations. Investigators have also examined the tissue produced by meniscal regeneration in animals. The mechanisms and the conditions that promote this type of repair and its functional importance, however, remain poorly understood. Meniscal regeneration can occur repeatedly in the same knee111 and occasionally occurs following total knee replacement.110 In rabbits, meniscal regeneration occurs more frequently on the medial side of the knee than on the lateral side, and development of degenerative changes in articular cartilage following meniscectomy is inversely correlated with the extent of meniscal regeneration.102-104 Synovectomy appears to prevent meniscal regeneration, suggesting that synovial cells contribute to formation of meniscal repair tissue. The factors related to the predictability and frequency of meniscal regeneration remain unknown.

Repair in Avascular Portions of the Meniscus The response of meniscal tissue to tears in the avascular portion resembles the response of articular cartilage to lacerations in many respects.109 Experimental studies show that a penetrating injury to the avascular region of the meniscus causes no apparent repair or inflammatory reaction. Meniscal cells in the injured region, like chondrocytes in the region of an injury limited to the articular cartilage, may proliferate and synthesize new matrix, but appear incapable of migrating to the site of the defect or producing enough new matrix to fill it.

Improving Repair in the Avascular Portion of the Meniscus The ineffective response of meniscal cells in the avascular region of the meniscus has led investigators to develop several methods to stimulate repair. Some promising approaches include creation of a vascular access channel to the injury site and stimulation of cell migration to the avascular region using implantation of a fibrin clot, an artificial matrix, or growth factors.1,58 Creation of full-thickness vascular channels from the periphery of the meniscus to experimentally created bucket

63

handle–like lesions in the avascular portion stimulated healing. Blood vessels and presumably mesenchymal cells migrated through these channels and healed the meniscal lesions. Sheep and rabbit studies have shown that this approach may compromise the biomechanical function of the meniscus by destroying the integrity of the peripheral meniscal rim and the circumferential collagen fiber bundles. To avoid this problem, some investigators have used trephines to create vascular tunnels that connect the vascular peripheral portion of the meniscus to a lesion in the avascular region.1 The synovium can provide a source of vessels and cells for meniscal healing. Suturing a flap of vascular synovial tissue into a longitudinal lesion in the avascular portion of the meniscus brings a blood supply and new cells to the injury site.1 Synovial abrasion stimulates proliferation of the synovial fringe into the meniscus and allows blood vessels to enter the avascular regions. Although early results appear promising, the quality of the repair tissue, its biomechanical properties, and the long-term results of these methods have not been evaluated. Investigators have also attempted to stimulate meniscal repair without participation of blood vessels. This approach assumes that either the meniscal cells can repair defects in the avascular meniscal regions if they are appropriately stimulated or mesenchymal cells from the vascular region can migrate into the avascular region. Several experimental studies have shown that cultured meniscal fibrochondrocytes proliferate and synthesize matrix when exposed to chemotactic and mitogenic factors that are found in wound hematomas.60,61 These nonvascular methods include implantation of a fibrin clot, which presumably contains platelet-derived growth factor, and the use of other growth factors that stimulate mesenchymal cell migration, proliferation, and matrix synthesis. A fibrin clot may act as scaffolding for the migration of cells and could serve as a vehicle for implantation of specific growth factors. Implanting an exogenous fibrin clot in defects in the avascular regions of dog menisci stimulates proliferation of fibrous connective tissue that eventually assumes the appearance of fibrocartilaginous tissue, although it differed from normal meniscal tissue histologically and grossly.54 The source of the repair cells was not identified, but they may have arisen from adjacent synovium and from meniscal tissue. Clinical experience with injection of fibrin clots into meniscal defects also suggests that these clots stimulate repair. In one study, 92% of isolated meniscal tears treated with fibrin clots healed, compared with 59% of isolated meniscal tears treated without fibrin clots.58 Implantation of a synthetic matrix,52 possibly containing growth factors, might also stimulate repair. A recent study using a rat model has shown that implantation of bone marrow–derived mesenchymal stem cells in a fibrin glue can contribute to meniscal healing in the avascular zone by proliferating and producing extracellular matrix for up to 8 weeks following transplantation.114

Meniscal Grafts To prevent articular cartilage degeneration that may result from loss of a meniscus,14,16,26,43-45,102 many surgeons currently attempt to preserve or repair menisci whenever

64

DeLee & Drez’s Orthopaedic Sports Medicine

possible; it is not always possible. Intrinsic meniscal repair in the avascular portions of the meniscus is limited, and some injuries or degenerative diseases preclude successful repair even in the vascularized regions of the meniscus. In these circumstances, partial or total meniscectomy may be necessary. Several approaches to meniscal tissue replacement to protect articular cartilage from degeneration after meniscectomy have been investigated.

Meniscal Allografts Experimental studies have shown that allograft menisci will heal to host tissues. Transplanted cryopreserved dog allograft menisci caused no apparent rejection reactions, healed to the host tissues, and appeared to function normally after transplantation.56,57 Transplantation of lyophilized and deep-frozen menisci in sheep also showed that grafts can heal to the recipient site tissues.62 Studies of dog meniscal allografts55,56 showed that cryopreservation and short-term storage did not alter the shape or the mechanical properties of the menisci. These studies also showed, however, that only about 10% of the meniscal cells remained metabolically active, meniscal cell synthetic activity decreased to less than 50% of normal, and total metabolic activity of the tissue declined with increasing storage time. Experimental studies have shown that meniscal grafts remodel and may help decrease the probability of degenerative joint disease following removal of the original meniscus. With time, the cell density of dog meniscal grafts increased; 3 months after transplantation, the cell density and level of metabolic activity was similar to normal menisci. Six months following transplantation, small blood vessels penetrated the peripheral one third of the grafts. The articular cartilage underlying the menisci remained intact 6 months following transplantation, whereas exposed tibial cartilage had fissures and degenerative changes. These changes appeared less severe than those found in dog knees subjected to total meniscectomy,42 suggesting that the allografts provide some protection for the articular cartilage. The long-term results of meniscal allografts and the efficacy of allografts in decreasing the probability or the severity of degenerative joint disease remain unknown.59 A few clinical studies confirm that meniscal allografts heal with host tissues.62,115 One group of investigators using fresh meniscal allografts found that “most” of the grafts appeared structurally sound and functional at followup and reported survival of grafts for as long as 8½ years; no biomechanical tests were performed to assess their mechanical properties.115 Another group using lyophilized and deep-frozen allografts reported that operative complications and rejection reactions did not occur but that over time the grafts decreased in size.62 Although the potential for allograft meniscal replacement exists, the effects of meniscal allografts on the long-term frequency and severity of degenerative joint disease have not been determined.

Synthetic Matrix Meniscal Grafts Synthetic matrices, created from reconstituted collagen, fibrin, or other materials and shaped to fit specific meniscal defects, can replace lost or damaged meniscal tissue.52,59

C

r i t i c a l

P

o i n t s

l The

composition, ultrastructure, and mechanical properties of the knee meniscus allow it to serve as a load-bearing and shock-absorbing structure. In addition, the menisci contribute to the stability of the knee joint and may aid in joint lubrication. Loss of one or both menisci alters the loading of articular cartilage and increases the risk for degenerative joint disease. l Some regions of adult meniscus, such as the meniscal horns, have blood vessels and a nerve supply. Injuries in the vascular regions of the meniscus can heal, restoring the structural integrity of the tissue. Injuries in the avascular regions do not have a meaningful healing response. l Surgical interventions preserve the menisci when possible to prevent subsequent degeneration of articular cartilage. l Several promising treatment modalities are currently undergoing review, including meniscal allografts and synthetic matrices (sometimes with the addition of growth factors and mesenchymal cells), which can be shaped to fit specific meniscal defects.

Recipient-site cells and blood vessels might grow into these synthetic matrices and remodel them to resemble meniscal tissue. Initial experimental investigations suggest that synthetic collagen matrices may have the potential to replace menisci.52,59 These synthetic matrices, or polyurethane116 or biodegradable scaffolds,117 could also serve as vehicles for implantation of growth factors or cultured mesenchymal cells.

CONCLUSIONS The knee meniscus is a specialized intra-articular fibrocartilaginous structure of the knee. Its composition, ultrastructure, and mechanical properties support its function as a load-bearing and shock-absorbing structure. Along with the knee ligaments, the menisci help stabilize normal knees, and they may also assist in lubrication of the knee. Unlike articular cartilage, some regions of adult meniscus have blood vessels and a nerve supply, especially the meniscal horns. Loss of one or both menisci alters the loading of articular cartilage and increases the probability and the severity of subsequent degenerative joint disease. Tears through the vascular regions of the meniscus can heal, but tears through the avascular regions do not undergo a significant repair process. Repair and subsequent remodeling of the repair tissue in the vascular regions of the meniscus can restore the structural integrity of the tissue. To prevent degeneration of articular cartilage, surgeons preserve or repair the menisci when possible. Initial studies of meniscal allografts show that these grafts can heal to the host tissues and may help restore meniscal function. Synthetic matrices, created from reconstituted collagen, fibrin, and other materials and shaped to fit specific meniscal defects, also have shown promise as a method of replacing severely damaged or absent menisci. The addition of growth factors or mesenchymal cells to these synthetic matrices may improve results.

Basic Science and Injury of Muscle, Tendon, and Ligament

S U G G E S T E D

R E A D I N G S

Adachi N, Ochi M, Deie M, et al: Lateral compartment osteoarthritis of the knee after meniscectomy treated by the transplantation of tissue-engineered cartilage and osteochondral plug. Arthroscopy 22:107-112, 2006. Adams SB Jr, Randolph MA, Gill TJ: Tissue engineering for meniscus repair. J Knee Surg 18:25-30, 2005. Arnoczky SP, Adams M, DeHaven K, et al: Meniscus. In Woo SL, Buckwalter JA (eds): Injury and Repair of the Musculoskeletal Soft Tissues. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1988, pp 487-537. de Groot JH: Polyurethane scaffolds for meniscal tissue regeneration. Med Device Technol 16:18-20, 2005. Izuta Y, Ochi M, Adachi N, et al: Meniscal repair using bone marrow-derived mesenchymal stem cells: Experimental study using green fluorescent protein transgenic rats. Knee 12:217-223, 2005. McDevitt CA, Webber RJ: The ultrastructure and biochemistry of meniscal cartilage. Clin Orthop 252:8-18, 1990.

65

Moore EE, Bendele AM, Thompson DL, et al: Fibroblast growth factor-18 ­stimulates chondrogenesis and cartilage repair in a rat model of injury-induced osteoarthritis. Osteoarthritis Cartilage 13:623-631, 2005. Shoemaker SC, Markolf KL: The role of the meniscus in the anterior-posterior stability of the loaded anterior cruciate–deficient knee. J Bone Joint Surg Am 68: 71-79, 1986. Tuckman DV, Bravman JT, Lee SS, et al: Outcomes of meniscal repair: Minimum of 2-year follow-up. Bull Hosp Jt Dis 63:100-104, 2006. Zimny ML, Albright DH, Dabezies EJ: Mechanoreceptors in the human medial meniscus. Acta Anat 133:35-40, 1988.

R eferences Please see www.expertconsult.com

S ecti o n

A

Physiology of Injury to Musculoskeletal Structures 5. Bone Injury Mark R. Brinker, Daniel P. O’Connor, Louis C. Almekinders, Thomas M. Best, Joseph A. Buckwalter, William E. Garrett, Jr., Donald T. Kirkendall, Van C. Mow, and Savio L.-Y. Woo

HISTOLOGY OF BONE Types of Bone Normal bone is lamellar and can be cortical or cancellous. Immature bone and pathologic bone are woven and, in contrast to lamellar bone, are more random with more osteocytes, have increased turnover, and are weaker and more flexible. Lamellar bone is stress oriented; woven bone is not stress oriented. Cortical bone (compact bone) (Fig. 1A5-1; Table 1A5-1) makes up 80% of the skeleton and is composed of tightly packed osteons or haversian systems that are connected by haversian (or Volkmann’s) canals. These canals contain arterioles, venules, capillaries, nerves, and possibly lymphatic channels. Interstitial lamellae lie between the osteons. Fibrils frequently connect lamellae but do not cross cement lines (where bone resorption has stopped and new bone formation has begun). Cement lines define the outer border of an osteon. Nutrition is through the intraosseous circulation (canals and canaliculi [cell processes of osteocytes]). Cortical bone is characterized by a slow turnover rate, a relatively high Young’s modulus (E), and a high ­resistance

to torsion and bending. Cancellous bone (spongy or ­trabecular bone) (see Fig. 1A5-1) is less dense than cortical bone and undergoes more remodeling according to lines of stress (Wolff’s law). It has a higher turnover rate, has a smaller Young’s modulus, and is more elastic than cortical bone.

Cellular Biology Osteoblasts are responsible for bone formation and are derived from undifferentiated mesenchymal cells. These cells have more endoplasmic reticulum, Golgi apparatus, and mitochondria than other cells (to fulfill the cell’s role in the synthesis and secretion of matrix). More differentiated, metabolically active cells line bone surfaces, and less active cells in “resting regions” or entrapped cells maintain the ionic milieu of bone. Disruption of the lining cell layer activates these cells. Osteoblast differentiation in vivo is effected by the interleukins, plateletderived growth factor, and insulin-derived growth factor. Osteoblasts show high alkaline phosphatase activity, produce type I collagen, respond to parathyroid hormone (PTH), and produce osteocalcin [stimulated by 1,25(OH)2

66

DeLee & Drez’s Orthopaedic Sports Medicine

Cortical

Immature

Cancellous

Pathologic (giant cell tumor)

Haversian canal Cement line

Interstitial lamellae Canaliculi

Osteocyte

CORTICAL BONE DETAIL Figure 1A5-1  Types of bone. Cortical bone consists of tightly packed osteons. Cancellous bone consists of a meshwork of trabeculae. In immature bone, there is unmineralized osteoid lining the immature trabeculae. In pathologic bone, atypical osteoblasts and architectural disorganization are seen.���   (From Brinker MR, Miller MD: Fundamentals of Orthopaedics. Philadelphia, WB Saunders, 1999, p 2.)

vitamin D]. Osteoblasts have receptor-­effector interactions for (1) PTH, (2) 1,25(OH)2 vitamin D, (3) glucocorticoids,1 (4) prostaglandins, and (5) estrogen (Table 1A5-2).2 Certain antiseptic agents used as an irrigation solution, including hydrogen peroxide and povidoneiodine (Betadine), have been shown to be toxic to cultured osteoblasts3; bacitracin is generally believed to be less toxic to osteoblasts. Osteocytes (see Fig. 1A5-1) make up 90% of the cells in the mature skeleton and serve to maintain bone. These cells represent former osteoblasts that have been trapped within newly formed matrix, which they help preserve. Osteocytes have an increased nucleus-to-cytoplasm ratio with long interconnecting cytoplasmic processes and are not as active in matrix production as are osteoblasts. Osteocytes have an important role in controlling the extracellular concentration of calcium and phosphorus; they are directly stimulated by calcitonin and inhibited by PTH.

Osteoclasts are responsible for bone resorption.4 These multinucleated, irregularly shaped giant cells originate from hematopoietic tissues5 (monocyte progenitors form giant cells by fusion) and possess a ruffled (“brush”) border (plasma membrane infoldings that increase the surface area and are important in bone resorption) and a surrounding clear zone. Bone resorption occurs in depressions known as Howship’s lacunae and is more rapid than bone formation; bone formation and resorption are linked (“coupled”). Osteoclasts synthesize tartrate-resistant acid phosphate. Osteoclasts bind to bone surfaces by cell attachment (anchoring) proteins (integrins). The integrin attachment to bone effectively seals the space below the osteoclast. Osteoclasts produce hydrogen ions (through carbonic anhydrase) to lower the pH, which increases the solubility of hydroxyapatite crystals, and the organic matrix is removed by proteolytic digestion. Patients who are deficient in carbonic anhydrase cannot resorb bone

Basic Science and Injury of Muscle, Tendon, and Ligament

  TABLE 1A5-1  Types of Bone Microscopic Appearance Subtype Lamellar

Cortical

Woven

Immature

Characteristics

Examples

Structure is oriented along lines of stress Strong More elastic than cortical bone

Femoral shaft

Not stress oriented Pathologic

Random organization Increased turnover Weak Flexible

Distal femoral metaphysis Fracture callus Embryonic skeleton Osteogenic sarcoma Fibrous dysplasia

Modified from Brinker MR, Miller MD: Fundamentals of Orthopaedics. ­Philadelphia, WB Saunders, 1999, p l.

by this mechanism. Osteoclasts have specific receptors for calcitonin to allow them to directly regulate bone resorption (see Table 1A5-2). Osteoclasts are responsible for the bone resorption seen in multiple myeloma and metastatic bone disease. Interleukin-1 (IL-1) is a potent stimulator of osteoclastic bone resorption and has been found in the membranes surrounding loose total joint implants. IL-10 suppresses osteoclast formation.

Bone Matrix Bone matrix is composed of both organic and inorganic components. The organic components make up 40% of the dry weight of bone. Organic components include   TABLE 1A5-2  Bone Cell Types, Receptor Types, and Effects Cell Type

Receptor

Osteoblast PTH

1,25(OH)2 vitamin D3 Glucocorticoid Prostaglandin Estrogen

Osteoclast

Calcitonin

Effect Releases a secondary messenger (exact mechanism unknown) to stimulate osteoclastic activity. ­Activates adenylate cyclase Stimulates matrix and alkaline phosphatase synthesis and ­production of bone-specific proteins Inhibits the synthesis of DNA, ­production of collagen, and ­synthesis of osteoblastic proteins Activates adenylate cyclase and stimulates resorption of bone Anabolic (bone production) and ­anticatabolic (prevents bone ­resorption) effects on bone. ­Increases the levels of messenger RNA for alkaline phosphatase and inhibits the activation of adenylate cyclase Inhibits the function of osteoclasts (inhibits bone resorption)

PTH, parathyroid hormone. From Brinker MR: Basic sciences. In Miller MD, Brinker MR (eds): Review of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 2000, p 3.

67

collagen, proteoglycans, noncollagenous matrix proteins (glycoproteins, phospholipids, phosphoproteins), and growth factors and cytokines. Collagen is responsible for the tensile strength of bone. The principal type of collagen in bone is type I. The collagen structure consists of a triple helix of tropocollagen (two α1- and one α2-chains) that is quarter-staggered to produce a collagen fibril. Hole zones (gaps) exist within the collagen fibril between the ends of molecules. Pores exist between the sides of parallel molecules. Mineral deposition (calcification) occurs within these hole zones and pores (Fig. 1A5-2). Cross-linking decreases solubility and increases the tensile strength of collagen. Proteoglycans are partially responsible for the compressive strength of bone. Proteoglycans are composed of glycosaminoglycan-protein complexes. Noncollagenous matrix proteins promote mineralization and bone formation. Matrix proteins include osteocalcin,6 osteonectin, osteopontin, and others. Osteocalcin is produced by osteoblasts and is directly related to the regulation of bone density. Osteocalcin is the most abundant noncollagenous matrix protein of bone and accounts for 10% to 20% of the noncollagenous protein of bone. The synthesis of osteocalcin is inhibited by PTH and stimulated by 1,25-dihydroxyvitamin D. Osteocalcin levels can be measured in the serum or the urine as a marker of bone turnover. Osteocalcin levels in urine and serum are elevated in Paget’s disease, renal osteodystrophy, and hyperparathyroidism. Osteonectin (secreted by platelets and osteoblasts) is postulated to play a role in the regulation of calcium or the organization of mineral within the matrix. Osteopontin is a cell-binding protein (similar to an integrin). Growth factors and cytokines are present in small amounts in bone matrix. These include transforming growth factor-β (TGF-β); insulin-like growth factor; interleukins (IL-1, IL-6); and bone morphogenetic proteins (BMPs). These proteins aid in bone cell differentiation, activation, growth, and ­turnover. The inorganic or mineral component of bone matrix makes up 60% of the dry weight of bone. Calcium hydroxyapatite [Ca10(PO4)6(OH)2] is responsible for the compressive strength of bone. Calcium hydroxyapatite makes up most of the inorganic matrix and is responsible for matrix mineralization. Primary mineralization occurs in gaps in the collagen; secondary mineralization occurs on the periphery. Osteocalcium phosphate makes up the remaining portion of the inorganic matrix.

Bone Remodeling Bone remodeling is affected by mechanical stress according to Wolff’s law. Removal of external stresses can lead to significant bone loss, but this situation can be reversed to varying degrees on remobilization. In addition to remodeling in response to stress, bone remodels in response to piezoelectric charges. The compression side is electronegative, stimulating osteoblasts and bone formation; the tension side is electropositive, stimulating osteoclasts and bone resorption. Both cortical bone and cancellous bone are continuously remodeled by osteoclastic and osteoblastic activity. Bone remodeling occurs in small packets of cells known as basic multicellular units. This bone remodeling is modulated by

68

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 1A5-2  Microstructure of collagen. Collagen is composed of microfibrils that are packed in a quarter-staggered fashion (tropocollagen). Note hole and pore regions for mineral deposition (for calcification). Tropocollagen, in turn, is made up of a triple helix of α-chains of polypeptides.���   (From Brinker MR, Miller MD: Fundamentals of Orthopaedics. Philadelphia, WB Saunders, 1999, p 3.)

Cartilage

Bone

Ligament

Tendon

Microfibril Tropocollagen

Pore Hole

Triple helix

OH α chain

systemic hormones and local cytokines. Bone remodeling occurs throughout life. The Hueter-Volkmann law (compressive forces inhibit growth, and tensile forces stimulate growth) suggests that mechanical factors can influence longitudinal growth, bone remodeling, and fracture repair. This law is postulated to play a role in the progression of scoliosis and Blount disease. Cortical bone remodels by means of osteoclastic tunneling (cutting cones) (Fig. 1A5-3) followed by layering of

Figure 1A5-3  Mechanism of cortical bone remodeling via cutting cones.���   (From Simon SR [ed]: Orthopaedic Basic Science. Rosemont, Ill, American Academy of Orthopaedic Surgeons, 1994, p 142.)

osteoblasts and successive deposition of layers of lamellae (after the cement line has been laid down) until the tunnel size has narrowed to the diameter of the osteonal central canal. The head of the cutting cone is made up of osteoclasts, which bore holes through hard cortical bone. Behind the osteoclast front are capillaries, followed by osteoblasts, which lay down osteoid to fill the resorption cavity. Cancellous bone remodels by osteoclastic resorption followed by osteoblasts, which lay down new bone.

Basic Science and Injury of Muscle, Tendon, and Ligament

69

Figure 1A5-4  Intraoperative arteriogram (canine tibia), demonstrating ascending (A) and descending (D) branches of the nutrient artery. C, cannula.���   (From Brinker MR, Lipton HL, Cook SD, Hyman AL: Pharmacological regulation of the circulation of bone. J Bone Joint   Surg Am 72:964-975, 1990. Copyright The Journal of Bone and Joint Surgery, 1990.)

Bone Circulation As an organ, bone receives 5% to 10% of the cardiac output. The long bones receive blood from three sources: nutrient artery system, metaphyseal-epiphyseal system, and periosteal system.7 Bones with a tenuous blood supply include the scaphoid, the talus, the femoral head, and the odontoid. The nutrient artery originates as branches from major arteries of the systemic circulation. The nutrient artery enters the diaphyseal cortex (outer and inner tables) through the nutrient foramen and then enters the medullary canal. Once in the medullary canal, the nutrient artery branches into ascending and descending small arteries (Fig. 1A5-4), which branch into arterioles that penetrate the endosteal cortex to supply at least the inner two thirds of mature diaphyseal cortex through vessels that traverse the haversian system (Fig. 1A5-5). The nutrient artery system is a high-pressure system. The metaphyseal-epiphyseal system arises from the periarticular vascular plexus (e.g., geniculate arteries). The periosteal system is composed primarily of capillaries that supply the outer one third (at most) of the mature diaphyseal cortex. The periosteal system is a low-pressure system. Arterial flow in mature bone is centrifugal (inside to outside), a result of the net effect of the high-pressure nutrient arterial system (endosteal system) and the lowpressure periosteal system. In the case of a completely displaced fracture with complete disruption of the endosteal (nutrient) system, the pressure head is reversed, the periosteal system predominates, and blood flow is centripetal (outside to inside). Arterial flow in immature developing bone is centripetal because the periosteum is highly vascularized and is the predominant component of bone blood flow. Venous flow in mature bone is centripetal; cortical

capillaries drain to venous sinusoids, which drain in turn to the emissary venous system. Bone blood flow is responsible for the delivery of nutrients to the site of bony injury. The initial response is decreased flow to a fracture secondary to disruption of the vascular anatomy at the fracture site.8 Within hours to days, bone blood flow increases (as part of the regional acceleratory phenomenon) and peaks at about 2 weeks. Blood flow returns to normal between 3 and 5 months. Bone blood flow is a major determinant of fracture healing. Cortical bone blood flow is directly influenced by the method of fracture fixation (i.e., plate, intramedullary nail, or external fixation).9 Bone blood flow is under the control of metabolic, humoral, and autonomic inputs. The arterial system of bone has great potential for vasoconstriction (from the resting state) and much less potential for vasodilation. The vessels within bone possess a variety of vasoactive receptors (α-adrenergic, muscarinic, thromboxane/prostaglandin), which may be useful in the future for pharmacologic treatment of bone diseases related to aberrant circulation (e.g., osteonecrosis, fracture nonunion).10

Tissue Surrounding Bone The periosteum is the connective tissue membrane that covers bone. It is more highly developed in children because of its role in the deposition of cortical bone, which is responsible for growth in bone diameter. The inner, or cambium, layer of periosteum is loose and more vascular and contains cells that are capable of becoming osteoblasts (to form bone). These cells are responsible for enlarging the diameter of bone during growth11 and forming periosteal callus during fracture healing; the outer, fibrous layer is less cellular and is contiguous with joint capsules.

70

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 1A5-5  Blood supply to bone.   (From Brinker MR, Miller MD: Fundamentals of Orthopaedics. Philadelphia, WB Saunders, 1999, p 4.)

Nutrient artery Emissary vein

Bone marrow is a source of progenitor cells. Red marrow is hematopoietic and slowly changes to yellow marrow with age, beginning in the appendicular skeleton and occurring later in the axial skeleton. Yellow marrow is inactive and has a lower water content and higher fat content than red marrow.

NORMAL BONE MINERAL METABOLISM Calcium Bone serves as a reservoir for more than 99% of the body’s calcium. Calcium is also important in muscle and nerve function, the clotting mechanism, and many other areas. Plasma calcium (less than 1% of total body calcium) is about equally free and bound (usually to albumin). It is absorbed from the gut (duodenum) by active transport (adenosine triphosphate and calcium-binding protein required), which is regulated by 1,25(OH)2 vitamin D and by passive diffusion (jejunum). Calcium is 98% reabsorbed by the kidney (60% in the proximal tubule). The dietary requirement for elemental calcium12 is about 600 mg/day

for children, increasing to about 1300 mg/day for adolescents and young adults. The requirement for adult men and women (25 to 65 years of age) is 750 mg/day. Pregnant women require 1500 mg/day, and lactating women require 2000 mg/day. Postmenopausal women and patients with a healing long-bone fracture require 1500 mg/day. Most people have a positive calcium balance during the first three decades of life and a negative balance after the fourth decade. About 400 mg of calcium is released from bone on a daily basis. The primary ­homeostatic regulators of serum calcium are PTH and 1,25(OH)2 vitamin D.

Phosphate In addition to being a key component of bone mineral, phosphate has an important role in enzyme systems and molecular interactions (metabolite and buffer). About 85% of the body’s phosphate stores are in bone. Plasma phosphate is mostly in the unbound form and is reabsorbed by the kidney (in the proximal tubule). Dietary intake of phosphate is usually adequate; the daily requirement is 1000 to 1500 mg/day. Phosphate may be excreted in urine.

Basic Science and Injury of Muscle, Tendon, and Ligament

71

  TABLE 1A5-3  Regulation of Calcium and Phosphate Metabolism

Parameter

Parathyroid Hormone (PTH) (Peptide)

Origin

Chief cells of parathyroid glands

Factors promoting production

Decreased serum Ca2+

Factors inhibiting production

Elevated serum Ca2+ Elevated 1,25 (OH)2D

Effect on end organs for hormone action Intestine

Kidney

Bone

Net effect on Ca2+ and Pi concentrations in extracellular fluid and serum

No direct effect. Acts indirectly on bowel by stimulating production of 1,25 (OH)2D in kidney Stimulates 25(OH)D-1αOHase in mitochondria of proximal tubular cells to convert 25(OH)D to 1,25(OH)2D. Increases fractional resorption of filtered Ca2+ Promotes urinary excretion of Pi Stimulates osteoclastic resorption of bone Stimulates recruitment of ­ preosteoclasts Increased serum Ca2+ Decreased serum Pi

1,25 (OH)2 Vitamin D (Steroid)

Calcitonin (Peptide)

Proximal tubule of kidney Elevated PTH Decreased serum Ca2+ Decreased serum Pi Decreased PTH Elevated serum Ca2+ Elevated serum Pi

Parafollicular cells of thyroid gland

Strongly stimulates intestinal absorption of Ca2+ and Pi ?

?

Strongly stimulates osteoclastic resorption of bone Increased serum Ca2+ Increased serum Pi

Elevated serum Ca2+

Decreased serum Ca2+

?

Inhibits osteoclastic resorption of bone? Role in normal human physiology Decreased serum Ca2+ (transient)

1,25(OH)2D, 1,25-dihydroxyvitamin D; 25(OH)D, 25-hydroxyvitamin D; Pi, inorganic phosphate; PTH, parathyroid hormone. Adapted from an original figure by Frank H. Netter. From the Ciba Collection of Medical Illustrations. Vol 8, Part 1, p 179. Copyright by Ciba-Geigy Corporation.

Parathyroid Hormone PTH is an 84–amino acid peptide synthesized in and secreted from the chief cells of the (four) parathyroid glands. PTH helps regulate plasma calcium, directly activates osteoblasts, and modulates renal phosphate filtration. Decreased calcium levels in the extracellular fluid stimulate release of PTH, which acts at the intestine, kidney, and bone (Table 1A5-3). PTH may also have a role in bone loss in the elderly.

Vitamin D Vitamin D is a naturally occurring steroid that is activated by ultraviolet irradiation from sunlight or is used from dietary intake. It is hydroxylated to 25(OH) vitamin D in the liver and is hydroxylated a second time in the kidney.13

Conversion to 1,25(OH)2 vitamin D activates the ­hormone, whereas conversion to 24,25(OH)2 vitamin D inactivates it. The active form works at the intestine, kidney, and bone (see Table 1A5-3).

Calcitonin Calcitonin is a 32–amino acid peptide hormone made by the clear cells of the thyroid gland; this hormone also has a limited role in calcium regulation. Increased calcium levels in the extracellular milieu cause secretion of calcitonin. Like PTH, calcitonin secretion is controlled by a β2-­receptor. The main biologic effect of calcitonin is to inhibit osteoclastic bone resorption. (Osteoclasts have calcitonin receptors.) Calcitonin acts to decrease serum ­calcium levels. Calcitonin may also have a physiologic role in fracture healing and the treatment of osteoporosis.

72

DeLee & Drez’s Orthopaedic Sports Medicine

Other Hormones Estrogen prevents bone loss by inhibiting bone resorption. (A decrease in urinary pyridoline cross-links is observed.) Because bone formation and resorption are coupled, however, estrogen therapy also decreases bone formation. Supplementation is helpful in postmenopausal women but is most beneficial if it is started within the first 5 to 10 years after the onset of menopause. The risk for endometrial cancer for patients taking estrogen is reduced when it is combined with cyclic progestin therapy. Corticosteroids increase bone loss. (They decrease gut absorption of calcium by decreasing binding proteins, and they decrease bone formation [cancellous bone is more affected than cortical bone] through inhibition of collagen synthesis.) Adverse effects may be reduced with alternateday therapy. Thyroid hormones affect bone resorption more than bone formation, leading to osteoporosis. (Large [thyroid-suppressive] doses of thyroxine can lead to osteoporosis.) Growth hormone causes a positive calcium balance by increasing gut absorption of calcium more than its increase in urinary excretion. Insulin and somatomedins participate in this effect.

CONDITIONS OF DECREASED BONE MINERAL DENSITY Bone mass is regulated by the relative rates of deposition and withdrawal. Bone loss occurs at the onset of menopause, when there is both accelerated bone formation and accelerated resorption. Markers of bone resorption include urinary hydroxyproline and pyridoline cross-links. (When there is bone resorption, both of these are elevated.) Serum alkaline phosphatase is a marker for bone formation. (It is elevated when bone formation is increased.) Estrogen therapy for osteoporosis results in a decrease in urinary pyridoline (decreased bone resorption) and a decrease in serum alkaline phosphatase (decreased bone formation).

Osteoporosis Osteoporosis is an age-related decrease in bone mass ­usually associated with loss of estrogen in postmenopausal women.14 Osteoporosis is responsible for more than 1 million fractures per year (vertebral body most common). The lifetime risk for fracture in white women after 50 years of age is about 75%; the risk for hip fracture is 15% to 20%. Osteoporosis represents a quantitative, not a qualitative, defect in bone. Sedentary, thin white women of northern European descent, particularly smokers, heavy drinkers, and patients on phenytoin (which impairs vitamin D metabolism), with diets low in calcium and vitamin D, who breast-fed their infants, are at greatest risk. Cancellous bone is most markedly affected. Clinical features include kyphosis and vertebral fractures (compression fractures of T11–L1, which create an anterior wedge-shaped defect or result in a centrally depressed “codfish” vertebra), hip fractures, and distal radius fractures. Two types of osteoporosis have been characterized: type I (postmenopausal) and type II (age related). Type I osteoporosis (postmenopausal) affects primarily trabecular bone; vertebral and distal radius fractures

are common. Type II osteoporosis (age-related) is seen in patients older than 75 years of age, affects both trabecular and cortical bone, and is related to poor calcium absorption. Hip and pelvic fractures are common. Laboratory studies, including analysis of urinary calcium and hydroxyproline and serum alkaline phosphatase, are helpful in evaluating osteopenic conditions. Results of these laboratory studies are usually unremarkable in osteoporosis, but hyperthyroidism (with reversible osteoporosis), hyperparathyroidism, Cushing’s syndrome, hematologic disorders, and malignancy should be ruled out. Plain radiographs usually are not helpful unless greater than 30% bone loss is present. Special studies used for the work-up of osteoporosis include single-photon (appendicular) and double-photon (axial) absorptiometry, quantitative computed tomography, and dual-energy x-ray absorptiometry (DEXA). DEXA is most accurate with less radiation. A biopsy (after tetracycline labeling) may be performed to evaluate the severity of osteoporosis and to identify osteomalacia. Histologic changes in osteoporosis are thinning of trabeculae, decrease in size of osteons, and enlargement of haversian and marrow spaces. Physical activity, calcium supplements (more effective in type II [agerelated] osteoporosis), estrogen-progesterone therapy (in type I [postmenopausal] osteoporosis; best when initiated within 5 to 10 years of menopause), and fluoride (inhibits bone resorption, but bone is more brittle) have a role in the treatment of osteoporosis. Bisphosphonates bind to bone resorption surfaces and inhibit osteoclastic membrane ruffling without destroying the cells. Other drugs, such as intramuscular calcitonin, may also be helpful but are expensive and may cause hypersensitivity reactions. The future of bone augmentation with PTH, growth factors, prostaglandin inhibitors, and other modes of therapy remains to be determined. The best prophylaxis for patients at risk for developing osteoporosis is diet with adequate calcium intake; a weight-bearing exercise program; and an estrogen therapy evaluation at menopause.

Osteomalacia In osteomalacia, a defect in mineralization results in a large amount of unmineralized osteoid (qualitative defect). Osteomalacia is caused by vitamin D–deficient diets, gastrointestinal disorders, renal osteodystrophy, and certain medications (aluminum-containing phosphate-binding antacids [aluminum deposition in bone prevents mineralization] and phenytoin [Dilantin]) (see Box 1A5-1 on p. 79). Like osteoporosis, osteomalacia is also associated with chronic alcoholism. It is commonly associated with Looser’s zones (microscopic stress fractures), other fractures, biconcave vertebral bodies, and trefoil pelvis seen on plain radiographs. Biopsy (transiliac) is required for diagnosis. (Histologically, widened osteoid seams are seen.) Femoral neck fractures are common in patients with osteomalacia. Treatment usually includes large doses of vitamin D.

Scurvy Vitamin C (ascorbic acid) deficiency produces a decrease in chondroitin sulfate synthesis, which leads to defective collagen growth and repair as well as impaired intracellular

Basic Science and Injury of Muscle, Tendon, and Ligament

73

  TABLE 1A5-4  Overview of Clinical and Radiographic Aspects of Metabolic Bone Diseases Disease

Etiology

Clinical Findings

Radiographic Findings

PTH overproduction—adenoma PTH overproduction—MEN/renal

Kidney stone, hyperreflexia Endocrine/renal abnormalities

Osteopenia, osteitis fibrosa cystica Osteopenia

Hypoparathyroidism

PTH underproduction—idiopathic

Calcified basal ganglia

PHP/Albright’s Renal osteodystrophy Rickets (osteomalacia) Vitamin D dependent

PTH receptor abnormality CRF—↓ phosphate excretion

Neuromuscular irritability, eye symptoms Short MC/MT, obesity Renal abnormalities

↓ Vitamin D diet; malabsorption

Bone deformities, hypotonia

Vitamin D dependent (types I and II) Vitamin D resistant (hypophosphatemic) Hypophosphatasia

See Table 1A5-5

Total baldness

“Rachitic rosary,” wide growth plates, fractures Poor mineralization

↓ Renal tubular phosphate resorption ↓Alkaline phosphatase

Bone deformities, hypotonia

Poor mineralization

Bone deformities, hypotonia

Poor mineralization

Osteoporosis

↓ Estrogen—↓ bone mass

Kyphosis, fractures

Scurvy

Vitamin C deficiency—defective collagen

Fatigue, bleeding, effusions

Compression vertebral fractures, hip fractures Thin cortices, corner sign

Osteoclastic abnormality—↑ bone turnover Osteoclastic abnormality—unclear

Deformities, pain, CHF, fractures

Hypercalcemia

Hyperparathyroidism Familial syndromes Hypocalcemia

Brachydactyly, exostosis “Rugger jersey” spine

Osteopenia

Osteodense

Paget’s disease Osteopetrosis

Hepatosplenomegaly, anemia

Coarse trabeculae, “picture frame” vertebrae Bone within bone

↓, Decreased; ↑, increased; CHF, congestive heart failure; CRF, chronic renal failure; MC, metacarpal; MEN, multiple endocrine neoplasia; MT, metatarsal; PHP, ­pseudohypoparathyroidism; PTH, parathyroid hormone. From Brinker MR: Basic sciences. In Miller MD, Brinker MR (eds): Review of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 2000, p 25.

hydroxylation of collagen peptides and a deficiency in collagen cross-linking. Clinical features include fatigue, gum bleeding, ecchymosis, joint effusions, and iron deficiency. Radiographic changes may include thin cortices and trabeculae and metaphyseal clefts (corner sign). Laboratory studies are normal. Histologic changes include replacement of primary trabeculae with granulation tissue, areas of hemorrhage, and widening of the zone of provisional calcification in the physis. The greatest effect on bone formation occurs in the metaphysis.

Osteogenesis Imperfecta Osteogenesis imperfecta is caused by abnormal collagen synthesis (failure of normal collagen cross-linking). The abnormality is primarily due to a mutation in the genes that produce type I collagen.15

CONDITIONS OF BONE MINERALIZATION Conditions of bone mineralization* include hypercalcemic disorders, hypocalcemic disorders, and hypophosphatasia. An overview of these disorders is shown in Tables 1A5-4 to 1A5-6. *See references 16-29.

CONDITIONS OF BONE VIABILITY Osteonecrosis Osteonecrosis (ON) represents death of bony tissue (usually adjacent to a joint surface) from causes other than infection. ON is usually caused by loss of blood supply due to trauma or other causes (e.g., a slipped capital femoral epiphysis). Recent studies suggest that idiopathic ON of the femoral head and Legg-Calvé-Perthes ­disease occur in patients with coagulation abnormalities— ­deficiency of antithrombin factors protein C and protein S and increased levels of lipoprotein(a). ON commonly affects the hip joint, leading to eventual collapse and flattening of the femoral head. The condition is associated with steroid and heavy alcohol use; it is also associated with blood ­dyscrasias (e.g., sickle cell disease), dysbarism (caisson ­ disease), excessive radiation therapy, and Gaucher’s disease. Theories regarding the cause of ON vary.30 It may be related to enlargement of space-occupying marrow fat cells, which leads to ischemia of adjacent tissues. Vascular insults and other factors may also be significant. Idiopathic ON is diagnosed when no other cause can be identified. Idiopathic/alcoholic and dysbaric ON are associated with multiple insults. ON may arise secondary to an underlying hemoglobinopathy (such Text continues on p 77

74

Disorder

Serum Serum Alkaline Calcium Phosphate Phosphatase PTH

25 (OH) Vit D

1,25(OH)2 Urinary Other Findings/ Vit D Calcium Possible Findings

Primary hyper- ↑ parathyroidism

No change No change or ↓ or ↑

↑

No No change change or ↑

↑

Malignancy with ↑ bony metastases

No change No change or ↑ or ↑

No change No No or ↓ change change or ↓

↑

Hyperthyroidism ↑

No change No change

No change No No or ↓ change change

↑

Vitamin D intoxication

↑

No change No change or ↑ or ↑

No change ↑↑↑ or ↓

↑

Hypoparathyroidism

↓

↑

No change

↓

No ↓ change

↓

Pseudohypo↓ parathyroidism

↑

No change

No change No ↓ or ↑ change

↓

No change

Treatment

Comments

Active turnover seen Surgical excision of parathyroid Most commonly due to parathyroid on bone biopsy with adenoma adenoma peritrabecular fibrosis Treat hypercalcemia Because PTH stimulates conversion of the inactive form to the active form [1,25 (OH)2 vitamin D] in the kidney, ↑ production of PTH leads to ↑ levels of 1,25 (OH)2 vitamin D Brown tumors Treat cancer and hypercalcemia ↑ Calcium levels may lead to ↓ PTH Destructive lesions production via feedback mechanism. ↓ in bone 1,25(OH)2 D levels are due to ↓ PTH (which is responsible for conversion of the inactive to the active form of vitamin D in the kidney). Multiple myeloma will display an abnormal urinary and serum protein electrophoresis ↑ Calcium levels due to ↑ bone turnover ↑ Free thyroxine (hypermetabolic state) index ↓Thyroidstimulating hormone Tachycardia, tremors History of excessive vitamin D intake Dietary vitamin D is converted to 25(OH) vitamin D in the liver. This results in very high concentrations of 25(OH) vitamin D, which cross-react ���������������������������� with intestinal vitamin D receptors to ↑ resorption of calcium to cause hypercalcemia Basal ganglia ↓ PTH production most commonly follows calcification surgical ablation of the thyroid (with the Hypocalcemic findings parathyroid) gland. ↓ PTH leads to ↓ serum calcium and ↑ serum phosphate (due to ↓ urinary excretion of phosphate). Because PTH stimulates conversion from the inactive to the active form of vitamin D (in the kidney), 1,25(OH)2 vitamin D is also ↓ Hypocalcemic findings PTH has no effect at the target cells (in the kidney, bone, and intestine) due to a PTH receptor abnormality. This leads to a ↓ in the active form of vitamin D. Therefore, serum calcium levels are ↓ due to: (1) the lack of effect of PTH on bone; (2) ↓ levels of 1,25 (OH)2 vitamin D

DeLee & Drez’s Orthopaedic Sports Medicine

  TABLE 1A5-5  Laboratory Findings and Clinical Data Regarding Patients with the Various Metabolic Bone Diseases

Renal osteodys- ↓ (or No ↑↑↑ trophy (high- change) turnover bone disease of renal disease [secondary hyperparathyroidism])

↑

Renal osteodys- ↑ (or No No change ↑ trophy (lowchange) or ↑ turnover bone disease of renal disease [aluminum toxicity])

↑↑↑

↓

—

No change No ↓ (or mildly change elevated

—

↓

No change

↓ or No ↓ change

↑

↑

↓

↓

Nutritional rickets— calcium ­deficiency Hereditary vitamin D–dependent rickets type I (deficiency)

↓ or No ↓ change

↑

↑

No ↑ (or No ↓ change change)

↓

↓

↑

↑

No ↓↓↓ change (or ↑)

↓

Hereditary vitamin D–dependent rickets type II (hereditary resistance to 1,25(OH)2 vitamin D)

↓

↓

↑

↑

No ↑↑↑ change (or ↑)

↓

Continued

Basic Science and Injury of Muscle, Tendon, and Ligament

Nutritional rickets— vitamin D deficiency

↓ Renal phosphorus excretion leads to (1) Correct underlying renal hyperphosphatemia abnormality. Phosphorus retention leads to ↓ serum (2) Maintain normal serum calcium and ↑↑↑ PTH (which can lead to phosphorus and calcium (3) Dietary phosphate restriction secondary hyperparathyroidism). (4) Phosphate-binding antacid Elevated BUN and creatinine Associated with long-term hemodialysis (calcium carbonate) (5) Administration of the active form of vitamin D—1,25(OH)2 vitamin D (calcitriol) “Rugger jersey” PTH levels are ↓ because of: spine (1) Frequent episodes of hypercalcemia Osteitis fibrosa (2) Direct inhibitory effect of aluminum Amyloidosis on PTH. Osteomalacia may be No secondary hyperparathyroidism is seen present. Elevated BUN and creatinine Associated with long-term hemodialysis Oral administration of vitamin D With ↓ vitamin D intake, intestinal calcium Osteomalacia, (1500-5000 IU/day) and phosphate absorption are reduced, hypotonia leading to hypocalcemia. ↓ Serum calcium Muscle weakness, tetany stimulates ↑ PTH (secondary hyperparaBowing deformities of thyroidism) that leads to bone resorption the long bones and to ↑ serum calcium (toward or to Rachitic rosary normal levels) Sources of vitamin D include: (1) Sunlight (2) Fish liver foods (3) Fortified milk Similar clinical Oral administration of calcium Hypocalcemia leads to secondary hyperfindings as for (700 mg/day) parathyroidism. ↑ PTH leads to enhanced vitamin D deficiency renal conversion of 25(OH) vitamin D to 1,25(OH)2 vitamin D. There is a defect in renal 25(OH) vitamin Osteomalacia Oral administration of physiD 1α-hydroxylase. This enzymatic defect Clinical findings ologic doses (1-2 μg/day) of inhibits conversion from the inactive similar to (but more 1,25 (OH)2 vitamin D form [25(OH)] of vitamin D in the severe than) nutritional kidney. rickets—vitamin D deficiency Osteomalacia Long-term (3-6 months) daily There is an intracellular receptor defect Alopecia administration of high-dose for 1,25(OH)2 vitamin D. Clinical findings vitamin D analogue [1,25 Patients with this disorder have the highest similar to (but more (OH)2 vitamin D or 1α (OH) 1,25(OH)2 vitamin D levels observed severe than) nutritional vit D] plus 3 g/day of elemental in humans; this ↑↑↑ level of 1,25(OH)2 rickets—vitamin D calcium ­vitamin D distinguishes hereditary deficiency vitamin D–dependent rickets type II from type I (the level of 1,25(OH)2 vitamin D is ↓↓↓). Findings of secondary hyperparathyroidism “Rugger jersey” spine Osteitis fibrosa Amyloidosis

75

76

Disorder

Serum Serum Alkaline Calcium Phosphate Phosphatase PTH

25 (OH) Vit D

1,25(OH)2 Urinary Other Findings/ Vit D Calcium Possible Findings

No ↓↓↓ Hypophoschange phatemic rickets (vitamin D–resistant tickets) (phosphate diabetes) (Albright’s syndrome is an example of hypophosphatemic syndrome)

↑

No change

No No change change

Hypophospha- ↑ tasia

↓↓↓

No change

No No change change

↑

Treatment

Comments

No Osteomalacia Oral administration of elemen- There is an inborn error in phosphate change No changes of secondary tal phosphate (1-3 g/day) plus transport (probably located in the proxihyperparathyroidism high-dose vitamin D (20,000mal nephron). This leads to failure of Classic triad: (1) hy70,000 IU/day). Vitamin D reabsorption of phosphate in the kidney pophosphatemia, (2) administration is needed and spilling of phosphate (phosphate lower limb deformities, to counterbalance the diabetes) in the urine. (3) stunted growth rate hypocalcemic effect of Although the absolute levels of 1,25(OH)2 phosphate administration, vitamin D are normal, they are inapwhich otherwise could lead propriately low considering the degree of to severe secondary phosphaturia [production of 1,25(OH)2 hyperparathyroidism vitamin D is normally stimulated by ↓ serum phosphorus (see Table 1A5-3)]. This is the most commonly encoun­ tered form of rickets. ↑ Osteomalacia There is no established medical There is an inborn error in the tissue-nonEarly loss of teeth therapy specific (kidney, bone, liver) isoenzyme of alkaline phosphatase. Elevated urinary phosphoethanolamine is diagnostic.

BUN, blood urea nitrogen; PTH, parathyroid hormone. From Brinker MR: Basic sciences. In Miller MD, Brinker MR (eds): Review of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 2000, pp 26-27.

DeLee & Drez’s Orthopaedic Sports Medicine

TABLE 1A5-5  Laboratory Findings and Clinical Data Regarding Patients with the Various Metabolic Bone Diseases—cont’d

Basic Science and Injury of Muscle, Tendon, and Ligament

77

  TABLE 1A5-6  Differential Diagnosis of the Metabolic Bone Diseases Based on Blood Chemistries ↑ Calcium

↓ Calcium

Normal Calcium

↑ Phosphorus

↓ Phosphorus

Normal Phosphorus

Primary hyperparathyroidism Hyperthyroidism Vitamin D intoxication Malignancy without bony metastasis Malignancy with bony metastasis Multiple myeloma Lymphoma Sarcoidosis Milk-alkali syndrome Severe generalized ­immobilization Multiple endocrine neoplasia Addison’s disease Steroid administration Peptic ulcer disease Hypophosphatasia

Hypopara­ thyroidism Pseudohypopara­ thyroidism Renal osteodystrophy High-turnover bone disease Nutritional ­rickets— vitamin D deficiency Nutritional rickets—calcium ­deficiency Hereditary vitamin D–dependent rickets (types I and II)

Osteoporosis Pseudohypoparathyroidism Nutritional rickets— vitamin D deficiency Nutritional rickets— calcium deficiency Nutritional rickets— phosphate deficiency Hypophosphatemic rickets

Malignancy with bony metastasis Multiple myeloma Lymphoma Vitamin D intoxication Hypoparathyroidism Pseudohypopara­ thyroidism Renal osteodystrophy Hypophosphatasia Sarcoidosis Milk-alkali syndrome Severe generalized immobilization

Primary ­hyperpara­thyroidism Malignancy without bony metastasis Nutritional rickets — vitamin D deficiency Nutritional rickets—calcium deficiency Nutritional rickets— phosphate deficiency Hereditary vitamin D–dependent rickets (types I and II) Hypophosphatemic rickets

Osteoporosis Primary hyperthyroidism Malignancy with bony metastasis Multiple myeloma Lymphoma Hyperthyroidism Vitamin D intoxication Renal osteodystrophy (only low-turnover bone disease) Sarcoidosis Milk-alkali syndrome Severe generalized immobilization

↑ PTH

↓ PTH

Normal PTH

↑ 1,25(OH)2 Vitamin D

↓ 1,25(OH)2Vitamin D

Primary hyperparathyroidism Pseudohypopara­ thyroidism Renal osteodystrophy Nutritional rickets— vitamin D deficiency Nutritional rickets— calcium deficiency Hereditary vitamin D– dependent rickets (types I and II)

Malignancy with bony metastasis Malignancy without bony metastasis Multiple myeloma Lymphoma Hyperthyroidism Vitamin D ­intoxication Hypoparathyroidism Sarcoidosis Milk-alkali syndrome Severe generalized immobilization

Osteoporosis Malignancy with bony metastasis Multiple myeloma Lymphoma Vitamin D intoxication Pseudohypoparathyroidism Renal osteodystrophy (only low-turnover bone disease) Nutritional rickets— phosphate deficiency Hypophosphatemic rickets Hypophosphatasia Sarcoidosis Hyperthyroidism Milk-alkali syndrome Severe generalized immobilization

Primary hyperparathyroidism Nutritional rickets— calcium deficiency Nutritional rickets— phosphate deficiency Hereditary vitamin D–dependent rickets (type II) Sarcoidosis

Malignancy with bony metastasis Malignancy without bony metastasis Multiple myeloma Lymphoma Hypoparathyroidism Pseudohypoparathyroidism Renal osteodystrophy Nutritional rickets— vitamin D ­deficiency Hereditary vitamin D–dependent rickets (type I)

Normal 1,25(OH)2 Vitamin D Osteoporosis Primary hyperparathyroidism Malignancy with bony metastasis Multiple myeloma Lymphoma Hyperthyroidism Vitamin D intoxication Nutritional rickets— calcium deficiency Hypophosphatemic rickets Hypophosphatasia

PTH, parathyroid hormone. From Brinker MR: Basic sciences. In Miller MD, Brinker MR (eds): Review of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 2000, pp 28-29.

as sickle cell disease) or a ­ marrow disorder (such as ­hemochromatosis). The incidence of ON of the femoral head in renal ­transplant recipients has been reduced by the use of cyclosporine. In ON, grossly necrotic bone, fibrous tissue, and subchondral collapse may be seen. Histologically, early changes involve autolysis of osteocytes (14 to 21 days) and necrotic marrow, followed by inflammation with invasion of buds of primitive mesenchymal tissue and capillaries. Later, new woven bone is laid down on top of dead trabecular bone. This stage is followed by resorption of the dead trabeculae and remodeling during a process of “creeping substitution.” The bone is weakest during this process, and collapse (crescent sign, seen on radiographs) and fragmentation can occur.31

Osteochondrosis Osteochondrosis can occur at traction apophyses in ­children and may or may not be associated with trauma, inflammation of the joint capsule, or vascular insult and

secondary thrombosis. The pathologic process is similar to that described for ON in the adult.

BONE AS BIOMATERIAL Mechanical Properties Bone is a composite of collagen and hydroxyapatite. Collagen has a low Young’s modulus (E), good tensile strength, and poor compressive strength. Calcium apatite is a stiff, brittle material with good compressive strength. The ­combination is an anisotropic material that resists many forces; bone is strongest in compression, weakest in shear, and intermediate in tension. Mineral content is the main determinant of the elastic modulus of cortical bone. Cancellous bone is 25% as dense, 10% as stiff, and 500% as ductile as cortical bone. Bone is a dynamic material because of its ability to self-repair, to change with aging (becoming stiffer and less ductile), and to change with prolonged immobilization (becoming weaker).

78

DeLee & Drez’s Orthopaedic Sports Medicine

  TABLE 1A5-7  Relationship between the Mode of Bone Loading and the Resulting Fracture Configuration Loading Mode

Fracture Configuration

Tension Compression Torsion High energy

Transverse Oblique Spiral Comminution

With aging, there is a decline in the material properties of bone. In an attempt to offset the loss in material properties, bone remodels its geometry such that the inner and outer cortical diameters increase. The increased diameter of bone increases the area moment inertia and results in a decrease in the bending stresses on bone. Stress concentration effects, which occur at points of defects within the bone or the implant-bone interface (stress risers), reduce the overall strength of bone for loading. Stress shielding by implants results in osteoporosis of adjacent bone owing to lack of normal physiologic stresses. This situation occurs commonly under plates and at the femoral calcar in high-riding total hip arthroplasties. A hole of 20% to 30% of the bone diameter, regardless of whether it is filled with a screw, reduces overall strength up to 50%, which does not return to normal until 9 to 12 months after screw removal. Cortical defects can reduce strength 70% or more (less with oval defects, as compared with rectangular defects, owing to a smaller stress riser). Bone is anisotropic and viscoelastic. The strength of cortical bone is excellent versus torque; the strength of cancellous bone is good versus compressive and shear forces. A fracture’s configuration is determined by the mode of loading (Table 1A5-7).

TYPES OF BONE FORMATION Endochondral Bone Formation and Mineralization In the process of endochondral bone formation, undifferentiated cells secrete cartilaginous matrix and differentiate into chondrocytes. This matrix mineralizes and is invaded by vascular buds that bring in osteoprogenitor cells. Osteoclasts then resorb calcified cartilage, and osteoblasts form bone. Bone replaces the cartilage model; cartilage is not converted to bone. Examples of endochondral bone ­formation include embryonic long-bone formation, longitudinal growth (physis), development of fracture callus, and the formation of bone with the use of demineralized bone matrix. Embryonic long bones are formed through endochondral bone formation. In the process of embryonic long-bone formation, bone is formed from a mesenchymal anlage that is usually present at 6 weeks in utero. Vascular buds invade the mesenchymal model, bringing in osteoprogenitor cells that differentiate into osteoblasts and form the primary centers of ossification at about 8 weeks in utero. The cartilage model grows through appositional (width) and interstitial (length) growth. The marrow is formed by resorption of the central portion of the cartilage anlage by invasion of myeloid precursor cells brought in by the capillary buds.

Secondary centers of ossification develop at the bone ends, forming epiphyseal centers of ossification (growth plates), which are responsible for longitudinal growth of immature bones. During this developmental stage, there is a rich arterial supply composed of an epiphyseal artery (which terminates in the proliferative zone), metaphyseal arteries, nutrient arteries, and perichondrial arteries. Two growth plates exist in immature long bones: the horizontal growth plate (the physis) and the spherical growth plate that allows growth of the epiphysis. The spherical growth plate has the same arrangement as the physis but is less organized. Physeal cartilage is divided into zones based on growth and function.

Intramembranous Bone Formation Intramembranous bone formation occurs without a cartilage model. Undifferentiated mesenchymal cells aggregate into layers (or membranes). These cells differentiate into osteoblasts and deposit organic matrix that mineralizes to form bone. Examples of intramembranous bone formation include embryonic flat-bone formation (pelvis, clavicle, vault of skull), bone formation during ­distraction ­osteogenesis, and blastema bone formation (occurs in young children with amputations).

Appositional Ossification In the process of appositional ossification, osteoblasts align themselves on an existing bone surface and lay down new bone. Examples of appositional ossification include periosteal bone enlargement (width) and the bone formation phase of bone remodeling.

BIOLOGY OF FRACTURE HEALING Overview Fracture healing involves a series of cellular events: inflammation, fibrous tissue formation, cartilage formation, and endochondral bone formation.32 The cellular events of fracture healing are influenced by undifferentiated cells in the area of the fracture and osteoinductive growth factors released into the fracture environment.

Fracture Repair The response of bone to injury can be thought of as a continuum of histologic processes, beginning with inflammation, proceeding through repair (soft callus followed by hard callus), and finally ending in remodeling.33-36 Fracture repair is unique in that healing is completed without the formation of a scar. Fracture healing may be influenced by a variety of biologic and mechanical factors (Table 1A5-8). In the inflammation phase, bleeding from the fracture site and surrounding soft tissues creates a hematoma, which provides a source of hematopoietic cells capable of secreting growth factors. Subsequently, fibroblasts, mesenchymal cells, and osteoprogenitor cells gather at the fracture site, and fibrovascular tissue forms around the fracture

Basic Science and Injury of Muscle, Tendon, and Ligament

Box 1A5-1 Causes of Rickets and Osteomalacia

79

  TABLE 1A5-8  Biologic and Mechanical Factors Influencing Fracture Healing

Nutritional Deficiency • Vitamin D deficiency • Dietary chelators (rare) of calcium • Phytates • Oxalates (spinach) • Phosphorus deficiency (unusual) • Antacid (aluminum-containing) abuse leading to severe ­dietary phosphate binding Gastrointestinal Absorption Defects • Postgastrectomy (rare today) • Biliary disease (interference with absorption of fat­soluble vitamin D) • Enteric absorption defects • Short bowel syndrome • Rapid-transit (gluten-sensitive enteropathy) syndromes • Inflammatory bowel disease • Crohn’s disease • Celiac disease Renal Tubular Defects (Renal Phosphate Leak) • X-linked dominant hypophosphatemic vitamin D–­resistant rickets (VDRR) or osteomalacia • Classic Albright’s syndrome type II • Phosphaturia and glycosuria • Fanconi’s syndrome type III • Phosphaturia, glycosuria, aminoaciduria • Vitamin D–dependent rickets (or osteomalacia) type I (a genetic or acquired deficiency of renal tubular 25­hydroxyvitamin D 1-α hydroxylase enzyme that prevents conversion of 25-hydroxyvitamin D) • Vitamin D–dependent rickets (or osteomalacia) type II (this entity represents enteric end-organ insensitivity to 1,25-dihydroxyvitamin D and is probably caused by an abnormality in the 1,25-dihydroxyvitamin D nuclear ­receptor)  enal Tubular Acidosis R • Acquired—associated with many systemic diseases • Genetic • Debré-de Toni-Fanconi syndrome • Lignac-Fanconi syndrome (cystinosis) • Lowe’s syndrome Renal Osteodystrophy  iscellaneous Causes M • Soft tissue tumors secreting putative factors • Fibrous dysplasia • Neurofibromatosis • Other soft tissue and vascular mesenchymal tumors • Anticonvulsant medication (induction of the hepatic P-450 microsomal enzyme system by some anticonvulsants—phenytoin, phenobarbital, primidone (Mysoline)—causes increased degradation of vitamin D metabolites) • Heavy metal intoxication • Hypophosphatasia • High-dose diphosphonates • Sodium fluoride Adapted from Simon SR: Orthopaedic Basic Science, 2nd ed. Rosemont, Ill,   American Academy of Orthopaedic Surgeons, 1994, p 169.

Biologic Factors

Mechanical Factors

Patient age Comorbid medical conditions Functional level Nutritional status Nerve function Vascular injury Hormones Growth factors Health of the soft tissue envelope Sterility (in open fractures) Cigarette smoking Local pathologic conditions Level of energy imparted Type of bone affected Extent of bone loss

Soft tissue attachments to bone Stability (extent of immobilization) Anatomic location Level of energy imparted Extent of bone loss

From Brinker MR: Basic science. In Miller MD, Brinker MR (eds): Review of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 2000, p 18.

ends. Osteoblasts from surrounding osteogenic precursor cells, fibroblasts, or both proliferate. In the repair phase, primary callus response occurs within 2 weeks. If the bone ends are not in continuity, bridging (soft) callus occurs. (Fibrocartilage develops and stabilizes the bone ends.) The soft callus (fibrocartilage) later is replaced, through the process of endochondral ossification, by woven bone (hard callus). Another type of callus, medullary callus, supplements the bridging callus, although it forms more slowly and occurs later. The amount and type of callus formation are related to the method of treatment (Fig. 1A5-6; Table 1A5-9).37 Primary cortical healing, which resembles normal remodeling, occurs with rigid immobilization and anatomic (or near-anatomic) reduction (bone ends in continuity). With rigidly fixed fractures (such as with a compression plate), direct osteonal or primary bone healing occurs without visible callus formation. With closed treatment, “endochondral healing” with periosteal bridging callus formation occurs. The remodeling phase begins during the middle of the repair phase and continues long after the fracture has clinically healed (up to 7 years). Remodeling allows the bone to assume its normal configuration and shape based on the stresses to which it is exposed (Wolff’s law). Throughout the process, woven bone formed during the repair phase is replaced with lamellar bone. Fracture healing is complete when there is repopulation of the marrow space.

Growth Factors of Bone Growth factors of bone include BMPs, TGF-β,38 insulinlike growth factor II, platelet-derived growth factor, and others.39 BMPs are osteoinductive40 and induce ­metaplasia of mesenchymal cells into osteoblasts; the target cell for BMPs is the undifferentiated perivascular mesenchymal cell. BMPs stimulate bone formation and can be used clinically to stimulate bony healing in challenging injuries such as open tibial fractures41 and nonunions of the long bones.42 TGF-β induces mesenchymal cells to produce type II collagen and proteoglycans. TGF-β also induces osteoblasts to synthesize collagen. TGF-β is found in fracture hematoma and is believed to regulate cartilage and bone formation

80

DeLee & Drez’s Orthopaedic Sports Medicine

A

C

B

D

Figure 1A5-6  The amount of callus formation and type of healing are related to the method of fracture treatment. Two very similar fractures presented here were treated using different methods and healed by different biologic processes. A, Presenting radiograph of a 50-year-old man with a displaced midshaft clavicle fracture. B, Following nonoperative treatment with a figure-  of-eight dressing, abundant periosteal callus formation is seen (inset). C, Presenting radiograph of a 25-year-old woman with a displaced midshaft clavicle fracture. D, Following plate and screw stabilization, the mode of healing is direct osteonal healing (remodeling) without external callus formation.

in fracture callus. Insulin-like growth factor II stimulates type I collagen, cellular proliferation, cartilage matrix synthesis, and bone formation. Platelet-derived growth factor is released from platelets and attracts inflammatory cells to the fracture site (chemotactic). The production and actions of these growth factors appear to be related to heparan ­sulfate, which is expressed on the cell surface and is known to increase the effect of many growth factors.43

Ultrasound and Fracture Healing Clinical studies show that low-intensity pulsed ultrasound accelerates fracture healing44 and increases the mechanical strength of callus, including torque and stiffness. The postulated mechanism of action is that the cells responsible for fracture healing respond in a favorable manner to the mechanical energy transmitted by the ultrasound signal. Ultrasound has also been shown to decrease consolidation time during distraction osteogenesis45 and to stimulate bony healing in nonunions.46

Electricity and Fracture Healing Electrical properties of cartilage and bone are dependent on their charged molecules. Devices intended to stimulate fracture repair by altering a variety of cellular ­activities have been introduced. Direct current stimulates an inflammatory-like response (stage I). Alternating current (capacity-coupled generators) affects cyclic adenosine monophosphate, collagen synthesis, and calcification during the repair stage. Pulsed electromagnetic fields initiate calcification of fibrocartilage but cannot induce calcification of fibrous tissue.

Problems of Fracture Healing A delayed union represents a fracture that has failed to unite as anticipated but continues to show some biologic activity. A nonunion (Fig. 1A5-7) represents a fracture without clinical or radiographic evidence of healing, and without evidence of the ability for progression to healing, at

Basic Science and Injury of Muscle, Tendon, and Ligament

  TABLE 1A5-9  Type of Fracture Healing Based on Type of Stabilization Type of Immobilization

Predominant Type of Healing

Cast (closed treatment) Compression plate

Periosteal bridging callus Primary cortical healing (remodeling) Early—periosteal bridging callus Late—medullary callus Dependent on extent of rigidity Less rigid—periosteal bridging callus More rigid—primary cortical healing Hypertrophic nonunion

Intramedullary nail

Inadequate

Comments Enchondral ossification Cutting cone–type remodeling Enchondral ossification

Failed endochondral ossification Type II collagen ­predominates

From Brinker MR: Basic sciences. In Miller MD, Brinker MR (eds): Review of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 2000, p 19.

6 to 9 months after injury. Atrophic nonunions are avascular and lack the biologic capacity to heal. The ends of the bone are typically narrowed (such as a pencil point) and appear to be avascular. The treatment of atrophic nonunion is stimulation of the local biologic activity, such as with a bone graft or a corticotomy for bone transport. Hypertrophic nonunions are hypervascular and possess the biologic capacity to heal but lack mechanical stability. The ends of the bone are typically hypertrophied and give the appearance that the fracture has “attempted to heal.” The treatment of hypertrophic nonunion is to add further mechanical stability, such as with plate and screw fixation; bone grafting is not needed. The initial biologic response of hypertrophic nonunion to plate stabilization is mineralization of fibrocartilage.

BONE GRAFTING Bone grafts are an important adjunct in the treatment of fractures, delayed unions, and nonunions.47-49 Bone grafts have four important properties: (1) osteoconductive matrix (acts as a scaffold or framework into which bone growth occurs), (2) osteoinductive factors (growth factors such as BMPs and TGF-β that promote bone formation), (3) osteogenic cells (include primitive mesenchymal cells, osteoblasts, and osteocytes), and (4) structural integrity (Table 1A5-10).

Specific Bone Grafts Type of Graft Cortical grafts incorporate through slow remodeling of existing haversian systems by a process of resorption (which weakens the graft) followed by deposition of new bone (restoring its strength). Resorption is confined to the osteon borders, and interstitial lamellae are preserved. Cortical bone is slower to turn over than cancellous bone and is used for structural defects.50 Cancellous grafts are commonly used for grafting nonunions or cavitary defects because cancellous bone is quickly

81

remodeled and incorporated. Cancellous bone is rapidly revascularized; osteoblasts lay down new bone on old trabeculae, which are later remodeled (“creeping substitution”). Vascularized bone grafts, though technically difficult, allow more rapid union with preservation of most cells. Vascularized grafts are best used for irradiated tissues or when large tissue defects exist.51 Donor site morbidity is a risk of vascularized grafts (e.g., fibula). Osteoarticular (osteochondral) allografts are being used with increasing frequency for tumor surgery. These grafts are immunogenic52-54; cartilage is vulnerable to inflammatory mediators of immune response (cytotoxic injury from antibodies and lymphocytes). The articular cartilage is preserved with glycerol or dimethyl sulfoxide treatment. Cryogenically preserved grafts leave few viable ­ chondrocytes. Tissue-matched fresh osteochondral grafts produce minimal immunogenic effect and incorporate well. Demineralized bone matrix is both osteoconductive and osteoinductive. Several studies using animal models have suggested that the use of a composite of ceramic and osteoinductive protein may be effective in the treatment of segmental bone defects.55-62 In addition, bone marrow cells have been shown to be clinically useful in the treatment of mechanically stable nonunions.

Source of Graft Autograft bone is harvested from the same person. Allograft bone is harvested from a cadaveric donor. All allografts must be harvested with use of sterile technique, and donors must be screened for potential transmissible diseases. Synthetic grafts are composed of calcium, silicon, or aluminum.

Allograft Preparation Allograft bone can be prepared using a variety of methods.63 Fresh allograft has the highest antigenicity. Fresh frozen allograft is less immunogenic than fresh allograft and preserves BMPs. Freeze-dried (lyophilized) allograft loses structural integrity and depletes BMPs. Freeze-dried allograft is the least immunogenic and is purely osteoconductive, with the lowest likelihood of viral transmission.

DISTRACTION OSTEOGENESIS Distraction osteogenesis (Fig. 1A5-8) involves the use of distraction to stimulate the formation of bone. Clinical applications include limb lengthening, treatment of hypertrophic nonunions, deformity correction (by differential lengthening), and repair of segmental bone loss (by bone transport).64,65 Under optimal stability, bone is formed by intramembranous ossification. In an unstable environment, bone forms by endochondral ossification, or, in an extremely unstable environment, pseudarthrosis may occur.66 Conditions that promote optimal bone formation during distraction osteogenesis include the following: a low-energy corticotomy/osteotomy; minimal soft tissue stripping at the corticotomy site (preserves the blood supply); stable external fixation to eliminate torsion, shear, and bending moments; a latency phase (no lengthening) of

82

DeLee & Drez’s Orthopaedic Sports Medicine

5 to 7 days; a distraction phase with 0.25 mm of distraction 3 or 4 times per day (0.75 to 1.0 mm per day, about 1 inch per month); a neutral fixation interval (no distraction) or consolidation phase (typically twice as long as the distraction phase); and normal physiologic use of the extremity, including weight-bearing.

HETEROTOPIC OSSIFICATION Heterotopic ossification (HO) is the formation of ectopic bone in the soft tissues, most commonly in response to an injury or a surgical dissection. Myositis ossificans is a form of HO that occurs specifically when the ossification is in muscle. Patients with traumatic brain injuries are particularly prone to developing HO, and recurrence after operative resection is likely if the neurologic compromise is severe. When resecting HO that has formed after total hip arthroplasty (resection should be delayed for a minimum of 6 months after total hip arthroplasty), radiation is a useful adjuvant therapy to prevent the recurrence of HO. Irradiation (usually in doses of 700 rad) prevents the proliferation and differentiation of primordial mesenchymal cells into osteoprogenitor cells that can form osteoblastic tissue. Oral diphosphonates inhibit mineralization of osteoid but do not prevent the formation of osteoid matrix; when the oral diphosphonate therapy is discontinued, mineralization with formation of HO may occur.

BONE INFECTIONS Osteomyelitis is an infection of bone and bone marrow that may be caused by direct inoculation of an open traumatic wound or by blood-borne organisms (hematogenous). It is impossible to predict the microscopic organism that is causing osteomyelitis based on the clinical picture and the age of the patient; therefore, a specific microbiologic diagnosis through deep cultures is essential. Organisms isolated from sinus tract drainage typically do not accurately reflect the organisms present deep within the wound and within bone.67

Acute Hematogenous Osteomyelitis Acute hematogenous osteomyelitis is a bone and bone marrow infection caused by blood-borne organisms. Children are commonly affected; boys are more commonly affected than girls. In children, the infection is most common in the metaphysis or the epiphysis of the long bones and is more common in the lower extremity than the upper extremity. Radiographic changes of acute hematogenous osteomyelitis include soft tissue swelling (early), bone demineralization (10 to 14 days), and sequestra (dead bone with surrounding granulation tissue) and later involucrum (periosteal new bone). Pain, loss of function of the involved extremity, and a soft tissue abscess may be present. Patients commonly have an elevated white blood cell count and an elevated erythrocyte sedimentation rate. Blood culture results may also be positive. The most sensitive monitor of the course of infection in children with acute hematogenous ­osteomyelitis is the C reactive protein. Nuclear medicine studies may be ­helpful in equivocal cases. Magnetic resonance imaging shows

Figure 1A5-7  Anteroposterior radiograph of a 60-yearold man 15 months after a closed tibia fracture. This fracture shows no evidence of progression to healing and is therefore considered a nonunion.

changes in bone and bone marrow before plain films do. Empirical therapy should be started after cultures have been obtained, either by aspiration or by surgical drainage, when indicated. The indications for operative intervention include (1) drainage of an abscess, (2) débridement of infected tissues to prevent further destruction, and (3) treatment of refractory cases that fail to show improvement after nonoperative treatment. The management of acute osteomyelitis may be summarized as follows: (1) identification of the organisms, (2) selection of appropriate antibiotics, (3) delivery of antibiotics to the site of infection, and (4) halting of tissue destruction. Empirical therapy before definitive culture results are available is based on the patient’s age and special circumstances.

Acute Osteomyelitis (after Open Fracture or after Open Reduction with Internal Fixation) Clinical findings in acute osteomyelitis may be similar to those in acute hematogenous osteomyelitis. Treatment includes radical irrigation and débridement with removal of orthopaedic hardware as necessary. Open wounds may require rotational or free flaps. The most common offending organisms are Staphylococcus aureus, Pseudomonas aeruginosa, and coliforms. Empirical therapy before definitive culture results are available is nafcillin with ciprofloxacin; alternative therapy is vancomycin with a third-­generation cephalosporin. Patients with acute osteomyelitis and vascular insufficiency and those who are immunocompromised generally show a polymicrobial picture.

Basic Science and Injury of Muscle, Tendon, and Ligament

83

  TABLE 1A5-10 ����������������������������������������������������� Types of Bone Grafts and Bone Graft Properties Graft

Osteoconduction

Osteoinduction

Osteogenic Cells

Structural Integrity

Other Properties

Autograft Cancellous Cortical Allograft

Excellent Fair Fair

Good Fair Fair

Excellent Fair None

Poor Excellent Good

Fair Fair

None Good

None None

Fair Poor

Rapid incorporation Slow incorporation Fresh has the highest immunogenicity Freeze-dried is the least immunogenic but has the least structural integrity (weakest) Fresh frozen preserves BMP

Poor

Poor

Good

Poor

Ceramic Demineralized bone matrix Bone marrow

BMP, bone morphogenetic protein. Modified from Brinker MR, Miller MD: Fundamentals of Orthopaedics, 3rd ed. Philadelphia, WB Saunders, 1999, p 7.

Chronic Osteomyelitis Chronic osteomyelitis may arise as a result of inappropriately treated acute osteomyelitis, trauma, or soft tissue spread, especially in the Cierny type C elderly host or in those who are immunosuppressed, diabetic, or ­intravenous drug abusers (Fig. 1A5-9). Chronic ­ osteomyelitis may

Figure 1A5-8  Anteroposterior radiograph of the tibia of a 71-year-old man undergoing bone transport for infected tibial nonunion. The arrows indicate the bony regenerate of distraction osteogenesis.

Figure 1A5-9  Radiograph of the forearm of a young woman with chronic osteomyelitis of the radius after an open forearm fracture. The large arrow indicates periosteal new bone formation. The small arrow indicates the sequestrum.

84

DeLee & Drez’s Orthopaedic Sports Medicine

drainage only if pus is present; otherwise, 48 hours of intravenous antibiotics followed by 6 weeks of oral antibiotics is curative.

Chronic Sclerosing Osteomyelitis Medullary

Superficial

Chronic sclerosing osteomyelitis is an unusual infection that involves primarily the diaphyseal bones of adolescents. Typified by intense proliferation of the periosteum leading to bony deposition, it may be caused by anaerobic organisms. Insidious onset, dense progressive sclerosis on radiographs, and localized pain and tenderness are common. Malignancy must be ruled out. Surgery and antibiotic therapy usually are not curative.

Chronic Multifocal Osteomyelitis Localized Diffuse Figure 1A5-10  Cierny’s anatomic classification of adult chronic osteomyelitis.   (From Cierny G III: Chronic osteomyelitis: Results of treatment. Instr Course Lect 39:495, 1990.)

be classified anatomically (Fig. 1A5-10). Skin and soft tissues are often involved, and the sinus tract may occasionally develop squamous cell carcinoma. Periods of quiescence of the infection are often followed by acute exacerbations. Nuclear medicine studies are often helpful in determining the activity of the disease. Operative sampling of deep specimens from multiple foci68 is the most accurate means of identifying the pathologic organisms. A combination of intravenous antibiotics (based on deep cultures), surgical débridement, bone grafting, and soft tissue coverage is often required. Unfortunately, amputations are still necessary in certain cases. S. aureus, Enterobacteriaceae, and P. aeruginosa are the most frequent offending organisms. Treatment is based on culture and sensitivity testing; empirical therapy is not indicated in chronic osteomyelitis.

Subacute Osteomyelitis Subacute osteomyelitis is usually discovered radiologically in a patient with a painful limp and no systemic (and often no local) signs or symptoms. Subacute osteomyelitis may arise secondary to partially treated acute osteomyelitis and occasionally develops in a fracture hematoma. Unlike in acute osteomyelitis, white blood cell count and blood culture results are frequently normal. Erythrocyte sedimentation rate, bone cultures, and radiographs are often useful. Subacute osteomyelitis most commonly affects the femur and the tibia; unlike acute osteomyelitis, it can cross the physis, even in older children. Radiographic changes include Brodie’s abscess, a localized radiolucency usually seen in the metaphyses of long bones. It is sometimes difficult to differentiate Ewing’s sarcoma from subacute osteomyelitis. Treatment of Brodie’s abscess in the metaphysis includes surgical curettage. When the process is localized to the epiphysis, other lesions (e.g., chondroblastoma) must be ruled out. Epiphyseal osteomyelitis is caused almost exclusively by S. aureus. Epiphyseal osteomyelitis requires surgical ­

Chronic multifocal osteomyelitis is caused by an infectious agent; it appears in children without systemic symptoms. Normal laboratory values, except for an elevated ­erythrocyte sedimentation rate, are common. Radiographs demonstrate multiple metaphyseal lytic lesions, ­especially in the medial clavicle, the distal tibia, and the distal femur. Symptomatic treatment only is ­recommended because this condition usually resolves spontaneously.

C l Bone

r i t i c a l

P

o i n t s

is a composite of collagen (good tensile strength, poor compressive strength) and hydroxyapatite (high stiffness, good compressive strength). The combination results in an anisotropic material that is strongest in compression, weakest in shear, and intermediate in tension. l Pathology of bone can be classified into conditions affecting bone mineral density, conditions affecting bone mineralization, conditions affecting bone viability, fracture, and infection. l Osteoporosis is an age-related decrease in bone mineral density usually associated with loss of estrogen in postmenopausal women or poor calcium absorption in elderly persons. Osteoporosis is a defect of bone quantity, not quality, that greatly increases the risk for fracture. l Osteomalacia is a defect of bone quality rather than quantity; specifically, a high proportion of osteoid remains unmineralized. Osteomalacia is caused by vitamin D deficiency (dietary or metabolic), gastrointestinal disorders (impaired vitamin D absorption), renal osteodystrophy, and some medications. Deformities and fractures are common. l Osteonecrosis represents bone death caused by a loss of blood supply. Grossly necrotic bone, fibrous tissue, and subchondral collapse may occur. A similar pathophysiology can produce osteochondrosis in children at traction apophyses. l Fracture healing consists of a continuum of histologic processes: inflammation, repair (soft callus followed by hard callus), and remodeling. A variety of biologic and mechanical factors affect this process.

Basic Science and Injury of Muscle, Tendon, and Ligament l During

inflammation, growth factors are secreted and attract fibroblasts, mesenchymal cells, and osteoprogenitor cells. Fibrovascular tissue forms, and osteoblasts and fibroblasts proliferate. During repair, primary callus response occurs within 2 weeks. The amount and type of callus formation are related to the method of treatment. Remodeling begins during the repair phase and continues for years. The bone assumes its configuration and shape based on the stresses to which it is exposed (Wolff’s law). l Heterotopic ossification is the formation of ectopic bone in soft tissues; myositis ossificans is when the ossification occurs in muscle, which may arise following a severe muscle contusion. l Osteomyelitis is an infection of bone and bone marrow. Acute hematogenous osteomyelitis is caused by bloodborne organisms and is most common in children. Acute osteomyelitis occurs following open fracture or surgical reduction with internal fixation. Chronic osteomyelitis may arise as a result of inadequately treated acute osteomyelitis, trauma, or soft tissue spread, especially in people with immunosuppression, diabetes, or abuse of intravenous drugs.

S U G G E S T E D

85

R E A D I N G S

Baltzer AW, Lieberman JR: Regional gene therapy to enhance bone repair. Gene Ther 11:344-350, 2004. Ciombor DM, Aaron RK: The role of electrical stimulation in bone repair. Foot Ankle Clin 10:579-593, 2005. Dimitriou R, Dahabreh Z, Katsoulis E, et al: Application of recombinant BMP-7 on persistent upper and lower limb non-unions. Injury 36(Suppl 4):S51-59, 2005. Gamradt SC, Lieberman JR: Genetic modification of stem cells to enhance bone repair. Ann Biomed Eng 32:136-147, 2004. Khan SN, Cammisa FP Jr, Sandhu HS, et al: The biology of bone grafting. J Am Acad Orthop Surg 13:77-86, 2005. Lane JM: Bone morphogenic protein science and studies. J Orthop Trauma 19: S17-22, 2005. Malizos KN, Hantes ME, Protopappas V, Papachristos A: Low-intensity pulsed ultrasound for bone healing: An overview. Injury 37(Suppl 1):S56-62, 2006. Nordsletten L: Recent developments in the use of bone morphogenetic protein in orthopaedic trauma surgery. Curr Med Res Opin 22(Suppl 1):S13-17; S23, 2006. Termaat MF, Den Boer FC, Bakker FC, et al: Bone morphogenetic proteins. Development and clinical efficacy in the treatment of fractures and bone defects. J Bone Joint Surg Am 87:1367-1378, 2005. Westerhuis RJ, van Bezooijen RL, Kloen P: Use of bone morphogenetic proteins in traumatology. Injury 36:1405-1412, 2005.

R eferences Please see www.expertconsult.com

S ecti o n

B

Fundamentals of Biomechanics Richard E. Debski, Alexis C. Wickwire, and Jon K. Sekiya

Biomechanics is an interdisciplinary field that uses the principles of mechanics to improve the human body through conception, design, development, and analysis of equipment and systems in medicine and biologic systems. This knowledge can help in understanding the loading of the musculoskeletal system and its mechanical responses, which can be used in sports medicine to further understand normal function, predict changes, and propose interventions to athletes. More specifically, basic biomechanics explores forces and moments required for movement and balance of the human body by muscle recruitment and the consequence of internal forces on loading of soft tissues. This chapter will explore biomechanics in terms of three different topics: statics, dynamics, and mechanics of materials. Additionally, throughout the chapter there are examples of these concepts to illustrate how they can be directly applied to sports medicine.

BASIC CONCEPTS Units of Measure Understanding the basic units of measure for biomechanics is important in describing the dimensional or spatial analyses. There are three primary dimensions (length, time, and

  TABLE 1B-1  Dimensional Quantities and Corresponding Units in the SI System Dimensional Analysis Primary Dimensions Length Time Mass Secondary Dimensions Area Volume Velocity Acceleration Force Pressure/stress Moment/torque Work/energy Power

SI Unit meter (m) second (s) kilogram (kg) m2 m3 m/s m/s2 Newton (N) = kg·m/s2 Pascal (Pa) = N/m2 N·m Joule (J) = N·m J/s

mass) whereby secondary dimensions have subsequently been derived (Table 1B-1). Metric unit measurements are commonly used in the field of biomechanics to describe such dimensions and will also be used in this chapter. One measurement not described in Table 1B-1 is that used for angular descriptions. The unit for angles is typically defined in terms of radians or degrees.

86

DeLee & Drez’s Orthopaedic Sports Medicine

Scalars and Vectors Most of the physical quantities encountered in mechanics are either scalars or vectors. Scalar quantities describe only magnitude and are used for concepts such as time, speed, mass, temperature, volume, work, power, and energy. In contrast, vector quantities have both a magnitude and a direction component associated with them. Examples of vector quantities are displacement, force, moment, velocity, and acceleration. Graphically, a vector is represented as an arrow, with the orientation of the arrow indicating its line of action or direction. Typically, vectors are represented in a Cartesian coordinate system. Depending on the analysis being performed, the coordinate system can be in either two or three dimensions. Two perpendicular axes are identified for a twodimensional coordinate system that is usually illustrated as x- and y-axes. Each axis has a positive and negative linear component. Similarly, angles and moments have a positive and negative direction about a specified axis (clockwise or counterclockwise).

between the point and line of action of the force. This distance is known as the moment or lever arm. A larger moment arm requires less force to achieve angular motion about the axis of rotation, which also means that the moment will be of greater magnitude. Although a moment can realistically be calculated about any point, typically it is calculated about a joint axis of rotation during biomechanical analyses. For example, when tension is generated in the biceps muscle during a biceps curl exercise, the tendon pulls on the forearm at a point distant from the elbow axis of rotation (Fig. 1B-1). This tension from the biceps muscle generates a positive moment about the elbow that can either help maintain the posture of the elbow or increase the amount of flexion.

Newton’s Laws Three basic rules of physics—Newton’s laws of motion— are used to describe the relationship between the forces applied to the body and the consequences of those forces on human motion (Table 1B-2).

STATICS

Newton’s Third Law (Action-Reaction)

Static analyses evaluate the external effects of forces on a rigid body at rest or during motion with a constant velocity. Applied to the body, static analyses are used to further determine the magnitude and nature of forces at joints and in the muscles. Forces provide both mobility and stability to the body, but also introduce the potential to deform and injure the body. Typically, healthy tissues are able to withstand changes in their shape, but a tissue structure that has been injured by disease or trauma may not be able to adequately sustain the same loads.

Newton’s third law states that for every action there is an equal but opposite reaction. This explains the idea that if a person pushes against a wall, it will in essence push back. Forces of action and reaction are equal in magnitude but in the opposite direction. This is important when upholding the principle of equilibrium and using tools that assess forces being applied to the body of interest. Reaction forces act to constrain motions by reacting to an applied force. A daily application of this concept is simple walking and running. Every time a person places a foot on the ground, there is a force exerted from the foot on the ground throughout the gait cycle. Simultaneously, however, a force of equal magnitude is exerted in the opposite direction from the ground up to the foot. This is termed a ground reaction force. Ground reaction forces are the forces applied by the ground due to the weight of the body. Similarly, within joints, there are muscles and connective tissues that create a joint reaction force (Fig. 1B-2). For the glenohumeral joint, there are both passive and active stabilizers that maintain the humeral head within the glenoid fossa.1 This has implications for surgical repair of Bankart lesions by changing the arc of the glenoid and tendon transfers in changing the line of muscle action. If these changes of the joint structure result in a smaller range of joint force angles acting through the humeral head, continual or increased instability could result. When a body is balanced by equal but opposite reactions for every action, the system is considered to be in a condition of equilibrium, which states that the net effect of the applied forces is zero as well as the net moment about any point. Although the sum of the force and moment vectors is zero, the body may still be moving at a constant velocity. Assuming a state of static equilibrium with a set of known force vectors, it is possible to determine unknown forces within a system. To do so, each force vector can be resolved into its individual components, such as the vertical (Fy) and horizontal (Fx) component for a two-dimensional coordinate system (Fig. 1B-3; see Fig. 1B-1D). Therefore, the

Force Vectors The most common vector quantity in mechanical systems is a force. A force is applied on an object to create either a pushing or pulling response. Depending on the original state of the object, a force can cause a stationary object to move or to alter the state of an object already in motion. Such forces can be either internal or external. Internal forces include those that hold a rigid body together, such as muscle tension generated within an extremity, whereas external forces are those applied to a rigid body. An example of an external force is the weight of an object being held in a person’s hand. Forces can be further categorized into two subgroups: contact (tension, friction, normal, external, internal) and distance (gravitational) forces. Distance forces act at a distance from the object or body with no direct physical contact. Additionally, forces can act in either a normal (perpendicular) or tangential (parallel) direction to the surface to which they are being applied.

Moment / Torque Vectors Vector quantities not only can represent translational motions in response to an applied force but also can represent rotation, twisting, and bending of an object. A moment (or torque) is determined by the magnitude of the force acting about a point and the length of the shortest distance

Basic Science and Injury of Muscle, Tendon, and Ligament

resulting equations for force and moment equilibrium in three-dimensions are: � Fx = 0

� Mx = 0

� Fy = 0

� My = 0

� Fz = 0

� Mz = 0

For static equilibrium, no linear or rotational motion occurs.

Ligament and Joint Contact Forces These concepts of static analyses can be applied not only to whole-body analyses but also at the joint and tissue level. For the joint, typically forces can be related to compression and shear. For example, the tibial plateau and femoral condyles experience compressive forces in the normal direction to each articular surface while standing with the knee in full extension. An increased shear force, however, is experienced during an anterior drawer test or when stopping quickly in a sporting event. Shear forces are experienced in the tangential direction along the tibial plateau. These forces not only are experienced between the bony structures but also are transmitted through the soft tissue structures. During an anterior drawer test or quick stop, the anterior cruciate ligament (ACL) can become significantly loaded to resist anterior tibial translation and provide stability at the joint. The ACL becomes loaded as a result of a generated tensile force within the ligament to maintain equilibrium. However, with an injury such as an ACL rupture, opposing forces cannot be transferred through the ligament, and thus equilibrium cannot be maintained at the joint without excessive translation, indicating anterior instability.

Free-Body Diagrams To better evaluate a biomechanical system, such as forces being applied to a specified part of the body, free-body diagrams are an effective tool to simplify a complex analysis. Free-body diagrams allow for visualization and ease of calculation of the problem by properly identifying all the forces and moments acting on the body of interest in order to successfully achieve equilibrium. This is done by first drawing the body of interest, then isolating the body from its environment, and only including the forces acting on the body. Again using the example of a biceps curl exercise to evaluate the elbow joint, the system should be drawn as only the forearm (radius and ulna combined as one) because this is the body segment of interest with the elbow as the axis of rotation (see Fig. 1B-1B). It is important to note the position and orientation of the object by defining a coordinate system. Applied forces are then identified and considered. Arrows representing the force vectors are drawn at every point where two (or more) bodies interact or join (see Fig. 1B-1C). Recalling that internal forces are those that hold a rigid body together, two examples within the human body as a whole are muscle contractions and bony contact at a joint line. Internal forces will be further examined with the concept of stress, which will be introduced in a later section of the chapter. External forces can be the weight of an object, friction, or gravity applied at the center of mass of a

87

body segment. In a simplified situation, flexion of the elbow is counteracted by the weight of the forearm itself and the weight of the dumbbell. The point of application of the forearm weight is at the segment’s center of gravity. Commonly, this point is determined from anthropometric data. The point of application of the biceps force on the forearm is its tendon insertion. The remaining force is the result of contact with the distal end of the humerus and the proximal end of the forearm whereby the point of contact is approximately at the joint surface of the bones. These forces can then be further resolved into their respective components along the previously identified x- and y-axes (see Fig. 1B-1D). In addition to forces, the forearm experiences various applied moments. Because the points of application for the forces are known, the distances of each force along the forearm from the axis of rotation—the elbow—is also known (see Fig. 1B-1E). Using these distances as moment arm values, the next step is to identify moments created by each force. The direction of the moment is relative to the applied force and its relationship to the axis of rotation. No moments are generated by the forces at the elbow because their moment arms are zero (see Fig. 1B-1F). Once all the forces and moments applied to the body or body segment of interest are identified, they can be summed to determine the resultant force and moment. When the body is in a state of static equilibrium, the resultant forces and moments sum to zero.

DYNAMICS Dynamic analyses evaluate bodies in motion and can be divided into subgroups: kinematics and kinetics. For dynamic systems, the forces and moments do not have a net value of zero, thus not satisfying the principle of equilibrium. Therefore, a different approach must be taken. Kinematics simply describes the motion of bodies without regard to the factors that cause or affect the motion by including the geometric and time-dependent aspects of the motion. Conversely, kinetics is based on kinematics but includes the effects of forces and moments. Motion analyses and sports mechanics typically involve dynamic systems.

Kinematics Simply stated, kinematics deals with motions without regard to forces and moments. These motions include translations and rotations. Translations are simply the linear motions in which all the parts of a rigid body move simultaneously in the same direction as every other point in that body. Rotations are the angular motions of a rigid body along a circular path and about an axis of rotation. During passive knee flexion, the tibiofemoral joint undergoes both linear and angular motions. As the knee progresses through the range of motion, the tibia experiences rotation in the sagittal plane, and simultaneously the point of contact between the tibial and femoral articular surfaces translates in the posterior direction. Although flexion is the primary angular motion, internal tibial rotation also occurs. This exemplifies the complex joint motion that can occur at a single joint.

88

DeLee & Drez’s Orthopaedic Sports Medicine

FR

FBiceps

A

FExt

FWArm

B y

y x

FR

x

FR FBiceps

FBiceps

FWArm

FWArm

C

FExt

D

FExt

y

y x

FR

x

FR

FBiceps

MBiceps MWArm MExt

a a b

E

b

FWArm c

c

FExt

F

Figure 1B-1   A, Simulation of a person performing a biceps curl. B, The dumbbell applies an external force (FExt) downward in addition to the downward force due to the weight of the arm (FWArm). The biceps muscle generates a force (FBiceps) to the forearm and causes a joint reaction force (FR) at the elbow to keep the joint stabilized. C, A free-body diagram of the forearm representing each force as an arrow, with the head of the arrow pointing in the direction of the applied force. D, These forces are then decomposed into component vectors. E, The varying moment arms for each force vector are identified with respect to the axis of rotation at the elbow. F, The biceps muscle creates a counterclockwise moment �� (MBiceps) �������������������������������������������������������������� to resist the clockwise moments due to the weight of the arm � (MWArm) ��������������� and dumbbell��� (MExt).�������   

Basic Science and Injury of Muscle, Tendon, and Ligament

  TABLE 1B-2  Newton’s Laws of Motion Applied to Static and Dynamic Analyses Law

Definition

Equations

An object at rest tends to stay at rest and an object in motion tends to stay in motion with the same speed and in the same direction ­unless acted upon by an ­external unbalanced force. Second The rate of change of the (acceleration) ­moment of a body is directly proportional to the applied force and takes place in the ­direction in which the force acts. Third For every action there is (action-reaction) an equal but opposite reaction.

SF = 0 SM=0

First (inertia)

FNET = m × a M NET = I × a

Degrees of Freedom Degrees of freedom are a set of independent movements that are needed to describe the position and orientation of a body. The musculoskeletal system has numerous degrees of freedom through which countless movements are accomplished. These movements can be performed in three-dimensional space through a set of translations and rotations. For example, the glenohumeral joint has a total of 6 degrees of freedom: 3 translations (superior-inferior, medial-lateral, anterior-posterior) and 3 rotations (internal-external, abduction-adduction, flexion-extension). However, depending on the analysis, the glenohumeral joint can be assumed to be a ball-and-socket joint, thus reducing the number of degrees of freedom to 3 rotations by constraining the translations. Similarly, the elbow only

89

allows for 1 degree of freedom when being considered as a hinge joint. Motion of joints is constrained by the shape of the articular surfaces or by stabilizing ligaments. Some joints, such as the knee and ankle, also allow for small amounts of passive translation that is usually achieved during a clinical exam or upon an injury of the joint. With an ACL injury, the knee may gain the ability to translate in the anterior direction, therefore adding a translational degree of freedom to the joint. Additionally, using reduced degrees of freedom is strategic when modeling or characterizing joints of the body. For example, when a person picks up an object from the floor and subsequently returns to an erect posture, each vertebra moves in three-dimensional space to achieve the desired posture, representing a large number of degrees of freedom. To evaluate lower back injuries using modeling techniques, the degrees of freedom can be significantly reduced by only considering the L5–S1 joint and assuming that the motion occurs only in the sagittal plane resulting in a two-dimensional analysis. Similarly, when assessing the tibiofemoral joint during a kicking motion, it could be assumed that the tibia only moves in two dimensions within the sagittal plane (Fig. 1B-4A).

Relative Motion When considering forces and moments applied to a body, it is important to know the relative relationship in terms of both position and orientation for the individual bodies to each other in addition to the overall environment. An illustration of this concept is that a moving object may appear to have a different motion from two different observers depending on their location to the moving object. Knowing relative relations allows for an accurate description of the system when solving for unknown forces by establishing a frame of reference. If a frame of reference is not identified, the measurements become irrelevant. A frame of reference

Figure 1B-2  Resulting joint reaction force at the glenohumeral (A) and patellofemoral (B) joints.

A

B Muscle/Ligament Force Joint Reaction Force

90

DeLee & Drez’s Orthopaedic Sports Medicine

Y

X

FR

FR

FRy

FRx

Figure 1B-3  Force vector (FR) representing the joint reaction force between the femoral head and acetabulum, and FR resolved into its individual component forces (FRx, FRy) in the x- and y-directions.

describes the position of one body with respect to another whereby a relative measurement can be performed. This measurement is made by comparing the change in ­position and orientation of the object to the reference frame. For example, the position and orientation of the femur can be defined relative to the pelvis when describing motions of the hip, or the tibia with respect to the femur during kicking (see Fig. 1B-4B).

Linear and Angular Kinematics The spatial components used to describe linear kinematics are position, displacement, and distance. Position is a vector that simply defines the location of the object in space relative to a reference frame, which could be the center of a joint or a point of contact. Once the initial position changes, the object has moved a specified distance and displacement. Distance is a scalar quantity that describes the length of the entire path traversed. Displacement is a vector quantity (D) defining the shortest distance (straight line) between the starting and ending positions of the object independent of the path taken (see Fig. 1B-4C). Additionally, temporal descriptions of linear kinematics are speed, velocity, and acceleration. Velocity is a vector quantity (v) describing the rate of change of the position with respect to time, whereas speed is a scalar quantity equal to the magnitude of the velocity vector (see Fig. 1B-4D). Furthermore, acceleration is also a vector quantity (a) that describes the

rate of change of velocity with respect to time but is commonly used to express an increase or decrease in speed (see Fig. 1B-4E). Angular kinematics is experienced when there is a change in angular position such that the body undergoes rotational motion about an axis of rotation. The angular distance through which the body moves is equal to the length of the angular path. Angular kinematics can describe rotatory motion for a body segment, such as the lower leg during a kick (see Fig. 1B-4A), in addition to the body as a whole, such as for gymnasts swinging on the bar to perform a giant circle. Descriptors used for these motions with respect to time are angular position, displacement, distance, velocity, and acceleration. These terms are similar to those used for linear kinematics but during a rotatory motion (see Fig. 1B-4C to E). The lower leg undergoes large changes in angular velocity (ω) and acceleration (α) throughout the range of motion for activities such as running, swimming, or kicking a soccer ball. Most general movements in the body are a combination of both linear and angular motion. For example, during a normal gait cycle, the lower extremities translate and rotate. Similarly, during athletics, combined motions are clearly illustrated during pitching of a ball as the glenohumeral joint creates an instantaneous axis of rotation as the ball is swung along an imaginary arc about the shoulder in addition to being propelled forward.

Joint Motions Motion at an articular surface can be described in terms of three motions that exist as a result of convex surfaces moving on a concave surface (Fig. 1B-5). Rolling motion occurs when the bone with the convex surface rotates to cause a change in the point of contact for both articular surfaces in addition to a corresponding linear motion. A ball experiences this kind of motion when it is rolled across a smooth surface. A sliding motion is experienced when one articular surface translates across the other with no rotation and progressively changes the point of contact. Sliding occurs when a box is pushed across a smooth surface. Spinning motion occurs when there is a single point of contact on the fixed surface and the point of contact changes on the rotating surface that does not undergo any linear motion. Spinning can be experienced on an automobile tire when the tire is rotating but the automobile is not moving forward or backward because of ice on the road. There are certain joints in which there is a combination of two or all three of these motions. The motion achieved at a joint is based on the shape of the articular surface. At the tibiofemoral joint during flexion, the knee experiences femoral condylar rollback on the tibial plateau. This phenomenon encompasses all three motions (rolling, sliding, and ­spinning) through flexion rotation, posterior translation, and external rotation, respectively, of the femur ­relative to the tibia.

Kinetics Kinetics is the branch of mechanics that describes the effect of forces and moments on the body by utilizing Newton’s second law of motion. More specifically, by using kinetics, it is possible to determine the forces and moments on a joint produced by mass, muscle tension, soft tissue loading, and externally applied loads. It is also possible to ­determine

Basic Science and Injury of Muscle, Tendon, and Ligament

91

A

y

y 1

1 x

x

D 2

2

�

C

B

y

y

x

x v

2

a

2

E

D

y y

x 2

F

x

M F

I·� 2

m·a

G

Figure 1B-4  A, A sequence of leg positions during kicking of a ball. B, The motion of the leg in the acceleration phase of kicking in the sagittal plane at position 1 and position 2. C, Displacement (D) and rotation (θ) of the tibia during the motion. D, The linear (v) and angular (ω) velocity vectors. E, The linear (a) and angular (α) acceleration vectors in the acceleration phase of kicking. F, The linear and angular inertial forces with the applied force (F) resisted by the inertial force (m·a). G, The applied moment (M) resisted by the angular inertia (I·α).

92

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 1B-5  Three fundamental motions that occur between articular surfaces. The point of contact changes on both articular surfaces during rolling motion. The point of contact on the moving surface remains constant during sliding motion. A single point of contact occurs on the fixed surface during a spinning motion. Some joints, such as the tibiofemoral joint, experience up to all three   of these motions simultaneously. Rolling

Sliding

Combined

Spinning

the appropriate range of loading to a joint that can be ­considered noninjurious. However, it is also possible to identify situations that produce excessively high moments and forces that exceed these limits, thus leading to musculoskeletal injuries. If an excessive valgus moment is experienced at the knee due to contact during a sport such as football, the medial collateral ligament could become overloaded and rupture in an attempt to maintain stability and resist the applied load. High forces can also be applied internally by overloading tendons as a result of extreme muscle tension.

Newton’s First Law (Inertia) Newton’s first law states that an object at rest tends to stay at rest and an object in motion tends to stay in motion with the same speed and in the same direction unless acted upon by an unbalanced force. This is also commonly referred to as the law of inertia. Therefore, a nonzero resultant force must act on a rigid body to change its velocity. Headrests are placed in cars to prevent whiplash injuries during rearend collisions by stopping motion of the head. This is an example of decreasing linear velocity and acceleration by application of an unbalanced force. Furthermore, if the

resultant moments and forces are zero, then the body will also have no rotational or linear acceleration. Linear acceleration can exist as an extremity is sent through a range of motion during a task (see Fig. 1B-4F ). In sports, athletes frequently control their mass moment of inertia or center of mass of their entire body by altering the positioning of their individual body segments to achieve stability or a particular motion. Gymnasts and divers use this concept to achieve multiple somersaults while in the air by tucking in their head and limbs closer to the center of their body. This is an example of angular acceleration due to changes in angular inertia (see Fig. 1B-4G).

Newton’s Second Law (Acceleration) Newton’s second law pertains to the behavior of objects for which all forces are not balanced. Therefore, the resultant force is not equal to zero, and acceleration in the direction of the applied force will occur. The magnitude of the acceleration is proportional to the magnitude of the resultant forces or moments applied to the body. Thus, in essence, the first law is a special case of the second law: � F = m · a � M = I · α

Basic Science and Injury of Muscle, Tendon, and Ligament

93

F

Ao

F

L Lo

F

Figure 1B-6  Tensile loading along the longitudinal axis of the anterior cruciate ligament causes a change in length from its initial length (L0) to an elongated state (L). A0, cross-sectional area; F, applied force.

Again, gymnasts and divers use the second law to increase or decrease the angular velocity of their spin when completing a somersault motion.

MECHANICS OF MATERIALS To evaluate both static and dynamic systems, it is assumed that the bodies or body segments are rigid. This section better describes biologic materials of the musculoskeletal system in a more realistic manner by considering that load is experienced at the tissue level and that these tissues are deformable. These biologic materials include both soft (articular cartilage, tendons, ligaments, capsular tissues) and hard (bone) tissues. Tissue structures that have been weakened by disease or trauma may not be able to ­adequately resist loads applied. To evaluate these structures, mechanics of materials can be addressed in two analytical approaches: structural and mechanical properties.

Structural Properties The mechanics of a structure that include multiple materials or tissues represent its structural properties, and the response of these complexes to tensile, shear, and compressive loading can be examined. For example, the structural properties of the femur–anterior cruciate ligament–tibia complex can be determined in response to a tensile load to assess its load-elongation behavior. To do this, a tensile force (F) is applied to the bone-ligament-bone

complex, causing the tissue to become stretched until the complex ruptures. While loading is applied, the corresponding increase of length is measured (Fig. 1B-6). The resulting nonlinear load-elongation curve that is typical of biologic soft tissues consists of four primary regions (Fig. 1B-7A). As the tensile load is first applied, the relationship between load and elongation is nonlinear and is referred to as the “toe” region. This is the level of loading typically ­experienced during a clinical exam as normal fiber recruitment occurs. However, with increasing loads, the relationship becomes more linear and represents loading experienced during daily and sporting activities. The slope of this ­ linear region of the curve is known as the stiffness. When the soft tissue begins to sustain more load than the structure can support, plastic deformation begins to occur. Finally, the entire structure approaches the failure region and completely ruptures. The point at which failure occurs is also known as the ultimate load. Another structural property is the energy absorbed as a result of loading of the tissue, which is determined by the area under the load-elongation curve.

Mechanical Properties It is also important to understand the mechanical response of an individual tissue or material, which is independent of specimen geometry, specifically of the cross-sectional area and initial length, by using normalized load and ­deformation parameters. Mechanical properties can be used to ­evaluate

94

DeLee & Drez’s Orthopaedic Sports Medicine

7

3

8 6 4

1

Stress (Pa)

Load (N)

2

5

A Strain Elongation (mm) B Figure 1B-7  A, Load-elongation curve of a bone-ligament-bone complex in response to tensile loading that characterizes its structural properties. Regions 1, 2, 3, and 4 correspond to the toe region, linear region, partial failure of the complex, and complete rupture of the complex, respectively. B, Stress-strain curve of the ligament substance that characterizes its mechanical properties. Regions 5, 6, 7, and 8 correspond to the toe region, linear region, partial failure of the material, and complete rupture of the material, respectively. the quality of the tissue when making comparisons between normal, injured, and healing states and are represented by the stress-strain relationship. Stress is defined as the amount of force applied per unit area and is one of the most basic engineering principles. Strain can be considered a measure of the degree of deformation and is defined as the change in length per unit length. Specimens with a greater length can withstand more total deformation, whereas specimens with a larger cross-sectional area can carry loads of greater magnitude. The mechanical properties can be derived from a plot of the stress and strain data and may include modulus, ultimate strength, ultimate strain, and strain energy density. Stress-strain relationships are obtained experimentally during tensile, compressive, or shear loading of the excised tissue. Four distinct regions exist along a stress-strain curve for biologic tissues (see Fig. 1B-7B). A typical stress-strain curve begins with a nonlinear toe region (region 5). Stretching of the crimped collagen fibrils occurs within this region as the fibers are being drawn taut before significant tension can be measured. Strain becomes linearly proportional to stress in region 6, and the slope of the curve in this region can be calculated to determine the tangent modulus of the tissue. The area under the curve within this region can be referred to as the elastic deformation energy. The tissue returns to its original length or shape once the stress is removed within this zone that is usually reached during most daily activities. Furthermore, the energy used to deform the tissue is Figure 1B-8  Applying a shear force (F) to the top of a cube results in deformation from the y-axis through an angle, γ, in the direction of the load.

released when the applied stress is removed. When the tissue experiences extreme and abnormally large strain, the tissue undergoes only a marginal increase in corresponding stress (region 7). It is at this point that the tissue begins experiencing microscopic failures. The area under this region of the curve represents plastic deformation energy. Once the tissue undergoes this amount of deformation, the tissue does not recover and return to its original state in its entirety upon release of the deforming stress. If the tissue continues to deform, it will eventually experience complete failure (region 8). The mechanical properties of a tissue can be used to evaluate injured and healing states. When evaluating ligament repair, the medial collateral ligament has much poorer mechanical properties during the healing process after rupture, although it shows some improvement over time without full recovery.2 To determine the mechanical properties of a knee cruciate ligament, a tensile load is applied along the long axis of the excised tissue sample after measuring the original crosssectional area of the tissue (see Fig. 1B-6). When a load is applied in the direction normal to the tissue structure such that there is only a linear change in deformation, the load and deformation are considered to be axial stress and strain, respectively. As the load is applied along the long axis of the ACL in Figure 1B-6, a decrease in the cross-­sectional area at the mid-substance occurs in response to an increase in the overall length (L) of the ligament from the initial length (L0). Compressive loading is commonly used for bone or articular

y

F

y

�

x

x

Basic Science and Injury of Muscle, Tendon, and Ligament

95

positive x- and y-axes, the shear strain is said to be positive. Experimental shear loading of knee articular cartilage has shown that loading parallel to the articular cartilage surface may lead to cartilage degeneration and osteoarthritis.3 Load (N)

Viscoelasticity

Elongation (mm) Figure 1B-9  A, Load-elongation curve of a biologic soft tissue in response to the application of a tensile load. The area between the loading and unloading curves represents the energy absorbed by the tissue during this loading regimen, or hysteresis.

Load (N)

Elongation (mm)

cartilage, whereas tensile and shear testing may be used for connective structures (­tendons, ligaments, capsular tissue). Shear stress and strain act parallel to the surface of the tissue (Fig. 1B-8), and the body is observed to deform into a rhomboid with sides equal to one if the body is originally a unit cube. Shear strain can be measured as the angular change (γ) at any point in the body owing to the applied shear force (F). When the deformation involves a reduction of the angle formed by the two faces oriented toward the

A tissue is viscoelastic when it possesses to some degree both solid-like characteristics, such as elasticity and strength, and liquid-like characteristics, such as flow depending on ­temperature, time, rate, and amount of loading. Most tissues within the musculoskeletal system demonstrate at least some degree of viscoelasticity. After plotting the load-elongation results of a nondestructive tensile load, there remains a region between the loading and ­ unloading curves that is known as hysteresis and clearly shows the time-dependent effects that viscoelasticity introduces (Fig. 1B-9). Preconditioning with repeated loading and unloading of the tissue decreases this area of hysteresis and maximizes elongation of the tissue. This is important in sports because athletes ­perform ­ preconditioning of the tissues in their bodies by completing repetitive stretching activities. A viscoelastic material experiences the phenomenon of creep and stress-relaxation. Creep describes a progressive increase in elongation of a material when exposed to a constant load over time (Fig. 1B-10A). Simply, creep can be considered the tendency of a material to move or deform in response to a constant stress. Conversely, with stress-relaxation there is a decrease in load over time upon application of a constant elongation (see Fig. 1B-10B).

Time (sec)

Load (N)

Elongation (mm)

Time (sec)

A B Time (sec) Time (sec) Figure 1B-10  Viscoelastic phenomena exhibited by biologic tissues include creep (A), response due to a constant applied load and (B), ������������������������������������������������������������������������������������ stress-relaxation, response due to a constant applied elongation over some time.

96

DeLee & Drez’s Orthopaedic Sports Medicine

Stress-relaxation is the phenomenon that occurs in a material to relieve stress under a constant strain due to the liquid-like characteristic of viscoelasticity. For both creep and stress-relaxation, after a period of time the tissue will reach an equilibrium state of elongation and load, respectively. A practical use of both stress-relaxation and creep in the clinic is for initial graft tensioning of ligament or tendon reconstructions. Over time it is impractical to expect that the initial graft tension will be maintained after being fixed. The length of the graft will inevitably increase from its original length due to a creep response. Moreover, the rate of loading is important because a faster loading rate will result in a greater stiffness. Viscoelasticity is also experienced in the articular cartilage of the knee. For example, as the rate of compressive loading increases, such as ­during running activities, the tissue becomes much stiffer.4 This increase in stiffness allows for improved protection to the underlying bone while forces at the joint are high. Overall there is not a substantial strain rate dependency on tendons and ligaments; however, bones exhibit large changes in stiffness with increases in the rate of loading. In addition, bone has a greater compressive strength than its tensile strength. This is advantageous during sporting activities when a sudden powerful blow may occur to a portion of the body.

C

r i t i c a l

P

o i n t s

quantities describe only magnitude and are used for concepts such as time, speed, mass, temperature, volume, work, power, and energy. In contrast, vector quantities have both a magnitude and a direction component associated with them. l Static analyses evaluate the external effects of forces on a rigid body at rest or during motion with a constant ­velocity. l Three basic rules of physics—Newton’s laws of ­motion— are used to describe the relationship between the forces applied to the body and the consequences of those forces on human motion. l Newton’s third law describes ground reaction forces, which are forces applied by the ground due to the weight of the body and joint reaction forces, which are created by muscles and connective tissues within the joints. l Dynamic analyses evaluate bodies in motion and can be divided into subgroups: kinematics (motion without regard to forces and moments) and kinetics (includes the effect of forces and moments). l The law of inertia states that an object at rest tends to stay at rest and an object in motion tends to stay at the same speed and direction unless acted on by an unbalanced force, such that a nonzero resultant force must act on a rigid body to change its velocity. l Newton’s second law pertains to the behavior of objects for which all forces are not balanced, thus the resultant force does not equal zero, and acceleration in the direction of the applied force will occur. The magnitude of the acceleration is proportional to the magnitude of the resultant forces or moments applied to the body.

l To evaluate biologic materials, including both soft and hard

tissues, mechanics of material can be addressed in two analytical approaches: structural and mechanical ­properties. l The mechanics of a structure that include multiple materials or tissues represent its structural properties and the response of these complexes to tensile, shear, and compressive loading can be examined. l Mechanical properties can be used to evaluate the ­quality of the tissue when making comparisons between normal, injured, and healing states and are represented by the stress-strain relationship.

CONCLUSIONS This chapter introduced the basic terminology and concepts of statics, dynamics, and mechanics of materials using examples relevant to sports medicine. With these foundations, the reader should be able to communicate using common units of measure to describe various physical quantities. The concepts also present how to approach both a static and dynamic system to calculate unknown forces and moments experienced on the body by use of a free-body diagram. A description of biologic materials within the musculoskeletal system is also presented. Although biomechanics has great depth and breadth to it, these concepts should allow the reader to begin making the link between sports medicine and biomechanics.

l Scalar

S U G G E S T E D

R E A D I N G S

Abramowitch SD, Papageorgiou CD, Debski RE, et al: A biomechanical and histological evaluation of the structure and function of the healing medial collateral ligament in a goat model. Knee Surg Sports Traumatol Arthrosc 11(3):155-162, 2003. Beaupré GS, Stevens SS, Carter DR: Mechanobiology in the development, maintenance, and degeneration of articular cartilage. J Rehabil Res Dev 37(2):145-152, 2000. Inman VT, Saunders M, Abbott LC: Observations on the function of the shoulder joint. J Bone Joint Surg Am 26:1-32, 1944. Lippitt S, Matsen F: Mechanisms of glenohumeral stability. Clin Orthop 291: 20-28, 1993. Nordin M, Frankel VH: Basic Biomechanics of the Musculoskeletal System, 2nd ed. Philadelphia, Lea & Febiger, 1989.

R eferences Please see www.expertconsult.com

97

Basic Science and Injury of Muscle, Tendon, and Ligament

S ecti o n

C

Design and Statistics in Sports Medicine S. Raymond Golish, Carl G. Mattacola, Darren L. Johnson, and Mark D. Miller

The purpose of this chapter is to present elements of study design and data analysis useful for conducting sports medicine research. The generation of a question, the application of a design, and the systematic collection and analysis of data are principle components of the research process. The practical importance of understanding study design and statistics is to allow clinicians the opportunity to understand (1) what characteristics should be considered in the planning stage, (2) how to collect data, and (3) how to draw inferences by analyzing the data. The importance of understanding statistics and research design has been supported by a supplemental issue of the American Journal of Sports Medicine.1 This supplement is a primer for clinicians who are not immersed in statistical concepts on a daily basis. Similarly, generation of a clinical research agenda has been undertaken by the American Physical Therapy Association (APTA).2 The first step in initiating sports injury research is to formulate a question.

THE RESEARCH PROCESS Developing the Question Clinical experience and an understanding of the literature are vital in developing the research question. Characteristics of a good research question have been described by the mnemonic FINER: feasibility, interesting to the investigator, novel, ethical, and relevant.3 The feasibility of the study must be addressed: Are the resources and the expertise available, and can the study be completed in a timely manner? The primary investigators must be committed and interested in the question because completing research can be arduous. The research topic should be novel or unique so that the question contributes to the body of knowledge in sports medicine. Ethical considerations related to the question and the intended study population must also be addressed. Finally, the relevance of the question to the profession may lead the researcher to decide on its importance and the possible need for further study.

Developing the Research Team As the researcher is deliberating over methods and procedures, care should be given to developing a sound research team. As in sports, the presence of a good leader and a strong and capable surrounding team is vital in research. It is important to include members whose strengths will contribute to the overall effectiveness of the study. Each member can and should have unique characteristics that

strengthen the research team. In many sports medicine facilities, a research scientist may oversee all aspects of the research process. The research scientist would work closely with the group of physicians who evaluate, refer, and treat those subjects that meet inclusionary criteria. Similarly, a group of athletic trainers or physical therapists are available to oversee all data collection and to provide rehabilitation services. The cooperation of all members of the research team is vital to ensure quality science.

Subject Selection When one is considering subject selection and methodologic procedure, the goals should be to minimize random and systematic error and to increase the chance of finding a difference by considering sample size, number of variables measured, and extent (time) of measurement.4 Determining what subjects will be used is akin to determining what the selection criteria will be. Inclusion and exclusion criteria should be considered a priori. Careful attention to inclusion and exclusion criteria allows the researcher to focus efforts on the hypothesis and to decrease extraneous competing variance. Sampling of the study group should be considered. The ideal standard is a randomized study; however, in most sports injury research, the sample may be randomized from a sample of convenience (i.e., a sample of available patients during a defined period). The sample of subjects should be representative of a larger population; this population may be equivalent to patients from a specific sports medicine clinic or may represent a specific age or activity group for a region or nation.

Selection of Variables The selection of dependent and independent variables must be given careful consideration and should be part of the hypothesis and research design phase (Fig. 1C-1). All too often, a sports injury researcher will have an important clinical question that once defined is unfeasible or immeas­ urable. Determination of the dependent variable (the meas­ urable variable) and the independent variable (that which is manipulated) is an important first step when designing a protocol. For example, to assess the effect of previous ankle sprains on postural stability scores, one could study a team of soccer players from preseason to postseason. The dependent variable would be the equilibrium score measured on a NeuroCom Balance Master, and the independent variable could be the presence of previous second-degree or greater ankle sprains. After outlining the question of interest, the dependent variables, and the ­ independent variables, other competing variables can be eliminated, thus

98

DeLee & Drez’s Orthopaedic Sports Medicine Idea

Develop idea

Draft statement of problem

Revise statement

No

Review the literature

Analyze results

Verify uniqueness of the question

Make a decision regarding the hypothesis

Feasible question ?

Yes Develop implementation plan

Research design

Integrate findings with accepted practice

Draw conclusions based on current knowledge base

Disseminate results

Execute plan

Pilot test

Collect data

Figure 1C-1  Implementation of the research process. The research process begins with an idea that is transformed into   a statement of the problem. Following a review of the literature, a research design is developed, and the study is implemented. The study involves collection of data, analysis of results, and a decision regarding the original hypothesis; the findings then can be integrated and compared with accepted practice in order to form conclusions.

Basic Science and Injury of Muscle, Tendon, and Ligament

r­ educing potential error. Population refers to the collection of all possible measurements (or data) that could be used to address a study question.5 Determining characteristics of the population of interest is important. Patients reporting to a specialty clinic may possess different characteristics than patients reporting to a primary care facility (e.g., they may be more familiar with their orthopaedic disorder, more motivated, or possibly better educated).

Study Design Once the question, team, subjects, and variables have been identified, a critical question is which study design can or should be used. Randomized clinical trials, case series, and cohort studies are some of the choices. The section on study design examines these choices in detail.

INTRODUCTION TO STATISTICS Populations and Samples A population is a collection of all possible measurements that could be used to address a hypothesis. A sample is a subset of a population that is analyzed with statistics because it is usually not possible to study the whole population. Inclusion and exclusion criteria define the sample and the population to which the study applies. Statistics inference is the process of drawing conclusions about the population from analysis of the sample. Ideally, the sample is a random subset of the population so that any member of the population is equally likely to be represented. To the extent that the sample is not random, the inferences will be biased. Several different types of bias are discussed later.

Random Sampling Simple random sampling results in determining all members of a given population and randomly choosing a sample. Systematic sampling involves selecting subjects based on some systematic criteria, such as selecting every third subject from a list of the whole population. Stratified sampling involves separating the population into subgroups based on some accepted criteria (injured versus uninjured) and then taking a random sample from the subgroup. Cluster sampling involves taking a random sample of clusters from a larger population, such as sampling clusters of subjects based on their zip code.

Nonrandom Sampling Nonprobability sampling is common when assessing samples from a single hospital or clinic. A nonprobability sample must be as representative of a population of interest as possible.3 Consecutive sampling involves taking all subjects who meet inclusion criteria on a consecutive basis. Convenience sampling involves taking subjects who are most accessible. Although this method is most common in sports medicine research, the representativeness of the sample must be weighed and considered a possible limitation of the study. Judgmental sampling involves selecting patients based on the desire of the primary investigator and is common in many

99

surgical studies. For example, one group of orthopaedic surgeons (group 1) might critically question the results of another group (group 2) when the results of the reported surgical study seem unusually high and are not reproducible. The fact, however, in many of these comparisons, is that group 2 may have very strict guidelines for inclusion criteria for the surgical group. Likewise, the inclusion criteria may often depend on the judgment (gut feeling) of the attending physician. Although group 2 may have listed their inclusion criteria and stated the limitations of the study, it is often difficult to replicate the judgmental component of whether a patient is a successful candidate for surgery.

Inference, Hypothesis Testing, Decision Analysis In statistical inference, the goal is to estimate a population quantity from a sample. A point estimate is a single number that provides the best estimate in some sense. A confidence interval is a range of values that represents an interval in which the value is likely to lie (e.g., with 95% confidence). Inference is one large branch of statistics. In hypothesis testing, the goal is to derive a statistical judgment on whether a given question is true (e.g., Is the treatment effective?). In decision analysis, the goal is to combine the results of inference and hypothesis testing with information about outcomes to determine an optimal treatment.

Clinical Research Studies Clinical research studies are addressed in detail in a subsection. In observational studies, the management of clinical patients as it currently exists is the object of study. In experimental studies, the management of patients is altered in order to conduct an experiment. By far the most important experimental study is the randomized controlled trial (RCT), in which a single variable (e.g., a treatment) is randomly assigned. The goal is to assess the effect of the treatment on some outcome variables. Other variables than the treatment may also affect the outcome. These other factors in the design are often called confounding variables because they may also influence the conclusions if not properly addressed. In experimental studies, randomization controls confounding variables.

Clinical Instruments Instruments are outcome variables designed by experts to assess function, pain, or other clinically important judgments. They usually have multiple components involving multiple-choice questions, scales, or ratings completed by patient or clinician. An instrument is valid if it measures on average the outcome of interest, such as the familiar WOMAC score for lower extremity function. An instrument is reliable if it produces similar scores in similar situations. Validity in clinical instruments is an example of a general notion of accuracy in statistics. A measure is accurate if it measures the quantity of interest on average. A measure that is not accurate is biased. A measure is precise if it produces very similar outcomes under similar circumstances. A measure that is not precise has high variability. In a diagnostic study, intraobserver variability may occur

100

DeLee & Drez’s Orthopaedic Sports Medicine

between successive observations by the same ­ surgeon. Interobserver variability occurs between observations by different surgeons.

Bias and Error No study is free of bias or error, but researchers should attempt to reduce error by considering and managing all controllable error. Random error is a wrong result due to chance.1 Occasionally, chance may produce a number of cases with phenomena that are not representative of a true occurrence. The simplest and best way to decrease random error is to increase the sample size and the effect size (see Appendix 1C-1). Systematic error is the wrong result due to bias or can include types of error that can influence or distort the findings in one direction.3,5 Random and systematic error can result from sampling and the measurement techniques. It is important to draw a random sample of the study population to reduce bias. Bias is a condition or ­phenomenon that, when introduced, influences the meas­ urement and analysis.5 Bias is a flaw in impartiality that alters the manner in which measurement, analysis, or assessment is recorded, conducted, or interpreted. Randomization controls the bias that would otherwise be introduced by confounding variables, but other types of bias exist. Blinding controls bias that arises when orthopaedic surgeons have information about study subjects that may compromise their impartiality, however subconsciously or subtly. Meticulous attention to the study protocol is also important in minimizing bias. Selection bias is introduced by a nonrandom selection of a sample from a given population. Observational or informational bias occurs when a measurement or outcome is affected by the characteristics of the study group itself. There are numerous potential sources of bias, which are given special names. The most important issue is to understand that potential biases exist and research ways to minimize or assess their effect.

Validity Validity is the ability to state with confidence that the results were due to the experiment and that those results have some larger inherent generalizability. External validity, which is defined as the generalizability of the study, addresses the extent to which the measurable variables can be generalized to the population. Internal validity is the extent to which the findings are the result of the intervention.6 Did the variable that was manipulated cause the effect that was found?7 Validity may be further differentiated into four categories: face validity, criterion-related validity, content validity, and construct validity. Face validity is a subjective assessment of whether a measurement makes sense intuitively. Criterion-related validity relates to assessing a technique by comparing it with a gold standard measurement.6 Content validity refers to whether an instrument accurately reflects the domain of interest. In orthopaedics, the assessment of pain and function are important concepts. Because there is no universal method of assessment, content validity of a new pain and function measurement questionnaire can be assessed in two ways: (1) a panel of experts could assess the instrument until they agreed that

the scale covered the domain of pain and function, or (2) the ­instrument could be compared with a scale that is currently in use and already validated. In either case, there is no universal agreement of the content in each domain, and therefore each scale is suspect to error. Construct validity is the comparison of a theoretical concept with other concepts that have been supported. A construct is a theoretical concept that has been given a title to describe it, such as pain and personality. In this example, pain is a concept that has various components. Workers try to understand and classify pain with ­descriptors that they think are representative of the pain experience.

Reliability Reliability refers to the degree of reproducibility of a meas­ urement. It is the consistency of measurement or the extent to which the measurement yields the same measurements on repeated episodes by either the same tester or different testers.8 The measurement of any phenomenon always contains a certain amount of chance error. The goal of errorfree measurement, although laudable, is never attained in any area of scientific investigation.9 It is necessary to realize that because repeated measurements never exactly equal one another, unreliability is always present to at least a limited extent.8 Experimentation in orthopaedics and sports medicine typically uses data that are continuous.10 Pearson’s product-moment correlation and the intraclass correlation coefficient (ICC) are used to assess agreement between two or more groups of continuous data. If the type of data is ranked, Spearman’s rank order correlation or Kendall’s tau can be used. If the data are nominal, Cohen’s kappa coefficient can be used to assess agreement. When examining agreement between continuous data, it is more prudent to use an ICC. An ICC is derived from the product of an analysis of variance (ANOVA) equation. Appendix 1C-2 includes an example that demonstrates how to calculate Pearson’s product-moment correlation and an ICC for a hypothetical data set. For a more detailed description of how to calculate and choose the correct formula, see Denegar and Ball,11 and Shrout and Fleiss.12 In brief, the ICC is more sensitive to changes between trials or subjects and is not fooled by systematic error like Pearson’s productmoment correlation and therefore is recommended as the process of choice when calculating reliability.

Accuracy and Precision Accuracy is determined by how closely the individual meas­ urement comes to the true phenomenon.13,14 Precision is the degree of error that accompanies a measurement.14 These concepts are closely related to the ideas of reliability and validity discussed earlier but most often serve as the terms used in a numerical context. Figure 1C-2 demonstrates the relationships between reliability and validity, similar to accuracy and precision.

Correlation and Causation Probability and statistics in general are concerned with dependency and correlation. Dependency occurs when two or more variables are known to affect each other,

Basic Science and Injury of Muscle, Tendon, and Ligament

1

2

101

3

Figure 1C-2  The bull’s-eye example can be used to document validity and reliability. In screen 1, the cluster of shot is reliable but not valid if the goal was to reach the center circle. In screen 2, the cluster of shot is more valid but not reliable. In screen 3, the cluster of shot is both valid and reliable.���   (Adapted from Sim J, Arnell P: Measurement validity in physical therapy research. Phys Ther 73:102-110, 1993; with permission from the American Physical Therapy Association.)

without any knowledge of which is the cause and which is the effect. Correlation is the term given to two dependent continuous variables. Probability and statistics most often do not involve judgments about causation, with one important exception. Randomized controlled trials (RCTs) can prove causation directly, by isolating the randomized variable from the web of dependencies and correlations. There are sophisticated methods for attempting to prove causation without RCTs, but these are beyond our scope.

STUDY DESIGN Research can be partitioned into four categories: analytical, observational, experimental, and qualitative.15,16 Observational and experimental research are the most important types in sports medicine. A taxonomy of observational and experimental studies is presented in Figure 1C-3.

Observational Studies In observational studies, the researcher observes patients and management as they exist currently. There is no experimental intervention to alter management. These study types account for the majority of research in the surgical literature.

Observational studies may be further divided into ­ rospective and retrospective studies. A prospective obserp vational study is usually called a cohort study. In a cohort study, a group of patients are identified to receive a specific treatment, then treated, and their results are followed serially over time. The essential feature of a cohort is the prospective nature. The study design is considered first and a priori power analysis (see previous definition) is performed to determine the necessary sample size. The study is announced publicly (in a fashion) by institutional review board for human subjects. The treatment is administered. The outcomes of interest are followed at well-defined intervals. All data are gathered up to a minimum interval (e.g., 2 years’ follow-up) for analysis. In practice, all cohort studies involve a control group. A control group consists of patients who have the same diagnosis but different treatment, such as a gold standard treatment or conservative management. Ideally, the control group is defined prospectively as well, and follows an identical clinical protocol, except for the treatment. It is possible to use a historical control group by identifying control patients from medical records. In some cohort studies, matching is used to identify control patients. In matching, patients are compared who have similar demographic or clinical variables such as gender, age, and comorbidity. Matching is a way of controlling confounding variables in observational studies.

Experimental Studies

RCT partial compliance/ randomization

Randomized Controlled Trial (RCT)

RCT with crossover

Observational Studies

Retrospective

Prospective

Cohort Case Control (+/– matched)

Case Series

Matched Cohort

Figure 1C-3  A taxonomy of experimental and observational studies. RCT, randomized controlled trial.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 12.)

102

DeLee & Drez’s Orthopaedic Sports Medicine

Prospective studies have the advantage of being a well-defined, principled observational study. But note that there is no randomization in the choice of treatment between the cohort and the control group. All the potential biases and confounders still exist in this type of study. The most common type of retrospective observational study is a case series. A case series presents information from multiple patients with a similar disease or condition. The data usually come from a review of charts looking back in time at what occurred with each patient. An advantage of case series is that they can generate speculative associations, which can be further tested by other study types. The major drawback is a lack of any control group to which to compare results. Statistical associations may be sought within the series patients, but exploratory statistics are most relevant here. An improvement on case series is the case control study, a case series with a historical control group to which comparisons can be made. Because many case control studies have relatively small numbers of patients, it is useful to choose control patients matched by gender, age, pathology, and so forth. Epidemiology is the study of the frequency and cause of disease. Such terminology must be clearly defined and stated in the methodologic approach to avoid confusion when distinguishing between rate and risk.

Rate Rate can be defined as the change in one phenomenon with respect to another variable, usually time.17 It is reported as a ratio, and in sports medicine, the numerator is defined as an injury or injured player. It is important to define the numerator so that similar data can be compared. If no reference is made to the population at risk, the rate is expressed as an absolute rate. Put another way, it is the instantaneous rate of change in the numerator per change in unit of time (the denominator).17 If the size of the population at risk is available (number of players), the rate would be expressed as a relative rate.17 A relative rate would be expressed in person-periods or, more specifically, in person-years. In sports medicine, a relative rate might be expressed as the number of injuries sustained per player-game. Two commonly published methods of calculating the rate are the absolute incidence rate and the relative incidence rate.17 The absolute incidence rate is the number of new cases of a disease that occur, usually expressed per year (Table 1C-1).17 One weakness of the absolute incidence rate is that it does not take into consideration the population size.17 Table 1C-2 provides hypothetical data for a soccer team over a period of 10 games. There were a total of eight lateral ankle injuries in 10 games, or an absolute incidence rate of 0.80 injuries per game. The absolute incidence rate can also be used to assess the number of players injured per game. In this example, six players were injured in 10 games for a 0.60 players per game incidence rate. The absolute incidence rate does not reflect the size or the experience of the at-risk population.18,19 If the characteristics of the population are not known, it is difficult to form a comparison with other studies. The relative incidence rate might be better used.

  Table 1C-1 Expression of Frequency of Injury Occurrence Prevalence:

Number of people who have the disease/injury at one point in time Number of people at risk at that time Incidence:

Number of new cases of the disease/injury over a period of time Number of people at risk during that period Absolute incidence rate :

Number of injuries Number of expossure-events (games)

Relative incidence rate :

Number of new cases Population tiime

Rate:

Change in a phenomenon Time Risk:

Probability that a player will incur an injury Time

The relative incidence rate is the number of new cases divided by the population-time.20 The relative incidence rate depends on the number of injured players or injuries (the numerator) and the population and the time period of player participation (the denominator).18 It can be defined as the instantaneous potential for change in disease or disorder status per unit of time.19 To calculate the relative incidence rate, use the number of cases (injured players, injuries), the population, and the period of interest (time). A number of methods can be used to assess the population-time. This example involves the exact population-time. In the soccer example, 10 players participated for a total of 79 player-games (populationtime). The total of eight injuries during the 79 playergames yielded a relative incidence rate of 0.10 (8/79) injuries per player-game results. The number of injured players can also be calculated. There were 6 players injured during the 10 games. The number of injured players (6) is divided by the player-game exposures (79) to yield the relative incidence rate of 0.07 (6/79). Although this information is useful for describing the number of lateral ankle injuries that occurred, it still may be difficult to compare this to other sports or teams that have a greater or lesser number of contests. Exposure variables should be specific and should permit the exposure of interest to be distinctly separated from other confounding variables.21 For these occasions, the relative incidence rate can be expressed as the number of injuries or injured players per 100 or 1000 player-exposures. In this example, the number of injuries per player-games can be converted to 10.1 injuries per 100 exposures and 7.59 injured players per 100 exposures.

Risk The risk for injury is clearly an important statistic for the sports injury researcher. Workers often want to know the risk of participating in a sport or the risk for reinjury following surgery or rehabilitation. Risk for injury is a proportion that is an estimate of a probability.19 Proportions are relative frequencies or fractions and often estimate or are probabilities of certain events. They are often identifiable

Basic Science and Injury of Muscle, Tendon, and Ligament

103

  TABLE 1C-2  Hypothetical Data (First-Degree Ankle Sprains) for a Soccer Team over 10 Games Status per Game Player

Previous Injury*

 1 Y  2 N  3 N  4 N  5 N  6 N  7 N  8 Y  9 Y 10 N Total games at risk New case(s) of ankle sprains Total new cases = 6

Position

1

2

3

4

5

6

7

8

9

10

Number of Games at Risk

F F C C G S G S D D

+ ⊕ X + + + + + O +

+ O X + + + + + + +

+ + X + + ⊕ + + + +

+ + + + + O + + + +

+ + + + + + + + ⊕ +

⊕ + ⊕ + + + + + O ⊕

O + O + + ⊕ + + O O

O + O + X O + + O +

O + + + X + + + + ⊕

O + + + X + + + + O

0

2

1

 6  9  5 10  7  8 10 10  6  8 79

1

1

1

+, Player is at risk; O, player is injured; X, player is not at risk owing to nonmedical reasons; ⊕, player is injured during game. *Identifiable anterior talocrural laxity during preseason physical. Modified from Schootman M, Powell JW, Albright JP: Statistics in sports injury research. In DeLee JC, Drez D (eds): Orthopaedic Sports Medicine. Philadelphia, WB Saunders, 1994, pp 160-183.

as a ratio in which the numerator is also included in the denominator (e.g., + b). Risk for injury is often incorrectly defined as an incidence rate, but it is in fact a proportion and would probably be better termed an incidence proportion.22 Incidence is the number of new cases that develop over a specific period of time.3,22 There are a number of techniques for calculating incidence. One method, the cumulative incidence, is the number of new cases of a disease during a given duration of time divided by the total cases without the disease at the start of the time period.20,22 The cumulative index is dependent on the time period of the study; subjects are followed for a certain period of time or until an injury occurs. In our example, 3 of the 10 athletes had previously suffered a lateral ankle sprain. At the conclusion of the season, there were a total of eight lateral ankle sprains; six were new sprains. The cumulative incidence for the season was 6/10 or 60%. Although helpful, the cumulative index does not account for changes in time. Important to consider is that incidence data compared between different studies may have very different time periods. Prevalence is the number of existing cases of a disease or an injury divided by the total population at risk at a given point in time.20 For example, during preseason physical examination, anterior instability of the talocrural joint was identified with the use of an anterior draw test. In Table 1C-2, 3 of the 10 athletes had previously suffered a lateral ankle sprain. At the time of preseason physical examination, the prevalence of lateral ankle instability due to previous injury was 30% (3/10). Sequentially, the descriptive study offers the most basic of needed information. Information regarding the frequency and the measures of central tendency (mode, median, and mean) can be obtained. A review of the literature will reveal these values from similar studies. If this information is not available or not universally known, the descriptive study is often the first step when designing an area of research.

Experimental Research Experimental designs are characterized by the inclusion of an independent variable that is manipulated in a controlled

manner.15 The manipulation of the independent variable allows the researcher to know if the treatment was effective. There are a number of designs that can be considered when conducting sports medicine research. The strength and the weakness of each design should be appreciated a priori so that the final outcome can be appropriately appreciated. This chapter describes various experimental and quasiexperimental designs. The depiction of the design is based on the work of Campbell and Stanley.7 In depicting the research design, an X is representative of an exposure, treatment, or manipulation. An O is representative of an observation or measure. This discussion reviews the various designs and highlights the advantages and disadvantages of each one. For a more detailed discussion, readers are encouraged to read the classic work by Campbell and Stanley.7

Pre-experimental Designs Pre-experimental or quasi-experimental designs are so described because they lack a control. Such a design involves studying one group of subjects without a comparison group. The most basic of the pre-experimental design is the one-shot case study. X O1

The case study has no control; therefore, the result of the treatment must be examined with a great degree of scrutiny, and findings can be considered speculative.15 The case study is invaluable in describing an unusual case or the treatment of an unusual case but lacks all scientific control. The one-group pretest-posttest design is better equipped to describe whether X caused the difference in O1. It is suspect to a multitude of rival hypotheses, however. The ability of a design to counter rival hypotheses is dependent on the ability of the design to account for factors that affect internal and external validity (see later discussion of validity). O1 X O2

In the one-group pretest-posttest design, confounding factors such as maturation, history, testing, and ­instrument

104

DeLee & Drez’s Orthopaedic Sports Medicine

decay are not accounted for. We cannot be assured that any change between O1 and O2 was not the result of getting older, an historical event, a learning effect due to testing, or instrument error. Although the one-group pretest-posttest design does not offer strict control, it is often used because the choice of withholding treatment with a control group is not a consideration. In the static-group comparison, there is no pretest score. In this design, it is not known if the difference between O1 and O2 is the result of selection. For example, if there is a difference between O1 and O2, it is not known whether the groups may have been different from the start. The intervention may have exacerbated or reduced the effect, yet one could never be sure.

is that the effect of testing and selection is not controlled. Therefore, the external validity of a study using this design is suspect. For example, scores might change as a result of being tested twice or as a result of selection to each group. The posttest-only control group design eliminates the initial pretest in the control group. As such, it helps eliminate any error or effect due to testing. Although it is important to assign subjects randomly in this design, there is no verification that subjects were equal at the beginning of the experiment. Therefore, one cannot be assured that a difference was due to the treatment. Research in sports medicine often involves repeated measurements of patients over time; therefore, this design would not be appropriate or used very frequently.

X O1

O2

O2

The static-group comparison does not control for maturity, selection, and mortality. This design is often used when a physician wants to compare one group of subjects who have received a surgical treatment with another group who did not. In conclusion, pre-experimental designs lack sufficient control of numerous threats to internal and external validity. The pre-experimental design lacks a control group and randomization, and by adding these components, the research design is strengthened considerably. Factors that, when included, encompass a true experiment are manipulation, control, and randomization. A true experiment involves the manipulation of an independent variable to assess if the manipulation caused a measurable change. The measurable change must be to a variable that can be manipulated (strength and pain perception) compared with an attribute such as age or gender.23 A control group is a necessary component of a true experiment to help reduce extraneous factors such that the investigator is assured that differences were due to the treatment alone and not to differences between the groups. Assignment of subjects to a control and an experimental group should be based on the concept of randomization.15 A true experiment and the statistical analyses that accompany it are dependent on the assumption that all subjects had an equal or a known chance of being assigned to a certain group.15 Randomization is necessary to decrease the potential for systematic bias.

Experimental Designs For all experimental designs discussed here, it is given that all groups are randomly assigned. A pretest-posttest control group design is the more popular of the experimental designs. Employing a control group that is followed prospectively allows control for factors that affect internal validity. O1 X O2 O3 O 4

By employing a control group and including a pretest, the researcher can be better assured that a difference between O4 and O2 was the result of the intervention (X). Another advantage is that this design is easy to employ and is not subject intensive as compared with the Solomon four-group design (see later paragraphs). The disadvantage in this design

X O1

The advantage of this design is that it requires fewer subjects and less time to complete. It can be an appropriate design if the effect of the treatment is large. If there is a large difference between O1 and O2, then a stronger case can be made that the treatment was effective. The inclusion of randomly selected subjects separates this design from the static-group comparison. The Solomon four-group design is a variation of the pretest-posttest control group design presented earlier. The Solomon four-group design employs two additional groups (as compared with the pretest-posttest control group design): (1) one group receives treatment and posttest, and (2) one group receives only the posttest. The advantage of this design is that the effect of testing is controlled. O1 X O2 O3 O 4 X O5 O6

In the Solomon four-group design, the first group assesses the change between O1 and O2 as a result of the treatment. The second group assesses the effect of (testing) no treatment between O3 and O4. The third group assesses the effect of the pretest on the outcome of O5, and the last group assesses the effect of no intervention or pretest on the outcome. The effect of the treatment is evident if there was a difference between O1 and O2, O2 and O4, O5 and O6, and O5 and O3.7 The strength of this design is that it effectively controls for all the threats to internal validity. The disadvantage is that it would be time-consuming, subject intensive, and expensive to perform a study with this design. A suggestion would be to design a study with the first three groups. By eliminating the fourth group, subject recruitment could be reduced, and an effect from the intervention in this design would provide strong evidence that the treatment caused a change in O5 and O2 when compared with O3 and O1. Other factors to consider when designing a study are whether the experimenters or subjects will be blinded. Blinding is a condition whereby the experimenter or the subject, or both, does not know what the intervention is or does not know who is receiving the intervention. A singleblinded study is one in which only the patient or the clinician is blinded. A double-blinded study is one in which

Single-Subject Research Design Single-case experimental designs offer a research approach that closely mimics a typical rehabilitation process.24 For example, when the patient/athlete reports with a problem, clinicians identify the disorder and then offer an intervention. The identification of a problem is similar to establishing a baseline. As the patient/athlete returns for treatment, the clinician continues to re-evaluate and then proceeds with the current treatment or sometimes alters the treatment. The repeated measurement and interventions used in everyday practice offer an ideal arrangement for performing single-case examinations. A single-case or single-subject experimental design is characterized by the following: identification of a baseline measure, repeated measurement of the dependent variable, repeated manipulation of the independent variable, a comparison within the individual across differing conditions, one or only a few subjects, a replication of effects, and ideally a measurement that is objective.25 In contrast, case studies are frequently reported in the literature and are considered a prescientific mode of reporting findings.26 Case studies are not classified as experimental designs because they lack control groups that would eliminate or minimize other possible sources of variations.27 Single-case designs are most commonly evaluated by examining the effects of the intervention over time using visual analysis. Statistical analysis may also be used in single-case designs, but such usefulness is often a subject of debate.28 A recent discussion of the statistical considerations in single-case designs, specifically related to the discipline of sports medicine, was undertaken by Bates29 and by Reboussin and Morgan.30 Similarly, interested readers may find a more comprehensive review of the topic in reports by Kazdin26 and Kratochwill and Levin.28 A number of different research designs are employed in single-case research, but the most common are the multiple baseline and reversal designs. In the multiple baseline design, the intervention is staggered in time (Fig. 1C-4)24 so that any effect that is evident can be linked clearly to the treatment and will not be an effect of testing. In a reversal design, the intervention is withdrawn, and the effect is plotted to assess whether the measure is affected (Fig. 1C-5).24 The interesting consideration is the addition and removal

30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0

Baseline

2

4

105

Treatment

8 10 12 14 16 18 20 22 24 26 28 30 32 34 Time Figure 1C-4  A multiple baseline design for single-subject research. Initiation of the treatment intervention is staggered   to demonstrate the effect of the treatment.

6

of the intervention. The effect of the design is most clear when the dependent variable changes with the intervention and returns to baseline or in an opposite direction during reversal. The disadvantages of the reversal design are obvious; the ethical dilemmas of removing an effective or a helpful treatment must be considered. The effect of the intervention is clear when systematic changes in behavior occur during each phase in which the intervention is being withdrawn or presented.24 Thus, the magnitude and the rate of change are evaluated. The magnitude of change is assessed by a change in mean or a change in level. The change in mean refers to a change in the arithmetic mean from one phase or condition to another (Fig. 1C-6A and B).24 A change in level refers to a shift or a discontinuity of performance from the end of one phase to the beginning of the next.24 A consistent change in level following the implementation or the withdrawal of an intervention indicates that the changes were a result of the treatment (see Fig. 1C-6C and D).24 The rate of change is accomplished by examining changes in trend and latency of the change. A change in trend or slope details systematic increase or decrease over time (see Fig. 1C-6E and F).24 Therefore, a change in trend is demonstrated by a change in the direction

Dependent Variable

both the patient and the clinician are blinded. Another consideration is that of reactive arrangements. Reactive arrangements are the potential source of unrepresentativeness due to the artificiality of the experimental setting.7 The presence of a laboratory setting or the intense scrutiny of the experimental procedures can create an arrangement in which the subject tries to outguess the physician or to perform according to a preconceived conception. The less obvious the connection between the testing and the experimental process, the less likely this reaction will be. Although paying careful consideration to the research design provides the researcher with more control over the independent variable and more ammunition to state that any change was the result of the treatment, the significance of having a sterile and perfectly designed study must be weighed against development of a study that lacks clinical relevance.

Dependent Variable

Basic Science and Injury of Muscle, Tendon, and Ligament

30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0

Baseline Treatment

Reversal

Treatment

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Time Figure 1C-5  A reversal design for single-subject research. Removal of the treatment (reversal) demonstrates the effectiveness of the treatment.

106

DeLee & Drez’s Orthopaedic Sports Medicine Change in Mean 10

Dependent Variable

10 9

9

8

8

7

7

6

6

5

5

4

4

3

3

2

2

1

1

0

0

1

2

3

4

5

6

7

8

9

10

0

0

1

2

3

4

5

6

7

8

9 10 11

2

3

4

5

6

7

8

9 10 11

B

A

Change in Level 12 11 10 9 Dependent Variable

8 7 6 5 4 3 2 1 0

1

2

3

4

5

6

7

8

9 10 11

C

0

0

1

D

Time Figure 1C-6  The magnitude of change is assessed by a change in mean. In A, the mean decreases from time 1 to time 2.   In B, the mean increases from time 1 to time 2. A change in level refers to a shift or discontinuity of performance from the end   of one phase to the beginning of the next phase. C and D show a change in level as treatment is applied or withdrawn.   (Adapted   from Mattacola CG, Lloyd JW: Effects of a 6 week strength and proprioception training program on measures of dynamic balance:   A single-case design. J Athletic Training 32:127-135, 1997.) Continued

in which the data pattern is moving.31 Latency of the change refers to the period between the onset or the termination of one condition and changes in performance (see Fig. 1C-6G and H).24,26 The sooner that changes in performance follow an intervention, the greater the confidence in the treatment effect. A researcher or clinician can be confident that a treatment is effective when there is a change in mean, level, trend, and latency of change following the intervention.

Qualitative Research Qualitative research involves intensive observation, which often includes a comprehensive interview process

in a natural setting.16 Qualitative research is also termed naturalistic inquiry that is ethnographic, interpretive, grounded, phenomenologic, subjective, and participant observational.16 Qualitative research designs are naturalistic because the researcher does not attempt to manipulate the research setting. Therefore, this type of research is in contrast to experimental research, which focuses on controlling and manipulating an independent variable. Qualitative research is inductive, and quantitative research is deductive. Inductive research involves beginning with observations and formulating a hypothesis, then generating explanations that support a theory. Deductive reasoning works in the opposite direction. In deductive

Basic Science and Injury of Muscle, Tendon, and Ligament Change in Trend 14

11 10

13

9

12

Dependent Variable

8

11

7

10

6

9

5

8

4

7

3 2

6

1

5

0

107

0

1

2

3

4

5

6

7

8

9 10 11

4

0

1

2

3

4

5

6

7

8

9 10 11

F

E

Latency of Change 9

8

Dependent Variable

8

7

7

6

6

5

5 4

4

3

3

2

2

1

1 0

0

1

2

3

4

5

6

7

8

9 10 11

G

0

0

1

2

3

4

5

6

7

8

9

10

H

Time Figure 1C-6—cont’d ­ A change in trend refers to a systematic increase (E) or change in the direction of data (F) over time. Finally, a latency of change refers to the period between the onset or termination of one condition and changes in performance.   The sooner the change (G and H) in performance following an intervention, the more confidence in the treatment effect.

reasoning, the researcher starts with a theory, develops a hypothesis, tests the hypothesis, and then compares this to reality (other studies in the literature). The point of understanding and using qualitative research is to understand phenomena in their naturally occurring states.32 The qualitative researcher attempts to replace the fixed treatment/ outcome emphasis of the controlled laboratory setting with the dynamic and never-changing environment of the real world.32 The experimental researcher is concerned with controlling the research setting, reducing threats of internal and external validity, and verifying precision of measurement; the qualitative researcher focuses on the capturing process, documenting variations, and exploring differences in experiences and outcomes.32 Although qualitative research is not formally done in the sports medicine setting,

it occurs when a medical history is recorded, it occurs when the patient is observed entering or leaving the facility, and it occurs when the patient undergoes therapy. Clinicians can learn from qualitative methodology to become skilled observers, to reduce bias during the interview process, and to relate to the patient without being sterile. All these qualities will improve the data collection process in the clinical setting and will make the interaction between clinician and patient a richer and more inclusive process.

Analytical Research An example of analytical research includes historical research, philosophical research, and meta-­analysis. In historical research, the researcher arrives at the answer to a

108

DeLee & Drez’s Orthopaedic Sports Medicine

question by the use of primary and secondary ­ literature, published and anecdotal facts, and personal recollection. Therefore, the historian must create a map from all the available facts to support a specific question that addresses the “how” and “why.” The historian works to disprove rather than to prove findings.16 Because the historian is working from a retrospective position, that researcher must strive to explain change without the advantages of controlling variables, settings, and interventions. Philosophical research is a type of research that is characterized by critical inquiry in which the researcher establishes hypotheses, examines and analyzes existing facts, and synthesizes the evidence into a workable theoretical model.16 Philosophical research involves the generation of ideals, questions, and values, compared with the empirical collection of variables related to performance and physiologic function.33 It is obviously clear to the physician that one must treat not only the malady but also the mind. Therefore, the philosophical researcher’s goal is to examine and understand human phenomena through reflective techniques. Philosophical research involves analyzing a problem through inductive, deductive, descriptive, or speculative reasoning.16

Two-by-Two Table Analysis

Positive Predictive Value Positive predictive value is the probability of disease, given a positive test. It is a property of the test interacting with the prevalence of disease in the community (see later). It is computed row-wise (see Fig. 1C-7D): positive predictive value = a / (a + b).

Negative Predictive Value Negative predictive value is the probability of health, given a negative test. It is a property of the test interacting with the prevalence of health in the community (see later). It is computed row-wise (see Fig. 1C-7E): negative predictive value = d / (c + d).

Accuracy Accuracy of a test is the probability of a correct test result. It is a property of the test interacting with the prevalence of disease and health in the community (see later). Accuracy is computed from both rows and columns (see Fig. 1C-7F): accuracy = (a + d) / (a + b + c + d).

Tests for Disease

Prevalence of Disease

A two-by-two table is a statistical comparison of two categorical variables. Table analysis is familiar to most orthopaedic surgeons. We approach this discussion systematically so that the many relationships among the concepts are more apparent. Figure 1C-7 demonstrates the quantities graphically. Often, the cells of the table are filled with counts when the data are initially gathered. Counts are simply the ­number of patients in each situation, for example, 15 patients with a positive test who also have disease. Because the properties of tables are probabilities (or frequencies), the table must be normalized by dividing each cell by total number of patients.

Prevalence of disease is the probability of disease in the community, given no test information. Prevalence is computed from both rows and columns (see Fig. 1C-7G): prevalence of disease = (a + c) / (a + b + c + d).

Basic Two-by-Two Table The basic two-by-two table is shown in Figure 1C-7A. It is important to write the table the same way each time. Label the columns ± for disease (D). Then label the rows ± for the test (T). Label the cells of the table a, b, c, d as you would read: left to right and top to bottom. Some authors like to label the cells as a = true positives (TP); b = false negatives (FN); c = false positives (FP); d = true negatives (TN). However, this notation is cumbersome and can be distracting. Once the patterns are clear, the T/F and P/N notation is easy to recreate, if desired.

Sensitivity Sensitivity is the probability of a positive test, given the presence of disease. It is a property of the test itself. It is computed column-wise (see Fig. 1C-7B): sensitivity = a / (a + c).

Specificity Specificity is the probability of a negative test, given the absence of disease. It is a property of the test itself. It is computed column-wise (see Fig. 1C-7C): specificity = d / (b + d).

Prevalence of Health Prevalence of health is the probability of health in the community, given no test information. Prevalence is computed from both rows and columns (see Fig. 1C-7H). The prevalence of health = 1 − prevalence of disease, but may also be calculated as prevalence of health = (b + d) / (a + b + c + d).

False-Positive Rate False-positive rate is the probability of a positive test, given the absence of disease. It is computed column-wise (see Fig. 1C-7I): false-positive rate = b / (b + d).

False-Negative Rate False-negative rate is the probability of a negative test, given the presence of disease. It is computed column-wise (see Fig. 1C-7J): false-negative rate = c / (a + c).

Relationships Four numbers uniquely define a two-by-two table. In the preceding discussion, we have listed seven different summary measures of a table; therefore, there must be relationships between these quantities. These relationships are called constraints, and they should accord with your intuition of the relationships and their clinical meaning.

Basic Science and Injury of Muscle, Tendon, and Ligament

• Write the table down • Write disease on top • Letter left to right and top to bottom • You are asked for probabilities

Disease D+ D– T+

a

b

T–

c

d

Disease D+ D– • Accuracy T+ • (a + d) (a + b + c + d) Test • Correct test results T– irrespective of positive/ negative results

Test

A

a

b

c

d

109

Figure 1C-7  Two-by-two contingency table analysis explicated in a graphical format.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 3.)

B Disease D+ D–

• Sensitivity • a / (a + c) • Probability of a positive test given presence of disease

T+

a

Disease D+ D– • Specificity • d / (b + d) • Probability of a negative test given presence of health

b

Test T–

c

d

C

T+

a

b

T–

c

d

Test

D Disease D+ D–

• Positive Predictive Value T+ • a / (a + b) Test • Probability of T– disease given a positive test

a

b

c

d

Disease D+ D– • Negative Predictive Value T+ • d / (c + d) Test • Probability of T– health given a negative test

E

a

b

c

d

F Disease D+ D–

• Prevalence of Disease T+ • (a + c) (a + b + c + d) Test • Probability of T– disease given no test information

a c

Disease D+ D– • Prevalence of Health • (b + d) (a + b + c + d) • Probability of health given no test information

b d

T+

a

b

T–

c

d

Test

H

G

• False-positive Rate • b / (b + d) • Probability of a positive test given the absence of disease

Disease D+ D– T+

a

b

Test T–

c

d

I

• False-negative Rate • c / (a + c) • Probability of a negative test given the presence of disease

Disease D+ D– T+

a

b

T–

c

d

Test

J

False-positive rate = 1 - sensitivity. False-negative rate = 1 - specificity. Positive predictive value is proportional to prevalence of disease ´ sensitivity. Negative predictive value is proportional to prevalence of health ´ specificity.

The two relationships among the predictive values, prevalences, sensitivities, and specificities are fundamental to understand. They are a form of Bayes’ theorem, a basic result in probability theory.

Screening and Confirmation with Diagnostic Tests Different types of tests are used for screening and confirmation of a diagnosis. Screening involves testing a large number of individuals, presuming that there is a low prevalence of disease in the population and many healthy individuals. Sensitive tests are used for screening because they have a low false-negative rate. They are unlikely to miss an affected individual. A sensitive test increases the positive predictive value in the face of a low prevalence of disease.

110

DeLee & Drez’s Orthopaedic Sports Medicine

  TABLE 1C-3  Calculation of Sensitivity, Specificity, Prevalence, and Positive and Negative Predictive Values Test Criteria

Ankle Sprain

No Ankle Sprain

Total

Positive (bilateral difference ≥ 25%) Negative (bilateral difference ≤ 25%) Positive (bilateral difference ≥ 25%) Negative (bilateral difference ≤ 25%) Totals A+B Prevalence = ´ 100 A +B+C+D

A B 35 (A) 10 (B) 45

C D 5 (C) 30(D) 35

40 40 80

45 ´ 100 = 56.25 80

Sensitivity =

A ´ 100 A+B

35 ´ 100 = 77.77 45

Specificity =

D ´ 100 C+D

30 ´ 100 = 85.7 35

Positive predictive value =

A ´ 100 A +C

Negative predictive value =

D ´ 100 B+D

Confirmation of a diagnosis involves testing a smaller number of individuals, presuming that there is a high prevalence of disease in the population and many affected individuals. Specific tests are used for confirmation because they are unlikely to result in false treatment of a healthy individual. A specific test increases the negative predictive value in the face of a low prevalence of health.

Example: Sensitivity and Specificity For example, to test the hypothesis that bilateral comparison of balance scores in a group of soccer players yielding a difference equal to or greater than 25% is predictive of future ankle injury (Table 1C-3), one might obtain balance scores on each soccer player and record the number of ankle injuries for all participants. For those athletes who had a bilateral difference of greater than or equal to 25%, one would then determine whether this value was predictive of sustaining an ankle injury. From these data, a researcher can assess the sensitivity, the specificity, and the positive and negative predictive values of the test. In this example, the sensitivity of the test is 77%, and the specificity is 85.7%. The ideal test would have a high sensitivity and a high specificity. There are occasions when sensitivity is more important than specificity, however. Thus, for a new test for multiple sclerosis, researchers would want a high sensitivity. In that case, a false-positive result would be a temporary inconvenience for the patient because other tests could be used to confirm the error. Patients in whom the disease would be confirmed need a sensitive test so that follow-up and treatment could be expedited, however. In this example, the positive predictive value of the screening test was 87.5%, and the negative predictive value was 75%. The positive predictive value is the probability that the individual has or will get the disorder if the test is positive.13 The negative predictive value is the probability that the individual does not have or will not get the ­disorder if the test is negative.14 Both are important, but in this example,

35 ´ 100 = 87.5 40 30 ´ 100 = 75.0 40

it would be more useful for the physician to detect those individuals who would be most likely to suffer a future ankle sprain and to place them on a preventive conditioning program. Prevalence is the proportion of the population that has a disease or a disorder at any one time. Incidence is the proportion of the population that has the disease through a measured period of time. In this example, 56% of the population studied had suffered an ankle sprain.

HYPOTHESIS TESTING Null and Alternate Hypotheses Statistics revolves around the null hypothesis and collecting data regarding this hypothesis. The null hypothesis (HO) is the hypothesis to be tested, often that there is no difference between two sets of data or between subsets of a data set. The alternative hypothesis is just the opposite of the null hypothesis. An important point is that data are gathered to comment on the null hypothesis. The null hypothesis is never truly proved, but data can cast doubt on it. The null hypothesis is finally accepted when a preponderance of data have failed to cast doubt on it. The P value is the probability that the data arose by chance, supposing that the null hypothesis is true. As such, a small P value casts doubt that the null hypothesis is true. In a schematic, the P value is the area under the tails of the probability distribution for the data given the null hypothesis. The process of accepting or rejecting the null hypothesis involves choosing a cutoff for the P value. In practice, P < .05 is usually termed a significant difference, and the experimenter rejects the null hypothesis. In practice, P > .05 is usually termed no significant difference, and the experimenter accepts the null hypothesis. This process constitutes the core procedure of hypothesis testing. The next section discusses various tradeoffs and characteristic values in greater detail.

Basic Science and Injury of Muscle, Tendon, and Ligament

Type I Error Rate By setting a cutoff for the P value as discussed previously, the experimenter is choosing the allowable rate of falsepositive errors. By setting the cutoff for the P value at .05, the experimenter is saying that is it acceptable to detect a false-positive difference that occurs by pure chance 5% of the time, in an attempt to detect true-negative differences. A type I error is a false-positive difference. This corresponds to detecting a difference when in fact there is no difference, or rejecting the null hypothesis when one should accept it. The type I error rate is the false-positive rate, sometimes called alpha (α), and it is the cutoff for the P value chosen by the experimenter.

Type II Error Rate Choosing the type I error rate is familiar to most readers. The type II error rate is equally important; however, it is possible to do an entire study and be quite ignorant of the type II error rate inherent in the study design, although this is ill-advised. A type II error is a false-negative difference. This corresponds to detecting no difference when in fact there is a difference, or accepting the null hypothesis when one should reject it. The type II error rate can be calculated given the type I error rate and the sample size (for most experiments). The conventional value for the type II error rate is 0.2. By setting the type II error rate at 0.2, the experimenter is saying that is it acceptable to conclude a false-negative difference that occurs by pure chance 20% of the time, in an attempt to detect true-negative differences. The type II error rate is the false-negative rate, sometimes called beta (β), and it is chosen by the experimenter. The power is 1− β, or the probability of detecting a true-positive difference. The experimenter chooses the false-negative rate or power (usually 80%).

Experiment Design There is an intimate relationship among the type I error rate, type II error rate (or power), and sample size. Ideally, wouldn’t we like the type I error rate to be very low and the power to be very high? One must trade off type I error rate and power for a fixed sample size. The experimenter can only decrease type I errors and increase power simultaneously by increasing the sample size. Increasing the sample size costs time, money, and study complexity. It may entail patient morbidity, so the tradeoffs must be managed carefully in designing an experiment. A priori power analysis is the preferred method for experiment design. In a priori power analysis, the type I error rate is chosen (e.g., 0.05), the power is chosen (e.g., 80%), and then the sample size needed to achieve the desired power is calculated. The experiment is executed with the needed number of samples. Exactly how the calculations are performed is beyond our scope, but they are well defined for many statistics. To perform the calculations, one must have an estimate of the effect size. The effect size is the expected size of the difference between the null and alternate hypotheses, an idea that can be made mathematically precise.

111

Post hoc power analysis is how many experiments are performed in practice. In post hoc power analysis, the type I error rate is chosen, then a number of samples is chosen that is reasonable to do. The experimenter asks “How many patients are in my series?” or “How much funding do I have?” and obtains the number of samples that is practical under this constraint. From the type I error rate and number of samples, the power is calculated. If the number of samples is too small, the power to detect true-positive differences will be small, and the type II error rate will be large. A negative result will be less meaningful. The asymmetry between type I and type II errors and the prevalence of post hoc power analysis has several implications for orthopaedic science. Literature reviews have been performed that claim that many negative results are from underpowered studies; these may be false-negative results. Many studies omit even post hoc power analysis. In such reports, not only may the study be underpowered, an estimate of the effect is not offered. Finally, the type I error rate is often set at .05, whereas the type II error rate is often set at .20, even in well-designed studies. Are type I errors worse? Is it worse to pollute the scientific literature with a claim of an effect where none exists? Or is it worse to claim there is no effect to a beneficial therapy? These are difficult questions. Designing good experiments takes experience and resources.

Examples: Power Analysis One component of the research design process that is often overlooked is the calculation of a priori power analysis.34 Statistical power is the probability that a significant difference between or among groups will be detected with a statistical test, or it can be stated as the ability to reject the null hypothesis.35 The importance of obtaining power is evident when upon completion of a study there are changes between groups, yet there is no statistically significant difference. On examination of the data, however, the changes between groups appear to be clinically encouraging. ­Statistical power has been likened to a magnifying glass. As statistical power is increased, it becomes easier to see more details, or smaller differences between groups.36 The use of statistical procedures to test a hypothesis is based on the premise that the researcher is willing to accept that a difference found between two or more sets of data would not be as great more than 5% of the time owing to chance (P = .05).37 Therefore, the researcher can confidently state that 95% of the time, differences that great would occur because of something other than chance. The importance of power analysis is determining the probability of finding a difference if one should exist. The power of a statistical test is determined from the following components: the significance criterion, the effect size, the standard deviation, and the number of subjects. The significance criterion (P) is the standard of proof that a difference exists, or the risk for mistakenly rejecting the null hypothesis.35 It is expressed as a probability. Usually in the sports medicine literature, a probability of P = .05 is accepted as being adequate.37 The effect size (d) is the magnitude of difference between the two groups (means) of interest. The standard deviation (SD) is the variation between the means, and the number of subjects (n)

112

DeLee & Drez’s Orthopaedic Sports Medicine

is calculated as the quantity needed per each group. An example of how to calculate power is found in Appendix 1C-1. For example, if it is determined that in a group of anterior cruciate ligament–deficient knees (ACLd), a difference in quadriceps strength of 10 ± 8 lb is clinically significant, researchers can design an experiment to assess differences from participating in different rehabilitation protocols. They may want to examine quadriceps strength in two groups of ACLd patients, with each group receiving a different rehabilitation protocol. A mean difference of 10 ± 8 lb is clinically significant, but this hypothesis must be tested in a clinical trial. In other words, by determining the effect size, researchers can reference a power table (see Cohen’s report for a full listing of power tables35) to calculate the power of the study based on the number of subjects available or potentially available. In Example 1, the effect size is 1.25. By examining a table that lists power values, the power can be calculated by knowing the number of subjects available, or the number of subjects needed to achieve a given power value can be determined. In ­Example 1, to conduct a study with a power of .80, about 15 subjects would be needed in each group (see Appendix 1C-1). Because the calculation of power is based on the effect size, the significance criterion, the number of subjects, and the standard deviation of measurement, it makes sense that by varying each of these factors, power can be increased or decreased. For example, increasing the number of subjects from 15 to 30 can cause the power of this study to increase from .80 to .99. It is a general assumption that a power of .80 is an acceptable value when designing a study. This means that 80% of the time, the investigator will be correct when accepting the conclusion that there was a difference between the treatment groups.38 Examples 3 and 4 demonstrate that power can be increased by reducing the standard deviation of measurement (Example 3) or by increasing the effect size (Example 4). In conclusion, the power of a study increases with an increase in the difference the investigator is trying to detect, when the standard deviation of measurement is decreased, or when the number of subjects is increased.38 From this discussion of reliability, validity, and accuracy of measurement, it can be appreciated that (1) obtaining concise error-free measurements results in decreased standard deviations, (2) offering a controlled and potent treatment intervention results in an increased mean difference, and (3) increasing the number of subjects helps to increase the effect size.

STATISTICS AND TESTS There are numerous statistics and statistical tests for various situations and distributions. Here we cover some general principles of statistics and several of the most commonly used and tested statistical procedures. Descriptive statistics are used to summarize data in a few numbers, such as the mean and standard deviation. Often the goal of statistics is inference, to learn something about the properties of the population from the sample. The mean and standard deviation are also used for inference because these statistics are estimates of the population mean and standard deviation if the distribution is normal. Statistics are also useful for hypothesis testing, as described

previously. A common pattern is to perform inference and hypothesis testing in sequence. We might calculate the correlation coefficient between two variables, then test the hypothesis that the correlation is significant. Statistics are also used for decision analysis, which is discussed briefly. Most of the basic distinctions of statistics were reviewed earlier (see Introduction to Statistics). To review, ­ statistics are relevant to either continuous or ­categorical distributions. Parametric statistics are relevant to normal (bell-curve) distributions, and nonparametric statistics apply when the distribution is non-normal or unknown. Most statistical procedures apply when the variables are all assumed to be dependent, although it is occasionally possible to control one or more variables experimentally (the independent variables) and measure the other variables (the dependent variables). The most common statistics compare two variables. However, sophisticated procedures exist for analyzing the relationships among many variables simultaneously. These are discussed briefly. Some procedures exist for comparing matched samples, samples that correspond value by value rather than just as a whole group. Statistics such as the median and mean are measures of central tendency or location; they give a single summary measure. Statistics such as the standard deviation and range are measures of variability or spread; they give a summary measure of the width of a sample.

One-Variable and Descriptive Statistics The mean is the well-known average of a group of numbers. The median is the number that divides the sample into two groups: half above and half below. The mode is the most common single number in a sample. The standard deviation is the square root of the average squared deviation from the mean. The variance is just the standard deviation squared. The interquartile range is the relationship among the minimum, median, and maximum. A histogram is a graphic representation of all the numbers in a sample divided into bins. The histogram can be considered an estimate of the probability distribution ­function.

Two or More Variables Comparing Means Comparing means involves comparing continuous variables and a categorical variables. Not all the tests involve means per se, but the general term is a useful indicator of the procedure. Figure 1C-8 is a schematic of the procedure, and Figure 1C-9 summarizes the tests. The t-test compares the mean of a sample to a single fixed value. The two-sample t-test compares the means of two samples. The t-test can be paired. Analysis of variance (ANOVA) is a procedure for comparing the means of three of more variables. It can be modified to compare paired samples, or to compare three or more variables with respect to multiple categorical variables (MANOVA). There are nonparametric methods for most of these approaches; the nonparametric statistics are occasionally asked on examinations in addition to

Basic Science and Injury of Muscle, Tendon, and Ligament 10

113

6.5 6

8

5.5 5

6

4.5 4

4 A

B

C

Figure 1C-8  A schematic for the comparison of means by   t-test, ANOVA, or other methods.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 7.)

being useful in practice. The method of discriminant analysis is related to ANOVA and is useful for predicting groups.

Linear Regression Comparing two continuous variables is given the general name linear regression. Figure 1C-10 is a schematic of the procedure, and Figure 1C-11 is a summary of the tests. By far the most common situation involves just two ­variables. Pearson’s correlation coefficient compares two groups of numbers so that +1 is a perfect correlation (straight-line agreement), 0 is uncorrelated (independent in a normal distribution), and −1 is a perfect negative correlation (high values of one predict low values of the other). Another measure is the R-squared (or the coefficient of determination). R-squared is the correlation coefficient squared, and denotes the proportion of one variable that is “explained” by another. One tests the significance of a correlation coefficient with a t-test. The null hypothesis is that there is no correlation. (This illustrates that one statistical test can be used for many different purposes). Note that the correlation coefficient tests only a linear correlation. Two variables can be closely related (e.g., one is simply the square of the other), but poorly linearly correlated. Strictly speaking, the correlation coefficient is a parametric statistic most applicable to normal distributions. Normal

3.5 3 2.5 2.5

3

3.5

4.5

5

5.5

6

6.5

Figure 1C-10  A schematic of the comparison of continuous variables by correlation coefficients or other methods.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 9.)

Spearman’s rank correlation is a nonparametric measure of correlation. The case of two variables is a highly simplified case of the general process of linear regression for any number of variables. The analysis of linear relationships is a broad field with a variety of methods, some quite sophisticated. Options also exist for the nonparametric analysis of linear relationships among several continuous variables. The calculation of partial correlations among several variables is part of linear regression as well.

Logistic Regression The prediction of a categorical variable from a continuous variable is termed logistic regression. A basic approach involves an S-shaped function that approximates a cutoff value between two groups. Figure 1C-12 is a schematic of the procedure. More generally, logistic regression methods may predict a categorical variable from many continuous or categorical variables simultaneously. This general class of approaches also includes related methods such as generalized regression, probit, and logit models. Despite technical distinctions, logistic regression is the most widely used term. One of the most important aspects of logistic regression is the idea of model building. Model building is the selection or elimination of predictor variables in a systematic way using both statistical criteria and expert opinion for

Two categories

t-test

Two categories (paired) Three or more categories Three or more (paired)

Paired t-test

Nonparametric Wilcoxon rank sum Mann-Whitney U Wilcoxon signed rank

ANOVA

Kruskal-Wallis

Two variables

ANOVA (two-way)

Friedman

Three or more variables

Figure 1C-9  A summary of the statistical tests used for comparison of means.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 8.)

4

Normal Pearson correlation Linear regression

Nonparametric Spearman correlation Nonparametric regression

Figure 1C-11  A summary of statistical methods used for the comparison of continuous variables.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 10.)

114

DeLee & Drez’s Orthopaedic Sports Medicine

B

A

1.0 2.0 3.0 4.0 5.0 6.0 7.0 Figure 1C-12  A schematic for the comparison of continuous   and categorical variables by logistic regression or other methods.   (Adapted from Miller MD: Review of Orthopaedics, 4th ed. Philadelphia, Elsevier, 2004, Chapter 12, Figure 11.)

variable selection. Model building is applicable to many multivariate statistical approaches, and many orthopaedic surgeons first encounter this process in logistic regression. In general, variables are either sequentially added or sequentially removed based on whether the model’s ability to predict the data improves with a predictor variable. This is an important procedure in many practical clinical studies.

Table Analysis We have already considered the analysis of two-by-two tables in terms of various measures (e.g., sensitivity, ­specificity) at some length. We now consider hypothesis testing in tables. By far the most common approach is the chi-square statistic, which can be used for two-by-two tables or tables with more categories per variable (e.g., three-bythree table). The chi-square statistic tests the null hypothesis that the cells of the table are random, and that there are no dependencies among the cell counts. In the case that there are relatively few total cases (the sum of all cells is less than about 50), the Fisher exact test is a useful substitute for the chi-square statistic. Occasionally, there is a need to analyze the relationships among many categorical variables. There are advanced methods such as log-linear analysis and Bayes’ networks for such problems, the details of which are beyond our scope. However, these methods highlight an important aspect of model building in general, the question of covariate selection and how to control for confounding. In a prior section, we discussed the use of randomization to control for confounding. However, it is sometimes necessary to estimate the effect of a treatment using observational data (e.g., retrospective data) alone, without any randomization. In these cases, it is necessary to control for confounders that can affect the outcome so that the effect of the treatment on the outcome can be isolated. There is a folklore practice and a conventional practice about how to select the variables that must be controlled. Also, there are advanced theoretical studies of this question. It is accepted that one should control for variables that affect both the treatment and the outcome, and one should not control for variables that are affected by the treatment. Details may be found in the references.

SUMMARY The adaptation of an idea into a scientific study is multifaceted. The development of a research team and the refinement of a question are strengthened with a collaborative approach. The practical importance of understanding statistics is to allow physicians and clinicians the opportunity to understand (1) what characteristics should be considered in the preplanning stages of a study, (2) what procedure to use following data collection, and (3) how to draw conclusions and inferences with other studies while reviewing the literature and during the preparation of the discussion and conclusion sections of a manuscript. The selection of the dependent and independent variables must be given careful consideration and should be part of the hypothesis and research design phase. Determination of an assessable dependent variable (the meas­ urable variable) and a potent independent variable (that which is manipulated) strengthens the scientific merit of the study. Random sampling of subjects should be used because statistical tests are based on the assumption that all samples are randomly selected. Bias is a condition or phenomenon that when introduced influences the meas­ urement and the analysis.5 Bias should be reduced or eliminated. Research can be partitioned into four categories: ­analytical, descriptive, experimental, and qualitative.15,16 In analytical research, historical researchers arrive at the answer to a question by the use of primary and secondary literature, published and anecdotal facts, and personal recollection. An example of descriptive research, epidemiologic research is the study of the frequency and the cause of disease. Rate and risk are commonly used as the measure in epidemiologic studies. Rate is the ratio when the change in one phenomenon is paired with another variable such as time. Risk is the probability that an injury will occur during a specified time period. Experimental designs are characterized by the inclusion of an independent variable that is manipulated in a controlled manner.15 Quasi-experimental designs are studies in which there is a lack of control of the independent variable. A single-case or single-subject experimental design is characterized by the following: identification of a baseline measure, repeated measurement of the dependent variable, repeated manipulation of the independent variable, a comparison within the individual across differing conditions, one or only a few subjects, and a replication of effects; ideally it should include a measurement that is objective.25 In contrast, qualitative research involves intensive observation that often involves a comprehensive interview process in a natural setting.16 Qualitative research designs are naturalistic because the researcher does not attempt to manipulate the research setting. Consideration of validity and reliability must be incorporated when designing a study. Validity is the ability to state with confidence that the results were due to the experiment and that those results have some larger inherent generalizability; reliability refers to the degree of reproducibility of a measurement. Finally, the probability that a significant difference can exist between or among groups (the power of the test) must be considered.

Basic Science and Injury of Muscle, Tendon, and Ligament

C l Choosing

r i t i c a l

P

o i n t s

the correct study design is essential: experimental versus observational and prospective versus ­retrospective studies. l Although the randomized controlled trial is an ideal ­design, most clinical studies in orthopaedic surgery are case series or case control series. l Correct choice of statistics allows the data to shed light on the hypothesis: is the hypothesis improbable given the data? l Power analysis allows one to design studies of appropriate size and cost to answer the question at hand. l When the data fit a bell-curve distribution, normal statistics are appropriate (e.g., mean, standard deviation, t-test). When they do not, nonparametric statistics are ­important (e.g., median, range, Wilcoxon test). l Two-by-two table analysis allows the determination of important measures of a diagnostic test: sensitivity, ­specificity, and positive and negative predictive values.

S U G G E S T E D

115

R E A D I N G S

American Orthopaedic Society for Sports Medicine: Statistics Primer. Birmingham, Ala, The American Journal of Sports Medicine, 2000. Greenfield ML, Kuhn JE, Wojtys EM: A statistics primer. P values: Probability and clinical significance. Am J Sports Med 24:863-865, 1996. Greenfield ML, Kuhn JE, Wojtys EM: A statistics primer. Power analysis and ­sample size determination. Am J Sports Med 25:138-140, 1997. Greenfield ML, Kuhn JE, Wojtys EM: A statistics primer. Tests for continuous data. Am J Sports Med 25:882-884, 1997. Greenfield ML, Kuhn JE, Wojtys EM: A statistics primer. Correlation and regression analysis. Am J Sports Med 26:338-343, 1998. Greenfield ML, Kuhn JE, Wojtys EM: A statistics primer. Validity and reliability. Am J Sports Med 26:483-485, 1998. Kuhn JE, Greenfield ML, Wojtys EM: A statistics primer. Prevalence, incidence, relative risks, and odds ratios: Some epidemiologic concepts in the sports medicine literature. Am J Sports Med 25:414-416, 1997. Kuhn JE, Greenfield ML, Wojtys EM: A statistics primer: Statistical tests for ­discrete data. Am J Sports Med 25:585-586, 1997. Kuhn JE, Greenfield ML, Wojtys EM: A statistics primer. Types of studies in the medical literature. Am J Sports Med 25:272-274, 1997. Rothman KJ, Greenland S: Modern Epidemiology, 2nd ed. Philadelphia, ­Lippincott Williams & Wilkins, 1998.

R eferences Please see www.expertconsult.com

116

DeLee & Drez’s Orthopaedic Sports Medicine

A P P E N D I X

1 C - 1

Statistical Power Examples Power Exercise

Example 2

d = effect size Ma, Mb = population means in raw units SD = standard deviation of either population

Increasing the number of subjects increases the power of a research design. Doubling the number of subjects in each group with the same effect size (d = 1.20) would increase power from .80 to .98. (See table at the end of this ­appendix.)

Example 1 Calculation of the effect size (d) for the two groups of anterior cruciate ligament–deficient knee (ACLd) patients.

Example 3 Decreasing the standard deviation of the measurement ­increases the effect size and the power of the study. Mean a = 20 Mean b = 10 Mean difference = 10 SD = 7 d = 1.43 n = 10 d = (mean a –���������������� mean b) / SD d = 10/7 d = 1.43 Power = .91

a. Specify the significant difference between the two groups: 10 ± 8 lb. b. Using a study that is similar or based on pilot data, identify the standard deviation (SD) for the measure of interest. In this example, it is ± 8 lb. c. Using the difference and SD, calculate the effect size index (d) for this study. Mean a = 20 Mean b = 10 Mean difference = 10 SD = 8 d = 1.25 d = (mean a – mean b) / SD d = 10/8 d = 1.25

Example 4 Increasing the mean difference of the measurement increases the effect size and the power of the study. Mean a = 30 Mean b = 10 Mean difference = 20 SD = 8 d = 2.5 n = 10 d = (mean a –���������������� mean b) / SD d = 20/8 d = 2.5 Power > .91

In order to conduct a study with a power of .80, approximately 10 subjects should be included in each group. (See table at the end of this appendix.) d. Using the effect size and table, find the row that ­corresponds to n for each group. Then find the power of this study. Power = .83 based on each group containing approximately 10 subjects. Power of t-test of m1 = m2 at a1 = .05 d n

de

.10

.20

.30

.40

.50

.60

.70

.80

1.00

1.20

1.40

 8  9 10 11 12 13 14 15 16

.88 .82 .78 .74 .70 .67 .64 .62 .60

07 07 08 08 08 08 08 08 09

10 11 11 12 12 13 13 13 14

13 15 16 17 18 18 19 20 21

19 20 22 23 25 26 27 28 30

25 27 29 31 33 34 36 38 40

31 34 36 39 41 44 46 48 51

38 41 45 48 51 54 57 59 62

46 50 53 57 60 63 66 69 72

61 66 70 74 77 80 83 85 87

74 79 83 86 89 91 93 94 95

85 88 91 94 96 97 98 98 99 Continued

Basic Science and Injury of Muscle, Tendon, and Ligament

Power of t-test of m1 = m2 at a1 = .05—cont’d d n

de

.10

.20

.30

.40

.50

.60

.70

.80

1.00

1.20

1.40

  17   18   19   20   21

.58 .56 .55 .53 .52

09 09 09 09 09

14 15 15 15 16

22 22 23 24 25

31 32 33 34 36

42 43 45 46 48

53 55 57 59 60

64 66 68 70 72

74 76 78 80 82

89 90 92 93 94

96 97 98 98 99

99 99 *

  22   23   24   25   26   27   28   29   30   31   32   33   34   35   36   37   38   39   40   42   44   46   48   50   52   54   56   58   60   64   68   72   76   80   84   88   92   96 100 120 140 160 180 200 250 300 350 400 450 500 600 700 800 900 1000

.51 .50 .48 .47 .46 .46 .45 .44 .43 .42 .42 .41 .40 .40 .39 .39 .38 .38 .37 .36 .35 .35 .34 .33 .33 .32 .31 .31 .30 .29 .28 .28 .27 .26 .26 .25 .24 .24 .23 .21 .20 .18 .17 .16 .15 .13 .12 .12 .11 .10 .10 .09 .08 .08 .07

09 10 10 10 10 10 10 10 10 10 11 11 11 11 11 11 11 11 11 12 12 12 12 12 13 13 13 13 13 14 14 15 15 15 16 16 17 17 17 19 21 23 24 26 30 34 37 41 44 47 53 59 64 68 72

16 16 17 17 18 18 18 19 19 19 20 20 20 21 21 21 22 22 22 23 24 24 25 26 26 27 28 28 29 30 31 33 34 35 36 37 38 40 41 46 51 56 60 64 72 79 84 88 91 93 97 98 99 *

26 26 27 28 28 29 30 30 31 32 33 33 34 34 35 36 36 37 38 39 40 41 43 44 45 46 47 49 50 52 54 56 58 60 61 63 65 66 68 75 80 85 88 91 96 98 99 *

37 38 39 40 41 42 43 44 46 47 48 49 50 50 51 52 53 54 55 57 59 60 62 63 65 66 68 69 70 73 75 77 79 81 82 84 85 87 88 93 95 97 98 99 *

50 51 53 54 55 57 58 59 61 62 63 64 66 67 68 69 70 71 72 74 75 77 79 80 81 83 84 85 86 88 90 91 92 93 94 95 96 96 97 99 99 *

62 64 66 67 69 70 72 73 74 76 77 78 79 80 81 82 83 84 84 86 87 89 90 91 92 93 93 94 95 96 97 97 98 98 99 99 99 99 *

74 76 77 79 80 82 83 84 85 86 87 88 89 89 90 91 91 92 93 94 95 95 96 97 97 98 98 98 98 99 99 99 *

83 85 86 88 89 90 90 91 92 93 93 94 95 95 96 96 96 97 97 98 98 99 99 99 99 99 99 *

95 96 96 97 97 98 98 98 99 99 99 99 99 99 99 *

99 99 99 99 *

*

*

*Power values below this point are greater than .995. From Cohen J: Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Hillsdale, NJ, Lawrence Erlbaum Associates, 1988.

*

117

118

DeLee & Drez’s Orthopaedic Sports Medicine

A P P E N D I X

1 C - 2

Comparison of Pearson’s Product-Moment ­Correlation and Intraclass Correlation Coefficient for Data with Systematic Error

   

Time

Time

Time

Time

Data

1

2

3

4

1 2 3 4 5 6 7 8 9 10

30 32 34 35 38 40 42 44 46 48

29 31 34 35 39 41 43 45 46 48

30 32 34 35 38 40 42 44 46 48

130 132 134 135 138 140 142 144 146 148

Analysis of Variance Summary Table for Time 1 and Time 2*

Betweensubjects effect Error Time (T1 & T2) Error

SS

DF

MS

F

Sig of F

30420.00

1

30420.0

377.63

.00

725.00 .20

9 1

80.56 ← BMS .20 ← TMS

2.80

9

.31 ← EMS

.64

.443

*The analysis of variance table is used for calculating the Intraclass Correlation Coefficient. DF, degrees of freedom; F, F value; Sig of F, significance of F value; MS, mean square; SS, sums of squares. ICC for Time 1 and Time 2: Between mean square (BMS) = 80.56 Error mean square (EMS) = .31 Trial mean square (TMS) = .20 Mean (X) = 39 Standard deviation (S) = 6.033 k = No. of examiners BMS - EMS ICC (2.1) = BMS + (k - 1) EMS + k[TMS - EMS/N] 80.56 - .31 = 80.56 + (1) .31 + 2[.20 - .31/10] 80.56 .31 = 80.56 + .31 + ( - .022) = .9926 Standard error of measurement (SEM) SEM = s 1 - r = 6.033 1- .9926 = .52

Analysis of Variance Summary Table for Time 3 and Time 4* SS Betweensubjects effect Error Time (T3 & T4) Error

DF

MS

F

Sig of F

158064.0 1

158064.0

2111.276

.000

673.80 9 50000.00 1

74.867 ← BMS 50000.067 ← TMS

.00

.00 ← EMS

9

*The analysis of variance table is used for calculating the Intraclass Correlation Coefficient. Abbreviations defined in summary table for times 1 and 2. ICC for Time 3 and Time 4: Between mean square (BMS) = 74.87 Error mean square (EMS) = .00 Trial mean square (TMS) = 5000 Mean (X) = 88.9 Standard deviation (S) = 50.33 BMS - EMS ICC (2.1) = BMS = (k - 1)EMS + k[TMS - EMS/N] 74.87 - .00 = 74.87 + (1) .00 + 2[50000 - .00/10] 74.87 - .00 = 74.87 + .00 + 1000 = .00743 Standard error of measurement (SEM) SEM = S 1 - r = 50.33 1 - .00743 = 49.95

Basic Science and Injury of Muscle, Tendon, and Ligament

Pearson’s Product-Moment Correlation for Time 1 and Time 2 Data

T1

T2

X2

Y2

XY

1 2 3 4 5 6 7 8 9 10

30 32 34 35 38 40 42 44 46 48

29 31 34 35 39 41 43 45 46 48

900 1024 1156 1225 1444 1600 1764 1936 2116 2304

841 961 1156 1225 1521 1681 1849 2025 2116 2304

870 992 1156 1225 1482 1640 1806 1980 2116 2304

∑X =  38.9 ����   ��∑Y = 39.1������   ∑X = 15469������   ∑Y = 15679������   ∑XY = 15571  s = 5.8  s = 6.2 N = No. of pairs df = N - 2 = 8 Critical r at the .05 level = .6319 Note: Data indicate that with an r of .6319 or higher, there is a less than 5 out of 100 chance that the null hypothesis would be rejected incorrectly. NSXY - ( SX) ( SY) r = NSX 2 - ( SX)2 ´ NSY 2 - ( SY)2

Pearson’s Product-Moment Correlation for Time 3 and Time 4: Determination of r When There Was a Systematic Change upon Retest 1 2 3 4 5 6 7 8 9 10

T3

T4

X2

Y2

XY

30 32 34 35 38 40 42 44 46 48

130 132 134 135 138 140 142 144 146 148

900 1024 1156 1225 1444 1600 1764 1936 2116 2304

16900 17424 17956 18225 19044 19600 20164 20736 21316 21904

3900 4224 4556 4725 5244 5600 5964 6336 6716 7104

∑X =  38.9   ��∑Y = 138.9   ��∑X2 = 15469   ��∑Y2 = 193269   ��∑XY = 54369  ���s = 5.8   ���s = 5.8 r= =

NSXY NSX 2 - ( SX )2

- ( SX )( SY ) ´

NSY 2 - ( SY )2

(10 ) 54369 - (38.9)(138.9 ) 10(15469 ) - ( 38.9 )2

´

10(193269 ) - (138.9 )

=

10(15571) - (38.9) (39.1) 10(15469) - (38.9)2 ´ 10(15679) - (39.1)2

543690 - 5403.21 = 154690 - 1513.21 ´ 1932690 - 19293.21

=

155710 - 1520.99 154690 - 1513.21 ´ 156790 - 1528.81

=

=

154189.01 153176.79 ´ 155261.19

=

=

154189.01 (391.3) ´ (394.03)

=

154189.01 154183.939

= 1.00

119

538286.79 153176.79 ´ 1913396.79

538286.79 541267.28 = .994

C H A P T E R�  

  �2

Surgical Principles S ect i o n

A

Basic Arthroscopic Principles Mark D. Miller and Jennifer Hart

Arthroscopy has its roots in Japan, when in 1918 Takagi used a cystoscope to look inside a knee.1 Watanabe, another Japanese surgeon, is credited with further refinements of the arthroscope and development of the concept of triangulation.1 North American arthroscopic pioneers include Jackson, Joyce, McGinty, Cassceles, Dandy, O’Connor, and Johnson. Arthroscopy has grown rapidly and is the standard of care for the treatment of many orthopaedic injuries. Textbooks, journals, societies, and subspecialization have expanded the scope of arthroscopic surgery further. Arthroscopy often can be done more quickly, with increased accuracy, lower complication rates, decreased hospitalization time, and shorter recovery periods than many operative techniques. The effective use of arthroscopy is based on the understanding of benefits and use of arthroscopy as well as its limitations. In the first issue of Arthroscopy, the Journal of Arthroscopic and Related Surgery, guidelines were elucidated for the practice of arthroscopic surgery: 1. The arthroscopist should perform an adequate history and physical examination as well as obtain radiographs or other pertinent laboratory evaluations for each patient or determine that they already have been performed. 2. The risks, benefits, alternatives of treatment, and potential complications should be outlined carefully to each patient before an arthroscopic evaluation is carried out. 3. The arthroscopist should carefully select the correct arthroscopic procedure for a particular condition. 4. A detailed report of the procedure should be prepared, including the arthroscopic findings and description of the operation.

OPERATIVE SUITE Arthroscopy is performed most commonly in a standard surgical suite in either a hospital or an ambulatory surgical facility. Office arthroscopy, although advocated by some, may have significant limitations and is used primarily for

diagnosis in selected patients. Newer flexible fiberoptic catheter systems resulted in under-recognition and underestimation of the severity of intra-articular knee disease in one study.2 A relatively large room that can accommodate the bulky arthroscopic equipment is preferred. Storage space for video equipment, shaver blades, instruments, and other equipment should be available in the room or nearby. Dedicated, well-trained operating room personnel are required. Adequate electrical support, lighting, suction, and other logistic concerns should be considered. Equipment sterilization is important. Standard autoclaving is not appropriate for arthroscopes, cables, and many of the instruments used in arthroscopy. Gas sterilization with ethylene oxide is effective, but because of the long turnover times associated with this method, instruments cannot be reused during the same operative day.3 Because of these concerns, most centers have elected to accept high-level disinfection in lieu of sterilization. A newer system that uses peracetic acid (Steris, Menor, Ohio) is convenient and efficacious, and it largely has replaced other methods of sterilization.4

ARTHROSCOPIC EQUIPMENT Arthroscopy requires an arthroscope, a camera, a light source and fiberoptic cable, an irrigation system, cannulas, and various hand and motorized instruments. An ­ arthroscope is a small-diameter fiberoptic instrument that allows direct visualization of joints (Fig. 2A-1).5 It is designed to fit into a sleeve, or cannula. The cannula first is inserted into the joint with a sharp or blunt rod called a trocar, and the trocar is then exchanged for the arthroscope. A fiberoptic cable and camera are attached to the arthroscope, and irrigation fluid is attached to one or more cannulas. A specially adapted video camera usually is attached to the eyepiece of the arthroscope, and images can be recorded directly onto video or onto an accompanying printer (Fig. 2A-2). Arthroscopes are classified according to their diameter and viewing angle. The most commonly used arthroscope is 4 mm with a 25- to 30-degree angle. Smaller ­arthroscopes 121

122

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 2A-1   Top to bottom: Thirty-degree arthroscope with attached cable, 70-degree detached arthroscope, detachable camera cable (left) and sleeve (cannula) (right), trocar for introduction of sleeve, and 30-degree detached arthroscope. (From DiGiovine NM, Bradley JP: Arthroscopic equipment and setup. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 543-556.)

(2.5 and 1.9 mm) can be used for smaller joints (e.g., wrist). Different angles (70 or 90 degrees) sometimes are useful when visualization can be difficult (e.g., posterior knee). Instruments used during arthroscopy include those that are hand operated (e.g., probes, baskets, grabbers) and those that are motorized (e.g., shavers) (Fig. 2A-3). Commercially available shavers of various sizes, shapes, angles, and blades are useful in a variety of situations (Fig. 2A-4). Intra-articular cautery can be useful, and specially adapted tips that can be used with normal irrigating systems are available. Other available systems include special heat probes and lasers. Irrigation is used during arthroscopy for joint distention, improved visualization, and removal of debris. ­ Lactated Ringer’s solution usually is used because it is more physiologic than normal saline.6 A study has suggested, however, that 5% mannitol may prevent excessive loss of proteoglycan from hyaline cartilage.7 Cannulas are used for fluid egress or ingress. Hydrostatic pressure is necessary to maintain joint distention. This pressure can be attained by gravity or through commercially available pumps (Fig. 2A-5).

ARTHROSCOPIC VISUALIZATION One must understand several fundamental concepts to interpret arthroscopic images accurately. First, the typical arthroscope does not look straight ahead but instead is directed at a 25- to 30-degree angle off the axis.8 This angle allows greater visual control and improves the field. The arthroscopist can keep the arthroscope stationary and rotate the viewing angle, allowing a 60-degree view from a 30-degree scope (Fig. 2A-6). Another consideration is that an object’s size as seen on the monitor is magnified. The degree of magnification varies with the distance of the object from the lens of the arthroscope. Judging distances and size requires practice. Knowing the dimensions of an instrument, such as an arthroscopic probe, helps in determining the relative size

Figure 2A-2   Top to bottom: Arthroscopic cart showing camera, light source with cable, video equipment, and printer. A monitor (not shown) can be placed on top of the cart. (From DiGiovine NM, Bradley JP: Arthroscopic equipment and setup. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 543-556.)

of the object being visualized. It is helpful to know about normal versus abnormal findings in each joint.

ARTHROSCOPIC TECHNIQUE Extreme care should be taken in positioning the extremity for arthroscopy to avoid any compression that can result in neurapraxia. The nonoperative side is padded and protected, and excessive traction or unnecessary motion of the operative extremity is avoided. Numerous positioning and traction devices are commercially available and are often helpful. Tourniquets, when used, should be as wide as possible, and tourniquet time should be minimized. Surgical preparation and draping should be performed carefully to seal the operative field and to create a sterile environment. A joint-specific systematic diagnostic examination should be done before any therapeutic procedures are performed. A complete arthroscopic examination usually can be performed in a few minutes, and then a complete operative plan can be confirmed or modified according to these findings.

ARTHROSCOPIC COMPLICATIONS Any operation carries a risk for complications, and arthroscopy is no exception.9 Perhaps the most common complication is inadvertent damage to the intra-­articular

Surgical Principles

Figure 2A-3   Probes (top two instruments) and baskets (bottom three instruments) come in a variety of sizes and angles. (From Ciccotti MG, Shields CL, El Attrache N: Meniscectomy. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 591-613.)

123

Figure 2A-5   Commercially available arthroscopy pump allows precise control of flow rate and hydrostatic pressure. (From Ciccotti MG, Shields CL, El Attrache N: Meniscectomy. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 591-613.)

structures.10 This risk is inversely proportional to the surgeon’s experience and the care with which the surgeon performs the procedure. Proper portal placement, gentle technique, and attention to detail are crucial. The exact prevalence and long-term sequelae of this lesion are unknown, but studies involving second-look arthroscopy and animal models have shown that this is a true risk and that the lesions do not tend to fill with time.11,12 Certain joints such as the hip are at increased risk for iatrogenic cartilage injury, and techniques of portal placement and traction have been described to help decrease the risk in this joint.13 Instruments can break, especially when older, unserviced instruments are used.14 Nerve or vessel injury can result from improper portal placement. A thorough knowledge of local anatomy is needed before performing arthroscopy. In joints at particular risk (e.g., the elbow), the nick and spread method is preferred. The skin only is cut with a blade, and a small hemostat is used to clear the way before inserting instruments with this method. Tourniquet

paresis can be reduced with a wide cuff and by limiting tourniquet time to less than 90 minutes whenever possible.15 Positioning devices (e.g., leg holders) can have a tourniquet effect, even when the tourniquet is not inflated. Fluid extravasation has been reported,16 although the incidence of this complication can be reduced with careful placement of inflow cannulas and proper use of inflow pumps. Synovial fistula formation is a rare complication of arthroscopy and is usually remedied by 7 to 10 days of immobilization and, occasionally, delayed closure. Infection is an extremely rare complication of arthroscopy, usually the result of a break in sterile technique.17 High-dose antibiotics and arthroscopic irrigation and débridement may be indicated in severe cases.18 An unusual, but potentially life-threatening, complication of knee arthroscopy is the development of a postoperative deep vein thrombosis (DVT) or pulmonary embolism (PE). Although the risk for these events is low, careful screening of risk factors must be done to identify patients who require routine prophylaxis. These risk factors include things such as increased body mass index (BMI), personal or family history of clotting disorder, advancing age, and

Figure 2A-4   Arthroscopic shavers come in a variety of sizes, angles, and blades. (From Ciccotti MG, Shields CL, El Attrache N: Meniscectomy. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 591-613.)

Figure 2A-6   Rotation of a 30-degree arthroscope allows a 60-degree view. (From Crane L, Sullivan DJ: Instrumentation. In McGinty JB [ed]: Operative Arthroscopy. New York, Raven Press, 1991, p 6.)

124

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 2A-1  Relative Risk Factors for ­Postoperative Deep Venous ­Thrombosis or Pulmonary  Embolism in Knee Arthroscopy One-Point Factors Age 40-59 yr Surgery shorter than 45 min Body mass index > 25 kg/m2 Pregnant or 1 mo postpartum Chronic obstructive pulmonary disease Varicose veins Inflammatory bowel disease Preoperative bed rest Two-Point Factors Age 60-70 yr History of malignancy in past Surgery longer than 45 min Postoperative immobilization > 72 hr Three-Point Factors Age 70-75 yr Personal or family history of clot Factor V Leiden deficiency Thrombophilia Five-Point Factors Hip, pelvis, or leg fracture Multiple trauma History of stroke Limb paralysis

history of malignancy. We use the system described in Box 2A-1 to identify patients who need prophylactic treatment for thromboembolic events. Any patient with more than three points as defined by these guidelines is considered to be at increased risk and should be treated.19,20

ARTHROSCOPIC APPLICATIONS Arthroscopy can be used for a variety of diagnostic and therapeutic purposes. It is useful in almost all major joints for irrigation and débridement, loose or foreign body removal, synovectomy, débridement of loose soft tissue, addressing osteoarticular lesions or fractures, and diverse other procedures. The use of arthroscopy in the knee and shoulder has increased steadily. Arthroscopy has become the standard for treatment of meniscal tears. Applications continue to expand, and the future scope of arthroscopic applications is limited only by the imagination of the arthroscopist. A brief introduction of arthroscopy for each joint follows.

Knee Arthroscopy Indications Arthroscopy has diverse application in various forms of knee disease. Diagnostic arthroscopy helps to confirm suspected knee injuries. Arthroscopic synovectomy can be useful for synovial biopsies to aid the diagnosis of ­ rheumatologic

Figure 2A-7   Arthroscopic positioning with a commercially available leg holder for the operative leg and a well-padded gynecologic leg holder for the nonoperative extremity. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 50.)

­ isorders, to remove diseased synovium, and to resect d synovial folds or plicae. A six-portal technique is favored for complete synovectomy. Treatment of meniscal disease is perhaps the most common application of arthroscopy. Meniscal tears account for about half of knee injuries that require surgery.21 Osteochondral lesions commonly are addressed arthroscopically. Injuries to the cruciate ligaments can be diagnosed easily with arthroscopy; endoscopic reconstruction of these ligaments is one of the most common orthopaedic procedures. Other procedures that sometimes are aided with arthroscopy include tibial plateau fracture reduction, reduction and fixation of tibial eminence fractures, loose body removal, anterior fat pad débridement, lateral release for patellar malalignment, and irrigation and débridement of septic arthritis.

Positioning and Portal Placement Two different forms of positioning are commonly used for knee arthroscopy. The patient can be placed supine on the operating table, and a lateral post can be used for countertraction. Alternatively, the operative leg can be positioned in a commercially available leg holder (Fig. 2A-7). The operative leg is allowed to hang freely over the end of the operating table, and the opposite leg is positioned in a well-padded leg holder, with care taken not to compress the peroneal nerve. Standard arthroscopic portals for knee arthroscopy have traditionally included a superomedial or ­superolateral

Surgical Principles

125

Patella

PSM SL

Trochlea

SM

PM PL FM FL

PCL ACL

IM

MP IL Figure 2A-9   Normal arthroscopic anatomy of the knee. (Inset shows posteromedial aspect of the knee as seen through the notch.) ACL, anterior cruciate ligament; PCL, posterior cruciate ligament. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 51.) Figure 2A-8   Arthroscopic portals for knee arthroscopy. FL, far lateral; FM, far medial; IL, inferolateral; IM, inferomedial; MP, midpatellar; PL, posterolateral; PM, posteromedial; PSM, proximal superomedial; SL, superolateral; SM, superomedial. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 50.)

­ ortal for fluid inflow and outflow and inferomedial and p ­inferolateral portals positioned just above the joint line on both sides of the patellar tendon for arthroscopy and instrumentation (Fig. 2A-8). Newer arthroscopic fluid control systems have now made the use of superior portals optional. The use of a far proximal superior portal can still be helpful for visualization of patellar tracking. The inferolateral portal usually is used during diagnostic arthroscopy. Accessory portals for the knee include the posteromedial, posterolateral, far medial and lateral, and proximal superomedial portals. The posteromedial portal is often helpful for visualizing the posterior cruciate ligament22 and the posterior horn of the medial meniscus.23 The posterolateral portal, located between the iliotibial band and the biceps tendon, sometimes is helpful, but extreme care should be taken to ensure that the portal is anterior to the biceps tendon to avoid injury to the peroneal nerve. Other portals include the midpatellar portal; far medial and lateral portals (sometimes helpful for instrument placement in hard to reach areas); and the proximal superomedial portal, located 4 cm proximal to and in line with the medial edge of the patella (for assessment of patellar tracking).

Arthroscopic Anatomy As with any joint, systematic examination of the knee is appropriate. Before positioning the patient, a complete examination under anesthesia is conducted to assess

i­nstability in all planes. An arthroscopic cannula is placed in the superomedial or superolateral portal for inflow and outflow (although the use of these superior portals is now optional with many of the new pump systems), and the arthroscope is introduced into the inferolateral portal. Although many examination sequences are possible, it is important to visualize the suprapatellar pouch, patellofemoral joint, medial and lateral gutters, medial and lateral compartments (meniscus and articular cartilage), and intercondylar notch (cruciate ligaments) in all patients. Accessory viewing portals are established as necessary if other areas need to be evaluated. A posteromedial portal can be helpful whenever medial meniscus pathology is suspected but is unable to be identified from the anterior portals. This portal is established by introducing the arthroscopic cannula into the back of the knee by directing it from anterior to posterior on the notch side of the medial femoral condyle. Care must be taken to avoid the saphenous nerve and vein while using the spinal needle to establish the portal. Once the arthroscope is in the posterior aspect of the knee, the posterior horn of the medial meniscus can be visualized. A 70-degree scope may be helpful. After a complete evaluation of the joint (Fig. 2A-9), all surgical pathology is addressed.

Hip Arthroscopy Indications The indications for hip arthroscopy are more limited than for other joints. Because of its anatomy, the hip joint is difficult to access through arthroscopy, and maneuverability is difficult. Arthroscopy may be indicated for cases of loose bodies, torn labral tissue, articular cartilage lesions, synovial or ligamentum teres impingement, and refractory pain of undetermined cause. Arthroscopy can be used in

126

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

young arthritic patients to delay the need for total joint arthroplasty.24 A more recent indication for hip arthroscopy includes the treatment of internal and external snapping hip. Results have proved perhaps even superior to the more traditional open techniques of lengthening Z-plasty and iliopsoas tendon release.25,26 Benefits again include shorter operation times and recovery periods with fewer surgical complications, with the added benefit of allowing the arthroscopist to evaluate for the presence of intra-articular pathology at the same time.

Positioning and Portal Placement Two positions are popular for hip arthroscopy: the supine and the lateral decubitus positions. The latter, developed by Glick and colleagues27 after their initial dissatisfaction with the supine position, is performed with the patient on his or her side, with the involved hip on top. The patient is placed in the lateral decubitus position on a fracture table with a peroneal post, and traction is applied through the footplate, skeletal traction, or a commercially available external traction device. The leg is abducted 45 degrees and forward-flexed 10 degrees.27 About 50 pounds of traction is necessary to distract the hip 8 to 10 mm for arthroscopic visualization.24 The supine position was reintroduced by Byrd,28 who also used a fracture table. A padded peroneal post is lateralized to the operative side, the hip is positioned in extension and 25 degrees of abduction, and traction is applied. Fluoroscopy is recommended for either position to confirm entry into the joint. Portals are similar for both positions and include the anterior, anterolateral (anterior trochanteric), and posterolateral (posterior trochanteric) portals (Fig. 2A-10). The anterior portal is established at the intersection of a sagittal line drawn distally from the anterosuperior iliac

spine and a transverse line across the superior margin of the greater trochanter. The arthroscope can be inserted into this portal in a direction 45 degrees cephalad and 30 degrees toward the midline. Although this portal places the lateral femoral cutaneous nerve at risk, studies show that the nerve, which arborizes before this point, lies lateral to this portal.29 The other two portals are established at the anterior and posterior margins of the superior edge of the greater trochanter. These portals lie about 4 cm below the superior gluteal nerve, and the posterolateral portal is about 3 cm superior to the sciatic nerve.29 The surgical approach to the iliotibial band for the treatment of snapping hip is made with laterally based portals to access the superficial aspect of the band. The iliopsoas tendon can be released either from the lesser trochanter or accessed from the peripheral compartment.30 Special extra-long cannulas and sheaths (5.25 inches) have been developed for hip arthroscopy. Surgical instruments and the arthroscope can be exchanged easily with these cannulas. Cannulated trocars are commercially available to allow insertion over a guidewire that can be introduced through a large-diameter spinal needle, which facilitates portal placement.

Arthroscopic Anatomy Most of the hip joint can be visualized with the arthroscope (Fig. 2A-11).31 Eighty percent or more of the femoral head can be seen, and the insertion of the ligamentum teres into the anteromedial portion of the femoral head is a readily identifiable landmark. The acetabulum is well visualized with the horseshoe-shaped, lunate surface of the acetabulum that extends to the peripheral labrum and central acetabular fossa. The acetabular fossa extends inferiorly and is filled with vascular adipose tissue extending to the transverse ligament. Viewing the joint through different portals helps evaluate all forms of hip disease.

AL

AW

L

FH

LT

FH

PW PL

Figure 2A-10   Portals for hip arthroscopy include the anterior portal and two portals adjacent to the superior edge of the greater trochanter. (From Miller MD, Cooper DE, Warner JJP: Review of Sports Medicine and Arthroscopy. Philadelphia, WB Saunders, 1995.)

Figure 2A-11   Arthroscopic anatomy of the hip. AL, anterior labrum; AW, wall of acetabulum; FH, femoral head; L, labrum; LT, ligamentum teres; PL, posterior labrum; PW, posterior wall. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRIArthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 101.)

Surgical Principles

127

risk, the anterocentral (dorsalis pedis artery and deep peroneal nerve) and posteromedial (posterior tibial artery and tibial nerve) portals usually are not recommended.

Ankle Arthroscopy Indications The most accepted indication for ankle arthroscopy is removal of loose bodies. This situation is uncommon but may be a result of osteochondritis dissecans or traumatic osteochondral injuries. These lesions can be débrided or repaired with arthroscopic techniques. Synovial or osteophytic impingement is perhaps the most common indication for ankle arthroscopy.32-34 Another procedure that is gaining acceptance is arthroscopically assisted tibiotalar arthrodesis.35 Other relative indications for ankle arthroscopy include adjunctive treatment of fractures, stabilization procedures, and synovectomy for rheumatologic conditions.

Arthroscopic Anatomy

Positioning and Portal Placement

Indications for shoulder arthroscopy are evolving, but there are proponents and opponents for each proposed application, and the controversy continues. Shoulder arthroscopy has been applied to the diagnosis and treatment of shoulder instability, impingement syndrome, distal clavicle problems, rotator cuff disease, inflammatory and degenerative diseases of the shoulder, adhesive capsulitis, sepsis, and other diagnoses. Arthroscopy has been advocated for débridement of loose tissue and washout of degenerative arthritis,38 treatment of adhesive ­ capsulitis,39,40 irrigation and débridement of septic joints,41 removal of foreign bodies, and a variety of other procedures.

Most surgeons perform ankle arthroscopy with the patient supine. Several commercially available ankle distracters are available and provide enough distraction for adequate visualization. Alternatively, the patient’s knee and leg can be positioned over the end of the operating table, and a gauze bandage loop can be fashioned as the surgeon’s foot provides a distraction force.36 The standard 4.0-mm arthroscope can be used for most procedures, although small arthroscopes (2.7 mm) may be helpful for areas that are more difficult to visualize (e.g., the posterior ankle). A 3.5-mm-diameter shaver is often helpful for ankle ­arthroscopy. The most commonly used portals are the anteromedial and anterolateral portals.37 The anteromedial portal is placed just medial to the tibialis anterior tendon at the level of the ankle joint. A spinal needle helps to localize this site and instill saline into the joint. The anterolateral portal is established just lateral to the peroneus tertius tendon at the joint line level (Fig. 2A-12). The nick and spread technique helps to establish these portals to ensure that superficial nerves and veins are avoided. A posterolateral portal, placed just lateral to the Achilles tendon, can be used for outflow or occasionally to visualize the posterior ankle. A 70-degree arthroscope placed through an anterior portal usually provides adequate visualization of the posterior ankle, however. Because of the significant neurovascular

Anterolateral

Anteromedial

The anterolateral portal is used most commonly for visualization. The joint should be evaluated systematically (Fig. 2A-13). The lateral sulcus and tip of the fibula usually can be visualized. The medial sulcus and tip of the medial malleolus is best seen with the arthroscope placed in the anteromedial portal. Pathology is addressed as necessary.

Shoulder Arthroscopy Indications

Positioning and Portal Placement Two methods of positioning are popular: the lateral decubitus position and the beach-chair position (Fig. 2A-14). The lateral decubitus position involves placement of the patient with the affected arm up; traction is used to suspend the arm. Transient neurapraxia can be a common problem with this positioning.42 Other disadvantages of this position are that regional anesthesia is poorly tolerated, conversion to an open procedure is difficult, the capsular anatomy is distorted, and arm positioning can be more

Posterolateral

Figure 2A-12   Portals for ankle arthroscopy are anteromedial, anterolateral, and posterolateral. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 134.)

Figure 2A-13   Arthroscopic anatomy of the ankle. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 134.)

128

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 2A-14   Lateral decubitus and beach-chair positioning for shoulder arthroscopy. (From Miller MD, Chhabra A, Hurwitz S, et al: Orthopaedic Surgical Approaches. Philadelphia, Saunders, 2008.)

­ ifficult.43 Because of these problems, the beach-chair posid tion was developed by Warner.44 This positioning allows easy access for arthroscopy without the disadvantages of other positions. The most commonly used arthroscopic shoulder portals are the posterior, anterosuperior, anteroinferior, and lateral portals. The superior, or Neviaser, portal and the posterolateral portal (of Wilmington) are used by some surgeons as accessory portals (Fig. 2A-15). The posterior portal, placed 2 cm medial and 2 to 3 cm inferior to the

Arthroscopic Anatomy

5 2 1

posterolateral corner of the acromion, is used primarily for arthroscopic visualization. The anterosuperior portal, located lateral to the coracoid and just distal to the anterior edge of the acromion, is used most commonly for instrumentation. The anteroinferior portal, located about 2 cm below the anterosuperior portal just above the subscapularis tendon (visualized arthroscopically), is used primarily for arthroscopic Bankart repair. The lateral portal is used for instrumentation during bursoscopy. The superior (or supraspinatus or Neviaser) portal is located at the corner of the supraspinatus fossa and is oriented slightly anteriorly and laterally. This portal is used for visualization of the anterior glenoid and may be helpful in addressing superior labral and biceps injuries. The posterolateral portal (port of Wilmington) is used most commonly for superior labral repairs.

4

3

Clavicle 5

1

Acromion

2

3

Coracoid 4 Figure 2A-15   Arthroscopic portals for shoulder arthroscopy. 1, Posterior; 2, anterosuperior; 3, anteroinferior;   4, lateral; 5, superior or Neviaser portal. (From Miller MD, Cooper DE, Warner JJP: Review of Sports Medicine and Arthroscopy. Philadelphia, WB Saunders, 1995.)

Although many sequences are recommended for arthro­ scopic evaluation of the shoulder, it is important to visualize and palpate the biceps tendon, labrum, glenohumeral articular surfaces, glenohumeral ligaments, and rotator cuff at a minimum (Fig. 2A-16). During bursoscopy, the superior surface of the rotator cuff, the inferior acromion, and the acromioclavicular joint should be evaluated. The interface of the biceps tendon and superior labrum most commonly is visualized first to allow the surgeon to become oriented; the surgeon can then address the possibility of a superior labral anterior-to-posterior lesion (SLAP). The labrum is inspected and probed to detect any labral injury (Bankart tear), especially in a patient with anterior instability. Separation of the superior labrum from the glenoid may be normal.45 The articular surfaces of the glenoid and humerus are examined for any articular injuries or degenerative joint disease. By rotating the arm, one can visualize most of the proximal humerus. With the arthroscope positioned posterior and inferior to the superior surface of the glenoid,

Surgical Principles

129

Direct posterior Proximal medial Posterolateral Anteromedial

Proximal lateral Anterolateral

Figure 2A-16   Arthroscopic anatomy of the shoulder. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 160.)

internal rotation of the humerus allows visualization of a Hill-Sachs defect, or impression fracture, associated with anterior instability. The glenohumeral ligaments, which represent thickenings of the joint capsule, can then be evaluated. The superior glenohumeral ligament arises just anterior to the long head of the biceps tendon in the rotator interval (between the biceps and subscapularis). The middle glenohumeral ligament drapes over the subscapularis tendon. The inferior glenohumeral ligament complex is composed of two bands, anterior and posterior, and attaches to the inferior labrum. Arthroscopy of the inferior axillary pouch, the rotator cuff insertion (superiorly), and the subacromial bursa completes the examination.

Elbow Arthroscopy Indications Although the indications for elbow arthroscopy are ­evolving, it is a well-accepted technique for removal of loose bodies, irrigation of an infected joint, synovectomy, and osteophyte excision. Loose or foreign bodies can be removed easily from the joint using the contralateral portal.46-48 A septic joint can be irrigated with only two portals (proximal-medial and posterolateral).49 Osteophytes can be removed using arthroscopic burs and generous irrigation.50 Synovectomy can be accomplished through anterior and posterior portals and can provide considerable pain relief in patients with chronic synovitis.51 Other proposed procedures include arthroscopic capsular52 and radial head excision.49

Positioning and Portal Placement Elbow arthroscopy can be done with the patient supine53 or prone.54 The prone position, which is more popular, allows improved arthroscopic manipulation and better visualization without the use of an overhead suspension device.49

Figure 2A-17   Arthroscopic portals for elbow arthroscopy. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRIArthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 197.)

The soft spot, or anconeus triangle (defined by the lateral epicondyle, tip of the olecranon, and radial head) should be palpated, and 30 to 50 mL of fluid should be injected to distend the capsule before establishing arthroscopic portals. Three portals commonly are used for elbow arthroscopy: the proximal-medial, posterolateral, and anterolateral portals. A fourth portal, the direct posterior portal, can be established when posterior instrumentation is necessary. A final portal, the proximal-lateral portal, has been advocated for improved safety (Fig. 2A-17).55-57 The nick and spread method should be used when establishing all arthroscopic portals around the elbow because of the attendant neurovascular risk. The proximal-medial portal is the primary diagnostic arthroscopic portal with the patient in the prone position.49 This portal is located 2 cm proximal to the medial epicondyle, just anterior to the intermuscular septum. The arthroscopic sheath should contact the anterior surface of the humerus and is directed toward the radial head. The posterolateral portal is made in the anconeus triangle and can be used for inflow and outflow. The arthroscope can be introduced into this portal to allow visualization of the posterior joint. Some authors favor establishing this portal proximal and posterior to this location, 2 cm proximal to the tip of the olecranon and adjacent to the lateral edge of the triceps tendon. The anterolateral portal is established 3 cm distal and 2 cm anterior to the lateral epicondyle with the elbow flexed 90 degrees. The portal can be established with an inside-out technique using a blunt rod. The radial nerve is at risk with this portal and is immediately adjacent to the portal in 50% of cases.57 The direct posterior portal is

130

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

made 2 cm proximal to the tip of the olecranon, through the triceps tendon, and is used primarily for instrumentation in the posterior joint. The proximal-lateral portal was developed to reduce the risk associated with other portals.56 It is located 1 to 2 cm proximal to the lateral epicondyle, directly on the anterior surface of the humerus. As with the proximalmedial portal, the trocar is kept in direct contact with the anterior humerus during insertion. The proximal-lateral portal has been advocated in lieu of the more hazardous anterolateral portal.58 The radial head and capitellum can be well visualized through this portal. The anteromedial portal, located 2 cm distal and 2 cm anterior to the medial epicondyle, can jeopardize the anterior branch of the medial antebrachial cutaneous nerve and the median nerve and usually is not recommended. The standard anterior portals used in elbow arthroscopy should be proximalmedial and proximal-lateral portals. The anteromedial and anterolateral portals, if necessary, should be established proximal to the radial head.

Arthroscopic Anatomy With the arthroscope placed in the proximal medial portal, the humeroulnar and radiocapitellar joints, the coronoid fossa, and the medial and lateral gutters can be well visualized (Fig. 2A-18). The posterior joint (olecranon and olecranon fossa) can be well visualized through the posterolateral portal.

Wrist Arthroscopy Indications Wrist arthroscopy is becoming increasingly useful in the diagnosis and treatment of wrist disorders. Treatment of injuries to the triangular fibrocartilage complex

(TFCC), treatment of ligament injuries, fracture management, and treatment of wrist disease are enhanced with ­arthroscopy. Type 1C volar rim tears of the TFCC that involve the origins of the ulnotriquetral and ulnolunate ligaments are managed with an open procedure after arthroscopic identification. The treatment of these injuries is controversial, but two good choices currently exist. Bednar and Osterman59 advocated direct repair augmented with a strip of the flexor carpi ulnaris tendon. The other option is to advance the origin of the ulnotriquetral and ulnolunate ligaments into the triquetrum with a bone anchor. Type 1D tears of the radial attachment of the TFCC by definition involve the dorsal radioulnar ligament or the volar radioulnar ligament, or both. Tears close to the radial attachment of the TFCC, which spare the dorsal ­ radioulnar ligament and volar radioulnar ligament, are best considered type 1A tears and are treated with débridement. Arthroscopic reduction and fixation of intra-articular distal radius fracture is being employed more frequently. In the search for the most appropriate and beneficial treatment for this difficult injury, the arthroscope has some important potential advantages.52,58

Positioning and Portal Placement Wrist arthroscopy is performed with the patient supine. Wrist distraction (10 to 12 pounds) usually is achieved with a commercially available traction device. A small 1.9- to 2.7-mm arthroscope and small joint instruments with a maximal diameter of 3.0 mm are used. Arthroscopic portals around the wrist are designated based on their location in reference to the dorsal extensor compartments (Fig. 2A-19). The standard portals for any diagnostic wrist arthroscopy are the 3-4, the 4-5, and

STT MCR 1-2

MCU 6U 6R 4-5

1 6

Figure 2A-18   Arthroscopic anatomy of the elbow. (Inset, posterior view.) (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 198.)

5

4

3

3-4

2

Figure 2A-19   Arthroscopic wrist portals. Numbers indicate wrist extensor compartments and associated portals. MCR, midcarpal radial; MCU, midcarpal ulnar; R, radial; STT, scaphotrapeziotrapezoid; U, ulnar. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 220.)

Surgical Principles

the radial midcarpal portal. Accessory radiocarpal portals include the 1-2, 6R, and 6U. The 3-4 is the primary portal for arthroscopic visualization. Inflow usually is established through the scope. Outflow can be done with an 18-gauge catheter through either the 1-2 or 6U portal. Alternatively, a pressure-sensitive inflow-outflow pump device can be used. The 4-5 and 6R portals are primarily for instrumentation. The optimal configuration for working on the TFCC is to place the arthroscope in the 4-5 portal and to introduce instruments through the 6R portal. The midcarpal portals include the midcarpal radial, midcarpal ulnar, and scaphotrapeziotrapezoid (STT) portals. The midcarpal radial portal is 1 cm distal to the 3-4 portal, and the midcarpal ulnar portal is 1 cm distal to the 4-5 portal. Generally, visualization is accomplished through the midcarpal radial portal. Instruments are introduced through the midcarpal ulnar portal, although the steps can be reversed. The STT portal is located just ulnar to the extensor pollicis longus tendon over the STT joint. When débriding the STT joint, visualization is accomplished through the midcarpal radial portal. Instruments are introduced through the STT portal.

Arthroscopic Anatomy A thorough and complete wrist arthroscopy details the anatomy of the articular surfaces, intrinsic carpal ligaments, extrinsic carpal ligaments, and TFCC (Fig. 2A-20). ­Proceeding from radial to ulnar in the radiocarpal joint, the following structures are observed: radial styloid, radioscaphocapitate ligament, long radioulnate ligament, scaphoid proximal pole, radius scaphoid facet, scapholunate interosseous ligament, radioscapholunate ligament (ligament of Testut), lunate proximal pole, radius lunate facet, short radiolunate ligament, TFCC (including the dorsal and volar radioulnar ligaments), ulnolunate ­ ligament, ulnotriquetral ligament, lunatotriquetral ­ interosseous ligament, and triquetrum proximal pole. Proceeding from radial to ulnar in the midcarpal joint, examination of the following structures is accomplished: STT, capitate proximal pole, scaphoid distal pole, scapholunate interval stability, lunate distal pole, lunatotriquetral interval stability, triquetrum distal pole, capitohamate interval, hamate proximal pole. The ligaments that cross the midcarpal joint often are difficult to visualize because of overlying synovium. If a midcarpal instability is suspected, visualization can be accomplished by débriding the synovium carefully. The ligaments in question include the radioscaphocapitate, ulnocapitate, triquetrocapitate, and triquetrohamate.

CONCLUSION Arthroscopy is a useful tool in the orthopaedic surgeon’s operative approach. Arthroscopy is not the only tool, however, and well-founded open techniques should be available and understood. New applications of arthroscopy in the future may include cartilage regeneration, ligament reconstruction and repair, and genetic engineering.

131

Figure 2A-20   Arthroscopic wrist anatomy. (From Miller MD, Osborne JR, Warner JJP, Fu FH [eds]: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 221.)

C

r i t i c a l

P

o i n t s

• Appropriate diagnosis with physical examination and imaging studies remains of particular importance. • Arthroscopy has the advantage of less invasive surgical approach, shorter operative time, and potentially faster recovery. • Arthroscopic surgery is not without risk, and extreme caution should be used in patient selection, portal placement, and careful technique. • Indications for arthroscopic surgery are variable, depending on the joint, and are constantly evolving. • Arthroscopic techniques should never replace the surgeon’s ability to perform standard open approach surgical procedures.

S U G G E S T E D

R E A D I N G S

Coward DB: General principles and instrumentation of arthroscopic surgery. In Chapman MW (ed): Operative Orthopaedics. Philadelphia, JB Lippincott, 1988, pp 1549-1559. DiGiovine NM, Bradley JP: Arthroscopy equipment and setup. In Fu FH, Harner CD, Vince KG (eds): Knee Surgery. Baltimore, Williams & Wilkins, 1994. Jackson RW: Quovenis quovadis: Evolution of arthroscopy. Arthroscopy 15: 680-685, 1999. Johnson LL: Arthroscopic Surgery: Principles and Practice, 3rd ed. St. Louis, CV Mosby, 1986. Olsen EJ, King NA: Arthroscopic anatomy. In Fu FH, Harner CD, Vince KG (eds): Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 77-99.

R eferences Please see www.expertconsult.com

132

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n

B

Suture Materials Marius von Knoch and Christian Gerber

Essentially, no orthopaedic procedure is carried out without the use of sutures. The selection of the type and size of the suture material is, however, often related to what a surgeon has learned empirically during training rather than to logical conclusions based on material properties and imposed demands. The role of the type and size of suture material and the surgical suturing technique in the development of fixation failure are not uniformly ­established. The technique of tying knots in combination with arthroscopic techniques has been given particular attention recently.1 It is established that the success of orthopaedic operations (e.g., tendon or ligament repairs) depends on the type of suture material used and the technique.2-9 Thus, a more precise analysis of the role of sutures and suturing ­techniques is needed to allow a more scientifically based selection of material and of suturing and knotting technique. Among the several forms in which surgical sutures are available today are braided, unbraided, absorbable, nonabsorbable, hybrid, polyblend, and sutures of different materials.10 This chapter aims to describe different types of sutures and discuss advantages of and indications for ­different forms.

IMPOSED DEMANDS ON SUTURE MATERIALS Different demands are imposed on surgical suture materials. Their importance or priority may change with the specific application of the suture material considered. A variety of mechanical, biomechanical, and biologic properties should be considered when a specific suture material is selected. Such properties include the following: 1. Biologic characteristics a. Biocompatibility b. Antimicrobial characteristics 2. Mechanical characteristics a. Ultimate tensile strength b. Elasticity and deformation under load (gap formation under tensile load) c. Abrasion resistance d. Adequate maintenance of mechanical properties during healing (absorbable sutures) e. Knotting properties (ease of knotting, loss of strength after tying knots, knot slippage, number of knots necessary for stability, knot prominence) 3. Handling characteristics a. Ease in practical use

BIOLOGIC CHARACTERISTICS AND BIOCOMPATIBILITY A number of studies have assessed the compatibility of ­different suture materials, either by means of semiquantitative analysis of the histologic foreign body reaction to the implanted material or by means of experimental or clinical testing of the healing properties of the sutured tissues.7,11-22 Biocompatibility depends on the type of suture material, its structure (braided versus monofilament), the amount of material implanted, and the site of implantation.23 Three principal phases of healing of soft tissues to bone have been identified in animal experiments: inflammation, repair, and remodeling.10 Each phase can be influenced by specific characteristics of the sutures used, and a rational choice leads to clinical success. The initial inflammatory reaction is characterized by the presence of polymorphonuclear cells, lymphocytes, and monocytes. This acute inflammatory foreign body reaction peaks between days 2 and 7. By day 4, mononuclear cells start to predominate, and fibroblasts appear. By day 7, mature fibroblasts are present; the foreign material becomes encapsulated in a fibrous mantle by day 10.23 At that point, no further tissue reaction is expected if the implanted material is nonabsorbable. Most currently used nonabsorbable materials are inert and are therefore extremely well tolerated. Absorbable materials elicit a “second” boost of inflammatory reaction at the time of their resorption. The intensity of this reaction depends on the specific chemical process that leads to resorption and on the amount of material to be resorbed. Previous reports of lytic response after implantation of PGA (polyglycolic acid) sutures have led to the use of less reactive polymers. A higher rate of material degradation has been associated with an increased cellular response.24 Among currently used degradable materials are PGA, PLLA (poly-l-lactic acid), PDS (polydioxanone), poly-D, PDLLA (l-lactic acid), and their copolymers.24 In general, a higher percentage of PLA is associated with slower resorption than with PGA.10 Suture materials such as polyglactin (Vicryl), polyglycolic acid (Dexon Plus), polyglyconate (Maxon), poliglecaprone 25 (Monocryl), polydioxanone (PDS), and poly(l-lactide/glycolide) (Panacryl) are absorbed by simple hydrolysis, whereas catgut requires enzymatic degradation, which tends to provoke a much more intense soft tissue reaction. Of the commonly used materials, monofilament stainless steel provokes the least amount of foreign body reaction in skin and other musculoskeletal tissues. Almost no foreign body reaction is seen after implantation of nylon,

Surgical Principles

polypropylene (Prolene), polyester (Ethibond, Tevdek, Ti-Cron), polybutester (Novofil), or absorbable materials (Maxon and PDS). Polyglycolic acid (Dexon Plus) and polyglactin (Vicryl), which are dissolved by simple hydrolysis, also elicit a minimal foreign body response but are probably tolerated somewhat less well than polydioxanone (PDS) and polyglyconate (Maxon).16,17,19,25 The tissue response may be more pronounced if these materials are used in the skin. Catgut and, at the time of its resorption, chromic catgut cause a moderate to intense inflammatory response.14,26 Silk, which used to be the standard material for skin closure, is probably the least well tolerated of all materials still in use, and its application experimentally has been proved to compromise the results of intra-articular ligament repairs.7,19 Optimal selection of materials for skin closure has also been shown clinically to reduce the incidence of wound infection (polydioxanone plus polypropylene versus chromic catgut plus silk).22 A monofilament structure has a variety of biologic advantages. There is less surface area exposed to the body so that the foreign body reaction is less intense than that seen with multifilament sutures.23 Because less suture material is exposed to hydrolysis, monofilament sutures retain their mechanical properties longer.12 There is increasing concern that braided capillary materials may favor the propagation of infection, whereas noncapillary materials or monofilament sutures do not.11,23 It appears that bacteria can colonize these materials, not so much on the surface of the suture as within it, where immunocompetent cells have insufficient access. This accounts for the lack of support for the use of multifilament sutures in situations of potential contamination.13 Sutures with antimicrobial coatings are under development.27 A silver-doped bioactive glass powder was used to coat resorbable Vicryl (polyglactin 910) and non-resorbable Mersilk surgical sutures, thereby imparting bioactive, antimicrobial, and bactericidal properties to the sutures. Laboratory testing showed that the bioactive glass coating did not affect dynamic mechanical and thermal properties of the sutures. Resorbable sutures with bioactive coatings may open new opportunities for the use of antimicrobial sutures in surgery. Because the implantation of braided suture material very close to the skin incites a much more intense and lasting reaction than burying the suture material in well­vascularized tissue, and because contamination very close to the epidermis is always possible, we suggest that the use of braided suture materials immediately beneath the epidermis should be avoided. Clearly, the presence of large amounts of suture material incites more intense foreign body reactions. Therefore, sutures with optimal strength and knotting characteristics are needed so that small sutures requiring few throws for a stable knot can be used.

MECHANICAL CHARACTERISTICS Suture Strength From an engineering standpoint, ultimate tensile strength should be measured in relation to the cross section of the material tested. Material properties become system

133

­ roperties when the suture material is knotted. For pracp tical purposes, however, the orthopaedic surgeon selects the suture size exclusively according to the United States Pharmacopeia (USP) size, which is designed in numbers ranging from 12-0 to 6. When comparing data from ­different suture materials of different sizes, it should be understood that the cross section (or the diameter) of one material with a specific USP number may be different from the cross section of another material with the same USP number.26 In addition, we never use unknotted sutures in daily practice. This system appears reasonable because the surgeon is not interested in verifying the diameter of a certain suture and because essentially all sutures invariably fail at the site of the knot.28,29 Until recently, only thin suture materials had been widely tested, and the mechanical properties of heavier materials (e.g., sizes 0 to 6) were only sparsely documented.12,15,16,25,29-32 Repairs of large musculotendinous units are performed with the use of thick sutures, which have so far not been proved adequate, let alone optimal, for their respective ­applications. We recently tested the mechanical in vitro properties of the heavier sutures (gauges 0 to 3).28 Not all suture materials were available in all sizes. Maximal tensile in vitro strength of comparably sized sutures was historically found for monofilament stainless steel sutures and for absorbable monofilament materials (polyglyconate [Maxon] and polydioxanone [PDS]). Braided absorbable polyglactin (Vicryl) and polyglycolic acid (Dexon Plus), as well as braided polyester (Mersilene, Ethibond, Tevdek), showed excellent ultimate tensile strength, whereas nylon (Dermalon, Prolene) was somewhat weaker. Our own previous data were in rough agreement with those found in the literature. In contrast to the study of Bourne and colleagues, we did not find a decrease in the ultimate tensile strength of wet sutures compared with dry material.12 Recently, high-strength polyethylene polyblend sutures have been introduced that are less prone to breakage during knotting. The first of these sutures was introduced under the trade name FiberWire and has rapidly been accepted and used by the orthopaedic community. These sutures are characterized by a core of fine strands of UHMWPE (ultra–high-molecular-weight polyethylene), which is surrounded by braided polyester. Other brand names are Magnum Wire, Ultrabraid, Force Fiber, ­ MaxBraid PE, and Hi-Fi. All these latter sutures are composed exclusively of UHMWPE.24 A partially resorbable suture of this new kind is Orthocord, which consists of a UHMWPE sleeve covering a resorbable PDS core. While the PDS core dissolves over time, the nonresorbable UHMWPE sleeve retains some of the initial stability.24 Suture abrasion is mainly observed during knot tying.33 Different modes of abrasion have been described, among which are abrasion against bone, abrasion during sliding through a knot pusher or through an anchor eyelet, or inherent abrasion at the knot site.34 A biomechanical study showed that polyblend sutures may be advantageous over conventional Ethibond with respect to abrasion resistance.33 As a result, failure of tendon to bone repair with new polyblend sutures can occur at the suture-­tendon margin, as slippage of the anchor, or as eyelet ­failure of an

134

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

absorbable anchor.35 The cutting characteristics of different sutures through tendon have not been well established to date. In a mechanical study performed at our institution, four types of polyblend sutures were tested: FiberWire, Herculine, Orthocord, and Ultrabraid.36 Fretting resistance was tested on eyelets of metallic and absorbable suture anchors. The ultimate strength of a polyblend suture material was 2- to 2.5-fold compared with Ethibond and PDS sutures. The resistance to fretting was up to 500fold better than that of Ethibond or PDS sutures. This makes polyblend sutures particularly useful with metallic edges of anchors or prostheses or with absorbable anchor eyelets.

DEFORMATION UNDER TENSILE LOAD Although monofilament sutures tend to be more compliant, they are favored for certain arthroscopic techniques because they can be passed more easily through arthroscopic instruments.37 Both suture elongation under load and thread or knot failure, however, may lead to gap formation and may impair successful healing and functional recovery.6,33 The very strong, absorbable monofilament polyglyconate (Maxon) and polydioxanone (PDS II) sutures are very compliant under tensile loads, as opposed to the also very strong and absorbable, but braided, polyglycolic acid (Dexon) and polyglactin 910 (Vicryl) sutures, which are very stiff. Among the commonly used nonabsorbable sutures, only monofilament stainless steel sutures are stiff; the other monofilament sutures (polypropylene and nylon) are very compliant. The most commonly used nonabsorbable braided suture material has been polyester and has been partially replaced by polyblend sutures. Clinically, it may be advantageous to use a very compliant suture, especially a running type, because it may yield rather than break.38 For tendons, suture repair techniques that prevent gapping are considered optimal because scar and adhesion formation is reduced and early functional treatment, which promotes remodeling, can be undertaken.4,36 An optimal tendon-to-bone repair should not allow gap formation under tensile load, but moderate extensibility may be beneficial for healing. If a rotator cuff tendon is sutured to a trough in the greater tuberosity, transosseous suture loops of roughly 7 cm are tied over the proximal humeral cortex or a miniplate. If such repairs are brought under a tension of, for example, 200 N, which is possible when the arm is lowered, the elasticity of the suture material alone may allow gapping of the repair. Under a load of 200 N, the suture material properties alone would allow the following gaps: 2.2 mm for Ethibond No. 3, 3.2 mm for Ethibond No. 1, 6.7 mm for Surgilon No. 1, and 9.1 mm for PDS II No. 1.28 Therefore, suture material alone may prevent stable anchoring of a tendon in bone if the repair is subjected to load; the selection of appropriate suture material is critical. Also, it is extremely important to know the loads to which a repair will be subjected to determine the optimal suture material.

MAINTENANCE OF MECHANICAL PROPERTIES IN VIVO It is widely believed that nonabsorbable suture materials maintain their strength during healing. This is true for braided polyester (Ethibond), which remains stable after implantation.39 It is surprising, however, that Ethilon and Prolene, which both are nonabsorbable, lose 56% and 74%, respectively, of their initial ultimate tensile strength 6 weeks after subcutaneous implantation.39 Even greater changes in the mechanical properties of absorbable suture materials are demonstrated after implantation.39 Catgut loses its strength essentially within 1 week. Polyglactin 910 (Vicryl) loses about 50% of its initial ultimate tensile strength after 2 weeks, and then loses 75% by 3 weeks and 100% by 4 weeks.12 The material is absorbed after 56 to 90 days.14,40 Polyglycolic acid (Dexon Plus) has similar mechanical properties in vivo; 20% of its initial strength is lost within the first week, 50% is lost by 2 weeks, and essentially 100% is lost by 4 weeks.21 Dexon Plus remains in the wound longer than Vicryl and is resorbed only after about 120 days.40 Poly­ glyconate (Maxon) loses its strength distinctly less rapidly; its very high initial strength is reduced by roughly 50% after 3 weeks and is almost completely lost by 6 weeks.12,39 Polydioxanone (PDS) is initially somewhat less strong. Because it seems to be hydrolyzed more slowly than Maxon, it is already distinctly stronger at 3 weeks after implantation and maintains about 14% to 50% of its initial breaking strength for 6 weeks.12,39 Its resorption requires roughly half a year.14 Of particular interest is another absorbable suture material, poly(l-lactide/glycolide) (Panacryl). According to the manufacturer, Panacryl maintains 80% of its ultimate tensile strength for 3 months after in vivo implantation and maintains 60% of its strength for half a year. Full absorption occurs after 1.5 o 2.5 years. For absorbable sutures, Panacryl preserves its mechanical properties longer than any other suture material; it almost behaves as a nonabsorbable suture. The in vivo behavior of new polyblend sutures is not well established to date. The selection of the appropriate suture material depends on the expected type and rate of wound healing. For simple adaptation of subcutaneous tissue that is not under tension, the rapid resorption of polyglycolic acid or of polyglactin 910 may be desirable. Because the fibroblastic response dictates that a healing wound will rapidly regain strength between days 5 and 14, and because collagen content increases until day 42 with subsequent remodeling of the wound, a suture material such as polyglyconate (Maxon) may have optimal resorption characteristics for a wound that is under slight tension.23 If a tendon or fascia is repaired under moderate tension and a longer period of protection is desired, a material such as polydioxanone (PDS) may be optimal. If, however, prolonged holding power is required and gapping is to be prevented, braided polyester (Ethibond, Tevdek, Mersilene) or a polyblend polyethylene suture may be the best choice. Any recommendation in favor of polyblend polyethylene sutures is based on current knowledge, whereas clinical proof of its superiority remains to be established.

Surgical Principles

135

KNOTTING PROPERTIES Tera and Åberg have introduced an internationally accepted terminology for knotting techniques.29 They distinguished between parallel and crossed knots and established that in general, parallel knots are more stable than crossed knots. Knotting properties appear to be similar for different sizes of the same material.25,41 Knots are indeed of great importance because suture material almost invariably fails at the site of the knot.15,28,29 Stainless steel loses little strength when knotted, whereas catgut and silk lose a large part of their strength when knotted.29,41 Indeed, the failure strength of suture material may be reduced by 10% to 70% by tying a knot.15 Slipping of the knot depends on the number of throws and on the material. In general, any knot with six throws should not slip.12 We were, however, able to show that the 2 = 1 = 1 configuration leads to stable knots for all tested sutures, and this is our preferred technique, although some sutures may require even fewer throws.28 Zechner and colleagues have shown that the reduction in tensile strength can be partly prevented if knots are tied in the horizontal branch of the suture rather than in the region where linear stress is applied.42 Although one may believe that knots of monofilament sutures slide easily, therefore necessitating multiple knots, this is not true; polyglyconate has excellent knotting properties, as have other monofilament sutures.12,17,21,28 The braided absorbable sutures tend to slip less when wet.1 Knotting is more delicate in coated materials such as Ti-Cron (braided

A

Figure 2B-2   Roeder’s knot security depends on the number of initial turns around the standing part.

polyester) because coating reduces friction and therefore favors slipping of the knot. Coated and particularly stiff sutures are better knotted with double throws.15,21 Several new arthroscopic sliding knots have been introduced. Ten commonly used knots were tested in a mechanical study: Dines knot, Duncan loop (Fig. 2B-1A), Tennessee slider (see Fig. 2B-1B), field knot, giant knot, Roeder’s knot (Fig. 2B-2), Nicky’s knot (Fig. 2B-3), SMC knot, Snyder knot, and Weston knot.43 Knots were tested for forward and backward sliding characteristics, loop security before securing with half-hitches, resistance to sliding, knot security after securing with three alternating half-hitches on alternating posts, resistance to sliding, and distance to failure. The Dines knot performed best, exhibiting superior biomechanical characteristics in six of seven measured categories. The only category in which the Dines knot did not perform superiorly was forward sliding, indicating that the Dines knot can be rated as intermediate in ease of placement.43 Another study investigated the optimal knot configuration for maximal knot and loop security.44 A static surgeon’s knot provided the best balance of loop security and knot security within the knot configurations tested in that study. A sliding knot without the addition of three reversing half-hitches on alternating posts had both poor loop

B

Figure 2B-1   A, The Duncan loop is a typical sliding knot.   B, Tennessee slider. The loop strand is thrown around the post and loop strand one time and then around the post strand only. It is then brought up between the parallel limbs between the first and second loops.

Figure 2B-3   Nicky’s knot is a one-way sliding knot with high initial holding capacity. It maintains tension on soft tissue while additional hitches are being tied.

136

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

security and knot security and was not recommended. The addition of three reversing half-hitches on alternating posts improved knot security of all sliding knots tested and improved loop security of most of the sliding knots tested. The addition of three reversing half-hitches on alternating posts improved the knot security of all sliding knots to adequately resist predicted in vivo loads. The Roeder knot with three reversing half-hitches on alternating posts provided the best balance of loop security and knot security within the sliding knot configurations tested in that study regardless of suture type. Tying a surgeon’s knot or a sliding knot with three reversing half-hitches on alternating posts using No. 2 FiberWire increased knot security over the same knot tied with No. 2 Ethibond. In another mechanical study, performance characteristics of three different knots were compared between standard Ethibond suture and a braided, high-strength polyethylene suture and were in accordance with the aforementioned results.45 In contrast to this, it has been observed by other investigators under different testing conditions that FiberWire knots may be more prone to knot slippage and failure than Ethibond knots.46 At least two throws more than with conventional sutures have to be used for knot tying according to our experience.36 Accordingly, it has been described that up to seven knots are required to secure a polyblend suture.47 It was concluded that polyblend sutures may be most suitable in regions with subcutaneous fat in order to prevent frictional problems in tight spaces, such as the pulley space of the long head of the biceps tendon. In our hands, the sliding knot is the most convenient surgically; it corresponds to a parallel square knot and has the same properties.48 It should be noted, however, that a double throw for the first knot increases tensile strength in a statistically significant manner.28 If such a double throw is pinched with a ribbed needle holder, the tensile strength of sutures sized 0 to 3 is not significantly impaired. Pinching a smaller-sized thread reduces the tensile strength by up to 30%, however.12,28 As an arthroscopic knot, we prefer to use Nicky’s sliding knot.

HANDLING PROPERTIES Modern suture materials are available with atraumatic precision needles, which are adapted to the intended use of the suture and thus essentially solve needle problems. Among suture materials, stainless steel has little popularity despite its soft tissue tolerance because of its stiffness and its potential for breaking if the suture is kinked.2 Braided, high-strength polyethylene sutures have become very popular among orthopaedic surgeons. In vitro testing of these sutures has been very promising. Clinical data will show whether the increased costs for using these sutures are warranted. Handling of the sutures remains a matter of personal preference, and there are no clear-cut advantages of one suture over another in regard to handling properties. In our practice, we have seen no distinct handling disadvantages of these sutures.

A u t h o r s ’ P r e f e rr e d M e t h o d Personal Approach to the Selection of Sutures

To fix large tendons to bone, we prefer polyblend polyethylene sutures. As a sliding knot, we prefer to apply Nicky`s knot. For tendon sutures, we use fine polypropylene (Prolene) or polydioxanone (PDS). They are somewhat elastic but have extremely high tensile strength, and polydioxanone maintains its mechanical properties sufficiently long to allow tendon healing.24 If strength needs to be maintained, as in closures of aponeuroses (fascia lata), we prefer to use heavier elastic running sutures with polydioxanone. In subcuticular tissue, the breaking strength of the suture can be lost rapidly, so 4-0 polyglyconate (Maxon) appears optimal. For sutures placed very close to the skin, as well as in situations with questionable contamination, we try to avoid the use of braided suture materials. For closure of the skin, we prefer polypropylene (Prolene), 4-0 or 3-0. We do not use very rapidly absorbed polyglactin (Vicryl Rapid) or polyglycolic acid (Dexon), which have been recommended for skin closure, but they can be used for approximation of subcuticular tissues.49-51 Within the skin, these materials tend to cause irritation and may serve as a wick, promoting contamination. We have no use for any form of catgut.52 The selection of suture remains personal in every field of surgery but should be done on a rational basis.53

C

r i t i c a l

P

o i n t s

l High-strength polyethylene polyblend sutures are less prone to breakage during knotting. l These sutures are characterized by a core of fine strands of UHMWPE (ultra—high-molecular-weight polyethylene), which is surrounded by braided polyester. l Polyblend sutures may be advantageous over ­conventional Ethibond with respect to abrasion resistance. The ­ultimate strength of a polyblend suture material is 2- to 2.5-fold compared with Ethibond and PDS sutures. The resistance to fretting is up to 500-fold better than that of Ethibond or PDS sutures. l Failure of tendon to bone repair with new polyblend sutures can occur at the suture-tendon margin, as ­slippage of the anchor, or as eyelet failure of an absorbable anchor.

S U G G E S T E D

R E A D I N G S

Abbi G, Espinoza L, Odell T, et al: Evaluation of 5 knots and 2 suture materials for arthroscopic rotator cuff repair: Very strong sutures can still slip. Arthroscopy 22:38-43, 2006. Elkousy HA, Sekiya JK, Stabile KJ, McMahon PJ: A biomechanical comparison of arthroscopic sliding and sliding-locking knots. Arthroscopy 21:204-210, 2005. Mahar AT, Moezzi DM, Serra-Hsu F, Pedowitz RA: Comparison and perfomance characteristics of 3 different knots when tied with 2 suture materials used for shoulder arthroscopy. Arthroscopy 22:614. e1-e2, 2006. Wüst M, Meyer DC, Favre P, Gerber C: Mechanical and handling properties of braided polyblend polyethylene sutures in comparison to braided polyester and monofilament polydioxanone sutures. Arthroscopy 22:1146-1153, 2006.

R efer����� ��������� ences Plea����������������������������������������������� se see www.������������������������������������ expertconsult.com

Surgical Principles

S e c t i o n

137

C

Allograft Tissues Barton R. Branam and Darren L. Johnson

The use of allograft tissues in orthopaedic surgery is increasing yearly. The most commonly used musculoskeletal allografts are bone grafts and those used for ligament reconstruction. Osteochondral and meniscal allografts are also available to the surgeon for more complex joint reconstructions. The main driving force behind the keen interest in allograft tissue use is that it eliminates the need to harvest autogenous graft tissue. The reduction of donor site morbidity and rapid rehabilita­ tion are enticing, especially to the sports medicine surgeon. Meniscal transplantation or reconstruction relies totally on allograft tissue sources. Many find the idea of an ample supply of “spare parts’’ very appealing. Before implantation, we must understand the science behind the procurement, the steriliza­ tion, the storage, and the clinical use of allograft tissue before we can confidently offer them as an option to patients. The purpose of this chapter is to describe the current concepts re­ garding the science and use of allograft tissues for knee recon­ structive surgery.

Besides the elimination of donor site morbidity, the ben­ efits of using allografts instead of autografts include smaller incisions as well as the ability to choose among various graft types and sizes. This allows the surgeon to “customize” the specific graft to the exact patient’s needs. The ability to prepare graft tissues for implantation before the surgical procedure starts is also a benefit in saving tourniquet time, total operating room time, and overall surgical morbidity. This is a great asset for the surgeon working without an experienced assistant. Furthermore, the reduction in oper­ ating room time, the elimination of autogenous graft har­ vesting, and the decreased need for postoperative analgesia arguably offset the cost of the allograft tissue used. There are risks that go along with the use of allograft tissue, which must be explained accurately to the patient during preoperative counseling. Although remote, there is an estimated risk for disease transmission with musculo­ skeletal allografts of about 1 in 1.5 million.1-8 This risk has been reduced and more clearly defined as allograft tissue processing has become more controlled with the establish­ ment of the standards of the American Association of Tissue Banks9 (AATB) and the U.S. Food and Drug Adminis­ tration (FDA). The other risks with the use of allograft tissue are a slower remodeling incorporation process, with the potential for an adverse immune response.10-13 The immune response is often described as subclinical, although further research is needed in this area.

HISTORY Allograft tissue has been used in orthopaedic surgery since the late 19th century. MacEwen reported using allograft bone in 1880.14 Lexer first reported osteoarticular allograft

use in 1908.15 In 1925, Lexer reported his results on 23 whole-joint and 11 hemijoint transplantations around the knee, with a declared 50% successful outcome.15 Both Shino and coworkers and Noyes and associates began using allograft tissues for ligament reconstruction in 1981 and have subsequently provided long-term data from these cases.16-20 Finally, meniscal allograft transplantation was first reported in 1984 by Milachowski and colleagues.21 The operative techniques and the postoperative reha­ bilitation protocols in knee reconstructive surgery have evolved significantly. For example, ligament fixation meth­ ods and rehabilitation protocols after knee ligament recon­ struction continue to evolve and improve outcomes. With this in mind, it is difficult to compare clinical outcome data from earlier studies on allografts to those done more recently. It is especially important to give critical attention to the methods of procurement, sterilization, and storage of the allografts when reviewing the literature. Allografts are not all processed in the same way. Differ­ ences in sterilization and storage have a tremendous bear­ ing on clinical outcome. For example, the use of ethylene oxide to sterilize allograft tissue was popular in the 1980s because of its effectiveness in killing bacteria, viruses, and fungi. It was later found, however, to cause a synovial reac­ tion as well as bone changes after implantation. Ethylene oxide residue levels on graft tissues are now strictly regu­ lated.22,23 Similarly, the use of gamma irradiation to sterilize allograft tissues has been regulated owing to a dose-related weakening of the collagen. This effect was not appreci­ ated fully until the late 1980s and the early 1990s, when a number of studies proved and quantified the effects of different radiation levels on the biomechanical properties of allograft tissues.12,16,18,24-29 With these examples alone, it is easy to understand the need to regulate the processing of allograft tissues. Today, the two major regulatory bod­ ies that monitor the processing of allograft tissues are the FDA and the AATB. The first bone bank reported on was established in 1942.30 Bone banks later increased in number as regional tissue banks became popular in the 1950s. The AATB was founded in 1976 and has been the major influence on the standardization of allograft tissue processing.9 Accredita­ tion by the AATB requires a periodic examination and review of the tissue bank, its methods, and its employees. The AATB provides accreditation for the recovery, process­ ing, and distribution of allograft tissue. Accreditation cer­ tification lasts 3 years. Standards are reviewed and updated annually. On-site inspections by the AATB were started in 1986. Consequently, a number of tissue banks have failed to meet certain requirements on occasion. Membership in the AATB is voluntary, but without its accreditation,

138

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

a ­tissue bank lacks credibility in the medical community. It is highly recommended that orthopaedic surgeons use only tissue banks accredited by the AATB. It is estimated that there are more than 150 tissue banks; less than 100 of them are currently accredited by the AATB. Lack of accredita­ tion should be a red flag to the surgeon concerning the safety and quality of allograft tissue. Conversely, compliance with FDA standards is a strict requirement. In 1993, the FDA issued regulations for allograft donor screening.12 In 1994, FDA personnel con­ ducted on-site evaluations that ultimately led to some tissue bank closures. All the tissue banks in the United States today have been evaluated by the FDA and must remain in com­ pliance with FDA regulations to stay open. This process has been a tremendous step forward in bringing more unifor­ mity to allograft studies in the orthopaedic literature. Allograft safety is being addressed by other groups besides the FDA and AATB. In 1987, the Musculoskeletal Transplant Foundation (MTF) was founded by academic orthopaedic surgeons determined to provide allograft tis­ sue of the highest quality and safety. In the founding year, the MTF distributed about 1500 units of tissue from about 100 donors. In 2002, almost 300,000 units of tissue were distributed from 4431 donors. In more than 15 years and more than 2 million units of tissue transplanted, the MTF has had no confirmed cases of allograft-associated infec­ tion.31 The Tissue Banking Project Team (TBPT) was formed by the American Association of Orthopaedic Sur­ geons in February 2002. The goal of the TBPT is to work together with the FDA and Centers for Disease Control and Prevention (CDC) to formulate guidelines for the safe use of musculoskeletal allografts by evaluating cur­ rent tissue banking procedures in relation to orthopaedic practice.32 Effective July 1, 2005, the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) released new hospital standards and requirements perti­ nent to the tissue banking industry.33 This is applicable to hospitals and specimens found in clinical laboratories, surgical centers, and outpatient centers. This requires overseeing of the individual tissue program within the institution. This essentially includes evaluation, monitor­ ing, and documentation of all steps of the process within the institution, including the reporting and prompt inves­ tigation of any adverse event.34 Allograft use in orthopaedics is becoming increas­ ingly popular. In 2006, about 1.5 million bone and tissue allografts were implanted. Donors have increased from about 6000 in 1994 to more than 22,000 in 2005.34 Despite the recent marked increase in musculoskeletal allograft implantation, there is currently no consensus among or­ thopaedic surgeons regarding the exact indications for use of allograft tissue, particularly in the sports medicine com­ munity. The greatest tool in determining the efficacy of treatment efforts is the blinded randomized prospective study of similar patient populations, in significant num­ bers, treated by different methods with long-term followup. There is a paucity of this type of literature to help us come to a consensus regarding the use of allografts simply because these types of studies involving allograft tissues are extremely hard to do. Surely, as more organizations take an interest and active role in ensuring the safe use of allografts, the number of adverse events will diminish.

Double-blinded, randomized, prospective studies com­ paring allograft to autograft tissue for knee ligament reconstructive surgery are virtually impossible. Because the patient is required to give informed consent regarding graft choice before the surgery, the choice is not blinded to the patient. The patient’s decision on which graft type to use also eliminates randomization. Furthermore, it is impractical to blind the surgeon to graft choice during the operation. Independent examiners and patients would see any differences in surgical incisions as an obvious indica­ tion of the type of graft used. Therefore, it is important to understand the potential for bias in any clinical outcome data on allografts because the studies are not blinded and are without true randomization.

PROCUREMENT The AATB printed its first edition of Standards for Tissue Banking in 1984. In keeping up with the rapid changes in allograft science, the publication has gone through mul­ tiple revisions. The 11th edition of Standards9 was pub­ lished in 2006 and can be ordered online at the AATB’s website, http://www.AATB.org. This information should be understood thoroughly by the clinician to assist in edu­ cating patients about musculoskeletal allografts. The screening of donors is the first line of defense in preventing the transmission of disease with allograft tissue. A questionnaire filled out by the donor, the next of kin, the significant life partner, or other relevant individuals is used to detail the medical, social, and sexual history of the donor. A check for drug use, neurologic diseases, autoim­ mune disorders, metabolic diseases, collagen disorders, and exposure to communicable diseases or unprotected sex is included. If any one of these risk factors is positive, it disqualifies the potential donor. A physical examination and, if available, an autopsy report from the donor are used to detect signs of infectious disease, such as hepatosplenomegaly, lymphadenopathy, oral thrush, or cutaneous lesions (e.g., Kaposi’s sarcoma). The donor is also checked for evidence of sexually trans­ mitted disease, and the genitals and anus are examined for cutaneous lesions or condyloma. Any suggestion of sexually transmitted disease or anal sex disqualifies the potential donor. As outlined in the most recent Standards, several medical conditions, including rheumatoid arthritis, systemic lupus erythematosus, polyarteritis nodosa, sar­ coidosis, and clinically significant metabolic bone disease, preclude musculoskeletal tissue donation in addition to the general exclusion criteria.9 The donor’s blood and serum are tested with aerobic and anaerobic cultures. The AATB requires several infectious disease tests: anti–HIV-1, anti– HIV-2, nucleic acid test (NAT) for HIV-1, ­ hepatitis B surface antigen, total antibody to hepatitis B core antigen, antibodies to the hepatitis C virus (HCV), NAT for HCV, antibodies to human T-lymphotropic virus type I and type II, and syphilis. The harvested tissue is also cultured for aerobic and anaerobic organisms. If a donor is exposed to certain pathogens close to the time of death, a window of vulnerability exists in which an infection can go undetected with serum antibody tests. Initially, nucleic acid testing was not a requirement of tis­ sue banks. As of March 9, 2005, the AATB requires NAT

Surgical Principles

screening for HIV and HCV.35 Rather than looking for antibodies, NAT uses a highly sensitive polymerase chain reaction (PCR) test to look for the actual genetic material of the viruses.34 PCR testing decreases the window period from about 4 to 6 weeks to 10 days for viral RNA.36 The AATB outlines very specific guidelines in regard to the timing and technique of tissue harvesting. Tissue excision for musculoskeletal and osteoarticular allografts should begin within 24 hours of asystole provided that the body was cooled or refrigerated within 12 hours of asystole. However, harvesting should commence within 15 hours of death if the body was not cooled. The harvesting of tis­ sues should be performed under the same standard sterile technique used in the operating room for all surgical cases. Some institutions allow the procurement of tissues to be carried out in a “clean’’ or substerile fashion. These insti­ tutions rely on tissue sterilization at a later date. There is no provision for a substerile technique in the AATB Standards, however.9 After harvesting, the grafts are aseptically wrapped and appropriately labeled. The tissues are usually cooled and taken to the tissue bank. The maximal allowed time for the tissues to remain at wet ice temperatures before final processing or freezing is no longer than 72 hours. Osteo­ articular allografts may remain at wet ice temperatures for 5 days before processing. Osteoarticular and musculoskeletal allograft tissue are kept in a bacteriologically and climatecontrolled environment while processed.9

STERILIZATION Tissue processing and sterilization is a delicate balance between preserving the biologic function of the tissue and removing potentially infectious agents. The ideal method of allograft sterilization would eliminate all poten­ tial pathogens from the graft without compromising the viability of the graft or its biomechanical properties and would not cause any morbidity to the eventual recipient. Currently, there is no ideal method. Gamma irradiation and antibiotic soaks are acceptable methods of second­ ary sterilization after graft procurement. Less common methods include heat and ultraviolet radiation. All these methods are less than ideal and therefore make it neces­ sary to adhere to strict sterile technique during procure­ ment and initial storage phases. Processing methods must be validated to reduce the risk for tissue contamination and cross-contamination. Bacteriostasis can cause false-negative culture results of allograft tissues, which can be ­problematic with spore-forming bacteria.31 However, unless a sporicidal method is used to process the tissue, the tissue should not be considered sterile, and there is risk for possible ­bacterial infection.37 Sterile is considered to be the absence of all living or potentially living microorganisms at the steril­ ity assurance level (SAL) of 10-6. This means that there is a 99.99% probability that the tissue is sterile, which is the same percentage for the total hip or knee implant that might be needed later. Ethylene oxide is an effective agent against bacterial, fungal, and viral agents on nonporous surfaces and is com­ monly used to sterilize surgical instruments. Ethylene oxide was previously a popular and accepted method for sterilizing musculoskeletal allografts. However, Jackson

139

and ­associates found a 6.4% incidence of persistent intraarticular reaction in patients who received an allograft for anterior cruciate ligament (ACL) reconstruction that was sterilized with ethylene oxide.22 The AATB requires elimination of residual ethylene oxide or its breakdown products to less than specified levels.9 Increased graft fail­ ure and chronic synovitis associated with ethylene oxide sterilization have caused most tissue banks to discontinue its use.32,34,36 Gamma irradiation historically has been the most widely used method of secondary sterilization of musculoskeletal allograft. Bacillus pumilis spore control strips are used to confirm that the radiation load is effective in sterilizing the tissue against bacteria. This typically requires a radiation load of at least 1.5 megarad. The elimination of HIV with gamma irradiation requires higher doses of radiation, esti­ mated to be greater than 3.5 megarad.5 Gamma irradiation levels of 3.0 megarad have been proved to significantly alter the mechanical properties of a goat infrapatellar ten­ don.26 In this model, a 27% reduction of maximal failure force and a 40% reduction in the energy-to–maximal force ratio were observed in infrapatellar tendons treated with 3 megarad of radiation when compared with untreated control models.24 Furthermore, human infrapatellar tendons treated with 4 megarad have a significant reduction in lin­ ear stiffness and maximal force compared with untreated control models.27 Therefore, HIV cannot be eliminated from allograft tissues using gamma irradiation without compromising the mechanical properties of the graft. Currently, 2.5 megarad is the recommended maximal dose to avoid altering the biomechanical structure of the graft while providing maximal bacterial eradication.36 Antibi­ otic soaks are typically used synergistically with low-dose gamma radiation. Antibiotic solutions kill bacteria and viruses; however, the effect is limited by incomplete tissue penetration.32 There is no one best way to sterilize allograft tissue. All sterilization processes have the potential to affect bio­ mechanical and biologic properties. Techniques simply vary with each tissue bank. The graft is swabbed and is checked for bacterial contamination at the conclusion of the sterilization process in most cases. Several major pro­ cessing companies use specific and unique techniques to disinfect allograft tissue. Allowash formula (LifeNet, Vir­ ginia Beach, Va) combines irradiation, ultrasonics, cen­ trifugation, and negative pressure with reagents, including biologic detergents, antibiotics, alcohols, and hydrogen peroxide to increase solubilization and the removal of lip­ ids, bone marrow, and blood elements.34 The BioCleanse technique (Regeneration Technologies, Gainesville, Fla) uses a low-temperature chemical sterilization process with liquid sterilants that perfuse the inner matrix of the tissue. This is followed by irradiation.34 More than 300,000 grafts have been implanted without a known infection. The pro­ cess has been validated by the FDA to kill implanted spores and viruses.38 The Clearant Process (Clearant, Los Ange­ les, Calif) freezes the tissue, extracts the water, and adds dimethlysulfoxide as a radioprotectant. The tissue is then treated with 50 kGy of radiation, which is 2 to 4 times the dose recommended to avoid damaging cells.34 The search for a perfect mechanism of sterilization is ongoing, with further refinement techniques expected in the future.

140

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

STORAGE The ideal method of allograft preservation would main­ tain the viability of the cells, would not alter the collagen matrix, and would enable indefinite storage in a conve­ nient, cost-effective manner. Reducing the antigenicity of the graft tissue and secondary sterilization would be desir­ able as well. An ideal method does not exist, similar to our current situation with sterilization. Deep freezing, freeze drying, cryopreservation, and fresh (osteochondral) are the current methods employed to preserve musculoskeletal allograft tissue. Deep freezing is the most commonly used method of preservation for allografts used in ligament reconstruc­ tion.39 Grafts preserved by deep freezing, referred to as fresh frozen grafts, undergo the freezing process twice: once during quarantine and then finally for storage on the shelf before use. Marrow elements and blood must be removed from the graft before the initial freeze. After the tissue is removed from quarantine, is is thawed to room temperature and soaked in antibiotic solution. Secondary sterilization with gamma irradiation may or may not be used, depending on the preferences of the individual tissue bank. Finally, the tissue is packaged without any solution and frozen rapidly to −80° C. The graft can be stored at this temperature for 3 to 5 years.40 One benefit of deep freezing is the reduced antigenic­ ity of the graft, which is accomplished by the destruction of cellular antigens. This process reduces the potential for the host to mount an immune response once the graft is implanted. Unfortunately, however, the viability of the cells in the graft is not maintained with deep freezing. Ice crystal formation kills the cells and is also believed to be detrimental to the collagen matrix. This problem calls into question the use of deep freezing as a potential method of preserving articular or meniscal cartilage allograft tissue. In these tissues, cell viability may be critical to long-term success. It is believed that multiple freezings are increas­ ingly detrimental to the graft tissue collagen matrix41; as more ice crystals are formed in the graft collagen matrix, more damage is done. Cryopreservation methods use chemicals to remove cel­ lular water and controlled-rate freezing to prevent crys­ tal formation. After procurement, tissues designated for cryopreservation are cooled to wet ice temperatures for up to 48 hours. During this short period of quarantine, the graft is cleared for final processing. The tissues are then soaked in antibiotic solution at ambient temperatures for 24 hours. The tissue is placed into a storage container filled with a cryoprotectant solution of dimethyl sulfoxide or glycerol. The cryoprotectant displaces cellular water. The controlled-rate freezing then ensues. The controlled-rate freezing is completed at a tempera­ ture of about −135° C. The final storage uses liquid nitro­ gen at a temperature of about −196° C. The grafts can be stored at this temperature for up to 10 years. The higher temperatures used in deep-freezing storage (−80° C) allow ice crystals to continue to grow and re-form. The liquid nitrogen storage temperature (−196° C) used for cryo­ preservation does not allow this type of crystal turnover. Cryopreservation is one method of storage for meniscal allografts because cellular viability is largely maintained.

Many surgeons prefer ligament reconstruction allografts preserved by this method as well. Jackson and associates, however, have shown that preserved cells in allografts used for ligament reconstruction do not survive long after implantation.42 Using DNA-probe analysis, they found that the donor cell DNA is no longer present in allograft ligament reconstructions after 4 weeks in a goat model. This absence of the donor tissue DNA occurred without any detectable sign of immune rejection. Their study sug­ gests that neither fresh nor cryopreserved allograft liga­ ments maintain cellular viability after implantation and rely totally on host cell repopulation for graft incorpo­ ration and remodeling. The most significant benefit of cryopreservation clearly is the elimination of ice crystal formation, which can damage and weaken the collagen in the graft. Freeze drying is also used to preserve allograft tissue used for ligament reconstruction and is distinctly different from cryopreservation or deep freezing. Freeze drying has the advantage of allowing ambient storage temperatures after processing is completed. The graft can be rehydrated for use up to 5 years later. Freeze drying is the most costefficient means of storage. Procured tissues designated for freeze drying are cleaned to remove marrow elements and blood and are frozen during a period of quarantine. After the tissues are cleared and removed from quarantine, they are thawed and soaked in antibiotic solution for 1 hour at ambient tem­ peratures. Secondary sterilization with ethylene oxide or gamma irradiation may be done at the discretion of the tis­ sue bank. Freeze drying, or lyophilization, is then carried out. In this process, alcohol replaces water within the tissue to a residual moisture level of 5% or less. A vacuum pro­ cess then removes the alcohol from the tissue. This process ultimately kills all the nucleated cells in the graft. The tissue is very dry and stiff after lyophilization. It is important not to manipulate the graft because of its vul­ nerability to fracture. The tissue is prepared for surgery by soaking it in a physiologic solution for at least 30 minutes before any type of handling. It may take up to 24 hours for ligament-sized tissue to rehydrate totally. A number of studies have evaluated the biomechanical and clinical properties of freeze-dried grafts used for liga­ ment reconstruction.22,23,25,43-46 Most studies show that the freeze-drying process alone has minimal effect on the bio­ mechanical properties of allograft bone, tendon, or liga­ ment as long as the specimens are not repeatedly frozen and thawed. More pronounced differences in freeze-dried grafts compared with other grafts have been demonstrated in clinical studies after implantation. For example, Indeli­ cato and associates showed a significantly higher incidence of both subjective instability and a pivot shift in ACL reconstructions done with freeze-dried grafts compared with fresh frozen grafts.43 One benefit of freeze drying may be that it potentially helps neutralize HIV. In 1985, three different recipients converted to HIV positive after receiving a fresh frozen allograft from the same infected donor.1,4,7 Recipients of freeze-dried grafts from the same donor did not become infected with HIV. Although anecdotal, there is a sugges­ tion that freeze drying assists in neutralizing HIV. This potential benefit cannot be totally relied on, however,

Surgical Principles

because HIV has been cultured successfully from freezedried bone. Crawford and associates recently performed a study on tendon and cortical bone from cats infected with retrovirus. The retrovirus was not inactivated by the lyophilization process, and the authors concluded that freeze drying should not be relied on to inactivate retrovi­ rus in musculoskeletal allografts.47 In the case of osteochondral allografts—grafts that involve articular cartilage and its subchondral bone—the viability of the chondrocytes may be critical. Cellular viability is best maintained when transplanted in a fresh fashion without long-term storage.13,18,29,48-52 Freezing of osteochondral allografts has been shown to decrease the viability of articular cartilage chondrocytes severely. Some tissue banks establish a 45-day shelf life for osteochondral allografts. There is a decrease in the percentage of viable chondrocytes after 24 hours. Fresh grafts have a higher risk for disease transmission because of the lack of steril­ ization that occurs with secondary sterilization and storage ­processing.

RISK FOR INFECTION The predominant concern regarding the ultimate safety of allograft use is the risk for infection. To date, there is only one reported case of HIV transmission using a musculo­ skeletal allograft,1,4,7 and it occurred in 1985. The donor tissue, a patellar tendon graft, came from an AATB-certified tissue bank. Donor screening was carried out by ­ family interview, medical history review, and enzyme-linked immunosorbent assay for HIV. All the screening test results were negative, and the tissue was accepted. The graft was not processed with removal of blood or bone marrow. The graft that caused the transmission of HIV was not second­ arily sterilized and was fresh frozen. Other tissues from the same donor (fascia lata, Achilles tendon, and patellar ten­ don) were processed and freeze dried and did not cause HIV infection after implantation into three other recipi­ ents. NAT testing for HIV is now an AATB require­ ment. HIV is an RNA virus; however, it infects the DNA of the white blood cell. Because white blood cell DNA is stable in cadaver blood for up to 48 hours, PCR testing for HIV is effective in detecting the infected white blood cell DNA and consequently highly effective in allograft ­screening.38 In 1995, Conrad and associates reported two separate cases of hepatitis C transmission through musculoskeletal allografts.6 Both patients were recipients of a fresh frozen allograft patellar tendon graft. Neither graft was processed with blood and marrow element removal, nor were the grafts secondarily sterilized. Twelve other musculoskel­ etal allografts from the same donor that were processed and irradiated did not transmit hepatitis C to the eventual recipients. Recently, 38 patients received allograft tissue from one infected donor, and at least 6 patients tested posi­ tive for HCV. One of the patients had received a patellar tendon allograft. The infection in the donor was unde­ tected by the tissue bank. Ultimately, the donor was found to be anti–HCV antibody negative, HCV RNA positive.53 A patient underwent anterior and posterior spinal fusion in March 1998 with cancellous allograft bone. Six weeks after surgery, he developed acute hepatitis, and in May 1998,

141

tests for hepatitis C antibodies were negative. In August 1998, he tested positive for hepatitis C RNA, and in November 1998, he tested positive for hepatitis C antibod­ ies. He had no known risk factors and it was believed that he acquired the virus through the bone graft.54 This illus­ trates that there is a clear window for negative serologies despite the presence of disease. The AATB now mandates NAT testing for HCV. It is an RNA virus but, in contrast to HIV, does not attach to white blood cell DNA. It is continuously cleared from the serum by immune system enzymes, and the half-life of HCV in cadaver blood can be a few hours unless the blood sample is obtained early after death and cooled. PCR testing for HCV is less effective in tissue banking than in blood banks and can produce a 15% to 25% false-negative rate in tissue banks.38 Further­ more, hepatitis C has been found in about 1.1% of donors, in comparison to HIV, which is present in only 0.03% of donors.38 Risk for allograft-associated bacterial infection is clearly a concern as well. As of March 2002, the CDC identified 26 cases of musculoskeletal allograft infection. Investiga­ tion ensued after the postoperative death of a 23-year-old man who received an osteochondral femoral graft. He developed pain at the surgical site on postoperative day 3, which rapidly progressed to shock; the patient died on postoperative day 4. Premortem blood cultures grew Clostridium sordellii. Another patient received a fresh femoral condyle graft and a frozen meniscal allograft from the same donor. He also developed septic arthritis; however, cul­ tures for anaerobic bacteria were not obtained. The body of the donor was refrigerated 19 hours after death, and the tissue was procured 23.5 hours after death. The CDC sub­ sequently cultured C. sordellii from nonimplanted donor tissue. The tissue bank had cultured the aseptically har­ vested tissue only after soaks in an antibiotic and antifungal solution. Thirteen of the 26 patients with a musculoskel­ etal allograft–associated bacterial infection were infected with Clostridium. Eleven of these patients, including the 2 previously described patients, received tissues processed by the same tissue bank. Eight of the allografts were used for ACL reconstruction. Eleven were frozen, and 2 were fresh. All were processed aseptically, but none was termi­ nally sterilized. Of the 13 other patients, 11 patients were infected with gram-negative bacilli; 5 were polymicrobial and 2 were negative by culture. Eight of the 13 patients had evidence implicating the allograft. Eight patients received grafts that were not terminally sterilized, and 3 patients received grafts that had reportedly undergone gamma irradiation.37 It is clear after this investigation that sporeforming bacteria are potential pathogens. The CDC made several recommendations based on this investigation and ultimately stated that “Unless a sporicidal method is used, aseptically processed tissue should not be considered ster­ ile. Health-care providers should be informed of the pos­ sible risk for bacterial infection.”37 The general risk for postoperative wound infection with musculoskeletal allografts is negligible with small grafts. Large musculoskeletal allografts have an infection rate sim­ ilar to that of other sterilized prosthetic implants. Tomford reported no wound infections in 287 patients who received a small bone or soft tissue allograft.8 A large intercalary bone graft used for joint revision or tumor reconstruction

142

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

was involved in all 13 wound infections found in this series. All the grafts in this series had negative culture results before implantation, and the cause of infection was not thought to be a result of graft contamination. The infec­ tion rate in the large revisions and reconstructions with allograft was found to be comparable to similar cases using metallic prosthetic implants. None of the pathogens could be traced to the donor or the tissue bank. Clearly, musculoskeletal allograft–associated viral and bacterial infection is still a concern. This underscores the importance of choosing a tissue bank that is accredited by the AATB. Surgeons should be familiar with the processes of the tissue bank that they choose in order to ensure the safe use of allografts. The surgeon has the ultimate choice between autograft and allograft tissue. This fortunately is not life or death surgery. Therefore, all risks must be con­ sidered before allograft use.

ALLOGRAFTS FOR LIGAMENT RECONSTRUCTION There is no absolute indication for the use of allograft tissue in knee ligament reconstruction other than a lack of autogenous tissue. In multiple-ligament knee recon­ structions, allograft is often the tissue of choice because of the need for more than one ligament graft.55-60 This choice reduces donor morbidity and operative time sig­ nificantly. Otherwise, the choice to use allograft tissue instead of autograft is at the discretion of the surgeon and the patient. Most of the controversy regarding the use of allografts concerns the use of allograft tissue for primary reconstruc­ tion of the ACL. This debate largely arises because of the frequency of this procedure as well as the familiarity that most orthopaedic surgeons have with it. Autograft ACL reconstruction is a procedure with an enormous amount of outcome data with good and excellent results. Surgeons who prefer autogenous tissue also cite the risk for infection, the potential for an immune response, and a potentially higher failure rate as other reasons not to use allograft for primary ACL reconstruction. Surgeons who prefer allograft tissue for ACL reconstruction cite no donor morbidity, less post­ operative pain, smaller incisions, less operative time, and comparable results in terms of knee stability as reasons for their graft choice.61 The difficulty in reaching a consensus is largely due to a lack of long-term data on the outcome of primary ACL reconstruction with allograft compared with auto­ graft using similar surgical and rehabilitation techniques in a uniform patient population. Surgeons, therefore, rely on animal studies and the available clinical outcome data, which are difficult to interpret because of differences in patient populations. Allograft studies cannot be “lumped’’ together because of differences in surgical and rehabilita­ tion techniques, tissue processing, patient populations, and outcome measurement tools. Also, multiple, different types of allograft tissues are used for ligament reconstruc­ tion, including patellar tendon, Achilles tendon, and all soft tissue allografts. It is not recommended that different tissue types be lumped together as allografts in terms of outcome after ligament reconstruction. One must be very

critical when analyzing the literature on allograft outcome studies because not all “allografts” are the same. Animal studies have been done to evaluate allograft and autograft ligament histologic incorporation and bio­ mechanical properties at various stages of healing.10,28,44,45 The process of graft revascularization and incorporation for allograft tissue has been found to be similar to that for autograft tissue. This was demonstrated in the classic study by Cordrey and associates using a rabbit model.10 They found that the revascularization and the collagen turnover in the allograft ligaments were the same as in autografts, except that the allografts took longer to go through the process. Infrapatellar tendon allograft ACL reconstruc­ tions in dogs have been shown to be grossly and histologi­ cally similar to the native ACL 1 year after implantation in two separate studies.62,63 Biomechanical differences between allograft and auto­ graft ACL reconstructions have been demonstrated using a goat model.42 In this study, Jackson and coworkers found that the allograft ACL reconstructions had a maximal loadto-failure ratio that was 27% of normal, compared with 62% of normal for autograft reconstructions. This find­ ing is in contrast to the study by Nikolaou, who demon­ strated 90% of normal strength in cryopreserved allograft ACL reconstructions after sacrifice at 36 weeks in a dog model.64 It is likely that allograft ligament reconstructions are weaker than autografts during graft incorporation, but this has yet to be proved clinically significant in terms of increased graft failures.12 Despite a slower rate of graft incorporation, the clini­ cal results of allograft ACL reconstruction have been promising. Shelton and associates compared bone–patellar tendon–bone allograft and bone–patellar tendon–bone autograft ACL reconstructions and found no difference in pain, effusion, stability, range of motion, patellofemo­ ral crepitus, or thigh circumference after 24 months.65 Harner and colleagues found no statistical difference in the 3- to 5-year outcome for ACL reconstructions using fresh frozen nonirradiated allograft tissue compared with autograft ACL reconstruction, with the exception that the autograft patients had a higher incidence of terminal extension loss.61 Interestingly, the allograft group in Harn­ er’s study had better knee scores than the autograft group in two different ratings systems, but this was not statisti­ cally significant. In a later publication, Harner stated that his indications for allograft ACL reconstruction include increased age, low activity level, and patient preference.32 Also, Noyes and coworkers reported 89% good to excellent results in ACL reconstructions done with bone–patellar tendon–bone and fascia lata allografts after 2 years and noted that bone–patellar tendon–bone allografts had bet­ ter arthrometric results.66 Revision ACL reconstruction is a situation in which allograft tissue may be particularly useful. Often, in a failed ACL reconstruction, previous autogenous tissue has been used, limiting options for graft selection during revi­ sion surgery. Revision ACL reconstruction often requires increased tunnel size (because of tunnel lysis) and the need for larger bone plugs or tissue.32 Allograft tissue may pro­ vide for larger soft tissue grafts (increased tensile strength),32 which could be optimal in the setting of a salvage operation with gross instability or combined ­ instability in multiple

Surgical Principles

directions. Tissue choices for revision ACL reconstruction include patellar tendon, Achilles tendon, and all soft tis­ sue grafts, including anteroposterior tibialis tendons and iliotibial band. Fewer published outcome studies address the use of allograft tissue for posterior cruciate or collateral ligament reconstruction. Allograft tissue is often the graft of choice for these complex cases. Noyes and Barber-Weston, as well as Bullis and Paulos, have provided studies on the use of allografts for these types of reconstructions, with good results.58,59,67,68 In addition, the data available on allograft for ACL reconstruction appear to support its use for poste­ rior cruciate ligament or collateral ligament reconstruction or augmentation after primary repair. Consideration should be strongly given to allograft use in reconstruction of the multiligamentous knee. Not only are multiple ligaments injured with knee dislocations but the injury also often includes damage to menisci, articu­ lar cartilage, and neurovascular structures. Advantages of allografts include a sufficient amount of tissue to recon­ struct all injured ligaments, decreased surgical time, imme­ diate motion, and decreased donor site morbidity in an already traumatized knee. Minimizing incisions, intraop­ erative tourniquet time, postoperative pain, and postopera­ tive knee stiffness are additional benefits.32

MENISCAL ALLOGRAFTS The role of the meniscus in the preservation of articular cartilage has been increasingly appreciated since Fairbank’s 1948 article on postmeniscectomy degenerative changes to the knee.69 Lee and colleagues70 evaluated changes in tib­ iofemoral contact areas and stresses after serial meniscec­ tomies of the posterior horn of the medial meniscus. They concluded that the peripheral portions of the medial menis­ cus provide greater contribution to increasing contact areas and decreasing mean contact stresses than does the central portion and that the increase in peak contact stresses are proportional to the amount of the removed meniscus. Fur­ thermore, they showed that in terms of load bearing, loss of hoop tension (i.e., segmental meniscectomy) is equal to total menisectomy.70 Most surgeons make a concerted effort to minimize the removal of meniscal tissue during partial meniscectomy and to repair the meniscus whenever possible, particularly in the knee with an ACL injury. An injury to the meniscus resulting in near-total meniscec­ tomy was without a good solution until meniscal allograft reconstruction was pioneered. Milachowski was the first to attempt meniscal allograft reconstruction in 1984.21,71 A few years later, ­ Arnoczky used a dog model to demonstrate peripheral healing, ­cellular repopulation, and the lack of immune response after meniscal allograft reconstruction and therefore proved the feasibility of the procedure.72 Since then, knee sur­ geons have been defining the indications and refining the procedure.48,73-75 Selecting the appropriate patient is critical for a success­ ful outcome after allograft meniscal transplantation. The ideal candidate is physiologically young, has had a previous meniscectomy, has developed pain over the involved tib­ iofemoral compartment, and has minimal articular carti­ lage changes without significant bipolar disease.32,36 Limb

143

alignment and ligament stability are critical and must be addressed before meniscal transplantation or at the index procedure. Meniscal transplantation can also be combined with osteochondral allograft when necessary.32,76 Patient compliance and realistic expectations are important in obtaining a good outcome. Contraindications include obe­ sity, rheumatoid arthritis, metabolic diseases, gout, and infection.36 The success of meniscal allograft reconstruction has been most predictable in patients with ligamentous sta­ bility, physiologically normal alignment, and minimal articular cartilage damage. It is not uncommon to find significant irreparable damage to a meniscus with little to no arthritic change after an acute ACL tear. Meniscal allograft transplantation may be planned in conjunction with ACL reconstruction. Implantation of the meniscal allograft is technically much easier to perform before the ACL reconstruction is completed and should precede the ACL reconstruction when combining the procedures. Reconstructing the meniscus and the ACL may have a synergistic effect on the stability of the knee and the survival of both grafts. Sekiya and associates77 recently performed a retrospective review of 28 patients who underwent combined ACL reconstruction and menis­ cal allograft transplantation. Nearly 90% had normal Lachman and pivot shift scores at an average 2.8 years after surgery. Joint space narrowing was not significantly different from that of the contralateral knee. They con­ cluded that restoration of meniscal function could pro­ tect the cartilage and improve joint stability.77 Yoldas and coworkers78 evaluated meniscal allograft transplanta­ tion with and without combined ACL reconstruction in 31 patients. No difference was found between which meniscus was transplanted, concurrent ACL reconstruc­ tion, or degree of arthrosis found at surgery. Twentytwo patients were greatly improved, 8 were somewhat improved, and 1 was unchanged.78 Sizing of the meniscal allograft is critical to outcome. Plain radiographs of the patient’s knee are sent to the tissue bank to compare to donor radiographs. Any magnification in the radiographs must be accounted for. Without proper sizing, the meniscus cannot share and distribute the load across the articular cartilage. Central hypocellularity and shrinkage of meniscal allografts have been a problem in 15% to 30% of the cases reported.12,18,29,75,79 It is believed that this occurs because the more central chondrocytes in the meniscus do not receive enough nutrition. The more peripheral cells in the meniscal allograft are more likely to survive. Graft selection is important for meniscal allograft suc­ cess. Most meniscal allografts are fresh frozen or cryo­ preserved. In 1993, Jackson and colleagues80 used a goat model to show that host cell repopulation of the graft takes place within several weeks after transplantation. Given that host cells repopulate the graft, fresh frozen grafts are appropriate.32,36,80 Fresh frozen grafts are cheaper than cryopreserved grafts, and freezing the graft kills the cells and diminishes the potential for an immune response. The advantages of cryopreserved tissue are maintenance of cellular viability and the possibility of prolonged storage. This allows ample time for serologic testing of the graft and appropriate sizing.32,36

144

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The technique of transplantation has been refined sig­ nificantly. Early meniscal allograft transplantation was done through a moderate-sized arthrotomy that included releasing the collateral ligament. The modern techniques are done arthroscopically, assisted by small arthrotomies without disruption of the collateral ligaments. The menis­ cus is attached to bone plugs (medial) or to a bone bridge (lateral) fashioned to key into drill holes (medial) or a slot (lateral). The periphery of the meniscus is fixed with cur­ rent meniscal repair suture techniques. Multiple studies have been published regarding the clin­ ical results of meniscal allograft transplantation.16,18,81-89 There is significant variability in mean length of follow-up, assessment techniques, and results. Cook90 concisely sum­ marized conclusions from meniscal allograft transplanta­ tion. In general, patient satisfaction exceeds 75%, and about 90% of patients would have the procedure again. Fresh frozen and cryopreserved allografts perform equally well. Functional outcomes are better for lateral meniscal allografts than for medial meniscal allografts. All treatable pathologies of the knee should be treated before or simul­ taneous with the meniscal allograft transplantation. Finally, rehabilitation is important for a successful outcome.90 Meniscal allograft surgery continues to evolve. The untoward effects of a meniscal-deficient knee are well doc­ umented. Advances in tissue processing increase the safety of allograft surgery. There is currently no simple solution for symptomatic patients with a previous meniscectomy. At this time, meniscal allograft transplantation is the most viable option for young patients with localized symptoms who have had a prior meniscectomy.

OSTEOCHONDRAL ALLOGRAFTS Treatment of articular cartilage injury and defects remains a challenging problem for the orthopaedist, especially in physiologically younger and active patients. Osteochon­ dral injury can be disabling with pain, mechanical symp­ toms, and swelling. These lesions can generate irregular surfaces and predispose the joint to further articular carti­ lage damage and meniscal pathology.32,36 If symptomatic, some surgical treatment options include microfracture, abrasion arthroplasty, and drilling. These procedures result in healing of the defect with fibrocartilage. More aggressive surgical techniques include autologous chon­ drocyte implantation, osteochondral autograft transplan­ tation, or osteochondral allograft. Factors determining the appropriate procedure include the size, depth, location of the defect, activity level, physiologic age of the patient, and associated knee pathology.36 Osteochondral allografts are indicated for patients with full-thickness articular cartilage defects larger than 2 cm.32 This deficit is usually the result of trauma, osteochondri­ tis dissecans, or avascular necrosis.32 Contraindications include rheumatoid arthritis, generalized arthritis, and corticosteroid-induced osteonecrosis.36 As with meniscal transplantation, associated meniscal pathology, ligamen­ tous instability, or limb malalignment must be addressed.32 Advantages of osteoarticular allograft over autograft include the ability to harvest a much larger graft and the ability to obtain tissue from a younger donor with healthier cartilage.32,36 The graft can be sized with radiographs with

a disparity in size of less than 10%.32 The goal of trans­ plantation is to achieve an anatomic articular surface with minimal step-off at the host-donor cartilage interface.32 Osteochondral allografts are typically implanted fresh or cryopreserved because the viability of the chondrocytes is critical to success. Fresh typically means that the graft was harvested within 24 hours of the donor’s death and implanted within 7 days.91 Fresh grafts preserve chondro­ cyte viability at the expense of increased risk for disease transmission, increased immunogenicity, and more diffi­ culty in sizing.36 Recent studies have been performed to evaluate chondrocyte viability after storage of the grafts. Williams and associates92 looked at hypothermically (4° C) stored sheep knee condyles at various time inter­ vals between 1 and 60 days. Mean chondrocyte viability decreased over the storage interval, with a preponderance of nonviable chondrocytes in the superficial layer. Decreases in matrix proteoglycan and matrix dynamic modulus were seen as well. Allen and colleagues93 evaluated chondrocyte viability and extracellular matrix quality in unused cartilage from 16 consecutive allografts after tissue bank processing and storage. They found that after a mean storage time of 20 days, there was a decrease in cell viability, especially in the superficial zone, as well as a decrease in cell density and metabolic activity. Matrix and biomechanical properties were preserved. Pearsall and coworkers91 assessed 16 refrig­ erated (2° to 8° C) osteochondral allografts that had been refrigerated for an average of 30 days from donor expira­ tion to implantation. The grafts were evaluated at the time of implantation. No significant correlation was noted with chondrocyte viability. The authors concluded that refrig­ erated osteochondral allografts can be maintained for up to 44 days, with an average chondrocyte viability of 67%. This last study is encouraging because the goal is to implant a graft with maximal chondrocyte viability while placing the patient at the least possible risk.36 Alternatively, cryopre­ served grafts maintain up to 80% viability of cells, have greater storage time, and are less immunogenic.36 How­ ever, cryopreserved grafts reveal early articular degenera­ tion and decreased chondrocyte viability compared with fresh grafts32,94 and yield inferior results.36 Zukor and coworkers have reported on more than 100 cases and showed that the best results were in patients with focal traumatic unipolar lesions.95 Osteochondritis disse­ cans and bipolar lesions that were treated had less favor­ able results. Of the 92 knees treated in this series, a 75% success rate was found at 5 years, and the rate declined to 63% at 14 years. An 85% success rate in the treatment of small (2 to 4 cm2) lesions of osteochondritis dissecans was reported by Garret.96 Success was determined by the appearance of normal articular cartilage on second-look arthroscopy. The subchondral bone did not incorporate, and the articu­ lar cartilage fragmented in those that failed. This finding is further proof that the incorporation of the subchondral bone is critical to the outcome of these grafts. Several studies have recently evaluated osteochondral allografts. Gross and colleagues97 performed a prospec­ tive, nonrandomized study of fresh allografts for unipolar ­femoral condyle defects (60 patients with average 10-year follow-up) and fresh allografts for reconstruction of the tib­ ial plateau (65 patients with average 11.8-year ­follow-up).

Surgical Principles

Kaplan-Meier survivorship showed 95% femoral graft ­survival at 5 years and 85% at 10 years. Twenty-one of 65 patients with a tibial plateau allograft were converted to total knee arthroplasty at an average of 9.7 years. KaplanMeier survivorship revealed 95% survival at 5 years, 80% at 10 years, and 65% at 15 years. Spak and Teitge98 looked at 14 fresh osteochondral allografts of the patella or patel­ lofemoral joint for patellofemoral arthritis in 11 patients younger than 55 years with a mean follow-up of 10 years. Knee scores and functional scores improved. At last follow-up, 8 of 14 grafts were in place, and 3 nonsurviv­ ing grafts had lasted more than 10 years. Emmerson and associates99 looked at 66 knees in 64 patients who under­ went osteochondral allograft transplantation for osteo­ chondritis dissecans of the femoral condyle. The average patient age was 28.6 years (15 to 54 years), and the average follow-up was 7.7 years (range, 2 to 22 years). Seventy-two percent of patients had good or excellent results, and only 1 patient had a fair result. Ten patients underwent reop­ eration. Finally, McCulloch and coworkers100 performed a prospective evaluation of prolonged fresh osteochon­ dral allografts of the femoral condyle in 25 patients with a mean follow-up of 35 months. The grafts were stored at 4° C and implanted at a mean of 24 days after procure­ ment. Statistically significant improvements were seen for the Lysolm and International Knee Documentation Com­ mittee scores. Patients reported 84% satisfaction and felt their knee functioned at 79% of their contralateral knee. This study validates the basic science study performed by Pearsall and colleagues.91 It is important to consider these promising results in light of the fact that there is no per­ fect solution to articular cartilage defects in young active patients and that these are very difficult problems to treat.

SUMMARY There is a great deal of science in the processing of muscu­ loskeletal allograft tissues and in their use in patients. More groups are becoming interested and involved in the moni­ toring and assurance of the safe use of allografts. Allograft safety has improved significantly during the past 15 years. It is critical that the orthopaedic surgeon use grafts only from AATB-accredited tissue banks. The surgeon must be familiar with the processes of the bank supplying their allograft. Not all tissue banks use the same donor screen­ ing, tissue harvesting, tissue processing, safety purification methods, and secondary sterilization techniques. Implan­ tation of allograft tissue can cause infection with ­significant morbidity and mortality. It is critically important for the surgeon to weigh the risk and benefits of allograft tissue compared with autograft tissue and discuss this with the patient. Ultimately, it is you as the surgeon who is the “tissue banker,” at least in the patient’s view. Encouraging results from recent studies indicate that musculoskeletal allografts will be used more in the future. There is currently a clear role for the safe use of musculoskeletal allografts in ligament reconstruction, meniscal ­transplantation, and chondral defects.

C

r i t i c a l

P

145

o i n t s

l

 rthopaedic surgeons should only use allograft tissues O obtained from an AATB-accredited tissue bank. The use of allograft tissue is only as safe as your tissue source or tissue bank. l It is important to be familiar with the processes of the tis­ sue bank from which one obtains the tissue grafts. Not all tissue banks are the same. l It is important to obtain grafts that were appropriately stored to preserve the critical properties necessary for successful transplantation, that is, fresh or cryopreserved storage for osteochondral allografts. l Disease transmission can be a problem with musculoskel­ etal allografts, and unless a sporicidal method of second­ ary sterilization is used, a graft should not be considered sterile. Secondary sterilization does not guarantee sterility or safety. l Allografts may be used for any ligament reconstruction, but revision ACL surgery and multiligamentous knee injuries are situations in which allograft tissue may be particularly indicated. l Patient selection is critical to the outcomes of menis­ cal allograft transplantation and osteochondral allograft transplantation. Associated knee pathology should be treated before these procedures or at the index surgery. l Outcomes of osteochondral allograft surgery are encour­ aging given this difficult problem in a relatively young patient population.

S U G G E S T E D

R E A D I N G S

American Association of Tissue Banks: Standards for Tissue Banking. MacLean, Va, American Association of Tissue Banks, 2006. Caldwell PE 3rd, Shelton WR: Indications for allografts [review]. Orthop Clin North Am 36(4):459-467, 2005. Centers for Disease Control and Prevention: Update: Allograft-associated bacterial infections—United States, 2002. MMWR Morb Mortal Wkly Rep 51:207-210, 2002. Cook JL: The current status of treatment for large meniscal defects [review]. Clin Orthop 435:88-95, 2005. Gross AE, Shasha N, Aubin P: Long-term follow-up of the use of fresh osteochon­ dral allografts for posttraumatic knee defects. Clin Orthop 435:79-87, 2005. Lee SJ, Aadalen KJ, Malaviya P, et al: Tibiofemoral contact mechanics after se­ rial medial meniscectomies in the human cadaveric knee. Am J Sports Med 34(8):1334-1344, 2006. Rihn JA, Harner CD: The use of musculoskeletal allograft tissue in knee surgery. Arthroscopy 19(Suppl 1):51-66, 2003. Shelton WR: Arthroscopic allograft surgery of the knee and shoulder: Indications, techniques, and risks. Arthroscopy 19(10 Suppl 1):67-69, 2003. Spak RT, Teitge RA: Fresh osteochondral allografts for patellofemoral arthritis: Long-term followup. Clin Orthop 444:193-200, 2006. Vangsness CT, Wagner PP, Moore TM, Roberts MR: Overview of safety issues concerning the preparation and processing of soft-tissue allografts. Arthroscopy 22(12):1351-1358, 2006.

R eferences Please see www.expertconsult.com

C H A P T E R  

 3

Nonorthopaedic Conditions S ect i o n

A

Infectious Disease and Sports Rob Johnson and William Knopp

Infectious disease in athletes poses unique circumstances and challenges. The circumstances of an ill-timed infection can prevent training or competition with little advanced warning. The challenges are to promptly recognize the problem, initiate appropriate treatment, return the athlete to practice and play safely, and prevent spread amongst teammates, coaches, and contacts (family, friends, and spectators). The average adult experiences one to six upper respiratory tract infections (URTIs) per year.1 This translates into about 100 million URTIs and 250 million days of activity restriction.2 And this only represents the risk for the common rhinovirus! Added to common infections, the infectious exposures posed by the athletes’ training regimen, travel, and daily communal training and competition environments virtually guarantee that infectious disease will affect the athlete.3 Pathogens can single out the recreational athlete, the regular exerciser, the citizen competitor, or the team-sport athlete at every level of competition. Consequently, the sports and team physician must be knowledgeable of the more common threats and possess the skills to diagnose and treat them.

EFFECTS OF TRAINING AND COMPETITION ON THE IMMUNE SYSTEM The effects of exercise and training are addressed in two different ways: the effects of a single bout of exercise and the long-term effects and observations. A brief review of the components of the immune system provides a context for interpreting the data regarding the acute and chronic effects of exercise on the immune system and illness. The innate immune system is the body’s broad, nonspecific, first line of defense consisting of the skin, mucous membranes, upper and lower respiratory systems, specific cell types such as phagocytes and natural killer (NK) cells, humoral factors such as cytokines, and various complement factors.4 Within the mucous membrane is secretory

i­mmunoglobulin A (IgA) that identifies and alters viral ­particles to enhance control and removal of the virus. The release of cytokines and complement is the initial response of the immune system to an infectious stimulus. This response activates and controls the T cells and B cells that represent a specific response of the acquired immune system to the infectious agent. T- and B-cell activity is specific to the infectious agent, viral or bacterial. Both T cells and B cells have memory for the specific infectious agent maintaining antibodies against the agent. The duration of memory and antibody activity is highly variable. An initial response of the immune system to an acute, intense bout of exercise is an increase in leukocytes, predominantly composed of neutrophils, with a mild increase in lymphocytes and monocytes.5,6 This acute response is triggered by catecholamines and cortisol. NK cells increase 150% to 300% immediately after an exercise bout of less than 1 hour, but then fall to less than baseline within 30 minutes after exercise.7 An exercise session lasting more than 1 hour fails to incite a similar NK-cell response. Macrophages also increase with an acute exercise bout. This response is blunted in athletes who train regularly, but levels are still higher than those in untrained individuals.7 Secretory IgA concentrations are impaired by long, intense exercise sessions.8,9 Based on this information, there is no correlation of these depressed levels with a higher rate of respiratory infection. Another study of secretory IgA levels over the course of a football season, however, demonstrated both significantly diminished levels of IgA and a diminished secretory rate of IgA.10 The outcome measured was the incidence of URTIs. The lower IgA levels and secretion were associated with more URTIs. The level of IgA was a predictive factor in the risk for developing URTI. Cytokines such as tumor necrosis factor-α (TNF-α), interleukin-1 (IL-1), and IL-6, another factor within the innate immune system, increase with a single bout of intense exercise.11 Some factors show dramatic increases. A summary of the effects of a single bout of exercise is shown in Table 3A-1. 147

148

DeLee & Drez’s Orthopaedic Sports Medicine

  TABLE 3A-1  Immune System Function Changes

High

with Exercise

Increased after Acute Exercise Bout

Increased after Training Period

Neutrophil concentration Monocytes (chemotaxis, adhesion) Dendritic cell concentration (T-cell inducer) Th2 dominance (bacterial protection)

NK-cell function Monocyte function Vaccination response

Decreased after Acute Exercise Bout

Th1 dominance (viral ­protection) Decreased after Training Period

NK-cell function Delayed-type ­hypersensitivity Lymphocyte function Mucous immunoglobulin A production Monocytes (major histocompatibility ­complex enzymes) (Adapted from Malm C: Exercise immunology: The current state of man and mouse. Sports Med 34[9]:555-566, 2004.)

The response of the immune system to a single bout of exercise demonstrates acute increases during or immediately after exercise, followed by levels decreasing to or below baseline, giving support to the open-window ­theory.7 This hypothesis suggests that the athlete is at increased risk for infection if exposed to pathogens ­during the “window” of decreased immune surveillance. The concern for the athlete is to observe precautions in exposure and hygienic practices to reduce risk during this ­vulnerable period. Animal studies have looked at this very issue. Inoculation of rats after an exhaustive bout of exercise showed that the rats were protected against infection compared with another group that was already infected and performed a similar bout of exercise. The infected group had progression to more severe symptoms.12,13 Another study comparing trained mice with untrained mice showed less damage to the myocardium after infection with either bacterial or viral pathogens in the trained mice.14 The trained mice, however, also showed more severe symptoms and outcomes when exercising after they were infected. The effects of long-term training on the immune system are largely reflected in the epidemiologic studies of training and infection. Generally, individuals who train aerobically on a regular basis perceive themselves as having fewer infections than untrained people.15 Several randomized studies by Nieman support this perception.7 These studies demonstrated that near-daily physical activity reduces the number of days of reported illness. Other research shows a 23% decrease in URTIs in people who train on a regular basis, and another found that URTI development is inversely related to the level of moderate physical activity.16,17 Marathoners training in excess of 97 km (60 miles) per week self-reported twice the rate of URTI than those training at distances of less than 97 km.18 Marathoners also self-reported URTI at a rate of 33% within 2 weeks of completing a marathon, compared with controls who reported a rate of 15%.19 Using total training time or volume to explain infection risk is probably oversimplification because other factors, including stress, malnutrition, and weight loss, are likely contributors to infection risk.20

Average

Low Sedentary

Moderate

Very high

Exercise workload Figure 3A-1  Infection risk (J-curve). (Adapted from Nieman DC: Current perspective on exercise immunology. Curr Sports Med Rep 2:239-242, 2003.)

The effects of regular exercise on the immune system are summarized in Table 3A-1. As a result of randomized and epidemiologic studies of regular exercisers, Nieman proposed a J-curve theory for infection risk.7 He hypothesized that the untrained and elite athletes at opposite ends of the training spectrum are at greater risk for infectious illness than those who train moderately (Fig. 3A-1). The data to support this, however, are either conflicting or insufficient for those at the upper end of the training spectrum. Interestingly, athletes diagnosed with overtraining syndrome, presumably at the high end of the training spectrum, demonstrated normal cell counts and fewer infections than well-trained athletes.21,22 Similar immune function benefits can be observed in the older population. Training in the elderly population results in improved immune function compared with those who do not train.23 Vaccination response in elderly people is also improved in those who exercise regularly.24 In summary, those who exercise regularly have fewer infections than those who are sedentary and those who train at high levels and for long durations. If exposed to infection after an intense bout of exercise, the athlete is relatively protected. If already symptomatic with an infection, intense exercise may increase the severity of the ­illness. For those training on a regular basis or competing at a high level, preventing illness is a priority. Nieman offers ­common-sense recommendations to reduce infection risk7: 1. Minimize other stressors. 2. Ensure adequate nutritional intake, especially carbohydrates. 3. Avoid fatigue from “excessive” training. 4. Ensure adequate rest during training cycles. 5. Avoid rapid weight loss. 6. Avoid infectious exposure and inoculation. 7. Minimize or avoid potential infectious exposures before important competitions. 8. Obtain appropriate vaccines.

Nonorthopaedic Conditions

EPIDEMIOLOGY OF OUTBREAKS The most typical scenario for infectious disease is a single athlete contracting an infection by droplet, aerosolized, fecal-oral, or more rarely, blood-borne transmission. The athlete is diagnosed and treated with minimal disruption of the training cycle and competition. A more troublesome development is the occurrence of an outbreak of infectious illness. In a review of infectious disease outbreaks in athletes (1966-2005), Turbeville and colleagues found three modes of transmission: direct (skin-to-skin), indirect (respiratory, fecal-oral, body fluids), and common source (e.g., equipment, water bottles, soap, towels, whirlpools).3 Football, accounting for 34% of the reported infectious outbreaks, wrestling (32%), and rugby (17%) were the most common sports reporting outbreaks. A variety of other sports reported infrequent outbreaks. The responsible pathogens varied much more than the affected sports. Herpes simplex virus (22%) and Staphylococcus aureus infection (22%) accounted for most of the football, wrestling, and rugby outbreaks. Enteroviruses (19%) and tinea (14%) ranked next in frequency. The only aerosolized pathogen implicated in sports-related outbreaks was measles, responsible for 5% of the outbreaks. The sports and organisms most often involved support the finding of skin as the most commonly affected site of infectious outbreaks.

RESPIRATORY INFECTIONS Upper Respiratory Infection The “common cold” is just that—common. The average adult experiences one to six URTIs per year, caused 40% of the time by rhinoviruses.1 Athletes experience the same potential exposure and consequently will be infected at some time, likely during their competitive seasons. Cold exposure, especially in outdoor sports during the fall and winter seasons and indoor ice arenas, may increase the risk by drying the mucosal surfaces of the respiratory tract, adversely affecting the cilia and inhibiting the ability to serve as an effective barrier to viral illness.25 Parainfluenza and influenza illnesses also place athletes and physically active people at risk. These agents occur as more seasonal phenomena. The URTI, a self-limited illness, usually presents with onset of nasal congestion, rhinorrhea, lethargy, sore throat, cough, and low-grade fever. The symptoms may intensify over 3 to 5 days, then slowly resolve. Analgesics and antipyretics (acetaminophen, ibuprofen) and decongestants or decongestant-antihistamine combinations may be used for symptomatic treatment. Decongestant nasal sprays, such as oxymetazoline, may also be used. Be careful to limit their use to 4 to 5 days to minimize the likelihood of rhinitis medicamentosa (mucosal dependence on topical decongestants). Sinusitis and otitis media are potential, but infrequent, complications of URTIs. The effects of both individual exercise and training sessions and long-term training on infectious disease have previously been discussed. The important consideration, then, becomes whether to train or compete during the

149

i­llness or to rest and resume training and competition when symptoms permit. Based on animal studies, exercising after acquiring a bacterial infection may intensify the symptoms.12,13 Does this same risk apply to the common cold? Certainly, we have all witnessed media reports or actual contests where “sick” athletes performed at high levels with no apparent adverse effects. Are these performances exceptions or are they the rule? Sedentary subjects who performed exercise after naturally acquiring a URTI did not experience an effect on the symptoms or duration of the illness compared with nonexercising controls.26 A 3-year study of elite swimmers with mild illness and the effect on competitive performance demonstrated that female swimmers had a 32% chance of a beneficial effect, 31% chance of a trivial effect, and 37% chance of harmful effect on performance.27 Compare this to male swimmers, who had a 17% chance of beneficial effect, 31% chance of trivial effect, and 52% chance of harmful effect on performance. These results imply confounding factors of forced rest or modifying training because of the illness or, perhaps, that the individual’s perception of illness and effects or lack of effects can be overcome by effort. Viral and bacterial pharyngitides are other common URTIs. Certainly, identifying group A, β-hemolytic streptococcal infections is important for the potential sequelae of glomerulonephritis or rheumatic fever with associated valvular involvement. Viral pharyngitides, in association with rhinoviruses and influenza viruses, are far more common than streptococcal infections in adults (about 10% of adult sore throats) and more common in childhood and adolescence (about 50%). Participation with these illnesses can be treated as other viral illnesses, a topic that is discussed later. Sinusitis is a potential complication of, or comorbidity with, URTIs. The initial symptoms may resemble those of URTI. Consider the diagnosis of acute bacterial sinusitis in those who have URTI symptoms, usually lasting longer than 7 days and less than 4 weeks, accompanied by two of the following findings: purulent nasal discharge; maxillary, tooth, or facial pain or tenderness; unilateral maxillary sinus tenderness; or worsening of URTI symptoms after an initial course of improvement.28 Infectious mononucleosis poses additional return-to-play considerations and is discussed in a separate section. Available data suggest that participating with a URTI is athlete dependent. The risk appears to be only performance related without significant disease sequelae.

Lower Respiratory Infection Lower respiratory infections are less common than URTIs. Acute bronchitis and pneumonia constitute the common spectrum of lower respiratory infectious illness. Acute bronchitis, the most common lower respiratory illness in adults, is inflammation of the respiratory tree (trachea, bronchi, bronchioles).29 The responsible pathogens are most commonly of viral origin, including influenza A and B, parainfluenza viruses, coronaviruses, rhinoviruses, and adenoviruses. Bacterial sources are also implicated in bronchitis but are less common. The typical presenting symptoms are dry cough (lasting less than 3 weeks) and mild lassitude. Physical examination findings may be scant.

150

DeLee & Drez’s Orthopaedic Sports Medicine

Auscultation of the chest can occasionally yield course breath sounds or wheezes. The diagnosis is made based on history and examination. A chest radiograph is rarely necessary. Treatment is symptomatic. The cough, which may be exacerbated with exertion or lying down, can be treated with β2-agonist inhalers because bronchoconstriction is often the cause. Antibiotics are rarely indicated in treating acute bronchitis. If the cough persists for more than 3 weeks, further diagnostic investigation is warranted.29 Pneumonia is the other common cause of lower respiratory infection. Although the symptoms may be similar to those of bronchitis, the cough is productive of purulent sputum and is accompanied by fever, myalgias, and fatigue. Chest auscultation reveals crackles. Chest radiograph confirms the clinical suspicion. If influenza A or B is suspected, amantadine, rimantadine, or oseltamivir is an appropriate treatment choice. If bacterial causes are likely, a macrolide or doxycycline is an appropriate choice.29 Return-to-play decisions have been made based on tradition, individual experience, and expert opinion. Solid evidence is lacking as we guide athletes back to play after respiratory infections. An elevated temperature (commonly cited as 100.6° F, 38° C) can impair muscle strength, pulmonary perfusion, cognitive function, increase metabolic parameters, and increase insensitive fluid losses, all factors that can impair athletic performance.4,30 Fortunately, this is an objective measure, whereas many of the other recommendations involve subjective interpretations. The elevated body temperature is a relative guide for the clinician to place in the context of the history, symptoms, and physical examination. A common guideline developed by Eichner is a reasonable, practical approach to safely returning athletes to participation.31 He labels this process, the “neck check.” If the symptoms are all occurring above the neck (sore throat, congestion, rhinorrhea), participation is probably okay. If the symptoms include myalgias, elevated body temperature, or significant cough, participation is probably unwise. Included as a part of the neck check, probably the most practical of the recommendations, is a brief warm-up. If after a brief warm-up there is no worsening of symptoms, participation may be considered. If symptoms worsen, participation should be terminated. Animal studies raise the possibility of worsening symptoms of an illness when performing high-intensity exercise after symptoms of illness have occurred.13

INFECTIOUS MONONUCLEOSIS Infectious mononucleosis carries clinical implications (splenomegaly, fragile spleen) for athletes that other infectious illnesses lack. For that reason, careful consideration is necessary in the diagnosis and decision making regarding return to play. Infectious mononucleosis is a viral illness caused by the Epstein-Barr virus (EBV) and is spread by salivary contact. The incubation period is long, ranging from 30 to 50 days after exposure.32 The typical clinical presentation consists of a 3- to 5-day prodrome of lethargy, headache, and loss of appetite. For the next 5 to 15 days, the athlete experiences the more typical symptoms of sore throat, fever, swollen glands (particularly the posterior submandibular nodes), and

fatigue.33 The peak incidence of infectious ­mononucleosis occurs in the 15- to 21-year-old group, the most common years of competitive athletic participation. In fact, in the college age group, 1% to 3% are infected annually.34 Commonly, younger children may have infectious mononucleosis with minimal or no symptoms. As a result, many adults, without a specific recollection or clinical history of infectious mononucleosis, are seropositive (by heterophil antibody test, or Mono-Spot test) when tested.35,36 When presented with an athlete with a sore throat and fatigue, a history consistent with mononucleosis, and physical findings suspicious for mononucleosis, further evaluation is necessary to confirm the diagnosis. Laboratory results consistent with mononucleosis include a leukocytosis (white cell count, 15,000 to 25,000) with an absolute lymphocytosis (10% to 20% atypical) and a positive heterophil antibody test (typically the Mono-Spot).37,38 The heterophil antibody test usually becomes positive 5 to 7 days after the onset of the prodromal symptoms. The sensitivity of the heterophil antibody test is 63% to 85%, and the specificity is 84% to 100%.39 The low sensitivity is thought to be related to performing the test too early in the course of the illness before the test is expected to yield positive results. When the heterophil antibody test is positive, but the clinical course and physical examination are inconsistent with the diagnosis of infectious mononucleosis, the next step is to obtain the EBV IgG virally encoded antigen (VCA) and EBV IgM VCA serologies to determine whether the illness is, indeed, an acute case of infectious mononucleosis. EBV IgG VCA levels begin to rise within the first few days of the illness and decline from a maximum after 2 to 3 months to a steady-state level that may remain present for life. EBV IgM VCA represents an acute response appearing during the early symptoms and declining to undetectable levels after 1 to 3 months. Because the heterophil antibody screening tests for infectious mononucleosis are IgG based, a person’s heterophil antibody test may remain positive for a lifetime.40 If the implications for return to play were insignificant, confirming the diagnosis in atypical cases would be unimportant. Because splenomegaly associated with infectious mononucleosis has the potential for a substantial delay in returning to play interfering with a considerable portion of an athlete’s season, testing for the EBV serologies to confirm the diagnosis is recommended. If EBV IgG is present, but IgM is not, the illness is not infectious mononucleosis, and the athlete can return to play when symptoms permit. Liver function tests may increase 2 to 3 times above normal, usually peaking by the second or third week and returning to normal by the fifth week.33 These tests are probably not clinically relevant unless the athlete appears icteric. They seldom affect the clinical course and returnto-play decisions. Obtaining a “rapid strep” test to evaluate for concomitant group A, β-hemolytic streptococcal infection is advised. Studies of infectious mononucleosis have shown a coexisting group A, β-hemolytic streptococcal infection in 30% of cases.33 Treatment of infectious mononucleosis is supportive. Analgesics can be used to treat pain and fever. Rest or relative rest and no sport or training participation for at

Nonorthopaedic Conditions

least 3 weeks from illness onset is recommended because ­spontaneous splenic rupture is thought to be a risk during this time frame. In practice, if the tonsils are swollen enough to cause the individual to have difficulty handling oral secretions, a short course of prednisone can be used to reduce tonsillar hypertrophy. Use of corticosteroids does not ­otherwise alter the course of infectious ­mononucleosis. The controversial issues with infectious mononucleosis, then, involve splenomegaly and return-to-play decisions. During the course of mononucleosis, the spleen undergoes lymphocytic infiltration with enlargement within the first few weeks. The lymphocytic infiltration distorts the normal anatomy and support structures of the spleen, causing both enlargement and increased fragility.41 These changes result in an increased risk for splenic rupture but otherwise have no effect on the course of the illness. Splenomegaly in infectious mononucleosis is estimated to occur in 50% to 100% of cases.38,42 Splenic rupture associated with infectious mononucleosis occurs in 0.1% to 0.2% of cases.33 Case reports have documented both spontaneous rupture and rupture as a result of athletic activity. These same case reports showed that the timing of rupture occurs between day 4 and day 21 from the onset of symptoms. Clinical determination of splenomegaly is unreliable.43 As a result, various imaging techniques have been assessed for accuracy in determining spleen size. Computed tomography, magnetic resonance imaging, and ultrasound of the spleen are accurate in assessment, but defining the normal spleen size proves more difficult. A recent study of college athletes performed during their preparticipation examinations showed that 7% of asymptomatic athletes had spleens that exceeded the accepted normal size range.44 In general, male athletes had larger spleens than female athletes, and white males had larger spleens than African American males. Because of the variability of normal spleen size, the most effective means of using ultrasound is to obtain a baseline splenic ultrasound at admission to either high school or college for comparison in the event of mononucleosis occurring at a later time. However, this is both logistically and economically impractical given the low incidence of mononucleosis in these populations. Thus, we are left with clinical decision making based on experience, or the best evidence available. The usual standard of care in regard to return to play applies. The athlete should be free of symptoms, have a normal energy level, be well hydrated, and have no palpable splenomegaly; the athlete participating in a noncollision sport should be at least 21 days beyond the onset of symptoms. Rarely does splenomegaly continue beyond the fourth week. Although the timing of return to play for collision sports remains controversial, safe return is possible by the third to fourth week after the onset of symptoms. Many team physicians continue to use some form of imaging to assist in these decisions, but the wide variability of normal spleen size makes interpretation of the imaging unreliable. If a preseason, baseline splenic size has been determined, following the infectious mononucleosis spleen with serial ultrasounds to peak spleen size and resolution provides the clinician with objective return-to-play information.32 Again, however, this is expensive and impractical.

151

After confirming the diagnosis of infectious mononucleosis by a combination of clinical and laboratory information, the athlete should refrain from physical activity until at least 3 weeks from the onset of symptoms. Return to play is permitted if the athlete is completely symptom free and has no palpable spleen. Return to ­collision sports may be entertained at this time, but more safely, based on available information and practice tradition, at 4 weeks.

CARDIAC INFECTIONS Myocarditis Myocarditis is an uncommon phenomenon, in general, and is even less common in athletes. The consequences, however, have serious potential. Myocarditis, in most studies of exertional sudden death, is cited as a cause of death 10% to 42% of the time.45-47 As a result, physicians working with athletes with illness must be vigilant for symptoms suggestive of cardiac involvement. Myocarditis is defined as an inflammation of the myocardium with myocellular necrosis.48 The clinical diagnosis may be difficult because of variable symptoms at presentation. Some patients may be completely asymptomatic but then rapidly develop symptoms of congestive heart failure, syncope, and even sudden death as a result of a rhythm disturbance.48 Although this course is atypical, one must be suspicious of any symptoms of chest pain, rapid heart rate, and tachypnea following a recent, seemingly innocuous, viral illness. The incidence of myocarditis is low, ranging in frequency from 0.001% to 0.2 %.47 Men are diagnosed more frequently than women (62% versus 38%), and the affected individuals are usually younger than 40 years.47 The infectious cause is most often Coxsackie B virus, followed by human immunodeficiency virus (HIV), adenovirus, and cytomegalovirus (CMV). Medications such as doxorubicin may also be responsible as a chemical cause of myocarditis. Ampicillin, sulfa, lithium, hydrochlorothiazide, and indomethacin may cause myocarditis mediated by a hypersensitivity reaction.48 When myocarditis occurs, the acute phase may last up to 3 days. The subacute phase (days 4 to 14) is mediated by the immune response to the infectious process. It is this response that causes the myocardial damage.48 If, during the chronic phase (variably from days 15 to 90), the immune response continues, causing further myocardial damage, cardiomyopathy may result. If the immune response is moderate, complete recovery is likely. In fact, there are estimates suggesting that 5% of asymptomatic people have myocarditis with viral illnesses but recover completely.49 Diagnosing myocarditis is largely based on history and physical examination findings of tachypnea, tachycardia, S3, apical murmur, rales on chest auscultation, and distended neck veins. Information supporting the diagnosis includes elevations of several blood tests, including ­creatine kinase, troponin, erythrocyte sedimentation rate (ESR), and C-re­ active protein. The electrocardiogram is usually abnormal, but the changes are nonspecific. The chest radiograph may demonstrate increased left ventricular size. If the ­athlete

152

DeLee & Drez’s Orthopaedic Sports Medicine

has findings consistent with a diagnosis of ­myocarditis, consult a cardiologist for possible biopsy and further treatment recommendations. For most, the treatment is supportive with symptom management and diuretics and angiotensinconverting enzyme inhibitors when necessary.48 Although complete recovery is the usual result, 10% to 30% of patients may have a dilated cardiomyopathy. As many as 10% may be at risk for death due to congestive heart failure or arrhythmias.50 According to the guidelines published by the 36th Bethesda Conference, there should be 6 months of recovery before considering return to play.51

Pericarditis Pericarditis is defined as an inflammation of the pericardial sac. As with myocarditis, pericarditis has return-to-play implications. Thus, the sports physician must be familiar with the problem and its athletic implications. Pericarditis is typically infectious in origin, with viral pathogens most often responsible. URTIs often precede pericarditis.52 Of all causes of chest pain presenting to emergency departments, 5% are diagnosed as pericarditis.53 Of those diagnosed, most are in young adults.54 When pericarditis occurs, the inflammatory process, infectious disease, or other cause (e.g., hypersensitivity reaction, autoimmune disease) results in vascular permeability, causing fluid to accumulate within the pericardial sac. Given enough fluid within the pericardial sac, the fluid serves as a constraint to left ventricular filling and causes subsequent cardiac dysfunction. Clinical signs and symptoms suggestive of pericarditis include flu-like symptoms with chest pain and friction rub at the left middle and lower sternal border with ST elevation in all leads of the ECG. Sinus tachycardia may also be seen.55 Other confirmatory tests include echocardiography and elevated ESR. Cardiac enzymes are typically normal. Because most episodes of pericarditis are self-limited, the primary treatment is supportive. Nonsteroidal antiinflammatory drugs can be given in uncomplicated cases. For more severe situations, corticosteroids can be used, but these are rarely necessary. Most symptoms resolve within 2 weeks.55 Pericardiocentesis is rarely indicated. The 36th Bethesda Conference recommendations for athletes and physically active individuals urge no activity during the acute phase (usually 2 weeks). After the acute phase, athletes may return to activity when there is no evidence of active disease.51 If there is evidence of myocardial involvement, return-to-play decisions should be the same as those recommended for myocarditis, that is, 6 months of recovery. Recurrence of pericarditis is rare in the situation of infectious pericarditis. However, when no cause can be identified (idiopathic pericarditis), the recurrence rate may be as high as 15% to 30%.56

INFECTIOUS DIARRHEA Athletes and physically active adults develop infectious diarrhea with a frequency similar to a sedentary population. Rare outbreaks may develop as a result of specific sports competition.

Common causes of community-acquired diarrhea are Salmonella, Shigella, and Campylobacter species, Escherichia coli O157:H7, and Clostridium difficile infection.57 Most are self-limited illnesses and need only supportive treatment such as rehydration, dietary modification, and, possibly, antidiarrheal agents. If the diarrhea is associated with a significant fever or blood in the stool, stool cultures are obtained to identify the causative agent. When the responsible pathogen is identified, treatment is begun with the appropriate antibiotic, including a quinolone for suspected Shigella species infection or a macrolide for suspected ­Campylobacter species infection.57 For a diarrheal illness without fever or bloody stool that persists for more than 7 days, consider other pathogens such as parasites (especially with travel outside the athlete’s usual locale), such as Giardia, Cryptosporidium, or Cyclospora species infection. In this situation, obtain a stool sample for ova and parasites. If symptoms warrant, consider evaluation for inflammatory bowel disease. Treatment is based on the organism identified by the examination for stool pathogens.57 Specific gastrointestinal outbreaks associated with water-sport activities have been reported. Leptospirosis (spirochete of various Leptospira species) has been identified in tri-athletes exposed to contaminated fresh water during the swimming portion of the event.58 Paddle-sport athletes are also at risk if exposed to contaminated water. Symptoms of fever, nausea, vomiting, and diarrhea occur following an incubation period of 2 to 20 days. A late phase of the illness may cause aseptic meningitis, rash, or uveitis. Those with severe symptoms may have to be hospitalized for parenteral therapy. Those with mild to moderate symptoms may be treated with doxycycline, 100 mg twice daily for 7 days, or amoxicillin, 500 mg 4 times a day for 7 days.58 Giardia species outbreaks have also been associated with fresh-water paddle and swimming sports.58 Persistent, intermittent diarrhea is the prominent physical complaint with this illness. Stool samples for ova and parasites yield the diagnosis 70% of the time. Direct stool antigen assay has a sensitivity of 98%.58 The recommended treatment is metronidazole, 250 mg 3 times a day for 5 to 7 days.

URINARY TRACT INFECTIONS Diagnosis of urinary tract infection (UTI) is one of the most common diagnoses of infectious disease, and athletes are also be afflicted with this annoying problem.59 Half of all women in the United States and Canada report at least one UTI over the course of a lifetime.60 Based on these data, women athletes are more likely than male athletes to report symptoms consistent with UTI during the course of training and competition. Typical presenting symptoms are urgency, dysuria, and urinary frequency. Dipstick urinalysis positive for leukocyte esterase (specificity, 94% to 98%; sensitivity, 75% to 96%) or nitrites accompanying the classic symptoms is usually adequate for the diagnosis.61 A urine culture with sensitivities is confirmatory but not necessary in the uncomplicated UTI. For someone with a history of recurrent UTIs, the culture and sensitivities become increasingly valuable to identify specific pathogens.

Nonorthopaedic Conditions

Treatment of an uncomplicated UTI is with t­ rimethoprim-sulfamethoxazole, 160/800 mg twice a day for 3 days. For a treatment failure or history of resistant organisms, ciprofloxacin, 250 mg twice daily for 3 days, or nitrofurantoin, 100 mg twice daily for 7 days, is an acceptable choice. For postmenopausal female athletes, the treatment choices are identical, except the duration of therapy is extended to 7 days.62 Men have a low incidence of UTIs, especially in the most active age group, 15 to 50 years of age. As a result, research and recommendations for treatment are limited. Diagnosis and treatment are similar to that in women. Treatment, however, should be for at least 7 days because the likelihood of complications is higher in men. Nitrofurantoin has poor tissue penetration and is not recommended in men.62,63

BLOOD-BORNE INFECTIONS Human Immunodeficiency Virus About 1 million Americans are infected with HIV. The acute infection of HIV is similar to that of mononucleosis or CMV and includes fever, malaise, and lymphadenopathy, but not all individuals develop symptoms during the acute infection. There is a long asymptomatic state lasting for many years during which the person is healthy and unaware that he or she is infected and allowing spread of the disease.64 As the infected person’s CD4 counts decline, he or she will develop opportunistic infections and ­neoplasms and, ultimately, overt acquired immunodeficiency syndrome (AIDS). Because of its extremely long asymptomatic period, the potential for transmission is high. HIV is present in body fluids, but only blood poses a significant risk for transmitting the virus. Direct contact of body fluids with open wounds or mucous membranes is required for infection. There is no evidence that an individual can acquire HIV through intact skin. There is only one suspected case of possible transmission of HIV in a professional soccer player in Italy in 1990, but there is insufficient evidence to confirm transmission.64 Given the prevalence of the disease and no confirmed transmission of HIV in the athletic setting, the risk for transmission from one athlete to another is exceedingly low. One study calculated the risk to an NFL football player to be 1 per 85 million games.64 The risk for an athlete acquiring HIV is greatest off the field of play through unsafe sexual practices; shared needles with the use of ergogenic aids such as anabolic steroids, growth hormone, and erythropoietin; and illicit parenteral drug use. Because an effective vaccine for HIV is not available and treatment at this point is not curative, prevention is paramount. Universal precautions should be used without exception by all coaches and medical personnel when caring for athletes. Athletes with open wounds that can bleed in competition should be protected in such a way that they can withstand the demands of competition. If an athlete has anything more than a superficial scratch or abrasion, he or she should be removed from competition until the wound is appropriately covered and blood

153

soaked clothing removed and replaced. A detailed list of preventive meas­ures has been documented by the American Academy of Pediatrics65 and the American Medical Society for Sports Medicine (AMSSM) and American Orthopedic Society for Sports Medicine (AOSSM).64   The detailed measures recommended by the AMSSM and AOSSM follow: 1. Pre-event participation includes proper care for existing wounds. Abrasions, cuts, or oozing wounds that may serve as a source of bleeding or as a portal of entry for blood-borne pathogens should be covered with an occlusive dressing that will withstand the demands of competition. Likewise, care providers with healing wounds or dermatitis should have these areas adequately covered to prevent transmission to or from a patient. 2. Necessary equipment or supplies important for compliance with universal precautions should be available to caregivers. These supplies include latex or vinyl gloves, disinfectant, bleach (freshly prepared in a 1:10 dilution with tap water), antiseptic, designated receptacles for soiled equipment or uniforms (with separate waterproof bags or receptacles appropriately marked for uniforms and equipment contaminated with blood), bandages or dressings, and a container for appropriate disposal of needles, syringes, or scalpels. 3. During the sports event, early recognition of uncontrolled bleeding is the responsibility of officials, athletes, and medical personnel. Participants with active bleeding should be removed from the event as soon as this is practical. Bleeding must be controlled and the wound cleansed with soap and water or an antiseptic. The wound must be covered with an occlusive dressing that will withstand the demands of the activity. When bleeding is controlled and any wound properly covered, the player may return to competition. Any participant whose uniform is saturated with blood, regardless of source, must have that uniform changed before returning to competition. 4. The athletes should be advised that it is their responsibility to report all wounds and injuries in a timely manner, including those recognized before the sporting activity. It is also the athletes’ responsibility to wear appropriate protective equipment at all times, including mouth protectors, in contact sports. 5. The care provider managing an acute blood exposure must follow the guidelines of universal precautions. Appropriate gloves should be worn when direct contact with blood, body fluids, and other fluids containing blood can be anticipated. Gloves should be changed after treating each participant, and as soon as practical after glove removal, hands should be washed with soap and water or antiseptic. 6. Minor cuts and abrasions commonly occur during sports. These types of wounds do not require interruption of play or removal of the participant from competition. Minor cuts and abrasions that are not bleeding should be cleansed and covered during scheduled breaks in play. Likewise, small amounts of bloodstain on a uniform do not require removal of the participant or a uniform change.

154

DeLee & Drez’s Orthopaedic Sports Medicine

7. Lack of protective equipment should not delay emergency care for life-threatening injuries. Although HIV is not transmitted by saliva, medical personnel may prefer using airway devices. These devices should be made available whenever possible. 8. Any equipment or area (e.g., wrestling mat) soiled with blood should be wiped immediately with paper towels or disposable cloths. The contaminated areas should be disinfected with a solution prepared with 1  part household bleach to 10 parts water and should be prepared fresh daily. The cleaned area should be dry before reuse. Persons cleaning equipment or collecting soiled linen should wear gloves. 9. Postevent considerations should include re-evaluation of any wounds sustained during the sporting event. Further cleansing and dressing of the wound may be necessary. Also, blood-soiled uniforms or towels should be collected for eventual washing in hot water and detergent. 10. Procedures performed in the training room are also governed by adherence to universal precautions. Care providers should wear gloves. Any blood, body fluids, or other fluids containing blood should be cleaned in a manner as described previously. Equipment handlers, laundry personnel, and janitorial staff should be advised to wear gloves whenever contact with bloody equipment, clothing, or other items is anticipated. Appropriate containers for the disposal of needles, syringes, or scalpels should be available. 11. Members of the athletic health care team are considered to be covered by Occupational Safety and Health Administration guidelines. Assessment of the application of these guidelines must be made on an individual basis. This application may include consideration for hepatitis B virus (HBV) immunization for some personnel who are involved with the athletic health care team. No recommendation has been specifically made for the immunizations against HBV for athletes in particular. However, several groups now recommend universal immunization against HBV of the newborn and college-age groups.64 The decision to allow the participation of an athlete who is HIV positive should be individualized and should involve the athlete, his or her parents (if appropriate), the athlete’s personal physician, and the team physician. Factors affecting the decision to participate should include the athlete’s current state of health, the status of the HIV infection, the nature and intensity of training, the potential contribution of stress from athletic competition, and the potential risk for transmission. Although the data are limited, the following exercise guidelines have been suggested66: 1. Exercise is a safe and beneficial activity for the HIVinfected person. 2. HIV-infected individuals should begin exercising while healthy and adopt strategies to help them maintain an exercise program throughout the course of their ­illness. 3. HIV-infected persons, through the use of exercise, can play an important role in the management of their illness, while improving quality of life.

4. Exercise has the potential of other subtle and effective behavioral therapeutics benefits regardless of ethnicity, exposure category, or gender and is particularly promising in areas in which pharmacologic treatments are not readily available (e.g., underdeveloped countries). Mandatory testing is not recommended by any organization because of the extremely low risk for athletic transmission and because mandatory testing would not result in preventing transmission in sport. The ethical, moral, legal, medical, financial, and logistical considerations in mandatory testing for HIV or other blood-borne illnesses are challenging and beyond the scope of this discussion. Testing for HIV should be performed if there is a known exposure or the athlete has had multiple sexual partners, uses intravenous (IV) drugs or ergogenic aids, has sexual contact with at-risk persons, has had other sexually transmitted diseases including HBV, or had a blood transfusion before 1985.64 If an athlete is HIV positive, this information can be used to protect and treat the infected athlete, his or her sexual partner, and other athletes. Because of patient-doctor confidentiality matters, the athlete’s physician should not reveal an athlete’s HIV status to the coach, the school, or anyone involved with the athlete. This is to protect the athlete’s ability to participate in athletics safely and without discrimination. The present laws protect the physician from any legal responsibility for not revealing an athlete’s HIV status.65

Hepatitis A, B, C, D, and E Five viruses cause the majority of infectious hepatitis in the United States: hepatitis A, B, C, D, and E. The prevalence of infectious hepatitis among competitive athletes in not known.67 The risk for acquiring hepatitis as a result of active participation in sports is extremely low, and the athlete is much more likely to acquire these infections from off-field activities. Although there are no official recommendations requiring immunization, it is recommended that all athletes be immunized against HBV, and athletes who will compete in areas at high risk for hepatitis A virus (HAV) should also receive HAV vaccination. As of 2007, it is recommended that all children born in the United States be immunized against HAV and HBV. The symptoms and signs of all forms of hepatitis are similar and may include asymptomatic infection, anorexia, malaise, fatigue, myalgia, arthralgia, diarrhea (more common in children), jaundice, headache, right upper quadrant pain and fever (more common in HAV), serum sickness syndrome, pharyngitis, hepatomegaly, splenomegaly, and in rare cases, fulminant hepatitis. Lymphadenopathy is not a common symptom of infectious hepatitis but is commonly seen in infectious mononucleosis, CMV infection, and HIV infection. HAV is endemic in developing countries, and about 40% of adults in the United States show serologic evidence of prior infection. It is an RNA picornavirus that is transmitted by fecal-oral, direct person-to-person contact, or through contaminated drinking water. Incubation is 15 to 60 days, and individuals are most contagious in late incubation. Fifteen percent of infected individuals have prolonged or relapsing hepatitis. Fulminant hepatitis and death rarely occur. Because of genetic stability throughout

Nonorthopaedic Conditions

the world, chronic infection with HAV rarely occurs. An effective vaccine is available. Hepatitis E virus (HEV) is a single-stranded RNA virus that, like HAV, is enterically acquired by fecal-oral ­transmission through consumption of contaminated drinking water or food, but not by person-to-person transmission. HEV is most common in Asia, Mediterranean countries, and ­Central America. The prevalence of HEV is low in the United States, where virtually all cases of HEV are among travelers returning from endemic areas.67 Incubation is 15 to 60 days. The symptoms, signs, and course of illness are similar to those of HAV. There is no immunization for HEV. HBV, a DNA hepadnavirus, is transmitted parenterally, by sexual contact, perinatally, and potentially during athletic activities in which an athlete comes in direct contact with another athlete’s blood or body secretions. The incubation period is 45 to 160 days. Most individuals recover completely, but 6% to 10% progress to chronic infection each year. The peak age of infection is 20 to 39 years. The most common mode of transmission is heterosexual contact, but about one third of infected individuals have no known risk factors.67 Hepatitis C virus (HCV), an RNA flavivirus, occurs when there is contact with blood or blood products, primarily through transfusion (now less than 1% who receive a blood or blood product transfusion), injecting drugs (including anabolic steroids with shared needle use), and needle-stick exposure. Sexual transmission is less common. Incubation is 14 to 180 days. Seroprevalence of HCV in the United States is 1.8%.68 The concern about acquiring hepatitis C is that up to 70% of individuals develop chronic infection, and about one fourth of these develop cirrhosis. In fact, HCV is the most common reason for liver transplantation in the United States. Unlike HAV, HCV mutates rapidly and escapes immune surveillance by the host. There is no effective vaccine for HCV. Hepatitis D virus (HDV), a defective single-stranded RNA virus, requires coinfection with HBV for viral replication. An individual who has detectable hepatitis B surface antigen is at risk for coinfection with HDV. Incubation is 42 to 180 days. HDV is prevalent worldwide but is more commonly seen in Africa, Central Asia, Italy, and the Middle East. The overall prevalence in the United States is low except in IV or intramuscular drug users. Transmission is primarily parenteral and less frequently through sexual contact. Coinfection with HBV and HDV results in a more severe acute hepatitis and essentially guarantees progression to chronic hepatitis. As with hepatitis C, there is no effective vaccine for HDV. The AMSSM and AOSSM have each published position statements that state that acute viral hepatitis should be viewed like other viral infections and that specific activity and athletic participation recommendations should be based on the individual’s clinical condition.64 In athletes who have chronic hepatitis, multiple studies have examined the effect of endurance exercise on liver function, and no studies have found significant changes in liver function.67 Chronic persistent viral hepatitis should be viewed much like HIV. Although no firm recommendations have been made for chronic persistent hepatitis, the ­recommendations used for exercise and competition should be similar to those for athletes with HIV.67 Using hygienic practices such as not sharing food and water and thorough hand washing before meals can ­prevent

155

transmission of HAV and HEV. Outbreaks of HAV have occurred in multiple sports because of poor hygienic practices.67 Only one case of athletic transmission of HBV has been documented: multiple high school sumo wrestlers in Japan were infected after being exposed to a wrestler with known HBV who had multiple scars that often bled during competition. The only other documented case of bloodborne pathogen transmission occurred in Swedish crosscountry athletes who acquired HBV after ­cleaning their superficial skin wounds with water from a common, contaminated source.65 To date, there is no known transmission of HBV or HCV in the United States during practice or competition.67 Therefore, because the risk for transmission is extremely low, when appropriate precautions are taken, an athlete with chronic persistent hepatitis should be allowed to practice and compete in all sports.

Lyme Disease Lyme disease is a tick-borne systemic illness caused by the spirochete Borrelia burgdorferi. This spirochete is carried by the Ixodes scapularis tick (black-legged or deer tick) in the eastern United States and the Ixodes pacificus (western black-legged tick) in the western United States. Lyme disease is most prevalent in the Northeast and Upper Midwest but has a widespread distribution throughout the United States. Heavily infested tick habitats, such as wooded areas containing trees, brush, leaf litter, woodpiles, and long grass, pose a risk to the outdoor athlete.69 Lyme disease has three stages. Stage 1 (early localized) presents with influenza-like symptoms, regional lymphadenopathy, myalgia, headache, and in most cases, the classic erythema migrans rash. Stage 2 (early disseminated) pre­sents weeks to months later, and any of the following can occur: atrioventricular block, myopericarditis, pancreatitis, malaise, fatigue, regional or generalized lymphadenopathy, migratory pain in joints, bone, muscle, brief arthritis, meningitis, Bell’s palsy, cranial neuritis, radiculoneuritis, and secondary annular lesions. Stage 3 (late chronic) can present months to years later as fatigue, prolonged arthritis, encephalopathy, polyneuropathy, leukoencephalitis, lymphocytoma, and acrodermatitis chronica atrophicans.69 The symptoms of arthralgia and myalgia may mimic musculoskeletal injury. Consequently, Lyme arthritis must be included in the differential diagnosis of joint and muscle problems that present atypically. Laboratory testing must be used appropriately to avoid excessive false-positive and false-negative test results. The two-step approach of the Centers for Disease Control and Prevention (CDC) combines the pretest probability, the time since onset of the disease, and the Western blot test IgM (symptom onset less than 4 weeks) and IgG (all tests) titers. If the CDC criteria are used, the specificity of the two-step approach is 99% to 100%.70 Treatment of stage 1 Lyme disease includes oral doxycycline, 100 mg twice a day; amoxicillin, 500 mg 3 times a day; cefuroxime, 500 mg twice a day; or erythromycin, 250 mg 4 times a day—all for 14 to 21 days. Treatment of stage 2 or 3 or resistant stage 1 includes ceftriaxone, 2 g IV daily; cefotaxime, 2 g IV every 4 hours; or penicillin G, 24 million units IV every 24 hours—all for 14 to 21 days. There are other oral and IV regimens that can be used based on local resistance patterns.70

156

DeLee & Drez’s Orthopaedic Sports Medicine

The Lyme disease vaccine (LYMErix) has been removed from the market, and at present no vaccine for Lyme disease exists. To avoid Lyme disease, insect repellants ­containing N,N-diethyl-3-methylbenzamide (DEET), permethrin applied to clothing, or permethrin-impregnated clothing can prevent tick bites; also useful are protective clothing, such as boots and socks. Because the tick usually must be attached for at least 48 hours before transmitting the infection, frequent tick checks are essential to prevent the disease. C

r i t i c a l

P

o i n t s

l Moderate

exercise enhances the immune system. an acute, intense bout of exercise, the athlete may be transiently at increased risk for infection. This is the open-window hypothesis. l The neck check is a simple clinical tool to assist the team physician in determining return to play in an athlete with illness. If symptoms occur above the neck, ­participation is possible if the athlete feels okay after warm-up. If ­symptoms occur below the neck, participation is ­discouraged. l Concerns regarding splenomegaly and fragile spleen ­accompanying infectious mononucleosis suggest no ­activity for 3 weeks after onset of symptom, with return to ­collision sports 3 to 4 weeks after onset of symptoms. l HIV transmission during sports activity has not been ­reported. l Athletes, athletic trainers, and physicians should ­observe universal precautions in the training room and on the s­ ideline. l Transmission of HBV during sports activity is rare. No cases of transmission among athletes have been reported in the United States. l Following

S U G G E S T E D

R E A D I N G S

American Medical Society for Sports Medicine (AMSSM) and the American ­Orthopedic Society for Sports Medicine (AOSSM): Human immunodeficiency virus (HIV) and other blood-borne pathogens in sports. Joint position statement, 1995. Beck CK: Infectious disease in sports. Med Sci Sports Exerc 32(7 Suppl):S431-S438, 2000. Brenner I, Shek P, Shephard B: Infection in athletes. Sports Med 17:86-107, 1994. Lorenc TM, Kernan MT: Lower respiratory infections and potential complications in athletes. Curr Sports Med Rep 5:80-86, 2006. Malm C: Exercise immunology: The current state of man and mouse. Sports Med 34(9):555-566, 2004. Metz JP: Upper respiratory tract infections: Who plays, who sits?. Curr Sports Med Rep 2:84-90, 2003. Nieman DC: Nutrition, exercise and immune system function. Clin Sports Med 18:537-548, 1999. Nieman DC: Current perspective on exercise immunology. Curr Sports Med Rep 2:239-242, 2003. Turbeville SD, Cowan LD, Greenfield RA: Infectious disease outbreaks in competitive sports: A review of the literature. Am J Sports Med 34(11):1860-1865, 2006. Waninger KN, Harcke HT: Infectious mononucleosis and return to play. Clin J Sport Med 15(6):410-416, 2005.

R eferences Please see www.expertconsult.com

S ect i o n

B

Management of Hypertension in Athletes Shafeeq Ahmed and Paul D. Thompson Hypertension is the most common cardiovascular ­ roblem in athletes, although the prevalence of hypertenp sion among athletes is about 50% lower than that in the general population.1 The risk for developing hypertension in both athletes and the general population is increased among blacks and with age, a family history of hypertension, diabetes, obesity, and renal disease. Although awareness, treatment, and control of blood pressure have improved, blood pressure control remains inadequate in many patients.2 According to the World Health Organization, inadequate blood pressure control is responsible for 49% of cases ischemic heart disease and 62% of cerebrovascular disease events, with little variation by sex. In addition, inadequate blood pressure control is the number one ­attributable risk factor for death throughout the world.3 This chapter describes the diagnosis and treatment of hypertension in athletes based on the Seventh

report of the Joint National Committee on Prevention, ­Detection, Evaluation and Treatment of High Blood Pressure (JNC 7) and the 36th Bethesda Conference Task Force 5 ­recommendations.

CLASSIFICATION OF BLOOD PRESSURE JNC 7 classifies hypertension into prehypertension and stage 1 and stage 2 hypertension (Table 3B-1). Most athletes with hypertension have either prehypertension or stage 1 hypertension. Hypertension can be classified into primary, or idiopathic, and secondary hypertension. Primary hypertension refers to that with no detectable underlying cause, whereas secondary ­hypertension refers to hypertension secondary to some other disease process.

Nonorthopaedic Conditions

  Table 3B-1  Classification of Blood Pressure in Adults Type

SBP (mm Hg)

Normal Prehypertension Stage 1 hypertension Stage 2 hypertension

<120 120-139 140-159 ≥160

DBP (mm Hg) and or or or

<80 80-89 90-99 ≥100

DBP, diastolic blood pressure; SBP, systolic blood pressure. Data from the Seventh report of Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. JAMA 289:2560-2571, 2003.

157

cuffs overestimate the true pressure. Blood pressure can also be underdiagnosed and overdiagnosed by the use of inadequately calibrated aneroid or electronic devices. Mercury sphygmomanometers are the gold standard for blood pressure measurement, and other devices should be calibrated using a mercury instrument approximately every 6 months. If an athlete is found to be hypertensive on screening, he or she should return for more formal measurements using standard measurement techniques (Box 3B-1). The diagnosis and classification of hypertension should be based on blood pressure measurements ­performed on at least two separate occasions.

WHITE COAT HYPERTENSION BLOOD PRESSURE MEASUREMENT AND THE DIAGNOSIS OF HYPERTENSION Hypertension should not be diagnosed cavalierly in young athletes because such a diagnosis can have emotional, selfimage, and insurability implications. Screening blood pressure measurements can be obtained without careful regard to measurement guidelines because blood pressure is only rarely underestimated by inadequate measurement techniques. The exceptions include the use of too large a blood pressure cuff because large cuffs underestimate and small

White coat hypertension is defined as elevated blood pressure measurements when these are determined by a ­doctor or nurse, but normal when measured at home, work, or by ambulatory blood pressure monitoring. White coat hypertension may be found in 20% of people with mild hypertension.4 There are no outcome trials to determine whether or not white coat hypertension should be treated, but it is possible that individuals who are hypertensive during blood pressure measurement are also hypertensive at other stressful times. Many young individuals have white coat hypertension, so this possibility should be considered in athletes.

Box 3B-1  Guidelines for Blood Pressure Measurement Posture Blood pressure should be obtained after 5 minutes in the seated position, with the back supported by the chair, feet on the floor, and arm supported at the level of the heart. Circumstances No caffeine during the hour preceding the reading. No smoking during the 30 min preceding the reading. The s­ etting should be quiet and the room warm. Equipment Cuff size: The bladder should encircle at least 80% of the arm circumference. Small cuffs elevate the reading. Large cuffs reduce the reading. Manometer: Use a mercury, recently calibrated aneroid, or validated electronic device. Technique Number of Readings At least two readings, separated by as much time as practical, should be taken. If readings vary by >5 mm Hg, a­ dditional readings are required. If the arm pressure is elevated, leg measurements should be obtained, especially in individuals <30 years of age, to detect coarctation. Initial pressures should be obtained in both arms; if the pressures differ, use the arm with the higher pressure. If the initial values are elevated, obtain two other sets of readings at least 1 week apart. Performance The first step is to obtain a systolic measurement using palpation of the radial pulse. The bladder should then be quickly inflated to a pressure 20 mm Hg more than the palpated systolic pressure. The bladder is then deflated at 2 mm Hg per second, and the Korotkoff phase I (appearance) and phase V (disappearance) pressures recorded. If the Korotkoff sounds are weak, have the patient raise the arm and open and close the hand 5 to 10 times; reinflate the bladder quickly. Recordings Blood pressure, patient position, arm and cuff size From Pickering TG, Hall JE, Appel LJ, et al: Recommendations for blood pressure measurement in humans and experimental animals. Part 1: Blood p ­ ressure measurements in humans: An AHA scientific statement from the Council of High Blood Pressure Research, Professional, and Publication Subcommittee. ­Hypertension 45:142-161, 2005.

158

DeLee & Drez’s Orthopaedic Sports Medicine

SECONDARY HYPERTENSION IN ATHLETES Secondary hypertension refers to hypertension caused by some other disease and occurs in less than 5% of athletes and physically active people.5 Secondary ­hypertension may be due to a variety of conditions that can be diagnosed by a variety of tests (Table 3B-2). The most common cause of secondary hypertension in the general population is ­parenchymal and vascular renal disease.5 Secondary causes of hypertension should be sought when hypertension occurs in younger patients, in adults with severe hypertension, in patients with rapid-onset hypertension, and in patients who respond poorly to conventional medical treatment.

CLINICAL EVALUATION The clinical evaluation of athletes with hypertension documented on two different visits should include a medical history, physical examination, and limited laboratory ­testing to evaluate secondary causes of hypertension and to assess target-organ damage. The history should focus on behaviors that can increase blood pressure, including alcohol and drugs such as cocaine, amphetamines, and anabolic steroids. The athlete may be unaware of the composition of dietary supplements and should be asked whether he or she takes supplement to increase energy or facilitate weight loss. Such supplements may contain ma huang, ­ephedra, and guarana, which contain adrenergic compounds or ­caffeine-like stimulants.6 Athletes with hypertension should also be queried about the use of nonsteroidal medications and cold remedies because both may increase blood ­pressure. The physical examination of an athlete with hypertension should focus on detecting secondary causes of hypertension and on judging the severity of the blood pressure elevation. The critical elements include a funduscopic examination to detect retinal signs of persistently elevated pressure, even though such findings are rare in athletes; a thyroid examination to detect thyromegaly; auscultation of the heart; auscultation of the abdomen to detect renal artery bruits; and a determination of the radial-femoral pulse delay. Significant aortic coarctation should produce a delay in the femoral impulse compared with the radial impulse, even when the femoral pulse is preserved because of collateral blood flow.

Consequently, presence of a ­femoral pulse does not exclude coarctation of the aorta if there is radial-femoral pulse delay. ­Aortic insufficiency can cause systolic hypertension. Typically the blood pressure in aortic insufficiency shows high systolic and low diastolic values with a wide pulse pressure. When the hypertension is primarily systolic, the patient should be examined for aortic insufficiency by listening for a diastolic murmur at the cardiac base both over the sternum and to the right of the sternum in the aortic area. When in doubt, the patient should undergo an echocardiographic examination with Doppler ­interrogation. Routine laboratory testing should include urinalysis and serum measurements of electrolytes and renal function. A low serum potassium level without treatment or a rapid reduction in potassium with diuretic therapy may indicate hyperaldosteronism. Hyperaldosteronism due to adrenal hyperplasia in the absence of defined adenomas is increasingly recognized among patients with difficult-­ to-treat hypertension. This responds promptly to aldosterone antagonists such as spironolactone. Additional testing for other secondary causes should be based on the clinical impression and physical examination.

TREATMENT OF HYPERTENSION IN ATHLETES Nonpharmacologic therapy is the first step in managing hypertension in all patient groups (Table 3B-3). Weight loss, alcohol restriction, aerobic exercise, and special diets reduce blood pressure levels. The Dietary Approaches to Stop Hypertension (DASH) eating plan is one such diet. This diet is rich in fruits, vegetables, and low-fat dairy products; has reduced cholesterol as well as saturated and total fat6; and is rich in calcium and potassium and low in sodium.7,8 Although such measures are effective in the ­general ­population, their utility is often less in athletes because athletes are already physically active and often lean, or because some, such as American football linemen, resist weight loss because of their athletic activity. Alcohol use, in contrast, is prevalent among college athletes, and alcohol cessation may be effective in reducing hypertension in such patients. Nevertheless, most athletes with persistent hypertension will require pharmacologic therapy.

  Table 3B-2  Screening Tests for Secondary Hypertension Diagnosis

Diagnostic Test

Chronic kidney disease Coarctation of aorta Cushing’s syndrome and other glucocorticoid-excess states, including chronic steroid therapy Drug induced or drug related* Pheochromocytoma Primary aldosteronism and other mineralocorticoid-excess states

Estimated glomerular filtration rate Computed tomography angiography History, dexamethasone suppression test

Renovascular hypertension, including fibromuscular dysplasia in women Sleep apnea Thyroid or parathyroid disease

History, drug screening 24-hour urinary metanephrine and normetanephrine 24-hour urinary aldosterone level or specific measurements of other mineralocorticoids Doppler flow study; magnetic resonance angiography Sleep study with O2 saturation Thyroid-stimulating hormone; serum parathyroid hormone

*Drugs commonly associated with secondary hypertension in athletes include nonsteroidal anti-inflammatory drugs, cocaine, amphetamines, sympathomimetics (­decongestants, anorectics), oral contraceptive hormones, anabolic steroids, adrenal steroid hormones, erythropoietin, licorice (including some licorice-flavored chewing tobacco), selected over-the-counter dietary supplements, and medicines such as ephedra, ma huang, and bitter orange. Data from the Seventh Report of Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. JAMA 289:2560-2571, 2003.

Nonorthopaedic Conditions

159

  Table 3B-3  Possible Reductions in Blood Pressure with Hygienic Methods Modification* Weight reduction Adopt DASH eating plan Dietary sodium reduction Physical activity Moderation of alcohol consumption

Recommendation

Approximate Systolic Blood Pressure Reduction kg/m2)

Maintain normal body weight (body mass index, 18.5-24.9 Consume a diet rich in fruits, vegetables, and low-fat dairy ­products with a reduced content of saturated and total fat. Reduce dietary sodium intake to no more than 100 mmol per day (2.4 g sodium or 6 g sodium chloride). Engage in regular aerobic physical activity such as brisk walking (at least 30 min per day, most days of the week). Limit consumption to no more than 2 drinks (e.g. 24 oz beer, 10 oz wine, or 3 oz 80-proof whiskey) per day in most men, and to no more than 1 drink per day in women and lighter weight persons.

5-20 mm Hg/10 kg 8-14 mm Hg 2-8 mm Hg 4-9 mm Hg 2-4 mm Hg

*The effects of implementing these modifications are dose and time dependent and could be greater for some individuals. DASH, Dietary Approaches to Stop Hypertension; SBP, systolic blood pressure. Data from the Seventh Report of Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. JAMA 289:2560-2571, 2003.

Management of athletes with hypertension is more difficult than management in other patients because some medications may impair exercise performance or not be allowed in certain sports. For example, β-adrenergic ­blockers are prohibited in sports such as archery and shooting because they slow the heart rate.9-11 A slower heart rate allows more time to aim between shots and reduces the possibility that a systolic surge in pressure will alter aim. Diuretics are also restricted in many sports because they dilute the urine and thereby mask the ­presence of other prohibited substances. These considerations apply only to drug testing in competitive sports. Table 3B-4 shows the commonly used antihypertensives and their usual doses.

Diuretics Thiazide and loop diuretics decrease plasma volume, cardiac output, and peripheral vascular resistance. ­Thiazide diuretics decreased mortality and morbidity in ­various randomized controlled clinical trials.12-18 They are generally safe and inexpensive. They are effective in black patients whose hypertension is often sensitive to salt loading and to volume depletion. The adverse effects of thiazides include hypovolemia, orthostatic hypotension, hypokalemia, and hypomagnesemia. These side effects can increase the risk for muscle cramps and even heat stroke and rhabdomyolysis in athletes engaged in intense exercise during warm weather. Consequently, their use in athletes routinely subjected to such conditions should be avoided, including American football players during warm weather workouts. Stopping such medicines for 2 days before major competition in other athletes, such as distance runners, is also advised. Loop diuretics such as furosemide and ethacrynic acid are more potent than thiazide diuretics but have a shorter duration of action and are less useful than thiazides in managing hypertension. It must be stressed that diuretics cannot be used in athletes subjected to drug testing.9-11

Angiotensin-Converting Enzyme Inhibitors Angiotensin-converting enzyme (ACE) inhibitors are an excellent choice for managing hypertension in athletes because they do not diminish exercise performance, they

do not deplete electrolytes or reduce intravascular volume, and their use is not restricted by athletic governing ­bodies. These agents block the conversion of angiotensin I to angiotensin II by inhibiting ACE. ACE inhibitors have a more pronounced effect on blood pressure when combined with a thiazide diuretic19 but can be used alone. A dry, hacking cough is the most common adverse effect of ACE inhibitors. Other side effects include angioedema, which is markedly less common but more serious and can be life threatening. Both are mediated by bradykinin,20 an effect not seen with angiotensin receptor blockers (ARBs) because ARBs do not increase bradykinin levels. ACE inhibitors and ARBs can produce hyperkalemia by reducing aldosterone production, but this is an unusual side effect in patients with normal renal function. ACE inhibitors are contraindicated in pregnant women because they increase the risk for congenital anomalies when used in pregnancy.21 Sexually active women should use contraception when taking ACE inhibitors. ACE inhibitors are our first choice for hypertensive treatment in athletes.

Angiotensin Receptor Blockers ARBs have similar effects on hypertension as ACE inhibitors but do not increase bradykinin and therefore do not induce cough. They are an excellent choice for athletes who have had cough or angioedema during ACE inhibitor treatment. The contraindications for ARBs are similar to those for ACE inhibitors.

β-Adrenergic Blockers β-Adrenergic blockers should generally not be used in treating athletes unless there is some other compelling indication for their use.22 They are not highly effective in reducing blood pressure. They can reduce performance anxiety and heart rate and are therefore banned by the United States Olympic Committee for use in athletes participating in archery, shooting, diving, and iceskating.9-11 β-Blockers can also cause fatigue and depression and can decrease exercise capacity when compared with ACE inhibitors. In addition, β-blockers can exacerbate reactive airways disease, which occurs frequently among athletes.

160

DeLee & Drez’s Orthopaedic Sports Medicine

  Table 3B-4  Commonly Used Antihypertensive Drugs and Their Usual Doses Drug (Trade Name)

Usual Dose Range (mg/d)

Usual Frequency (per day)

Thiazide Diuretics

Chlorothiazide (Diuril) Chlorthalidone (generic) Hydrochlorothiazide (Microzide, HydroDIURIL) Polythiazide (Renese) Indapamide (Lozol) Metolazone (Mykrox) Metolazone (Zaroxolyn)

125-500 12.5-25 12.5-50 2-4 1.25-2.5 0.5-1.0 2.5-5

1-2 1 1 1 1 1 1

25-100 5-20 2.5-10 50-100 50-100 40-120 60-180

1 1 1 1-2 1 1 1

200-800 10-40 10-40

2 1 2

200-800

2

10-40 25-100 5-40 10-40 10-40 7.5-30 4-8 10-80 2.5-20 1-4

1 2 1-2 1 1 1 1 1 1 1

8-32 400-800 150-300 25-100 20-40 20-80 80-320

1 1-2 1 1-2 1 1 1-2

180-420 120-540 80-320 120-480 120-360

1 1 2 1-2 1

Amlodipine (Norvasc) Felodipine (Plendil) Isradipine (DynaCirc CR) Nicardipine sustained release (Cardene SR) Nifedipine long-acting (Adalat CC, Procardia XL) Nisoldipine (Sular)

2.5-10 2.5-20 2.5-10 60-120 30-60 10-40

1 1 2 2 1 1

α1-Blockers Doxazosin (Cardura) Prazosin (Minipress) Terazosin (Hytrin)

1-16 2-20 1-20

1 2-3 1-2

0.1-0.8 0.1-0.3 250-1000 0.1-0.25 0.5-2

2 1 weekly 2 1 1

25-100 2.5-80

2 1-2

β-Blockers

Atenolol (Tenormin) Betaxolol (Kerlone) Bisoprolol (Zebeta) Metoprolol (Lopressor) Metoprolol extended release (Toprol XL) Nadolol (Corgard) Propranolol long-acting (Inderal LA) β-Blockers with Intrinsic Sympathomimetic Activity

Acebutolol (Sectral) Penbutolol (Levatol) Pindolol (generic)

Combined α- and β-Blockers

Labetalol (Normodyne, Trandate) ACE Inhibitors

Benazepril (Lotensin) Captopril (Capoten) Enalapril (Vasotec) Fosinopril (Monopril) Lisinopril (Prinivil, Zestril) Moexipril (Univasc) Perindopril (Aceon) Quinapril (Accupril) Ramipril (Altace) Trandolapril (Mavik) Angiotensin II Antagonists

Candesartan (Atacand) Eprosartan (Teveten) Irbesartan (Avapro) Losartan (Cozaar) Olmesartan (Benicar) Telmisartan (Micardis) Valsartan (Diovan) Calcium Channel Blockers—Nondihydropyridines

Diltiazem extended release (Cardizem CD, Dilacor XR, Tiazac) Diltiazem extended release (Cardizem LA) Verapamil immediate release (Calan, Isoptin) Verapamil long acting (Calan SR, Isoptin SR) Verapamil (COER, Covera HS, Verelan PM) Calcium Channel Blockers—Dihydropyridines

Central α2-Agonists and Other Centrally Acting Drugs

Clonidine (Catapres) Clonidine patch (Catapres-TTS) Methyldopa (Aldomet) Reserpine (generic) Guanfacine (Tenex) Direct Vasodilators

Hydralazine (Apresoline) Minoxidil (Loniten)

Data from the Seventh Report of Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. JAMA 289:2560-2571. Adapted from the Physicians Desk Reference, 57th ed. Montvale, NJ, Thomson PDR, 2003.

Nonorthopaedic Conditions

Calcium Channel Blockers Calcium channel blockers are divided into the dihydropyridines (DHPs), such as amlodipine, nifedipine, felodipine, and isradipine, and the nondihydropyridines (NDHPs), such as verapamil and diltiazem. Calcium channel blockers decrease calcium concentrations in vascular smooth muscle, thereby producing generalized vasodilation. NDHPs also decrease heart rate and cardiac contractility. Calcium channel blockers do not affect exercise energy metabolism or reduce exercise performance.23 All calcium channel blockers are highly effective antihypertensive agents. The DHPs are generally well tolerated in physically active patients and are a good selection for blood pressure treatment in black athletes, with or without a diuretic. The major adverse effect of calcium channel blockers is fluid retention, so combining these agents with diuretics is preferred, depending on the sport and training situation.

α-Adrenergic Blockers α1-Adrenergic blockers are used infrequently to treat ­hypertension because the α1-blocker doxazosin significantly increased the risk for heart failure compared with the diuretic chlorthalidone (8.1% versus 4.5%; relative risk, 2.04) in the Antihypertensives and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT).23 Nevertheless, α1-blockers may have occasional use in athletes. These agents reduce blood pressure by decreasing peripheral vascular resistance by blocking α1-receptors in arterial smooth muscle. They do not alter energy metabolism during exercise or reduce exercise capacity. Therefore, they have no major effect on training or athletic performance.22 Common adverse effects of α1-blockers are dizziness, headache, and weakness. They can also cause first-dose syncope because of their potent vasodilation ability, so the first dose of any α-blocker should be administered at bedtime to avoid postural hypotension. Dizziness can also occur with increasing the dose. α1-Blockers are commonly used to reduce the symptoms of prostatic hypertrophy in older men and may be considered for use in older hypertensive male athletes with this condition. Because of the ALLHAT results, however, all patients treated with an α-blocker should also be treated with a diuretic.

α-Adrenergic Receptor Agonists α-Agonists, such as clonidine, act centrally and have no major effect on exercise or athletic performance. They frequently cause dry mouth, mild to moderate drowsiness, and impotence, making them useful used only when blood pressure is refractory to other medications. Rebound hypertension can also occur if these drugs are stopped ­suddenly.

Combination Therapy Most antihypertensive medicines, except diuretics, produce a plasma volume expansion that reduces their effectiveness in reducing blood pressure. Consequently, the most effective antihypertensive regimens include a diuretic. Indeed, a common cause of refractory hypertension is failure to

161

reduce plasma volume with a diuretic. Diuretics should be avoided in athletes subjected to in- or out-of-competition testing, however, without clear, written permission from the athlete’s sport’s governing body. They should also be used judiciously in athletes subjected to volume depletion or to heat stress.

RECOMMENDATIONS FOR ATHLETIC PARTICIPATION IN HYPERTENSIVE ATHLETES The Hypertension Task Force (Task Force 5) of the 36th Bethesda Conference recommended that athletes have their blood pressure monitored before competition. ­Athletes with stage 1 hypertension should undergo ­echocardiography. If they have left ventricular hypertrophy (LVH) by echocardiography, they should be restricted from competition until their pressure is controlled. Athletes with stage 1 hypertension without LVH or other evidence of end-organ damage can participate in athletics without restriction. Their pressure should be monitored to ­evaluate the effect of exercise training. In contrast, patients with stage 2 hypertension should be restricted from sports with a high static or isometric component, such as weightlifting or weightlifting in training, even if there is no evidence of target-organ damage, until their pressure is controlled. Hypertensive athletes with other cardiac disease should follow the guidelines for athletic competition for their other heart disease. Any medication used by an athlete should be registered with his or her sport’s governing body to obtain a medical exemption.24

C

r i t i c a l

l Hypertension

P

o i n t s

is the most common cardiovascular problem in athletes, although the prevalence of hypertension among athletes is about 50% lower than that in the general population. l Hypertension can be classified into primary, or ­idiopathic, and secondary hypertension. l Hypertension should not be diagnosed cavalierly in young athletes because such a diagnosis can have ­emotional, selfimage, and insurability implications. l Nonpharmacologic therapy is the first step in managing hypertension in all patient groups. l Weight loss, alcohol restriction, aerobic exercise, and special diets reduce blood pressure levels. Among pharmacologic agents, ACE inhibitors are an excellent choice of medications in athletes. Other medications include ­calcium channel blockers, ARBs, α-blockers, β-blockers, and α-agonists. l Athletes with stage 1 hypertension without LVH or other evidence of end-organ damage can participate in athletics without restriction. Their pressure should be monitored to evaluate the effect of exercise training. In contrast, patients with stage 2 hypertension should be restricted from sports with a high static or isometric component, such as weightlifting or weightlifting in training, even if there is no evidence of target-organ damage, until their pressure is controlled.

162

DeLee & Drez’s Orthopaedic Sports Medicine

S U G G E S T E D

R E A D I N G S

Curb JD, Pressel SL, Cutler JA, et al: Effect of diuretic-based antihypertensive treatment on cardiovascular disease risk in older diabetic patients with isolated systolic hypertension. Systolic Hypertension in the Elderly Program Cooperative Research Group. JAMA 276:1886-1892, 1996. Pickering TG, Hall JE, Appel LJ, et al: Recommendations for blood pressure measurement in humans and experimental animals. Part 1: Blood pressure measurements in humans: An AHA scientific statement from the Council of High Blood Pressure Research, Professional, and Publication Subcommittee. Hypertension 45:142-161, 2005. Sacks FM, Svetkey LP, Vollmer WM, et al: Effects on blood pressure of reduced dietary sodium and Dietary Approaches to Stop Hypertension (DASH) diet. DASH-Sodium Collaborative Research Group. N Engl J Med 344:3-10, 2001.

Seventh Report of Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. JAMA 289:2560-2571,2003. World Health Report 2002: Reducing Risks, Promoting Healthy Life. Geneva, Switzerland, World Health Organization, 2002. Available at http://www.Who. int/whr/2002/.

R eferences Please see www.expertconsult.com

S ect i o n

C

Sudden Death in Athletes: Causes, Screening Strategies, Use of Participation Guidelines, and Treatment of Episodes Christine Lawless and Thomas Best

In the United States, it has been estimated that the risk for sudden death occurring during athletics is about 1 in 200,000 athletes per year.1 Because many of these events go unreported, this figure may be an underestimate of the actual number; but, even if that figure were doubled or tripled, it would still represent a very small segment of the population of young athletes. Nonetheless, such events are devastating for the staff involved with an athlete suffering such an event. The media tends to cover these events extensively because the public is keenly interested in what caused the event, why the athlete was allowed to participate with an underlying heart condition, and why the underlying heart condition had not been previously detected when the athlete underwent preparticipation physical examination (PPE) at the beginning of the season.

This chapter is devoted to the issues raised by such sudden events, and includes:

1. What causes sudden death in young athletes? 2. How efficacious are current preparticipation screening strategies for detecting the causes of sudden death in athletes? 3. What techniques might be used to prevent at least some of these episodes from occurring? 4. If a sudden death event occurs, what is the likelihood of resuscitating an athlete from such an event, and does the possibility of success warrant the placement of ­automatic external defibrillators (AEDs) at sporting events?

DEFINITION AND CAUSES There have been many highly publicized instances of sudden death due to a variety of causes in young competitive athletes. Among the more notable were basketball players

Reggie Lewis and Hank Gathers, professional soccer player Marc Vivian-Foe, volleyball athlete Flo Hyman, Olympic figure skater Sergie Grinkov, and basketball athlete Pete Maravich. The causes of death for Lewis, ­Gathers, and Vivian-Foe were various forms of myocardial disease.2-5 ­However, Hyman succumbed to Marfan syndrome, a disorder known to predispose to aortic dissection and aortic rupture during athletics,6 whereas both Grinkov and Maravich suffered from previously undetected coronary artery disease. Grinkov carried the gene for a platelet glycoprotein polymorphism known to predispose to acute ­thrombosis, causing him to occlude the left anterior descending ­coronary artery and succumb to sudden cardiac arrest;7 Maravich collapsed while playing a recreational game of basketball, experiencing cardiac arrest due to ischemia from previously undetected anomalous coronary artery.8 On occasion, recreational or ­performance-­enhancing drugs such as cocaine or ephedra may complicate the issue, as do other illnesses such as asthma, heat illness, or trauma.9 There have been a number of articles over the past 20 years describing the causes of sudden death in young ­athletes, military recruits, or exercising individuals (Table 3C-1).9-17 Specific causative factors depend on the age of the population being studied and the sport in which they are engaged. Athletes younger than 35 years are more prone to undetected myocardial disease, whereas those athletes older than 35 years are more prone to atherosclerotic coronary artery disease as the underlying cause of the sudden death episode.10 According to the most recently published update of an ongoing registry of sudden death in young athletes in the United States (Fig. 3C-1),1,15 of 387 athletes who suffered from sudden death, the most common cause is still hypertrophic cardiomyopathy, which accounts for 26%

Nonorthopaedic Conditions

163

  Table 3C-1  Major References for the Causes of Sudden Cardiac Death in Athletes Year

Reference No.

No. of Athletes

Population

Mean Age (yr)

Most common Cause of Death

Country

1982 1988 1990 1991 1995 1996 2003 2004

10 11 12 13 14 9 15 16

12 60 22 34 160 134 387 126

Joggers Young people Young athletes Athletes Athletes Young athletes Young athletes Military recruits

47 22.3 23 14-40 16.9 (males); 16.2 (females) 17 (median) N/A 19 (median)

USA Italy Italy USA USA USA USA USA

2006

17

55

Elite athletes

23.3

CAD ARVC ARVC CAD, HCM HCM HCM HCM Anomalous coronary artery and ­myocarditis ARVC

Italy

ARVC, arrhythmogenic right ventricular cardiomyopathy; CAD, coronary artery disease; HCM, hypertrophic cardiomyopathy.

of the sudden deaths. However, commotio cordis (sudden death caused by ventricular fibrillation due to a sudden blow delivered during the vulnerable period of the cardiac cycle) is increasing in frequency and is now responsible for about 19% of the episodes. Anomalous coronary artery accounts for 14% of the episodes. Although a small number of sudden death episodes in young athletes may be due to heat illness or trauma, in the great majority of instances, these events are due to sudden cardiac arrest (SCA) secondary to ventricular fibrillation (VF). Because the noncardiac causes of sudden death during athletics are covered elsewhere (see Chapters 11, 12, and 15), the remainder of this chapter discusses only the cardiac causes of sudden death. The term sudden death, when used in describing an episode of SCA due to VF, is probably somewhat of a misnomer in the modern era of sideline use of AEDs. Instances of what used to be true sudden death due to SCA from VF, with previously no hope of recovery, can now be treated if high doses of electricity are applied as soon as possible to the athlete. If such treatment is successful, it is possible to restore the athlete to a stable rhythm and hemodynamics. However, it is

crucial to deliver the defibrillation shock as soon as possible because the earlier that the rhythm can be treated with defibrillation, the higher the survival rate from such events.18 Figure 3C-2 represents the cardiac rhythm downloaded from an AED applied to a 17-year-old high school athlete who suffered SCA due to VF while playing dodgeball and was successfully restored to sinus rhythm by the device. SCA due to VF occurs in a variety of sports, but most commonly it occurs in basketball and football players in the United States and soccer players in Europe. There is a maleto-female ratio of 9:1, with 90% of the deaths occurring during training or competition.9 It has been postulated that the reason for the preponderance of rhythm disturbances occurring during practice or competition is that ­diseased hearts are more prone to VF when they are subjected to catecholamine excess, or the severe hemodynamic and metabolic stresses known to occur with intense athletics.19 This theory is supported by epidemiologic research indicating that those athletes with underlying heart disease are 2.5 times more likely to have an episode of sudden death on the playing field than nonathletes with heart disease.20 Figure 3C-1  Causes of sudden death in young athletes in the United States, 2003. ARVD, arrhythmogenic right ventricular dysplasia; AS, aortic stenosis; CAD, coronary artery disease; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; LQTS, long QT syndrome; LVH, left ventricular hypertrophy.  (From Maron BJ: Medical progress: Sudden death in young athletes. N Engl J Med 349:1064-1075, 2003.)

164

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 3C-2  A 17-year-old with history of aortic valvuloplasty at the age of 9 years collapsed while playing dodgeball. An automatic external defibrillator (AED) was applied. Download from the device showed successful defibrillation (arrow) from ventricular fibrillation, followed by asystole, and then restoration of sinus rhythm.   (Electrocardiographic tracing courtesy of Dr. Christine Lawless.)

Hypertrophic Cardiomyopathy Hypertrophic cardiomyopathy, an inherited disease of the myocardium, occurs at a rate of 1:500 in the general adult population with a male-to-female ratio of 1:1.21 The incidence is much lower in the young athletic population, presumably because of intense preparticipation screening efforts.1 The disease is characterized grossly by thickening and hypertrophy of the cardiac muscle, and histologically by myofibrillar disarray and fibrosis (Fig. 3C-3).22 In the past, it was thought that all cases of hypertrophic cardiomyopathy were associated with asymmetrical septal hypertrophy, with or without outflow obstruction, usually identified through two-dimensional echocardiography. However, advances in

A

imaging techniques and genetic testing have revealed various phenotypic forms of this disease, varying from disproportionate hypertrophy of the septum, to symmetrical hypertrophy of the entire left ­ventricle, to isolated hypertrophy of nonseptal segments of the ventricle.22 Hypertrophic cardiomyopathy is a markedly arrhythmogenic substrate, and once SCA due to VF has occurred, it has been suggested that it may be very difficult to defibrillate such a heart, especially given the hormonal, hemodynamic, and fluid and electrolyte demands of vigorous athletic activity.23 An athlete with hypertrophic cardiomyopathy may complain of shortness of breath, chest pain, syncope or dizziness, palpitations, and fatigue or reduced effort tolerance, or might have a positive family history of the disease.22

B

Figure 3C-3  A, Gross appearance of hypertrophic cardiomyopathy. B, Histologic appearance of hypertrophic cardiomyopathy. Note the myocardial fiber disarray.   (Reproduced with permission from Braunwald E: Essential Atlas of Heart Diseases, 3rd ed. Philadelphia, Current Medicine, 2000, pp 156-177.)

Nonorthopaedic Conditions Aortic valve

165

Circ

Rt cor

Figure 3C-4  Swine model of commotio cordis demonstrates that a blow delivered to the chest during the vulnerable period of the cardiac cycle results in ventricular fibrillation.   (Reproduced with permission from Link MS, Wang PJ, Pandian NG, et al: An  experimental model of sudden death due to low-energy chest-wall impact (commotio cordis). N Engl J Med 338:1805-1811, 1998.)

Although more than 400 gene mutations have been identified in hypertrophic cardiomyopathy, certain types are prevalent, accounting for about 50% of the cases. These include the β-myosin heavy chain, myosin-­binding ­protein C, and cardiac troponin T.24 The genetic ­mutations can also be classified into high-risk and low-risk mutations; high-risk mutations appear to be some of the β-myosin heavy chain mutations and troponin T, whereas low-risk mutations include cardiac myosin-binding protein C or α-tropomyosin. Genetic testing is now commercially ­available for about $4500 per test, and insurance carriers may cover it on a variable basis. It is possible for an athlete to be genotype positive, but phenotype negative. This is most likely to occur during childhood and early ­adolescence because gene expression of hypertrophic cardiomyopathy may not occur until after the teenage years.25 Thus, a normal screening test such as electrocardiography or echocardiography done during the first year of high school does not preclude an abnormal study several years later.

Commotio Cordis Commotio cordis is a condition of SCA due to VF that occurs in otherwise normal hearts following a blow to the chest during the vulnerable period of the cardiac cycle. Sports that place athletes at risk for commotio cordis are those that involve a blow to the chest from an opponent, a stick, or a ball; baseball, lacrosse, hockey, and martial arts are the most common.26 Recent studies in an ­experimental swine model (Fig. 3C-4) concluded that episodes of ventricular fibrillation are due to the impact of the ball hitting the chest during the vulnerable period of the cardiac cycle, and that the only effective treatment for this is defibrillation.27 Experimental evidence indicates that the condition is probably related to the consistency of the ball and the pliable chest of the young adolescent.26 The mean age at

Pulmonic valve Figure 3C-5  High-risk coronary anomalies.   (Redrawn from Hauser M: Congenital anomalies of the coronary arteries. Heart 91:1240-1245, 2005.)

which commotio cordis occurs is 13.6 years.26 Although survival is unusual, the Commotio Cordis Registry has found an overall 16% resuscitation rate from these events, but that figure increased to 25% when resuscitation and defibrillation were applied within 3 minutes.26

Coronary Artery Anomalies Coronary artery anomalies account for about 14% of the sudden cardiac deaths in athletes that occur in the United States.9,18 Although there are various types of coronary anomalies, the type that tends to be fatal, or associated with SCA due to VF, is that in which the left main coronary arises from the proximal segment of the right coronary artery or right coronary sinus, then courses between the main trunks of the pulmonary artery and the aorta (Fig. 3C-5).16,28 During exercise, when the great vessels dilate, the anomalous coronary can become compressed between the great vessels, with resultant ischemia. Combined data from the U.S. Sudden Death Registry and the Italian Sudden Death Registry are shown in Figure 3C-6.29 Of 12 young athletes who suffered sudden death, virtually all experienced antemortem symptoms 7 days to 24 months before the sudden death event. Electrocardiograms (ECGs) and stress testing were often normal in this cohort. Thus, one must be diligent in searching for this condition in symptomatic ­athletes because routine tests are often nonrevealing. ­Clinicians should consider other types of imaging such as computed tomography (CT) angiography or cardiac ­magnetic resonance angiography (MRA) if this diagnosis is ­suspected.

Myocarditis Myocarditis accounts for 5% of sudden death in athletes and is caused generally by an enterovirus, adenovirus, or cytomegalovirus. Other viral pathogens such as EpsteinBarr virus, human immunodeficiency virus, influenza

166

DeLee & Drez’s Orthopaedic Sports Medicine

27 Clinical data

No Clinical data

12

15

Previous symptoms

No symptoms

2

10 7

0

12-lead ECG 9

9

Normal

6

Normal

Maximal exercise ECG

Abnormal 0

Arrhythmogenic Right Ventricular Cardiomyopathy

2

Abnormal

6

Clinical diagnosis & sports disqualification 0 Figure 3C-6  Flow chart showing clinical data available and findings in a study group of young athletes who died of coronary artery anomalies. The majority of athletes had symptoms before death, but stress testing was generally normal. ECG, electrocardiogram.   (Redrawn from Basso C, Maron BJ, Corrado D, Thiene G: Clinical profile of congenital coronary artery anomalies with origin from the wrong aortic sinus leading to sudden death in young competitive athletes. J Am Coll Cardiol 35:1493-1501, 2000.)

A

A and B, and spirochetes and bacteria can be causative, as are recreational drugs such as cocaine.30 Acute myocarditis can be seen in up to 20% of young military recruits who die suddenly.16 There are scant reports of myocarditis occurring in athletes, with the majority of reports being postmortem series of young people who have suffered sudden cardiac death. Figure 3C-7 is taken from the clinical data of a 24-year-old cyclist who presented with a 10-day ­history of chest pain, dyspnea, and fever. The ECG showed ST segment elevation, and the cardiac biopsy was consistent with acute myocarditis.

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a rare fibrofatty infiltration of the myocardium (Fig. 3C-8). On occasion there can be myocarditis associated with this. It accounts for 2.8% of the sudden deaths in athletes in the United States, but a similar analysis in Europe results in a much higher figure, up to 22%.17 The extensive preparticipation cardiovascular Italian screening program has been able to eliminate causes due to other conditions such as hypertrophic cardiomyopathy, thus skewing the data toward causes like ARVC.17 Familial occurrence is seen in 30% to 50% of patients, and although the ­endomyocardial biopsy is the gold standard, it can be negative in 67% of ­cases.31 The genetics of this condition have been defined, and it is thought to be due a defect in the desmoplakin gene.31

Long QT Syndrome Long QT syndrome is an inheritable defect of the ­cardiac sodium channels, and it can be diagnosed in many instances by simple 12-lead ECG (Fig. 3C-9). Long QT occurs in at least seven different varieties, long QT1 through QT7, each having its own characteristic clinical, epidemiologic, and genetic features.32 The different types of long QT ­syndrome can be triggered by different stimuli. For instance, long QT1 syndrome is triggered by swimming; whereas long QT2 is triggered by auditory or emotional

B

Figure 3C-7  Acute myocarditis in a 24-year-old cyclist presenting with 10-day history of chest pain. Note the ST elevation in the inferolateral leads on the 12-lead electrocardiogram (A) and the lymphocytic infiltrate in the myocardium on the myocardial biopsy (B).   (A, Courtesy of Dr. Christine Lawless. B, Courtesy of Dr. Mark Halushka.)

Nonorthopaedic Conditions

A

167

B

Figure 3C-8  A, Transmural fibrofatty infiltration of the right ventricular free wall in arrhythmogenic right ventricular cardiomyopathy (ARVC). B, Cardiac magnetic resonance imaging shows that the right ventricle is dilated with bright signals from the thinned and fatty infiltrated right ventricular free wall.   (Reproduced with permission from Braunwald E: Essential Atlas of Heart Diseases, 3rd ed. Philadelphia, Current Medicine, 2000, pp 156-177.)

triggers such as loud alarm clocks or experiencing an intense emotion.33 The diagnosis can be suspected by the surface ECG, especially if the corrected QT interval is greater than 440 msec in a male or greater than 460 msec in a female.34 To make a proper diagnosis, QT interval on the resting ECG must be corrected for heart rate by dividing the QT interval from the beginning of the Q wave to the end of the T wave by the square root of the R-R interval (Bazett’s formula). The sensitivity and specificity of this formula is not known in athletes, in whom high vagal tone and tendency toward a longer QT interval of about 20 msec compared with nonathletic controls may result in a high false-positive rate. Genetic testing is commercially available for long QT syndrome, and in some instances, it has been conducted postmortem (“molecular autopsy”).35 Other less common causes of sudden death in young athletes are aortic stenosis, mitral valve prolapse, and ruptured aorta, usually due to Marfan syndrome or other collagen vascular disorders. Recent publications have suggested that there appears to be an increased number of aortic ruptures occurring during heavy weightlifting, but

that not all result in death. In one study of 33 cases of ruptured aorta, the mean diameter of the aorta was 4.7 cm, a diameter less than what is usually thought to predispose to rupture of the ­aorta.36 Although highly speculative, this suggests the amount of stress that it takes to rupture the aorta during sports like weightlifting is less than the usual amount required to do so in nonweightlifters. Further study is necessary to determine whether this is a trend, or whether participation guidelines require modification.

EFFICACY OF CURRENT SCREENING STRATEGIES Preparticipation History and Physical Examination Attempts to reduce or eliminate SCA from occurring in young athletes have led to intense screening efforts to detect the underlying cardiac conditions, with the primary screening tool being the PPE. When conducting the PPE, it is

25mm/s 10mm/mV 150Hz 005C 12SL 237 CID: 10

Figure 3C-9  Electrocardiogram of long QT syndrome.   (Electrocardiographic tracing courtesy of Dr. Raul Weiss.)

168

DeLee & Drez’s Orthopaedic Sports Medicine

important to include the standard cardiac screening questions recommended by the American Heart Association (AHA) and to ensure follow-up on any positive answers to the questions or abnormalities on the ­physical ­examination.37 Symptoms of chest pain, shortness of breath, syncope, dizziness, and palpitations require careful thought and evaluation by the team physician. Although lack of standardization of the PPE and differences in methodology in various ­published studies affect objective assessment of the value of the PPE alone as a screening tool, when the PPE is compared with ECG, echocardiography, or postmortem examination, its sensitivity to detect underlying cardiac disease appears to be quite low, in the range of 2.5% to 6%.9,38

Electrocardiography Given the limitations of the PPE, some authors and professional groups have recommended that ECG, echocardiography, or both be added to routine PPE to enhance its ability to detect disease.39,40 Cardiovascular screening with ECG alone appears to increase sensitivity of the screening examination to at least 50%.41 This appears to be an effective strategy, based on the epidemiologic data obtained from countries where such screening programs have been in place long enough to determine their effects on the incidence of sudden death in athletes. In Italy, for instance, where longstanding mandatory cardiac screening of elite athletes has been in place for more than 25 years, the incidence of sudden death appears to have decreased by 89%, whereas the epidemiology has shifted toward conditions that are more difficult to diagnose by standard cardiac ­testing.17 In 2004, the International Olympic Committee ­recommended that an ECG be performed on all elite athletes before Olympic sports participation,39 and in 2005, the European Society of Cardiology recommended implementation of a common European ECG-based screening protocol.40 Concurrently, the authors of the AHA and American College of Cardiology 36th Bethesda Guidelines for Sports Participation concluded that ECGs are a practical and cost-effective alternative to the routine use of echocardiography in preparticipation screening.19 They speculated that the ECG will be 75% to 95% sensitive in detecting hypertrophic cardiomyopathy, one of the common causes of sudden death in athletes. However, a similar expert panel convened by the AHA issued a recent clarification in 2007, stating that ECG-based screening protocols are currently not recommended in the United States primarily because of the cost of conducting such screening in such a large number of eligible athletes, lack of a randomized trial demonstrating clear superiority of the ECG over a standard PPE and use of the 12 AHA screening questions, lack of standardization for interpretation of ECGs in athletes, and lack of normative data in certain demographic and ethnic groups.37

Echocardiography Because ECGs are not 100% sensitive, some advocate the addition of echocardiography to the screening. However, there are many problematic issues inherent in this approach. Echocardiography requires special equipment and training, is less portable, and is more costly than ECG. Some would argue that there are quality concerns because it is not

likely that accredited echocardiographers would perform or interpret all the screening tests. Given the number of athletes in the United States, widespread use of this technique does not seem practical. Nonetheless, there are some who advocate this approach. Some collegiate programs have adopted an abbreviated echocardiogram for incoming athletes the first year they join their respective programs. Such abbreviated echocardiograms tend not to be complete studies, but are performed and interpreted by accredited laboratories and physicians, and screen for the major causes of sudden cardiac death in athletes. In a ­survey conducted in 2005 among 122 North American professional sports teams including Major League Baseball, National Hockey League, and National Football League, 13% performed preparticipation echocardiography.42 There are a number of charitable organizations whose members are parents who have lost their teenage athletes from sudden death during athletics. These groups promote echocardiographic screening of high school athletes and go as far as to advocate training of nonprofessionals in performance of inexpensive “screening” echocardiography.43 The quality of such screening programs and the sensitivity and specificity of the echocardiogram in this model have not yet been validated.

Practical Cardiovascular Screening Strategy To summarize, the routine PPE does not appear to be very sensitive in its ability to detect the underlying causes of sudden cardiac death in athletes. Echocardiography is probably the most practical gold standard for detection of hypertrophic cardiomyopathy, but because of the limitations described previously, widespread use of this technique is not likely to occur. Although clearly not advocated by consensus panels in the United States, if an athletic organization were to consider a cardiac screening p ­ rogram for its athletes, the simple 12-lead ECG appears to offer the most cost-effective way to increase the sensitivity of PPE. There are several excellent references on the use of the ECG for screening large populations of athletes for heart disease. Table 3C-2 summarizes the findings of the Italian researchers.41 They divided the ECG patterns seen in athletes into three categories: normal, mildly abnormal, and distinctly abnormal. The categories were based on degree of left ventricular hypertrophy and the presence or absence of Q waves, abnormal T waves, bundle branch blocks, atrial hypertrophy, and preexcitation (Wolf-Parkinson-White) patterns. About 60% (603/1005) of athletes demonstrated a normal ECG pattern, 26% of athletes (257/1005) demonstrated a mildly abnormal ECG, and 14% (145/1005) showed a distinctly abnormal ECG pattern. The combined power of the mildly abnormal and distinctly abnormal ECG (seen in about 40% of athletes) showed a sensitivity of 51%, specificity of 61%, positive predictive accuracy of 7%, and a very high negative predictive accuracy of 96%. All three types of ECG patterns described previously varied ­according to gender and sport; the abnormal ECG patterns were more likely to be seen in males compared with females and were more likely to be present in endurance sports such as cycling, cross-country skiing, tennis, ­canoeing, and basketball. The ECG does have its limitations. There is currently no standard for ECG interpretation in the athlete. Thus, the ECG is subject to marked variation in interpretation by

Nonorthopaedic Conditions

169

  Table 3C-2  Classification Scheme for Electrocardiograms in Athletes R or S Q waves T waves RBBB LBBB Axis WPW P-R interval ↑ Atrial size % With heart disease

Normal (n = 603)

Mildly Abnormal (n = 257)

Distinctly Abnormal (n = 145)

25-29 mm None Not included IRBBB Absent NL Absent ≤ 0.20 sec Absent 4%

30-34 mm, abnormal progression 2-3 mm, 2 leads Flat, tall, mildly inverted, 2 leads Can be present Absent NL Absent ≤ 0.12 sec Can be present 5%

≥ 35mm ≥ 4 mm, 2 leads Inverted ≥ 2 mm, 2 leads Can be present Can be present ≤−30° or ≥110° Can be present Not included Not included 10%

IRBBB, incomplete right bundle branch block; LBBB, left bundle branch block, RBBB, right bundle branch block; WPW, Wolff-Parkinson-White syndrome. Adapted from Pelliccia A, Maron BJ, Culasso F, et al: Clinical significance of abnormal electrocardiographic patterns in trained athletes. Circulation 102(3):278-284, 2000.

cardiologists and others involved in athlete care. It is not 100% sensitive for the detection of underlying heart disease; hence, the possibility of false-negative results exists. False-positive results exist as well because normal athletic adaptation to exercise can result in marked alterations of the surface ECG, resulting in delays because of the time it takes to follow up and perform further cardiac evaluation. Finally, although it appears to be a p ­ ractical, cost-effective alternative in most clinical settings, expert consensus panels in the United States currently do not recommend its widespread adoption as a screening method in athletes.37

PREVENTION OF SUDDEN DEATH AND SUDDEN CARDIAC ARREST DUE TO VENTRICULAR FIBRILLATION Team physicians can optimize their ability to prevent sudden death episodes due to SCA and VF in athletes by several strategies. Such strategies include the following: 1. Thoughtful execution of the PPE, with, at the very least, use of cardiac screening questions based on the 12-point AHA recommendations 2. Addition of cardiovascular screening in the form of an ECG 3. Evaluation of athletes with symptoms such as chest pain, shortness of breath, syncope, palpitations, and ­dizziness 4. Meticulous use of the 36th Bethesda guidelines for participation in sports, and adherence to the guidelines even when there is pressure to allow participation 5. Use of protective gear or modification of playing equipment, for example, softer baseballs to reduce the ­likelihood of commotio cordis44 6. Availability of AEDs at training sessions and during competition, especially for high-risk sports, such as football, basketball, hockey, lacrosse, martial arts, and baseball

Bethesda Guidelines for Participation in Sports Once the PPE has been conducted, the AHA questions have been asked, and any positive questions (symptoms or family history) have been addressed in the athlete

(as ­discussed in the previous section), team physicians have access to multiple sets of published guidelines for sports participation. The 36th Bethesda Conference for Sports Participation is based on consensus opinion from a panel of experts, primarily members of the American College of Cardiology and the AHA.19 The report from the most recent panel was published in April 2005, and recommendations for the most common causes of sudden death in athletes are summarized in Table 3C-3. Guidelines are divided into several task forces, including Preparticipation Screening, Congenital Heart Disease, Valvular Heart Disease, Cardiomyopathy, Marfan syndrome, Hypertension, Coronary Disease, Arrhythmias, Classification of Sports, Drugs and ­Performance-­Enhancing Substances, Use of the Defibrillator and the Implanted Defibrillator, Commotio Cordis, and Legal Aspects. The most recent panel ­significantly updated the guidelines in that they ­incorporated newer techniques such as genetic testing into the participation algorithm, and they addressed evolving treatments for cardiac disease and their impact on participation and return-to-play decisions. Treatments such as defibrillators, electrophysiologic ablation, alcohol septal ablation and cardiac surgery for hypertrophic cardiomy­ opathy, and β-blockers were addressed, particularly as all these pertain to the prevention of sudden cardiac death or SCA during athletics. The writers also discussed the legal implications of the document, citing that use of the 1996 guidelines to guide legal decision making in a court case in the 1990s set an important precedent regarding use of the guidelines in a court of law and the legal risk for deviation from the guidelines.19,45

Implanted Defibrillators Although implantable defibrillators (ICDs) in general are very effective in preventing sudden death due to SCA in certain cardiac diagnoses including hypertrophic cardiomyopathy,46 they have not been tested under the conditions of rigorous sports, in which fluid shifts, electrolyte abnormalities, and catecholamine excess may alter defibrillation thresholds.19 Thus, they cannot be depended on to reliably defibrillate under these conditions. The Bethesda writers are very clear that presence of an ICD does not change their recommendation that athletes with hypertrophic cardiomyopathy should not play vigorous dynamic sports even if an ICD is implanted.

170

DeLee & Drez’s Orthopaedic Sports Medicine

  Table 3C-3  Summary of 36th Bethesda Guideline Recommendations for Sports Participation for Athletes with Underlying Heart Disease Known to Predispose to Sudden Death HCM

Anomalous Coronary

ARVC

DCM

Long QT syndrome

Marfan Syndrome

Participation in all sports allowed

No

No

No

No

No

Participation allowed if genotype positive, phenotype negative

Yes

N/A

Yes

N/A

Participation allowed after ­corrective surgery

No

Yes

N/A

Low-intensity sports only

Participation allowed with ICD Participation allowed with β-blockers

No

No

No

Yes, after heart ­transplantation, ­provided no coronary luminal narrowing or ischemia No

Yes, but no ­swimming ­allowed for long QT1 N/A

With certain restrictions, depending on size of aorta (≤40 mm), absence of family history of ­dissection, and absence of significant valve disease Not specified

No

N/A

No

No

No

No

No

Not clearly addressed, but if LVEF has normalized, can consider

ARVC, arrhythmogenic right ventricular cardiomyopathy; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; ICD, implantable cardioverter-­defibrillator; LVEF, left ventricular ejection fraction.

­ onetheless, despite the Bethesda authors’ recommendaN tion, there are survey data from physicians indicating that up to 70% of athletes with implanted ICDs are participating in sports,47,48 despite what the guidelines recommend. The National Sports Safety ICD Registry has been established to determine outcomes of sports participation in athletes who ­participate with an ICD,49 but at the moment, ICDs are not a proven strategy to prevent sudden cardiac death d ­ uring athletics.

Ablations Athletes with Wolff-Parkinson-White (WPW) syndrome, symptomatic with palpitations, dizziness, or syncope, who wish to participate in sports require electrophysiologic testing, definition of the accessory pathways, and treatment with radiofrequency ablation. For those with WPW syndrome detected by ECG but without symptoms, treatments are a little more controversial, but studies suggest that sudden death in athletes occurs in those with a short refractory period of the accessory pathway.50 Therefore, in this instance, an athlete will need electrophysiologic testing to determine the refractoriness of the pathway, with subsequent ablation for those with short refractory ­periods, and return to play 2 to 4 weeks after ablation.19

Septal Ablation or Surgery for Hypertrophic Cardiomyopathy The writers of the Bethesda guidelines are clear in their recommendation that sports participation is not allowed for those with hypertrophic cardiomyopathy, regardless of presence or absence of outflow obstruction, or treatment with alcohol septal ablation or cardiac surgery. Because there are no demonstrable outcome data indicating that risk for ventricular fibrillation is decreased during ­athletics

for any of these treatments, they cannot be assumed to reduce the likelihood of a sudden death episode.

β-Adrenergic Blockers Team physicians may be tempted to allow play if an athlete with hypertrophic cardiomyopathy or long QT syndrome is taking β-blockers to reduce the incidence of sudden death episodes. Although β-blockade did appear to reduce the incidence of such episodes in a nonathletic population,51 the effectiveness of β-blockers has not been proved under the conditions of vigorous sports. If anything, the amount of β-blockade required to prevent sudden death episodes would most likely have a dramatic impact on maximal heart rate and athletic performance. Thus, use of β-­blockers as a preventive strategy does not appear to be practical. Additionally, the Bethesda writers are quite clear in their recommendations for long QT, stating, “No ­participation is recommended if syncope is present in ­athletes who survive cardiac arrest regardless of their ICD or beta-blocker ­status.”19

Use of Protective Gear or Modified Playing Equipment One of the more striking examples of use of sports medicine research to reduce the incidence of sudden death during athletics through use of modification of playing equipment has come from the research conducted in the swine model of commotio cordis. Because those most likely to experience this condition are adolescents whose mean age is 13 years, it has been postulated that their pliable chests more readily allow a blow delivered during the vulnerable period of the cardiac cycle to cause an episode of VF. Through a series of experiments conducted in the swine model, it has been demonstrated that chest protectors are

Nonorthopaedic Conditions

not likely to be protective, but use of a softer baseball is more likely to have the desired effect.44

TREATMENT OF SUDDEN DEATH DUE TO SUDDEN CARDIAC ARREST No matter how much preparticipation screening has taken place or how meticulous the team physician has been in evaluating symptoms and following guidelines for disqualification, there will always be a chance of SCA events occurring during athletics. Team physicians will have the greatest chance of successfully resuscitating an athlete from SCA due to VF if defibrillation is readily available. The rationale for use of the AED is clear, and it is recommended that an AED be available at sporting events in a distribution that can achieve a response rate of 5 minutes or less.19 Some team physicians have recommended that athletes with heart disease be considered protected from sudden death if there is an AED available at all practices and competition. However, the presence of an AED at a sporting event should not be construed as absolute ­protection against fatal outcome from a cardiac arrest, especially in light of the outcome data provided next. AEDs were developed and tested during the 1990s and have been widely adopted because of evidence that they provide the most rapid means of defibrillation, that bystanders can easily be instructed on their use through basic life support classes, and that outcomes in the general population have been superior to those obtained by traditional means of resuscitation.52 In nonathletes, studies have demonstrated marked improvement in resuscitation rates from SCA from 5% to as high as 63%.53 However, for the same reasons outlined in the discussion on ICDs previously, the AED cannot be expected to reliably defib­rillate for all diagnoses under the conditions of intense rigorous sport. Scant data are available regarding resuscitation in athletes, but from the limited data available, it appears that resuscitation does take place in a variety of settings from the school-aged and collegiate athlete to the professional and masters athlete, with variable success (Table 3C-4).23,26,54,55 Isolated case reports in the popular media indicate that it is entirely possible to resuscitate some athletes during intense athletic activity. In 2005, a professional hockey

171

player suffered SCA due to VF during a game and was successfully resuscitated by the team physician within 90 seconds through use of a sideline AED.56 Because SCA in athletics does not occur often, it has been difficult to collect meaningful detailed data. However, several recent small surveys and case series have provided some important information regarding resuscitation of the young athlete from an SCA event. The Commotio Cordis Registry found an overall survival rate of 16% in 128 cases analyzed. However, when defibrillation was applied, 17 of 68 athletes (25%) were resuscitated and survived.26 In a cohort analysis of all athletes participating in the Twin Cities and Marine Marathons over a 25-year period, a total of 9 SCA episodes were treated by the marathon physicians providing coverage of the event.54 Although the numbers are quite small, the overall defibrillation rate was 44%, about 33% in the younger athlete and 50% in the older athlete.54 Marathons of this type are typically staffed by highly trained physicians supplied with sophisticated advanced life support equipment who can quickly defibrillate with either manual or automatic external defibrillators. In a survey conducted among members of the American Medical Society for Sports Medicine, 21 cases of SCA in athletes were reported, with an overall resuscitation rate of 66%.55 The rates appeared to be higher in the older athlete in that 81% of older athletes and 50% of young athletes were successfully resuscitated to a hemodynamically stable rhythm. In a smaller survey conducted among athletic trainers at National Collegiate Athletic Association (NCAA) colleges and universities, nine athletes were reported to have experienced sudden death, with seven of nine documented to be due to VF.23 Only one young athlete who received prompt treatment for the VF survived, resulting in an overall resuscitation rate of 14%. However, it is important to note that five of nine of the athletes (55%) in this cohort were found to have an underlying diagnosis of hypertrophic cardiomyopathy, a proportion much higher than what was seen in the other studies of SCA in athletes.9 It is possible that this group, given the preponderance of hypertrophic cardiomyopathy, may have been more difficult to defibrillate from SCA ­episodes. To summarize, the resuscitation rate in the young athlete appears to lie somewhere between 14% and 50%, with an overall rate of about 25%. Although highly speculative, the differences in resuscitation rates appear to be dependent on

  Table 3C-4    Resuscitation in Athletes Investigator

Year

Type of Athlete

Total No. of Athletes

Resuscitation Rate for Young Athlete

Resuscitation Rate for Older Athlete

Maron

2002

Commotio cordis

Overall survival 16%; increased to 17/68 (25%) if defibrillated within 3 minutes

N/A

Roberts

2005

Marathon

1/3 (33%)

3/6 (50%)

Lawless

2006

3/6 (50%)

13/15 (80%)

Drezner

2006

Various sports Various sports

Total cohort =128; various types of ­treatment on the scene 9 cases resuscitated by marathon physicians 21 athletes resuscitated by team physicians 9 NCAA athletes ­resuscitated by athletic trainers

1/7 (14%)

71% in officials, spectators, and nonathletes

NCAA, National Collegiate Athletic Association.

172

DeLee & Drez’s Orthopaedic Sports Medicine

the population being studied, the underlying cardiac diagnosis, the circumstances of the event, the length of time it takes for the defibrillator to be applied, and perhaps the experience of the person performing the defibrillation. Availability of the AED at sporting events ought to be encouraged, but the data are clear in that presence of the AED cannot guarantee absolute protection against death due to SCA. C

r i t i c a l

P

o i n t s

l There

are numerous causes of sudden death in athletes, with the most common being a sudden cardiac arrest (SCA) due to ventricular fibrillation (VF) in an athlete with p ­ reviously undetected heart disease. l The most common cause in the United States is still hypertrophic cardiomyopathy, but other causes appear to be increasing, especially commotio cordis. l Cardiovascular screening in the form of ECG might improve the ability of the PPE to detect the underlying causes of sudden death, but it has not yet been shown by rigorous scientific method to be superior to the standard PPE, and it has not yet been recommended for widespread use by expert consensus panels in the United States. l If a cardiac diagnosis is made through the preparticipation screening process, the team physician is advised to consult existing sports participation guidelines. l Even though every attempt has been made to prevent ­episodes of sudden death due to SCA, and team physicians have followed the Bethesda guidelines meticulously, there will continue to be instances of SCA in athletics. l Because studies have demonstrated that some athletes can be and are resuscitated from these episodes, it is highly recommended that an AED be readily available so that ­defibrillation can be administered within 5 minutes. l By following these recommendations, the athlete is af­ford­ed the best chance of survival from an SCA ­occurrence.

S U G G E S T E D

R E A D I N G S

Basso C, Maron BJ, Corrado D, Thiene G: Clinical profile of congenital coronary artery anomalies with origin from the wrong aortic sinus leading to sudden death in young competitive athletes. J Am Coll Cardiol 35:1493-1501, 2000. Corrado D, Basso C, Pavei A, et al: Trends in sudden cardiovascular death in young competitive athletes after implementation of a preparticipation screening program. JAMA 296:1593-1601, 2006. Hatzaras I, Tranquilli M, Coady M, et al: Weight lifting and aortic dissection: More evidence for a connection. Cardiology 107:103-106, 1007. Hauser M: Congenital anomalies of the coronary arteries. Heart 91:1240-1245, 2005. Kapetanopoulos A, Kluger J, Maron B, et al: The congenital long QT syndrome and implications for young athletes. Med Sci Sports Exerc 38(5):816-825, 2006. Lawless CE, Lampert R, Olshansky B: Sudden cardiac death in athletes: Rates of defibrillation. J Am Coll Cardiol 47:165A, 2006. Maron BJ, Pelliccia A: The heart of trained athletes: Cardiac remodeling and the risks of sports, including sudden death. Circulation 114(15):1633-1644, 2006. Maron BJ, Thompson P, Ackerman M, et al: Recommendations and considerations related to preparticipation screening for cardiovascular abnormalities in competitive athletes: 2007 Update: A scientific statement from the American Heart Association Council on Nutrition, Physical Activity, and Metabolism: Endorsed by the American College of Cardiology Foundation. Circulation 115(12):16431655, 2007. Maron BJ, Zipes DP: 36th Bethesda Conference: Eligibility recommendations for competitive athletes with cardiovascular abnormalities. J Am Coll Cardiol 45:1313-1375, 2005. Pelliccia A, Maron BJ, Culasso F, et al: Clinical significance of abnormal electrocardiographic patterns in trained athletes. Circulation 102(3):278-284, 2000.

R eferences Please see www.expertconsult.com

S ect i o n

D

Diabetes Mellitus Jennifer J. F. McVean and David B. Allen Diabetes mellitus is a group of diseases caused by an a­ bsolute or relative deficiency of insulin; it affects 1 in 400 to 1 in 600 children in the United States younger than 20 years.1 The predominant form of diabetes affecting children is type 1 diabetes mellitus (T1DM), a disorder in which the insulin-producing islet cells of the pancreas are selectively destroyed by the body’s immune system. The resulting absolute insulin deficiency invariably requires treatment with insulin delivered exogenously by injection or a pump and is the hallmark of T1DM. Type 2 diabetes

mellitus (T2DM), a disease of insulin resistance and relative insulin deficiency, is the most common form of diabetes in the United States. It accounts for 90% to 95% of the 20.8 million people in the United States who have diabetes.1 Until recently, T2DM was a disease that occurred during adulthood. However, with the ever-­expanding epidemic of childhood obesity and insulin resistance, the incidence of T2DM in children and adolescents is increasing as well.2 Physical exercise is one of the mainstays of treatment of diabetes mellitus. There are many benefits of exercise

Nonorthopaedic Conditions

Glucose Regulation during Exercise Maintenance of a normal plasma glucose level during exercise depends on a precise balance between fuel mobilization and utilization. The exercising muscle fiber can increase its metabolic rate and production of adenosine triphosphate (ATP) tremendously, with oxidative processes 50 times higher than resting levels and glucose uptake 35 times higher than resting levels.10 Primary sources of energy include (1) fat and carbohydrate present in muscle and (2) glucose released from the liver as a result of glycogenolysis, the breakdown and mobilization of stored glycogen. The relative contributions of these sources vary with the duration and intensity of exercise. When exercise begins, muscle glucose uptake increases, and muscle glycogen is the primary source of energy. After 20 to 30 minutes of activity, energy is derived from a combination of hepaticderived glucose and adipose-derived free fatty acids. With prolonged exercise (60 to 90 minutes), free fatty acids become the principal energy source. Glycogen remains an important fuel, with the depletion of muscle glycogen coinciding with the time of exhaustion.11 Intensity of exercise is measured by the ­percentage of the individual’s maximum oxygen consumption (Vo2max) required for fuel usage. During exercise at high intensity, oxidation of glucose for energy predominates; with less intense exercise, fat usage is preferred. Physical training enables the athlete to perform the same work at a lower percentage of Vo2max and to conserve glucose and improve endurance by using a greater proportion of free fatty acids. Insulin is an anabolic hormone that regulates carbohydrate and fat metabolism. Insulin increases the uptake of glucose into muscle and adipose tissue, suppresses the production of glucose by the liver, increases glycogen ­synthesis,

Muscle Plasma glucose concentration steady

Insulin secretion

FF A

inhibited

ol er

RELEVANT PHYSIOLOGY AND PATHOPHYSIOLOGY

Liver

yc gl

in patients with diabetes, including a greater sense of well-being, help with weight control, improved physical fitness with lower pulse and blood pressure, an improved lipid profile, and improved insulin sensitivity.3 Observations of exercise-associated reductions in blood glucose4 have been substantiated by in vivo and in vitro5 findings of increased insulin sensitivity as a result of enhanced insulin receptor binding after physical training; these effects can be seen after a single exercise session in an inactive individual,6 albeit probably not to the same degree as in a trained individual. Although the value of exercise in improving long-term metabolic control is controversial in T1DM7,8 (beneficial effects of exercise in T2DM are well established), athletic participation by young individuals with T1DM is encouraged to achieve the same health and cardiovascular9 benefits enjoyed by exercising nondiabetic individuals. Management of these athletes requires knowledge of the differences in exercise physiology in diabetes. Careful monitoring of blood glucose and adjustment of insulin doses and nutrition plans will allow the safe and successful participation of patients with diabetes in virtually any athletic activity.

173

Fat Figure 3D-1  The response to exercise in healthy individuals and in insulin-dependent diabetic patients. When plasma insulin is normal or slightly diminished, hepatic glucose production increases markedly, as does skeletal muscle usage of glucose, whereas blood glucose remains unchanged. FFA, free fatty acids.   (From Ekoe JM: Overview of diabetes mellitus and exercise. Med Sci Sports Exerc 21:353-368, 1989. © The American College of Sports Medicine.)

and inhibits fat and protein degradation, encouraging growth and preventing weight loss and tissue breakdown. Decreased levels of insulin lead to an increase in lipolysis or fat breakdown. Thus, the appearance of ketones (products of fat breakdown) in the blood or urine of a diabetic patient signifies marked insulin deficiency. When exercise begins, a number of hormonal responses occur to provide energy to the exercising muscles and maintain blood glucose levels in a normal range. The body decreases its release of insulin and increases its release of glucagon from the pancreas (Fig. 3D-1). This fall in insulin allows for increased hepatic glycogenolysis and facilitates lipolysis and the liberation of free fatty acids and glycerol (used with lactate, pyruvate, and alanine as precursors for gluconeogenesis). The rise in glucagon is required for hepatic glycogenolysis and gluconeogenesis.12 This suppression of endogenous insulin release and increased secretion of hormones, such as glucagon, that oppose the actions of insulin (counter-regulatory hormones), allow hepatic glucose production to increase, satisfying the demands of exercising muscle (Table 3D-1). Exercise stimulates insulin-independent glucose uptake and potentiates insulin action. These actions compensate for declining insulin concentrations and facilitate delivery of substrate to exercising tissues. The exact mechanism by which exercise independently stimulates glucose transport into muscle is unknown. Muscle contraction stimulates muscle glucose uptake in vitro in the absence of insulin.13 There is also some evidence that insulin-independent glucose uptake is related to an increase in plasma membrane glucose ­transporter number.14 Physical training increases the athlete’s sensitivity to insulin,15 an effect that predominates in skeletal muscle rather than liver. An increase in insulin binding16 resulting

DeLee & Drez’s Orthopaedic Sports Medicine

Table 3D-1  Actions of Major Counter-Regulatory

Liver

Muscle

Hormones

Hormone

Mechanism of Hyperglycemic Effect

Glucagon* Epinephrine*

Activates hepatic glycogenolysis and gluconeogenesis Stimulates hepatic glucose production, limits peripheral glucose use, suppresses insulin secretion After initial glucose-lowering effect, limits glucose transport into cells, mobilizes fat, and provides gluconeogenic substrate (glycerol) Initially inhibits glucose use; with time, mobilizes substrate (amino acids and glycerol) for gluconeogenesis

Growth hormone Cortisol

down

Insulin concentration

elevated

yc gl

*Hormones important in recovery from acute hypoglycemia.

Plasma glucose concentration

ol er

from augmentation of insulin receptor number and ­affinity has been shown after short periods of physical training. Reductions in body adipose tissue lead to greater insulin effect after prolonged training. Changes in insulin sensitivity correlate directly with improvements in Vo2max,15 although this may reflect primarily frequency of exercise rather than fitness per se.

Glucose Regulation during Exercise in Athletes with Type 1 Diabetes Mellitus Soon after the implementation of insulin therapy, it was observed that physical activity could reduce the insulin requirements of patients with T1DM and that the decrease in blood glucose after an insulin injection was magnified by subsequent exercise.17 Exercise increases blood flow to muscles and skin, leading to an increased rate of insulin absorption in the athlete with T1DM (Fig. 3D-2).18 This effect is most pronounced when insulin is administered less than 60 minutes before exercise. Because insulin is given exogenously in T1DM, the body cannot decrease its release of insulin, and increased serum insulin levels inhibit hepatic glucose production and peripheral lipolysis. At the same time, continued insulin-independent glucose uptake by exercising muscles depletes energy stores. Glucagon secretion in response to hypoglycemia is usually lost about 5 years after the diagnosis of T1DM; the epinephrine response to hypoglycemia is also attenuated in individuals with T1DM.19 Deficiencies of these counter-regulatory hormones further limit fuel availability during exertion. The balance between energy supply and demand is often disrupted in the diabetic athlete by excessive or inadequate insulin effect.

Consequences of Excessive Insulin Effect during Exercise Suppression of glycogenolysis by excessive insulin action, increased insulin sensitivity, and insulin-independent glucose uptake by working muscles can result in hypoglycemia during, immediately after, or several hours after exercise. Failure to anticipate the occurrence or appropriate duration or intensity of exercise and to make appropriate adjustments in insulin dosage and carbohydrate intake accounts for most instances of hypoglycemia during activity. Most diabetic athletes are aware of the effect of exercise on blood

FF A

174

Fat Figure 3D-2  The response to exercise in hyperinsulinemic insulin-dependent diabetic patients. When plasma insulin is increased, skeletal muscle usage of glucose during exercise increases markedly, but the increase in hepatic glucose production is smaller than normal: blood glucose levels fall. FFA, free fatty acids.   (From Ekoe JM: Overview of diabetes mellitus and exercise. Med Sci Sports Exerc 21:353-368, 1989. © The American College of Sports Medicine.)

glucose and alter their insulin dose and carbohydrate intake accordingly. If exercise is unplanned and insulin adjustments are unable to be made, extra carbohydrate intake is recommended. After exercise, insulin sensitivity is increased, and the body is repleting muscle and liver glycogen stores. If unaccompanied by a reduction in insulin dose or supplementation with extra carbohydrate, these actions may combine to cause postexercise late-onset hypoglycemia. Postexercise late-onset hypoglycemia may occur in the 24 hours following exercise, with the maximal risk occurring 6 to 10 hours after exercise.20 A 2-year prospective case study of 300 children and adolescents with T1DM revealed that 48 (16%) experienced such postexercise late-onset hypoglycemia.21 Distinctive characteristics of this phenomenon are summarized in Box 3D-1. In addition to its delayed occurrence, postexercise late-onset hypoglycemia is distinguished by its severity. Adrenergic symptoms are often absent, and the patient usually manifests neuroglycopenic symptoms such as altered level of consciousness or seizures. Patient age, duration of T1DM, and tightness of metabolic control are unrelated to the likelihood of experiencing Box 3D-1 Clinical Characteristics of Postexercise Late-Onset Hypoglycemia Occurs 6-10 hr after unusually strenuous or prolonged activity (frequently nocturnal) Frequently severe (stupor, coma, and/or seizure) with few warning signs Unrelated to age of patient or duration of diabetes Unrelated to strict metabolic control

Nonorthopaedic Conditions

High-intensity, short-term exercise normally is associated with transient rises in plasma glucose levels, which peak 5 to 15 minutes after exercise is stopped and return to baseline within 40 to 60 minutes.26 Suppression of insulin secretion, stimulation of the sympathetic nervous system with subsequent release of counter-regulatory hormones, and an increase in hepatic glucose production that exceeds the rise in peripheral glucose utilization combine to produce this glycemic response. When exercise is complete, insulin secretion is immediately stimulated by elevated arterial blood glucose levels. Consequently, transiently elevated glucose levels return rapidly to normal.27 In the athlete with T1DM, the postexercise rise in plasma insulin levels cannot occur, and hyperglycemia after intense exertion may be sustained and of greater magnitude.28 This effect is exaggerated when a pre-­exercise insulin deficiency is present and the plasma glucose concentration is elevated (Fig. 3D-3). In an insulin-deficient state, there is increased hepatic glycogenolysis and gluconeogenesis, increased release of counter-regulatory ­hormones, and increased lipolysis. Exercise in this setting leads to further impairment of peripheral glucose uptake and subsequent ­hyperglycemia, ketosis, and acidosis. In addition to prompt elevation in blood glucose concentrations, defective peripheral clearance of ketones leads to a rapid worsening of an already compromised metabolic state.29

CLINICAL EVALUATION Athletes with T1DM with good blood glucose control and no evidence of complications may engage in all levels of physical activity from recreational sports to ­competitive professional athletics.30 Children and adolescents with T1DM should adhere to the CDC and American Academy

Muscle Plasma glucose concentration up

Insulin concentration

FF A

diminished

ol er

Consequences of Insufficient Insulin Effect during Exercise

Liver

yc gl

late-onset hypoglycemia. Consuming carbohydrate within 30 minutes of exercise allows for more efficient glycogen repletion and decreases the risk for postexercise late-onset hypoglycemia.22 Diabetic athletes making the transition from an untrained to a trained state are more likely to experience delayed hypoglycemia, and it is recommended that frequent blood glucose monitoring be undertaken in the 12-hour postexercise period at the onset of a new sports season.3 The effects of acute exercise on glucose metabolism persist for several hours. Enhanced glucose uptake and glycogen synthesis by muscle,23 increased sensitivity of muscle to insulin,24 and increased binding of insulin to monocytes25 combine to improve glucose tolerance in untrained subjects for at least 18 hours after exercise.6 This increased sensitivity to insulin, combined with the avid extraction of glucose from the circulation for the repletion of muscle and liver glycogen stores, explains the occurrence of nocturnal postexercise late-onset hypoglycemia. Diabetic athletes with diminished glucagon and epinephrine responses to hypoglycemia are at increased risk for experiencing delayed hypoglycemia. Deficient counter-regulation contributes to the lack of warning signs and increased severity of postexercise late-onset hypoglycemia.

175

Fat Figure 3D-3  The response to exercise in insulin-deficient insulin-dependent diabetic patients. When plasma insulin is diminished markedly, hepatic glucose production is increased markedly during exercise, but the increase in skeletal muscle usage of glucose is smaller than under normal circumstances: blood glucose levels rise. FFA, free fatty acids.   (From Ekoe JM: Overview of diabetes mellitus and exercise. Med Sci Sports Exerc 21:353-368, 1989. © The American College of Sports Medicine.)

of Sports Medicine recommendations for a minimum of 30-60 minutes of moderate physical activity daily.3 Before beginning an exercise program, athletes with diabetes should undergo a thorough medical history and physical examination. The physician should ensure that the athlete is committed to frequent blood glucose monitoring and follow-up with the health care team and is comfortable with alterations in insulin and carbohydrate intake based on activity (Box 3D-2). The athlete’s awareness and knowledge of the likelihood of hypoglycemia related to exercise should be explored, and strategies for prevention (presented later) should be discussed. Because alterations in insulin sensitivity and episodes of hypoglycemia are observed more frequently during the transition from the untrained to the trained state, the individual’s current level of physical activity and fitness should determine the magnitude of changes made to the athlete’s insulin and nutritional regimen. The diabetic athlete should be questioned about his or her ability to recognize hypoglycemia. Lack of warning signs of hypoglycemia indicates an attenuation of the glucagon and epinephrine responses and increases the probability of occurrence of potentially dangerous hypoglycemia during and after exercise. Diabetic patients are predisposed to the development of coronary heart disease (CHD). Sixty-five percent of deaths in patients with diabetes are due to CHD or stroke.1 Elevations in serum total and low-density lipoprotein cholesterol and triglycerides and reductions in high-density lipoprotein cholesterol (together constituting a high-risk lipid profile) are seen frequently in patients with T2DM or in patients with T1DM when diabetes management is suboptimal. A detailed family history of premature CHD, stroke, or hypertension should be elicited and, if positive, should prompt a thorough evaluation of the athlete’s lipoprotein profile. Detection of

176

DeLee & Drez’s Orthopaedic Sports Medicine

Box 3D-2 Preparticipation Evaluation of the Diabetic Athlete    I. Has a routine of stable blood glucose control been established? Requirements for the athlete:    A. Recent satisfactory measurement of Hemoglobin A1c    B. Habitual (4-6 times/day) self-monitoring of blood glucose levels    C. Understanding of indications for and interpretation of urine ketone measurement  II. Has the possibility of hypoglycemia been ­anticipated? Requirements for the athlete:    A. Recognition of early warning signs of hypoglycemia    B. Knowledge of treatment strategies for mild hypoglycemia (e.g., glucose gel, glucose tablets, Gatorade)    C. Knowledge of insulin activity and ability to change insulin dose as needed for extra physical activity    D. Medic-Alert bracelet or necklace    E. Provision of glucagon (1 mg for intramuscular injection by trained personnel) for treatment of severe hypoglycemia III. Are complications of diabetes present? Requirements for the athlete:    A. Recent evaluation of blood pressure, neurologic function, joint mobility, and skin condition    B. Retinal examination by ophthalmologist in the past year    C. Screening laboratory evaluation for blood lipid abnormalities and diabetic nephropathy

familial lipid disorders and marked abnormalities in serum total cholesterol levels (e.g., >300 mg/dL) should not interfere with sports participation but should focus attention on the need for improvement in blood glucose control and referral for additional lipid-lowering therapy. Because significant complications of diabetes are extre­mely unlikely during childhood or adolescence for athletes with T1DM, participation in virtually any sport is possible for the diabetic child. It is important to remember that adolescents with T2DM may have complications of the disease that are present at diagnosis. A careful physical examination focusing on detecting any evidence of retinopathy, nephropathy, peripheral or autonomic neuropathy, or musculoskeletal impairment is important for all patients with diabetes. Particular attention should be directed toward individuals beginning a fitness program who have poor control or longstanding diabetes. Patients with active proliferative diabetic retinopathy may develop vitreous hemorrhage or retinal detachment when blood pressure is transiently elevated during vigorous exertion. Athletic activities requiring heavy lifting or straining should be avoided, whereas low-impact cardiovascular conditioning activities are preferred. Urinary albumin excretion, one of the manifestations of diabetic nephropathy, can become worse with exercise.31 This urinary albumin excretion may represent merely a transient hemodynamic response, and it has not been shown that ­exercise has any deleterious effect on the progression of renal

disease. There are no specific physical activity recommendations for patients with either microalbuminuria or overt diabetic nephropathy.30 Conversely, hypertension is known to accelerate the ­progression of diabetic nephropathy and should be controlled. Peripheral neuropathy may result in the loss of sensation in the feet. Even minor degrees of anesthesia predispose the athlete to soft tissue and joint injuries and require the limitation of weight-bearing exercise. Autonomic neuropathy may lead to a decreased cardiovascular response to exercise,32 a lower maximal aerobic capacity, and an increased risk for a cardiovascular event during physical activity. Limited joint mobility, a sign of poor metabolic control and increased risk for microvascular disease,33 can impair performance and increase the risk for injury. Although participation in specific sports rarely is ­prescribed, not all sports activities are equally beneficial to the prevention of long-term diabetic complications. ­Longevity and freedom from vascular and neurologic complications are enhanced by the attainment of ideal body weight and increased lean body mass (which increases sensitivity to insulin), good cardiovascular fitness (deceasing the risk for CHD) and avoidance of hypertension (which, if present, accelerates retinopathy, nephropathy, and CHD). Sports that emphasize aerobic conditioning, muscle tone, and endurance and that have the potential for lifelong participation are more likely to contribute to achieving these goals. Consequently, counseling the young diabetic patient about prudent choices of sports activities is a valuable role for the health care professional.

TREATMENT Guidelines for Preventive Management The challenge for the team physician supervising a diabetic athlete is to anticipate and prevent untoward events. Preparation begins with the establishment of good metabolic control through a consistent routine of blood glucose monitoring, insulin administration, and carbohydrate counting. Glycemic control should be documented by obtaining measurements of hemoglobin A1c levels every 3 months. The American Diabetes Association has developed age-specific glycemic goals that balance achieving near-normal blood glucose values while attempting to limit hypoglycemia (Table 3D-2). Gradual introduction of a fitness program should precede athletic participation when possible. Attainment of fitness before the athletic season allows insulin sensitivity to increase gradually and reduces the risk for hypoglycemia during enforced practice and   Table 3D-2  Age-Specific Glycemic Goals Age (yr)

Recommended Hemoglobin A1c

<6 6-12 13-19 ≥20

7.5%-8.5% ≤8% <7.5% <7%

From Silverstein J, Klingensmith G, Copeland K, et al: Care of children and adolescents with type 1 diabetes. A statement of the American Diabetes Association. Diabetes Care 28:186-212, 2005.

Nonorthopaedic Conditions

177

Box 3D-3 Prevention of Hypoglycemia or Hyperglycemia with Exercise Before Exercise Estimate intensity, duration, and energy expenditure needed for exercise Insulin   Administer insulin >1 hr before exercise   Decrease insulin that has peak activity coinciding with exercise period Check blood glucose   If blood glucose <100 mg/dL, ingest ≥15 g carbohydrate   If blood glucose >250 mg/dL, delay exercise and measure urine ketones   If urine ketones are positive, take insulin and delay exercise until negative During Exercise Add 15 g carbohydrate for every 30-60 min of exercise; during events of longer duration, 30-60 g/hr carbohydrate should be ­ingested at 15- to 30-min intervals Replace fluid losses adequately Monitor blood glucose at least hourly during exercise After Exercise Monitor blood glucose frequently, especially overnight Consume carbohydrate within 30 min of exercise completion to allow for efficient glycogen repletion and decreased risk for postexercise late-onset hypoglycemia Reduce insulin dosage that reaches peak effect in evening and night, according to intensity and duration of exercise From Vranic M, Berger M: Exercise and diabetes mellitus. Diabetes 28:147, 1979.

competition. This preparation reinforces development of a lifestyle with consistent exercise as a part of the diabetes treatment plan. Strategies for ensuring successful participation in sports for the diabetic athlete are summarized in Box 3D-3. The athlete should incorporate these procedures into an individualized regimen that is modified according to an estimation of the duration and intensity of exercise. In anticipation of heightened insulin sensitivity and insulin-independent glucose uptake, administration of insulin should be timed to avoid the occurrence of the peak activity of insulin during exercise. Table 3D-3 lists all current insulin preparations as well as their onset of activity, peak, and duration of action. A physician caring for a patient with diabetes must take into account the onset, peak, and duration of activity of all ­insulin preparations the athlete is taking when adjusting his or her insulin regimen. Because exercise can exacerbate hyperglycemia, ketosis, and acidosis when insufficient insulin is present in a diabetic athlete, elevated blood glucose levels and ketonuria should delay the start of exercise until insulin is given, blood glucose decreases, and ketosis resolves. Because insulin is given exogenously in diabetes, excessive insulin effect during exercise may suppress hepatic glycogenolysis and gluconeogenesis, leading to inadequate

glucose production and hypoglycemia. Consuming extra carbohydrate before, during, and after exercise is important in the athlete with diabetes. During moderate intensity exercise, glucose uptake is increased by 8 to 13 grams per hour. It is thus recommended to add 15 g of carbohydrate for every 30 to 60 minutes of activity. Exercise of longer duration or higher intensity may require additional carbohydrate intake.22 The failure of exercising T1DM patients to lose weight and gain improved metabolic control is largely due to this need for supplemental caloric intake.34 It is important to note that adjustments to the insulin dosage can be made to decrease the amount of supplemental ­carbohydrate required. Monitoring of blood glucose should occur at least every hour during exercise. Because fluid losses are increased by glycosuria, adequate fluid replacement is crucial for the athlete with diabetes. Awareness of and education concerning postexercise late-onset hypoglycemia is extremely important in the diabetic athlete. The lack of warning signs of impending postexercise late-onset hypoglycemia emphasizes the need to implement preventive measures in the hours after ­unusually strenuous or prolonged activity. Reductions in short-acting and long-acting insulin dosages, as well as increases in carbohydrate intake, are appropriate in

  Table 3D-3  Pharmacokinetics of Insulin Preparations Insulin Preparations

Onset

Peak

Duration

Aspart (NovoLog) Glulisine (Apidra) Lispro (Humalog) Regular NPH Detemir (Levemir) Glargine (Lantus)

5-15 min 5-15 min 5-15 min 30-60 min 1-3 hr 1 hr 1 hr

45-90 min 45-90 min 45-90 min 2-4 hr 4-6 hr 6-8 hr —

3-4 hr 3-4 hr 3-4 hr 5-6 hr 8-12 hr 12-24 hr 24 hr

178

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 3D-5  Insulin pump catheter insertion. Figure 3D-4  An insulin pump.

­ roportion to the duration and intensity of physical activp ity. The frequency of blood glucose monitoring should be increased to ­identify decreasing blood glucose values before hypoglycemia occurs. In general, athletes with T1DM who experience a significant increase in appetite or who feel weaker and more exhausted than usual in the evening after exercise can minimize or avoid hypoglycemia by consuming extra carbohydrate, decreasing insulin dose with peak activity during the evening or overnight, and checking blood glucose frequently during the night. Increasing numbers of athletes with diabetes are receiving insulin through continuous subcutaneous insulin infusion (CSII), or insulin pump. These devices contain short-acting insulin (Aspart, Glulisine, or Lispro) which is delivered to the subcutaneous tissue (usually the abdomen or buttocks) through a thin plastic catheter (Figs. 3D-4 and 3D-5). The pump provides insulin in two ways. The first is through a basal rate, which is preprogrammed background insulin delivered every few minutes in very small increments to meet insulin requirements while fasting or sleeping. The second manner of delivery is a bolus of insulin, which is a larger dose delivered before meals and snacks or any time when the blood glucose is elevated. The pump can be detached temporarily if not waterproof, allowing for participation in water sports. CSII has been shown to improve diabetes control in children35 and decrease the number and duration of hypoglycemic episodes.36 CSII also allows greater flexibility in timing of meals and a­ ctivity. When appropriate adjustments in insulin dosage and carbohydrate intake are made, the insulin pump can be used safely by the diabetic athlete. A recently published randomized trial of 49 children, 8 to 17 years of age, with T1DM on CSII demonstrated that suspending the basal rate of insulin for about 1 hour during exercise decreased the rate of hypoglycemia during exercise from 43% to 16%. However, it is important to note that there was an increase in postexercise hyperglycemia in the patients whose basal rates were suspended.37 On balance, for periods of exercise less than 1 hour, it is reasonable to suspend the basal rate. It is also wise to decrease the basal rate by 50% for the hour preceding exercise. In addition, a reduction in the preexercise meal bolus may be made. With more prolonged

exercise, a combination of increased carbohydrate intake, along with reduced boluses and basal rates, is appropriate. It is also important to consider decreasing the basal rate and meal boluses after exercise to avoid postexercise lateonset hypoglycemia. The following precautions apply to use of the insulin pump and exercise: Suitable protection of the device should be provided, which may be difficult during contact sports. If the pump is disconnected, either the exercise period should be brief (e.g., <60 minutes) or blood glucose monitoring with intermittent small boluses (e.g., about 50% of usual hourly basal rate) should be given every hour to replace missing basal insulin. Proper catheter placement must be monitored to ensure adequate insulin delivery during exercise. The athlete should temporarily suspend the insulin infusion from the pump if he or she experiences hypoglycemia during exercise (Box 3D-4).

Guidelines for Acute Management of Hypoglycemia Hypoglycemia during exercise may occur despite ­preventive efforts, particularly when the duration of activity is more prolonged than expected or the degree of exertion is exceptional. When blood glucose drops below 70 mg/dL, the diabetic athlete will first experience sweating, shakiness, heart racing, hunger, and irritability. Although these feelings are unique to each patient and readily ­identifiable, it is sometimes difficult to ­differentiate these signs and symptoms (e.g., sweating, heart racing) from a normal response to physical activity. If an athlete is having a mild reaction such as just Box 3D-4 General Pump Guidelines  The pump can be removed during exercise; wearing a pump during contact sports is not a good idea. l Do not remove the pump for more than 1 hour at a time. l Some, but not all, pumps are waterproof. l Insulin is sensitive to temperature extremes. l  Sweat may loosen and/or dislodge catheter. l The pump is a machine; it may fail to deliver insulin. l

Nonorthopaedic Conditions

Box 3D-5 Amount of Quick-Acting Sugar ­Needed to Produce 15 g Glucose tabs (4-5 g each): 3-4 Instant glucose (31 g tube): ½ tube Cake gel (small tube = 12 g): 1 tube Orange juice (1/2 cup = 15 g): ½ cup Sugar (tsp = 4 g): 4 tsp Gatorade (8 oz = 14 g): 8 oz Regular soda (1 oz = 3 g): 5 oz Milk (1 cup = 12 g): 1 cup Lifesavers (2.5 g each): 6 Skittles (1 g each): 15 Sweet tarts (1.7 g each): 7 From Chase HP: Understanding Diabetes, 10th ed. Denver, The Guild of the Children’s Diabetes Foundation, 2002.

described, blood glucose should be checked, and at least 15 g of fast-acting sugar should be given immediately. Box 3D-5 outlines several sources of fast-acting glucose. It is important to note that it takes at least 10 to 15 minutes for blood glucose to rise after consumption of quickacting carbohydrate; the athlete should wait at least this long before returning to physical activity. Fast-acting sugar should later be supplemented by food containing complex carbohydrate and protein, which produces a more sustained glycemic response. If hypoglycemia progresses untreated, the athlete may experience confusion, headache, behavioral change, and pallor and may progress to syncope, seizure, and loss of consciousness. These later signs and symptoms are due to neuroglycopenia, a lack of sugar to the brain. An athlete who is unconscious or impaired to such an extent that protection of the airway is uncertain should not be given oral preparations. Treatment should be initiated immediately, before confirmation of hypoglycemia using a blood glucose meter. A diabetic athlete using CSII should have the pump suspended. Parenteral glucagon (1 mg given intramuscularly) should be available to properly instructed trainers or other responsible persons and is the treatment of choice in these circumstances. Because the effect of glucagon is not long-lasting, supplemental carbohydrate and protein should be given when the mental status has improved. Not uncommonly, gastrointestinal upset, including vomiting, may occur after glucagon administration. This is an ­unfortunate side effect of glucagon administration but should not deter its use when indicated.

CRITERIA FOR SPORTS PARTICIPATION An individual with diabetes should meet the following criteria before participation in an athletic or conditioning program: 1. The individual should show knowledge, technical mastery, and consistent application of home blood glucose monitoring techniques.

179

2. The individual should show achievement of good metabolic control both chronically, as documented by measurements of hemoglobin A1c, and acutely with blood glucose levels of less than 250 mg/dL and the absence of ketonuria before exercise. 3. Evidence of ongoing follow-up with a diabetes care team and screening for diabetic complications. If such ­complications are present, the patient should show awareness of compensatory alterations in exercise choice or intensity needed for safe participation. 4. The individual should show knowledge of preventive strategies needed to avoid hypoglycemia during and after exercise. 5. If possible, prior arrangements with the team trainer or other responsible person should be made to ensure emergency treatment of hypoglycemia. Consideration of these guidelines will help ensure successful participation in sports for the athlete with diabetes. Athletic activity promotes a greater sense of well-being, better weight control, and improved physical fitness and insulin sensitivity. Athletics also provide an opportunity to pursue a meaningful personal goal. Participating with peers in sports promotes feelings of mastery, control, and individuality that significantly enrich the psychological and physical well-being of young persons with diabetes.

C l Exercise

r i t i c a l

P

o i n t s

is one of the mainstays of treatment in diabetes mellitus. l Exercise physiology is altered in diabetes because of reliance on exogenous insulin. l Before initiation of an exercise regimen, athletes with diabetes should undergo a thorough medical evaluation. l Anticipation and prevention of hypoglycemia and hyperglycemia is crucial in the management of the athlete with diabetes. l Successful participation in sports for diabetic athletes requires consideration of adjustment of insulin dosage, carbohydrate intake, and blood glucose monitoring. l Hypoglycemia is a common occurrence in the athlete with diabetes but can be prevented. l Postexercise late-onset hypoglycemia is a frequent occurrence in the athlete with diabetes, and its incidence can be decreased with (1) decreased insulin dose before, ­during, and after exercise; (2) increased carbohydrate intake; and (3) increased blood glucose monitoring. l Continuous subcutaneous insulin infusion or insulin pump therapy is commonly used by the athlete with T1DM and is a safe and effective manner in which to deliver insulin. l All physicians, trainers, and coaches should know how to manage hypoglycemia in the diabetic athlete. l Athletic activity should be encouraged and supported in the person with diabetes.

180

DeLee & Drez’s Orthopaedic Sports Medicine

S U G G E S T E D

R E A D I N G S

American Diabetes Association: Physical activity/exercise and diabetes. Diabetes Care 27(Suppl 1):58-62, 2004. Chase HP: Understanding Diabetes, 10th ed. Denver, The Guild of the Children’s Diabetes Foundation, 2002. Colberg S: The Diabetic Athlete. Champaign, Ill, Human Kinetics, 2001. Doyle EA, Weinzimer SA, Steffen AT, et al: A randomized prospective trial comparing the efficacy of continuous subcutaneous insulin infusion with multiple daily injections using insulin glargine. Diabetes Care 27:1554-1558, 2004. United States National Diabetes Information Clearing house: National diabetes statistics, November, 2005. http://diabetes.niddk.nih.gov/dm/pubs/statistics/index. htm. Raile K, Kapellen T, Schweiger A, et al: Physical activity and competitive sports in children and adolescents with type 1 diabetes. Diabetes Care 22:1904-1905, 1999. Richter EA: Glucose utilization. In Rowell LB, Shepherd JT (eds): Handbook of Physiology. New York, Oxford University, 1996, pp 912-951.

Ruderman N: ADA Handbook of Exercise in Diabetes. Alexandria, Va, American Diabetes Association, 2002. Silverstein J, Klingensmith G, Copeland K, et al: Care of children and adolescents with type 1 diabetes. A statement of the American Diabetes Association. Diabetes Care 28:186-212, 2005. Tsalikian ET, Kollman CK, Tamborlane WV, et al: Prevention of hypoglycemia during exercise in children with type 1 diabetes by suspending basal insulin. Diabetes Care 29:2200-2204, 2006.

R eferences Please see www.expertconsult.com

S ect i o n

E

Exercise-Induced Bronchospasm Jonathan P. Parsons, Kendra McCamey, and John G. Mastronarde

DEFINITION AND PREVALENCE Exercise-induced bronchospasm (EIB) describes acute, transient airway narrowing that occurs during and most often after exercise. EIB is characterized by symptoms of cough, wheezing, or chest tightness during or after exercise. Exercise is a common trigger of bronchospasm in asthmatic patients, and 50% to 90% of all individuals with chronic asthma have EIB.1 However, EIB also occurs in up to 10% of people who are not known to be atopic or asthmatic.2 These patients do not have the typical features of chronic asthma (i.e., frequent daytime symptoms, ­nocturnal symptoms, impaired lung function), and exercise may be the only stimulus that causes respiratory symptoms. EIB occurs quite commonly in athletes, and prevalence rates of bronchospasm related to exercise in athletes range from 11% to 50% (Table 3E-1).1 Holzer and colleagues3 found 50% of a cohort of 50 elite summer athletes had EIB. Wilber and associates4 found that 18% to 26% of Olympic winter sport athletes and 50% of cross-country skiers had EIB. The U.S. Olympic Committee reported an 11.2% prevalence of EIB in all athletes who competed in the 1984 Summer Olympics.5 Despite numerous studies that investigate the prevalence of EIB in athletes, few studies have investigated the prevalence of EIB in cohorts of athletes without known history of asthma or EIB. Mannix and associates6 found that 41 of 212 subjects (19%) in an urban fitness center, none of whom had a previous diagnosis of asthma, had EIB. Rupp and colleagues7 evaluated 230 middle and high school student athletes, and after excluding those with known EIB, found that 29% had EIB. These studies suggest that EIB occurs commonly in subjects who are not known to be asthmatic and likely is underdiagnosed clinically.

The prevalence of EIB may be further underestimated because patients with asthma and EIB have been shown to be poor perceivers of symptoms of bronchospasm.8,9 Specifically, athletes often suffer from lack of awareness of symptoms suggestive of EIB.10,11 Health care providers and coaches also may not consider EIB as a possible explanation for respiratory symptoms occurring during exercise. Athletes are generally fit and healthy, and the presence of a significant medical problem often is not considered. The athlete is often considered to be “out of shape,” and vague symptoms of chest discomfort, breathlessness, and fatigue are not interpreted as a manifestation of EIB. Athletes themselves are often not aware that they may have a physical problem. Furthermore, if they do recognize they have a medical problem, they often do not want to admit to health personnel that a problem exists because of fear of social stigma or losing playing time.12

SPECIFIC ATHLETIC POPULATIONS AT RISK Athletes that compete in high-ventilation or endurance sports (Table 3E-2) are more likely to experience symptoms of EIB than those who participate in low-ventilation sports13; however, EIB can occur in any setting. It is especially prevalent in endurance events in which ventilation is increased for long periods of time during training and competition such as such as cross-country skiing, swimming, and long-distance running.13 There is also increased prevalence of EIB in winter sports athletes.4 In addition, environmental triggers may predispose certain populations of athletes to an increased risk for development of EIB. Chlorine compounds in swimming pools14 and chemicals related to ice-resurfacing machinery in ice rinks,15 such as carbon monoxide and

Nonorthopaedic Conditions

Table 3E-1  Prevalence of Exercise-Induced ­Bronchospasm in Selected Studies

Reference No.

Athletes

 4  5 25

Winter Olympians Summer Olympians Elite figure skaters

 3

Elite athletes

50 51

College athletes College athletes

EIB Prevalence ­(Bronchoprovocation ­Technique) 18%-26% (exercise) 11% (exercise) 41% (EVH) 31% (exercise) 50% (EVH) 18% (methacholine) 46% (exercise) 50% football, 25% ­ basketball (methacholine)

EIB, exercise-induced bronchospasm; EVH, eucapnic voluntary hyperventilation.

nitrogen dioxide, may put exposed athletic populations at additional risk. These environmental factors may act as triggers and exacerbate bronchospasm in athletes who are predisposed to EIB. Thus, it is important for athletes, coaches, and trainers supervising athletes in these high-risk sports to be aware of the increased incidence of EIB.

CLINICAL PRESENTATION The clinical manifestations of EIB are extremely variable and can range from mild impairment of performance to severe bronchospasm and respiratory failure. Common symptoms include coughing, wheezing, chest tightness, and dyspnea. More subtle evidence of EIB includes fatigue, symptoms that occur in specific environments (e.g., ice rinks or swimming pools), poor performance for conditioning level, and avoidance of activity (Box 3E-1). Generally, exercise at a workload representing at least 80% of the maximal predicted oxygen consumption for 5 to 8 minutes is required to generate bronchospasm in most athletes.16 Typically, athletes experience transient bronchodilation initially during exercise, and symptoms of EIB begin later or shortly after exercise. Symptoms often peak 5 to 10 minutes after exercise ceases and can remain significant for 30 minutes or longer if no bronchodilator therapy is provided.17 However, some athletes spontaneously recover to baseline airflow within 60 minutes, even in the absence of intervention with bronchodilator therapy.17 Unfortunately, it is currently impossible to predict which athletes will recover without treatment. Athletes who experience symptoms for extended periods often perform at suboptimal levels for significant portions of their competitive or recreational activities.

181

Box 3E-1 Common Symptoms of Exercise-Induced Bronchospasm Dyspnea on exertion Chest tightness Wheezing Fatigue Poor performance for level of conditioning Avoidance of activity Symptoms in specific environments (e.g., ice rinks, ­swimming pools)

DIAGNOSIS History and Differential Diagnosis The presence of EIB has been shown to be difficult to diagnose clinically because symptoms are often nonspecific.13 Complete history and physical examination should be performed on each athlete with respiratory complaints associated with exercise. However, despite the value of a comprehensive history of the athlete with exertional dyspnea, the diagnosis of EIB based on self-reported symptoms alone has been shown to be inaccurate. Hallstrand and colleagues18 found that screening history identified subjects with symptoms or a previous diagnosis suggestive of EIB in 40% of the participants, but only 13% of these persons actually had EIB after objective testing. Similarly, Rundell and associates10 demonstrated that only 61% EIBpositive athletes reported symptoms of EIB, whereas 45% of athletes with normal objective testing reported symptoms. The poor predictive value of the history and physical examination in the evaluation of EIB strongly suggests that clinicians should perform objective diagnostic testing when there is a suspicion of EIB. Other medical problems that can mimic EIB and that need to be considered in the initial evaluation of exertional dyspnea include vocal cord dysfunction, cardiac arrhythmias, cardiomyopathies, gastroesophageal reflux disease, and pulmonary or cardiac shunts (Box 3E-2). A comprehensive history and examination is recommended to help rule out these other disorders, and specific testing such as echocardiography may be required. A history of specific symptoms in particular environments or during specific activities should be elicited. Timing of symptom onset in relation to exercise and recovery is also helpful. A ­thorough family and occupational history should be obtained because a family history of asthma increases the risk for other family members developing asthma.19

Table 3E-2  Examples of High- and Low-Ventilation Sports

High-Ventilation

Low-Ventilation

Box 3E-2 Common Mimics of Exercise-Induced Bronchospasm

Swimming Soccer Lacrosse Cross-country skiing Distance running Cycling

Baseball Volleyball Football Tennis Golf Downhill skiing

Arrhythmias Cardiomyopathy Vocal cord dysfunction Gastroesophageal reflux disease

182

DeLee & Drez’s Orthopaedic Sports Medicine

Objective Testing Objective testing should begin with spirometry before and after inhaled bronchodilator therapy, which will help identify athletes who have asthma. However, many people who experience EIB have normal baseline lung function.20 In these patients, spirometry alone is not adequate to diagnose EIB. Significant numbers of false-negative results may occur if adequate exercise and environmental stress are not provided in the evaluation for EIB. In patients being evaluated for EIB who have a normal physical examination and normal spirometry, bronchoprovocation testing is recommended. A positive bronchoprovocation test indicates the need for treatment of EIB. Specific tests have varying positive values, but in general, a change (usually ≥10% decrease in forced expiratory volume in 1 second [FEV1]) between pretest and posttest values is suggestive of EIB.21 In a patient with persistent exercise-related symptoms and negative physical examination, spirometry, and bronchoprovocation testing, we recommend reconsidering alternative diagnoses. Not all bronchoprovocation techniques are equally valuable or accurate in assessing EIB in athletes. The International Olympic Committee recommends eucapnic voluntary hyperventilation (EVH) challenge to document EIB in Olympians.22 EVH involves hyperventilation of a gas mixture of 5% CO2 and 21% O2 at a target ventilation rate of 85% of the patient’s maximal voluntary ventilation in 1 minute (MVV). The MVV is usually calculated as 30 times the baseline FEV1. The patient continues to ­hyperventilate for 6 minutes, and assessment of FEV1 occurs at specified intervals up to 20 minutes after the test. This challenge test has been shown to have a high specificity23 for EIB. EVH has also been shown to be more sensitive for detecting EIB than methacholine3 or field- or lab-based exercise testing.23 It is portable, is relatively inexpensive, and has protocols21 that allow standardization between laboratories. Mannix and colleagues24 used EVH to screen 79 high school athletes and found that 38% of them had EIB. It is possible that the increased sensitivity of EVH for detecting EIB will demonstrate prevalence rates in athletes that are higher than previously reported with other, less sensitive tests. Field-exercise challenge tests that involve the ­athlete performing the sport in which he or she is normally involved and assessing FEV1 after exercise have been shown to be less sensitive than EVH25 and allow for little standardization of a protocol. Pharmacologic challenge tests, such as the methacholine challenge test, have been shown to have a lower sensitivity than EVH for detection of EIB in athletes3 and are also not recommended for first-line evaluation of EIB. Therefore, EVH currently is the best validated bronchoprovocation technique available given its sensitivity and specificity, portability, and standardization of protocol.

TREATMENT OPTIONS Pharmacologic Therapy Pharmacologic therapy for EIB (Table 3E-3) has been studied extensively; however, most studies have included asthmatic participants, and there are no guidelines ­currently

Table 3E-3  Treatment and Prevention of ­ Exercise-­Induced Bronchospasm

Pharmacologic Therapy

Nonpharmacologic Therapy

Short-acting β-agonists Inhaled corticosteroids Long-acting β-agonists Leukotriene modifiers Cromolyn compounds

Adequate pre-exercise warm-up Wearing a mask in cold environment Avoidance of triggers Nasal breathing

available to guide pharmacotherapy in nonasthmatic EIB. It is not known whether recommended therapy for EIB in asthmatic patients is as efficacious in nonasthmatic patients who experience EIB. The most common therapeutic recommendation to minimize or prevent symptoms of EIB is the prophylactic use of short-acting bronchodilators (selective β-­adrenergic receptor agonists) such as albuterol shortly before exercise.26 Treatment with two puffs of a short-acting β-­agonist shortly before exercise (15 minutes) will provide peak bronchodilation in 15 to 60 minutes and protection from EIB for at least 3 hours in most patients. Long-acting bronchodilators work in a similar manner pharmacologically as short-acting bronchodilators; however, the bronchoprotection afforded by long-acting β-agonists has been shown to last up to 12 hours, whereas that of short-acting agents is no longer significant by 4 hours.27 Ferrari and associates28 demonstrated inhalation of formoterol, a long-acting β-agonist, is effective in ­protecting asthmatic athletes as early as 15 minutes after dosing. However, tachyphylaxis also has been shown to occur after repeated use of long-acting β-agonists29; thus, close follow-up is recommended when using these medications. In addition, recent controversy about use of longacting β2-agonists as monotherapy in asthmatic patients should caution health care providers about the use of these agents alone.30,31 Inhaled corticosteroids are first-line therapy in terms of controller medications for athletes who have asthma and experience EIB.26 Airway inflammation is also often ­present in nonasthmatic athletes who have EIB14,32; therefore, inhaled corticosteroids may be an effective medicine for treatment, but efficacy of corticosteroids in this cohort has not been studied. Inhaled corticosteroids are thus more commonly reserved for athletes refractory to short-acting β2-agonists. Leukotriene modifiers have also been shown to be effective in treating EIB.33-35 Leff and colleagues36 evaluated the ability of montelukast, a leukotriene receptor antagonist, to protect asthmatic patients against ­EIB. Montelukast therapy offered significantly greater protection against EIB than placebo therapy and was also associated with a significant improvement in the maximal decrease in FEV1 after exercise. In addition, tolerance to the medication and rebound worsening of lung function after discontinuation of treatment were not seen. In another study, daily zafirlukast treatment protected against EIB for at least 8 hours after regular dosing.34 Based on the data from a limited number of studies, leukotriene modifiers are an effective second-line agent for treatment of EIB.

Nonorthopaedic Conditions

183

  A u t h o r s ’ P r e f e r r e d M e t h o d     Our preferred method for diagnosis and treatment of EIB is shown in Figure 3E-1. The diagnosis of EIB based on symptoms alone is extremely inaccurate. Objective testing is necessary to make a confident diagnosis of EIB. We recommend using EVH as the bronchoprovocation test of choice to document EIB; however, EVH may not be available to many health care providers. If EVH is not easily accessible, spirometry before and after an adequate exercise challenge is our second-line recommendation. It is essential to ensure that the exercise challenge is strenuous enough to generate adequate ventilation rates in patients who have excellent physical fitness. In our experience, both pharmacologic and nonpharmacologic approaches are essential to minimizing the adverse effects of EIB. We recommend starting athletes who have

clinical evidence of EIB on short-acting bronchodilators before exercise and instructing them on the importance of adequate warm-up and avoidance of known triggers. This regimen will prevent significant EIB in more than 80% of athletes.26 If symptoms persist, especially in athletes with asthma, we recommend adding corticosteroids as maintenance therapy. Although the efficacy of inhaled steroids in nonasthmatic athletes has not been evaluated, we recommend using them in nonasthmatic athletes whose symptoms are not completely controlled with short-acting bronchodilators because there is evidence of increased inflammation in the airways of subjects without known asthma as a result of hyperventilation and exercise.14,46 Alternatively, leukotriene modifiers or cromolyn compounds can be used in athletes inadequately controlled with β2-agonists.

Symptoms suggestive of EIB?

History, Physical Examination, and Spirometry

Examination Normal?

No

Yes Obstruction on spirometry?

Consider further work-up based on history/examination findings, e.g.: • Echocardiogram • Videolaryngostraboscopy • Holter monitor • Pulmonary function tests

No

Yes

EVH Testing • If EVH unavailable: Alternative bronchoprovocation

Consider maintenance asthma therapy

Positive

Negative

• Albuterol before exercise • Proper warm-up • Avoidance of triggers

Reconsider diagnosis

• VCD • Cardiac arrhythmias • Shunts • Other pulmonary diseases

Symptoms improved? No Yes Continue

No Add controller medication: • Inhaled corticosteroid • Leukotriene modifier • Cromolyn compound

Symptoms improved? Yes Continue current therapy

Figure 3E-1  Evaluation and management of exercise-induced bronchospasm (EIB). EVH, eucapnic voluntary hyperventilation; VCD, vocal cord dysfunction. (Redrawn from Parsons JP, Mastronarde JG: Exercise-induced bronchoconstriction in athletes. Chest 128:3966-3974, 2005.)

184

DeLee & Drez’s Orthopaedic Sports Medicine

Box 3E-3 Symptoms of Respiratory Distress Increase in wheezing or chest tightness Unable to speak in full sentences A respiratory rate greater than 25 breaths per minute Persistent cough Breathing with nostril flaring Breathing with paradoxical abdominal movements

Mast cell stabilizers have been studied extensively for the prophylaxis of EIB. These medications prevent mast cell degranulation and subsequent histamine release. In a recent meta-analysis of the prevention of EIB in asthmatic patients, nedocromil sodium was found to improve FEV1 by an average of 16% and to shorten the duration of EIB symptoms to less than 10 minutes.37 Although these agents are effective, they are often used as a second-line treatment because of their cost and their decreased duration of action and efficacy compared with β2-agonists.

Nonpharmacologic Therapy Many athletes find that a period of precompetition­ warm-up reduces the symptoms of EIB that occur during their competitive activity. Athletes often draw this conclusion without any guidance from health care specialists. It has been shown by investigators that this refractory period does occur in some athletes with asthma and that athletes can be refractory to an exercise task performed within 2 hours of an exercise warm-up.38-40 However, the

Figure 3E-2  Sideline management of exerciseinduced bronchospasm (EIB) and criteria for return to play.

refractory period has not been consistently proved across different athletic populations and has not been well documented in EIB-positive athletes who are not asthmatic; in addition, it is currently not possible to identify which athletes will experience this refractory period.41 Other nonpharmacologic strategies (see Table 3E-3) can be employed to help reduce the frequency and severity of symptoms of EIB. Breathing through the nose rather than the mouth will also help ameliorate EIB42 by warming, filtering, and humidifying the air, which subsequently reduces airway cooling and dehydration. Wearing a facemask during activity warms and humidifies inspired air when outdoor conditions are cold and dry and is especially valuable to elite and recreational athletes who exercise in the winter.43 In addition, people with knowledge of triggers (e.g., freshly cut grass) should attempt to avoid them if possible.

SIDELINE MANAGEMENT Acute, sideline management of EIB requires trainers and coaches to be prepared to intervene if an athlete experiences an acute episode of EIB. All athletic trainers should have pulmonary function measuring devices such as peak flow meters at all athletic events, including practices.44 In addition, a rescue inhaler should available during all games and practices. Spacers are recommended to be used with the rescue inhalers, and nebulizers should be readily available for emergencies in which inhalers do not work.44 On-field management of asthma begins with being aware of the signs and symptoms of respiratory distress (Box 3E-3). Any athlete presenting with any of these symptoms should be removed from competition and get

Symptoms suggestive of EIB

Remove athlete from activity

Assess peak expiratory flow rate (PEFR)

If PEFR is ≥10% below baseline, treat with two puffs of albuterol

Reassess PEFR in 5-10 minutes

If PEFR back to baseline, may return to activity

If PEFR not back to baseline, repeat albuterol treatment Reassess PEFR in 5-10 minutes

If PEFR back to baseline, may return to activity

If PEFR not back to baseline, transfer from sideline to higher level health care facility

185

Nonorthopaedic Conditions

i­mmediate evaluation by a physician.44 It is recommended that any athlete with a peak expiratory flow lower than 80% of personal best be removed from activity until their peak flow returns to at least 80% of personal best.45

POTENTIAL COMPLICATIONS The goals of treating an athlete with EIB are to optimize pulmonary function before starting athletic competition and to attempt to prevent significant episodes of EIB from occurring during exercise. Unfortunately, EIB often goes unrecognized, and consequences of unrecognized or inadequately treated EIB are significant. Becker and associates47 identified 61 deaths secondary to asthma over a 7-year period occurring in close association with a sporting event or physical activity. Of these deaths, 81% occurred in subjects younger than 21 years, and 57% occurred in subjects considered elite or competitive. Strikingly, almost 10% of deaths in this review occurred in subjects with no known history of asthma. Similarly, Amital and colleagues48 found that asthma was the single greatest risk factor for unexplained death in a review of Israeli military recruits’ data over a 30-year period. Results from these reviews suggest that all individuals involved in organized sports or physical activity should be cognizant of the risk for EIB. Coaches, trainers, parents, and team physicians who care for competitive athletes who have asthma or EIB should be specifically trained in the recognition and treatment of EIB.

CRITERIA FOR RETURN TO PLAY Criteria for safe return to play after an acute episode of EIB are based on expert opinion only. Most experts agree that no athlete should return to play until lung function returns to baseline.44,49 However, there is no consensus return-toplay protocol, and each athlete must be evaluated on an individual basis for fitness for returning to play after an acute episode of EIB. An algorithm outlining acute, sideline management of EIB and suggested criteria for return to play are shown in Figure 3E-2.

C

r i t i c a l

P

o i n t s

l Exercise-induced

bronchospasm occurs in athletes with and without chronic asthma. l Exercise-induced bronchospasm occurs more commonly in athletes than the general population. l The symptoms of exercise-induced bronchospasm are often subtle and difficult to differentiate from normal manifestations of intense exercise. l Diagnosis of exercise-induced bronchospasm based on subjective symptoms alone is extremely inaccurate. l Objective testing is strongly recommended to document a diagnosis of exercise-induced bronchospasm. l The consequences of unrecognized or inadequately treated exercise-induced bronchospasm can be significant. l Treatment of exercise-induced bronchospasm with a shortacting bronchodilator before exercise is 80% effective. l Coaches and trainers should be prepared to manage an athlete with an acute episode of exercise-induced bronchospasm at all practices and competitive events.

S U G G E S T E D

R E A D I N G S

Anderson SD, Argyros GJ, Magnussen H, et al: Provocation by eucapnic voluntary hyperpnoea to identify exercise induced bronchoconstriction. Br J Sports Med 35:344-347, 2001. Anderson SD, Holzer K: Exercise-induced asthma: Is it the right diagnosis in elite athletes?. J Allergy Clin Immunol 106:419-428, 2000. Becker JM, Rogers J, Rossini G, et al: Asthma deaths during sports: Report of a 7-year experience. J Allergy Clin Immunol 113:264-267, 2004. McFadden ER Jr, Gilbert IA: Exercise-induced asthma. N Engl J Med 330:13621367, 1994. Miller MG, Weiler JM, Baker R, et al: National athletic trainers’ association position statement: Management of asthma in athletes. J Athl Train 40:224-245, 2005. Parsons JP, Mastronarde JG: Exercise-induced bronchoconstriction in athletes. Chest 128:3966-3974, 2005. Rundell KW, Im J, Mayers LB, et al: Self-reported symptoms and exercise-induced asthma in the elite athlete. Med Sci Sports Exerc 33:208-213, 2001. Rundell KW, Jenkinson DM: Exercise-induced bronchospasm in the elite athlete. Sports Med 32:583-600, 2002.

R eferences Please see www.expertconsult.com

S ect i o n

F

Sports and Epilepsy Mary L. Zupanc

HISTORICAL BACKGROUND Epilepsy is a common neurologic disorder. Between 0.5% and 1.0% of the general population have epilepsy.1 Despite the recent surge in sports and exercise in the United States and other countries, individuals with epilepsy are less likely to exercise than their peers. The restrictions on exercise are often imposed by either overprotective families or

­ hysicians who are concerned about the risk for seizures p during exercise. Some individuals with epilepsy do not have the desire to participate in sports because of fear of embarrassment if they were to have a seizure; depression; anxiety; low self-esteem; and social isolation. Unfortunately, this has resulted in a large number of people with epilepsy who lead relatively sedentary lives and are not obtaining the exercise that they need to remain healthy. In past years,

186

DeLee & Drez’s Orthopaedic Sports Medicine

the American Academy of Pediatrics and the American Medical Association actively discouraged patients with epilepsy from participating in physical fitness and athletic sports. Between 1968 and 1974, both the American Medical Association and the American Academy of Pediatrics published recommendations against sports participation for individuals with epilepsy.2,3 These guidelines were reversed in 1978 and 1983, respectively. In 1983, the Committee on Children with Handicaps and Committee on Sports Medicine came out with the following recommendations: “Proper medical management, good seizure control and proper supervision are essential if children with epilepsy are to participate fully in physical education programs and ­interscholastic athletics. Common sense dictates that situations in which a seizure could cause a dangerous fall should be avoided.… Epilepsy per se should not exclude a child from hockey, baseball, football, basketball, and wrestling.”4 This overprotective attitude has changed in recent years because of the lack of clinical data supporting restrictions and other studies that actually show a beneficial response, both physical and psychological. The evidence demonstrates that individuals with good seizure control can be allowed to participate in both contact and noncontact sports without adversely affecting seizure control.5-7 In fact, exercise can increase feelings of self-worth, decrease depression, and reduce stress. It is rare that physical activity precipitates seizures.8 Children who have breakthrough seizures can participate in some sports activities, depending on the nature of the sports activity, the severity of the seizure, and the treatment. The most valid approach is to individualize the decision so that the health care provider, the patient, and the family weigh the risks versus benefits of sports participation and make a reasonable judgment as to whether or not a particular sports activity should be allowed.

TERMINOLOGY Epilepsy refers to unprovoked, recurrent seizures. An epileptic seizure is the clinical manifestation of an abnormal synchronous discharge of a group of neurons in the cerebral cortex. The features of a clinical epileptic seizure differ according to the anatomic location of the neurons in the cortex. Depending on where the seizure focus resides in the brain, an epileptic seizure can cause an altered state of consciousness, tonic-clonic movements, stereotypic or repetitive movements, loss of muscle tone, sensory or psychic experiences, or autonomic or visual dysfunction. These disturbances usually, but not always, are accompanied by electrographic epileptogenic activity during scalp electroencephalography (EEG). There are many different types of epileptic seizures. In the past, the classification of epileptic seizures was confusing, and many redundant terms were used. The International League against Epilepsy has published an international classification of epileptic seizures.9 According to this classification, epileptic seizures are divided into two categories: generalized and partial (Box 3F-1). Generalized seizures are characterized by bilaterally synchronous discharges involving the entire cortex of the brain. They can be subdivided further into nonconvulsive and convulsive seizures. Nonconvulsive generalized seizures include absence seizures (petit mal) and myoclonic seizures. Convulsive generalized seizures

Box 3F-1 Seizure Types Partial (Focal) Simple partial seizures—without alteration of consciousness (e.g., déjà vu, jamais vu, abdominal queasiness, autonomic symptoms) Complex partial seizures—with alteration of consciousness: (1) can begin as simple partial seizures and then pro­gress to impairment of consciousness; (2) can ­secondarily ­generalize Generalized (Convulsive or Nonconvulsive) Absence seizures (previously called petit mal ) Myoclonic seizures Clonic seizures Tonic seizures Atonic seizures Generalized tonic-clonic seizures Unclassified Neonatal seizures Febrile seizures

are the more familiar tonic-clonic seizures (grand mal). Partial seizures are seizures that begin in one part of the brain. Partial seizures can be either simple partial seizures (without impairment of consciousness) or complex partial seizures (with impairment of consciousness). Simple partial seizures are what used to be called the aura before the more recognizable complex partial seizures. Simple partial seizures can encompass a range of symptoms, including motor, sensory, autonomic, and psychic phenomena. As the epileptogenic discharges spread across the neocortex, these simple partial seizures typically transform to complex partial seizures. A typical complex partial seizure ­emanating from the temporal lobe is characterized by staring, diminished responsiveness, and stereotypic automatisms, such as repetitive swallowing or lip smacking. If a partial seizure evolves into a generalized tonic-clonic seizure, it is called a secondarily generalized seizure.10 Epileptic seizures, if recurrent and unprovoked, are termed epilepsy. Epilepsy syndromes also have been identified and classified.11 The two big subdivisions are ­localization-related epilepsies and generalized epilepsies (Box 3F-2). Simplistically, generalized epilepsy syndromes Box 3F-2 Examples of Epilepsy Syndromes Generalized, Typically Hereditary Childhood absence epilepsy Juvenile absence epilepsy Juvenile myoclonic epilepsy Localization Related Symptomatic—secondary to injury or abnormality in the brain Idiopathic (hereditary): (1) benign rolandic epilepsy, (2) benign occipital epilepsy

Nonorthopaedic Conditions

often are idiopathic, contain a genetic predisposition, are ­associated with bilateral ­synchronous discharges on EEG, and often have a good prognosis. Localization-related epilepsy syndromes are divided into idiopathic and symptomatic. The idiopathic category implies a hereditary predisposition, and all diagnostic studies are negative. In the ­idiopathic category, the most common ­localization-­related epilepsy is benign rolandic epilepsy. This epilepsy is a ­self-limited condition with a hereditary predisposition and is outgrown by the time the child reaches puberty. Benign rolandic ­epilepsy is associated with temporal-central spikes seen ­during drowsiness and light sleep on the EEG; the seizures can be either partial or secondarily generalized tonic-clonic seizures. The seizure semiology typically is characterized by sensorimotor symptoms and clonic activity of the face, arm, and leg, usually associated with hypersalivation and speech arrest. Seizure frequency varies, but seizures usually are rare, occur nocturnally, and are precipitated by sleep deprivation. Treatment with antiepileptic drugs (AEDs) is optional, depending on seizure frequency and the psychosocial impact of the seizure disorder itself. The symptomatic category consists of epilepsy syndromes caused by focal brain abnormalities, such as malformations of cortical development, trauma, and tumor. Central nervous system infections, toxic or metabolic abnormalities, and neurodegenerative disorders are categorized under this heading. These abnormalities typically result in unifocal or multifocal damage to the brain. Symptomatic localization-related epilepsies tend to have a poorer prognosis for complete seizure control or cure. In some of the symptomatic localization-related epilepsies, resective epilepsy surgery can be curative or reduce the seizure burden considerably.

EXERCISE AND SEIZURE CONTROL Historically, many experts have felt that physical activity does not affect seizure frequency and may, in fact, be protective against seizures. Furthermore, studies have shown that physical activity may decrease epileptiform activity on EEG during exercise. One classic study, by Gotze and colleagues, showed that muscular exercise caused a decrease in voltage production and desynchronization of the rhythmic background of spontaneous resting activity.12 Nakken and associates, in a 1997 report, found that most patients with epilepsy experienced a decrease in the frequency of epileptiform discharges on EEG during exercise.13 Many other studies have confirmed that exercise either leads to fewer seizures or does not change seizure frequency.14-17 In another study by Nakken and associates, patients with epilepsy who engaged in a regular, intense physical exercise program over 4 weeks did not experience change in seizure frequency during the active or inactive phases of the study.14 Nakken also confirmed that more than half of patients with epilepsy never experienced a seizure during exercise and that only 2% had documented exerciseinduced seizures.8,18 Another study, examining physical activity in women with intractable epilepsy, found that the average seizure frequency decreased from 2.9 to 1.7 seizures per week with exercise.19 The mechanism by which seizure frequency is reduced with physical exercise is not understood. We do know

187

that during exercise, one is typically more vigilant, alert, and attentive. Often, epileptiform discharges are activated with drowsiness. Studies have documented a decrease in epileptiform discharges when patients with epilepsy are engaged in an interesting task or activity.20 Exercise can also reduce stress and increase endorphin levels.21 Stress is a known precipitant for seizures in individuals with epilepsy. Endorphins may function as natural antiepileptic agents; β-endorphins can inhibit epileptiform discharges. In addition, vigorous exercise can produce metabolic acidosis, which reduces the irritability of the cortex, perhaps by altering the enzymatic function for γ-aminobutyric acid, the primary inhibitory neurotransmitter of the brain. Hyperventilation in a clinical neurophysiology laboratory is often used to activate epileptiform discharges and precipitate seizures. This type of hyperventilation causes a decrease in carbon dioxide, resulting in cerebral vasoconstriction, decreased cerebral blood flow, and hypoxia. Hypoxia in hyperexcitable neurons may be enough to induce epileptiform discharges and even seizures. Hyperventilation during exercise does not produce the same phenomenon. Increased ventilation during vigorous exercise is a compensatory mechanism to avoid hypercapnia and meet increased oxygen demands. It does not produce a respiratory alkalosis.22 The most common provoking factors for seizures include stress, mental strain, and physical fatigue.23 All these can occur with intense physical exercise and competition. Stress may activate seizures through sympathetic stimulation.15 Physical fatigue may activate seizures by producing a chronically drowsy state, with a concomitant increase in underlying epileptiform activity.24 Prolonged, strenuous exercise can produce metabolic disturbances, including hyponatremia, dehydration, and overhydration, which can aggravate an underlying epilepsy. In reality, however, studies have not documented an increase in seizure activity with exercise, even contact sports.5-7 The existing data demonstrate that patients with epilepsy are at no greater risk for seizures following collisions and are unlikely to have their epilepsy exacerbated by contact. In one of the most notable studies, Livingston and colleagues reported on their experience with 15,000 young children spanning over 36 years.7 Hundreds of their patients participated in all types of sports, including contact sports such as lacrosse, wrestling, and soccer. There was not a single instance of recurrence or aggravation of epilepsy related to head injury in their patient athletes. Patients with epilepsy can experience accidents. About 7% of individuals with epilepsy die as a result of accidents. Only 5% of these deaths are attributable to seizures.1 The most common seizure-related injuries include fractures of the humeral neck, femur, clavicle, and ankle. Chipped teeth, as well as shoulder and hip dislocations, can also occur. However, in a study that spanned over 16 years, Aisenson and colleagues documented that the accident rate in patients with epilepsy was similar to their control patients without epilepsy.5 A more recent study by Fischer and Daute also found no difference in the number of accidents that children with epilepsy had during exercise compared with children without epilepsy.25 Additionally, data from the Special Olympics has shown a lower risk of injury for the athlete compared with other sporting activities.

188

DeLee & Drez’s Orthopaedic Sports Medicine

Swimming is the one activity where caution appears to be prudent. However, a well-known study by Pearn and colleagues from Hawaii found that over a 5-year period, no children with epilepsy in Hawaii died from accidents in the sea or in swimming pools.26 They concluded that the risk for drowning is small. There is evidence to suggest that the overall risk for drowning in individuals with epilepsy is 4 times the risk in their peers without epilepsy—with the greatest risk being bathing.26-29 The physical and psychological benefits of exercise are well documented in the literature. Steinhoff and associates found that patients with epilepsy have greater body mass index than controls and have significant deficits in aerobic endurance, muscle strength endurance, and physical flexibility.30 Nakken, who investigated the effects of a 4-week controlled exercise program, found that his patients increased their Vo2max to an average value that was 95% of the normal population.14 Eriksen and associates also studied women with intractable epilepsy. After 15 weeks of a twice-weekly, 60-minute exercise program, these women had significant reductions in cholesterol, overall health complaints, muscle pain, and fatigue.19 Exercise has also been shown to provide mood benefits, aid in the treatment of depression, and reduce anxiety and stress.31 Depression and anxiety are more common in patients with epilepsy. Suicide rates are estimated to be 5 times that of the normal population.31 Multiple studies in patients with epilepsy have documented an improvement in mood and psychosocial functioning with exercise.14,19,32,33

TREATMENT It is not necessary to treat the first unprovoked seizure. The chance of a recurrent seizure after a first seizure is about 30% to 40%. If the EEG shows temporal epileptiform discharges or generalized spike and slow wave discharges, the chance of recurrence is much higher, bordering on 90%.34 If a child has a second seizure, the risk for continued seizures is much higher. Antiepileptic medication generally is recommended after a second seizure. There are exceptions to this, particularly if a benign epilepsy syndrome is identified (discussed previously). There also are epilepsy syndromes that are malignant. When identified, treatment should begin without delay, regardless of seizure frequency. When discussing AED therapy, it is important to recognize the seizure type as well as the epilepsy syndrome. The identification of the appropriate epilepsy syndrome is the most important criterion in making a decision about antiepileptic medication. There are basic principles to remember in choosing AED therapy, as follows: 1. AED monotherapy is effective in most patients and avoids undesirable drug interactions. About 60% of patients will be controlled with the first AED that is chosen for treatment. An additional 10% will respond to a second AED monotherapy. The remaining 30% of patients are typically refractory, with a 10% chance or less of responding to AED therapy.35 2. AEDs should be titrated slowly and only to the point of seizure control, if possible. 3. Seizure control should not be achieved at the expense of side effects.

4. Drug compliance is enhanced when medication is given once or twice daily. Sustained-release medication should always be considered. 5. Therapeutic blood levels are not absolute. They are formulated on the basis of trough levels and represent a statistical range of efficacy. Various medications commonly are used to treat epilepsy. Older medications include phenobarbital, carbamazepine, phenytoin, valproate/valproic acid, ethosuximide, clonazepam, and primidone. Since 1993, nine new AEDs have been approved by the U.S. Food and Drug Administration (FDA): felbamate, gabapentin, lamotrigine, topiramate, oxcarbazepine, tiagabine, zonisamide, levetiracetam, and pregabalin. Phenobarbital, phenytoin, and primidone are among the oldest AEDs. Most child neurologists do not use these medications nearly as frequently as they did previously. There are newer, better tolerated AEDs that have fewer side effects. Phenobarbital, phenytoin, and primidone all have been reported to produce significant cognitive and behavioral side effects. In one study, phenobarbital reduced IQ scores in children, an effect that outlasted administration of the drug.36 Phenobarbital and primidone have also been noted to produce sedation in adults, irritability and poor attention and concentration in children, depression in adolescents, mood lability, and sleeping disorders. Phenytoin has been reported to depress cognitive function, slow overall performance, and produce sedation.10 Phenytoin is also difficult to dose because of its pharmacokinetics. It also can cause osteoporosis, as documented in a recent study by Pack and colleagues.37 The advantages of these older medications, however, are that they are relatively inexpensive and can be given once or twice daily, increasing patient compliance. Carbamazepine is the most widely used AED in the treatment of partial seizures. Carbamazepine is generally well tolerated but should be introduced slowly. Gradual introduction reduces the risk for toxicity and enhances compliance. Autoinduction of carbamazepine metabolism through the P-450 enzyme system occurs within the first month of therapy, often necessitating an increase in the total dosage of carbamazepine. If possible, when the dosage has been adjusted, attempts should be made to change to a sustained-release preparation (i.e., Tegretol-XR or Carbatrol). The toxic side effects of carbamazepine include dizziness, diplopia, sedation, ataxia, and nausea. Rare idiosyncratic reactions include aplastic anemia and hepatic dysfunction. Transient leukopenia occurs in 10% of children, usually during the first month of therapy. Allergic rash occurs in about 8% to 10% of patients; cases of ­Stevens-Johnson syndrome have been reported. Other rare side effects include irritability and dystonia. The antibiotic erythromycin and the antidepressant fluoxetine alter the kinetics of carbamazepine, resulting in significant increases in carbamazepine levels.34 Valproate is a broad-spectrum AED with demonstrated efficacy for partial and generalized seizures. Valproate comes in several formulations, including the liquid ­valproic acid, sodium valproex tablets, and sodium valproex sprinkle capsules.

Nonorthopaedic Conditions

The most common side effects of valproate include an increase in appetite with concomitant weight gain and tremor. The weight gain may be troublesome for athletes. Rarely, valproate causes an encephalopathy with sedation and cognitive impairment. With high doses, tremor, transient alopecia, and thrombocytopenia (with easy bruising and bleeding) can occur. The most publicized and serious side effect of valproate is hepatotoxicity. This is a rare, idiosyncratic reaction that occurs in the first few months of therapy. The highest-risk group is infants and toddlers less than 2 years old who have an abnormal neurologic examination and are on multiple AEDs.38 The risk in a schoolaged child is less than 1 in 30,000. Another rare side effect of valproate is pancreatitis.10 Literature has implicated valproate in the development of polycystic ovarian syndrome.39 This syndrome is associated with infertility, dyslipidemia, and insulin-resistant diabetes mellitus. The exact mechanism by which this syndrome occurs is still being investigated. However, preliminary research indicates that women who have generalized epilepsy and have been treated with valproate have a greater than 50% chance of developing anovulatory cycles or polycystic ovarian syndrome.40 Because of the strength and consistency of the data, the American Academy of Neurology has stated that valproate is relatively ­contraindicated in women with epilepsy who are in the reproductive years. Further research needs to be done to confirm these preliminary findings and their extent in pediatric and adolescent patients. Gabapentin (Neurontin) is one of the newer AEDs that is used commonly in the treatment of partial seizures. One of the biggest advantages of gabapentin is the lack of drug interactions. In contrast to other AEDs, gabapentin is not metabolized by the liver, does not induce the P-450 enzyme system, and is not highly protein bound. It is excreted through the kidney. Gabapentin is generally safe and well tolerated. There are no known fatal side effects. In pediatric patients, particularly in children with an underlying encephalopathy, irritability, aggressiveness, agitation, and other behavioral side effects have been reported.34 Lamotrigine (Lamictal) was released in 1994. It has been approved by the FDA as adjunctive therapy for the treatment of partial seizures and generalized tonic-clonic seizures. Lamotrigine is probably another broad-spectrum AED, effective in the treatment of partial and generalized seizures. Lamotrigine comes in several formulations, including tablets and sprinkle capsules. Lamotrigine carries a risk for an allergic rash, especially when combined with valproate. Valproate inhibits the metabolism of lamotrigine, resulting in an increase in the half-life from 12 hours to 72 hours. When initiating lamotrigine, the dosage must be increased slowly. If the appropriate titration schedule is used, the risk for an allergic rash is no greater than for other AEDs. If a rash is reported, the patient should be seen immediately because the rash can progress rapidly. Patients who have reported skin allergies to other drugs, especially to carbamazepine, are at a higher risk for an allergic rash from lamotrigine.34 Other less common side effects of lamotrigine include dizziness, headaches, diplopia, sedation, and movement disorders, including tics, choreoathetosis, and dystonia. Positive behavioral side effects, including antidepressant effects, have been reported with lamotrigine.

189

Oxcarbazepine is a derivative of carbamazepine. It has the same side-effect profile. However, it is not as potent a stimulator of the P-450 enzyme system. The metabolism of oxcarbazepine does not produce 10-11 epoxide, which is the metabolite often cited as the culprit for many of the side effects of carbamazepine. The epoxide is also theoretically thought to be responsible for the teratogenic effects of carbamazepine; however, the pregnancy registry data for oxcarbazepine are still pending. Oxcarbazepine is provided in liquid and tablets and can be administered twice daily. Topiramate (Topamax) is another broad-spectrum AED. Topiramate comes in several formulations, including tablets and sprinkle capsules. Topiramate is approved by the FDA as adjunctive therapy for partial seizures and generalized tonic-clonic seizures. The major side effects of topiramate are cognitive dysfunction, oligohidrosis, fevers, and metabolic acidosis. A slow titration process can potentially minimize the cognitive side effects. Topiramate can also cause renal stones, owing to its weak action as a carbonic anhydrase inhibitor. Athletes should be kept well hydrated if they are taking topiramate.34 Zonisamide is similar to topiramate, with identical side effects. Tiagabine (Gabitril) has been approved by the FDA for the treatment of partial seizures, as adjunctive therapy. There are no known drug interactions. The major side effects of tiagabine include lethargy, irritability, aggressive behaviors, dizziness, headache, and tremor. It is 96% protein bound and is metabolized by the liver.34 Levetiracetam is a new AED that is being used increasingly in the pediatric population. It is FDA approved as adjunctive therapy for partial seizures and myoclonic seizures. It has no known drug interactions. It does not produce bone marrow or liver dysfunction. Its most common side effect is behavioral disinhibition.

Alternatives to Antiepileptic Drugs The ketogenic diet as a treatment of epilepsy has been known since Biblical times. It became repopularized in the modern era by Dr. Keith from the Mayo Clinic and more recently by Johns Hopkins Medical Center. Patients with a particularly intractable form of epilepsy, called LennoxGastaut syndrome, have the best chance of responding to this diet. The diet can be difficult for children who already have dietary preferences because most of the calories are obtained from fat. Carbohydrates are extremely limited in this diet. Typically, there also is a fluid restriction. The diet would be difficult to maintain for a high-performance athlete but could be used in patients who are competing in intramural and neighborhood sports. The vagal nerve stimulator was approved by the FDA for use as adjunctive therapy in the treatment of medically refractory localization-related epilepsy in adults and children older than 12 years. The vagal nerve stimulator is a pacemaker implanted in the chest pocket below the clavicle and delivers pulses from a bipolar electrode connected to the vagus nerve. The exact mechanisms of action by which the vagal nerve stimulator works remain an enigma. Most likely, the vagal nerve stimulator influence over the EEG is mediated by brainstem pathways with concomitant ­cortical projections. Stimulation of the vagus nerve appears to result in EEG desynchronization and eventual attenuation

190

DeLee & Drez’s Orthopaedic Sports Medicine

of seizure frequency. Side effects include bleeding, infection, voice alteration or hoarseness when the vagal nerve stimulator cycles on, cough, throat pain, dyspepsia, and nausea. There are no reports of cardiac arrhythmias with this device. Immunotherapy and steroids have been used in the treatment of a variety of rare and unusual epileptic syndromes, such as Rasmussen’s syndrome and Landau-­Kleffner syndrome. Epilepsy surgery is another therapeutic option for children who have intractable epilepsy and who have an identifiable, easily resectable epileptogenic focus or lesion on neuroimaging. Most individuals who undergo epilepsy surgery have a significant reduction in seizure frequency, and many are completely cured of their epilepsy. If a patient has mesial temporal sclerosis and seizures emanating from this temporal lobe, the chance of surgery producing a seizurefree outcome is 85% to 90%. Athletic participation can be resumed within months after surgery in most cases.1

Management of Seizures Although epilepsy generally is controlled easily, children and adolescents with epilepsy occasionally have a seizure while participating in athletic events. This situation can be embarrassing for the individual, frightening to those who observe it, and at times dangerous, particularly if it occurs during a swimming event. For the team physician, coach, and other supervisors of the event, the main tenet to remember is to stay calm and keep others calm. A seizure almost always is a self-limited event that requires minimal intervention. The next recommendation is to prevent the individual with epilepsy from self-injury. This goal can be accomplished by removing objects close to the individual, particularly during a generalized tonic-clonic seizure. If the individual has been swimming, a skilled lifeguard should remove the individual from the water as quickly as possible. A tongue blade or other object never should be inserted between the teeth of an individual who is having a seizure; the individual will not swallow his or her tongue, and chances are that the individual will bite the object in two or injure the person who is attempting to insert the object. If a mouthpiece is present, it should be removed only if this can be accomplished easily. When a seizure is over, usually within minutes, the individual often is tired. The individual should be allowed to lie down. The individual should be turned onto his or her side to prevent aspiration of vomitus. If the episode has been a nonconvulsive seizure (e.g., partial complex seizure consisting of stereotypic movements or bizarre behaviors), the patient may not need to be placed in a lateral decubitus position, although the individual often is lethargic and semiresponsive after the seizure. If a seizure occurs during a sporting event, the individual should not participate further that day. The one exception to this rule may be the child who has absence epilepsy (petit mal) and experiences occasional breakthrough seizures. When the physical needs of the individual who has experienced a seizure have been met, attention should be turned to the emotional needs of this individual and others. The embarrassment that many people feel after a seizure should be acknowledged and addressed; this implies listening to the individual and his or her feelings. A simple “How

are you feeling?” may be enough to elicit a conversation. Alternatively, the person who had the seizure may not want to discuss it at all. The significant adults in the individual’s life may want to let him or her know that they are willing to listen and talk at any time in the future. The same thing applies to other people who witnessed the event. These people need to know that seizures are not harmful, that the person will recover, and that most individuals with epilepsy lead normal lives with only rare or no breakthrough seizures. The lay public still has many misconceptions about epilepsy.

Medications and Exercise Exercise does not significantly affect drug metabolism, absorption, or serum drug concentrations.14 Physical training can have a liver enzyme–inducing effect, but this has not been shown to change AED levels. Therefore, a patient’s dosage does not need to be altered before the start of an exercise program. There can be cognitive side effects to many of the AEDs. Most of these side effects are mild and transient and should not interfere with sports and exercise. AEDs can produce cognitive dysfunction, lethargy, slow processing time, and impaired learning and memory. This is especially true of some of the older AEDs, such as phenobarbital, phenytoin, and the benzodiazepines (Table 3F-1). However, if these side effects are significant, the patient should be changed to a different AED, one that does not produce the same side effects. The athletic governing bodies, such as the International Olympic Committee, the National Basketball Association, Major League Baseball, National Football League, and the National Hockey League do not ban the use of AEDs or any of their metabolites by competitors.

CLINICAL EVALUATION OF THE SPORTS PARTICIPANT There are only a few questions that need to be answered in assessing the ability of an individual with epilepsy to participate in sports: (1) How well controlled are the individual’s seizures? (2) Is the individual experiencing significant side effects from the antiepileptic medication? (3) In what sport activity does the individual wish to participate? (4) How great is the individual’s desire to participate? What is the risk versus benefit analysis?

Treatment Options and Criteria for Sports Participation Individuals with epilepsy, even those whose seizures are not under complete control, should be allowed to participate in a variety of sports. Severely limiting sports participation may impose unduly harsh restrictions on the individual with epilepsy, resulting in stigmatization, low self-esteem, social isolation, and depression. As stated by the American Academy of Pediatrics in their ­Committee Report on this subject, “in today’s culture, sports and athletics are extremely important, and unnecessarily strict interpretations of medical conditions may do more harm than good.”4

Nonorthopaedic Conditions

  Table 3F-1  Side Effects of Antiepileptic Drugs Drug

Side Effects

Phenobarbital

Sedation, cognitive slowing Allergic rash Drug interactions (including oral ­contraceptives) Sedation, cognitive slowing Complicated pharmacokinetics Drug interactions (including oral ­contraceptives) Highly protein bound Allergic rash Cosmetic effects Drug interactions (including oral ­contraceptives) Allergic rash Sedation, cognitive slowing Hyponatremia (in adults) Leukopenia (usually asymptomatic) Aplastic anemia (rare) Drug interactions Highly protein bound Hepatic toxicity and pancreatitis, ­especially in children Weight gain Sedation (rare) Polycystic ovarian syndrome Anovulatory cycles Irritability Dose-dependent absorption Allergic rash Slow titration Drug interactions (including oral ­contraceptives) Slow titration Cognitive effects Drug interactions (including oral ­contraceptives) Renal stones (rare) Oligohidrosis Metabolic acidosis Slow titration Cognitive effects Gastrointestinal effects Irritability Interferes with oral contraceptives Allergic rash Hyponatremia Allergic rash Renal stones (rare) Oligohidrosis

Phenytoin

Carbamazepine

Valproate

Gabapentin Lamotrigine

Topiramate

Tiagabine Levetiracetam Oxcarbazepine Zonisamide

Most authors agree that individuals with epilepsy can participate in sports even if incomplete control of seizures exists.7,27,41,42 There is widespread agreement that some sports by their nature pose situations that could be dangerous even if only one breakthrough seizure were to occur. These sports include rock-climbing, rope-climbing, activity on parallel bars, high-diving, and prolonged underwater swimming.4 If fatigue is a major precipitant of an individual’s seizures, re-evaluation of the appropriateness of a particular sports activity may be necessary. In most circumstances, the sport can be allowed, but with certain restrictions. There are two areas of controversy among various authors: body contact sports and swimming. Individuals with epilepsy used to be universally denied participation

191

Author’s Preferred Method of Treatment Most children and adolescents with epilepsy should be ­encouraged to participate in athletics and sports (Table 3F-2). The opportunities to participate are much greater than they used to be. Individuals who are physically or cognitively impaired can become active in sports. Special Olympics provides opportunities for these children and ­adolescents, often with positive results in self-esteem. Children and adolescents can participate in contact sports such as football. There must always be a balance between the patient’s medical condition and his or her perceived need to participate in contact sports. Contact sports can result in head trauma. Mild head trauma can result in attention or concentration problems, memory problems, headache, and other somatic complaints. Moderate-to-severe head trauma can ­result in further brain injury and a decreased seizure threshold. Individuals with epilepsy, similar to the general population, should be outfitted with the proper protective equipment for the particular contact sport involved, including a mouthpiece. An individual with epilepsy will not choke on the mouthpiece in the event of a seizure. Recommendations for swimming must be conservative because of the risks for drowning. If an individual with epilepsy can be supervised closely (one on one) by a competent lifeguard who is trained in CPR and who is aware of the individual’s condition, swimming can be allowed. Swimming can be unrestricted if the individual has been seizure free for more than 6 months or if the individual’s usual seizure is brief and results in no or only brief alteration of consciousness. If an individual with epilepsy has chosen a sport that is thought to be unsafe (based on the criteria outlined previously), attempts should be made to steer the individual into an alternative sport or to set limitations or modifications on the sport chosen.

  Table 3F-2   Sports and Risks Risk Sport Sky diving Scuba diving Rock climbing Parallel bars Motor racing Swimming Bicycling Horseback riding Rollerblading Skateboarding Waterskiing Surfing Sailing, canoeing Cross-country running Track

High

Moderate

Low

No Known Risk

X X X X X X X X X X X X X X X

Contact Sports

Baseball Basketball Soccer Running

X X X X

192

DeLee & Drez’s Orthopaedic Sports Medicine

in body contact sports such as football. It was hypothesized that repeated head trauma would result in further brain injury with eventual aggravation of the epileptic condition. There are no studies in the medical literature that prove that chronic head trauma increases the frequency of epileptic seizures.5,6,17 Authors have argued against restriction, stating that it does not happen in reality. Their conclusion is that the decision to participate in body contact sports must be made on an individual basis. Frequent seizures (occurring daily or weekly) are considered a relative contraindication by these authors because of the possibility that a seizure would incapacitate the patient at an inopportune moment, for example, a football lineman experiencing a seizure right after the beginning of a play. Swimming represents a special case. Fatalities of individuals with epilepsy in swimming pools, the ocean, and bathtubs are well-documented but relatively rare events. An individual with epilepsy is 4 times more likely to drown than an individual without epilepsy.26-29 Various contraindications for swimming have been proposed by different authors,26,28,29,43,44 and include the following: 1. Frequent seizures, more than once a month 2. A seizure-free status that has lasted for less than 3 months 3. A patient who is going through a period of adjustment to AED therapy 4. Noncompliance with drug regimen 5. Unstable blood levels of AED 6. Lack of one-to-one supervision 7. Murky water in oceans and lakes 8. Mental retardation Frequent seizures pose a danger to the individual with epilepsy who swims. Oceans and lakes with murky water are particularly dangerous because it may be impossible to find a drowning individual before the onset of disastrous consequences. Most authors concur that participation in swimming can be allowed if an individual with epilepsy has been seizure free for 6 months, is mentally normal, has stable and therapeutic AED levels, and is supervised closely by a lifeguard who is aware of the individual’s condition and knows cardiopulmonary resuscitation (CPR).27,28,42 The buddy system should be used. If these requirements are met, swimming poses little risk. Bathing is a much bigger risk. Several studies have shown that taking a bath presents a more serious risk for drowning than swimming.27,28,45,46 Older adolescents and adults run the highest risk. They tend to bathe alone, whereas younger children are supervised closely during bath time by parents. C l Individuals

r i t i c a l

P

o i n t s

with epilepsy, even those whose seizures are not under complete control, should be allowed to participate in a variety of sports. Severely limiting sports participation may impose unduly harsh restrictions on the individual with epilepsy, resulting in stigmatization, low self-esteem, social isolation, and depression.

l Exercise

can increase feelings of self-worth, decrease depression, and reduce stress. It is rare that physical activity precipitates seizures. Children who have breakthrough seizures can still participate in some sports activities, depending on the nature of the sports activity, the severity of the seizure, and the treatment. The most valid approach is to individualize the decision so that the health care provider, the patient, and the family weigh the risks and benefits of sports participation and make a reasonable judgment about whether a particular sports activity should be allowed. l There is evidence to suggest that the overall risk for drowning in individuals with epilepsy is 4 times the risk in their peers—with the greatest risk being bathing. Participation in swimming can be allowed if an individual with epilepsy has been seizure-free for 6 months, is mentally normal, has stable and therapeutic AED levels, and is supervised closely by a lifeguard who is aware of the individual’s condition and knows cardiopulmonary resuscitation. l Children and adolescents with epilepsy can participate in contact sports such as football. There must always be a balance between the patient’s medical condition and his or her perceived need to participate in contact sports. ­Contact sports can result in head trauma. Mild head trauma can result in attention or concentration problems, memory ­problems, headache, and other somatic complaints. Moderate-to-severe head trauma can result in further brain injury and a decreased seizure threshold.

SUMMARY Children and adolescents with epilepsy should be allowed and encouraged to participate in sports and athletic events. These individuals can participate in most sports activities with no or only a few restrictions or modifications. The sense of satisfaction and the growth in self-esteem that result from participation usually far outweigh the risks.

S U G G E S T E D

R E A D I N G S

American Academy of Pediatrics Committee on Children with Handicaps and Committee on Sports Medicine: Sports and the child with epilepsy. Pediatrics 72(6):884-885, 1983. Commission of Pediatrics of the International League against Epilepsy: Restrictions for children with epilepsy. Epilepsia 38(9):1054-1056, 1997. Eriksen HR, Ellertsen B, Gronningsaeter H, et al: Physical exercise in women with intractable epilepsy. Epilepsia 35(6):1256-1264, 1994. Livingston S, Berman W: Participation of the epileptic child in contact sports. J Sports Med 2:170-174, 1974. Nakken KO: Physical exercise in outpatients with epilepsy. Epilepsia 40:643-651, 1999. Nakken KO, Bjorholt PG, Johannessen SI, et al: Effect of physical training on aerobic capacity, seizure occurrence and serum level of antiepileptic drugs in adults with epilepsy. Epilepsia 31:88-94, 1990. Pearn J, Bart R, Yamoaka R: Drowning risks to epileptic children: A study from Hawaii. Br Med J 2:1284-1285, 1978. Roth DL, Goode KT, Williams VL, et al: Physical exercise, stressful life experience, and depression in adults with epilepsy. Epilepsia 35(6):1248-1255, 1994. Steinhoff BJ, Neussuss K, Thegeder H, et al: Leisure time activity and physical fitness in patients with epilepsy. Epilepsia 37:1221-1227, 1996. Zupanc ML: Epilepsy in infants and children. In Rakel R, Bope E (eds): Conn’s Current Therapy. Philadelphia, WB Saunders, 2002.

R eferences Please see www.expertconsult.com

Nonorthopaedic Conditions

S ect i o n

193

G

Dermatologic Disorders Michael D. Pleacher and William W. Dexter An athlete’s ability to safely and comfortably compete in sport can clearly be compromised by skin injuries and infections. The skin is the body’s largest organ and its first line of defense against injury and infection. Certain sports place athletes in direct skin-to-skin contact, increasing the risk for skin injury and infection. Professionals providing medical care for athletes must be familiar with the diagnosis and management of common skin infections, injuries, and inflammatory conditions. Many conditions can be easily diagnosed by simple visual inspection, but some conditions may require more advanced testing. Many athletic skin conditions can be easily treated with medications or simple office procedures. Medical providers should be familiar with potential side effects of the medications prescribed because these may affect an athlete’s ability to compete in sport. Additionally, it is imperative that medical personnel be familiar with the return-to-play guidelines for athletes affected by skin injuries and infections.

ANATOMY OF THE SKIN The skin and its appendages function to protect the body from the environment. The skin itself is composed of two distinct layers: the epidermis and the dermis. The epidermis is the outermost layer and is composed of epithelial cells. The cells on the outer surface are dead, and are called the stratum corneum. The stratum corneum varies in thickness and is typically thickest on surfaces of the body that are exposed to repetitive friction forces such as the sole of the foot. The dermis is composed of dense connective tissue and is the layer of skin in which nerve endings and vascular structures are found. The skin has a variety of appendages as well, including the eccrine and apocrine sweat glands, sebaceous glands, hair, hair follicles, teeth, and nails. Injury and infection can affect any of these skin structures.1

been rendered and the particular return-to-play criteria of the athlete’s sport governing body have been met.

BACTERIAL DERMATOSES The skin is host to a variety of typically nonpathogenic bacteria, including species of Staphylococcus, Streptococcus, and Corynebacterium. However, these normally nonpathogenic bacteria may become pathogens under the correct circumstances, and these bacteria are commonly implicated in sports-related bacterial dermatoses. Most of these bacterial infections are easily treated with topical or oral antibiotics; however, the recent spread of methicillin-resistant strains of Staphylococcus aureus (MRSA) into the community pre­ sents a new danger to athletes and a new set of diagnostic and treatment challenges to medical personnel.

Impetigo Impetigo is a common bacterial infection most commonly caused by species of Staphylococcus and Streptococcus. Once infected, an athlete can spread the infection by skin-toskin contact. Outbreaks of impetigo are common among wrestlers, rugby players, football players, and other athletes who are frequently in close contact with their competitors. Impetigo may present with bullous or nonbullous lesions. Bullous impetigo presents as large coalescent vesicles filled with clear or honey-colored fluid. Once the bullae rupture, a flat erythematous plaque is left behind (Fig. 3G-1). In nonbullous impetigo, small vesicles cluster

SKIN INFECTIONS AND INFESTATIONS The skin can be infected with a wide variety of bacteria, fungi, viruses, and parasites. The skin of athletes is often an ideal setting for infection because the close personto-person contact and tight, occlusive athletic uniforms may lead to the development of moist, warm, abraded skin where pathogens can proliferate. Once the protective layer of stratum corneum has been breached, the skin is vulnerable to infection. When infections of the skin are present, it is important to accurately diagnose and treat the offending pathogen. It is equally important to protect other ­athletes from the spread of infection by disqualifying infected athletes until appropriate medical treatment has

Figure 3G-1  Large ruptured lesions of bullous impetigo. (Courtesy of D. Stulberg.)

194

DeLee & Drez’s Orthopaedic Sports Medicine

Folliculitis and Furunculosis

Figure 3G-2  Nonbullous impetigo.    (Courtesy of D. Stulberg.)

on an erythematous base. These lesions are typically covered in a honey-colored crust (Fig. 3G-2). The diagnosis of impetigo is typically made by simple observation of the classical clinical appearance, although culture of the fluid from the bullae may be helpful in identifying potentially resistant bacteria and directing appropriate antimicrobial therapy. Current research favors treatment with topical antibiotics such as mupirocin ointment or fusidic acid ointment. Oral antibiotics such as erythromycin, cephalexin, or dicloxacillin may be considered in widespread cases. Pen­ icillin has been shown to be ineffective in treating impetigo (Table 3G-1).2 Athletes with active, untreated impetigo must be restricted from competition until appropriate antibiotic therapy has been instituted. The National Collegiate Athletic Association (NCAA) requires wrestlers to complete 72 hours of appropriate antibiotic therapy and to be free of new lesions for 48 hours before returning to competition. Additionally, all lesions must be covered with a nonpermeable bandage that will not dislodge during competition.3 The National Federation of State High School Associations (NFHS) recommends 48 hours of oral antibiotic therapy before return to competition (Table 3G-2).4

Bacterial infection of the hair follicles may manifest as folliculitis, in which the infection is confined to the upper part of the follicle (Fig. 3G-3), or as furunculosis, in which the infection penetrates deeper, to the lower part of the hair follicle. The pathogens leading to these infections are typically species of Staphylococcus and Streptococcus and are commonly spread from person to person by direct contact. The lesions of folliculitis are usually small, tender papules and pustules found on glabrous skin. Furuncles are typically large, tender, firm pustules or abscesses. Furuncles are commonly surrounded by an area of cellulitis (Fig. 3G-4). The microorganisms responsible for folliculitis and furunculosis typically include S. aureus and Streptococcus species. The antibiotic sensitivity of these organisms is changing, with the emergence of community-acquired methicillinresistant S. aureus (CA-MRSA). Outbreaks of CA-MRSA infections have been widely reported among members of collegiate and professional football teams.5,6 The spread of CA-MRSA folliculitis and furunculosis is facilitated by skin injuries in abrasions. In one case series, the practice of cosmetic body shaving, the presence of abrasions from artificial turf, and the use of a common hot tub were identified as risk factors for the development of CA-MRSA infection.7 Diagnosis of folliculitis and furunculosis is typically made by visual inspection alone. However, given the rising frequency of CA-MRSA infections, bacterial identification through routine culture and sensitivity should be considered. Abscesses may be incised with a sterile scalpel under local anesthesia, with the abscess fluid collected and sent for culture. Folliculitis can be treated topically with mupirocin ­ointment if the infection is well-localized. More widespread infection can be treated with antistaphylococcal oral antibiotics such as cephalexin. Treatment of furunculosis combines antimicrobial therapy with incision and drainage. Antistaphylococcal antibiotics may be employed after drainage of the abscesses. If CA-MRSA is suspected, empirical use of alternative antibiotics such as trimethoprim-­sulfamethoxazole, clindamycin, or fluoroquinolones may be considered. A subset of folliculitis, termed hot-tub folliculitis, is caused by use of hot tubs colonized with Pseudomonas aeruginosa. Hot-tub folliculitis typically erupts between

  Table 3G-1  Summary of Topical and Systemic Antibiotics Commonly Used to Treat Impetigo, Folliculitis, Furunculosis, and Erythrasma Condition

Agent

Dose, Frequency

Duration of Treatment

Impetigo

Mupirocin ointment Cephalexin Dicloxacillin Mupirocin ointment Cephalexin Dicloxacillin Mupirocin ointment Cephalexin Dicloxacillin Trimethoprim-sulfamethoxazole Clindamycin Erythromycin

Apply locally, bid 500 mg, tid 500 mg, tid Apply locally, bid 500 mg, tid 500 mg, tid Apply locally, bid 500 mg, tid 500 mg, tid 1 DS tablet, bid 300 mg, tid 500 mg, qid

10 days 10 days 10 days 10 days 10 days 10 days 10 days 10 days 10 days 14 days 14 days 14 days

Folliculitis Furunculosis, non-MRSA MRSA furunculosis Erythrasma

Nonorthopaedic Conditions

195

  Table 3G-2  Summary of Return-to-Play Guidelines from the National Collegiate Athletic Association (NCAA) and the National Federation of State High School (NFHS) Associations for Skin Infections in Athletes Condition

NCAA Guidelines

NFHS Recommendations

Bacterial skin infections

1.  Minimum 72 hr of an appropriate antibiotic 2. No new lesions for at least 48 hr Minimum 2 wk of oral griseofulvin before return to play 1. Minimum 72 hr of a topical fungicide 2.  Lesions must be covered with a gas-permeable membrane 1.  Free of systemic symptoms for 72 hr 2.  No new lesions for at least 72 hr 3.  Oral antiviral treatment for at least 120 hr 1.  No moist lesions 2.  120 hr of oral antiviral treatment 1.  Local lesions may be covered with gas-permeable dressings 2.  Extensive lesions must be curetted or removed Lesions must be adequately covered Pediculicide treatment with complete response before return to play Negative scabies preparation at time of meet

1.  Oral antibiotic treatment for 48 hr 2.  No draining, oozing, or moist lesions Minimum 2 wk of oral griseofulvin before return to play 1.  Oral or topical treatment for 7 days 2.  Written release from team physician to coach 1.  120 hr of oral antiviral treatment 2.  No new lesions appearing while on antiviral treatment 1.  120 hr of oral antiviral treatment 2. No new lesions appearing while on antiviral treatment 1. Curettage followed by 24 hr before returning to play

Tinea capitis Tinea corporis Primary herpes gladiatorum Reactiviation herpes gladiatorum Molluscum contagiosum Verrucae vulgaris Pediculosis Scabies

1 and 5 days after exposure. The lesions are pruritic, ­erythematous papules located on skin that was submerged in the colonized water and are usually self-limited, lasting 7 to 14 days (Fig. 3G-5). Numerous outbreaks of hot-tub folliculitis have been reported, with one third of outbreaks occurring in health club hot tubs.8 Inadequate water chlorination, the use of cyanuric acid to stabilize chlorine, and poor monitoring of disinfectant levels in hot tubs have been implicated as factors contributing to colonization of hot tubs with P. aeruginosa.9 Treatment should be supportive, employing antihistamines to control pruritus. Antibiotic treatment may be considered with a fluoroquinolone antibiotic; however, antibiotic use may increase the risk for recurrence of hot-tub folliculitis.10 Athletes with folliculitis and furunculosis are restricted from competition in the NCAA until appropriate antibiotic therapy is instituted. The NCAA requires a minimum of 72 hours of antibiotic therapy before athletes may return to play. Additionally, athletes must not have had any new lesions develop in 48 hours, and they may not return to play with any moist, exudative, or draining lesions.3 As with

Figure 3G-3  Superficial folliculitis.    (Courtesy of D. Stulberg.)

No official position 24 hr after appropriate treatment with a pediculicide 24 hr after appropriate topical treatment

impetigo, the NFHS recommends 48 hours of oral ­antibiotic therapy for athletes with folliculitis and furunculosis before return to competition.4

Erythrasma Erythrasma is a chronic bacterial infection of the skin, generally localized in the groin and commonly mistaken for tinea cruris. Erythrasma may also occur in the intertriginous areas of the toes and axillae, again mimicking dermatophyte infection. Athletes may have a higher incidence of this condition, although there are no good epidemiologic studies demonstrating this. Erythrasma is caused by Corynebacterium minutissimum, a gram-positive diphtheroid. Erythrasma appears clinically as a sharply marginated erythematous patch that may be scaly and pruritic (Fig. 3G-6). A unique feature of erythrasma that helps distinguish it from dermatophyte infection is that erythrasma fluoresces, producing a coral-red color under Wood’s lamp examination (Fig. 3G-7). If a potassium hydroxide (KOH) ­preparation is done, erythrasma will not have any hyphae present. Dermatophyte infections will not fluoresce under Wood’s lamp and will have hyphae present on KOH preparation.

Figure 3G-4  Abscess—MRSA.    (Courtesy of W. Dexter.)

196

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 3G-5   Pseudomonas hot-tub folliculitis.    (Courtesy of D. Stulberg.)

Treatment of erythrasma may be undertaken with oral or topical antibiotics. Treatment with either erythromycin tablets for 14 days or with fusidic acid applied topically for 14 days is acceptable.11 Neither the NCAA nor the NFHS makes specific return-to-play recommendations for athletes with erythrasma.

Pitted Keratolysis Pitted keratolysis is a bacterial infection affecting the feet of athletes. The infection is generally caused by a member of the genus Corynebacterium. The growth of corynebacteria is enhanced by warm, moist environments as might be found in an athlete’s shoe. Corynebacteria have a predilection for infecting the stratum corneum, and therefore, pitted keratolysis most commonly affects the heel and the ball of the foot where the stratum corneum is thickest. Pitted keratolysis has a unique appearance, with discrete pits found on the plantar surface of the foot. There may be brownish discoloration surrounding the pits (Fig. 3G-8). These lesions are not typically painful or pruritic. The athlete may notice a slimy sensation over the affected area, and he or she may complain of particularly malodorous feet. As with erythrasma, Wood’s lamp examination may produce a coral-red fluorescence in pitted keratolysis.

Figure 3G-6  Erythrasma.    (Courtesy of D. Stulberg.)

Figure 3G-7  Erythrasma seen by Wood’s lamp.    (Courtesy of D. Stulberg.)

Initial treatment of pitted keratolysis involves simple hygienic measures, including scrubbing the feet with antibacterial soaps and blowing them dry with a blow dryer. Absorbent socks and well-ventilated shoes may help reduce the moisture in the environment. Utilizing aluminum chloride 20% cream may reduce foot perspiration. Use of topical antibiotics such as erythromycin may also be helpful.12 Athletes with pitted keratolysis are not specifically restricted from activity by either the NCAA or NFHS.

VIRAL DERMATOSES A variety of viruses may infect the skin of athletes. The same factors (sweating, warmth, abrasions) that predispose athletes to developing bacterial skin infection increase the chances of viral skin infection. In particular, the close ­body-to-body contact in certain sports such as wrestling facilitates the transmission of viral agents. Physicians caring for athletes should be familiar with the common presentations of viral skin infections and must intervene appropriately to prevent widespread outbreaks.

Figure 3G-8  Pitted keratolysis.    (Courtesy of D. Stulberg.)

Nonorthopaedic Conditions

197

Figure 3G-9  Herpes gladiatorum.    (Courtesy of W. Dexter.)

Herpes Simplex Virus Herpes simplex virus (HSV) includes two distinguishable types: HSV type 1 is the most common cause of herpes labialis, whereas HSV type 2 more commonly causes urogenital outbreaks of herpes. Primary infection with either HSV-1 or HSV-2 may produce systemic symptoms including fever, malaise, and adenopathy. Following the primary infection, the virus remains dormant in neural ganglia until it is reactivated, commonly triggered by stress, illness, or even sunlight exposure. Symptoms of reactivation include a tingling or burning at the site followed by the development of a cutaneous lesion. Reactivations of HSV infections are not typically characterized by systemic symptoms. HSV outbreaks commonly occur among wrestlers and rugby players. When the infection occurs in wrestlers, it is termed herpes gladiatorum. HSV infections among rugby players are often referred to as scrumpox. Outbreaks of HSV infection have occurred at wrestling camps and on wrestling teams with very high attack rates. In one such outbreak, more than one in three wrestlers developed HSV infection.13 Wrestlers who spar with an actively infected partner have a 32.7% chance of developing HSV infection.14 The lesions of HSV are typically vesicles on an erythematous base. The vesicles then ulcerate, creating a painful erythematous ulcer. Ultimately, these ulcers then crust over and heal. Common sites of HSV outbreak among athletes include the lips, body, and hands (Figs. 3G-9 and 3G-10). Occasionally the eyes can be infected, resulting in a severe keratoconjunctivitis, which requires ophthalmologic referral. Diagnosis of HSV infection is typically made based on the characteristic appearance of the lesions. A Tzanck smear and viral culture of the fluid from a vesicle may be helpful in confirming the diagnosis. Treatment of herpes gladiatorum involves the use of antiviral medications such as acyclovir and valacyclovir. Numerous acceptable dosing regimens exist for these medications, and both medications are effective at reducing the duration of outbreaks and speeding the time to healing of clinically apparent lesions.15 Suppressive therapy with valacyclovir has been shown to reduce the numbers of outbreaks, particularly in athletes with a longer than 2-year history of herpes gladiatorum.16 To prevent spread of HSV, athletes with active HSV outbreaks must be restricted from play until appropriate

Figure 3G-10  Herpes labialis.    (Courtesy of D. Stulberg.)

antimicrobial therapy is instituted. According to NCAA rules, athletes with a primary HSV infection must be clear of systemic symptoms, free of new blisters for a minimum of 72 hours, and free of moist lesions and have been on systemic antiviral therapy for at least 120 hours before returning to competition. Reactivation infections must be treated similarly, with no new lesions, no moist lesions, and systemic antiviral treatment for 120 hours before return to play is allowed.3 The NFHS endorses this requirement, recommending that athletes have no moist lesions and be on antiviral therapy for 120 hours before returning to competition.4

Molluscum Contagiosum Molluscum contagiosum is a common viral skin infection among children that may also be encountered in athletes. This virus is transmitted by direct skin-to-skin contact with an infected individual. The lesions of molluscum contagiosum are most commonly found on the trunk and face and appear as round, skin-colored papules with a characteristic umbilicated center (Fig. 3G-11). Lesions may be solitary but more frequently are quite numerous. Smaller lesions may appear similar to warts, but the larger lesions have the easily identifiable umbilicated center characteristic of molluscum. Molluscum may be treated with a wide variety of nonsurgical and surgical means. Application of a 0.7% cantharidin solution is a widely accepted means of chemical destruction. Cryotherapy with liquid nitrogen is another acceptable treatment option. Imiquimod 5% cream, an immune modulator, has been used successfully to treat molluscum.17 Curettage of the lesions may also be effective. The NCAA requires curettage or surgical removal of molluscum lesions before returning to competition. Solitary lesions may be covered with a gas-permeable membrane before returning to competition.3 The NFHS allows return to competition 24 hours after curettage of the lesions.4

Verrucae Vulgaris Verrucae vulgaris (common warts) are caused by various types of human papillomavirus (HPV). Certain HPV types carry the potential for malignant transformation, although

198

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 3G-12  Tinea capitis with kerion scalp.    (Courtesy of D. Stulberg.)

Figure 3G-11  Molluscum contagiosum.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

these particular HPV types are far more commonly found in genital warts than in common warts. Warts have a low rate of infectivity and are spread by direct skin-to-skin contact with an infected individual. Common warts appear as firm, skin-colored, hyperkeratotic papules. They may be present anywhere on the body. Warts are not typically painful or pruritic. Plantar warts may have a flat, skin-colored, hyperkeratotic appearance, and they may be painful. Diagnosis is typically made by simple visual inspection. Although many warts will resolve spontaneously over a number of months or years, many patients desire treatment. Two common methods of treatment are surgical removal and destruction of the lesions with cryotherapy. Topical salicylates are available over the counter and are effective in treating warts. Both cryotherapy and topical salicylates have been shown effective in a recent systematic review of wart treatments.18 The immune-modulating agent imiquimod has been effective in the treatment of anogenital warts, and it is gaining acceptance in the treatment of common warts as well. The NCAA requires warts to be adequately covered before returning to competition. If possible, the warts should be treated with curettage before return to play.3 The NFHS does not state a position on the treatment of warts.

FUNGAL DERMATOSES A diverse group of organisms termed dermatophytes may infect the skin of athletes. These organisms are typically found in soil and are of three genera: Trichophyton, Microsporum, and Epidermophyton. Infection with dermatophytes is limited to the stratum corneum. Transmission of these organisms is through direct contact, either with an infected individual, with a fomite, or directly with the soil. Fungal dermatoses are named based on the area of the body infected; tinea capitis is fungal infection of the hair and scalp, tinea corporis or ringworm is fungal infection of the body, tinea cruris is fungal infection of the groin, and tinea

pedis is the term for dermatophyte infection of the feet. Tinea gladiatorum is a term used to describe fungal infections of the skin and scalp of athletes. Outbreaks of tinea gladiatorum have been widely reported among wrestlers, with the scope of outbreaks reaching up to 75% of athletes on a team in one case series.19 The most commonly identified dermatophyte in these outbreaks is Trichophyton tonsurans.20 Over the course of one wrestling season, more than 80% of Pennsylvania high schools had at least one wrestler with a dermatophyte infection.21 Dermatophyte infections produce a characteristic clinical appearance. The lesions of tinea capitis are generally round, gray, scaly hyperkeratotic plaques with associated alopecia (Fig. 3G-12). Tinea corporis produces a round, red, scaly patch with a raised red border (Fig. 3G-13). The raised leading edge is where the active dermatophytes are found, and it is from this edge that skin scrapings should be examined under KOH preparation. Tinea cruris typically presents as very erythematous, pruritic plaques in the groin (Fig. 3G-14). There are a variety of appearances in tinea pedis. Tinea pedis may be found in a moccasin distribution, with a reddened hyperkeratotic plaque covering the plantar aspect of the foot (Fig. 3G-15). Interdigital tinea pedis may produce moist, macerated skin in the web spaces between toes (Fig. 3G-16). Tinea pedis may also result in formation of bullae filled with clear fluid. Cutaneous fungal infections are typically diagnosed clinically based on the location and characteristic appearance of the lesions. Diagnosis can be confirmed with direct microscopy of a KOH preparation or fungal culture. A positive KOH preparation reveals septated hyphae on the slide (Fig. 3G-17). Fungal culture can be performed, particularly if the lesions are suspicious for fungal infection but the KOH preparation is negative. Sabouraud’s glucose medium should be used for the fungal culture. Dermatophyte infections may be treated with topical or oral medications. The three major classes of topical agents used to combat dermatophyte infections include the imidazoles, allylamines, and naphthiomates. Griseofulvin, as well as oral preparations of allylamines and imidazoles, may be used for dermatophyte infections, particularly for

Nonorthopaedic Conditions

199

Figure 3G-15  Tinea pedis instep.    (Courtesy of D. Stulberg.) Figure 3G-13  Tinea corporis.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

­ idespread infections or for tinea capitis. The allylamines w are ­fungicidal drugs and therefore are favored over the fungistatic imidazoles and griseofulvin. Fungicidal medications require shorter courses of therapy and are the preferred agents when treating athletes (Table 3G-3). Treatment of tinea capitis requires an oral agent. The only such agent with a specific U.S. Food and Drug Administration (FDA) indication for tinea capitis is griseofulvin, typically given once daily for 6 to 12 weeks at a dose of 15 to 25 mg/kg/day. Although lacking a specific FDA indication for treatment of tinea capitis, oral preparations of imidazoles such as ketoconazole and itraconazole are often used, requiring only 2 to 4 weeks of treatment. Treatment of tinea cruris and tinea corporis should be initiated with a topical allylamine such as terbinafine. Topical terbinafine 1% cream may be applied once or twice daily for 2 to 4 weeks. A shorter 1-week course of terbinafine

Figure 3G-14  Tinea cruris.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

resulted in clinical cure in more than 80% of cases.22 The imidazole creams may be applied daily for 2 to 4 weeks, although in athletes these are not preferred because they are fungistatic rather than fungicidal medications. Tinea pedis may be appropriately treated with topical allylamines. A recent systematic review noted that tinea pedis responds well to treatment with the allylamines naftifine and terbinafine. The imidazole agents are also effective in treating tinea pedis when taken for 4 to 6 weeks.23 Refractory or chronic cases of tinea corporis, tinea cruris, and tinea pedis may require oral antifungal medications. Griseofulvins, as well as oral preparations of itraconazole, fluconazole, and terbinafine, have been used to treat refractory tinea corporis. Oral medications are more likely to induce adverse drug effects, with up to 12.5% of patients taking griseofulvin experiencing side effects. Common adverse effects of griseofulvin include headache, photosensitivity, and rash. More serious adverse effects include leukopenia, paresthesias, and transaminase elevations.24 Itraconazole taken orally at doses of 100 mg daily results in cure or improvement in 96% of treated patients, with only mild side effects of headache, dyspepsia, and flatulence affecting

Figure 3G-16  Tinea pedis toes.    (Courtesy of D. Stulberg.)

200

DeLee & Drez’s Orthopaedic Sports Medicine

regions. Lice are easily spread from person to person by direct ­contact; sports such as wrestling that place athletes in close contact may facilitate the spread of this parasite. Certain venues may predispose athletes to acquisition of parasites; athletes who compete on beaches may become infested with a specific nematode worm, resulting in cutaneous larva migrans (see later).

Pediculosis Capitis

Figure 3G-17  Fungal hyphae in KOH wet mount.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

20% of those treated.25 Fluconazole, taken orally at doses of 150 mg weekly for 2 to 4 weeks, produces cure in more than 80% of patients, with only mild adverse side effects reported.26 The most effective regimen in terms of producing mycologic cure uses the fungicidal allylamine terbinafine, taken orally at a dose of 250 mg daily for 1 week. This resulted in 100% mycologic cure, with no reported adverse events.27 The NCAA requires a minimum of 72 hours of topical treatment using a fungicidal agent for cases of tinea corporis before an athlete may return to competition. Tinea capitis must be treated with a systemic antifungal agent for 2 weeks before an athlete is cleared for participation.3 The NFHS echoes these recommendations, requiring 72 hours of topical fungicidal treatment for tinea corporis and 2 weeks of systemic antifungal therapy for tinea capitis.4

PARASITIC INFESTATIONS Athletes may acquire parasitic infestation through contact with infested competitors or family members, or through contact with parasites in the environment. Lice can be acquired and can affect the head, body, and genital

Head lice are commonly found in school-aged children but can infest people in any age group. Lice can be transmitted by direct contact with an infested individual or by fomites such as hats, brushes, and combs. Athletes infested with head lice may complain of pruritus on the scalp. Lice produce egg nits, which are easily identifiable white, firmly adhered capsules found at the base of the hair shaft (Fig. 3G-18). Diagnosis is made by visualizing the lice and their nits in the affected athlete. Treatment of head lice involves a combination of physical removal of nits through combing with a fine-toothed nit comb and the use of pediculicides. During the course of treatment, athletes with head lice should comb their hair when damp at least twice a week with the nit comb. Bed linens must be washed and dried at high temperatures to kill lice present on them. In addition to physical removal, a topical pediculicide should be applied. Permethrin 1% cream rinse is the currently recommended first-line agent for head lice. It is applied to freshly shampooed hair and is left in place for 10 minutes before being rinsed. If this fails to completely rid the hair of lice, a repeat application may be done in 1 week. For infestations that do not respond to permethrin, lindane 1% shampoo may be used in a similar manner, with one 10-minute application followed by rinsing and a repeat application in 7 to 10 days if not completely free of lice. Lindane is associated with human neurotoxicity if used incorrectly and therefore should be used with caution. Malathion is an organophosphate insecticide that is available as a 0.5% lotion. It is applied to the hair and is washed off after 8 to 12 hours.28 Athletes with pediculosis capitis are restricted from competition by the NCAA until they have completed an appropriate course of pediculicide treatment and have

  Table 3G-3    Summary of Treatment Regimens for Tinea Capitis, Tinea Corporis, Tinea Cruris, and Tinea Pedis Condition

Agent

Dose, Frequency

Duration of Treatment

Tinea capitis

Griseofulvin Itraconazole Fluconazole Ketoconazole Terbinafine 1% cream Ketoconazole 2% cream Clotrimazole 1% cream Griseofulvin Itraconazole Terbinafine Fluconazole Ketoconazole 2% cream Clotrimazole 1% cream

15-25 mg/kg/day, qd 200 mg, qd 6 mg/kg/day, qd 200 mg, qd Topical, qd-bid Topical, qd Topical, qd 500 mg orally, qd 100 mg orally, qd 250 mg orally, qd 150 mg orally, weekly Topical, qd Topical, qd

6-12 wk 2-4 wk 3-6 wk 2-4 wk 2-4 wk 2-4 wk 2-4 wk 2-4 wk 2 wk 1 wk 2-4 wk 4-6 wk 4-6 wk

Tinea corporis and tinea cruris

Tinea pedis

Nonorthopaedic Conditions

Figure 3G-18  Pediculosis capitis with secondary pyoderma.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

exhibited a complete response to therapy.3 The NFHS permits wrestlers to return to competition after 24 hours of appropriate pediculicide treatment.4

Scabies Scabies, caused by mites of Sarcoptes scabiei, are transmitted by direct person-to-person contact. Outbreaks are common among those in crowded living conditions. Scabies mites burrow into the epidermis, making it difficult to visualize the parasite directly. Symptoms of scabies infestation develop 3 to 4 weeks after exposure. Pruritus develops as the body reacts to the saliva, ova, and feces of the mites. Linear burrows are produced, typically between the fingers and toes, in the axillae, and in the groin. Diagnosis is made by the observation of these characteristic burrows (Fig. 3G-19). A scabies preparation may be made by scraping the burrows and immersing the contents in mineral oil on a microscopic slide. The slide is then examined under low power and may reveal mites and their fecal material. Treatment of scabies is similar to treatment of pediculosis capitis. The recommended first-line agent is permethrin 5% cream, applied to the entire body and then rinsed off in 8 to 12 hours. A repeat application in 1 week is recommended. Lindane 1% lotion is an acceptable alternative treatment and is applied as described earlier. The NCAA requires athletes known to be infested with scabies to have a negative scabies preparation before returning to competition.3 The NFHS requires 24 hours of appropriate topical therapy before returning to competition.4

Cutaneous Larva Migrans Beach volleyball players and beach soccer players who compete in tropical locales are susceptible to infestation with the larvae of nematode worms, which can lead to the

201

Figure 3G-19  Scabies.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

development of cutaneous larva migrans. The worm most commonly associated with the development of cutaneous larva migrans is Ancylostoma braziliense. The larvae burrow into the skin and, after 72 hours of infestation, produce raised, red, pruritic, serpiginous tunnels in the skin (Fig. 3G-20). The infestation is typically self-limited but may take 6 months to 1 year for complete resolution. Treatment with topical or oral thiabendazole is recommended. Oral albendazole has been used successfully to treat this condition as well.29 Neither the NCAA nor the NFHS has specific restrictions for athletes with cutaneous larva migrans.

SKIN INJURIES A variety of injuries to the skin can be induced by direct trauma, repetitive friction forces, recurring pressure, and environmental forces such as cold and sun exposure. Skin injuries can range from mild, self-limited conditions to more severe injuries that threaten the long-term function and appearance of the affected area of skin. Physicians providing on-site medical coverage must be properly prepared to address the variety of skin injuries that athletes may incur.

Abrasions and Lacerations Abrasions and lacerations are common skin injuries among athletes who compete in contact and collision sports. Abrasions are the result of trauma that strips the epidermal layer from the dermis. This trauma results in a denuded appearance with small foci of pinpoint bleeding commonly referred to as a “raspberry” or “turf burn.” Abrasions are commonly produced when an athlete falls on the mat, roadway, or turf on which he or she is competing. Abrasions may become infected with common skin bacteria. To reduce the chance of infection, ­abrasions should be

202

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 3G-21  Talon noir.    (Courtesy of W. Dexter.)

Figure 3G-20  Cutaneous larva migrans—severe.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

thoroughly cleansed with a saline solution. Application of an antibacterial ointment and the use of a semipermeable dressing will allow for a moist environment, which will encourage healing while minimizing scar formation.30 Lacerations are full-thickness injuries to the skin, involving both the epidermis and dermis. Blunt trauma to the skin overlying a bony prominence is the typical mechanism of injury leading to a laceration. The use of injectable local anesthetics will make the irrigation, exploration, and closure of lacerations more tolerable for the injured athlete. After adequate analgesia is attained, the wound should be irrigated with sterile saline solution to remove any debris. Suture, skin staples, and tissue adhesives may be used to close lacerations. For superficial lacerations, a monofilament suture or a tissue adhesive may be considered. For deeper lacerations, absorbable sutures may be used. Skin staples have the advantage of being quick and easy to apply, but the cosmetic results when closing a wound with skin staples are typically inferior to those wounds closed with sutures or tissue adhesives.31 After the wound is closed, a topical antibiotic ointment is applied once or twice daily. Abrasions and lacerations should be monitored for signs of wound infection. Staples and sutures used to close wounds will need to be removed when the wound has completely healed. Athletes with lacerations and abrasions should have their tetanus status assessed, and if necessary, a tetanus booster should be administered.

Friction-Induced Skin Injury Repetitive friction forces can induce skin injuries such as blisters, talon noir, acne mechanica, and contact dermatitis. Blisters commonly appear on the soles of the feet. Runners, especially long-distance runners, are prone

to developing blisters. Athletes who use poorly fitted footwear are at higher risk for developing blisters. Blisters are vesicles that are filled with clear fluid. Small blisters with an intact roof may be left alone. Larger blisters may be unroofed with a sterile needle or scalpel, allowing the fluid to drain. After drainage, the use of a membrane-type dressing may be applied to reduce friction in the area. Using properly fitted footwear can reduce the risk for developing blisters.30 Talon noir, or “black heel,” is another common frictioninduced skin injury among athletes. Talon noir appears as a black punctate plaque on the sole of the foot, commonly on the posterolateral heel (Fig. 3G-21). This lesion is typically not painful and does not usually impair athletic performance. Talon noir may be confused with plantar warts. Because of its black color, talon noir often raises concern for malignant melanoma. For talon noir, no specific therapy is required.30 Acne mechanica is a friction-induced eruption that commonly occurs on the shoulders and backs of athletes who wear protective equipment such as shoulder pads. The lesions of acne mechanica appear similar to the inflammatory lesions of acne vulgaris. Treatment of acne mechanica is aimed at reducing the friction forces that caused the eruption. Medications used for acne vulgaris are ineffective at clearing eruptions of acne mechanica. Athletes may wear a clean cotton T-shirt under the athletic pads to prevent the development of acne mechanica.32 Contact dermatitis may occur from friction irritation and sensitivity to various metals (nickel is a common offender) or materials such as neoprene. This appears as areas of ­erythema, often with papules and vesicles (Fig. 3G-22). Often pruritic, the areas may become excoriated with areas of secondary infection. Treatment is aimed at removing the offending agent and using emollients and low- to ­mid-potency topical corticosteroids. Infrequently, the dermatitis may be severe enough to warrant oral ­corticosteroids.

Environmentally Induced Skin Injury Athletes competing outdoors are susceptible to skin injuries caused by exposure to sun, wind, and cold. Clinicians caring for outdoor athletes should be familiar with the

Nonorthopaedic Conditions

203

Figure 3G-23  Mild frostbite of ear.    (From McCaulley RL, Smith DJ, Robson MC, et al: Frostbite. In Auerbach PS (ed): Wilderness Medicine, 4th ed. Philadelphia, Mosby, 2001.)

Figure 3G-22  Contact dermatitis.    (Courtesy of W. Dexter.)

diagnosis and management of common environmentally induced dermatologic problems, including frostbite, chilblains, and sunburn.

Frostbite Athletes competing outdoors in cold environments may develop frostbite. Frostbite commonly occurs on the athlete’s hands, feet, and face. Frostbite is caused by freezing of the intracellular and extracellular water within the skin. There is a spectrum of severity with frostbite, ranging from minor superficial frostnip to frank ischemic changes in the affected area, which can ultimately lead to permanent tissue damage or loss of tissue. Clinically, frostbite presents as a burning sensation accompanied by numbness or tingling in the affected area. As the depth of freezing progresses, the pain of frostbite generally resolves. The skin becomes firm, pale, and nonpliable. As the skin is warmed, blisters may develop. Superficial frostbite is characterized by numbness, edema, and erythema of the skin, and clear blisters may form (Fig. 3G-23). The blisters of deep frostbite are filled with dark, hemorrhagic fluid. Deep frostbite often progresses to necrosis of the skin and underlying soft tissues (Fig. 3G-24). Frostbite is treated by warming the skin, but it is essential to delay treatment until the athlete is removed from the cold environment. Once the cold stress has been eliminated, athletes may rewarm affected areas by immersing them in a warm water bath for 15 to 30 minutes. As the skin thaws, patients may experience more pain, which can be managed with oral analgesics. Oral anti-inflammatory drugs may limit tissue damage associated with frostbite. The blisters of superficial frostbite may be débrided and treated with aloe vera cream. Hemorrhagic, deep frostbite

blisters should be left intact. Intravenous penicillin is often administered to patients with deep frostbite to try to prevent infection. Athletes with frostbite should be removed from competition and taken to a warmer environment to begin prompt treatment. Once the blisters of frostbite have healed completely, athletes may return to competition. Athletes may develop long-term sequelae of frostbite, including altered sensation and increased cold sensitivity in the affected area. Athletes who have had frostbite are at increased risk for recurrence.33

Chilblains Chilblains are local inflammatory lesions leading to the development of a chronic vasculitis induced by prolonged exposure to cool but not freezing temperatures. Chilblains appear on the toes and fingers most commonly. Chilblains appear as violaceous colored papules, plaques, or nodules. The lesions are typically painful or pruritic and develop within 24 hours of prolonged exposure to cool temperatures. Treatment involves warming the affected area. Topical steroids may reduce pruritus, and calcium channel blockers such as nifedipine have been used to reduce pain. Preventive measures aimed at keeping the extremities warm and dry during prolonged cold exposure may reduce the risk for developing­ chilblains.34

Figure 3G-24  Moderately severe frostbite of foot.    (From McCaulley RL, Smith DJ, Robson MC, et al: Frostbite. In Auerbach PS (ed): Wilderness Medicine, 4th ed. Philadelphia, Mosby, 2001.)

204

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 3G-25  Eczema detail.    (Courtesy of D. Stulberg.)

Figure 3G-26  Psoriasis of knee (detail).    (Courtesy of D. Stulberg.)

Sunburn

Eczema

Athletes competing outdoors may develop ­sunburn, which is damage to the skin induced by exposure to ultraviolet radiation. Sunburn occurs commonly in pale-skinned individuals who have a limited ability to produce melanin in response to ultraviolet light exposure. Athletes taking certain medications, including sulfonamides, tetracyclines, and phenothiazines, may be at increased risk for developing sunburn owing to the photosensitizing effects of these medications. Sunburn appears as confluent erythema of the skin exposed to sunlight. Athletes may develop vesicles and bullae in more severe cases of sunburn. Treatment of sunburn may include topical steroids, oral nonsteroidal anti-inflammatory drugs, and topical emollients. Prevention of sunburn is essential for athletes competing outdoors. Repeated blistering sunburns increase the chance of developing malignant melanoma. Using sunscreen lotions with a sun protection factor (SPF) of 15 is recommended. Athletes should consider using a waterresistant sunscreen during practice and competition outdoors. Loose-fitting clothing may be used to cover skin as well to reduce the risk for sunburn.33 Athletes, particularly those who have sustained recurrent blistering sunburns, should have periodic skin evaluations to screen for melanoma and other skin cancers.

Eczema is an inflammatory skin disorder that may be seen in conjunction with other atopic conditions, such as allergic rhinitis and asthma. Eczema has been referred to as the “itch that rashes.” Intense pruritus and dry skin are hallmarks of eczematous eruptions. The rash that develops appears as poorly defined erythematous plaques with lichenification or excoriations (Fig. 3G-25). Eczema tends to wax and wane, with general improvement in the moister summer months and exacerbation during the dry ­winter months. Secondary bacterial infection is common in eczema, which may limit an athlete’s ability to participate in sports. Maintenance of skin moisture and suppression of inflammation are important facets of eczema ­treatment. Application of topical moisturizing lotions 2 to 3 times daily is often recommended. Topical corticosteroids have long been used to treat eczema, although prolonged use of topical and fluorinated corticosteroids may induce skin atrophy and skin pigment changes. The use of other immunosuppressants has been studied as well. Oral cyclosporine can be used in severe cases of eczema recalcitrant to other treatment methods.35 Topical immunosuppressant therapy with pimecrolimus and tacrolimus has been effective in the treatment of mild to moderate ­eczema.36 Ultraviolet exposure has been shown to improve eczema as well.35

INFLAMMATORY SKIN CONDITIONS A variety of inflammatory skin conditions may impair an athlete’s ability to compete comfortably in sports. Eczema, psoriasis, urticaria, and acne vulgaris may affect athletes. Although these disorders may not directly disqualify athletes from competition, they may lead to skin breakdown, increasing the chance for secondary skin infection, which may disqualify an athlete from competition. Appropriate diagnosis and management of inflammatory skin disorders will ensure that athletes can continue to participate fully in sports.

Psoriasis Psoriasis is an autoimmune skin disorder characterized by the development of erythematous plaques with an overlying shiny silver scale (Fig. 3G-26). Psoriasis has a predilection for affecting the extensor surfaces of extremities, the scalp, and the nails. Koebnerization, that is, the development of new lesions at areas of friction or trauma, is a common feature of psoriasis. Athletes may be particularly affected by this phenomenon, developing new psoriasiform lesions at points of contact with their protective athletic gear. Psoriasis is sometimes associated with an ­inflammatory arthritis

Nonorthopaedic Conditions

205

This condition is characterized by pruritus and the development of large hives. Respiratory symptoms, angioedema, and hypotension may develop in true exercise-induced anaphylaxis. Treatment of exercise­induced anaphylaxis includes attention to airway, breathing, and circulatory support. Rapid administration of subcutaneous epinephrine may be lifesaving if laryngeal edema has developed. Transfer to an emergency facility for ongoing management of this problem is recommended. Use of antihistamines and mast cell stabilizing agents may prevent recurrence.38

Acne Vulgaris

Figure 3G-27  Cholinergic urticaria.    (From Habif TP: Clinical Dermatology, 4th ed. Philadelphia, Elsevier, 2004.)

that primarily affects the axial skeleton and the small joints of the hands. Various treatments exist for psoriasis. Topical treatments, including coal tar preparations, topical cortico­ steroids, selenium, and ultraviolet light, can be effective in controlling the dermatologic symptoms of psoriasis. Systemic therapies, including cyclosporine, methotrexate, and sulfasalazine, may be effective in controlling psoriasis as well. Newer biologic agents, including etanercept, infliximab, and alefacept, have been effective in controlling the skin manifestations of psoriasis. Additionally, infliximab and etanercept have been shown efficacious in treating psoriatic arthritis.37 As with eczema, secondary bacterial infection of psoriatic lesions is possible and may limit an athlete’s ability to compete in sport. Additionally, athletes affected by psoriatic arthritis may not be able to compete fully in sports until their arthritis symptoms are under adequate control.

Urticaria and Exercise-Induced Anaphylaxis Several types of urticaria may develop in athletes during exercise. Cholinergic urticaria develops as an individual’s core body temperature rises quickly. Typically, the hives of cholinergic urticaria are small, measuring 2 to 4 mm in diameter (Fig. 3G-27). Cholinergic urticaria is not associated with anaphylaxis or hypotension. Cold-induced urticaria also appears as small-diameter hives, which appear on cold-exposed skin. Cold-induced urticaria appears without other evidence of anaphylaxis. A diagnosis of cold-induced urticaria may be made by a cold challenge, where hives develop on skin that has rewarmed following exposure to an icepack or ice cube. Treatment of cholinergic urticaria and cold-induced urticaria includes the use of antihistamines and mast cell stabilizers. Exercise-induced anaphylaxis is a rare but potentially life-threatening condition seen in some athletes.

Acne vulgaris is a ubiquitous skin disorder among adolescents that may persist into adulthood. There are several subtypes of acne vulgaris. Accurate recognition of the particular subtype of acne will allow the clinician to choose an appropriate and effective treatment regimen. The clinician must also keep in mind that acne may develop in the setting of anabolic steroid abuse or certain systemic diseases such as polycystic ovary syndrome. A complete medical history and a thorough physical examination should be performed on athletes who present for treatment of acne. Acne is a disorder of the pilosebaceous unit. The sebaceous glands produce sebum in response to the rise in adrenal androgens during puberty. Plugging of the sebaceous gland with keratin allows for the retention of keratin and sebum, leading to the formation of a comedone. The comedone may become colonized with an anaerobic diphtheroid called Propionibacterium acnes. As the body responds to this infection, pustules or cysts may be produced. Acne is classified based on the predominant lesion present. Acneiform lesions are broadly categorized as inflammatory or noninflammatory. Open and closed comedones are noninflammatory lesions, commonly referred to as blackheads and whiteheads, respectively. Inflammatory acneiform lesions develop in the setting of infection of a closed comedone. Varying degrees of inflammation may be present, leading to development of pustules, papules, cysts, and nodules. A tremendous number of medications exist for the treatment of acne. Topical retinoids act as keratolytics and are therefore effective first-line medications for the treatment of noninflammatory lesions. Topical retinoids may also be important as adjuncts in treating the inflammatory lesions. Adapalene, tazarotene, and tretinoin are the most commonly used retinoids. All can cause erythema, pruritus, peeling, and photosensitivity as side effects. Topical antibiotics such as benzoyl peroxide, clindamycin, and erythromycin may be used to treat inflammatory lesions of acne. In widespread cases, oral antibiotics such as tetracycline, erythromycin, and minocycline may be used to suppress growth of P. acnes. Photosensitivity is a concern when using the tetracycline-derivative antibiotics. Oral isotretinoin is a systemic retinoid that may be used to treat severe cases of acne. Isotretinoin is teratogenic, so two forms of appropriate birth control must be used in females treated with isotretinoin. Additionally, isotretinoin can cause arthralgias and myalgias, which athletes may not tolerate.39

206

DeLee & Drez’s Orthopaedic Sports Medicine

S U G G E S T E D

SUMMARY Skin infections and injuries can affect an athlete’s ability to compete comfortably and safely in sports. Prompt recognition of skin conditions and appropriate treatment will allow athletes to return to sports safely. Medications used to treat some of the common skin conditions that affect athletes may produce side effects that interfere with an ­athlete’s ability to participate in sports. C l The

r i t i c a l

P

o i n t s

NCAA and NFHS have strict guidelines governing return to play for athletes with skin infections. l Athletes are at risk for developing community-acquired MRSA infections; abscesses and furuncles should be ­incised and drained, with the abscess fluid sent for ­culture and sensitivity. l Herpes simplex virus infections are common among athletes competing in wrestling. Treatment with antiviral medications and close monitoring of athletes for new lesions is required before returning an athlete to competition or practice. l Fungal infections of the skin may require long periods of treatment with topical antifungals. Fungicidal ­medications should be used preferentially over fungistatic drugs. l Abrasions and lacerations are common injuries in sports; physicians providing sideline medical care for athletic events should be prepared to manage these problems.

R E A D I N G S

Adams BB: Tinea corporis gladiatorum. J Am Acad Dermatol 47:286-290, 2002. Adams BB: Sports Dermatology. New York, Springer, 2006. Adams ES: Identifying and controlling metabolic skin disorders: Eczema, psoriasis, and exercise-induced urticaria. Physician Sports Med 32:29-40, 2004. Anderson BJ: The epidemiology and clinical analysis of several outbreaks of herpes gladiatorum. Med Sci Sport Exerc 11:1809-1814, 2003. Basler RSW, Hunzeker CM, Garcia MA: Athletic skin injuries: Combating pressure and friction. Physician Sports Med 32:33-40, 2004. Begier EM, Frenette K, Barrett NL, et al: A high-morbidity outbreak of methicillinresistant Staphylococcus aureus among players on a college football team, ­facilitated by cosmetic body shaving and turf burns. Clin Infect Dis 39:1446-1453, 2004. Bouchard M: Sideline care of abrasions and lacerations: Preparation is key. ­Physician Sports Med 33:21-29, 2005. National Federation of State High School Associations. Retrieved December 3, 2006, from http://www.nfhs.org/core/contentmanager/uploads/PDFs/Wrestling/ Physician_Release_for_Wrestlers.pdf. NCAA Rules Committee: Appendix D: Skin infections. In Wrestling: 2007 Rules and Interpretations, 2006. Retrieved December 3, 2006, from http://www.ncaa. org/library/rules/2007/2007_wrestling_rules.pdf. Snowise M, Dexter WW: Cold, wind, and sun exposure: Managing and preventing skin damage. Physician Sports Med 32:26-32, 2004.

R eferences Please see www.expertconsult.com

C H A P T E R  

 4

Exercise Physiology Joseph M. Hart, Donald E. Fowler, and David J. Lunardini Exercise physiology is the study of human systems’ response to acute and prolonged exercise. The field of exercise physiology evolved from anatomy and physiology, the study of human body structure and function. Scientists in this field study neuromuscular, cardiovascular, metabolic, and hormonal adaptations to physiologic stresses such as exercise or training across the life span, or in the presence of injury and disease. This chapter discusses the major concepts that guide the field of exercise physiology.

has been characterized by histochemical analysis of muscle fiber biopsy.2-4 Although a muscle will contain more than one or all types of motor units, the percentage of each fiber type within a muscle contributes to its function and fatigability. A muscle with more high-tension, fast-twitch motor units will be stronger and more powerful, whereas a muscle with fatigue-resistant, slow-twitch motor units will be able to sustain a contraction for longer periods of time.

Structure of a Skeletal Muscle

SKELETAL MUSCLE PHYSIOLOGY The human body contains three categories of muscle: smooth muscle, cardiac muscle, and skeletal muscle. Because the structure, location, and function of these muscle types differ, the focus of this section will be on skeletal muscle only. Grossly, skeletal muscles work together in coordinated fashion to generate force and cause joint motion, provide joint stability and protection, and attenuate external forces. Although it seems that skeletal muscles work as a single unit, their function is far more complex. Each skeletal muscle comprises different fiber types, categorized by structure and function. Individual muscle fibers have different rates of contraction, tension development, and susceptibility to fatigue.1 In the following sections, we classify muscle fibers as type I, type IIA, and type IIB. In general, type I muscle fibers operate at a slower twitch rate and are responsible for longer term, endurance-type activities owing to their ability to resist fatigue. Type IIA and IIB muscle fibers operate at a faster twitch rate and are responsible for high-force contractions that dominate during explosive, power maneuvers. Table 4-1 describes the distinguishing characteristics of these muscle fiber types. Muscle fiber content is homogenous among individual motor units, meaning that a single α-motoneuron will innervate muscle fibers with the same contractile characteristics in an all-or-none fashion. The distribution of the different fiber types within a skeletal muscle largely depends on genetic history and the function of that muscle1 and

Skeletal muscles are composed of bundles of fascicles and are grossly enclosed in connective tissue called epimysium. Each fascicle is encased by connective tissue called ­perimysium and contains multiple muscle fibers. The muscle fiber itself is an individual skeletal muscle cell. It is cylindrical in shape, multinucleated, and composed of bundles of myofibrils, each bundle surrounded by connective tissue called endomysium (Fig. 4-1). Myofibrils contain thick and thin protein filaments that form repeating dark and light bands along the length of the myofibril. Each repeated section is called a sarcomere, which is arranged end-toend within each myofibril. Sarcomeres are the functional and contractile component of skeletal muscle through a dynamic interaction between the muscles’ contractile proteins, actin and myosin. Alternating patterns of light and dark bands are created by alternating orientation of thin and thick contractile proteins, respectively, giving skeletal muscle its striated appearance (Fig. 4-2). Each sarcomere contains a structural lattice of thin protein filaments composed of spherical actin molecules in helical arrangement. Each actin molecule contains a binding site that is well protected by a thread-like protein filament, tropomyosin, and a stabilizing protein, troponin. Thick filaments are centered in the sarcomere and are composed of myosin molecules grouped together, each containing a globular end (myosin cross-bridges), that are staggered toward opposite ends of the thick filament. Each myosin molecule has a binding site located on the crossbridge with high affinity for binding with actin molecules.

   TABLE 4-1  Muscle Fiber Types and Distinguishing Characteristics Fiber Type

Contraction Speed

Force/Tension Production

Fatigue Rate

Mitochondria Content

Blood Supply

Fiber Diameter

Slow (type I)

Slow twitch

Low

Slower

High

Abundant

Small

Fast (type IIA)

Fast twitch

Moderate

Slower

Intermediate

Less abundant

Large

Fast (type IIB)

Fast twitch

High

Quicker

Sparse

Sparse

Large

Metabolism/ATP Production Oxidative; capable of aerobic activity Glycolytic/oxidative; capable of both aerobic and anaerobic activity Glycolytic; capable of anaerobic activity

207

208

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Axon of motor neuron Perimysium

Muscle fiber Sarcolemma Blood capillary

Fasciculus

Blood vessel Epimysium Perimysium Endomysium Blood vessels

Figure 4-1   Basic structure and organization of skeletal muscle with associated connective tissues, blood vessels, and motor neuron. (Redrawn from Palastanga NP, Field D, Soames R: Anatomy and Human Movement—Structure and Function. Edinburgh, UK, Butterworth Heinemann, 2006, p 17.)

The thick protein filament is surrounded at each end by a lattice of thin filaments where actin and myosin binding sites are in close proximity. This orientation of the thick filaments and thin filaments provides the structural framework for sarcomere shortening. Figure 4-2   Illustration of repeated sarcomere structure and organization within each individual muscle fiber. Each sarcomere contains thick (myosin) and thin (actin) contractile protein filaments that are responsible for sarcomere shortening. Alternating light and dark bands characterizing skeletal muscle are due to repeated bands of thick (A-band) and thin (I-band) filaments. The T-tubule system is contiguous with the sarcoplasmic reticulum and sarcolemma. Once an action potential from a motor neuron is initiated, it propagates along the sarcolemma and is internalized to the myofibrils within a muscle fiber through the T-tubule system. (Redrawn from Seeley RR, Stephens TD, Tate P: Anatomy and Physiology, 3rd ed. St. Louis, Mosby, 1995.)

Physiology of Skeletal Muscle Contractions Skeletal muscle contraction originates at the sarcomere. According to the sliding filament theory, active shortening of the sarcomere and hence the muscle results from the relative movements of the actin and myosin filaments past one another while each retains its original length. In its resting state, myosin cross-bridges and actin molecules are not in contact because of the orientation of troponin and tropomyosin molecules relative to binding sites on actin. When an action potential is propagated along a motor axon, ­acetylcholine is released from the axon terminal at the neuromuscular junction, increasing permeability to sodium and potassium ions, causing an end-plate potential. This endplate potential propagates along the muscle cell membrane (sarcolemma), which communicates throughout all myofibrils of the muscle fiber through the transverse tubule system, contiguous with the sarcolemma. Calcium ions are released from the terminal cisternae of the sarcoplasmic reticulum and quickly bind with troponin molecules located on the thin contractile filament, causing tropomyosin molecules to expose binding sites on actin. Myosin cross-bridges can now bind with actin molecules and collectively pull the actin filaments closer to the center of the sarcomere (Fig. 4-3). Myosin cross-bridges do not work in unison; rather, they each produce a “power stroke” that causes whole muscle shortening and force production. Each myosin cross-bridge power stroke causes the release of hydrolyzed adenosine triphosphate (ATP; adenosine diphosphate [ADP] + phosphate). The linkage between myosin and actin is broken by the hydrolysis of a new ATP molecule to prepare the myosin molecule for another power stroke or for relaxation. If no new ATP is available, myosin cross-bridges cannot break their connection, resulting in rigor mortis.

A

ba

nd

Ib

an

Sarcoplasmic reliculum

d

Sarcolemma

Myofibrils

Terminal cisterna Transverse tubule (T-tubule) Terminal cisterna Triad

Capillary

Mitochondrion

Exercise Physiology

209

Action potential Ca2+

Ca2+ ion

Sarcolemma Sarcoplasmic reticulum

Ca2+

Ca2+

T-tubule Myosin

Ca2+ ion

Actin

ADP P

ADP P

ADP

P

B Ca2+

Sarcomere

Ca2+

Ca2+ binds to troponin

Ca2+

ADP

ADP

Tropomyosin

ADP

Troponin

Ca2+

G-actin molecule

P

P

P

C

Ca2+

Ca2+

Ca2+

Myosin Ca2+ AD

AD

P

P

Active site

Cross bridge

AD

P

D Ca2+

Ca2+

Ca2+

A ATP

ATP

ATP

E Ca2+

Ca2+

ADP

P

ADP P

Ca2+

ADP

P

F Figure 4-3   Sequence of skeletal muscle contraction: sarcolemma depolarization causes calcium release from the sarcoplasmic reticulum. A, Calcium binds with troponin and shifts tropomyosin molecules to expose myosin-binding sites on actin. Myosin ­  cross-bridges bind to actin, producing a “power stroke” of contraction. Adenosine triphosphate is needed to break the link and prepare for the next cycle. Cycles (B to F) continue as long as sufficient calcium is present to inhibit the troponin-tropomyosin system from blocking actin-binding sites. (Redrawn from Seeley RR, Stephens TD, Tate P: Anatomy and Physiology, 3rd ed. St. Louis, Mosby, 1995.)

Cycles of myosin-actin binding and the resultant power strokes continue until the concentration of calcium ions is insufficient to inhibit the troponin-tropomyosin mechanism of blocking binding sites on actin molecules. Specific mechanisms of task failure during exercise will be discussed in a later section. The motor unit is the smallest portion of a muscle that can contract independently and consists of all the muscle fibers it contracts. A motor unit is innervated by a single motor neuron; therefore, all muscle fibers within a motor unit contract in an all-or-none fashion. Force production by a skeletal muscle can be voluntarily graded through recruitment of more or less motor units. As more motor

units are recruited in a particular skeletal muscle, the muscle produces more force.5 During a graded contraction, motor units are recruited from smaller to larger (i.e., motor units with fewer muscle fibers are recruited before motor units with more muscle fibers). This is known as the Henneman principle of motor unit recruitment. All muscle fibers within a motor unit have the same metabolic characteristics (fiber type). However, skeletal muscles can be composed of different types of muscle fibers based on the metabolic demand of the muscle.5 Muscle fiber type composition in human skeletal muscle is genetically determined.5 There is little information regarding the ability of a muscle fiber type to transform in response

210

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 4-4   Overall depiction of energy metabolism illustrating how carbohydrates, fats, and proteins enter the pathway to produce adenosine triphosphate (ATP). FFA,   free fatty acid; G6P, glucose6-phosphate; CoA, coenzyme A; TCA, tricarboxylic acid, also known as citric acid cycle or Kreb’s cycle. (Redrawn from Robergs RA, Roberts SO: Exercise Physiology—Exercise, Performance, and Clinical Applications. Dubuque, Iowa, William C. Brown Publishing, 1997, p 60.)

Amino acids

Glucose

FFA

Glycogen Glycogenolysis

G6P Glycolsis

Lipid droplets

Glycerol FFA

ATP Lactate

Pyruvate

CO2 AcetylCoA

AcetylCoA

Mitochondria

β-oxidation

e– H+

TCA cycle

ATP ATP

e– H+ CO2

ATP O2 H2O

to training or exercise. However, heavy resistance training caused increased type IIA and decreased type IIB fibers, whereas type I fiber composition in human skeletal muscle was unchanged.6 Structural and genetic characteristics of muscle fiber types have been modulated with fiber-specific stimulation in vitro.7 However, it is unknown whether such changes occur in all muscle fiber types or if the transformation will be sustained over time in vivo. Because muscle fiber composition is genetically determined, athletes may participate in sports or activities that involve muscle contractions that are more “natural” to their muscle fiber contribution. Whether a distance runner can train to be a successful powerlifter or vice versa remains an unanswered but interesting question.

Energy Metabolism Efficient production, storage, and utilization of energy are essential for skeletal muscle contraction. To perform exercise, the cells in the body must convert energy from food and body stores into a specific form of usable energy, ATP. ATP is the immediate source of energy for body functions. Whatever the fuel source—carbohydrate, protein, or fat—the primary goal is to make ATP. Energy in the form of ATP is created through three pathways: the ATP-phosphocreatine system and glycolytic and oxidative systems (Fig. 4-4).

ATP-Phosphocreatine System At rest, a muscle cell has only small amounts of ATP that can be used for very short duration contractions. In the ATP-phosphocreatine system, ATP is quickly released from phosphocreatine (also called creatine phosphate) as the enzyme creatine kinase facilitates the release of a phosphate group. Phosphocreatine is available in limited quantity in muscle cells to quickly replenish ATP stores. Because of these constraints, the ATP-phosphocreatine production system is short-lived and typically used for short-­duration, high-intensity activity. When ATP production from phosphocreatine is insufficient for the imposed demands, glucose (carbohydrate) is metabolized. Further ATP production is dependent on proper metabolic and oxidative pathways to sustain muscle activity after intrinsic stores of high-energy compounds have been exhausted.

Glycolytic System ATP is also created through the breakdown of glucose (glycolysis), which ultimately produces pyruvate. With exercise, skeletal muscle glucose uptake can increase as much as 28-fold, depending on the intensity of activity.8 Glucose is supplied by either circulating glucose or muscle glycogen, with transport into cells facilitated by

Exercise Physiology

c­ arrier-mediated diffusion, driven by a concentration gradient. Although transformation of glucose to pyruvate does not require oxygen, the fate of pyruvate depends on oxygen availability. The glycolytic system involves the process of energy production in the absence of oxygen, termed anaerobic glycolysis. Specifically, when oxygen is not available, energy is created by breaking down pyruvate into lactic acid. The net gain is 3 moles of ATP per mole of glycogen (muscle stores of glucose) and 2 moles of ATP per mole of circulating glucose. Unfortunately, this energy system is short-lived and serves energy needs, supported by the ATP-­phosphocreatine energy system, for about 1 to 2 minutes of high-intensity exercise. Pyruvate conversion into lactic acid lowers the pH and creates a relative state of muscle fiber acidification, where glycolytic enzymes are unable to continue catalyzing the production of ATP. During high-intensity and strenuous exercise, skeletal muscle cells derive most of their energy from this anaerobic glycolysis.9 Fatigue is believed to develop during high-intensity performance, in part owing to this lactate accumulation, along with depleted stores of glycogen and plasma glucose.10 The burning sensation felt after short-duration, highintensity exercise is believed to come from lactic acid accumulation in anaerobic glycolysis. The point during exercise when production exceeds the ability of the body to metabolize the byproducts of anaerobic glycolysis (i.e., lactic acid) is termed lactate threshold and can be measured invasively (blood) or noninvasively (respiratory gases).

Oxidative System The process by which the body breaks down dietary fuels into usable energy with the aid of oxygen occurs in cell mitochondria. In muscle cells (fibers), mitochondria are adjacent to myofibrils and located throughout the sarcoplasm. In this section, we discuss the pathway of each nutrient—carbohydrates, fats, and proteins—in creating energy through the oxidative system.

Carbohydrates Glucose is the major fuel source for the body during exercise, obtained mainly through dietary carbohydrates such as wheat flour and sugars. Carbohydrates typically provide 50% of daily calories in the average American diet. Larger disaccharides (sucrose and lactose) are broken down into their simpler units (glucose, fructose, and galactose) by amylases to allow for intestinal absorption. Fructose and galactose are quickly converted to additional glucose, and the substrate enters the circulation for either utilization or storage, depending on the body’s state of activity. Under resting conditions, glucose is primarily stored in the form of multiple, branching polymers called glycogen. The two major depots for glycogen are the liver and skeletal muscle. Although the liver has a higher concentration of glycogen (65 g/kg tissue versus 15 g/kg tissue), skeletal muscle has a much larger quantity because of its overall mass.11 Glycogen stores have been shown to be higher in actively trained individuals compared with those living a sedentary lifestyle.12

211

Fats Dietary fats typically constitute 40% of caloric intake. Fats are mainly stored in adipose tissue as triglycerides, constituting the body’s primary energy reserve. Triglycerides are formed when three molecules of free fatty acid (FFA) are attached to a glycerol backbone. Transfer of these triglycerides from the intestine to adipocytes is accomplished through the bloodstream by chylomicrons and lipoproteins. These droplets are coated by cholesterol and phospholipids and vary in fat composition. During prolonged exercise, fat utilization increases through the process of β-oxidation. Triglycerides are initially broken down through lipolysis into individual FFAs. FFAs undergo β-oxidation producing one molecule of acetyl-coenzyme A (CoA), and the process repeats itself until the entire fatty chain is cleaved into two-carbon acetylCoA units. Metabolism then proceeds in the same fashion as glucose oxidation, with acetyl-CoA entering the citric acid cycle and reducing equivalents entering the electron transport chain, to drive ATP synthesis through oxidative phosphorylation. The energy yield from this pathway is high. For example, one 16-carbon FFA would generate a total of 129 molecules of ATP.

Protein Proteins include 20 directly translated amino acids and come mainly from meat, milk, and wheat. They are not considered a primary fuel source at rest or during exercise of any intensity, unless the individual’s nutritional status is impaired.13 Under such circumstances, protein can be converted to glucose through gluconeogenesis, or into other intermediates of energy metabolism such as pyruvate or acetyl-CoA to enter the oxidative energy system. However, nitrogen, a large component of protein, cannot be oxidized, and therefore energy yield from protein is fairly small compared with carbohydrates and fats. In the presence of oxygen, pyruvate is transported into the mitochondria and oxidized by pyruvate dehydrogenase into acetyl-CoA. This process of creating acetyl-CoA requires oxygen and is hence termed aerobic glycolysis. Each acetyl-CoA enters the mitochondria for a complex series of chemical reactions collectively called the citric acid cycle (also known as Krebs’ cycle). Krebs’ cycle is the central oxidative pathway for all metabolic fuels, including carbohydrates, fats, and proteins. By the end of the cycle, each acetylCoA is broken down into carbon dioxide and hydrogen, and 2 moles of ATP are created. Hydrogen from the initial process of glycolysis (transformation of glucose to pyruvic acid) and from Krebs’ cycle combine with two coenzymes, which act to carry hydrogen atoms to the electron transport chain. Hydrogen atoms are split: protons combine with oxygen to form water (preventing acidification), and electrons are transported through a series of reactions for the phosphorylation of ADP into ATP. Because this process requires oxygen, it is termed oxidative phosphorylation. Overall, depending on the specific mechanism of aerobic metabolism, the sum total of energy produced from one glucose molecule is up to 39 ATP. The balance of aerobic and anaerobic breakdown of glucose is ultimately related to the oxygen content of the muscle mitochondria and

212

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

therefore is dependent on the fitness level of the person and the intensity of activity performed.14

be measured indirectly through ­ respiratory exchange of ­carbon dioxide and oxygen, termed indirect calorimetry (Fig. 4-5). During intense exercise testing, maximal oxygen uptake (Vo2max) is used in diagnostic and experimental settings and is considered the best measurement of cardiorespiratory endurance and aerobic fitness. In review, the need for ATP is met through stored or dietary fuel sources: carbohydrates, fats, and proteins. Intensity and duration of exercise dictate the proportion of energy drawn from each of these nutrients, with carbohydrates and fats serving more prominent roles in energy production. When oxygen is abundant, as with prolonged exercise at lower intensities, fatty acids serve as the predominant fuel type.15 However, during times of higher-­intensity exercise with oxygen depletion, glucose

Aerobic Capacity A muscle’s oxidative capacity is determined by mitochondria content (see muscle fiber type descriptions earlier in this chapter) and presence of oxidative enzymes. Metabolism of nutrients into energy depends on oxygen availability, and produces carbon dioxide and water. Therefore, the amount of oxygen and carbon dioxide exchanged in the lungs equals that used and released by body tissues. In laboratory settings, the volume of oxygen consumed (Vo2) and carbon dioxide expired (Vco2) are measured breath by breath, or over a finite time period. Thus, energy ­expenditure can

A/D convertor Computer

Inspired air

Expired air

O2 analyzer Flow meter

Moisture remover

CO2 analyzer

Mixing chamber

A

Subject

Pneumotach

Inspired air

Expired air Moisture remover Computer High Performance

CO2 analyzer

Subject

O2 analyzer

A/D convertor

B Figure 4-5   Indirect calorimetry is used to determine oxidative capacity through time-averaged (A) or breath-by-breath (B) systems. Results can be used to project aerobic fitness levels. (Redrawn from Robergs RA, Roberts SO: Exercise Physiology—Exercise, Performance, and Clinical Applications. Dubuque, Iowa, William C. Brown Publishing, 1997, p 140.)

Exercise Physiology

is preferentially used because its oxidation requires less ­oxygen per mole than do fatty acids. Essentially, glucose tends to be spared in favor of fatty acids until oxygen availability is limited.14

213

contraction is characterized by controlled and resisted lengthening of a muscle against a load. Eccentric muscle contractions (often referred to as “negatives” in weightlifting) are more effective in producing strength gains and hypertrophy than concentric contractions.16 However, both eccentric (negative) and concentric (positive) contractions elicit gains in skeletal muscle strength and size.16 Eccentric muscle contractions produce greater muscle force and more myofibrillar disruption than concentric exercise.17

Types of Skeletal Muscle Contraction Skeletal muscle contractions produce muscle tension and control body and joint movement and are capable of different types of contractions based on activity or demands placed on the skeletal system. Isometric muscle contractions produce muscle tension without joint movement; for example, pushing against a wall or contracting the quadriceps muscle while holding the knee motionless at a particular point in the knee range of motion. Isotonic muscle contractions produce muscle tension and joint movement against a constant load, at which rate of movement is variable. For example, a dumbbell curl is a contraction against a constant load that can be voluntarily moved at a self-selected rate. This is the most typical contraction in weightlifting. Isokinetic muscle contractions involve a constant rate of joint displacement that is maintained by varying amounts of resistance based on muscle effort. This is uncommon in weightlifting or athletic settings. Isokinetic exercise requires expensive machinery and is usually most applicable in the rehabilitation setting. Isotonic and isokinetic muscle movements can be performed through concentric or eccentric muscle contractions. Skeletal muscle can generate torque to produce joint movements through concentric and eccentric movements (Fig. 4-6). A concentric contraction generates force and movement as the muscle shortens. Concentric contractions describe a voluntary muscle contraction characterized by muscle shortening against a load. Conversely, an ­eccentric

MUSCLE RESPONSE TO TRAINING Improvements in muscle strength, power, or endurance are best achieved by overloading the muscle being trained. Skeletal muscles respond to training with improved strength and endurance, depending on the imposed demands of the training exercise. Skeletal muscle adaptations to training may include initial neuromuscular changes, hypertrophy, hyperplasia, and fatigue (Table 4-2). The overload principle states that when a muscle is exposed to a stress or load that is greater than what it usually experiences, it will adapt so that it is able to handle the greater load.5,18,19 Similarly, the SAID principle (specific adaptations to imposed demands) states that a muscle or body tissue will adapt to the specific demands imposed on it. For example, if a muscle is overloaded, its fibers will grow in size so that it is able to produce enough force to overcome the imposed load.5,19 Observed strength gains within the first few weeks of a weightlifting program are mostly due to neuromuscular adaptations.20 As exercise intensity increases and muscles begin to fatigue, the nervous system recruits larger motor units with higher frequencies of stimulation to provide the force necessary to overcome the imposed resistance.5 Early

Flexion Concentric

Isokinetic contraction 120 degrees/sec

Flexion

Torque (Nm)

Extension

20 degrees/sec

Eccentric Extension

90 75 60 45 30 15 0 15 0 Joint angle (degree) Figure 4-6   Concentric and eccentric muscle contraction of the biceps brachii through a dumbbell curl. Concentric contractions occur with flexion (muscle shortening), whereas eccentric contractions arise from extension (muscle lengthening). Greater force and strength gain are produced by eccentric contractions. Isokinetic contractions typically require specialized machinery to measure torque. The torque curves are recordings of isokinetic muscle contractions at two consistent speeds. Torque is greatest at the midrange of motion and least at the start and end of the contraction, when joint angles affect muscle angle of pull, limiting force generation. (Redrawn from Robergs RA, Roberts SO: Exercise Physiology—Exercise, Performance, and Clinical Applications. Dubuque, Iowa, William C. Brown Publishing, 1997, p 160.) 90

75

60

45

30

214

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 4-2  Characteristics of Common Skeletal Muscle Responses to Exercise Neuromuscular adaptations Hypertrophy

Hyperplasia

Fatigue

Early strength gains are largely due to neuromuscular adaptations typically involving more efficient motor unit recruitment and coordination. Increased muscle cross-sectional area is due to increased contractile protein synthesis in response to imposed demands of resistance training and muscle overloading. Muscle cell “splitting” contributes little to increased muscle cross-sectional area in response to training; however, long-term, ­very-high-intensity resistance training may involve some degree of hyperplasia. Reduced force production during prolonged or intense exercise has central nervous system and peripheral causes leading ultimately to task failure.

strength gains and increased muscle tension production from training result from a more efficient neural recruitment process.21

Muscle Hypertrophy and Hyperplasia Human muscle hypertrophy occurs when the cross­sectional area of a muscle group increases.21,22 As a skeletal muscle hypertrophies, contractile proteins are synthesized,20 and the muscle is therefore capable of producing more force. Type IIA fibers exhibit the greatest growth, whereas type IIB and type I exhibit the least amount of growth in response to heavy resistance training.20 Muscle hypertrophy is more common in fast-twitch than slow-twitch muscles. Strength training leads to muscle hypertrophy, which increases muscle mass21 and typically occurs after weeks of such training.18,20 Immediate improvements in strength observed shortly after initiation of a training ­program are likely due to neuromuscular adaptations (Fig. 4-7). However, there appears to be a gender difference in the rate at which muscles hypertrophy favoring males, and the rate at which muscle mass is lost due to detraining favoring females.23 Resistance-trained muscles hypertrophy to adapt to greater imposed loads.24 A less common muscular adaptation contributing to increased muscle cross-sectional area is muscle cell division or hyperplasia. Extremely heavy weight training caused increased muscle fiber number through fiber splitting in cats.25 In theory, split muscle fibers can hypertrophy independently and contribute to muscle size increase; however, most increases in muscle size and mass in humans is attributed to muscle fiber hypertrophy. Early strength and endurance gains following the initiation of a structured exercise program appear to be more influenced by neural factors. In the absence of hypertrophy; improved ­coordination, learning, increased and more synchronized voluntary motor unit activation, and other central and peripheral nervous system factors likely contribute to short-term strength gains. Because it takes time to synthesize new contractile proteins, later strength gains are largely due to muscle hypertrophy.

Fatigue Muscle fatigue is described as the reduced force-producing capacity of a muscle during and following exercise that may lead to impairment of performance.26,27 The reduction of force leads to task failure and may result from a combination of factors occurring at or distal to the neuromuscular junction (peripheral fatigue) or from other physiologic or psychological factors (central fatigue) leading to the progressive reduction in voluntary activation of muscle during exercise.26,28 Peripheral causes of force reduction may be due to substrate accumulation from metabolism, such as the accumulation of inorganic phosphate from phosphocreatine metabolism or lactic acid accumulation from glycolysis during prolonged or intense exercise. Reduced availability of ATP and other factors leading to a disturbance in calcium release from the sarcoplasmic reticulum may also contribute to force reductions at the muscle.1,26,29 Central factors contributing to task failure include spinal mechanisms, such as reduced α-motoneuron excitability or a decrease in firing frequency of individual motor units, and supraspinal factors such as motivation, motor cortical excitability, or other biochemical influences.1,26,29 Noninvasive measures of central versus peripheral fatigue can be performed through artificial, electrical stimulation of fatigued and nonfatigued muscle. If the characteristics of an electrically induced muscle contraction are different before and after fatigue, the causes are primarily peripheral, whereas if the force of a voluntary maximal contraction is reduced before and after fatigue and the characteristics of an electrically induced maximal contraction are similar, the causes are most likely central.29 Differentiation between spinal and supraspinal influences on central fatigue is more complicated and may involve psychological or biochemical influences that are more difficult to measure. When a muscle is fatigued, there are fewer motor units available to call on during muscle contractions so that force reduction will result from reduced motor unit activation.30

NEUROMUSCULAR ADAPTATION TO EXERCISE The motor unit is the functional unit of movement and consists of a motor neuron and the muscle fibers that it innervates (Fig. 4-8). Increased force generation is achieved with greater numbers of motor units and rates of firing, both of which correlated to the response seen after resistance exercise.31,32 Strength gains may occur within days of starting resistance training, despite the fact that muscle hypertrophy does not occur for weeks or months. This phenomenon is explained by neural adaptations causing increased generation of force. Reduction in the coactivation of antagonist muscles has also been associated with increased strength in response to exercise,33 as have adaptations leading to increased central motor drive, elevated motoneuron excitability, and reduced presynaptic inhibition (Table 4-3).34 Resistance training for a period of 6 to 8 weeks or longer leads to two primary changes within muscle. The first change results from increased protein synthesis and leads to

Exercise Physiology Strength gain due to neural factors

Strength gain due to hypertrophy

BEFORE TRAINING

IEMG

INCREASED ACTIVATION NO CHANGE IN RATIO E/F

IMPROVED RATIO E/F NO CHANGE IN ACTIVATION

215

Figure 4-7   Strength gain broken down into components of neural factors (A) and muscle hypertrophy (B). Diagram C shows combined effect. Neuromuscular adaptations are likely responsible for immediate gains, whereas muscle hypertrophy produces stronger effects after several weeks of strength training exercise. IEMG, integrated electromyogram. (Redrawn from Moritani T, deVries HA: Potential for gross muscle hypertrophy in older men. J Gerontol 35:672-682, 1980.)

AFTER TRAINING

A

B

force

IEMG

Evaluation of % contributions of neural factors (N.F.) vs hypertrophy (M.H.)

C

A force

B

% M.H. =

B–A × 100 C–A

% N.F. =

C–B × 100 C–A

C

muscular hypertrophy, as seen by increased cross-­sectional area. This occurs in all fiber types within the muscle and results in increased overall force output. The second change involves switching between the types of fast twitch fibers from IIb to IIa fibers.35 This change results in increased fatigue resistance at the expense of some force output. The net result of resistance training is typically increased ­muscular strength in the form of improved force/tension development during contractions, with small improvements in endurance seen as resistance to fatigue. The neural response to endurance training involves strategies to improve efficiency such as reduced number of motor units activated to maintain force and increasing activation of synergistic muscles. The goal of endurance training is to improve aerobic performance. Endurance training causes increased capillary density and decreased muscle cross-sectional area to facilitate the delivery of oxygen to tissues. In addition, endurance training leads to an increase in cellular mitochondrial content and oxidative enzyme activity, also increasing aerobic capacity. Other changes found include increased contraction velocity and power of slow-twitch fibers, as well as an overall increase in slow type I fiber concentration.36 Although it was stated

previously that type I fiber levels are generally static, there are small amounts of hybrid muscle fibers that can be converted to a pure slow- or fast-twitch fiber. The net results of long-term endurance training are increased metabolic capacity of muscle, increases in muscular force output, and muscle hypotrophy.

Delayed-Onset Muscle Soreness Immediately after exercise, the muscle swells and may reduce resting length and lose the ability to generate force, particularly as the muscle fatigues. Muscle soreness following intense exercise may be delayed for days as ­intramuscular edema collects, hence the phrase delayed-onset muscle soreness (DOMS) used to describe this typical discomfort. Immediate muscular swelling is believed to result from increased pressure within the extracellular matrix and the myofibrils themselves. It has long been shown that myofibril protein structure is damaged by exercise.37 This damage may be accompanied by an influx of calcium ions to the myofibril, with resultant pressure and edema. Additionally, cellular damage and the acute inflammatory response contribute to leakage of intracellular proteins to the ­extracellular space,

216

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 4-8   The motor unit consists of a motor neuron and all the muscle fibers it innervates. Muscle fiber type is homogenous within a motor unit so that firing of the neuron produces contraction in an all-or-none fashion. (Redrawn from Palastanga NP, Field D, Soames R: Anatomy and Human Movement—Structure and Function. Edinburgh, UK, Butterworth Heinemann, 2006, p 21.)

Motor neuron Muscle fibers

Myofibrils Muscle cell nucleus Axon Schwann cell Sarcoplasm Myelin sheath Axon

Collagen fibers

Axon

Schwann cell

Mitochondrion Primary synaptic cleft

Prejunctional membrane

Sarcoplasm

Postjunctional membrane

Secondary synaptic cleft

most notably creatine kinase. Serum creatine kinase levels peak several days after exercise with increasing levels of inflammation. Decreased resting muscle length a­ccompanies this initial muscular edema, as the swollen tissue pushes against the fascia and passively shortens muscles.38 Increased muscle edema and Doms constitute the primary skeletal muscle response in the days to weeks after exercise. Although precise ­mechanisms for these reactions are not fully understood, the ­inflammatory response plays a key role. DOMS is the feeling of soreness after highforce, eccentric exercise39-41 and usually peaks about 24 to 48 hours after exercise39,41-43 with resolution at about 5 to 7 days.39,41 DOMS is associated with muscular pain, swelling, and decreased muscle endurance and force production that can be detrimental to athletic performance.44 Eccentric exercise has been reported to significantly contribute to the onset of DOMS.43-46 The onset of the chronic inflammatory phase correlates with the onset of delayed muscle soreness, and the peak of chronic inflammation generally correlates with maximal muscle edema on about postexercise days 5 to 7. Although several theories on the underlying mechanism of DOMS following eccentric exercise have been suggested,39,43-45 the development of DOMS may resemble the sequence of events observed in the acute inflammatory cycle.43,47,48 Within a few hours after muscle damage, biochemical infiltrates begin to collect in the area of muscle injury.43,45 ­Macrophage accumulation, which is ­dependent

Mitochondria

on this ­ initial white blood cell migration to the injury site, ­contributes considerably to activation of nociceptors through prostaglandin production and edema accumulation.45 Postexercise edema is considered secondary to increased vascular permeability associated with pronounced cytokine release. There are some data correlating pain to mast cell degranulation and histamine within the muscle tissue,49 but a definitive cause of pain remains unclear.

   TABLE 4-3  Neuromuscular Adaptations to Exercise Resistance Training

Endurance Training

Early

Increased number of motor units firing Increased rate of motor unit firing Decreased coactivation of ­antagonistic muscles

Fewer motor units are required to maintain a given force Increased activation of synergistic muscles

Later

Muscular hypertrophy IIb → IIa fiber type switching

Muscular hypotrophy Increased mitochondrial density and aerobic enzyme activity Increased muscular capillary density Increased type I fiber concentration

Exercise Physiology

HORMONAL ADAPTATIONS TO EXERCISE Vast arrays of changes occur in response to exercise, many of which are hormonally mediated. Adrenal catecholamines, glucocorticoids, and mineralocorticoids drive most key changes, although many other hormones are involved. In this section, we discuss hormone-mediated responses to exercise (Table 4-4).

Pituitary Hormones Serum levels of human growth hormone (GH) have been shown to increase proportionally to exercise intensity, regardless of training modality.50 Levels increase within a few minutes of exercise onset, and return to normal by 1 hour after cessation.51 GH functions to stimulate protein synthesis and promote lipid metabolism, which may serve to conserve blood glucose levels during exercise. Because norepinephrine stimulates GH release, increased levels could be secondary to rising catecholamines described later. Increased β-endorphin release has been demonstrated with both brief and extended exercise above 60% Vo2max.52 β-Endorphins are active at opioid receptors and are responsible for analgesic effects as well as the “natural high” that can be experienced with exercise. Levels peak within 10 minutes of exercise completion and can reach up to 5 times resting levels. Posterior pituitary secretion of ADH is strongly stimulated by exercise, which makes physiologic sense as a preventive mechanism to avoid excessive dehydration with exercise. Prolactin levels are increased acutely in response to anaerobic exercise,53 although the role of such increases is unclear. Some support exists for acute increases with resistance training, but research is conflicting, and a clear understanding does not exist.

Adrenal Hormones Glucocorticoids, primarily cortisol, are released from the adrenal cortex in response to pituitary adrenocorticotropic hormone (ACTH). Levels of both ACTH and cortisol have been shown to rise in response to resistance and endurance training. Rates of increased secretion are proportional to exercise intensity, beginning above 50% Vo2max.54,55 Cortisol functions to stimulate liver gluconeogenesis, adipose lipolysis, and protein degradation in both liver and predominantly type II muscle fibers.56 No changes in basal rates of ACTH or cortisol have been shown with long-term training, although like many other hormones, the response to a single episode of exercise is diminished in the trained athlete as compared with an untrained control.57 Aldosterone levels also increase in an intensity-related fashion with exercise,58 and function at the kidney to prevent sodium and water loss. Aldosterone secretion is driven by kidney renin production, which in turn is driven by decreased renal blood flow accompanying dehydration and shunting of blood to skin and muscle during exercise. Response can be significant, with marathon runners having documented increases in aldosterone levels for almost a full day after completion of a race.59 Catecholamine (epinephrine and norepinephrine) ­se­c­retion from the adrenal medulla increases with exercise. Like the other adrenal hormones, increased secretion is directly related to exercise intensity.60 Norepinephrine and epinephrine secretion begins at low exercise intensities (below 50% Vo2max) and correlates directly with exercise intensity (Fig. 4-9).61 These hormones have a wide range of effects that are beneficial in the exercise setting, including increased heart rate and respiratory stimulation to meet increased oxygen demands, generalized ­vasoconstriction with vasodilatory effects on cardiac and skeletal muscle vessels to improve perfusion of active

   TABLE 4-4  Hormonal Responses to Exercise Hormone

Change

Function

Increases proportionally to exercise intensity Increase with exercise above 60% Vo2max Increases with exercise Increases with exercise

Stimulates metabolism Analgesic, “natural high” Maintains hydration Unclear

Increase with exercise above 50% Vo2max

Stimulates metabolism

Increases proportionally to exercise intensity Increase proportionally to exercise intensity*

Maintains hydration Mediate cardiovascular responses, maintain

Decreases proportionally to exercise duration (and blood glucose levels)* Increases proportionally to exercise duration

Guards against hypoglycemia

Increases proportionally to exercise intensity* Increase with exercise*

Unclear Unclear

Pituitary

Growth hormone β-Endorphins Antidiuretic hormone Prolactin Adrenal

Glucocorticoids, adrenocorticotropic hormone Aldosterone Epinephrine and   norepinephrine Pancreas

Insulin Glucagon

Increases serum glucose levels

Gonadal

Testosterone Estrogen and progesterone

*Basal levels also decrease with long-term endurance training.

217

218

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

to exercise, which may be a result of decreased catecholamine release.67

2.5 Norepinephrine

Gonadal Hormones

Catecholamines (mg/mL)

2

Testosterone increases acutely in response to resistance training or acute bursts of maximal energy, increasing with the intensity of exercise.60 Alternatively, endurance training shows unclear results in the acute setting, but results in reduced basal testosterone levels to 60% to 85% of normal with extended periods of training.68 Estrogen and progesterone levels have been found to increase acutely with exercise, independent of changes in follicle-stimulating hormone and luteinizing hormone levels.69 These increases are modified and confounded by the luteal phase at the time of exercise. Chronic training has been associated with decreased levels of estrogen and progesterone, along with subsequent menstrual irregularities or amenorrhea.

1.5

1

Epinephrine

0.5

0

0

20

40 60 % Vo2max

80

100

Figure 4-9   Catecholamine response in relation to exercise intensity. (Redrawn from Robergs RA, Roberts SO: Exercise Physiology—Exercise, Performance, and Clinical Applications. Dubuque, Iowa, William C. Brown Publishing, 1997, p 355.)

t­ issue, decreased renal blood flow (RBF) ­ stimulating renin and aldosterone to protect against dehydration, decreased insulin and increased glucagon secretion to maintain blood glucose levels, and increased lipolysis for additional energy. Unlike many other hormones discussed thus far, basal levels of catecholamines do change with prolonged training. Basal catecholamine levels decrease, and maximal hormone output level increases. This leads to decreased heart rate and blood pressure both at rest and at a given level of exercise, as well as increased capacity for high-intensity exercise. Levels begin to decline almost immediately with training, and reach their nadir after 4 to 5 weeks.62

Pancreatic Hormones Levels of insulin secretion decrease acutely with all forms of exercise, and the decrease is proportional to the duration of exercise rather than the intensity.63,64 Decreased insulin production is due to reduced blood glucose levels. Muscle cell uptake of glucose is enhanced during exercise despite these decreased insulin levels. This response involves catecholamine-driven uptake, exercise-induced receptor sensitivity, and increased delivery of glucose secondary to increased muscular blood flow.65 Longer-term training leads to decreased basal insulin rates and increased tissue sensitivity to insulin.66 Glucagon levels increase during exercise to maintain higher levels of blood glucose through stimulation of liver gluconeogenesis and glycogenolysis. Like insulin, changes in glucagon levels are related to exercise duration and are mediated, at least in part, by catecholamine effects on α-adrenergic pancreatic receptors. Long-term ­ training leads to decreased secretion of plasma glucagon in response

CARDIORESPIRATORY RESPONSE TO EXERCISE Regular exercise has profound effects on the human body, decreasing cardiovascular risk and improving overall quality of life. Physiologic response to dynamic exercise is ­primarily focused on supplying sufficient fuel and oxygen to activated muscle. Subsequently, increases in ventilation, oxygen extraction, and cardiac output accompany a redistribution of blood flow to the skeletal muscle to maximize performance.

Exercise Capacity Maximum oxygen uptake, or Vo2max, is generally used to provide an overall assessment of exercise capacity (Fig. 4-10). The Fick equation defines Vo2max as the product of maximal cardiac output and the difference between arterial and venous oxygen. In healthy individuals, cardiac output and peripheral oxygen extraction are tightly coupled to facilitate oxygen delivery and utilization within active skeletal muscle.70 Oxygen consumption increases linearly with exercise intensity, before reaching a plateau where no further oxygen uptake occurs despite continued increases in workload.71 Vo2max increases with training and decreases with age.72 Some studies have shown lactate threshold to be a more accurate predictor of sustained exercise performance when compared with Vo2max.73 The lactate threshold is defined as the point in oxygen uptake (Vo2) at which pyruvate level exceeds the ability to be metabolized through Krebs’ cycle, leading to a constant rise in blood lactate.74 This threshold varies with cardiovascular fitness, occurring at 40% of predicted Vo2max in normal individuals, compared with 80% to 90% of Vo2max in endurance athletes.75

Cardiovascular Response to Exercise The heart and blood vessels are responsible for increased transport of oxygen to the periphery, satisfying muscles’ nutrient demand and waste disposal requirements during

Exercise Physiology 80

75

Vo2 MAX (ml/kg-min)

70 58

60

58

55

50

45

40

35

30

Endurance Runners 18-28

Sprinters 18-24

Athletes 18-28

Sedentary Endurance Sedentary 18-28 Runners 40-50 40-50

YEARS Figure 4-10   Contrast of maximum oxygen uptake (Vo2max) among men with various ages and aerobic fitness levels. (Data from Pollock ML, Wilmore JH, Fox SM: Health and fitness through physical activity. New York, John Wiley and Sons, 1978.)

exercise. Pacing this system is the cardiac cycle. Cardiac output is defined as the product of heart rate and stroke volume, with stroke volume classified as the amount of blood ejected per ventricular beat. Aerobic capacity is largely determined by cardiac output, increasing about 5 times

219

resting level during heavy exertion. Normal adults exhibit a baseline cardiac output of 5 L/min, escalating to 25 L/min with maximal exercise.76 Cardiac output increases primarily as a function of increased heart rate, and to a lesser extent increased stroke volume. Resting heart rate is typically between 60 and 100 beats per minute, and increases linearly with exercise intensity to a maximal heart rate of 200 beats per minute in the young adult (Fig. 4-11A).77 Maximal heart rate decreases with age and is commonly predicted by the formula: maximal heart rate = 220 − age (in years). Rapid increases in heart rate during exercise result initially from decreased vagal input, followed by subsequent increases in sympathetic tone and circulating catecholamines.14 Stroke volume increases early in exercise, but unlike heart rate, plateaus before maximal intensity at about 40% to 60% of Vo2max (see Fig. 4-11B).78 Increases in stroke volume are largely attributed to increased myocardial contractility through sympathetic stimulation and to increased venous return with improved ventricular ­filling.71 Although systolic and mean blood pressures show modest increase with exercise intensity as a function of cardiac output, total peripheral resistance actually decreases. The significant decrease in total peripheral resistance is a result of overwhelming vasodilation in the activated skel­etal muscle, leading to increased blood flow.71 Levels of skeletal muscle blood flow increase up to 20 times resting value during exercise to maximize oxygen delivery.70 ­ Visceral

200

120

100

80 Stroke volume (mL)

Heart rate (beats per min)

150

100

60

40 50 20

0 0

50 100 Normal maximal exercise intensity (%)

120

0

0

35 50 Maximal exercise intensity (%)

100

Heart disease Congestive heart failure Normal, sedentary Normal, sedentary A B Healthy, exercise trained Regular exercise Figure 4-11   A, Heart rate increase in proportion to exercise intensity among groups in various states of health. B, Stroke volume change in response to exercise intensity among groups in various states of health. Note that levels plateau prior to maximal intensity, unlike heart rate. (A and B, Data from Hasson S: Clinical Exercise Physiology. St. Louis, Mosby, 1994, p 117.)

220

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 4-5  Cardiovascular Response to Exercise Function

Response

Cardiac output (CO) Heart rate

Large increase Large increase (primarily responsible for CO increase) Modest increase Modest increase Modest increase Modest increase Large decrease (massive skeletal muscle vasodilation dominates over visceral vasoconstriction) Increase due to oxygen ­consumption

Stroke volume Systolic blood pressure Pulse pressure Mean blood pressure Total peripheral resistance Arteriovenous oxygen difference

vasoconstriction likewise aids in redistributing blood flow to the skeletal muscle (Table 4-5).

Respiratory Response to Exercise The lungs play the central role in oxygenating blood during exercise and manage to maintain consistent arterial oxygen concentrations in the presence of increased demand and expenditure by working muscle. Normal respiratory rate is 12 to 20 breaths per minute in adults, but can rise to about 50 breaths per minute with heavy exercise.79 Tidal volume also increases with effort, and ultimately ventilation increases linearly with oxygen consumption. Living and training at high altitudes has drawn ­considerable attention from the athletic and medical communities due to the effects on the cardiovascular and respiratory ­systems. Initially, the body responds to lower partial ­pressures of oxygen at higher elevation by increasing blood pressure, heart rate, and respiratory rate. The kidneys increase secretion of erythropoietin (EPO), leading to ­subsequent increases in red blood cell (RBC) and hemoglobin counts. The oxyhemoglobin dissociation curve shifts to the right, facilitating improved oxygen unloading in the muscle. Ultimately, maximal heart rate and ­cardiac output decrease once the body has fully acclimated to higher elevation. Proponents of altitude training cite the increased RBC count and improved oxygen delivery as advantages for its practice. However, other researchers are quick to point out that these changes are only temporary, and reverse with any significant duration spent at sea level. Additionally, Vo2max declines with altitude, and exercise intensity suffers given the lower concentration of oxygen in the atmosphere. Newer programs advocate living at altitude while training at lower elevations to maximize performance. Although most people adapt well to elevation, some suffer side effects ranging from mountain sickness to excessive polycythemia. Hyperviscous blood from high-altitude training can lead to serious complications such as pulmonary hypertension, cerebral hypoperfusion, right heart failure, and even death.80

C

r i t i c a l

P

o i n t s

l Muscle fibers are capable of various forms of energy metabolism, contraction type, and exercise based on structural and metabolic characteristics. l Skeletal muscle contraction originates at the sarcomere. According to the sliding filament theory, active shortening of the sarcomere and hence the muscle results from the relative movements of contractile protein (actin and myosin) filaments past one another while each retains its original length. l ATP-phosphocreatine, glycolytic, and oxidative energy metabolism systems work together during prolonged exercise to produce useable muscle energy in the form of ATP from stored or dietary nutrients such as carbohydrate, protein, and fat. l Early adaptations to exercise training are largely due to neural adaptations in the form of more efficient motor unit recruitment and coordination. Later strength and endurance gains, including hypertrophy and increased aerobic capacity, take weeks to months of training. l Mechanisms of skeletal muscle task failure consist of central and peripheral fatigue due to reduced central drive or muscle activation failure, buildup of metabolic byproducts during exercise, or other supraspinal factors such as effort and pain. l Delayed-onset muscle soreness is the discomfort felt in the days following high-intensity, typically eccentric, exercise. l Hormonal responses to exercise act to maintain body systems under high demand and tend to vary with exercise intensity and duration and may change with training. l Cardiovascular and respiratory systems adapt to support body needs and oxygen requirements during exercise.

S U G G E S T E D

R E A D I N G S

Aagaard P, Simonsen EB, Andersen JL, et al: Neural adaptation to resistance training: Changes in evoked V-wave and H-reflex responses. J Appl Physiol 92(6):2309-2318, 2002. Folland JP, Williams AG: The adaptations to strength training: Morphological and neurological contributions to increased strength. Sports Med 37(2):145-168, 2007. Gandevia SC: Spinal and supraspinal factors in human muscle fatigue. Physiol Rev 81(4):1725-1789, 2001. Kidgell DJ, Sale MV, Semmler JG: Motor unit synchronization measured by crosscorrelation is not influenced by short-term strength training of a hand muscle. Exp Brain Res 175(4):745-753, 2006. McCall GE, Byrnes WC, Dickinson A, et al: Muscle fiber hypertrophy, hyperplasia, and capillary density in college men after resistance training. J Appl Physiol 81(5):2004-2012, 1996. Monti RJ, Roy RR, Edgerton VR: Role of motor unit structure in defining function. Muscle Nerve 24(7):848-866, 2001. Proske U, Morgan DL: Muscle damage from eccentric exercise: Mechanism, mechanical signs, adaptation and clinical applications. J Physiol 537(Pt 2):333-345, 2001. Raastad T, Bjoro T, Hallen J: Hormonal responses to high- and moderate-intensity strength exercise. Eur J Appl Physiol 82(1-2):121-128, 2000. Sargeant AJ: Structural and functional determinants of human muscle power. Exp Physiol 92(2):323-331, 2007. Trappe S, Harber M, Creer A, et al: Single muscle fiber adaptations with marathon training. J Appl Physiol 101(3):721-727, 2006.

R eferences Please see www.expertconsult.com

C H A P T E R�  

 5

Rehabilitation and Therapeutic Modalities S ect i o n

A

Language of Exercise   and Rehabilitation Russ Paine

Successful results of surgical procedures for sports injuries are heavily dependent on postoperative and injury rehabilitation. The catabolic effect of surgery and injury places the injured athlete in a position of diminished functional ability and inhibition. For many athletes, this impaired position is a new adventure, many times testing the individual’s patience, mental drive, and work ethic. The team approach to solving the injury now involves the rehabilitation specialist. As is true with any successful team, communication with all members of the team allows a more accurate, timely, and positive outcome for the injured athlete. The function of this section is to describe language used among sports medicine professionals and elaborate on how this can be used during development and implementation of rehabilitation programs.

KINETIC CHAIN To begin our discussion, the kinetic chain is used to help describe the relationship of exercises to the healing constraints placed on the development of a particular rehabilitation protocol. Using mechanical engineering concepts, the kinetic chain concept was described first by Steindler in 1955.1 Steindler sought to describe various types of exercise conditions with this new concept. The mechanical engineering definition of the kinetic chain describes a series of interconnected joints, fixed proximally and distally. In this system, movement of one joint has an effect on all joints above and below. Steindler1 separated the mechanical engineering definition into two states—open and closed kinetic chain. He described the open chain state as a peripheral extremity able to move freely, such as waving the hand or the swing phase of the gait pattern. The closed kinetic chain state was

described as when the distal segment meets considerable resistance, as it does with a pull-up or squat. Considerable ambiguity exists when describing various exercises by open and closed chain definitions. Confusion sometimes exists when determining what is a true closed kinetic chain state. A more simple and precise description may be weight-bearing (WB) and non–weight-­bearing (NWB) exercises and the relationship to the anatomic structure: Is the ligament loaded or unloaded during the WB exercise? How do these two conditions relate to other anatomic structures, such as the muscles and the patellofemoral joint? Because of the ambiguity of closed and open chain states, this chapter describes more specifically what is occurring to pertinent anatomic structures. During progression of the exercise program, the injured athlete is placed in different positions that could cause harm to the healing tissue. In a previous chapter, we described these conditions and outlined various states that could cause an impact to the healing graft. The tables presented below describe how functional exercise might have an impact on the reconstructed knee. The information relates to the anterior cruciate ligament (ACL) and the posterior cruciate ligament (PCL). Rather than stating the commonly used terms of open and closed kinetic chain, we elect to divide the exercises into WB and NWB conditions. Many authors have reported ACL and PCL load values during NWB resisted knee extension. It has been concluded that strain on the ACL substantially increases in the last 60 degrees of knee extension and peaks during NWB knee extension.2-7 Knee extension exercises performed in ranges of 110 to 60 degrees place dramatically higher loading on the PCL.2,3,6,7 This range is opposite of the ACL loading range. As seen in Tables 5A-1 and 5A-2, the peak loads do not approach the ultimate tensile strengths of these 221

222

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 5A-1  Peak Loads on the Anterior Cruciate Ligament with Common Rehabilitation Exercises Activity

Study

Mean Peak Loads

Nisell and Ekholm (1986)20 Beynnon et al (1997)9 Erickson and Nisell (1986)30 McCoy and Gregor (1989)43

500 N (20 degrees) 136 N (10 degrees) 37 N (65 degrees)

Grood et al (1984)25 Lutz et al (1993)3 Wilk et al (1996)6 Nisell et al (1986)14 (30 degrees/sec)

350 N (0 degrees) 285 N (30 degrees) 248 N (14 degrees) 700 N (25 degrees)

Weight-Bearing

Squat Stationary bicycle

125 N (70 degrees)

Non–Weight-Bearing

Knee extension Isokinetic knee extension

l­igaments (when healthy). These loads may be deleterious in the early phases of rehabilitation, when the strength of the reconstructed graft is a concern. Studies have been performed on the magnitudes of tibial translation during NWB knee extension. Using varying methods, each of the investigators documented that NWB knee extension elicits significant anterior tibial translation during the terminal degrees of the movement and imposes excessive stress and strain on the ACL. Similarly, many of these studies documented significant posterior tibial displacements when extending the knee from full flexion to about 60 degrees against resistance.4,8-12 Resultant posterior tibial translation provides evidence for the stresses placed on the PCL. Researchers have documented the effects of performing NWB isokinetic knee extension exercises. As with NWB isometric or isotonic knee extension activities, increases in anterior tibial translation were seen in the terminal degrees of extension.11,13-15 Nisell and coworkers14 and Wilk and Andrews15 reported significantly higher anterior-directed shear forces with a more distal pad placement (versus a single, proximally placed pad). Wilk and Andrews15 documented significantly higher anterior tibial displacement at lower isokinetic speeds (60 degrees/second compared with 180 and 300 degrees/second). At 60 degrees/second, the    TABLE 5A-2  Peak Loads on the Posterior Cruciate Ligament with Common Rehabilitation Exercises Activity

Study

Mean Peak Loads

Beynnon et a (1997)9 Escamilla et al (1998)24 Lutz et al (1993)3 Nisell and Ekholm (1986)20 Stuart et al (1996)22 Wilk et al (1996)6 Wilk et al (1996)6

136 N (≈0 degrees) 1868 N (63 degrees) 538 N (90 degrees) ≈1800 N (≈130 degrees) 295 N (93 degrees) 1783 N (90 degrees) 1667 N (94 degrees)

Lutz et al (1993)3 Wilk et al (1996)6 Lutz et al (1993)3

387 N (90 degrees) 1178 N (91 degrees) 1780 N (90 degrees)

Weight-Bearing

Squat

Leg press Non–Weight-Bearing

Knee extension Knee flexion

isokinetic exercise loads to the ACL are not different than an isotonic knee extension. Numerous authors have documented resultant shear forces during NWB knee flexion. Each investigator reported posterior shear forces throughout the entire range of NWB knee flexion.2,3,7-9,11 Kaufman and associates11 reported that posterior shear forces increased linearly with increased knee flexion while performing isokinetic knee flexion at 60 degrees/second and 180 degrees/second. Resisted knee flexion places possible deleterious stresses on a healing PCL graft and should be avoided in the early and intermediate phases of PCL reconstruction rehabilitation. Paine and colleagues17 showed there to be no anterior tibial displacement while performing a straight leg raise. If the leg raising was performed with a flexed knee (extensor lag), however, there was a significant increase in anterior tibial translation. Adding a 5-pound ankle weight increased the anterior tibial displacement further. When considering PCL loading, the resultant posterior shear forces seen in some WB exercises should be taken into account. Several authors concluded that as the flexion angle increases during a WB exercise, posterior shear forces increase, with maximal loads occurring around 90 degrees.6,18-22 WB exercises such as the squat, leg press, lunge, and front squat are considered detrimental to a healing PCL when performed in ranges greater than 60 to 70 degrees of flexion in the early phases of rehabilitation. According to the literature, the therapeutic exercises that produce the greatest amount of load to the ACL are (1) NWB resisted terminal knee extension, (2) slowspeed isokinetic terminal knee extension, and (3) resisted straight leg raises with a flexed knee. When considering loads on the PCL, the following exercises appear to apply the most stress: (1) NWB resisted knee flexion, (2) NWB knee extension from full flexion to about 70 degrees, and (3) WB exercises with depths greater than 70 degrees of knee flexion. These exercises should be avoided in the early stages of a respective ACL or PCL reconstruction ­rehabilitation program.

LIGAMENT UNLOADING EXERCISES Ligament unloading exercises are exercises that transmit minimal or no strain to the ACL, PCL, or healing graft. Ligament unloading exercises can be used during the initial phases of rehabilitation when early protection to the graft is desired. These exercises should not be harmful to different graft fixation techniques, a crucial factor during the first 6 weeks. Realistically, exercises fall somewhere on a continuum of ligamentous loading. On one end lie the most offending exercises, and on the other are exercises that minimally stress the ligamentous structures. The exercises should be added to the rehabilitation protocol in a manner that begins with the least loading and advances to greater loading exercises as the graft healing allows. Tables 5A-3 and 5A-4 outline a continuum of common therapeutic exercises, categorizing them from relative low to high ligamentous loading on the ACL or PCL. In accordance with the findings of Butler and colleagues,23 our interpretation

Rehabilitation and Therapeutic Modalities

223

   TABLE 5A-3  Continuum of Stress Placed on the Anterior Cruciate Ligament in Relation to Various Parameters of Common Rehabilitation Exercises

Lower Stress ↔ High Stress Activity

Ligament Unloading

Ligament Unloading

NWB resisted knee extension8,23,25,28

90-60 degrees

30-0 degrees*

NWB resisted knee extension—external load placement13-15 Straight leg raise17 Isokinetic knee extension—speed11,15 Vertical squat—depth6,21,22,24 Vertical squat—trunk lean6,21 Vertical squat—stance width26 Leg press—depth6 Leg press—foot placement26 Stationary bike—workload30 Stationary bike—foot placement30 Stationary bike—seat height43 StairMaster29

Proximal Knee extended >180 degrees/sec Safe throughout Forward Wide Safe throughout Lower Lower Anterior High Safe throughout

Distal* Knee flexed with resistance* <60 degrees/sec* Vertical Narrow High High Posterior Low

*Considered unsafe for the healing anterior cruciate ligament graft. NWB, non–weight-bearing.

of the research literature is that anterior shear forces primarily strain the ACL, and posterior shear forces strain the PCL. Two studies reported anterior shear forces during the squat exercise,9,20 whereas several other investigations found posterior shear forces and resultant PCL loading throughout the squat.6,7,21,22,24 Similarly, posterior shear forces were seen in the leg press6 as well as the front squat and lunge exercises.22 Several authors concluded that as the flexion angle increases, shear and compressive forces also increase.6,18-22 As seen in Tables 5A-1 and 5A-2, however, these shear forces do not approach the ultimate tensile strengths of the cruciate ligaments. This situation may be due in part to the increased stability gained by the increasing tibiofemoral compressive forces during WB exercises.3,6,25 In 1997, Beynnon and coworkers9 published an article that described ACL strain during WB exercises. Before

   TABLE 5A-4  Continuum of Stress Placed on the   Posterior Cruciate Ligament in Relation to Various Parameters of Common Rehabilitation Exercises

Lower Stress ↔ High Stress Activity NWB resisted knee extension3,8,38,42 NWB resisted knee flexion11 Isokinetic knee flexion11 Vertical squat—depth3,15 Vertical squat—trunk lean6,14 Vertical squat—stance width26 Leg press—depth6 Leg press—foot placement26 Stationary bike—seat height43 StairMaster29

Ligament Unloading

Ligament ­Unloading

60-0 degrees

100-60 degrees*

0-45 degrees Vertical Narrow 0-45 degrees High Low Safe throughout

Throughout* Throughout* 70-100 degrees* Forward Wide 70-100 degrees* Low High

*Considered unsafe for the healing posterior cruciate ligament graft. NWB, non–weight-bearing.

this publication, it was the general consensus that because of the compression forces present during WB exercises, the amount of anterior tibial translation and calculated shear forces was reduced.9 Beynnon’s strain gauge study showed similar strain in the ACL during WB and NWB exercises such as the squat (with body weight) and active knee extension. Strain did not increase during the squat when resistance was added. Strain did increase, however, during the active NWB knee extension as resistance was increased.8 Increasing external loads to the squat exercises could be considered safe when strengthening patients after ACL reconstruction surgery. Likewise, adding resistance to NWB knee extensions may cause excessive strain on a healing ACL graft. Several investigators studied squat mechanics further and noted how variables in squat performance affect shear forces in the knee joint. Escamilla and colleagues26 studied the effects of stance width (narrow, wide, normal) and foot angle (parallel, turned out 45 degrees) on tibiofemoral shear and compressive forces during the squat and leg press. These investigators analyzed the effects of foot placement on the footplate during the leg press (high, low, natural). No significant differences in shear forces were found with these varied parameters. Escamilla and colleagues26 did document higher compressive forces with the wider stance, which may be beneficial in limiting shear forces. Results of a study by Ariel27 determined that when the knees move forward during the squat, and when the individual bounces at the end of the descent, higher shear forces occur. Stationary cycling has been studied for its effects on the cruciate ligaments. Henning and coworkers28 and Fleming and associates29 determined that ACL strain was evident during cycle ergometry. These values were given as low, however, in comparison with other commonly prescribed exercises. It has been determined that ACL forces can be decreased further with stationary cycling by lowering the workload,14 using an anterior foot position,30 and using a high seat height.30 To decrease posteriorly directed shear forces, seat heights should be lowered.30

224

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The strain produced in the normal ACL has been measured during stair climbing.29 Strain values were shown to increase as the knee moved from flexion to extension. Although ACL strains increased with the stair-climbing motion, the forces were considered moderate in comparison with other previously tested rehabilitation activities, including squatting. The authors of this study were unable to make clinical recommendations for stair climbing because the ACL strain values were highly variable across their subjects. Early WB exercises should begin with exercises that produce minimal ligament loading and no rotational forces to the knee, such as the Total Gym and leg press. In the intermediate and advanced phases of strengthening, additional NWB exercises, such as squats, lunges, and step-ups, can be employed. These exercises are effective during rehabilitation after an ACL reconstruction and can be used after PCL reconstruction with knee flexion depths less than 70 degrees. Although there may be confusion regarding whether WB exercises are safer than NWB exercises, there is little argument that WB exercises are more functional and provide a greater proprioceptive input to the joint. An example of a new exercise device that combines joint position sense and muscular activation is the Monitored Rehab Systems functional squat (Fig. 5A-1). This device allows limiting range of motion through computerized software. For the optimal quadriceps building mode, we recommend that the knee not be allowed to drift into full extension. Allowing the knee to exercise between 120 and 40 degrees allows a greater fatigue of the quadriceps. The sports medicine patient performs a series of exercises that mimic computer games requiring control of the curser position on the computer screen. A typical exercise session involves six sets of 60-second contractions with 45 seconds of rest intervals. Knowledge of the kinetic chain has allowed us to further advance safety and protection while rehabilitating the injured athlete. Although the language of closed and open kinetic chain may be ambiguous, it allowed a method of

Figure 5A-1    Functional squat (Courtesy of Monitored Rehab Systems, Haarlem, The Netherlands.)

communication among physicians and therapists during the early years of ACL and PCL reconstructions. The Monitored Rehab Systems leg press allows ­interaction of a patient and computer screen to permit muscle re-education and proprioception. This device mimics a functional squat as the sled glides in a curvilinear direction as the patient performs the leg-pressing motion. Combining closed chain strengthening with proprioceptive training using computerized interaction is an exciting method of training and testing. Devices that make objective testing more functional will provide us with more ­predictable patient status.

MUSCLE ATROPHY—INHIBITION During the initial stages of knee and shoulder rehabilitation, the primary goal is to restore muscle function. Without the protection of the dynamic forces supplied by muscle, undue forces will be placed on the static restraints that act in concert with muscle to provide joint stability and equilibrium. Altered mechanics could lead to undesired forces to the articular cartilage, healing graft, and overloaded tendon tissue. Muscle inhibition appears to be focused to our antigravity muscles. One possible explanation is that the antigravity muscles have a higher percentage of slow-twitch type I muscle fibers that may be more sensitive to the inhibitory impulse that is present when swelling and pain are present in the joint.31 Immobilization has been proved detrimental to type I muscle fibers. It has been shown that the vastus medialis obliquus (VMO) of the quadriceps has a richer saturation of type I muscle fibers over the remaining quad muscles. This may be a cause of rapid shrinking and inhibition of the VMO while under the influence of pain or swelling, leading to decreased excitability of quadriceps motoneurons. The Ruffini endings are stimulated when an effusion is present,32 alerting and activating central structures regarding the current state of the knee joint. ­Palmieri-Smith and colleagues described the influence on a step-down maneuver during quadriceps inhibition.33 This demonstrated that greater ground reaction forces were induced to the knee while under the influence of experimentally induced effusion in the normal knee. This further supplants the theory that a weak knee will be unable to slow down forces external to the joint. A lack of dynamic shock absorption from the quadriceps may lead to breakdown of the “golden” protection of the joint, the articular cartilage. The same effects are present to the antigravity muscles of the shoulder: the posterior rotator cuff ­muscles—­infraspinatus and teres minor. To combat the inhibitory effects on muscle atrophy and resultant weakness, several modalities can be implemented. My primary method of reversing muscle inhibition is the use of biofeedback. The device is applied over the VMO of the quadriceps. A threshold of activation is set according to the patient’s ability to elicit an electrical signal through nerve depolarization. A goal is then set on the instrument for the patient to attempt to reach. The visual feedback to the patient is the use of light-emitting diode (LED) lights on the face of the instrument. When the goal has been reached, the lights change color, indicating a successful contraction. I believe that it is more effective for the

Rehabilitation and Therapeutic Modalities

patient to use his or her own ability, rather than electrical stimulation, to recruit motor units that result in muscle contraction. Muscle stimulation without active contraction has not been shown to be beneficial. Muscle stimulation in coordination with active muscle contraction has been shown to help combat muscle atrophy.34 Biofeedback has been shown to be more effective than electrical muscle stimulation in restoring muscle strength to postoperative patients.35

ADHESIONS AND ARTHROFIBROSIS A stiff joint alters the rehabilitation protocol. The most unhappy and debilitated sports medicine patient is one who has lost function due to stiffness of his or her joint. It is believed that some individuals have a greater propensity to develop adhesions in the joint. Many studies have documented ill effects that result from postoperative joint adhesions. One of the factors that help to prevent loss of motion is muscle activation. Paulos and colleagues described the altered position of the patella that occurs postoperatively after ACL reconstruction.36 Patella infra is the result of immobilization and a lack of quadriceps contraction. If the quadriceps is firing appropriately after surgery, there is a natural superior glide of the patella that occurs, preventing the patella from drifting inferiorly because of fat pad contracture to the proximal tibia. Manual superior glide mobilizations are required in the severely quadriceps-inhibited knee. This is performed with the knee in full extension, with no quadriceps activity present. Aggressive force may be used to glide the patella superiorly until quadriceps function takes over the superior gliding movement of the patella (Fig. 5A-2). Other structures that prevent motion of the knee are the medial and lateral gutters of the knee and the suprapatellar pouch. Without immediate motion of the knee, these fascial planes do not glide together, and scarring may

obliterate the normal space, presenting as arthrofibrosis. The joint capsule is joined to the joint surface and anatomy by the scar. This in turn limits motion in both directions (flexion and extension) and creates increased compression forces to the patellofemoral joint. Rehabilitation devices have been created to help combat the restrictions associated with the development of a stiff joint. The ERMI devices are shown in Figure 5A-3. These devices use overpressure in a controlled manner to help lengthen the scar tissue or prevent scar formation. These devices put the patient in control and allow more relaxation of the musculature so that the scar tissue can be addressed. If pain is too severe, a muscular co-­contraction limits the ability to address the culprit that is limiting motion—­adhesion formation. Long-duration stretching is performed to fatigue scar tissue. Another factor governing knee condition after surgery is maintaining a normal gait pattern. Substitution patterns must be avoided that might accentuate the occurrence of arthrofibrosis. I suggest that patients continue to ambulate with crutches until they (1) are able to ambulate without a limp, (2) have no giving way sensation, and (3) achieve adequate control of swelling.

ARTICULAR CARTILAGE PROTECTION There is no more important topic when discussing rehabilitation than preserving and protecting articular cartilage. Degenerative joint disease is frequently involved in shortening an athlete’s career. Swelling leading to muscle weakness, lack of vertical leap, and inability to accelerate are all common symptoms of an athlete in the throws of articular cartilage degeneration. The structure of articular cartilage is an amazing, complex, extremely tough component of joint protection. Although articular cartilage is able to resist tremendous compressive forces, once an injury occurs, it is unable to repair itself. It is up to the rehabilitation specialist to make certain that therapeutic exercises employed do not place undue shearing forces on healing articular cartilage.

1 ERMI/Knee Ankle Flexionater 3

2

ERMI/Knee Extensionater

9

7 4 5

A

8

6

B

Figure 5A-2    Normal recesses that allow gliding to occur. Loss of these recesses limits motion and increases patellofemoral compression. (From Millett PJ, Wickiewicz TL, Warren RF: Motion loss after ligament injuries to the knee. II. Prevention and treatment. Am J Sports Med 29:822-828, 2001).

225

Figure 5A-3    ERMI Knee/Ankle Flexionater. (Courtesy of ERMI, Inc., Atlanta, Ga.)

226

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Inability of articular cartilage to heal has been recognized since the 17th century. Although replacement techniques are evolving, total recovery from articular cartilage cell implantation has not been achieved. Protecting and preserving the healing graft or injury is a primary goal of the rehabilitation specialist. Articular cartilage lesions often occur in particular regions of the knee. Lesions are commonly present involving the patellofemoral and tibiofemoral joint. The WB location of the patellofemoral location is 20 to 45 degrees.16 Thus, avoiding the 20- to 45-degree range of motion, especially when using open chain knee extensions, is imperative during the healing phase of articular cartilage ­rehabilitation. In a study by Lewandrowski and coworkers, tibiofemoral lesion locations were identified on 1740 knee joints by arthroscopy.37 Articular cartilage lesions were found in the medial femoral condyle in 1212 knees, on the medial tibial plateau in 749 knees, on the lateral femoral condyle in 493 knees, and on the lateral tibial condyle in 612 knees. WB locations where these lesions impact each other are also in the 20- to 45-degree range. It is important to establish a working knowledge of patellofemoral mechanics to understand muscle forces and resultant impact to the articular cartilage. The mechanics of the patellofemoral joint are discussed subsequently by describing the WB and NWB states.

Patellar Mechanics during WeightBearing and Non–Weight-Bearing Exercises The primary functions of the patella are to increase the distance of the quadriceps muscle force from the center of rotation of the knee and to increase the moment arm and resultant knee extension torque.10 When the knee is extended in an NWB state, the patella is pushed forward by the trochlear groove of the femur. When extending the knee beyond 45 degrees, the lever arm function of the patella diminishes (height of the patella decreases as the patella begins to exit the trochlear groove). Because of the decrease in mechanical advantage, increased quadriceps muscle force is required to achieve terminal extension.10 In a patient with diminished quadriceps function, this inability to handle the additional demand is represented by an extensor lag. Goodfellow and coworkers38 identified the different contact regions of the patella at various flexion angles, corresponding to the articular facets viewed on the posterior surface of the patella. The first consistent contact of the patella is made at 10 degrees of flexion. As the knee is flexed from full extension to 90 degrees, the area of contact moves steadily from the inferior articular portion to the superior portion of the patella. With further flexion beyond 90 degrees, the odd medial facet obtains contact for the first time. At 135 degrees of flexion, contact is on the patella’s lateral and odd facets. Actually, the quadriceps tendon becomes the primary contact surface with the femur at these deeper flexion angles. Any imbalance in contact and compression that has occurred can lead to articular degenerative changes.

An analysis of the forces applied to the patellofemoral joint throughout the range of flexion reveals important information regarding compressive loads and stresses to the articular surfaces of the patella. As knee flexion proceeds from an extended position, the pull of the quadriceps tendon, either active or passive, and the pull of the patellar tendon further compress the patella into the femur. This patellofemoral joint reaction (PFJR) force can be calculated as a resultant vector force equal and opposite to the pull of the quadriceps and the patellar tendon (Fig. 5A-4). As the knee flexes, the angle between the quadriceps tendon and the patellar tendon becomes more acute and increases the resultant vector force. Excessive compressive forces can produce damaging stresses on the articular cartilage of the patella.

Weight-Bearing Exercises and Patellar Function Many studies have quantified patellofemoral compressive forces during WB exercises.18,20,24,39 As with ligament loading studies, these investigators noted how variables in exercise performance affect the forces in the patellofemoral joint. When considering the amount of knee flexion, patellofemoral compressive forces have been shown to increase as knee flexion increases during the squat,20,24,39 leg press,40 and stationary cycling.30 Nisell and Ekholm20 further determined that squatting in flexion ranges greater than 90 degrees produces significant compressive forces between the quadriceps tendon and the intercondylar notch of the femur. Currently, research has not provided evidence as to what compressive loads are detrimental to the patellofemoral joint. For individuals with patellofemoral arthritic symptoms, limiting the magnitude of knee flexion to an appropriate symptom-free range can minimize patellofemoral stresses. Wrentenberg and associates studied the effects of lowbar and high-bar positioning during the squat. They concluded that the high-bar position produced significantly FQ

PFJR

FPT

Figure 5A-4    The pull of the quadriceps (FQ) and the pull of the patellar tendon (FPT) determine the patellofemoral joint reaction (PFJR) force.

Rehabilitation and Therapeutic Modalities

227

Figure 5A-5    Leg press.

greater patellofemoral compressive forces. They also noted that the lifters who employed the high-bar position maintained a more upright trunk posture and required more knee extensor torque to complete the squat (and less hip extensor torque). These investigators believed the highbar position was ultimately responsible for the increased compression forces in the patellofemoral joint. In the research by Escamilla and colleagues,26 in which the effects of stance width, foot angle, and leg press foot placement were studied, patellofemoral compressive forces were calculated. No significant differences were found with these varied parameters. Reilly and Martens39 calculated forces present with normal activities of daily living. Walking on level ground showed minimal PFJR force (PFJR = 0.5 × body weight [BW]) because minimal knee flexion is required. When knee flexion angles increase, such as when squatting to 90 degrees, the PFJR forces increase to 2.5 to 3 times body weight. Descending stairs always is difficult for an individual with patellofemoral pain. During descent, the peak PFJR forces reach 3.3 × BW at 60 degrees of knee flexion. Although the PFJR force increases with increased flexion, the area of contact also increases. This increased area of contact can respond effectively to disperse these forces.

leg extension occur when the patellofemoral contact is the least (0 and 30 degrees) (Fig. 5A-6). Although compressive forces may be lower for leg extensions, patients with patellar articular degeneration and arthritic changes often experience pain during NWB extensions of 30 to 0 degrees as a result of the relatively large compressive forces being applied to minimal contact areas. Steinkamp and coworkers40 concluded that patients with patellofemoral joint arthritis may tolerate leg press exercises (or similar WB squatting exercises) better than leg extension exercises in the final 30 degrees of knee extension because of the

Non–Weight-Bearing Exercises and Patellar Function Understanding the forces that act on the patellofemoral joint at different positions (WB and NWB) allows the clinician to develop a program based on empirical evidence. Biomechanical knowledge in action is illustrated by the interaction between the patellofemoral contact area and the PFJR force during the leg press (and other WB exercises) and leg extension (NWB) exercises. Steinkamp and coworkers40 showed mathematically that maximal PFJR forces during the leg press occur when contact between the patellofemoral surfaces is greatest (60 and 90 degrees) (Fig. 5A-5), whereas the maximal PFJR forces during a

Figure 5A-6    Leg extension.

228

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

lower patellofemoral joint stresses. These investigators also determined that the PFJR forces are less in open chain exercises than closed chain exercises from 90 to 60 degrees, which Brownstein and associates41 found to be the range in which the greatest vastus medialis electromyographic activity was registered. Immobilization has been proved detrimental to articular cartilage. Loss of protein elements known as glycosaminogly­ cans (GAGs) decreases the ability of articular cartilage to resist compression forces. The GAGs slow down the fluid flow in and out of the articular cartilage matrix and act as a hydraulic shock-absorbing system. With immobilization, articular cartilage becomes soft owing to the decrease in GAGs and is prone to injury at this state. Articular cartilage needs motion for nutrients to be absorbed into the matrix. A lack of motion, or the excessive pressure that might occur with arthrofibrosis, affects the nutrition of articular cartilage. Early motion protocols are now commonplace in sports medicine practice to prevent articular cartilage injury. The efficacious use of continuous passive motion machines is still an important aspect of rehabilitation for the purpose of maintaining articular cartilage health as well as restoring range of motion.

SUMMARY Knowledge of biomechanics and healing constraints continue to be a mainstay of successful rehabilitation specialists. Surgical techniques help restore static restraints to promote stability of sports injuries. Sports medicine rehabilitation uses control of muscle forces to affect the dynamic stability of the injured joint. Without a working knowledge of surgical healing constraints and dynamic muscle forces, a properly implemented rehabilitation program is not possible. In addition to restoring motion and strength, a primary goal of current sports rehabilitation must be to protect articular cartilage. Alterations in the postoperative rehabilitation program to protect healing or malnutritioned articular cartilage result in prolonged athletic careers and prevention of degenerative joint disease. After education, the second most important link in sports medicine rehabilitation is communication among the sports medicine team. Physical therapists, athletic trainers, and orthopaedic surgeons must be interconnected for the injured athlete to achieve his or her preinjury level. This means that all parties must be available for communication at any point in time.

C

r i t i c a l

P

o i n t s

l Closed kinetic chain rehabilitation provides a more functional form of therapeutic exercise but should be combined with open kinetic chain exercise for the most complete strategy of recovery. l Ligament unloading exercises should be used in the early postoperative ­period to protect the healing graft. l Ligament loading exercises may be instituted after graft incorporation is complete. l Muscle atrophy can be effectively treated with use of biofeedback to help the patient learn to recruit motor units, enhancing muscle activation. l Recognition of the beginning stages of arthrofibrosis is the key to preventing its occurrence. Devices such as the flexionator and extensionator provide excellent tools in the treatment of arthrofibrosis. l Articular cartilage protection is paramount during the rehabilitation process and is especially important during the transition to functional exercise and plyometrics. l Patellofemoral symptoms are controlled by ­ increasing quadriceps strength and restoring normal hamstring flexibility. Progression to functional activities should be avoided until the above are achieved.

S U G G E S T E D

R E A D I N G S

Escamilla FR, Fleisig GS, Barrentine SW: The effects of technique variations on knee biomechanics during the squat and leg press. Med Sci Sports Exerc 29:S156, 1997. Fleming BC, Beynnon BD, Renstrom PA, et al: The strain behavior of the anterior cruciate ligament during bicycling: An in vivo study. Am J Sports Med 26: 109-118, 1998. Markolf KL, Gorek JF, Kabo M, et al: Direct measurement of resultant forces in the anterior cruciate ligament. J Bone Joint Surg Am 72:557-567, 1990. Steinkamp LA, Dillingham MF, Markel MD, et al: Biomechanical considerations in the patellofemoral joint rehabilitation. Am J Sports Med 21:438-444, 1993. Wilk KE, Escamilla RF, Fleisig GS, et al: A comparison of tibiofemoral joint forces and electromyographic activity during open and closed kinetic chain exercises. Am J Sports Med 24:518-527, 1996.

R eferences Please see www.expertconsult.com

Rehabilitation and Therapeutic Modalities

S ect i o n

229

B

Use of Modalities in Sports Robert Mangine, Marsha Eifert-Mangine, and W. Andrew Middendorf

Historically, the use of modalities for the treatment of sports-related injuries has played a significant role in a variety of forms and interventions. The foundation of athlete injury management continues to be the basic use of ice and heat to influence the effect of the injury and promote healing within injured tissue. As the health care system has evolved into the practice it is today, a plethora of agents have been developed and brought to market for clinical practice. The purpose of this chapter section is to review common devices in today’s world of sports medicine, describe the authors’ experiences with each, and present relevant literature. This section will concentrate on the following: • Electrical currents • Iontophoresis • Ultrasound • Laser • Cryotherapy devices

ELECTRICAL CURRENTS The use of electrical currents has been incorporated into physical therapy practice for several decades. The field of sports medicine can be traced back as far as the ancient Olympic Games in Greece, where the earliest predecessors of what today are referred to as athletic trainers used the application of electric eels to treat athletes’ injuries. The clinical use of electrotherapy in sports medicine takes many forms in the management of a wide range of pathologies. The electrical currents used to treat athletes and their injuries have evolved to a complex and varied mix of devices and applications. Electrical currents have been described to perform multiple functions, including pain management, muscle contraction, functional retraining, stimulation of tissue healing, and edema control, although these claims are not always directly supported in the literature. Because of the lack of direct, thoroughly randomized controlled trials, the clinical application of electrical currents is problematic. The most common uses of electrical currents in sports medicine are to control inflammation, modulate pain, and maximize muscle function. Over the course of our practice, the use of electrical stimulation for these purposes has evolved through the development of various wave forms, wavelengths, and frequencies used for therapeutic benefits. These electrical currents are commonly applied to the management of athletes after injury or surgery. Other uses of electrical currents include promotion of tissue healing (especially in management of open wounds, movement of fluid stasis with peripheral vascular disease, stimulation of denervated muscle, and functional retraining ­following

cerebrovascular accident or spinal cord injury. In this section, focus is on the following applications of electrical ­currents: • Pain modulation • Management of soft tissue swelling • Restoration of muscle function following injury or ­surgery

Pain Modulation Although transcutaneous electrical nerve stimulation (TENS) technically refers to any electrical current delivery through the skin to stimulate a sensory or motor nerve, the term TENS has clinically been adapted to refer to stimulation of sensory nerves for the purpose of pain control. In sports medicine, the use of TENS for acute pain management is less common than in other areas of rehabilitation; however, it can be an effective treatment intervention that facilitates a quicker return to athletic competition for the injured athlete. The use of TENS in the management of chronic pain patients is widely accepted in clinical practice, but there is little agreement in the literature on its ­effectiveness.1,2 Electroanalgesia can be accomplished using larger clinical stimulation units that can produce premodulated or interferential currents. Premodulated current has continuous sinusoidal waveform with sequentially increasing and decreasing current amplitude, which create an effect similar to burst mode of a traditional TENS unit. Interferential current involves two medium-frequency currents that cross, producing envelopes of pulses. Interferential current is said to be more comfortable because it allows greater amplitude to be delivered through the skin by adjusting the sweep and scan (personal communication, James Pomonis Ph.D., Senior Manager of Clinical Programs, EMPI, Inc., St. Paul, Minn). There are several theories about how electrical currents result in pain modulation, including the gate control theory3 and the endorphin theory.4 The gate control theory is based on stimulation of the larger A fibers with electrical current, which inhibits synaptic transmission along the smaller C fibers. According to the endorphin theory, prolonged electrical stimulation (40 to 60 minutes) of small afferent fibers is thought to trigger the release of β-­endorphin. Pain control is most likely a combination of both these theories. Multiple studies have described the benefit of TENS for pain management in surgical and nonsurgical patients.5,6 In sports medicine, the primary use of electrical stimulation for pain management is for acute injury ­management or

230

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5B-1    Electrode setup following an acute brachial plexus injury for upper trapezius pain control.

for the postsurgical athlete. Based on the current literature, a protocol of 20 to 30 minutes of cycling, 80 to 150 pulses per second, at a strong but comfortable intensity up to six cycles per day is used to stimulate the region surrounding the injury or surgical incision. Figure 5B-1 demonstrates the use of stimulation for electroanalgesia after an acute brachial plexus injury to a football player. The pad placement is in the region of the upper trapezius to control residual pain and spasm. Figure 5B-2 demonstrates positions commonly used with postsurgical shoulder and knee patients. In a study involving postoperative shoulder patients (rotator cuff repairs and stabilization procedures), the authors found electroanalgesia with a TENS unit to be effective when initiated within the first 24 hours after surgery and used for the 7 to 21 days following the surgical procedure.7

Decrease Swelling Acute soft tissue inflammatory response is the most common limiting factor to early return to sport participation following injury to muscles, tendons, ligaments, and joint capsule. The ability to limit the magnitude of this inflammatory response and to facilitate the dissipation of accumulated swelling is a critical component of the ­rehabilitation. Reduction of acute swelling can have an effect on other

A

impairments, including range of motion, muscle performance, and proprioception.8 Electrical currents used clinically to address soft tissue edema are supported by two different theories: contractile and noncontractile. The noncontractile theory is based on ion movement to increase lymphatic drainage. Ion movement is facilitated by a twin-peak monophasic current, clinically known as high-voltage pulsed galvanic stimulation, or high-voltage pulsed current (HVPC).9 The clinical use of HVPC has been supported since 1966 and is described as effective for joint swelling as well as soft tissue inflammation.10 The protocol we use implements the use of HVPC in the acute phase of injury as part of a comprehensive program, including medication and active procedures. Based on the type of injury, HVPC treatment can be applied in 30-minute cycles or continuously for 24 hour cycles. With the standard of care in sports medicine today, the need for aggressive treatment interventions to return the athlete to sport expediently is essential. Figure 5B-3 demonstrates the use of clinical and portable units for the delivery of HVPC. The contractile theory is based on muscle pumping contractions, which influence proper circulatory action.11 This influence of the electrical current on the skeletal muscle contraction results in a pumping action to encourage fluid movement toward the heart and away from the extremity being treated. In addition to skeletal muscle contraction, smooth muscle tone in the lymphatic and venous system can also be influenced by the drainage pattern. Muscle contraction is generated with an interrupted alternating current, such as variable muscle stimulation (VMS), premodulated, or Russian current, the same currents used for strengthening or functional electrical stimulation. A brief 1:1 ratio of contraction to relaxation between 5 and 10 seconds can be used without risk for muscle fatigue because of submaximal contraction. Recommended treatment time is 20 to 30 minutes multiple times per day; in our postsurgical patients, we recommend a minimum of six cycles per day.

Restoration of Muscle Function Of all the physical agents in sports injury management, especially after a surgical procedure, the use of electrical currents to reeducate muscle contractile sensing by

B

Figure 5B-2    A, Common setup for the postsurgical shoulder with the electrodes placed parallel to the wound for pain management. B, Common setup for the postsurgical knee for pain management. This position can also be used for swelling.

Rehabilitation and Therapeutic Modalities

A

231

B

Figure 5B-3    A, Clinical application of high-voltage galvanic stimulation. B, Portable unit application of high-voltage galvanic stimulation.

s­ timulating the motor end plate of the involved muscle has gained popularity in the past 20 years. The type of electrical current is an AC, symmetrical or asymmetrical biphasic pulsed current or a medium-frequency burst current with an on-duty, off-duty cycle. Examples include Russian, VMS, and VMS Burst currents. The advancement of these units in the past 5 years has made the use of electrical currents a standard of care when muscle shutdown has occurred. Many published rehabilitation programs advocate the use of such currents to facilitate neuromuscular reeducation in the acute phase of postsurgical rehabilitation.12-14 Noyes and colleagues reported the use of neuromuscular stimulation as a means to avoid motion complications in anterior cruciate ligament reconstruction patients.15 The basis for the use of stimulation after knee surgery had its foundation in neurologic adaptation of the mechanoreceptor system in a joint that was negatively influenced by joint pressure resulting from surgery.16 Kennedy reported the inhibitory effect that joint receptors would have on the quadriceps muscle group.17 Direct stimulation to the motor end plate may correct this inhibiting effect. Figure 5B-4 demonstrates the setup of a study that produced an artificially induced effusion in a normal knee. This study

Figure 5B-4    Demonstration of a study by Mangine and Brownstein in 1982,49 which showed that an injection of 70 mL   of saline in a normal knee resulted in a 70% decrease in electromyographic activity of the vastus medialis obliquus.

c­ onfirmed the influence of simulated joint effusion on muscle performance. The literature is far more complete in its support for neuromuscular stimulation and the frequently used term functional electrical stimulation (FES). Similar to the knee, the influence of injury and surgery to the shoulder presents a unique clinical observation of the rotator cuff shutdown after surgery or injury. In 2005, the results were published of a study on a controlled group of shoulder patients undergoing neuromuscular stimulation to the deltoid and rotator cuff muscles after rotator cuff and open capsular shift surgery.7 An effective outcome was demonstrated in regard to reestablishing motion in a shorter time frame than was seen in a nonstimulated group of patients, although both groups eventually regained the same motion. The ability to regain motion in a shorter time period permitted the initiation of an active strengthening program in an earlier phase of rehabilitation. The key in regaining motor control around a joint, after surgery, aids return of functional motion. The first level of training at the shoulder is the concept of coactivation. Figure 5B-5A demonstrates the pad placement in a patient 24 hours after labral repair to excite the deltoid and supraspinatus muscles. Simultaneous to current activation, the patient is instructed to contract the total cuff structure to pull the humeral head into the glenoid, that is, to perform a cuff shrug. The coactivation concept controls translation of the humeral head, reducing the risk for stress on the repaired tissue. The total treatment time is only a 10-minute cycle, which results in muscle fatigue, repeated for six cycles per day. Parameters for stimulation are set at a 10seconds-on, 10-seconds-off time and at a frequency dependent on the unit from 60 to 2000 Hz. The ramp time is unit dependent, but our protocol suggests a minimal ramp time to simulate normal physiologic muscle-firing capacity. After control of the supraspinatus is obtained, the protocol switches focus to the external rotators. Figure 5B-5B demonstrates the position used to isolate training to the external rotators. It is believed that the placement of the postsurgical shoulder into internal rotation causes the muscle spindles of the external rotators to be influenced by placing them in a lengthened position. The sequencing of training initially is to reestablish neutral position, then to gradually titrate into a 45-degree position over a 3- to 4-week period dependent on the procedure.

232

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 5B-5    A, Pad placement for excitation of the deltoid and supraspinatus muscles. B, Positioning used to isolate the external rotators of the shoulder.

The third phase of the protocol is designed to incorporate closed chain exercises into the clinical pathway to facilitate the mechanoreceptors while also minimizing the forces on the rotator cuff and capsule. Figure 5B-6 demonstrates two positions for closed chain training. This sequence gradually takes the athlete from a nonfunctional to a functional position. Treatment timing is the same as previously discussed. Literature support for the clinical use of electrical stimulation is ample for rehabilitation of the knee after injury and surgery. Delitto and colleagues reported on the use of electrical stimulation to the quadriceps muscle after ACL injury and its ability to hasten neuromuscular redevelopment of quadriceps motor control and to produce increased strength gains.12 Patterson recently reported the influence of using electrical neuromuscular stimulation in chronic quadriceps weakness and increased strength achieved with electrical stimulation.18 Others have reported similar findings, and although the long-term benefit with or without

A

electrical stimulation may show no difference, the fact that electrical stimulation assists with an earlier return of motor control and motion makes it a clinical standard. The protocol outlined follows a progression with a multistage sequence. In Figure 5B-2, we described the first step of the postsurgical protocol in which pain was managed with TENS and also established the position for pad placement. Again, this is a beneficial modality to control oral narcotic use in the first 3 to 5 days after surgery. Initial neuromuscular training program setup is to the quadriceps muscle with activation to regain control of extension range of motion and patella position. The concept of isolated training of one fiber type has not been supported in the literature; therefore, our pad placement is over the vastus medialis obliquus and femoral nerve to stimulate the quadriceps in its entirety.19 The position progresses from extension isometrics, to active extension from 40 to 0 degrees, and gradually to 90 to 0 degrees, with speed of progression

B

Figure 5B-6    A, Closed kinetic chain exercise for the upper extremity in conjunction with neuromuscular reeducation of the deltoid and supraspinatus muscles. B, Lower extremity closed chain activity with quadriceps reeducation during terminal knee extension.

Rehabilitation and Therapeutic Modalities

233

Figure 5B-7    Pad placements to elicit quadriceps reeducation.

based on pathology. Stimulation is applied with the same parameters previously described for the shoulder, and pad placement is demonstrated in Figure 5B-7 for quadriceps reeducation. The next cycle of programming is advancement to closed chain positioning to replicate gait position as a training mechanism and advancing functional positions. Gait training is done in the last 15 degrees in the closed chain position to emphasize the eccentric requirement of the quadriceps after heel strike to advance to midstance, which is often a portion the patient experiences giving way in the early phases of rehabilitation. In Figure 5B-8 the patient is postsurgical day 5, and we are stressing the terminal extension angle. The cycling of the treatment session can be customized to a 5-seconds-on, 5-seconds-off cycle, and again we recommend a decreased ramp time to produce replicable firing patterns. The final phase of functional training with neuromuscular stimulation centers on the adaptation of angles that are highly stressed. The goal at this point is to retrain joint proprioceptive sensing, which is disrupted after injury or surgery. Common angles include 15- to 30-degree flexionextension positions and the control of knee valgus moment, which has been found to have an influence on injury. ­Figure 5B-9B demonstrates an eccentric and concentric activity to control knee position to avoid valgus angulations.

BIOFEEDBACK Biofeedback is an effective modality for muscle reeducation. Biofeedback uses an amplifier attached to an electrode that is applied over the affected muscle. Clinically, biofeedback is very useful in quadriceps muscle reeducation. It is common to have muscle inhibition with joint effusion after knee injury or surgery.10 With even 30 mL of normal saline in a normal joint, the ability to contract the quadriceps may be reduced up to 50%. Once a contraction is available, biofeedback helps the patient to recruit motor units to improve strength through neuromuscular recruitment. Normal treatment time is 10 ­ seconds hold and 10 seconds rest. This is repeated for 10 minutes. Instructions to the patient are to force

Figure 5B-8    Use of transcutaneous electrical nerve stimulation unit to stress terminal knee extension.

the knee straight while ­contracting the quadriceps. This allows both muscle reeducation and active gain in extension (see Fig. 5B-9C). Biofeedback may also be used over the shoulder and scapular musculature for enhanced contraction. Placing the electrodes over the posterior cuff (teres minor, infraspinatus) is effective in reversing the effects of muscle atrophy and inhibition. Biofeedback over the shoulder can be used during traditional rotator cuff exercises such as dumbbell lifts and external rotation using Thera-Band (see Fig. 5B-9D). Another form of visual biofeedback may be applied with the use of the Monitored Rehab Systems cable column (Ft. Worth, TX) (see Fig. 5B-9E). This device combines visual input with joint positioning to encourage joint position awareness with muscle recruitment. An excellent position for posterior cuff recruitment is the 90/90 position with the patient seated. External rotation is controlled using both concentric and eccentric muscle contractions. A typical exercise session includes six sets of 60-second bouts of exercise.

IONTOPHORESIS Iontophoresis is an electrically based delivery system to transport medication and has gained wide popularity in the field of sports medicine during the past 20 years. The concept of using a direct electrical current for transdermal delivery of substances has been reported for centuries. Although the purported mechanism is still not clear, there is a theorized direct transfer of ions through the dermal layer to underlying tissue.20 Recent literature has demonstrated the effectiveness of iontophoresis in a variety of pathologies, including tendinitis, contusions, ligament sprains, myositis ossificans, muscle spasm points, and chronic edema. Puttemans demonstrated the system’s ability to deliver about 10% of the drug applied.21 Multiple studies have reported that the direct current could deliver a drug to a depth of 1 to 17 mm.22

234

A

C

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

D

E

Figure 5B-9    A, Functional closed chain training with quadriceps reeducation. B, Eccentric and concentric activity with emphasis on avoiding valgus angulation of the knee. C, Biofeedback electrode placed over the vastus medialis obliquus to reeducate the muscle: 10-second hold, 10-second rest, for 10 minutes. D, Biofeedback placed over the posterior cuff to facilitate contraction. E, Monitored Rehab Systems cable column—visual feedback to help recruit posterior cuff and enhance joint position sense: 90/90 position,   60 seconds of concentric and eccentric muscle contractions.

The most common drugs used are dexamethasone, acetic acid, lidocaine, and salicylates. In a clinical study on plantar fasciitis, Gudeman and associates demonstrated the effectiveness of dexamethasone after six visits.23 Mangine and colleagues, in a series of unpublished studies presented at a national conference, compared iontophoresis to ultrasound in rotator cuff patients and found a decrease in pain by two levels compared with the ultrasound group.24 These data are consistent with Delacerda’s reported results in shoulder patients.25 Early iontophoresis units were limited by U.S. Food and Drug Administration approval on current delivery to 40 milliampere-minutes, secondary to concerns about dosage level and complications. The most frequent complication noted is skin reaction due to negative electrode buildup of histamine resulting in a reaction. This reaction was minimized by the development of the pH-buffered pads. One of our initial clinical studies on iontophoresis set out to determine whether there existed a dosage unit relationship between controlled groups. Two groups were studied; one was treated for six visits at 40 milliampereminutes, and the second received an increased dosage to 80 milliampere-minutes. Patients who received the higher

dosage consistently reported lower pain level on an analog pain scale. Anderson and associates suggested that magnitude and duration of iontophoresis should be considered factors when treating musculoskeletal dysfunction.26 Flexibility in providing this procedure has improved with the advancement of the single-use disposable patch system. This mode can deliver medication during athletic activity, whether practice or competition. Figure 5B-10 demonstrates the use of iontophoresis in both a clinical and game situation. Our experience is that the delivery benefits the athlete by modulating the pain with activity, as part of a comprehensive rehabilitation approach.

CRYOTHERAPY The most commonly applied modality and staple of sports medicine is cryotherapy. Few procedures are as supported as the use of a cooling mechanism after injury to reduce the inflammatory effects of the injury. The principles of ice, compression, and elevation are time tested and have been proven valid through numerous studies. The effects of cryotherapy include a reduction in edema through decreased blood flow by the mechanism of

Rehabilitation and Therapeutic Modalities

A

235

B

Figure 5B-10   A, Clinical application of iontophoresis. B, Application of iontophoresis for use during athletic participation.

v­ asoconstriction, a reduction in the inflammatory response through a decreased metabolic rate, a reduction in pain through the gate control theory, and temporary inhibiting effects to the neuromuscular system related to spasticity, nerve conduction velocity, and muscle strength.27-30 Cryotherapy can be administered through a variety of methods, including ice bags, cold water immersion, ice cups, thermoelectric cooling, cold packs, chemical spray (Cryostretch), and continuous cold flow through a device with intermittent compression.28,29 Numerous studies have demonstrated an increased reduction in edema during the acute phase, with the administration of continuous cryotherapy using intermittent compression.31,32 The most practical method of administration is with an ice bag or vasopneumatic device. Figure 5B-11 demonstrates application of ice clinically. The treatment time for ice application should not surpass 30 minutes because of risk for possible frostbite and nerve palsy. A treatment period of 20 minutes has been proven adequate to sustain the desired tissue temperature. Two cycles of 10 minutes on and 10 minutes off after the initial 20 minutes produced a significantly greater effect on decreased blood flow. Indications for use of cryotherapy include the following: • Acute trauma • Postoperative treatment • Chronic inflammation

A

• Spasticity • Pain modulation • Edema control • Inflammatory reduction • Cryokinetics Contraindications include cold hypersensitivity, cold intolerance, Raynaud’s disease, and use over an area of vascular compromise. Precautions include placement over an open wound, hypertension, and mental impairment or decreased sensation.

LASER The term laser is an acronym for light amplification by the stimulated emission of radiation. In the early 1900s, Albert Einstein spoke in theory of a process called stimulated emis­ sion, which would eventually lead the way to the invention of the laser. During the late 1960s in Hungary, a researcher named Endre Mester conducted numerous studies on the effects of the low-level laser. The studies were performed to confirm if the laser would cause cancer, but to Mester’s surprise, he discovered a biostimulatory effect that led the way for further research of the laser.33 In constructing the laser, there must be a medium that combines the atoms together to form the laser beam. Most low-level lasers are HeNe (helium-neon) lasers consisting of a helium-neon

B

Figure 5B-11    A, Cryotherapy through use of a vasopneumatic device. B, Cryotherapy applied in conjunction with electrical stimulation for pain modulation.

236

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

gas as the medium. The HeNe laser was originally invented in the 1960s and emits laser light at a wavelength of 632.8 nanometers (nm). The GaAs (gallium-arsenide) laser was later invented in the 1980s and emits laser light at a wavelength of 904 nm. For clinical use, the intensity of laser light is measured in watts per centimeter squared (W/cm2) or joules per centimeter squared (J/cm2).34,35 Both the HeNe laser and GaAs laser are used in today’s practice. Effects of low-level laser therapy include the following36: • Biostimulation and photostimulation • Reduction of pain through opiate production • Stimulation of cellular healing through increased ionic transport of sodium and potassium • Inhibition of the inflammatory process • Increase in cellular adenosine triphosphate production Indications for low-level laser include a variety of conditions such as wounds, fractures, acute and chronic soft tissue musculoskeletal injuries, and pain control. Most of the published literature on lasers pertains to low-level laser and the effect on wound healing. Numerous studies have demonstrated positive cellular responses to promote cell migration and proliferation through the stimulation of mitochondrial activity as a result of low-level laser application.37,38 The low-level laser causes a significant increase in fibroblast proliferation.39,40 Unfortunately, only a limited number of studies exist on the treatment of musculoskeletal injuries, and of those studies, both positive and negative outcomes have been demonstrated. A 2003 study by Saunders looked at supraspinatus tendinosis treated with the low-level laser compared with ultrasound.41 The results found that low-level laser was a more effective treatment in reducing symptoms. The proposed effect in treating chronic tendinosis with respect to reducing the inflammatory process of a possible acute episode was through the biostimulatory effects of the low-level laser treatment. Fibroblastic proliferation40,41 and pain modulation37 were two other hypothesized effects. On the contrary, other studies have demonstrated no positive outcomes from use of the low-level laser. A 2006 study by Maher found the laser to be an ineffective treatment intervention for lateral epicondylitis.42 It has been hypothesized that the parameters used in this study were too high to stimulate fibroblastic proliferation but rather inhibited the process. The general recommendation of low-level laser therapy has been indicated for use in rehabilitation due to the biostimulatory effects of the light energy. The recommended dosage for acute injuries is between 2 and 4 J/cm2 and up to 40 J/cm2 for chronic conditions.27 Currently, clinical anecdotal experience has often been reported as favorable when using laser therapy for muscle contusions, acute and chronic ligament injuries, acute and chronic tendon injuries, and pain modulation. The patient population treated with low-level laser therapy within our clinical setting generally has reported a pain reduction using the visual analog scale. The current recommendation for this type of treatment with the conditions listed previously, in conjunction with a rehabilitation program, use parameters of J/cm2 and a GaAs laser in the subacute and chronic stages of healing. In the acute phase, dosages between 2 and 4 J/cm2 are recommended. It is important

Figure 5B-12    Laser application for treatment of a deep calf contusion.

to keep in mind that overexposure may cause inhibitory effects; thus, the clinician should regularly assess the effects of treatment to adjust the dosage accordingly.35,37 Absolute contraindications for the use of laser therapy include pregnancy, photosensitivity, immunosuppression, and exposure to cancerous cells, as well as the need to avoid the thyroid gland and the epiphysis. Precautions should be taken with patients receiving botulinum toxin (Botox) injections, anti-inflammatory medications, or steroid injections. Avoid direct exposure with the eyes because retinal damage may occur. The clinician administering the laser and the patient should be wearing appropriate protective glasses.35 Figure 5B-12 demonstrates the application of a laser to an athlete with a deep calf contusion. The limited and unreliable research on the laser makes its use very controversial. The evidence supporting the laser shows promising results; however, more randomized ­controlled trials are needed to establish definitive recommendations for clinical use in rehabilitation programs. Further, the need for studies, which define the parameters of application to improve clinical efficacy, would be ­beneficial.

ULTRASOUND Ultrasound is the most commonly used modality in general clinical practice; however, in sports medicine, its use pales in comparison with other modalities such as cryotherapy and electrical stimulation. Therapeutic ultrasound is used for two common purposes: nonthermal tissue healing effects and thermal effects on tissue extensibility. Therapeutic ultrasound energy is generated by acoustic sound waves at high, inaudible frequencies (1 and 3.3 MHz). These acoustic sound waves are generated by applying an electrical current through a transducer composed of a metal plate adhered to a piezoelectric crystal. The crystal converts the electrical energy into ultrasonic energy through a process known as the piezoelectric effect. The two major frequencies used clinically—1 MHz (low frequency) and 3 MHz (high frequency)—are effective in treating both deep and superficial soft tissues, respectively. Low-frequency ultrasound is effective in treating

Rehabilitation and Therapeutic Modalities

s­ tructures deeper than 2 cm to the point of application, whereas high-frequency ultrasound is effective for more superficial tissues—those 1 to 2 cm in depth.43,44 The thermal effects of ultrasound are most frequently applied to an athletic patient population for purposes of enhancing tissue extensibility before stretching or to address joint contractures and loss of range of motion after injury or surgery. To achieve therapeutic effects through heating, tissue temperatures must rise to and be maintained at 40° to 45° C for at least 5 minutes with a continuous duty cycle. These tissue temperature changes can be accomplished through manipulation of the intensity of the sound energy delivered and the duration of the application. Although the depth of penetration is less, tissue temperature rises at a faster rate at 3 MHz than 1 MHz, indicating that a lower intensity or shorter duration of the application is necessary when using higher frequencies.44 The goal of nonthermal ultrasound when applied to the athletic population is to accelerate tissue healing by altering cellular activity during all three stages of tissue healing. These effects have been demonstrated in bench research studies in all three phases of healing: inflammatory, proliferation, and remodeling. In the inflammatory stage of healing, ultrasound has been demonstrated to stimulate the activity of macrophages and neutrophils, facilitating their phagocytotic function of dispensing devitalized tissue and debris from the area of acute injury.45 Young and Dyson, in an animal model, concluded that the use of ultrasound in the acute phase of tissue healing resulted in a shorter time span of this phase of healing.46 In the proliferation stage of tissue healing, ultrasound has been demonstrated to increase fibroblastic mobility and proliferation47 as well as stimulate collagen production.48 In addition to increased collagen production, animal studies have revealed that when ultrasound is applied to healing tendons, the newly laid collagen is better organized and demonstrated greater tensile strength with intensities of 0.5 W/cm2.45 In the remodeling phase of tissue healing, thermal ultrasound can be employed for its ability to improve tissue extensibility with the goal of scar remolding and tissue reorganization. Ultrasound can be effectively applied in treatment of the athletic population for both thermal and nonthermal effects on tissue healing. The choice of ultrasound as a modality includes the selection of appropriate parameters for treatment and the desired therapeutic results. For thermal effects, a higher intensity (1.0 to 2.5 W/cm2) and a

237

continuous duty should be selected to heat the desired tissue to temperatures as previously described. For nonthermal effects, a pulsed duty cycle (20% to 50%) and an intensity of 0.5 to 1.0 W/cm2 can be used to stimulate cellular activity in the inflammatory and proliferation phases. Selection of the sound head size is determined by the treatment area and should not exceed 2 times the area of the sound head, and an effective coupling agent should be used to deliver the treatment.

C

r i t i c a l

P

o i n t s

l The use of modalities in rehabilitation should constitute only a portion of an intervention program. l The application of modalities in rehabilitation requires sufficient knowledge of biomechanics and principles of tissue healing in order to maximize outcomes. l It is important to remain current with the literature when using modalities to provide best practice through correct application and methods. l The overall goals of a rehabilitation program should be functional and based on patient desires. l All patients should be thoroughly secreened and evaluated prior to the ­application of modalities to ensure safety.

S U G G E S T E D

R E A D I N G S

Andrews JR, Harrelson GL, Wilk KE: Physical Rehabilitation of the Injured ­Athlete, 3rd ed. Philadelphia, Saunders, 2004. Cameron M: Physical Agents in Rehabilitation: From Research to Practice, 3rd ed. Philadelphia, Saunders, 2003. Kisner C, Colby L: Therapeutic Exercise: Foundations and Techniques, 5th ed. Philadelphia, FA Davis, 2007. Manske R: Postsurgical Orthopedic Sports Rehabilitation: Knee and Shoulder. Philadelphia, Mosby, 2006. Michlovitz SL, Nolan TP: Modalities for Therapeutic Intervention, 4th ed. Philadelphia, FA Davis, 2005. Prentice W: Therapuetic Modalities in Rehabilitation. New York, McGraw-Hill, 2005.

R eferences Please see www.expertconsult.com

238

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n

C

Therapeutic Exercise Prescription Ron M. Johnson

The purpose of this chapter is to provide rehabilitation specialists with a resource that will help them use therapeutic exercises to correct impairments in motor function and muscle performance. Evidence points to the fact that the use of appropriate therapeutic exercise has been the foundation for most successful rehabilitation programs for most musculoskeletal impairments. Indeed, systematic reviews have concluded that therapeutic exercise is the treatment of choice (and often the only proven therapeutic modality) for disorders of the neck,1 low back,2 shoulder,3-6 knee,7 and ankle.8 Historically, exercise strategies and techniques have varied in their scope and effectiveness. It is the goal of this chapter section to present and outline therapeutic exercises that have been validated through research and through consensus of treating clinicians. It is also the goal to provide this information in a very practical and applicable manner. Whether or not rehabilitation specialists employ these practices, this chapter section will undoubtedly provide an excellent resource, practical guide, and rationale for the inclusion of a number of these exercises in their rehabilitation arsenal.

REHABILITATION CONCEPTS FOR THE APPLICATION OF THERAPEUTIC EXERCISE To provide the necessary framework for the clinician, the rehabilitation concepts for the application of therapeutic exercise are first outlined and discussed. These concepts are then applied to specific rehabilitation programs for the shoulder, elbow, knee, and foot and ankle. Additionally, a section is devoted to the concept of “core strengthening” in the treatment of low back disorders. • The application of therapeutic exercises must be based on a thorough and continual clinical evaluation. • The treating clinician should be acutely aware of which tissues are affected, whether through injury or surgical intervention. The clinician should first ask, What tissues are affected? What exercises, motions, and forces should be avoided? Modifications and specific applications of exercises are then employed that avoid overstressing the healing ­tissue. • Each joint in the body has a specific function and is prone to spe­ cific, predictable levels of dysfunction. As a result, each joint has specific training needs. It is beyond the scope of this chapter section to detail exercises for specific protocols of injuries and disorders. However, as noted previously,

the general needs of each joint are outlined and thus provide the framework for most rehabilitation protocols. This leads to the simple but most profound idea that our efforts in rehabilitation, and performance enhancement, are to effectively restore the optimal movement patterns for which the body was designed. Period. We find that this approach is best for all joint pathologies considered. Table 5C-1 outlines the general and overriding function of each joint. • The rehabilitation specialist should follow a continuum of therapeutic exercise application. The acronym AIR can be used to help outline this approach to rehabilitation and training. • Activate (A): Initially, therapeutic exercises are designed to isolate, activate, and strengthen weak muscles. In most cases, an injured athlete with a dysfunctional joint needs to target and train certain muscles first to allow for the most safe and efficient return to function. Primarily from electromyography (EMG) research, exercises that allow maximal activation of the muscle and rely little on synergistic activity of surrounding muscles are performed. • Integrate (I): Once these weak muscles are “activated” and strengthened in isolation, exercises are then provided that integrate the muscles through movements in more functional patterns—pushing, pulling, squatting, lunging, and so forth. • Reinforce (R): Finally, therapeutic exercises are given that reinforce the proper motor pattern throughout the functional and sport-specific needs of the individual. This encompasses the specific movement patterns, neuromuscular control needs, and energy demands of the activity. • Proper therapeutic exercise form and motor control should never be compromised. It should go without saying, but proper technique and form should never be compromised. Improper intensity and insufficient supervision most often lead to poor exercise performance. This may    TABLE 5C-1  General and Overriding Functions   of Each Joint Joint

Primary Function and Training Needs

Ankle Knee Hip Lumbar spine Thoracic spine Scapulothoracic Glenohumeral

Mobility Stability Mobility, stability Stability Mobility Stability Mobility, stability

Rehabilitation and Therapeutic Modalities

overstress healing structures and slow the individual’s recovery. • Provide an appropriate amount of variety in therapeutic exer­ cise prescription. Doing the same exercise routine repeatedly leads to psychological and orthopaedic burnout; however, too much variability leads to little or no progress. Providing variety helps to keep things interesting, improves compliance, and helps to ensure that the program is encompassing all aspects of training. • Stick to the basics. In most cases, there is a core group of exercises that should be adhered to and built on when designing a rehabilitative or strength and conditioning program. These are the exercises that have been ­established in research and in the clinical settings. As these exercises are mastered, different variations of these exercises and new innovative training ideas can be introduced. It is not the goal of this chapter to focus on those “special” exercises that may look impressive and unique but that do not lend themselves to advancing the most important needs of the individual. Before proceeding, a general overview of important basic rehabilitation principles needs to be outlined. Wilk and colleagues have outlined in detail a four-phase approach to rehabilitation.9 The phases are progressive and sequential in nature and should be implemented as such in order to safely and most effectively return the individual to his or her activities. Table 5C-2 briefly details the phases and general goals for each phase in a rehabilitating setting.

SHOULDER COMPLEX Scapulothoracic Joint In most cases, when designing a rehabilitation program, therapeutic exercises first focus on restoring proper function of the scapulothoracic joint. This is the case for a number of reasons. The scapula is the base from which the rotator cuff originates and allows for more normal scapular positioning and, therefore, more normal patterns of activation of the rotator cuff. In addition, sufficient scapular control is needed to elevate the acromion and decrease the chances of rotator cuff impingement within the coraco­ acromial arch. Any altered scapular muscle function, weakness, or inability to position and stabilize the scapula results in a direct effect on the shoulder joint with often dire consequences. Furthermore, performance of scapular training is often tolerated well and rarely exacerbates conditions related to the glenohumeral joint, shoulder capsule, or rotator cuff; it is therefore a safe way to begin the exercise program. Logic dictates that this is a joint that needs stability. Although most rehabilitation programs include exercises for the rotator cuff, there are a number of individuals who are not training the scapulae to stabilize sufficiently. It has been said that having a strong rotator cuff without a stabilized scapula is like trying to shoot a cannon from a canoe. One common scapular dysfunction is the inability to maintain the scapula in a stable and neutral position when the arm is performing dynamically. The scapula must provide strong retraction and downward rotation forces to counteract the pull of the arm in space, such as during the

239

   TABLE 5C-2  General Goals for Each Phase   in a Rehabilitating Setting Phase

Goals

Focus of Training

Acute

Diminish pain and inflammation Advance, normalize motion Address postural and flexibility limitations Modify activity Activate, isolate, and strengthen weak muscles Progress strength and neuromuscular training Begin to integrate strength into functional movements Advance, normalize motion Address postural and flexibility limitations Promote dynamic stability Reinforce therapeutic exercises to (1) include more aggressive strength training, (2) advance dynamic stability training, and (3) improve strength, power, and muscular endurance Progress activity and sport-specific strength and conditioning program Return to activity or sport

Modalities as needed Flexibility, stretching Strength and neuromuscular training

Intermediate

Advanced strengthening

Return to activity

Progress above as indicated Initiate kinetic chain flexibility, stretching Initiate core training

Progress above as indicated Initiate plyometric training Initiate interval return-to-activity program

Progress above as indicated Strength and conditioning Progress plyometric training Progress interval return-to-activity program

c­ ocking and acceleration phases of the throwing motion, or with simple reaching and lifting tasks. Training programs should aim to lock the scapula in place in a retracted and depressed position with the rhomboids, middle trapezius, and lower trapezius for the glenohumeral joint to function properly. The scapula must also be trained to provide stability of the shoulder through proper mobility. That is, the scapula must move in a coordinated manner to allow a congruent connection with humeral movement. Of the typical 180 degrees of overhead reach in a healthy shoulder, the scapula’s upward rotation is responsible for about 60 degrees of it. It achieves this through the synergistic efforts of the “upward” rotators: the upper trapezius, lower trapezius, and serratus anterior. Insufficiencies in these efforts may lead to abnormal scapular motion and the inability to maintain proper congruity with movement of the arm. In most cases, the abnormal pattern is an overactivation of the upper trapezius combined with decreased control or strength of the lower trapezius and the serratus anterior.10-16 For this reason, therapeutic exercises that

240

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B

Figure 5C-1    Early therapeutic exercises for the lower trapezius. A, Prone rowing. B, Prone horizontal abduction. C, Prone extension.

activate and train the lower trapezius and serratus anterior are often vital in rehabilitating and training the shoulder complex. Furthermore, it is an added benefit when these muscles can be trained in a manner that minimizes contributions from the upper trapezius. Research identifying activation of individual scapular muscles is discussed, and indications for therapeutic applications are given. Later in the chapter section, thoughts and discussion turn to how to best design a practical rehabilitation program for the entire shoulder complex.

Lower and Middle Trapezius We have found that the inability to maintain scapular downward rotation and retraction is often due to weakness or inhibition of the lower trapezius muscle. Early rehabilitation efforts should focus on restoring function of this key scapular muscle. EMG studies have shown high activations of the lower trapezius with a number of exercises. Standing scapular retraction with simultaneous bilateral shoulder external rotation,17 rowing,18 prone horizontal abduction,18,19 and prone extension18 can all be introduced early in the rehabilitation program without overstressing the shoulder (Fig. 5C-1). Additionally, these exercises also recruit the middle trapezius effectively. To further target the middle trapezius and rhomboid muscles, rowing exercises can be advanced to allow for complete scapular muscle training.18 The emphasis with rowing exercises is to allow the individual to train for scapular retraction. Rowing can be performed sitting or

A

standing with elastic tubing or bands20,21 or with cable column–type machines (Fig. 5C-2). Once the individual has progressed with these initial lower and middle trapezius training exercises, or when symptoms allow, prone training advances to include horizontal abduction with external rotation in line with the muscle fibers of the lower trapezius. This movement has been proved the greatest lower trapezius activator (Fig. 5C-3).22-24

Serratus Anterior Generally, the studies report increasing EMG activity in the serratus anterior as the shoulder is actively elevated to end range. Exercises that create upward rotation of the scapula have demonstrated much more EMG activity in the serratus anterior than straight scapular protraction exercises. Donatelli and colleague’s EMG analysis of the serratus anterior determined that a diagonal exercise with a combination of shoulder flexion, horizontal flexion, and external rotation (D1 Flexion proprioceptive neuromuscular facilitation [PNF] pattern) and shoulder abduction in the plane of the scapula above 120 degrees generated the highest levels of EMG activity (Fig. 5C-4).22 Moseley and coworkers also found that shoulder elevation exercises from 120 to 150 degrees produced maximal EMG activity in the serratus anterior.18 Because the trapezius and serratus anterior muscles work synergistically to produce upward rotation of the scapula,

B

Figure 5C-2    Rowing variations. A, Rowing with elastic resistance. B, Seated rows on a cable column.

Rehabilitation and Therapeutic Modalities

241

exercise (Fig. 5C-5) and the push-up plus (protraction) exercise. Lear and Gross documented by dynamic EMG the changes in serratus anterior with a common push-up progression.27 Their results demonstrated increased serratus anterior activation as the feet were elevated and the upper extremity weight-bearing surface became less stable (Fig. 5C-6). Uhl and coworkers documented the increased shoulder musculature activations when weight-bearing forces through the upper extremities increased.28

Other Scapulothoracic Muscles

Figure 5C-3    Isolation of the lower trapezius. Horizontal abduction with external rotation in line with the muscle fibers of the lower trapezius.

the exercises that activate these two muscles simultaneously are valuable to both strengthen and train proper scapular function. The prone arm raise overhead (horizontal abduction with external rotation at 135 degrees) was found to work these muscles together most effectively.22 Furthermore, this exercise was shown by Cools and investigators not only to effectively target the serratus anterior and lower trapezius but also to effectively minimize activation of the upper trapezius.10-14 The horizontal abduction with external rotation exercise frequently is promoted for optimal shoulder rehabilitation. Townsend and associates,25 as well as Moseley and coworkers,18 included this exercise in their selection for glenohumeral and scapulothoracic muscle strengthening programs. Many authors recommend scapular protraction exercises for serratus anterior muscle strengthening. Decker and colleagues26 and Hintermeister and associates20 performed scapular protraction-type exercises and found maximal activity in the serratus anterior in the dynamic hug

Figure 5C-4    Serratus anterior activation. D1 flexion combines shoulder flexion, horizontal adduction, and external rotation (D1 flexion proprioceptive neuromuscular facilitation pattern).

Although not often found to be primary contributors to shoulder dysfunction, researchers have also examined EMG amplitudes of the latissimus dorsi, pectoralis major, and pectoralis minor during common therapeutic exercises for the shoulder complex. Strengthening of the entire shoulder complex, particularly when recovering from injury or surgery, is important for optimal return to function and performance. The press-up (Fig. 5C-7) was found to be the most effective exercise that was included in the EMG research by both Townsend and associates25 and Bradley and Tibone29 in recruiting the pectoralis major and latissimus dorsi muscles. Likewise, Moseley found the press-up to be the exercise that most activated the pectoralis minor.18 The push-up was found to be the second most activating exercise for the pectoralis major and minor muscles.18,25

Rotator Cuff The glenohumeral joint is designed for maximal mobility to allow for the wide ranges of shoulder function. Unfortunately, this mobility comes at the expense of stability. The rotator cuff muscles play a vital role in normal shoulder arthrokinematics. The rotator cuff musculature maintains stability by compressing the humeral head into the glenoid fossa during upper extremity motion. Furthermore, a ­ sufficiently compressed humeral head allows for ­appropriate clearance of the rotator cuff within the subacromial space, reducing the potential for impingement.

Figure 5C-5    Dynamic hug. This movement includes shoulder horizontal adduction with an emphasis on maximal scapular protraction.

242

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 5C-6    Push-up plus progressions. A, Feet elevated. B, On an unstable surface.

Figure 5C-7    Press-up. This exercise trains scapular downward rotation and depression.

Beyond stabilization, each of the muscles also individually contributes to humeral motion.

Infraspinatus and Teres Minor Sufficient strength of the external rotators (infraspinatus and teres minor), in particular, is integral during the overhead motion to develop an approximation force on the upper arm at the shoulder to prevent joint distraction and impingement.30 Exercises designed to strengthen the

A

B

r­ otator cuff with minimal deltoid involvement are often desired to minimize the amount of superior humeral head migration, thus reducing the chance of subacromial impingement. Reinold and colleagues measured the EMG activity of the infraspinatus, teres minor, supraspinatus, and deltoid muscles during commonly prescribed shoulder rehabilitation exercises.31 Of the seven common shoulder exercises, they determined that side-lying external rotation exercise produced the greatest amount of EMG activity for the infraspinatus and teres minor. For both posterior rotator cuff muscles, standing external rotation in the scapular plane and prone external rotation at 90 degrees of shoulder abduction were the following two most effective ­exercises regarding their activations (Fig. 5C-8). Note that the scapular plane external rotation has been modified by performing the external rotation with the elbow resting on the knee in a supported side-lying position. We believe this allows better stabilization and isolation. In another similar study, Cools and his team showed that the side-lying external rotation exercise provided high EMG recording for the infraspinatus and teres minor muscles.10-14 Blackburn and coworkers analyzed shoulder rehabilitation exercises with EMG electrodes in the supraspinatus, infraspinatus, and teres minor muscles.32 The infraspinatus and teres minor muscle activity was greatest with external rotation with the shoulder abducted 90 degrees and the elbow flexed 90 degrees in the prone position (see Fig. 5C-8C). The second highest recording of these muscles was found with prone horizontal abduction at

C

Figure 5C-8    External rotation training. A, Side lying. B, External rotation in the scapular plane. We have found best results with the individual supporting the elbow on the knee. C, Prone external rotation with the shoulder abducted 90 degrees.

Rehabilitation and Therapeutic Modalities

243

Figure 5C-9    Prone horizontal abduction at 90 degrees with arm externally rotated.

90 degrees with arm externally rotated (Fig. 5C-9). Interestingly, the only exercise in which the teres minor was more active than the infraspinatus was the prone shoulder extension with the arm externally rotated (Fig. 5C-10). In each of the exercises tested by Blackburn and colleagues, higher posterior cuff activation was seen when the arm was placed in an externally rotated position (compared with a neutral or internally rotated position).32 Likewise, Townsend and colleagues25 and Bradley and Tibone29 recorded the activity of the glenohumeral muscles in 17 identical shoulder rehabilitation exercises for professional baseball players. Overall, they each concluded that the horizontal abduction exercise with the arm externally rotated and standing external rotation with the arm at the side (Fig. 5C-11) most highly activated the posterior rotator cuff ­musculature.

Supraspinatus Several authors have analyzed the EMG activity of the supraspinatus musculature and deltoid muscles to ­determine exercises that produce the most supraspinatus activity with the least deltoid involvement.33-38 This has been ­theorized to avoid potentially impinging effects from

Figure 5C-10    Prone shoulder extension with the arm externally rotated.

Figure 5C-11    Standing external rotation with the arm at the side.

superior humeral head migration associated with high deltoid ­activity. Restoration and maintenance of ­supraspinatus strength are important components in achieving optimal shoulder function, and strengthening of the supraspinatus should be an important component of shoulder ­rehabilitation. Studies regarding optimal supraspinatus muscle activation have been somewhat conflicting, and debates between research methods have been seen in the literature. Jobe and Moynes first suggested that abduction in the scapular plane with internal rotation, the so-called empty-can exercise, is the optimal exercise position for isolating the supraspinatus muscle (Fig. 5C-12).39 However, this conclusion was based on the surface EMG data of a single subject without comparison to other exercises. Townsend and associates reported that this exercise is second to the military press for exercising the supraspinatus muscle and recommended its inclusion in the rehabilitation program because it better isolated the muscle.25 However, Blackburn and colleagues found that the prone position with the elbow extended and the shoulder abducted to 100 degrees and externally rotated (Fig. 5C-13) produced the greatest amount of EMG activity in the supraspinatus muscle.32 It is noted that they did include the empty-can exercise in their research for comparison. Malanga and coworkers

Figure 5C-12    Empty-can exercise. Shoulder elevation in the scapular plane with internal rotation.

244

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

because of its inherent safety and its ability to sufficiently activate and isolate the supraspinatus. One final exercise to note was introduced by Horrigan and investigators.33 They compared the emptycan ­ exercise, military press, and side-lying abduction to 45 degrees in their ability to activate the supraspinatus muscle (Fig. 5C-15). MRI before and after the exercises showed the greatest increase in signal intensity of the supraspinatus with the side-lying abduction to 45 degrees. It was their conclusion that this can be a key exercise for supraspinatus isolation, especially because of its ability to avoid positions that can potentially lead to subacromial impingement. Figure 5C-13    Prone horizontal abduction at 100 degrees with arm externally rotated.

sought to directly compare these two positions and found that both positions resulted in significant activity of the supraspinatus, but the difference between these two positions was not statistically significant.35 Similarly, Kelly and associates reported that there was no significant difference in supraspinatus muscle EMG activation between the empty-can exercise and abduction in the scapular plane with external rotation, the so-called full-can exercise (Fig. 5C-14).34 Kelly and colleagues did note that the full-can position demonstrated less activation of the infraspinatus and deltoid muscles, allowing for superior isolation of the supraspinatus. In addition, magnetic resonance imaging (MRI) evaluation has not revealed differences in supraspinatus muscle relaxation time37 or signal intensity33 when comparing the full-can and empty-can exercises. The safety of performing the empty-can exercise should be questioned. Humeral abduction coupled with humeral internal rotation has been shown to decrease the size of the subacromial space40-42 and maximally approximate the supraspinatus to the anterior acromion.42-44 Thigpen and coworkers studied the scapular kinematics of these two exercises and found increased scapular internal rotation and anterior tipping with the empty-can exercise decreased the supraspinatus outlet during the empty-can exercise.45 To conclude this debate on the full-can versus the empty-can exercise, we include the full-can version

Figure 5C-14    Full-can exercise. Shoulder elevation in the scapular plane with external rotation.

Subscapularis The subscapularis receives quite a bit of work with the high volume of internal rotator work (pectoralis major, latissimus dorsi, anterior deltoids, and teres major) in most training programs. Therefore, in most cases, direct subscapularis training is not necessary unless a specific strength deficit is present. The perceived relative unimportance of subscapularis training is noted by the scarce amount of research done on training this muscle. However, subscapularis and subsequent internal rotation deficits have been noted after anterior shoulder stabilization procedures.46,47 In the immediate and early strength training phases, direct subscapularis training is warranted. Tubing or cable column internal rotation with the arm at the side should be included. In the most noted subscapularis study, Decker and colleagues demonstrated that the push-up plus is superior to traditional internal rotation exercises for activating both functional portions of the subscapularis muscle.48

Advanced Dynamic Shoulder Training As the individual progresses to the advanced strengthening and return-to-activity phases, more dynamic and forceful therapeutic exercises are introduced. The rehabilitation specialist is cautioned to resist the temptation to initiate these exercises too soon into the program for the sake of doing something more novel and unique to impress the individual. Remember, progressing the already established “most important” exercises are still the key to successfully completing the rehabilitation program.

Figure 5C-15    Side-lying shoulder abduction to 45 degrees.

Rehabilitation and Therapeutic Modalities

A

245

B

Figure 5C-16    Advanced training. A, D2 proprioceptive neuromuscular facilitation. Combination of shoulder flexion, horizontal abduction, and external rotation. B, External rotation at 90 degrees of abduction.

Elastic resistance and cable column exercises can be advanced to include overhead training in more functional and dynamic movement patterns. PNF patterns and external and internal rotation training at 90 degrees of abduction are introduced to progressively challenge the scapular and rotator cuff muscles (Fig. 5C-16). Dependent on the individual’s needs, movements mimicking throwing, golf swinging, and swimming can be included as well (Fig. 5C17). Myers and associates demonstrated high rotator cuff and scapular muscle activation in simulated throwing exercises with the use of rubber tubing.21 Implementation of the Body Blade in the advanced phases can also help to advance shoulder strength training and muscle activation. Indeed, Lister and colleagues found higher EMG activity of the lower trapezius and serratus anterior with the use of the Body Blade, as directly compared with Thera-Band and cuff weights.49 Figure 5C-18 demonstrates two of a number of common exercises that can be performed with the Body Blade. A more recent trend in rehabilitation and strength training is the use of unstable training implements, such as stability balls, wobble boards, foam pads, and balance disks. A common assumption is that an unstable surface places an

A

B

increased demand on the neuromuscular system to provide stability due to the labile surface. The purported benefits of training with this instability are improvements in joint proprioception and greater muscle activation requirements. Indeed, unstable surface training with the torso on the unstable surface has been shown to increase core musculature activation when compared with stable-surface exercises.50,51 However, studies have not shown that unstable training increasingly activates the rotator cuff or scapular muscles.52 Nonetheless, we believe that individuals can perform the previously discussed traditional exercises in such a manner that equally targets these muscles with the added benefit of challenging and training trunk and core stability. Individuals can progress their prone dumbbell training from the table or bench to training on the stability ball (Fig. 5C-19). Likewise, push-ups and other scapular protraction exercises are advanced through their performance on unstable surfaces such as physio balls, wobble boards, and medicine balls, as well as elevating the feet (Fig. 5C-20). Direct scapular retraction training should be advanced as well. Single-arm rowing with cable columns, bent-over rowing with dumbbells, standing cable rows with rotation (Fig. 5C-21), and pull-down exercises (Fig. 5C-22) can all

C

Figure 5C-17    Advanced training to include sports-specific motions. A, Throwing. B, Golf swing. C, Swimming.

246

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

Figure 5C-18    Body Blade training. A, D2 flexion. B, Shoulder abduction. Virtually any shoulder motion can be performed.

be performed to allow for a complete shoulder complex rehabilitation. Full devotion to therapeutic exercise selection for the shoulder complex would not be complete without mentioning the consideration for manual resistance training as well. The application of manual resistance can be vital to the development of appropriate muscle and movement ­activations and can be included within the individual’s

­ rogram at any time during rehabilitation. Manual resisp tance allows the clinician to apply the appropriate resistance at specific ranges during the exercises. The clinician is also better able to assess and provide feedback to the individual regarding muscle activation and input. PNF patterns for the scapula and shoulder and specific exercises for each muscle can all be conducted through manual resistance (Fig. 5C-23).

A

B

C

D

Figure 5C-19    Prone stability ball training. A, Prone horizontal abduction at 100 degrees with arm externally rotated. B, Prone horizontal abduction at 90 degrees with arm externally rotated. C, Prone extension with arm externally rotated. D, Prone external rotation with the shoulder abducted 90 degrees and the elbow flexed 90 degrees.

Rehabilitation and Therapeutic Modalities

A

B

247

C

Figure 5C-20    Push-up plus variations. A, Feet elevated on ball. B, Hands on ball. C, Closed chain upper extremity walk-ups and walk-overs on step.

A

B

C

Figure 5C-21    Advanced rowing variations. A, Single-arm rowing with cable columns. B, Bent-over rowing with dumbbells. C, Standing cable rows with rotation.

Gym Training Considerations In the advanced strengthening and return-to-activity phases, it is the goal of many individuals to return to their previously performed strength routines in the commercial gym setting. Athletes also need to be reintegrated into their team’s strength and conditioning programs. For these individuals, certain guidelines and suggestions have proved helpful in their full recovery and full return to strength training.

stresses on the shoulder complex. The use of dumbbells allows the individual to select the arm positions that are most comfortable, and the floor acts to limit the stresses to the anterior shoulder (Fig. 5C-24). Once the individual has progressed with the floor press, and no other clinical

Bench Press Modifications to the bench press need to be considered for the individual recovering from a shoulder injury or surgery. The heavier loads used with pressing place high demands on the rotator cuff to stabilize the glenohumeral joint, particularly as the bar is lowered to the chest. Research has demonstrated that bench pressing may increase the risk for shoulder injury, including anterior shoulder instability, atraumatic osteolysis of the distal clavicle, and pectoralis major rupture.53 High glenohumeral capsulolabral stresses can also occur with bench pressing. Anterior shoulder and capsular stresses occur as the bar approaches the chest, whereas posterior capsular stresses occur when the arms are extended in front of the body. To safely reintegrate the bench press, a return-tobench press progression should be followed. The dumbbell floor press is first used to protect and lessen the

Figure 5C-22    Pull-down.

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

248

A

B

Figure 5C-23    Manual resistance. A, Side-lying external rotation. B, D2 flexion and extension proprioceptive neuromuscular facilitation patterns.

symptoms are of concern, the barbell bench press can be introduced. To minimize capsular stresses, in almost all cases, a narrower grip is safest. Fees and colleagues recommend a grip distance 1.5 times the biacromial width or less.54 Reducing grip width to less than 1.5 times the biacromial width appears to reduce this risk and does not affect muscle recruitment patterns, resulting in only a ±5% difference in one repetition maximum. If posterior forces need to be minimized, a grip distance should be a symptom-free distance of at least 1.5 times the acromial width. Initial training with the bench press is done with a towel (or half-foam roller) on the chest (Fig. 5C-25). In this manner, the stresses to the shoulder as the bar is lowered are lessoned.

positioning of the scapulae and a strong and healthy rotator cuff that can depress the humeral head effectively, overhead pressing can be included if it is deemed necessary. If the individual insists on performing overhead presses, initiate training with dumbbells in the scapular plane and limit the movement overhead to reduce the likelihood of subacromial impingement.

Behind-the-Neck Training

Overhead pressing can be problematic because the subacromial space is compromised as the weight is lifted overhead. This is particularly true if the rotator cuff is weak or the scapula is anteriorly tilted. For those who have ­normal

The concurrent extreme shoulder external rotation and abduction involved in behind the neck training has been termed the at-risk (or 90/90) position by many practi­tioners. Behind-the-neck pull-downs and shoulder presses place undue stresses on the glenohumeral joint. Potential problems with anterior glenohumeral instability, internal impingement, acromioclavicular joint degeneration, and even the risk for intervertebral disk injuries (owing to the flexed neck position) have been reported. Simply modify the lift by performing these exercises in front of the neck to pose much less risk. In the case of the

Figure 5C-24    Dumbbell floor press.

Figure 5C-25    Bench press modifications. Note half-foam roll and hand spacing to minimize anterior shoulder stresses.

Overhead Press

Rehabilitation and Therapeutic Modalities

pull-down, going behind the neck only shortens the range of motion and reduces activation of the latissimus dorsi.55 Traditional back squats can also be irritating and problematic for individuals with shoulder problems, especially when the bar placement is low on the back. This low bar position increases the amount of shoulder external rotation that is required to hold and stabilize the weight. Modifications for shoulder health may include a high bar placement, performing front squats, or use of a cambered bar that does not require behind-the-neck positioning of the arms. Or, other suitable leg exercises can be substituted for the squat.

Upright Rows A barbell upright row places the shoulder in a maximally internally rotated position as the arm is abducting or flexing to a position at or above shoulder level. This, not dissimilar to the Hawkins-Kennedy impingement maneuver, impinges the anterior rotator cuff and long head of the biceps. Eliminating the exercise is the recommendation of choice; however, if some sort of upright rowing variation is required in the athlete’s program, encourage the use of dumbbells, which allow the individual to adjust the plane of motion and grip to some extent and thus reduce the aforementioned risks.

Anterior Capsule Stresses Other exercises that are of concern are those that horizontally abduct or extend the glenohumeral joint behind the frontal plane of the body. Exercises such as dumbbell flies and triceps dips can overstress the anterior capsule and long head of the biceps, which, in turn, exacerbates the rotator cuff musculature. The advice of “you should be able to see your hands at all times” is good to follow in this context. The individual should limit the range of shoulder motion to a position that is symptom free and preferably in front of the body’s frontal plane.

Putting It All Together—Therapeutic Exercise Program Design for the Shoulder Complex Although expensive gym machines, intricate and detailed routines, and other gimmick devices are available, therapeutic exercises for the shoulder complex do not always need to be elaborate. Programs incorporating body weight, dumbbells, and elastic tubing resistance are very effective and can easily be implemented. The components of a successful rehabilitation program include proper warm-up, strength training, power enhancement, skill training, flexibility enhancement, and cardiovascular conditioning. If the individual fails to train each component, optimal return to function and performance may not be achieved. For the purposes of this chapter section, concepts for proper warm-up activities are introduced and suggestions for optimal implementation of therapeutic exercises are outlined for each phase of the rehabilitation program.

249

Therapeutic Exercise Outlines Acute Phase or Immediate Postoperative Phase In the acute phase, or immediate postoperative rehabilitation phase, our emphasis is to protect healing tissues and to negate the effects of possible immobilization while controlling pain and inflammation. This is accomplished by introducing activities such as pendulums and passive and active-assisted exercises to aid in reestablishing range of motion. Submaximal, pain-free isometric training of the rotator cuff and scapulothoracic muscles are prescribed to help promote scapulothoracic and glenohumeral joint stability. Therapeutic prescription in this phase is often protocol and physician specific. The rehabilitation specialist should obviously be aware of the specific tissues involved and to what extent protective measures are to be followed to avoid overstressing the healing structures.

Subacute or Intermediate Strengthening Phase In the subacute or intermediate phase, the exercise emphasis is to improve strength and neuromuscular control of the glenohumeral and scapulothoracic muscles. This is likely the phase in which an individual who is performing rehabilitation and training for a shoulder injury or syndrome that does not require surgical intervention would begin. This phase contains movements, mostly in isolation, whose purpose is to correct problems such as muscle imbalances or specific weak points. Remember the acronym AIR because this phase typically calls for the activation of inhibited and weak periscapular and rotator cuff muscles. Scapular and rotator cuff isolating exercises with arms below shoulder level in pain-free and safe ranges are first prescribed. These exercises are then progressed, again in pain-free ranges, to advance strengthening of the scapular and rotator cuff muscles.

Advanced or Dynamic Strengthening Phase In the advanced or dynamic strengthening phase, the objective of the exercise protocol is to improve strength, power, and muscular endurance. These parameters need to be reestablished in parallel with the functional activities to which the individual will return. Therapeutic exercises are given to further enhance and integrate targeted rotator cuff, deltoid, and scapular muscles. For the overhead athlete, the clinician can advance the exercises to include more functional sport-specific ranges. The treating rehabilitation specialist must ensure that the individual can sufficiently maintain dynamic stabilization of the shoulder complex before advancing to such activities. Functional eccentric muscle training is also emphasized as needed. Traditional gym strength training exercises are introduced under the supervision of the rehabilitation specialist at this time.

Return-to-Activity Phase The final phase in the rehabilitation program is the returnto-activity phase. The objective of the therapeutic exercise outline during this period is to progressively implement functional demands on the shoulder complex and return the individual to full sport or work activities. Typically, the

250

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 5C-3  Dumbbell Routines Exercise Component

Targeted Muscles

Intermediate Strengthening

Advanced Strengthening

Dumbbell routine (choose one routine)

General rotator cuff and scapula

Standing flexion, scaption with external rotation, abduction Prone horizontal abduction with external rotation at 100 and 90 degrees, prone external rotation at 90 degrees of abduction

Prone stability ball horizontal abduction with external rotation at 100 and 90 degrees Prone external rotation at 90 degrees of abduction

previously prescribed exercises are advanced in intensity through alterations in resistance and volume. The strength, power, endurance, and functional stability must continually be advanced as sport or work activities are introduced. Tables 5C-3 through 5C-7 provide the therapeutic exercises for each phase of rehabilitation. These exercises, mentioned and described earlier, are grouped into a number of categories. These categories are listed and defined below as an outline for a comprehensive therapeutic exercise rehabilitation program for the shoulder. This outline is provided in such a way that the clinician can easily create a thorough, effective, and successful rehabilitation program. 1. Dumbbell routine—choose one routine. These routines group related dumbbell shoulder complex exercises together that activate a large number of the rotator cuff and scapular muscles. 2. Rotator cuff–specific exercises—choose the appropriate exercises. These are designed to isolate and strengthen specific rotator cuff muscles as necessary. 3. Scapular specific: retraction, downward rotation exercises— choose one exercise. These are designed to train the scapular stabilizing muscles that aid in functional pulling movements and act to isometrically hold the scapula when the arm is dynamically moving. 4. Scapular specific: protraction, upward rotation exercises— choose one exercise. These exercises are intended to help the shoulder complex in functional pushing motions and help to maintain proper scapulohumeral rhythm with active elevation of the arm. 5. Dynamic shoulder complex training—choose one or two exercises. These are advanced training exercises for the shoulder complex in functional and sport-specific movement patterns. In Tables 5C-3 through 5C-7, simply select the most appropriate therapeutic exercises from each category, according to the specific needs and goals of the individual, to provide a complete, thorough, and realistic training program to most effectively treat the shoulder complex.

Another important consideration in exercise selection is to avoid exercise redundancy, both within each training session and between different training days. The shoulder complex has a limited capacity to recover from physical stresses, particularly when recovering from injury or surgery. It is therefore wise to include the least number of exercises that are necessary in a single training day to pro­gress and build activity tolerance and conditioning, not overwork and potentially overstress the shoulder. Redundant exercises are those that work the same muscle group using similar movement patterns. If performance of a number of exercises for one muscle group is necessary, then choose the ones that complement—not copy—one another. Also, avoid redundancy when comparing training days to allow for a variety of training stimuli and to help maintain the individual’s interest in the program.

Summary The shoulder is a dynamically mobile joint that relies heavily on the dynamic stability of the scapular and rotator cuff musculature to function safely and effectively. To successfully rehabilitate the shoulder complex, a thorough and complete program is often necessary. Through a detailed clinical assessment, specific weaknesses and dysfunctional movement patterns can be first targeted and corrected. Then, the individual is advanced to a more inclusive program that seeks to train the entire shoulder complex in a progressively more dynamic and functional manner. The program outlined previously will certainly allow the rehabilitation specialist to fully and effectively return the injured individual to an optimal level of ­function.

ELBOW COMPLEX The elbow is an inherently stable joint that receives its stresses typically from movements and stresses at the shoulder and repetitive stresses from wrist and hand tasks. The guidelines that detailed the general progression of rehabilitation programs for the shoulder apply as well to elbow ­disorders. The protocols for specific diagnoses actually vary

   TABLE 5C-4  Therapeutic Exercises to Target Specific Rotator Cuff Muscles Exercise Component

Targeted Muscles

Exercises

Rotator cuff specific (choose one exercise from each as needed)

Supraspinatus, infraspinatus, teres minor, subscapularis

Supraspinatus—side lying Posterior rotator cuff—side-lying external rotation, tubing external rotation, cable column external rotation, side-lying external rotation in scapular plane, tubing external rotation at 90 degrees of abduction Subscapularis—tubing internal rotation, tubing internal rotation diagonal

Rehabilitation and Therapeutic Modalities

251

   TABLE 5C-5  Therapeutic Exercises to Train Scapular Retraction and Downward Rotation Stability Exercise Component

Targeted Muscles

Intermediate Strengthening

Advanced Strengthening

Scapular-specific retraction, downward rotation (choose one exercise)

Lower trapezius, rhomboids, middle trapezius

Scapular retraction with bilateral external rotation, prone horizontal abduction with external rotation at 100 to 130 degrees and 90 degrees Prone extension; tubing, dumbbell, cable column rowing, pull-downs

Prone stability ball horizontal abduction with external rotation at 100 to 130 degrees and 90 degrees Prone extension Single-arm rowing, rowing with trunk rotation

little. When designing a rehabilitation program, the rehabilitation ­specialist must simply keep in mind what tissues or structures need to be protected and what motions or activities need to be avoided or modified. Armed with that information, the treating clinician can confidently advance anyone through the program. As the rehabilitation program advances through the AIR stages of rehabilitation ­(activate, integrate, reinforce), efforts are directed at achieving dynamic stability through therapeutic exercises that enhance strength, muscular endurance, and neuromuscular control.

Total Arm Strengthening All of the muscles of the elbow joint complex play vital roles in providing stability and enhancing function. Furthermore, the rehabilitation program would be deficient if aggressive strengthening techniques of the entire shoulder complex were not advocated. Too many times an individual or athlete with elbow symptoms is found to have underlying shoulder complex disorders that inherently lead to compensations and overstresses at the elbow. To ensure that the rehabilitation and training of the elbow is thorough and complete, an outline detailing the different components of the elbow complex therapeutic exercise categories is given next. Within each of the programs, specific exercises are given. 1. Elbow flexor training—choose one or two exercises. These exercises focus on elbow flexion strength deficits and movement complications. Particular attention is given to symptomatic combinations of elbow flexion and specific forearm rotations. 2. Elbow extensor training—choose one or two exercises. Like elbow flexor training, the elbow extensors are strengthened in accordance to weakness and movement difficulties.    TABLE 5C-6  Therapeutic Exercises to Train Scapular Protraction and Upward Rotation Stability Exercise Component

Targeted Muscles

Intermediate Strengthening

Advanced Strengthening

Scapular-specific Serratus Push-up (+), Stability ball and protraction, anterior press-up, unstable surface upward protraction, push-ups, closed rotation (choose dynamic hug, chain upper one exercise) D1 proprioceptive extremity step neuromuscular variations facilitation Bench-press progression

3. Forearm supination and pronation training—choose one exercise. These exercises train the forearm supinators and pronators in a variety of ways. 4. Wrist flexion and extension training—choose one exercise. Wrist flexion and extension training variations allow for specific needs of the individual. 5. Wrist radial and ulnar deviation training—choose one exercise. Likewise, these exercises are performed to strengthen the forearm as necessary. 6. Shoulder complex training. The total arm strengthening concept is a must to best prepare the individual for return-to-function activities. This is especially vital after surgical procedures to the elbow when the entire upper extremity has been inactive. 7. Rhythmic stabilization/perturbation training—choose one or two exercises. These manual PNF techniques help to enhance muscular co-contractions and train stability of the elbow complex.

Elbow Flexor Training The biceps brachii (biceps), brachialis, and brachioradialis musculature is important in almost all daily lifting and pulling tasks and is an important elbow stabilizer and decelerator in the throwing athlete. During the throwing motion, the biceps contract during the deceleration phase to prevent hyperextension forces to the elbow and pronation forces of the forearm after the ball has been released.56,57 EMG analysis of these elbow flexors reveals that their contributions to elbow movements are dependent on forearm positions. Investigators have concluded that the biceps is most active with resisted elbow flexion training with the forearm in slight supination.58-61 The brachialis is noted to be active with resisted elbow flexion training when the forearm is supinated, in neutral, or pronated, seemingly making it a primary elbow flexor. Similarly, the

   TABLE 5C-7  Shoulder Complex Training to Enhance Dynamic Strength, Stability, Power, and Neuromuscular Control Exercise Component

Targeted Muscles

Exercises

Dynamic shoulder complex training (choose one or two exercises)

General rotator cuff and scapular

D1/D2 proprioceptive neuromuscular facilitation, external rotation of 90 degrees abduction, golf diagonals, wall dribbles, Body Blade

252

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5C-26    Arm curl with a pronated grip.

­ rachioradialis contributes more to resisted elbow ­flexion b when the forearm is pronated or in a neutral position. Although it is active, it contributes less when the forearm is supinated.58 Schoenfeld demonstrated that the position of the shoulder also plays a role in the contribution of the biceps during resisted elbow flexion.62 He found that the long head of the biceps was more active as an elbow flexor with the shoulder in a hyperextended position. He noted that this allows the long head to be placed to exert more force and carry out a greater range of motion and workload during elbow flexion. The short head has a more dominant role by placing the shoulder into increasing degrees of flexion. Clinically, we apply these anatomic and EMG principles to the evaluated weaknesses of the individual. The biceps brachii is trained with arm curls with a comfortable supinated grip, whereas the brachioradialis is exercised with arm curls in a neutral or comfortable pronated grip (Fig. 5C-26).

A

Because the brachialis is active with all arm curls, the assumption is that it does not need to be isolated like the other elbow flexors. The elbow flexors can be sufficiently trained with the use of dumbbells, elastic resistance, or a long cable column. If it is deemed necessary to isolate the head of the biceps, arm curls with the shoulder extended while sitting on an incline bench can be performed. For a more direct isolation of the short head, train arm curls with the elbow by the side or with the upper arms supported on a bench or stability ball (Fig. 5C-27). Bankoff and coworkers, however, documented through EMG the activities of the brachialis, biceps brachii long portion, biceps brachii short portion, and brachioradialis during resisted elbow flexion in supination and pronation positions as well.63 Their results showed similar muscle activations in both supination and pronation varieties. In another interesting study, Naito and investigators observed increased biceps brachii activity with increased forearm supination, like the previously mentioned research.59,60 However, the patterns of changes varied from individual to individual. They concluded that each subject has an ­individual pattern of use of the biceps brachii for supination during elbow flexion movements. In light of these findings, and because rarely is it found that only one of the elbow flexors is deficient in strength, we train the elbow flexion movement that is weak or symptomatic. In this case, we activate the muscles responsible for the movement, not a specific muscle, in hope that it will remedy a movement problem. The appropriate elbow flexor training that we first employ is determined by evaluation of the following motions: • Elbow flexion beginning in a pronated position and progressively supinating during the full range of motion (Fig. 5C-28). This same motion can be performed with the shoulder in a more extended position with the arm off the table or flexed with the arm on a bolster or foam roll. • Elbow flexion beginning in a supinated position and progressively pronating during the full range of motion. • Elbow flexion with the forearm supinated.

B

Figure 5C-27    Arm curls to target. A, Long head of biceps. B, Short head of biceps.

Rehabilitation and Therapeutic Modalities

A

253

B

Figure 5C-28    Evaluation of elbow flexion strength and symptoms. A, With forearm supination. B, With forearm pronation.

• Elbow flexion with the forearm pronated. • Elbow flexion with the forearm in neutral. The specific type of elbow flexion training that is prescribed is based on the specific movement found to be problematic. If no type of motion is noticeably different from the others, vary the exercises throughout the rehabilitation program to allow more complete elbow flexor training.

Elbow Extensor Training The triceps is an important muscle that provides support in most pushing activities and in activities that require the upper extremity to bear weight. For the athletic population, training of the triceps should be included to maximize strength and performance enhancement.57 Altcheck and Andrews reported that triceps are recruited sequentially.64 The medial head is activated first and the lateral head second. Interestingly, the long head is finally activated third in an as-needed manner. Like the biceps, the activation of the heads of the triceps is also ­ dependent

A

on the position of the arm. The lateral and medial heads are more active as an elbow extensor with the elbow at the side. The long head is more isolated as an elbow ­extensor with the shoulder flexed.59 Clinically, triceps push-downs target the lateral and medial heads. The long head is recruited with elbow extension with the shoulder in a flexed position (Fig. 5C-29). In the same manner that the elbow flexors are evaluated, elbow extension is assessed throughout the range of motion in the following movement patterns: • Elbow extension beginning in a pronated position and progressively supinating during the full range of motion. This same motion can be performed with the shoulder in a more extended flexed position with the arm at 90 degrees of elevation. • Elbow extension beginning in a supinated position and progressively pronating during the full range of motion. • Elbow extension with the forearm supinated. • Elbow extension with the forearm pronated. • Elbow extension with the forearm in neutral.

B

Figure 5C-29    Triceps extension to target. A, Lateral and medial heads of the triceps. B, Long head of the triceps.

254

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

C

B

A

D

Figure 5C-30    Forearm training. A, Wrist flexion. B, Supination and pronation. C, Radial deviation. D, Ulnar deviation.

Elbow extension should be trained to replicate and remedy the problematic movement combinations. As symptoms resolve, train elbow extension in a variety of manners to ensure that a thorough approach is being applied.

Forearm Muscle Training The forearm muscles also play vital roles in the various functions of the elbow in the throwing athlete. The forearm pronator–wrist flexor muscle group has been cited for its importance in providing medial stabilization against the valgus forces during throwing, especially during the acceleration phase when they are activated at extremely high levels.56,57 Davidson and associates provided evidence that the flexor digitorum superficialis and flexor carpi ulnaris are best suited to provide a stabilizing effect for the ulnar collateral ligament because these muscles overlay the anterior band of the ulnar collateral ­ligament.65 Therefore, wrist flexor and forearm pronation training are key in the strengthening and recovery of the throwing athlete’s elbow. Strengthening of the forearm supinators and wrist extensors are also of key importance in the rehabilitation and conditioning of the throwing athlete because they demonstrate moderately high activation during the cocking phase of throwing.57 Forearm supination, pronation, and radial and ulnar deviation can all be strengthened effectively with dumbbells and elastic tubing (Fig. 5C-30).

A

B

O’Sullivan and Gallwey determined that the angle of elbow flexion affects supination torque more than pronation torque.66 They conclude that supination torque is stronger for the midrange of elbow flexion but that ­pronation torque increases with increasing elbow extension. With this in mind, supination training may be best carried out with the elbow in midrange flexion, whereas pronator training can be prescribed with the elbow in an extended position. Special emphasis should be placed on the eccentric components of the movements to help enhance the musculotendinous portions of the forearm muscles. This is typically the area of concern with the various tendinopathies that appear to plague the elbow complex. Indeed, eccentric training of the wrist extensors has been documented to aid in the recovery of subjects with lateral epicondylitis.67 Special importance is placed on medial elbow strengthening because these muscles act to resist valgus tensile forces and actually overlay the anterior band of the ulnar collateral ligament. The athlete should train the flexorpronator muscles in both an eccentric and dynamic manner to stabilize against the inevitable valgus loads encountered in sports. Various wrist rollers can be used to train the wrist flexors and extensors, supinators and pronators, and radial and ulnar deviators (Fig. 5C-31). We have also found that all the wrist and forearm movements can be trained somewhat isokinetically in a bucket of beans or rice. The individual

C

Figure 5C-31    Wrist roller training. A, Wrist flexion and extension. B, Radial and ulnar deviation. C, Supination and pronation.

Rehabilitation and Therapeutic Modalities

255

Summary The elbow joint complex is often exposed to tremendous forces in the overhead athlete and to repetitive stresses in many working environments. Luckily, most individuals with such pathologies benefit from conservative treatment consisting of appropriate therapeutic exercises.

KNEE

Figure 5C-32    Bean bucket wrist and forearm training.

simply penetrates his hand into the beans and carries out the movements. The resistance is continuous and fairly aggressive (Fig. 5C-32).

Rhythmic Stabilization and Perturbation Training Juul-Kristensen and colleagues documented decreased joint position sense and detection of a passive movement in elbows with lateral epicondylitis.68 Proprioception, therefore, appears to be poorer in elbows with lateral epicondylitis. This needs to be taken into consideration in the management of lateral epicondylitis. We have found that PNF techniques are beneficial for training the elbow in the overhead athlete. The clinician can perform rhythmic stabilization drills in various arm positions to enhance more functional and complete muscular contractions to protect the joints. This is particularly effective when training the medial elbow stabilizers after injury to the ulnar collateral ligament or flexor-pronator muscle units (Fig. 5C-33).

The knee joint complex is relatively simple and straightforward—knees need stability. They are hinge joints with minimal rotary and transverse components. They must stabilize against and endure tremendous forces from the ground below and from movements of the entire body above. The knee and its musculature must provide for deceleration, acceleration, and stabilization forces repeatedly between these two components. When these forces become excessive, ligamentous, meniscal, or patellar injuries often occur. Before discussing the specific applications of therapeutic exercise, principles related to the different phases of rehabilitation of knee disorders are first outlined. These principles act as important guidelines for the development and implementation of appropriate therapeutic exercise prescription throughout the rehabilitation process.

Acute Phase or Immediate Postoperative Phase The primary goal of the immediate postoperative phase is to protect healing tissues while working to minimize the negative effects of immobilization and decreased activity levels. Further goals are to decrease pain and inflammation, increase range of motion and flexibility as necessary, increase weight-bearing tolerance, and increase strength, balance, and neuromuscular control of the involved lower extremity. The therapeutic exercises prescribed in this phase are designed to activate and reeducate the muscles that have been found to be inhibited and weak.

Subacute or Intermediate Strengthening Phase The major goals of this phase are to increase strength, endurance, and functional activities while protecting the healing structures. Single-leg balance and proprioceptive exercises are initiated, and flexibility of the entire lower extremity is addressed. The exercises previously performed in the first phase are continued and progressed as tolerated. More progressive closed kinetic chain exercises are gradually initiated, and resistance is advanced. It is stressed that exercises must be pain free and performed in stable ranges of knee motion.

Advanced or Dynamic Strengthening Phase

Figure 5C-33    Rhythmic stabilization training to train the medial elbow stabilizers during the throwing motion.

Phase three, the advanced strengthening phase, is directed toward progression of activities to prepare for full return to work or sporting events. The primary objectives of this phase are to increase the individual’s strength, power, and endurance. Total lower extremity strength training is continued and progressed during this phase. More advanced loading patterns and schemes are introduced to facilitate continued strength gains.

256

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Return-to-Activity Phase The goals of this final phase are to progressively return the individual to unrestricted activities and to establish a proper strength and conditioning program that helps to ensure that the individual will be able not only to tolerate the rigors of his or her activities but also to optimize and ultimately enhance performance levels. As preinjury activity levels are approached, the formal rehabilitation is decreased to avoid overtraining and overuse complications. The repetitions for the strength training exercises can be lowered, with the athlete performing the repetitions at increased speeds to facilitate power development for sport activity. Difficult or strenuous aspects of the patient’s sport or job should be isolated and trained for separately.

Putting It All Together—Therapeutic Exercise Program Design for the Knee Remember, one of the basic fundamentals of therapeutic exercise application is to follow the AIR principle. Activate the weak and inhibited muscles first, then integrate these muscles into more functional movements of the lower kinetic chain, and then reinforce the proper motor pattern in progressively more functional and specific skills. If the therapeutic exercises are advanced inappropriately, overcompensations of the stronger and more dominant muscle groups will lead to faulty and inefficient movement patterns. This not only fails to allow for optimal performance but also will likely lead to continued overloading and possible injuries. The following outline details an all-encompassing list of therapeutic exercises that help to ensure that appropriate training is performed to most effectively rehabilitate an individual following an injury to the knee complex. 1. Neuromuscular activation exercises—choose as needed. These include exercises used in the early stages of the program to best isolate the key musculature of the lower extremities as they relate to the rehabilitation of knee injuries. 2. Quadriceps-dominant squatting exercises—choose one or two. These exercises include closed chain squatting movements that are functional and allow for a progression to heavy resistances, allowing important strength gains to be established. 3. Single-leg training—choose one or two exercises. Training on one leg allows the individual to train functionally and emphasize stabilizing musculature to a greater extent. 4. Hamstrings- and gluteal-dominant bent leg exercises—choose one exercise, alternating with category D on alternate training days. 5. Hamstrings- and gluteal-dominant straight leg exercises— choose one exercise, alternating with category C on alternate training days. 6. Proprioceptive and neuromuscular control exercises—choose one exercise. These exercises train and further enhance the individual’s stability in progressively more challenging and unstable environments. Special attention is given to the ability to resist hip internal rotation­adduction forces and knee valgus forces.

The exercises are outlined in accordance with their level of difficulty and overall stress to the knee and lower extremity musculature. These progressions are initiated from a stable system that does not stress stabilizers or neutralizers to so-called functional exercises that are designed ­specifically to stress the stabilizing and neutralizing muscles. Finally, we discuss briefly how the categories of these exercises should be modified and how they can best be included (or excluded) in rehabilitation programs for the disorders of the anterior cruciate ligament (ACL), posterior cruciate ligament (PCL), and patellofemoral (PF) joint.

Neuromuscular Activation Exercises One of the first clinical considerations when considering therapeutic exercise prescription is the evaluation of the strength and ability to activate key muscle groups of the lower extremity. In most cases, we have found the quadriceps and gluteal muscles to be negatively affected or deficient in cases of knee pathology. Specific exercises used to activate these muscles are discussed and outlined.

Quadriceps It is the accepted norm that the quadriceps muscle group, especially the vastus medialis obliquus (VMO), is inhibited and poorly functioning from the pain and effusion that accompany a knee injury. The quadriceps is obviously a key contributor to extension of the knee and to the stabilization of the patella centrally within the trochlea. Earlystage neuromuscular activation exercises for the quadriceps include isometric quadriceps setting and straight leg raises (Fig. 5C-34). These exercises are particularly relevant in the immediate postoperative stage and after acute injury to the knee when the quadriceps muscles are inhibited and weak and provide minimal support to the knee. Although straight leg raises seem benign in their applications, the rehabilitation specialist must ensure that these exercises are performed correctly and applied appropriately to avoid detrimental effects on healing ligaments and patellar stabilizers. If hip flexion straight leg raises are performed with the knee bent, a sign that the quadriceps is unable to achieve or maintain full extension, excessive stresses may be placed on a healing ACL graft. For a healing medial collateral ligament, hip adduction straight leg raises are avoided because of their stresses on the medial knee. Likewise, hip abduction straight leg raises are not included when lateral structures need to be protected. Early in the rehabilitation process, initial attempts to contract the quadriceps may be incomplete, resulting in inadequate performance of quadriceps setting and straight leg raise exercises. To maximize the patient’s ability to contract the quadriceps and enhance potential force production, electrical stimulation (ES) and EMG biofeedback can be used. ES artificially activates the motor nerve and helps to recruit a greater number of contracting fibers and increase maximal contractile forces.69,70 Biofeedback through surface electrodes gives visual or audible feedback to the user and aids in facilitation of increased levels of muscle activity.71 Indeed, researchers have concluded that both ES72-74 and biofeedback71,75 are more effective in

Rehabilitation and Therapeutic Modalities

A

B

C

D

257

Figure 5C-34    Straight leg raises. A, Hip flexion. B, Hip abduction. C, Hip adduction. D, Hip extension.

regaining quadriceps muscle force after knee surgery than are voluntary isometric contractions alone. Other early exercises that may be implemented are short arc terminal knee extensions (short arc quads) and fullrange knee extensions (long arc quads). The effectiveness of these exercises lies in their ability to isolate the quadriceps and train their activation.76-79 Of these exercises, Anderson and investigators demonstrated higher EMG amplitudes for the knee extension exercises than quadriceps setting.76 Likewise, they found higher EMG amplitudes with the knee extension exercise than the leg press or squat. However, these open chain knee extension exercises may produce excessive PF compressive forces on small patellar and trochlear contact areas.80 Each patient must be monitored to determine whether these exercises are appropriate. We typically use short arc and long arc quads only in the early stages of the rehabilitation program, with minimal resistance applied and when surgical inspection has revealed articular damage, causing weight-bearing progression to be slowed. In consideration of stresses and strains to healing cruciate ligaments, Lieb and Perry demonstrated that a 60% increase in quadriceps muscle force is needed to extend the knee the last 15 degrees of extension in an open chain environment.81 Unfortunately, the final 60 degrees of motion is also the range in which maximal strain is placed on the ACL.16,79,82-85 Furthermore, knee extension exercises performed in ranges of 110 to 60 degrees place dramatically higher loading on the PCL.16,79,82,85 It is assumed that these loads are deleterious in the early phases of rehabilitation, when the strength of a reconstructed graft is a concern. Because of the VMO’s role in stabilizing the patella against lateral tracking, and because it is more easily inhibited after injury or surgery, special attention has been placed on the effects of exercise on the VMO. Several researchers have recommended the performance of specific quadriceps

exercises for their suggested preferential activation of the VMO. Some investigators have reported that simultaneous contraction of the quadriceps and hip adductors would increase VMO activity.86-89 Others have stated that activity of the VMO is enhanced with straight leg raises if the hip is maintained in external rotation.86,89 However, recent studies using normalized EMG values failed to demonstrate preferential VMO activation with these and other similar exercises.90-92 In light of these studies, we have found that emphasis placed on activating the VMO is best accomplished by introducing an effective generalized quadriceps strengthening program and that attempts to isolate this region are not always necessary.

Gluteal Musculature Recent research has suggested the importance of gluteal muscle strengthening in the rehabilitation of knee dysfunction.93-99 Jaramillo and colleagues’ examination of patients who had undergone unilateral knee surgery revealed significant hip weakness in all four hip muscle groups of the surgical extremities relative to the nonsurgical extremities.95 In separate but related studies, Cichanowski and colleagues and Ireland and coworkers each found that women with PF pain syndrome were significantly weaker in hip abduction and hip external rotation when compared to controls.100,101 In addition to weaker hip abduction and external rotation values, Rowe and associates found hip extensor weakness in females seeking physical therapy treatment for unilateral PF pain syndrome when compared with the weaker limbs of control subjects.102 Likewise, it has shown that females are weaker (normalized to body weight) in hip strength than their male counterparts.103,104 These weaknesses may cause increases in hip internal rotation and valgus force vectors at the knee, a combination that may further increase lateral PF forces. Furthermore, this valgus

258

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5C-35    Hip abduction with slight hip and knee flexion of 20 degrees. This greater activates the gluteal muscles on the stance leg.

position of the lower extremity is viewed as hazardous for the ACL.83,105-109 ­Leetun and colleagues actually demonstrated that hip external rotation weakness was a significant predictor of future knee injury in collegiate basketball and track athletes.104 Early-phase gluteal exercises are targeted at hip abduction, hip extension, and external rotation isolation. Through EMG analysis of common hip abduction exercises, Bolgla and Uhl demonstrated that side-lying hip abduction best activated the gluteus medius muscle.110 The gluteus medius muscle was next most activated in the weight-­bearing leg when hip abduction was performed on the contralateral side with slight hip and knee flexion of 20 degrees (Fig. 5C-35). Their results also indicated that standing hip abduction exercises with the non–weightbearing leg required the least amount of muscle activity. Therefore, weaker patients or patients that have weightbearing restrictions may be able to use these exercises early in the rehabilitation process. These initial hip-strengthening exercises may be performed with cuff weights, with Thera-Band, or on machines. As patients become proficient with the non–weight-bearing standing hip abduction exercises, they may progress to non–weight-bearing side-lying hip abduction, particularly if there are weight­bearing precautions. If patients can apply full weight, they can obtain a similar amount of resistance to the hip abductors by standing on the involved lower extremity and abducting the contralateral limb. Hip extension training can be included early in the rehabilitation phases through bridging variations. Bridges can be performed bilaterally or unilaterally and can be combined with hip abduction training (Fig. 5C-36). The focus of teaching hip extensions should be on emphasizing extension from the hip, versus substituting with hyperextension of the lumbar spine. Hip external rotation training is introduced with exercises such as side-lying hip abduction and external rotation (the “clam”) and hip external rotation in sitting and in prone. Thera-Band, cable columns, and machines can each be used to allow for the appropriate resistances (Fig. 5C-37).

Figure 5C-36    Bridge with hip abduction to help ensure proper gluteal training.

Another method used to activate the proximal stabilizing muscles is band walking. Once the individual has demonstrated a more normal gait pattern without pain or swelling, band walking is an effective way to teach how to engage the gluteal muscles in a functional movement ­pattern. Minibands around the lower legs apply the appropriate resistance while the individual performs various walking patterns. We use forward and backward walking with the knees relatively extended (“toy soldier march”), medial and lateral side steps while maintaining an “athletic position,” and forward and backward walking with the feet spread apart wide (“gunslinger” or “monster walk”) (Fig. 5C-38). The miniband can also be applied above the knee to reinforce hip external rotation contractions. A longer band can be use around the feet to target hip abduction as well (Fig. 5C-39). Early in the rehabilitation process, band walking is performed as a neuromuscular activation exercise, whereas in the later stages, we have used them as an active warm-up.

Quadriceps-Dominant Squatting Exercises The development of lower-body squatting strength should be a cornerstone of a well-developed rehabilitation and training program for individuals after knee injury. The continuum of knee dominant squatting exercises focuses first on squatting movements that are stable and allow high muscle

Figure 5C-37    The “clam” combines hip abduction and external rotation.

Rehabilitation and Therapeutic Modalities

A

B

259

C

Figure 5C-38    Band walking. A, Forward and backward walking. B, Medial and lateral walking. C, “Monster” walks.

activation from the quadriceps. Lying leg press machines and variable incline machines (Total Gym, Engineering Fitness International, San Diego) are typically introduced first. As with most machine exercises, the individual acts as force producer while the stability is provided by the machine. The next exercise on the continuum is the standing wall squat. The standing position is obviously more functional than lying and may incorporate additional muscles not stressed in the leg press or Total Gym. However, stability is still provided by the wall in the wall squat exercise. The third and final step up the functional continuum in this category is to teach and train standing squatting.

Figure 5C-39    Placing the miniband above the knees emphasizes hip external rotation, whereas the longer band targets hip abduction.

Leg Press and Total Gym The leg press and Total Gym allow individuals to train in a squatting movement with their body weight taken out of the equation and allow the quadriceps to be targeted without overloading the knee joint structures. Wilk and colleagues recorded the EMG activity of the quadriceps and hamstring muscles for the leg press exercise and determined that the maximal quadriceps activity was between 88 and 102 degrees of knee extension during the ascent phase of the lift.79 They further determined that minimal hamstring activity was present in the leg press exercise, making it a relative quadriceps-dominant exercise. Escamilla and colleagues investigated how variations in foot placement on the leg press would affect muscle activity of the quadriceps and hamstrings.111,112 In their study, 10 experienced male lifters performed a high foot placement leg press and a low foot placement leg press employing a wide stance, narrow stance, and two foot angle positions (feet straight and feet turned out 30 degrees). No differences were found in muscle activity between foot angle and foot placement variations. Similarly, Tassi and associates showed similar quadriceps activity in leg press variations of high and low foot placements.113 However, Escamilla and researchers did demonstrate greater hamstring activity in the wide stance–high foot placement variation than the narrow stance exercise.111 No ACL, PF, or tibiofemoral joint forces were deemed hazardous in the leg press.111 In comparison to squatting, tibiofemoral compressive forces, PCL tensile forces, and PF compressive forces were all less in the leg press variations. However, posterior shear forces were documented in the leg press, particularly in the deeper ranges of knee flexion.9,79 We have found these exercises to be effective during rehabilitation after an ACL reconstruction, and they can be used following PCL reconstruction with knee flexion depths less than 70 degrees. Early in the rehabilitation ­process, leg

260

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 5C-40    A, Leg press. B, Total Gym.

pressing allows the individual to perform a squatting-type maneuver safely. As the individual progresses in the training, the leg press becomes a way in which progressively more resistance can easily be administered, allowing for the appropriate amount of intensity to be applied. Also, single-leg training can easily be introduced in the more stable environment given by the machine. In light of these findings, we encourage the individual to perform these exercises with a foot position that is most comfortable and to train in ranges that are nonoffending to the PF joint or healing PCL (Fig. 5C-40).

Wall Squats The wall squat provides an effective transition from machine squatting to a squatting activity in an upright position with the feet on the ground (Fig. 5C-41). When the squat is performed while leaning against a wall, the individual is allowed to position the feet in front of the body’s base of support. This allows for increased muscular activations and decreased joint stresses on the knee, especially at the PF joint. Blanpied demonstrated the muscular advantages of the wall squat as he documented gluteal, quadriceps, hamstring, and plantar flexor muscle activations during different variations of this exercise.114 The activations were

Figure 5C-41    Wall squat.

varied in support location and in foot position. The feet were placed in line with the hip and with the feet forward. The results indicated that placing the foot forward, versus in line with the hip, caused an increase in quadriceps, gluteal, and hamstring muscle activations. The wall squat exercise can be advanced in intensity through progression in repetitions, resistance, and depth of the squat. It is assumed that the effects and ultimate precautions to protect the ligaments and joint surfaces for the wall squat are the same as other closed chain exercises. Limiting painful ranges of motion to protect the PF joint and limiting the depth of the squat when considering PCL stresses are typically the main concerns.

Squats The final and ultimate step up the functional continuum of knee-dominant squatting exercises is teaching and training standing squatting variations. The squat is typically regarded as the “king of lower body exercises” by strength coaches and trainers alike. In the rehabilitation setting, squatting provides a functional activity that requires contributions from most lower body, hip, and core muscle groups. Squatting also reveals important information regarding an individual’s overall strength, kinetic chain flexibility, and injury potential. And obviously, squatting is a movement that is encountered in many different forms each and every day. Whether or not the rehabilitation specialist chooses to pursue heavy back squatting as a therapeutic exercise is a personal choice in coaching ability and in consideration of the individual’s goals and needs. Squatting, although seemingly a simple task, is at times technically difficult to perform for an individual in a rehabilitation and training program. Proper squatting minimizes knee, hip, and low back stresses and maximizes its tremendous benefits. The functional body weight squat is first introduced, followed by dumbbell squats and possibly hex bar squats. Depending on the individual’s goals and needs, front or traditional back squatting is finally introduced with a Smith machine or barbell. The functional body weight squat emphasizes proper use of a “hip hinge,” incorporation of the powerful gluteal musculature, and maintenance of proper lower extremity alignment. A dowel rod may be used to reinforce maintenance of proper spine angles because the rod should

Rehabilitation and Therapeutic Modalities

A

B

261

C

Figure 5C-42    Functional squat training. A, Proper form. B, Excessive anterior migration of the knees. C, Poor lower extremity alignment control. Note the resultant valgus stresses to the knee.

remain in contact with the head, thoracic spine, and sacrum throughout the squatting motion (Fig. 5C-42A). Without going into great detail, the following cues and tips should allow the rehabilitation specialist to effectively teach proper squatting. • When initiating the movement, have the individual concentrate on sitting back and placing the body weight on the back half of the foot. The individual should still be able to wiggle the toes. Essentially, try to find a happy medium between sitting back and sitting down. A common squatting error is when the individual bends only at the knees and thus relies on the calf muscles for support and places undue stress on the anterior knee (see Fig. 5C-42B). • The knees are going to come in front of the toes simply because this is the only way to get proper squatting depth when the trunk is more upright and the lumbar spine is protected. • Throughout the descent, and ascent, the knees should remain in line with the feet. Do not allow the knees to dip inward. Remember, this hip adduction–internal rotation and knee valgus position is often a likely contributor to knee stresses and injuries (see Fig. 5C-42C). • As the individual begins to ascend from the bottom position of the squat, have him or her concentrate on driving upward with the chest and extending the hips, driving them forward. Often the question arises of whether the knees coming in front of the toes during the squat is dangerous. Fry and colleagues examined joint kinetics during back squats in two conditions.115 In the first condition, a board placed in front of the participants’ shins restricted the forward displacement of the knees. In the second condition, movement was not restricted at all; the participants squatted normally, and the knees slightly passed the toes. The researchers found that restricting the forward excursion of the knees during the squat increased anterior lean of the trunk and promoted

an increased “internal angle at the knees and ankles.” The results were a 22% decrease in knee torque (particularly in the PF joint) and a 1070% increase in hip torque! Yes, there was decreased stress at the knee, but those forces were transferred more than 10-fold to the hips and lower back. Therefore, appropriate joint loading during squatting would allow the knees to move slightly past the toes. Once the individual has developed proper squatting technique, he or she can be advanced to dumbbell (“goblet”) squats in which the weight is held in front of the body. The dumbbell effectively counterbalances the proper sitting-back technique and allows the individual to gain confidence in his or her ability to squat (Fig. 5C-43). Hex bar (diamond bar, trap bar) squatting, front squatting, and Smith Press squatting can all be integrated to further advance general lower body strength effectively. The hex bar squat allows for the weight to remain close to the ­lifter’s

Figure 5C-43    “Goblet” squat.

262

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5C-44    Hex bar squat.

base of support and center of gravity without overloading the spine (Fig. 5C-44). Front squatting also requires perfect body position, reinforcing proper trunk and spine angles. Simply stated, if an erect and stable trunk is not maintained in the front squat, the weight will fall forward, and the lift will not be completed. Individuals performing the front squat can use a clean grip or a crossed-arm grip (Fig. 5C-45). Finally, back squatting can be performed on a Smith Press machine or with a barbell (Fig. 5C-46). Remember, squatting is a safe movement when done properly. However, of all the lifts, back squatting requires especially close supervision and feedback for optimal performance and safety. Numerous authors have studied the effects various exercise variables have on muscle activity for the squat. These variables include magnitude of knee flexion, foot angle, amount of forward trunk lean, and bar placement. Several investigators have quantified increased quadriceps activity

Figure 5C-45    Front squat.

with depth of the squat, with the most activity generally seen near 90 degrees of knee flexion.16,79,92,116-120 Peak values for hamstring activity occur between 60 and 10 degrees of knee flexion during the squat ascent.16,79,116,117,119 Hence, it can be concluded that squats promote a co-contraction of the muscles around the knee. When the lifter squats with a more forward trunk lean, the larger hip flexion moment requires an increased amount of hamstring activity to perform the lift.79,118 Information gathered from studies on the effects of varying foot angles during the squat revealed that quadriceps and hamstring muscle activities are not significantly altered when the feet are turned in, turned out, or kept parallel.59,111,121,122 When bar placement is considered, greater quadriceps and hamstring activity was seen in the low-bar squat (3 to 5 cm below the level of the acromion) than the high-bar squat (about level with the acromion).120 Several investigators have further studied squat mechanics and have noted how variables in squat performance affect shear forces in the knee joint. Along with their study on the leg press, Escamilla and colleagues studied the effects of stance width (narrow, wide, normal) and foot angle (parallel, turned out 45 degrees) on tibiofemoral shear and compressive forces during the squat.111,112,123 No significant differences in shear forces were found with these varied parameters. They did document higher compressive forces with the wider stance, which may be beneficial in limiting shear forces. Investigations have determined posterior shear forces and resultant PCL loading throughout the squat.79,85,112,118,119 Several authors have concluded that as the flexion angle increases, both posterior shear and compressive forces also increase.79,118,119,124-126 Even though these shear forces do not approach the ultimate tensile strengths of the PCL, we again recommend limiting the flexion angle to less than 70 degrees to minimize stresses after PCL reconstruction or PCL injury. Proper squatting depths are also a concern when PF stresses are considered. Salem and Powers looked at PF joint kinetics in female collegiate athletes at three different depths: 70 degrees (above parallel), 90 degrees (at parallel), and 110 degrees (below parallel) of knee flexion.127 The researchers found that peak knee extensor moments, PF joint reaction forces, and PF joint stresses did not vary significantly between the three squatting trials. They concluded that there was no support for the idea that squatting below parallel increases stress on the PF joint. Understanding the forces that act on the PF joint throughout ranges of motion in both closed and open chain exercises allows the clinician to develop a program based on empirical evidence. Steinkamp and coworkers demonstrated mathematically that maximal PF joint reaction (PFJR) forces during the leg press occur when contact between the PF surfaces is greatest (60 and 90 degrees), whereas the maximal PFJR forces during a leg extension occur when the PF contact is the least (0 and 30 degrees).80 Therefore, patients with patellar articular degeneration and arthritic changes often experience pain during open chain knee extensions from 30 to 0 degrees as a result of the relatively large compressive forces being applied to minimal contact areas. Steinkamp and colleagues concluded that patients with PF joint arthritis may tolerate leg press

Rehabilitation and Therapeutic Modalities

A

263

B

Figure 5C-46    Squat. A, Smith press. B, barbell.

exercises (or similar closed chain squatting exercises) better than leg extension exercises in the final 30 degrees of knee extension because of the lower PF joint stresses.80 It is also important to note that the squatting depth in any closed chain exercise should be determined by the individual’s ability to maintain proper form at increasing depths. And of course, painful ranges of squatting should be avoided. In 1997, Beynnon and colleagues published an article that described ACL strain during weight-bearing exercises.128 Before this publication, it was the general consensus that because of the compression forces present during weight-bearing exercises, the amount of anterior tibial translation and calculated shear forces was reduced.83,129-131 Beynnon’s strain gauge study showed similar strain in the ACL during closed and open chain exercises such as the squat (with body weight) and active knee extension. It is important to note that strain did not increase during the squat when resistance was added. However, strain did increase during the active knee extension as resistance was increased.128 Thus, increasing external loads to the squat exercises could be considered safe when strengthening after ACL reconstruction surgery. Conversely, adding resistance to open chain knee extensions may cause excessive strain on a healing ACL graft. Although the squat exercise is considered essential by most strength and conditioning professionals and is a primary leg exercise for powerlifters and weightlifters, controversy exists regarding its long-term effects on knee stability. Panariello and associates132 studied the effect of squat exercises on knee joint stability of professional football players, whereas Chandler and coworkers133 investigated the effects of squatting in powerlifters and weightlifters. Overall, no effect of squat training on knee stability was demonstrated in any of the groups tested. Actually, the powerlifters and weightlifters demonstrated significantly less ligamentous laxity than the controls.

Single-Leg Strength Training The next step in the progression and in the thought process should be to work on one leg. From a functional and anatomic standpoint, it is absolutely critical to include ­single-leg

training. In the rehabilitation setting, the individual must isolate and train the specific weaknesses seen in the recovering leg. Single-leg strength simply cannot be developed through double-leg exercises. When performing double-leg exercises, the stronger leg always controls the movement. Furthermore, the muscles that support the lower leg in single-leg stance (quadratus, gluteus medius, and adductors) are not nearly as active in double-leg exercise. Finally, single-leg training seeks to better mimic ­functionally the fact that most real-world movements are done while either on one leg or with one leg bearing more weight. There is no doubt about the efficacy of the traditional double-leg lifts for positively affecting total body strength and hypertrophy. However, when the mechanisms behind noncontact athletic injuries or dissipation of force ­production are examined, it becomes obvious that singleleg forces are primary present most of the time. Therefore, to follow the principle of “specificity,” single-leg training produces excellent results without extreme loads or ­positions. It is essential to note that the shear force and ligamentous considerations seen in the other forms of closed chain squatting (including the leg press) apply to single-leg training as well. The depth of the squat in each of these exercises should be limited to avoid posterior shear forces when considering protection of the PCL. PF stresses should be limited by performance of these exercises in pain-free ranges.

Single-Leg Presses The earliest and easiest way to introduce single-leg training is with the leg press or Total Gym. The use of the machines provides the individual with the stability and confidence to perform the exercise through a more complete range of motion and with better lower extremity control and alignment. Also, as the individual advances in training, the leg press can still be used to allow for an environment that more safely applies progressive resistance. Remember, heavier and heavier loads are necessary to provide continued strength gains. The leg press is the most effective way of providing this in a single-leg movement (Fig. 5C-47).

264

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

a­ lignment. If necessary, reinforce this proper alignment through slight resistance of a Thera-Band (Fig. 5C-48). • In the top position, have the individual balance on the supporting foot. Ultimately, have the individual end with the unsupported hip flexed (Fig. 5C-49). • Teach the individual to slowly return the unsupported leg to the ground, versus “falling back.” This allows for a controlled eccentric movement of the exercise leg.

Figure 5C-47    Single-leg press on Total Gym.

To advance step-up training, first increase the height of the step. Ultimately advance to a position of where the hip is parallel to the knee. The higher the step, the better the contribution will be from the hamstrings and gluteal muscles, and the ability to cheat will be lessened. Dumbbells can also be added to provide for increasing resistance.

Single-Leg Wall Squats Step-Ups Step-up variations are an easy way to introduce singleleg training outside the confines of a machine. Step-ups are also an obvious functional movement that needs to be addressed early in the rehabilitation program. Forward and lateral step-ups are introduced on a smaller box with possible hand support and progressed to higher boxes without support. Of major emphasis in these exercises is minimizing the assistance of the ground leg in the step-up. A few helpful technique points are listed here. • To help minimize push-off from the ground leg, initiate the movement with the exercise leg on the box and the ground leg close to the box. Have the individual try to maintain a “flat foot” (hold the foot dorsiflexed) as the ground foot comes up. • Teach the individual to push down through the heel and midfoot on the box. • Emphasize hip extension and bringing the hips up and forward. Minimize the individual’s attempt to only extend the knee. • Especially in the lateral step-up version, focus on maintaining a level pelvis and proper lower extremity

Figure 5C-48    Lateral step-up with Thera-Band to elicit and train proper gluteal activations.

Another effective single-leg exercise performed in somewhat of a supported position is the wall squat’s single-leg version. The wall squat can be performed in a modified single-leg position and ultimately as a single-leg exercise (Fig. 5C-50). Ayotte and colleagues compared the EMG signal amplitudes of the single-leg wall squat with a number of commonly prescribed single-leg exercises—the lateral step-up, forward step-up, mini-squat, and retro step-up.134 Their results demonstrated that the single-leg wall squat produced the highest levels of muscle activations for the gluteus maximus, gluteus medius, and VMO. Considering the high levels of muscle activation for all muscles for the wall squat, this exercise can be effective in strengthening these muscles individually or training a cooperative effort for the entire lower extremity kinetic chain.

Lunges Lunges should be a staple of any knee rehabilitation program. Lunges are great because individuals of virtually any capacity can perform some variation of them. They allow for recruitment of all key lower extremity muscle groups in functional movement patterns without the need for machines or heavy resistance. Also, lunges add real value

Figure 5C-49    Step-up end position. Teach and train full hip extension of the stance leg.

Rehabilitation and Therapeutic Modalities

A

265

B

Figure 5C-50    Single-leg wall squats. A, Supported. B, Unsupported.

to the program because they allow for the application of the “same but different” principle. Meaning, the lunge can be varied in a number of ways without really changing the exercise completely. This allows the individual to ­maintain proficiency in the lunge maneuver and at the same time receive the benefits that the lunge variations provide. Regardless of the type of lunge being performed, the following techniques apply:

There are two basic ways to progress on the lunge: through the type of lunge and through the use and ­placement of external loading. The type of lunge is actually the easiest way to dictate proper performance. This is the progression we are currently using: static lunge (split squat) → reverse lunge → walking lunge → lateral lunge.

• Keep the front knee in the lunge in line with the foot and teach the individual to lower the hips by sitting back (similar to the proper squatting). Do not allow the knee to dip inward or migrate too far in front of the foot. • Have the individual keep the chest and head up. • The knee of the back leg should be slightly flexed, with the feeling of a light hip flexor stretch. • Like mentioned in most of the previous exercises, ascend from the lunge position through a focused hip extension contraction (Fig. 5C-51).

• Static Lunge—The static lunge minimizes typical movement faults and allows the focus to be on proper movement quality. Simply, the individual starts in the split stance lunge position and squats up and down. This lunge exercise can be advanced through their performance with the back foot elevated (Fig. 5C-52). • Reverse Lunge—In this variation, the plant leg is kept in place while taking a step backward. This allows the individual to properly feel “sitting back” and also allows for the proper hip extension drive to be taught in the return to starting standing positions. (No, the individual is not walking backward.)

Figure 5C-51    Proper lunge ascension with hip extension emphasis.

Figure 5C-52    Lunge with back foot elevated.

266

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

­ istakenly trained only as knee flexors. In running, jumpm ing, or walking, the function of the hamstrings and gluteal muscles is not primarily to flex the knee but to extend the hip. Simply stated, for the most complete and proper knee rehabilitation, the muscles of the posterior chain—the gluteal muscles and hamstrings—need to be trained both as hip extensors and knee flexors. Therefore, we exercise these actions separately through two distinct types of hip extension movements. These exercises are categorized as hamstring- and ­ gluteal-­dominant bent leg exercises and hamstring- and gluteal-dominant straight leg exercises. It is most beneficial to use exercises from both these categories to properly train the posterior chain muscles.

Figure 5C-53    Lateral lunge.

Hamstrings- and Gluteal-Dominant Bent Leg Exercises Hamstring Curls

• Walking Lunge—The walking lunge brings it all together. More force is transferred to the front leg, requiring an eccentric and stabilizing contraction than the other lunges. Also, the front leg is required to produce more force to take the body into the next lunge step in a smooth and seamless fashion. • Lateral Lunge—The lateral lunge is an advanced lunge progression that trains lateral movement control. The individual simply steps laterally and sits back into a front lunge posture. Essentially, the nonlunge leg is located at the side of the body, versus behind the individual as in the other forms of lunging (Fig. 5C-53). The individual returns to the standing position to enhance lateral training. After mastering the body weight variations, lunges can be advanced and varied with external loading (adding weight). Here’s a typical and logical progression to use: body weight → dumbbells → Smith Press machine or barbells. As long as proper form and lower extremity alignment can be maintained and sufficient progress has been made without external loading, the individual is allowed to advance and ultimately receive the benefits of the various loading parameters. A well-designed rehabilitation and training program wisely includes a consistent dose of lunges in its design. It is mentioned in this section that each of the ­hamstringand gluteal-dominant exercises can be performed as singleleg exercises as well. Each of these exercises is discussed in more detail in the following sections. The rehabilitation specialist should be mindful of the total training volume that is being performed in the single-leg fashion to avoid overtraining and overstressing the recovering lower extremity.

Hamstrings and Gluteal Therapeutic Exercises The muscles that extend the hip, primarily the gluteus maximus and hamstring group, are often neglected in many rehabilitation and training programs. Programs frequently place excessive emphasis on the quadriceps and neglect the hip extensors. Even more disturbing, the muscles that extend the hip, especially the hamstrings, are often

Lutz and associates recorded relatively high activity of the hamstring muscles throughout the hamstring curl, with maximal EMG activity occurring at 90 degrees of flexion.135 Similarly, Andersen and coworkers determined that there was higher EMG amplitudes in the hamstring curl exercise than in the leg press or squat.76 Therefore, hamstring curl exercises are effective at isolating and training the hamstrings early in the rehabilitation process, especially if there are weight-bearing concerns. However, as stated previously, the function of the hamstrings and gluteal muscles in most functional movements are not to flex the knee but to extend the hip. A leg curl exercises the muscles in a pattern that is never used in sports or daily activities. As a result, we generally do not perform lying, sitting, or standing leg curls in the middle and later stages of rehabilitation. Similar to resisted open chain knee extension exercises, numerous authors have documented resultant shear forces during open chain knee flexion training. Each investigator reported posterior shear forces throughout the entire range of non–weight-bearing knee flexion.16,82,85,107,128,136 Hence, resisted knee flexion places possible deleterious stresses on a healing PCL graft and should be avoided in the early and intermediate phases of PCL reconstruction rehabilitation.

Stability Ball Bridges The stability ball allows for training to occur in a closed chain environment with little risk to the low back and offers the additional benefits of training for core stability. Remember, emphasize abdominal bracing (covered later) and hip extension and avoid lumbar hyperextension with these exercises. Stability ball bridges (Fig. 5C-54) help to train the individual to think gluteal muscle contraction and hip extension properly.

Stability Ball and Slide Board Leg Curls The stability ball and slide board leg curls effectively develop core stability through hip extension and lumbar stabilization while training the hamstrings to flex the knee

Rehabilitation and Therapeutic Modalities

267

(Fig. 5C-57). This is a great exercise to develop the knee flexion function of the hamstrings with simultaneous gluteal contraction and trunk stabilization.

Hamstrings- and Gluteal-Dominant Straight Leg Exercises

Figure 5C-54    Stability ball bridges. Reinforce gluteal squeeze and hip extension. Closely monitor for lumbar extension substitution patterns.

(Fig. 5C-55). These kinematic combinations more correctly simulate the actions of the posterior chain during running (especially sprinting) as the ground leg strives to “pull the ground” and help propel the runner forward. Each of these exercises offers different advantages. The slide board leg curl offers more resistance through the friction of the board, whereas the stability ball requires more trunk stability and musculature coordination.

Gluteal Muscle and Hamstring Raise Gluteal muscle and hamstring raises use a gluteal-hamstring bench and are designed to target simultaneous hip extension and knee flexion contraction with the lower body fixed. This exercise is effective for athletic populations and individuals with advanced training (Fig. 5C-56). Forceful knee flexion and hip extension is performed to extend the trunk. Again, it is vital that the gluteal muscles and hamstrings allow for the extension, not the lumbar spine.

Natural Gluteal Muscle and Hamstring Raise The natural gluteal muscle and hamstring raise can be used alone or as a substitute when a gluteal-hamstring bench is not available. Although simple in concept, this is a difficult exercise. In this case, the feet are anchored (often held) while the individual attempts to lower the torso to the ground under control, and then pull the trunk back up. Very few individuals can actually bring themselves up at first, so help is given by pushing themselves off the ground

A

B

This key therapeutic exercise category for the knee is often neglected. Because of the inherent muscular loads and potential stresses on the lumbar spine, many rehabilitation specialists are hesitant to include these exercises in a rehabilitation and training program for the knee. However, we believe that with proper supervision, the benefits of ­training the gluteal muscles and hamstrings in this functional pattern outweigh any risks. As mentioned earlier, ensure that these exercises are performed with a hip extension emphasis while maintaining a braced core.

Hip Extensions Hip extension training with the lower extremities off the edge of a treatment table, with the trunk stabilized, is effec­tive for an early training modality. With the knees extended, the individual raises his or her legs with forceful gluteal contractions. The legs are raised to a neutral position, taking care not to hyperextend the lumbar spine.

Hyperextension Despite the name, the emphasis should not be on ­hyperextending the lumber spine but rather on using the gluteal muscles and hamstrings as hip extensors. This exercise is similar to the hip extension exercise in muscle actions, with the difference being that the legs are now fixed and the trunk extends. In this case, extend the trunk upward to a neutral position, once again, avoiding lumbar ­hyperextension.

Romanian Deadlift The Romanian deadlift (RDL), or modified straight leg deadlift, is an extremely safe and beneficial exercise when performed correctly with an appropriate load. However, like the squat, the RDL can be dangerous when performed improperly or with too heavy a weight. Please note that this is an extremely difficult lift to teach and should be learned with a dowel or weight bar before loading. The following

C

Figure 5C-55    Stability ball leg curl. A, Beginning position. B, End position. C, Slide board leg curl.

268

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

too much weight is being used. Remember that this is an isometric exercise for the spinal erectors and a concentric exercise for the gluteal muscles and hamstrings. Movement should come from the hip, not from the lumbar spine.

Single-Leg Romanian Deadlift Possibly a better alternative to the RDL is its single-leg ­version. Single-leg RDLs develop the entire posterior chain, enhance balance, and decrease loads and stresses on the lower back owing to the lighter weights being used. For this exercise, the dumbbell is held in the hand opposite the stance leg. As the weight is lowered, the free leg is extended to the rear, staying in line with the torso (Fig. 5C-59). When the individual progresses to the point at which holding the heavy dumbbell becomes too difficult, dumbbells in each hand allow for the appropriate resistance.

Figure 5C-56    Gluteal hamstring raise.

technique points should help the rehabilitation specialist effectively teach proper RDL form.

Proprioceptive and Neuromuscular Control Exercises

• Feet should be about hip width apart. Knees are slightly bent. • Keep the back arched, the shoulder blades retracted, and the chest up. • Teach the individual to bend from the hip and push the butt back, not lean forward, while maintaining an arched back. • While maintaining your back position, slide the bar down your thighs until you reach the end of your hamstring range of motion (Fig. 5C-58). For dumbbell RDLs, the dumbbells are held with the palms in toward the thighs, and the hands should move down the outside of the thigh to the shin. • Return to standing by extending the hips, not the lumbar spine. Have the individual concentrate on forcefully contracting the gluteal muscles and pushing the hips forward.

The final step in the knee rehabilitation continuum is to perform proprioceptive and stability-challenging exercises. These therapeutic exercises are more dynamic and challenging by varying the surface’s stability, applying external perturbating forces, and increasing the difficulty of the exercise’s balance component. In these exercises, the individual must engage the prime movers, stabilizers, and neutralizers while dealing with the additional proprioceptive input. The goal of this “functional training” category is to provide “hybrid” knee-hip-pelvis-trunk strengthening and improve stability and coordination in an attempt to properly control force and maintain balance and posture. As mentioned previously, it is of vital importance that proper lower extremity alignment is maintained throughout the training continuum. It is in this category of exercises that we address, further correct, and train the individual’s ability to resist hip adduction–internal rotation forces and knee valgus forces. And remember, this is especially true when training females because research has repeatedly

Obviously, maintaining back position is important. If the individual begins to flex the spine, he or she has reached the end of the active range of motion of the hamstrings, or

A

B

Figure 5C-57    Natural gluteal hamstring raises. A, Beginning position. B, Midportion in which the individual forcefully contracts the gluteal muscles and hamstrings to eccentrically control the descent.

Rehabilitation and Therapeutic Modalities

269

   TABLE 5C-8  Exercise Modifications to Increase Challenge

Traditional, Less Challenging Double-leg Single-leg supported Double-leg, single-leg stable surface Double-leg, single-leg Double-leg, single-leg single-plane Double-leg, single-leg slow, controlled Double-leg, single-leg

Figure 5C-58    Romanian dead lift.

­ emonstrated their inability to correctly maintain this d proper posture during squatting maneuvers, step-downs, and jumping tasks.137 Markolf and colleagues have shown that muscular contraction can decrease both varus and valgus laxity of the knee threefold.130 There are several ways that the previously covered traditional exercises can be altered and advanced to become appropriate proprioceptive and neuromuscular control exercises. Listed in Table 5C-8 are simple suggestions on how to achieve the results desired from these types of exercises. Truthfully, this category of training is only limited to the imagination and creativity of the rehabilitation specialist. However, it is suggested that only a handful of exercises be introduced throughout the individual’s rehabilitation program because too much variability does not allow for an optimal training effect. Not much is gained if the individual is always trying to learn how to perform a new exercise. However, before proceeding to describe a number of these training variations, the evidence for their inclusion

Figure 5C-59    Single-leg Romanian dead lift.

More Proprioceptive and Neuromuscularly Challenging Single-leg Single-leg unsupported Double-leg, single-leg unstable surface Double-leg, single-leg with perturbing external force Double-leg, single-leg multiplane Double-leg, single-leg fast and dynamic Double-leg, single-leg with upper extremity activities

in the knee rehabilitation program should be explored. The use of unstable training implements such as stability balls, wobble boards, foam pads, and balance disks has grown in popularity in recent years. Like most new trends in rehabilitation, training programs often are used in extreme variations and amounts before clinical experiences and empirical evidence more accurately evaluate their effectiveness. Unstable training implements reduce, or eliminate, an individual’s points of contact with solid ground. Training in a more unstable environment is proposed to enhance rehabilitation outcomes by improvement of balance, kinesthetic awareness, and proprioception. However, increased lower extremity muscle activity has not been demonstrated when performing exercises on unstable surfaces when compared with their stable counterpart. Anderson and Behm were able to demonstrate increased EMG activity of the abdominals and lumbar erector spinae while performing squats on two balance disks.138 However, McBride and colleagues demonstrated that isometric squatting in an unstable condition significantly reduces peak force, rate of force development, and agonist muscle activity with no change in antagonist or synergist muscle activity.139 Basically, it is difficult to progress resistance levels sufficiently to elicit significant strength gains when training on unstable surfaces. As a result, there are few discernable strength benefits when performing a resistance exercise in an unstable condition. At our facility, we do advocate unstable surface training, not to add resistance to the prime movers, but rather to add additional stress to the stabilizers and ­neutralizers of the lower extremity kinetic chain. It should be discussed whether the development of balance, proprioception, and neuromuscular control through special training is a wise choice when viewed as an investment-versus-benefit endeavor. The question must be asked if the time devoted to excessive balance development will yield improved outcomes. Would time be better served by working on more aggressively progressing strength training to remedy imbalances and deficits rather than spending extra efforts on unique and innovative balance and proprioceptive challenging exercises? The answer is relatively straightforward. If there are significant strength deficits present, make the training emphasis on improving strength. Then, as strength levels improve, shift the training emphasis to an ideal integrated training approach that

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

270

A

B

C

D

Figure 5C-60    Multiplanar single-leg squats. A, Anterior. B, Lateral. C, Posterior. D, Posterior medial.

best improves neuromuscular control and develops strength that the individual will use. The wise rehabilitation specialist incorporates a measured dose of stability training (i.e., using an unstable training environment) within any knee rehabilitation and training program. Listed and described next are a number of typical exercise variations that can be used to enhance the individual’s proprioceptive and neuromuscular control. Be creative and functional when applying this type of training. But remember, too much variety in the program does not allow for sufficient progress. Choose a few of the types of exercises described here and stick with their variations. Using the “same but different” principle in this category of therapeutic exercises allows for optimal benefits.

Multiplanar Squats The dynamic multiplanar environment of sports demands that a single leg apply the force in a proprioceptively driven manner. That is, the ground-based leg has to control forces (i.e., concentrically, isometrically, and eccentrically) while the joint angles in the leg are continuously changing in all

planes of motion. By varying the location and movement of the free leg while single-leg squatting, the ground leg must account for forces in a number different planes. Cones and a Functional Training Grid (Engineering Fitness International, San Diego) both work well because they provide targets and, in the case of the Functional Testing Grid, objectivity in directions of motions (Fig. 5C-60). Resistance can be added by dumbbells or cuff weights if ­necessary.

Squat and Reach The squat and reach is similar to the multiplanar squats illustrated earlier. However, in this version, the individual reaches with his or her upper body to different targets. The targets can be varied in height and location. Again, resistance can be applied (Fig. 5C-61).

Manual Perturbations There may be certain sports (e.g., surfing, snowboarding, and beach volleyball) for which unstable surface training may offer appreciable carryover to performance. However,

Rehabilitation and Therapeutic Modalities

271

Sport Cord Activities Another method of applying instability forces to the individual is through the use of sport cords and elastic resistance. Step-ups, cone walking, and single-leg training variations can all be performed while secured and challenged by the resistance around the waist (Fig. 5C-63). Functional sports motions, such as basketball passes, volleyball digs, or soccer heading, can also be performed while tethered to the elastic resistance.

Unstable Surface Training Most therapeutic exercises can be advanced to become more proprioceptively and neuromuscularly challenging through their performance on an unstable surface. Halffoam rollers (round side up, round side down), balance disks, foam pads, Bosu balls, and rocker boards are several examples of numerous devices that are at the disposal of the rehabilitation specialist. Please remember the risk-versusbenefit ratio of these exercises and that safety comes first. No individual should be asked to exercise through potentially compromising motions or positions. Figure 5C-64 provides a few examples of unstable surface training.

Figure 5C-61    Single-leg squat and reach.

in most cases, the support surface is fixed. Even if the surface is uneven, it is still unyielding. For most individuals, especially athletes, instability forces are applied further up the kinetic chain, not at the feet. In the clinical setting, this type of ­unstable environment can be created through manual perturbations and resistances or with sport cord elastic resistance. Manual perturbations can be performed to appropriately challenge the individual in conjunction with most traditional exercises. By simply applying short and quick pushes from multiple directions and in different locations, the individual is required to provide the counterforces ­necessary to stabilize the entire kinetic chain. We have found that this training can be initiated relatively early in the program with double-leg stance and then advanced to perturbations during fairly stable exercise such as body weight squats and step-ups. Perturbations can also be applied with single-leg standing and then advanced to challenging balance during single-leg stance with ball tosses (Fig. 5C-62).

A

Single-Leg Balance with Upper Extremity Activities Exercises Single-leg balance training is initiated as early as tolerated by the individual and can be advanced to balancing on unstable surfaces as noted previously. Also, single-leg balance is challenged through such activities as playing catch, throwing and catching against a Rebounder, or performing functional activities like those listed previously.

Summary Many different aspects of rehabilitation, strengthening, and neuromuscular training are required to best enhance the recovery of an individual from a knee injury. This somewhat difficult task requires advanced knowledge of

B

Figure 5C-62    Manual perturbation training can be progressed from simple standing positions (A) to single-leg unstable activities (B).

272

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

Figure 5C-63    Sport cord training. A, With step-ups. B, With dynamic gait activities.

the entire knee complex’s anatomy and physiology. The exercises that were provided and organized will help the rehabilitation specialist provide an optimal training environment for a successful return to previous function.

ANKLE The ankle is a commonly injured joint that requires a detailed therapeutic exercise outline. Although much variability exists, most ankle rehabilitation programs use therapeutic exercises to enhance ankle stability (most commonly, lateral stability). Early rehabilitation and training of the ankle includes range-of-motion activities and isotonic strengthening exercises. In the intermediate stage of rehabilitation, a progression of proprioceptive training exercises is incorporated. Advanced rehabilitation focuses on sport-specific activities to prepare the athlete or ­individual

A

B

for return to competition and daily work and activities of daily living requirements. Although it is important to individualize each rehabilitation program, a well-structured template for ankle rehabilitation can be adapted easily to address individual needs. Once again, to ensure that the rehabilitation and training for the ankle is most thorough, an outline with the different components of therapeutic exercises for the ankle is provided. Within each of the programs, specific exercises are given. Simply include those areas as suggested, or adapt the routine to the specific deficits and needs of the ­individual. 1. Neuromuscular activation exercises—choose as needed. These exercises are used early in the training program to activate the key musculature of the lower leg and ankle.

C

Figure 5C-64    Examples of unstable surface training. A, Rocker board squats. B, Single-leg balance on half-foam roller. C, Foam padding.

Rehabilitation and Therapeutic Modalities

2. Gastrosoleus training—choose one or two exercises. These train the often weakened plantar flexors in progressively more functional closed chain activities. 3. Single-leg training—choose two or three exercises. These hybrid exercises allow for total lower extremity kinetic chain training and aid in strength and neuromuscular control development. 4. Proprioceptive and neuromuscular control exercises—choose two or three exercises. These are again functional in nature and train more specifically for balance, control, and ultimate stability in unstable movements and ­postures.

Neuromuscular Activation Exercises Early rehabilitation efforts typically focus on initiation of strengthening to isolate and train the plantar flexors, dorsiflexors, inverters, and everters in an open chain, conservative, and nonoffending environment. These exercises are simple and easy to perform and allow for a good home exercise program. Thera-Band and cuff weights can be used effectively (Fig. 5C-65). Of particular importance is the progression of eversion strength after a lateral ankle sprain. Therapeutic exercises designed to strengthen the ankle everters have typically been an integral part of the rehabilitation process after lateral ankle sprains. ­ Willems and ­associates investigated subjects with a history of ankle sprains and chronic instability and determined that the instability group showed significantly lower relative eversion muscle strength (percent body weight).140 It was their conclusion that the possible contributor of chronic ankle instability is evertor muscle weakness. Likewise, AshtonMiller and colleagues concluded in their research that fully activated and strong ankle evertor muscles are the best protection for a near-maximally inverted ankle at ­footstrike.141

Gastrosoleus Training Effective strengthening of the ankle plantar flexors is vital to a full return to function, particularly in athletic events that require a powerful and explosive push-off in sprinting and jumping. These calf muscles are also essential to the proper deceleration forces at the knee. It has been established by research that the gastrocnemius and soleus muscles are affected in their ability to fully contribute to ankle plantar flexion by the degree of knee flexion. The gastrocnemius, a two-joint muscle that originates proximal to the knee, is more active when the knee is extended, whereas the soleus contributes more when the knee is flexed.142,143 Miaki and associates determined that the soleus acts most selectively with the ankle in a neutral position and the knee flexed between 90 and 130 degrees.142 Calf raises are initiated with the individual in a recumbent position on a leg press or Total Gym, decreasing his or her body weight. The amount of dorsiflexion stretch allowed is dependent on the individual tolerance or by postoperative restrictions. To better target the soleus musculature, seated calf raises are initiated. A seated calf machine or dumbbells or plates can be used effectively (Fig. 5C-66).

273

Single-Leg Training The benefits of single-leg training were espoused earlier in our discussion of the knee. The benefits of training on one leg cannot be overemphasized. The essential benefit of single-leg training during an ankle rehabilitation program is the balance, proprioceptive, and neuromuscular input that it requires—throughout the entire lower extremity kinetic chain. Of particular interest is a study conducted by Beckman and Buchanan.144 Like many previous studies, their research subjects stood on a platform constructed such that either foot and ankle could be instantaneously inverted. However, not only did they monitor the response of the lateral ankle muscles to stabilize against this destabilizing incident, they also looked at the gluteus medius muscle’s response. They were able to determine that there is decreased latency of hip muscle activation after ankle inversion in the hypermobile population. They concluded that when treating ankle instability, clinicians must address the hip muscle recruitment patterns as well. No different than what was discussed in the rehabilitation of the knee, we employ single-leg training exercises that challenge the individual’s ability to maintain proper lower extremity alignment in all the joints. The various forms of step-ups and lunges are all included in the rehabilitation and training of an individual with an ankle disorder. The reader is encouraged to review these exercises described earlier. We employ additional modifications of the single-leg therapeutic exercises to enhance their benefits on the ankle complex. We do advance these closed chain exercises through the addition of unstable surfaces. Performing lunges and step-ups on a foam surface or rocker board can easily be integrated (Fig. 5C-67). Also, to better train the gastrosoleus muscles, having the individual perform a calf raise with the exercise makes the exercise that much more effective (Fig. 5C-68).

Proprioceptive and Neuromuscular Control Exercises The most common strategies that have been prescribed to prevent ankle sprains are proprioceptive training and strength training. The loss of proprioceptive sense following an ankle injury has been well documented in the literature.114,140,145-147 Likewise, we strongly emphasize proprioception and strength training in the rehabilitation program for ankle instability. Dynamic joint stabilization is achieved by a co-contraction of the muscles surrounding the ankle. During activities that involve the lower limb, such as running, cutting, and jumping, the athlete relies on these muscular co-contractions, in particular eccentric control and plantar flexioninversion control, to provide stability to the ankle complex. We believe that those who are lacking or imbalanced in these functional movements may be susceptible to injury because they do not have the ability to dissipate and control these forces. Dynamic proprioceptive and neuromuscular control exercises are employed in much the same manner as those discussed for the knee. Wobble boards, stability disks, foam mats, rocker boards, stability balls, and sport cords can all be used to aid in providing unstable training ­environments.

274

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E Figure 5C-65    Ankle muscle activation and isolation training. A, Dorsiflexion. B, Plantar flexion. C, Inversion. D, Eversion. E, Use of cuff weights.

Rehabilitation and Therapeutic Modalities

A

275

B

Figure 5C-66    Seated calf raises.

Figure 5C-68    Calf raise emphasis with step-ups (A) and lunges (B).

We mention an additional few exercises that we have found to be helpful in specifically enhancing the stability of the ankle. We believe that additional specific evertor muscle training is warranted for the individual with ankle instability. Rehabilitation strengthening exercises for the ankle evertors often use resistive bands or cuff weights with motions performed in an open chain manner and with the uniplanar movement of eversion. However, a more functional and meaningful way to enhance the evertor’s stabilizing function is to train in a closed chain and multiplanar environment. A controlled way to introduce this exercise is through a calf raise movement that emphasizes the ability of the peroneus longus to stabilize the first metatarsal. With insufficient stabilization from the peroneus longus, the first metatarsal head loses contact with the ground, the foot supinates, and weight shifts to the lateral foot and ankle. This could, in turn, overstress the lateral ligament complex. To emphasize stabilization of the first metatarsal by the peroneus

l­ongus, a coin is placed under the first metatarsal head of the foot to serve as a tactile cue. The goal is to perform the heel raise motion while the first metatarsal head maintains firm pressure on the coin, which demands stabilization from the peroneus longus (Fig. 5C-69). To advance this exercise, a resistive band is positioned around the midfoot and anchored laterally, or away, from the body, perpendicular to the long axis of the foot. The resistance from the band provides a supination-inversion force at the ankle that opposes the function of the peroneus longus (Fig. 5C-70).

Figure 5C-67    Step-up onto foam surface.

Figure 5C-69    Peroneus longus training.

Eccentric Training Special attention should be given to eccentric training in this section on therapeutic exercises for the ankle. Although there is substantial evidence on the benefits of therapeutic exercises for the treatment and prevention of musculoskeletal disorders, possibly the strongest case for specific exercise intervention is seen in the treatment of

276

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Summary Therapeutic exercise prescription after an ankle injury must include strength and neuromuscular training for the entire lower extremity kinetic chain. A thorough and complete rehabilitation program for the ankle initially includes isolation and activation exercises and then progresses rapidly to a functional ground-based training regimen that serves to train the co-contraction of the muscles that stabilize the ankle-foot complex in real-world movement patterns.

EXERCISE APPLICATION

Figure 5C-70    Calf raise targeting peroneus longus stabilization on first metatarsal.

Achilles tendinopathy. Research has provided evidence in clinical trials that eccentric training obtained overall better results in regard to pain reduction,148-151 successful return to activity,135,148,150,151 increases in calf strength,148,151 and decreases in tendon thickness.118,152 The protocol that is typically followed is the eccentricbased treatment regimen provided by Alfredson and colleagues in which the subjects performed eccentric-only calf raises for three sets of 15 repetitions twice daily for 12 weeks (Fig. 5C-71).148 The subjects were informed to exercise through pain so long as it was not “disabling.” Resistance was advanced from body weight to an added load when the training session could be completed pain free. Taunton and colleagues actually determined that performing three sets of 10 (as well as 20) repetitions was just as beneficial as Alfredson’s protocol of three sets of 15 repetitions.153 Fahlstrom and coworkers suggested that treatment with eccentric calf muscle training may be more beneficial for patients with chronic painful mid-portion Achilles tendinosis, not for patients with chronic insertion Achilles tendon pain.149

Figure 5C-71    Eccentric calf training. The concentric portion of the calf raise is performed with both legs.

There are a few other topics that must be addressed when considering proper therapeutic exercise prescription. A proper warm-up certainly enhances the rehabilitation and training program. Likewise, applying the appropriate loading parameters ensures that the concept of progressive resistive training can be used most effectively.

Proper Warm-up It is generally believed that preparing the body before training through proper warm-up activities benefits performance and decreases the risk for injury. Common warm-up techniques have included general full body training, specific joint-related movements, and dynamic flexibility drills. General and specific static stretching has been used to prepare individuals for the activities to follow. Also, thermal modalities have been cited as appropriate means to increase local tissue temperatures and prepare for ­training. The proper means of warming up and preparing for the therapeutic exercise routine that follows is dependent on the current rehabilitation phase and the specific needs of the individual. Early in the rehabilitation program, when controlling pain levels and increasing joint range of motion and soft tissue mobility are primary objectives, thermal modalities and static self-stretching are the focus. Once pain levels have diminished, the individual may benefit from a general active warm-up, such as upper body ergometer training, brisk walking, or bicycle and elliptical training, followed by static self-stretching. In the more advanced rehabilitation phases, more dynamic and skillspecific training can be implemented without the need for specific static stretching. In regard to static stretching, a recent meta-analysis of 361 research papers examining the relationship between stretching and injury prevention revealed that static stretching before activity is not associated significantly with a reduction in injuries.154 However, if restrictions in joint and soft tissue mobility are present, self-stretching and manual therapy techniques should obviously be implemented. Basically, an appropriate warm-up should include modalities if necessary and a period of light exercise to increase body temperature (in the later phases of rehabilitation) to allow for improved dynamic flexibility and prepare the shoulder complex for manual therapy or for the training to follow. For strength training, repetitions with a lighter resistance should be performed to prepare the muscles and joints for that particular ­exercise.

Rehabilitation and Therapeutic Modalities

Loading Parameters When designing resistance training programs, the rehabilitation specialist has to consider a number of variables that can be manipulated to make programs different. These include choosing (a) the exercise, (b) the repetitions, (c) the sets, (d) the resistance, (e) the speed of performing the exercise, (f) the order of exercises, and (g) the rest periods between sets and exercises. We proposed the use of a training cycle whereby the intensity is increased each week of the cycle in a linear method of intensification. The initial loading parameter used in our facility is based on the processes of accumulation and intensification. For example, the individual begins with a resistance in which he or she can perform three sets of 10 repetitions with perfect form and can complete without negative clinical symptoms. Advance to three sets of 12 repetitions, and ultimately to three sets of 15 repetitions, with the same resistance in an as-tolerated fashion. Once the individual can perform three sets of 15 repetitions, advance in weight, tubing, or band resistance. This simple progression of loading allows for the most comprehensive training effects. Strength, strength endurance, neuromuscular control, and tissue remodeling are all positively affected in this manner. The appropriate amount of resistance implemented during a shoulder rehabilitation program has traditionally been a topic of discussion. Some believe that resistance levels must remain light (e.g., 1 to 5 pounds for dumbbell training for the rotator cuff) to ensure that the targeted shoulder muscles remain activated. We believe that as long as the exercise is performed correctly, the exercise’s targeted muscles will continue to respond anatomically as they were designed. There should be no reason for a muscle group to no longer function during a movement just because the resistance increased (again, as long as the exercise is performed correctly). Dark and associates demonstrated this point in their study on simple internal and external resistance training.155 As the intensity of both internal and external rotation exercises increased, the activity of the appropriate rotator cuff muscles increased in a systematic manner. For the large phasic muscles of the lower extremity, progressively heavier weights need to be applied for those compound motions that involve a number of the gluteal and thigh muscle groups. Individuals must be encouraged to provide more intense efforts in exercises such as leg presses, squats, and lunges to gain optimal strength training effects. As the individual’s strength and exercise tolerance increases, we have found that a subtle linear pattern of progression is simple and effective in progressing strength levels. A linear pattern is characterized by fairly regular progressions in resistance that are relatively equivalent and in small increments in training intensity each week. These subtle variations in intensity (and workload) enable the individual to better maintain proper exercise technique and help to avoid overstressing the recovering structures. Table 5C-9 provides an example of how an individual’s resistance might be progressed over a 4-week period of time.

277

   TABLE 5C-9  Linear Pattern of Progression Set Repetitions Week 1 Week 2 Week 3 Week 4

1 10 30 35 40 40

2 10 30 40 40 45

3 10 35 40 45 50

The key to continual strength increases does not rely solely on the special program of exercises prescribed or the training tools used. Although these are important, the key to ongoing progress is the consistent effort put into training. Realistically, as long as the program attempts to progressively add more resistance to the exercise over time, practically any training method used will lead to gains in strength. In the advanced and return-to-activity phases, particularly with the multijoint, large muscle strengthening exercises, more advanced loading parameters may be introduced.

CORE TRAINING Core training has become the topic of debate and discussion and often the focus of both clinical practice and strength and training programs in recent years. Core work has been prescribed for both its therapeutic benefits and proposed performance enhancement capabilities. Walk into any physical therapy clinic, training room, gym, or health club and the virtues of core training will most often be exclaimed. Seemingly, the term core training has become a label for any exercise that addresses some aspect of lumbar spine and hip stability. In this view, core training could refer to any mode of exercise that addresses any one of the number of different muscles associated with the lumbar spine, pelvis, and hips. Our definition of core training involves both stabilization and strength training. Core stabilization training could be defined as training in which the exercise aim is to minimize or eliminate movement of the spine. In other core exercises, movement of the spine is incorporated into the exercise. These exercises would be termed core strengthening. Simply put, core stabilization + core strengthening = core training. Core training is most often applied to the treatment and prevention of low back disorders but is also now being touted as a vital component in the rehabilitation and care of injuries of the extremities. Furthermore, strength and conditioning specialists have integrated core training into their programs with the thought that a more stable core translates into improved performance on the playing field. As a result, core training has become considerably more complex than the broad use of the term. Although this section is slanted toward therapeutic exercise applications for low back disorders, we do not want to tackle the most complex issue of how to treat low back pain. However, we do mention that core training has become accepted as a standard component in the treatment of lumbar spine disorders. Presently, there is growing evidence that the use of a classification approach to physical

278

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

therapy results in better clinical outcomes than the use of alternative management approaches. In 1995, Delitto and colleagues proposed a classification system intended to aid in the evaluation and rehabilitative management of patients with low back pain.156 They described four classifications for patients with low back pain: manipulation, stabilization, specific exercise, and traction. Fritz and colleagues more recently reviewed this classification system and essentially updated and modified the previous recommendations in light of new evidenced-based practices.157 They updated the examination criteria and intervention strategies with the intent of further enhancing positive treatment outcomes. Relevant to our discussion on therapeutic exercise prescription, core training that promotes isolated contraction and co-contraction of the deep stabilizing muscles, as well as strengthening of the larger global stabilizing muscles, is the current accepted standard of therapeutic exercise programming for those requiring core stabilization. Research has subsequently proved that adherence to these guideline recommendations is associated with better clinical outcomes.86,158,159 It has more recently been suggested that general trunk exercises alone may be better suited for patients with recurrent episodes of nonspecific subacute or chronic low back pain. These patients are in contrast to those with signs and symptoms of instability. In line with evidence from other studies on patients with nonspecific recurrent low back pain, it could be suggested that a general exercise program is more successful in the management of patients with recurrent nonspecific subacute or chronic low back pain.160,161 This section does not attempt to detail a single, specific, or ideal approach to core training. Although some seem confused because there has yet to be found an optimal approach to training the core, we do not see it that way. Is there an optimal approach for the rehabilitation and training of the shoulder, knee, ankle, or any musculoskeletal disorder? Not really. We would like to first describe the concepts related to the original development of spinal segmental stabilization of the low back, then detail the components that make up lumbopelvic stability, and ultimately suggest an approach to core training that is integrated in its design. Although it has been concluded that therapeutic exercises are beneficial for chronic, subacute, and postsurgery low back pain, there appears to be no such thing as an ideal set of exercises.7 However, there are general suggestions for exercises that emphasize trunk stabilization in a neutral spine while also emphasizing mobility at the hips and knees. McGill suggested that an ideal exercise would challenge the particular muscles while imposing minimal spine loads with a neutral posture and elements of whole body stabilization.162,163 The key is to provide training modalities in a sequential manner that would develop core strength and stability as a part of the total training program. Among clinicians and strength and conditioning specialists, it is generally agreed that all the muscles surrounding the spine provide stabilization to some degree during physical activity. Although we believe that a comprehensive approach to spinal stabilization and conditioning is necessary, specific muscles require attention in the rehabilitation of an individual with low back pain.

In the healthy spine, the trunk musculature functions to control and initiate movement, respond to loading and postural perturbations, provide stiffness, minimize aberrant movements, and provide a stable base for activity. Although all muscles of the trunk play a role in stability to some degree, certain muscles have a more specialized function than others. Stabilization of the lumbar spine is achieved through muscles classified as either having deep stabilizing or global stabilizing function. The deep and intrinsic spinal stabilizers have received considerable attention in the literature owing to their ability to prevent movement outside the spine’s neutral zone. However, the focus of most rehabilitation and training programs is often on the large global stabilizers and is deficient in the area of local stabilization. Researchers have found that after acute episodes of low back pain, important deep stabilizing muscles of the back become inhibited and stop functioning normally. These muscles atrophy significantly,164,165 become less resistant to fatigue,166 change their muscle fiber composition,167 and decrease their ability to contract.167,168 Unfortunately, these deep stabilizing muscles do not spontaneously recover—even if patients are pain free and are able to return to normal activity levels.169,170 Therefore, the goal of the first phase of core training is to train, through a high level of awareness, the specific isometric co-contraction of the transversus abdominis with the lumbar multifidi. These deep stabilizers are activated with low levels of volitional contraction and controlled breathing while maintaining a neutral lordotic posture. The term now commonly used to describe this co-contraction is ­abdominal bracing. The deep stabilizers are those muscles with intervertebral attachments that are capable of providing intersegmental stability.171 The multifidus, transversus abdominis, and internal oblique muscles are classified as deep stabilizers. The deep stabilizers fulfill the role of stabilizing the spine when the integrity of the spine’s neutral zone is challenged. In the healthy spine, contraction of the deep stabilizers is automatic and precipitated by movement of the extremities or trunk, unlike the injured spine in which the activation is suppressed or delayed. The global stabilizers include the rectus abdominis, spinal extensors, external oblique, quadratus lumborum, and psoas muscles. The global stabilizers function in response to voluntary effort during the initiation of spinal movement and during challenging activities that require a stiff spine. Many rehabilitation specialists mistakenly believe that strength is the key to lumbar spine stability. In most cases, however, back pain is not about a weak back. In fact, McGill’s research has determined that those with a bad back generally have stronger back extensors than those with a weak back.172,173 Back pain is often a result of cumulative microtrauma and overuse. Evidence has now led us to the understanding that motor control coordination is actually the key to stability and that trunk muscle endurance (not strength) is related to low back pain symptom reduction.174 When patients require rigidity under load, they must be trained to function under those conditions, but most patients who experience low back pain need an intrinsic retraining program first to ensure control of the

Rehabilitation and Therapeutic Modalities

joint neutral position. Although this intrinsic system can be more time consuming and is difficult to teach at first, the system cannot be ignored any longer in the future of exercise rehabilitation. Many prominent researchers and clinicians have detailed their approach to core training.162,172,175-178 It is important to note that in each of these approaches, the general premises of our previously mentioned AIR principle still resound. Core training must adhere to the same philosophy as other modes of exercise. With this in mind, core training progresses through a number of phases, should be progressive in nature, needs to be applied in a systematic fashion, and needs to be specific to the individual’s needs. No different than training to rehabilitate other joints of the body, we employ the AIR (activate, integrate, and reinforce) principle in our core training approach. The outline and terminology for the core training we employ are discussed next.

Neuromuscular Activation and Stabilization (A) In this initial phase of training and rehabilitation, mastery of core contraction of the deep stabilizers is the primary emphasis. The therapeutic exercises prescribed seek to educate and train the individual, through conscious awareness, how to maintain a stable system through static holds and slow movements in a stable environment. Retraining of these muscles to produce continuous, low-grade forces over long periods of time is also a key component in core training.

Dynamic Stabilization (I) Once these weak and inhibited stabilizing muscles are “activated” through abdominal bracing, exercises in the dynamic stabilization phase are provided that integrate the deep stabilizing muscles in functional patterns while in a relatively pain-free and stable environment. The aim is to integrate these patterns of co-contraction in more functional environments so they will eventually occur automatically. Static holds in progressively more unstable environments and dynamic movements in a stable environment are the foundation of this phase of training.

Advanced Stabilization and Strength Training (R) The advanced stabilization and strength training stage is the aim of the specific exercise intervention, whereby individuals can dynamically stabilize their spines more automatically during the functional demands of daily living. It is at this stage that the individual should be able to cease the formal specific segmental stabilization program. In this final phase, therapeutic exercises are given that reinforce the proper motor patterns, coordination, and endurance of deep and global stabilizing muscles in progressively more unstable and dynamic surroundings. Resisted, dynamic movements in an unstable environment are a focus at this point in the training program. Again, this encompasses the specific movement patterns, neuromuscular control needs, and energy demands of the activity.

279

Putting It All Together—Therapeutic Exercise Program Design for Core Training To encompass and target the essential components of a core training program for rehabilitation, injury prevention, and performance enhancement, an outline of different exercise categories has been provided (once again). The benefits and aims of these subgroups of training are described next. Within each category of training, specific exercise progressions are offered. Although some individuals may require training in each of these categories (particularly after injury to the low back), others would benefit from a modified version to meet their particular needs in training. 1. Bridging progression—Bridging helps train the individual to maintain co-contractions of the deep stabilizers while activating the gluteal muscles (not the spinal erectors) to extend the hip. A key component that has been stressed (and will continue to be stressed) is training for motion to occur at the hip while the lumbar spine remains stable. 2. Quadruped progression—The purpose of the quadruped progression is to teach the individual to stabilize the spine and trunk with the deep stabilizers while simultaneously putting the upper and lower extremities through controlled motions. 3. Lateral flexion progression—These variations allow for enhanced activations of the oblique muscles, transverse abdominis, latissimus dorsi, and quadratus lumborum— all vital for core stability. 4. Rhythmic stabilization and perturbation training—These PNF techniques help to reinforce abdominal bracing and whole body stabilization in meaningful and functional postures. 5. Curl-up progression—These exercises allow for strength and stability training of the rectus abdominis, internal and external oblique muscles, and abdominal wall. 6. Functional gluteal muscle training—Training the gluteal muscles in functional squatting patterns helps to engrave safe lifting and movement patterns. These exercises also allow the deep muscles to be trained as stabilizers and the gluteal muscles to be trained as primary movers in more dynamic and functional movements. 7. Functional latissimus and scapular training—The latissimus dorsi helps to both generate lumbar extension motion and stabilize the lumbar spine through its origin at the lumbar spinous processes by way of the lumbodorsal fascia. 8. Torso rotation training—Early in the rehabilitation process, the rotatory muscles are trained more as anti­rotators because they attempt to counteract and stabilize against rotation forces. As the individual advances, rotation training becomes more powerful and explosive in nature, with the continued goal of minimizing shear forces to the spine.

Loading Parameters For the core stabilization exercises, a body weight progression is followed in which each exercise repetition is held for 5 seconds and progressed sequentially from 10 to

280

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 5C-72    Teaching abdominal bracing. A, Palpate and feel tension in the lumbar spinal erectors in a slightly flexed position. B, Return to standing until the erectors quiet down and then contract the abdominals.

15 ­repetitions. After 15 repetitions can be achieved symptom free with proper form, the intensity of the exercise is increased, or a more demanding exercise is introduced. The rhythmic stabilization and lateral flexion core stabilization are progressed in duration of holds. Work up to 60 seconds before advancing the exercise. For the core strength training (functional gluteal, latissimus and scapular, and torso training), progress from 10 to 15 repetitions for three total sets.

Abdominal Bracing Teaching abdominal bracing can be challenging at times, but there are a few techniques that have been suggested to help teach this most important musculature activation. McGill suggests that the individual bend forward slightly, enough to feel the lower spinal erectors contract while palpating the lower back.162 The individual then slowly extends to standing position until the individual can feel the extensors shut off. This should be the neutral spine position. Without moving, have the individual contract his or her abdominals slightly until the extensors turn back on. This should be a braced position (Fig. 5C-72). Having the individual attempt to slightly pull the bellybutton “up and in” or to contract the abdominals as if to prepare for the impact of a punch to the stomach may help elicit the correct abdominal brace. Some clinicians teach the individual to achieve abdominal hollowing by drawing in, in an attempt to draw the bellybutton to the spine (Fig. 5C-73). Drawing in is used in our program with initial quadruped and bridging exercises to reeducate and activate the transverse abdominis. As the individual progresses to stabilization training with degrees of compressive loading in standing, much more than just activating the transversus abdominis is necessary. The abdominal brace is emphasized to use all the muscles of the core to protect the spine so that it will not buckle or shear in an unstable way. A landmark study in segmental stabilization training showed that when specific exercises for the deep

s­ tabilizing muscles of the back and abdominal wall were not performed following a first episode of acute low back pain, those subjects were 12.4 times more likely to have recurrence of back pain within 3 years.109 Seventy percent of the participants in the study who retrained these deep stabilizing muscles in a 4-week program reported 3 years later that their lower back pain had not returned. Other studies show that these specific back exercises are also beneficial for those suffering from chronic lower back pain.179 Other studies demonstrate that the brain reacts to pain and pathology with altered and delayed firing patterns of the deep stabilizing muscles.180,181 The mechanism of preparatory spinal control provided by transversus abdominis and multifidus is impaired. This results in decreased muscle stiffness and poor spinal segmental control in response to movements or situations requiring increased spinal ­stability.

Figure 5C-73    Abdominal hollowing by drawing in the stomach.

Rehabilitation and Therapeutic Modalities

Figure 5C-74    Bridging.

Bridging Progression

281

Figure 5C-75    Unilateral bridging.

The bridging progression is a key component in the core training program. Training to fire both the gluteal muscles and the hamstrings while maintaining core co-contractions of the deep stabilizers cannot be overemphasized. The individual must be taught to first contract the gluteal muscles and then, with the back “locked,” extend the hips.

leg pushes through the floor at the heels and the hip is extended by the gluteal muscles, the contralateral leg flexes at the hip and knee. If necessary, the individual can pull the knee to the chest. Hold the top position. If the individual feels pain in the low back or senses the hamstrings “cramping,” he or she is most likely ineffectively firing the gluteal muscles.

Bridging

Bridging with Lower Extremity Movements

This is where the individual should first be educated about the critical difference between hip range of motion and lumbar spine extension range of motion. Beginning in the hook-lying position, have the individual “squeeze” the gluteal muscles, then extend the hips to a position that creates a straight line through the hips to the shoulders (Fig. 5C-74). Again, avoid lumbar spine extension! At this top position, reinforce the deep stabilizers’ contribution by having the individual perform an abdominal brace and hold the position for 5 to 10 seconds.

The final step in our bridging progression is to add alternating lower extremity movements. These progressive exercises truly enhance multifidi activity. Have the individual push down through the heel of the stance leg and squeeze the gluteal muscles. The individual must work to maintain a level pelvis while only one foot is contacting the surface. Again, ensure that the individual is striving to maintain an abdominal brace position. The first progression is to add a small alternate marching action to the bridge (Fig. 5C-76). The stability is further challenged by performing a knee extension and holding the leg while the pelvis remains level (Fig. 5C-77). In this version of the bridge series, relative to other common core

Unilateral Bridging Unilateral bridges help to activate the multifidi muscles and reinforce their ability to stabilize against rotation forces. Simply perform the bridge as described earlier with a single leg (Fig. 5C-75). Make sure the individual doesn’t “help” extend the hips with the straight, nonexercise leg. Also, this type of bridge does not allow as much hip extension range of motion. Do not let the individual hyperextend the lumbar spine in an attempt to achieve the amount of hip lift he or she thinks is necessary. Interestingly, Stevens and coworkers demonstrated that internal oblique muscle activity increased when bridging advanced to the single-leg versions.182

Single-Leg Hip Lifts The hip lift is a great exercise and neuromuscular training tool to help teach proper hip extension with contralateral hip flexion—just as the lower extremities act during walking, stair climbing, and especially running. As the ground

Figure 5C-76    Bridge with alternate lower extremity marching.

282

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5C-77    Bridge with alternate lower knee extension.

training exercises, Ekstrom and colleagues recorded high activities for the gluteus medius and maximus and the lumbar multifidus muscles of the supporting lower extremity.183 Ultimately, these bridge progressions can be performed on a stability ball or other unstable ­ surface. This allows for increased hip ranges of motion and further challenges the individual’s ability to stabilize in an unstable environment (Fig. 5C-78).51

Quadruped Progression The purpose of the quadruped progression is to teach the athlete to stabilize the spine and trunk with the deep stabilizers while simultaneously putting the upper and/or lower extremities through controlled motions. In proper quadruped posture, the knees are positioned under the hips, the hands are under the shoulders, and proper spinal curves throughout are maintained.

Quadruped Drawing In Initially, the individual is taught the drawing-in maneuver to help elicit the proper transverse abdominis activation. In some cases, those who struggle with this activating pattern

Figure 5C-79    Quadruped stabilization with alternate upper extremity lift.

in the supine position better “feel” this contraction owing to the slight resistance provided by the weight of the internal organs in this gravity-dependent posture. However, as the quadruped exercises advance, the shift in stabilization focus transitions from drawing in to abdominal bracing.

Quadruped Upper Extremity Lift The initial progression in all-four position is through active elevation of the arm. This three-point stance now applies a rotatory load to the trunk, encouraging increased ­multifidi and abdominal oblique activations. A dowel can be balanced along the spine to reinforce the maintenance of proper spinal curves (include the cervical spine). The position is held for 5 seconds. The arm is then lowered, and the exercise is repeated in alternate fashion with the other arm (Fig. 5C-79).

Quadruped Hip Extension The next step in the quadruped sequence is the quadruped bent leg extension. It is of utmost importance that the individual is taught to use the hip extensors, not the lumbar spine extensors, to extend the hip. The aim is to extend the hip to a position of neutral, as long as the lumbar spine remains neutral and stabilized. As the individual is able, the hip can be extended with the leg remaining straight, effectively increasing the lever arm and intensity of the exercise (Fig. 5C-80). Use of the dowel rod exposes the inability to maintain proper spinal curves. During both the hip extension and the alternating leg and arm exercise (described later), the contralateral internal oblique and ipsilateral external oblique reached high levels.184,185

Quadruped Alternating Arm and Leg (Bird Dog)

Figure 5C-78    Bridge with contact on unstable surface.

This exercise represents the ultimate goal for quadruped training. Too often this exercise is prescribed too early in the rehabilitation protocol and, as a result, has become an exercise that is often poorly performed (Fig. 5C-81). Slow and controlled movements with a braced core are

Rehabilitation and Therapeutic Modalities

283

a must. The individual should not be allowed to drop, rotate, or hyperextend the pelvis at any time. As necessary, limit the ranges of upper and lower body motions to allow for proper maintenance of stabilization. Ekstrom and ­ associates ­ demonstrated in their EMG research that the gluteus maximus, spinal erectors (longissimus thoracis) and lumbar multifidi were activated well.183 Progressions of the bird dog include upper extremity sweeping (Fig. 5C-82) and use of unstable surfaces. The “sweeping” ­further enhances upper extremity motions and time spent in two-point stance.

Lateral Flexion Progression Figure 5C-80    Quadruped stabilization with alternate lower extremity lift.

A

The lateral oblique muscles and quadratus lumborum are often neglected in core training programs. When considering spine loads and stability, the side bridge is the optimal technique to train the quadratus lumborum, transverse

B

Figure 5C-81    “Bird dog” quadruped stabilization with alternate upper and lower extremity lifts. A, Proper form. B, Typical improper form.

A

B

Figure 5C-82    “Bird dog” with upper extremity sweeping. The individual “sweeps“ the hand on the surface, from A to B, eventually attaining full “bird dog“ position.

284

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5C-85    Side bridge.

Figure 5C-83    Standing side bridge. While maintaining proper abdominal brace, the body is slowly lowered to the wall and back.

abdominis, latissimus dorsi, and internal and external oblique muscles simultaneously.162

Standing Side Bridge The standing side bridge is a remedial starting position for those unable to perform these exercises on their side. The standing position eliminates a majority of the body weight and allows some training effect. Simply lower the trunk to the wall while maintaining a neutral and braced spine. Proper muscle control is necessary to avoid any type of lateral flexion or rotation (Fig. 5C-83).

Bent Leg Side Bridge In this beginning version, the individual is supported by the elbow and knee, with some help from the lower leg. Encourage a strong abdominal brace and ensure that the head and neck are maintained in proper alignment as well (Fig. 5C-84).

Figure 5C-84    Bent leg side bridge.

Side Bridge The side bridge is the next progression in the sequence as the resistance and intensity are increased by straightening the legs. At the top position, the body is supported by the elbow and the feet (Fig. 5C-85). The side bridge exercise produced high EMG signals in the gluteus medius muscle and in the external obliques.183

Rolling Side Bridge Our most advanced lateral flexion core training is the rolling version of the side bridge. Placing the upper leg and foot in front of the other allows for small amounts of ­rolling of the trunk. This trunk unit rolling challenges both the anterior and posterior portions of the abdominal wall (Fig. 5C-86).162

Curl-Up Progression Curl-up training allows for training of the anterior abdominal wall and rectus abdominis effectively. Curl-ups are performed in lieu of traditional sit-ups because sit-ups are characterized by high hip flexor activations and subsequent high lumbar compressive loads. These loads actually exceed National Institute for Occupational Safety and Health

Figure 5C-86    Rolling side bridge.

Rehabilitation and Therapeutic Modalities

285

Figure 5C-87    Beginner’s curl-up.

guidelines.162 Also, full sit-ups (and improperly performed curl-ups) may contribute to unwanted increased thoracic kyphosis and cervical spine forces.162

Beginner’s Curl-Up Properly performing the curl-up is essential when considering spine loads.186 The initial starting position is the singleleg hook-lying position with one or two hands supporting the lumbar spine. The hands and bent leg help to support the low back while ensuring that the back is not flattened to the floor. While leaving the elbows on the floor, the head and shoulders are elevated only a short amount (Fig. 5C-87). The rotation of the curl-up is in the mid-thoracic region while the head and neck are fixed. No cervical motion should occur. The intent of the curl-up is to elicit rectus abdominis activity, not to produce spine motion.

Curl-Up against Abdominal Brace Curl-ups up while maintaining proper abdominal bracing is an advanced exercise. First brace the abdomen and then curl up against the brace. The resistance is actually provided by the prebracing of the abdominals. If the individual can seemingly perform this with ease, he or she is probably not correctly bracing. While in the contracted position, have the individual perform deep breathing. Curl-ups may ultimately be performed on unstable surfaces, such as a stability ball and balance disks, to elicit higher levels of core muscle co-contractions (Fig. 5C88).187 It must be stated that along with the increased cocontraction, spine loads also increase.162 Therefore, the practitioner should proceed with caution when introducing unstable surface training with patients who have low back pain. We believe that, too often, rehabilitation specialists are too quick to introduce these advanced exercises.

Figure 5C-88    Curl-up against abdominal brace on unstable object.

bug exercise, the rectus abdominis and abdominal obliques are highly active. Remember, these exercises should never look easy when performed while the individual is sufficiently bracing the core.

Functional Gluteal Training When studying the potential contributors to low back pain, we believe that one recurring theme is excessive movements, strains, and loads on the lumbar spine because of poor gluteal and pelvic control. As stated earlier, substituting lumbar extension for hip extension is a major culprit in training and daily function. When properly designing and implementing core training programs, gluteal training is emphasized from the onset. The individual should be trained in both bridging and quadruped positions to set the core and fire the gluteal muscles.

Neuromuscular Activation For those who are deficient in their gluteal activation and have significant hip weakness, the previously mentioned side-lying hip abduction and clam exercises are appropriate training components in the initial neuromuscular

Dead Bug The dead bug is the final step in the progression. In this case, both arms and legs are unsupported and slowly flexed and extended while maintaining an abdominal brace (Fig. 5C-89). This is a very demanding exercise that should be closely monitored for technique and pain. During the dead

Figure 5C-89    Dead bug.

286

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Functional Latissimus Dorsi and Scapula Training Again, the latissimus dorsi is a key stabilizer of the lumbar spine through its lumbodorsal fascia attachments to the lumbar spinous processes. Training the latissimus, middle and lower trapezius, and rhomboid muscles also challenges the deep stabilizers to maintain neutral lumbar postures during the various exercise movements. Training progressions include rowing and pull-down exercises in both sitting and standing postures (Fig. 5C-91). The exercises can also be advanced to be performed with single-arm variations.

Rhythmic Stabilization and Manual Perturbation Training Figure 5C-90    Squat and reach.

a­ ctivation and stabilization phase of the core training program (see Figs. 5C-34B and 5C-37).

Functional Squat The functional squat, provided in the knee section, is a critical functional movement that must be mastered by the individual (see Fig. 5C-42A). This proper hip hinge maneuver helps to spare the spine while using the large gluteal muscles for appropriate support and control of squatting motions. The squat and reach is an appropriate teaching drill to help ensure proper hip hinging and that gluteal support is occurring (Fig. 5C-90).

Single-Leg Training All the single-leg training exercises mentioned earlier are appropriate to include in a core training program. The benefits as previously espoused help to ensure that the individual is appropriately prepared to protect the lumbar spine and improve overall functional capacity. Make sure the exercises are advanced to movements that are ­multiplanar and meaningful to the individual.

A

B

Throughout the entire core training program, rhythmic stabilization techniques can be performed to help engage and reinforce abdominal bracing and whole body ­stability. Early in the rehabilitation or training program, proper quadruped, bridging, sitting, and standing postures are reinforced (Fig. 5C-92). In the dynamic and advanced phases of core training, dowel rods and stability balls are used to help provide perturbing stimuli in functional ­postures (Fig. 5C-93). Kofotolis and Kellis concluded that the inclusion of these PNF techniques increased the muscle endurance of people with chronic low back pain as well as significantly decreased back pain intensity and functional disability.188 We have found these techniques to be valuable in training and reinforcing abdominal bracing in functional postures.

Rotation Training The initial rotation work with the core is more anti-rotator and stabilizing in design. We believe that anti-rotation training is a key concept in core training. In most functional destabilizing challenges in everyday life, forces on the trunk are off center and therefore require a counteracting anti-rotation contraction. Remember, the core muscles are stabilizers first. Early in the core training outline, the bridging and quadruped stabilization exercises that are done with decreased

C

Figure 5C-91    A-C, Three-position rowing. Each position trains different scapular muscle activations and trains for lumbar stabilization with varying forces.

Rehabilitation and Therapeutic Modalities

A

287

B

Figure 5C-92    Manual perturbation training. A, Sitting. B, Standing.

point of support effectively target the core to act in an antirotator manner. And truthfully, all single-arm and singleleg exercises activate the core stabilizers and are considered to be rotation training exercises as well. These exercises can be emphasized to use the core in its primary function—the prevention of rotation. As the patient progresses to the dynamic and advanced core training phases, specific exercises can be introduced to enhance trunk rotation training. Effectively, they are either rotational stabilization exercises or rotational power activities.

Chop-and-Lift Exercises PNF training of the trunk allows for the combination of flexion and extension patterns with rotation. Chopping is basically a pattern of flexion and rotation, whereas lifting is the pattern of extension and rotation. In line with our strong belief that individuals must be able to sufficiently prevent rotation before we allow them to produce it, we first initiate these exercises through the arms with a stable trunk. Once the individual is proficient with stable trunk

A

rotation, more complete whole body chop-and-lift patterns can be introduced. However, spine rotation forces should always be minimized as much as possible. Instead, rely on the upper trunk, shoulders, and lower extremity to produce the power and forces around the stable spine (Fig. 5C-94).

Pushing and Pulling Exercises with Rotation Ultimately, the pulling and pushing exercises can include rotational components and single-leg variations to train those who require such advanced programming (Fig. 5C-95).

Rotational Plyometric Training Medicine ball training is essential to provide a powerful and functional torso, especially for the athletic population. Medicine ball training helps to convert the previous stabilization and strength training to useable power. The focus of these exercises is to provide velocity of movement with a stable base of support. When considering proper medicine

B

Figure 5C-93    Dynamic perturbation training. A, Dowel rod. B, Stability ball.

288

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Advanced Functional Training Ultimately, therapeutic exercises all can be advanced to include dynamic, unstable, and functional movements. With thoughts on core training, the core always is involved in these advanced modes of training. Whether or not it is the primary objective of the training, proper abdominal bracing and spine protection measures should be emphasized. If the individual’s primary objective of core training is for the rehabilitation of a low back disorder, specific advanced exercises should be introduced in accordance with his or her needs.

Summary Figure 5C-94    Proprioceptive neuromuscular facilitation trunk lift exercise.

ball weight, the individual should not struggle to throw the ball. When in doubt, a lighter ball should be used. Medicine ball training is best performed with throws against a wall. The front twist throw and side twist throw (Fig. 5C-96) are best for beginning torso plyometric tosses. The individual is encouraged to throw from the lower extremities and hips and eventually through the arms and hands. This ground-up approach decreases the likelihood that the throws will be initiated and performed primarily through the lumbar spine. A lumbar spine throw would obviously impart tremendous loads and torques on the spinal segments. The medicine ball tosses can be advanced to include more sport-specific movements and tolerated by the individual. Progress to three sets of 10 throws on each. Do not attempt to increase the number of throws. Instead, focus on throwing harder with more trunk control. After 3 to 4 weeks of training, increasing the weight of the ball may be considered. Be careful not to increase resistance at the expense of decreased velocity and form.

A

Comprehensive and thorough core training effectively enhances the stability of the spine and pelvis and aids in the transfer of energy from the torso to the extremities during various daily activities and sports movements. We believe that active control of spine and trunk can be achieved through the co-contraction of the deep and global ­stabilizing muscles to stiffen the lumbar spine and increase its ­stability. The multifaceted approach to core training that was offered emphasizes core and whole body stabilization while training mobility at the hips and lower extremities with minimal spine loads. The key to successful and effective core training is to sequence the training appropriately to develop core strength and stability in progressively more dynamic, functional, and real-world environments (Table 5C-10).

STRETCHING TECHNIQUES FOR PATHOLOGY Stability and muscle recruitment are extremely important when restoring function to the injured athlete. Equally important is restoring normal muscle length. Shortened muscle can be a result of several factors. Muscle injury can lead to interstitial muscle scarring and resultant loss of normal length. Joint contracture or adhesions can also lead

B

Figure 5C-95    A, Dynamic rowing (“pulling”) exercise with rotation. B, Dynamic punch (“pushing”) with rotation.

Rehabilitation and Therapeutic Modalities

289

B

A

Figure 5C-96    Medicine ball plyometric rotational training. A, Front twist throw. B, Side twist throw.

to muscle tightness. There are several conditions on which abnormal muscle length can have an adverse effect. Abnormal tightness of the hamstrings is thought to be detrimental to athletes suffering from PF pain. This is thought to be caused by the increased force of the quadriceps required during knee extension to overcome the hamstring resistance. Another result of hamstring tightness is the inability to reach full extension during the gait cycle. If the athlete continues to assume a flexed knee posture, resultant PF pain often ensues. There are two hamstring stretches that are effective in our clinical practice, as shown in Figure 5C-97. The first technique places the individual on the floor, with one leg through the doorway and

the stretching leg against the wall. This position is especially effective for individuals with low back pain who have ­difficulty in the traditional hamstring stretch position by flexing at the hip. The goal of this exercise is to be able to rest the opposite leg completely flat on the floor, the stretch leg completely flat against the wall, and the buttocks touching the wall. After this position is achieved, a towel can be placed over the foot to pull the leg away from the wall, accentuating the stretch. We recommend that the stretch be held for 30 seconds, performing five repetitions. The other stretch we teach to our patient population is sitting with one leg outstretched and the other off the table or plinth. Technique is extremely important in that ­lumbar

   TABLE 5C-10  Core Training Outline Category Bridging

Quadruped: (1) dowel parallel, (2) dowel perpendicular Lateral flexion Flexion Functional gluteal Functional latissimus, trapezius: (1) seated, (2) standing Rhythmic stabilization

Rotation Advanced functional training

Neuromuscular Activation and ­ tabilization S

Dynamic Stabilization

Advanced Stabilization and Strength Training

Draw-in, abdominal brace Bridge Unilateral bridge Bridge with marching Draw-in, abdominal brace Arm elevation Hip extension (bent leg, straight leg) Standing side push-up Bent-leg side bridge Draw-in, abdominal brace Beginner’s curl-up

Cook hip lift Bridge with knee extension

Cook hip lift elevated, unstable Bridge with knee extension elevated, unstable

Bird dog Bird dog sweeping

Bird dog geometric Bird dog unstable

Side bridge

Side bridge rolling

Curl-up against abdominal brace

Side-lying clam Side-lying hip abduction Rows Pull-downs Extension Sitting Standing Bridges Quadruped Unilateral bridges Quadruped

Functional squat Squat and reach Single-arm or alternating extensions Single-arm or alternating rows

Curl-up against abdominal brace—unstable surface Dead bug Single-leg training Multiplanar single-leg training Single-arm dead bug rows Rows with rotation

Standing Dowel rod Stability ball

Functional postures

Bird dog Side bridges

Chop and lift Plyometric training Per individual needs

290

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 5C-97    A, Seated hamstring stretch: maintain lumbar lordosis, reach toward toe, hold for 30 seconds, do 5 repetitions.   B, Wall stretch—advancement: use towel around the foot and pull away from wall, hold for 30 seconds, do 5 repetitions.

A

B

C Figure 5C-98    A, Genie stretch: control rotation as horizontal adduction force is applied, pulling elbow across chest. B, Traditional horizontal adduction stretch: rotation is not controlled. C, Sleeper stretch with contract-relax technique: manually resist external rotation, then relax and apply overpressure into internal rotation. Repeat.

Rehabilitation and Therapeutic Modalities

A

291

B

Figure 5C-99    A, Gastrocnemius stretch: heel on ground, knee straight, lean forward to feel stretch and hold for 30 seconds, do 5 repetitions. B, Soleus stretch: knee in slight flexion, heel on ground, lean forward and hold for 30 seconds, do 5 repetitions.

lordosis is maintained as the subject reaches toward the toe. This technique removes the lumbar flexion that might contribute to the reach during this stretch. Another area in which specific stretching can greatly improve pathology is the shoulder. Posterior and internal impingement of the shoulder has been described by several authors. This condition is thought to be exacerbated by a tight posterior cuff (infraspinatus, teres minor) and is common in the throwing athlete but can also be present in the normal population. When the tightness is present, the humeral head is no longer resting in the inferior position but is placed to a position that is superior and posteriorly oriented. As the athlete comes to a fully externally rotated position, the posterior cuff is pinched on the posterior rim of the glenoid. This results in pain in the back of the shoulder. An excellent method of stretching this tight structure has been developed by Paine (R. Paine, Roger Clemens Institute, Memorial Herman, Houston, Texas, personal communication, 2008). Paine describes this stretch as the “genie” stretch. In this position, the patient starts with one arm over the other as a genie would be granting a wish. Horizontal adduction stretching is then performed by pulling the elbow across the chest. The normal elbow rests above the involved hand as the elbow is pulled across the chest. This allows control of rotation during the stretch and gives a more localized and focused stretch to the infraspinatus–teres minor muscle complex (Fig. 5C-98A). This stretch differs from the traditional posterior cuff stretch (see Fig. 5C-98B). During the traditional posterior cuff stretch, rotation is not controlled. The genie stretch is held for 10 seconds, with 10 repetitions performed. Another popular stretch for increasing the length of the posterior cuff is the sleeper stretch. This stretch is performed in the side-lying position. The subject slides the arm to 90 degrees abduction and applies an internal rotation force. This stretch can be painful. To combat the reflex activation of the antagonist muscle due to pain, a PNF self contract-relax technique is performed to gain internal rotation with much less discomfort (see Fig. 5C-98C). The calf is another area for which stretching can be effective at preventing and treating injuries. The use of a

slant board is very effective when performing calf stretches. These are performed with the heel on the ground and with the knee in two positions: slightly flexed and straight. Tensecond holding is performed in each position, with 10 repetitions each (Fig. 5C-99). The iliotibial tract is a structure that is often associated with running injuries. The distal insertion of the iliotibial tract may become irritated with long-distance running and can present with acute pain over the distal insertion. Many stretching techniques have been described to treat this malady, but we have found the foam roller to be a more effective tool than stretching. The individual attempts to relieve the tension in the iliotibial band by rolling the iliotibial band along the foam roller while lying on the side on the floor (Fig. 5C-100). This exercise may be painful, but an important technical point is to remove the body weight from the exercise by lifting with the weight-­bearing arm. The athlete is to work the entire length of the iliotibial band from the hip to the knee. If a tight area is encountered, the athlete is instructed to spend a few oscillations with the foam roller to help “knead” this broad taught structure.

Figure 5C-100    Foam roller used for iliotibial band restriction.

292

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5C-101    ERMI Flexionator is used to increase flexion. A hydraulic system is used to control flexion moments produced to the knee. The patient is in control.

As we age, we lose the elastin content in our muscles. This may be a precursor to stiffness of the muscle and can be overcome with stretching techniques. Maintaining a normal muscle length allows the muscle to operate efficiently and maintain the normal length-tension relationship. This allows the elastic and nonelastic components of the muscle to share the load of an external force. If the elastic component is in an abnormally shortened state, the nonelastic component (muscle sheath) may be injured with a rapid force. Many factors are related to muscle physiology, but their discussion is beyond the scope of this chapter. Stretching of adhesions and contractures is another area in which the sports medicine rehabilitation specialist can have a great impact on patient recovery. Knee flexion and extension contractures are common maladies that are

A

Figure 5C-102    ERMI Extensionator is used to reverse flexion-contractures.

often present with combined knee ligament injuries that have scarring and immobilization associated with recovery. The ERMI devices (ERMI, Inc., Atlanta, Georgia) have been developed to overcome the lack of motion that is not resolved with traditional physical therapy measures. These devices use hydraulic pressure to apply carefully controlled but forceful stretching to the scar tissue that has been developing in the knee. The Flexionator (Fig. 5C-101) is used to aid in increasing knee flexion. We have found this device to be superior to manual mobilizations and forcing of the knee by the physical therapist. The advantage of these devices is that they place the patient in control of the amount and time that the stretch is applied to the joint. The ERMI Extensionator applies a forceful load to the knee using a blood pressure cuff. The heel is lifted by a cushioned platform, allowing the overpressure to be applied above the knee to the anterior thigh (Fig. 5C-102).

B

Figure 5C-103    A, Prone stretching using biofeedback (B) over the hamstring allows relaxation of the hamstring in the prone position to focus on lengthening the tight posterior capsule. Stretching is done for 10 minutes. Small (2- to 5-lb) ankle weights are used when the patient is able to tolerate and maintain hamstring relaxation.

Rehabilitation and Therapeutic Modalities

Guidelines vary, but we recommend 30-­second holds at the extremes of motion with a rest period of 15 seconds between each set. This is repeated for 10 minutes, 4 times per day. Another technique that has shown clinical success is the prone hang using biofeedback with the patella off the end of a plinth or table. Biofeedback is used over the hamstrings to promote relaxation as the knee is stretched into extension (Fig. 5C-103). It is very common for the hamstrings to contract because of the pain caused when the posterior scarring is stretched. Unless the hamstrings are relaxed, prone hangs are ineffective at achieving full extension in the knee. After the subject has learned to maintain this position for 10 minutes with no hamstring activity, a 2pound weight may be applied to help force extension. Biofeedback must continue to be used to monitor any increase in hamstring activity. This section describes several specific stretching techniques associated with very common sports medicine ­injuries. Proper stretching technique application results in much faster and effective results.

THERAPEUTIC EXERCISE SUMMARY This chapter section has detailed comprehensive therapeutic applications for the typical musculoskeletal systems seen in orthopaedic and sports medicine facilities. Therapeutic exercises are invaluable tools that should be used to help successfully correct impairments in motor function and muscle performance. The rehabilitation specialist has been provided with numerous practical exercises and training tips that can easily be applied to most individuals treated in the clinical setting.

Acknowledgment I would like to thank Russ Paine for his special contribution to this chapter in his addition of the section on Stretching Techniques for Pathology. I would also like to acknowledge my deep appreciation for his mentorship, encouragement, and friendship throughout my career. He has provided me with countless opportunities for growth as a physical therapist and to be involved in special endeavors such as this one. Thanks again.

C

r i t i c a l

P

293

o i n t s

l From

the information first gathered in a thorough clinical evaluation, the treating clinician should initially seek to activate inhibited or weak muscles through the application of evidence-based exercises that have been proven to target the affected muscles. The clinician is cautioned to avoid overstressing the healing tissue at all times. l As the individual sufficiently gains neuromuscular activation and as symptoms allow, therapeutic exercises are ­initiated that help the individual integrate the affected ­musculature into appropriate movement patterns. l T  he individual should be advanced to performing more dynamic and goal-specific tasks. Skill, power, and endurance characteristics should be addressed in accordance with the needs of the individual. The clinician should never forget that the ultimate goal of rehabilitation efforts is to effectively restore the optimal movement patterns for which the body was designed. l Although numerous therapeutic exercises have been presented in this chapter, the rehabilitation specialist should be reminded to first master a certain percentage of these evidence-based training ideas before attempting to apply too much variety and “spice” into a rehabilitation and training program. Also remember, the clinician would be wise to avoid introducing exercises that he or she is unable to teach. l As a final note, therapeutic exercise application is an everevolving piece in the rehabilitation process. Keeping abreast of current evidenced-based research is a must, and efforts to further advance clinical knowledge should never cease.

S U G G E S T E D

R E A D I N G S

Baechle TR, Earle RW (eds): Essentials of Strength Training and Conditioning, 2nd ed. Champaign, Ill, Human Kinetics, 2000. Boyle M: Designing Strength Training Programs and Facilities. Available at: http:// www.michaelboyle.biz. Kisner C, Colby LA: Therapeutic Exercise: Foundations and Techniques, 5th ed. Philadelphia, FA Davis, 2007. McGill S: Ultimate Back Fitness and Performance, 2nd ed. Waterloo, Ontario, Canada, Backfitpro, 2006. Verstegen M, Williams P: Core Performance. Emmaus, Penn, Rodale, 2004.

R eferences Please see www.expertconsult.com

294

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n

D

Proprioception and Joint Dysfunction Russ Paine

PROPRIOCEPTION OF THE KNEE Anterior cruciate ligament (ACL) injury results in functional instability. This is due to the increased translation of the tibia on the femur as well as rotatory instability of the knee. Stability of the knee is also dependent on awareness of joint position in space.1 Proprioception is the ability to perceive a joint’s position even without a visual reference. Nerve endings provide sensory feedback to the central nervous system that allows us to reference joint position. These nerve endings, called proprioceptors, are located in the joint capsule, cruciate ligaments, and menisci.2 Without this feedback loop, success of ligament reconstruction is diminished. In the 1980s with the advent of ACL reconstruction, rehabilitation efforts were focused on gaining range of motion and strength. At that time in sports medicine history, we were unaware that proprioception should be addressed during rehabilitation. Common complaints after knee ligament reconstructions during the 1980s were that the patient felt fatigue and the inability to cut and jump as well as the preinjury level. Most likely, this was due to inappropriate functional rehabilitation techniques. Kennedy demonstrated that proprioceptors were present in the knee, and this discovery allowed us to change our rehabilitation strategy to include balance and coordination exercises into the regimen of postoperative care.2 Many authors have studied the diminished proprioceptive function that is present after ACL injury.3,4 All conclude that there is in fact a deficit that is present in the knee. This is probably due to the lack of feedback information that the ACL was providing.5 Delayed hamstring firing could be a result of this lack of feedback loop. Capsular tension may also provide feedback to the central nervous system through proprioceptors present. When the capsule is injured during a knee injury, we may lose the normal capsular tension that is disrupted with an ACL injury.6 The challenge of rehabilitation specialists is to stimulate other receptors of the surrounding area to overtake the loss of the feedback loop provided by injured proprioceptors. This is done by beginning to stimulate the proprioceptive system as early as possible after ACL reconstruction. Joint position sense involves several systems. The auditory, visual, tactile, and proprioceptive systems all contribute to joint position sense. To overload the damaged proprioceptive system, one of the normal systems can be removed. An example of this could be removing the auditory input when performing a ball toss to a patient (Fig. 5D-1). As the patient accepts the pass, he or she no longer hears the auditory input of the ball leaving the thrower’s hand or the sound of the ball hitting the hand as it is caught. This results in a more concerted effort to overload

the proprioceptive system in hopes of recovering a degree of lost sensory feedback. Proprioceptive input may begin early after knee injury or surgery. ACL-reconstructed patients still maintain a degree of proprioceptive loss after the reconstruction. Even though capsular and ligament tension are restored, the feedback loop remains deficient. It is the job of the clinical rehabilitation specialist to safely overload the injured neuromuscular system to help regain function through improved proprioceptive control. It is unknown how proprioceptive loss is regained, but improved dynamic function by increased muscle control may be a factor. Proprioceptive nerve endings do not return to the graft material used to reconstruct the knee. Although the ACL and capsule may have not been injured, there still exists a proprioceptive deficit in the postoperative knee. This is due to the hemarthrosis and swelling present in the knee. The initial proprioceptive exercise performed after ACL reconstruction of the knee is instituted on postoperative day 2. This exercise involves simple weight shifting from normal to involved side and back again. The patient is asked to assume a standing position supported by the hands by leaning on a table or plinth. The knees are maintained in about 30 degrees of flexion. The patient then shifts the weight from side to side so that the involved knee accepts the full body weight alone, supported by the upper extremity. This activity is performed for 5 minutes. It is important to ­maintain

Figure 5D-1    Plyoball toss to stimulate balance and coordination. Removing the auditory input by use of headphones could help overload the proprioceptive system to enhance recovery.

Rehabilitation and Therapeutic Modalities

Figure 5D-2    Side-to-side weight shifting.

a slightly bent position of the knee so that the quadriceps is made to fire to support the body weight (Fig. 5D-2). Flexed knee ambulation is an abnormal postoperative gait that many patients assume. This may be due to lack of eccentric quadriceps control during the single-limb ­balance of the weight-bearing extremity during the swing phase of the gait cycle. The specific lack of control of the quadriceps occurs between the beginning of midstance through beginning of push-off phase of the gait cycle (Fig. 5D-3). It is during this phase that the quadriceps must eccentrically control the knee flexion moment that is created during this phase of the gait cycle. It is critical that this eccentric control be established early in the rehabilitation process. This is helpful in maintaining a normal gait pattern that will reduce altered neuromuscular modulation of the muscles that allow us to propel. If this is not addressed early in rehabilitation, patients will continue to assume this flexed knee posture because this position provides a stable base of support during the stance phase owing to the lack of need for muscle activity at

Phases

Periods % Cycle

30 degrees of knee flexion. Another problem with flexed knee posture is increased patellofemoral (PF) compression. By maintaining this posture during gait, there is constant pressure to the PF joint that could cause undue stress to the articular cartilage and soft tissue structures that support it. Another gait control activity is cone ambulation. This can be instituted 2 or 3 days after ACL reconstruction (Fig. 5D-4). The patient is required to use crutches for this activity. Of course, patients who have weight-bearing restrictions due to associated knee injuries should not perform this activity. Cone ambulation begins by placing 4 to 6 Styrofoam coffee cups on the ground about 18 inches apart. The subject is asked to step over the cone (the cup) with the involved extremity and place the foot between the next two cones. It is important that the rehabilitation specialist instruct the patient to maintain a slightly flexed position at the knee during this activity (Fig. 5D-5). The patient balances on the involved extremity alone, with crutches for 5 seconds, then proceeds to the next cone. We recommend four cycles of cone ambulation—up and back completes one cycle. Additional training can be performed by turning 90 degrees and performing lateral cone ambulation. As the knee flexion and quadriceps control improve, cones may be used to increase the challenge of gait cycle. Another important cue to maintenance of a normal gait pattern is to instruct patients to flex the knee as they begin lift-off of the weight-bearing extremity (preswing phase of gait—Fig. 5D-3). Criteria for discarding crutches are twofold. First, the patient must be able to ambulate with a normal gait with crutches. Second, the patient must have no sensation of giving way of the knee. Too often patients have crutches removed before achieving a normal gait pattern in hopes that this will improve their strength. We have found the opposite to be true. Progression should be from two crutches to one crutch to no crutch. Proper gait training is the first phase of forcing the proprioceptive system to become engaged. Continued proprioceptive input may be enhanced by use of specialized equipment. The MR Systems (Haarlem, The Netherlands) Functional Squat provides stimulation of joint position sense along with strength training. This

Stance Phase

Initial Loading Contact Response 0%

Midstance

12%

295

Swing Phase

Terminal Stance 50%

PreSwing

Initial Swing 62%

MidSwing

Terminal Swing 100%

Figure 5D-3    Perry’s five phases of the gait cycle. (From Perry J: Gait analysis: Normal and pathologic function. New York, McGraw-Hill, 1992.)

296

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5D-4    Emphasize knee flexion during gait.

system allows the clinician to preset the range of motion to be used during the activity. The patient then controls the cursor position on a computer screen by performing a leg press. As he or she extends the knee the cursor moves horizontally on the screen. Six computer games may be performed to stimulate proprioceptive input. Accuracy is measured by calculation of error, and a score is given at the end of each exercise session. A typical session includes 60-second bouts of exercise and six sets (Fig. 5D-6). Continued proprioceptive training evolves as lower extremity strength is increased. Functional strengthening continues with lunging activities using weighted balls as shown in Figure 5D-7. Added proprioceptive stimulation can be done using a foam cushion, as shown.

Figure 5D-5    Cone ambulation: balance on anterior cruciate ligament of the knee for 5 seconds.

Cone reaching is another method of providing ­ roprioceptive training. This requires the patient to reach p across the midline and control balance and coordination while performing a single-leg squatting maneuver (Fig. 5D-8). The sports cord is another proprioceptive control exercise. This device is attached at the waist and provides ­resistance that must be overcome to control balance. Cones may be used as stepping obstacles during this activity (Fig. 5D-9). Double-leg jumping is performed at the point in the rehabilitation protocol when running is allowed. Figure 5D-10 demonstrates initial jumping. The subject is asked to jump laterally across a jump rope and to try to land with both feet. It is normal for the uninvolved leg to touch the ground first until proprioceptive control is achieved. As the patient gains control, double-leg jumping proceeds to single-leg hopping. Control of valgus is emphasized (Fig. 5D-11). Functional bracing may afford protection to the healing graft at low loads.7 We are advocates of functional knee braces, especially during this transition period of rehabilitation when muscle strength and proprioceptive input are not fully restored. Proprioceptive control is a necessary precursor to plyometric training. If sufficient strength is not achieved before plyometric training, overload of the musculotendinous tissue will most likely occur and result in tendonitis. At this point during rehabilitation, plyometric training can proceed, followed by functional training and sport-specific exercises.

PROPRIOCEPTION OF THE SHOULDER Damage to shoulder stability can cause a lack of feedback information, resulting in the inability to properly position the shoulder during activities. Smith and Brunolli have observed deficits in shoulder kinesthesia and joint position sense in subjects with recurrent anterior shoulder instability.8 Lephart and coworkers also demonstrated proprioceptive deficits in the pathologic shoulder compared with the contralateral normal shoulder.9 Use of reflexive-mediated activities such as plyoball tossing is based on the recent findings of Guanche and coworkers, who observed three different articular branches of the axillary nerve innervating the shoulder capsule that provide a primary reflex arc to the biceps, supraspinatus, infraspinatus, deltoid, and subscapularis muscles in a feline model.10 Providing cocontraction through neuromuscular reactive exercises may help promote dynamic stability in the unstable shoulder. Restoring normal capsular tension through surgical intervention may help with enhancing the proprioceptive feedback loop by tightening the capsule that has been loosened by a shoulder dislocation or atraumatic instability. Techniques to provide shoulder proprioceptive training include wall perturbations. Figure 5D-12 demonstrates this technique. The subject holds a small plyoball against the wall at 90 degrees in the scapular plane. The rehabilitation specialist then applies quick external perturbations as the athlete tries to maintain the original position of the ball.

Rehabilitation and Therapeutic Modalities

A

297

B

Figure 5D-6    A, MR Systems Functional Squat. B, Computer screen image of the MR Systems Functional Squat.

A

B

Figure 5D-7    A and B, Functional training in preparation for plyometrics.

Figure 5D-8    Cone reaching enhances the proprioceptive control by not allowing the subject to look downward. Emphasize a straight-ahead gaze.

Figure 5D-9    Sports cord lateral lunges and cones force the patient to control balance using resistance.

298

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5D-12    Plyoball closed chain perturbations Figure 5D-10    Double-leg jumping to begin transition to plyometrics.

Another proprioceptive activity for the shoulder is the wall bounce using a small plyoball. Figure 5D-13 demonstrates the wall bounce. The athlete makes a semicircular path on the wall to stimulate the proprioceptive system. Use of a Rebounder is an effective beginning proprioceptive training tool in the shoulder. Figure 5D-14 shows the 90/90 position of the shoulder as the throwing athlete assumes the functional position and tosses the ball into the Rebounder. Additional proprioceptive training is achieved using the MR Systems Cable Column (Fig. 5D-15). A ­modified ­position is shown to focus on posterior rotator cuff strengthening. The patient is required to position the cursor on the computer screen by performing eccentric control using the external rotators of the shoulder. This promotes both strengthening and joint position sense simultaneously. Proprioceptive training of the shoulder is important in achieving maximal recovery after shoulder injury. This type of training is especially important when instability of the shoulder is present that produces a lack of the normal tension of the capsular ligaments. Although tension may be

Figure 5D-11    Transition to single-leg hopping.

restored through surgical reconstruction, proprioceptive training should be performed to stimulate nerve endings and promote quicker response time in muscle activation.

PROPRIOCEPTION OF THE ANKLE Ankle stability is dependent on both static and dynamic components. Instability of the ankle can be greatly improved by proprioceptive training. Loss of the lateral ligament structures of the ankle increases the propensity for inversion ankle sprains. Improving dynamic stability of the ankle through proprioceptive training can often overcome the instability and allow the athlete to resume near preinjury levels. Konradsen and Ravn showed that chronic ankle instability may result in reduced peroneal muscle reaction time.11 Forcing patients with ankle instability into single-limb balance while performing balance and coordination activities provides proprioceptive training to improve the reaction time of dynamic stabilizing muscles of the ankle (see Fig. 5D-1). Ankle proprioceptive training is performed using exercises similar to the aforementioned knee exercises, with special emphasis on the ankle joint.

Figure 5D-13    Wall bounce: small, quick bouncing of a small plyoball in a circular motion around the wall.

Rehabilitation and Therapeutic Modalities

A

299

B

Figure 5D-14    The 90/90 position tossing a 2-pound plyoball into the Rebounder.

C

r i t i c a l

P

o i n t s

l ACL

Figure 5D-15    Proprioceptive training using the MR Systems Cable Column.

tears result in instability and decreased joint position sense. l Proprioceptive training after ACL injury/reconstruction should begin on the second postoperative day. l Early institution of eccentric quadriceps control is essential for achieving a normal gait pattern in early ­rehabilitation. l Cone ambulation allows forced single-limb balance to begin the proprioceptive training regimen. l Progression of the rehabilitation program should force the patient to perform single-limb activities. Conquering the last hurdle to achieve normal quadriceps strength is greatly eased by instituting single-limb jumping ­activities. l Although not as dominating to functional performance, shoulder proprioception is a side-effect of shoulder injury and should be addressed during rehabilitation. l Successful ankle rehabilitation is heavily dependent on balance and coordination exercises to stimulate proprioceptive control.

S U G G E S T E D

SUMMARY Muscle strengthening has been proved to be a mainstay of rehabilitating sports medicine injuries of the upper and lower extremities. Muscle firing patterns and reflex reaction have recently been shown to enhance recovery. Proprioceptive training stimulates damaged receptors in hopes of recovering a greater degree of dynamic ­stability that may have been lost because of ligament and capsular injury. When designing a rehabilitation program, proprioceptive training must be combined with muscle ­ reeducation and strengthening to achieve the ultimate outcome in ­recovery.

R E A D I N G S

Fremerey R: Proprioception after rehabilitation and reconstruction in knees with deficiency of the anterior cruciate ligament: A prospective longitudinal study. J Bone Joint Surg Br 82:801-806, 2000. Johansson H, Sjölander P, Sojka P: Receptors in the knee joint ligaments and their role in the biomechanics of the joint. Crit Rev Biomed Eng 18:341-368, 1991. Kennedy J, Alexander IJ, Hayes KC: Nerve supply of the human knee and its functional importance. Am J Sports Med 10:329-335, 1982. Konradsen L, Ravn JB: Ankle instability caused by prolonged peroneal reaction time. Acta Orthop Scand 61:388-390, 1990. Lephart SM, Warner JP, Borsa PA, et al: Proprioception of the shoulder in normal, unstable and post-surgical individuals. J Shoulder Elbow Surg 3:371-380, 1994.

R eferences Please see www.expertconsult.com

300

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n

E

Return-to-Sport Plyometric Training in the   Rehabilitation of Athletes Following Anterior Cruciate Ligament Reconstruction Gregory D. Myer, Mark V. Paterno, Kevin R. Ford, and Timothy E. Hewett

Advances in fixation methods and other graft reconstruction techniques have dramatically improved surgical success with anterior cruciate ligament (ACL) reconstruction.1-3 These advances in surgical technique have resulted in consistent, quality surgical outcomes. The increased consistency in surgical outcomes may have shifted the limiting factor for an athlete to return to his or her prior level of sports participation to differences in postsurgical rehabilitation rather than variance in surgical outcomes.3-5 Traditional ACL rehabilitation that once included prolonged immobilization, non–weight-bearing, and slow progression to activity now emphasizes immediate motion, early weight-bearing, and accelerated return to sports participation for athletic patients. Compared with past protocols, ­rehabilitation ­ programs are now more aggressive and advocate the release of athletes to sports activities as early as 8 weeks after surgery.6,7 An athlete’s return to play is often dictated by graft stability (anteroposterior tibiofemoral motion), patient confidence, postsurgical timeline, and subjective medical team opinion. Appropriate objective criteria that consider both graft strength and objective functional criteria to determine advancement through the end stages of rehabilitation and readiness to return to sport after ACL reconstruction remain elusive.4 Rehabilitation and ultimate return to sport using objective tests that quantitatively measure functional ability may increase athlete reintegration to sports at the same competitive level as before the injury. Rehabilitation after ACL reconstruction is commonly divided into early (immediate postoperative and subacute strengthening) and late rehabilitation (functional progression and return-to-sport) phases, with specific goals and time since surgery as determinants for phase progression. Early phases of postoperative ACL reconstruction often use stringent, criteria-based guidelines for range of motion (ROM), progression to full weight-bearing, and exercise selection. In contrast, the final phases of rehabilitation prescriptions are typically more broad, with general categorizations of appropriate exercises and progressions, without specific milestones for when it is safe to introduce highrisk and high–joint-loading activities.6,8,9 In addition, more conservative therapeutic approaches may limit progression to later stages of rehabilitation and possibly delay successful return to sport.

Exercise prescription for an athlete’s progression through rehabilitation and back to sport participation should avoid overstressing the graft in athletes who do not possess the strength and functional abilities necessary to protect the healing joint while undertaking high–jointloading activities. Structurally, animal studies indicate that the graft’s strength may reach its weakest point about 6 to 8 weeks after surgery10 and may only reach failure loads between 11% and 50% of the native ACL 1 year after surgery.11 Controlled loading may enhance ligament and tendon healing,12,13 whereas excessive loading can potentially damage the healing graft and lead to increased anteroposterior knee laxity.14 Graft healing properties have been studied primarily in animal models, including rabbits,11 canines,14,15 and primates.15 These types of animal models present important information related to histologic properties, stiffness, and ultimate load to failure intermittently over a 1-year span in mammals. However, the properties of a healing graft in animal models may be limited in their generalizability to outcomes in the human ACL.11,14,15 The limited data in humans makes the determination of the optimal load to place on the healing ACL reconstruction, as well as the optimal timing to place that load, difficult to determine.16 It is possible to return to pivoting, twisting, and rotational sports as early as 3 to 4 months after surgery6,7; however, this early return to sport may not be safe for athletes who do not have sufficient functional stability to protect the weakened, healing graft. Healing ACL grafts may be better protected if more aggressive postoperative ACL reconstruction rehabilitation protocols use objective measures of functional status to drive rehabilitation progression. Progression should be based on variables that determine functional stability and neuromuscular control. This may improve successful early (2 to 3 months) return to sport and good long-term outcomes.17 Concomitant with decreased biomechanical strength of the ACL graft relative to the native ligament, athletes may demonstrate decreased muscular strength, joint position sense, postural stability, and force attenuation (significant limb-to-limb landing ground reaction force differences during bilateral tasks) for 6 months to 2 years after reconstruction.18-22 Deficits evident in the early stages of rehabilitation (unique to the patient and possibly to the graft type), if left unaddressed, will likely persist beyond the late rehabilitative stages.20,23 Ongoing biomechanical

Rehabilitation and Therapeutic Modalities

deficits, which contribute to neuromuscular performance during competitive sport, may limit dynamic support and may compromise the already weakened graft. This may increase the risk for ipsilateral ACL injury in the first year after.24,25 In addition to reduced graft strength and altered functional joint control, there are other factors that make late-phase ACL rehabilitation a high-risk period for the athlete. During this phase of rehabilitation, clinicians must be especially cognizant of the potential gap between the athlete’s perceived versus actual sports readiness because subjective scores often do not correlate to quantified function and strength scores in patients with ACL injuries and reconstructions.26-28 Without objective measures that identify potential deficits, it may be difficult for therapists to justify sport restriction and the associated limitations as well as to address additional physical areas of concern. Specific progressive guidelines, based on objective measures, can provide a goal-oriented rehabilitation process that may be an appealing approach for athletes.3 The presented objective criteria offer the clinician an example of a standardized, criteria-driven progression through the return-to-sport phase of rehabilitation for athletes after ACL reconstruction. The outlined progression has yet to be validated; however, both documented and empirical evidence is provided for each component, and the clinical rationale for the algorithm is outlined. The authors acknowledge that further validation is needed to formalize the use of our criteria-driven algorithm into the mainstream clinic, but we desire to use the current clinical commentary to initiate critical evaluation of our current practice.29

CRITERIA FOR PROGRESSION INTO THE RETURN-TO-SPORT PHASE Our return-to-sport neuromuscular training incorporates a progression through specific criteria designed to provide structure and objective standardization to latephase rehabilitation after ACL reconstruction. Figure 5E-1 presents an algorithmic flow chart used to track the athlete’s progress through the late rehabilitation stages. Before initiation of return-to-sport training, our recommendation is that the patient meets the following minimal baseline criteria29: 1. Minimal International Knee Documentation Committee (IKDC) knee subjective rating form score of 70 2. Either no postsurgical history of giving way or a negative pivot shift 3. A minimal baseline-level strength knee extension peak torque/body mass at least 40% (male) and 30% (female) at 300 degrees per second and 60% (male) and 50% (female) at 180 degrees per second Carefully documented and validated subjective assessment of the patient’s ability to progress in rehabilitation may be a key factor for determination of an athlete’s readiness to enter a return-to-sport program. The IKDC knee subjective rating form is a reliable and valid tool

301

for the determination of a patient’s rating of their knee symptoms, function, and ability to participate in sport following knee injury, specifically ACL injury.30 The constructs validated for the IKDC were swelling, pain level, and functional ability. Initial scoring of at least 70 on the IKDC knee subjective rating form on the involved limb is one of the requirements for our athletes after ACL reconstruction to enter return-to-sport training. An IKDC ­rating of 69 or more would put athletes within 1 standard deviation of a population-based average for males and females aged 18 to 24 years (1079 limbs).31 Athletes with increased functional abilities may achieve an IKDC rating of 70 or greater and be prepared to progress into the return-to-sport phase more rapidly (2 to 4 months). An IKDC knee rating below 70 may indicate that an athlete is in need of additional recovery time from postsurgical trauma and improvement in their functional status before beginning return-to-sport training. Incorporation of a validated subjective knee rating system like the IKDC may bridge the gap between patient perceived function and objectively measured function to enhance progress through the proposed algorithm in the return-to-sport phase of rehabilitation. Functional stability, or the ability to avoid giving way of the knee using dynamic muscular restraints, protects the healing graft after ACL reconstruction. Although mechanical stability may be restored by surgical reconstruction, the patient may continue to experience functional instability (giving-way episodes or perceived instability) or functional impairments.32-34 Return to high level sports is a high-risk time period for athletes during the first year after reconstruction.24,25,35 Return of a patient to high-level sports before functional stability is achieved may increase the potential for poor outcome. In addition, inadequate functional stability may be related to decreased confidence in the injured knee and to decreased ability to return to preinjury sports participation.5 The inability of the patient to develop dynamic muscular joint stabilization through neuromuscular control during walking and activities of daily living should exclude the patient from progression into an aggressive return-to-sport rehabilitation phase.36,37 Therefore, the athlete should have no giving-way episodes before entering the return-to-sport phase. However, a ­giving-way episode may represent a deficiency in active (neuromuscular restraint) or passive (static restraint) stability, or a combination of both. A positive pivot shift indicates mechanical instability and is related to subjective reports of poor functional outcome.38,39 A patient who reports a previous history of giving-way episodes, and a negative pivot shift, likely possesses sufficient mechanical stability to safely enter into advanced rehabilitation exercises designed to address the functional instability.40 Measurable functional deficits that may relate to past giving-way episodes may be correctable if the athlete participates in safe and progressive return-to-sport training. Before initiation of return-to-sport training, the athlete should demonstrate a sufficient baseline of strength to improve potential for success.41 The absence of sufficient strength may result in an inability to initiate dynamic movements, attenuate ground reaction forces, or achieve high levels of performance during dynamic tasks.41,42 Normative ranges for postpubescent adolescents and adults for

302

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� ACL-R Patient

Postsurgical Therapy

Entrance Criteria Met NO YES

1. Minimum International Knee Documentation Committee (IKDC) Subjective Knee Form score of 70 2. Either no postsurgical history of giving way or a negative pivot shift 3. A minimum baseline level strength knee extension peak torque/body mass >40% (male) and 30% (female) at 300°/sec and 60% (male) and 50% (female) at 180°/sec

Dynamic Stabilization and PATH Strengthening

Stage I

NO Minimum Criteria Met

YES

1. Single limb squat and hold symmetry (minimum of 60° knee flexion with 5-second hold) 2. Audibly rhythmic foot strike patterns without gross asymmetries in visual kinematics when running (treadmill 6-10 mph; 10-16 km/hr) 3. Acceptable single-limb balance scores on Stablometer (Females < 2.2° of deflection and males < 3.0° of deflection total sway tested for 30 seconds at level 8)

Functional Strengthening

Stage II

NO Minimum Criteria Met

YES Stage III

Power Development

NO Minimum Criteria Met

YES

1. Side-to-side symmetry in peak torque knee flexion and extension (within 15% at 180 and 300°/sec) and hip abduction peak torque symmetry (within 15% at 60 and 120°/sec) 2. Plantar force total loading symmetry measured during squat to 90° knee flexion (<20% discrepancy between sides) 3. Single limb peak landing force symmetry on a 50-cm hop (< 3 × body mass and within 10% in side-to side measures)

1. Single-limb hop for distance (within 15% of the uninvolved limb) 2. Single-limb cross over triple hop for distance (within 15% of the uninvolved limb) 3. Single-limb timed hop over 6 meters (within 15% of the uninvolved limb) 4. Single-limb vertical power hop (within 15% of the uninvolved limb) 5. Re-assessment of tuck jump (15 percentage points of improvement or an 80-point score)

Sports Performance Symmetry

Stage IV

NO Minimum Criteria Met

YES

1. Drop vertical jump landing force bilateral symmetry (within 15%) 2. Modified Agility T-test (MAT) test time (within 10%) 3. Single-limb average peak power test for 10 (bilateral symmetry within 15%) 3. Re-assessment of Tuck Jump (20 percentage points of improvement from initial test score or perfect 80-point score)

Reintegration to Interval Sport Participation Figure 5E-1   Return-to-sports algorithm after anterior cruciate ligament reconstruction. Before progression to the next rehabilitative stage in the program, the athlete is required to meet the minimal progression criteria listed.   (Modified and redrawn from Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

Rehabilitation and Therapeutic Modalities

isokinetic knee extension peak torque–to–body mass ratio at 300 degrees per second are 40% to 55% for men and 30% to 45% for women and at 180 degrees per second are 58% to 75% for men and 50% to 65% for women.43 We use a minimal quadriceps torque–to–body mass ratio of 40% for males and 30% for females at 300 degrees per second and 60% for males and 50% for females at 180 degrees per second for return-to-sport training for the athletic population. These values are the low ranges of normative data, which we hypothesize are the baseline levels of strength that athletes should demonstrate for a safe and successful introduction into the initial stages of the return-to-sport program.

PROGRESSION INTO THE RETURN-TO-SPORT STAGES The rehabilitation progression, especially the plyometric component, should take the athlete through a combination of both low-risk and high-demand maneuvers in a controlled environment. Traditional rehabilitation exercises are performed at slower speeds, with low to moderate forces and often in single planes of motion. These exercises promote muscle recruitment, improve muscle strength, and increase muscle endurance; however, they might not simulate the speed, forces, or planes of movement that are encountered during athletic competition or provide opportunity for skill reacquisition.44 Consequently, plyometric exercise has been recommended to bridge the gap between traditional rehabilitation exercises and sport-specific activities.44,45 The training should balance attempts to progressively increase load and develop the functional abilities of the athlete to control these loads with minimal exposure to potential injury risk positions. The introduction to this type of training into the rehabilitation program may create acute muscle soreness.44 The rehabilitation team must use discretion in all phases of return-to-sport training to avoid adverse reactions, such as excessive pain or joint swelling.44 Contraindications for initiating plyometric exercise are acute inflammation or pain, immediate postoperative status, insufficient strength to control joint movement, or joint instability.44,46 Joint pathologies such as arthritis, bone bruise, or chondral injury, are relative contraindications, depending on the ability of the tissue to tolerate the high forces and joint loading required in many plyometric activities. Musculotendinous injury is also a relative contraindication until the tissue is able to handle rapid and high forces of a plyometric exercise.44 Many plyometric exercises, even at low intensities, expose joints to substantial forces and movement speeds47 and are not appropriate in the early phases of rehabilitation after ACL reconstruction. Guidelines for initiating plyometric exercise are poorly developed. Most of the criteria have been established for high-intensity exercise in uninjured athletes and are grounded in opinion rather than research.44 Continual assessment of tissue tolerance and swelling is necessary for the rehabilitation specialist to determine the needed modifications to the outlined protocol to facilitate appropriate intensity and progression of exercises to their individual athletes.44 The presented criteria-driven

303

guidelines may facilitate the decision-making approach toward intensity and exercise mode. The ultimate goal of the ACL return-to-sport algorithm of rehabilitation is to ­identify and address deficits that may inhibit the athlete from improving his or her neuromuscular function and to raise them to a performance status that will minimize the risk for reinjury. In addition, this approach may provide the potential for athletes after ACL reconstruction to improve their ability to manage dangerous forces and torques that may have incited the initial injury and hindered performance before injury.48

Stage I: Dynamic Stabilization and Pelvis, Abdomen, Trunk, and Hip (Core) Strengthening: Points of Emphasis The first stage of return-to-sport training should focus on the initiation of dynamic lower extremity stabilization techniques and the institution of a core strengthening regimen. More specifically, stage I of the return-to-sport protocol focuses on the following goals (See Appendix 5E-A)29,48: 1. Improvement of single-limb weight bearing at increasingly greater knee flexion angles 2. Improvement of side-to-side symmetry in lower extremity running mechanics 3. Improvement of closed chain single-limb postural ­balance Rehabilitation specialists should modify guidelines to address deficiencies of each athlete, with a secondary focus on increasing the athlete’s potential to meet the minimal criteria required to exit stage I and progress to stage II of return-to-sport training.29 A rationale for exercise selection is provided later to aid exercise prescription modifications necessary to meet the needs of individual athletes after ACL reconstruction.48 The goal of the dynamic stabilization training and core strengthening is to develop a baseline level of core stability and coordination that allows the athlete to control the deceleration of his or her center of mass, maintain balance and posture, and subsequently accelerate mass by rapidly generating force in the desired direction. Ultimately, the goal is to control the position of the center of mass (COM) in relation to the base of support to minimize the excessive forces on the lower extremity related to excessive movement of the COM. Exercises to strengthen core stability specifically address the musculature around the torso and hip.49,50 Decreased core strength and muscle synergism may reduce performance in power activities and may increase the incidence of injury secondary to lack of control of the COM, especially in female athletes.51,52 Zazulak and colleagues reported that factors related to core stability predicted risk for knee, ligament, and ACL injuries with high sensitivity and moderate specificity in female athletes.53 A logistic regression model that incorporated measures of core stability and trunk proprioception predicted knee, ligament, and ACL injury risk in females with 84%, 89%, and 91% accuracy, respectively.53 Increased hip adduction with dynamic tasks and decreased hip muscle strength can contribute to lower extremity valgus.54 Lower extremity valgus

304

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

combined with limb asymmetries is related to increased risk for ACL injury in young female athletes.48,55 Successful core strengthening requires a multifaceted approach to athlete preparation that includes total body strength and power training, fundamental movement and technique training, plyometrics, balance and stability, and speed and agility training.49,56,57 Core strength and stability are related to the body’s ability to actively control the body’s COM in preparation to control forces generated from distal body parts during athletic competition. Neuromuscular training and rehabilitation geared toward increasing core strength and stability may both reduce the risk for injury through more effective control of the athlete’s COM and prepare the athlete to achieve optimal performance levels.50,55,57,58 The program design incorporates dynamic stabilization and core strengthening into every facet of an athlete’s return-to-sport training. However, the initiation of high-level return-to-sport training requires the athlete to have an adequate level of balance and strength; thus, it is an early focus of return-tosport training. In addition, balance board proprioceptive training should be used well past the early return-tosport training stage, not only for restoration of function but also for the potential prophylactic effect on ligament reinjury.48,50,55,58-64 In the early stages of rehabilitation after ACL reconstruction, rehabilitation specialists focus on developing a proficient walking gait for the athlete. However, small gait deviations may still be present in the late phases of accelerated ACL rehabilitation protocols.9,65,66 These small gait deficits during walking may be exaggerated into pronounced gait deviations in athletes after ACL reconstruction who attempt to run and sprint.67 Running gait training performed on a treadmill may allow the clinician to provide simultaneous and continuous verbal feedback cues to improve the athlete’s running techniques (Fig. 5E-2).68 An early goal in gait retraining is to normalize ROM in the involved and the noninvolved limbs.69,70 In addition, there should be a focused effort to improve symmetry of the lower extremity musculature, which may prevent abnormal side-to-side loading of the ligaments and soft tissue.70,71 The involved limb often demonstrates limited joint ROM during functional activities, especially at the hip, despite full anatomic ROM.41,65 Inclined treadmill running can force the athlete to increase hip flexion power and functionally use hip ROM during running.72 Care should be taken when initiating running gait training to monitor the athlete for signs of patellofemoral pain, which should be addressed accordingly with exercise ­modifications.48,73 Increasing sprint speed (running velocity) on the treadmill will continue to require movement through larger joint ROM, especially at the hip and the ankle. Therefore, attention should be directed toward obtaining a normal rhythmic stride. A nonrhythmic foot contact pattern during sprinting may be indicative of unbalanced limb contribution and is evident through the audible monitoring of foot contacts. If the athlete demonstrates unbalanced sprinting gait, the contributing factors are likely either pain or failure to use full ROM in the involved leg. If patellofemoral pain and decreased joint mobility are determined to be the limiting factors, then increased focus on backward

F

I

A B E G

D

C H

Figure 5E-2   Representative figure of running techniques instituted for athletes during their return-to-sport training. Having the athlete run on a treadmill during stage I allows the rehabilitation specialist to provide feedback on potential deficits and to control the intensity and volume of running bouts. Athletes are instructed to flex the elbow to 90 degrees and maintain this position (though slightly increased elbow extension is acceptable at the end-point range of motion) during the back swing (A). Athletes are instructed to maintain 90-degree position of elbow (though slightly increased elbow flexion is acceptable at the end-point range of motion) during the forward arm swing (B). Athletes are instructed to extend their shoulder to a point where their wrist would swing past their hip. In addition, during the back swing, athletes are encouraged to minimize shoulder horizontal abduction by keeping the “elbows in” and “brushing the hip pocket with the wrist” (C). D, During the forward arm swing, athletes are told to continue to keep the elbows close to the body, but not to horizontally adduct or flex the shoulder to a point that the “wrist should not cross the midline of the torso” or “take the wrist higher than the chin”(D). Athletes are instructed to “maintain upright position of the torso” during sprint training (E). Both the inclined treadmill training and resistive groundbased training can influence forward trunk flexion beyond optimal positions. This teaching point for torso alignment was cued often for both sprint groups during training. Athletes are encouraged to “keep the shoulders square to the direction of travel” to limit torso rotation during training (F). Athletes are encouraged to “drive the thigh through” and “attempt to get the thigh parallel to the ground” (90 degrees of hip flexion) during the forward leg swing (G). H, Athletes are instructed to initiate foot strike “on the balls of the feet” and “push off with full ankle extension” (H). Athletes are encouraged to be relaxed and to avoid trying to “strain” through the sprint training bouts. They are instructed to “relax the upper torso” and to “avoid clinching or straining the jaw and neck” during training bouts (I).   (From Myer GD, Ford KR, Brent JL, et al: Predictors of sprint start speed: The effects of resistive ground based vs inclined treadmill training. J Strength Cond Res 21[2]:491-496, 2007, with permission from the National Strength and Conditioning Association, Colorado Springs.)

Rehabilitation and Therapeutic Modalities

running may limit patellofemoral loads and assist the athlete through this stage of progression.73,74 In addition, backward running may be used to increase work and decrease patellofemoral joint loads, which may effectively increase quadriceps strength.48,73,75,76 Once the athlete has increased functional lower extremity mobility to normal levels and has attained lower extremity symmetry when jogging at lower intensity, treadmill speed can be increased to assess the athlete’s sprinting form near functional speeds. Pain-free symmetrical sprinting gait should be the ultimate goal of this treadmill training. The stage I focus on core strengthening and running should be tailored to provide an appropriate balance between developing the proprioceptive abilities of the athlete and exposing the athlete to inadequate joint control.48 This may help prepare the athlete for safer participation in plyometric activity, later in the rehabilitation progression. The rehabilitation exercises should take the athlete through a combination of low- and high-demand maneuvers in a controlled situation.49 The intensity of the exercises can be modified by changing the arm position, opening and closing the eyes, changing support stance, increasing or decreasing surface stability with balance training devices, increasing or decreasing speed, adding unanticipated movements or perturbations, and adding sports-specific skills (Fig. 5E-3).50 Core strengthening and dynamic stabilization should provide the athlete with baseline levels of both torso and hip strength and coordination that are adequate to safely progress onto more dynamic sports-related training.48 Simultaneous running gait retraining and a progressive core strengthening program introduce athletes after ACL reconstruction to strategies that allow them to properly initiate, control, and decelerate ground reaction forces that they will encounter in competitive play when jumping, landing, and cutting. Before progression to stage II of the ACL return-to-sport program, it is recommended that the athlete demonstrate minimal unilateral balance and functional strength measures described later.29 Athletes who have decreased neuromuscular control of the core measured during trunk repositioning and sudden load release tasks are at increased risk for ACL injury.53 Athletes should be evaluated for trunk and hip positioning and postural stability deficits before return to competition and perform targeted core neuromuscular training. The implementation of dynamic stabilization and core strengthening, including proprioceptive exercise, perturbation, and correction of body sway, has the potential to prevent the occurrence and to reduce the reoccurrence of ACL injury in athletes after ACL reconstruction.48

305

Figure 5E-3   Partner perturbations are used to advance balance and postural control strategies. Testing for neuromuscular deficits helps physicians determine interventions. (From Myer GD, Ford KR, Hewett TE: Preventing ACL injuries in women. J Musculoskel Med 23:12-38, 2006.)

Stage I: Criteria for Progression 1. Single-limb squat and hold symmetry (minimum of 60 degrees of knee flexion with 5-second hold) (Fig. 5E-4) 2. Audibly rhythmic foot strike patterns without gross asymmetries in visual kinematics when running (treadmill, 6 to 10 mph; 10 to 16 km/hr) (Fig. 5E-5) 3. Acceptable single-limb balance scores on stabilometer (females, <2.2 degrees of deflection; males, <3.0 degrees of deflection; total sway tested for 30 seconds at level 8) (Fig. 5E-6)

Figure 5E-4   Single-limb squat test with a goniometer. The athlete is instructed to squat to 60 degrees or more of knee flexion and hold for 5 seconds. Athletes must perform this task on both their involved and non-involved limbs. (From Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

306

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5E-5   Demonstration of a clinician assessing running gait mechanics. The clinician evaluates the athlete to determine whether he or she demonstrates audibly arrhythmic foot strike patterns or gross asymmetries in visual kinematics when running that would limit progression onto subsequent stages.   (From Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

Single-limb postural stability deficits may be present bilaterally after ACL rupture and well into the postoperative rehabilitation period.29,77,78 Measures of postural stability can be used as a means of assessing an athlete’s recovery of functional stability after ACL reconstruction and are related to their subjective knee rating.61,77-79 Postural balance is a complex function that relies on the interplay of several factors, including proprioception, strength and function of dynamic joint restraints, static joint restraints, and postural equilibrium. Dynamic joint restraints are muscle-tendon units that maintain limb and joint position and react to changing loads and forces. Static joint restraints include ligaments and bony architecture that limit joint motion. Measures of postural sway are often assessed on stable surfaces during single-limb standing with the knee extended or nearly extended and with variations in visual input. These measures can be accurately assessed by clinicians but may not always be sensitive to determining deficits after ACL reconstruction.80 Increasing knee flexion positions during dynamic tasks may be useful for the determination of side-to-side differences because greater knee flexion may be more challenging if an athlete has continued strength deficits. In addition, improved proficiency in performing tasks at increased knee

Figure 5E-6   Single-limb balance measurement taken on a Biodex Stabilometer. (From Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

flexion may limit exposure to excessive anterior tibial shear loads that can overload weakened grafts when performing dynamic loading tasks.81-83 Patients with ACL injury may not demonstrate differences when compared with controls or their contralateral limb. However, during more dynamic cutting tasks, these patients use a cutting strategy with decreased knee flexion.84 Progression through subsequent return-to-sport stages will incorporate safe biomechanical techniques in deep knee flexion angles that may protect the ACL graft.49 Based on empirical evidence, we require that athletes be able to squat to 60 degrees of knee flexion in single-limb stance and maintain postural control for a minimum of 5 seconds. During this test, the athlete should demonstrate the ability to maintain the hip and trunk in an upright position during descent in order to maintain the position of the COM along a vertical axis. Clinicians can use a goniometer to demonstrate the desired knee flexion

Rehabilitation and Therapeutic Modalities

angle and cue the athlete once the angle is achieved to start ­timing. Straight line jogging is often initiated early in rehabilitation programs,6 but frequently the athlete’s technique is altered owing to underlying deficits.85 Without direct assessment of a patient’s running gait pattern, it is difficult to develop objective tests that can be used for progression after ACL reconstruction. However, we delay progression into the next stage of rehabilitation if the patient visually demonstrates grossly evident limb asymmetries during treadmill running. Patients with abnormalities in running gait after ACL reconstruction may benefit from biofeedback training. Pilot work demonstrates that visual and verbal biofeedback can influence desired kinematic gait variables in normal and in patient populations.86,87 Clinicians rehabilitating patients after ACL reconstruction may use biofeedback techniques with treadmill training to assess and treat gross abnormalities in straight line running technique.86,87 Specifically, we focus on improvement of the symmetry of loading the extremity and ROM at lower extremity joints, and obtaining normal rhythmic strides during forward running on a treadmill.86,87 We recommend that patients demonstrate an audibly rhythmic foot strike pattern without gross asymmetries in visual kinematics when running (treadmill, 6 to 10 mph; 10 to 16 km/hr) before progression to stage II of return-to-sport rehabilitation. These recommendations for running gait assessment and training, gained primarily from empirical evidence, warrant further investigation to determine their clinical effectiveness and validity. Stabilometry (see Fig. 5E-6) is an objective method for evaluating postural stability on more functional unstable platforms and can be diagnostic for remaining neuromuscular deficits in athletes after ACL reconstruction.61,78 The athlete’s ability to control the platform’s tilt is quantified as a variance from center of pressure (increased variance scores indicating decreased postural stability). The Biodex Stability System (BSS, Biodex Corp., Shirley, NY) gives reliable measures of body sway.88,89 Schmitz and Arnold examined the intrarater and intertester reliability of the BSS in a cohort of 19 healthy subjects.89 The authors implemented a repeated measures design with two testers on consecutive days. They reported intratester (ICC = 0.82) and inter­ tester (ICC = 0.70) reliability for total ­stability. In a similar study, which evaluated test to retest reliability on the same day, Pincivero and associates reported good intrarater reliability at level 8 for the nondominant (ICC = 0.78) and dominant (ICC = 0.95) limbs.88 Mizuta and colleagues examined subjective complaints of functional instability following ACL injury with a stabilometric measurement of postural balance on a force plate and found an impaired standing balance in the group with functional instability.90 They concluded that stabilometry was a useful method for evaluating functional instability of the knee. Tropp and coworkers reported that athletes who could not demonstrate postural balance within 2 standard deviations of normal had a significantly higher risk for lower extremity injury.91 Normal female subjects demonstrate greater single-limb postural stability than males on a stabilometer.61 However, in subjects with ACL-deficient knees, males had greater stability than females preoperatively on the involved and noninvolved limb.61 On

307

­ ostoperative examination, the males continued to have p greater total stability than females with significant differences remaining between these groups 6, 9, and 12 months after surgery. The males’ instability on the involved limb peaked 3 months after surgery, whereas the females had the most instability 6 months after surgery. If females have greater deficits in single-limb balance than males after ACL rupture, they may require more rehabilitation to recover from ACL reconstruction, to regain functional stability, and to be prepared to return to peak function. These findings also indicate that the clinician should stress balance and functional stability exercises in the rehabilitation of female patients to aid in a progressive return of proprioceptive abilities.61 On average, males and females demonstrate 3 months after ACL reconstruction about a 20% deficiency relative to normal controls in postural control as assessed on BSS for 30 seconds. Our athletes who desire to progress to stage II of the return-to-sport phase must demonstrate a postural stability deficits that are within a 20% range of control values. We require females to score less than 2.2 degrees of deflection and males less than 3.0 degrees of deflection of total sway for both the involved and uninvolved limb to progress to stage II.

Stage II: Functional Strength: Points of Emphasis The second stage of return-to-sport training should focus on improvement of the athlete’s functional strength.48 More specifically, stage II of the return-to-sport phase focuses on the following (see Appendix 5E-A)29: 1. Improvement of lower extremity non–weight-bearing strength 2. Improvement of force contribution symmetry during activities involving bipedal stance 3. Improvement of single-limb landing force attenuation strategies During this stage, we recommend that rehabilitation specialists continue lower extremity weight-bearing strengthening activities and high-intensity balance and perturbation training. In addition, the return-to-sport training program can now progress non–weight-bearing lower extremity exercises such as knee extension exercises.92 Additional emphasis is placed on improving the athlete’s strength with squatting techniques, focusing on equal side-to-side limb contribution. Increased focus on appropriate force attenuation strategies with landing on a single limb may also be incorporated into the training regimen. Exercise prescription should be targeted to address other identified deficits specific to the individual athlete.48 The prophylactic effects of increased strength and the use of resistance training have not been shown in isolation to reduce ACL injury in normal populations or reinjury in athletes after ACL reconstruction. Interestingly, assessment of quadriceps strength has traditionally been used as the gold standard to release athletes to return to sport.4 There is inferential evidence that resistance training improves functional strength and reduces injury based on the beneficial adaptations that occur in bones, ligaments,

308

A

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

Figure 5E-7   A, Box heel touches can be performed on a 12-inch box. B, Difficulty can be increased by using an Airex pad to decrease stability. Testing for neuromuscular deficits helps physicians determine interventions. (From Myer GD, Ford KR, Hewett TE: Preventing ACL injuries in women. J Musculoskel Med 23:12-38, 2006.)

and tendons following training,93,94 and strength measures are related to increased functional outcome after an ACL injury.4,41,95 Lehnhard and colleagues reported significantly reduced injury rates with the addition of strength training in men’s soccer.96 They monitored injuries for 2 years without training and 2 years with strength training. Although they did not report a reduction of ACL injuries, they reported a decrease in percentage of ligament sprains in the study group in which knee injuries accounted for up to 57% of the total injuries in a given year.96 In addition, Cahill and Griffith incorporated weight training into their preseason conditioning for football teams.97 They found a reduction both in reported knee injuries and in knee injuries that required surgery over four competitive seasons in the

trained groups.97 Protocols that supplement plyometric and technique training with strength training may significantly reduce ACL injuries in female athletes.98Thus, it appears that exercises designed to induce functional strength gains, especially those exercises that involve strength and balance (see Figs. 5E-5 and 5E-6), may be effective at reducing knee injuries when combined with other training components. However, the efficacy of a single-faceted resistance training protocol on ACL injury or reinjury risk reduction has yet to be demonstrated in the literature. Thus, once functional strength level progresses, more dynamic exercises that teach appropriate lower extremity control may be warranted.48 The initial functional strength training can be performed with body weight only (Figs. 5E-7 and 5E-8), with an initial high-volume, low-intensity protocol.99-101 When appropriate, external weight can be added to increase exercise intensity (Fig. 5E-9). Rehabilitation specialists should take the time to prescribe the appropriate weight to be used before each session based on the workload achieved in the prior session to safely progress the athlete. The weight used or repetitions prescribed must be increased between sessions to ensure progression of exercise intensity and strength adaptation. However, intensity progression must not sacrifice proper technique or safety. If technique is not near perfect, then resistance should be decreased until proper technique is restored. The goals of the functional strength training component of the protocol are to strengthen major muscle groups through the complete ROM and to provide adequate muscular power to progress to more advanced plyometric components included in later stages of the protocol. In addition, the progressive core strength and dynamic stabilization techniques should be continued to ensure competence for progression to stage III of the protocol. Again, appropriate care should be taken to limit patellofemoral pain during training, similar to the early stages of rehabilitation. For example, evidence suggests that the sumo squat (see Fig. 5E-9) alters the mechanics and muscular activation9,102,103 and therefore may sufficiently alter patellofemoral kinematics during closed chain activities to reduce patellofemoral stress with increased (>45 degrees) knee flexion. Modified squat ­exercises with increased hip abduction may allow athletes after ACL reconstruction to increase knee

Figure 5E-8   Single-leg pelvic bridge (A) and the single-leg straight leg deadlift (B) can be performed on a BOSU training device with or without the addition of external weight. Testing for neuromuscular deficits helps physicians determine interventions. (From Myer GD, Ford KR, Hewett TE: Preventing ACL injuries in women. J Musculoskel Med 23:12-38, 2006.)

Rehabilitation and Therapeutic Modalities

Figure 5E-9   The sumo squat is performed with the feet wider than shoulder width apart. The athlete should focus on maintenance of an upright posture with minimized trunk flexion during the exercise.

flexion, with potentially decreased anterior knee pain compared with shoulder-width squatting exercises.48 High-intensity retrograde incline running (Fig. 5E-10) can be used to facilitate functional knee ROM and to increase quadriceps functional strength with limited relative joint loading.74,75,104 The ability of inclined retrograde training to increase functional quadriceps activation and to limit patellofemoral stress was shown by Flynn and coauthors, who reported increased concentric quadriceps activation with decreased relative patellofemoral compressive forces in backwards treadmill training.73,104 In addition to the ROM and strength benefits from retrograde training, it may help the athlete regain cardiorespiratory fitness without increased knee joint stress when compared with other

A

309

forward running.73,105 Finally, retrograde treadmill training has been used in protocols to improve performance measures that may benefit athletes in sports that require speed, agility, and backward motion.29,48,105 Sprint training can be accomplished through interval resistive band running or high-intensity treadmill training (Fig. 5E-11).50,68,106,107 The important component with interval speed training is to emphasize short-duration and high-intensity running bouts. Performing excessive endurance training may interfere with explosive strength development needed for running and cutting sports.108,109 Proper running technique should continue to be emphasized as the intensity of training increases (see Fig. 5E-2). In addition to the strength, power, and anaerobic capacity gains achieved from sprint training, athletes after ACL reconstruction can improve their muscular endurance and delay fatigue during high-intensity activities through mechanisms of improved efficiency of movements, improved aerobic energetics, and improved buffering capacity.109,110 To perform interval partner resistive band running, two medium bands (Jump Stretch Inc., Youngstown, Ohio) can be tied together and anchored around the waists of two athletes (see Fig. 5E-11).111 The athlete in the forward position should be instructed to quickly transition from this starting stance to full running with proper ­biomechanics for the allotted time period. The trailing athlete provides a light, medium, or heavy resistance as instructed by the rehabilitation specialist. During the initial session, the athletes should be instructed by the clinician how to vary the resistance. The rehabilitation specialist should provide ­biomechanical feedback during each training bout.111 The final running of each session should include a ­ nonresisted ­ maximal effort run of varying distance. If available, speed training can also be performed on high-performance treadmills that can accommodate high speeds and inclines to adjust protocol intensity (see Fig. 5E-11).106 Utilization of both inclined treadmill training and band resistive ­techniques in return-to-sport training may be best to achieve the goals

B

Figure 5E-10   Retrograde training performed on level (A) and inclined (B) treadmill positions. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008).

310

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 5E-11   Treadmill training (A) and resistive running (B) are used to help athletes regain functional speed and functional strength, respectively. The athlete is encouraged to improve symmetry in gait cycles when training. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

of improved running mechanics (increased stride length and frequency, decreased vertical displacement), improved short distance speed, increased explosiveness, and increased muscular resistance to fatigue.48,109-111 The global effects of strength training for athletes after ACL reconstruction may be best achieved when combined with progressed dynamic stabilization and core strengthening as well as resistive and retrograde movement training.62 Before progression to the third stage, it is recommended that the athlete demonstrate the following minimal strength measurements.29,48

Stage II: Criteria for Progression We recommend that the athlete demonstrate the proficiency in the following criteria before progression to stage III39:

­ eakness is a limiting factor in rehabilitative progression, w and failure to achieve adequate strength can potentially result in increased risk for future injury, including acute and overuse injuries such as anterior knee pain.9,117 Decreased quadriceps function following injury and reconstruction, coupled with the evidence of compensation with functional tasks such as stair ascent and landing from a jump following ACL reconstruction, is of concern.18,23 Global assessments of function, although critical to the overall assessment of the athlete, may fail to detect isolated weakness in the knee extensor muscle group. Because quadriceps and hamstrings co-contraction is an important dynamic stabilizer of the knee joint,118 adequate quadriceps and hamstrings strength is important for the safe progression of the athlete after ACL reconstruction to the next stage of the return-tosport program as well as to the ultimate discharge of the

1. Side-to-side symmetry in peak torque knee flexion and extension (within 15% at 180 and 300 degrees/second) and hip abduction peak torque (Fig. 5E-12) side-to-side symmetry (within 15% at 60 and 120 degrees/second) 2. Plantar force total loading symmetry measured during squat to 90 degrees of knee flexion (<20% discrepancy between sides) 3. Single-limb peak landing force symmetry on a 50 cm hop (<3 times body mass and within 10% in side-to-side measures) Historically, isokinetic strength has been used as a criterion to progress to return-to-sport activities.2,112,113 Despite reports that quadriceps deficit may persist up to 2 years after ACL reconstruction22,42,114 and that only a low to moderate correlation of isokinetic strength to knee function may exist,114-116 isolated quadriceps strength is still considered a critical component to safely progress athletes back to dynamic activities.4 Typically, quadriceps

Figure 5E-12   Example of hip abduction test performed on an athlete after anterior cruciate ligament reconstruction.

Rehabilitation and Therapeutic Modalities

athlete to sport. Therefore, we recommend that patients demonstrate peak torque symmetry of the involved limb within 15% of the contralateral limb for both quadriceps and hamstring strength at 180 and 300 degrees/second to progress to the power development stage of the program. Hip abduction strength is likely to be important for both dynamic knee stability and decreasing reinjury risk. In a cohort of young healthy female athletes, Hewett and coworkers determined that measures of dynamic lower extremity valgus and asymmetries in hip abduction torque were predictive of ACL injury in this population.55 Padua and associates showed that hip abduction strength was a significant predictor of initial contact and peak knee ­valgus angles during a drop landing task.54 Zazulak and colleagues reported that female athletes who are at higher risk for ACL injury demonstrate decreased hip muscle activation during single-limb landing tasks when compared with males.119 Although differences between limbs have been demonstrated in side-lying isometric hip abduction torque,120 pilot data in a larger cohort (n = 152) of young athletes indicated that these side-to-side differences are not evident when hip abduction is evaluated isokinetically in the standing position.121 Adequate hip abduction strength is likely necessary to safely return to sport after ACL reconstruction. Therefore, we recommend that the patient demonstrate peak torque symmetry within 15% of the contralateral limb for hip abduction strength at 60 and 120 degrees/second (see Fig. 5E-12). Side-to-side asymmetries in dynamic functional tasks such as jumping and cutting are risk factors for ACL injuries in young healthy athletes.55 After ACL reconstruction, patients frequently do not demonstrate the ability to balance forces bilaterally in the lower extremities with both high-level tasks such as landing19 and less dynamic tasks such as squatting.21 Neitzel and colleagues demonstrated that with squatting, patients were unable to balance sideto-side loading response equal to that of controls until 12 to 15 months after surgery.21 These patients demonstrated side-to-side deficits between 33% and 48% at 1.5 to 4 months after surgery and between 21% and 28%

311

6 to 7 months after surgery. Considering the potential for side-to-side biomechanical differences to increase risk for ACL injury, it may be necessary to train the patient to balance these forces before progression to stage III.55 Force platforms, insole foot pressure systems, or standard bathroom scales may be used to determine relative side-to-side loading discrepancies during this activity. We suggest that the patient demonstrate less than a 20% side-to-side difference in total loading during a 90-degree knee flexion squat maneuver.21 Inability to attenuate forces during a single-limb maneuver may be related to increased risk for ACL injury.122 Specifically, athletes who use decreased landing knee flexion may subject the limb to an abrupt bone-to-bone stress at the knee. Decreased knee flexion at landing may be evidenced by increased landing forces and could be a result of decreased thigh muscle strength.122 The single-limb landing force symmetry test (<3 times body mass and within 10% in side-to-side measures) is performed a total of six times (three randomized trials on each side) using an AccuPower portable force platform (Advanced Mechanical Technology, Inc., Watertown, Mass). Subjects are instructed to initiate the movement while balancing on one foot, to hop forward 50 cm, and to balance for 10 seconds after landing on the same foot. Maximal vertical ground reaction force is calculated for each trial. Maximal vertical ground reaction force shows high within-session reliability on both the dominant (r = 0.823) and nondominant (r = 0.877) sides.123 The average maximal vertical ground reaction force for the normal athlete is 2.4 times body mass.123 It is recommended that athletes perform this test with less than 3 times body mass and a side-to-side discrepancy of less than 10%. Finally, before progression to stage III, we recommend that the athlete’s lower extremity plyometric techniques be assessed. Specifically, we have the athlete perform repeated tuck jumps for 10 seconds. A standard two-dimensional camera in the frontal and sagittal planes may be used to assist the clinician. The athlete’s technique (Fig. 5E-13) is then subjectively rated on an 80-point scale (See Appendix 5E-C).

Figure 5E-13   Examples of techniques that clinicians evaluate during the tuck jump assessment. For this test, the athlete is positioned with the feet shoulder width apart and is asked to jump pulling the thighs parallel to the ground for 10 seconds. 1, Example of knees not in neutral alignment. 2, Example of thighs not reaching parallel at top of jump. 3, Example of thighs not equal side to side throughout the flight sequence. 4, Example of inappropriate foot placement at landing, not shoulder width apart. 5, Example of foot placement in appropriate parallel position not staggered during ground contact. (From Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

312

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The athlete’s baseline performance is determined at this time. This assessment will be repeated at the conclusion of each subsequent stage to objectively track improvement with jumping and landing technique. Pilot work in our laboratory with 10 physical therapists was conducted to determine the reliability of scoring the tuck jump assessment. These data demonstrated that intrarater, within-session reliability was high (r = 0.84; range, 0.72 to 0.97). Pilot data using the tuck jump assessment tool suggest that it may be adequate for a single clinician to reassess athletes to determine changes in technical performance of the tuck jump exercise. The authors acknowledge that further work is needed to validate this tool when used for progression in patients after ACL reconstruction. The tuck jump exercise may be useful for the clinician to identify lower extremity valgus and side-to-side differences of landing mechanics during plyometric exercise (Fig. 5-14; see Fig. 5E-13).49 The tuck jump requires a high effort level from the athlete. Initially, the athlete may place most of his or her cognitive efforts solely on the performance of this difficult jump. The clinician may readily identify potential deficits, especially on the first few repetitions.49 Additionally, the tuck jump exercise may be used to assess improvement in lower extremity biomechanics as the athlete progresses through the return-to-sport ­training.3,49 Specifically, correction of lower extremity valgus at landing and improvement of side-to-side differences in lower extremity movements and foot placements are the focus of the tuck jump assessment tool. The link between valgus knee loading and resultant increases in ACL strain is demonstrated through cadaver, in vivo, and computer

modeling experiments.79,124-126 Physiologic valgus torques on the knee can significantly increase tibial subluxation and load on the ACL.126 A prospective combined biomechanical-epidemiologic study showed that knee abduction moments (valgus torques) and angles were significant predictors of future ACL injury.55 Knee abduction moments, which directly contribute to lower extremity dynamic valgus and knee joint load, predicted ACL injury risk with high sensitivity and specificity.55 It may be even more important to address potential lower extremity valgus measures if demonstrated by patients after ACL reconstruction because it may have been a predisposing factor to their initial injury.55 By improving neuromuscular control and biomechanics during this difficult jump and landing sequence, the athlete may gain dynamic neuromuscular control of the lower extremity and create a learned skill that can be transferred to competitive play.

Stage III: Power Development: Points of Emphasis The third stage of return-to-sport training focuses on return of the athlete back to sport and improvement beyond his or her lower extremity preinjury power levels.48 More specifically, stage III of the return-to-sport phase focuses on the following (See Appendix 5E-A)29: 1. Improvement of single-limb power production 2. Improvement of lower extremity fatigue resistance 3. Improvement of lower extremity biomechanics during plyometric activities

Figure 5E-14   Tuck jumps are an example of an exercise used to train the athlete to increase lower body power. The tuck jump can also be used as an assessment to grade improvement in technique. To perform the tuck jump, the athlete starts in the athletic position with the feet shoulder width apart. The athlete initiates the jump with a slight crouch downward while extending the arms behind. The athlete then swings the arms forward and simultaneously jumps straight up and pulls the knees up as high as possible. At the highest point of the jump, the athlete is in the air with the thighs parallel to the ground. When landing, the athlete should immediately begin the next tuck jump. Encourage the athlete to land softly, using a toe-to-midfoot rocker landing. The athlete should not continue this jump if he or she cannot control the high landing force or demonstrates a knock-knee landing. (From Myer GD, Ford KR, Hewett TE: Rationale and clinical techniques for anterior cruciate ligament injury prevention among female athletes. J Athl Train 39[4]:352-364, 2004.)

Rehabilitation and Therapeutic Modalities

During stage III of the return-to-sport training, we recommend the incorporation of mid-intensity double-limb plyometric jumps and the introduction of low-intensity single-limb repeated hops into the training regimen. We focus on proper and safe technical performance of the plyometric activities. The athlete’s ability to properly perform the plyometric tasks can be used to guide the volume and intensity of the exercises selected.44,48 The goal of the power development stage of the returnto-sport program is to progress increased strength into sports-related power. Plyometric training can be used to improve power measures and force dissipation strategies.57,62,127,128 Most of the initial plyometric exercises should involve both legs to safely introduce the athlete to the training movements (see Fig. 5E-14).44 The appropriate intensity for plyometric exercise is based on the ability of the healing tissue to handle loading and the ability of the patient to perform an activity with proper technique. Similar to other forms of training and rehabilitation techniques, intensity of plyometric exercise should follow a gradual progression from low- to high-intensity activities to avoid adverse responses.44 Early training emphasis should be on balanced athletic positioning (Fig. 5E-15) that can help create dynamic control of the athlete’s c­enter

Figure 5E-15   The athlete is in a functionally stable position with the knees comfortably flexed, shoulders back, eyes up, feet about shoulder width apart, and body mass balanced over the balls of the feet. The knees should be over the balls of the feet, and the chest should be over the knees. This is the athleteready position and is the starting and finishing position for most of the training exercises. During some of the exercises, the finish position is exaggerated with deeper knee flexion (deep hold; see Fig. 5E-16) to emphasize the correction of targeted biomechanical deficiencies. (From Myer GD, Ford KR, Hewett TE: Rationale and clinical techniques for anterior cruciate ligament injury prevention among female athletes. J Athl Train 39[4]:352-364, 2004.)

313

of gravity.50,58,129 Soft, athletic landings that stress deep knee flexion with coronal plane knee control should be employed with verbal feedback from the rehabilitation specialist to make the athlete aware of biomechanically undesirable positions (Fig. 5E-16).49 Later training sessions use explosive double-leg movements focused on maximal performance in multiple planes of motion (Fig. 5E-17). The plyometrics and dynamic movement training components should progressively emphasize double-leg, then reciprocal single-leg, movements through training stages (Fig. 5E-18).50 A greater number of single-leg movements can be introduced gradually while still maintaining the focus on correct technique. For example, the single-leg hopand-hold exercise can be used as a teaching tool to help the athlete develop proper force attenuation strategies on a single limb (Fig. 5E-19).49 Volume of the initial plyome­ tric bouts should be low, owing to the extensive technique training required and decreased ability of the athlete to perform the exercise with proper technique for the given durations. Progression of volume should occur only when technique is maintained and there are no adverse events. In general, patients must demonstrate tolerance of a lowintensity, high-volume activity before progressing to a high-intensity, low-volume activity.44,58,130 Volume can be increased as technique improves to the midpoint of training, followed by a progressive decrease in volume during the final sessions to allow for concomitant increase in exercise intensity.44,48,50,58 Continued progression of the functional strength and core training combined with plyometrics may provide additional benefits (Figs. 5E-20 to 5E-22).58 Subjects who underwent a combined plyometric and squat training program had significant increases in vertical jump over subjects who trained with squats or plyometrics alone.131 Additionally, Fatouros and colleagues found that the combinative effects of plyometrics and resistance training increased not only jump performance but also leg strength.132 Myer and colleagues evaluated the effects of combined neuromuscular training including resistance, plyometric, core, and speed training on basketball, soccer, and volleyball players.50 After training, the athletes demonstrated improvements in performance measures (back squat, single-leg hopand-hold distance, vertical jump, speed) as well as several biomechanical factors related to increased lower extremity injury risk (increased knee flexion-extension ROM, decreased abduction moments during the landing phase of a vertical jump, and increased single-leg postural stability).50,55,56,91 Partner perturbation training was an important component of the training protocols used to improve measures of sports-related performance and reduce ACL injury risk factors and is suggested to be a critical training tool for returning athletes to full function after ACL injury.50,58,133 At this point in return-to-sport training, the partner perturbation training (see Fig. 5E-3) may be progressed into single-leg activities. In addition, higher-intensity resistance training exercises, especially those that target increased knee flexor strength and power (Fig. 5E-23), are critical in this stage of training. Hence, stage III incorporates multiple training components, which may be beneficial for the athletes after ACL reconstruction to help facilitate return to sport with improved performance measures and lower injury risk.48

314

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5E-16   Line broad jump deep hold and broad jump deep hold exercise used to teach the athlete to achieve and maintain knee flexion when landing from a jump. The athlete prepares for this jump in the athletic position with the arms extended behind at the shoulder. The athlete begins by swinging the arms forward and jumping horizontally and vertically at about a 45-degree angle to achieve maximal horizontal distance. The athlete is encouraged to stick the landing with the knees flexed to about 90 degrees in an exaggerated athletic position. The athlete may not be able to stick the landing during a maximal effort jump in the early phases. In this situation, have the athlete perform a submaximal broad or line jump in which the he or she can stick the landing with the toes straight ahead and no inward motion of the knees. As the athlete’s technique improves, encourage the athlete to add distance to the jumps, but not at the expense of perfect technique. (From Myer GD, Ford KR, Hewett TE: Rationale and clinical techniques for anterior cruciate ligament injury prevention among female athletes. J Athl Train 39[4]:352-364, 2004.)

Stage III: Criteria for Progression In order to advance to the next stage, the athlete must achieve the following criteria related to athletic power development and symmetry29: 1. Single-limb hop for distance (within 15% of the uninvolved side) 2. Single-limb crossover triple hop for distance (within 15% of the uninvolved side) 3. Single-limb timed hop over 6 meters (within 15% of the uninvolved side) 4. Single-limb vertical power hop (within 15% of the uninvolved side) 5. Reassessment of tuck jump (15% improvement or an 80-point score) (See Appendix 5E-C; see Fig. 5E-13).

Figure 5E-17   The athlete performs a broad jump and immediately progresses into a maximal effort vertical jump. During the broad jump, the athlete should attempt to attain maximal horizontal distance. Encourage the athlete to provide minimal braking during the transition from the broad jump to the maximal vertical jump. Coach the athlete to go directly vertical on the vertical jump and not move horizontally. Use full arm extension to achieve maximal vertical height. (From Myer GD, Ford KR, Hewett TE: Rationale and clinical techniques for anterior cruciate ligament injury prevention among female athletes. J Athl Train 39[4]:352-364, 2004.)

One tool to assess unilateral lower limb functional power, while assessing limb symmetry, is functional performance testing. Tests such as the single-limb hop for distance, single-limb triple hop for distance, single-limb crossover triple hop for distance, and single-limb timed hop over 6 meters have been previously described as tools to use with athletes after ACL reconstruction or with an ACL-deficient knee.115,134,135 The ability to demonstrate limb symmetry within 15% on these tests may be an effective tool for the evaluation of patients for progression following ACL reconstruction.135 Therefore, the patient’s ability to successfully attain 15% limb symmetry on these

Rehabilitation and Therapeutic Modalities

315

Figure 5E-18   Example of an exercise used to improve lower limb symmetry during an explosive plyometric task. The athlete begins this jump by bounding in place. Once the athlete attains proper rhythm and form, encourage the athlete to maintain the vertical component of the bound while adding some horizontal distance to each jump. The progression of jumps progresses the athlete across the training area. When coaching this jump, encourage the athlete to maintain maximal and symmetrical bounding height on both legs. (From Myer GD, Ford KR, Hewett TE: Rationale and clinical techniques for anterior cruciate ligament injury prevention among female athletes. J Athl Train 39[4]:352-364, 2004.)

Figure 5E-19   The single-leg hop and hold exercise can help teach the athlete appropriate force attenuation and postural control strategies on a single leg. The starting position for this jump is a semicrouched position on a single leg. The athlete should hold the arms fully extended behind at the shoulder. The jump is initiated by swinging the arms forward while simultaneously extending at the hip and knee. The jump should carry the athlete up at about a 45-degree angle and attain maximal distance for a single-leg landing. Athletes are instructed to land on the jumping leg with deep knee flexion (to 90 degrees). The landing should be held for a minimum of   3 seconds. Coach this jump with care to protect the athlete from injury. Start with a submaximal effort using line jumps and progress to a single-leg broad hop. Continue to increase the distance of the broad hop as the athlete improves the ability to stick and hold the final landing. Have the athlete keep his or her visual focus away from the feet, to help prevent too much forward trunk flexion at the waist. (From Myer GD, Ford KR, Hewett TE: Rationale and clinical techniques for anterior cruciate ligament injury prevention among female athletes. J Athl Train 39[4]:352-364, 2004.)

Figure 5E-20   The athlete starts by sitting balanced on the center of the BOSU and then flexes the trunk simultaneous with hip flexion. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

hop tests may be important to safely progress in the returnto-sport phase of rehabilitation. Side-to-side imbalances in muscular strength, flexibility, and coordination may be important predictors of increased injury risk.55,136,137 Knapik and colleagues demonstrated that side-to-side balance in strength and flexibility is important for the prevention of injuries and that when side-to-side differences are present, the athlete is more injury-prone.136 Baumhauer and coworkers also found that individuals with muscle strength imbalances exhibited a higher incidence of injury.137 Hewett and associates developed a model to predict ACL injury risk with high sensitivity and specificity.55 About half of the parameters in the predictive model

316

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

within 15% to meet the requirements for stage progression with this test. In addition, the athlete can be reassessed with the standardized tuck jump assessment tool (See Appendix 5E-C; see Fig. 5E-14). At this point, we recommend that the athlete demonstrate 15% improvement (or a perfect score of 80 points) relative to the initial test to meet the requirements for progression of this task.

Stage IV: Sport Performance Symmetry: Points of Emphasis

Figure 5E-21   The athlete starts lying on the side with the hip located at the edge of the table. The athlete’s feet and legs must be anchored during this exercise by the trainer or a stationary object. The athlete proceeds to flex and extend laterally at the waist for the prescribed repetitions. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

were side-to-side differences in lower extremity kinematics and kinetics. Side-to-side imbalances may increase risk for both limbs. Over-reliance on the dominant limb can put greater stress and torques on that knee, whereas the weaker limb may be at risk owing to an inability of the musculature on that side to effectively absorb the high forces associated with sporting activities. The single-limb vertical power hop test can be used to assess the athlete’s ability to perform work (force × displacement) over a given time. The athlete is instructed to stand in unilateral stance, hop as high as possible from a single limb, and land on both limbs. Peak power generated during the push-off phase is calculated from each hop and averaged over three trials to assess asymmetries among limbs. The athlete must attain an average symmetry value

The final stage of return-to-sport training focuses on movement skills related to the athlete’s sport and maximization of athletic development.48 More specifically, stage IV of the return-to-sport protocol focuses on the following (See Appendix 5E-A)29: 1. Equalization of ground reaction force attenuation strategies between limbs 2. Improvement of confidence to maintain dynamic knee stability with high-intensity change-of-direction ­activities 3. Improvement of symmetry to produce power endurance between limbs 4. Use of safe biomechanics (increased knee flexion and decreased knee abduction angles with symmetrical forces and motions between limbs) when performing high-intensity plyometric exercises In this stage, we recommend that the rehabilitation specialist incorporate power, cutting, and change-of-­direction tasks that are modified to be related to the athlete’s sport.49,50 We suggest emphasis of the performance of power movements equally well in both directions, with appropriate hip and knee flexion angles and decreased knee abduction.49,50 Extensive verbal and visual feedback should be used to help athletes after ACL reconstruction to develop safe biomechanics during power movements.48 The final progression of the plyometric and movement training in stage IV of return-to-sport training should use unanticipated cutting movements during training. ­

Figure 5E-22   The athlete positions the resistive band below the fifth cervical vertebra and stands with both knees slightly flexed and feet on the band. The movement is initiated with the trunk flexed to about 90 degrees. A neutral spine should be maintained as the athlete extends the trunk from 90 to 0 degrees (an erect posture). (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

Rehabilitation and Therapeutic Modalities

Figure 5E-23   In this exercise, the rehabilitation specialist anchors the athlete by standing on the arch of the feet and provides lift assistance with a strap that is wrapped around the chest. The athlete performs full eccentric and concentric movement with the assistance of the rehabilitation specialist. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

Single-faceted sagittal plane training and conditioning protocols that do not incorporate cutting maneuvers do not provide similar levels of external valgus-varus or rotational loads that are seen during sport-related cutting maneuvers.125 Training programs that incorporate safe levels of valgus-varus stress may induce more muscle-dominant ­neuromuscular adaptations.138 Such adaptations can better prepare an athlete for multidirectional sports activities that may improve performance and reduce the risk for lower extremity injury.50,55 Female athletes perform cutting techniques with increased valgus angles.139 Valgus loads on the knee can double during unanticipated cutting maneuvers similar to those used in sport.140 Teaching the athlete to

317

use movement techniques that produce low knee abduction moments during movements that typically induce high loads on the joint can ultimately reduce the risk for injury.55,140,141 Training that incorporates techniques to focus on unanticipated cuts reduces knee joints loads.50 In addition, by improving reaction times to provide more time to voluntarily precontract muscles and make appropriate kinematic adjustments, ACL loads may be reduced.118,140 Figure 5E-24 presents an athlete who demonstrates excessive dynamic knee valgus positions during agility and unanticipated cutting drills. Extensive ­ verbal and, potentially, visual feedback (by videotape) is used to help athletes after ACL reconstruction correct unsafe biomechanics during these movements.48 Before teaching unanticipated cutting, athletes should first be able to attain proper athletic position (see Fig. 5E-15). The athletic position is a functionally stable position with the knees comfortably flexed, shoulders back, eyes up, feet about shoulder width apart, and body mass balanced over the balls of the feet. The knees should be over the balls of the feet, and the chest should be over the knees.49,50 The athletic “ready position” should be the starting and ­finishing position for several of the training exercises. Further, this is the goal position before initiation of a directional cut. Addition of directional cues to the unanticipated training can be as simple as the ­ rehabilitation ­ specialist pointing out a direction or as sports specific as using partner mimic or ball retrieval drills. In addition to the development of safe biomechanics during unanticipated cutting, athletes should work to master techniques during high-intensity plyometrics. During the tuck jump (see Fig. ­ 5E-14), athletes after ACL reconstruction often unload their involved side, as is visually evidenced by uneven foot placement (Fig. 5E-25) and asymmetric limb alignment during flight of jumping (Fig. 5E-26). The rehabilitation specialist should provide real-time feedback to encourage the athlete to equalize lower extremity biomechanics. Focused effort to improve jumping and landing symmetry may alleviate deficits demonstrated by athletes up to 2 years after ACL reconstruction.142 Training the athlete to employ safe

Figure 5E-24   Examples of dynamic valgus positions that athletes after anterior cruciate ligament reconstruction may demonstrate during agility and unanticipated cutting techniques. The rehabilitation specialist should provide active feedback to the athlete to encourage him or her to perform reactive training with limited knee valgus positions. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

318

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 5E-25   Example of an athlete after anterior cruciate ligament reconstruction demonstrating the tuck jump with staggered foot placement during landing. The rehabilitation specialist should start the athlete in the desired foot position and encourage him or her to land in the same footprint. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

c­ utting and landing techniques in sports-related situations may help instill technique adaptations that more readily transfer onto the field of play. The ligament-dominant and leg-dominant athlete may become muscle dominant and symmetrical if the desired training adaptations are achieved, thus ultimately reducing the athlete’s risk factors for future ACL injury.48,50,51,55,57 At this stage, it may become difficult to keep the athlete motivated to train for return to sport. The increased function attained may increase the athlete’s desire to get back into game situations and cause him or her to sacrifice late-stage return-to-sport training sessions for competitive play. Offering more performance-oriented training may influence the athletes after ACL reconstruction, especially the high-risk female athlete, to maintain the return-to-sport training.48 Neuromuscular training programs for young women can improve performance measures of speed, strength, and power.57,94,99,100 Female athletes may especially benefit from neuromuscular training because they often display decreased baseline levels of strength and power compared with their male counterparts.127,143 Dynamic neuromuscular training also reduces gender-related differences in force absorption, active joint ­stabilization, muscle imbalance, and functional biomechanics and increases strength of structural tissues (bones, ligaments, and tendons).57,93,127,144-146 These ancillary effects of neuromuscular training, which likely reduce the risk for injury in female athletes, are positive results of training; however, without the performance-enhancement training effects, athletes may not be motivated to undertake neuromuscular training. Training that is oriented toward the reduction of lower extremity injuries, even in elite female athletes, may have compliance rates as low as 28%.129 However, training targeted toward the improvement of performance measures can have better compliance, ranging from

Figure 5E-26   Example of an athlete after anterior cruciate ligament reconstruction performing the tuck jump with staggered limb alignment during flight. The rehabilitation specialist should encourage the athlete to achieve symmetrical thigh placement at the top of the jump to decrease this symmetry deficit. (From Myer GD, Paterno MV, Ford KR, Hewett TE: Scientific commentary: Neuromuscular training techniques to target deficits prior to return to sport following anterior cruciate ligament reconstruction. J Strength Cond Res 22:987-1014, 2008.)

80% to 90%.95,99,100,108,147 In addition, high-risk athletes may be more responsive to neuromuscular training effects if the training protocols are targeted to address ACL injury risk factors.111 Therefore, if the protocol is designed to focus on safe performance enhancement techniques during late-stage return-to-sport training and incorporates proven ACL injury prevention exercises, combined performance and reinjury preventive training may be instituted with high compliance in ACL reconstruction athletes.48

Stage IV: Criteria for Progression Successful completion of stage IV and ultimate clearance for integration back into sporting activities is dependent on the athlete’s ability to achieve the following criteria related to sport-specific movements29: 1. Drop vertical jump landing force bilateral symmetry (within 15%) (Fig. 5E-27) 2. Modified agility t-test (MAT) time (within 10%) (Fig. 5E-28) 3. Single-limb average peak power test for 10 seconds (bilateral symmetry within 15%) (Fig. 5E-29) 4. Reassessment of tuck jump (20% improvement from initial test score or perfect 80-point score) (see Appendix 5E-C; see Fig. 5E-13) Limb asymmetries with athletic tasks may be potential risk factors for ACL injury and therefore should be minimized before return to sport. All stages of the ACL returnto-sport program attempt to minimize these asymmetries, not only with strength but also with athletic maneuvers.

Rehabilitation and Therapeutic Modalities

B

A

C

D

E

319

F

Vertical ground reaction force (N)

3000 F 2500 2000

C D

1500 1000 B 500 A

0 0

E 200

400

600

800

1000

Uninjured side ACL injured side

Time (msec)

Figure 5E-27   Example of an athlete after anterior cruciate ligament reconstruction performing a drop vertical jump maneuver. The image sequence on the top shows the higher ground reaction forces on the uninjured side (right) at specific time points. Two force curves (right and left side) over time are displayed on the bottom. Force measures are customarily taken from the first initial contact, but the second landing can also be analyzed for a more in-depth analysis. (Redrawn from Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

shuffle

shuffle

15’

15’

15’ backward

backward

Finish

run

run

Start (Right Leg Test)

Figure 5E-28   Layout of modified agility t-test (MAT) used as part of the progression criteria back into interval sport activities. The athlete should perform the maneuver from each direction by starting from the finish line and repeating the pattern. (Redrawn from Hickey KC, Quatman CE, Myer GD, et al: Dynamic field tests used in an NFL combine setting to identify lower extremity functional asymmetries. Strength Condit Res 2009 [in press.])

320

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Power (W/kg)

25

Uninjured ACL

20

15

10

Average peak power (W/kg) ACL 16.6 Uninjured 20.9 Difference 20.6% 1

2

3

4

5

Power decrement (%) ACL 18.8 Uninjured 1.3

6 7 Jump number

8

9

10

11

12

Figure 5E-29   Example of single-limb vertical jump power degradation over time for an athlete in return-to-sport training after anterior cruciate ligament reconstruction. The athlete starts the trial standing on a single limb and repeatedly executes maximal vertical hops for 10 seconds. (Redrawn from Myer GD, Paterno MV, Ford KR, et al: Rehabilitation after anterior cruciate ligament reconstruction: Criteria-based progression through the return-to-sport phase. J Orthop Sports Phys Ther 36[6]:385-402, 2006, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.)

Athletes who demonstrate side-to-side differences in ­biomechanical measures during a drop vertical jump are at increased risk for ACL injury compared with subjects with more symmetrical lower extremity biomechanics.55 Limbto-limb asymmetries are evident during drop landing19 and drop vertical jump22 in patients after ACL reconstruction. Addressing side-to-side differences may decrease risk for injury once athletes are allowed unrestricted returnto-sport after ACL reconstruction. Thus, we assess an athlete’s ability to demonstrate symmetry (within 10%) with vertical ground reaction force during a drop landing maneuver at this stage of progression for return-to-sport rehabilitation. Although the T-test is standard for the assessment of agility,148-152 it may not adequately measure side-to-side differences because of the equalization of cutting directions in performance of the test. For this reason, the MAT was developed. The MAT may more accurately identify the differences between an athlete’s involved and noninvolved side than the standard t-test. This modification to the standard t-test incorporates four 90-degree cuts isolated to a single direction during the trial. This modification is aimed to isolate involved-side deficits during a multidirectional agility test. Agility testing similar to the MAT can provide high retest reliability (r = 0.94 to 0.98).148 The goal of this test is for the athlete to attain a 10% symmetry in the time taken to complete the task. The ability to generate and maintain single-limb power is important during single-limb cutting maneuvers in sport. Also, improved ability to attenuate force on a single limb and regeneration and redirection of motion may be relevant to a reduction in injury risk in various single-limb positions in sports.62,122,150 The single-limb average peak power test for 10 seconds, which can be used to measure single-limb ground reaction force attenuation and force generation, may be useful to identify athletes who are prone to injury and possible reinjury after ACL reconstruction.41,55,151 To execute this test, the athlete performs a single-limb vertical hop for 10 seconds, and the average of the peak power generated during push-off for each jump is calculated. The

single-limb power hop for 10 seconds demonstrates high within-session (r = 0.97) and moderate between-session reliability (r = 0.76).152 Athletes after ACL reconstruction demonstrate deficits up to 20% or more on their involved limb.142 We recommend that athletes achieve 85% or better side-to-side symmetry in the average force production to progress beyond this stage of rehabilitation. Athletes unable to perform the single-limb power hop with symmetry (<15% deficit) may be affected by residual pain or strength deficits that warrant further rehabilitation before reentry into sport activities.142

Return to Sport After athletes meet the stage IV criteria, they should be prepared to begin reintegration into their respective sport.48 However, we do not suggest that this is the time for unrestricted full participation in competitive events. Rather, it is suggested that athletes resume practice activities and begin to prepare themselves for competitive play. Return to sport after ACL reconstruction can be a high-risk period for athletes both because of the risk for graft failure as well as the increased risk for injury to the contralateral limb, which may be higher than the involved side.7,153 Reinjury to either the contralateral or ipsilateral knee may reach as high as 20% in young athletes who return to competitive activities (Shelbourne KD: Incidence of recurring contralateral or ipsilateral anterior cruciate ligament tears after initial reconstruction in patients 18 years or younger [12 years of data with a minimum of 5 years follow-up]. Personal correspondence, April 2005). However, athletes who attain sports performance symmetry in both limbs before sports reintegration after ACL reconstruction may significantly reduce their potential for future ACL injury.55,70 Successful execution of the suggested criteria of return-to-sport training may more objectively determine an athlete’s readiness to return safely to sports participation. Systematic progression through these objective testing protocols may provide the athlete with both increased neuromuscular control and increased confidence, both of

Rehabilitation and Therapeutic Modalities

which will facilitate successful and safe return to sport after ACL injury.3,5,48

CONCLUSION Late-stage rehabilitation and return-to-sport training after ACL reconstruction without the inclusion of progressive sports-related exercise may fail to address deficits in lower extremity neuromuscular control, strength, and ground reaction force attenuation and production.19,21,62,65,78,113,154 These deficits may continue into competitive play and increase the risk for reinjury or limit the achievement of optimal performance levels. The developed protocol, which incorporates progressive core strengthening and explosive lower extremity plyometric exercise into late-phase rehabilitation, has the potential to target postsurgical deficits and address them through systematic progression during the stages of the return-to-sport training. Ultimately, this approach may translate into successful return to sport; however, long-term outcome studies are necessary to validate the described criteria-based progression and to confirm the relationship of plyometric exercise inclusion into rehabilitation and achieving targeted goals to successful outcomes after the athlete returns to sport.

Acknowledgments The authors would like to acknowledge funding support from National Institutes of Health Grant R01-AR049735. The authors thank Jensen Brent, Carmen Quatman, Jane Kirwan, Catherine Quatman and Kim Foss for the input into the chapter content. The authors would also like to thank the athletic trainers, strength coaches, physical ­therapists, and physicians from Cincinnati Children’s Hospital Sports Medicine Biodynamics Center for their critical input towards the development of the proposed protocols. The authors would also like to acknowledge that portions of the text were reprinted with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association (Alexandria, VA)(Myer, Paterno et al. 2006)and the National Strength and Conditioning Association (Colorado Springs, CO).(Myer, Paterno et al. 2008)�� C

r i t i c a l

P

o i n t s

l Advances in fixation methods and other graft reconstruction techniques have dramatically improved surgical success with anterior cruciate ligament reconstruction. These advances in surgical technique have resulted in consistent quality surgical outcomes. This increased consistency in surgical outcomes may have shifted the limiting factor for an athlete to return to prior level of sports participation to differences in postsurgical rehabilitation rather than variance in surgical outcomes. l Rehabilitation following ACL reconstruction has undergone a relatively rapid and global evolution over the past 25 years. Traditional ACL rehabilitation that once included prolonged immobilization, non-weight bearing, and slow progression to activity now emphasizes immediate motion, early weight bearing, and “accelerated” return

321

to sports participation for athletic patients. Compared to past protocols, rehabilitation programs are now more aggressive and advocate the release of athletes to sports activities as early as 8 weeks after surgery. l There appears an absence of standardized, objective criteria to accurately assess an athlete’s ability to progress through the end stages of rehabilitation and safely return to sport. l Late stage rehabilitation and return to sport training following ACL reconstruction without the inclusion of progressive sports-related exercise may fail to address deficits in lower extremity neuromuscular control, strength, and ground reaction force attenuation and production. l Deficits may continue into competitive play and increase the risk of reinjury or limit the achievement of optimal performance levels. l Return to sport rehabilitation, progressed by quantitatively measured functional goals, may improve the athlete’s integration back into sport participation. l The developed protocol, which incorporates progressive core strengthening and explosive lower extremity plyometric exercise into late phase rehabilitation, has the potential to target postsurgical deficits and address them through systematic progression during the stages of the return to sport training. l Ultimately, this approach may translate into successful return to sports; however, long-term outcomes studies are necessary to validate the described criteria-based progression and to confirm the relationship of plyometric exercise inclusion into rehabilitation and achieving targeted goals to successful outcomes after the athlete returns to sport.

suggested

R eadings

Beynnon BD, Johnson RJ, Fleming BC: The science of anterior cruciate ligament rehabilitation. Clin Orthop 402:9-20, 2002. Chmielewski TL, Myer GD, Kauffman D, Tillman SM: Plyometric exercise in the rehabilitation of athletes: Physiological responses and clinical application. J Orthop Sports Phys Ther 36:308-319, 2006. Hewett TE, Schultz SJ, Griffin LY: American Orthopaedic Society for Sports Medicine: Understanding and Preventing Noncontact ACL Injuries. Champaign, IL, Human Kinetics, 2007. Kvist J: Rehabilitation following anterior cruciate ligament injury: Current recommendations for sports participation. Sports Med 34:269-280, 2004. Lewek M, Rudolph K, Axe M, Snyder-Mackler L: The effect of insufficient quadriceps strength on gait after anterior cruciate ligament reconstruction. Clin Biomech (Bristol, Avon) 17:56-63, 2002. Myer GD, Ford KR, Hewett TE: Tuck jump assessment for reducing anterior cruciate ligament injury risk. Athl Ther Today 13(5):39-44, 2008. Paterno MV, Ford KR, Myer GD, Hewett T: Vertical hopping and isokinetic strength assessment to determine ACL rehabilitation status. J Orthop Sports Phys Ther 36(1):A75, 2006. Paterno MV, Myer GD, Ford KR, Hewett TE: Neuromuscular training improves single-limb stability in young female athletes. J Orthop Sports Phys Ther 34: 305-317, 2004. Risberg MA, Holm I, Myklebust G, Engebretsen L: Neuromuscular training versus strength training during first 6 months after anterior cruciate ligament reconstruction: A randomized clinical trial. Phys Ther 87:737-750, 2007. Wilk KE, Arrigo C, Andrews JR, Clancy WG: Rehabilitation after anterior cruciate ligament reconstruction in the female athlete. J Athl Train 34:177-193, 1999. Wilk KE, Reinold MM, Hooks TR: Recent advances in the rehabilitation of ­isolated and combined anterior cruciate ligament injuries. Orthop Clin North Am 34: 107-137, 2003.

R eferences Please see www.expertconsult.com

322

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A ppe n d i x

5 E - A

Return-to-Sport Phase Training Protocol   and Teaching Cues Stage I Dynamic Stabilization and Core Strengthening ­Session 1

Time (sec)

Jogging gait retraining (treadmill @ 7 mph 5% grade) Deep hold position Box butt touch squat Line jump (forward)—deep hold Line jump (lateral)—deep hold Single-leg Airex balance (knee slightly flexed) Single-leg squat hold BOSU (flat)—deep hold Single-leg dumbbell bend-over deadlift (focus on balance) Walking lunges BOSU (round) bilateral knee balance BOSU (round) crunches BOSU (round) swivel crunch (feet planted) BOSU (round) double-leg pelvic bridges BOSU (round) superman Running mechanics (treadmill @ 8 mph 10% grade) Running mechanics (treadmill @ 9 mph 10% grade)

20 5 5 5 10 5 5 20

15 15

Stage I Dynamic Stabilization and Core Strengthening Session 2

Time (sec)

Jogging gait retraining (treadmill @ 7 mph 5% grade) BOSU (flat) deep hold partner perturbations BOSU (flat) drop off—deep hold BOSU (flat) rapid squat—deep hold BOSU (flat) athletic position—partner ball toss BOSU (round) single-leg step hold Single-leg Airex step (front/back)—hold Single-leg Airex step (side/side)—hold BOSU (round) single knee—hold BOSU (flat) single straight leg bend-over Lateral stepping with band resistance Wall squats with Swiss ball BOSU (round)—reverse crunches BOSU (round)—swivel ball touches (feet up) BOSU (round)—trunk extensions Running mechanics (treadmill @ 8 mph 10% grade) Running mechanics (treadmill @ 9 mph 10% grade)

20 20 5 5 20 5 5 5 20

15 15

Stage II Functional Strength Session 1

Time (sec)

Jogging gait retraining (treadmill @ 7 mph 0% grade) Box drop off—deep hold BOSU (round) jump-up—deep hold BOSU (flat) single-leg squat—hold 12-inch box lateral step-down (heel touch) Split squats BOSU (round) single-leg step—hold Double-leg bend-over deadlift Sumo squat dumbbell pick-up Resisted lateral shuffling Table double crunch Table double swivel crunch Prone table manual resisted hip extension BOSU (round) swimmers BOSU (round) single-leg pelvic bridges Resistive band running (heavy resistance) Resistive band running (light resistance)

20 5 5 5 5

15 10

Repetitions 5 8 8 4 4 6 8 12 16 steps 25 40 12 15

Repetitions

8 8 6 4 4 10 20 steps 12 25 40 12

Repetitions 10 10 6 12 10 6 10 10 6 Passes 15 8 12 10 12

Sets 2 1 1 1 R and L R and L R and L 1 R and L 2 2 2 1 1 1 3 3

Sets 2 3 1 1 3 R and L R and L R and L R and L R and L R and L 2 2 1 1 3 3

Sets 2 1 1 R and L R and L 2 R and L 2 2 R and L 2 R and L 2 R and L R and L 4 4

Rehabilitation and Therapeutic Modalities Stage II Functional Strength Session 2

Time (sec)

Jogging gait retraining (treadmill @ 7 mph 0% grade) BOSU (round) jump-up—deep hold Single-leg Airex hop (front/back)—hold Single-leg Airex hop (side/side)—hold Double BOSU (flat) rapid squats—deep hold Single-leg X hop 12-inch box Airex lateral step-down (heel touch) Split squats Supine Swiss ball hamstring curl Lateral lunges BOSU (flat) single-leg balance—hold Table double crunch Table double swivel crunch BOSU (round) lateral crunch BOSU (round) toe-touch, swimmers Retrograde training (treadmill @ 3-4 mph 10% grade) Retrograde training (treadmill @ 4-5 mph 5% grade)

20 5 5 5 3

10

20 12

Stage III Power Development Session 1

Time (sec)

Jogging gait retraining (treadmill @ 8 mph 0% grade) Wall jumps Line jumps (side/side)—speed Line jumps (front/back)—speed Line jumps—maximal vertical (four-way) 180-degree jumps (height) BOSU (flat) drop off, single-leg—hold BOSU (round) jump-up, single-leg—hold Single-leg X hop (reaction) Barbell back squats Assisted Russian hamstring curls BOSU (round) butt balance (feet up) partner ball toss BOSU (round) V-sit partner toe-touch Table lateral crunch BOSU (round) toe-touch, swimmers, partner perturbations Bounding in place Running mechanics (treadmill @ 8-10 mph 15% grade)

20 15 10 10 10 5 10

25

15 10

Stage III Power Development Session 2

Time (sec)

Jogging gait retraining (treadmill @ 8 mph 0% grade) BOSU (flat) drop off, 75% maximal vertical Tuck jumps Broad jump, jump—deep hold Broad jump, maximal vertical Single-leg 90-degree hop—hold Crossover hop, hop, hop (distance)—athletic position BOSU (round) single-leg (four-way) hop—hold Dumbbell bent-leg deadlift pick-up Band good mornings BOSU (flat) single-leg maximal depth squat (opposite leg extended forward) BOSU (flat) single-leg hold (partner perturbations) BOSU (round) double crunch BOSU (round) opposite swivel crunch (feet up) Swiss ball trunk extensions Retrograde training (treadmill @ 4-6 mph 5% grade) Retrograde training (treadmill @ 4-8 mph 0% grade)

20 10 3 3 3 3

10

10 12

Repetitions 10 6 6 12 3 10 10 10 15 steps 6 15 8 10 10

Repetitions

3 5 5 4 8 8 10 8 10

Repetitions 8 8 6 8 4 2 8 12 8 4 15 12 12

Sets 2 1 R and L R and L 1 R and L R and L 2 2 R and L R and L 2 R and L R and L R and L 3 3

Sets 2 2 1 1 1 2 R and L R and L R and L 2 2 2 2 R and L R and L 2 3

Sets 2 1 2 1 1 R and L R and L R and L 2 2 R and L R and L 2 R and L 1 3 3

323

324

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Stage IV Sport-Performance Symmetry Session 1

Time (sec)

Ground base warm-up (carioca, lateral shuffle, forward jog, backward jog) Box drop off, athletic position Wall jumps Tuck jumps Lunge jump 180-degree jump—broad jump Power box steps Bounding for distance Box drop off—reaction Broad jump, maximal vertical—reaction step Forward barrier jumps—reaction Forward barrier jumps with middle box—reaction Box drop off, maximal vertical—reaction step Assisted Russian hamstring curl Partner-assisted single-leg box butt touch squats Four corners drill W-drill

Stage IV Sport-Performance Symmetry Session 2

15 10 10 10

Time (sec)

Ground base warm-up (carioca, lateral shuffle, forward jog, backward jog) Box drop off 180 degrees—reaction Wall jumps 10 Tuck jumps 8 Jump into bounding Box drop off, maximal vertical Box drop off, maximal broad jump—athletic position Hop, hop, hop (distance)—hold 3 Crossover hop, hop, hop (distance)—athletic position Forward barrier hops with staggered box—reaction Lateral barrier hops with staggered box—reaction Box drop off 180 degrees, box touch maximal vertical—reaction Lateral box drop off, maximal vertical Assisted Russian hamstring curl Dumbbell overhead squats Wheel drill V-drill

Repetitions

Sets

10 6 8 6 6 6 10 8 8 6 6

4 1 2 2 R and L 2 R and L 1 1 1 1 1 1 2 R and L 1 1

Repetitions

Sets

5

4 1 2 2 1 1 1 R and L R and L 1 R and L 1 R and L 2 2 1 1

5 6 6 6 4 5 6 4 6 6 8 8 6 6

A ppe n d i x

5 E - B

Glossary 12-Inch Box Airex Lateral Step-Down (Heel Touch) (see Fig. 5E-7):  The athlete balances on one leg on a 12-inch box with an Airex pad placed on top of the box. With the contralateral foot dorsiflexed, the involved knee is flexed until the contralateral heel makes contact with the surface of the floor, trying to keep the hips level; then the athlete ascends back up to starting position. 12-Inch Box Lateral Step-Down (Heel Touch) (see Fig. 5E-7):  The athlete balances on one leg on a 12-inch box. With the contralateral foot dorsiflexed, the involved knee

is flexed until the contralateral heel makes contact with the surface of the floor; then the athlete ascends back up to starting position. 180-Degree Jump, Broad Jump:  The jump is initiated by a direct vertical motion combined with a 180-degree rotation. Once landed, a broad jump is immediately initiated to achieve maximal horizontal distance. 180-Degree Jumps (Height):  The jump is initiated by a direct vertical motion combined with a 180-degree

Rehabilitation and Therapeutic Modalities

325

r­ otation. Once landed, the jump is initiated immediately to the opposite direction.

then drops off the BOSU and lands on the same leg with the knee flexed.

Airex:  Two-inch foam balance pad. (Perform Better Inc., 11 Amflex Dr., PO Box 8090, Cranston, RI 02920-0090)

BOSU (Flat) Single-Leg Balance—Hold:  The athlete assumes a single-leg stance on the flat side of the BOSU with knee and hip flexed and attempts to maintain this ­position for the duration of the exercise.

Assisted Russian Hamstring Curl (see Fig. 5E-23):  The athlete begins in a kneeling position with a partner providing foot support and torso support (with band assistance). The athlete extends at the knee while maintaining a neutral spine. The rehabilitation specialist should provide enough assistance so that the exercise can be performed without flexing at the hip. Athletic Position (see Fig. 5E-15):  The athletic ­position is a functionally stable position with the knees comfortably flexed, shoulders back, eyes up, feet about shoulder width apart, and body mass balanced over the balls of the feet. The chest should be aligned over the knees, which are over the balls of the feet. This is the athlete ready position and should be the starting and finishing position for most of the training exercises. During some of the exercises, the finishing position is overexaggerated with deeper knee flexion to emphasize the correction of certain biomechanical deficiencies. Band:  Resistive tubing, heavy Thera-Band or Jump Stretch band. (Jump Stretch Inc., Youngstown, Ohio) Band Good Mornings (see Fig. 5E-22):  The athlete positions the resistive band below the 7th cervical vertebra and stands with both knees slightly flexed and feet on band. The movement is initiated with the trunk flexed to about 90 degrees. A neutral spine should be maintained as the athlete extends the trunk from 90 to 0 degrees (an erect posture). BOSU:  Double-sided balance device. (Team BOSU, 1400 Raff Rd, Canton, OH 44750) BOSU (Flat):  Flat side of domed balance apparatus is turned upward. BOSU (Flat) Deep Hold Partner Perturbations (see Fig. 5E-3):  The athlete balances in deep hold position while standing on flat surface of a BOSU while the clinician ­perturbs the BOSU or the torso of the athlete. BOSU (Flat) Drop Off, 75% Maximal Vertical:  The athlete begins standing on the flat side of the BOSU in athletic position, then drops off the BOSU simultaneously with both feet and, upon landing on the ground, performs a vertical jump with 75% of maximal effort. BOSU (Flat) Drop Off—Deep Hold:  The athlete begins standing on the flat side of the BOSU in athletic position, then drops off the BOSU simultaneously with both feet and, upon landing on the ground, immediately assumes the deep hold position. BOSU (Flat) Drop Off—Single-Leg Hold:  The athlete begins standing on one leg on the flat side of the BOSU,

BOSU (Flat) Single-Leg Hold (Partner Pertur­ bations):  The athlete assumes a single-leg stance on the flat side of the BOSU with knee and hip flexed and ­attempts to maintain this position for the duration of the exercise while a partner or trainer perturbs the BOSU. BOSU (Flat) Single-Leg Maximal Depth Squat (Opposite Leg Extended Forward):  The athlete assumes a single-leg stance on the flat side of the BOSU with knee and hip flexed as much as possible within the limits of control and attempts to maintain this position for the duration of the exercise. Opposite leg is to be extended forward during the exercise. BOSU (Flat) Single-Leg Squat—Hold:  The athlete assumes a single-leg stance on the flat side of the BOSU and attempts to squat to a position with the knee flexed to 90 degrees and torso erect and then returns to the original position. BOSU (Flat) Single Straight Leg Bend-Over (see Fig. 5E-8):  Balancing on one leg on the flat side of the BOSU with knee slightly flexed and maintaining neutral spine, the athlete flexes the trunk to 90 degrees, reaching for the front of the BOSU. BOSU (Flat), Athletic Position, Partner Ball Toss:  The athlete begins standing with both feet on the flat side of the BOSU in athletic position, and a ball is tossed between the athlete and the partner or trainer. When tossing the ball to the athlete, attempt to place it in positions that will perturb the athlete’s center of mass. BOSU (Flat)—Deep Hold:  The athlete assumes the deep hold position while standing on the flat side of the BOSU. BOSU (Flat) Rapid Squat—Deep Hold:  The athlete rapidly descends into a parallel squat position with the feet shoulder width apart on the flat side of the BOSU. BOSU (Round):  The round side of the domed balance apparatus is turned upward. BOSU (Round) Bilateral Kneel:  The athlete begins this exercise by balancing in a kneeling position with the knees shoulder width apart in the middle of the round side of the BOSU. The athlete will maintain this balanced position with the hips slightly flexed for the duration of the exercise. BOSU (Round) Butt Balance (Feet Up) Partner Ball Toss:  The athlete begins sitting on the round side of the BOSU in a balanced position (see Fig. 5E-20) with the feet

326

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

held in the air. A trainer or partner provides perturbations by tossing a ball back and forth with the athlete. BOSU (Round) Crunches:  The athlete begins sitting on the round side of the BOSU in a balanced manner with the feet planted on the ground. The exercise is performed by extending the spine in such a way that the athlete allows the back to touch the ground followed by flexing the spine to allow the elbows to touch the knees. BOSU (Round) Double Crunch:  The athlete starts by sitting balanced on the round side of the BOSU and then flexes the trunk simultaneously with hip flexion. BOSU (Round) Double-Leg Pelvic Bridges:  The athlete lays supine with the hip and knees flexed and the feet planted on the round side of the BOSU. The athlete then extends the hips and elevates the trunk off the ground to execute a pelvic bridge. This position should be held for 3 seconds before the next repetition (see Fig. 5E-8 for ­singleleg pelvic bridge). BOSU (Round) Jump-Up—Deep Hold:  The athlete starts on the ground and jumps onto the round side of the BOSU and lands in a deep hold position. BOSU (Round) Jump-Up, Single-Leg—Hold:  The athlete starts on single leg on the ground and jumps up onto round side of the BOSU and lands on that same leg with the knee flexed. BOSU (Round) Lateral Crunch:  The athlete starts lying on the side with hip located in the center of the round side of the BOSU. The athlete’s feet and legs must be anchored during this exercise by the trainer or a stationary object. The athlete proceeds to bend laterally at the waist back and forth for the prescribed repetitions. BOSU (Round) Opposite Swivel Crunch (Feet Up):  The athlete begins sitting on the round side of the BOSU in a balanced position with the feet held in the air (similar to Fig. 5E-20). The athlete begins the exercise by twisting the trunk so that he or she can touch the ground with the hands. The movement is reversed, and the athlete swivels the torso to touch the ground on the other side of the body. BOSU (Round) Single-Knee—Hold:  The athlete begins this exercise by balancing in a kneeling position with one knee directly in the middle of the round side of the BOSU and the other knee extended out to the side. The athlete maintains this balanced position with the hip slightly flexed for the duration of the exercise. BOSU (Round) Single-Leg (Four-Way) Hop—Hold:  The athlete starts in a single-leg athletic position immediately behind the BOSU. The athlete hops forward onto the round side of the BOSU and lands in a balanced position. After achieving a balanced single-leg stance on the BOSU, the athlete proceeds to hop off the BOSU laterally and assumes this same stance on the floor immediately next to the BOSU. The athlete then continues to hop on and

off the BOSU, achieving a balanced athletic position, in each of the four directions: forward, backward, lateral, and ­medial. BOSU (Round) Single-Leg Pelvic Bridges (see Fig. 5E8):  The athlete lies supine with the hip and knees flexed and a single foot planted on the round side of the BOSU and the contralateral leg fully extended. The athlete then extends the hips and elevates the trunk off the ground to execute a pelvic bridge. This position should be held for 3 seconds before repeating the next repetition. BOSU (Round) Single-Leg Step—Hold:  The athlete starts off of the BOSU in athletic position. The movement begins with the athlete stepping onto the round side of the BOSU and continuing to balance with knee flexed to about 90 degrees. BOSU (Round) Superman:  The athlete begins in prone position with the arms overhead and legs extended and abdomen centered on the round side of the BOSU. The movement is initiated by extending the hip and trunk while maintaining the shoulders in flexed position. The position is held for 3 seconds and then repeated. BOSU (Round) Swimmers:  The athlete begins in prone position with the abdomen centered on the round side of the BOSU and with the arms overhead and legs extended. The movement is initiated by elevating the opposite arm and leg and is held for 3 seconds. BOSU (Round) Swivel Crunch (Feet Planted):  The athlete starts out balancing supine on the round side of the BOSU with the lower back/butt centered on the BOSU. The athlete rotates at the spine while flexing the trunk for the crunch. BOSU (Round) Toe-Touch, Swimmers:  The athlete begins in a prone position with the abdomen centered on the round side of the BOSU and the arms overhead and legs extended. The athlete reaches back with one arm to touch the opposite foot and returns to the outstretched ­superman position. BOSU (Round) Toe-Touch, Swimmers, Partner Perturbations:  The athlete begins in a prone position with the abdomen centered on the round side of the BOSU and the arms overhead and legs extended. The rehabilitation specialist should perturb the BOSU while the athlete reaches back with arm to touch the opposite foot, performing swimmers technique. BOSU (Round) V-Sit Partner Toe-Touch:  The ­athlete starts out on the round side of the BOSU with the lower back/butt centered on the BOSU, leaning the shoulders back on the floor with the arms reaching overhead and the feet extended upward at a 45-degree angle. The partner holds the feet and gives support while the athlete crunches forward, reaching to touch the toes. BOSU (Round) Reverse Crunches:  The athlete starts out balancing supine on the BOSU with the lower

Rehabilitation and Therapeutic Modalities

back/butt centered on the BOSU. The athlete flexes the hip while attempting to maintain a balanced position on the BOSU. BOSU (Round) Swivel Ball Touches (Feet Up):  The athlete starts by sitting balanced on the round side of the BOSU with the feet up and with the athlete leaning ­slightly back. The athlete rotates at the spine while flexing the trunk for the crunch. BOSU (Round) Trunk Extensions:  The athlete begins in a prone position on the round side of the BOSU and performs the exercise by extending the upper torso. Bounding (see Fig. 5E-18):  The athlete jumps hori­ zontally off one foot and lands on the other. Once proper rhythm is attained, the vertical component of the bound should be maximized. Box Butt Touch:  A box is placed behind the athlete, and the athlete starts with the feet shoulder width apart and performs a squat down to the height of the box, softly touches the box without resting, and then ascends up to initial starting position. Box Drop Off 180 Degrees—Reaction:  The athlete drops off the box, performs a 180-degree jump, lands in an athletic position, and follows with a lateral reaction to a cue such as the rehabilitation specialist pointing out a random cut direction, using defender reaction cut or ball retrieval drills. Box Drop Off 180 Degrees, Box Touch Maximal ­Vertical—Reaction:  The athlete drops off the box, performs a 180-degree jump, and then jumps back up on the box and immediately drops down forward off the box, performs a maximal vertical jump, lands in an athletic position, and follows with a reaction to a cue such as the rehabilitation specialist pointing out a random cut direction, using ­defender reaction cut or ball retrieval drills. Box Drop Off, Athletic Position:  The athlete drops down from a box, landing with both feet simultaneously in the athletic position (see Fig. 5E-15). Box Drop Off—Deep Hold:  The athlete drops down from a box landing with both feet simultaneously in the deep hold position (ending position of Fig. 5E-16). Box Drop Off, Maximal Broad Jump, Athletic ­Position:  The athlete drops down from a box, landing with both feet simultaneously and immediately jumping horizontally to achieve maximal horizontal distance. The ­athlete should stick the landing in athletic position. Box Drop Off, Maximal Vertical:  The athlete drops down from a box landing with both feet simultaneously in the athletic position and immediately performs a maximal vertical jump and lands in the athletic position. Box Drop Off, Maximal Vertical—Reaction Step:  The athlete drops down from a box, landing with both feet

327

s­ imultaneously in the athletic position. Immediately ­after landing, the athlete performs a maximal vertical jump, lands in the athletic position, and reacts to the rehabilitation specialist’s directional cue with a submaximal effort cut. Focus is on the desired technical performance and not speed of movement. Box Drop Off—Reaction:  The athlete drops off, lands in athletic position, and follows with a reaction to a cue such as the rehabilitation specialist pointing out a random cut direction, using defender reaction cut or ball retrieval drills. Broad Jump, Maximal Vertical—Reaction Step:  The jump is initiated horizontally to achieve maximal horizontal distance. Immediately after landing, the athlete performs a maximal vertical jump (see Fig. 5E-17), lands in athletic ­ position, and reacts to the rehabilitation specialist’s ­ directional cue with a submaximal effort cut. Focus is on the desired technical performance and not speed of ­movement. Broad Jump, Jump—Deep Hold (see Fig. 5E-16):  The athlete prepares for this jump in the athletic position with the arms fully extended behind the back at the shoulder. The athlete begins by swinging the arms forward and jumping horizontally to achieve maximal horizontal distance. The athlete must stick the landing with the knees flexed to about 90 degrees in an overexaggerated athletic position. The athlete may not be able to stick the landing during a maximal effort jump in the early phases. In this situation, have the athlete perform a submaximal broad jump and stick the landing with the toes straight ahead and no inward motion of the knees. As the athlete’s technique improves, encourage him or her to add distance to the jumps, but not at the expense of technique ­perfection. Broad Jump, Maximal Vertical:  The athlete performs a broad jump, immediately progresses into a maximal effort vertical jump, and lands in athletic position. When teaching this jump, the athlete may have a tendency to “float” in a horizontal direction during the vertical jump; in this case, encourage the athlete to quickly transfer from the broad to vertical jump. Crossover Hop, Hop, Hop (Distance)—Athletic Position:  The starting position for this jump is with the athlete in a semi-crouched position on the single limb being trained. The arms should be fully extended behind the athlete at the shoulder. The athlete initiates the hop by swinging the arms forward while simultaneously extending at the hip and knee. The hop should carry the athlete up at a 45degree angle laterally toward the opposite leg and should be for maximal distance. The athlete lands on the leg opposite of the initial stance leg, then immediately hops in a 45­degree angle laterally toward the other leg. This is repeated for one hop, with the exception that the final landing is on two feet, and the athlete maintains the athletic position. Deep Hold Position:  The athlete squats with the feet shoulder width apart and holds a position with the knees flexed to 90 degrees and the torso erect.

328

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Double BOSU (Flat) Rapid Squats—Deep Hold:  The athlete places each foot on the flat side of separate BOSUs. The athlete then rapidly descends into a parallel squat position with the feet shoulder width apart, then ascends slowly back to the start position. Double-Leg Stretch Bend-Over Deadlift:  With the knees slightly flexed and a neutral spine, the athlete flexes the trunk to 90 degrees. The weight is held in front of the shins and targeted to the shoe tops. Dumbbell Bent-Leg Deadlift Pick-Up:  The athlete starts in a stance that has the feet twice shoulder width apart, then descends in squat position to pick up the dumbbell. The athlete secures the dumbbell in an alternated grip and ascends upward. Dumbbell Overhead Squats:  The athlete holds the dumbbells overhead and squats to 90 degrees of knee flexion while maintaining the dumbbells in the overhead ­position. Four Corners Drill:  Four cones are lined up in a shape of a square about 5 yards apart in each direction. The athlete performs the basic pattern of sprint to first cone, lateral slide to second cone, backward sprint to third cone, and lateral slide to first cone. Ground Base Warm-Up (Carioca, Lateral Shuffle, Forward Jog, Backward Jog)—Hold:  The knee is flexed to greater than 60 degrees in single-leg stance and flexed to greater than 90 degrees in bipedal stances. The athlete must stabilize the center of mass and maintain postural stability during the specific agility activity, for the prescribed durations. Hop:  Single-leg jump. Hop, Hop, Hop (Distance)—Hold:  The athlete performs three single-leg hops for distance with no pause between jumps and performs a hold at the end. Jump:  Double-leg jump with the feet shoulder width apart. Jump into Bounding:  The athlete begins by doing a single maximal effort broad jump. After landing on a single leg, the athlete should immediately begin the bound exercise. The bounding should emphasize achieving vertical height with minimal horizontal distance. Coach the athlete to drive the non–weight-bearing leg forward and vertically to help achieve the maximal vertical height. Do not allow the athlete to perform an exaggerated strideout jog. Lateral Box Drop Off—Maximal Vertical:  The athlete starts with both legs on a box (12 inches or less) and drops off the box laterally with both legs simultaneously, landing with both feet shoulder width apart, then immediately performs a maximal vertical jump and lands in athletic ­position.

Lateral Lunges:  The athlete starts standing with the feet shoulder width apart. The athlete lunges with one foot out at a 45-degree angle and returns to the starting position. Line Jumps (Side/Side)—Speed:  The athlete prepares for this exercise by standing with the feet close together and the knees slightly bent on one side of the line. The athlete jumps sideways over the line, keeping the knees bent and staying close to the line. After landing on the opposite side, the athlete immediately redirects back to the initial position. This sequence is repeated as quickly as possible while maintaining proper form. When teaching this exercise, encourage the athlete to achieve as many repetitions as possible in the allotted time by jumping close to the line, shortening the ground contact time, and not using excessive height on the jumps. Do not allow the athlete to perform a double hop on the side of the line. Early in the training, the athlete may focus on the line. As the athlete’s technique improves, encourage the athlete to shift his or her visual focus away from the line to outside cues. Line Jumps (Front/Back)—Speed:  The athlete prepares for this exercise by standing with the feet close together and the knees slightly bent behind the line. The athlete jumps forward over the line keeping the knees bent and staying close to the line. After landing on the opposite side, the athlete immediately redirects back to the initial position. This sequence is repeated as quickly as possible while maintaining proper form. Teach this jump by having the athlete keep the eyes up as much as possible. Looking down at the line will cause the athlete to lean too far forward on the forward jump, making it difficult for the athlete to redirect backward. The athlete can improve speed and efficiency of this jump by learning to maintain the core center of gravity control and by preparing to change direction in midflight. Encourage the athlete to jump directly over the line and not around the sides. Lunge Jump:  The athlete starts in an extended stride position with the hips pushed forward and the front knee positioned directly above the ankle and flexed to 90 degrees. The back leg is fully extended at the hip and knee, providing minimal support for the stance. The athlete should jump vertically off of the front support leg, maintaining the starting position during flight and landing. The jump is repeated as quickly as possible while still achieving maximal vertical height. To coach this jump, encourage the athlete to keep the back leg straight and use it only for balance support. Vertical power is obtained by the front leg. Stance support percentages are 90% for the front leg and 10% for the back. Partner-Assisted Single-Leg Box Butt Touch Squats:  A box is placed behind the athlete, and a band (held by a partner and the athlete) is provided to assist with the exercise. The athlete starts on a single leg, performs a squat down to the height of the box, softly touches the box without resting, and then ascends up to initial starting ­position.

Rehabilitation and Therapeutic Modalities

Power Box Steps:  The athlete stands with the ball of one foot on top of the 6- to 12-inch box. The athlete performs a maximal effort vertical hop up and off of the box using the foot that was placed on the box and landing on both feet in the athletic position. Prone Table Manual Resisted Hip Extension:  The athlete begins in a prone position with the pelvis and lower extremity stabilized on the table and the trunk flexed forward off the edge of the table with the hands on the floor in front of him or her. The movement is initiated by extending hip and trunk to a neutral position while maintaining shoulders in an overhead position. Hold the position for 3 seconds and repeat. Reaction:  The athlete reacts to a cue such as the rehabilitation specialist pointing out a random cut direction, using defender reaction cut or ball retrieval drills. Reaction Step:  The athlete reacts to the rehabilitation specialist’s directional cue with a submaximal effort cut. Focus is on the desired technical performance and not speed of movement. Resisted Lateral Shuffling:  The athlete begins in athletic position with a resistive Thera-Band anchored to the ankles. The athlete is instructed to maintain the athletic position and shuffle in the prescribed direction. The rehabilitation specialist can have the athlete move quickly during exercise or can use increased resistance and have the athlete move slower and more directed to focus on ­improved strength. Single-Leg Airex Balance (Knee Slightly Flexed):  The athlete balances on a single leg with the knee slightly flexed and attempts to maintain postural stability for the duration of the exercise. Single-Leg Airex Hop (Front/Back)—Hold:  The athlete starts behind the Airex pad and hops up onto the Airex. The athlete should maintain balance and hold the knee in a flexed position. The athlete then hops forward off the Airex, maintains balance with the knee in a flexed position, and then hops backward onto the Airex pad. After regaining balance and holding the knee in a flexed position, the athlete hops backward off the Airex onto the ground and maintains balance in a flexed knee position. Single-Leg Airex Hop (Side/Side)—Hold:  The athlete starts on one side of the Airex pad and hops laterally onto the Airex. The athlete should maintain balance and hold the knee in a flexed position. The athlete then hops off the other side of the Airex onto the ground, maintains balance, and then repeats the exercise in the other direction. Single-Leg Dumbbell Bend-Over Deadlift (Focus on Balance):  Balancing on one leg with the knee slightly flexed and maintaining a neutral spine, the athlete flexes the trunk to 90 degrees. The weight is held in front of the athlete’s shins and targeted to the shoe top as the athlete descends during the exercise.

329

Single-Leg Hop—Hold (see Fig. 5E-19):  The starting position for this jump is with the athlete in a semicrouched position on the single limb being trained. The arms should be fully extended behind the athlete at the shoulder. The athlete initiates the jump by swinging the arms forward while simultaneously extending at the hip and knee. The jump should carry the athlete up at a 45-degree angle and provide the maximal distance the athlete can handle while still maintaining an upright stance on the single landing. The landing is on the jumping leg and occurs with deep knee flexion (to 90 degrees). The landing should be held for a minimum of 3 seconds to be counted as a successful landing. Coach this jump with care to protect the athlete from injury. Start the athlete with a submaximal effort on the single-leg broad jump so that the athlete can experience the difficulty of the jump. Continue to increase the distance of the broad jump as the athlete improves his or her ability to stick and hold the final landing. Have the athlete keep his or her focus away from the feet to help prevent too much forward lean. Single Leg Squat—Hold:  The athlete squats on single leg, attempting to achieve 90 degrees or more of knee ­flexion. Single-Leg X Hop:  The athlete begins facing a quadrant pattern standing on a single limb with the support knee slightly bent. The athlete hops diagonally, landing in the opposite quadrant, maintaining forward stance, and holding the deep knee flexion landing for 3 seconds. The athlete then hops laterally into the side quadrant, again holding the landing. Next the athlete hops diagonally backward, holding the landing. Finally, the athlete hops laterally into the initial quadrant holding the landing. The athlete should repeat this pattern for the required number of sets. Encourage the athlete to maintain balance during each landing, keeping the eyes up and focusing away from the feet. Single-Leg X Hop—Reaction:  The athlete performs the single-leg X hop as described previously with the exception that each landing must be held until the athlete receives an unanticipated cue from the rehabilitation specialist to hop to the next quadrant. Split Squats:  The athlete starts in lunge stance with full support on the front limb with the opposite limb resting on a box behind. The athlete then squats to 90 degrees of knee flexion on the front limb. Encourage the athlete to lunge the front limb far enough out so that the knee does not cross over the ankle when performing the squat exercise. Sumo Squat Dumbbell Pick-up (see Fig. 5E-9):  The stance is wide (about double shoulder width) so that weight can be lifted between the legs. The athlete should focus on maintaining an upright posture with minimized trunk flexion when descending to pick up the dumbbell. After reaching the dumbbell, the athlete grasps it an alternated grip and ascends back to the start position. Supine Swiss Ball Hamstring Curl:  The athlete begins lying in a supine position with the shoulders and back on

330

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the floor, with the hips flexed and both feet on top of the Swiss ball. The athlete then extends at the hip and flexes at the knee, attempting to pull the heels to the buttock. Swiss Ball Bilateral Kneel:  The athlete kneels and balances on the Swiss ball with the feet off the ground. A spotter should be available at all times in front of the athlete. Swiss Ball Hip Extensions:  The athlete begins in a prone position on the Swiss ball with the hands and elbows on the floor in front. The movement is initiated by extending both hips while maintaining the shoulders in flexed position. Hold the torso and lower extremity in the overhead position for 3 seconds and repeat. Table Double Crunch:  The athlete starts out supine on a table and flexes trunk simultaneous with hip flexion. Table Double Swivel Crunch:  The athlete starts in a supine position on a table with the arms placed on the back of the head. The athlete flexes the trunk simultaneously with hip flexion, and as the trunk and hip are maximally flexed, the athlete rotates at the trunk, touching each elbow to the opposite knee. Table Lateral Crunch   (see Fig. 5E-21):  The athlete starts lying on the side with the hip located at the edge of the table. The athlete’s feet and legs must be anchored during this exercise by the trainer or a stationary object. The athlete proceeds to flex and extend laterally at the waist for the prescribed repetitions. Tuck Jumps (see Fig. 5E-14):  The athlete starts in the athletic position with the feet shoulder width apart. The athlete initiates a vertical jump with a slight crouch downward while extending the arms behind. The athlete then swings the arms forward while simultaneously jumping straight up and pulling the knees up as high as possible. At the highest point of the jump, the athlete is positioned in the air with the thighs parallel to the ground. When landing, the athlete immediately begins the next tuck jump. Encourage the athlete to land softly, using a toe-to-midfoot rocker landing. The athlete should not continue this jump if he or she cannot control the high landing force or keep the knees aligned while landing.

V-Drill:  The athlete starts at the base of three cones that are set up in a “V” shape 5 to 8 yards apart. The pattern is initiated by sprinting to the left cone, backpedaling back to the middle cone, turning 90 degrees, sprinting to the right cone, and then backpedaling to the starting position. Walking Lunges:  The athlete performs a lunge and instead of returning to the start position, stepping through with the back limb and proceeding forward with a lunge on the opposite limb. Encourage the athlete to lunge the front limb far enough out so that the knee does not advance beyond their ankle during the exercise. Wall Jumps:  The athlete stands erect with the arms semiextended overhead. The athlete then executes ­repeated quick vertical jumping while reaching upward. This vertical jump requires minimal knee flexion because the gastrocnemius and soleus muscles create the vertical height. The arms should extend fully at the top of the jump. Use this jump as a warm-up and an important interactive coaching exercise because this relatively low-intensity movement can reveal abnormal knee motion in athletes with poor side-to-side knee control. Wall Squats with Swiss Ball:  A squat exercise that is performed with the aid of a Swiss ball positioned between the back and a stable wall. W-Drill:  The athlete starts at the left-hand side of five cones that are positioned in the shape of a “W.” The athlete first backpedals at a 45-degree angle to the next cone, turns and sprints to the next cone, and repeats through the series of cones. Wheel Drill:  The athlete stands next to a cone that is encircled by seven other cones that are 3 to 5 yards away. The athlete moves through the cones using a series of sprints, lateral slides, and backpedals. The athlete should keep the shoulders square to the starting position during the drill.

Rehabilitation and Therapeutic Modalities

A ppe n d i x

331

5 E - C

Tuck Jump Assessment Scoring Sheet Athletes perform repeated tuck jumps for 10 seconds while the clinician scores the performance for each category on the visual analog scale. Each category is measured in centimeters from the NEVER (“0” score) to the ALWAYS (“10” score) with each incremental increase

(1 cm) toward ALWAYS equal to 1-point increase in their score. Scores for each category are summed to provide the “total” score for the assessment. The scores will range from 0 to 80 total points.

Tuck Jump Assessment KNEE AND THIGH MOTION (1) KNEES NEUTRALLY ALIGNED AT LANDING NEVER

ALWAYS

(2) THIGHS REACH PARALLEL (Observed at highest point of jump) NEVER ALWAYS

(3) THIGHS EQUAL SIDE-TO-SIDE (Throughout sequence) NEVER

ALWAYS

FOOT POSITION (4) FOOT PLACEMENT SHOULDER WIDTH APART NEVER

ALWAYS

(5) FOOT PLACEMENT NOT STAGGERED (Side view) NEVER

ALWAYS

(6) TOE-TO-MIDFOOT ROCKER UTILIZED (No heel strike) NEVER

ALWAYS

PLYOMETRIC TECHNIQUE (7) RAPID REBOUND BETWEEN JUMPS (No visible delay) NEVER

ALWAYS

(8) LANDS IN SAME FOOTPRINT (From point of take-off) NEVER

ALWAYS

C H A P T E R�  

  �6

Principles of Injury Prevention Holly J. Silvers, Roald Bahr, Eric Giza, and Robert G. Watkins IV Athletic participation has a multitude of positive benefits, including, but not limited to, weight control, decreased hypertension, decreased systemic disease, lower rates of teen pregnancy, and an overall sense of psychological wellbeing. There is strong scientific evidence that supports the effectiveness of regular physical activity in the primary and secondary prevention of several chronic diseases, such as cardiovascular disease, diabetes, cancer, hypertension, obesity, depression, osteoporosis, and premature death.1 The positive impact of athletic participation—competitive or recreational—cannot be overstated. However, there is a direct correlation between the increased level of participation in sports and the number of sports-related injuries that are incurred by those participants.2,3 According to a prospective study analyzing injury rates nationally between 1997 and 1999, an estimated 7 million Americans received medical attention each year for sports-related injuries (25.9 injuries per 1000 population).4 As sports participation continues to be promoted as part of a healthy lifestyle, sports-related injuries are becoming quite ubiquitous and remain paramount in the minds of the medical community. Prevention efforts aimed at reducing the rate of injury continue to be investigated as a result of this surge

in injury. This chapter examines several common injuries related to sports participation and the prevention of such: anterior cruciate ligament injuries, foot and ankle injuries, hamstring injuries, and sports-related spine injury. Collectively, the authors hope to shed some light on the subject of injury and prevention and hope to inspire a continued effort toward investigating the notion of injury prevention across sport, age, and gender.

S U G G E S T E D

R E A D I N G S

Conn JM, Annest JL, Gilchrist J: Sports and recreation related injury episodes in the US population, 1997-99. Inj Prev 9(2):117-123, 2003. Fernandez WG, Yard EE, Comstock RD: Epidemiology of lower extremity injuries among U.S. high school athletes. Acad Emerg Med 14(7):641-645, 2007. McQuillan R, Campbell H: Gender differences in adolescent injury characteristics: A population-based study of hospital A&E data. Public Health 120(8):732-741, 2006. Warburton DE, Nicol CW, Bredin SS: Health benefits of physical activity: The evidence. CMAJ 174(6):801-809, 2006.

R E F E R E N C E S Please see www.expertconsult.com.

S ect i o n

A

Anterior Cruciate Ligament Tear Prevention   in the Female Athlete Holly J. Silvers

In 1972, about 300,000 girls participated in high school sports. This figure has grown exponentially to 2.36 million girls as of 2005. Globally, more than 21 million females are registered with the Fèdèration Internationale de Football Association (FIFA).1 Several prevalence studies have indicated that the number of female athletes incurring a serious anterior cruciate ligament (ACL) injury exceeds that of males by 2 to 8 times.2-5 The incidence of noncontact ACL injuries remains greater in sports requiring rapid deceleration, cutting, and pivoting, such as soccer, basketball, volleyball, team handball, and netball.6 The incidence of noncontact ACL injuries appears to be greatest in athletes who are less than 25 years of age.6 A system for ACL tear prevention is particularly important with regard not only to ligamentous

instability but also to prevention of early-onset osteoarthritis. Twelve years after injury, 34% of previous female soccer players in Sweden who suffered an ACL injury had radiographic changes consistent with osteoarthritis.7

ETIOLOGY Many studies have addressed the extrinsic and intrinsic factors related to ACL tears. The variation in morphology between males and females has been well examined, including variances in pelvic size and shape, ­intercondylar notch width, size of ACL, and Q angle. Estrogen, progesterone, and relaxin receptor sites have been found to be present within the ACL, and menstrual cycle hormonal 333

334

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

fluctuations have been studied to determine their effect on the integrity of the ACL.8-10 The increases in estrogen and relaxin hormones coincide with a subsequent decrease in the rate of collagen synthesis; however, there is no conclusive evidence linking an increase in ACL injury to a predictable time within the menstrual cycle.11,12 In a select, athletic, college-aged population, a combination of increased body mass index (BMI), narrow notch width, and increased joint laxity (as defined by KT-2000 or hyperlaxity measures) was directly correlated to ACL injury.6 No studies have demonstrated conclusive evidence on the effectiveness of functional knee braces for preventing noncontact ACL injuries.13 In 1974, Torg and colleagues developed a quantitative measurement called release coeffi­ cient to describe the force-to-weight ratio of shoe and surface interaction.14 Heidt and associates found that 73% of the 15 different types of athletic shoes tested demonstrated an “unsafe” or “probably unsafe” rating.15 When considering shoe design, it is important not to forgo safety for the sake of enhanced performance. Ekstrand and associates noted that the optimal shoe design should include a sole that will minimize rotational friction to avoid injury yet optimize transitional friction to allow peak performance.16 The current consensus is that no single environmental, anatomic, or hormonal risk factor correlates with an increase in ACL injuries in female athletes.13 Therefore, the emphasis has turned to biomechanical risk factors and the use of intervention programs to address potential biomechanical deficits.

RATIONALE FOR PREVENTION PROGRAMS The incidence of ACL injuries can be reduced through comprehensive neuromuscular training methods. A large number of noncontact ACL injuries occur during the deceleration phase of a cutting maneuver, when a rotation torque in concert with a varus and valgus moment is applied to a knee that is flexed 10 to 30 degrees.17,18 Markolf and colleagues noted that both a varus moment and internal rotation moment at the knee will place the ACL at a greater risk for injury as opposed to valgus and external rotation moments.19 This “pathokinetic chain” is described as a combination of an increased hip adduction moment, decreased hip abduction control, and increased hip adduction angles, thus placing the lower extremity in a valgus position. Theoretically, when this is combined with an increase in internal rotation moment and motion at the knee joint, the ACL incurs an increase in tension. The ground reaction force falls medial to the knee joint during a cutting maneuver, and this added force may tax an ACL ligament already under tension and lead to failure.

PREVENTION PROGRAM RESULTS Prevention programs focusing on skiing, basketball, and soccer have been performed in the past with results of overall reduction of severe ACL injuries ranging from 72% to 89%.20-24

Caraffa and colleagues implemented a proprioceptive balance training program on 600 semiprofessional and amateur soccer players using a 20-minute training program divided into five phases of increasing difficulty.20 They demonstrated an 87% decrease in ACL injuries compared with the control group. Ettlinger and associates implemented the “guided discovery” technique that focused on avoiding high-risk behavior and positioning (i.e., “phantom foot”), recognizing potentially dangerous skiing situations, and responding quickly to unfavorable conditions.21 They found a 62% serious knee injury decrease among the trained individuals compared to controls. The PEP (Prevent Injury, Enhance Performance) ACL Prevention Program used at our institution focuses on biomechanical risk factors and stresses avoidance of high-risk behaviors and increase in kinesthetic awareness.25 During the 2000 season, 1041 female club soccer players (52 teams) performed the PEP program, and 1902 players (95 teams) served as the age- and skill-matched controls. There were two ACL tears (0.2 ACL injuries per athlete) in enrolled subjects, compared with 32 ACL tears (1.7 ACL injuries per athlete) in the control group; an 88% decrease in ACL ligament injury. A repeat of the study in the 2001 season, demonstrated a 75% reduction in ACL tears. The next important study was a randomized control study involving collaboration between the Centers for Disease Control and Prevention (CDC), FIFA, and the National Collegiate Athletic Association (NCAA).26 In 2002, 61 NCAA Division I teams with 1394 athletes were enrolled (833 control athletes and 561 intervention). Control teams (CT, 833 total players) performed routine warmups and practices, whereas intervention teams (IT, 561 players) received a videotape and instruction manual for an alternative warm-up completed 3 times per week during the season. Injury rates were calculated based on athleteexposure (AE), and the overall ligament knee injury rate was found to be 0.66 per AE in IT, compared with 0.75 in CT (P > .05). The ACL injury rate was 0.15 per AE in IT versus 0.28 in CT (P > .05). The practice session ACL injury rate was 0.0 in IT compared with 0.125 per 1000 AEs in CT (P < .01).

CONCLUSIONS All recent ACL prevention investigations have found that a neuromuscular training program, such as PEP, may significantly reduce the incidence of severe ACL injuries in the female athlete. A prophylactic training program that focuses on developing neuromuscular control of the lower extremity though strengthening exercises, pylometrics, and sport-specific agilities may address the proprioceptive and biomechanical deficits that are demonstrated in the high-risk female athletic population. Future studies should strive to identify which of the components of present-day prevention programs are most significant in decreasing the rate of noncontact ACL injuries.6

R E F E R E N C E S Please see www.expertconsult.com.

Principles of Injury Prevention

S ect i o n

335

B

Preventing Hamstring Strains Roald Bahr

Muscle injuries occur frequently as contusion injuries in contact sports and as strains in sports involving maximal sprints and acceleration. Among sprinters, hamstring strains represent about one third of all acute injuries.1 Because the different football codes (soccer, rugby, ­American football, Australian-rules football) combine maximal sprints with frequent player-to-player contact, it is not surprising that a sizable proportion of injuries are thigh injuries. In fact, recent studies from the professional level show that hamstring strains alone rank as the first or second most common injury in soccer,2-5 Australian-rules football,6,7 rugby,8,9 and American football,10,11 in most studies accounting for one in every five to six injuries. There also appears to be a trend with a gradual increase in the proportion of hamstring strains as opposed to other injury types, such as ankle sprains, when comparing data from studies from the 1980s.12 However, quadriceps strains are common in soccer,4 and muscle contusion injuries to the quadriceps muscles account for a significant proportion of all football injuries at the elite level. Hamstring strain injuries are also common in sports in which the muscles may be stretched past the usual range of movement, including dancing and water-skiing.13

the net moment developed by the hamstrings is thought to be maximal in the late swing phase, right before heelstrike, this is thought to be a vulnerable position.15,16 In this instance, the hamstring muscles work eccentrically. Another suggestion is at push-off. Strain injuries to the quadriceps muscles have been studied less but are thought to result mainly from kicking the ball.

CAUSES—RISK FACTORS A number of candidate risk factors have been proposed for hamstring strains, the most prominent being the following four internal factors (Box 6B-1): age, previous injury, reduced hip range of motion (ROM), and poor hamstring strength.17 In theory, limited ROM for hip flexion could mean that muscle tension is at its maximum when the muscle is vulnerable close to maximal length. However, this hypothesis has yet to be confirmed because several studies of soccer players have suggested that hamstring flexibility is not a risk factor for strains.18,19 However, other studies from soccer and Australian-rules football have shown low quadriceps flexibility to ­represent a risk

CAUSES—INJURY MECHANISMS AND RISK FACTORS Two main mechanisms are involved in thigh muscle injuries: direct (contusion) and indirect (distension or strain) injuries. The contusion mechanism is straightforward: the player is typically injured by a direct blow from an opponent, usually the knee hitting the lateral thigh in a tackle (so-called charley horse or cork thigh). The muscle is thereby crushed between the opponent’s kneecap and femur (Fig. 6B-1). The injury mechanism for hamstring strains is less well understood. The hamstring muscle group is composed of three muscles: semimembranosus, semitendinosus, and biceps femoris. All of these (except the short head of the biceps) have their origin at ischial tubercle on the pelvis and insert at the inside and outside of the lower leg right below the knee. This means that they overlap two joints: they straighten the hip joint and bend the knee joint. Muscle strains usually occur in the interface between the muscle and its tendon (the myotendinous junction), but avulsion injuries from the ischial tubercle are also seen. Hamstring strains most often occur during maximal sprints. It is difficult to document exactly at what time during the running cycle injuries occur.14 However, because

Figure 6B-1   Typical injury mechanism for quadriceps contusions: crushing of the muscle belly between the femur and the opponent’s patella, resulting in an intramuscular quadriceps hematoma. (From Bahr R, Maehlum S (eds): Clinical Guide to Sports Injuries. Champaign, Ill, Human Kinetics, 2004. ©Gazette Bok/ T. Bolic.)

336

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 6B-1 R  isk Factors for Hamstring Strains Well Documented • Previous hamstring strain • Increased age • Low hamstring muscle strength Less Well Documented • Low hamstring muscle flexibility • Low quadriceps muscle flexibility • Race (black players and players of aboriginal descent at greater risk) • Sex • High level of play • Insufficient warm-up • Muscle fatigue • Player position

factor not only for hamstring strains20 but also for quadriceps strains.21 Low hamstring strength would mean that the forces necessary to resist knee flexion and start hip extension during maximal sprints could surpass the tolerance of the muscletendon unit. Hamstring strength is often expressed relative to quadriceps strength as the hamstrings-to-­quadriceps ratio because it is the relationship between the ability of the quadriceps to generate speed and the ­ capacity of the hamstrings to resist the resulting forces that is believed to be critical. Several studies show that players with low hamstring strength, low hamstrings-to-quadriceps strength ratio, or side-to-side strength imbalances may be at increased risk for injury.17 A history of previous hamstring strains greatly increases injury risk, as documented in numerous studies.17,22,23 Injury can cause scar tissue to form in the musculature, resulting in a less compliant area with increased risk for injury. A previous injury can also lead to reduced ROM or reduced strength, thereby indirectly affecting injury risk. Soccer players with a history of previous hamstrings injury have a 7 times higher risk for injury than healthy players; as many as 13% can expect to suffer a new injury during one season. Older players are at increased risk for hamstring strains, and although older players are more likely to have a previous injury, increased age is also an independent risk factor for injury.19,23 Other risk factors that have been suggested but are less well studied include race, sex, level of play, player position, improper running technique, superior running speed (peak performance), low back pain, increases or changes in the training program (particularly intense periods of training), insufficient warm-up, and muscle fatigue. Players of black or aboriginal origin sustain significantly more hamstri ng strains than white players,23 and it has been suggested that these players may be faster runners when compared with their white counterparts, possibly because of a higher proportion of type II muscle fibers. A faster running speed generates higher hamstring torques, which may explain the increased injury risk.

METHODS TO PREVENT HAMSTRING STRAINS Research on injury prevention methods for hamstring strains is limited, and the evidence available has mainly been collected from observational studies. None of the prevention methods described here have been tested in largescale randomized clinical trials with hamstring strains as the main end point. Studies to date have examined intervention methods targeting the key risk factors for hamstring strains: hamstrings strength, hamstrings flexibility, and previous injury. In addition, one observational study of South ­African rugby players suggested that the use of thermal pants might reduce the risk for hamstring strain.24 The consistent finding that a history of previous injury leads to a several-fold increase in the risk for new strains has led to the suggestion that this is at least partly due to inadequate rehabilitation and early return to sport. A study of Swedish soccer players25 documented that a coach­controlled rehabilitation program consisting of information about risk factors for reinjury, rehabilitation principles, and a 10-step progressive rehabilitation program including return to play criteria reduced the reinjury risk by 75% for lower limb injuries in general. Although the specific effect on hamstring strains could not be assessed in this study, it seems reasonable to recommend including functional and specific rehabilitation programs and careful screening of players before return to play. There are no intervention studies of elite athletes on the preventive effect of flexibility training on hamstring strains. However, one study of military basic trainees indicates a reduced number of lower limb overuse injuries after a period of hamstring stretching,26 whereas another military-based study found no effect of stretching.27 However, these studies were designed to examine the effect of general stretching on lower limb injuries in general, not a specific hamstring program on hamstring strain risk. Questionnaire-based data on flexibility training methods collected from 30 English professional football clubs, in which the stretching practices of the teams were correlated to their hamstrings strain rates, indicated that using a standard stretching protocol reduces injury risk.28 Also, one study from Australian-rules football observed a reduction in the incidence of hamstring strains with a three­component prevention program, and stretching while fatigued was one of the components.29 The other factors in the program were sport-specific training drills and high­intensity anaerobic interval training. Thus, it is not possible to determine which of these factors are responsible for the observed effect. The best evidence for injury prevention is available for programs designed to increase hamstring strength, particularly eccentric hamstrings strength. Several studies indicate that low hamstring strength is a risk factor for sustaining hamstring strains.18,30,31 Electromyographic studies have shown that activity is highest late in the swing phase and during heel-strike, when the hamstrings work eccentrically or transfer from eccentric to concentric muscle action.16,32 It is assumed that most hamstring strains occur during eccentric muscle actions, when the muscle activity is highest.33,34 It is well documented that strength training

Principles of Injury Prevention

337

  TABLE 6B-1  Training Protocol for Nordic Hamstring Group* Week

Sessions per Week

1 2 3 4 5-10

1 2 3 3 3

Sets and Repetitions 2 × 5 2 × 6 3 × 6-8 3 × 8-10 3 sets, 12-10-8 repetitions

*Load is increased as subject can withstand the forward fall longer. When managing to withstand the whole range of motion for 12 repetitions, ­ increase load by adding speed to the starting phase of the motion. The partner can also increase loading further by pushing at the back of shoulders (see Fig. 6B-2).39

Figure 6B-2   The Nordic hamstrings exercise. Subjects are instructed to let themselves fall forward and then resist the fall against the ground as long as possible by using their hamstrings. (Courtesy of Oslo Sports Trauma Research Center.)

is mode specific.35-39 Based on this, it may be argued that, to be specific, strength training for the hamstring muscles should be eccentric. It has been suggested that an indicator of susceptibility for the damage from eccentric exercise is the optimal angle for torque.40 When this is at a short muscle length, the muscle is thought to be more prone to eccentric ­damage, and by means of isokinetic dynamometry, it has been shown that mean optimal angle in previously injured muscles is at a shorter length than in uninjured muscles. Recent studies from Scandinavia have shown that replacing the traditional hamstrings strength exercise used by teams—hamstring curls—with exercises to develop eccentric strength reduces the risk for hamstring strains.2,41 Traditional hamstring curls have been shown to be ineffective in increasing eccentric hamstring strength among elite athletes.39 In contrast, a simple partner exercise—the ­Nordic hamstring lower (Fig. 6B-2)—has been demonstrated to be effective in improving eccentric strength.39 A pilot study has also shown that using a special apparatus—the

YoYo flywheel ergometer—also increases eccentric hamstring strength.41 Both these methods have been shown to prevent hamstring strains in studies of soccer players2,41 and rugby players.8 Because the Nordic hamstring lowers are easily implemented in a team setting, this exercise is recommended as a specific tool to prevent hamstring injuries. However, to avoid delayed onset muscle soreness, it is important to follow the recommended exercise prescription with a gradual increase in training load when introducing a program of Nordic hamstring lowers (Table 6B-1). By the end of a 10-week training period, many players are able to stop the downward motion completely before touching the ground (i.e., at about 30 degrees of knee flexion), even after being pushed by their partner at a considerable speed. When a player can reach this stage, the characteristics of the Nordic hamstring lower exercise appear to resemble the typical injury situation: eccentric muscle action, high forces, and nearly full knee extension. The program has been implemented in several different sports and younger age groups, and injuries from the exercise have not been recorded.

R E F E R E N C E S Please see www.expertconsult.com.

338

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n

C

Ankle Instability Prevention Eric Giza and Holly J. Silvers

Athletic participation places high demands on the foot and ankle and inevitably leads to chronic changes that are related to exposure time. Ankle instability comprises a spectrum of pathology from mechanical laxity to functional tissue impairment, which can include ligament instability, tendinopathy, articular cartilage damage, and tibiotalar osteophytes. Recognition of the individual facets of ankle instability is necessary for effective diagnosis, treatment, and prevention. Chronic problems associated with laxity of the lateral or medial ligamentous structures of the ankle are called mechanical instability. In other cases, the pain or sensation of instability is a result of intra-articular or periarticular changes without ligamentous laxity, which is called func­ tional instability. Many athletes have a combination of the two conditions and are in the middle of the spectrum of pathology. The sports medicine professional should be cautious in the use of the diagnosis of a “simple ankle sprain.” Verhagen and associates studied 577 patients with sprains who were treated at the same institution with 6.5 years follow-up.1 Thirty-nine percent of patients had residual symptoms of swelling, pain, stiffness, instability, and fear of giving way. Nine percent of patients had interference with physical activity, 5% changed sports, 4% stopped sports, and 6% patients could not return to their presprain employment. The authors found no correlation between the extent of the handicap and the severity of the sprain, nor between the frequency of the residual complaints and the severity of the sprain.1 Ankle instability can result from a single injury or represent a spectrum of injuries that are accrued over the course of an athlete’s career. In separate studies, Larsen and associates2 and Drawer and Fuller3 found that the risk for knee and ankle osteoarthritis in professional soccer players is significantly greater than in the general population. Others found that the incidence of degenerative arthritis after chronic lateral ankle instability ranges from 13% to 78%4; therefore, it is important to be aware of the factors that lead to chronic ankle instability. There is one inversion event to the ankle per 10,000 people per day, which amounts to about 23,000 ankle injuries a day in the United States.5 Lateral ankle sprains account for up to 45% of all basketball injuries and 17% to 20% of soccer injuries.5,6 Ankle sprains are also the most common cause of acute injury in volleyball and represent the leading cause of time loss in the National Football League.7 A two-season survey of 91 English soccer clubs showed that 11% of all injuries were ankle sprains and that 77% of

sprains involved the lateral ligament complex.8 Fortyeight percent of footballers with a first-time sprain have another sprain, and up to 26% have recurrent sprains.5 Many ankle sprains in soccer are a result of unavoidable, traumatic situations9,10; however, the sports medicine professional should be aware that risk factors such as ligamentous laxity, posterior position of the fibula relative to the tibia, and peroneal muscle weakness can lead to chronic instability.11-13

ANATOMY There are four important ligaments of the lateral ankle complex. The anterior talofibular ligament (ATFL) restricts internal rotation of the talus in the mortise and is elongated in plantar flexion. The ATFL undergoes greater plastic deformation than the calcaneofibular ligament (CFL), and failure occurs at 138 Newtons (N).14 The CFL is a diarthrodial ligament that prevents hindfoot adduction and is stiffer than the ATFL. It is elongated in dorsiflexion and fails at 345 N.14 The posterior talofibular ligament (PTFL) restricts external rotation when the ankle is dorsiflexed and fails at 261 N.14 The inferior extensor retinaculum (IER) is composed of lateral, intermediate, and medial roots. It blends with the lateral talocalcaneal ligament and plays an important role in stability by linking the lateral ligament complex with the subtalar joint.14 Although the ATFL spans only one joint, the CFL and IER play an important role in subtalar ­stability.14,15 The superior peroneal retinaculum (SPR) spans from the posterior and distal aspect of the fibula to the Achilles fascia and os calcis. Along with the retromalleolar sulcus and cartilage rim, the SPR provides stability for the peroneal tendons.12 The deltoid ligament complex is composed of six separate ligaments and has both a deep and superficial layer.16 The deltoid prevents external rotation of the talus in the mortise and prohibits abduction of the ankle.12,17

MECHANISM OF INJURY Cutting sports place high impact on the ankle and foot, and athletes must maintain a balance between strength and agility in a joint that has multiplanar ranges of motion. The most common mechanism for ankle injury in sports is an inversion and plantar flexion mechanism10,18; however, aggressive cutting and direction changes can place large demands on the anterior, posterior, and medial ankle.4,8

Principles of Injury Prevention

Video analysis studies have provided information regarding the mechanisms of ankle injuries during soccer matches. Giza and coworkers reviewed injuries from four Fédération Internationale de Football Association (FIFA) football competitions and found that direct contact between players occurred in 72 of 76 injuries.10 They also demonstrated a significantly higher risk for time lost from play if the footballer is weight-bearing at the time of injury. Andersen and colleagues reviewed 26 match injuries and found that lateral ankle injuries are most commonly a result of a tackle that places a laterally directed force on the medial aspect of the leg of the injured player, leading to an inversion injury.9 Basketball and volleyball are high-risk sports for ankle inversion injuries.5,7 Severe ankle sprains can occur if an athlete lands on the foot of an opposing player during landing after a jump.

ASSOCIATED INJURIES Mechanical instability can often have associated intraarticular pathology that leads to functional tissue impairment. Van Dijk and associates found acute injuries to the tibia and talus in 20 of 30 patients who had arthroscopy after acute ankle sprain.19 In a study of 61 patients who underwent stabilization for lateral ankle laxity, Di Giovanni and associates found no patients with isolated instability.20 Takao and colleagues found that patients who had continued disability for more than 2 months after ankle sprain had intra-articular pathology. Moreover, they found that arthroscopy was particularly useful for the diagnosis of osteochondral lesions because the prearthroscopic sensitivity of magnetic resonance imaging and physical examination was only 82.4%.21 Chronic lateral laxity and stretching of the SPR can lead to subluxation of the peroneus brevis and longus. Recognition of SPR laxity and tendon pathology associated with lateral ligament laxity is paramount because failure to do so can lead to chronic pain after a lateral stabilization procedure.22-24

DIAGNOSIS The hallmark of mechanical instability is mobility beyond the physiologic range of motion, which has been traditionally characterized by a positive talar tilt or anterior drawer test.5 There is some variability in the radiographic criteria for mechanical instability owing to differing amounts of laxity among patients; however, Mann showed that 81% of patients with radiographic instability had recurrent sprains.25 Mechanical instability is considered present if there is more than 10 mm of anterior translation on the effected side or more than 3 mm on side-to-side difference, or if talar tilt is more than 9 degrees or more than 3 degrees on side-to-side difference.5 Another method to determine instability is to place the patient prone on the examination table and allow both ankles to hang over the edge of the table. This relaxes the peroneals, which are often in spasm to compensate for the instability. The distal leg is stabilized with one hand and the calcaneus is translated medially to reveal the amount of lateral stability. The anterior drawer can also be easily

339

measured in this fashion. Scranton has described a method by which the distance between the center of the Achilles and the tip of lateral and medial malleoli is measured. A ratio of Achilles/lateral tip to Achilles/medial tip of greater than 1.3 is indicative of lateral instability.26 A recent study by Hiller and coworkers validated the Cumberland Ankle Instability Tool for use as a simple method to measure the severity of ankle instability.27 Numerous systems have been used to grade acute lateral ankle sprains12; however, treatment can be guided by an understanding of the structures injured. A grade 1 injury involves only a tear in the ATFL. In grade 2 injuries, the CFL is involved, and in grade 3 injuries, the entire lateral complex is disrupted.

MEDIAL ANKLE INSTABILITY Although injuries to the medial ligament complex are less common than lateral sprains, Woods and associates found that 14% of all ankle injuries in English professional soccer were medial ligament injuries.8 Medial instability can coexist with lateral instability. In a series of 52 patients with medial ankle instability, 40 (77%) were found to have concomitant lateral instability.28 Hintermann and associates found that deltoid ligament instability is often missed on examination and is demonstrable on arthroscopic evaluation of the ankle. Moreover, they found that cartilage damage was present in 98% of cases with medial instability but only in 66% of cases with lateral instability, indicating that pathology to the medial ligaments may result from a more traumatic injury.4 The clinical characteristics of medial ankle instability are a feeling of giving way and a valgus and pronation deformity of the foot that can be actively corrected by firing the posterior tibial muscle.28,29 Often the athlete will have failed some weeks of treatment for a lateral ligament injury and have continued pain. Examination may reveal tenderness of the medial gutter of the ankle and a valgus and pronation deformity of the foot.29 Anterior drawer testing of the ankle in plantar flexion and internal rotation may reveal a slight increase in translation compared with testing in plantar flexion and external rotation.

INITIAL TREATMENT After an ankle injury, the general treatment principles include maintaining a normal environment to keep the athlete involved and interested and protecting injured ligaments while they heal. Depending on the severity of the sprain, patients should be allowed progressive weight­bearing and maintenance of range of motion. After the initial injury phase is complete, the focus is on muscle strengthening, fitness, retraining proprioception, and regaining sport-specific skills. Operative treatment is rarely, if ever, indicated for acute lateral ligament injuries, and the mainstay of treatment is rest, ice, compression, and elevation (RICE.) Players with excessive swelling or tenderness of the distal fibula should have a radiograph to exclude ankle fracture.30 Return to play ranges from 1 to 8 weeks depending on the severity,5 and athletes should undergo a systematic rehabilitation program that includes peroneal strengthening and

340

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

­ roprioceptive exercises to prevent future injury and prop gression to chronic laxity.31,32 Most athletes are completely pain free 8 weeks after the injury, and those who have continued pain should be referred for evaluation of other possible ankle pathologies.

PREVENTION OF MECHANICAL ANKLE INSTABILITY It has been demonstrated that a history of a previous ankle injury is correlated with a 2.3 times greater risk for reinjury.5 About 20% of ankle sprains lead to a recurrent injury, and 20% to 50% of recurrent ankle sprains have subsequent chronic pain or instability. Thacker and colleagues performed a meta-analysis of 113 ankle sprain studies and found that the most common risk factor for an ankle sprain is a history of prior sprain.33 Woods has shown that repeat ankle injury is more often due to a noncontact situation compared with the initial injury, which indicates that ankle instability develops as the lateral ligament complex develops laxity.8 Others have shown that major injuries of the ankle are often preceded by inadequately rehabilitated or treated injuries.34-37 In 1983, Ekstrand and Gillquist demonstrated a 75% reduction in the rate of injuries when a professional Swedish soccer team had meticulous supervision by an athletic trainer and a physician.34 More recently, FIFA showed that preseason and intraseason injury prevention programs can reduce injury in up to 21% of cases, and introduced the FIFA Eleven, which consists of 10 exercises and a commitment to fair play and good sportsmanship.38 Physiotherapy is aimed at improvement of coordination and strength of the muscles around the ankle to restore proprioception and thus stability.39 Tape and braces provide direct mechanical support for an unstable ankle, but it has also been suggested that the beneficial effect is explained by enhancement of proprioception through skin pressure.40 Preparticipation taping has been found to decrease peroneal reaction time and increase firing of peroneus brevis during the swing phase of gait; however, taping does not significantly reduce talar tilt or anterior translation and typically loosens after 20 to 30 minutes of play.41,42 Bracing has been shown to be more effective than taping for athletes with one to three episodes of recurrent sprains. Surve and colleagues evaluated the effect of a semirigid ankle orthosis on the incidence of ankle sprains in soccer players during season.43 They compared 258 soccer players without a history of sprain with 246 players with a history of sprain. They found a significant reduction in the incidence of ankle sprains in the orthosis group with previous sprains compared with the nonbraced group with previous sprains. The incidence of ankle sprains was significantly higher in the nonbraced group with previous sprains compared with the nonbraced group without previous sprains. They concluded that the use of a semirigid orthosis leads to a significant reduction in sprains only for the group with a history of sprains.43 Rosenbaum and colleagues compared the effects of 10 different ankle braces (1 rigid, 5 semirigid, and 4 soft models) in a comprehensive evaluation with multiple testing

procedures in 34 subjects with self-reported chronic ankle instability.44 The multiple testing procedures evaluated objective performance-related parameters and subjective parameters related to comfort and stability. The subjects performed a jumping and cutting agility course, a single-leg hopping test on level and inclined planes, and a combined straight and curve sprint. Their results for the objective parameters showed no significant differences between the braces except for the rigid brace, which showed decreased values for the vertical jump and longer times for the other tests. The subjective evaluation of the braces revealed significant differences with respect to comfort and handling. Because there was no difference in objective stability, the authors concluded that patients could choose a brace model according to their individual needs for comfort and type of sport.44 Verhagen and associates performed a prospective study of 116 male and female volleyball teams that were randomly divided into control and intervention groups.32 The intervention groups performed 5 minutes of proprioception or ankle strengthening exercises before each practice session during a 36-week season. A significant risk in reduction of ankle sprain was found only for players with a history of previous sprain; however, the authors recommended the use of a proprioceptive balance board program for prevention of ankle sprain recurrences.32 Stasinopoulis compared the effects of technical training (peroneal strengthening), proprioception, or orthosis use in three separate groups of volleyball players.45 They demonstrated that technical training and proprioception were more effective in reducing recurrent sprains better than the use of an orthosis only if the player had greater than four sprains during his or her career. Moreover, the orthosis use was only effective when the player had less than four sprains during his or her career.45 In 2001, Eils and Rosenbaum investigated the effects of a 6-week multistation proprioceptive exercise program that can be integrated into normal training programs.46 Patients with chronic ankle instability were compared with controls before and after three testing procedures: joint position sense, postural sway, and muscle reaction times to sudden inversion events on a tilting platform. A total of 30 subjects with 48 unstable ankles were evaluated, and significant improvements were seen in joint position sense, postural sway, and muscle reaction times. The authors concluded that a multistation proprioceptive exercise program is recommended for the prevention and rehabilitation of recurrent ankle inversion injuries.46

OPERATIVE TREATMENT The operative treatment of mechanical ankle instability is necessary for athletes who have had multiple sprains and have continued episodes of instability despite taping and rehabilitation. Surgical stabilization is an invasive yet powerful method of injury prevention and protects the ankle cartilage from repeated trauma. Many different procedures have been described and fall into the categories of anatomic or nonanatomic reconstruction. Nonanatomic reconstructions utilize a graft from the peroneal tendons, Achilles, or allograft and can lead to excessive tightness of the lateral ankle and subtalar joint.7,47,48

Principles of Injury Prevention

Footballers need to preserve as much ankle and subtalar motion as possible to maintain a “touch” on the ball; therefore, the anatomic reconstruction described by Broström is the preferred treatment for chronic footballer’s ankle (CFA) with mechanical instability.48,49 The procedure involves exposure of the attenuated ATFL, CFL, and IER, with advancement of the ligaments back to their anatomic insertions on the fibula using bone tunnels or suture implants. Krips and associates compared the functional outcome and sports activity level of 41 patients who underwent anatomic reconstruction with 36 patients who underwent nonanatomic reconstruction and found superior results in return to play in the anatomic reconstruction group.50 Karlsson and colleagues found good to excellent results in 132 of 152 ankles that were operated on for chronic lateral instability, and the 20 with poor results were found to have generalized ligamentous laxity or multiple surgeries.42 DeVries and coworkers showed that early functional rehabilitation is superior to 6 weeks of immobilization following stabilization.40 The immediate postoperative ­protocol includes immediate weight-bearing, stabilization in an Aircast brace (Aircast, Inc., Summit, NJ), and range of motion exercises. After 4 weeks, the brace is removed, and the patient begins peroneal strengthening and an ankle exercise program.51,52 Jogging begins at 6 weeks, and return to football is often possible between 12 and 16 weeks.

C

r i t i c a l

P

341

o i n t s

l Athletic participation places high demands on the foot and ankle and inevitably leads to chronic changes that are related to exposure time. l Ankle instability comprises a spectrum of pathology from mechanical laxity to functional tissue impairment, which can include ligament instability, tendinopathy, articular cartilage damage, and tibiotalar osteophytes. l Recognition of the individual facets of ankle instability is necessary for effective diagnosis, treatment, and ­prevention. l Proper rehabilitation is important after injury to prevent recurrent instability. l Prevention exercises have been found to be most effective in athletes with prior history of sprain. l Bracing with a sports orthosis is necessary after injury, and proprioceptive exercise programs are the key to prevention of continued instability.

R E F E R E N C E S Please see www.expertconsult.com.

S ect i o n

D

Spine-Related Injury Prevention in the Athlete: Trunk Stabilization Robert G. Watkins IV

The goal of a spinal exercise program is to produce functional coordinated core strength for the body. Athletic functions such as throwing, swinging, and lifting, as well as activities of daily living, require coordinated muscle strength to achieve maximal performance while protecting the spine. In testing of professional golfers, major league baseball players, and other athletes, it has been well demonstrated that the coordination of trunk musculature produces maximal control of the spine. Coordinated strength is more effective than uncoordinated strength. Each of the trunk muscles fires in an exact sequence in relation to each other for particular actions. The coordinated strength protects the spine from injury and produces the desired athletic result.

Recently, there has been much discussion in the physical therapy literature regarding the best way to train the spinal musculature. Is it better to train the local stabilizers of the spine or the global stabilizers? It is our experience that starting in a neutral pain-free position is the most functional way to begin an exercise program, regardless of whether the focus is on the local or global stabilizers. This allows injury-free athletes to maintain a solid, neutral position for the prevention of injuries and allows an injured athlete a pain-free position to begin rehabilitation. The trunk stabilization program we use is a five-level strength and coordination program. The patient or athlete progresses through eight different exercises rated 1 through 5 in difficulty (Table 6D-1). The entire program

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

342

  Table 6D-1  Watkins-Randall Scale of Trunk Stabilization Exercises Dead Bug

Partial Sit-Ups

Bridging

Prone

Quadriped

Wall Slide

Ball

E

F

G

H

Less than 90 degree; ­repetitions; 10×

Balance on ball; leg press

Walk; land and water

90 degree; hold 20 sec; 10×

Leg press with 10-min; cycle arms over head; water sit-ups; forward; no hold; run

A

B

C

D

1

Supported; arms over; 2 min marching

Forward; hands on chest; 1 × 10

Slow ­repetitions; double leg; 2 × 10

2

Unsupported; arms over head; one leg ­extended; hold for 3 min Unsupported; arms over; alternate leg extended; with weights; hold for 7 min Unsupported; bilateral upper extremity with bilateral lower extremity extended; with weights; hold for 10 min; alternate leg extension Unsupported; bilateral lower extremity ­extension; 15 min total; increased weights; bilateral upper extremity with bilateral lower extremity extension

Forward; hands on chest; 3 × 10

Slow ­repetitions; double leg; weight on hips; 2 × 20 Single leg; 3 × 20; hold; double with weights; double on ball On ball; single leg; 4 × 20; hold; double on ball with weights; feet on ball double bridge

Gluteal squeeze; Upper alternating ­extremity arm or leg or lower lifts; ­extremity; 1 × 10 repetitions hold; 1 × 10 Alternating Arm and leg; arm and leg 2 × 10; hold lifts; 2 × 10 hold

3

4

5

3 × 10 forward; 3 × 10 right; 3 × 10 left 3 × 20 forward; 3 × 20 right; 3 × 20 left; weights on chest

3 × 30 forward; 3 × 30 right; 3 × 30 left Unsupported; weights; overhead and behind

On ball; single leg; 5 × 20; weights; ­holding double with feet on ball and ­bilateral knees flex

Ball flies; swim; superman; 2 × 10 Ball; 10 × 20; hold; superman with weights; prayer; push-ups; walk-outs

Ball; all exercises with weights; 4 × 20; body blade

starts with finding a neutral pain-free position for the spine and holding that position while performing the exercises. This makes it possible to begin postoperative conditioning earlier because it avoids the extremes of motion through the injured spine. The entire program can be performed with relatively simple exercise equipment: exercise balls, hand weights, and pulleys. The rehabilitation of the injured athlete’s spine begins with level 1 core stabilization training. The key is to learn the proper technique, like learning to hit a baseball, ride a bicycle, or other coordinated activities. It is not a matter of brute strength; it is matter of doing the technique properly. Once the athlete has established the proper technique at level 1 of the program, he or she is advanced through the

Arm and leg; 90 degree; Ball sit-ups; 3 × 20; hold 30 ×20; hold 5 sec; sec; 10×; forward with weights lunges with right/left no weights Arm and leg; 90 degree; Ball sit-ups 2 × 20; hold hold 15 sec; forward, 5 sec; with with weights; right/left with weights ×10; lunges weights; 3 × 20; with weights Wand Manual ­Resistance Pulleys Arm and leg; 3 × 20; hold 15 sec; with weights; body blade

90 degree; Ball; overhead hold; arms and lateral pullextended; through Sports with weights; stick; pulleys; ×10 body blade Lunges with weights; hold 1 min

Aerobic

20-30 min; swim and NordicTrack 45 min Versiclimber; also step, skip rope

60 min; also run

five levels of increasing difficulty. These exercises should be performed under the supervision of a therapist or trainer to ensure proper technique and to prevent injury. After establishing mastery of level 5, the athlete begins a series of sport-specific exercises that have been developed for virtually every sport. Return to play depends on (1) achieving the proper level of the stabilization program (for recreational golfers and tennis players, it is level 3; for professional athletes, it is level 5); (2) obtaining good aerobic conditioning (the key to aerobic conditioning is to diversify the exercise); (3) performing the sport-specific exercises; (4) returning gradually to the sport; and (5) continuing the stabilization exercises once the athlete returns to the sport.

Principles of Injury Prevention

Dead Bug

343

STAB�������������������� I������������������� LIZATION EXERCISES Partial Sit-Ups

We begin our identification of the neutral spine position with the dead-bug exercises. Dead-bug exercises are done supine with the knees flexed and feet on the floor. With the assistance of the trainer or therapist, the player pushes the lumbar spine toward the mat until exerting a moderate amount of force on the examiner’s hand. This is not exaggerated, back-flattening, extreme force, but a mild to moderate amount of painless force on the examiner’s hand. The player is then taught to maintain this same amount of force through abdominal and trunk muscle contraction while doing the following: 1. Raising one foot 2. Raising the other foot 3. Raising one arm 4. Raising the other arm 5. Raising one leg 6. Raising the other leg 7. Doing a leg flexion and extension with one foot 8. Doing a leg flexion and extension with the other foot The same exercises can be performed with weights on arms or legs.

The feet are placed firmly on the floor, arms beside the body with palms to the floor, and the abdominal bracing is begun. The arms are then placed across the chest, and the shoulders and back are raised off the floor while maintaining the neutral pain-free position of the spine. The shoulders are held off the ground for a count of 5 and then returned. The amount of time the shoulders are held off the ground may vary from 2 to 10 seconds. The speed with which the maneuver is done may vary from a resting count of 1 to 2 seconds. The exercise is repeated in three sets of 30 times each. Weight may be added to the chest for additional contracture in the neutral pain-free position and is the key to increasing abdominal tone and strength. This exercise may be done with the arms behind the head, alternating right elbow to left knee and left elbow to right knee.

344

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Bridging

Raise the hips off the floor about 3 inches and hold for a count of 10, then return the hips to the floor. Starting in the supine neutral position, raise the hips 1 inch off the floor and maintain the neutral, pain-free position for a count of 10, then return hips to the floor.

Raise the hips further off the floor to the maximal height allowed while maintaining the neutral position and hold for a count of 10, then return hips to the floor. This is not meant to be a back-arching exercise; maintain trunk control in the neutral, pain-free position throughout the exercise.

Raise the hips off the floor about 3 inches and hold. Extend one leg while maintaining the back in the neutral pain-free position. Hold for a count of 10. Place the foot back on the floor and relax the hips back to the start position. Repeat with the other leg. Weights can be added to the leg in this position, and the legs may also be crossed over with flexion, abduction, and external rotation of the leg while maintaining the neutral pain-free position. This exercise can be progressed to include balancing on an exercise ball under the middle back.

Principles of Injury Prevention

345

Prone Exercises

Prone with Single Leg Lifts

Neutral Position Because the prone position may be painful in certain back conditions, it is suggested that the prone exercises begin with a pillow under the trunk to prevent too much lumbar extension. Rigidly tighten the trunk musculature into the neutral, pain-free position while maintaining the arms and legs in an extended position. Hold for a count of 10 and relax.

Prone with Single Arm Lifts Maintain the original abdominal bridging position in the neutral pain-free position while extending one arm off the ground. Hold for a count of 10 and relax. Repeat with other arm.

Maintain the original abdominal bridging position in the neutral pain-free position while extending one leg off the ground. Hold for a count of 10 and relax. Repeat with other leg. This exercise can be progressed to include double arm and leg lifts, weights, and an exercise ball.

Prone—Ball

Relax prone with the abdomen resting on the ball and with feet apart with toes on the floor in the push-up position. Arms are flexed at the shoulder and down to the floor. Roll forward slowly, extending the trunk out into midair while maintain tight trunk control. Hold for 10 seconds, then roll back to start position. Extension of the arms parallel to the shoulder can be added. Roll out slowly, hold for 10 seconds, and roll back. Weights can be held in the hands to increase the difficulty of the exercise. This exercise can be expanded to doing superman, swimming, prayer, and push-up exercises.

346

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Quadriped Exercises

In the all-fours position, with the knees and hands on the floor, tighten the trunk musculature and hold the spine in the neutral, pain-free position for a count of 10, then relax.

In the all-fours position, with the knees and hands on the floor, tighten the trunk musculature and hold the spine in the neutral, pain-free position. Extend one arm, hold for a count of 10, and then relax. Repeat with other arm.

In the all-fours position, with the knees and hands on the floor, tighten the trunk musculature and hold the spine in the neutral, pain-free position. Extend one leg, hold for a count of 10, and then relax. Repeat with other leg.

In the all-fours position, with the knees and hands on the floor, tighten the trunk musculature and hold the spine in the neutral, pain-free position. Extend one arm and the opposite leg; hold for a count of 10 and relax. Repeat with opposite arm and leg. Difficulty of these exercise can be increased with the use of weights on the extremities or the balancing of the bar across the back.

Principles of Injury Prevention

Wall Slides

347

Stabilization Exercises with Supine Green Ball

Supine Quad Press Sit on the ball with the ball placed in the small of your back. Keep the chest and stomach tight. Keeping the feet in the same position, roll back on the ball by straightening the legs. Keep the chin tucked in so as not to strain the neck. Keep the back in neutral and the chest up off of the ball. Return back to the starting position by bending the knees and rolling back down the ball.

A green exercise ball is positioned behind the back against the wall. Hold the legs slightly apart and arms at the side. The body rolls down the ball into the sitting position and maintains this sitting position for a count of 10, returning to the initial semistanding position. This exercise should begin with only a slight knee flexion, a partial squat, and eventually can proceed to a full 90/90 position (90 degrees of hip and knee flexion). Throughout the procedure, the trunk should be maintained in the neutral, pain-free position with tight abdominal bridging. This exercise combines trunk strengthening with a functional quadriceps strengthening maneuver. After being able to maintain a full 90/90 position for three sets of 30 times each, holding the position for 10 seconds, the maneuver can be done while standing on the toes, and additionally can be done while holding a weight in the arms.

Supine Shoulder Flexion with the Green Ball Alternately flex the shoulders, arms over the head, first right arm hold, left arm hold; do this with or without weights.

348

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Repeat the maneuver of rolling the chest out off the ball. With arms positioned behind the head, rotate the left elbow toward the right knee. Alternate with the right elbow toward the left knee, again maintaining tight, rigid trunk control.

Supine Ball Sit-Ups Maintain a supine position with the lower back on the ball, arms folded across chest, knees bent, and feet flat on the floor. Tighten the trunk into the neutral, pain-free position. Keep the pelvis stabilized and level using your abdominal and buttock muscles. Lift your shoulder blades and upper back off the ball, keeping your lower back in a neutral position. Walk backward on the ball so that more of the trunk is off the ball, projecting out into the air. Hold for a count of 4 to 8 while keeping the trunk rigid. Weights may be held to the chest to increase resistance.

Resistive exercises using a baton or a towel can be done with the aid of the trainer or therapist by pulling against the person on the ball and providing resistance for a count of 8 to 10. This resistance can be provided alternately across the chest, to the side, or over the head with a baton, weighted stick, or pulleys.

UPPER EXTREMITY POSTURAL EXERCISES The slip-shouldered round-forward posture is probably the most typical cause or extenuating factor in delay of recovery from neck and arm pain. This position shifts the head anterior to the trunk, producing a lever-arm effect and effectively increasing the weight of the head. This position also closes the intervertebral foramina because of extension of the cervical spine and closes the thoracic outlet. The basis of our cervical treatment is the same as our lumbar spine treatment—the Trunk Stabilization Program. We start our cervical treatment with the same lumbar-neutral dead-bug exercises. Isometric trunk exercises are essential to produce a chest-out posture. We frequently use a basic group of preventative exercises designed for neck and shoulder problems. The key to these exercises is emphasizing the chest-out posture. By emphasizing the chest-out posture during upper extremity, shoulder and neck exercises, proper head and neck alignment is enhanced. The chest-out posture does three things: 1. It increases the thoracic outlet. This is the area through which the artery, veins, and nerves pass from the trunk out to the arm. 2. It puts the weight of the head over the neck. This eliminates the lever arm effect of the distance from the center of gravity of the head to the spine and decreases the neck strain required to resist that weight. 3. It opens the intervertebral foramina and provides more space for the nerve as it leaves the spine.

Exercises should be designed to produce the isometric strength necessary to maintain this position, and all upper strengthening exercises should be done emphasizing this position. Do not start neck therapy by stretching or moving a painful neck. Use careful head control, positioning, modalities, and posture realignment. A general exercise program could include the shoulder and rotator cuff exercises as well as dorsal glides, midline neck isometrics, shoulder shrugs, arm rolls, and a weight program. Remember, just stick the chest out, do not attempt to hold the shoulders back or forcefully tuck the chin. Do it with the chest, abdominal, and buttock muscles. The important factor is the chest-out posture.

Principles of Injury Prevention

349

AEROBIC CONDITIONING

Shoulder Shrugs and Shoulder Rolls Shrug the shoulders and relax, shrug the shoulders and relax. Roll the shoulders and relax, roll the shoulders and relax. Weights may be added to increase the difficulty.

Water Running

Arm Roll Standing in a properly aligned spinal position and facing the mirror, the arms are extended out to the side, and a fine arm roll is done with the arms: first with the fingers pointed out, second with the thumbs pointed down, and third with the thumbs pointed up. These exercise can be expanded to include elbow touch, arm abduction, and chest pull.

WEIGHT TRAINING In the transition from the trunk stability exercises to the use of a weight-type machine, it is essential to maintain the neutral pain-free position while using the different types of weight machines such as pectoralis laterals or bench press machine. When using the machine, tighten the trunk in the neutral pain-free position, perform the particular type of exercise on the machine, and relax the position between sets. As for free weights, the control is a vital part of any free weight program. One example of this can be seen in the forward lunge. This can be done with or without weights, but maintenance of a proper neutral position is of paramount importance while performing this exercise.

Water running is a non–weight-bearing activity done in 8 to 10 feet of water. The stride is that of a full sprint. A pair of old tennis shoes can be worn. A kitchen timer is used right beside the pool. Begin with 15-second intervals of full-out sprinting in the water. Usually a static position in the water can be maintained with the face out of the water. A buoyancy vest or life jacket can be of great help. Keep the back straight. Bring the knees up in a high step sprint. Water running allows no stress on lower extremities or the spine and should be an excellent conditioning method not requiring the jarring of running. C

r i t i c a l

P

o i n t

l Coordinated

trunk strength protects the spine from injury and produces the desired athletic result.

S U G G E S T E D

R E A D I N G S

Richardson CA, Jull GA, Hodges PW: Therapeutic Exercise for Spinal Segmental Stabilization in Low Back Pain. Edinburgh, Churchill Livingstone, 1999. McGill SM, Grenier S, Kavcic N, Cholewicki J: Coordination of muscle activity to assure stability of the lumbar spine. J Electromyogr Kinesiol 13(4):353-359, 2003. Watkins RG: The Spine in Sports. St. Louis, Mosby–Year Book, 1996.

C H A P TE R  

 7

Complications S e c t i o n

A

Complex Regional Pain Syndromes Including Reflex Sympathetic Dystrophy and Causalgia David L. Saxton and Thomas N. Lindenfeld

INTRODUCTION AND TERMINOLOGY Controversy and confusion surround nearly every aspect of complex regional pain syndrome (CRPS).1-9 ­ Historically, this condition has been plagued by delayed or missed diagnosis and haphazard treatment protocols.10 Although the nomenclature currently recommended is new, the ­constellation of signs and symptoms was first recognized in ­ American Civil War soldiers in 1864.11 The medical literature regarding this syndrome is difficult to interpret because of the vague nature of the condition and the ­inconsistent use of terminology (Box 7A-1).2 Despite this, CRPS is a debilitating condition affecting 5.5 per 100,000 patients in the community with increased incidence in postsurgical orthopaedic patients.12 This CRPS incidence is a subset of the estimated 3.75 million cases of chronic neuropathic pain in the United States.13 CRPS is most succinctly described as a chronic pain syndrome characterized by severe, diffuse, nondermatomal pain associated with painful responses to nonpainful stimuli (allodynia), accompanied by autonomic and trophic changes, and usually following tissue injury. The pain is associated with changes in skin color, temperature changes, sudomotor dysfunction (sweating), edema, and reduced range of motion. The pain is sharp and burning and often out of proportion to the inciting event.14 In 1988, Amadio15 wrote about pain dysfunction syndrome as a descriptive term to incorporate all former related diagnoses under one heading. Pain dysfunction syndrome includes patients with pain that is excessive, nondermatomal, and out of proportion to the inciting event as their unifying symptom. Pain dysfunction syndromes are generally thought to have three primary contributing components and one secondary component: a trigger; a personality disorder; systemic factors that exacerbate pain; and sympathetic dysfunction. A local trigger commonly begins the process and should be identified and eliminated.

Examples of common local triggers are painful organic conditions such as patellofemoral pain, fracture, and nerve injury.16 Second, psychological factors such as secondary gain issues, substance abuse, and psychiatric conditions and personality disorders are to be identified. Personality disorders may be objectively evaluated with a Minnesota Multiphasic Personality Inventory (MMPI).17,18 Psychiatric conditions, such as somatization disorders, malingering, factitious injury, and conversion reactions, are thought to contribute to the syndrome (Table 7A-1).15 Amadio15 concedes that although psychiatric consultation is necessary in such cases, it is often unsuccessful at resolving the issues. Third, systemic factors are thought to cause or exacerbate pain. These factors include such entities as diabetic peripheral neuropathy, lupus erythematosus, polymyalgia rheumatica, giant cell arteritis, multiple sclerosis, ischemic heart disease, Pancoast’s tumors, and others. These first three components do not involve the sympathetic nervous system and therefore are named sympathetically independent pain (SIP). Sympathetic dysfunction is the fourth factor in this syndrome. Pain due to sympathetic dysfunction is termed sympathetically maintained pain (SMP). Amadio15 includes reflex sympathetic dystrophy, causalgia, traumatic dystrophies, shoulder-hand syndrome, and Sudeck’s atrophy as examples of the sympathetic component of pain dysfunction syndrome. Despite Amadio’s attempt to standardize the nomenclature, the International Association for the Study of Pain established a standard classification scheme in an effort to eliminate confusion.19 The new terminology was established to eliminate the inaccuracy of the term reflex ­ sympathetic dystrophy. Many authors point out that this syndrome is not mediated by a true reflex, does not always reflect ­sympathetic nervous system dysfunction, and does not consistently include dystrophic changes.6,20 The new term, complex regional pain syndrome (CRPS), has become the standard nomenclature and is now being widely used.2,3,6,8,9,18,19,21-41 Merskey and Bogduk19 grouped the multitude of former 351

352

DeLee & Drez’s Orthopaedic Sports Medicine

Box 7A-1 Historical Terms Used to Describe Excess Pain with or without Sympathetic Dysfunction Erythromelalgia Chronic traumatic edema Sympathalgia (postsympathectomy pain) Post-traumatic pain syndrome Hyperpathic pain Reflex neurovascular dystrophy Algodystrophy Sudeck’s atrophy Peripheral acute trophoneurosis Traumatic angiospasm/vasospasm Shoulder-hand syndrome Postinfarction sclerodactyly Major and minor causalgia Mimocausalgia Major and minor dystrophy Sympathetically maintained pain Reflex sympathetic dystrophy Pain dysfunction syndrome Saphenous neuralgia/neuritis

pain syndromes into this new broad label and then subclassified CRPS into types I and II. CRPS type I broadly corresponds to classic reflex sympathetic dystrophy; CRPS type II includes the causalgias19 (derived from the Greek words kausis, burning, and algos, pain)40 and therefore is defined as a burning pain typically associated with a nerve injury.42 SMP is defined as pain that is sustained by sympathetic innervation or by circulating catecholamines, which stimulate the sympathetic system.37,43 Merskey and Bogduk19 reported that SMP is associated with CRPS at some point in the course of the disease and is defined as that component of pain that is relieved by sympathetic blockade. Any pain remaining after true sympathetic blockade is SIP.19 In addition, variable amounts of SMP and SIP contribute to both types of CRPS (Fig. 7A-1).37 Differentiation of these two components of pain is clinically useful because it can influence treatment.41 The presence of autonomic dysfunction in CRPS does not guarantee that all patients will respond to sympathetic blocks. SMP was introduced

   TABLE 7A-1  Definition of Psychiatric Diagnoses   Associated with Pain Dysfunction Syndrome Somatization disorders Malingering Factitious injury Conversion reactions

Preoccupation with pain (exaggerated) without organic disease for more than 6 months Intentional misrepresentation of symptoms to escape a duty or obligation Intentional misrepresentation of symptoms with no secondary gain incentives; psychological need to ­assume the role of a sick person; the patient is often unaware that injury may be self-inflicted A response to psychological conflicts with ­unintentionally produced signs of physical disorder without physical cause

100% Patient B SIP Proportionate reduction in total pain

Patient A SMP

Response to Sympathetic Block Figure 7A-1   The relative contribution that sympathetically maintained pain (SMP) may have to the overall pain. Patient A is a person whose pain is predominantly unresponsive to sympatholysis. Patient B has pain that is almost totally sympathetically maintained. Points A and B may represent different patients or the same patient at different times. SIP, sympathetically independent pain. (Redrawn from Boas RA: Complex regional pain syndromes: Symptoms, signs, and differential diagnosis. In Janig W, Stanton-Hicks M [eds]: Reflex Sympathetic Dystrophy: A Reappraisal. Progress in Pain Research and Management, vol 6. Seattle, IASP Press, 1996, p 83.)

to explain the favorable response some patients have to sympathetic blockade. Conversely, the term SIP was introduced to explain the lack of response some patients have to sympathetic blockade. Although patients can display SMP or SIP characteristics, they often have contributions of both, explaining the varying response to sympathetic blockade.44 SMP is diagnosed in patients who experience improvement during or after treatment with medications that modify the sympathetic nervous system. Classically, SMP must respond to an epidural, intrathecal, lumbar plexus block, or peripheral nerve block. Rapid response to such blocks is pathognomonic for SMP. It must be emphasized, however, that there is no single pathognomonic test for CRPS.45 Specific criteria must be satisfied to make the diagnosis of CRPS (Box 7A-2).37 The main symptom necessary to establish the diagnosis of CRPS is an exaggerated pain response. The sole differentiating factor between the CRPS subtypes is the presence of a known nerve injury in type II, whereas type I is generally attributed to a noxious event (other than nerve injury). Thus, the type I designation is assigned when no known peripheral nerve injury exists. Electromyography and nerve conduction velocity studies may help the clinician discover occult peripheral nerve injury and thereby differentiate between type I and type II46 (negative test results indicate CRPS type I, and positive results indicate CRPS type II). Both CRPS types are characterized, at some point in the course of the disease, by alterations of sympathetically controlled functions such as vasomotor and sudomotor activity.37 Several attempts to validate and refine the CRPS diagnostic criteria have been published since the original criteria were reported in 1994.25,27,37 It is likely that the diagnostic definition will continue to evolve as our understanding of the syndrome improves.

Complications

Box 7A-2 Diagnostic Criteria of the ­International Association for the Study of Pain for CRPS Type I and Type II CRPS Type I (Reflex Sympathetic Dystrophy) An initiating noxious event is present. Spontaneous pain or allodynia/hyperalgesia occurs beyond the territory of a single peripheral nerve and is disproportionate to the inciting event. There is or has been evidence at some time of edema, skin blood flow abnormality, or abnormal sudomotor activity in the region of the pain since the inciting event. This diagnosis is excluded by the existence of conditions that would otherwise account for the degree of pain and dysfunction. CRPS Type II (Causalgia) This syndrome follows nerve injury. It is similar in all ­other respects to type I. A more regionally confined presentation around a joint (e.g., ankle, knee, or wrist) or area (e.g., face, eye) is ­associated with a noxious event. Spontaneous pain or allodynia/hyperalgesia is usually limited to the area involved but may spread variably distal or proximal to the area, not in the territory of a dermatomal or peripheral nerve distribution. Intermittent and variable edema, skin blood flow changes, temperature change, abnormal sudomotor activity, and motor dysfunction, disproportionate to the inciting event, are present around the area involved. CRPS, complex regional pain syndrome.

PERTINENT ANATOMY AND PHYSIOLOGY REVIEW The human nervous system consists of somatic and autonomic branches. The somatic system includes the central nervous system (brain and spinal cord) and the peripheral nervous system. At the vertebral level, sensory nerve cell bodies are contained in the dorsal root ganglion. The ventral and dorsal rootlets join together to form the root of a peripheral nerve.47 Specialized somatic sensory nerve endings known as nociceptors, mechanoreceptors, and thermoreceptors perceive pain stimuli, pressure stimuli, and temperature stimuli, respectively.48 These nerves transmit signals from the peripheral nervous system to the central nervous system and are termed afferent nerves. Conversely, nerves that transmit commands from central to peripheral are termed efferent nerves. The autonomic nervous system is composed of the sympathetic and parasympathetic divisions. Anatomically, the parasympathetic nerve cell bodies are located in ganglia that are primarily cranial or sacral. In contrast, sympathetic nerve cell bodies are mainly located in paravertebral chains at the thoracolumbar levels. There are three sympathetic chain ganglia located at the cervical level, and they are known as the superior, middle, and inferior cervical ganglia. The inferior cervical ganglion is also known as the stellate ganglion and is commonly thought to supply the

353

­ ajority of sympathetic innervation to the upper extremity. m The sympathetic nervous system is responsible for the classic fight-or-flight response needed in crisis situations.47 The catecholamines are a class of biologic chemicals including epinephrine and norepinephrine. These chemicals function as neurotransmitters that carry messages across synaptic clefts in the sympathetic nervous system. The term adrenergic refers to those synapses in which epinephrine is used. Conversely, the term noradrenergic refers to those synapses in which norepinephrine is used. These catecholamines both have two receptors called α-receptors and β-receptors. α-Receptor stimulation produces skin vasoconstriction, pilomotor contraction, cardiac acceleration, and intestinal relaxation. Excitation of β-receptors causes muscle vasodilation, bronchial relaxation, and cardiac acceleration.49 The catecholamines also have roles as sympathetic hormones released from the adrenal glands into the systemic circulation. In contrast to the primarily adrenergic sympathetic nervous system, the somatic nervous system is cholinergic. The term cholinergic refers to the use of the neurotransmitter acetylcholine to carry an impulse across a synapse. One notable exception to these general rules is that the sweating mechanism, which is sympathetically controlled, is a cholinergic system.50 This fact becomes important later when determining why some CRPS symptoms must be centrally controlled.

THEORIES OF THE PATHOPHYSIOLOGIC MECHANISM Much of what is known about the pathophysiologic mechanisms of CRPS is based on a multitude of animal and human studies. These studies have been compiled in an attempt to support a unifying theory.50 Great strides in our understanding of this complex disorder have been made since Mitchell’s time, but many of our current hypotheses are still unproved and highly controversial.1,2,4-7 The following summary of the proposed mechanisms of disease is intended to provide the reader with an overview of the most plausible current theories in the literature based on the experimental evidence to date. Current leading theories are based on the supposition that a pathologic interaction occurs between the somatic (central and peripheral) nervous system and the sympathetic branch of the autonomic nervous system. This interaction is believed to occur between somatic sensory nerves and sympathetic efferent nerves, when an abnormal synapse is formed.10,50 The abnormal synapse is termed an ephapse.51 The location of the ephapse may be peripheral, central, or both. A short circuit produced by an ephapse allows sympathetic discharge to stimulate sensory nerves directly.52,53 This pathologic sensory nerve stimulation is believed to produce symptoms associated with CRPS, including SMP.50 It has been hypothesized that the most symptomatic regions of an extremity with CRPS correlate with those regions where sympathetic innervation density is greatest.54 A study by Eisenburg, using trans­ cranial magnetic stimulation, showed evidence in CRPS for ­ sensory and motor central nervous system (CNS) hyperexcitability.55 The CNS region affected involved

354

DeLee & Drez’s Orthopaedic Sports Medicine

only the region represented by that extremity and not the entire hemisphere. Under normal conditions, somatic pain receptors (nociceptors) are unaffected by sympathetic outflow. Although somatic sensory neurons do contain the messenger ribonucleic acid necessary to produce α-receptors, sensory nerves normally do not express adrenergic receptors and, therefore, have no interaction with circulating or local catecholamines.50 In CRPS type II, a known traumatic nerve injury is, by definition, the inciting event that leads to a pathologic interaction between the two nervous systems. In this scenario, the pathologic interaction occurs both in the periphery and centrally in the dorsal root ganglion.10,50 Peripheral pathologic interaction occurs when there is complete nerve transection and subsequent development of a neuroma. The transected sensory nerve endings then develop catecholamine sensitivity through the expression of α2-receptors.56 These newly expressed receptors can be stimulated by circulating catecholamines as well as by sympathetic efferent nerves that grow into the neuroma. Central pathologic interaction in the dorsal root ganglion also follows complete nerve injury. Sensory nerve cell bodies in the dorsal root ganglion begin to express α2-receptors, whereas sympathetic efferent nerves that normally innervate blood vessels in the dorsal root ganglion begin to sprout and surround the injured nerve cell bodies (Fig. 7A-2A). It is through these peripheral and central mechanisms that an abnormal coupling between the sensory (pain) nerves and the sympathetic nervous system occurs.50 In simpler terms, this describes a pathologic interaction in which a patient’s anxiety triggers a catecholamine response. Catecholamines would normally stimulate the sympathetic nervous system, producing fight-or-flight responses like accelerated heart rate, increased blood pressure, and vasodilation. Instead, the pathologic interaction of CRPS allows the catecholamines to stimulate the dorsal root ganglion, and the signal is now also interpreted as pain. In CRPS type II, the nerve injury may be complete or partial, as would occur with a nerve contusion or incomplete transection. In this scenario, a slightly different mechanism produces the abnormal interaction. Injured sympathetic nerve fibers lead to a decrease in the sympathetic innervation density in the periphery. This may incite the production of α-receptors by intact sensory nerve endings. This α-receptor expression constitutes an abnormal synapse between sympathetic and somatic nerves (responding to catecholamines) and results in pain (see Fig. 7A-2B).50 The clinical findings of light touch pain (allodynia) and cold hypersensitivity (hyperalgesia) are postulated to occur by a central sensitization mechanism known as the Raja model.57 Nociceptors up-regulate their production of α-receptors in response to injury through the pathway previously described. These α-receptors are persistently stimulated by sympathetic efferent nerve endings, which results in reduced threshold of central pain neurons.58,59 Thus, changes in the dorsal horn cells can occur.13 Light touch–sensing mechanoreceptors or cold-sensing ­thermoreceptors produce the abnormal painful response by stimulating the sensitized central pain neurons. In the Raja model, sympathetic blockade produces a temporary inhibition of norepinephrine release, thereby preventing nociceptor activation. This leads to desensitization of the central pain neurons, thereby reducing the ­painful response to light

�2

NA PAN

PGN

A

SPGN

Nerve Transection NA

�2 NA

Partial Nerve Lesion �2 NA

B

Nociceptor Sensitization NA �2 Bradykinin NGF

PG �2

C NA Figure 7A-2   Influence of sympathetic activity and catecholamines on primary afferent neurons (PAN). A, Nerve transection. The sympathetic-afferent interaction is located in the neuroma and in the dorsal root ganglion. It is mediated by norepinephrine (NA) released from sympathetic postganglionic neurons (SPGN) and α2-adrenoreceptors expressed at the plasma membrane of afferent neurons. PGN, preganglionic neuron. B, Partial nerve lesion. Partial nerve injury is followed by a decrease of the sympathetic innervation density (stippled sympathetic postganglionic neuron). This induces ­up-regulation   of functional α2-adrenoreceptors at the membrane of intact afferent fibers. C, After tissue inflammation, intact but sensitized primary afferents acquire norepinephrine sensitivity. Norepinephrine is not acting directly on afferents; rather, it induces the release of prostaglandins (PG) from sympathetic terminals that sensitize the afferents. Accordingly, bradykininand nerve growth factor (NGF)-induced nociceptor sensitization is also mediated by the release of prostaglandins from sympathetic postganglionic neurons. (Redrawn from Baron R, Levine JD, Fields HL: Causalgia and reflex sympathetic dystrophy: Does the sympathetic nervous system contribute to the generation of pain? Muscle Nerve 22:678-695, 1999. © 1999, John Wiley & Sons. Reprinted by permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.)

touch. This is the reason continuous or repeated ­sympathetic blockade may ultimately lead to down-­regulation of the α-receptors and elimination of the SMP.50 SMP also occurs without known nerve injury, as in CRPS type I. There is evidence that inflammation plays a role in the early stages of CRPS. An initiating noxious event produces a peripheral inflammatory tissue reaction that sensitizes the sensory nerve ending through an indirect mechanism. In this inflammatory induction model,

Complications

norepinephrine and inflammatory mediators such as bradykinin and nerve growth factor lead to the release of prostaglandins from the sympathetic terminals. The prostaglandins then stimulate the nociceptor, producing the pain response. Thus, an indirect action of norepinephrine on sensory nerve endings is postulated (see Fig. 7A-2C).50 The inflammatory induction theory is supported by experimental evidence of nociceptor sensitivity to catecholamines after thermal or chemical stimulus application in animals. In addition, if norepinephrine is experimentally injected into animals, surgical postganglionic sympathectomy can prevent sensitization of the nociceptor. This occurs because norepinephrine can no longer stimulate the release of prostaglandins from the transected sympathetic nerve. Because of these and other experimental observations, it is thought that norepinephrine-induced afferent sensitization is accomplished through an indirect ­mechanism.50

SYMPATHETIC VASOMOTOR AND SUDOMOTOR ABNORMALITIES Both peripheral and central mechanisms have been postulated to explain symptoms of CRPS that are classically thought to be under sympathetic control. These symptoms include swelling, skin color, and local temperature changes that are under vasomotor control, and sweating abnormalities, a sudomotor function. The peripheral mechanism by which these abnormalities may occur is called denervation supersensitivity. After a peripheral nerve injury (CRPS type II), some of the sympathetic fibers that normally innervate blood vessels to control vasoconstriction are injured. This local blood vessel denervation leads to acute vasodilation. Over time, the chronic loss of sympathetic tone leads to increased α-receptor concentration (up-regulation). The α-receptor up-regulation increases blood vessel sensitivity to sympathetic innervation, which induces late vasoconstriction. Consistent with this theory is the clinical finding that some patients initially display warm, swollen, red skin that is followed later by a cold and mottled appearance with less or no swelling.50 Some evidence exists for a central mechanism of vasomotor dysfunction as well. This mechanism may more readily explain the development of sympathetic dysfunction without known nerve injury (CRPS type I). The central control theory is known as decentralization supersensitivity. Although the peripheral and central theories agree that supersensitivity of the end organ does occur, they differ in their respective explanations of how this occurs. The central theory contends that CNS inhibition of sympathetic vasoconstrictor activity is the initiating event. This inhibition leads to vasodilation and subsequent skin redness, warmth, and swelling. Continuous central inhibition can lead to up-regulation of blood vessel α-receptors. Increased α-receptor concentration induces a supersensitive state, causing late vasoconstriction. The central theory thus accounts for clinical signs of sympathetic overactivity despite experimental evidence of reduced sympathetic output. Indeed, measurements of norepinephrine in venous drainage taken proximal to areas with sympathetic

355

dysfunction have shown an overall lower than normal level.60 ­Further support for the up-regulation of adrenergic receptors comes from the finding of increased levels of α-receptors in skin biopsy specimens from involved areas. Whereas the peripheral and central mechanisms differ in their explanations, there is also some evidence that these both play a role in modulating the abnormalities of sympathetic vasoconstrictor function.50 Sweat glands, in contrast to blood vessels, do not exhibit denervation or decentralization supersensitivity; therefore, a direct central mechanism of increased sudomotor function is believed to play a role here as well. A centrally triggered increase in sympathetic outflow to sweat glands may occur, causing the clinical finding of increased sweating.24 That vasomotor dysfunction and sudomotor dysfunction occur by separate mechanisms is confirmed by lack of correlation between these two clinical findings.50 There is increasing evidence that localized neurogenic inflammation might be involved in the generation of acute CRPS (edema, vasodilation, and increased sweating). Venules are more responsive to the constrictive effects of circulating catecholamines than arterioles. This relative outflow obstruction of the extremity explains the initial edematous appearance of CRPS limbs. It also may explain why scintigraphic investigations with radiolabeled immunoglobulins show extensive plasma extravasation in patients with acute CRPS type I. Furthermore, synovial fluid is enhanced in affected joints, as measured by magnetic resonance imaging (MRI). In the fluid of artificially created skin blisters, significantly higher levels of cytokines (e.g., interleukin-6 and tumor necrosis factor-α) were observed in the involved extremity than in the uninvolved extremity.61

CLINICAL PRESENTATION The clinical presentation of CRPS is highly variable and difficult to characterize.62 Classic findings are described, but these are not necessarily the most common clinical findings. Delayed or failed diagnosis is common, partially because of the syndrome’s wide variability and nonspecific set of signs and symptoms. Epidemiologic factors are not particularly helpful in diagnosis because there is no sex predilection in adults, all ages may be affected, and differences in left and right side occurrence are not seen. In addition, there is no particular predilection for involvement of the dominant limb over the nondominant limb.2 The defining criteria of CRPS dictate that a noxious event or peripheral nerve injury initiates the symptoms. The common noxious conditions include blunt trauma, surgical intervention, inflammatory reactions, fractures, arthritis, and others.49 According to the International Association for the Study of Pain the diagnosis of CRPS requires four elements: (1) an initiating noxious event or course of immobilization; (2) continuing pain, allodynia, or hyperalgesia disproportionate to any inciting event; (3) evidence at some time of edema, changes in skin blood flow, or abnormal sudomotor activity; and (4) the exclusion of medical conditions that would otherwise account for the degree of pain or dysfunction. However, a precipitating event may not be detected in 10% of cases. This definition is entirely descriptive in

356

DeLee & Drez’s Orthopaedic Sports Medicine

nature and has failed to achieve universal acceptance for lack of an underlying mechanism.63 The clinical history and physical examination findings of CRPS can be classified into several overlapping categories including sensory disturbances, sympathetic dysfunction, motor abnormalities, trophic changes, and psychological issues. Initial testing of patients should assess four major questions: (1) Is the condition CRPS? (2) Is the ­extremity hot and swollen, cold and atrophic, or in between? (3) Is there a neural or mechanical nociceptive focus? (4) Is the process sympathetically maintained or sympathetically independent?12

Sensory Disturbances Perhaps the hallmark sensory finding of CRPS is pain that is disproportionate to the expected response for the inciting condition. The disproportionality applies to the duration of pain response, the severity of the response, and the distribution of the painful area. Although pain is a perceptual event normally associated with cellular injury or death, CRPS is not associated with ongoing cellular damage in the traditional sense. In fact, persistent pain in the absence of a cellular insult is a hallmark of CRPS. Segmental ischemia secondary to arteriole-venous shunting in the cutaneous circulation may cause cell death and may be partly responsible for the development of arthrofibrosis, osteopenia, abnormal neuroreceptor function, and central pain imprinting.12 Within hours or days after the initial injury, the pain becomes more diffuse and unrelated to the site of the injury. As the syndrome progresses, the painful area expands in a nondermatomal distribution, instead following thermatomes. Progression will also have a preference for distal limb involvement (i.e., stocking or glove distribution). The character of pain is also an important diagnostic clue. Patients commonly describe a burning, shooting, or deep, constant aching. A classic finding is pain with light pressure (allodynia). Patients commonly report an inability to tolerate the faint touch of bed sheets, clothing, or air currents. Night pain is an occasional complaint. The symptoms of hyperalgesia, hyperpathia, hyperesthesia, and dysesthesia may also be present (Table 7A-2). The pain of CRPS may also fluctuate and recur, depending on several common factors. These aggravating factors include overaggressive physical therapy (active and passive motion), environmental and local temperature changes (particularly cold intolerance), dependent limb position, and emotional excitement.41 In addition, up to 50% of patients with chronic CRPS I develop hypoesthesia and hypoalgesia on the entire half of the body or in the upper quadrant ipsilateral to the affected extremity.61

Sympathetic Dysfunction Sympathetic dysfunction is commonly manifested through skin color and temperature abnormalities, swelling, abnormal sweating, and local cold intolerance. The classic skin pattern changes from acutely red, warm, and dry skin to the chronic appearance of bluish or mottled, cold, and moist skin (Fig. 7A-3). One author reports that about 80% of patients will have side-to-side differences in limb temperature averaging 3.5°�� C. ���41

   TABLE 7A-2  Definition of Pain-Related Terminology Pain Allodynia Hyperalgesia Hyperesthesia

Hyperpathia

Dysesthesia

An unpleasant sensory and emotional experience associated with actual or potential tissue damage or described in terms of such damage Pain due to a stimulus that does not normally provoke pain (e.g., light touch) An increased response to a stimulus that is ­normally painful Increased sensitivity to stimulation, excluding the special senses. Hyperesthesia may refer to various modes of cutaneous sensibility, ­including touch and thermal sensation without pain, as well as to pain (e.g., cold hypersensitivity). A painful syndrome characterized by a delayed ­reaction and reaction that outlasts the stimulus and spreads beyond the site of the stimulus (commonly a repetitive stimulus) An unpleasant abnormal sensation, whether ­spontaneous or evoked

Swelling is a common finding, although objective evidence of swelling occurs less frequently than subjective reports.35 The swelling of CRPS is generally painful and usually extra-articular, although joint effusions are also reported.64 Swelling is most pronounced acutely in the course of the disease and usually becomes less pronounced with chronicity. The soft, puffy edema seen acutely is eventually replaced with tight, shiny skin that lacks normal creases.58 The character and severity of limb swelling vary with recurrent acute flares of sympathetic dysfunction. Acutely decreased and chronically increased sweating of the palmar hand and plantar foot can occur.41 A final manifestation of sympathetic dysfunction is cold intolerance.65 This finding is nonspecific but particularly sensitive with regard to diagnosis of SMP. Cold intolerance is commonly first discovered when cryotherapy is employed by the physical therapist to control swelling and a significant painful reaction is exhibited by the patient.49 We recommend office use of the ice test, in which ice is applied to the involved area and the patient is questioned about the sensation. The CRPS response to the ice test is that the ice produces an intolerable burning sensation, whereas the application of ice to the normal limb is described as cold. Cold weather commonly precipitates recurrent flares of sympathetic dysfunction in CRPS patients.49 Patients often report the need to wear socks to bed because of cold feet. The natural history of sympathetic dysfunction has classically been divided into three sequential but overlapping stages.66 The acute stage is characterized by sympathetic hyperfunction manifested by disproportionate pain; red, warm, and dry skin; and extra-articular swelling. The classic presentation of the acute stage lasts less than 6 months. The dystrophic stage begins when the increased sympathetic output succumbs to a period of reduced sympathetic activity, typically 3 to 9 months after onset. This stage is characterized by cyanotic or mottled, cold, moist skin; muscle wasting; thick nails and coarse hair; and other early trophic changes. After a period of chronic sympathetic dysfunction, the atrophic stage begins. This stage is characterized by thin, tight, glossy skin; osteoporosis; and joint

Complications

A

357

B

Figure 7A-3   A and B, In the early stages of complex regional pain syndrome (CRPS), the color of the extremities often changes to blue or dusky red. This 14-year-old girl presented to the emergency department 2 weeks after sustaining an inversion ankle sprain. Her pain, swelling, and dysfunction had increased despite a treatment program of wraps, ice, elevation, and home-based range of motion exercises. A venogram was obtained in the emergency department to rule out deep venous thrombosis, and CRPS was diagnosed soon after. Because an initial course of oral corticosteroids relieved symptoms only partially, the patient had a series of five paravertebral blocks, each of which alleviated pain for a progressively longer time. The patient also participated in a program of physical therapy that included massage to control swelling, active and active-assisted range of motion and strengthening exercises, and weight-bearing as tolerated. After about 4 months, she was able to resume normal activities.

contractures. The clinical usefulness of this classic staging system has been called into question because of the profoundly variable appearance of symptoms from patient to patient as well as the variable time course of any particular patient’s disease.24 The usefulness of the staging system is best realized when the clinician understands that the disease does not progress systematically through stages. Instead, the common scenario is the frequent toggling back and forth between stages in almost random order. The staging system does improve our understanding of the classic presentation as long as the clinician recognizes that exceptions not only exist but are common. For example, many patients with CRPS may present with disproportionate pain as the only symptom. The astute clinician will correctly consider the diagnosis of CRPS even in the absence of other signs of sympathetic dysfunction. One specific exception to the classic presentation, termed cold reflex sympathetic dystrophy, may have implications for prognosis. Van der Laan and colleagues67 found that CRPS type I patients with cold initial skin temperature were much more likely to develop a severe complication, such as infection, ulceration, chronic edema, dystonia, or myoclonus. Sympathetic denervation and denervation hypersensitivity in an injured nerve cannot account for all vasomotor and sudomotor abnormalities observed in CRPS. First, in CRPS type I, there is no overt nerve lesion, and second, in CRPS type II, the autonomic symptoms spread beyond

the territory of the damaged nerve. In fact, there is direct evidence for reorganization of central autonomic control in these syndromes. For example, hyperhidrosis occurs in many CRPS patients, yet sweat glands do not develop denervation supersensitivity. Norepinephrine levels were found to be lower in the affected side. Data support the idea that CRPS type I is associated with a pathologic unilateral inhibition of cutaneus sympathetic vasoconstrictor neurons leading to a warmer affected limb in the acute stage. The locus must be in the CNS. Secondary changes in neurovascular transmission may induce severe vasoconstriction and cold skin in chronic CRPS.61

Motor Abnormalities Motor impairment is a more recently appreciated phenomenon associated with CRPS.68 Although not essential for diagnosis, motor dysfunction does occur. The spectrum of motor abnormality ranges from weakness and disuse to increased tone and hyperfunction.35 In the absence of traumatic nerve injury (CRPS type I), findings of electromyography and nerve conduction velocity studies are usually normal, suggesting that motor abnormalities are centrally mediated, presumably at the spinal cord level.41 Weakness of the affected limb occurs for several reasons, including disuse and muscle wasting. The finding of pseudoparalysis in patients with CRPS suggests that some motor abnormalities may be psychogenic or pain related. Paresis and limb

358

DeLee & Drez’s Orthopaedic Sports Medicine

neglect are also seen. Motor abnormalities ­characterized by hyperfunction include action tremor, myoclonus, hyperreflexia, muscle spasm, and voluntary guarding.40 Dystonia, apraxia, and lack of coordination are findings typically associated with long-standing disease.41 About half of patients with CRPS show a decrease in active range of motion, an increased amplitude of physiologic tremor, and a reduction of active motor force of the affected extremity. In about 10% of cases, especially chronic cases, dystonia of the affected hand or foot develops.61

Trophic Changes Persistent or recurrent CRPS symptoms may result in trophic changes.40 Atrophy of the tissues probably occurs from disuse and as a direct result of sympathetic dysfunction. Tissues known to undergo atrophy in response to chronic CRPS include skin, subcutaneous fat, muscle, tendon, and bone. Thin, shiny, smooth, tight skin is the typical end result of chronic CRPS. Atrophy of subcutaneous fat is particularly evident in the digits and results in a pencil-like appearance. Tendon atrophy is known to occur. The classic finding of patchy periarticular osteoporosis is known as Sudeck’s atrophy, but it is relatively uncommon and nonspecific.69 Chronically increased blood flow to the limb leads to local hypertrichosis, coarsening of the hair, and thickening of the nails. Chronic joint stiffness ultimately leads to fixed contractures.

Psychological Issues In the face of such severe and debilitating symptoms, it is no surprise that psychological disturbances are observed. Great debate exists about whether psychological issues are a cause or a result of CRPS. Ochoa and Verdugo5 have gone so far as to suggest that this disease is a “pseudoneuropathy of psychogenic origin.” One must take care to exclude from the diagnosis of CRPS those patients with factitious disorders, malingering, somatization disorders, and conversion reactions (see Table 7A-1).2,6 These ­psychiatric disorders represent examples without an organic pathologic basis and therefore must be considered separate diagnostic entities. Chronic pain patients frequently have comorbid psychiatric disease. Ranked from most common to least common, they include affective disorders (depression), psychoactive substance use–related disorders, somatoform disorders, and anxiety disorders. The prevalence is reported between 18% and 64%. However, attempts to establish a “CRPS personality” have been unsuccessful. But, is this cause or effect? In daily diaries, it was shown that yesterday’s depressed mood contributed to today’s increased pain and that yesterday’s pain also contributed to today’s depression, anxiety, and anger.63 Whereas some authors have suggested that a particular personality or psychological profile puts some patients at risk for the development of CRPS,26 most authors currently reject this notion.70,71 It is more likely that the emotional and behavioral disturbances are the result of chronic pain and loss of function.18 Psychological disturbances have been shown to resolve when the patient is relieved of the physical symptoms of CRPS. Moreover, Ciccone

and associates70 found no evidence to suggest that CRPS type I patients were psychologically unique compared with patients with other chronic pain disorders. The presence of secondary gain issues, such as worker’s compensation claims, disability cases,26 and pending litigation, also complicates matters. When secondary gain represents a patient’s sole motivation, the individual is considered a malingerer by definition. A patient’s motivation, however, may be difficult or impossible to discern. Patients with apparent secondary gain potential may also have true CRPS organic disease; therefore, they must not be dismissed even when alternative motivations are suspected to play a role. Few would dispute the existence of psychological symptoms in conjunction with many chronic pain patients (and chronic illnesses, in general), including CRPS.70 Symptoms such as depression and anxiety are often observed, and certain subscales of the MMPI evaluation are commonly elevated in chronic pain states.36

DIAGNOSTIC PROCEDURES CRPS is primarily a clinical diagnosis. There is no increased white blood cell count, erythrocyte sedimentation rate, or autoimmune marker with which to definitely diagnose this condition. Thus, there is no diagnostic gold standard. Objective methods used to aid diagnosis of CRPS are aimed primarily at assessing the degree of sympathetic dysfunction. These methods are fraught with high levels of inaccuracy and poor sensitivity or specificity. Because of these factors, the diagnosis of CRPS is made primarily on clinical grounds and then secondarily confirmed with objective tests. Confirming the presence of sympathetic dysfunction (pain, vasomotor, and sudomotor abnormalities) is useful because of the implications for treatment. The following diagnostic tests attempt to relate objective findings to the diagnosis of CRPS.

Hematologic Evaluation Blood studies, such as white blood cell count, erythrocyte sedimentation rate, and C-reactive peptide determination, are indicative of inflammatory processes but are neither sensitive nor specific in the diagnosis of CRPS. These and other studies, including rheumatoid factor and antinuclear antibody titers, are more useful in ruling out other inflammatory conditions.

Radiography Objective assessment of virtually any extremity pain begins with plain radiographs. Standard radiographic series of the affected area typically reveal soft tissue swelling and osteoporosis. Sudeck’s atrophy is classically described as patchy periarticular osteoporosis of the long bones, whereas the osteoporosis seen in the small bones of the hand and foot is more diffuse. Subchondral bone margins are usually retained. These radiographic findings are nonspecific; they may be subtle and generally indicate chronicity of disease.69 Comparison films of the contralateral extremity are recommended. These findings are likely positive only in chronic cases.61

Complications

Bone Scanning Use of the triple-phase bone scan is highly controversial.72-76 The scan consists of images obtained seconds (arterial phase), minutes (soft tissue phase), and hours (bone phase) after the intravenous injection of the radionuclide tracer.40 Increased periarticular uptake of the tracer in the soft tissue phase and bone phase images is the classic bone scan finding in early CRPS. This occurs because of both increased blood flow and increased bone turnover in early CRPS. The increased tracer uptake seen acutely eventually gives way to normal or reduced uptake that is characteristic of the dystrophic and atrophic stages. Whereas the sensitivity and specificity of the bone scan have been reported to be as high as 96% and 97%, respectively,73 these numbers drop significantly 6 months after onset of disease. In addition, a small percentage of patients have bone scans early in the course of disease that show an abnormally decreased tracer uptake. This may account for the wide variability in the literature with regard to reported sensitivity rates (60% to 100%) and specificity rates (80% to 98%).72-74,76 Use of bone scans to track the success of treatment or spontaneous resolution of disease is not recommended.72 There is a natural tendency for the bone scan findings to return to normal over time despite persistence of disease. The clinician must understand that a normal bone scan finding does not rule out CRPS.40 In addition, bone scans taken after successful sympatholysis procedures show a paradoxical intensification of the tracer uptake. A normal scan finding, therefore, may indicate progression of disease beyond the acute phase, and a hyperintense scan finding may be the result of successful sympatholysis.72,75 For these reasons, bone scans should not be used to monitor treatment. Like the finding of osteoporosis on radiographs, a hypointense or normal bone scan finding in a patient with CRPS may be indicative of chronicity of disease and a poorer prognosis. The real usefulness of the triple-phase bone scan, however, is to confirm clinical suspicion of the diagnosis of CRPS early in the course of disease and to help localize symptoms. Some authors have observed that the bone scan can show abnormal findings even before symptoms appear.72

Magnetic Resonance Imaging MRI is generally not used to confirm the diagnosis of CRPS. Typically, MRI is used to rule out other pathologic processes and narrow the differential diagnosis or to discover the inciting event that triggered CRPS. Several nonspecific findings of CRPS have been described on MRI. The characteristic findings acutely include thickening of the skin, soft tissue swelling that enhances with contrast, and intra­articular effusions. Over time, these findings abate, and chronic changes, including muscle atrophy and skin thinning, are observed. Bone marrow edema (once thought to be characteristic) is usually not seen as part of CRPS.64,69,77

Paravertebral Sympathetic Ganglion Blockade The diagnosis of SMP as a component of CRPS is made by the degree of the patient’s response to sympatho­ lytic procedures. Several methods of achieving diagnostic

359

s­ ympathetic blockade are available. The most effective method and the gold standard in diagnosis of SMP is the paravertebral sympathetic ganglion block.39,41,49,78,79 Under fluoroscopic control, the lumbar paravertebral sympathetic ganglion for lower extremity CRPS, or the stellate ganglion for upper extremity CRPS, is injected with local anesthetic. The accuracy of the injection is assessed by evaluating vasomotor and sudomotor functions (skin color, skin temperature, and sweating) in the affected limb. Skin temperature changes, therefore, must be monitored and recorded. One cannot be sure a technically adequate block has been accomplished unless local skin temperature approaches core body temperature. In addition, the induction of Horner syndrome (ipsilateral ptosis, miosis, and anhidrosis) is required to confirm adequate stellate ganglion blockade. Inadvertent spread of local anesthetic to sensory nerve roots is screened by a carefully documented sensory examination. Partial or complete pain relief with intact sensation indicates the presence of SMP.40 Several limitations of this technique exist and must be recognized. First, the technique is dependent on the accuracy of the injection’s location, and the result can be influenced by the anesthetic agent selected. The systemic uptake of the anesthetic may also adversely bias results. Finally, placebo responses may be as high as 33% because of the high expectations of the patient and physician.5 Despite these limitations, the paravertebral block remains the diagnostic procedure of choice to quantify the degree of SMP objectively. Pain that remains after successful paravertebral sympathetic block is by definition SIP. In cases of chronic CRPS, most of the patient’s pain may be SIP; therefore, the diagnosis of CRPS cannot be excluded solely on the basis of the paravertebral block.

Differential Spinal and Epidural Blockade Physicians not skilled in the technique of paravertebral block may substitute a more common interventional method for the diagnosis of SMP, that is, spinal or epidural blockade. In this common technique, a spinal or epidural puncture is performed, and variable concentrations of an anesthetic agent are injected over time. At first, a saline placebo may be injected. The patient is then questioned about symptoms. Increasingly concentrated solutions of the agent are then injected, and after each injection, the patient’s symptoms are again recorded. Low concentrations should penetrate the least myelinated sympathetic nerve fibers and relieve SMP. A moderate concentration of anesthetic agent will additionally block moderately myelinated sensory fibers, and a high concentration adds motor blockade. The weakness of this method is that there is a variable amount of sensory blockade even at low anesthetic concentrations. As a result, it may be inaccurate to attribute the pain relief solely to sympathetic blockade. Therefore, the more sympathetic-specific paravertebral ganglion block is preferred.49

Phentolamine Testing Phentolamine (Regitine) is a nonspecific α-­antagonist with a short (17 minutes) serum half-life. By blocking α-­receptors (mediators of sympathetic pain), the drug ­ theoretically

360

DeLee & Drez’s Orthopaedic Sports Medicine

­ rovides temporary relief to the patient and valuable p diagnostic information for the clinician. The results of the phentolamine test may correlate with the patient’s responses to paravertebral ganglion block, and therefore some use the test to predict outcomes for patients after paravertebral blocks.56,80 The phentolamine test for the diagnosis of SMP was described independently by two different groups in 1991.56,80 Several advantages of phentolamine testing support the use of this test to confirm diagnosis. The test is safe for the patient and simple to administer in the office because adverse reactions are rare, fluoroscopy is not required, and injection into the painful affected area and the use of a tourniquet are not required. Because of the short duration of action of phentolamine, it is not an effective form of treatment. In 1994, a placebo-controlled study was performed comparing saline infusion to phentolamine with or without phenylephrine (77 patients, double blinded). The results showed no significant pain relief with phentolamine. Also, there was no worsening of the pain with intravenous phenylephrine (which would have the theoretical opposite effect of phentolamine and be expected to increase pain).81 In fact, only 7 patients experienced decreased pain with phenylephrine, whereas 22 had a significant decrease in pain with saline infusion.82 These results bring doubt to the validity of this test, and its use is not recommended.

Regional Intravenous Sympathetic Blockade Regional intravenous sympathetic blockade is another means of confirming one’s clinical diagnosis, and the blocking effect is sometimes used for treatment as well. Despite not being available in the United States, guanethidine was frequently used for this technique; however, reserpine and bretylium blocks have also been described.40 The common denominator of these medications is their mechanism of action. The drugs are taken into sympathetic terminals where they deplete norepinephrine for up to 2 days by stimulating its release and inhibiting its reuptake.54 This explains the initial burning pain on injection experienced by many patients, indicating the presence of SMP. Significant pain relief lasting 2 weeks to 6 months may occur and constitutes a positive test result. Little or no pain relief for less than 5 days favors the diagnosis of SIP. Central ephapses (in the dorsal root ganglion) as a source of SMP cannot be excluded on the basis of tourniquet-controlled regional α-blockade. Intravenous administration of guanethidine is not currently approved for this use in the United States, but it has been used extensively in other countries for more than 25 years. Conflicting studies exist in the literature, and the usefulness of regional intravenous sympathetic blockade is highly debated.3,83-85 We do not recommend intravenous regional sympathetic blockade for several reasons. First, the specificity of these medications for sympathetic functions is undetermined. Second, local anesthetics are commonly injected in conjunction with these medications and may confuse results. Third, injection of these medications is often painful for patients and often poorly tolerated. Fourth, the use of a tourniquet on a limb with profound sympathetic dysfunction is generally ­ unfavorable.40,86

Paravertebral ganglion injection is our preferred method of achieving prolonged sympathetic blockade and has the added advantage of blocking central ephapses for a more complete sympatholytic effect.

Vasomotor and Sudomotor Measurements Accurate quantitative assessment of sympathetic functions has proved difficult. A concept that is not necessarily obvious is that sympathetic dysfunction may exist without the presence of pain (SMP or SIP) and may ultimately lead to dystrophic and atrophic changes. This phenomenon has specifically been observed in the contralateral extremity of patients with CRPS.23 As a result of this observation, studies of autonomic function using side-to-side comparisons may be inherently flawed. Accurate quantification of skin temperature, sweat output, blood flow, and edema volume compared with normative values from control groups is a better way to evaluate autonomic dysfunction. Such normative values have not been clearly established; therefore, side-to-side comparison remains the current standard by which autonomic dysfunction is quantified. The most commonly used method of sympathetic dysfunction measurement is thermography.24,35,87-90 Infrared cameras that sense heat are connected to a computer and produce colorized images of affected and unaffected limbs compared with those of control groups. These images provide valuable area-specific information about extremity surface temperature variations. Surface temperature is thought to correlate with surface blood flow, a sympathetically controlled parameter. Some authors have reported on the ability of thermography to provide earlier detection of CRPS.87-89 Although early diagnostic information may be obtained with this method, several drawbacks have prevented its widespread implementation. First, video thermography requires a carefully controlled environment consisting of a draft-free, temperatureand humidity-­controlled room. It also requires a period of acclimatization before testing that can range from 20 ­minutes to 2 hours. Another difficulty in thermography is that the amount of side-to-side temperature difference that constitutes dysfunction is unknown. When this temperature difference threshold is set at 0.4������������������� ° C, the test is sensitive but nonspecific. If, on the other hand, the threshold is set at 1.2����������������������������������������������� ° C, the test is specific but not sensitive. Most authors currently use 1.0��������������������������� ° C as the cutoff value.87 Sympathetic dysfunction exists when the affected extremity is either warmer or cooler than the opposite extremity. When patients are evaluated at all stages of disease, however, most demonstrate cooler affected limbs, reflecting the frequency with which patients with subacute and chronic CRPS are encountered. A somewhat more specific but less sensitive method of sympathetic function analysis is the measurement of sweat output.24,35,91 Dysfunctional sweating is thought to occur independently of dysfunctional vasomotor activity even though both are under sympathetic control.24 This independence of vasomotor and sudomotor control may reflect the different pathophysiologic mechanisms by which each occurs. The most common method of sweat output meas­ urement is the quantitative sudomotor axon reflex test (QSART).24,35,91 In this test, the sweat response to ­topical

Complications

application of acetylcholine or another sweat-inducing agent is measured and compared with resting sweat output levels. The QSART response is mediated entirely by the postganglionic sympathetic neuron. A criticism of this test is that the increased sudomotor activity in CRPS is thought to be centrally mediated. Therefore, the QSART response may not be an accurate reflection of CRPS sympathetic overactivity. As a result, the thermoregulatory sweat test, which stimulates sudomotor function by activation of thermal neurons in the ventral hypothalamus, has been ­ advocated.24 Neither sweat test is widely available, and they are currently used primarily for research purposes. Some authors have advocated measurement of other sympathetic functions, including regional blood flow and edema volume.91,92 Blood flow measurements are made by transcutaneous oxygen pressure measurements, and edema is quantified with volume displacement methods.40 These tests are not widely used, and their sensitivity and specificity for sympathetic dysfunction are unknown.

TREATMENT PRINCIPLES AND METHODS Many treatments have been proposed, yet few are supported by randomized controlled trials (RCTs). The relative benefit of oral medications, compared with widely used treatments of physical therapy, nerve blocks, sympathectomy, intraspinally administered drugs, and neuromodulatory therapies, remains uncertain. Data on CRPS treatment are insufficient and remain largely empirical.81 A conspicuous deficiency of blinded, randomized, placebo-controlled clinical trials evaluating the efficacy of specific treatments exists in the CRPS literature.3 Therefore, many of the current recommendations are based on the art of medicine more than on the science of medicine. Significant placebo effects have been observed in up to 33% of patients with CRPS, emphasizing the need for ­ placebo control. Uncontrolled trials must be interpreted with extreme caution for this reason. Furthermore, because the diagnostic criteria of CRPS have only recently become standardized, comparison between studies is virtually impossible. Despite these limitations, several general statements about treatment principles and specific treatment methods can be made with confidence. Perhaps the most important principle is that of early recognition and treatment.20,29,40,93-99 Ideally, treatment should begin within 2 to 3 weeks of onset. Many authors have emphasized that the long-term prognosis is critically dependent on the amount of time from disease onset to the beginning of treatment. Delay of longer than 6 months is associated with a poor long-term prognosis.94,100,101 More than 80% of patients diagnosed and treated within 1 year of onset of CRPS experience significant improvement of their symptoms; if treatment is initiated after 1 year, permanent functional impairment occurs in half of patients.12 The diagnosis of CRPS mandates a full and thorough investigation in search of a persistent painful focus. Any such painful focus will aggravate and perpetuate SMP. Elimination of this painful focus is a critical step in

361

t­ reatment of CRPS. Whereas correction of a painful focus alone will not always eliminate CRPS, SMP will generally not resolve without elimination of the painful lesion. When the patient is scheduled to undergo sympathetic blockade, ask the patient to think about the character and location of the pain that remains while the block is working. When the burning, nonspecific pain is gone, an underlying pain may be unmasked. The patient can often describe a specific location and may now point to the medial joint line, for example, rather than the entire medial knee. The clinician should remember that MRI and other diagnostic tests may miss small or large lesions. Correctable lesions, such as meniscal tears, chondral lesions, infections, painful neuromas, and many others, that were not previously revealed by MRI are often discovered by arthroscopy.49 Surgery is generally not recommended for patients with active CRPS because of the risk for worsening the pain syndrome. In the senior author’s experience (T.N.L.), however, patients with chronic refractory CRPS with complaints that suggest a painful non-neurologic condition frequently require arthroscopy to correct a painful condition that exacerbates CRPS. Eliminating the painful focus of CRPS can eliminate the condition. Treatment often requires a multidisciplinary team approach. Patients must often seek the cooperative advice of an orthopaedic surgeon, anesthesiologist or pain management specialist, physical or occupational therapist, psychologist or psychiatrist, neurologist or physiatrist, and primary care physician. An open line of communication among health care providers is essential to provide the most effective and efficient care. The primary care physician, orthopaedic surgeon, or pain management specialist will often serve as the team manager to help coordinate the treatment program. Nevertheless, for patients with recurrent exacerbations or flares of CRPS, the patient’s direct access to the pain management specialist improves efficiency and reduces the patient’s frustration. It is imperative always to remain the patient’s advocate. Personality conflicts between patients and health care providers must be set aside, and confrontations are almost always counterproductive. Patients must believe that their physicians, nurses, and therapists are on their side to lend credibility to the prescribed treatment programs. Therefore, the development of rapport and the establishment of a therapeutic alliance between physician and patient are critical to success.36 A sound physician-patient relationship can also help to eliminate the “doctor shopping” tendency that many patients display.17,22 Because specific treatment protocols have not been validated, good clinical judgment must be used to individualize treatment on the basis of the patient’s signs and symptoms.41 The decision to use medications, injections, interventional techniques, therapy, psychiatric treatments, or surgery must be made individually on the basis of all the available medical information. The myriad available treatment options reported in the literature stand as testimony to the fact that few are backed by conclusive scientific support. The widely ranging treatment techniques are generally separable into categories of patient education, physical therapy, medications, psychiatric management, moderately invasive techniques, and surgical procedures.

362

DeLee & Drez’s Orthopaedic Sports Medicine

Physical Therapy Physical therapy with or without occupational therapy is generally regarded as the mainstay of CRPS treatment.32 On occasion, physical therapy alone can reverse CRPS. Many of the medical, interventional, and surgical treatment options are intended to improve patients’ symptoms with the purpose of facilitating an active physical therapy program. Controlled trials demonstrating efficacy of physical therapy are scant,31,32,102 but several guiding principles must be followed in designing therapy programs for patients with CRPS. These principles are based on science, experience, and the recommendations of consensus panels. Confrontations between aggressive therapists and resistant patients are to be avoided. The physical therapy program must never induce intolerable pain because this is always counterproductive. The program must work within the limits of a patient’s pain tolerance, supplemented by medications and blocks. The goals of any therapy program in the treatment of CRPS are prevention and treatment of pain, swelling, stiffness, weakness, and disuse and restoration of function. A four-step program emphasizing gradual progressive return to function has been outlined.36 The first step involves the development of a patienttherapist treatment alliance. By remaining a strong advocate for the patient in all matters, the therapist will gain the confidence and trust of the patient, lending credibility to the treatment program.36 The second step involves motivation of the patient and desensitization and mobilization of the affected limb. Desensitization is accomplished through high- and lowfrequency vibration,103 gentle textured massage,104 contrast baths using temperatures within the pain-free range,103 and transcutaneous electrical nerve stimulation.104 Movement phobias must be overcome. Motion exercises in CRPS are best accomplished through active and activeassisted exercises. Passive motion takes control of the limb away from the patient and is often poorly tolerated. Passive stretching exercise is thought to increase sympathetic output and can thereby aggravate pain, swelling, and sympathetic dysfunction.103 Techniques of immobilization (splints and casts) are generally counterproductive and are avoided when possible. Active motion, elevation, massage, and compression, if tolerated, will also help with edema control.104 The third step involves strengthening and stress loading. Muscle strengthening is best done in an isometric fashion to minimize unnecessary motion. The patient is gradually progressed to isotonic methods. Stress loading is encouraged to prevent disuse and restore functionality. Upper and lower extremity load-bearing activities are implemented gradually through the use of weight-carrying or water exercises. Postural training and balanced use of bilateral extremities are then undertaken with use of BAPS boards. General aerobic conditioning helps maintain range of motion, strength, and balance.36 The last step emphasizes return to normal function. Patients are enrolled in vocational rehabilitation with work hardening. Eventually, a functional capacity evaluation is performed, and job modifications are implemented as

needed. Patients are encouraged to return to work, school, sports, or other daily activity. This program is supplemented by sympathetic blocks and medications whenever appropriate to allow the patient to make steady progress without intolerable pain.36

Medications for Symptomatic Relief Many different medications are used in the treatment of CRPS, but few have more than anecdotal support. The perfect combination of medications for each patient is still based on a trial-and-error approach.36 The correct dosages must be titrated to allow maximal benefit with minimal side effects (Table 7A-3).41 Symptom-relieving medications are intended to increase the patient’s comfort and to facilitate the implementation of a physical therapy program. The following medications are thought to provide some benefit to patients with CRPS,36 although their use must be individualized.41 RCTs for medical treatment of CRPS have been performed for bisphosphonates, adrenergic active drugs, and steroids (and treatment aimed at α-adrenergic receptors and the sympathetic nervous system has not been proved effective in RCTs). Other medications (like gabapentin, tricyclic antidepressants, and ­opioids) have proved effective in RCTs for neuropathic pain conditions other than CRPS. The evidence to date is in favor of use of these medications for CRPS even though RCTs have not been performed for this specific condition.81

Corticosteroids Corticosteroids are particularly beneficial in the early stages of the disease. Their anti-inflammatory effect is most pronounced when the clinical signs of redness, warmth, and swelling are present.3,105,106 We have found that trial of a corticosteroid, such as methylprednisolone (Medrol Dosepak), early after onset may result in complete resolution of the syndrome. Christensen and associates studied 23 patients and reported that 30 mg/day of oral prednisone was significantly better than placebo.105

Antidepressants The use of antidepressants for neuropathic pain is well established.3 The nonspecific serotonin and norepinephrine reuptake inhibitors, such as amitriptyline (Elavil) and desipramine (Norpramin), often provide relief of burning pain. In addition, these drugs can help patients with sleep difficulties and mood and anxiety disturbances. In general, the dose required for neuropathic pain control is much lower than the dose necessary for antidepressant effect. The onset of action of these medications occurs within 2 to 3 weeks, and the peak effect is not felt for 4 to 6 weeks. In our experience, amitriptyline used at bedtime provides effective symptom relief in many patients with burning pain or sleep disturbances. If the side effects (including sedation, dry mouth and eyes, urinary retention, and constipation) are tolerable, the dose can be titrated for maximal benefit.107 Selective serotonin reuptake inhibitors (SSRIs) are not effective for neuropathic pain relief.36

Complications

363

   TABLE 7A-3  Medications Drug

Mechanism of Action

Use

Side Effects

Blocks peripheral α-receptors; ­decreases sympathetic tone

Causalgia

Hypotension

α-Blockers

Phenoxybenzamine (Dibenzyline): 100 to 300 mg (100 mg qhs or tid) Prazosin (Minipress): 1 mg bid to tid Terazosin (Hytrin) Clonidine (Catapres)

α2-Agonist; blocks α1 transmission peripherally and centrally

Hypotension; tachycardia; decreased sexual function; nasal congestion May have fewer side effects Pain relief can be obtained Hypotension with clonidine patches

β-Blockers*

Propranolol (Inderal): up to 320 mg/day β-Adrenergic response presynaptically Central RSD; migraine;   (centrally)   facial pain ����

Timolol (Blocadren) Atenolol (Tenormin)

Hypotension; bradycardia; ­depression; aggravation of asthma and cardiac arrhythmias; decreased libido; decreased memory; sudden withdrawal Sensitivity to catecholamines; resultant cardiac arrhythmias and myocardial infarction

Antidepressants†

Trazodone (Desyrel): 50 mg tid

Blocks serotonin receptors; activates descending pain-inhibitory fibers

Most chronic pain syndromes

Multiple

Tricyclic Antidepressants

Desipramine (Norpramin): 100-200 mg/day Doxepin (Sinequan): 10-25 mg tid Amitriptyline (Elavil): 25 mg tid

Myocardial ischemia; urinary ­retention Orthostatic hypotension

Calcium Channel Blockers

Nifedipine (Procardia): 10-30 mg tid Diltiazem (Cardizem): 30 mg qid, increase to 60-90 qid Verapamil (Calan) Nicardipine (Cardene)

Relaxes smooth muscle; increases peripheral blood flow; decreases ­discharges from ephaptic scars

Peripheral nerve injury, ­especially in decreased blood flow states

Headaches; hypotension

Suppresses pathologic electrical ­discharges in CNS and PNS

Ephapses; burning pain; sharp discharges

CNS depression; drowsiness; multiple

Inhibits bone resorption; direction of action on SMP ­unknown

SMP

Nasal irritation

Anticonvulsants

Carbamazepine (Tegretol): 100 mg PO bid up to 800 mg/day Clonazepam (Klonopin) tricyclic: 0.5 mg bid to tid, may increase Gabapentin (Neurontin) Valproic acid (Depakene) Pregabalin (Lyrica) Other Medications

Salmon calcitonin (Miacalcin) nasal spray: 1 puff daily Bisphosphonates (alendronate IV) Narcotics and Benzodiazepines

Exogenous sources of narcotics and benzodiazepines cause decreases in endorphins and endobenzodiazepines. In brainstem and limbic systems, this may lead to drug dependence, depression, and increased pain. NSAIDs

Cyclooxygenase inhibitor—three actions: 1. Decreased peripheral inflammation is mediated by prostanoids. 2. Spinal—prostanoids facilitate substance P and glutamate pain fiber transmission. 3. Central (supraspinal effects)—present mechanism is unclear. *Not currently accepted for treatment of reflex sympathetic dystrophy. †Question of usefulness of Paxil, Prozac, and others—there are some proponents of these selective serotonin reuptake inhibitors, but they are probably no better than other antidepressants, except for migraine headache. CNS, central nervous system; IV, intravenously; NSAIDs, nonsteroidal anti-inflammatory drugs; PNS, peripheral nervous system; PO, orally; RSD, reflex sympathetic dystrophy; SMP, sympathetically maintained pain.

364

DeLee & Drez’s Orthopaedic Sports Medicine

Membrane Stabilizers Membrane stabilizers include medications from several therapeutic categories, including anticonvulsants, antiarrhythmics, and local anesthetics. The use of many different membrane-stabilizing medications is reported in the literature with mixed results. Controlled trials, however, are rarely found. The anesthetic agent lidocaine (Xylocaine) has been administered locally through several different routes, including intravenous, subcutaneous, and topical, and as the oral analogue mexiletine (Mexitil).30,37,108 Results have been favorable in some articles,3 but only short-term follow-up is commonly reported.30,108 Perhaps the most promising drug in the membrane stabilizer category is the anticonvulsant agent gabapentin (Neurontin). This drug has recently shown promise in the relief of burning neurogenic pain and improvement in sleep patterns. The mechanism of action of gabapentin is unknown, but it is believed to act centrally. It is administered orally in doses up to 1200 mg/day in divided doses. Major advantages of gabapentin are the absence of significant drug interactions and the more tolerated sideeffect profile compared with antidepressants.109 Experimental data suggest that gabapentin acts at multiple central sites and most likely modifies calcium currents, ultimately causing a decrease in firing of the transmission cell or a decrease in the release of certain monoamine neurotransmitters. These mechanisms might underlie the effects of gabapentin on allodynia. In a pilot study, gabapentin at 2400 mg daily was effective in reducing cold and tactile allodynia.13 A recent addition is the medication pregabalin (Lyrica). Both gabapentin (Neurontin) and pregabalin (Lyrica) block the static and dynamic components of mechanical allodynia induced by streptozocin in rats.110 Allodynia may also be treated with drugs that antagonize the N-methyl-d-aspartate (NMDA) receptor. This receptor is thought to play a key role in the windup and central sensitization involved in the induction and maintenance of CRPS or neuropathic pain. Some studies suggest the NMDA antagonist ketamine is effective in treating allodynia in patients with postherpetic neuralgia, chronic posttraumatic pain, and chronic neuropathic pain.13 In selected patients with severe refractory CRPS, epidural administration of ketamine, as well as the adrenoceptor agonist clonidine, induced analgesia associated with marked side effects like sedation and hypotension.61 A study of 33 CRPS patients treated with subanesthetic intravenous infusions of ketamine showed 76% experienced relief of pain. This increased to 100% response with up to three repeat infusions. Relief lasted for 3 months to 3 years, and patients who responded for shorter durations received repeat infusions. Ketamine is the only potent NMDA-blocking drug available for clinical use, and although it may have more than one mechanism of action, it is thought to work by blocking NMDA receptors that support central sensitization.111

Calcium Channel Blockers The calcium channel–blocking agent nifedipine (Procardia), in doses of 10 to 20 mg given 3 times daily, has been associated with pain reduction and improvement in signs of sympathetic dysfunction in uncontrolled trials. They act by decreasing arteriole-venous shunting, minimizing

segmental ischemia, and decreasing pain in the atrophic extremity.12 Care must be taken to warn patients of the possibility of adverse effects, including orthostatic hypotension and headaches.29 In general, nifedipine is safe and effective in our experience, particularly when it is used early in the course of the disease.

Nonsteroidal Anti-inflammatory Drugs Nonsteroidal anti-inflammatory medications may be useful in mild cases of CRPS that display typical inflammatory signs of rubor, swelling, and warmth. These medications irreversibly inhibit the enzyme cyclooxygenase, thereby reducing production of a prostaglandin. Their use as analgesic agents is generally reported to be ineffective for SMP,3 but they may be helpful when arthritis, tendinitis, or other inflammatory disease is a painful focus. Side effects include gastrointestinal ulceration and renal or hepatic failure.36

Capsaicin Topical capsaicin cream in concentrations of 0.025% and 0.075% may be useful for localized areas of hyperalgesia from neurogenic causes.3,36 Capsaicin is a protein derived from a common hot pepper plant and acts by depleting stores of substance P from sensory neurons. Because of the initial release of substance P, the cream produces an initial burning sensation. Repeated application results in desensitization and inactivation of depleted sensory neurons. Early reports indicate that the desensitizing effect may be reversed within 2 to 4 weeks after discontinuation of the cream.112

Opioids and Benzodiazepines Opioid use in CRPS is highly controversial. No controlled trial demonstrating long-term efficacy and safety has been undertaken.36 Few would dispute the efficacy of narcotic agents in temporarily reducing pain. Chronic use, however, leads to the development of drug tolerance, rendering opioids ineffective for pain control. Tolerance to opioids also creates problems of perioperative pain management when surgical intervention is necessary. The potential for drug abuse and the development of physical drug dependency are also strong arguments against the use of opioids and benzodiazepines in any nonterminal chronic pain state. In rare cases, chronic opioid administration is necessary. To minimize the potential for abuse, it is recommended that one physician dispense all narcotics.49 We recommend avoidance of the routine use of opioid analgesics and benzodiazepines by patients with CRPS. When narcotic use is unavoidable, such as in patients already addicted, pain management consultation is appropriate. Tramadol has dual action on serotonin and norepinephrine receptors (and weak action on μ-opioid receptors). It has been proved effective in two RCTs for peripheral neuropathy and may also be useful in CRPS.81

Calcitonin and Bisphosphonates Calcitonin has been shown to be effective in increasing function in CRPS limbs and for a mild decrease in pain.81 Its efficacy in the subcutaneous or intranasal form has yet

Complications

to be established. When it is effective, however, the result is usually observed soon after the initiation of treatment.36 However, Bickerstaff and Kanis treated 40 patients with intranasal calcitonin and reported no significant improvement compared with patients receiving placebo after 3 months of therapy.81 Bisphosphonate therapy improved some patients’ pain and swelling in several open trials.113-115 It is one of the most thoroughly studied therapies and, in a placebo controlled trial of 118 patients, showed statistically significant improvement with active therapy. Bisphosphonates are powerful inhibitors of bone resorption through the inhibition of osteoclasts. Relief of pain and swelling may be caused by an effect on prostaglandins.115 Also, CRPS can be associated with increased bone resorption and patchy osteoporosis, which might benefit from bisphosphonates. One study treated patients with intravenous alendronate and reported decreased pain, increased function, and increased bone density. Interestingly, the increase in bone density occurred in the CRPS limb, but not in the contralateral limb.116

Psychotherapy Early in the course of disease (0 to 2 months), psychological changes are not present because patients expect to be cured. Only mild abnormalities are seen on MMPI evaluations. After 2 to 6 months, patients become anxious about their disease, and this is accompanied by a progression of the MMPI abnormalities. Education of the patient is critical in this phase to lessen anxiety. Psychological counseling and stress management training are appropriate,95 and low-dose antidepressants may be necessary. After 6 months, nearly all patients demonstrate a degree of depression because of sleep deprivation, chronic pain, and anxiety. If the response to suicide risk testing is abnormal, a 10-fold increase in the risk for suicide is present, and psychiatric hospitalization is required. Antidepressants in higher doses may be required. Group therapy with patients with chronic pain can be beneficial. After 8 months, depression wanes as patients begin to accept the chronic disease state and adapt to it. Attendance of family members at group therapy sessions can be helpful.36

Electroconvulsive Therapy Electroconvulsive therapy (ECT) has been reported to relieve refractory chronic depression for more than 50 years. It is also reported to relieve neuropathic pain, trigeminal neuralgia, phantom pain, and CRPS. It appears most effective in neuropathic pain and CRPS, but few authors advocate its use when major depression is not also present. The mechanism of effect is not well understood but may be rooted in its effect on the thalamus. Using single-­photon emission computed tomography (SPECT), Fukui and coworkers showed that ipsilateral thalamic cerebral regional blood flow (rCRBF) was reduced in both neuropathic and CRPS pain patients. When this pain was relieved during ECT for comorbid depression, it was associated with a normalization of rCRBF.117 Fukui and coworkers postulated that thalamic perfusion increases immediately after onset of symptoms as a reaction to the

365

pain, then decreases as an adaptive response or as a possible feedback mechanism to minimize pain ­perception.14

Sympatholysis Sympatholytic measures may be used for diagnosis as described previously, but they are more often used in treatment of CRPS. Sympatholytic procedures for diagnosis must have specificity for the sympathetic nervous system, but sympatholysis in treatment need not be specific. In general, continuous sympatholysis is probably more effective than pulsed sympatholysis, but this has not been verified in the literature. Continuous or repetitively pulsed sympathetic blockade may lead to complete resolution of the syndrome in many cases, particularly when it is used early.49 As a result, sympatholytic techniques are commonly implemented soon after diagnosis when a short trial of conservative measures fails. Sympatholysis can be accomplished through topical patches, oral medications, interventional techniques, and surgical procedures.

Adrenergic Active Drugs (Clonidine and Oral Sympatholytics) Topical Clonidine Topical clonidine (Catapres) patches may have efficacy in reducing localized hyperalgesia.118,119 Clonidine is an α2-receptor agonist that inhibits presynaptic release of norepinephrine.40 Clonidine has also been administered orally and by epidural injection in patients with CRPS, but further research is needed to evaluate efficacy and safety.3

Oral Sympatholytics Clinical usefulness of oral α-blockers such as prazosin (Minipress), terazosin (Hytrin), phenoxybenzamine (Dibenzyline), and reserpine is suspect. In theory, by blocking α-receptors, the drugs interrupt sympathetic outflow.29 These medications have significant cardiovascular side effects because their α-receptor–blocking effects are systemic, and their long-term efficacy is unknown.36

Interventional Sympatholysis Paravertebral sympathetic chain ganglion blockade techniques were described earlier. Controlled trials demonstrating significantly better results than placebo injections have not been done. Upper extremity CRPS requires stellate ganglion blockade, and lower extremity CRPS is addressed through lumbar ganglion paravertebral blockade. Continuous blockade can be achieved through the use of a constant catheter infusion. More commonly, however, sympatholysis is achieved by repetitive injections of the ganglion.100,120 Administration of the blocks within 6 months of onset of disease is associated with a much better longterm success rate.100 Duration of relief obtained from each injection is recorded.121 Progressively increasing duration of relief with each subsequent block is an ­indication that repeating the blocks as many as 10 to 15 times may be beneficial.79 Ultimately, the longer the pain relief after sympathetic blockade, the better the ­ prognosis.49 We advocate

366

DeLee & Drez’s Orthopaedic Sports Medicine

early use of multiple paravertebral ganglion blocks when a short course of medications and physical therapy fail to demonstrate significant improvements. Interestingly, one prospective study showed that, in patients with CRPS episodes, perioperative stellate ganglion blocks can significantly reduce the recurrence rate of this disease.61 Continuous cervical or lumbar epidural infusion can be used to treat SMP, and it is the anesthetic method of choice when surgical intervention is necessary.96 Bupivacaine 0.25% (Marcaine) with or without an opioid such as fentanyl (Sublimaze) is administered through an indwelling catheter in the epidural space. Optimal dosage allows sensory and sympathetic blockade while allowing motor function so that an active physical therapy program can be performed.94 The infusion is continued for 2 to 5 days postoperatively to minimize the risk for perioperative exacerbation of disease.96 Continuous epidural infusion requires meticulous nursing care to prevent complications. Sacral and heel sores are prevented by log-rolling the patient every 2 hours, using heel pads and an egg crate or air mattress, and requiring patients to get out of bed at least 3 times daily. Physical therapy should start on the first postoperative day. Patients must be protected against deep venous thrombosis with mechanical means such as sequential compression devices and compression stockings because anticoagulant medication is contraindicated. Urinary retention may require insertion of a Foley catheter, and bladder voiding must be carefully monitored after removal of the catheter. Our procedure is to stop the infusion the night before planned discharge. The patient’s pain is evaluated on the morning of discharge. If symptoms are quiescent, the catheter is removed, and the patient is discharged. This protocol generally prevents iatrogenic recurrence of SMP when surgical intervention is necessary. Other methods of regional neural blockade are also possible, including brachial or lumbar plexus blocks, axillary blocks, and isolated peripheral nerve block (e.g., saphenous nerve block).40 The efficacy of these techniques is unknown in the treatment of CRPS. Implantable pumps and intrathecal catheters are being investigated for long-term administration of a variety of medications including bupivacaine and opioid analgesics. These techniques are currently used in severe or refractory cases.28,40

Surgical Sympathectomy Surgical sympathectomy has been used in chronic unrelenting cases of SMP with mixed results. Surgical sympathectomy is considered only when regional sympathetic blockade provides relief, but the effect is short-lived. As with many of the treatments of CRPS, results are best when it is performed early after onset of disease.42,99 Relief after sympathectomy may initially be dramatic. The effect may be short because the symptoms tend to recur within 2 to 5 years. Multiple explanations for this recurrence have been postulated. Incomplete surgical removal of all sympathetic innervation to an extremity may be a common cause of failure.122 Upper extremity sympathectomy must include not only the stellate ganglion but also the T2 and T3 ganglia for complete sympathetic denervation.99,122

Lower extremity complete sympathectomy should include removal of L2, L3, and L4 ganglia and also extend above the diaphragm, possibly as high as T10.49 Failures have also occurred because of collateral reinnervation from the contralateral ganglion.123 Therefore, contralateral sympathectomy is sometimes required. Bilateral interruption of sympathetic outflow carries risk for bowel, bladder, or sexual dysfunction. Another cause for concern from sympathectomy is the sudden development of deep muscle ache that has a character different from CRPS pain. Postsympathectomy neuralgia is usually self-limited and typically resolves in 2 to 3 months. In addition to open surgical disruption, ablation of the sympathetic chain ganglia may be accomplished in a number of ways. Percutaneous chemical ablation of sympathetic ganglia with phenol or alcohol is possible but risks damage to adjacent structures.36 Radiofrequency percutaneous sympathectomy is reported as a safe means of sympathetic interruption, but efficacy is unknown.124 Endoscopic sympathectomy is also advocated because of reduced morbidity compared with open sympathectomy and equal efficacy.125,126 We do not recommend surgical sympathectomy except as a last resort in chronic refractory cases.

Neuromodulation Techniques Neuromodulation techniques, including peripheral nerve stimulation and spinal cord stimulation, are hypothesized to work by one of two theories. In the gate control theory, stimulation of large myelinated nerve fibers blocks transmission in the smaller pain nerve fibers. An alternative theory postulates that nerve stimulation causes the release of endogenous opioids.40 Most of this literature is based on small prospective and retrospective observational studies. Peripheral implanted nerve stimulation may provide relief in patients with symptoms relatively confined to the distribution of one major peripheral nerve. Stimulating electrodes are surgically placed directly on the involved nerve.127 Hassenbusch and colleagues128 documented pain relief in 63% of patients with 2- to 4-year follow-up after peripheral nerve stimulator implantation. Spinal cord stimulation (dorsal column stimulation) has been used for a variety of chronic pain conditions for several years.129-131 Electrodes are implanted through laminotomy or percutaneously in the epidural space at C4 or C5 for upper extremity treatment and at T9 to T11 for lower extremity treatment. A trial period allows confirmation of symptom relief before surgical implantation of the stimulator.132 In one published report, 12 of 12 patients achieved pain relief with an average 41-month followup.133 Stanton-Hicks and coworkers36 stated that 70% of patients with CRPS will respond to peripheral nerve or spinal cord stimulation. Spinal cord stimulation (SCS) has no long-term effect on experimental pain thresholds (thermal or mechanical allodynia) and only limited effect on dynamic and static hyperalgesia. A consensus statement from a panel of experts on SCS states that it should be considered when traditional therapies have failed. The only RCT of SCS is by Kemler, who studied 36 patients and reported a success rate of 56% with an average decrease from 7.1 to 3.5 on the visual analog pain scale.134

Complications

In the later stages of CRPS, it has been reported that results with stimulators are superior to those achieved by using either sympathetic blocks or sympathectomy procedures.12 Spinal cord stimulation in properly selected patients may be more cost-effective, provide better outcomes than standard treatment, and be more effective compared with physical therapy alone as demonstrated in a randomized, placebo-controlled trial.12 Complications include infection, nerve damage, and electrode displacement or breakage.132 The complication rate is between 30% and 64%, and ­ battery changes are required. The percentage of patients with at least one complication is between 9% and 50%. The reoperation rate is between 11% and 50%. Some studies report patients with multiple complications requiring several reoperations. Only three of the studies reviewed had a low complication rate between 0% and 13%.134 The best results were observed in patients with CRPS type II limited to a single nerve. The longterm results of patients treated with spinal cord stimulation are unknown.12

Surgical Intervention Surgical intervention in patients with CRPS should be avoided whenever possible. In patients with chronic refractory CRPS with complaints that suggest a painful non-neurologic condition, however, arthroscopy (or other surgery) is frequently necessary to correct a painful condition that exacerbates CRPS. In those cases, surgical intervention becomes unavoidable and should be performed as soon as CRPS symptoms are quiescent.93,135 Optimal control of CRPS symptoms is desirable at the time of surgical intervention. Continuous postoperative epidural infusions for 2 to 5 days will help prevent exacerbations of CRPS.94,96 In general, tourniquet hemostasis is to be avoided.86 Exacerbation of CRPS after surgical procedures is reported to occur in 13% to 47% of patients.86,135 In the senior author’s experience, when a specific pathologic focus for CRPS is identified and corrected, and postoperative blockade is performed for 2 to 3 days after the surgery, the CRPS stimulus can be eliminated. None of the senior author’s patients have had a worsening of their symptoms with this approach, and many have achieved significant improvements. CRPS of the knee commonly requires surgical intervention to correct intra-articular disease or to rule out intraarticular derangement definitively. Patellofemoral surgery is associated with a high rate of exacerbation of CRPS symptoms.136 Injury to the saphenous nerve or its branches can produce a saphenous neuritis, which is a form of CRPS type II.62,137,138 Operative resection of painful neuroma or release of nerve entrapment has been shown to be effective in selected cases of CRPS type II.46,101 To avoid injury to the infrapatellar branch, a medial portal incision can be made horizontally and above the joint line. In addition, the concomitant vein can be transilluminated with the arthroscope in the lateral portal to avoid the nerve.139 Chronic CRPS of the knee is associated with the development of patella infra and arthrofibrosis.140 Lankford states that sympathetic blocks should be performed and that the reflex sympathetic dystrophy process must be allowed to “cool down” for at least 1 year, during

367

which time the patient should actively engage in physical therapy before any further surgical procedures. In postarthroplasty patients with CRPS, Katz and colleagues state that the elective surgery to correct coexistent mechanical dysfunction (aseptic loosening, ligament imbalance, or component malalignment) should be delayed until the CRPS symptoms are “under good control.” They recommend that the patient undergo a series of sympathetic blocks before the anticipated surgery. When proceeding with surgery, specific actions may help prevent or mitigate the recurrence of CRPS. First, using regional anesthesia at the time of surgery has been reported to decrease the recurrence rate when compared with general anesthesia. The regional blocks used were epidural anesthesia for lower extremities and brachial plexus blocks for upper extremities. Second, many surgeons use local anesthetic infiltration for carpel tunnel surgery. It is unlikely that this provides a perioperative sympathectomy, and this technique is associated with a high recurrence rate. A stellate ganglion block may be beneficial in this instance. Third, intravenous regional anesthesia with lidocaine or clonidine can manage both acute postoperative pain and the symptoms of CRPS. The complication rate is low, and the procedure is less technically demanding than a stellate ganglion block.141 Although frustrated patients with chronic unrelenting CRPS may demand amputation, results of this treatment are alarming. Dielissen and associates142 reviewed the results of 34 amputations in patients with CRPS with intractable pain. Only two patients obtained complete pain relief after amputation. Twenty-eight patients experienced CRPS symptoms in the stump, and phantom pain and phantom sensations occurred in 24 and 29 patients, respectively.142 Because of these appalling statistics and poor outcomes, amputation for treatment of CRPS should be avoided.

COMPLEX REGIONAL PAIN SYNDROMES IN CHILDREN Complex regional pain syndrome is a commonly missed diagnosis in children.143 Wilder and colleagues144 found that the average time between disease onset and diagnosis in children was 12 months. CRPS affects children in a girl-to-boy ratio of 3:1 to 7:1.143,145 The most common age group affected is the 9- to 15-year-old cohort.40,146 In contrast to CRPS in adults, children have lower extremity involvement more commonly than upper extremity disease (lower-to-upper extremity ratio of 6:1).40,143,145 Bone scan findings in children with CRPS are typically normal; therefore, bone scans are used primarily to rule out other causes of extremity pain, such as infection, tumor, osteoid osteoma, and stress fracture.40 In addition, this disease in children is typically triggered by emotional or psychological stress or a minor trauma.143 The treatment program usually centers on stress management, psychological counseling, and behavioral modification techniques. Psychiatric referral is recommended early as a matter of routine.143 Children with CRPS commonly have dysfunctional family situations; therefore, family therapy is often needed as well.36

368

DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method

of

Treatment

Orthopaedic surgeons are frequently the first to see the signs and symptoms of patients with CRPS. This places the surgeon in the best position for early diagnosis and successful treatment of these patients (Fig. 7A-4). Some of the simpler interventions, when used early, have the highest chance of success; therefore, the orthopaedic surgeon should be comfortable using some of these techniques. When patients present with chronic CRPS, the orthopaedist’s role may be limited to finding and correcting a painful non-neurologic cause that perpetuates the CRPS. Long-term management of patients with chronic CRPS is often best undertaken in a specialized pain clinic. Patients may often present in the office with early signs of CRPS, such as stiffness of the hand and upper extremity, burning pain, or changes in sweating. In patients with symptoms of short duration, massage and gentle active range of motion will occasionally cause these symptoms to resolve. If physical measures alone are not enough to resolve symptoms, our preference is to provide the patient with a steroid burst and taper (Medrol Dosepak). Once the course of steroid is completed, the patient should return and report symptoms to the physician. When taken early enough in the disease course, this treatment will often prevent progression and actually reverse the effects of CRPS. If this provides partial but incomplete relief, a second steroid burst and taper will sometimes be prescribed. For those patients not responding to steroids who are in a later stage of CRPS, we recommend a quick referral to a skilled anesthesiologist for cervical or lumbar sympathetic blockade. Skin temperature should always be monitored, and repeated blocks should automatically be scheduled if the patient obtains good but temporary relief. An appointment should be arranged following the sympathetic blocks to examine the patient when the SMP has been suppressed and only the SIP remains. During this interval, the examiner may now be able to determine a pathologic painful stimulus. For example, a patient may have global knee

pain before sympathetic blockade, but after the blockade may localize the pain to the medial joint line. This suggests a meniscal tear as a possible focus for the CRPS condition. ­Arthroscopy under these conditions can be very helpful. The CRPS symptoms should be under control and preoperative blockade performed. After surgery, the patient is placed on a continuous sympathetic block for 2 to 3 days. Each postoperative day the dose is decreased, to be increased again only if the patient reports a recurrence of SMP symptoms. Using this approach, surgical intervention is as safe as possible, and the senior author has never seen a patient’s CRPS worsen. Orthopaedists who are familiar with the use of several drugs may continue their use in patients with persistent symptoms of CRPS after treatment. Salmon calcitonin (Miacalcin) can be used with few side effects and may help prevent bone loss in the female patient as well. Nifedipine (Procardia) has relatively few and easily controllable side effects and appears to help patients, particularly when they have flares of symptoms during cold weather. Most orthpaedists are not comfortable using multiple-drug combinations; therefore, these patients would be better served by referral to a neurologist or other physician experienced in multipledrug therapy. Patients who do not respond to these methods need specialized care and should be referred to a pain center experienced with the treatment of CRPS. The exact cause and nature of CRPS are still not completely understood. Its many manifestations, however, are well described. Orthopaedic surgeons can perform a tremendous service to the patient by recognizing the symptoms and initiating treatment as early as possible. Orthopaedists can also play a major role in the treatment of CRPS by finding underlying painful organic causes and correcting these through methods that are as painless as possible. The orthopaedist is both the primary care physician and the specialist dealing with patients with CRPS. Appropriate medical and surgical treatment works best to bring about recovery in these patients.

Clinical suspicion of acute SMP Technically adequate? Initial work-up • Radiography • MRI • Bone scan • Pain questionnaire

Oral steroid burst and taper

No

Yes

Pain relief

No

Yes Paravertebral block Symptoms do not resolve or quickly recur

Motion loss

Yes

Continue work-up Phentolamine Continue work-up Phentolamine

No Parvertebral series

P A I N R E L I E F

Full

Partial

Stop and observe

Medication TENS

Symptoms resolve

Observe Figure 7A-4   An algorithm showing the decision-making process in the diagnosis and treatment of sympathetically maintained pain (SMP). MRI, magnetic resonance imaging; PT, physical therapy; TENS, transcutaneous electrical nerve stimulation.

Complications

Children with CRPS typically respond to conservative forms of treatment better than adults do.36,40,146 Aggressive physical therapy employing exercise programs that build up to 5 hours per day are the mainstay of treatment.146 Parents should not be allowed to be present during exercise sessions because parental presence may hinder the child’s progress.146 The judicious use of medications, such as nonsteroidal anti-inflammatory drugs, tricyclic antidepressants, or anticonvulsants, is sometimes helpful.36,144 Sympathetic blocks become necessary only when physical therapy is intolerable or unsuccessful at resolving symptoms.40 When sympathetic blockade techniques are necessary, a single block is generally sufficient. Repeating sympathetic blocks multiple times, as is commonly done in adults, is usually not required in children.36 In general, the prognosis of the disease in children is for an excellent recovery in most cases.36,40,145,146 Murray and coworkers143 reported that the median time from diagnosis to recovery was 7 weeks in most of their patients, although late recurrences were common. In rare cases in children, however, a severe and debilitating form of the syndrome may occur that may be refractory to traditional conservative treatment methods.143 These patients commonly require more aggressive forms of treatment and typically experience a chronic or recurring form of the disease. As an alternative to repeated injections, a continuous infusion catheter technique for sympathetic blockade is preferable. Only in the most severe and refractory cases is spinal cord stimulation or sympathectomy required.36 In dealing with athletic children who develop CRPS, the clinician must be sensitive to the possibility that a child’s disease may offer an escape from intense competition and parental expectations. Therefore, return to athletic participation must not be the primary goal; rather, treatment should focus on pain relief, functional rehabilitation, and improvement in school attendance.40 Overall, Wilder and colleagues144 reported a 50% rate of return to sports in children with CRPS who were involved with athletics before the onset of disease. Many of those who did not return to athletic participation required long-term management plans.144

PATIENT EDUCATION AND INFORMATION Many patients report that the establishment of a definitive diagnosis provides significant emotional relief from the torment of not knowing what is wrong. Once the diagnosis is made, it is the job of the clinician to help educate the patient about the disease through careful explanation and question answering. This can be reinforced in several ways, including the distribution of printed information booklets on CRPS. In addition, the patient can be provided with the names of support organizations, which are now easily and widely accessible through the Internet. Several support groups currently exist on the World Wide Web, which provides information and support to patients with CRPS. These sites are the Reflex Sympathetic Dystrophy Syndrome Association of America (www.rsds.org), the International Reflex Sympathetic Dystrophy Foundation (www.rsdinfo.com), and the Reflex

369

Sympathetic Dystrophy Coalition (www.rsdcoalition. com). Finally, information can be found at WebMD and Yahoo health groups (http://health.groups.yahoo.com/ group/RSD). C

r i t i c a l

P

o i n t s

l CRPS is a clinical diagnosis. A high level of suspicion is required.

l A complete and integrated mechanism underlying CRPS is unknown.

l Implement treatment before the limb becomes ­discolored

l l l

l l l l

and has temperature changes. Success rates are improved the earlier you start treatment. Start treatment when pain that seems out of proportion is the only manifestation of the disease. In the clinic, an ice test can be used to diagnose CRPS. The patient will report intense burning pain to the affected limb, whereas ice to the unaffected limb will be reported as simply cold. Randomized, controlled trials have been performed for treatment with paraspinal blocks, vitamin C, bisphosphonates, corticosteroids, and adrenergic active drugs. Author’s preferred treatment: Prescribe a Medrol Dosepak and physical therapy for desensitization and mobilization. If the Medrol Dosepak is partially ­ successful, prescribe a second Dosepak. Should this fail to provide complete relief, prescribe Neurontin and ­refer to pain management for paraspinal sympathetic ganglion blocks. Surgery should be avoided while a CRPS flare is ­ongoing. Some reports state that tourniquets should be avoided in limbs with CRPS. We believe the important factor is to minimize tissue trauma as much as possible. Surgery may be necessary to eliminate the painful focus initiating or exacerbating the CRPS condition. If surgery is performed, it should be done under ­epidural anesthesia. Continuing the epidural for 2 to 5 days after surgery can help reduce the incidence of postoperative CRPS flare.

S U G G E S T E D

R E A D I N G S

Adami S: Bisphosphonate therapy of reflex sympathetic dystrophy syndrome. Ann Rheum Dis 56(3):201-204, 1997. Grabow T: Complex regional pain syndrome: Diagnostic controversies, psychological dysfunction, and emerging concepts. Adv Psychosom Med 25:89-101, 2004. Harden RN: Chronic neuropathic pain: Mechanisms, diagnosis, and treatment. Neurologist. 11(2):111-122, 2005. Koman LA, Smith BP, Ekman EF, Smith TL: Complex regional pain syndrome. Instr Course Lect 54:11-20, 2005. Mellick GA, Mellick LB: Reflex sympathetic dystrophy treated with gabapentin. Arch Phys Med Rehabil 78:98-105, 1997. Reuben S: Preventing the development of complex regional pain syndrome after surgery. Anesthesiology 101:1215-1224, 2004. Rowbotham M: Pharmacologic management of complex regional pain syndrome. Clin J Pain 22:425-429, 2006. Stanton-Hicks M, Baron R, Boas R, et al: Consensus report. Complex regional pain syndromes: Guidelines for therapy. Clin J Pain 14:155-166, 1998. Teasdall R: Complex regional pain syndrome (reflex sympathetic dystrophy). Clin Sports Med 23:145-155, 2004. Wang JK, Johnson KA, Ilstrup DM: Sympathetic blocks for reflex sympathetic ­dystrophy. Pain 23:13-17, 1985.

R eferences Please see www.expertconsult.com

370

DeLee & Drez’s Orthopaedic Sports Medicine

S e c t i o n

B

Deep Venous Thrombosis and Pulmonary Embolism Marc M. DeHart The ability to form a clot after injury prevents bleeding, which can threaten local tissues, and if extensive, can be fatal. Clotting at improper sites (thrombosis) or clots that migrate (emboli) can block vessels and cause tissue damage or organ failure. Disorders of the balance between the transport function of liquid blood and the hemostatic function of solid thrombus or clot can threaten the survival of an organism. Thrombophilia and hemophilia represent disease states at opposite ends of the spectrum of clotting function. Thromboembolic disease is a common important complication of orthopaedic surgery (Table 7B-1). Clinical problems of venous thromboembolic disease include symptomatic deep venous thrombosis (DVT), increased risk for future recurrent thromboembolic disease, chronic postthrombotic syndrome (PTS), and pulmonary emboli (PE). PE is the leading cause of readmission and the most common cause of death occurring within 3 months of major hip surgery. Fatal PE has been reported after minor orthopaedic procedures such as arthroscopy and ankle fractures. However, most thromboembolic disease is asymptomatic or causes so few symptoms that it is clinically unrecognized. This dichotomy in clinical severity leads to the controversy that surrounds selecting the most appropriate method of venous thromboembolic disease prophylaxis.

RELEVANT ANATOMY AND PHYSIOLOGY The venous system carries blood from capillaries through venules toward veins of increasing diameter. It is a lowpressure system composed of thin-walled vessels that are larger than their corresponding arteries and holds about two thirds of all the blood. Liquid blood is moved toward the heart by the pumping functions of muscles and the presence of valves in the extremities that help prevent reverse flow. Veins are little more than flexible tubes of smooth muscle lined by epithelial cells (Fig. 7B-1). A single-cellthick lining of vascular endothelial cells maintains a critical nonthrombogenic barrier between blood and tissue. Endothelial cells maintain a permeability barrier, regulate inflammation and immunity, modulate blood flow and vascular reactivity, and regulate coagulation and fibrinolysis.1 A thrombus is the pathologic production of clotting products inside blood vessels. The term emboli (from the Greek for “throwing in”) was coined by Rudolph Virchow in the late 1800s. He first depicted the pathology of venous thrombosis: “The detachment of larger or smaller fragments from the end of the softening thrombus which are carried

   TABLE 7B-1  Risk for Venous Thromboembolism   by Hospital Service Patient Group Medical General surgery Major gynecologic surgery Major urologic surgery Neurosurgery Stroke Major orthopaedic surgery Major trauma Spinal cord injury Critical care patients

Deep Venous Thrombosis Prevalence without Prophylaxis (%) 10-20 15-40 15-40 15-40 15-40 20-50 40-60 40-80 60-80 10-80

Data from Geerts WH, Pineo GF, Heit JA, et al: Prevention of venous ­thromboembolism: Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 126(3 Suppl):338-400, 2004.

along by the current of blood and driven into remote vessels.”2 For his contribution, the three primary influences on thrombosis formation were later called ­Virchow’s triad: endothelial damage, stasis, and hypercoagulability. Problems with the lining endothelium of vessels play a key role in thrombotic disease. Endothelial tissue carefully balances hemostasis, to prevent bleeding, and thrombolysis, to allow free flow of liquid blood. An intact endothelial cell lining normally inhibits thrombosis by mechanisms that block platelet adhesion and aggregation, interfere with the coagulation cascade, and actively lyse blood clots (Fig. 7B-2A). When the endothelial lining is disrupted by surgery or injury, collagen is exposed, and endothelial cells are triggered to make tissue factor, which activates the extrinsic clotting cascade. The negatively charged collagen of the exposed subendothelial extracellular matrix serves to activate the intrinsic pathway (through factor XII) of the clotting cascade and is the most important stimulant for platelet adhesion. Platelets normally circulate as smooth membrane–bound disks that are loaded with granules containing fibrinogen, adenosine diphosphate (ADP), adenosine triphosphate (ATP), calcium ions, clotting factors (V and VIII), the vasoconstrictor thromboxane A2 (TXA2), and other raw material for hemostasis. Platelets have three functions that play a critical role in clot development: (1) Platelet adhesion: Von Willebrand factor present in endothelium links platelets to ���������������������������������� exposed collagen �������������������������� (see Fig. 7B-2B). A defective gene for this protein is inherited in 1% of the population and leads to spontaneous mucosal bleeding, menorrhagia,

Complications Endothelium

371

Intima

Internal elastic lamina Media PGI2 External elastic lamina

Adventitia

A

NO

A

B

V TXA2 ADP Fibrinogen vWF

A

B Figure 7B-1   A, A single-cell-thick layer of endothelium lines blood vessels, balancing hemostasis and thrombolysis. B, Relative to arteries (A)�, veins (V) have larger diameters with thinner and less well-organized walls. (From Schoen FJ: Blood vessels. In Kumar MV, Abbas AK, Fausto N [eds]: Robbins and Cotran Pathologic Basis of Disease. Philadelphia, Saunders, 2005, p 512.)

e­ xcessive bleeding from wounds, and prolonged bleeding time despite normal platelet counts (von Willebrand’s disease). (2) Secretion: Adhesion activates the platelet to change shape and release the thrombogenic contents of its granules (see Fig. 7B-2C). Phospholipid complexes exposed by platelet shape change activate the clotting cascade, which is supported by the released calcium ions. (3) Aggregation: Release of ADP and TXA2 creates a ­catalytic reaction that forms an enlarging mass of platelets (see Fig. 7B-2D). Platelet aggregation is initially reversible, but the production of thrombin from the clotting cascade converts soluble fibrinogen to insoluble fibrin, which glues the growing platelet and surrounding red blood cells into a stable clot (Fig. 7B-3). Stasis of the venous system is another important component of Virchow’s triad. Normal leg activity promotes venous flow by increasing heart rate, increasing arterial inflow, contraction of intramuscular calf veins, and intermittent weight-bearing compression of the feet venous plexus. All act to push blood upward as healthy functioning valves help prevent blood from pooling more distally. ­Factors that slow blood flow or encourage turbulence favor the formation of thrombosis. Blood normally flows through vessels in a laminar pattern. Slower moving plasma is found near the endothelial lining, whereas platelets and red blood cells tend to move faster in the lumen center. Stagnant venous blood allows concentration of coagulation factors, production of a hypoxic environment for ­endothelial cells,

C

D

Figure 7B-2   A, Normal: Healthy endothelial cells protect from thrombus formation by secreting platelet-inhibitory mediators such as PGI2 (prostacyclin) and nitric oxide (NO). B, Adhesion: Exposed collagen of damaged endothelium is the most important stimulant for platelet adhesion. C, Secretion: Activated platelets release contents of their granules: fibrinogen, adenosine diphosphate (ADP), calcium ions, clotting factors (V and VIII), the vasoconstrictor thromboxane A�2 (TxA2). D, Aggregation: Occlusive platelet plug forms from additional recruited platelets. vWF, von Willebrand factor. (From Konkle BA, Schafer AI: Hemostasis, thrombosis, fibrinolysis, and cardiovascular disease. In Zipes DP, Libby P, Bonow RO, Braunwald E [eds]: Braunwald’s Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia, Saunders, 2005, p 2073.)

and increased platelet-endothelial contact. Tourniquets and extreme positions of extremities during surgery create venous stasis and put orthopaedic patients at higher risk for thromboembolic disease. Low-flow states can be a result of immobility (postoperative pain, cast, limb paralysis, stroke), increased blood viscosity (cancer, estrogens, polycythemia), and increased venous pressure (venous scarring from postthrombotic syndrome, varicose veins, heart ­failure).3 Hypercoagulability is the final component of Virchow’s triad. A brief review of both the coagulation and fibrinolytic physiology is helpful in understanding thrombophilic states and the chemoprophylaxis of thromboembolic disease. Normally, coagulation pathways are kept in check by inhibitors from endothelial cells or circulating anticoagulants. The coagulation cascade describes a catalytic system of plasma clotting proteins (Fig. 7B-4). The intrinsic pathway is named for its ability to clot in vitro with only a negatively charged surface and proteins intrinsic to plasma. This pathway’s function is measured using the activated partial thromboplastin time (aPTT). In vivo, exposed collagen activates the first protein of this pathway—­Hageman factor (XII). The extrinsic pathway is initiated with tissue trauma and the exposure of a lipoprotein called tissue factor. The prothrombin time (PT) is used to measure the function of vitamin K–dependent factors VII, IX, and X, and prothrombin (factor II). These two parts of the

372

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 7B-3   A, Scanning electron micrograph (SEM) of free platelets. B, SEM of platelet adhesion. C, SEM of platelet activation. D, Transmission electron micrograph of aggregating platelets.   1. Platelet before secretion.   2, 3. Platelets secreting contents of granules. 4. Collagen of endothelium. E, SEM of fibrin mesh encasing colorized red blood cells. (A to E courtesy of James G. White, MD, Regents’ Professor, Department of Laboratory Medicine and Pathology, University of Minnesota School of Medicine.)

A

D

B

C

c­ oagulation cascade share a common pathway beginning with factor X, which helps convert prothrombin to the main product of the clotting cascade: thrombin (factor IIa). This web of interdependent enzyme-mediated reactions occurs in the presence of calcium on the phospholipid surfaces of activated platelets and damaged endothelial cells to focus clot formation where needed. The biologic goal is to limit bleeding at sites of damage by rapidly stabilizing the initial platelet plug with insoluble fibrin. To balance the effectiveness of the clotting cascade and avoid thrombosing the entire vascular tree, systems of anticoagulants exist (Fig. 7B-5). Antithrombin III binds to heparin-like receptors on healthy endothelial cells and inhibits the serine proteases (factors IXa, Xa, XIIa). Endothelial cells also secrete tissue factor pathway inhibitor (TFPI) to rapidly inactivate Xa and VIIa. The vitamin K–dependent proteins C and S function as anticoagulants by inactivating factors Va and VIIIa. Finally, another group of proteins known as the fibrinolytic cascade produces the end product plasmin, which breaks down fibrin. ­ Endothelial cells ­produce tissue-type plasminogen ­activator (t-PA) that

E

c­ onverts circulating precursor plasminogen into active plasmin. Plasmin enzymatically cleaves fibrin into fibrin split products including D-dimer, which is a marker for thrombosis, PE, and ­disseminated intravascular ­coagulation (DIC). The d-dimer test has little clinical utility in diagnosing thromboembolic disease in traumatic or postsurgical patients because its levels are already elevated by the soft tissue damage. Thrombophilia is the predisposition to venous thromboembolism and is caused by inherited (primary) and acquired (secondary) factors, alone or in combination. Primary hypercoagulability is usually a result of genetic mutations causing abnormal amounts or dysfunction of clotting proteins. These prothrombotic states can result from a deficiency of a normal antithrombotic protein or an increase in levels of prothrombotic factors (Fig. 7B-6). More than half of all patients with clinical characteristics of thrombophilia are now diagnosed with an inherited disorder.4 The most common genetic prothrombotic conditions include elevated factor VIII, hyperhomocysteinemia, factor V Leiden (activated protein C resistance), prothrombin G20210A,

Complications

INTRINSIC PATHWAY

Kallikrein

Tissue Injury

HMWK collagen

Prekallikrein

Tissue Factor (Thromboplastin)

XIIa VII XIa

XI

IX

Tissue Factor VIIa

IXa VIII

Figure 7B-4   Coagulation cascade: Two enzymatic pathways that combine in a complex web of proteins that ends forming an insoluble fibrin polymer clot. Note the common link between the intrinsic and extrinsic pathways at the level of factor X activation. Factors in red boxes represent inactive molecules; activated factors are indicated with a lower case “a” and a green box. HMWK, high-molecular-weight kininogen; PL, phospholipid surface.   (From Mitchell RN: Hemodynamic disorders, thromboembolic disease, and shock. In Kumar MV, Abbas AK, Fausto N [eds]: Robbins and Cotran Pathologic Basis of Disease. Philadelphia, Saunders, 2005, p 128, Fig. 4-9.)

EXTRINSIC PATHWAY

XII (Hageman Factor)

VIIIa

Thrombin (IIa)

X Ca2+

373

Ca2+

Xa

V

Va Ca2+ XIII Thrombin (IIa) Ca2+ II IIa (Prothrombin) (Thrombin)

XIIIa

Phospholipid surface Ca2+ Active Inactive

Fibrinogen (I)

Fibrin Cross-linked Fibrin (Ia)

COMMON PATHWAY

and ­deficiencies of antithrombin III and proteins C and S.5 Although the prevalence of these disorders is usually low, the relative risk for any one ­ specific disorder can be high (Table 7B-2). Combinations of two or more inherited factors or combinations of genetic and acquired factors have even higher risks.6 Screening for ­prothrombotic defects has not yet been shown to be effective in ­selecting a ­strategy for FAVOR THROMBOSIS

thromboembolic ­ prophylaxis.7 Secondary acquired clinical factors are found in nearly every ­ orthopaedic patient and probably have a more significant role in their perioperative thrombosis (Fig. 7B-7). The incidence of venous thromboembolism (VTE) rises exponentially with age, with ­substantial increases seen after age 40 years (Fig. 7B-8).8 Cancer, pregnancy, oral contraceptives, lupus antiphospholipid

INHIBIT THROMBOSIS Inactivates thrombin and factors Xa and IXa

Extrinsic coagulation sequence

Proteolysis of factors Va and VIIIa Active protein C

Exposure of membrane-bound tissue factor Platelet adhesion: Held together by fibrinogen

Thrombin

Fibrinolytic cascade

Inhibit platelet aggregation

Inactivates tissue factors VIIa and Xa Antithrombin III

vWF

Protein C

Thrombin

PGI2, NO, and adenosine diphosphate t-PA

Endothelial effects Thrombomodulin Collagen

Heparin-like molecule

Thrombin receptor

Tissue factor pathway inhibitor

Endothelium

Figure 7B-5   The balance of prothrombotic and antithrombotic activities of the endothelium. Antithrombin III, protein C and S, tissue-type plasminogen activator (t-PA), and the fibrinolytic cascade all help to inhibit thrombosis.    (From Mitchell RN: Hemodynamic disorders, thromboembolic disease, and shock. In Kumar MV, Abbas AK, Fausto N [eds]: Robbins and Cotran Pathologic Basis of Disease. Philadelphia, Saunders, 2005, p 126, Fig. 4-7.)

374

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 7B-6   Common primary (genetic) hypercoagulable states: relative risks and prevalence.   (From Bankier A, Herold C, Minar E, Watzke H: Pulmonary embolism. In Albert RK, Spiro SG, Jett JR [eds]: Clinical Respiratory Medicine. St. Louis, Mosby, 2004, p 643, Fig. 53-1.) *The prevalence of PS deficiency is unknown. †Includes acquired defects. AT, antithrombin III; FV, factor V; PC, protein C, PS, protein S.

Genetic risk factors for venous thromboembolism

AT deficiency PC deficiency PS deficiency* FV Leiden Prothrombin G20210A Hyperhomocysteinemia† 6 5 4 3 Relative risk

antibodies, obesity, smoking, diabetes, and ­hypertension all increase the risk for VTE (Table 7B-3).4,6,8,9 The natural history of both thrombi and emboli rests in the balance of factors favoring clotting and those that encourage lysis. Once a thrombus forms, it can dissolve by the fibrinolytic system (dissolution), remain stationary in a vein and incorporate into the vein wall (organization and recanalization), continue to grow (propagation), or break free to travel downstream to lodge in the pulmonary vessels (embolization).3 Where a clot forms can have significant clinical implications. Although thrombosis can occur in any vessel, more than 90% of thromboses form in the veins of the lower extremity. Distal clots (below the popliteal space) that occur in the smaller veins of the calf pose little clinical threat because most dissolve spontaneously. Larger ­diameter veins in the proximal thigh are associated with thrombus and rarely completely lyse. Obstruction of larger proximal veins leads to greater risk for postphlebitic syndromes. A larger burden of products of coagulation also increase the risk for clinically significant embolism. Half of patients with known proximal clots have asymptomatic PE by ­ventilationperfusion scan. Seventy percent of patients with ventilationperfusion scan–proven PE have proximal VTE.10 When a

2

1 0

1

2

3 4 5 6 7 Prevalence (%)

thrombus embolizes to the lung, the size of the clot becomes a critical issue. Massive saddle emboli block all cardiopulmonary function and cause immediate death.

EVALUATION: CLINICAL PRESENTATION—HISTORY AND PHYSICAL EXAMINATION Appreciating the silent nature of most thromboembolic disease and the potential fatal implications stimulates a desire to develop skills for early diagnosis and treatment of this common problem. Clinical diagnosis most ­heavily relies on the history of the primary and secondary risk ­factors discussed earlier because the physical examination for venous thrombosis is notoriously unreliable. Symptoms of VTE arise from venous obstruction, inflammation of vessel walls, and embolization to the lungs. Although VTE can cause leg pain, tenderness, ­ swelling, or a palpable cord, most cases are asymptomatic. The classic physical examination tests include Homans’ sign (calf pain with foot dorsiflexion when the knee is flexed) and Moses’ sign (pain with calf compression against

   TABLE 7B-2  Primary Clotting Disorders Most Common Hypercoagulable Genetic (Primary) Disorders

Relative Risk

Prevalence (%)

Venous Thromboembolism Patients (%)

20 6.5 5

0.02 0.3 0.003

2 4 2

80 7 5 3 3

0.02 4.8 11 6 2.7

20 25 10 7

Decreased Antithrombotic Factors

Antithrombin III deficiency Protein C deficiency Protein S deficiency Increased Prothrombotic Factors

Factor V Leiden—homozygous (C resistance) Factor V Leiden—heterozygous (C resistance) Elevated factor VIII Hyperhomocysteinemia Prothrombin G20210A (increased factor II)

Data from Ginsberg MA: Venous thromboembolism. In Hoffman R, Benz EJ, Shattil SJ, et al (eds): Hematology: Basic Principles and Practice, 4th ed. Philadelphia, Elsevier Churchill Livingstone, 2005, pp 2225-2236; and Perry SL, Ortel TL: Clinical and laboratory evaluation of thrombophilia. Clin Chest Med 24(1):153-170, 2003.

Complications

The Secondary Hypercoagulable States Abnormalities of blood flow

Abnormalities of blood composition

Abnormalities of vessel wall

Hyperviscosity Venous stasis Obesity

375

Figure 7B-7   Secondary hypercoagulable states based on Virchow’s triad of thrombogenesis: stasis—abnormalities in blood flow; hypercoagulability—abnormalities in blood composition; and endothelial damage— abnormalities of the vessel wall.   (From Schafer AI: Thrombotic disorders: Hypercoagulable states. In Goldman L, Ausiello D [eds]: Cecil Textbook of Medicine. Philadelphia, Saunders, 2004, pp 1082-1087.)

Postoperative state Trauma Pregnancy Myeloproliferative disorders Cancer Oral contraceptives Nephrotic syndrome Paroxysmal nocturnal hemoglobinuria Hyperlipidemia Heparin-associated thrombosis Diabetes mellitus Thrombotic thrombocytopenic purpura Antiphospholipid syndrome Vasculitis

Figure 7B-8   Exponential increase in venous thromboembolism with age. (Data from Anderson FA Jr, Wheeler HB, Goldberg RJ, et al: A population-based perspective of the hospital incidence and casefatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study. Arch Intern Med 151[5]:933-938, 1991.)

Thrmoboembolic Disease by Age 6

Annual Incidence per 1000

5

4

3

2

1

0 0

10

20

30

40 50 Age

60

70

80

376

DeLee & Drez’s Orthopaedic Sports Medicine

   TABLE 7B-3  Relative Venous Thromboembolism Risks of Common Conditions Condition Age > 70 yr Cancer Pregnancy Lupus antiphospholipid antibodies Oral contraceptive pills Obesity (body mass index ≥ 29) Smoking Diabetes Hypertension

Relative Risk 10-15× 6× 5× 4× 3-4× 3× 3× 2× 2×

Data from references 4, 6, 8, and 9.

the tibia).11 Fewer than half of patients with these classic signs of VTE have objective evidence.12 Only half of patients with venographically proven VTE have the ­classic clinical signs.13 PTS, also called postphlebitic syndrome, is the result of residual venous injury and incompetent valves. Similar symptoms appearing without documented VTE are known as chronic venous insufficiency. PTS presents as chronic postural dependent leg swelling, pain, hyperpigmented skin, dermatitis, and when severe, skin ulcers near the ankles, which are a potential source of infection.14 Twenty to 50% of patients with objectively proven VTE develop PTS. Long-term treatment (2-5 years) with elastic compression hose with a pressure of 30 to 40 mm Hg at the ankle can reduce this rate by half.15,16 The risk for PTS after DVT from orthopaedic surgery is controversial and poorly studied. Most orthopaedic patients with PE are asymptomatic. Yet, the first clinical sign of VTE can be a symptomatic or fatal PE. PE can present with acute-onset dyspnea (shortness of breath), pleuritic chest pain, hemoptysis, or circulatory collapse. PE is the most common cause of death and the most frequent cause of readmission after major orthopaedic surgery.14 Embolic symptoms are directly related to the extent of pulmonary vascular blockage. Large ­occlusions cause greater increase in pulmonary arterial resistance, which can lead to right heart failure and hypoxemia. Patients with significant cardiopulmonary disease can have severe consequences with smaller embolic burdens. Healthy cardiopulmonary function can allow substantial embolic load. If less than 60% of the pulmonary ­circulation is obstructed, a healthy patient may remain asymptomatic.17 In symptomatic patients diagnosed with pulmonary angiograms, the most common symptoms are chest pain (often pleuritic) and sudden onset of shortness of breath (dyspnea). Examination findings encountered in more than half of patients are tachypnea (>20 breaths/­ minute) and crackles (Table 7B-4).17

TESTING FOR VENOUS THROMBOEMBOLISM The nonspecific nature of the signs and symptoms of VTE demand a high clinical suspicion in patients at high risk. There is no ideal objective test for thromboembolic disease,

   TABLE 7B-4  Incidence of Symptoms and Signs   in ­Angiographically Proven Pulmonary Embolism Presenting Feature

Incidence (%)

Symptom

Chest pain* Pleuritic chest pain* Dyspnea* Cough* Hemoptysis Syncope

88 74 84 53 30 13

Sign

Tachypnea* Crackles* Rales Tachycardia (>100 beats/min) Fever (>37.8° C/100° F) Gallop Phlebitis Edema

92 58 48 44 43 34 32 24

*Finding encountered more than half the time. Modified from Bankier A, Herold C, Minar E, Watzke AB: Pulmonary ­embolism. In Albert RK, Spiro SG, Jett JR [eds]: Clinical Respiratory ­Medicine. St. Louis, Mosby, 2004, p 643, Table 53.2.

but contrast venography, duplex compression ­ultrasound, spiral computed tomography (CT) venography, CT pulmonary angiography, and d-dimer may be useful in various populations. Contrast venography continues to be the research gold standard for the diagnosis of DVT and is usually required by the U.S. Food and Drug Administration (FDA) for studies of VTE. It continues to be the most predictable test for the diagnosis of distal thrombosis (below the popliteal fossa) and can also demonstrate iliac thrombosis. However, venography is almost never used on a clinical basis because it is invasive and requires renal toxic radiographic dye, and because of the questionable clinical significance of distal thrombus. The most reliable criterion for the diagnosis of venous thrombosis is an intraluminal filling defect evident in two or more views (Fig. 7B-9).18 Duplex ultrasound serves as the most practical diagnostic tool for most patients because it is inexpensive, noninvasive, and easily repeatable at the patient’s bedside. The inability to visualize compressibility of a vein with real-time B-mode Doppler ultrasound is more than 95% sensitive and specific for the detection of proximal DVT.12 Diagnostic accuracy is operator dependent and may be limited in after-hour examinations. Evaluation of vessels above the inguinal ligament is limited. For most outpatient surgical procedures, duplex ultrasound serves as the best first-line test in a stable patient (Fig. 7B-10).19 CT and magnetic resonance (MR) venography demonstrate good sensitivity and specificity for the diagnosis of clinically significant DVT. Both have the advantage of better visualization of pelvic and inferior vena cava ­vessels than sonography, and both do not require compression, which is helpful in cases with nearby wounds, burns, or the presence of plaster casts. CT venography may not be appropriate for patients with renal insufficiency or allergies to radiographic dyes. MR venography is limited

Complications

377

Figure 7B-9   The gold standard diagnosis for deep venous thrombosis. Intraluminal filling defects seen on two or more views of venogram. A, Below knee. B and C, At knee. D, Proximal, above popliteal area. (From Jackson JE, Hemingway AP: Principles, techniques and complications of angiography. In Grainger RG [ed]: Grainger & Allison’s Diagnostic Radiology: A Textbook of Medical Imaging, 4th ed. Philadelphia, Churchill Livingstone, 2001, p 161.)

B

A

C

D

by its decreased availability, higher cost, slower image acquisition times, and image compromise near metal implants.20 When chest pain, dyspnea, or cardiovascular collapse raises the clinical suspicion of PE, the work-up includes chest radiograph, electrocardiogram (ECG), and arterial blood gas. The chest plain film is usually abnormal with subtle and nonspecific findings but helps to exclude other diagnoses. An ECG is also frequently abnormal with

A

B

­ on-specific findings. The most common ECG findn ings in PE are sinus tachycardia, T-wave inversion, and ST abnormalities. Arterial blood gas analysis is rarely of help—except as an indicator of the size of the embolism when massive PE demonstrates extreme hypoxemia. Most frequently, PEs have no hypoxemia but show hypocapnia (low CO2) from hyperventilation.21 High-sensitivity enzyme-linked immunosorbent assay (ELISA) d-dimer levels are not helpful in the ­orthopaedic

C

Figure 7B-10   Doppler ultrasound for proximal deep venous thrombosis in femoral vein. A, Longitudinal Doppler display shows the presence of flow (blue) in the more superficial vein over an occlusive thrombus (dark gray). B, Transverse view without compression shows open superficial vein (black oval) and thrombosed deeper vein (gray circle). C, Transverse view with compression shows compressible superficial vein and noncompressible thrombosed deeper vein. (Courtesy of Austin Radiological Association and Seton Family of Hospitals.)

378

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 7B-11   Helical computed tomographic pulmonary angiography. A, Patient with symptoms of syncope and hypoxia. Large clots (arrows) are present at the bifurcation of the main pulmonary artery and extend into the left and right pulmonary arteries. B, Normal scan in a different patient at same level. C, Patient with embolisms (arrows) present in smaller pulmonary arteries. D, Normal scan in a different patient at same level. (Courtesy of Austin Radiological Association and Seton Family of Hospitals.)

A

B

setting because the trauma from fracture and surgery can cause prolonged elevations with or without thromboembolic disease. A negative d-dimer excludes DVT or PE in patients with low probability of thromboembolic ­disease.10 The first-line study to provide objective diagnosis of acute PE is CT pulmonary angiography (Fig. 7B-11A and B). Multidetector spiral (helical) CT scanners offer shorter scan times, better resolution with less radiation and are available in most hospitals in the United States. PE is confirmed when spiral CT shows an intravascular filling defect in a pulmonary artery that occludes all or part of a vessel and is often associated with increased diameter of the affected vessel (see Fig. 7B-11C and D). ­ Sensitivities and specificities of CT pulmonary angiography over 95% and improved imaging of other intrathoracic pathology have made ventilationperfusion scintigraphy scans a second-line diagnostic study. Although a negative perfusion scan excludes significant PE, ventilation-perfusion scans are best reserved for patients with contraindications to radiographic dye (Fig. 7B-12). Pulmonary angiography may still be the gold standard for the diagnosis of isolated subsegmental PE; however, the clinical significance of these smaller emboli is controversial. Although MR imaging has great potential for PE diagnosis, future improvements in the resolution, acquisition times, and costs will be required before MR angiography can be recommended in the community.20

C

D

THROMBOEMBOLIC TREATMENT OPTIONS Prophylaxis The need for prophylaxis should be based on the risks for developing venous thrombosis and PE balanced by the dangers of prophylaxis. Location, extent, and duration of surgery can influence size, location, and frequency of thrombosis (Table 7B-5). Knee arthroplasty has the higher rate of clots, but most are distal, small, and asymptomatic. Hip arthroplasty has a higher rate of more significant larger proximal clots and PE. Hip fracture surgery has a higher rate of proximal clots and fatal pulmonary embolism, but older patients may also be at higher risk for major bleeding complications. Outpatient sports medicine practice includes less invasive procedures on younger and healthier patients, who are often motivated to return to function at an accelerated pace. Despite a population with low thromboembolic risk, fatal pulmonary embolisms can occur and can command considerable attention.22-25 Because the clinical diagnosis of VTE is unreliable, judging the effectiveness of various prophylactic strategies requires the use of an objective end point. The obvious objective clinical outcome to prevent is fatal PE. Fortunately, the incidence of fatal PE is so low it is impractical to measure. In place of symptomatic or fatal PE, the most common end points have been both symptomatic and asymptomatic

Complications

Figure 7B-12   High-probability pulmonary ventilation-perfusion scan. Ventilation scan (above) demonstrates full lung fields. Perfusion scan (below) shows multiple areas lacking tracer. (Courtesy of Austin Radiological Association and Seton Family of Hospitals.)

Ventilation

L Post R

RPO

R Ant L

LAO

379

RLAT

RAO

LLAT

LPO

Perfusion

L Post R

L RPO R

R Ant L

R LAO L

RLAT

R RAO L

LLAT

L LPO R

­ istal and proximal DVT by venogram and proximal DVT d by Doppler ultrasound (total DVT). Total DVT is a highfrequency event (see Table 7B-5) that is practical to objectively measure but may not be as clinically relevant. Bleeding is a predictable complication of surgery and the most common complication of anticoagulants. Higher rates of bleeding are seen when more effective prophylactic drugs are used and when drugs with more rapid onset of action are used closer to the time of surgery.26 Bleeding complications can be classified as minor and major, with major events including fatal or life-threatening bleeds (e.g., intracranial or retroperitoneal) or bleeding with a defined drop in hemoglobin, leading to transfusion of a

specified number of units of blood, or to hospitalization or return to surgery.27 In reference to increased bleeding with prophylaxis, caution is warranted regarding statements of “no statistical significance.” It takes a much smaller study group to statistically demonstrate a decrease in a high­frequency event such as total DVT (about 50% of total knee arthroplasty [TKA] patients) than to show an increase of a low-frequency event such as bleeding (3%-5% of TKA patients). To help keep the risks for bleeding in perspective, a review of vascular surgery literature is helpful. In a randomized, prospective study on surgery directly on large arteries comparing low-­molecular-weight heparin (LMWH) and low-dose unfractionated heparin (LDUH),

   TABLE 7B-5  Rates of Venous Thromboembolism Complications in Various Orthopaedic Procedures When No   Prophylaxis Is Used Procedure THA TKA Hip fracture Cast ≥ 5 wk Arthroscopy

Total DVT (%)

Proximal DVT %)

Total PE (%)

Fatal PE (%)

42–57 41-85 46-60 19 4-18

18-36 5-22 23-30 5 0-5

1-28 1.5-10 3-11 1

0.1-2 0.1-1.7 2.5-7.5 0

DVT, deep venous thrombosis; PE, pulmonary embolism; THA, total hip arthroplasty; TKA, total knee arthroplasty. Modified after data from Tables 7, 8, and 11 in Geerts WH, Pineo GF, Heit JA, et al: Prevention of venous thromboembolism: Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 126(3 Suppl):338-400, 2004; and from Section 3.7 in Lassen MR, Borris LC, et al: Use of the low-molecularweight heparin reviparin to prevent deep-vein thrombosis after leg injury requiring immobilization. N Engl J Med 347(10):726-730, 2002.

380

DeLee & Drez’s Orthopaedic Sports Medicine

both groups had a 2% rate of bleeding.28 Bleeding complications associated with neuraxial blockade can be devastating, are well described in the ­literature, and demand special mention. Paraplegia can result from spinal compression because of perispinal hematomas. Recommendations include avoiding neuraxial analgesia in patients with bleeding disorders and in those taking antithrombotic drugs. To avoid bleeding complications, manufacturers often recommended waiting 8 to 12 hours after wound closure before chemoprophylaxis.29 Although individual surgeons must ultimately choose the prophylaxis for each patient, sorting out the extensive literature can be challenging. The literature on thromboprophylaxis is strongly supported by industry and highly weighted with hip and knee arthroplasty patients; however, these studies include some of the highest level of ­ evidence available in orthopaedic surgery. An understanding of the existing consensus guidelines helps provide a starting point to make educated decisions on this controversial topic. The American College of Chest Physicians (ACCP) publishes the most frequently cited evidence-based clinical practice guideline on DVT prophylaxis. Internal medicine specialists and orthopaedic surgeons periodically summarize the existing knowledge base during the ACCP Conference on Antithrombotic and Thrombolytic Therapy.14 ­ Procedure-specific recommendations are made and graded. Numbers are used to rank the strength of the recommendations. Grade 1 is used when the expert panel is very certain that benefits clearly outweigh the risks. The quality of the studies supporting the recommendations is ranked with letters. Randomized controlled trials with consistent results provide unbiased grade A ranking. Studies that are overwhelmingly compelling but less rigorous receive a C+ rating (Table 7B-6).30 ACCP recommendations for prophylaxis are based on studies using DVT incidence as a proxy for PE risk. Because DVT incidence is high (40% to 60%), the statistical level of evidence of DVT prophylaxis is high (ACCP Grade IA). Guidelines focusing on high-incidence event (DVT) may not be powered for less frequent ­ complications (bleeding, PE). Using these guidelines may increase costs and bleeding complications with no measurable decrease in symptomatic PEs. Although chemoprophylaxis clearly decreases the incidence of DVT, its reduction does not appear to have a significant effect on the PE rate. Another evidence-based guideline has been produced by the ­American ­Academy of Orthopaedic ­ Surgeons (AAOS). AAOS guidelines question the use of DVT incidence as a proxy for PE and focus on the low incidence of life-­threatening ­pulmonary embolic events (0.15% to 0.93%) and risk of major bleeding episodes after arthroplasty (1% to 3%). Orthopaedic surgeons are acutely aware of how bleeding complications threaten patient outcome. Hematomas may cause pain, slow rehabilitation, stiffness, unplanned return to the operating room for evacuation, and prolonged wound drainage, risking deep infections and possibly removal of implants. Although the most clearly significant outcomes of interest to surgeons and patients are fatal PEs and ­bleeding ­complications, none of the studies was deemed to be of “good” quality regarding the outcomes of interest (no level I randomized controlled studies).

   TABLE 7B-6  American College of Chest Physicians’ ­Evidence-Based Ranking Classification Classification

Balance of Benefits vs. Risk, Harms, and Cost

Recommendation Strength

1. Recommend 2. Suggest

Clear Unclear

Quality or Type of Evidence

A C+ B C

RCTs: consistent Overwhelmingly compelling grade C data RCTs: inconsistent or flawed methods Observational studies Generalizations from randomized trials of other group

RCT, randomized controlled trial. Modified from Guyatt G, Schunemann HJ, Cook D, et al: Applying the grades of recommendation for antithrombotic and thrombolytic therapy: Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 126(3 Suppl):179-187, 2004.

Consensus opinions only serve as a convenient starting point for surgeons and hospitals to make daily care decisions based on current evidence. A specific patient’s risk may vary from the averaged results cited in multiple-study meta-analyses. The prudent surgeon’s “review of systems” will include specific questions regarding the presence of prior history or family history of bleeding disorders, thrombosis, or embolism. Mechanical methods of prophylaxis include aggressive range of motion and early weight-bearing activity, graduated compression stockings, venous foot pumps, and ­intermittent pneumatic compression hose.31 Promoting blood flow should help prevent thrombosis by diluting activated coagulation factors. The importance of active flowing blood is demonstrated by higher rates of clots in patients with spinal cord injury, stroke, and other causes of immobility. Decreased inflow of blood, measured as decreased ankle-brachial index, is an independent risk factor for DVT.32 Mechanical methods appeal because of their low bleeding potential. Fewer studies that are difficult to blind and poor compliance with routine use frustrate strong recommendations. Mechanical devices are best reserved for use in patients at high risk for bleeding and as adjuncts for other prophylactic techniques. When bleeding risks allow, chemoprophylaxis is recommended for all major orthopaedic surgery, including pelvic fractures, multiple trauma, hip fractures, and joint replacement of the hip and knee. For selected high-risk sports procedures and for patients with multiple risk factors, chemoprophylaxis can minimize VTE complications. The most common medicines used in the prophylaxis and ­ treatment of VTE are aspirin, unfractionated heparin (UFH), LMWH, vitamin K antagonist (VKA), and fondaparinux (Table 7B-7). Aspirin is the salicylic ester of acetic acid and blocks platelet aggregation. Aspirin’s mechanism of action is to irreversibly inactivate platelet cyclooxygenase (COX) ­activity.

Complications

381

   TABLE 7B-7  Common Drugs for Venous Thromboembolism Prophylaxis and Treatment Drug

Mechanism of Action

Complications

Notes

Reversal

Aspirin

Irreversibly blocks platelet COX-1 production of TXA2

Least effective chemical agent

None

Heparin

Binds AT to inactivate Xa, II (thrombin), and IXa, XIa, XIIa Binds AT to inactivate Xa, IIa (thrombin) Binds AT to selectively inactivate only Xa Blocks γ-carboxylation of II, VII, IX, X

Bleeding NSAID, GI issues, rare allergies Bleeding, HIT, ­osteoporosis Bleeding

Must monitor aPTT or anti-Xa activity Monitor anti-Xa activity if BMI > 50, CrCl < 30 Avoid in thin (<50 kg) or elderly (>75 yr); CRI: CrCl < 30 Oral, long-term therapy Multiple reactions: diet, drugs, disease

Protamine: 1 mg IV per 100 U UFH Protamine: 1 mg IV per 1 mg LMWH (enoxaparin) No antidote

LMWH Fondaparinux VKA Coumadin

Bleeding Bleeding, skin necrosis Fetal warfarin ­ syndrome

Vitamin K1 (phytonadione): Fresh frozen plasma, 8-10 mL/kg

aPTT, activated partial thromboplastin time; AT, antithrombin; BMI, body mass index; CrCl, creatinine clearance; COX-1, cyclooxygenase-1; CRI, chronic renal ­insufficiency; GI, gastrointestinal; HIT, heparin-induced thrombocytopenia; LMWH, low-molecular-weight heparin; NSAID, nonsteroidal anti-inflammatory drug; TXA2, ­thromboxane A2; UFH, unfractionated heparin; VKA, vitamin K antagonist. Data from references 29, 33, 34, and 35.

Platelet COX-1 produces thromboxane (TXA2), which causes platelet aggregation and vasoconstriction. Aspirin’s effect is 100-fold more potent in inhibiting COX-1 (platelet functions) than COX-2 (hypoalgesia and inflammation). The permanent platelet-inhibitory effects of aspirin last for the 10-day life span of the platelet. Clinically, these platelet-inhibitory effects can be measured with a bleeding time.33 Aspirin has a long tradition in thromboembolic prophylaxis and has been shown to decrease VTE events. Although its efficacy on the arterial side is well established, its effect on the venous system is so low that its use has been discouraged. ACCP issued grade IA recommendations against using aspirin alone for the prevention of VTE in any patient group.26 UFH is a mixture of long- and short-chain glycosaminoglycan molecules that inactivates factors IIa (thrombin), Xa, IXa, XIa, and XIIa. One small region of the UFH ­molecule is a pentasaccharide sequence that binds the plasma cofactor antithrombin (AT), and this heparin-AT complex specifically inhibits factor Xa. Longer heparin chains bind to inactivate thrombin (IIa), but also nonspecifically bind to various plasma proteins, endothelial cells, macrophages, and platelets, which cause its two most important limitations. First, nonspecific binding of heparin leads to less predictable pharmacokinetics and the requirement of careful monitoring to avoid bleeding. Second, an antigen is formed when the long chain of heparin binds to platelet factor 4 (PF4). Repeated or long-term exposure (>1 week) can result in an immune-mediated platelet activation known as heparin-induced thrombocytopenia (HIT) in 1% to 5% of patients. Long-term therapy with heparin has a negative effect on bone metabolism by suppressing osteoblast formation and activating osteoclasts, leading to heparininduced osteoporosis. Historically, heparin was monitored using an aPTT range (often 1.5 to 2.5 times the control). A more precise but more expensive technique is to monitor the anti–­factor Xa units directly (0.3 to 0.7 IU/mL). The anticoagulant heparin-AT effects can be reversed with protamine sulfate (1 mg IV per 100 U UFH).34 The enzymatic treatment of UFH molecules creates lowmolecular-weight fragments (LMWH) that maintain the specific AT-binding pentasaccharide sequence but have

less nonspecific binding to thrombin (IIa), proteins, and cells (Fig. 7B-13). This provides a more predictable dose response, better bioavailability, longer half-life, lower incidence of HIT (<1%), and less osteoporosis. Superior pharmacokinetic properties allow standard dosing without monitoring except in cases of obesity (body mass index > 50) and renal failure (creatinine clearance < 30). Monitoring of anti–factor Xa activity is commonly performed at peak effect 4 hours after subcutaneous injection with a therapeutic range of 0.6 to 1.0 IU/mL. Protamine sulfate is a recommended reversal agent at doses of 1 mg of protamine sulfate for every 1 mg of enoxaparin.34 Fondaparinux is a synthetic creation of the AT-specific pentasaccharide sequence in heparin. It selectively inhibits factor Xa through AT and has no effect on thrombin (IIa). It has good bioavailability and predictable pharmacokinetics and no binding to PF4. Routine monitoring is not ­recommended but can be accomplished with direct measure of anti–factor Xa activity. Because of its high risk for accumulation and bleeding, its use is contraindicated in patients who are thin (<50kg), elderly, or have renal insufficiency (creatinine clearance <30). Fondaparinux is an effective anticoagulant for which no convenient antidote exists.35 Warfarin is the most common VKA in use. Warfarin blocks the γ-carboxylation of glutamate residues of vitamin K–dependent proteins II, VII, IX, and X, inhibiting their clotting effect. VKA also blocks carboxylation of the anticoagulant proteins C and S. VKA advantages of oral ­dosing, rapid bioavailability, and long half-life are offset by its narrow therapeutic window, variable dose response among patients, and many interactions with drugs, diet, and diseases. Less predictable pharmacodynamics require monitoring, which is accomplished with a prothrombin time (PT) with the usual goal international normalized ratio (INR) of 2.0 to 3.0.29 VKA is contraindicated in pregnant patients because of the classic teratogenic effects known as fetal warfarin syndrome (FWS). Characteristics of FWS include stippled epiphyses and vertebrae (chondrodysplasia punctata) and hypoplasia of the midface and nose (Fig. 7B-14).36 The most common side effect after bleeding is warfarin-induced skin necrosis. This ­uncommon ­complication presents on days 3 to 8 of

382

DeLee & Drez’s Orthopaedic Sports Medicine

AT Xa AT Xa

PS

AT

5

LMWH (15)

Xa

IIa UFH

PF4 IIa

IXa

XIa

XIIa

(45)

40 50 60 70 80 90 100 30 Ranges of Length in Saccharide Molecules and Clotting Cascade Proteins Affected Figure 7B-13   Unfractionated heparin (UFH) includes a large range of long molecules (averaging 45 monosaccharides) with the specific antithrombin (AT) binding sequence (pentagon) and longer portions that nonspecifically bind thrombin (IIa) and other plasma proteins, endothelial cells, and platelets (PF4). Low-molecular-weight heparin (LMWH) includes the specific AT binding pentasaccharide sequence (PS) attached to shorter sugar chains (averaging 15 monosaccharides) that can also bind thrombin (IIa). Fondaparinux is a synthetic copy of the specific PS that binds AT to inactivate only Xa. 0

10

20

t­ herapy and is caused by extensive thrombosis of the veins of subcutaneous fat (Fig. 7B-15).37 Warfarin reversal can be ­accomplished by stopping the medicine (requires about 5 days), giving vitamin K (requires about 24 hours), or giving fresh frozen plasma (immediate).

ACCP guidelines give level IA recommendations for chemoprophylaxis in major orthopaedic surgery. Hip arthroplasty recommendations are for at least 10 days of LMWH, VKA, or fondaparinux. Recommendations are made against the use of aspirin, dextran, low-dose heparin,

Figure 7B-14   Fetal warfarin syndrome includes stippled epiphysis (chondrodysplasia punctata) and hypoplasia of the midface and nose. (From Hou JW: Fetal warfarin syndrome. Chang Gung Med J 27[9]:691-695, 2004.)

Figure 7B-15   Warfarin skin necrosis. (From Stewart AJ, Penman ID, Cook MK, Ludlam CA: Warfarin-induced skin necrosis. Postgrad Med 75[882]:233-235, 1999.)

Complications

or mechanical methods as the only prophylactic measures. Knee arthroplasty recommendations include 10-day use of LMWH, VKA, or fondaparinux. Intermittent pneumatic compression devices (not isolated venous foot pumps) are an acceptable alternative. Hip fracture surgery is associated with the highest risk for fatal PE, which is related to the proximal location of surgery, frequent comorbidities, older age, delays between fracture and admission, and delays between admission and surgery. Fewer studies have been completed in this population. Despite being a population that may need special consideration for bleeding complications, recommendations are strong for routine prophylaxis with fondaparinux, LMWH, or VKA. If bleeding risks are high, mechanical prophylaxis is recommended.26 Extended prophylaxis up to 28 to 35 days is recommended for highrisk patients with hip fracture or total hip replacement. High-risk patients include those with a prior history of VTE, delayed mobility, advanced age, obesity, or cancer.26 AAOS guidelines encourage surgeons to assess patients for elevated risk of bleeding (history of bleeding disorder, gastrointestinal bleed, or hemorrhagic stroke) and elevated risk of PE (history of hypercoagulable disorder or ­ previous PE). No objective laboratory measurement exists for the reliable stratification of the risks of PE, and serologic tests to screen bleeding risks are not routinely recommended. AAOS guidelines rely on each patient’s recollection of prior medical details (history), the quality of the surgeon’s physical examination, and the expert clinical judgment used for each case. Mechanical prophylaxis is always recommended. Patients with high risk of embolism are discouraged from using aspirin alone unless they have higher bleeding risks. Recommendations are to avoid the ­ fast-onset chemoprophylaxis agents (LMWH and synthetic pentasaccharides) in patients with high risk of bleeding. For the most common patient category (those with standard risks of both bleeding and embolism) recommendations include aspirin, LMWH, synthetic pentasaccharides, and warfarin (Table 7B-8).

383

Specific recommendations by surgical procedure for most sports medicine procedures are limited. Despite arthroscopy or arthroscopically assisted cases being the most commonly performed orthopaedic surgery in the world, with more than 3 million cases done annually, few studies exist on the need for prophylaxis.38 Arthroscopy and arthroscopically assisted cases are a heterogeneous group that range from simple partial meniscectomies that can be accomplished in less than 30 minutes to multiple ligament reconstructions requiring several hours of surgery. Increased time and extent of the procedure, manipulation of the knee for therapeutic cases, and use and duration of use of a tourniquet can alter the risk profile in any specific patient. Transesophageal ultrasound has been used to demonstrate embolic debris in the right atrium during hip arthroplasty and after tourniquet release in operations of the knee.39 The volume of material peaks within 1 minute of release and varies with the length of tourniquet time and procedure.40 Total knee operations with violation of medullary canal have a higher volume of embolic debris than anterior cruciate ligament reconstruction.41 A prospective Doppler study of patients younger than 40 years undergoing anterior ligament reconstruction showed one DVT in 67 patients.42 Specific literature on thromboembolic complications of high tibial osteotomy, unicondylar knees, posterior cruciate reconstruction, and multiple ligament injuries is still forthcoming. Rates of VTE after arthroscopy are generally quite low. A review of the prospective venographic studies showed a rate of total DVT between 4% and 18% and proximal DVT rates between 0% and 5%. Prospective duplex studies had proximal rates up to 2%.26 A meta-analysis of six level I or II studies including only patients without ­prophylaxis undergoing unilateral knee arthroscopy without ligament work or open procedures demonstrated total DVT of 10% and proximal DVT rates of 2%.43 Two randomized clinical trials using LMWH in knee arthroscopy patients showed a decrease in DVT and no major bleeding complications.

   TABLE 7B-8  AAOS Consensus Guideline for Prevention of Pulmonary Embolisms After Arthroplasty Risk of PE*

Risk of Bleed

Chemoprophylaxis Options

Quality of Evidence

Standard

Standard

Fair (IIIB)

High

Standard

Standard High

High† High†

ASA, LMWH, synthetic pentasaccharide, VKA LMWH, synthetic pentasaccharide, VKA No chemoprophylaxis, ASA, VKA No chemoprophylaxis, ASA, VKA

Fair (IIIB) Poor (IIIC) Poor (IIIC)

ASA: Aspirin for 6 wk 325 mg BID (81 mg per day if GI symptoms) LMWH: Low molecular weight heparin, start 12–24 hours postoperatively; use for 7–12 days Enoxaparin, 30 mg SC BID Dalteparin, 5000 IU SC QD Ardeparin, 50 anti-Xa units/kg SC BID Synthetic pentasaccharide fondaparinux, start 12-24 hr postoperatively; use for 7-12 days 2.5 mg SC QD VKA: vitamin K antagonist - warfarin Start night before or day of surgery and continue for 2-6 wk Dose daily and monitor to INR goal of ≤2.0 *AAOS guidelines r���������������������������������������������������������������������� ecommend considering mechanical prophylaxis for all patients (IIIB). †Vena cava filter should be considered: (very low evidence level VC) if contraindications to chemoprophylaxis and high risk for PE when high risk of bleeding in patient with symptomatic PE. Data from AAOS����������������������������������������������������������������������������������� Clinical Guideline on Prevention of Symptomatic Pulmonary Embolism, available at http://www.aaos.org/Research/guidelines/PE_guideline.pdf.

384

DeLee & Drez’s Orthopaedic Sports Medicine

   TABLE 7B-9  Treatment of Venous Thromboembolism Drug

Dose

Duration

Monitoring

Intravenous unfractionated heparin

≥5 days

Low-molecular-weight heparin

80 U/kg IV bolus 18 U/kg/hr IV 5000 U IV bolus 17,500 U SC bid 1 mg/kg bid

Vitamin K antagonist

5 mg PO q.a.m. ������

≥3 mo

aPTT: 1.8-2.5× control; or anti-Xa activity: 0.3-0.7 U/mL aPTT: 1.5-2.5× control q.a.m. 6 hr after morning dose None unless renal insufficiency or pregnancy Anti-Xa activity: 0.6-1.0 IU/mL q.a.m. 4 hr after morning dose INR: 2-3

Subcutaneous unfractionated heparin

≥5 days ≥5 days

aPTT, activated partial thromboplastin time. Data from Buller HR, Agnelli G, Hull RD, et al: Antithrombotic therapy for venous thromboembolic disease: Seventh ACCP Conference on Antithrombotic and ­Thrombolytic Therapy. Chest 126(3 Suppl):401-428, 2004.

Because of very low rates of total VTE after arthroscopic cases, however, ACCP suggests that no routine prophylaxis be used other than early mobilization. For patients with higher than the usual risks who undergo long or complicated procedures, prophylaxis with LMWH is safe and effective. In sports medicine practices, concern often arises regarding smaller procedures, more distal fractures, tendon ruptures, and the need for plaster casting. The risks for VTE decrease with more distal injuries and surgeries. Studies of isolated fractures demonstrate a decreasing rate of DVT between proximal tibial plateau fractures, shaft fractures, and plafond fractures (43%, 22%, and 13%, respectively).44 In patients with lower extremity injuries managed in plaster casts, prophylaxis has been effective in decreasing the rate of VTE when diagnosed by venogram (18% placebo to 9.6% LMWH)26 and duplex ultrasound (17% control versus 5% LMWH).45 The importance of these distal and almost universally asymptomatic clots is unclear. The ACCP guidelines suggest that no chemoprophylaxis is routinely needed for patients with isolated lower extremity injuries. History of thrombosis after these relatively minor events may serve as a forewarning when more proximal surgery is undertaken. The association between long-distance travel and VTE has become common in the popular press. Athletes frequently enjoy the opportunity of world travel for competition and are at increased risk for VTE on flights longer than 8 hours. At peak condition, significant health compromise is unlikely in young elite athletes; however, subclinical thromboembolic disease could potentially compromise performance. Injuries during competition and their treatment may put athletes at increased risk on their return journey. Prolonged air travel before surgery increases the risk for perioperative VTE.46 The risk for VTE in the general flying population is related to an individual’s risk factors and the duration of the flight. The cause of increased risk is likely related to the decreased vascular flow of immobility, but dehydration, decreased cabin pressure, and relative hypoxia may play a role. Travelers with 8-hour or ­longer flights, compared with nontravelers, have more than double the number of ultrasound-diagnosed thrombosis (total VTE 2.8% versus 1.0%, proximal VTE 0.7% versus 0.2%, respectively).47 The absolute risk for fatal PE is very low (2.57 per 1 million flights longer than 8 hours),48 but air travel for longer than 8 hours carries 8 times the risk for fatal PE in nontravelers.49 Most patients who

develop travel-related thrombosis have one or more other risk ­ factors. Recommendations to help prevent VTE in patients traveling long distances include avoiding constrictive clothing, avoiding dehydration, doing calf-­stretching exercises, and when possible, taking frequent walks in the cabin. A review of six randomized trials of graduated compression stockings providing 15 to 30 mm Hg pressure at the ankle demonstrated decreases in VTE rates.48 For long-distance travelers with additional risk factors, graduated compression stockings or a single preflight injection of LMWH is suggested by the ACCP.26

Treatment of Deep Venous Thrombosis and Pulmonary Embolism� Prompt treatment of VTE can help prevent progression of thrombosis and minimize mortality from embolic disease. Orthopaedic surgeons frequently enjoy the consultation of a hematologist or internal medicine specialist for the definitive treatment of VTE. Knowledge of current treatment guidelines can be helpful to provide reassurance to patients and to expedite treatment when needed (Table 7B-9). For objectively confirmed acute DVT, LMWH or monitored UFH should be started in tandem with oral VKA. Heparin can be given as a continuous intravenous drip or intermittent subcutaneous injection schedule. At least 5 days of LMWH or UFH is recommended and continued until VKA achieves a stable effective level (INR, 2.0-3.0). LMWH has more predictable pharmacokinetics, which allows twice-daily dosing without monitoring unless the patient is pregnant or has severe renal insufficiency. Early discharge with home management is possible and is a cost-effective approach for proximal thrombosis.16 Randomized trials have demonstrated that long-term VKA anticoagulation decreases thrombosis propagation, recurrent VTE, and PE. The duration of anticoagulation is based on the number of previous VTEs and other risk factors. Most first-time postsurgical thrombi are treated as clots caused by a transient risk factor, and only 3 months of anticoagulation is recommended. Patients with cancer, thrombophilia, or recurrent episodes of VTE require longer or indefinite anticoagulation. Patients with cancer have less bleeding and more effective treatment with LMWH.16 ACCP guidelines recommend against the routine use of thrombolytic medicines (urokinase, streptokinase), catheter-directed thrombolytic therapy, surgical venous thrombectomy, and vena caval filters. Vena caval

Complications

filters may be indicated if there is a contraindication to or complication of anticoagulation therapy and in the rare case of recurrent VTE while on adequate anticoagulation.16 Traditionally, surgeons have questioned mobilization after the diagnosis of lower extremity venous thrombosis for fear of dislodging a thrombus and creating an embolism. Ventilation-perfusion scanning studies have demonstrated no decrease in PE with the use of bed rest. Pain and swelling have been shown to resolve more rapidly when patients are mobilized with early ambulation.16 The prompt diagnosis and treatment of PE with anticoagulants results in a mortality rate of about 2%. Current recommendations parallel treatment of DVT: prompt initiation of both LMWH and VKA. LMWH should be continued for at least 5 days and long enough to stabilize VKA levels to an INR of 2.0 to 3.0. VKA therapy should last at least 3 months after PE but should be extended in patients with cancer, multiple thrombophilic conditions, or recurrent PE. The routine use of thrombolytic therapy is discouraged except in acute massive embolism in hemodynamically unstable patients at low risk for bleeding. Thrombolytic therapy for VTE carries a 1% to 2% risk for intracranial bleeding. Because the dose of thrombolytic agents is designed to create systemic fibrinolysis, use in the immediate postoperative period is not recommended. Heroic procedures, including catheter extraction and pulmonary embolectomy, should be reserved for highly compromised patients who are unable to receive thrombolytics.16

SUMMARY Tipping the balance of bleeding and clotting with prophylaxis can decrease thromboembolic complications in patients undergoing major orthopaedic surgery. PE is the most common preventable cause of hospital death, and its prevention has become an indicator of quality of care for federal as well as private health care programs. The evaluation and treatment of nonfatal symptomatic DVT and PE are expensive, and most studies demonstrate that thromboprophylaxis decreases costs while improving the overall quality of care. Although the rationale for thromboprophylaxis in hospitalized patients undergoing major orthopaedic procedures is overwhelmingly supported by the literature and now considered standard of care, a gray zone exists for less invasive procedures and techniques. Minor surgeries in young patients who are motivated toward aggressive mobilization carry minimal risk for DVT and PE. Efforts in identifying patients who have a history or family history of VTE may be rewarded with a lower rate of thromboembolic complications. As the quality of orthopaedic trials advances, more specific recommendations can be made regarding DVT prophylaxis for specific surgical procedures, and new screening techniques may be developed to select the highrisk patients undergoing low-risk procedures who may benefit from prophylaxis.

C

r i t i c a l

P

385

o i n t s

•��� DVT

prophylaxis is the cost-effective standard of care for major orthopaedic surgery (total hip and knee arthroplasty, hip fracture work, multiple trauma, pelvic ring fractures) and is effectively accomplished with LMWH, fondaparinux, or warfarin. High-risk patients may benefit from extended (28-35 days after surgery) therapy after hip procedures. •��� Patients with high bleeding risk may benefit from ­mechanical methods of prophylaxis. •��� Arthroscopy and arthroscopically assisted cases (meniscectomies, synovectomies, osteochondral repair, chondroplasty, and ligament reconstruction) are at low risk, and routine prophylaxis is not recommended. •��� Isolated lower extremity fractures, tendon ruptures, and elective surgery on the foot and ankle, with or without the use of casting, are considered low risk, and routine prophylaxis is not recommended. •��� Patients who present with VTE after minor or lowrisk procedures may benefit from evaluation for ­ genetic ­thrombophilic conditions and should be counseled about their future increased risk for VTE and need for ­prophylaxis. •��� Most DVTs and PEs are asymptomatic. •��� Leg pain and swelling after orthopaedic surgery should be promptly evaluated with duplex ultrasound. •��� Chest pain, shortness of breath, and tachypnea after ­orthopaedic surgery warrants prompt evaluation with ­spiral CT pulmonary angiography. If contraindications exist to radiocontrast (allergy or renal insufficiency), ­duplex ­ ultrasonography and ventilation-perfusion scanning can be used. •��� Prompt treatment of VTE will save lives and prevent ­morbidity of recurrent events. Effective treatment ­includes using both LMWH and warfarin until INR ­stabilizes (INR, 2.0-3.0) and then continuing warfarin for at least 3 months.

S U G G E S T E D

R E A D I N G S

Geerts WH, Pineo GF, Heit JA, et al: Prevention of venous thromboembolism: Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 126(3 Suppl):338-400, 2004. Hull R: Peripheral venous disease. In Goldman L, Ausiello D (eds): Cecil Textbook of Medicine. Philadelphia, WB Saunders, 2004, pp 477-482. Lieberman JR, Hsu WK: Prevention of venous thromboembolic disease after total hip and knee arthroplasty. J Bone Joint Surg Am 87(9):2097-2112, 2005. Mitchell RN: Hemodynamic disorders, thromboembolic disease, and shock. In ­Kumar MV, Abbas AK, Fausto N (eds): Robbins and Cotran Pathologic Basis of Disease. Philadelphia, Elsevier Saunders, 2005, p 136.

R eferences Please see www.expertconsult.com

386

DeLee & Drez’s Orthopaedic Sports Medicine

S e c t i o n

C

Infection: Prevention, Control, and Treatment Chealon D. Miller, Jennifer A. Hart, and Mark D. Miller

Several infection-related topics should be reviewed from the field of sports medicine. This chapter will focus on ­superficial wound infections, shoulder infections, and infected anterior cruciate ligament grafts. Special ­consideration will also be given to the nuances of ­ methicillin-­resistant Staphylococcus aureus (MRSA) and unusual infections encountered by the orthopaedist. Before these specific topics are explored, general information about types of infection and antibiotic properties and administration will be presented.

ANTIBIOTIC PROPERTIES AND ADMINISTRATION Orthopaedic surgeons constantly face the threat of infection. Both superficial and deep infections have detrimental sequelae. Prevention of infection requires an understanding of antibiotic pharmacokinetics, common organisms encountered, and appropriate administration of antibiotics preoperatively, intraoperatively, and postoperatively. With an understanding of these topics, orthopaedic surgeons can better prevent, control, and treat sports-related infections. Elderly, obese, diabetic, and immunocompromised individuals have a higher risk of infection, and prophylactic antibiotics can help to reduce this risk. In immunocompetent individuals, this benefit has been debated. Many authors have shown that the benefits of prophylaxis can outweigh the costs even in immunocompetent patients. D’Angelo and colleagues reported nine cases of septic arthritis, and their cost/benefit analysis showed it was less expensive to give prophylactic antibiotics to all than to treat a few patients with septic joints.1 Kurzweil, after a thorough review of the literature, also advocated for prophylactic antibiotics in conjunction with proper preoperative hygiene.2 Others have proposed that the financial costs and adverse physiologic effects on the host outweigh potential prophylactic effects. Wieck and colleagues concluded that the costs and the risks for allergic reaction outweigh the protective benefits of prophylaxis and recommended the cessation of prophylactic antibiotics for arthroscopic surgery.3 Despite the continuing arguments, the current standard of care regarding antibiotic prophylaxis for arthroscopy and other orthopaedic surgery comes from the American Academy of Orthopaedic Surgeons (AAOS) advisory statement entitled, “Recommendations for the Use of Intravenous Antibiotic Prophylaxis in Primary Total Joint Arthroplasty,” and its recommendations are reviewed later.4 Despite conflicting opinions on the administration of antibiotics prophylactically, it is undisputed that with

c­ orrect dosing, bactericidal effects can be achieved for common microbes.5 Antibiotics must be able to adequately penetrate tissues in order to be effective in the eradication of infection. Intravenous or intramuscular administration depends on diffusion properties of the drug and requires concentration gradients for efficacy, whereas direct administration to tissues can occur through irrigation or antibiotic-­impregnated implants.6 Dosing depends on the pharmacokinetics of the antibiotic used, intraoperative blood loss, increased body mass index, and the duration of the surgical procedure. For intravenous administration, the lag time between the administration of the antibiotic and its intended destination must be considered. Studies have shown that this lag time ranges from 30 to 60 minutes after administration of antibiotics until it reaches the drainage from a wound. On the other hand, it takes 20 to 30 minutes for antibiotics in serum to penetrate and ­produce bactericidal levels in bone. Adequate bone-serum antibiotic concentration must be maintained through milligram per kilogram dosing and continuous administration of antibiotics intraoperatively.7 In North America, S. aureus and Staphylococcus epidermidis are the most common organisms isolated from postoperative wound infections. The next most common are streptococci and gram-negative bacilli (see Table 7C-2). If prophylactic antibiotics are used, they should cover the most common organisms. If signs and symptoms of ­infection are present postoperatively, broad-spectrum antibiotics should be used until culture results are available. Once speciation and sensitivity have been obtained, the clinician should ­tailor antibiotics to achieve optimal results.6 Cephalosporins are the antibiotic of choice for long procedures (>2 hours), especially those involving implants. Hill and colleagues showed that the use of first-­generation cephalosporins such as cefazolin is associated with a reduced incidence of deep wound infection following total hip arthroplasty.8 This information can be extrapolated to include shoulder implants as well. However, there is some debate as to whether the use of prophylactic ­ antibiotics in shorter procedures (<2 hours) is necessary. Henley and colleagues reviewed multiple cases of varying lengths involving the use of implants and determined that there was no statistically significant difference in the development of operative-site infection in procedures of less than 2 hours’ duration when the second-generation cephalosporin cefamandole was used. However, a significant benefit was found in cases of longer duration.9 Postoperative management should include vigilant surveillance of the wound for infection. Babcock and colleagues found an overall incidence of postoperative

Complications

387

   TABLE 7C-1  Antibiotics and Frequency of Administration Antibiotic

Frequency of Administration

Cefazolin Cefuroxime Clindamycin Vancomycin

Every 2-5 hr Every 3-4 hr Every 3-6 hr Every 6-12 hr

From American Academy of Orthopaedic Surgeons: Advisory statement: ­Recommendations for the use of intravenous antibiotic prophylaxis in ­primary joint arthroplasty. Available at http://www.aaos.org/about/papers/ adristmt/1027.asp.

i­nfection of 1.3% in their study and found two major causes: (1) intra-articular steroid injection and (2) the use of razors for hair removal. The study also ­ suggests that operating room personnel themselves may contribute to contamination, underscoring the importance of adherence to sterile procedure. Other studies have suggested prolonged surgery time, increased number of procedures during surgery, conversion to an arthrotomy, and previous procedures as causes for postoperative ­infections.10,11 Despite debates from various authors, recommendations from the AAOS are usually followed by today’s orthopaedist. These recommendations call for cefazolin or cefuroxime for all patients undergoing orthopaedic procedures.4,12-15 Clindamycin or vancomycin should be used for patients with a β-lactam allergy. Vancomycin may be used with confirmed MRSA colonization and in facilities with recent MRSA outbreaks. To optimize the efficacy of therapy, prophylactic antibiotics should be administered within 1 hour of skin incision. Vancomycin should be started within 2 hours of the start of the procedure. Dose amount should be proportional to patient weight, and antibiotics should be redosed if the procedure exceeds 1 to 2 times the antibiotic’s half-life or if there is greater than 1500 mL of blood loss (Table 7C-1).4,16-18 Postoperatively, antibiotic administration should not exceed 24 hours from wound closure even if a drain is in place.4,19 C

r i t i c a l

P

Figure 7C-1   Photograph of abscess. (Courtesy of University of Virginia Department of Dermatology, Kenneth Greer, MD, Chairman.)

abscesses are both considered to be types of ­ superficial wound infection. Cellulitis demonstrates features of both skin erythema and increased warmth and can be more diffuse in nature, whereas an abscess involves a localized collection of purulent material (Figs. 7C-1 and 7C-2).20 Whether these are seen as postoperative complications or sequelae of localized trauma, the problem should be addressed promptly to prevent the development of necrotic tissue, which may require significant surgical débridement.21 To prevent the negative sequelae of a soft tissue infection, the surgeon should be aware of the risks and

o i n t s

•��� S. aureus and S. epidermidis are the most common organ-

isms isolated from postoperative wound infections. or cefuroxime is recommended as prophylaxis for all patients undergoing orthopaedic procedures. Vancomycin is used for patients with an allergy to β-­lactam antibiotics. •��� Antibiotics should be administered within 1 hour of skin incision and should not extend beyond 24 hours of wound closure. •��� Antibiotics should be redosed if the procedure exceeds 1 to 2 times the antibiotic’s half-life or if there is greater than 1500 mL of blood loss.

•��� Cefazolin

SUPERFICIAL WOUND INFECTIONS Superficial infections are defined as those that involve the subcutaneous fascial layer, whereas deep infections are those that involve the muscle or bone. Cellulitis and

Figure 7C-2   Photograph of orthopaedic cellulitis. (Courtesy of University of Virginia Department of Dermatology, Kenneth Greer, MD, Chairman.)

388

DeLee & Drez’s Orthopaedic Sports Medicine

Box 7C-1 Risk Factors for Soft Tissue Infections

• Patients with immunocompromised states • Patients with diabetes • Patients who smoke • Hypoxic, ischemic, or poorly perfused tissue causes, know how to diagnose the problem, and ­ provide proper treatment. Certain patient populations are at greater risk for soft tissue infection than others (Box 7C-1). In particular, immunocompromised patients, diabetics, and smokers are more likely to develop soft tissue wound infections.20,22,23 The diabetic patient with poor glucose control allows a rich media for the proliferation of bacteria. All diabetic patients should be counseled before surgery about the importance of maintaining good glycemic control in the perioperative period to prevent infection. If necessary, a referral to a nutritionist or an endocrinologist may be obtained preoperatively to achieve this goal. In addition to these preoperative factors, intraoperative and postoperative factors can have an effect on postoperative wound healing. Besides disease states in the host, there are other risk factors, associations, and causes of soft tissue wound infections. Soft tissue infections tend to occur in ischemic, hypoxic, poorly perfused, or devitalized tissues.20,24 To avoid these complications, intravenous hydration and supplemental oxygen should be given during invasive orthopaedic procedures.24,25 Vince and colleagues showed that in joint arthroplasty around the knee, more lateral incisions or incisions through previous or the most recently healed incision allowed for fewer wound complications.21 Others have shown that high tourniquet pressures and aggressive knee flexion in the early postoperative period can induce hypoxia in the area around the operative site and lead to infection.26 Meticulous intraoperative hemostasis is also imperative. Hematoma formation along with persistent postoperative drainage are both significant factors in the development of superficial wound infections.27 Particular attention should be paid to smoking and its effect on wound complications. Smoking has been shown to be the single most important risk factor for the development of postoperative wound complications. Moller and colleagues illustrated that smokers are likely to experience wound complications 2 times more frequently than nonsmokers.24 Smoking appears to have its effect through cells in the immune system, but smoking has also been shown to accelerate the aging processes that naturally occur in the musculoskeletal system.28-30 Sports medicine orthopaedists will also encounter soft tissue infections in the outpatient setting. Acute traumatic wounds in athletic play, accidental or intentional human bites, and intravenous (IV) drug abuse are the most common causes of soft tissue infections in the immunocompetent individual. A complete history and physical examination on initial presentation can help identify skin lesions that may later develop into soft tissue infections. For very competitive athletes, any suspicious aspects of the history warrant a thorough physical examination of vascular access areas for possible IV steroid or performance-enhancing drug use.20

With consideration of at-risk populations and factors that may be related to soft tissue infection, the orthopaedist has the proper tools to approach the problem. The classic signs of tumor, calor, dolor, and rubor of the soft tissues should prompt concern. Rubor is unusual, and the presence of the other three signs of inflammation is enough to prompt further investigation.31 Vince and colleagues classified wound infections into three categories: (1) superficial without intra-articular sepsis, (2) deep with drainage through a defect in the arthrotomy, and (3) deep as the result of an infected arthroplasty. Differentiating among these categories based on peripheral laboratory values may be difficult because there may be no differences in erythrocyte sedimentation rate (ESR), C reactive protein (CRP), or peripheral leukocyte counts. Therefore, diagnosis of a superficial infection must be made with joint aspiration, with a negative aspiration indicative of only a superficial infection (Fig. 7C-3). If the first aspirate is negative and the suspicion of infection remains, at least two more aspirations should be performed.31 Drainage through the arthrotomy site can be tolerated for the first few days after surgery and may be managed with immobilization and sterile dressing changes for 3 to 5 days, but should be considered a joint infection if it continues past 5 days.21 Aside from soft tissue infections that occur in the community secondary to traumatic wounds, which require automatic referral to surgical services, treatment of soft tissue infections depends on the outcome of diagnostic procedures. Suspected soft tissue infections after surgical procedures require a diagnostic aspiration to rule out septic arthritis before the institution of antibiotics. If the patient is on antibiotic therapy, an initial aspiration should be performed, followed by cessation of antibiotics for 10 days, and then repeat aspiration.31 Antibiotics should be given only ����������������� ������������ prophylactically if the culture and cell count of the aspirate are negative for intra-articular sepsis (Table 7C-2). If sepsis is present, immediate arthrotomy with irrigation and débridement is necessary.21 Outcomes depend on how quickly treatment is instituted. Potential complications of delayed treatment include osteomyelitis and even amputation. These devastating conditions may lead to significant morbidity in a condition that can be easily managed if found early. C

r i t i c a l

P

o i n t s

•��� Soft tissue infections tend to occur in ischemic, hypoxic,

poorly perfused, or devitalized tissues. patients, diabetic patients, and smokers are at greatest risk for development of a soft ­tissue infection. •��� Smoking has been shown to be the single most important risk factor for the development of superficial wound ­infections. •��� Acute traumatic wounds in athletic play, human bites, and intravenous drug use are the most common causes of soft tissue infections in immunocompetent individuals. •��� Diagnosis of a superficial infection should occur only ­after a negative joint aspiration and should be treated with antibiotics. • Joint infection should prompt immediate arthrotomy with irrigation and débridement.

•��� Immunocompromised

Complications

389

Joint aspiration

Positive aspiration

Negative aspiration

Septic arthritis

Treat with arthrotomy and débridement

No drainage through arthrotomy

Drainage through arthrotomy

Superficial wound infection

Manage with immobilization and sterile dressing change

Treat with antibiotics

>5 days of drainage

<5 days of drainage

Septic arthritis

Superficial wound infection

Treat with arthrotomy and débridement

Treat with antibiotics

Figure 7C-3   Treatment algorithm for superficial wound infections.

SHOULDER INFECTIONS Spontaneous shoulder infections are rare but may occur after shoulder arthroplasty. Prevalence of shoulder infection has been cited between 0% and 3.9% for unconstrained shoulder arthroplasties and 0% to 15.4% for constrained

   TABLE 7C-2  Most Common Organisms Isolated from Postoperative Wound Infections and Treatment Organism

Treatment

Staphylococcus aureus

Penicillinase-resistant synthetic penicillins (PRSPs, e.g., nafcillin or oxacillin) MRSA species treated with vancomycin Alternative therapies: erythromycin, ­first-­generation cephalosporins, amoxicillin/­ clavulanate, azithromycin, and clarithromycin Alternative to vancomycin for MRSA: ­teicoplanin, trimethoprim/sulfamethoxazole (Bactrim), doxycycline, minocycline, fusidic acid, fosfomycin, rifampin, and novobiocin) Same as above High likelihood of resistance to β-lactams; hence, some suggest vancomycin even if ­susceptibility to β-lactams is reported4 Nafcillin (oral dicloxacillin) For known streptococcal infection, use penicillin G Quinolone antibiotics

Staphylococcus epidermidis Streptococcus species Gram-negative bacilli

Modified from Miller M: Review of Orthopaedics. Philadelphia, Saunders, 2004.

arthroplasties.32,33 Suggested treatment options include antibiotic suppression of infection, débridement with prosthesis retention, direct exchange of the prostheses, delayed reimplantation (a two-stage operation), resection arthroplasty, arthrodesis, and amputation (Table 7C-3).32-36 Deciding on treatment for a postarthroplasty infection depends on many patient factors including age, patient expectations, soft tissue deficits, and patient health. Moreover, the virulence of the pathogen should be taken into consideration.37 Risk factors for postoperative glenohumeral joint infections include oral steroid use, advanced age, rheumatoid arthritis, diabetes mellitus, history of previous remote bone infection, history of septic arthritis in the affected joint, history of previous infected prosthesis, malnutrition, immunosuppressed states (human immunodeficiency virus, hypogammaglobulinemia), and chronic disease states.33,37 Clinical symptoms include pain, stiffness, fever, night sweats, and chills. Signs include a draining sinus, ­ erythema, and a localized effusion. Diagnosis is usually made by ­ aspiration of the joint and obtaining laboratory values, including a peripheral white blood cell count, ESR, and CRP. The most common organisms encountered in postarthroplasty infection are S. aureus, S. epidermidis, ­ Propionibacterium acnes, and ­ Streptococcus viridans (Table 7C-4).33,37 P. acnes is thought to have a propensity for the shoulder and has been shown in patients with rotator cuff repair. Heightened awareness for this organism must be present when a patient presents with a shoulder ­infection.38

390

DeLee & Drez’s Orthopaedic Sports Medicine

   TABLE 7C-3  Common Treatments of Postarthroplasty Shoulder Infection and Reported Outcomes Treatment

Reported Outcomes

Suggested Use

Resection arthroplasty

Successful infection eradication High patient satisfaction Poor Neer criteria for pain goals

Revision arthroplasty (delayed reimplantation)

Successful infection eradication High patient satisfaction Increased risk for morbidity Increased risk for persistent infection Decreased risk for recurrent deep infection

Low-demand patients Elderly patients Patients with intractable infections Patients who do not want large reconstructive procedures High functional–demand patients Young patients Patients with significant bone stock, but compromised soft tissues

Prosthetic removal with débridement and immediate reimplantation Débridement with prosthetic retention

Increased risk for persistent infection

Braman and colleagues reviewed a series of patients receiving resection shoulder arthroplasties and compared them to patients who underwent revision arthroplasty (Fig. 7C-4).37 Outcome measures included pain relief, function, and infection eradication. Despite poor Neer criteria for pain goals, infection was eradicated in all of their patients, and patient satisfaction was high.39 The authors concluded that resection arthroplasty was a beneficial procedure that should be reserved for low-demand elderly patients who do not want large reconstructive procedures and patients with intractable infections.3 In the same study, revision arthroplasty was presented as a more technically difficult procedure, causing increased morbidity and increased risk for persistent infection.3 However, when the patient is young, has high functional demands, and has significant bone stock, a revision arthroplasty in a two-stage procedure provides the best hope for pain relief and eradication of shoulder infection.33 Proper counseling should be given to the patient in either situation about the expectations of the procedure. All these procedures should be accompanied by proper initial broad-spectrum antibiotics and an infectious disease consultation.

Patient has no significant comorbidity Patient with intact surrounding soft tissue Patient with organisms that are not difficult to treat Acute infections (<1 month after arthroplasty) Patient with an acute hematogenous infection Infection with organism susceptible to IV antibiotics No prosthesis pathology

Two additional treatments include (1) prosthetic removal with débridement and immediate reimplantation and (2) débridement with prosthetic retention. Sperling and colleagues compared these two additional approaches to postarthroplasty shoulder infection with revision and resection arthroplasty and showed similar results to the Braman study.33 Although débridement with implant retention showed high rates of reinfection, success was more likely when débridement and retention were performed in an acute infection (first month after arthroplasty), for a patient with an acute hematogenous infection, infection by organisms susceptible to intravenous antibiotics, or infection with no evidence of prosthetic pathology (well-fixed

   TABLE 7C-4  Common Organisms in Postarthroplasty Shoulder Infection and Suggested Treatment Organism

Treatment

Staphylococcus epidermidis

Rifampin plus nafcillin or ­cloxacillin for 2 wk followed by rifampin plus quinolone Same as above MRSA: rifampin plus vancomycin for 2 wk followed by quinolone (or teicoplanin, fusidic acid, cotrimoxazole, or minocycline) Penicillin G or ceftriaxone for 4 wk followed by amoxicillin Clindamycin or penicillin G or ceftriaxone

Staphylococcus aureus

Streptococcus species Propionibacterium acnes

Reprinted with permission from Trampuz A, Zimmerli W: Antimicrobial agents in orthopaedic surgery: Prophylaxis and treatment. Drugs 66(8):1089-1105, 2006.

Figure 7C-4   Radiograph of antibiotic spacer with antibiotic beads and anchoring device. (Courtesy of University of Virginia Department of Orthopaedic Surgery, Division of Sports Medicine, Chairman.)

Complications

prosthesis and good soft tissue coverage).33 Immediate reimplantation using antibiotic-impregnated cement for prosthetic fixation showed a decrease in recurrent deep infection, but further studies are warranted.33,40 In most cases, after the diagnosis of postarthroplasty shoulder infection has been made, intravenous antibiotics should be administered. Recommendations are for 4 to 6 weeks of intravenous antibiotics followed by a course of oral antibiotics. Although duration of oral antibiotic course is patient dependent, treatment should be continued for at least 6 weeks. Spontaneous infection of the glenohumeral joint can occur, even though septic arthritis usually affects weightbearing joints.41-43 Septic arthritis of the shoulder usually occurs in very young infants or elderly patients with chronic diseases such as liver disease, cancer, diabetes, and alcoholism. Infants have immature immune systems, whereas sick, elderly patients have compromised immune systems. Shoulder septic arthritis is also more likely to develop in a joint with chronic arthritis than in a normal, healthy joint. If primary septic arthritis is suspected in a young adult, IV drug use should be in the differential diagnosis. Patients with glenohumeral infection usually present with fever and other constitutional symptoms. Other common findings at presentation are pain, limited range of motion, swelling, deformity, and subluxation of the glenohumeral joint. If an acute infection is suspected, instability testing should not be performed. Diagnosis involves joint aspiration and fluid analysis. Elevated leukocyte counts with low levels of glucose in the synovial fluid can be seen in crystalline synovitis and other acute inflammatory joint diseases. Therefore, Gram stain and culture should be obtained for a definitive diagnosis. A purulent aspirate should spawn treatment with empirical antibiotics for 4 to 6 weeks. Lossos and colleagues’ review of the literature showed that S. aureus, Salmonella species, gram-negative bacilli (Klebsiella pneumoniae, Proteus)����������������������� , and nonenterococcal and Streptococcus species were the most common pathogens found in primary bacterial arthritis (Table 7C-5). Lossos found no evidence of Neisseria gonorrhoeae or Mycobacterium tuberculosis as the sole pathogen causing monarthritis of the shoulder. Therefore, empirical treatment should focus on treating the most common organisms. In most

391

cases, staphylococcal coverage is adequate, but in elderly, immunocompromised individuals and in neonates, gramnegative coverage should be considered. These antibiotics should be administered intravenously and not directly into the joint because of the risk for chemical synovitis.44 When there is some question about the reliability of Gram stain and culture, the surgeon’s clinical experience is invaluable in directing the first plan of treatment. The triple therapy of articular rest, intravenous antibiotics, and occasional drainage of the joint are the mainstays of treatment for primary septic arthritis of the glenohumeral joint. Drainage can be performed through closed-needle aspiration, arthroscopy, or open arthrotomy. Needle aspirations can be difficult, and if there is no sign of clinical or laboratory improvement within 5 to 7 days of the first needle aspiration, open surgical drainage should be performed. If there is a history of previous joint disease, long duration of symptoms, or coexistent osteomyelitis, the joint should be drained without delay (Fig. 7C-5).44 Primary or spontaneous septic arthritis of the shoulder must be treated early, with attention paid to host defense mechanisms and virulence of the bacteria. Virulent organisms such as Clostridium perfringens warrant an immediate arthrotomy, irrigation, débridement, and prolonged anti­ biotics (Fig. 7C-6). Symptoms and laboratory ­abnormalities warrant further investigation, with aspiration of the joint and fluid analyses providing definitive diagnosis. The most common pathogen is S. aureus, whereas gram-­negative bacilli are more prevalent in children.44

C

r i t i c a l

P

o i n t s

• Treatment options for postarthroplasty shoulder infec-

tion include antibiotic suppression of infection, débridement with prosthesis retention, direct exchange of the prostheses, delayed reimplantation, resection arthroplasty, arthrodesis, and amputation. •��� P. acnes has a propensity for the shoulder. •��� Spontaneous glenohumeral infections usually occur in very young infants or elderly patients with chronic diseases. Suggested treatment is the triple therapy of articular rest, intravenous antibiotics, and occasional drainage of joint.

   TABLE 7C-5  Common Treatments of Spontaneous   (Primary) Shoulder Infection and Reported Outcomes Organism

Treatment

Staphylococcus aureus

Penicillinase-resistant synthetic penicillin plus a third-generation cephalosporin MRSA: penicillinase-resistant synthetic penicillin plus an antipseudomonal ­aminoglycoside Penicillinase-resistant synthetic penicillin plus a third-generation cephalosporin Ampicillin Quinolone antibiotic Clindamycin or penicillin G or ceftriaxone

Streptococcus species Salmonella species Gram-negative bacilli Streptococcus species

Modified from Miller M: Review of Orthopaedics. Philadelphia, Elsevier, 2004; and Bratzler DW, Houck PM, Richards C, et al: Use of antimicrobial ­prophylaxis for major surgery: Baseline results from the National Surgical Infection Prevention Project. Arch Surg 140(2):174-182, 2005.

INFECTED ANTERIOR CRUCIATE LIGAMENT GRAFTS Anterior cruciate ligament (ACL) tears and their reconstruction are a significant part of any sports practice. Although infection after arthroscopic ACL reconstruction is rare, its presence can have devastating consequences. The management of these infectious complications has been the source of controversy. The most recent literature supports joint lavage, débridement, and antibiotics as well as early recognition of infection as key components to successful treatment. However, before a treatment plan is derived, one must be aware of the risk factors, presenting signs, diagnostic tools, and likely offending organisms in intra-articular and periarticular infections.

392

DeLee & Drez’s Orthopaedic Sports Medicine

Spontaneous (primary) shoulder infection

History of previous joint disease Long duration of symptoms Coexistent osteomyelitis

Intravenous antibiotics Articular rest Needle drainage of joint

Open surgical drainage and antibiotics

5-7 days of above “triple therapy”

No clinical or laboratory improvement

Clinical or laboratory improvement

Open surgical drainage and antibiotics

Antibiotic treatment for 4 to 6 weeks

Figure 7C-5   Suggested treatment algorithm for spontaneous (primary) shoulder infection.

Infection risks have been curtailed with the implementation of meticulous sterile technique, with the current risk for postoperative septic arthritis having a reported incidence of only 0.14% to 0.78%.7,45-49 Unrecognized septic arthritis can lead to arthrofibrosis and loss of joint space. Several risk factors have been shown to increase the risk for infectious complications. Judd and colleagues showed an increased incidence of intra-articular infections in ACL reconstructions performed with hamstring autografts and in patients with previous knee surgery.49,50 However, Indelli and colleagues found no difference in the use of allografts versus autographs, attributing infection to the

Figure 7C-6   Photograph of necrotizing fasciitis caused by more virulent organism. (Courtesy of University of Virginia Department of Dermatology, Kenneth Greer, MD, Chairman.)

presence of a foreign body with absent blood supply rather than the type of graft.46 An increase in the number of procedures, longer operative time, larger or additional incisions, and increased hardware burden are also risk ­factors for the development of infection.45 Other associations that have been suggested but not proved to be of consequence include age, gender, surgical side, tourniquet use, hardware type, and suture use.45-47 Early recognition of postoperative septic arthritis and its likely presentation should be appreciated in order to minimize morbidity. The most common signs of both intra-­articular and extra-articular infection include fever, knee swelling with effusion, erythema and warmth, pain with limited range of motion, and local incisional drainage. Still, other signs of infection may be more subtle, such as decreased physical therapy performance. These same symptoms may mimic normal postoperative symptoms, but duration of symptoms is key to making the diagnosis. Typical post–knee arthroscopy pain symptoms last about 1 to 2 days,10 whereas most intra-articular postsurgical infections occur after 14 days, and most extra-articular infections occur after 23 days.50 Clinical suspicion of infection should prompt immediate laboratory investigation and aspiration. Synovial fluid characteristics, most notably turbid, yellow, or blood-tinged fluid, are red flags for infection. Subsequent Gram stain and cell count are helpful in identifying a possible organism and instituting a focused antibiotic regimen. Evidence of intra-articular infection is generally considered to be a positive culture from knee aspiration or more than 10,000 cells/mL in patients with symptoms. The white cell count from the aspirate itself can be alarmingly high, reaching values as high as 100,000 cells/mL, with a predominance of polymorphonuclear cells. ESR and CRP values are usually elevated above normal, although serum white blood cell count does not usually increase substantially, and normal to high-normal values are common.49

Complications

   TABLE 7C-6  Common Organisms Associated  

with Infected Anterior Cruciate Ligament Grafts   and Suggested Treatments Organism

Treatment

Methicillin-sensitive Staphylococcus aureus

Antistaphylococcal penicillin such as nafcillin or oxacillin or first­generation cephalosporin Same as above Penicillin G

Staphylococcus epidermidis Peptostreptococcus species

If there is clinical suspicion of infection, immediate treatment is most effective in preserving the joint space. The most common organisms isolated from infected knee aspirations include MRSA, S. epidermidis, and Peptostreptococcus species (Table 7C-6).45-47,49 Therefore, antibiotic treatments should provide appropriate coverage with agents such as parenteral first-generation cephalosporins (e.g., cefazolin) and nafcillin. Vancomycin should be included to cover MRSA until speciation and sensitivities are obtained. Antibiotics should be tailored when sensitivities are available. Duration of IV antibiotic use is variable, ranging from 2 to 6 weeks.45-47,50 Oral antibiotics may then be implemented for a period of 3 weeks following intravenous treatment. Arthroscopic lavage, incision and drainage, débridement, synovectomy, and evaluation of the graft should coincide with the implementation of antibiotics to minimize articular cartilage loss.51 Débridement should be deep and aggressive to interrupt any loculations that may have formed within the infected joint.47 Also, pulsatile, high-pressure lavage with either saline or lactated Ringer’s solution should be used.46,47 The addition of nonsteroidal anti-inflammatory medications may help to temper the inflammatory destruction associated with infection.46,51 However, if this intervention fails, open arthrotomy, débridement, and lavage must be performed.49 Continued physical therapy is also critical to retaining full function of the infected joint. Continuous passive motion with active ankle pumps and passive and active motion exercises have been shown to slow progressive degeneration of the joint space versus joint immobilization or intermittent motion. These therapies help to prevent the formation of synovial adhesions and keep the synovial medium fluid.45,46,52 The question then becomes whether to remove the graft (Box 7C-2). There has been support for several approaches, including retaining the graft with aggressive arthroscopic débridement and antibiotics or early graft removal. Studies have shown that the presence of less virulent organisms such as S. epidermidis can allow for graft retention. However, if

Box 7C-2 Reasons for Graft Removal

• Virulent organism encountered on culture • Reconstructed anterior cruciate ligament graft appears grossly infected

393

a more virulent organism is present (usually due to delayed treatment), if the reconstructed ACL appears grossly infected (i.e., a fibrinous exudate on the graft itself or soft and swollen articular cartilage at the time of débridement and lavage), or if symptoms suggesting persistent infection continue after initial débridement, then strong consideration must be given to graft removal.46,49,50,53,54 In a study by McAllister and colleagues, the grafts were successfully retained in patients who had early intervention (within 24 hours of the onset of symptoms).47 However, these patients subsequently developed joint space loss, diffuse thinning of the articular cartilage, longer overall hospitalizations, and moderate quadriceps muscle strength deficits, most likely a result of postinfectious changes. Burks and colleagues demonstrate some advantages to early removal and early reimplantation of grafts in the management of postoperative septic arthritis. In their studies, grafts were removed in all of the affected patients with the idea that retained grafts may serve as foreign bodies serving as media for postinfectious synovitis and ongoing chondrolysis.45 While the Matava survey suggested that graft replacement should not be performed before 6 to 9 months after graft removal for optimal results,53 the patients in the Burks study had reimplantation surgeries within 6 weeks of completing antibiotic treatment. The follow-up evaluations revealed that these patients had no joint-space narrowing or osteophyte formation and had full range of motion in the affected knee. However, given the small number in the study and no controls in place for associated risk variables (e.g., virulence of offending organism), interpretation of these results is limited. Although postoperative septic arthritis following ACL reconstruction is a rare complication, failure to respond to this infection in a timely manner may result in loss of normal joint function. The earlier the intervention, the higher the likelihood of preserving the joint space and retaining the graft. Poor prognosis is associated with delay in treatment and the virulence of the offending organism, with S. aureus having the most devastating consequences and S. epidermidis being minimally destructive. In approaching the patient, Indelli and colleagues offer the following algorithm: protect the articular cartilage first, then protect the graft.46 Prioritizing the preservation of articular cartilage may prevent subsequent thinning, which would eventually necessitate the need for total knee replacement.

C

•��� Risk

r i t i c a l

P

o i n t s

factors for ACL graft infection include previous knee surgery, long operative time, large or numerous incisions, and multiple additional procedures during graft placement. •��� Treatment of ACL graft infection should include joint lavage, débridement, and antibiotics. •��� Physical therapy should be included in the treatment regimen of an infected ACL graft to prevent degeneration of the joint space. •��� Rule for treatment of infected ACL graft: protect the ­articular cartilage first, then protect the graft.

394

DeLee & Drez’s Orthopaedic Sports Medicine

HARDWARE INFECTIONS Hardware infections are a major consideration in sports medicine. Total shoulder arthroplasties, hemiarthroplasties, and other hardware used in sports procedures pre­sent a new dilemma in infection prevention, control, and treatment. The overall risk for infection with all orthopaedic implants is presently quoted at 0.5% to 5%. These are relatively low numbers, but considering the volume of hardware implantations each year, this figure is significant. Studies have shown that in the first 2 years after joint replacement, infections have been reported as the second main cause of revision after instability.55-60 These reports were for knee replacements, but could likely be extrapolated to shoulder arthroplasties. The first considerations in the control of hardware infections are efforts toward prevention and an understanding of hardware infection pathogenesis in order to direct ­ treatment. Prevention should target improving operating standards, minimizing contamination during surgery, using perioperative antibiotics, and using isolation for patients with pathogenic strains.61,62 The pathogenesis of hardware infection requires a more detailed approach and a greater understanding of microbiology and hardware properties. In particular, the factors that allow for microbial adhesion, colonization of implant surfaces, and evasion of host defenses are important investigative topics. The pathogenesis of hardware infection differs from other postsurgical infections because of the presence of the implant. The implant affects the surrounding interstitium and creates what has been called a locus minoris resistentiae, or a region of local immune depression. This environment is an immunocompromised, fibroinflammatory zone that is more likely to have microbial colonies and infection.63 Studies have also shown that when a foreign material such as a prosthesis is present, fewer microorganisms are needed to cause infection.64,65 Naturally occurring small movements in the implant can also release debris that can damage the surrounding soft tissues, causing further damage and nidus for infection. The interaction between bacteria and the implant that allows infection is initially passive. However, actual infection is the result of an active ­process,

   TABLE 7C-7  Common Organisms Found in Hardware Infections and Suggested Treatment Organism

Treatment

Staphylococcus epidermidis

Rifampin plus nafcillin or ­cloxacillin for 2 wk followed by rifampin plus quinolone Same as above MRSA: rifampin or vancomycin for 2 wk followed by quinolone (­teicoplanin, fusidic acid, cotrimoxazole, or minocycline)

Staphylococcus aureus

involving bacterial adhesion, receptor proteins, and the development of a biofilm by the microbes on implant surfaces (Fig. 7C-7).66-68 Most implant infections are caused by staphylococci species, with S. aureus and S. epidermidis accounting for 67% of all implant infections. Antibiotic resistance is becoming an increasing concern in implant infection for a number of reasons: (1) Staphylococcus is the organism most often seen in infections, and this organism in particular is showing increased rates of resistance; and (2) bacteria that form biofilms on the implant are more resistant to antibiotics. Campoccia and colleagues showed that 4 of 5 strains of Staphylococcus species showed resistance to penicillin drugs, and 4 of 10 species showed resistance to methicillin­oxacillin combinations. Fortunately, in their series, they found no vancomycin resistance among these organisms (Table 7C-7).55 In an effort to minimize infection, control of the environment and personnel contamination should also be a focus. Whyte and colleagues performed a study and found that patient’s skin and airborne particles from operating room personnel accounted for most intraoperative contamination, with operating room personnel accounting for 98% of cases and patient’s skin 2% of cases (Fig. 7C-8).69 Knobben and colleagues suggested preventing airborne particles by improving airflow systems with laminar flow systems. The authors also suggested limiting the number of operating room personnel and restricting their movements (to prevent agitation of particles) as a solution for contamination from personnel.70 Recently, efforts have focused on the interface between the biomaterial surface and the surrounding tissues. In particular, efforts are directed toward interference with bacterial adhesion or the chemical properties of the outer layer of the device by coating materials with surfactants, proteins, or negatively charged polysaccharides such as hyaluronan and heparin.71,72 Coatings that have proved effective are those incorporating chlorhexidine–silver sulfadiazine and 2% Operating room personnel Patient’s skin

Figure 7C-7   Electron micrograph of Staphylococcus epidermidis as biofilm. (From Zimmerli W, Trampuz A, Ochsner PE: Prosthetic-joint infections. N Engl J Med 351[16]:1645-1654, 2004.)

98% Figure 7C-8   Causes of intraoperative contamination.

Complications

those with minocycline-rifampin. ­Alternatives to coatings include incorporating active substances into bulk materials and cavity-filling materials (e.g., antibiotic-impregnated cements). Although these materials have been effective, new concern over the selection of resistant strains of microbes has risen.55 The future of control of implant infection resides in a number of products. Studies have been performed on chitosan, a bioactive material with broad-spectrum antimicrobial properties.73 Moreover, because staphylococci are the most common organisms, researchers have developed a coating composed of a staphylolytic endopeptidase called lysostaphin.74 Because bacteria rely on iron sequestration for growth, they have developed siderophores to sequester iron. Researchers are developing inhibitors to block these siderophores and further prevent microbial advance.55 C

r i t i c a l

P

o i n t s

•��� In the first 2 years after joint replacement, hardware in-

fections have been reported as the second most common cause of revision after instability. I  •��� mplants affect the surrounding interstitium and ­create a locus minoris resistentiae (region of local immune depression). •��� Control of the operating room environment and ­prevention of personnel contamination should be the ­focus of hardware infection prevention. •��� Efforts to treat hardware infections should be directed ­toward interference with bacterial adhesion or the chemical properties of the outer layer of the device.

COMMUNITY-ACQUIRED STAPHYLOCOCCAL AUREUS AMONG ATHLETES Although previously thought to reside behind hospital walls and affect the critically ill, MRSA is a growing nemesis in the community, especially among healthy, young athletes. Athletes have long been known to have great susceptibility to infectious diseases, most commonly herpes simplex virus and MRSA.75,76 Enteroviruses, Tinea trichophytina, Streptococcus pyogenes, hepatitis A and B, measles, Leptospira species, and Neisseria meningitides also contribute to disease among athletes. Close contact, contaminated quarters, and poor hygiene allow these organisms to fester and make transmission rather easy. Disease transmission occurs most commonly through person-to-person contact (e.g., cutaneous infections, fecal-oral spread), airborne droplet spread (e.g., viral infections such as colds), or exposures to common sources of infection, including contaminated water sources, food, or athletic equipment.75-78 Cutaneous viral, bacterial, and fungal infections are most common and occur in contact sports, such as wrestling, rugby, and football, in which physical trauma to the skin allows a portal for entry of the infectious agent. Furthermore, certain sports have a predilection for a particular infectious agent. For example, wrestlers are more likely to develop tinea corporis (Trichophyton

395

tonsurans is the most common strain), whereas MRSA infections are found in the football player population, in which skin trauma is significant risk factor for infection. The disturbing rise of MRSA infection among athletes requires astute recognition of possible infection, those at risk for infection, and appropriate prevention and treatment. The organism is harbored in the anterior nares by attachment to the nasal mucin and keratinized epithelial cells of asymptomatic carriers who may otherwise have no risks for MRSA infection.79 The resistant strains integrate the staphylococcal cassette chromosome (SCCmec) into their genomes, of which the type 4 SCCmec has been associated with community-acquired MRSA strains and the PantonValentine leukocidin gene. Colonization of the skin may occur through minor skin abrasions, injection sites, or catheter devices. Among athletes, skin-to-skin contact is the most significant mode of transmission.76,80 For this reason, wrestlers and rugby and football players are at greatest risk. Even sports with minimal contact, such as fencing, have an elevated risk because sharing of equipment and protective clothing is common.81 Skin trauma, such as turf burns, shaving, and chafing, significantly increase risk for transmission.80-83 Begier and colleagues showed that infection was 7 times greater in players who had sustained turf burns than those without injury and 6 times greater in those who practiced body shaving (shaving a site other than the face) compared with those who did not.82 Both turf burns and body shaving produce microabrasions on the skin that serve as an entry point for the bacteria. Furthermore, practice mats, shared towels, and contaminated locker rooms may contribute to spread.84 Poorly sanitized whirlpools may facilitate bacterial spread if players fail to shower or enter whirlpools with open wounds. Players with greatest face-to-face play are also at considerable risk. In one study of a MRSA outbreak in a rugby team, all infected individuals were team forwards who have close contact during the game.85 In a review of community-acquired MRSA among professional football players, linemen and linebackers were at higher risk for infection than those in backfield positions.83 Cornerbacks and wide receivers have greater person-to-person contact during scrimmages and drills and were also more likely to develop infection.82 The key to treatment is identifying the resistant strain early in the infection and providing the appropriate antibiotic course to prevent more serious complications, such as septicemia, necrotizing pneumonia, and necrotizing fasciitis. Community-acquired MRSA infections are cutaneous in nature, appearing initially as an abscess with or without accompanying cellulitis in an otherwise healthy individual with no known risks for MRSA (e.g., hospital exposure).6 The abscess may then become exquisitely tender, erythematous, and indurated, and there may be multiple lesions. The infection most commonly originates from the site of a skin abrasion or open wound.80,83 If a skin furuncle or abscess persists despite coverage for S. aureus with agents such as dicloxacillin or cephalexin, MRSA should be considered.84 Furthermore, the presence of pain out of proportion to the physical examination, even likened to a spider bite, should also cause suspicion for MRSA infection (Fig. 7C-9).86,87

396

DeLee & Drez’s Orthopaedic Sports Medicine Figure 7C-9   A and B, Examples of   community-acquired MRSA infections of the skin. Lesions that have signs of necrosis, resemble spider bites, or have pain and erythema out of proportion to visual inspection are highly suggestive of community-acquired MRSA. (Courtesy of University of Virginia Department of Dermatology, Kenneth Greer, MD, Chairman.)

A

B

Treatment begins with incising the skin abscess and draining the purulent fluid during the initial presentation of a patient with a persistent skin infection. Obtaining a culture of methicillin-resistant strain of S. aureus early on will help to direct antibiotic therapy and curb the spread of infection.83 Gram stain of the fluid will reveal gram-positive cocci in clusters. Cultures are typically positive, and pulsedfield electrophoresis analysis may show a common clone of community-acquired MRSA among coinfected teammates. Although smaller abscesses may heal with incision and drainage alone, MRSA abscesses generally require systemic antibiotics for complete eradication. Unlike its hospital counterpart, community-acquired MRSA is generally susceptible to clindamycin, sulfamethoxazole-trimethoprim (SMX-TMP), fluoroquinolones, and rifampin.80,83,86 However, if susceptibility testing reveals that the strain isolated from the wound is resistant to erythromycin, it is possible for the strain to also have resistance to clindamycin and additional testing (e.g., erythromycin-clindamycin double-disk diffusion test) is required. For cases in which clindamycin susceptibility is questionable, SMX-TMP is a suitable agent. The use of topical mupirocin ointment to decolonize the nasal passages and skin lesions of MRSA is somewhat controversial given the development of resistance with long-term use.81 However, considerable efficacy in eradicating MRSA colonization in hospital patients has been demonstrated and may be of some use for short-term use in conjunction with systemic antibiotic treatment. Although difficult to eradicate completely, infectious risk can be significantly curtailed by implementing preventive measures across all athletic teams (Box 7C-3). First, players should all have up-to-date immunization histories, including MMR, tetanus, influenza, and hepatitis B.77 Second, limiting exposure to sick individuals, frequent hand washing (with soap and water or alcohol-based hand sanitizer), and showering well with soap after (and before) activities all help to reduce bacterial burden and risk for

infection.77,80,81 Skin wounds should be cleaned with soap and water and covered until the wound has completely healed. Noninfected players should avoid the lesions and drainage of infected teammates.6,81,84 Players, coaches, and other athletic personnel should be instructed in proper wound first aid and should notify a physician if an infection does not heal, if symptoms worsen, or a fever develops.81 Facilities should be routinely sanitized, which involves disinfecting showers, practice areas, uniforms, and water sources such as drinking bottles and water coolers. Players should be discouraged from sharing water bottles, towels, Box 7C-3 Measures for Preventing ­Staphylococcal Skin Infections among Sports Participants

• Cover

all wounds. If a wound cannot be covered adequately, consider excluding players with potentially infectious skin lesions from practice or competitions until the lesions are healed or can be covered adequately. • Encourage good hygiene, including showering and washing with soap after all practices and competitions. • Ensure availability of adequate soap and hot water. • Discourage sharing of towels and personal items (e.g., clothing or equipment). • Establish routine cleaning schedules for shared ­equipment. • Train athletes and coaches in first aid for wounds and recognition of wounds that are potentially infected. • Encourage athletes to report skin lesions to coaches and encourage coaches to assess athletes regularly for skin lesions. From Centers for Disease Control and Prevention: Methicillin-resistant ­Staphylococcus aureus infections among competitive sports participants: ­Colorado, Indiana, Pennsylvania, and Los Angeles County. MMWR Morb Mortal Wkly Rep 52:793-795, 2003.

Complications

and soaps and encouraged to report to coaches the presence of suspicious skin lesions. The epidemiology of certain infections may also help to focus preventive measures. Because the prevalence of tinea infection is higher among wrestlers, these players should be systematically screened ������������������������������� by coaching staff ������������� for the presence of infection and placed on prophylactic antifungal agents, such as fluconazole, if suspicious lesions are present.77 The presence of MRSA in rugby and football players should also warrant routine skin screening of players during the season as well as application of antibacterial agents on skin abrasions after games.85 If players are found with suspicious lesions, immediate wound care should be administered, and the player should avoid other teammates and common areas.85 Ultimately, players, coaches, and team physicians will benefit greatly from practicing and encouraging good hygiene, knowing the signs and symptoms of infection, and intervening with appropriate wound care early. Early intervention is the key to implementing protective meas­ ures so as to minimize the spread of disease.

C

r i t i c a l

P

o i n t s

•��� Community-acquired MRSA infection is likely to ­colonize

the nares of an asymptomatic person with no other risks for exposure. •��� Community-acquired MRSA among athletes is most commonly spread through skin-to-skin contact. •��� Skin lesions that originate from a site of trauma, such as skin abrasion, that are red and indurated, and that are exquisitely tender out of proportion to the examination may suggest community-acquired MRSA skin infection. •��� Incision and drainage and appropriate antibiotic treatment can prevent serious complications from MRSA infection, such as necrotizing pneumonitis and ­septicemia. •��� Clean clothing, sanitized athletic facilities, and good personal hygiene are all preventive measures to help curb the incidence of MRSA infection among athletes.

UNUSUAL INFECTIONS Infection is a problem when encountered in the sports practice. The preceding sections have outlined the common presentations of infection, key diagnostic methods, therapeutic options, and preventative measures. However, when confronted with persistent infection, further investigation into patient history and host characteristics must be explored. It may be possible that benign, even commensal, organisms are the culprits. Such infections are particularly problematic because they are overlooked until late in the infectious work-up. Initial presentation is usually misleading, with fever, erythema, and swelling most closely resembling bacterial septic arthritis. Furthermore, standard culture media are usually inappropriate to garner a positive result. Taking a detailed patient history, including the ­presence of other illnesses, traumas, or illicit drug use, may steer the diagnosis in the direction of unusual, rare infectious

397

agents. For example, a history of water-related injury in the setting of chronic infection and septic arthritis may make a clinician suspicious for Aeromonas hydrophila, a gram­negative bacillus found in warm climates and freshwaters,88 or Erysipelothrix rhusiopathiae, a gram-­positive anaerobe found in waters contaminated with feces of infected animals.89 History of a cat bite or scratch in a patient with fever and a swollen, warm joint is suggestive of septic arthritis caused by Pasteurella multocida.90,91 Immunosuppressed individuals, particularly those receiving chemotherapy for leukemic disease, are particularly susceptible to Candida species infection of the joints.92,93 Although Candida albicans is the most common ­fungal agent to cause septic arthritis, Candida tropicalis is typically found in patients who are granulocytopenic.92 In a case study of a patient with acute lymphoblastic ­leukemia, Sim and colleagues isolated C. tropicalis from the joint of a patient who was neutropenic after a second phase of chemotherapy and on broad-spectrum antibiotics to treat bacterial infection prophylactically. The authors postulate that immunosuppression from the chemotherapy made the patient leukopenic and unable to defend against infection. Broad-spectrum antibiotics additionally destroyed normal bacterial flora, allowing the fungal species to flourish.92 Young infants with signs of septic arthritis and history of congential immunodeficiencies, such as chronic granulomatous disease, may be infected not only with likely pathogens, such as S. aureus, but also with unusual bacterial infections, such as Nocardia species and atypical mycobacteria.94 Immunosuppression from alcoholism, human immunodeficiency virus infection, chronic corticosteroid use, or immunomodulator treatment for rheumatoid arthritis (e.g., tumor necrosis factor-α inhibitors) should also cause suspicion for a fungal or other unusual source of infection, such as Mycobacterium species.95-97 Mycobacteria have also been associated with aquatic reservoirs. Veitch and colleagues found a link between a local irrigation system and 29 cases of Mycobacterium ulcerans.98 This discovery underscores how important the source of water supplies is to the orthopaedic surgeon in the operating room. The sterility of water must be scrutinized and proper steps taken to prevent contamination of pipes. If Mycobacterium infection is suspected, polymerase chain reaction analysis is the best test for diagnosis.99 Treatment of Mycobacterium should consist of combined rifampin, isoniazid, pyrazinamide, and ethambutol therapy and in some cases surgical débridement.100 This treatment regimen is usually effective, but there have been case reports of resistance to traditional treatments. If resistance is suspected, consulting with infectious disease specialists, obtaining information from microbiologists, and determining sensitivities are critical in eradicating infection.101 In patients with no history of trauma or of immunosuppression, the presence of unusual pathogens may represent the presence of an undiagnosed malignancy. Rallot and colleagues described the diagnosis of prostate cancer in a gentleman who initially presented with Campylobacter fetus septic arthritis.102 Primary meningococcal arthritis was the presenting illness in a patient who was later diagnosed with multiple myeloma.103 It is likely that the immunosuppression caused by the ensuing malignancy predisposes individuals to these unusual infections.

398

DeLee & Drez’s Orthopaedic Sports Medicine

   TABLE 7C-8  Infection and Associated Pathogens Association

Unusual Pathogen

Injury sustained in fresh water in warm climates Water possibly contaminated with animal feces Recent cat bite or scratch Immunosuppressed patients receiving chemotherapy Congenital immunodeficiency Significant alcohol history Intravenous drug use

Aeromonas hydrophila

Periodontal disease Previous joint surgery or prosthesis

Erysipelothrix rhusiopathiae Pasteurella multocida Candida species (e.g., Candida tropicalis) Nocardia species Mycobacterium species Serratia species (e.g., Serratia marcescens) Streptococcus sanguis Mycobacteria kansasii

Other lifestyle factors can also influence the diagnosis of an unusual pathogen. A patient with a history of intravenous drug use and persistent septic arthritis despite appropriate treatment should create concern for Serratia marcescens, a well-documented pathogen in this population.104 Septic arthritis in a patient with periodontal disease or recent dental repair for caries may suggest flora typically inhabitant to the oral cavity and upper respiratory tract, such as Streptococcus sanguis.105,106 History of joint surgery or prosthesis may predispose an individual to unusual infection with Mycobacteria kansasii.107 The diagnosis of such unusual pathogens on initial presentation is quite unlikely. The number of pathogens that can cause disease is quite high, and having the foresight to provide the appropriate culture media for isolation of such pathogens is difficult. However, when persistent infection is present in a compromised host or an individual with a distinguishing condition or event history, then the suspicion for more indolent and less common pathogens may lead to appropriate patient care (Table 7C-8). C

r i t i c a l

P

o i n t s

•��� The patient history helps to identify environments and other risk factors for exposure to unusual pathogens. •��� Immunosuppressed individuals, elderly patients, and infants with congenital immunodeficiency are particularly susceptible to unusual infections, such as Candida, ­Serratia, and Mycobacterium species. •��� Lifestyle factors, such as alcoholism and intravenous drug use, should broaden the infection differential diagnosis to include unusual pathogens. •��� When unusual orthopaedic infections are found in an otherwise healthy individual, an undiagnosed malignancy may be present.

S U G G E S T E D

R E A D I N G S

American Academy of Orthopaedic Surgeons: Advisory statement: Recommendations for the use of intravenous antibiotic prophylaxis in primary joint arthroplasty. Rosemont, Illinois, AAOS, 2004. Bergmann KC: Effect of smoking on immune function. Allerg Immunol 26:3-14, 1980. Burks RT, Friederichs MG, Fink B, et al: Treatment of postoperative anterior cruciate ligament infections with graft removal and early reimplantation. Am J Sports Med 31(3):414-418, 2003. Campoccia D, Montanaro L, Arciola CR: The significance of infection related to orthopedic devices and issues of antibiotic resistance. Biomaterials 27:2331-2339, 2006. Centers for Disease Control and Prevention: Methicillin-resistant Staphylococcus ­ aureus infections among competitive sports participants: Colorado, Indiana, Pennsylvania, and Los Angeles County. MMWR Morb Mortal Wkly Rep 52:793-795, 2003. Charalambous CP, Zipitis CS, Kumar R, et al: Soft tissue infections of the extremities in an orthopaedic centre in the UK. J Infect 46:106-110, 2003. Condon RG: Surgical Infections: Selective Antibiotic Therapy. Baltimore, Williams & Wilkins, 1981, pp 125–130. Howard RS: Surgical Infectious Diseases. East Norwalk, Conn, Appleton & Lange, 1995���������������������������� , pp ����������������������� 1227–1229. Judd MA, Bottoni LT, Kim D, et al: Infections following arthroscopic anterior cruciate ligament reconstruction. Arthroscopy 22(4):375-384, 2006. Matava MJ, Evans TA, Wright RW, Shively RA: Septic arthritis of the knee following anterior cruciate ligament reconstruction: Results of a survey of sports medicine fellowship directors. Arthroscopy 14(7):717-725, 1998. Neer CS, Watson KC, Stanton FJ: Recent experience in total shoulder replacement. J Bone Joint Surg Am 64:319-337, 1982. Ross GN, Baraff LJ, Quismorio FP: Serratia arthritis in heroin users. J Bone Joint Surg Am 57(8):1158-1160, 1975. Saleh K, Olson M, Scott R, et al: Predictors of wound infections in hip and knee joint replacement: Results from a 20 year surveillance program. J Orthop Res 20(3):506-515, 2002. Sperling JW, Kozak TK, Hanssen AD, Cofield RH: Infection after shoulder arthroplasty. Clin Orthop Relat Res 382:206-216, 2001. Trampuz A, Osmon DR, Hanssen AD, et al: Molecular and antibiofilm approaches to prosthetic joint infection. Clin Orthop 414:69-88, 2003. Turbeville SD, Cowan LsD, Greenfield RA: Infectious disease outbreaks in competitive sports: A review of the literature. Am J Sports Med 34(11):1860-1865, 2006. Vicari P, Feitosa Pinheiro R, Chauffaille ML, et al: Septic arthritis as the first sign of Candida tropicalis fungaemia in an acute lymphoid leukemia patient. Braz J Infect Dis 7(6):426-428, 2003.

R eferences Please see www.expertconsult.com

C H A P T E R�  

  8�

Nutrition, Pharmacology, and ­Psychology in Sports S ect i o n A

Nutrition for Sports Leslie Bonci

Sports nutrition is a critical component of training and competition. To optimize performance and safeguard health, an athlete must be adequately and appropriately fueled and hydrated. Nutrition is important not only on the playing field but also in the classroom and work settings because of the physiologic benefits of promoting muscle growth, enhancing recovery, preventing injury, and supporting rehabilitation. The science of sports nutrition is constantly expanding through field and laboratory studies of athletes in all sports and of all ages. Of particular interest are studies of fluid and sodium balance, the optimal mix of nutrients to enhance recovery, techniques to help the athlete decrease body fat while sparing muscle, and maximizing muscle gain while minimizing increase in body fat. However, the ever expanding world of supplements and fad diets can be a source of misinformation and confusion for athletes, which is why the need for credible, updated information is so critical. Certainly athletes are always looking for the edge to become faster, stronger, and leaner. As a result, every health professional involved in the care of athletes needs to understand the basics of sports nutrition, from macronutrient and micronutrient requirements to weight management and supplements. Athletes need to understand and buy into the concept that sports nutrition is as important a component of training as field and weight room training, and the athletes who fuel and hydrate optimally will notice the greatest improvement in performance and will most likely be healthier. The collection of baseline data for athletes with a basic nutrition screening form can be invaluable in troubleshooting, making referrals, and enhancing performance. Figure 8A-1 details a nutrition screening form for athletes. The few minutes it might take for an athlete to complete this form can help to identify the individual who may be at risk for disordered eating, injury, and overtraining. Having the athlete complete this questionnaire can provide valuable information to the health ­professional.

GOALS OF SPORTS NUTRITION • To safeguard the athlete’s health and optimize performance • To achieve and maintain ideal body mass • To maintain proper hydration and electrolyte balance • To provide adequate carbohydrates to optimize respiratory metabolism • To preserve lean body mass with essential amino acids • To maximize oxygen delivery systems and oxidative phosphorylation with trace elements • To develop high-density skeletal structure • To promote recovery from training • To prevent injury and maximize the efficacy of rehabilitation through nutrition support • To improve performance by increasing the speed of muscle fiber contraction and the number of muscle fibers that contract • To tailor the eating plan to promote sustained energy over the course of the competitive season • To prepare for life beyond sports and adjust caloric intake accordingly Qualitative similarities exist across the spectrum of athletes, but there may be quantitative differences based on type of sport, gender, body composition, and age. More athletes are competing in senior games, and athletes are competing at a younger age; therefore, sports nutrition recommendations must be tailored accordingly.

ENERGY SUBSTRATES The two major types of exercise are aerobic and anaerobic. Anaerobic exercise is that performed without oxygen and involves maximal intensity, short-duration activities (1 to 2 minutes) such as 100- to 200-m track or swim events, or high-intensity bursts of activity such as in football, 399

400 1. 2. 3. 4. 5. 6.

7. 8. 9.

10.

11. 12.

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Name: Male Female Sport: Position: Age: years How would you describe your eating habits? (Check one) a. Good b. Fair c. Poor How many times a day do you eat? number of times per day How often do you eat out? number of times per week When you go out to eat, where are the three most common places you go? a. b. c. Do you avoid any of the following foods? (Check all that apply) a. Red meat b. Poultry (chicken, turkey) c. Fish d. Dairy (milk, cheese) e. Vegetables f. Fruits g. Fried foods h. Breads i. Grains (pasta, rice) j. Fast foods k. Sweets (candy, desserts) l. Alcohol m. Fats/oils (mayonnaise, salad dressings, butter) Do you currently take any dietary supplements? Yes No If yes, which ones? (Check all that apply) a. Creatine b. Protein shakes c. Amino acids d. HMB e. Vitamins f. Minerals g. Herbs h. “Andro”/DHEA i. Pyruvate j. Energy boosters (e.g., ephedra, ma-huang) k. Other, specify

13. Do you know which dietary supplements are banned or restricted by our sports organization? Yes No 14. In a typical workout, about how many cups of water, juice, sports drink, or noncaffeinated beverages do you drink before or during exercise? (Check one) a. None b. One or two cups c. Three to five cups d. More than five cups 15. Overall, how satisfied are you with the physical appearance of your body? (Check one) a. Very satisfied b. Somewhat satisfied c. Somewhat dissatisfied d. Very dissatisfied 16. How easy or difficult is it for you to maintain your in-season weight? (Check one) a. Very easy b. Somewhat easy c. Somewhat difficult d. Very difficult 17. Do you have any personal goals for body composition? Yes No If yes, which ones? (Check all that apply) a. Gain lean mass/weight b. Decrease body fat c. Lose weight d. Maintain current body composition e. None 18. Would you be interested in having your body fat tested? Yes No 19. Please indicate the topics you would like to learn about by placing a check mark next to them. a. Nutrition programs for peak performance b. Weight control c. Weight gain d. Grocery store tour e. Hiring a personal chef f. Cooking demonstrations/meal preparation g. Tips on eating out

Figure 8A-1   Nutrition screening form for athletes HMB, beta-hydroxy-beta-methylbutyrate: DHEA, dehydroepiandrosterone.

­ asketball, and soccer. Aerobic exercise is oxygen depenb dent and is therefore a lower intensity type of activity but can be performed for a longer period of time, such as distance running, cycling, or a triathlon. To be fueled optimally during activity, the body must have adequate stores of the macronutrients used as energy substrates. Glycogen, the storage form of carbohydrate, is stored in the muscle and liver. Most exercise is fueled by carbohydrate and fat; protein provides a fuel source if carbohydrate stores are inadequate. Fuel use is determined by the intensity and the duration of activity as well as by the level of training. ­Muscle glycogen is the major source of carbohydrate, followed by

liver glycogen and then blood glucose. Aerobic training and diet manipulation can significantly increase muscle glycogen stores.1 Intense training, especially in the athlete who is new to the sport or who may be making the transition from high school to college where training sessions are much more intense, frequent, and of longer duration, can result in depleted glycogen stores resulting in fatigue and subpar performance. Muscle glycogen is used for intense, short-duration activity as well as for endurance exercise. The rate of muscle glycogen use is most rapid during the early part of exercise and is related to exercise intensity. Muscle glycogen declines with continued exercise and is

Nutrition, Pharmacology, and Psychology in Sports

selectively depleted from the muscles that are involved in physical work. As muscle glycogen stores decline, blood glucose and liver glycogen become important fuel sources. Plasma free fatty acids are used as a fuel source during endurance-type activities through the process of adipose tissue lipolysis. As mentioned earlier, amino acids can be broken down to glucose to provide energy during activity, but only when carbohydrate stores are low. One of the main concerns of athletes is to prevent fatigue and maintain adequate energy levels for training and competition. Certainly many athletes do not get enough sleep, but even if athletes are well rested, suboptimal nutrition can be a significant cause of fatigue through one or more of the following mechanisms: • Inadequate fluid intake • Inadequate calorie intake • Inadequate carbohydrate intake • Inadequate protein intake • Iron deficiency • Vitamin and mineral deficiency

Hydration Fluid balance is essential to cardiovascular function, thermoregulation, injury prevention, optimal performance, and recovery from exercise. Fluid loss can be significant during exercise (up to 3 L/hr).2 Fluid loss increases heart rate by 8 beats per minute and impairs performance when it is more than 1.8% of total body water; mental functioning is impaired through a decrease in sustained attention, response time, and task accuracy, and error rate is increased.3 Heat-related injuries increase with increased body water loss. In addition, an athlete can become dehydrated because of changes in altitude, increases in training intensity and frequency, sudden climate changes, and long plane flights. The body cannot tolerate even slight dehydration, but unfortunately, thirst sensation is dulled by exercise, and voluntary fluid consumption is insufficient to meet fluid needs. Dehydration also reduces the gastric emptying rate, complicating the rehydration process.4 Curtailing fluid intake is still a common practice in certain sports. The chronic dehydration that often accompanies weight-class sports can impair the athlete’s ability to train and compete optimally. The goal of fluid intake is to prevent dehydration. A flawed hydration plan can lead to overdrinking or underdrinking, and either can result in health issues and puts the athlete in jeopardy of heat illness and need of medical care. It is also imperative for the athletes to drink smart by considering his or her size, sweat rate, pace, and the environmental conditions. In 2004, the Institute of Medicine Dietary Reference Intake for Fluid recommended the adequate intake of water as 3.7 L/day in males (130 ounces per day) or 16 cups, and 2.7 L/day for females (95 oz/day), or 12 cups.5 From 1985 to 2003, several position stands have recommended fluid guidelines before, during, and after activity. There is no doubt that the athlete who begins training or competition in a dehydrated state will jeopardize performance and health, but each athlete needs to individualize a hydration plan. The American College of Sports Medicine (ACSM) published a position stand on fluid replacement in 1996

401

providing the rationale and the guidelines for hydration.6 The ACSM has issued a revised position stand that was published in February 2007.7 The National Athletic Trainers position stand on fluid was released in 2000,8 and most recently, USA Track and Field has released fluid guidelines for competition.9 Athletes require a minimum of 14 to 40 ounces of fluid per hour of exercise, but most athletes consume only 8 ounces per hour.6 A larger fluid intake during exercise leads to greater cardiac output, greater skin blood flow, lower core temperature, and reduced perceived effort of exertion.10 The overall goal is clear and copious urine as a sign that the body is well hydrated. Athletes may ask for a recommendation about the types of fluid to consume. It is important to consider the sport, duration, calorie needs, and taste preferences. Water is a noncalorie fluid that works well for short-duration activities, but it not as beneficial for exercise lasting longer than 60 minutes, or for more intense activity lasting shorter than 30 minutes. Juices can provide calories and carbohydrate, but they contain fructose, which has a decreased absorption rate and may cause gastrointestinal distress; juices are generally not advised before exercise. Carbonated beverages before activity may cause gastrointestinal distress and often confer a feeling of fullness before fluid needs have been met. Caffeine-containing beverages do not have a diuretic effect, but caffeine alone is not a source of energy and may not be necessary fuel, and caffeinated products such as carbonated beverages and energy drinks may be too concentrated in carbohydrate. Alcoholic beverages have a diuretic effect, causing the body to lose valuable fluids before activity begins as well as impairing reaction time. The sports drinks can be an appropriate option for longer-duration sports and are certainly extremely popular with young athletes. The sports drinks contain a dilute glucose solution that stimulates water and sodium absorption so that more fluid is absorbed than from plain water.11,12 Because sports drinks have a fairly low carbohydrate content, they empty more rapidly from the stomach than a more concentrated beverage does. For optimal gastric emptying to be achieved, fluids should be cold or cool. A large volume of fluid empties more rapidly than smaller amounts. One liter of fluid empties from the stomach and is absorbed by the intestine within 1 hour.6 Athletes need to practice drinking during training to determine a comfort level and to learn to drink proactively, instead of reactively. The recommended fluid intake is 2 to 3 quarts of fluid per day for basic needs plus 1 liter of fluid for every 1000 calories expended.6 Fluid guidelines for specific sports are listed in Table 8A-1.10 Strategies for fluid replacement before, during, and after exercise are as follows6: • Before exercise 500 mL (17 ounces) of fluid 2 hours before exercise 8 to 16 ounces of fluid 30 minutes before exercise • During exercise 4 to 8 ounces of fluid every 15 to 20 minutes during exercise with sports drink for sports lasting longer than 1 hour OR 14 to 40 ounces of fluid per hour depending on sweat rate

402

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 8A-1  Fluid Guidelines for Specific Sports Sport

Cups of Fluid per Hour

Hockey Football Volleyball Soccer   Males   Females Basketball   Males   Females

3-6 3-6 or more depending on sweat rate 2.5-4 3-5 2.5-4 3-6 2.5-4

• After exercise 24 ounces of fluid for every pound lost during exercise to achieve normal hydration within 6 hours after activity The most important fact is that athletes need to establish a schedule and consider hydrating before, during, and after exercise. For children younger than 10 years, the goal is to drink to satisfy thirst plus an additional 3 to 4 ounces of fluid. Older children and adolescents should also drink to satisfy thirst plus an additional 8 ounces of fluid.13 Some athletes may inquire about the use of glycerol for hyperhydration. Glycerol is a three-carbon molecule that is the structural core of triglycerides and phospholipids. Glycerol ingestion increases blood osmolarity, decreasing urine production and increasing fluid retention. Even though glycerol is easily absorbed, the increased weight may be a disadvantage, and glycerol loading can cause headaches, dizziness, bloating, and nausea.14 If an athlete wants to try glycerol, the best option may be to try a premixed solution, such as Pro Hydrator, or to purchase glycerate and mix it with water in the proportion of 0.45 g of glycerate per pound of body weight mixed in 48 ounces of water and consumed 1 to 1½ hours before exercise.15 This should be tried only in a practice situation, when the athlete can assess his or her tolerance to glycerol, without the stress of a competition situation. To encourage optimal hydration, coaches, athletic trainers, parents, and other health care professionals can assist the athlete in the following ways: • Encourage athletes to start exercise well hydrated. • Athletes should weigh in and out to determine sweat and fluid lost to replace appropriately. • Recommend that the athlete carry fluid with him or her. Dry mixes of lemonade, fruit punch, or sports drinks are lightweight and easy to carry, and they can be added to drinking water to provide fluid and carbohydrate. • Encourage athletes to drink during exercise. • Remind athletes of the importance of drinking enough with every meal to meet baseline fluid requirements. • Reinforce the need to drink on schedule, not sporadically. • Advise the athlete to drink beyond thirst. • Advise the athlete to drink enough before exercise to have a full stomach. Gastric emptying is more rapid and effi­cient when the stomach is somewhat full rather than empty. • Instruct the athlete that fluid is useful to the body only if it is swallowed, not poured over one’s head.

Sodium Requirements In addition to fluid, some athletes lose significant amounts of sodium during exercise. They are called salty sweaters; these athletes are more likely to experience muscle cramps, their sweat stings their eyes, and they may notice salt on their skin or clothes after exercise. These athletes must be vigilant with sodium intake in every way. Many athletes and coaches mistakenly believe that eating bananas or drinking orange juice will alleviate or prevent cramps, but these foods are high in potassium, which is lost in small amounts during exercise compared with sodium losses, which can exceed 8000 mg in a 2-hour practice. However, some athletes may voluntarily restrict sodium intake to lower the risk for hypertension, but most athletes require a liberal salt intake to maintain blood volume. An athlete will most likely require more than the 2400 mg sodium/day as recommended in the Dietary Guidelines. The athlete who drinks water to the exclusion of all else, loses significant sodium in the sweat, and does not consume adequate sodium in the diet is at risk for hyponatremia. Although this is more likely to occur in endurance events, there have been cases of other types of athletes who have suffered from inadequate sodium intake. Sodium consumption during and following dehydrating exercise will maintain or restore blood volume more completely than plain water. For salty sweaters, suggest the following: • Consuming sports drinks instead of water during exercise • Adding extra salt to food • Consuming salty condiments such as Worcestershire sauce, soy sauce • Eating salty foods: pickles, crackers, pretzels, soup, broth • Drinking salty beverages: V8 or tomato juice

Calorie Requirements Working muscles require fuel; if calorie needs are suboptimal, the body will fatigue earlier, and performance will be curtailed. Some athletes regularly undercut their energy needs, increasing the likelihood of early fatigue and risk for injury. Others eat in excess of need, resulting in excess stores of adipose tissue, which can adversely affect performance. Calorie needs are higher for an athlete than for a nonexercising person and need to be individualized according to gender and weight. An athlete who has been injured may require additional calories early in the recovery process to aid tissue repair, but fewer calories are often required as the frequency and intensity of activity decline. Athletes who retire from their sport need to learn how to eat less than in their playing days or suffer the consequences of carrying around excess weight. In general, the calorie requirements for active adults, children, and adolescents are as follows: • Men: weight (pounds) × 23 = minimum number of calories per day16 • Women: weight (pounds) × 20 = minimum number of calories per day16 • Children and adolescents: Boys and girls aged 7 to 10 years: 2000 calories per day16

Nutrition, Pharmacology, and Psychology in Sports

High-school boys: 3000 to 6000 calories per day16 High-school girls: 2200 to 4000 calories per day16 The composition of these calories influences performance. The primary fuel substrates for activity are carbohydrate and fat; protein plays more of a supporting role. The goal is to achieve a balance in the diet through a mix of carbohydrate, protein, and fat. This has become especially challenging in light of the popular eating plans recommending that entire categories of foods be limited or avoided. Endurance athletes have traditionally relied heavily on carbohydrate as the mainstay of diet, whereas in strengthtype sports (e.g., football), athletes have been told that a high-protein diet should be the focus.17 Although the overall amount of food consumed may vary, every athlete should aim to include protein, carbohydrate, and fat at every meal and snack.

Carbohydrate Requirements Achieving optimal carbohydrate nutriture is important to maintain the usual training intensity, to prevent hypoglycemia during exercise, to serve as fuel substrate for working muscles, and to assist in postexercise recovery. Carbohydrate use increases with increased exercise intensity but decreases with increased exercise duration. The higher the initial glycogen stores, the longer an athlete can exercise at a given intensity level. The goal of carbohydrate feeding is to fill carbohydrate stores in the muscles and liver. Eating increases glycogen stores, whereas exercise depletes glycogen stores. Glycogen depletion can occur in sports requiring nearly maximal bursts of effort. Athletes who do not optimally refuel may experience gradual and chronic glycogen depletion that can decrease endurance and performance. Training glycogen depletion is often accompanied by a sudden weight loss. For optimal glycogen stores to be maintained, carbohydrate needs must be estimated on the basis of the number of hours the athlete trains daily. Carbohydrate requirements are always higher for training than for competition. Athletes may consume inadequate amounts of carbohydrate because of calorie restriction, avoidance of certain foods (e.g., sugar), fad diets, sporadic or infrequent meals, and poor nutrition knowledge of good carbohydrate sources versus marginal choices. Needs can be estimated as in Table 8A-2.18,19 There has been much discussion about which type of carbohydrate is better for sports, simple or complex. The distinction is not that clear, and what matters most is the total amount of carbohydrate consumed on a daily basis. Athletes can use the nutrition facts panel on a food label to quantify the amount of carbohydrate ingested. Some athletes are more comfortable ingesting carbohydrates in   TABLE 8A-2  Carbohydrate Needs per Hour of Training Hours of Daily Training

Grams of Carbohydrate per Pound of Body Weight

1 2 3 4+

2.7-3 3.6 4.5 5.4-5.9

403

  TABLE 8A-3  Carbohydrate Content of Certain Foods Food

Amount

Carbohydrates (g)

Bagel Bagel Cheerios Corn Pops Granola, low-fat Swedish fish Orange juice Coke Gatorade Sports gel Gatorade Energy Drink Yogurt, fruit Raisins Pretzels

2 ounces 4 ounces 1 cup 1 cup 1 cup 1 handful 8 ounces 8 ounces 8 ounces 1 ounce 12 ounces 8 ounces ¼ cup 1 handful

38 76 23 28 82 39 27 27 14 28 78 42 31 22

a liquid form, such as Gatorade Energy Drink, Ultra Fuel, or gels. Table 8A-3 lists some of the most frequently consumed carbohydrate-containing foods. Athletes have recently begun experimenting with manipulating the type of carbohydrate consumed at various points during exercise according to the glycemic index of the food.20 The glycemic index indicates the actual effects of carbohydrate-rich foods and fluids on blood glucose and insulin levels. The glycemic index ranks foods by measuring the blood glucose response after ingestion of a test food that provides 50 g of carbohydrate compared with the blood glucose response to a reference food. The response reflects the rate of digestion and absorption of a carbohydrate-rich food. Foods are classified into the three categories of high, moderate, and low glycemic index, as outlined in Table 8A-4. Manipulating the meal choices on the basis of the glycemic index may enhance carbohydrate availability and improve athletic performance. Carbohydrate-rich foods of low glycemic index may help to promote sustained availability of carbohydrates when they are consumed before exercise. Carbohydrate-rich foods of moderate to high glycemic index may promote carbohydrate oxidation when they are ingested during exercise and may promote glycogen repletion when they are consumed after exercise. Athletes will need to experiment to find out which foods work well and, more important, do not cause gastrointestinal distress. Consuming dried beans or lentils before exercise may be fine for a cyclist, but it may not be desirable for a runner. Athletes have also experimented with the concept of carbohydrate loading or muscle glycogen supercompensation, which combines tapering of exercise with a high­carbohydrate intake to top off muscle glycogen stores.

  TABLE 8A-4  Glycemic Index of Various Foods High Glycemic Index Moderate Glycemic Index Low Glycemic Index Glucose White bread Potatoes Breakfast cereals Sports drinks Carrots

Sucrose Soft drinks Oats Tropical fruits Bagels, wheat bread Cookies, cake

Fructose Milk Yogurt Lentils, dried beans Pasta, rice Cold climate fruits

404

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The original method called for a depleting exercise protocol coupled with a low-carbohydrate diet, followed by 3 days of rest with an extremely high carbohydrate diet. This often made the athlete feel exhausted during the lowcarbohydrate intake phase and heavy during the carbohydrate-loading phase. Current guidelines recommend 3 to 5 days of carbohydrate loading to attain maximal glycogen levels, and the exercise done to lower glycogen stores must be the same as the athlete’s competitive event.21 Carbohydrate loading is advantageous only for endurance athletes whose event lasts longer than 90 minutes. Carbohydrate loading before a 10-K event is not helpful and may actually make the athlete feel heavier and stiff. Carbohydrate needs for activity are divided into three distinct time periods: before, during, and after exercise. The goal of pre-exercise carbohydrate is to provide energy for the athlete who exercises heavily in excess of 1 hour. Pre-exercise carbohydrate also helps prevent the feelings of hunger, which can be distracting, especially in a competition. The pre-exercise carbohydrate also elevates blood glucose levels to provide energy for the exercising muscles. Current guidelines recommend 1.8 g of carbohydrate per pound of body weight within 3 to 4 hours before exercise and 0.5 g per pound 1 hour before exercise.22,23 • Example 1.8 g carbohydrate 3 to 4 hours before exercise 120-pound athlete requires 216 g carbohydrate 12-ounce glass of cranberry juice = 54 g carbohydrate 8 ounces of yogurt flavored with ½ cup granola = 96 g carbohydrate English muffin with 1 tbsp peanut butter and 1 tbsp jelly = 46 g carbohydrate Total: 196 g carbohydrate 0.5 g carbohydrate 1 hour before exercise 120-pound athlete: 60 g carbohydrate 12 ounces of Gatorade Energy Drink = 78 g carbohydrate Carbohydrate consumption during exercise maintains the availability and oxidation of blood glucose late in exercise and improves endurance. During exercise, the ingestion of carbohydrate exerts a liver glycogen-sparing effect, resulting in delayed hypoglycemia. Ingesting carbohydrates may also be advantageous in stop-and-go sports and should be encouraged during breaks in play. Guidelines24 recommend 30 to 60 g of carbohydrate per hour during exercise in the form of: 5 to 10 ounces of sports drink every 15 to 20 minutes 2 gels per hour + water (average, 20 to 28 g carbohydrate per packet) A handful of gummy-type candy + water The goal of postexercise carbohydrate ingestion is to elevate glucose as quickly as possible. Waiting too long to refuel will reduce muscle glycogen storage and impair recovery. The recommendation is 0.7 g of carbohydrate per pound within 30 minutes after exercise and again 2 hours later for those training longer than 90 minutes at a time.25 Postexercise carbohydrate repletion can be from

solid or liquid sources. The body can store twice as much muscle glycogen with sucrose or glucose sources than with fructose sources.26 The following foods can be used for repletion: Lemonade Fruit punch A concentrated carbohydrate beverage Granola Cereal bars Sweetened cereal Gummy-type candy For the 120-pound athlete, 84 g of carbohydrate: 4-ounce bagel with 1 tbsp jelly = 89 g carbohydrate 16 ounces of Ultra Fuel = 100 g carbohydrate 16 ounces of fruit punch and a cereal bar = 87 g carbohydrate Some studies have suggested that consuming protein and carbohydrate together after exercise can enhance muscle glycogen resynthesis by stimulating insulin. The recommended guidelines call for a protein-to-carbohydrate ratio of 1:3.27,28,29 Examples of this are the following: Gatorade Nutrition Shake (carbohydrate-protein sports supplement beverage) Sports bars Trail mix of three parts cereal to one part nuts Yogurt and granola

Protein Requirements Although most athletes are aware of the need for carbohydrate as part of a good training diet, protein intake tends to run the gamut from minimal to excessive; the athletes who need the most consume the least, and those who need the least consume the most. Protein is important for muscle growth and aids in recovery and repair after muscle damage. As a fuel source, protein provides up to 15% of the fuel during activity, when muscle glycogen stores are low, and only 5% when muscle glycogen stores are adequate.30 As exercise increases in intensity and duration, so does the use of protein as a fuel source. Exercise promotes muscle protein loss because of reduced protein synthesis and increased protein catabolism during and immediately after exercise.30 With training, the breakdown and loss of muscle protein diminish, which is why protein needs are often higher in the initial phases of training than in an athlete who is well trained. As a result of training, protein anabolism is enhanced in the recovery period after exercise, and regular training increases the effectiveness of protein synthesis during recovery.30 If training sessions are too frequent or protein intake is insufficient to meet needs, protein catabolism will exceed anabolism, resulting in reduced gains or loss of body protein.30 Achieving optimal protein nutriture can be challenging. Athletes who do not meet their needs are more likely to have decreased muscle mass, suppressed immune system, increased risk for injury, and chronic fatigue. Conversely, athletes who routinely exceed protein needs may have increased risk for dehydration, increased body fat stores, calcium loss, and an unbalanced diet that is

Nutrition, Pharmacology, and Psychology in Sports

  TABLE 8A-5  Protein Requirements for Various Types of Athletes

Protein Requirements (g/pound body weight/day)

Type of Athlete Recreational athlete Competitive athlete Athlete building mass Teenage athlete Athlete restricting intake Maximal usable amount by athletes in weight-class sports (e.g., crew, wrestling)

405

National Collegiate Athletic Association permitted institutions to provide only non–muscle-building supplements to student athletes, so meeting protein needs through food is extremely important.33

Fat Requirements

0.5-0.75 0.6-0.9 0.7-0.9 0.9-1.0 0.7-1.0 1.0

often deficient in carbohydrate. Protein requirements are outlined in Table 8A-5.31 Strength-training athletes have traditionally emphasized protein, sometimes to the exclusion of other essential nutrients, and more protein is initially required to support an increase in muscle mass, but not in excess of 1 g of protein per pound of body weight. Surprisingly, endurance athletes also require more protein in the early stages of training to increase aerobic enzymes in the muscle, to form red blood cells and myoglobin, and to replace protein stores that are oxidized during exercise. Recommendations for protein intake can range from 0.55 to 0.77 g/pound body weight.32 Food sources of protein will ideally compose most of the protein in the diet, rather than protein powders or amino acid supplements. Table 8A-6 lists various food sources of protein. Interestingly, an anabolic effect can be seen with as little as 12 to 15 g of protein (6 to 8 g of essential amino acids) consumed before and after strength training, negating the need for megadoses of protein. Many athletes are drawn to protein powders or amino acid supplements as a means to increase protein intake. No studies have shown benefits of amino acid supplements as an ergogenic aid, and gastrointestinal distress can be a problem.30 Protein powders can be fairly costly; they require mixing, and palatability can be a problem. If an athlete wants to use a protein powder, nonfat dry milk powder is inexpensive, shelf stable, and tasteless, and it is an excellent source of calcium as milk. In August 2000, the

  TABLE 8A-6  Protein Content of Select Foods Food

Amount

Protein (g)

Chicken breast Chicken thigh Cod Hamburger Steak Pork chop Egg Soy burger Nuts Peanut butter Cheese Refried beans Milk Yogurt Protein powders Nonfat dry milk powder Amino acid pills

3 oz 3 oz 3 oz 3 oz 3 oz 3 oz 1 1 ¼ cup 2 tbsp 1 slice ½ cup 8 oz 8 oz per scoop ¼ cup 1 serving

21 21 21 21 21 21 7 15-18 10 8 7 7 8 9-11 32-45 8 10

Fat is an energy substrate for low-intensity, longer-­duration exercise. Fat supplies a concentrated calorie source to provide energy for the athlete. A diet that is too low in fat may limit performance by inhibiting intramuscular triglyceride storage, resulting in earlier fatigue during exercise.34 Excess fat intake can increase fat stores and cause gastrointestinal discomfort before exercise. Conversely, inadequate fat intake can decrease serum testosterone concentration, decreasing muscle mass.35 The recommendation for fat intake is weight (pounds) × 0.45 = number of grams of fat per day.36 Sources of fat are listed in Table 8A-7.

Micronutrient Requirements Athletes often inquire about the need for vitamin and mineral supplementation. The needs of an athlete may be slightly higher than those of someone who does not exercise regularly, but these requirements can be met through a multivitamin and mineral supplement, not a megadose tablet. In 1998, the Food and Nutrition Board of the National Academy of Sciences established Dietary Reference Intakes for select vitamins and minerals.37 These recommendations are listed in Box 8A-1. For athletes who need to boost iron stores through food or supplemental iron, it is important to note that some foods and nutrients interfere with iron absorption: • Phytates (bran, whole grains) • Oxalates (spinach, beer, nuts) • Polyphenols (coffee, tea) • Excess intake of calcium and magnesium

  TABLE 8A-7  Fat Content of Select Foods Food Item

Amount

Fat Content (g)

Olive oil* Soft margarine* Mayonnaise Salad dressings   Italian*   Ranch   Blue cheese Nuts* Peanut butter* Cream cheese Bacon Chips French fries Ice cream Prime rib Burger Big Mac Drumstick, fried Wing, fried

1 tbsp 1 tbsp 1 tbsp

14 11 11

1 tbsp 1 tbsp 1 tbsp ¼ cup 2 tbsp 2 tbsp 2 slices 1 ounce small 1 scoop 3 ounces 1 regular 1 1 1

7 10 8 32 16 11 6 10 12 20 23 10 32 11 15

*Heart-healthier choice.

406

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 8A-1 D  ietary Reference Intakes of ­Micronutrients for Adults Vitamin A: 700-900 μg Vitamin C: 75-90 mg; tolerable upper limit: 2000 mg Vitamin D: 10-15 μg Vitamin E: 15 mg Vitamin K: 90-120 μg Thiamin: 1.1-1.2 mg Riboflavin: 1.1-1.3 mg Niacin: 14-16 mg Vitamin B6: 1.3-1.7 mg Folate: 400 μg Vitamin B12: 2.4 μg Zinc: 8-11 mg Phosphorus: 700 mg Chromium: 25-35 μg Magnesium: 320-420 mg Iron Women: 15 mg Men: 10 mg Calcium, men and women 9-13 yr: 1300 mg 19-50 yr: 1000 mg >51 yr: 1200 mg Tolerable upper limit: 2500 mg

The following factors help to enhance iron absorption: • Heme iron (animal source) • Low body stores of iron • Consuming a food high in vitamin C with meals • Waiting 1 hour before or after meals to drink coffee or tea

BONE HEALTH Preventing stress fractures should be a key component of the athlete’s education. Many athletes think that calcium supplementation alone will ensure healthy bones. Even with this assumption, most athletes are not meeting calcium needs on a daily basis. To optimize bone health, the following factors should be considered: • Optimal protein intake: at least 0.5 g/pound body weight • Daily consumption of calcium-rich foods and calcium supplements to attain an intake of at least 1200 mg/day • Vitamin D intake of 800 to 1000 IU from food plus supplements • Ingestion of supplements with meals • Vitamin K from leafy greens as well as part of a calcium and vitamin D supplement

WEIGHT MANAGEMENT ISSUES Every athlete has to face issues surrounding weight at some point in his or her career. The question is whether body fat or weight is the most important variable, and the answer depends on the type of sport. Weight standards

Box 8A-2 Techniques for Making Weight

• Maintain weight within 2-5 pounds of weight class during training.

• Seven to 10 days before a competition, the athlete can

try the following: • Decrease sodium intake through added salt and salty snacks • Decrease high-fiber foods, which can cause the body to retain fluid

are used more in certain sports than in others, such as the ­following: • Sports based on skill (archery, bowling)—weight is generally not an issue • Sports with weight division (crew, wrestling, jockeys) • Sports with low body fat for optimal performance (distance runners) • Sports with appearance and aesthetic criteria (gymnastics, figure skating) Many athletes strive to attain the lowest body fat possible, jeopardizing health and performance in the process. Disordered eating behavior to make weight can decrease the metabolic rate, making the body less efficient at burning fat. Rapid weight loss reduces the plasma volume and blood distribution to active tissues and may adversely affect thermoregulation, which can impair performance. Because athletes will do whatever it takes to make weight, guidelines are presented here for making weight, for weight loss, and for weight gain. An athlete who is interested in losing or gaining weight should seek the expertise of a sports nutritionist who can customize a program to allow the athlete to meet his or her goals. Box 8A-2 provides guidelines for making weight in weight-class sports. Because athletes often try fad diets for weight loss that are often restricted in energy intake, the end result can be a tired, poorly nourished athlete who is overly hungry and does not exercise or recover efficiently. Box 8A-3 offers some commonsense guidelines for weight loss that can help the athlete to achieve his or her goals without sacrificing energy. Achieving success with weight management requires a minimal investment of 3 months to change the underlying habits that determine food choices. Some athletes seem unable to gain weight. The problem is often an erratic eating pattern and the assumption that eating as much as one can at one time will rectify the problem. Consistency with number of meals per day is the most important variable for weight gain. Box 8A-4 outlines techniques for weight gain.

SUPPLEMENT USE Athletes have always been in quest of products that will make them stronger, leaner, and faster. Genetics, talent, training, and optimal fuel are certainly the most important variables, but that will not stop athletes from looking for the edge. The major issues of the moment are the

Nutrition, Pharmacology, and Psychology in Sports

407

Box 8A-3 Weight Loss Guidelines

• Healthy weight loss is 1-2 pounds per week. • Smaller, more frequent meals are preferred to larger, spo-

radic meals. empty calories from beverages, high-calorie snacks, condiments, and dressings. • Make eating purposeful—know what is being consumed, and pay attention to the portion. • Eat foods with fiber to promote a feeling of fullness. They also take longer to eat. • Eat foods that promote satiety (foods with some fat enable the body to feel fuller for longer).

• Decrease

Meal Plan for Weight Loss Breakfast A sandwich on whole grain bread or English muffin A scrambled egg Two thin slices of ham or Canadian bacon Lettuce, tomato Very thin spread of mayonnaise OR A yogurt (150-200 calories) with granola added A cereal bowl of fruit salad A slice of whole grain toast with a thin spread of peanut butter OR Breakfast burritos: 3 eggs, scrambled with ¼ cup cheddar cheese spread into 2 corn tortillas with salsa A banana Lunch Turkey and provolone submarine sandwich: 6-inch hoa­ gie roll, mustard, ketchup, thin spread of mayonnaise (light), ¼ pound of turkey and 2 slices of cheese A small salad with light dressing Dish of cut-up fruit OR Grilled chicken sandwich: grilled chicken breast, BBQ sauce, on a whole grain bun

­ idespread availability of supplements, their lack of purity, w and the belief that more must be better. Health care providers must be aware of the products athletes take and why they are taken. These are some important facts about supplements: • Supplements are not “one size fits all.” • The terms natural and safe are not synonymous. • Supplements may interfere with prescribed medications. • Athletes may not willingly disclose information about supplements being used. • Supplement use is not necessarily according to package instructions. • Supplements may be sports specific. • Supplements do not confer the same effects in everyone who takes them.

Bowl of vegetable soup OR Tuna salad sandwich: 6-oz can of tuna (water-packed) with light mayonnaise, relish spread on 2 English muffin halves, top with 2 slices of cheese, broil until cheese melts Sliced tomatoes OR Pasta: 3 cups with sauce A salad with a grilled chicken breast, light on the dressing (2-3 tbsp at most) Afternoon Snack An 8-oz yogurt with granola OR A small bowl of cereal OR A smoothie with 8 oz yogurt, 8 oz skim milk, one cup of frozen or fresh berries or a small banana, blend together Dinner Meat of some kind: 6-8 ounces cooked weight (steak, pork loin, venison, fish, chicken) 2 cups of pasta or rice or potatoes (about one third of the plate) 2 cups of vegetables (rest of the plate), steamed, grilled Evening Snack Popcorn: microwave snack-size bag OR Yogurt with granola OR Small bowl of cereal, like Raisin Bran, Kashi, ­ Nutri-Grain

• The efficacy of any supplement depends on the underlying diet, hydration status, and training level. • The issue of stacking has become a problem with overthe-counter supplements. • The Internet has increased the availability and accessibility of supplements. The Dietary Supplement and Health Education Act of 1994 details labeling guidelines that were put into effect in March 1999. Labels must have a supplement facts panel, detailing contents and amount, and an ingredient list. This act also allows companies to make claims about their products that can cause consumer confusion. The Suggested Readings at the end of this chapter list helpful resources for the health professional and the athlete. Whether at the recreational, high school, collegiate, professional, or Olympic games level, athletes have turned

408

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 8A-4 Techniques for Weight Gain

• Aim for 500-1000 additional calories daily • Eat on a schedule, more frequent meals and snacks • The goal is ½ pound of weight gain per week • Foods and beverages need to contain calories (juice

stead of water) • Choose higher-calorie items (nuts instead of pretzels) • Eating needs to be a priority • Aim for 500-1000 additional calories daily

in-

The Plan Breakfast Orange juice: large glass (12 oz) Cereal: a large bowl of Cheerios with granola added, skim milk A bagel or 2 slices of wheat toast with 1 tbsp peanut butter on each piece OR 2 pieces of fruit A glass of juice (12 oz) A breakfast sandwich of an English muffin 2 scrambled eggs 2 pieces of ham 2 slices of cheese OR A smoothie made with 1 scoop of protein powder, 8 oz yogurt, 12 oz skim milk, 1 cup of frozen fruit A bagel with peanut butter After-Workout Food Choices (within 15 Minutes) A peanut butter and jelly sandwich and a sports drink A container of yogurt with granola A small bag of trail mix and Gatorade A bar such as Balance, Power, or Clif Bar and sports drink This helps you to recover as well as increase weight. Lunch Submarine sandwich (12 inch): chicken, tuna, steak, ham and cheese Baked chips and lemonade or juice to drink OR A cheeseburger or a grilled chicken sandwich Small order of fries Shake

to nutritional supplements as ergogenic aids or performance enhancers. Supplements are advertised to do anything from alter body composition to boost energy, improve memory, and eradicate pain. Many products are innocuous, but some can be harmful and ergolytic, or performance detracting. The most commonly used supplements are listed in Box 8A-5. With all these products, the major issues are the safety, purity, efficacy, cost, and in certain cases, legality of the supplement. Creatine is a cell volumizer that results in an increase in muscle size and volume, partly owing to increased fluid retention within the muscle. The results with creatine have not been consistent, and the added

OR Grilled chicken salad with a baked potato and juice to drink OR A wrap with a salad and a shake If you make food: Bagel sandwich with turkey, cheese, and fruit OR Pasta with sauce, 2 pieces of chicken, and a salad OR An omelet with 3 eggs, cheese, vegetables Hash browns 2 slices of toast Afternoon Snack Banana with peanut butter (2 tbsp) OR Trail mix with cereal, nuts, and dried fruit (1 cup or 2 handfuls) OR Cheese (2 oz) and crackers (12) OR Large bowl of cereal Every time you eat, have something to drink that contains calories, such as juice, low-fat milk, lemonade, or sports drink Dinner Protein always (half of the plate): steak, chicken, fish, pork, turkey Rice, pasta, potato, or corn (half of the plate) And then separately: a salad, cooked vegetables, or fruit Seconds should come from carbohydrate-containing foods and protein before vegetables To drink at dinner: milk, juice, or lemonade Snack Later at Night To add some extra calories: A large bowl of ice cream or frozen yogurt OR A smoothie OR A protein shake with ice cream added OR A sandwich

weight with a loading regimen may be a performance detractor more than an enhancer. In addition, some athletes may be hyper-responders to creatine, whereas others notice no effects whatsoever. Increasing protein intake may be as effective as creatine supplementation and is a source of calories as well. Another popular product is nitric oxide stimulators, also known as arginine α-ketoglutarate. Arginine is a precursor to nitric oxide, and claims for the product suggest a pumped-up feeling and increased muscle mass, which have not been observed in studies. Basically this product is a vasodilator but has no anabolic effects and is very costly.

Nutrition, Pharmacology, and Psychology in Sports

Box 8A-5 Most Commonly Used Supplements Muscle-building supplements Creatine Protein powders Amino acid supplements Prohormones: dehydroepiandrosterone (DHEA), androstenedione, norandrostenediol, Tribulus ­terrestris (Tribestan), yohimbe Nitric oxide stimulators Thermogenic products or “fat burners” Ephedra Synephrine Caffeine Chitosan l-Carnitine Pyruvate Hydroxycitrate Energy boosters Ginseng Carbohydrate supplements Energy drinks Vitamin-mineral supplements Products for pain management Glucosamine, chondroitin

Athletes are always looking for the easy way to decrease weight and body fat and often turn to fat burners. Ephedra, or ma huang, is a potent central nervous system stimulant that can cause elevated heart, blood pressure, and respiration rates in addition to insomnia, jitteriness, chest pain, and stroke. The safe level is 24 mg in a 24-hour period, and many of the products exceed this dosage.38 Ephedra is a banned substance in the National Collegiate Athletic Association (NCAA), the International Olympic Committee (IOC), and most recently, the National Football League (NFL) and Major League Baseball (MLB). Synephrine, or citrus aurantium or bitter orange, is another product found in fat burners that is also a central nervous system ­stimulant but is not effective at promoting body fat loss. It is also considered to be a banned substance in the NCAA, NFL, MLB, and IOC. Protein powders are costly and in many cases provide more protein to the body than can be used efficiently. An athlete who is adamant about taking a protein product would be well advised to try nonfat dry milk powder or an Instant Breakfast product as an excellent source of protein and calcium at a low cost. Amino acid supplements are an inefficient, costly fuel source for the body that can cause gastrointestinal distress. Prohormones such as dehydroepiandrosterone (DHEA), androstenedione, norandrostenediol, and Tribulus terrestris are substances banned by the NCAA, the NFL, and the United States Olympic Committee. Yohimbe may cause kidney damage, and chromium does not confer an anabolic effect. Some products have a laxative or diuretic effect, such as Dieter’s Tea, hydroxycitrate.

409

Caffeine in large doses (e.g., Taroxotone, guarana, mate, or kola nut) can cause nervousness, insomnia, anxiety, restlessness, and digestive distress. Although 100 to 200 mg of caffeine consumed 30 to 60 minutes before exercise may spare muscle glycogen, mobilize free fatty acids for fuel use, and subsequently delay fatigue, the greatest effect is seen with caffeine in pill form rather than as coffee or an energy drink. There are caffeine-containing energy drinks that exceed 300 mg in an 8-ounce can. In that type of dose, the athlete would increase the risk for jitteriness and nervousness, potentially impairing performance. Products for weight loss, such as pyruvate and hydroxycitrate, have been used clinically in doses far exceeding those of products sold over the counter. Conjugated linoleic acid, sold as Tonalin, has been touted as a weightloss agent, to help burn body fat, but the studies have been small in sample size and number. Chitosan and quercetin are ineffective weight loss agents. Although glucosamine and chondroitin may help with the pain, swelling, and tenderness associated with osteoarthritis, there may be a 2-month trial period before results are seen. Glucosamine may cause indigestion and nausea and may worsen insulin resistance in people with diabetes.39,40 Chondroitin sulfate may also cause indigestion and nausea and is contraindicated for hemophiliacs or individuals receiving blood-thinning medications or aspirin therapy because it is similar in molecular structure to heparin.41 Collagen hydrolysate may be an option to improve joint mobility. γ-Hydroxybutyrate and γ-butyrolactone are extremely dangerous; they induce coma, seizures, and death and should never be used. Health care providers need to realize that athletes will experiment with products and should follow these recommendations: • Recommend that athletes who are taking over-the­counter or prescription medications check with their health care provider before taking any supplement. • Inquire as to what supplements are used. • Recommend that athletes who notice any side effects as result of supplement use stop taking the product immediately. • Inquire as to the dose and frequency. • Ask to see the product label. • Document the information in the chart. • Remind athletes that supplements are used to enhance, not replace eating.

SUMMARY Being optimally nourished is an essential part of training and conditioning. The goals of sports nutrition should be reinforced at any opportunity, from the playing field to the physician’s office. The role of nutrition in maximizing performance, enhancing recovery, preventing injury, and supporting rehabilitation can only help the athlete perform at his or her peak as well as improve the quality of life. The time spent asking athletes about their nutritional habits will help them to realize their potential and also keep them in their sport performing at their optimum and potentially extending their athletic career.

410

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S U G G E S T E D

R E A D I N G S

PDR for Herbal Medicines. Montvale NJ, Medical Economics, 2000. PDR for Nutritional Supplements. Montvale NJ, Medical Economics, 2001. Sports Nutrition: A Practice Manual for Professionals, 4th ed. Chicago, American Dietetic Association, 2006. Sarubin A: Health Professional’s Guide to Popular Dietary Supplements, 2nd ed. Chicago, American Dietetic Association, 2003. Williams MH: Ergogenics Edge: Pushing the Limits of Human Performance. Champaign, Ill, Human Kinetics, 1998.

W E B

S I T E S

International Bibliographic Information on Dietary Supplements (IBIDS), http:// odp.od.nih.gov/ods/databases/ibids.html National Center for Drug Free Sport, (816) 474-8655, http://info@drugfreesport. com Office of Dietary Supplements (National Institutes of Health), http://dietary­supplements.info.nih.gov Product Evaluation: Consumer Labs, http://www.consumerlabs.com

R eferences Please see www.expertconsult.com

Dietary Supplements: An Advertising Guide for Industry, http://www.ftc.gov/bcp/ guides/guides.htm Dietary Supplements Quality Initiative, http://www.dsqi.org Gatorade Sports Science Institute, http://www.gssiweb.com

S ect i o n B

Pharmacology 1. Sports Pharmacology: ­Ergogenic Drugs in Sports Edward R. McDevitt and David E. Brown

There is no room for second place. There is only one place in my game, and that’s first place. Vince Lombardi1 Athletes are often willing to cheat to win. The desire to win is not limited to the elite athletes of the world. Our society places a high value on winning, and athletes who perform well are rewarded in many ways in our society. Yet the abuse of drugs to improve performance is increasing not only in elite athletes but also among the general population. Clinicians need to be aware of the effects of these drugs on their patients. Athletes are not looking to their physicians for advice and expertise. Athletes are searching for any means to improve their performance. The word ergogenic is derived from the Greek éaargon, to work, and gennan, to produce, and means increasing the ability to do work. Athletes use ergogenic drugs to help them perform at a higher level. Athletes have looked for ways to improve their performance in sports for more than 3000 years. Athletes at the first Greek Olympics ate substances that they thought would give them a competitive advantage, including dried figs, mushrooms, large amounts of meat, and strychnine.2 Strychnine, known from ancient times as a poison, is extracted from the seeds of the plant nux vomica and in small amounts is a central nervous system stimulant and potentially an ergogenic substance. In slightly larger amounts, however, ­strychnine

is deadly. This athletic risk-taking behavior is a critical point for team physicians to understand. The ancient Greek athletes were willing to risk taking a potentially deadly substance with the hope that the substance would help them win. Athletes today continue to take potentially deadly substances that perhaps can help them compete. If it seems far-fetched that athletes would be willing to take a poison like strychnine, athletes in the 1904 Olympics, as well as athletes in the 1992 Barcelona Olympics, used strychnine as an ergogenic drug. Athletes continue to take “poisons” in their desire to win. In the short time since the last edition of this book, the sports world has been rocked by accusations of illegal drug use by athletes in a variety of sports fields, including track, baseball, football, cycling, wrestling, and boxing. Fainaru-Wada and Williams’ book, Game of Shadows, documents many of the world’s greatest athletes using a variety of ergogenic aids to push themselves, no matter what the cost, to perform at the highest level possible.3 The book also documents the tremendous rewards and dangers that await athletes, coaches, and scientists who use ergogenic drugs to increase athletic ­performance. Mirkin4 conducted a poll of Olympic athletes in the 1980s that asked the question: If you could take a pill that would guarantee you the Olympic gold medal, but would kill you within a year, would you take it? The poll revealed that more than 50% of the athletes would take the pill.

Nutrition, Pharmacology, and Psychology in Sports

Goldman conducted a modified poll in the 1990s and asked 198 aspiring Olympians, all elite athletes, two questions6: 1. If you were offered a banned performance-enhancing substance that guaranteed that you would win an Olympic medal, and you could not be caught, would you take it? Of 198 athletes, 195 answered yes. 2. Would you take a banned performance-enhancing drug with a guarantee that you will not be caught; you will win every competition for the next 5 years, but then die from the side effects of the substance? More than 50% of the athletes said they would take the substance. The results of these two polls reveal that athletes, in high percentages, may be willing to use substances that are potentially fatal if they think those substances will help them win. Athletes want to win now and may not worry about the consequences of substance use in the future. Athletes and individuals entrusted with the care of athletes, including coaches, athletic trainers, therapists, physicians, parents, and spouses, have a difficult dilemma: Are they willing to do “whatever it takes” to compete at an elite level? Are they willing to use poison if it may help the athlete win? One of the first team physicians to address this dilemma was John Ziegler, an American weightlifting team physician in the 1950s. Concerned about the tremendous gains made by the Soviet Bloc nations in Olympic weightlifting, Ziegler was determined to discover the secret of the Soviets’ athletic success.7 A Soviet physician told Ziegler that the impressive gains made by the Soviets were not due to changes in training or diet but were due to a powerful new drug being injected into the Soviet athletes’ bodies, the male hormone testosterone. Ziegler was impressed with the strength-building effects of testosterone, but he was wary of potential problems with testosterone injections. There were rumors that the Soviet weightlifters used catheters to urinate because the injectable testosterone, although responsible for increases in their strength, had caused such enlargement of the prostate that normal urinary flow was possible only by catheter. Ziegler, as a dedicated team physician, was inspired to do something to help his American athletes compete.7 He wanted to manufacture a better, safer drug to give the American athletes an advantage in sport without the worrisome side effects of testosterone. Working with Ciba Pharmaceuticals, Ziegler helped develop a drug that was designed to have anabolic (muscle building) properties without the androgenic (male-like) side effects. The drug developed, methandrostenolone (Dianabol), was a great improvement over testosterone and is used widely by athletes today. As the developer of methandrostenolone, Ziegler was an early ardent advocate for its use. He later regretted the important role he played in the history of ergogenic drugs. Ziegler thought he was doing what was best for his American athletes, and methandrostenolone was a superior ergogenic drug when compared with testosterone. Responsible for gains in strength, it was realized later that methandrostenolone had adverse side effects related to its potent androgenic properties. Responsible for encouraging his American athletes to use drugs that had many unintended side effects, Ziegler was just the first

411

of many well-meaning scientists who have attempted to develop a pure anabolic agent, without androgenic features. This perfect anabolic drug has not yet been found. With androgenic properties come potentially deadly side effects. This situation emphasizes the importance of thorough, long-term testing of drugs before recommending their use by athletes. In their quest for improvement, today’s athletes continue to use drugs that have not yet been proved to be efficacious and safe. To underscore the important point that derivatives of testosterone have anabolic and androgenic properties, we categorize all ergogenic derivatives of testosterone in this chapter as anabolic-androgenic steroids (AASs). There are hundreds of ergogenic variants of testosterone now being used by athletes.

HISTORY OF TESTOSTERONE One of the first scientists to promote the ergogenic properties of testosterone was the prominent French physiologist Brown-Séquard.8 Although largely remembered in the medical world today for his description of a spinal cord syndrome, Brown-Séquard played an important role in the history of AASs. In June 1889, at the Société de Biologie meeting in Paris, the 72-year-old Brown-Séquard announced he had found a rejuvenating compound that had reversed many of his physical and mental ailments related to aging. He reported that he had injected himself with a liquid extract made from the testicles of dogs and guinea pigs and that these injections had dramatically increased his strength, improved his mental acumen, relieved his constipation, and increased the arc of his urinary stream. Although most of his colleagues greeted Brown-Séquard’s findings with disdain, he had made an important discovery with protean implications for the medical and athletic worlds. Brown-Séquard’s discovery of the positive properties of the testicular extract was based on the earlier work of Berthold. Berthold proposed in 1849 that the implantation of testicles in the abdomen of roosters reversed the effects of castration.9 There appeared to be some ingredient in testicular tissue that made male animals male. BrownSéquard realized through the experiments he performed on himself that internal secretions of organs could act as physiologic regulators. His postulation of physiologic regulators predated the discovery of hormones in 1905. Scientists continued the work of Berthold and Brown-Séquard in an attempt to isolate the active ingredient in testicular function. Several other scientists made important discoveries contributing to a better understanding of the properties of the testicular hormone. In the 1920s, Zoth and Pregl performed experiments to determine whether testicular extracts could increase muscle strength. They injected themselves with extracts of bull testicles and found that they had increased their finger strength.10 In an attempt to identify the strength-building compound in testicular tissue, in 1931 the German scientist Butenandt isolated 15 mg of a pure substance, androsterone, from the extract of 15,000 L of urine obtained from policemen.11 Studies with androsterone did not reproduce fully all of the anabolicandrogenic properties of testicular tissue. A Dutch scientist, David, and his colleagues, first isolated ­ testosterone

412

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

in 1935.12 German scientists soon published an article describing the synthesis of testosterone from cholesterol.13 Physicians, as well as veterinarians, soon started using testosterone and its intermediaries clinically.14 Testosterone first was used medically in humans to help chronically ill patients recover muscle mass.15 Testosterone was found to have positive effects on almost every organ of the body. Sixty years after Brown-Séquard, testosterone was promoted again as a rejuvenation compound. Boje, in 1939, suggested that male sex hormones might improve athletic performance.15 In 1941, a racehorse dramatically improved his performance after receiving injections of testosterone.16 Human competitors started experimenting with testosterone looking for an advantage in competition.17 The first systematic use of testosterone in the sporting world took place in the 1950s by the Soviet Olympic athletes. The use of testosterone and derivatives such as methandrostenolone spread rapidly to athletes on both sides of the Iron Curtain. In the 1960s, use of AASs became prevalent.18 During that period, there were no sanctions against the use of ergogenic drugs. Many athletes believed that they had to use ergogenic drugs to keep up with their competitors. In the 1960s, there was extensive use of AASs by athletes in the strength sports (shot put, hammer throw, and javelin) in track and field.19 By the 1968 Olympics, the use of AASs was ubiquitous. New drugs claiming to be more anabolic were being marketed openly. Male athletes from other sports soon began taking AASs. American professional football players were among the first groups to try the new muscle-building drugs in large numbers. Many of the early professional football players were given AASs without their knowledge.20 Similar so-called vitamins, which were later identified as AASs, were given to college football players in the 1970s. Even if some athletes were duped into taking drugs, however, many athletes would have taken the AASs willingly. The use of AASs escalated in the 1970s and 1980s. It was estimated that 40% to 90% of professional football players were using AASs.21 College football players soon followed the lead of the professionals.21 When a Notre Dame football player claimed in a Sports Illustrated article in 1990 that the use of AASs was commonplace among his teammates, his claim was denied vehemently by Notre Dame coaches, players, and officials. The use of AASs was not something that would be admitted openly. The positive effects experienced by the male track athletes were noted not only by male athletes in other sports but also by female athletes. Women taking AASs experience anabolic effects that can increase muscular strength as well as androgenic or masculinization effects, such as a deepening of the voice, increase in body hair, loss of breast tissue, and enlargement of the clitoris. Masculinization of female athletes from the Soviet Bloc nations was so dramatic that sex tests, or chromosomal analyses, were initiated at the 1967 European championships. There was concern that male athletes were masquerading as female athletes; these athletes were not males but female athletes taking male hormones. In the trial of former East German team physician Kipke, his testimony revealed a systematic prescription of supporting means to young East German female athletes. The supporting means turned out to be Oral-Turinabol

(dehydrochloromethyltestosterone), an AAS manufactured in East Germany.22 Kipke contended that he was just following orders, and he stated he was unaware of any potential side effects of the AAS on the young women other than a deepening of their voices. He admitted administering the drugs for decades to the East German swim team. Convicted by a German court and sentenced to 15 months in jail, Kipke was the 26th East German sports official to be convicted on charges of giving AASs to East German athletes in a wide variety of sports. Many female athletes from the former East Germany have complained that they suffer from a variety of medical maladies that they believe are secondary to their being given AASs without their knowledge. Steven Ungerleider’s Faust’s Gold: Inside the East German Doping Machine is a book that documents the East German national directive to systematically provide the East German male and female athletes with anabolic steroids. The athletes were told they were taking “vitamins.”22 Women in Western nations have been suspected of using ergogenic drugs. Florence Griffith Joyner became one of the most famous athletes in the world when she came out of semiretirement to become a tremendous success at the 1988 Olympic Games. Joyner had progressed from a good runner to a world champion whose record in the 100-m run was unsurpassed for 10 years. Her muscular physique, combined with her newfound athletic success, led to speculation that her gains were due to performanceenhancing drugs—a charge she vehemently denied.23 Joyner claimed her changes were due to relentless training supervised by her husband, Olympic athlete Al Joyner, and not to ergogenic drugs. When Joyner died in her sleep at the age of 38 years, speculation was raised that her death may have been related to side effects from taking ergogenic drugs. No proof exists that Joyner used ergogenic drugs or that her death was related to use of drugs. Doubts similarly were raised about the tremendous gains made by Irish swimmer, Michelle Smith. In 1993, Smith had not been ranked among the top 25 female swimmers in any stroke.24 At the 1996 Olympic games, at age 26 (considered old in women’s swimming), Smith won four Olympic medals. Smith claimed her amazing progression from a second-tier swimmer to Olympic champion was a result of intensive training and proper nutrition and not to ergogenic drugs. Her husband, Erik de Bruin, monitored her training. A two-time Olympian from the Netherlands, de Bruin competed in the shot put and discus, failed a drug test in 1993, and was suspended for 4 years. No evidence existed, however, that Smith used performance-enhancing drugs. Many young athletes apparently think that use of ergogenic drugs is a risk worth taking. Buckley’s study published in 1988 revealed that 6.6% of 12th graders had used anabolic steroids.25 Another disturbing finding was that 30% of users were nonathletes. These nonathletes were using ergogenic drugs to improve their appearance rather than their athletic performance. Evidence reveals that ergogenic drug use is starting as early as middle school age. The National Institute of Drug Abuse 2006 Monitoring the Future study, which surveyed drug abuse among adolescents in middle school and high school, found evidence that 2.7% of high school male students had taken AASs at least once in their lives.28 Ergogenic drug use is

Nutrition, Pharmacology, and Psychology in Sports

not limited to young men. Yesalis found that the percentage of 14- to 18-year-old girls using anabolic steroids has almost doubled in 7 years.26 Many women consider ergogenic drugs an effective means to obtain athletic scholarships to college.27 When a Danish cyclist died at the 1960 Olympics after ingesting a combination of nicotinic acid and amphetamines, there was concern that drug use by athletes not only was unfair but also dangerous. The authorities decided to do something about the problem with ergogenic drugs. One of the first steps in the process was to define the problem. In 1963, the Council of Europe established a definition of doping as “the administration or use of substances in any form alien to the body or of physiological substances in abnormal amounts and with abnormal methods by healthy persons with the exclusive aim of attaining an artificial and unfair increase in performance in competition.”28 The origin of the word doping was attributed to the Dutch word dop. Dop was a narcotic mixture of opium used to stimulate racing horses. The Council of Europe decided that it was imperative to ban these drugs and developed a list of doping drugs used by athletes. In 1967, the International Olympic Committee published a list of banned drugs and a medical code to “protect the health of athletes and to ensure respect for the ethical concepts implicit in Fair Play, the Olympic Spirit, and medical practice.”29 Sports governing bodies worldwide adopted rules regarding the use of ergogenic drugs. The World Anti-Doping Agency (WADA) was established on November 10, 1999 in Lausanne to promote and coordinate the fight against doping in sport internationally. WADA was set up as a foundation under the initiative of the International Olympic Committee with the support and participation of intergovernmental organizations, governments, public authorities, and other public and private bodies fighting against doping in sport. WADA projects include the development of strategies to tackle organized doping schemes and trafficking through enhanced cooperation between government agencies and the sports movement. Another key project that is being developed is to track biologic data from athletes over time in order to identify abnormal biologic profiles. Pilot projects have been developed to study the technical, scientific, and legal feasibility of the concept. WADA publishes a listing each year of banned substances. This list is invaluable for both athletes and medical providers.30 One of the greatest controversies in sport today involves the accuracy of drug testing for ergogenic drugs. Developing tests to determine who was using banned drugs was problematic. Few laboratories were able to perform the necessary testing. The cost of each test was more than $100, which made the testing of large numbers of athletes prohibitive. The fact that new AASs were being developed and a variety of other ergogenic and masking agents were being used by athletes required that laboratories develop new specific screening tests. Effective laboratory testing to detect ergogenic drugs was not readily available until the 1976 Olympics. Nineteen athletes were banned for using illegal drugs at the 1983 Pan-American games, and many athletes left voluntarily when faced with mandatory testing. When athletes knew ahead of time that there was to be testing at the conclusion of an event, the numbers of athletes detected using banned ergogenic drugs

413

was small, usually less than 2%.29 In 1984 and 1985, the U.S. Olympic committee conducted unannounced testing of Olympic-level athletes, however, and 50% of athletes tested positive for banned ergogenic drugs.31 Apparently athletes had learned to taper off AASs before major events at which announced drug testing was going to be done, or they used other drugs that masked their ergogenic drug use. Testing for ergogenic drugs has become a major industry and a major source of disagreement between athletes and sports organizations. It is critically important for those supervising drug testing to follow to the letter the rules of “chain of custody.” It is essential for the accuracy of the testing process that the sample obtained was provided by the athlete, not a surrogate. The sample must be sealed securely and signatures obtained by the athlete and the test observer attesting to the accuracy of the sample. The chain of custody must be maintained and continued at the laboratory site so that there can be no question of tampering. The sample must be treated with precision and accuracy at the testing laboratory. These painstaking steps are essential for the protection of the athlete and the testing process because the consequences of positive testing for an ergogenic drug are significant. Athletes taking ergogenic drugs put themselves at risk not only for serious sports sanctions but also for potential jail terms. The Anti–Drug Abuse Act of 1988 prohibited the distribution of AASs for any use other than treatment of a disease. It is illegal for physicians to prescribe AASs to enhance athletes’ performance. The Anabolic Steroids Control Act of 1990 placed all AASs in Schedule III of the Code of Federal Regulations Schedules of Controlled Substances. This law places AASs in the same legal class as amphetamines, methamphetamines, opium, and morphine. Simple possession of any Schedule III substance is a federal offense punishable by 1 year in prison or a minimum fine of $1000, or both. Possession by a person with a previous conviction for certain offenses, including any drug or narcotic crimes, requires imprisonment of 15 days to 2 years and a minimum fine of $2500. Individuals with two or more such previous convictions face imprisonment of 90 days to 3 years and a minimum fine of $5000. Selling AASs or possessing them with intent to sell is a federal felony. This crime is punishable by 5 years in prison or a $250,000 fine, or both.32 Athletes take risks to obtain and use banned drugs such as AASs because they believe the drugs will help them compete, and the athletes do not think they will be caught. Ziegler’s crucial role in the development of methandrostenolone was a result of his desire as a team physician to help his team compete against the Soviet athletes. If the competition was using drugs, some athletes, coaches, administrators, and team physicians thought that they had to use drugs as well or fall behind. When mainstream medical opinion was sought in the midst of the controversy, the early negative opinions from physicians regarding ergogenic drugs divided athletes from their physicians. The official medical response was that AASs did not work and that the apparent benefits were largely due to a placebo effect. The potential for side effects was also emphasized. Physicians were concerned that athletes using AASs would be destined for a life of steroid-related complications, which might threaten or end the athletes’

414

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

life. Athletes were skeptical of this advice. They could see that other athletes using AASs were making gains and did not appear to have serious physical problems. The official medical world lost credibility with athletes, and athletes turned to other sources, not only to get their drugs but also to get information regarding performance-­enhancing drugs. Magazines such as Muscle and Fitness and Muscle Media became valuable sources of information concerning performance-enhancing medicine. The Underground Steroid Users Handbook33 became a trusted source of information concerning ergogenic drugs. This book promised to tell the truth about AASs and belittled the negative views of physicians regarding AASs. The author stated that steroids do work. He stated that although there are definite side effects associated with their use, the dangers of AAS are exaggerated by physicians. The author made a valid point that most of the studies regarding AAS were poorly done. The studies did not use elite athletes as subjects and did not test the supraphysiologic doses used by athletes. The author’s view was that athletes should be allowed to use AASs in the same way that women are allowed to use female hormones in birth control pills. In defense of physicians, there was little scientific proof to enable them to make recommendations concerning the efficacy and safety of AAS. The early testing of ergogenic drugs was not done scientifically. Prospective randomized studies, with placebo controls, had not been performed. Side effects were not studied adequately to prove that AASs were safe. Many physicians believed that AASs could not be recommended because they had not been tested scientifically. In addition, the side effects seen in athletes taking AASs were worrisome. Another consideration was that no one had studied the long-term effects of AASs in animal or human studies. Many medical authorities considered it unconscionable to recommend any therapy that had not been studied sufficiently.

MECHANISM OF ACTION AND EFFICACY OF ANABOLICANDROGENIC STEROIDS The three main steroids in the body are the androgens, which are responsible for the development of male characteristics; the estrogens, which are responsible for the development of female characteristics; and the corticosteroids, which are compounds manufactured in the adrenal glands, such as cortisol. Females produce small amounts of androgens in the ovary and adrenal gland, and males produce small amounts of estrogen, but it is the production of androgens that makes males masculine and the production of estrogen that makes females feminine. The most abundant androgen in males is testosterone, produced primarily in the testes. In many target cells in the body, testosterone is reduced at the 5α position to dihydrotestosterone, which serves as the intracellular mediator of the actions of testosterone.34 Dihydrotestosterone binds to androgen receptors more tightly than does testosterone and is a more stable and more potent androgen. Precursors in the metabolic pathway to testosterone, dehydroepiandrosterone (DHEA) and androstenedione, bind weakly to the androgen receptors and are known as weak androgens. Etiocholanolone

and androsterone, metabolites of testosterone, also bind weakly to the receptors and are weak androgens. Weak androgens still can have definite anabolic effects, however. In the male, testosterone production peaks at three distinct phases. The first peak is in the fetal period, during the second trimester, and is responsible for the development of the fetus as a male. There is a smaller surge during the first year of life. The largest surge of testosterone production is during puberty. At puberty, androgens cause laryngeal enlargement, which deepens the voice. The penis and scrotum grow. Spermatogenesis is stimulated in the testicles. Testosterone stimulates growth of bone and muscle tissue. As muscular tissue increases, body fat decreases. The skin becomes thicker and oilier. Body hair in the face, axilla, and groin grows to adult levels. Testosterone, or its more active metabolite, dihydrotestosterone, works by binding to an intracellular androgen receptor in the cytoplasm. The steroid–androgen receptor complex is transported to the nucleus, where it attaches to a specific hormone regulatory complex on nuclear chromosomes. Here the complex stimulates the synthesis of specific RNAs and proteins. The RNA compounds are transported through the bloodstream to act on target organs to stimulate spermatogenesis, effect sexual differentiation, and increase protein synthesis in muscle tissues. The androgenic message is received in all organs that have androgenic receptors, which includes the hair follicles and the sebaceous glands. With androgens, in addition to muscle development, one gets oily skin, which can lead to acne; increased body hair; and male pattern baldness, with hairline recession and thinning and loss of central scalp hair. The affects on an athlete’s skin is a key clue for physicians who suspect anabolic steroid use. Acne can be profound, and the back is often significantly affected. The athlete’s skin can take on an unusual oily texture. There can be a unique smell from anabolic steroid effects on the sebaceous glands. Testosterone and its metabolites are metabolized quickly in the liver. Testosterone given by mouth is absorbed quickly and metabolized and is ineffective. Scientists have developed methods of altering the basic structure of the testosterone molecule to delay its metabolism and increase its half-life in the plasma. Alkylation at the 17α position with a methyl or ethyl group allows oral agents to be degraded slowly. Esterification of the 17β position allows parenteral agents to resist degradation. Testosterone esters, such as cypionate and enanthate esters, are more potent than testosterone. These compounds must be injected intramuscularly, usually at 1- to 3-week intervals. The ultra–long-acting testosterone bucculate is administered intramuscularly every 3 months. Newer preparations of testosterone can be administered transdermally.35 These preparations were originally applied to the scrotum, but new transdermal patches and creams can be applied to other parts of the body.

ANDROGENS AND ATHLETIC PERFORMANCE Despite the perceptions of athletes, the scientific data regarding the effects of anabolic steroids on athletic performance are mixed. When used at therapeutic doses,

Nutrition, Pharmacology, and Psychology in Sports

­ either strength nor performance would be expected to be n improved. Hormonal levels are maintained closely in the body in a homeostatic fashion. If a therapeutic dose of a hormone is given, the body halts endogenous production of the hormone to maintain homeostasis. Athletes have taken doses that may be 50 to 100 times the therapeutic dose. This supratherapeutic dosing shuts down endogenous production of AAS, but will it improve performance? Steroid users believe that in combination with training, AASs increase lean body mass and decrease body fat. Early studies looking at these effects were equivocal.36 Athletes believe that AASs increase their muscle strength. The early studies, mostly using low-dose AAS, were not conclusive.37 In 1996, Bhasin and colleagues studied the effect of supraphysiologic doses of testosterone on normal men.38 This was the first prospective randomized study that looked at the effect of super dosing of AAS. In this study of 43 healthy men, the steroid group had definite increases of strength and muscle size compared with the placebo group. No significant side effects were seen in the steroid group. This study was criticized, however, because it was a shortterm (6-week) study, with limited follow-up. This study was significant because it was the first study to provide scientific evidence for a theory that athletes have believed for decades: Steroids work, especially at the increased doses that athletes use. Studies of anabolic steroids on athletes have shown an increase of body weight and an increase in lean body mass, but no significant decrease in the percentage of fat in the body. Studies have shown that not only do muscle fibers gain in cross-sectional diameter with anabolic steroid use but also that new muscle fibers are formed. The upper regions of the body are more susceptible to gains from AAS because of the relatively larger number of androgen receptors in these areas. AASs do not appear to improve athletic ­endurance.39 AASs work not only because they stimulate protein synthesis but also because they stimulate the production of growth hormone (GH), a potent ergogenic agent. AASs have anticatabolic effects. Cortisol is released in response to physical and psychological stress and causes protein degradation and muscle atrophy.40 Testosterone acts by displacing the corticosteroids from receptor sites, reversing the effects.41 AASs may increase oxygen uptake, increase cardiac output, and increase stroke volume.42 They may improve athletic performance by increasing aggressive behavior. One unsubstantiated rumor in the history of AASs was that AASs were given to German troops in World War II to increase their aggressiveness to make them fiercer warriors.43 Although some studies show an association between testosterone levels and aggressive behavior,44 others do not.45 Several problems exist with most of the scientific studies regarding AASs. The available research does not study the drugs actually used by athletes. The doses used in most medical studies are not supratherapeutic. The studies are largely short term. Athletes take large doses for long periods, and this is largely not reflected in the medical literature. Most studies look at only the effect of one drug on athletic performance, whereas athletes often “stack,” or take more than one drug at a time. They also cycle drugs, taking various amounts for a certain time, then going on a “drug holiday.” Case studies in the medical literature

415

that report complications from AASs may overstate the risks. These case studies are also often flawed in that the side effects may have occurred after the athlete used drugs obtained from the black market, where quality control of the drug is not a priority.46 It is easy to see why there are difficulties making a clear medical judgment ­ concerning the efficacy and safety of AAS. There are also ethical questions for scientists seeking to study AASs. It is illegal to use AASs for sports performance. Institutional Review Boards often see a study of an illegal drug given at supratherapeutic doses as problematic. AASs are being used at doses that have received only limited scientific study at the supra­therapeutic doses being used by athletes. AASs have potential long-term adverse side effects that have not been studied adequately.

ADVERSE EFFECTS OF ANABOLIC-ANDROGENIC STEROIDS A wide variety of transient and permanent adverse effects can be seen with the use of AASs. Most side effects are temporary and are reversible when the drugs are stopped. Many athletes think that the medical community overstates the potential problems with anabolic steroids, trying to scare athletes from using the drugs. Physicians have legitimate concerns that long-term studies regarding the benefits and potential adverse effects have not been done in athletes. Athletes who have adverse reactions to ergogenic drugs often do not tell physicians that they are taking the illegal drugs. Many potential problems with AASs are underreported because athletes do not tell physicians that they are using illegal ergogenic drugs. A physician may not make the connection between a possible adverse reaction and ergogenic drugs. Athletes are reassured when they see that although many athletes are using ergogenic drugs illegally, they know few that have serious problems. A variety of side effects have been seen, and there are myriad other problems that might be seen with prolonged use of AAS. More information is needed from long-term scientific studies and from athletes communicating honestly with their physicians. Most male athletes understand that using AASs causes oily skin, acne, small testicles, gynecomastia, and changes in their hair patterns, often leading to baldness. Small doses of androgens increase sebaceous gland secretions, leading to the changes in the skin and acne.47 Gynecomastia is enlargement of the male breast characterized by the presence of firm glandular tissue, usually associated with increased production of estrogens or decreased levels of androgens. The reason men on AASs develop breast tissue is not known. One theory is that when men elevate their androgen levels through drug use, the body homeostatically shuts down production of endogenous androgens. When the user stops the drugs, there is an increase in the relative level of estrogens, which leads to the development of gynecomastia. Gynecomastia is not reversible when the normal levels of androgens return. One means of off­setting the estrogen-related side effects of AASs is to take an antiestrogen drug such as clomiphene citrate (Clomid). Clomid blocks the effects of estrogen increases in AAS

416

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

users by binding to estrogen receptors. Men using AASs also often develop small testicles. The testicles atrophy, responding to homeostasis. With high exogenous AASs, the body ceases the production of endogenous AASs, and the testicles shrink. Often the testicles remain smaller after stopping the drug. There are other adverse effects on the male reproductive system. Oligospermia (decreased number of sperm) or azoospermia (absence of sperm) are possibilities.48 Taking exogenous AASs not only decreases levels of endogenous testosterone but also decreases circulation of follicle-stimulating hormone and luteinizing hormone, which can lead to male infertility. As the Soviet weightlifting team in the 1950s discovered, use of AASs also causes prostate enlargement. AASs also cause prostate cancer to grow. Patients with prostate cancer are given antitestosterone preparations or are castrated to lower testosterone levels. An athlete with an undiscovered prostate cancer could cause his cancer to progress by taking AASs.49 Women who take AAS experience adverse virilization effects, including deepening of the voice, increased body hair, loss of breast tissue, and enlargement of the clitoris. Irregularities in the menstrual cycle, as well as infertility or early menopause, may develop. Many of these adverse effects in women are not reversible.50 The musculoskeletal system may be affected adversely, especially in the prepubertal athlete who takes AASs. Although bone growth is stimulated immediately, there may be premature closure of the growth plates of the long bones. The resultant short stature is permanent.51 Changes in the muscle-tendon unit increase muscle strength to a greater extent than tendon strength, and tendon ruptures may occur.52 The heart, the most important muscle in the body, may be affected adversely by AASs. Androgens are modulators of serum lipoproteins. Androgens increase the plasma levels of low-density lipoproteins (the bad lipoproteins) and decrease the levels of high-density lipoproteins (the good lipids). Oral AASs and synthetic AASs have more pronounced negative effects on lipid metabolism than injectable AASs or natural AASs.53 The altered lipid pattern may lead to atherosclerotic heart disease. Male and female AAS users show similar adverse changes in lipid metabolism.54 Animal studies have suggested that AASs might cause myocardial damage.55 A study investigated the relationship among resistance training, anabolic steroid use, and left ventricular function in elite bodybuilders. Concentric left ventricular hypertrophy was seen, but there was no effect on cardiac function. This study suggests that the heart might enlarge as a physiologic adaptation to intensive training potentiated by AAS.56 In a set of bodybuilding twins, one of the twins used AASs for more than 15 years, and the other twin did not use them at all. No significant difference in cardiac function was found between the twins.57 Significant changes in cardiac function with AAS use have been reported in other cases, however. Cases of myocardial infarction and death of young athletes taking AASs have been documented.58 These case reports are worrisome but are not scientific proof that AAS use directly led to myocardial infarction and death of the athlete. Nevertheless, a postmortem study of 34 AAS abusers aged 20 to 45 years who had died in accidents, suicides, and homicides showed cardiac pathology in more than one

third.59 Cardiac abnormalities may not resolve after stopping AAS use. A study of former AAS users showed that years after discontinuation of AAS abuse, strength athletes showed left ventricular hypertrophy in comparison with AAS-free strength athletes.60 Steve Courson is a former all-pro football player for the Pittsburgh Steelers who admitted to heavy use of AASs. He unfortunately died in an accident while awaiting heart transplantation for severe cardiomyopathy. Although there is no direct proof of a causal relationship between his use of AASs and his severe heart disease, Courson believed that there was a direct correlation and was a strong opponent of AAS use by athletes. Some athletes heard Courson’s message and listened to only part of the story. They are attracted to AASs when they hear that Courson used AASs and intensive training to become a highly successful collegiate and professional football player. They ignore the consequences of a damaged heart. Many rationalize that what happened to Courson was just bad luck and could never happen to them. The liver is the organ responsible for most of AAS metabolism. Potential damage to the liver by use of AASs is a major concern. Athletes with preexisting liver dysfunction may be at greatest risk for serious liver abnormalities with AAS use. The oral AASs (the 17-alkylated androgens) are absorbed rapidly after ingestion and transported quickly to the liver. Oral AASs are more potentially damaging to the liver than injectable AASs. Temporary liver disturbances are common in athletes who use oral androgens.61 AASs have a multiplicity of effects in the liver, which has multiple AAS receptor sites. These disturbances in liver function usually are transient, and function returns to normal after stopping the use of drugs.62 Some investigators believe that AAS-induced hepatotoxicity is overstated, noting that elevations in transaminase levels seen in users might be due to skeletal rather than liver damage.63 The most important blood test to be done to follow patients using AASs is the level of γ-glutamyltransferase, which is the most distinctive enzyme of the detection of hepatic dysfunction. A study of the effect of AASs on rats showed definite cellular damage to hepatocytes.64 Long-term studies on the effect of AASs on the liver in human subjects have not been conducted. Until these long-term studies are performed, hepatotoxicity must remain a major concern with use of AASs, especially when using oral agents. One of the most serious adverse effects of AAS use occurring in the liver is peliosis hepatis. This condition results in the development of blood-filled cysts in the liver. If the cysts rupture, peliosis hepatis can be life threatening. The first association between AAS use and peliosis hepatis was described in 1952.65 In contrast to most adverse effects of AASs, peliosis hepatis does not appear to be dose related and can occur at any time after starting AAS use.66 The proposed mechanism of action of AAS-induced liver toxicity, including peliosis hepatis, is through AASmediated hepatocyte hyperplasia.67 Enlarged liver cells block venous and lymphatic flow, producing cholestasis and peliosis cysts. A relationship between androgens and hepatic tumors was first suggested in 1971.68 Hepatocellular carcinoma is more common in men than in women, and the link between AASs and hepatocellular carcinoma has been studied. No definite proof exists, although there are

Nutrition, Pharmacology, and Psychology in Sports

multiple reports of hepatocellular carcinoma developing after AAS use.69 These tumors are pernicious in that they present silently with few symptoms until the late stages of tumor growth. The most commonly associated psychiatric effect of AASs is an increase in aggressiveness. Defendants accused of violent crimes have claimed that their crimes were committed under the influence of AASs. They have claimed that the AASs made them more aggressive and violent. There are few good prospective, randomized studies of AAS on behavior. Most studies using therapeutic doses of exogenous testosterone showed no adverse effects. The short-term study of supratherapeutic dosing of AASs in athletes revealed no significant psychiatric effects.70 Many of the studies reported only a positive effect on mood.71 In Brown-Séquard’s 19th century study of self-administration of testicular substrates, one of the reported positive effects was improvement in mood that accompanied improvement in his physical stature. Synthetic testosterone derivatives became available in the 1930s. One of the first uses of AASs was by physicians exploring the effects of AASs on mood as well as psychiatric maladies.72,73 AASs were used to treat a variety of conditions ranging from depression to psychoses, but the results were disappointing. Studies in animals and humans have linked high levels of endogenous AASs with aggressive behavior.74 A study of 41 athletes using supratherapeutic doses of AASs for an average of 45 weeks revealed that 34% experienced psychiatric symptoms based on criteria in the Diagnostic and Statistical Manual of Psychiatric Disorders, 3rd edition.75 This study linked major mood disorders, such as severe depression or mania, with AAS use. AAS users are more likely to use alcohol, tobacco, and illicit drugs.76 AAS users have been reported abusing the opioid agonist-antagonist nalbuphine.77 In a study of 227 men admitted to a private inpatient facility for substance-dependence treatment, 9.3% had a history of AAS abuse.77 None of the men had any form of substance abuse or dependence before their use of AASs; 86% of the men using opioids said they had started using them to counteract the side effects of the AASs; 81% had purchased the opioids from the same drug dealer who had sold them the AASs. This study suggested that AASs might be the gateway to opioid dependence. The subjects using opioids were suburbanites with an average income of $69,800. Disorders of body image have been reported with use of AASs. Many weightlifters who may appear muscular when compared with the average athlete consider themselves small or weak. This disorder is similar to women with anorexia nervosa who, despite being very thin when compared with the average woman, think that they are fat. The Adonis Complex: The Secret Crisis of Male Body Obsession addresses what the authors describe as an obsession by men to achieve physical perfection.78 The authors claim that men are struggling with the same pressures to achieve physical beauty that women have dealt with for centuries. The authors see the tremendous interest in AASs by the general population as a reflection of the glorification in the media of athletes and movie actors who are using AASs to achieve their muscular bodies. Although concerned about the medical hazards of AASs, the authors state that what is far more dangerous are the psychiatric effects of irritability

417

and aggression seen with AAS use and the depression present on AAS withdrawal. The authors also are concerned that billions of dollars are being spent not only on illegal AAS but also on muscle-building supplements legally available on the Internet and at health food stores.

STEROID SUPPLEMENTS One of the most important laws passed by the U.S. Congress that has protean implications for individuals involved in the care of athletes is the Dietary Supplement Health and Education Act of 1994 (DSHEA). This law allows a tremendous variety of substances to be sold without approval from the U.S. Food and Drug Administration (FDA) as long as they are sold as dietary supplements and not as drugs. The DSHEA established a formal definition of dietary supplement using several criteria. A dietary supplement is a product (other than tobacco) that is intended to supplement the diet. It bears or contains one or more of the following dietary ingredients: a vitamin, a mineral, an herb or other botanical, an amino acid, a dietary substance to supplement the diet by increasing the total daily intake, or a concentrate, metabolite, constituent, or extract, or combinations of these ingredients. The product is intended for ingestion in pill, capsule, tablet, or liquid form. The product is not recommended for use as a conventional food or as the sole item of a meal or diet, and it is labeled as a dietary supplement. Products have included approved new drugs, certified antibiotics, or licensed biologics that were marketed as dietary supplements or food before approval, verification, or license.79 Subsequent to this bill’s passage, many synthetic AASs have become commercially available as dietary supplements. As such, these products do not have to pass the efficacy and safety requirements of the FDA. They do not have to meet quality control standards.80 Claims made concerning their effectiveness do not have to be substantiated by scientific proof as long as a disclaimer is listed on the product. If scientifically unproven claims are made, the federal government is responsible for establishing that the claim is false or misleading. Two dietary supplements that are used widely and purchased legally by athletes are DHEA and androstenedione. DHEA is a weak androgen but is one of the main precursors of testosterone. DHEA is made in large quantities by the adrenal cortex. DHEA is the most abundant steroid in the body and is a precursor not only of testosterone but also of estrogen, progesterone, and corticosterone. DHEA is available as a nutritional supplement because it is found naturally in wild yams.81 DHEA levels are high in the prenatal period and in puberty and gradually and progressively decrease as individuals age. Studies done in patients older than 50 years, when DHEA levels are decreasing significantly, showed beneficial results when DHEA supplements were taken.82 DHEA has been described as a wonder drug, and it has been marketed especially to middle-aged and older individuals as an anti-aging drug. Although few scientific studies have been done with DHEA, it has been marketed as a fat-burning, muscle-building, sexual stimulant. One study found that high doses of DHEA decreased low-density lipoproteins.83 Many athletes use it for its androgenic and anticatabolic effects. One of the problems

418

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

with athletes taking DHEA in an attempt to increase their testosterone levels is that the DHEA may be converted to estrogen, which does not have the anabolic effects desired by the athletes. Feminization may occur in the male athlete. There is little evidence to support the claims for DHEA either as a potent anabolic agent or as an anti-aging drug.84 There are no published studies of the long-term effects of taking DHEA, particularly in the large doses used by athletes.85 Many athletes will use an antiestrogen drug such as Clomid when taking DHEA or the next compound in the testosterone synthesis pathway, androstenedione. Androstenedione is a potent AAS produced endogenously in the adrenal glands and gonads. In the liver, androstenedione is metabolized to testosterone. Similar to DHEA, androstenedione is purchased easily as a dietary supplement. Originally touted as the “secret weapon of the East Germans,” androstenedione first was used in the 1970s. Its reputation as an effective anabolic agent resulted in worldwide use. In March 2004, the U.S. Department of Health and Human Services announced that the FDA had requested that androstenedione manufacturers stop distribution. By October 2004, President Bush signed into law the Anabolic Steroid Control Act, which added androstenedione to the list of banned, nonprescription, steroidbased drugs. Major League Baseball, the National Football League, the Olympics, and the National Collegiate Athletics Association all now prohibit the use of androstenedione. Androstenedione received tremendous exposure in the media during the 1998 baseball season when St. Louis Cardinals slugger Mark McGuire was discovered to be using androstenedione, legally, as a nutritional supplement. Although McGuire stopped using the supplement after the discovery, sales of androstenedione rose dramatically.86 Studies indicate that androstenedione does not enhance muscle building and has some adverse effects.87 Similar to DHEA, as androstenedione is metabolized, it can be converted to testosterone or to estrogen. A male athlete taking androstenedione theoretically can elevate estrogen levels. The adverse effects seen with androstenedione in this study were associated with the conversion of androstenedione to estrogens. Male subjects taking androstenedione developed elevated estradiol levels. These estradiol levels are equal to the estradiol levels seen in women at the follicular phase, when estradiol levels are highest. This study was criticized as being a small study, with only 20 subjects. It also was a short-term study; the subjects were given androstenedione for only 6 weeks. The two groups might not have been truly comparable because they had significant differences in free testosterone levels of almost 50%. As a testosterone precursor, the possibility of adverse reactions as are seen with testosterone must be considered. Most studies have shown a decrease in high-density lipoprotein cholesterol, thus leading to potential cardiac risks. Priapism, or persistent painful erection, was seen in a healthy young man using androstenedione.88 No precipitating factors of priapism were found in the patient. Other studies, done in monkeys, found hyperplastic changes in the prostate after androstenedione administration.89 Good scientific prospective, randomized studies of athletes are needed to determine the efficacy and safety of androstenedione and other ergogenic drugs. No study has shown significant ergogenic improvement with androstenedione supplementation.

Creatine One of the most popular supplements currently embraced by athletes is creatine. Creatine is a nitrogenous compound synthesized in the body by the liver, pancreas, and kidneys. Creatine also is absorbed through the diet. Fish and meat are good sources of exogenous creatine. Chevreul, who named creatine after the Greek word for flesh, discovered the compound in 1832. In the 1920s, it was discovered that creatine played a crucial part in the structure and function of adenosine triphosphate (ATP), the body’s prime energy source. Creatine is found principally in skeletal muscle. In its free and phosphorylated form, creatine plays a crucial role in the regulation of skeletal muscle metabolism. Creatine is synthesized, largely in the liver, from the amino acids arginine, glycine, and methionine. The creatine is transported to the skeletal muscles, where it is absorbed into the muscle cells by a sodium-dependent transporter system. When in the muscle cell, creatine becomes phosphorylated by the activity of creatine kinase, and creatine is synthesized into creatine phosphate. Creatine phosphate serves as an energy substrate contributing to the resynthesis of ATP during strenuous exercise. Phosphocreatine is stored in the muscle cell until it is needed as an anaerobic energy source. When motion is initiated, the creatine phosphate bond is broken, and the phosphate is donated to adenosine diphosphate (ADP) to make ATP and provide instant energy.90 Later, with sufficient oxygen, creatine is reconverted to phosphocreatine in an aerobic reaction. In 1992, it was discovered that creatine supplementation increases the ability of skeletal muscles to accumulate creatine and phosphocreatine.91 More substrate is available for energy-sapping anaerobic activities. Creatine is vital for the heart, brain, kidneys, and retinae.92 Creatine acts as a lactic acid buffer during exercise. Lactic acid is generated during intense exercise. Hydrogen ions are consumed when phosphocreatine is used to generate ATP: PCr + ADP + H

+

Cr + ATP Creatine kinase (enzyme)

Not only does creatine provide energy for anaerobic activities but also helps buffer lactic acid, which prevents prolonged anaerobic exercise. With increased phosphocreatine available, there is more ability to buffer lactate. The athlete can exercise anaerobically for a longer period, with delayed fatigue. A third way that creatine may help athletic performance is through stimulation of protein synthesis and increasing muscle mass.93 Although creatine was believed to have bodybuilding effects in the 1920s, it was not until the 1970s that seminal research on rats indicated potential benefits for athletes.92 Creatine supplementation soon was seen to provide performance-enhancing and fatigue-delaying effects in athletes.93 Creatine supplementation increased the synthesis of the myosin heavy chain in skeletal muscles. These studies indicated that creatine not only might have benefits in energy production and lactate buffering but also might increase muscle cell mass.93 A study using creatine supplementation in patients with the ophthalmic condition of gyrate atrophy showed a significant increase in type II (fast-twitch) muscle fibers.94

Nutrition, Pharmacology, and Psychology in Sports

All these findings suggested that creatine might be used optimally in athletes who engage in sports that emphasize short bursts of intense anaerobic activity, such as football, soccer, lacrosse, hockey, basketball, and powerlifting. Short-term creatine supplementation enhanced the ability to maintain muscular forces during jumping,95 intense cycling,96 and weightlifting.93 Creatine use became widespread among athletes as the benefits of its use were spread by word of mouth. Not all athletes benefited from creatine supplementation, however. Athletes involved in aerobic sports did not seem to benefit from creatine supplementation.97 In studies in cyclists, creatine did not improve performance in repeated 700-m sprints or in longer cycling. Creatine did not increase maximal isometric strength or the rate of maximal force production. There appear to be athletes who are responders to creatine supplementation as well as athletes who are nonresponders, no matter what the sport.92 It is estimated that about 30% of athletes are nonresponders in that they already possess a maximal amount of creatine that can be stored by the muscle cell. Unfortunately, muscle creatine content is measured only by muscle biopsy. Early studies emphasized the importance of loading creatine. Athletes would use large doses (20 g/day of creatine for 5 days) before reaching a maintenance dose (2 g/day).98 More recent evidence revealed that loading is not necessary. A low-dose regimen of 3 g/day is just as effective.99 The low dose of 3 g of creatine could be obtained through the diet by eating 5 pounds of meat or fish. Athletes most commonly use creatine monohydrate as the mode of taking creatine supplementation. Studies by Greenhaff and colleagues indicated that creatine intake could be facilitated by combining creatine with insulin-releasing carbohydrates.93 Many athletes mix their creatine with a fruit drink to facilitate uptake. Commercial carriers have combined creatine with dextrose and a variety of amino acids and minerals to maximize creatine absorption.100 Recently, creatine ethyl ester has become available. It is absorbed more readily and does not require loading or concomitant carbohydrate ingestion. Accompanying creatine supplementation is an increase in body mass up to 2 kg.101 This gain is believed to be secondary to an increase in body water. The effect of creatine on body fluid balance requires further investigation. When college football teams first used creatine, there were reports of severe cramping. It was recommended that creatine be taken with large volumes of water. It also was recommended that creatine not be taken before or during exercise.102 Case reports of gastrointestinal problems, including diarrhea, were common, but no definite evidence points to creatine as the definitive cause. There has been a lot of excitement about the ergogenic potential for creatine in athletes because creatine supplementation is a legal means of improving performance during short-duration, repetitive bursts of intense exercise. There are potential problems with creatine, however. Creatine does not work for all athletes; 30% may be nonresponders. Athletes in anaerobic sports appear to benefit much more than athletes participating in aerobic sports. Creatine may be detrimental to aerobic performance.103 The ability of creatine to increase the intake of water into the muscle cells may affect the body’s fluid balance. ­Recommendations for

419

the use of creatine have been developed. Drinking large amounts of water when taking creatine is crucial. Creatine should not be used during periods of intense exercise. Creatine should be avoided during periods of high heat and humidity. No studies have been done on the effects of creatine on young athletes, and creatine is not recommended for athletes younger than 18 years of age.104 Long-term studies of the effect of creatine on athletes have not been performed.

Growth Hormone One of the most fascinating (although illegal) ergogenic drugs used by athletes is human GH. Human GH is a polypeptide produced and stored in the anterior pituitary gland. Pituitary cells called somatotropes make human GH. More than half of the anterior pituitary gland consists of somatotropes. Although it is well known that human GH secretion is high during puberty, human GH is secreted throughout a person’s lifetime. GH secretion occurs daily in a pulsatile fashion, with the highest levels seen immediately after going to sleep.105 GH was discovered in the 1920s when researchers found that after an injection of ox pituitary glands, normal rats reached abnormally large size. Not only were their skeletal muscles enlarged, but all organs and viscera were enlarged. Animal breeders soon used GH extracts to increase the size of animals. It was discovered that animals treated with GH developed increased muscle mass and decreased body fat.105 In the 1950s, GH was injected into children whose growth was stunted by absence of GH.106 The GH was extracted from the brains of cadavers, many of whom came from Africa and Asia. Thousands of short-stature children were being treated with human GH successfully until Creutzfeldt-Jakob disease, a disease causing progressive dementia and loss of muscle control leading to death, was transmitted to the children through the cadaver extracts. Transmitted through prions, proteins that contain neither RNA nor DNA, Creutzfeldt-Jakob disease is related to bovine spongiform encephalopathy.107 By the early 1990s, seven children had developed CreutzfeldtJakob disease after injection of human GH.107 The FDA stopped distribution of the drug. There was a need to develop a safe drug to treat children of short stature caused by low levels of human GH. The Genentech Company used recombinant DNA technology to manufacture a biosynthetic GH, somatrem (Protropin). Protropin had the identical sequence of 191 amino acids as human GH, with an additional amino acid, methionine, on the N-terminus of the molecule. Human GH now could be safely given to children with low GH levels, and studies of its effects were undertaken. GH affects almost every cell in the body. Because human GH levels drop to low levels as one ages, human GH has been promoted as an antiaging agent. However, because human GH secretion can be stimulated throughout the lifetime by sleep or exercise, combining exercise with rest may be beneficial. Long-term studies of the effectiveness of human GH have not been done. In the liver, human GH is rapidly converted into insulinlike growth factor-1 (IGF-1). IGF-1 is responsible for most of the actions of human GH in the body. Athletes have used human GH and IGF-1 as ergogenic supplements. Human

420

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

GH has many actions that would be attractive to athletes. Human GH stimulates protein metabolism. Nitrogen is retained, and urinary nitrogen excretion is decreased. Human GH stimulates amino acid uptake and protein synthesis. Human GH is a potent regulator of carbohydrate metabolism. Human GH decreases insulin sensitivity and decreases cellular uptake of glucose. Human GH stimulates the catabolism of lipids, making free fatty acids available for quick energy use and sparing muscle glycogen.108 Human GH stimulates bone growth, which is crucial in a prepubertal youth but may cause problems in an adult with fused growth plates. In adults, some bones enlarge with human GH, and acromegalic changes develop, with facial bone, hand, and foot enlargement. Other problems can occur with human GH. With decreased insulin sensitivity comes a predisposition to developing diabetes. Heart disease, including cardiomyopathy and congestive heart failure, has been seen with human GH administration.108 Lyle Alzado, an all-pro football player and self-admitted heavy user of human GH, died of a brain tumor. Alzado became an opponent of illegal ergogenic drugs for athletes, and he was sure his brain tumor was a direct result of his use of human GH. No proof for his contention exists. Human GH can cause enlargement of intracranial lesions.109 Intracranial hypertension with papilledema has been seen in patients taking human GH. Leukemia has been seen in patients with low human GH levels treated with recombinant human GH. There are no scientific studies that show improved athletic performance with human GH. Nevertheless, human GH has been used widely by athletes. A survey of American high school students in 1992 showed a 5% reported use of human GH.110 In patients with acromegaly, muscles become larger, but are weaker as a result of a human GH–induced myopathy. Because much of human GH is obtained clandestinely from overseas sources, quality control of the product is not ensured. Human GH is not available orally and must be administered by injection, which can cause problems for the athlete. Using injectable drugs subjects the athlete to a risk for infection. If athletes are sharing needles, they put themselves at a risk for exposure to human immunodeficiency virus and hepatitis viruses. Human GH is expensive, costing about $1000 per month of use. One reason that human GH is so appealing to athletes, despite the costs and plethora of adverse effects, is that there is no effective test to prove its use because there is currently no method of differentiating exogenous from endogenous GH. The WADA has made an effective test for human GH one of its highest priorities. Yet, Don Catilin, the dean of drug testing, reported in March 2007 that an effective test for human GH is not available “and may never be.”111 Scientists are unsure whether a reliable blood or urine test for human GH can be developed. Athletes in the past have felt that a tremendous advantage of using human GH is that they could take human GH without fear of detection.

Erythropoietin The sporting world was alerted to the importance of high-altitude training by the results of the 1969 Summer Olympics in Mexico City. Athletes who had trained at

high altitudes had an apparently significant advantage in the endurance sports. Because aerobic endurance athletic activities are limited by the ability of the blood to deliver oxygen to the working muscle, activities to increase red cell mass theoretically should improve aerobic performance.112 It has been known for centuries that people who reside at high altitudes adapt to the altitude and are able to do more work than people from lower altitudes. In 1893, Miesher discovered that lower oxygen saturation, which occurs in high altitudes, produced changes in the bone marrow to cause an increased number of red blood cells. In 1906, Carnot and DeFlandre discovered that injecting serum from anemic rabbits into normal rabbits caused an increased number of red blood cells in the normal rabbit. Carnot and DeFlandre named this red blood–stimulating factor hemopoietin. Further work by Bonsdorff and Jalavisto in 1943 led to a discovery of the blood-stimulating hormone they named erythropoietin (EPO). Miyake purified erythropoietin in 1977, and Lai discovered the molecular structure in 1986. Recombinant human EPO was synthesized in 1986. Red blood cells have an average life span of 120 days, and a percentage of the cells are replaced by the bone marrow. The production of red blood cells is regulated mainly by EPO, which is manufactured in the kidney. If a person develops anemia, EPO can stimulate the bone marrow to increase red blood cell production by 3 to 10 times.113 Berglund and Ekblom found a 17% increase of time to exhaustion in male athletes after 6 weeks of EPO administration.114 When EPO became safely available with the manufacture of recombinant EPO, athletes were quick to see the implications of EPO as an ergogenic drug. Many athletes had become interested in blood boosting, or increasing one’s blood cell mass by transfusion, after the 1968 Mexico City Olympics.115 EPO allowed increased red cell volume without the dangers associated with blood transfusions. Although the Olympic Committee banned recombinant EPO (rEPO) in 1990, only recently has there been a test to detect rEPO supplementation. The testing hinges on two factors. Recombinant EPO–stimulated red blood cells are smaller and have greater amounts of hemoglobin. Too many of these abnormal red cells would suggest rEPO use. However, the most conclusive test is evaluation of EPO molecular weight. Recombinant EPO has a single peak molecular weight, whereas endogenous EPO has a more variable molecular weight. Recombinant EPO has a short half-life of 20 hours, and its effects last for 2 weeks after its use.116 Athletes have been moving away from EPO for fear of being caught as new methods of testing for the compound have been developed and employed. Increasing the red blood cell mass in the blood can lead to increases in blood viscosity, which can harm the athlete. Increased red blood cells can lead to hypertension, stroke, headache, congestive heart failure, venous thromboses, and pulmonary emboli. Studies have not been done in a controlled, scientific fashion to determine the efficacy and safety of EPO. Because it is accepted that a high blood volume is important to cardiac function in aerobic athletes, it would seem intuitive that EPO should benefit endurance athletes in whom aerobic performance is crucial. It appears, however, that many highly trained endurance

Nutrition, Pharmacology, and Psychology in Sports

421

a­ thletes have a blood volume that places the heart near or at its ability to fill during diastole.117 A well-conditioned athlete ­ attempting to increase blood flow to a heart that is at its maximal ability to hold blood may put himself or herself in danger.

supplementation. Randomized placebo-controlled studies failed to show any benefit to overtraining during laboratory testing using cycle ergometer.120

Arginine and Nitric Oxide

Caffeine is the most widely used legal ergogenic drug in the world. Billions of people use caffeine yearly to improve physical and mental performance, many doing so on a daily basis. Caffeine is most commonly provided through ingestion of coffee and tea and cola soft drinks. Doses between 5 and 10 mg/kg 1 hour before an athletic contest are used. A typical 8-ounce cup of coffee contains 120 mg of ­ caffeine. Thus, two to four standard cups provide the necessary amount of caffeine. Caffeine is also available in 100-mg and 200-mg tablets. The United States Olympic committee (USOC) previously had placed caffeine in a restricted status, allowing no more than 12 μg/mL in urine. However, because of highly variable excretion rates, this restriction has been eliminated. Three potential methods of ergogenic improvement exist. A direct central nervous system effect occurs, increasing perception of physical effort and improving neural activation of muscle transport. Caffeine also increases calcium transport in muscle. Perhaps most important for endurance athletes is the caffeine-stimulated release of free fatty acids (FFAs) from adipose tissue, allowing metabolism of FFAs by exercising muscle, sparing muscle glycogen, and increasing blood glucose. Thus, the endurance athlete is able to perform for a longer period without depleting his or her intracellular muscle glycogen. It is the depletion of muscle glycogen that causes the athlete to “hit the wall” and suffer performance degradation at the end of prolonged athletic activity. Depletion of muscle glycogen also leads to anaerobic metabolism and lactic acidosis.121 Caffeine also stimulates potassium transport into tissue, maintaining muscle cell membrane excitability. This may decrease muscle cell reaction time after neural stimulus and reduce muscle fatigue after prolonged competition. The most common side effects include anxiety, restlessness, diuresis, panic attacks, gastritis, reflux, and heart ­palpitations.122 It is aerobic activity that been demonstrated in a controlled environment to improve with caffeine ingestion. In one study, 1500-m swim time was significantly improved in trained athletes.123 Cycle time to exhaustion at Vo2max was significantly improved in another placebo-controlled study.124 In another cycling study, a 7% increase in distance covered in 2 hours was noted.125 It thus appears that caffeine’s greatest effect is for anaerobic performance in cycling and running for periods greater than 2 hours or for exhaustive middle-distance performance. There are only minimal data demonstrating improvement in anaerobic or strength performance. One study noted improvement in 1RM bench press after short-term caffeine supplementation, but no improvement in Wingate, leg extension, and total volume of weight lifted during an endurance test.126 Thus, caffeine has limited ergogenic potential for resistance training.

l-Arginine, an amino acid, is available for human consumption as arginine HCl, arginine α-ketoglutarate, and arginine aspartate. Supplementation with oral arginine has been advised on the assumption that intracellular nitric oxide (NO) will be increased with elevated blood levels of arginine. l-Arginine is a direct precursor of NO, an important mediator of multiple cellular processes. However, it has not been proved that arginine supplementation actually increases NO production in vivo.118 There are several proposed mechanisms of action of ergogenic improvement. The first and most important is as an endothelial vasodilator (EDRF) through the NO pathway. This increases perfusion to muscle, increasing flow of muscle nutrients, O2, and amino acids for muscle performance and growth. The second potential mechanism for ergogenic improvement is the ability of arginine to stimulate endogenous GH release. Arginine is one of the agents used in the GH stimulation test. Intravenous infusion of arginine stimulates the functioning pituitary gland to release GH. This test is primarily used to evaluate for low GH levels in hypopituitarism. Third, arginine stimulates glucose uptake by muscle cells.119 In the laboratory, it has been shown that prolonged NO exposure triggers mitochondrial biogenesis in mammalian cells and that NO stimulates activation of muscle satellite cells, thereby increasing muscle repair and hypertrophy. Promoters of arginine supplementation have used these data to infer that oral arginine stimulates the same processes. Potentially validating the vasodilator effects of arginine ingestion are some of the current medical indications for arginine supplementation. It is used in certain cardiovascular disease states, including heart failure and angina. NO inhibits platelet aggregation and arginine inhibits atherogenesis in laboratory animals. The most common side effects are flushing, nausea, and diarrhea. There is no standardized daily dose for oral arginine supplementation. Doses between 6 and 15 mg/day have been studied. Only one study demonstrated an ergogenic effect with oral supplementation with arginine: 35 resistance-trained men were randomly assigned to ingest either placebo or 12 g of arginine α-ketoglutarate (AAKG) daily in three divided doses. The men performed resistance training 4 days per week for 8 weeks. Multiple blood markers, body composition tests, aerobic and anaerobic tests, 1 repetition maximal (1RM) bench press test, and isokinetic quadriceps endurance were measured. Significant differences were noted in blood glucose, plasma arginine, 1RM bench press, and Wingate peak power.120 The ergogenic effect is not related to the proposed GH stimulation. GH release has long been known to increase with exercise. In one study, GH secretory stimulation was actually greater with resistance exercise alone than with arginine supplementation and exercise in combination.118 Endurance athletes do not appear to benefit from arginine

Caffeine

422

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

β-Hydroxy-β-Methylbutyrate

FUTURE ISSUES: GENE THERAPY

β-hydroxy-β-methylbutyrate (HMB) is a metabolite of leucine and is largely used in conjunction with other ergogenic drugs. HMB is found naturally, in small amounts in catfish and grapefruit. It is thought to be beneficial to athletes by increasing lean body mass and increasing strength. Its proposed mechanism of action is as an anticatabolic agent. There is scientific evidence that HMB can protect skeletal muscle from breakdown during intense sports activity.127 In a meta-analysis of research on more than 250 supplements, HMB was shown to increase strength with exercise significantly.128 However, the lead author holds a patent for an HMB supplement, and his research has not been duplicated independently. Other studies have not shown gains with HMB. Side effects have not been identified.

Performance in sports is dependent on general health, training, conditioning, nutrition, and genes. An athlete can work diligently to improve his or her skills and condition, but until recently, could not do anything to improve genetic ability. Recent animal studies, in which muscle types and strength, aerobic capacity, and oxygen carrying capacity have been improved through genetic engineering, have caught the attention of athletes and coaches trying to improve their performance. These technologies have the potential of adding a synthetic gene that theoretically could last for years and produce high quantities of musclebuilding chemicals.131 Because these chemicals would be identical to the naturally produced chemicals, they would be undetectable. Theoretically, all existing proteins in the body could be altered by gene therapy. Gene therapy can be defined as the transfer of genetic material to human cells for the treatment or prevention of a disease. A therapeutic gene can be transferred to the human cell nucleus to compensate for an absent or malfunctioning gene. DNA is the usual genetic material that is encapsulated into an adenovirus, a retrovirus, or a liposome. The viruses are processed so that they are not theoretically dangerous to the host cell. The DNA encodes for a protein that will be therapeutic. The DNA is also processed to provide a specific length of cell production.132 The potential dangers of using viruses to deliver and integrate DNA into host cells in gene therapy, which has led to a death, have caused researchers to develop nonviral gene delivery approaches, such as small episomal plasmids or artificial chromosomes.133 More than 4000 patients have received some form of gene therapy, and potential benefits are astounding. The clinical use of gene therapy is closely regulated, but athletes are interested in the benefits of gene therapy and, not unlike their attitudes concerning ergogenic drugs, are ready to experiment despite reported risks. Athletes seek to gain advantage by increasing their muscle strength. Great progress has been achieved with the discovery of growth factors and molecular mechanisms involved in muscle development. One of these growth factors is myostatin, a member of the transforming growth factor-β family of proteins that has been demonstrated to play a fundamental role in the regulation of skeletal muscle growth. Blocking of the myostatin signaling transduction pathway by specific inhibitors and genetic manipulations has been shown to result in a dramatic increase of skeletal muscle mass. Theoretically, by blocking myostatin, an athlete could increase muscle mass and, theoretically, muscle strength.134 Additional research into myostatin metabolism has identified myostatin-binding proteins that compete with receptor sites and prevent myostatin activation. These myostatin-binding proteins are also potential targets of gene doping to increase muscle size and strength in athletes.135 When Finnish Nordic Skier Eero Mantyranta won two gold medals in the 1964 Winter Olympic games, he was accused of “cheating.” He was not cheating, but he had been born with an athletic advantage. Mantyranta had a naturally occurring gene mutation that gave him more red blood cells. He had a much higher oxygen carrying

γ-Hydroxybutyrate Unlike HMB, for which few side effects have been discerned, γ-hydroxybutyrate (GHB) is a drug with a plethora of known side effects and has been banned in athletes for almost 20 years. Street terms for GHB include cherry meth, liquid X, fantasy, organic Quaalude, salty water, Georgia home boy, great hormones at bedtime, grievous bodily harm, and liquid Ecstasy. GHB is a powerful, rapidly acting central nervous system depressant. First synthesized in the 1920s, GHB is produced naturally by the body in small amounts, but its physiologic function is unclear. It may act as a neurotransmitter or a neuromodulator of dopamine metabolism.129 Most GHB is found in the brain, but there are also muscle and heart receptors. GHB’s alleged positive effects on muscle physiology led to it use by athletes, especially in the 1980s, when it was easily obtainable. It was thought that GHB was an inhibitor of energy metabolism, protecting tissues when glucose supplies were low. It was also thought that GHB protected muscles from the damaging effects of lactic acid and anoxia. GHB was often used in conjunction with other ergogenic drugs. GHB was thought to be anabolic in its ability to stimulate the release of human GH.130 However, problems with GHB overdose and GHB abuse led to its prohibition. Banned by the FDA in 1990, GHB became a Schedule I Controlled Substance in March 2000. GHB is abused for its ability to produce euphoric and hallucinogenic states and for its alleged function as a GH that releases agents to stimulate muscle growth. GHB is usually taken orally as a light-colored powder that dissolves in liquids. It is clear, odorless, tasteless, and almost undetectable when mixed in a drink. It is rapidly absorbed by the body, and its effects can last up to 6 hours. At a low dose, GHB acts as a relaxant, causing a loss of muscle tone and inhibitions. At higher doses, GHB also interferes with motor coordination and balance. Interference with motor and speech control can occur. Alcohol and GHB can be a deadly combination, leading to respiratory depression, unconsciousness, and coma. GHB can become addictive with sustained use. Although many users relate that GHB is a pleasure enhancer that lowers inhibitions, unscrupulous criminals have used GHB as a date-rape drug, clandestinely adding it to the drinks of unsuspecting victims. The victim may lose the ability to consciously consent to sexual acts.

Nutrition, Pharmacology, and Psychology in Sports

c­ apacity and had genetically gifted endurance. Gene therapy has been used to insert a gene to boost production of the hormone EPO. An athlete given this ability to increase his production of EPO would theoretically have a higher oxygen delivery ability and endurance. Monkey studies using an adenovirus delivery system showed some monkeys had tremendous gains in their EPO levels and hematocrits, but a small group of monkeys developed a serious and lifethreatening anemia due to an autoimmune response to the transgene-derived EPO.136 Obviously, further studies need to be done before this type of gene therapy is released for legitimate use. Other potential uses of gene therapy in athletes involve the production of IGF-1 to increase muscle bulk and strength, vascular endothelial growth factor to improve blood vessel production, and theoretically oxygenation of muscle tissues and leptin to lower body fat. The potential for good is almost incalculable, but many long-term studies need to be done to determine the risk-to-benefit ratio.

CONCLUSION For centuries, athletes have been willing to risk death in an attempt to improve their ability to compete in sports. What can individuals entrusted with the health of athletes do to protect athletes from themselves? First, it is essential to become educated regarding ergogenic drugs. An important tenet in medicine is, “you only recognize what you know.” Most medical personnel get little information regarding ergogenic training in their schooling. Sports nutrition is not a high priority in medical school training. To keep up with the athletes, it is necessary to keep learning. It is helpful to read not only the scientific literature regarding these agents but also the lay literature that is being read by the athletes. The health care professional should be able to communicate openly with athletes in the area of ergogenic drugs as well as nutrition, vitamin supplementation, and training for sport. Second, the medical community should conduct scientific controlled studies on athletes using ergogenic drugs. Such studies would be difficult to conduct because many of these drugs are illegal. Many ergogenic drugs have significant side effects. Research oversight committees often are reluctant to approve studies using drugs with known or suspected adverse effects, but such studies need to be conducted. Third, if athletes are using banned ergogenic drugs, and society and the athletic community agree that drug use should be precluded, effective drug testing and appropriate punishments of athletes, coaches, teams, and nations ignoring these bans should be implemented. Athletes have competed in a variety of sports for centuries. Their quest for improved performance has brought many developments in training practices as well as the use of ergogenic drugs. Significant questions remain regarding the safety and advisability of the use of ergogenic drugs available today. Careful scientific studies are needed to enable athletes and their health care providers to make wise decisions regarding use of ergogenic drugs and to understand the risks involved in their use.

C

r i t i c a l

P

423

o i n t s

l Athletes

have sought drugs to improve their performance for centuries. l The use of ergogenic drugs is not limited to elite professional or Olympic athletes. Athletes as young as middle school have easy access to drugs that allegedly enhance performance. l There are good short-term scientific studies that most of the drugs used by athletes to enhance performance have potential benefits. There are also many studies showing that side effects are not rare and are often dangerous. These dangers will not deter most. l There are no long-term studies of the effects of ergogenic drugs. l Ergogenic drug users will rarely tell their physician that they take these drugs. l New designer drugs and new techniques including gene therapy are being used by athletes to improve their performance.

S U G G E S T E D

R E A D I N G S

Abel T, Knechtle B, Perret C, et al: Influence of chronic supplementation of arginine aspartate in endurance athletes on performance and substrate metabolism. Int J Sports Med 26:344-349, 2005. Brosnan J, Brosnan M: Creatine: Endogenous metabolite, dietary, and therapeutic supplement. Annu Rev Nutr 27:241-261, 2007. Eichner ER: Blood doping: Infusions, erythropoietin and artificial blood. 37(45):389-391, 2007. Gonzalez A, Nutt D: Gamma hydroxyl butyrate abuse and dependency. Psycho­ pharmacology 19(2):195-204, 2005. Haisma H, de Hon O: Gene doping. Int J Sports Med 27(4):257-266, 2006. Hauk J, Hosey R: Nitric oxide therapy: Fact or fiction? Curr Sports Med Rep 5(4):199-202, 2006. Magkos F, Kavouras S: Caffeine and ephedrine: Physiological, metabolic and ­performance-enhancing effects. Sports Med 34(13):871-889, 2004. Sweeny H: Gene Doping. Sci Am 291(1):62-69, 2004. Volek J, Ratamess N, Rubin MR, et al: The effect of creatine supplementation on muscular performance and body composition responses to short-term resistance training overreaching. Eur J Appl Physiol 10:628-637, 2003. Yesalis C (ed): Anabolic Steroids in Sport and Exercise, 2nd ed. Champaign, Ill, Human Kinetics, 2000.

R eferences Please see www.expertconsult.com

424

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n B

Pharmacology 2. Sports Pharmacology: Recreational Drug Use Edward R. McDevitt

Athletes are told to “play hard.” Athletes often feel that they deserve to “party hard.” Many studies document widespread use of substances consumed by athletes for recreational and social reasons. The Monitoring the Future survey is an ongoing study of the behaviors, attitudes, and values of American secondary school students, college students, and young adults.1 The 2006 survey shows that nearly half of American 12th graders are regular alcohol users, and nearly one fourth of American 12th graders regularly smoke. Almost 40% of American students have used an illicit drug during the past year. There is a tremendous variety of illicit drugs from which to choose. A great deal of conflicting research has been generated in an attempt to determine the prevalence of drug use in athletes when compared with nonathletes. It appears that there is little difference in the prevalence of recreational drug use in athletes and nonathletes. In both groups, use of recreational drugs is significant. Clinicians need to be aware of the effects of these drugs on their athletes. Athletes will not be looking to their physicians for advice and expertise regarding recreational drugs. Physicians taking care of athletes need to educate themselves about the commonly used recreational drugs.

ALCOHOL Alcohol is the number one substance abused by athletes from middle school through the professional ranks.2 Research has indicated that collegiate athletes, compared with nonathletes, had a higher rate of binge drinking and that athletes also experienced more alcohol-related problems such as academic problems, drinking and driving, gambling, illicit drug use, and sexual promiscuity.3,4 Heavy episodic or “binge” drinking for men was defined as consuming five or more alcoholic drinks on at least one occasion in the past 2 weeks, and for women, the number of drinks was four or more. According to the Harvard School of Public Health Alcohol Study, 80% of college students drink. Forty-eight percent of these students feel that drinking to get drunk is an important reason to drink. More than 1700 students die each year from alcohol-related incidents. More than 130 deaths from hazing at fraternities have been reported. Other consequences of alcohol abuse include sexual assault, violence, criminal activity, health problems, and academic problems.5 Of most concern is the relationship between alcohol and impaired driving. Alcohol is implicated in morbidity and mortality from trauma. Exposure to alcohol is generally measured in blood alcohol

concentration (BAC) rather than drinks per day or week. In most of the United States, the legal BAC limit for driving is 0.08%. This corresponds to about four drinks for a 200-pound man but only 2.5 drinks for a 150-pound woman. However, the risk for involvement in a collision while driving doubles at a BAC of only 0.05%. Furthermore, simulated driving ability is impaired with BACs as low as 0.02%.6 Inability to operate a motor vehicle safely can occur after even a single drink. Yet, most athletes drink a great deal more than one drink. Studies that analyze the reasons for binge drinking in college students report that there are a variety of reasons that student athletes give for drinking, some of which include the belief that alcohol can help in athletic performance and injury recovery and can improve muscle mass. In 1982, the American College of Sports Medicine issued a position statement regarding alcohol and sports performance, and even though the statement was written more than a quarter of a century ago, its conclusions continue to be validated by further studies.7 1. Alcohol adversely affects coordination, balance, and accuracy. 2. Alcohol does not improve athletic performance. 3. Alcohol will not improve muscle work performance and negatively affects the ability to perform. 4. Alcohol may impair temperature regulation during prolonged exercise in a cold environment. Despite these adverse effects of alcohol and athletic performance, the reality is that athletes will continue to drink, often excessively. The television show Friday Night Lights follows a Texas high school football team. In the initial fall 2007 episode, a coach warns his 16-year-old daughter to report to him concerning any athletes drinking at a post–football game party that is going to occur that evening. The very next scene is a large party where all the participants, including the coach’s child and the entire high school team, are drinking alcohol in large plastic cups. No parents are seen at this party. No point is made in this particular episode about the extensive underage drinking. There are no sad consequences such as a car fatality. It is all very blasé, all very commonplace.8 However, this episode stands as a powerful statement about the state of athletics and alcohol at all levels, including high school. Athletes are drinking, and drinking heavily, even in high school, and team physicians need to know how to deal with the consequences of alcohol use by their players.

Nutrition, Pharmacology, and Psychology in Sports

425

The Surgeon General’s Office recently released the first Call to Action against Underage Drinking.9 The Surgeon General’s Call to Action to Prevent and Reduce Underage Drinking documents alcohol use by American children and teenagers; it is not just about spring break, and it is not just about post–football game parties. As early as ages 8 and 9 years, children are confronted with decisions about alcohol on a regular basis in many settings, including at home and at school. Nearly 20% of 14-year-olds say they have been drunk at least once. The Call to Action attempts to focus national attention on the problem of underage drinking. Parents or coaches often do not believe that it can happen to their charges. Yet by age 15 years, 50% of American boys and girls have consumed alcohol; the highest prevalence of alcohol dependence in any age group is among individuals aged 18 to 20. Young people who start drinking before the age of 15 years are 5 times more likely to have alcohol-related problems later in life. Six goals have been identified in the Call to Action:

to drink because they believe that “everyone else is doing it.” The basic idea behind a social norms marketing campaign is to turn this dynamic around by using campus-based media to inform students about the true levels of alcohol consumption among their peers. The actual levels of alcohol consumption among college students are much lower than students perceive. Having accurate information about college alcohol use is hypothesized to lead to changes in perceptions of drinking norms on campus and, in turn, may lead to fewer students engaging in high-risk drinking.11 Despite campaigns to decrease binge drinking by ­athletes, the present reality is that excessive drinking is going to occur. Team physicians should work closely with coaching staffs and athletic departments to develop not only a program of preventive measures but also rules and ­consequences of alcohol-related offenses. No one wants to see a star athlete suspended, but it is far preferable to deal with a suspension than with an alcohol-related death or career-ending injury after a night of drinking.

1. Foster changes in society that facilitate healthy adolescent development and that help prevent and reduce underage drinking. 2. Engage parents, schools, communities, all levels of government, all social systems that interface with youths, and youths themselves in a coordinated national effort to prevent and reduce underage drinking and its consequences. 3. Promote an understanding of underage alcohol consumption in the context of human development and maturation that takes into account individual adolescent characteristics as well as environmental, ethnic, cultural, and gender differences. 4. Conduct additional research on adolescent alcohol use and its relationship to development. 5. Work to improve public health surveillance on underage drinking and on population-based risk factors for this behavior. 6. Work to ensure that policies at all levels are consistent with the national goal of preventing and reducing underage alcohol consumption.

MARIJUANA

One attempt to lower drinking problems at college campuses has been a “social norm” marketing campaign. Although high-risk drinking and its related negative consequences are a serious problem at institutions of higher education and a key concern for team physicians, few studies have examined the effectiveness of environmentally based prevention strategies. New programs and policies designed to reshape the physical, social, legal, and economic environments in which students make decisions about their alcohol use have been developed. Several colleges and universities have implemented social norms marketing campaigns with promising results, some of which have seen 20% reductions in high-risk drinking after just 2 years of campaign implementation.10 The theory regarding social norms marketing is that college students tend to overestimate the number of their peers who engage in high-risk alcohol consumption. This misperception is believed to influence students to drink more heavily by changing their perceptions of expected behavior concerning drinking. Students may feel pressured

Marijuana is the second most commonly abused drug by athletes.2 Marijuana is the most commonly used illicit drug.11 According to the 2005 National Survey on Drug Use and Health (NSDUH), an estimated 97.5 million Americans aged 12 years or older tried marijuana at least once in their lifetimes, representing 40.1% of the U.S. population in that age group. The number of past-year marijuana users in 2005 was about 25.4 million (10.4% of the population aged 12 years or older) and the number of past-month marijuana users was 14.6 million (6.0%).12 Among 12- to 17-year-olds surveyed as part of the 2005 NSDUH, 6.8% reported past-month marijuana use. Additional NSDUH results indicate that 16.6% of 18- to 25-year-olds and 4.1% of those aged 26 years or older reported past-month use of marijuana.12 The 2005 NSDUH results also indicate that there were 2.1 million persons aged 12 years or older who had used marijuana for the first time within the preceding 12 months.12 Marijuana has been used to produce euphoria for centuries, dating back to the Chinese, perhaps as long ago as 2737 bce.13 It is usually smoked as a cigarette called a “joint” or in a pipe or bong, although it can be taken as a drink or in foods. Marijuana is a green, brown, or gray mixture of dried, shredded leaves, stems, seeds, and flowers of the hemp plant (Cannabis sativa). Cannabis is a term that refers to marijuana and other drugs made from the same plant. Other forms of cannabis include hashish and hash oil. All forms of cannabis are mind-altering (psychoactive) drugs. The main active chemical in marijuana is THC (∆-9-­tetrahydrocannabinol). Short-term effects of marijuana use include problems with memory and learning, distorted perception, difficulty in thinking and problem solving, loss of coordination, increased heart rate, and anxiety. These side effects do not deter athletes from using marijuana. According to the National Collegiate ­Athletic Association (NCAA) studies, 53% of athletes using marijuana use it at least monthly, whereas a quarter of the respondents reported greater than 40 experiences per year.2 An ominous finding from the 2002 Substance Abuse and Mental Health Service Administration report, ­Initiation of

426

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Marijuana Use: Trends, Patterns and Implications, concluded that the younger children are when they first use marijuana, the more likely they are to use cocaine and heroin and become dependent on drugs as adults. The report found that 62% of adults aged 26 years or older who initiated marijuana use before they were 15 years old reported that they had used cocaine in their lifetime. More than 9% reported they had used heroin. This compares with a 0.6% rate of lifetime use of cocaine and a 0.1% rate of lifetime use of heroin in those who never used marijuana.14 Even though it is illegal to use marijuana, many consider it a harmless drug. A study released in 2006 by a pro–­marijuana legalization organization stated that marijuana is the United States’ most valuable crop. Contrasting government figures for traditional crops like corn and wheat against the study’s projections for marijuana production, the report cites marijuana as the top cash crop in 12 states and among the top three cash crops in 30 states. The study estimates that marijuana production, at a value of $35.8 billion, exceeds the combined value of corn ($23.3 billion) and wheat ($7.5 billion).15 Athletes have access to and are using ­marijuana, and there is controversy in the medical literature concerning the dangers and benefits of marijuana. Marijuana is absorbed readily and has effects throughout the body, most predominantly in the central nervous system (CNS), the respiratory system, and the cardiovascular system. Surprisingly few studies of neurotoxicity have been published, and the results have been equivocal. There is evidence that chronic administration of large doses of THC leads to residual changes in rat behaviors that are believed to be controlled by the hippocampus. There is evidence for long-term changes in hippocampal ultrastructure and morphology in rodents and monkeys exposed to THC. Animal neurobehavioral toxicity is characterized by residual impairment in learning, electroencephalographic and biochemical alterations, impaired motivation, and impaired ability to exhibit appropriate adaptive behavior. Although extrapolation to humans is not possible, the results of these experimental studies have demonstrated cannabinoid toxicity at doses comparable to those consumed by humans using cannabis several times a day. There is evidence from human research to suggest that cannabinoids act on the hippocampus, producing behavioral changes similar to those caused by traumatic injury to that region.16 Human research has revealed evidence of acute central nervous system changes following significant cannabis administration; whether or not these changes are permanent has not been established.16 The psychotropic effects of cannabis include general euphoria, a mild release from inhibitions, and in certain cases, some distortion of sensory perception. Some patients also experience drowsiness, a stimulated appetite, and a freedom from anxiety, whereas others anticipate a more intense experience such as an altered state of consciousness.17 Recent studies document a spectrum of respiratory ailments associated with marijuana smoking. A large epidemiologic study suggested that marijuana smoke can cause the same types of respiratory damage as tobacco smoke.18 In a study from New Zealand, a dose-response relationship was found between cannabis smoking and reduced lung function. For measures of airflow obstruction, one cannabis joint had a similar effect to between 2.5 and

6 tobacco cigarettes.19 One reason for the increased danger with marijuana is that users typically hold their breath 4 times as long as tobacco smokers after inhaling, and marijuana deposits significantly more tar and known carcinogens within the airways.20 The acute physiologic effects of marijuana on the cardiovascular system include a substantial dose-dependent increase in heart rate, associated with a mild increase in blood pressure. Orthostatic hypotension may occur acutely as a result of decreased vascular resistance. Smoking marijuana decreases exercise test duration in maximal exercise tests and increases the heart rate at submaximal levels of exercise. The cardiovascular responses that occur in response to THC are mediated by the autonomic nervous system. Although several mechanisms exist by which marijuana use might contribute to the development of chronic cardiovascular conditions or acutely trigger cardiovascular events, there are few data regarding marijuana or THC use and cardiovascular disease outcomes. A large cohort study showed no association of marijuana use with cardiovascular disease hospitalization or mortality.21 Few studies have looked at the effect of marijuana on sports performance. To evaluate the effects of marijuana smoking on exercise performance, 12 healthy young subjects underwent progressive exercise testing on an ergo cycle to exhaustion under two conditions: a nonsmoking control and 10 minutes after smoking a marijuana cigarette. Heart rate, arterial blood pressure, minute ventilation, breathing rate, oxygen uptake (Vo2), and carbon dioxide output (Vco2) were measured before, during, and for 4 minutes after the exercise. The exercise duration was also measured. Marijuana smoking reduced exercise duration but there were no differences in Vo2, Vco2, heart rate, and breathing ventilation between the two experimental conditions. Marijuana induced tachycardia at preexercise that was sustained up to 80% of maximal effort and during the recovery period. After marijuana, Vo2 and Vco2 were increased above control levels from 50% of maximal effort to the end of the test. Marijuana induced bronchodilation, which was still present after exercise. Marijuana had no effect on tidal volume, arterial blood pressure, and carboxyhemoglobin levels. Overall effects of marijuana were therefore a reduction of maximal exercise performance with premature achievement of maximal oxygen uptake.22 Discussions with athletes should not only emphasize the negative effects of marijuana on performance but also stress the fact, often forgotten, that marijuana is a Schedule I drug, and recreational use of marijuana is illegal. For the fourth year in a row, U.S. marijuana arrests set an all-time record in 2006, according to the Federal Bureau of Investigation Uniform Crime Reports. Marijuana arrests totaled 829,627, an increase from 786,545 in 2005. Similar to previous years, 738,916, or 89% of cases, were for possession, not sale or manufacture, and marijuana possession arrests again exceeded arrests for all violent crimes combined.23

TOBACCO Currently, about 48 million individuals in the United States are cigarette smokers, including 26% of men and 22% of women. Smoking is responsible for about 450,000 preventable U.S. deaths annually. A lifelong smoker has

Nutrition, Pharmacology, and Psychology in Sports

about a one in three chance of dying prematurely from a complication of smoking. Smoking is the major preventable cause of death in developed countries. Other forms of tobacco use include pipes and cigars and smokeless tobacco. Smokeless tobacco use in the United States is primarily oral snuff and chewing tobacco, whereas nasal snuff is used to a greater extent in the United Kingdom. Tobacco smoke is an aerosol of droplets containing water, nicotine, and tar. Tobacco smoke contains several thousand different chemicals, many of which may contribute to human disease. Tobacco smoke may produce illness by way of systemic absorption of toxins, or oxidant gases may cause direct pulmonary injury. There are more than 4000 chemicals found in the smoke of tobacco products. Of these, nicotine, first identified in the early 1800s, is the primary reinforcing component of tobacco that acts on the brain. By inhaling tobacco smoke, the average smoker takes in 1 to 2 mg of nicotine per cigarette.24 When tobacco is smoked, nicotine rapidly reaches peak levels in the bloodstream and enters the brain. A typical smoker will take 10 puffs on a cigarette over a period of 5 minutes. Thus, a person who smokes 20 cigarettes daily gets 200 doses of nicotine to the brain. Immediately after absorption, nicotine stimulates the adrenal glands to discharge epinephrine (adrenaline). The rush of adrenaline stimulates the body and causes a glucose burst as well as an increase in blood pressure, respiration, and heart rate. Tobacco use is motivated primarily by the desire for nicotine. Nicotine is absorbed rapidly from tobacco smoke to the brain through the pulmonary circulation. It acts on nicotinic receptors to produce its gratifying effects, which occur within 10 seconds after a puff. Smokeless tobacco is absorbed more slowly with less intense pharmacologic effects. With long-term use of tobacco, physical dependence develops. When tobacco is unavailable, even for only a few hours, withdrawal symptoms often occur, including anxiety, irritability, hunger, depression, and a craving for tobacco.25 An improved overall understanding of addiction and of nicotine as an addictive drug has been instrumental in developing medications and behavioral treatments for tobacco addiction. The nicotine patch and gum, readily available, have proved effective for smoking cessation, especially if combined with behavioral therapy. It is very difficult to stop using tobacco products, not only for the physical addiction but also for the social reasons that individuals smoke. One persistent image that permeates American popular culture is the sharing of a cigarette after a satisfying sexual encounter. There appears to be something sexy and dangerous about smoking that many people enjoy, and as with many sexy and dangerous activities in society, most individuals participating in them do not feel that something catastrophic will happen to them. Many athletes fall victim to this dangerous misconception. Although there are athletes who smoke cigarettes, the bigger problem in athletics has to do with smokeless or “spit” tobacco. According to the 2005 Monitoring the Future national survey, nearly 8% of high school seniors reported using smokeless tobacco in the previous 30 days. This rate has not declined substantially in the past several years.26 Baseball appears to be a sport with high spit tobacco use. In a study of California high schools, researchers found that 46% of senior players had tried spit tobacco

427

and that 12% were current users.27 The numbers of professional baseball players who use spit tobacco appears to be dropping, yet it is a common occurrence to watch a televised baseball game and see a player with a “pinch between the cheek and gum.” Unfortunately, that player is being observed by millions of television watchers around the world, and many may see spit tobacco as something positive because of its association with the player. At least 70% of all major league baseball players do not chew or dip. Surveys show that two out of three players who use smokeless (spit) tobacco would like to quit! More than half of the players who chew or dip report gum problems and dental disease. Recently, both professional ­baseball (minor leagues) and junior hockey (­Western Hockey League) have banned the use of spit tobacco by players, coaches, and officials.28 Yet many athletes are drawn to spit tobacco. In the mid-19th century, spit tobacco was the most popular form of tobacco, and as baseball became popular in those times, players used spit tobacco. The spit tobacco served a practical purpose. Players used it to keep their mouths moist in the dusty parks of the day, and the saliva was used to soften their gloves. The popularity of spit tobacco dropped later in the century when it was associated with spitting and the possible transmission of tuberculosis, a deadly disease at the time. In the 1950s, cigarette manufacturers started to become commercial sponsors of major league baseball, and many of the players smoked the “team brand.” In the 1970s, the dangers of smoking became clearer, and many players switched to what they thought was safer, spit tobacco. In those years when you made the major league roster, as you went to your locker for the first time, you would find a free case of spit tobacco, placed by a teammate on behalf of the tobacco company, right next to your glove, hat, and uniform.29 Teams received the spit tobacco free of charge as part of the tobacco company’s marketing campaign. The marketing was very effective. Teen use of spit tobacco grew substantially. However, it soon became apparent that spit tobacco was not a safe alternative to smoking and had its own unique and potentially fatal risks.30 In 1990, Major League Baseball issued a report on the hazards of smokeless tobacco and announced new efforts to help players beat the habit and to help prevent the next generation from getting “hooked.” Addiction to nicotine from spit tobacco has been clearly established.31 Chewing tobacco and snuff contain 28 carcinogens (cancer-causing agents). The most harmful carcinogens in smokeless tobacco are the tobacco-specific nitrosamines (TSNAs). They are formed during the growing, curing, fermenting, and aging of tobacco. TSNAs have been detected in some smokeless tobacco products at levels many times higher than levels of other types of nitrosamines found in foods, such as beer and bacon. All tobacco, including spit tobacco, contains nicotine, which is addictive. The amount of nicotine absorbed from spit tobacco is 4 times the amount delivered by a cigarette. Nicotine is absorbed more slowly from spit tobacco than from cigarettes, but more nicotine per dose is absorbed from spit tobacco than from cigarettes. Nicotine stays in the bloodstream for a longer time with spit tobacco.32 Aside from addiction, there are many other adverse effects of spit tobacco.

428

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Nicotine acts as both a stimulant and depressant. It increases bowel activity, saliva production, and bronchial secretions. It stimulates the nervous system and may cause tremors or convulsions with high doses. As a euphoric agent, nicotine often helps athletes deal with stressful ­situations. On average, tobacco increases heart rate 10 to 20 beats/ minute, and it increases blood pressure by 5 to 10 mm Hg through its vasoactive properties. Nicotine may also cause sweating, nausea, and diarrhea. Nicotine elevates the blood level of glucose and increases insulin production. Spit tobacco also contains a significant amount of glucose, which adds to the insulin release and often leads to dental caries. Nicotine also stimulates platelet aggregation, which may lead to blood clots. Nicotine temporarily stimulates memory and alertness. It also tends to be an appetite suppressant. Many individuals find they gain weight after stopping tobacco products, which leads many back to its use. Nicotine is considered mood and behavior altering. Tobacco is believed to have an addictive potential comparable to alcohol, cocaine, and morphine.33 Some of the other effects of smokeless tobacco use include oral leukoplakia (white mouth lesions that can become cancerous), gum disease, and gum recession. Possible increased risks for heart disease, diabetes, and reproductive problems are being studied. Smokeless tobacco users increase their risk for cancer of the oral cavity. Oral cancer can include cancer of the lip, tongue, cheeks, gums, and the floor and roof of the mouth. People who use oral snuff for a long time have a much greater risk for cancer of the cheek and gum. The American Cancer Society has published a poster to warn athletes of the dangers of spit tobacco (Fig. 8B2-1).

Figure 8B2-1   Poster published by the American Cancer Society to warn athletes of the dangers of spit tobacco.

The NCAA bans the practice of spit tobacco use by players, coaches, and officials during NCAA-sanctioned events. Teams have attempted to help players by emphasizing the importance of dental screening of users. Dental screening, which can often detect precancerous lesions, is highly recommended for all athletes. Addiction to nicotine is very difficult to break.34 Combinations of nicotine gum, patches, and newer medications such as bupropion (Zyban) and varenicline tartrate (Chantix) can be used. It is much better never to start. Adolescent brains differ from adult brains in their response to drugs. Epidemiologic studies indicate that there is an increased likelihood for the development of nicotine addiction when cigarette smoking starts during adolescence. There could be a critical period in the teen years during which drugs of abuse have distinct effects responsible for the development of dependence later in life. In an experiment in rats, exposure to nicotine during periadolescence, but no similar exposure in the postadolescent period, increased the intravenous self-administration of nicotine in adult animals. This study suggests an increased sensitivity to the addictive properties of nicotine in the teenaged rat. Adolescence appears to be a critical developmental period, characterized by enhanced ­vulnerability to the addictive actions of nicotine.35 Addressing the dangers of nicotine and spit tobacco in youth sports is recommended to help athletes before they develop a habit difficult to break.

COCAINE Much of the world’s supply of cocaine is produced in South America. The ancient Inca tribes of Peru considered cocaine a gift of the gods. Thousands of years ago, the ancient Incas of Peru chewed coca leaves because it made it possible for them to work in the high mountains of the land for longer periods. Ancient Incan athletes used the coca leaves to improve their endurance in games. In the 1880s, Bolivian soldiers were given the drug to help them gain endurance and overcome fatigue. Sigmund Freud in 1882 advocated the therapeutic effects of cocaine for a variety of medical reasons and was himself a frequent cocaine user.36 John Pemberton applied for a patent in 1885 for a drink combining cocaine and caffeine that eventually became known as Coca Cola.37 Cocaine was removed from the Coca Cola formulary in 1903, but athletes continue the love affair with Coca Cola. Athletes continue to find newer means of obtaining cocaine not for sport but for recreational reasons. The Monitoring the Future survey documented that in 2006, 8.5% of 12th graders had used cocaine.38 This percentage has stayed largely stable since 1991, but is nevertheless a sobering statistic. A 1997 NCAA survey revealed that recreational cocaine use in college athletes was 87%. Only 3.9% took cocaine for some perceived athletic benefit.39 The death of sports stars Don Rogers, Len Bias, and Reggie Lewis brought the world’s attention to the ­dangers of recreational use of cocaine in athletes.40 Why does cocaine have the potential to kill? Cocaine is a naturally occurring alkaloid present in the leaves of Erythroxylon coca. Cocaine is commercially available and can be applied to mucous membranes of the oral, laryngeal, and nasal cavities for use as a topical anesthetic.

Nutrition, Pharmacology, and Psychology in Sports

Cocaine causes significant euphoria, and abuse can lead to physical dependence. Despite being an excellent local anesthetic, the risk for abuse and the intense local vasoconstriction it produces prevent cocaine from being widely used clinically. It is a controlled substance (Schedule II) and is banned and tested for in athletes by the International Olympic Committee. Once topically applied, cocaine decreases nerve permeability to sodium. This stabilizes the nerve membrane, increasing the threshold of electrical excitability and inhibiting depolarization, and results in the failure to propagate an action potential and initiate or conduct nerve impulses.41 Unlike other local anesthetics, cocaine also affects the nervous system by potentiating catecholamines. Centrally, the actions of cocaine are presumed to include stimulation of presynaptic release of norepinephrine combined with inhibition of presynaptic reuptake of norepinephrine, dopamine, and serotonin. Cocaine causes an acute dopamine release and inhibits dopamine reuptake in the synapse, which leads to some of cocaine’s most striking effects. Cocaine induces euphoria, reduces fatigue, stimulates sexual desire, and increases mental ability and sociality. At higher increased doses, tremors and tonic-clonic convulsions may occur. Vomiting may be intense. Mental stimulation is soon followed by depression, with the medullary centers becoming depressed. Death may occur from respiratory failure. Yet, the dangers to the cardiovascular system are the most worrisome for athletes. By blocking the presynaptic uptake of catecholamines and dopamine, cocaine increases the postsynaptic sympathetic activation and dopaminergic receptor stimulation. These sympathomimetic effects result in an augmentation of ventricular contractility, blood pressure, heart rate, and escalating myocardial oxygen demand. Cocaine contributes to myocardial ischemia in other ways, including inducing coronary vasoconstriction, stimulating platelet aggregation and, in animal studies, accelerating atherosclerosis. The ensuing supply-demand deficit may manifest as angina.42 Cocaine also promotes platelet aggregation and affects endothelial cell function, which may potentiate endothelial damage and thrombosis at sites of cocaineinduced vasospasm.43 In vitro studies have demonstrated that cocaine increases platelet activation, enhances platelet aggregation, and augments thromboxane production. Cocaine-induced coronary spasm may lead to acute myocardial infarction and death. Another danger for athletes, especially if they take cocaine before participation in sports or if they participate in warm or humid conditions, is that cocaine is markedly pyrogenic. Cocaine increases muscular activity, which augments heat production, and vasoconstriction decreases heat loss. Cocaine can lead to heat stroke and death. Cocaine causes a plethora of side effects beyond its cardiopulmonary toxicities. In general, the adverse effects of cocaine are a result of its promotion of excessive sympathetic activity. Adverse reactions can occur with as little as 20 mg; the fatal dose has been reported to be 1.2 g orally. Patient sensitivity to cocaine is highly variable. Long-term use of cocaine can result in nasal congestion and chronic rhinitis, including sneezing or sniffling, which may lead to chronic sinusitis and increased risk for upper respiratory infection. Ischemic damage to the mucosa may occur and

429

may lead to atrophy, septal tissue necrosis, and septum ­perforation. CNS toxicity is extremely common with cocaine use. Symptoms of CNS stimulation may include agitation, anxiety, apprehension, confusion, headache, dizziness, emotional lability, euphoria, excitement, hallucinations, seizures, and psychosis. First-time use of cocaine can result in seizure, with single generalized motor seizures being most common, but multiple seizures and status epilepticus can occur immediately or within 2 hours of ingestion. Rhabdomyolysis can occur with myoglobinuria, causing renal tubular obstruction. Gastrointestinal adverse reactions to cocaine include abdominal pain, nausea, vomiting, and bowel ischemia. Cocaine may cause a transient erythrocytosis, increasing blood viscosity while maintaining tissue oxygenation during vasoconstriction. Cocaine, as a drug with the potential for abuse, may lead to tolerance, dependence, and drug addiction. Cocaine does not produce physical dependence, but psychological dependence does occur. A cocaine withdrawal syndrome is not observed with all patients and is more likely to include depressed mood, drug cravings, and other behavioral effects rather than physical symptoms on abrupt discontinuation of the drug. Cocaine dependence is a significant problem in our society, associated not only with a spectrum of medical complications but also with crime and violence. Prevention through use of an educational program that clearly outlines the dangers of cocaine to athletes is recommended.

INHALANTS Inhalant use is the deliberate inhalation of volatile substances through sniffing, snorting, or huffing, to induce a mind-altering effect. It is an important yet underrecognized form of recreational drug use in athletes. National surveys of adolescents in the United States have found that, after marijuana, inhalants were the second most widely used class of illicit drugs for 8th and 10th graders. The peak age of inhalant abuse is 14 years, with some children starting as young as 5 years of age.44 The Monitoring the Future survey results consistently show that inhalant abuse decreases during the high school years. This recrea­ tional drug is the only one that appears to be more common in younger age groups. Surveys show that young children are drawn to this behavior because of the low cost and easy availability of the inhalant agents, and a perception that inhalants are fun and safe. Inhalant agents are not safe. The most commonly used inhalants are glue, shoe polish, and gasoline. Young girls are as likely as young boys to use inhalants. Among all American racial and ethnic groups, Native Americans have the highest rates of use. Inhalant use is often associated with impoverished living conditions, single-parent families, and familial abuse and drug use.45 The American Academy of Pediatrics has taken a major stand to educate clinicians, parents, and children of the dangers of inhalant abuse.46 Inhalants are divided into three groups on the basis of the inhalant pharmacology. Type I agents include volatile solvents such as paint thinner, acetone, glue, rubber cement, butane, aerosols, hair spray, and gasoline. Type II agents are nitrous oxide and whipping cream aerosols, which contain nitrous oxide. Type III agents include volatile alkyl nitrites known as

430

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

­ oppers, snappers, boppers, and Amys. Media attention p was directed to inhalant agents in the 1950s when it was discovered that children were sniffing glue and other household products.47 Recent surveys reveal that inhalant abuse is increasing. National surveys indicate that more than 12.5 million Americans have abused inhalants at least once in their life. Initial use of inhalants often starts early, often in elementary school. According to the National Institute on Drug Abuse (NIDA), about one in five 8th graders has abused inhalants. Most inhalant abuse occurs after dinner between 6 pm and 8 pm.48 Inhalant agents are abused through a variety of methods. If the substance is a glue or solid substance, the user will empty the substance into a bag and hold the bag to the nose and inhale through the nose (snorting or sniffing) or the mouth (huffing). Putting your entire head into a bag is called bagging. Soaking a rag with the chemical and holding the rag to the mouth is another huffing technique. Glading is the inhalation of air-fresheners (GLADE), and dusting is inhalation of computer cleaning aerosols. A simple but potentially deadly technique is to spray substances directly into the mouth. Inhalants contain different solvents, each with its own unique toxicity. Inhalants are readily absorbed through the lungs, with immediate transient effects, and then cross cell membranes and quickly enter the CNS and other lipid-rich tissues. All inhalants cause CNS depression. Small hydrocarbons such as methane or propane are simple asphyxiants, producing CNS effects from hypoxia. Higher-molecularweight hydrocarbons and petroleum distillates directly affect the CNS. The precise mechanism by which acute abuse causes CNS effects is not known but may involve γ-amino butyric acid (GABA) agonism or altered neuronal membrane function.49 Some glues are metabolized to the axonal neurotoxin 2,5-hexanedione and cause a peripheral neuropathy characterized by muscle weakness and wasting, diminished deep tendon reflexes, decreased nerve conduction, and paresthesias. Inhalants are metabolized in the liver and can cause liver toxicity. Renal damage can be seen as a direct result of certain inhalants on renal tissue and can also occur from secondary inhalant-induced complications such as acidosis and hyperkalemia. Death can occur with inhalant agents and may result from cardiac complications secondary to asphyxia, ventricular fibrillation, or cardiac arrhythmia. Cerebral death may be secondary to asphyxia, cerebral edema, and hyperpyrexia.50 The sudden sniffing death was described by Bass.50 Inhalants sensitize the myocardium to epinephrine. A person then subjected to a sudden fright or shock may develop an acute and fatal arrhythmia. This unpredictable and unpreventable death leaves no specific autopsy findings. The sudden sniffing death can occur with a person’s initial use of an inhalant. The signs and symptoms of inhalant abuse are so varied and affect so many different systems that a high level of suspicion is required to make the diagnosis. Occasionally, odor from the abused substance may be detectable on the patient or his or her clothing. Patients who have been sniffing paint often have paint around the mouth, nose, and hands from huffing or bagging. Certainly, any adolescent who presents with unexplained mental status changes, cere­ bellar findings, cardiac dysrhythmia, syncope, or ­ cardiac

arrest should be evaluated for inhalation abuse. Burn injuries are seen in youths who inhale gasoline or ignite exhaled butane vapor (fire breathing). In general, there are no specific antidotes for inhalant toxicity. As for many recreational drugs, the best treatment involves prevention. Carrying the message of the dangers of inhalant agents to elementary and middle schools is needed to lower the incidence of this potentially deadly practice.

CONCLUSION We must be aware that recreational drug use in athletes is a reality. It is a great privilege to treat athletes, but with the privilege comes responsibility. We can be of greatest benefit to our athletes by not only repairing an anterior cruciate ligament tear or treating the stress fracture but also by doing all we can to educate our athletes, coaches, and parents on the value of preventive medicine. An athlete who is well informed about the dangers of recreational drugs may be best able to avoid the temptation of such drugs, which are all too easily available. If the coaches are also involved in the education process and emphasize the importance of sobriety and healthy habits as essential requirements of a successful team, it may be easier for the athlete to resist recreational drugs. All the drugs discussed in this chapter have been clearly demonstrated in the ongoing Monitoring the Future surveys to be used by our youth at very young ages, often in middle or elementary school. It is therefore critical that the education of athletes, coaches, and parents be started in the earliest recreational leagues. All the drugs being used have the potential for deadly consequences for the athlete and for society in general. The importance of team rules regarding recreational drugs can be invaluable, but it is also important that there are specific consequences to the rules. No one wants the star of the team to die tragically in an alcohol- or drug-related motor vehicular crash, but it may be very difficult to suspend the uninjured star athlete who breaks team rules regarding recreational drugs. Rules need to have specific consequences. Physicians, coaches, parents, and the athletes themselves need to “buy into” the importance of avoiding these potentially deadly temptations.

C

r i t i c a l

P

o i n t s

l Recreational drugs are substances used primarily for social reasons. l Athletes are going to use recreational drugs. l A wide variety of legal and illegal recreational drugs are readily available for athletes to use. l Many studies have delineated that dangerous side effects from recreational drugs are not rare and are often dangerous. These dangers will not deter most users. l Athletes rarely tell their physician that they take these drugs. l It is imperative that physicians who care for athletes recognize the signs and symptoms of recreational drug use.

Nutrition, Pharmacology, and Psychology in Sports

S U G G E S T E D

R E A D I N G S

Afonso L, Mohammad T, Thatai D: Crack whips the heart: A review of the cardiovascular toxicity of cocaine. Am J Cardiol 15:(100(6)):1040-1043, 2007. Avois L, Robinson N, Saudan C, et al: Central nervous system stimulants and sport practice. Br J Sports Med 40(Suppl1):i16-i20, 2006. Campos DR, Yonamine M, de Moraes Moreau RL: Marijuana as doping in sports. Sports Med 33(6):395-399, 2003. Lowinson JH, Ruiz P, Millman RB, et al: Substance Abuse: A Comprehensive Textbook, 4th ed. Philadelphia, Lippincott Williams & Wilkins, 2005. Martens MP, Dams-O’Connor K, Beck NC: A systematic review of college ­studentathlete drinking: Prevalence rates, sport-related factors, and interventions. J Subst Abuse Treat 31(3):305-316, 2006. Saugy M, Avois L, Saudan C, et al: Cannabis and sport. Br J Sports Med 40(Suppl1):i13-i15, 2006.

431

Sinusas K, Coroso JG: A 10-yr study of smokeless tobacco use in a professional baseball organization. Med Sci Sports Exerc 38(7):1204-1207, 2006. Sloboda Z, Butoski WJ (eds): Handbook of Drug Abuse Prevention: Theory, Science and Practice. New York, Plenum, 2003. Turrisi R, Mallett KA, Mastroleo NR, et al: Heavy drinking in college students: Who is at risk and what is being done about it? J Gen Psychol 133(4):401-420, 2006. Willams JF, Storck M, and the Committee on Substance Abuse and Committee on Native American Child Health: Inhalant abuse. Pediatrics 119(5):1009-1017, 2007.

R eferences Please see www.expertconsult.com

S ect i o n C

Psychological Adjustment to Athletic Injury William D. Parham

The injury experience that befalls some athletes and its subsequent rehabilitation process can trigger an interactively complex set of emotions that can make an unfortunate circumstance feel arduous and burdensome. At the same time, the injury-to-recovery process can also turn into an experience ripe with tremendous opportunity. Research efforts that address sport injury and the rehabilitation process are numerous.1-7 Yet, a clear picture of the nature, course, and resolution of the injury experience has not emerged. When current data on the psychology of athletic injury are synthesized into commonly accepted assumptions and observations, three foci frame the sets of factors that best describe the salient features of the injury experience. Understanding the injury experience involves developing an awareness of preinjury, during-rehabilitation, and post-rehabilitation variables. Preinjury variables include, but are not limited to, personality, psychological risk factors, and psychological states that predispose an athlete to injury. Duringrehabilitation variables include, but are not limited to, the kind, type, and degree of impairment imposed by the injury, the way in which the athlete views the recently sustained injury relative to self-perceptions, the athlete’s belief about his or her ability to successfully rehabilitate, and the athlete’s perception of available resources and ability and willingness to access help resources. Postrehabilitation factors include, but are not limited to, post-rehabilitation appraisal of abilities and talents and assessing the degree of self-determination to reintegrate back into the field of play.8

Identifying preinjury variables positions researchers and practitioners to maximize primary interventions, thereby decreasing injury attributable to behaviors and cognitions expressed by the athlete. Given myriad factors other than athlete-intrapersonal factors, including sport, environment, and equipment-related injury, prevention as a phenomenon is not likely. Identifying variables that potentially result in athletes knowing more about their susceptibility to injury and then using that information to make responsible decisions relative to managing their preinjury physical health, however, represents a more reasonable strategy. The bulk of the athlete injury research has illuminated during-rehabilitation variables. The kind, type, and degree of actual impairment provide some indication of what will be required physiologically for maximal healing to take place. The physiologic prognosis for an anterior cruciate ligament injury with respect to time needed for healing, for example, is different from that for an ankle sprain. Factors such as athlete self-perceptions and athlete awareness of and confidence in rehabilitation processes and resources provide a sneak peak into what will be required psychologically in order for injured athletes to manage their engagement of the injury and rehabilitation journey. To be clear, the cognitive appraisal of the injury event represents the key psychological component of the duringrehabilitation process. The meaning that an athlete assigns to the injury event in all of its manifestations determines the type of emotional response to the injury that the athlete will exhibit during the rehabilitation journey. The injury event stimulates focus on the degree to which an athlete’s

432

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

identity is connected to his or her involvement in athletics. For some athletes, involvement in athletics represents an all-consuming endeavor from which they receive ­abundant emotional nourishment. It is often not until an athlete incurs an injury that the degree of attachment to the sport becomes evident.9-19 This “attachment” realization is one of the more surprising features of the injury experience and one that can feed its emotional intensity. An injury can be experienced suddenly and unexpectedly, or it can have a gradual onset. Both kinds of injury experiences often stimulate a floodgate of competing emotions that can be dramatic and overwhelming. Injury is not an experience athletes plan for, despite knowing that injury is a reality that could be an experience at some point in their career. Athletes also have a tendency to believe they are invincible and that injury will not happen to them. Circumstances that influence the way in which an athlete thinks about a current injury include (1) the emotional reliance on the sport compared with other life domains, (2) the degree of perceived disruption in athletics (and perhaps in other areas of the athlete’s life), (3) the degree of perceived or real difference in abilities and skills upon re-entry to the sport, (4) the way in which the athlete has coped with past injuries with similar disruptive outcomes, and (5) the ability of the athlete to identify, access, and develop confidence in assistance resources. Physiologic, cognitive, and emotional factors also ­influence post-rehabilitation integration back into the sport as well as into life after sport in cases in which injury leads to termination of an athlete’s sports career. The actual stability and sustainability of the medical intervention is important in the post-rehabilitation phase for athletes to consider, as is an athlete’s concern about the full ­effectiveness of the intervention that must now be put to the test. When preinjury, during-rehabilitation and post­rehabilitation injury experiences are viewed within the context of an athlete’s current personal and environmental realities, the exponential complexity of the injury and rehabilitation experience can be more fully appreciated. Specifics of the three phases describing the injury and rehabilitation process are discussed later in this chapter. Finally, the injury-to-recovery process is also ripe with opportunities. From the onset of injury and through the completion of the rehabilitation process, the athletes will be afforded time to think critically about self in relation to the sport that heretofore has played such a pivotal role in their life. Athletes are also afforded an opportunity for self-assessment and stocktaking of goals achieved compared with goals unmet in other areas of their life, including academic, occupational, personal, familial, relational, and spiritual. This chapter offers for consideration a more detailed account of the injury and rehabilitation experience of athletes. The hope is to foster in the reader a conceptual appreciation of an athlete’s responses to injury juxtaposed against the physiologic healing that is also taking place. A second goal is to introduce the reader to the role of sport psychologists in terms of their participation as members of the athlete’s injury management team. Expanded roles for sport psychologists in helping athletes manage the injury experience are also identified.

PREVALENCE OF ATHLETIC INJURY Discussion of the athletic injury experience and concomitant rehabilitation process would be incomplete without an understanding of the frequency with which injury occurs. Despite enormous scientific and technologic advances, considerable improvement in the quality of athletic equipment and of sports and recreational facilities and in the way in which today’s athletes are conditioned and trained, sports participation and injury prevalence data across age and across sports yield some noteworthy findings. Competitive or recreational sports pursuits by children outside of their school experiences number almost 20 million. An additional 6 million students participate in high school sports. Annually, in excess of 3.5 million children aged 5 to 14 years present at hospital emergency departments, clinics, ambulatory surgery centers, and physicians’ offices for injuries related to their participation in sports.20 Participation of young children in football, basketball, baseball, soccer, hockey, gymnastics, and volleyball produces overuse injuries (small injuries to immature bodies) and acute injuries (caused by sudden trauma), including contusions, sprains, strains, and fractures.20 Children ages 5 to 14 years represent 40% of the sport injury–related visits to emergency rooms.21 About half of youth sports injuries occur during practice as opposed to during the game. Participation in high school sports from 2005 to 2006 accounted for 1.4 million injuries as a result of practice and competition.22 Football accounted for the highest injury rate, followed by wrestling, boys’ and girls’ soccer, girls’ basketball, boys’ basketball, volleyball, baseball, and softball. About 80% of the documented injuries were new and not recurring or previously experienced. More than half of the reported injuries called for more than 7 days away from participation. About 393,500 student athletes participate annually in National Collegiate Athletic Association (NCAA) sporting events.23 The average per-institution number of athletes was about 375, with more men than women participating. The NCAA Injury Surveillance System provides current data on injury trends in intercollegiate athletics across 16 sports. In the fall of 2006, for example, preseason injuries occurred at rates 2 to 3 times higher than during the regular season. More than half of the injuries occurred in the lower extremities. Rates of concussions and anterior cruciate ligament (ACL) injuries are on the rise. Rates of ­practice-related (4.7 injuries per 1000 athlete-­exposures) and game-related (44.9 injuries per 1000 athlete-­exposures) sports injuries in football were higher than the 25-year NCAA Injury Surveillance System football game (35.9 injuries per 1000 athlete-exposures) and practice (4.1 injuries per 1000 athlete-exposures) averages. The practice and game concussion rates have continued to rise during the past 4 years. Men’s practice (4.2 injuries per 1000 athleteexposures) and game (20.8 injuries per athlete-exposure) rates for soccer held constant during the past 14 years, and women’s soccer showed practice (5.1 injuries per 1000 ­athlete-exposures) injury rates slightly lower than the 14-year average of 5.8 injuries per 1000 athlete-exposures. Game (18.5 injuries per 1000 athlete-exposures) injury rates for women’s soccer showed a slight increase from the average

Nutrition, Pharmacology, and Psychology in Sports

of 17.7 injuries per 1000 athlete-exposures. Women’s volleyball and field hockey showed slightly lower practice (4.2 and 3.4 injuries per 1000 athlete-exposures, respectively) and game (4.3 and 5.9 injuries per 1000 athlete-exposures, respectively) injury rates compared with the 14-year averages of 4.5 (women’s volleyball) and 4.1 (women’s field hockey) injuries per 1000 athlete-exposures. In all these cases, 30% to 40% of the injuries required restricted or missed participation for 7 days or longer. The most prevalent body parts injured across these referenced sports were knee, ankle, and upper leg.24 Spring sports (e.g., softball, baseball, men’s and women’s lacrosse, and spring football) showed a similar practice and game injury rate pattern of slightly more or less than their respective 14-year averages. Sports that required restricted or missed participation for 7 days or more included softball (32%), baseball (43%), women’s lacrosse (33%), men’s lacrosse (30%), and spring football (48%). In addition to the knee, ankle, and upper leg injuries, spring sports recorded shoulder, head, and face injuries among the most prevalent.25 Data that profile sports-related injuries to people 65 years of age and older yield several interesting observations.26 In a 6-year span from 1990 to 1996, sports-related injuries to people in this age category increased significantly (54%) from 34,000 to about 53,000. In the same 6-year period, the increase in the population aged 65 years and older was just above 8%. Thus, increased injury to this age group cannot be attributable solely to the increase in the number of people. Athletes 75 years of age and older saw a 29% increase in sport-related injuries. The increases in the numbers of people in both age categories (65 years and older, 75 years and older) are in contrast to the 18% increase in sport-related injuries to athletes 25 to 64 years of age. Interestingly, most injuries to athletes 65 years and older were associated with bicycles. Equally interesting is the fact that injuries resulting from exercise activity increased 173% from 1990 to 1996. Finally, there is clear indication that some of the injuries to the population aged 65 years and older resulted from their participation in sports such as snowboarding and in-line skating. These injury data correspond to the increased number of people in their 70s, 80s, and 90s maintaining physically active lifestyles, including increased participation in organized and recreational sports.

THE INJURY EXPERIENCE AND THE REHABILITATION PROCESS Efforts to document the injury experience and the process of psychological adjustment to the athletic injury have increased considerably and have resulted in descriptions of the injury-to-recovery process that are less impressionistic than even 5 years ago. To date, however, a solid depiction and an agreed-on scenario of the injury rehabilitation experience has yet to emerge. Research has targeted injury prediction and injury response as foci for study.26a,27 Injury prediction research aims to identify variables that contribute to injury vulnerability. Included in this category are studies that took a look at injury proneness as it relates to ­personality,28-37 psychosocial risk factors,38-46 ­predictors

433

of susceptibility,47 and psychological states.48-52 Injury response research attempts to identify an athlete’s psychological reaction to athletic injury. Included in this category are studies that took a look at self-management concerns,5,53-55 psychological aspects of the return to participation in sports,3 injury and rehabilitation tools,56-59 perceptions of support,3,4,60,60a emotional responses,61-63 postinjury self-­perceptions,49,64,65 perceived benefits of injury,62,66-68 injury recovery and its relationship to a sequence of emotional stages through which injured athletes pass on the road to healing,63,69-75 psychological states not necessarily tied to stage or phase sequencing,42,43,64,76-95 and compliance with the rehabilitation regimen.71,83,96-109 Results in both injury prediction and injury response research are predictably and reliably mixed. Within the injury prediction category, for example, some studies conclude that factors such as personality,32,34,36,110 psychosocial risk factors,38-40,45,110-118 and various psychological states28,119 correlate highly with injury occurrence. In this vein, perhaps the most promising work within the psychosocial risk factors category are models presented by Andersen and Williams,38 Wiese-Bjornstal and colleagues,109 and Brewer and associates.6 These three models offer ways of understanding the relationship of athletic injury to a range of factors, including stress, personal, cognitive appraisal, situational, behavioral response, emotional response, sociodemographic, biologic, psychological, social and contextual, intermediate biopsychological, and sport injury and rehabilitation outcomes. The authors suggest a framework for predicting, and thus preventing, stress-related injuries, using cognitive, behavioral, physiologic, attentional, and interpersonal correlates as anchors to their model. Several other studies, however, using personality,33,35,39,76,120,121 psychosocial risk factors,110,117,122 and various psychological states,123 strongly suggest that injury occurrence has little to do with these factors or is questionable, at best. Likewise, research efforts that purport to address injury response70,72,73,91,94,124 suggest, for example, that factors such as perception of support, use of rehabilitation tools such as imagery, and self-management strategies positively influence management of the rehabilitation process. Injury-response research also suggests that the various stage models aid in understanding the process of psychological adjustment to injury.71,87,108,125 Several other studies suggest that stage models do a disservice because they do not accurately reflect universal psychological adjustment experiences.43,71,77 Caution needs to be exercised in asserting that either of these research domains merits favor over the other. Perusal of the collective works in the injury prediction and injury response categories reveals significant methodologic confounds along several dimensions, the sum of which leads to questions of generalizability. These confounds might also explain the tremendous variations in outcomes in the research. First, there is not a uniformly agreed-on definition of injury. Injury is conceptualized partly as forced time away from athletic participation. The challenge presented in both domains with use of this criterion is that time away from athletic participation varies from 1 day to more than 7 days. Variations in the number of days off make ­comparisons regarding physical and emotional ­adaptation

434

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

difficult. Research using injury severity as the topical focus (e.g., wherein injury is defined as catastrophic and traumatic126) alternates between stage models to explain the journey of the injured athlete and cognitive appraisal models.77 Although both models acknowledge physiologic as well as psychological adjustment correlates,125 they offer different ways of conceptualizing the emotional aftermath that the athlete experiences.83 This difference in conceptualization of the emotional aftermath comes irrespective of the way in which the injury or illness occurred (e.g., injury from competition, injury from physical illness, injury received outside of sport, or injury from a traumatic event). The net result, however, is a lack of agreement on how best to describe the injury-to-recovery process. A second confound has to do with the populations (e.g., college students, high school students, young children, and senior citizens) chosen for study.126a The challenges in this area stimulate the question of how representative the smaller sample is of the respective larger groups of athletes purportedly represented. Related to the population confound is the concern about the limited number of sport teams that have been studied. Collegiate football players have been the most frequently studied athlete population,27 but athletes representing other sports, such as gymnastics, swimming, alpine skiing, soccer, wrestling, baseball, volleyball, basketball, cross-country, track and field, snowboarding, skateboarding, and mountain biking, to name a few, have been studied as well. The research challenge from the standpoint of being able to generalize the results centers around addressing the degree to which athletic teams studied and those not studied share similar experiences. Careful review of the research in the injury prediction and injury response domains reveals concerns about the instruments used to test hypotheses, to measure injuryrelated and rehabilitation-related phenomena, and to evaluate outcomes.127 Each of the instruments used to test, measure, and evaluate the various populations is challenged psychometrically and is limited by its usual ability to assess only a single phenomenon. When multiple phenomena are likely to be contributing to the resultant observations of a given investigation, the limitations of the chosen instruments become even more salient. Finally, individual sport-related variables, such as size, strength, level of conditioning, and demands of the sport at the time the injury occurred, have not been factored into studies of athletic injury and rehabilitation. Individual personal variables, such as age, ethnicity, race, sexualaffectional orientation, and disability status, have also been conspicuously absent as factors that contribute to researchbased observations of injury and rehabilitation. The net result of the confounding elements associated with individual sport-related and individual personal variables points, again, to the compromised ability of sport psychology research to generalize to populations of athletes beyond the immediate teams under investigation. Despite these limitations in sport psychology research, cautious interpretation of the available empirically based data and of consistently cited anecdotal reports from sport psychology practitioners generates two sets of factors (e.g., time perspectives and cognitive appraisal) that seem to best characterize the driving force behind the injury and rehabilitation processes.

Two Sets of Factors and Contextual Considerations of the Injury Experience This section addresses the during-rehabilitation experience of athletes because this area dominates the scholarship domain in this area. That said, athlete preinjury research focuses largely on personal and external factors to which athletes can pay attention in hope of minimizing injury experiences. Most researchers caution against the use of the term injury prevention because involvement in sport is likely to produce minor to major injury during the course of an athlete’s career. Internal or personal factors, such as personality, level of commitment to the sport, biologic and physiologic predispositions, and existing or budding sport-specific skill sets, and external factors, mostly ­environmental, guide the pursuit of preinjury scholarship. Internal factors that serve as the foci of post­rehabilitation scholarship include athlete’s self-perception of ability to reintegrate into athletics, concerns upon reentry ­ regarding perceptions of others, level of commitment to become reestablished, and fears about reinjury. External factors that influence post-rehabilitation include actual perceptions of others and existing opportunities for re-entry to sport. Most athletic injury research focuses on the duringrehabilitation experience. The kind, type, and degree of impairment constitute the first set of factors that most sport psychology researchers and practitioners agree come into play in understanding the during-rehabilitation process. Actual physiologic compromise or damage provides sort of a diagnostic barometer of the injury-to-recovery process that the athlete must now experience. An injury in which the degree of impairment is slight, and for which the prognosis is relatively quick return to active participation in practice and game situations, might be experienced by the athlete as inconvenient, at best. On the other hand, an injury of significant magnitude, such as an ACL or back injury, which would require dismissal from active participation in the sport for 6 to 18 months, will generate a different, perhaps more intense emotional response. Personal factors (e.g., body build, level of conditioning, preexisting injuries) and structural factors (e.g., equipment failures, compromised playing fields or surfaces) influence injury predisposition. Intrapersonal factors (e.g., self-confidence, sense of control, confidence in the treatment team and treatment approach, commitment to comply with the treatment regimen) and resource access factors (access to nutritionist or nutrition information, availability of rehabilitation equipment, access to professionals trained to assist with a rehabilitation), coupled with an innately healing body, contribute to injury and rehabilitation. The actual degree of compromise sustained by the body and the kind of intervention and course of treatment needed to rehabilitate the injury are best determined by the orthopaedic surgeon or other physician. Trainers and physical therapists can assist in the administration of rehabilitation activities. The cognitive appraisal of an injury situation represents a second group of factors that most sport psychology researchers and practitioners agree contributes to the driving force behind the injury rehabilitation process.79,128,129 An injury event in and of itself is just an injury event. The way in which an athlete views the injury event determines

Nutrition, Pharmacology, and Psychology in Sports

the emotional valence of the situation as well as the degree of urgency with which the challenges imposed by the injury will be met. In essence, the cognitive appraisal of the injury event sets in motion all subsequent affective, behavioral, and cognitive responses. Furthermore, it is the key to understanding the complexity of the injury experience. For example, the cognitive appraisal of an injury event may result in an athlete’s believing that the event will have a traumatic, devastating, long-term, and adverse impact on his or her athletic career. If there had been no history of injury, and thus never a need to access rehabilitative services, what lies ahead with respect to the current rehabilitation challenges would be uncertain. In this instance, an injured athlete might also feel overwhelmed, helpless, even despairing. Subsequent feelings and behaviors might be seen in an athlete’s strict compliance with the established rehabilitation regimen. Compliance, in this case, might spring from the intent to ensure the quickest possible recovery to preinjury condition and form. The factors that influence the way an athlete views an injury event include (1) the degree to which the athlete relies on athletics as a source of emotional nourishment; (2) the degree of concurrent reliance on other life domains (e.g., academic, personal, familial, spiritual) for emotional nourishment; (3) the degree of perceived interference or disruption in the athlete’s current and future sport and other life domain–related activities; (4) the way in which the athlete has coped affectively, behaviorally, and cognitively with past injuries of similar emotional magnitude; (5) the ability to access available resources in the management of the current injury challenges; and (6) the confidence in the abilities and talents of the treatment team.

Emotional Nourishment With respect to the degree of emotional nourishment, some athletes receive applause and accolades abundantly, beginning from a young age when they are mastering basic athletic skills. Successful mastery, coupled with fine execution of the sport-related skills, results in praise and recognition from the family, peers, community, school, and media, which after a while begins to feel satisfying. Acknowledgment of fans and support from around the world through Internet communications add to the impact. Thus, as success in the sport increases, and as attention and recognition for athletic success also increase, the felt connection to the sport into which a lot of time and energy have been devoted strengthens, and an athlete’s identity with the sport becomes solidified. Not only does the athlete begin to identify emotionally with the sport, but family, peers, the immediate community, and the greater local, state, national, and international communities all find themselves attracted to a person over whose image they both marvel and fantasize. In time, athletes become accustomed to this rich source of emotional nourishment, and the slightest hint suggesting that this emotional nourishment will cease triggers anxiety, depression, and a host of other emotional responses. Athletic injury, therefore, represents a situation that can feel threatening to many athletes. For athletes whose emotional investment in the sport is considerable, injury often results in the perception that the emotional nourishment

435

on which they have come to rely will now be withdrawn because of their decreased ability to perform and, thus, earn it.

Reliance on Other Sources of Emotional Nourishment Reliance almost exclusively on athletics as a source of emotional nourishment carries a cost, namely, that other potential sources of praise, recognition, and favor for achieved success are overlooked and ignored. Family, friends, academic environments, work settings, and spiritual life activities represent other sources from which emotional nourishment and self-esteem energies are derived. Rarely, however, does the significance of these other sources of emotional support become evident until an athlete has an injury experience.

Interference and Disruption ����������� Success in athletics, particularly those for whom athletics has become a sole source of life satisfaction, breeds dreams and ambitions. These dreams and ambitions are fed by the lure of huge salaries and other perks that today’s professional sports careers seem to promise.130 The path on which these athletes choose to travel to achieve their athletic desires becomes clear and straightforward until injury occurs and the disruption in an athlete’s athletic life proves to be significant. Injury can introduce considerable interference in the life of an athlete, particularly one whose focus is almost exclusively on achieving “fortune and fame” according to a prescribed timetable. A dream deferred can be disheartening. Disruption is a relative experience, however, in that its degree is determined by the way the athlete thinks about the impact of the injury. An injury that is actually minor can seem to be just as devastating as one that is more significant. On the other hand, a major injury can seem not as disruptive when it is viewed from a perspective of considering injury, for example, an opportunity for stocktaking. In that same vein, an athlete who has experienced success in a sport yet finds extreme difficulty responding assertively to parents, family, and coaches regarding continued participation beyond the current season would conceivably experience the injury event as a sort of relief. In this instance, temptation to thwart ongoing compliance with the rehabilitation process might be high, as might an athlete’s anxiety, as return to practice and competition becomes more of a reality. Thus, teasing out the disruptive feature of the injury experience is important in understanding its emotional magnitude. Assessing the degree of disruption that an injury has imposed on an athlete’s life often provides the starting point for determining what needs to be worked out to achieve some measure of resolve about the injury.

Past Coping Athletes, like all people, are creatures of habit and pattern. Thus, how an athlete responded cognitively, affectively, and behaviorally to past injuries provides a clue to how the athlete will respond to a current injury situation of at least similar emotional magnitude. Let us assume, for example,

436

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

that past injuries resulted in an athlete’s approaching the injury rehabilitation challenge with a belief that injury is simply an unfortunate experience. Let us assume further that the athlete believed that a return to preinjury shape and conditioning is possible as long as there is compliance with prescribed treatment. Given these conditions, it seems reasonable to predict that the current injury and rehabilitation challenges will be approached in a similar mindset. Alternatively, let us assume that a cognitive, affective, and behavioral approach to past injuries resulted in the athlete’s feeling overwhelmed emotionally, including feeling angry and combative. Let it also be suggested that the athlete complied sporadically with rehabilitation. In this scenario, it seems reasonable to expect that the current injury rehabilitation challenge will be approached similarly.

Accessing Services and Confidence in the Treatment Team Injury creates an opportunity to access medical and other help resources. The kind, type, and degree of impairment initially influence where the athlete will turn for help. Factors such as perceived talents of the service providers and confidence in the proposed treatment plan contribute to athletes’ choosing a person or persons with whom they will work. Readiness to commit to the rehabilitation regimen and belief that the rehabilitation process will result in the desired outcome contribute to an athlete’s willingness to surrender to the rehabilitation process. Physicians are consulted most often about general medical conditions, and other allied health persons (e.g., nutritionists, trainers, and psychologists) enter the picture at various junctures of the rehabilitation process. The quality of various help resources varies considerably, as does access to the resources most desired and preferred. For example, financial status may contribute to an athlete’s feeling stymied by the ability to access resources of a desired caliber. For example, a young male Latino high school gymnast from an economically challenged background whose gymnastics talents and abilities could result in a full scholarship to college might experience considerable emotional distress with a recently received ACL injury. The athlete’s distress might be especially overwhelming when he acknowledges that the “expert” medical intervention and subsequent rehabilitation service that are available will be rendered from the county hospital where the family traditionally receives medical attention.

THE INJURY EXPERIENCE: A CLOSER LOOK The injury experience is additionally characterized by an immediate attentional shift in time perspective. The often pivotal nature of the injury event forces an athlete to address concerns about time played versus time not played and to reconcile any differences, especially when these differences draw consideration of having to defer sportrelated dreams and ambitions. An injury can be a sudden and unexpected minor setback (e.g., strains, sprains) or a traumatic and catastrophic event (e.g., spinal cord injury).

It can also be gradual in its onset (e.g., when the athlete has lived with injury pain and discomfort until the discomfort can no longer be tolerated). Irrespective of the way in which injury becomes manifest, the importance of the injury event usually triggers an immediate need to reassess time investments. Injury is not an experience that athletes plan for; thus, most are not prepared for the considerable self-involvement as well as the involvement of others that seems part and parcel of the injury experience. In addition, most athletes are not prepared to handle the unnatural feel of an injury. Walking with the assistance of crutches or with the support of a knee brace is not a natural experience or an innate movement. Holding an arm in a sling is not natural. Yet, ready or not, injury stimulates a quick need for resolute action regarding participation in future sport activity. When the type, kind, and degree of injury impairment and the athlete’s cognitive appraisal of an injury experience are viewed within the context of personal (e.g., age, race, ethnicity, sexual orientation, disability status, life circumstances, personal “baggage”) and environmental (e.g., community, fans, media, physical equipment, unsafe playing surfaces) realities, the focus on intrapersonal distress and other external consequential realities of an athlete’s injury and recovery becomes more acute. The contextual parameters also highlight more clearly the gaps in empirically and anecdotally based sport psychology information about the injury-to-recovery process. For example, the data are not clear on how the cognitive appraisal of an injury event varies by age, race, ethnicity, gender, sexual orientation, and disability challenge.129a The data are also not clear on how these contextual parameters influence the degree to which an athlete relies almost exclusively on sport or other sources for emotional nourishment. How do the contextual parameters affect how an athlete coped with past injuries? How do the contextual parameters influence the differential expression of feelings and emotions from injury onset through injury recovery? What impact do the contextual parameters have on the ability of an athlete to access resources in the management of the current injury’s challenges? What about the collegiate female cross-country Olympics aspirant who was recently declared athletically ineligible because of the severely compromised condition of her body owing to a never before treated eating disorder? Empirically and anecdotally based sport psychology data suggest that the emotional response to feeling so powerless and to feeling that an eating disorder treatment program would be an arduous undertaking could reach dramatic proportions—but could the declaration of athletic ineligibility also be a blessing in disguise? A female athlete expecting to receive positive results from a “well-trained and seasoned” sport psychologist who is treating her performance anxiety (after successful rehabilitation of a foot injury) with a “straightforward,” commonly implemented set of relaxation and visualization exercises is disappointed. After several weeks, the therapist concedes that the interventions did not work, and cessation of sessions or referral to another psychologist is suggested. The new therapist conducts a thorough investigation and discovers a past riddled by childhood molestation. What becomes clear is that nothing was wrong with the female

Nutrition, Pharmacology, and Psychology in Sports

athlete’s ability to implement the performance enhancement program in the presence of her male therapist. Her rigidity and tension proved to be a self-protective response to the memory of an emotionally painful time in her life. What mental or psychological cost was incurred when the first therapist chose not to do a thorough initial evaluation? Worse yet, what if the treating therapist did not even know to inquire about childhood abuse? Why did the belief in her status as a “gifted” athlete take precedence over her emotional well-being and psychological safety? What about the 68-year-old widower of 12 months, whose only passion now is golf, who learns that he will need hip surgery, a process that will keep him out of golf for several months, even a year? With what additional challenges, concerns, and issues might he be struggling? A female softball athlete just returning to practice after having completed a several-month rehabilitation program for her surgically repaired shoulder pushes herself “mentally and emotionally” to regain her position on the team. Although it is not an unheard of strategy, her hourlong before- and after-practice work habits did seem a bit unusual, especially given that she was just released by the team physician only 3 days ago to return to practice. When confronted by the psychologist she began seeing during rehabilitation, the athlete finally confesses her discomfort with homosexuality. She further shares that several lesbian teammates had in the past approached her sexually. What comes to light is that the athlete had been dealing with her discomfort with her teammates’ advances. To avoid future encounters, the athlete showed up to practice early, ahead of the team, and remained 1 hour after practice hoping that all of her teammates would have showered and left the premises. Thus, what appeared to be dedication to returning to top form and to reclaiming her position on the team was, in reality, the athlete’s method of coping with a difficult and sensitive situation. The foregoing scenarios suggest that actual injury and physiologic compromise to the body, as well as the subsequent rehabilitation, can be the least of the athlete’s challenges. Quality surgical intervention, active use of expert rehabilitation services, and an innately healing body position the athlete for maximal physiologic healing. The more critical feature of the injury experience, namely, the interactively complex set of emotions that are stimulated by the injury, is where the injury recovery work comes into play. The emotional aftermath of an injury can trigger a host of intrapersonal (e.g., confidence, self-esteem) and interpersonal (relationships with parents, peers, intimate others) issues and concerns. Because of the restrictive and in some cases immobilizing features of some injuries, athletes report feeling forced to come to terms with, instead of run away from, important aspects of their life. Without an adequate support system, forced confrontation of personal baggage can feel jarring, especially when solutions to the challenges that have arisen do not appear to be immediately obvious. The emotional aftermath of an athletic injury is also characterized in part by an attentional shift in time perspective from time played to time not played. Furthermore, the injury experience can feel significant and urgent. Questions that begin to surface include, Why me? Will I ever be as good as I once was? What are my coaches and

437

teammates saying about me? Emotions, including anger, depression, anxiety, grief, loss, and sorrow, flood the athlete’s senses. Issues, concerns, and challenges in other areas of an athlete’s life, such as academic, social, familial, occupational, and spiritual, are also stimulated by the injury experience. The emotional aftermath of athletic injury also surfaces questions regarding perceived benefits of athletic injuries.62,66 These scenarios represent just a sample of the cases that might present to a team of athletic injury and rehabilitation professionals, including the orthopaedic surgeon, trainers, physical therapist, and sport psychologist. The scenarios are by no means inclusive. Rather, they are offered as examples of injury situations in which the surface presentation might appear different from the truer picture that emerges after further examination and learning directly from the athlete how the injury has affected his or her life.

SPORT PSYCHOLOGY Simply stated, sport psychology is the science of human behavior and cognition applied to athletics. A sport psychologist possesses the education, background, and training in psychology and in the sport sciences and applies professional skills and competencies to the areas of teaching, research, and clinical practice. This final section addresses the role and function of the sport psychologist in athletic injury and the benefits of working with a sport psychology practitioner. A sport psychology practitioner moves flexibly in and out of at least three roles with the injured ­athlete—as the clinician, educator, and facilitator—and has a consulting role with the sports medicine team.

Sport Psychologist as Clinician A sport psychologist as clinician plays a pivotal role in assisting the injured athlete to navigate the emotional “twists and turns” of the injury and rehabilitation process. Because the myriad concerns and issues that surface with injury can be complex, referral to a well-trained psychologist is of utmost importance. Ways of identifying a sport psychologist who possesses the skills, knowledge, and abilities to work with injured athletes, their trainers, and the orthopaedic physicians are presented in the closing sections of this chapter. On referral to a sport psychologist, the referring parties and the injured athlete accepting the referral can expect the following responses from the sport psychologist-clinician who is now part of the athletic injury rehabilitation treatment team. The injured athlete can expect to participate in an extensive assessment process. It is a process that invites the athlete to talk about at least three substantive areas. The first area involves obtaining a better understanding of the injury event and the athlete’s experience of that event. A detailed description of the experience, including the concerns, issues, and realities that frame the injury experience for that particular athlete, will be solicited. Equally important to solicit is the degree to which the injury has interrupted or detracted from the athlete’s goals, dreams, and aspirations. Third, some determination needs to be made relative to the athlete’s willingness to participate in the impending injury and rehabilitation experience. Given

438

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the predicted ups and downs of a typical rehabilitation process, it would be important to ascertain the extent to which the athlete is willing to comply with the prescribed rehabilitation regimen, including unplanned modifications. The sport psychologist-clinician also prepares to assess the “convenience factor” that injury introduces. Injury is not an experience desired, and for most athletes, recovery cannot happen soon enough. The benefits of time away from athletics, however, cannot be denied, so the following question also merits consideration: “Does the athlete want to succeed at recovery or succeed at staying stuck and avoiding reintegration back into sport?” Thus, this first core set of questions will result in both the practitioner’s and the injured athlete’s being able to construct a big-­picture perspective of the injury and of its role in the athlete’s current and future activities. The second set of core questions targets the clinical or psychological profile of the athlete with whom the psychologist is working. A standard clinical interview, which allows a comprehensive and systematic evaluation of the athlete’s current mental health, is conducted during this juncture. Framed within a context of a biopsychosocial model, questions are designed to determine clinical disorders, personality disorders, general medical conditions, psychosocial and environmental problems, and level of functioning.131,131a The third set of core questions is designed to gather data that help the psychologist construct as clear a picture as possible of the athlete as a person. The picture that will emerge will be one born out of a context of race, ethnicity, gender, sexual orientation, disability status, and religious affiliation. To claim to understand the athlete without appreciating the contextual parameters that shape the athlete’s reality is to fall short in your knowledge about and respect for the individual with whom you purport to be working. The gathered information generates a “snapshot” of the challenges the athlete is now confronting. On the basis of the resultant diagnostic impressions and in collaboration with the athlete, the sport psychologist-clinician suggests interventions that are aimed at addressing the problems that have surfaced. Mechanisms for evaluating the effectiveness of the interventions are also set in place.12,35,36,125,132-137

Sport Psychologist as Educator Acquiring accurate information about the injury and the rehabilitation process (e.g., factors that predisposed injury to occur, specific features of the current injury, goals for injury treatment, detailed information about the medical and rehabilitation procedures that will be used, and prognosis for complete recovery) is an important and paramount concern for the athlete. Factual information about the injury and rehabilitation process contributes to decreased worry and anxiety, decreased cognitive distortions about injury severity and treatment course, expectations that are increasingly aligned with realistic outcomes, and increased compliance with suggested treatment interventions. The sport psychologist can be instrumental in gathering information about the injury and the rehabilitation process and in disseminating information in such a way that the athlete not only hears the information but also

makes plans to use it. The sport psychologist, with an athlete’s permission, can also serve to educate the injury and ­rehabilitation team of physicians, trainers, and nutritionists about the treatment and treatment responses of the athlete to the services the sport psychologist is rendering. Finally, the sport psychologist as educator can conduct in-service training for other professionals working with athletes who desire to learn more psychology and its application to athletics.

Sport Psychologist as Facilitator The injury and rehabilitation process generates multiple needs for the injured athlete. The injured athlete may need to access multiple resources to meet the needs that have surfaced. It may be determined, for example, that an injured athlete struggles with alcohol and substance abuse problems; thus, the athlete would benefit from involvement with groups like Alcoholics Anonymous or other outpatient substance use and abuse services. An athlete’s history of sexual assault may result in the need for a referral to a therapist or agency that offers specialized services for these concerns. Specialized services in the form of therapy or support groups are also suggested for the athlete struggling with disordered eating. Family or marital concerns may contribute to the ongoing stress that an injured athlete experiences. In these instances, referral to a therapist with expertise in family therapy or couples counseling might prove a useful adjunct to the services already prescribed for the physical rehabilitation. Access to vocational counseling services, referral to services for senior citizens, consultation regarding legal concerns, or guidance regarding spiritual matters may also be needed adjunctively. A sport psychologist working collaboratively with a sports medicine team of professionals could be instrumental in helping the injured athlete to alleviate some of the stresses that may help to exacerbate the injury. Use of these additional services may contribute positively to the injured athlete’s physical and psychological healing.

Sport Psychologist as Consultant Sport psychologists play a pivotal role serving as consultants to the team of sports medicine professionals, including orthopaedic surgeons, family practice physicians, athletic trainers, physical therapists, nutritionists, and strength and conditioning coaches. Practitioners in each of these disciplines are expert in their respective areas, yet they may not have mastered core learning material in the disciplines of their collaborative colleagues. Of the group of sports medicine professionals, athletic trainers138 and family practice physicians34 may have received an introduction to basic principles of psychology during their formal academic training. Most of the other disciplines have little or no formal education in psychology fundamentals. Given the person-focused nature of each professional domain, it seems reasonable to expect that each would have had at least a formal introduction to the field of psychology. Enter the sport psychologist, a professional who not only can provide mental health services to the injured athlete but also can communicate an opinion about the psychological well-being of an injured athlete to the team of sport

Nutrition, Pharmacology, and Psychology in Sports

medicine professionals. Input from the sport psychologist regarding an athlete’s psychological well-being could help the injury and rehabilitation team in their overall formulation, coordination, and implementation of an agreed-on intervention plan. A sport psychologist is also in a position to provide education by in-service training of the sports medicine staff in a variety of mental health topics. Athletes, like all people, struggle with the array of mental health challenges, so acquiring substantive information about mental health and mental illness sensitizes the sports medicine staff to be on the lookout for athletes who might feel challenged psychologically but whose struggle with their challenges might otherwise go undetected.

Caveat Emptor: Let the Buyer Beware Sport psychologists play an important role as members of athletic injury and rehabilitation teams. Their expertise, coupled with the expertise of the sports medicine team, creates an opportunity for the injured athlete to heal physiologically as well as psychologically. There is some debate, however, about who is a sport psychologist. A complete analysis of the debate is beyond the scope of this chapter; suffice it to say that the debate has been ongoing almost since the beginnings of sport psychology as a formal area of study, research, and practice.139-145 Essentially, the debate is fueled by the presence of two groups calling themselves sport psychologists. Differences in historical roots as well as in the education, training, and respective career pursuits account for the separate emphasis that defines each group. The first group of professionals using the term sport psychologist comes out of the academic disciplines of sport sciences, kinesiology, physical education, and human movement. The target focus of study of these groups centers on (1) achieving a better understanding of the mechanics of human movement and (2) identifying the factors that contribute to athletes’ participation in sports and exercise and the benefits that result from their participation in these activities. Research and teaching are the primary means used to carry out their career roles. The second group of sport psychology professionals are persons trained as psychologists who choose to apply their training and expertise to athletes across sports and athletic settings. The target focus of study of this group centers on achieving through scientific inquiry a better understanding of human behavior across settings and within the context of environmental realities. Applied work is the hallmark of this group of sport psychologists, although research and teaching characterize some of their professional activities. Historically, each of these groups came to the sport psychology arena with incomplete training and lacking in the comprehensive skill sets needed to offer effective consultations to athletes seeking their services. The professionals emanating from the academic traditions of physical education, kinesiology, human movement, and sport sciences contributed their knowledge and expertise in basic physiologic principles, biomechanics of exercise, motor movement, anatomy, muscle strength and endurance, nutrition, aerobic capacities, and the like to their sport psychology practice. This group was deficient in education and

439

­ ackground in psychology, as characterized by a knowledge b base rooted in courses having to do with theories of personality; social, biological, and cognitive bases of behavior; psychopathology; and basic counseling skills. On the other hand, psychologists who claimed the title sport psychologist were usually well trained in the discipline of psychology but were deficient in the areas traditionally seen in the sport sciences. Several factors undoubtedly contributed to each group proclaiming itself to be the real sport psychologists. The economics of athletics, however, seems to have played a key role in the development of the territoriality positions of the opposing groups.15 Over the years, large sums of money have been poured into sports and athletics, so much so that sports has become a multibillion-dollar entertainment business. Media, especially television, and ­ endorsementsponsorship arrangements contributed significantly to the marketing of sports as an industry. Highly skilled and talented athletes were and are paid handsomely in return for their production on the field. Successful athletes and teams bred more success, which often translated into a language of economic prosperity. Sport psychologists were introduced as a means of helping athletes to maximize their performance abilities, thereby protecting owners’ and sponsors’ investments in their sports commodities. Both groups claiming to be sport psychologists wanted a piece of the lucrative sports enterprise and thus marketed themselves as the consultants who could best take care of the owners’ and sponsors’ investments. Where does this leave sports medicine professionals with respect to their need to know with whom they can consult regarding the management of an injured athlete? This debate has spawned several professional organizations to grapple meaningfully with the myriad issues associated with the provision of sport psychology services. Although the debate about who is a sport psychologist is ever present, attention is being directed to identifying standards that address (1) competency to render psychological interventions, (2) basic knowledge of human behavior, (3) basic knowledge of sports and athletics, (4) applied training, and (5) ethics. The American Psychological Association, Division 47—Exercise and Sport Psychology and the Association for Applied Sport Psychology have confronted in earnest the task of developing recognized and accepted proficiency standards of the practice for sport psychology. The impact of their work on the respective professions and, more important, on the public and athletes on whose behalf sport psychology professionals are purporting to work remains to be seen. Given the emotional complexities inherent in an athlete’s experience of injury and rehabilitation, and given the varied contexts within which the athlete functions and operates, a professional trained with some combination of the described course work, including direct applied experience, seems essential. In the absence of an agreed-on set of standards, the following recommendations for working with a sport psychologist are offered. Sports medicine professionals wishing to access services of a sport psychologist might want to inquire about a prospective consultant’s credentials to practice psychology. Because psychology is a recognized area of service, and the term psychologist is a protected title, having a license to engage in the ­independent

440

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

practice of psychology might be one benchmark by which to assess minimal competence. A license to practice psychology is granted only to candidates who satisfy national and state requirements, as measured by written examinations covering psychology and law and ethics. It is also important for professionals wishing to consult with sport psychologists to assess the amount and kind of direct, hands-on experience a prospective sport psychologist has had working with the specific athletic group for which consultation is desired. Soliciting from the prospective sport psychology consultant information about direct experience with athletes is one means of gathering the data. Securing recommendations in support of the prospective sport psychologist from persons (coaches, colleagues, other athletes) who can attest to his or her work is another way of gathering data about a sport psychologist’s experience. Once a consultation agreement between the sport psychologist and the party with whom the sport psychologist is consulting has been entered into, it is important for the party who brought the sport psychologist on board to request or make provisions for a periodic assessment of the work being performed. Periodic stocktaking of promised and rendered services always makes for a good practice. These “rules of thumb” are offered in the absence of an accepted and acknowledged criterion-referenced document that would otherwise detail competency-based standards for the provision of sport psychology services.

CONCLUSION There are a lot of emotional twists and turns in an athlete’s experience of the injury and rehabilitation process. To reiterate a point made earlier, there are methodologic flaws inherent in the empirically based sport psychology research on the athletic injury and rehabilitation process that has been conducted to date. Coupled with the anecdotal descriptions of the injury and rehabilitation process authored by sport psychology practitioners, what emerges is an incomplete picture of the process from the perspective of the athlete. Researchers and practitioners alike, however, continue to chip away at the complexity of the process in the hope of developing a more accurate picture of the athlete’s experience of injury and rehabilitation. The less than accurate picture of the athletic injury and rehabilitation process that has emerged from existing literature is a byproduct of the way in which the research questions have been framed. Interestingly, the absence of clarity about the athletic injury and rehabilitation process, as demonstrated by available research and anecdotal accounts, actually directs attention to the kinds of questions that could serve as a jump-off point for future scientific inquiry. The following alternative question is offered for consideration by researchers and practitioners with the intent of stimulating a different research direction: How will this particular injury, to this particular athlete, at this particular time in the athlete’s life, affect how the individual sees himself or herself now and in the future in relationship to the sport and in relationship to interests and areas other than those related to the sport? This question sets a research direction in that it has the potential to yield responses that are uniquely reflective of the injury and rehabilitation

­ rocess. As a result, researchers can define terms, orgap nize concepts, and articulate research strategies that seem aligned more closely to an injured athlete’s actual experience. The alternative question also sets a clinical direction in that its more exacting focus has the potential to yield responses that are more reflective of the individual’s perception of the injury and rehabilitation experience. One goal of this chapter has been to provide a nonexhaustive synopsis of the current literature on the athletic injury and rehabilitation process in the hope that the presentation might stimulate discussion of the various concerns, issues, and realities that frame that experience. In presenting the overview and in distilling the current research findings into the least common denominators, what emerges is a three-factor conceptual premise that essentially highlights two themes. The three premises suggest the importance of understanding the (1) preinjury, (2) during-rehabilitation, and (3) post-rehabilitation physiologic and psychological processes. These premises are offered as a way of highlighting what is agreed on by sport psychology researchers and practitioners. The scenarios are shared to illustrate how what appears on the initial presentation of an injured athlete to be the main focus actually turns out to be a different agenda once further details are gleaned.

ONE FINAL NOTE Athletes, irrespective of age, gender, race, ethnicity, sexual orientation, and disability status, are first and foremost everyday people who, like their nonathlete peers, grapple with life’s everyday challenges. On the playing field and in the athletic arenas, they find opportunities to learn what challenge really means and how they can harness their talents and abilities to face the challenges they encounter. In a similar vein, nonathletes use their work or school settings as their challenge arenas. Whereas all athletes enjoy winning, few athletes engage in competition solely for the sake of achieving victory. The real victory comes from athletes’ knowing they pushed themselves to the limits of their abilities and skills and thereby overcame the challenges inherent in the victory just achieved. Injury, then, can represent a time of grief, despair, and sometimes helplessness as it removes the athlete from the challenge arena and from continued opportunities to test personal limits. It is an event that is never planned; thus, it catches athletes off guard, and it always comes at the wrong time. Injury can be a minor inconvenience, or it can represent an event that is humbling. Either way, its disruption is unmistakable. Returning athletes to preinjury condition and form is of paramount importance because it signals to them a return to the means that has allowed them to feel important, successful, and accomplished. Therefore, it is incumbent on sports medicine teams employed to assist athletes in the recovery process never to approach an athlete thinking that their intervention efforts are simply routine or that the athlete is “just another body.” Athletes comply with prescriptive treatment regimens (1) when they have confidence that the intervention efforts will return them to full form, (2) when they have trust in the service provider’s competence, and (3) when they believe the service provider has faith and confidence in them.

Nutrition, Pharmacology, and Psychology in Sports

“Accepting the challenge” is a theme that influences the push of an athlete to grow and mature. There are many opportunities on the playing field and in sports arenas for able-bodied, noninjured athletes to put themselves “to the test.” The physician’s office, training room, and rehabilitation facility, however, can serve as alternative venues wherein an injured athlete can continue to accept the challenge, albeit of a different kind. Opportunities to discover inner strength and wisdom are abundant, irrespective of the type of challenge, when athletes begin to see their injury event and subsequent rehabilitation experience as an opportunity to invest in themselves as people with boundless potential.

441

R eferences Please see www.expertconsult.com

S ect i o n D

Sleep and Chronobiology in Sports Teodor T. Postolache

This section discusses effects of biological rhythms and sleep on athletes’ performance and health. The first part presents certain operating concepts of chronobiology, the science of biological rhythms. The second part discusses sleep and sleep disorders. The third part discusses the sleep and chronobiology consultation and implications of chronobiology and sleep medicine for the athlete’s life, health, and performance, with special emphasis on diagnostic and monitoring tools and the arsenal of nonpharmacologic interventions to correct adversity-related abnormalities of biological rhythms and sleep. To summarize, although the ultimate role of sleep continues to be debated, an increasing body of literature supports an important role of sleep in memory consolidation and metabolic functions. Sleep can have impaired duration or quality. In addition to sleep, however, the internally generated rhythms of metabolic (including endocrine), neurophysiologic, and cognitive factors translate into fluctuations of motor and mental abilities. This may largely contribute to outcome of competitions between athletes of similar caliber. When internal rhythms are out of sync with external environmental rhythms, in addition to possible alignment problems between peak performance and timing of the match or competition, impairment in sleep duration and quality may have a negative effect on athletic performance, training, and postexercise recovery.

CHRONOBIOLOGY As Thomas Wehr summarized, “Because the Earth rotates, it regularly presents two rather different environments to the organisms that live on it: a world of light and a world of

darkness”1 Day and night are not only physically distinct but each is also reflected in the biology of the living organisms. Thus, external alternation between day and night is represented internally (i.e., “internal” or “biological” night,” “internal” or “biological” day”) to anticipate rather than only react to the temporal environment. Notice that we do not use the term fluctuation, which involves a slow quantitative transition, but rather use alternation, which means a qualitative change in states, with short periods of transition. Biological day and night are characterized by changes in gene expression, protein synthesis, electrophysiology, hormonal levels, and behavioral propensities and are separated by relatively short “biological dawn” and “biological dusk.” Living organisms have developed mechanisms to track changes in environmental light, to be active and engage their environment during a specific temporal niche (e.g., diurnal, nocturnal, crepuscular species), and to retreat at other times for rest and sleep. Because the axis of the earth is tilted, seasonal changes in the duration of day length, temperature, and food availability occur. Many species have developed mechanisms to track day length in anticipation of changes in seasons and manifest marked seasonal behavioral changes such as those associated with reproduction, rearing of the young, nest building, hibernation, and migration. These seasonal rhythms are thought to achieve the best alignment between environmental resources and energetic demands for sustaining life and reproducing. We evolved under the sun. Our ancestors lived in brighter days, darker and longer nights, and did not travel across time zones. Although we live in a relatively shielded microclimate, our physiology is anchored in ancestral evolutionary times.

442

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Like other mammals, plants, insects, fishes, and birds, humans have body clocks that maintain endogenous rhythms in constant conditions, and we adjust our internal time to the external time through light input through the retina. In this way, light is not just a vehicle for sight but is also the main synchronizer between the internal, approximate 24-hour rhythms in physiologic states and the exact 24-hour cycles in environmental conditions. The definition of chronobiology is readily seen in the Greek origin of the term. From chronos, time, bios, life, and logos, study, it describes the field of science that ­systematically studies the timing processes in organisms. Regular, robust, reproducible, and highly predictable ­biological fluctuations occur in all living things. Chronobiology investigates and quantifies these biological rhythms, including the study of their manifestations, mechanisms, and consequences and their experimental or clinical ­variances. In humans, these include measurable periodic variations in physical and mental performance, and applied in the world of sports, these variations may result in defining the slimmest of margins in athletic performance, potentially representing the difference between winning and losing. CBT min

DSST

(#correct)

B

PRM

(#correct)

C

(#msec)

PVT lapses PVT median

D

(#lapses) (units)

KSS

S

W

0

Addition calculation

–3 3 Digit symbol substitution

0 –3 0.5 0.0

Probed recalled memory

–0.5 –30

Psychomotor vigilance: reaction time

0 30 –5

E

F

W

3 (#correct)

ADD

A

Chronobiology and sleep medicine, despite efforts to incorporate it as an applicative sports science,2 have not yet had a great impact in the life of athletes. Nonetheless, the timing may now be good for chronobiologic and sleep medicine interventions to be considered for restoring athletic ability affected by insufficient or poor-quality sleep and biological rhythms out of sync with the environment and among themselves. Three significant factors are responsible for this new momentum. First, sophisticated, reproducible laboratory investigations have established the important role that sleep and wake cycles have on both measurable and qualitative performance. It has been shown that psychomotor vigilance and subjective alertness, as well as other abilities of particular relevance for athletic performance, show marked predictable variation in time. These often improve during the later part of the active circadian phase but will deteriorate beyond a certain point of wakefulness (Fig. 8D-1).3 Second, major advances have been made recently toward the discovery and understanding of molecular processes of the so-called biological clock. Polymorphisms in the CLOCK gene4 and Per genes5 are related to morning or evening activity preferences, that is, the morningness-eveningness chronotype. In ­addition,

Psychomotor vigilance: lapses

0 5 –1 Alertness: Karolinska Sleepiness Scale

0 1 0 120 240 0 120 240 0

3.3

13.3

3.3

13.3

CIRCADIAN PHASE AT HOURS SINCE START START OF EACH TEST OF WAKE EPISODE Figure 8D-1    Circadian variation in performance. Laboratory protocol consisting of 24 repetitions of a 20-hour rest-activity cycle, resulting in desynchrony between the sleep-wake cycle and the circadian rhythms of body temperature. Double plots of main effects of circadian phase relative to minimum of core body temperature (CBT) (left) and duration of prior scheduled wakefulness (right) on neurobehavioral measures. Plotted points show deviation from mean values during forced desynchrony section of protocol and their respective standard errors. For all panels, values plotted lower in the panel represent impairment on that neurobehavioral measure. Addition calculation test (ADD) (A), digit symbol substitution test (DSST) (B), and Probed recalled memory test (PRM) (C) scores are derived from total number of correct responses. PVT results represent median reaction time (D) and total number of lapses (E) (reaction time > 500 msec). KSS scores (F) represent responses on a 1 through 9 Likert-type scale. (Redrawn from Wyatt JK, Ritz-De Cecco A, Czeisler CA, et al: Circadian temperature and melatonin rhythms, sleep, and neurobehavioral function in humans living on a 20-h day. Am J Physiol 277[4 Pt 2]:R1152-63, 1999.)

Nutrition, Pharmacology, and Psychology in Sports

Circadian rhythms represent periodic phenomena with a period of about 24 hours. The circadian timing system generates and regulates these endogenously generated circadian rhythms. A circadian rhythm in body temperature (a circadian phase marker), for example, like other ­oscillating measures, has a period, an amplitude, a mean level (MESOR), a peak and a timing of the peak (acrophase), and a trough and a timing of the trough (Fig. 8D-2). Marked, about 24-hour fluctuations occur in parameters that are inherently related to athletic performance, such as attention and other cognitive functions (see Fig. 8D-1), and include self-chosen work rate and body flexibility (Fig. 8D-3) and pulmonary function.7 A rhythm that is not in sync with the 24-hour day, because of a longer or shorter period, is called free-­running (Fig. 8D-4). In constant dim light conditions, the 24-hour rhythms lose their relationship to the environmental 24-hour day and free-run according to the period of their biological clock. In certain animals (e.g., most rodents), the period is shorter than 24 hours; in others, it is longer than 24 hours. In humans, the circadian rhythm generated internally by an endogenous pacemaker has a period of slightly longer than 24 hours. The pacemaker or clock produces these rhythms through a transcription-translation loop of the clock genes and feedback by their proteins. A rhythm that has been reset to the 24-hour period of the earth’s rotation is called entrained (see Fig. 8D-4, upper part). To entrain an organism, the endogenous rhythms must be exposed to a zeitgeber, or ‘‘time giver,” which could be any environmental variable that can reset the body’s clock. The most critical zeitgeber is light, but other zeitgebers (e.g., time of food availability, exercise, ambient temperature, and social contact) may also play a role. Subjective or biological day is an internal (physiologic) expression of external daytime, and the subjective or biological night is peak (crest) amplitude mean level (mesor)

period

210 Self-chosen workrate (w)

Terminology of Circadian Rhythms

220

200 190 180 170 160 150

00.00

04.00

08.00

12.00 16.00 Time of day

20.00

00.00

04.00

08.00

12.00 16.00 Time of day

20.00

00.00

A Whole-body flexibility (% 24-hour mean)

a murine model has been developed of morningnesseveningness.6 Linking these basic genetic findings with circadian behaviors provides a powerful theoretical framework for practical applications of chronobiology. Third, technologic advances have led to small ambulatory devices to record rest-activity rhythms, tools to measure mental and physical performance at different times, and the ability to measure hormone changes in saliva to accurately determine circadian phase (internal time).

443

B

140

120

100

80

60 00.00

Figure 8D-3    The circadian rhythm represented for selfchosen exercise intensity (A) and for whole-body flexibility (B). (From Reilly T, Waterhouse J: Sport, exercise and environmental physiology. Edinburgh, Elsevier, 2005.)

an internal (physiologic) expression of external nighttime. Excluding shift workers and transmeridian travelers, for most humans biological night overlaps appropriately with the external (environmental) night and with sleep. Although it had once been thought that for humans social cues were more important than other zeitgebers, it is now ­established that light is, for humans along with other animals, the most important zeitgeber. Exposure to light changes the period and phase of the circadian pacemaker (see Fig. 8D-4). In fact, with the absence of light perception, many blind individuals have been shown to have a free-running circadian rhythm, resulting in symptoms similar to those of jet lag, for example, poor quality of sleep at night, daytime somnolence, and diminished wake-time alertness and ­performance.8,9

trough acrophase

trough time

Figure 8D-2    The circadian rhythm represented as a cosine function. (Redrawn from Reilly T, Waterhouse J, Edwards B: Jet lag and air travel: Implications for performance. Clin Sports Med 24[2]:367-380, 2005.)

The Suprachiasmatic Nucleus The suprachiasmatic nucleus (SCN) of the hypothalamus, known as the “body clock” or the primary circadian pacemaker, is in charge of regulating endogenous circadian rhythms. In animal models, the absence of a functional

444

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� 0

12

24

48 hr

LD

DD

Phase advance Phase delay SCN lesion

Light pulse Figure 8D-4    Schematic representation of a circadian rhythm as normally entrained to a 24-hour light-dark cycle (LD), free-running pattern in constant darkness (DD), and suprachiasmatic nucleus (SCN) lesion condition. The dark bars indicate the nocturnal active (moving, drinking, and eating in rodents) periods on successive days (the onset of activity is arbitrarily chosen), presented in a so-called double-plotted actogram. The free-running rhythm is phase-delayed by a light pulse at the beginning of the active period (CT12-18), and phase-advanced by a light pulse at the end of the active period (CT18-23). Note in particular the point at which there is a lesion of the SCN. This results in a dramatic change in circadian rhythm (i.e., a fragmented activity pattern) and is consistent with the loss of master pacemaker function of the SCN. (Redrawn from van Esseveldt KE, Lehman MN, Boer GJ: The suprachiasmatic nucleus and the circadian time-keeping system revisited. Brain Res Brain Res Rev 33[1]:34-77, 2000.)

SCN results in the loss of circadian rhythmicity, with sleep and wake alternating in erratic short intervals (see Fig. 8D-4, lower part). Light reaches the SCN from the retina through the neuronal connection of the retinohypothalamic tract. In the retina, light interacts with photosensitive retinal ganglion cells (pRGCs), which contain a novel photoreceptor called melanopsin, which, in humans, is more intensely stimulated by short-wavelength blue light (420 to 440 nm).10

Seasonal Rhythms Profound seasonal environmental changes are caused by the tilted axis of the earth in its yearly movement around the sun. The most predictable seasonal environmental cue is the duration of day length or photoperiod. The amplitude of the photoperiod change is related to latitude, increasing the farther away one goes from the equator. Photoperiodic organisms use changes in the duration of day length, rather than temperature or other environmental markers of season, to regulate seasonal physiologic changes. Some seasonal behavioral and physiologic changes in humans are manifested as decreases in energy level, motivation,

and socializing and as increases in sleepiness, appetite, and weight. Some of these changes are analogous to those of photoperiodic mammals, most notably reproduction, migration, and hibernation. Also, in photoperiodic mammals, the duration of melatonin secretion not only reflects the seasonal changes in environmental darkness but also mediates the effects of seasonal changes in winter. Mild seasonal changes in mood and behavior (e.g., decreased energy, sleepiness, increased carbohydrate craving, increased appetite, weight gain) occur in 10% to 20% of the population of the United States as subsyndromal seasonal affective disorder (SSAD). About 4% of the general population have pronounced seasonal changes in mood, including episodes of major depression in fall and winter with remission in spring and summer, manifesting as ­winter-type seasonal affective disorder (SAD).11 In patients with SAD, as well as in photoperiodic mammals, the duration of melatonin secretion is longer in winter than in summer.12 An alternative theory of SAD is a circadian phase delay hypothesis based on a hypothesized circadian phase shift due to progressively lower photoperiods in winter13; this theory has been supported by a study that showed an associated increase in mood scores with the resynchronization of circadian timing after appropriately timed administration of melatonin.14 SAD is more prevalent in females and in younger age groups. Certain populations have been shown to have a resilience to SAD, such as Icelanders15 and their Canadian descendants.16 The Chinese and Japanese may be more vulnerable to a less prevalent type of SAD, ­summertype SAD, with summer depression and remission in fall and winter, which represents an analogue of a less frequently displayed seasonal change in some animals, called estivation.17 In addition, marked seasonal rhythms in athletic performance may occur because of infectious factors that have a seasonal distribution, such as the human rhinoviruses (peak in winter), influenza viruses (winter, early spring), and coronaviruses (winter, early spring).18 Another factor is exposure to seasonal allergens such as tree pollen in early spring, grass from spring until fall, and weeds (ragweed in the United States) at the end of summer and beginning of fall. For a detailed discussion of seasonal allergies and their treatment, with specific implications for seasonal athletes, the reader is directed to Komarow and Postolache’s article.19

Bright Light Treatment After the first report of bright light suppressing melatonin in humans,20 Rosenthal and colleagues followed with the first clinical description of SAD along with its therapeutic response to bright light.21 The effectiveness and safety of bright light treatment has been further confirmed in placebo-controlled trials.22,23 Bright light treatment for seasonal depression has been recognized in the Clinical Practice Guidelines issued by the U.S. Department of Health and Human Services24 and is included in the American Psychiatric Association’s Treatments of Psychiatric Disorders.25 A recent meta-analysis on bright light treatment,26 including only randomized trials, reported that the effect size of bright light treatment for SAD was 0.84

Nutrition, Pharmacology, and Psychology in Sports

D A

B

C

E

445

F

G

Figure 8D-5    Light treatment devices. A, Standard light box. B, Light box combined with bright desk lamp. C, Full-spectrum light box with capability for exercise, practice, and reading. D, Portable bright light treatment devise. E, Visor. F, Low-intensity, lower wavelength green light device, which can be used adjacent to a computer screen. G, Portable, low-wavelength blue-light LED device. (Modified from Postolache TT, Oren DA: Circadian phase shifting, alerting, and antidepressant effects of bright light treatment. Clin Sports Med 24[2]:381-413, 2005.)

with a 95% confidence interval of 0.60 to 1.08, similar to the effect size of pharmacologic antidepressant treatments. The direction and amplitude of the effects of bright light are influenced by the timing of light exposure and the wavelength and intensity of the light. There is a consensus that bright light treatment with an intensity of typically 10,000 lux should be administered in the morning, immediately upon awakening, and starting with a dose of 20 to 30 minutes in concordance with the specific indications of the particular light treatment device. Devices for bright light administration include light boxes, light visors, dawn simulators, sunrise clocks, and facemasks (Fig. 8D-5). For details on the devices and administration parameters, as well as side effects, consult Postolache and Oren.27

Menstrual Cycles Menstrual cycles are, in fact, menstrual rhythms, because they do not consist of true oscillations but of alternations between five states: menstrual, follicular, periovulatory, progestative, and premenstrual. The menstrual physiologic and dysfunctional implications for sports and sports performance may represent a ­component of a particular chronobiology regimen for the female athlete; however, this would exceed the scope of this chapter. The reader is referred to the chapter by Constantini and associates, which covers this topic ­thoroughly.28

SLEEP AND SLEEP DISORDERS In the past, sleep was viewed as a passive, inactive state resulting from a reduction of the activity in the ­wakefulness-promoting regions of the brain. Now sleep is being understood as a very dynamic state, with certain sites in the brain increasing activity during sleep and, in balance

with arousing areas of the brain and environmental stimuli, determining the behavioral state (Fig. 8D-6). Regulation of the sleep-wake cycle is accomplished by the interaction of two endogenous systems: sleep homeostat (process S) and the circadian process (process C) (Fig. 8D-7).29 The sleep homeostat manages the tendency to sleep and to remain asleep, and it drives the effects of sleep deprivation and sleep debt. Process S increases during wakefulness30; however, during sleep, there follows a gradual decay in process S as the sleep pressure diminishes (see Fig. 8D-7). Process S can be perceived as sleep pressure (urgency to go to sleep), sleep debt (accumulated effect of not getting enough sleep), or simply the ability to fall asleep. It can be estimated by measuring the delta power during sleep and may be mediated by soluble neurotransmitters such as adenosine.31 Process C is the circadian process, which generates a signal of wakefulness with a temporal trajectory that is about a mirror image of process S. It increases the threshold for falling asleep during the day as process S accumulates, peaks in the late afternoon and evening (consolidating wakefulness), and decreases during the night (consolidating sleep). The underlying neurologic structure of process C is the SCN, also known as the “master clock” or the “master circadian pacemaker.” Process C can be measured indirectly by core body temperature or by recording hormonal parameters such as melatonin and cortisol. A more complex model of sleep, which adds the sleep inertia process, has been proposed32 (Fig. 8D-8). The physiologic state of sleep inertia is the process of slow awakening with cognitive impairment. Its duration and severity depend on various factors; however, sleep state at arousal and circadian time of sleep appear to have the greatest impact.33 The duration of sleep inertia has some interstudy variability ranging from a few minutes up to 35 minutes depending on the designated task.

446

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Orexin +

–

TMN

HIST

LC

NA

Raphé

5-HT

PPT-LDT

ACh

GABA VLPO SLEEP

Gal

WAKE

– Figure 8D-6    This is a simplified model (not differentiating between rapid eye movement and non–rapid eye movement sleep) that illustrates reciprocal interactions between sleep-promoting (blue) and wake-promoting (red) regions, which result in a flip-flop switch (simplified model). Aminergic regions such as the tuberomammillary nucleus (TMN), locus ceruleus (LC), and dorsal raphe (DR) promote wakefulness by direct excitatory effects on the cortex and by inhibition of sleep-promoting neurons of the ventrolateral preoptic nucleus (VLPO). During sleep, the VLPO inhibits amine-mediated arousal regions through GABAergic and galaninergic projections. This inhibition of the amine-mediated arousal system disinhibits VLPO neurons, further stabilizing the production of sleep. Orexin (ORX) in the lateral hypothalamic area further stabilizes behavioral state by increasing the activity of aminergic neurons, thus maintaining consistent inhibition of sleep-promoting neurons in the VLPO. (Adapted from Saper CB, Chou TC, Scammell TE: The sleep switch: Hypothalamic control of sleep and wakefulness. Trends Neurosci 24:726-731, 2001.)

For instance, ­ impairments in ­ reaction time and performance in ­ arithmetic tasks have been observed to last up to 15 ­ minutes, and in contrast, decision-­making ability was measured to be affected by sleep inertia for at least 30 minutes.34 Cognitive performance decrements are also significantly dysfunctional up to 30 minutes after awakening, whereas motor performance skills were seriously degraded for an extended period lasting longer than 75 minutes.35 Sleep inertia is a process with an unknown neurobiology but is believed to be the consequence of Homeostatic process (S) Sleep pressure Wakefulness

Sleep

Wakefulness

Sleep

Circadian process (c) Alertness

8

13

18

23

4 9 14 Time of day

19

24

5

9

Figure 8D-7    Interaction of process S and process C. The homeostatic pressure for sleep builds up during wakefulness and dissipates rapidly during sleep. The circadian process is related to the time of day, is independent of the amount of previous sleep, and opposes the homeostatic process. (Modified from Borbély AA: A two process model of sleep regulation. Hum Neurobiol 1[3]:195:204, 1982.)

sequential rather than simultaneous activation of multiple areas in the brain during awakening.

Stages of Sleep Sleep is divided into the four stages of non–rapid eye movement (NREM) sleep, making up 75% to 80% of sleep duration, and one stage of rapid eye movement (REM) sleep, which alternate several times throughout the night in a cycle, with REM episodes lasting progressively longer and NREM progressively shorter as the night continues. NREM sleep and REM sleep can be differentiated by electroencephalogram (EEG) (Fig. 8D-9), muscle tone, eye movement, and respiration rate. REM sleep, which cannot be differentiated from wakefulness based on EEG alone, is also characterized by dreams (which also occur in other stages) and paralysis. As an individual becomes sleepier during the wakefulness state, the body gradually approaches a less active, restful state, during which muscle relaxation and a decrease in eye movement occur. During NREM stage 1, a reduction of brain activity by half ­compared with wakefulness is associated with a conversion from EEG alpha waves (a marker of relaxed wakefulness) to theta waves (4 to 7 Hz). Stage 2 sleep is the most dominant sleep phase, accounting for an estimated 45% to 55% of sleep duration, and is characterized by unique waveforms, including sleep spindles (12 to 16 Hz) and K complexes, consisting of a brief high-voltage amplitude spike. Stages 3 and 4 of NREM sleep constitute deep or slow-wave sleep (SWS) and are differentiated by the length of delta waves (0.5 to 4 Hz), with delta waves accounting for less than half of the brain activity in stage 3 and more than half during stage 4. It is important to note that the deeper stages of sleep, characterized by slow wave sleep (such as stages 3 and 4 in

Nutrition, Pharmacology, and Psychology in Sports

Sleepiness

Awake H 0

Elapsed Time awake (hr)

Alpha activity

Stage 1 sleep C

24

R

0

2

–R =

0

2

Theta activity

Time (hr) Related to thermoregulatory changes

0

–

Beta activity

40

+

Sleepiness

Sleepiness

447

I

0

2

Stage 2 sleep

Sleep spindle

K complex

Seconds

Stage 3 sleep

Time (hr) Figure 8D-8    Schematic illustration of the three-process model of sleepiness regulation. Process H represents the homeostatic component of sleepiness, which increases with time elapsed awake, regulated by a sleep homeostat. H describes an exponential curve. The circadian component of sleepiness (C) is driven by the suprachiasmatic nucleus (SCN) producing a waking signal (−) during the subjective day and a sleepiness signal (+) during the subjective night. The sleep inertia process (I) describes a fast process that is active immediately after sleep and disappears in the following 1 to 2 hours in an exponential manner (purple area represents previous sleep duration). The exponential process R describes a relaxation-induced sleepiness, which starts immediately after lights-off or lying down (purple area). −R represents the inverse process of R occurring after lights-on or standing up (therefore, −R = I). Note that processes C, R, and −R (I) are coupled with thermophysiologic changes (distal vasodilation), whereas H is not coupled. (Redrawn from Kräuchi K, Cajochen C, Wirz-Justice A: Thermophysiologic aspects of the three-process-model of sleepiness regulation. Clin Sports Med 24[2]:287-300, 2005.)

Fig. 8D-9), tend to occur early during the sleep cycle, and it is during this deep sleep when many restorative processes and certain aspects of cognitive consolidation are believed to take place in the brain. Consequently, the early component of sleep is vital for the growth and repair of several of the body’s systems, including the immune and nervous systems. Protecting this sleep stage from disruption, alerting agents, and alcohol (known to reduce the depth of sleep) may thus be very important.

Insomnia Insomnia includes inability to fall asleep within 15 minutes of retiring, tendency to wake up during the night, inability to return to sleep following undesired awakening during the night, inability to remain asleep for desired full duration with awakening more than 1 hour before intended time of arousal, and consequent feelings of anxiety about being able to fall asleep. Insomnia is often temporary and is commonly associated with stress and anxiety, whose ­levels

Delta activity Stage 4 sleep

Delta activity REM sleep

Theta activity

Beta activity

Figure 8D-9    Electroencephalogram activity associated with stages of sleep. Awake: low-voltage, random, fast activity with superimposed alpha. Stage 1: 4- to 7-Hz theta activity. Stage 2: 11- to 15-Hz sleep spindles and K complexes. Stage 3: 20% to 50% high-voltage activity in the delta band (<2 Hz). Stage 4: >50%   high-voltage activity in the delta band (<2 Hz). Rapid eye movement (REM): low-voltage fast activity with superimposed theta (similar to the awake state). (Redrawn from Horne JA: Why we sleep: The functions of sleep in humans and other animals. Oxford, UK: Oxford University Press, 1988.)

tend to increase on pending competitive performances. Athletes should be asked about sleep difficulties resulting from the “pregame jitters” or from evening practices and competitions, and whether any self-correcting measures such as drinking alcohol are being used. It is also important to inquire about stiffness, pain, and urinary frequency, all of which may affect the duration and quality of sleep. The treatment of insomnia includes several nonpharmacologic approaches, including relaxation and meditation strategies, stimulus control, sleep restriction, and cognitive behavioral therapy (CBT), in addition to psychopharmacologic interventions, which are beyond the scope of this article.36 The effects of exercise for those with insomnia is

448

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

BOX 8D-1 Putative Acute Nonpharmacologic Methods for Promoting Sleep Onset Inverted Posture Neck, chest below heart Legs above heart Skin Warming and Core Cooling Prior hot bath Hot footbath Warm blankets, socks, and so forth Thermal biofeedback Ice consumption Motor Relaxation Prior stretching Prior inactivity Completely supported posture Voluntary muscular relaxation Sensory Withdrawal, or Masking and Habituation Visual: darkness (sleep mask, dark room) Auditory: quiet, rhythmic sound, or white noise Tactile, proprioceptive, vestibular: comfort, stillness Breathing Techniques Prolonged exhalation Carbon dioxide elevation Cognitive Relaxation Relaxation training Biofeedback From Cole RJ: Nonpharmacologic techniques for promoting sleep. Clin Sports Med 24(2):343-353, 2005.

discussed by Youngstedt.37 For a more indepth discussion of nonpharmacologic interventions for sleep induction, the reader is directed to Cole38 and Box 8D-1.

Sleep Apnea Apnea is defined as a complete cessation of airflow lasting at least 10 seconds, with the frequency of these episodes per hour determining the severity of the condition. Hypopnea is a respiratory event of at least 10 seconds with at least a 4% drop in oxygen saturations and a 30% reduction in airflow. For assessing the severity of the condition, a per hour of sleep apnea-hypopnea index (AHI) is calculated, with mildly abnormal values ranging from 5 to 15 per hour to severe cases that include more than 30 per hour.39 Certain athletes are predisposed to obstructive sleep apnea (OSA) based on the physical characteristics of the player. OSA, among the sleep-disordered breathing (SDB) group of disorders, is characterized by complete obstruction of the airway. Its prevalence in middle age is estimated to be 2% in women and 5% in men, while the prevalence of SDB in general is significantly higher, stated as 9% in women and 24% in men.40 It is estimated that 80% of OSA cases go undiagnosed. OSA is the consequence of repeated episodes of pharyngeal airflow obstruction, whether partial or

c­ omplete, during sleep despite continued respiratory effort, resulting in oxygen desaturations, increased sympathetic tone and arousals, and ultimately concluding with fragmented and nonrestorative sleep with daytime sleepiness. Snoring is extremely common in OSA, but it is not sufficient for the diagnosis. George and colleagues41 studied SDB in a sample of 302 professional football players from eight different teams. More than 20% of questionnaire respondents scored abnormal results on the Epworth Sleepiness Scale (ESS), a scale designed to measure daytime sleepiness and frequently used in the diagnosis of sleep disorders, and 92% had large neck circumferences, both of which are risk factors for OSA. Other factors include obesity (body mass index > 30), age, and male gender. The rate of SDB in these professional football players was greatly increased, especially in offensive or defensive linemen, easily the largest and strongest players on the gridiron. Excessive daytime sleepiness was present in a high percentage of players, but not all occurrences were accounted for by SDB. The diagnosis of OSA generally requires polysomnography during an overnight stay in a sleep clinic or lab. Ambulatory methods are growing and will ­ probably become predominant in sports medicine. The AHI score and the clinical presentations are critical for treatment decisions. Diagnosis and treatment of OSA, which is expected to improve performance and overall health by restoring regular breathing during sleep, reducing loud snoring, and minimizing daytime sleepiness, is discussed in detail by Emsellem and Murtagh.39 The continuous positive airway pressure (CPAP) system, first developed by Dr. Colin ­Sullivan in 1981, is still currently the gold standard in treating OSA and consists of the CPAP machine delivering a steady stream of compressed air up the nose while patient is wearing a facemask, forcing the air passages to remain open. CPAP, although proved an effective method in reducing the symptoms of OSA, has compliance estimates as low as 50%, as has been demonstrated in some studies, mainly due to discomfort and inconvenience. Besides the CPAP, other approaches have been studied with varying results, including oral appliances such as the mandibular advancement splint, which holds the tongue in a position that allows the air passages to remain open and which forces the lower jaw forward to allow more space for air to flow. Tonsillectomy and adenoidectomy surgery is now a commonly used method, especially for children whose swollen tonsils and adenoids impede healthy breathing particularly at night. Prescription medicine also may play a role in OSA in the treatment of the disorder and its symptoms. The mechanisms by which pharmacologic agents could potentially assist in the treatment of OSA include induced reduction in REM sleep (the phase of sleep in which sleep apnea tends to occur), increase in respiratory drive, and increase in muscle tone in the upper airway. Many medications have been studied as possible treatment methods of OSA, and the reader is directed to two literature review studies by Smith and colleagues42 and Abad and Guilleminault.43 Because one of the major symptoms and chief complaints of OSA is excessive daytime sleepiness, alerting agents such as stimulants have been used to counteract it. In a recent year-long study, Hirshkowitz and Black

Nutrition, Pharmacology, and Psychology in Sports

Memory Consolidation Sleep has an important role for consolidation of memories and motor skill learning. Memory consolidation is the process by which short-term memories or learned tasks are placed into long-term memory. In motor skill learning, the amount of training and practice exerts a powerful effect on further improvements in performance. Research studies have shown a delayed improvement effect over a 24-hour period and suggest that a post-training sleep period is required for optimal consolidation of motor skill learning tasks. This improvement was observed to be sleep stage specific with the amount of improvement significantly correlated to stage 2 NREM.45 Nishida and Walker replaced the overnight sleep in the previously mentioned study with 60- to 90-minute naps and detected a ­significant improvement in a motor skill task in the nap compared with the no-nap group (Fig. 8D-10).46 Here, most improvement also occurred during stage 2 NREM sleep, thus supporting the finding of Walker and Stickgold.45 It is also important to note that post-training sleep is not the only mechanism behind consolidation because functional magnetic resonance imaging studies have demonstrated that novel memories are actively processed in the brain during the first few hours of post-training ­wakefulness.47 In football, one of the most important tasks is learning the playbook. Learning typically more than 100 plays with descriptions and diagrams of each play can be a daunting task, and yet, when athletes on the field or court hear a play called, they had better know their role, as well as everyone else’s, in order to execute it properly. Similarly, in other team and individual sports, before a match, a coach will go over the game plan and provide a detailed analysis of the opponent. Not remembering this information could lead to confusion on the field or not properly anticipating what the opponent will do in a given situation. As just discussed, there exists a cognitive component in sports that is often overshadowed by the physical components. Indeed, success with either component is enhanced by improved sleep. Studies on sleep and memory consolidation focus on two types of memory: declarative (related to facts and information storage such textbook learning) and procedural (related to long-term memory of skills and task performance inherent to “how to” knowledge). NREM sleep, composed mostly of SWS, has been shown to improve ­hippocampus-dependent declarative memory performance, whereas REM and stage 2 sleep are reported to enhance hippocampus-independent procedural ­ memory.

*

30 % Improvement

studied the effect of modafinil, a psychostimulant alerting agent approved by the U.S. Food and Drug Administration and used in conjunction with CPAP in OSA participants.44 The results included a reduction in daytime sleepiness as measured by the ESS, an increase in functional status as assessed by the Functional Outcomes of Sleep Questionnaire, and an improvement in overall health as shown by the Short Form-36 Health Survey. Because of its capability of enhancing rather than just restoring performance, modafinil joined stimulants on the prohibited list of the World Anti-Doping Agency (WADA) in 2004.

20

*

449

*

*

10

0

Across Day

Overnight

Total

Time Course of Change = Daytime Nap Group

= Control Group (No Nap)

* = Significant change

Figure 8D-10    Daytime naps and motor skill learning. Subjects practiced the motor skill task in the morning and either obtained a 60- to 90-minute midday nap or remained awake across the first day. When retested later that same day, subjects who experienced a 60- to 90-minute nap (Across Day, dark purple bar) displayed significant performance speed improvements of 16%, whereas subjects who did not nap showed no significant enhancements (Across Day, pink bars). When retested a second time after a full night of sleep the next day, subjects in the nap group showed only an additional 7% increase in speed overnight (Overnight, purple bar), whereas subjects in the control group expressed a significant 24% overnight improvement following sleep (Overnight, pink bar). Therefore, 24 hours later, both groups averaged the same total amount of delayed learning (Total, purple and pink bars). Asterisks indicate significant improvement and error bars indicate SEM. (Redrawn from Walker MP, Stickgold R: It’s practice, with sleep, that makes perfect: Implications of sleep-dependent learning and plasticity for skill performance. Clin Sports Med 24[2]:301-317, 2005.)

In a study determined to examine the effect of a 45-­minute nap on the consolidation of procedural and declarative memory, sleep architecture in the participants was composed mostly of stage 2 sleep with low amounts of SWS and REM. Results concluded a positive correlation between procedural, but not declarative, memory im­provement and naps.48 In this regard, sleep should be just as important as practice and training because it is during the sleep process that memory consolidation takes place for both specific motor tasks and cognitive tasks such as the previously mentioned declarative memory; additionally, sleep is the recovery time when the process of repair and rejuvenation occurs in the body. Sleep deprivation and restriction studies have shown a variety of neurobehavioral impairments in attention span and working memory, cognitive deficiencies, and depressed mood. Not even extreme sleep restriction is required to observe abnormalities because research has shown that getting fewer than 6 hours of sleep can affect coordination, reaction time, and decision making. Prolonged wakefulness in a 17- to 19-hour sleep-deprived group displayed worse impairment in motor skill performance than the study group with a blood alcohol concentration (BAC) of 0.05%, and even longer periods without sleep reported performance task scores similar to individuals with a BAC of 0.1%, which is above the legal limit in the United States

450

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

of 0.08%.49 With a BAC of 0.1%, behavioral effects such as disinhibition and extroversion and impairment effects in reflexes and reasoning ability, among others, are moderately but noticeably handicapped.

GOALS OF A SPORTS CHRONOBIOLOGY CONSULTATION Minimizing Adversity The limited goals of a sports chronobiology consultation are to reduce or avoid altogether possible hindrances to performance as a result of circadian misalignments and sleep abnormalities. This includes addressing early morning, early afternoon, or very late evening dips in performance; jet lag, menstrual, or seasonal adversity; and minimizing or preventing the effects of less-than-optimal quantity or quality of sleep.50

Maximizing Performance Peak performance50-53 is achieved during certain intervals and durations of wakefulness, for instance during the late afternoon and early evening period of alertness known as the wake-maintenance zone. In particular, the loftier objective of a sports chronobiology consultation includes using natural means (e.g., timely light avoidance or exposure, naps) to align the endogenous circadian peak in athletic performance to the timing of the competitive event and aiming to perform with a minimal or, ideally, no “sleep debt.”

MEASUREMENT OF SLEEPINESS AND ALERTNESS Subjective Scales of Sleep, Sleepiness, and Alertness A number of scales have been used for assessing sleepiness (defined as the urge to sleep), such as the Stanford Sleepiness Scale and Karolinska Sleepiness Scale (KSS). The KSS54 is a 9-point sleepiness scale ranging from 1 (very alert) to 9 (very sleepy, difficulty staying awake, or fighting sleep). The ESS (Fig. 8D-11),55 for which normative data are available, is a self-report of sleepiness over a period of time, varying from several weeks to a month. It is particularly useful to quickly assess sleepiness and can be used to support a suspicion of certain sleep conditions such as sleep apnea, which is not uncommon in football players. The Pittsburgh sleep quality index (PSQI) is a validated questionnaire giving a global score of sleep quality and includes 19 questions grouped in seven component domains (sleep quality, latency, duration, efficiency, disturbance, medication, and daytime dysfunction). Cutoff points for disturbed sleep vary between 5 and 8. For an example of use of the PSQI in athletes, consult Samuels and colleagues.56

THE EPWORTH SLEEPINESS SCALE Name: Today’s date: Your sex (male–M; female–F):

Your age (years):

How likely are you to doze off or fall asleep in the following situations in contrast to feeling just tired? This refers to your usual way of life in recent times. Even if you have not done some of these things recently, try to work out how they would have affected you. Use the following scale to choose the most appropriate number for each situation: 0 = would never doze 1 = slight chance of dozing 2 = moderate chance of dozing 3 = high chance of dozing Chance of Situation dozing Sitting and reading Watching TV Sitting inactive in a public place (e.g., a theater or a meeting) As a passenger in a car for an hour without a break Lying down to rest in the afternoon when circumstances permit Sitting and talking to someone Sitting quietly after a lunch without alcohol In a car, while stopped for a few minutes in traffic Figure 8D-11    The Epworth Sleepiness Scale. (Redrawn from Emsellem HA, Murtagh KE: Sleep apnea and sports performance. Clin Sports Med 24[2]:329-341, 2005.)

Objective Tests of Sleep, Sleepiness, and Alertness Laboratory tests include multiple sleep latency tests, polysomnography, and the maintenance of wakefulness test. Polysomnography is an overnight monitoring of EEG, eye movements, chin electromyogram, limb movements, airflow (nasal and oral), electrocardiogram, chest and abdominal movements, and oxygen saturations. The polysomnograph assesses stages of sleep (“sleep architecture”), arousals, severity of disordered breathing, desaturations, and limb movements. Used since the mid-1970s, the Multiple Sleep Latency Test (MSLT) is an objective measure of daytime sleepiness and assesses the length of time the individual takes to fall asleep (sleep latency) at predetermined times throughout the day.57 Typically, the MLST is an 8-hour procedure consisting of five timed naps taken at 2-hour intervals starting in the morning. Sleep latency is measured during each of the naps in the absence of stimulating activity, with the subject lying quietly on a comfortable bed in a darkened, sound-attenuated room. According to protocol, the subject should have an empty bladder, should be neither thirsty nor hungry, should not have consumed any caffeine or stimulating medications, and should be wearing comfortable clothing during the testing. The sleep latency measurements are typically performed in a sleep laboratory bedroom. Brain wave patterns are scored in 30-second epochs, and the sleep latency is defined as the duration of the interval from the first recorded epoch until the first epoch scored as sleep. A conventional method of displaying MSLT is

SLEEP LATENCY (MINUTES)

Nutrition, Pharmacology, and Psychology in Sports

20 15 10 MEAN OF THE DAY 5 0 1000

1200

1400

1600

1800

HOUR Figure 8D-12    The results of an idealized Multiple Sleep Latency Test. The speed of falling asleep is related to the amount of sleep on prior nights. A standard series of calibrations is performed while the subject is awake that requires about 1 minute. The subject is then given the instruction, “close your eyes and try to go to sleep.” The precise second the lights are out, the sleep latency measure begins. The five individual test results connected by a line make up a “sleep latency profile.” Examining the profile can give an overall sense of daytime sleep tendency. Alternatively, the mean of the five individual tests can also be used as a descriptor of the overall strength of the daytime sleep tendency on this particular day. It must always be kept in mind that the individual test is terminated within seconds after the onset of sleep. It is extremely rare that it takes more than 30 seconds for an observer to make this decision. Each test is always terminated if the subject does not fall asleep in 20 minutes. (Redrawn from Dement WC: Sleep extension: Getting as much extra sleep as possible. Clin Sports Med 24[2]:251-268, 2005.)

shown in Figure 8D-12. The five values of the individual tests are often averaged into the mean for the day. For an interpretation of sleep latency values, see Figure 8D-13. Athletes may have short sleep latencies either because of sleep disorders (sleep apnea being the most common) or because of voluntary or involuntary curtailed sleep. In any case, for those with a short sleep latency, a ­restoration 20 Good Alertness 15 Sleep Latency 10 (Minutes)

Can Get By Borderline

5 Twilight Zone 0 Figure 8D-13    There is a general correspondence between the Multiple Sleep Latency Test (MSLT) mean score and the overall level or degree of daytime alertness. If an individual falls asleep in less than 5 minutes on every test (sometimes less than 1 minute), this indicates a very strong sleep tendency and a very strong sleep drive. The label “twilight zone” is meaningful in this respect because memory and clarity of thinking are usually substantially impaired in individuals whose MSLT score is less than 5. (Redrawn from Dement WC: Sleep extension: Getting as much extra sleep as possible. Clin Sports Med 24[2]:251-268, 2005.)

451

of sleep latencies toward the 15- to 20-minute average is the goal, with expected results of higher alertness and an increased energy level. This is achievable with specific interventions depending on the cause of sleepiness. The potential of sleep extension for improved athletic performance is suggested by Dement.58 Ambulatory tests include actigraphy and the psychomotor vigilance task (PVT).59 Activity monitors (e.g., actigraphs, actiwatches) are small devices that accurately record body movement, similar in concept to the seismograph. Unlike their original bulkier design, current actigraphs are water resistant and about the size of a wristwatch with sufficient memory to record continuously for up to several weeks depending on the sampling epoch length. Collected data are downloaded for display and analysis. Actigraph software derives periods of activity and inactivity, sleep-wake parameters such as total sleep time, percentage of time asleep, total wake time, number of awakenings, and levels of activity, and can also detect certain circadian parameters, such as acrophase (circadian rhythm peak). The PVT60,61 is administered usually at 2-hour intervals and measures the changes in reaction time and sustained attention resulting from sleep loss and circadian ­rhythmicity. With a standard duration of 10 minutes, PVT can quantify the effects of sleep deprivation as well as measure the effectiveness of naps or recovery sleep.62 Two time lapses, the slowest reaction time and the top 10% fastest reaction time, appear very sensitive to sleep loss and recovery sleep (Fig. 8D-14). These changes are further discussed by Van Dongen and Dinges.62 Considering its reliability, portability, and convenience, the PVT can be taken outside onto the field, and individualized graphs detailing performance can be shown to each athlete.

ALERTNESS-ENHANCING DRUGS Some athletes use performance-enhancing drugs, including stimulants, to circumvent decrements of performance that may be secondary to sleep loss or a misalignment between sleep and performance. The issue of stimulants as performance-enhancing drugs is discussed in more detail by Postolache and associates.63 Stimulant abuse can result in an irregular heartbeat, alarmingly high body temperature, and, potentially, cardiovascular failure or seizures. If an athlete discloses to his clinician the nontherapeutic use of stimulants, an unequivocal recommendation to stop use must be made immediately. A medical taper of the stimulants while the athlete is competing or training should never be considered because it would infringe on the basic tenet of the WADA code (i.e., that performance enhancers are never to be used by national or international athletes). Symptoms of withdrawal from stimulants are not life-threatening (as would be the case with alcohol or benzodiazepines) but are often hard to tolerate (as it would be the case with opiate ­withdrawal).

Caffeine Caffeine is present in coffee, tea, cocoa, chocolate, sodas, sports drinks, and many nutritional supplements (Table 8D-1) and is one of the most frequently used stimulants

452

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Psychomotor vigilance lapses

30 25 20 15 10 5 0

0

24

48

A

72

96

120

144

168

120

144

168

Time (hours)

10% fastest reaction times (msec)

300

280

260

240

220

200

B

0

24

48

72

96

Figure 8D-14    Performance data from 13 healthy young adult males (mean age ± SD: 27.3 ± 4.6 years) who spent 10 days in the controlled environment of a laboratory. After 1 adaptation day and 2 baseline days with 8 hours time in bed (23:30-07:30), they were assigned to a condition involving 88 hours of extended wakefulness. Thereafter, during the last 3 days of the experiment, they received recovery sleep each night. A subset of 7 subjects were allowed 7 hours time in bed (23:30-06:30) on the first 2 recovery days and 14 hours time in bed (23:30-13:30) on the last recovery day, whereas the other 6 subjects were allowed 14 hours time in bed on all 3 recovery days. Throughout scheduled wakefulness, subjects underwent cognitive testing every 2 hours. The cognitive test battery included a 10-minute psychomotor vigilance task (PVT). A, The number of lapses (reaction times =   500 msec) on the PVT. B, The average of the 10% fastest reaction times (in msec) on the PVT. In both cases, group averages are plotted against cumulative clock time. Purple bars indicate scheduled sleep periods; the 2 baseline nights and the first 2 recovery nights (7 hours time in bed) are shown. Dotted lines in the 88-hour sleep deprivation period indicate midnight. On the last baseline day (before the last baseline sleep period) and on the first day of sleep deprivation, psychomotor vigilance lapses were relatively rare, and fastest reaction times were relatively short. However, both psychomotor vigilance lapses and fastest reaction times increased significantly during the rest of the 88 hours of wakefulness. The progressive increases over days of sleep deprivation were modulated by a circadian rhythm: the number of lapses and the fastest reaction times were reduced during the diurnal hours compared with the nocturnal hours even after 3 days without sleep. Recovery sleep rapidly reduced the level of impairment; after 2 nights with 7 hours in bed, performance was almost back to the baseline level (for the recovery days, averages are shown for the 7 subjects who received 7 hours time in bed only). (Redrawn from Van Dongen HPA, Dinges DF.: Sleep, circadian rhythms, and psychomotor vigilance. Clin Sports Med 24[2]:237-249, 2005.)

Time (hours)

in the world. It has been widely reported to increase levels of cognitive performance and alertness, and when taken before exercise, has demonstrated ergogenic properties. Caffeine can be used to restore functional loss attributed to sleep debt or to boost performance and is often used as a countermeasure for sleep inertia. In a meta-analysis report covering 21 studies, the ergonomic effect of caffeine not only resulted in a reduction in estimated exertion effort but was also positively associated with an increase in performance by 11.2%.64 Caffeine has been downgraded from the WADA prohibited drug list, where it was once previously classified as a substance with both nutritional and doping properties, but is still part of WADA’s monitoring program. For an in-depth discussion about restoring performance, enhancing effects, and the safety implications of caffeine, see the review by Rogers and Dinges.65 The WADA website can be consulted for updates on the doping properties of caffeine and the implications of its excessive use.

SLEEP-WAKE CONSIDERATIONS IN ATHLETES WITH MOOD DISORDERS Dysfunctional sleep is common in depression. Although this often presents as insomnia,66 increased sleep length is not unusual because atypical depression is characterized by increased rather than decreased amounts of sleep. Disturbances of circadian rhythms are also commonly seen in depression,66 shift work, and jet lag, and a variety of chronobiologic interventions have been used targeting both the disturbances themselves as well as the underlying illness. Among these interventions, bright light treatment has become a first-line treatment for winter-type seasonal depression. Further discussion of seasonal depression and bright light treatment is provided in Postolache and Oren.27 Losing in competition, which is inherent to athletic life, may sometimes adversely affect an athlete’s sense of

Nutrition, Pharmacology, and Psychology in Sports

  TABLE 8D-1  Caffeine Content of Some Common Drinks and Food Items, Based on Typical Serving Sizes Drink or Food

Serving Size

Caffeine Content

Brewed coffee Instant coffee Double espresso Decaffeinated coffee Black tea Green tea Cocoa Coca-cola Diet Coke Pepsi Diet Pepsi Pepsi One Jolt cola Dr. Pepper Barq’s root beer Mountain Dew Sunkist Red Bull Snapple Nestea iced tea Nestea sweetened iced tea Milk chocolate Dark chocolate

237 mL (8 oz) 237 mL 60 mL (2 oz) 237 mL 237 mL 237 mL 150 mL (5 oz) 355 mL (12 oz) 355 mL 355 mL 355 mL 355 mL 355 mL 355 mL 355 mL 355 mL 355 mL 245 mL (8.3 oz) 355 mL 355 mL 355 mL 55 g (2 oz) 55 g

60-100 mg 50-80 mg 45-100 g 1-5 mg 30-100 mg 20 mg 30-60 mg 34 mg 45 mg 38 mg 36 mg 55 mg 72 mg 41 mg 22 mg 55 mg 41 mg 80 mg 32 mg 17 mg 27 mg 3-20 mg 40-50 mg

Reprinted from Rogers NL, Dinges DF: Caffeine: Implications for alertness in athletes. Clin Sports Med 24(2):e1-e13, 2005.

self-worth, self-competence, and can potentially trigger depression in vulnerable individuals. Moreover, a physical injury suffered by an athlete may bring about a significant decrease in social interaction, exercise, and light exposure, which, in confluence, may also trigger depression in a vulnerable individual. Successfully coping with an injury generally requires a supportive, positive atmosphere with the collective help of health care professionals, coaches, ­teammates, and family.

Mood Disorders An in-depth discussion of the treatment of mood and anxiety disorders and their relationship to sleep is beyond the scope of this article. However, sports physicians, coaches, and trainers should be mindful that excessive daytime sleepiness and sleep disturbances might be symptoms indicative of a larger problem, which, if properly diagnosed, may be treatable with appropriate psychotherapeutic and pharmacologic interventions. Therefore, rather than only attempting to correct the sleep complaints, a referral to either a sports psychologist or a psychiatrist should be considered.

Homeostatic and Circadian Impairments in Alertness and Sports Performance A homeostatic or circadian “hardship” may be the cause of impairment in alertness, which may be detected with a psychomotor vigilance task. The homeostatic hardship is a function of the appetitive nature of sleep; it increases during prolonged wakefulness and decreases as the need for sleep is fulfilled. The circadian hardship describes the possible misalignment between the circadian rhythm–dictated ideal performance time (measured historically as the hours

453

in which the athlete’s performance peaks) and the timing of the competition.

Homeostatic Adversity: Increased Appetite for Sleep and Consequent Impairment of Alertness Many athletes who have practices in the early morning do not go to bed early enough to allow for an adequate duration of sleep, often because of academic, occupational, or family social demands. With prolonged wakefulness, the homeostatic pressure for sleep continually builds in intensity following a saturating curve, but with the occurrence of sleep, the dissipation is very steep, observable even after a short nap of 10 to 20 minutes.62 In a study conducted to examine neurobehavioral functioning in sleep-deprived compared with sleep-restricted groups, the 4- and 6-hour chronic sleep restriction groups showed cumulative performance deficits after 2 consecutive weeks that matched those of the total sleep deprivation group.67 Voluntary sleep restriction is particularly relevant for young athletes, whose need for sleep never decreases but in fact may increase during adolescence, whereas personal lifestyle choices may result in the curtailing of sleep.68

The Power Nap Naps are an effective means to address homeostatic adversity in athletes, although there appears to be no consensus on the appropriate duration of the nap. In a recent study, Waterhouse and coworkers reported that a 30-minute postlunch nap between 1 pm and 1:30 pm, after a night of partial sleep restriction of 4 hours, improved sprint time, alertness, wakefulness, and short-term memory compared with those with the same sleep restriction but without the nap.69 Brief naps as short as 10 minutes have been shown to be effective in countering sleep restriction as measured by sleep latency, fatigue, alertness, and cognitive performance.70 A nap longer than 30 minutes may encroach on the deep sleep phase, where upon an abrupt awakening results in considerably more subjective grogginess than before the nap and places the individual in the physiologic state of sleep inertia. Reilly and Edwards suggest that the subjective improvement with a power nap in athletes is greater in those who habitually power nap rather than in those who are unfamiliar with this practice71 (Box 8D-2).

Sleep Gates, Wake Maintenance Zones, and Performance There are predictable “windows” of time when it is relatively easy to fall asleep that alternate with windows of time when it is difficult to fall asleep. The two intervals during which falling asleep is greatly enhanced are called sleep gates. The first sleep gate, called the postprandial dip, occurs during the early afternoon (Fig. 8D-15), is more frequently seen in morning-type individuals, and is further intensified by a high-carbohydrate meal. Monk72 associates the dip with an innate human propensity for sleep during the early afternoon hours. Postprandial dip is a misnomer, however, because this circadian dip occurs regardless of food intake and likely represents metabolic anticipation of a main meal

454

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 8D-2 R  ecommendations for Sleep ­Induction, Naps, and the Avoidance of Sleep Inertia

• Keep the room dark or use eye shades. • Keep the room quiet (if not possible, use earplugs). • Keep the air in the room cool and keep extremities

warm (covered with a blanket). horizontal position, ideally with the legs slightly ­elevated. • Take a hot bath shortly before nocturnal sleep (i.e., not before a nap). • Be sure that there is no discomfort associated with ­hunger, thirst, need to urinate, and so forth. • Turn phones off. • Have someone wake you up. The anxiety of not waking up on time may decrease the capability of falling asleep, and the need to hear an alarm clock may prohibit the use of earplugs or the reduction of noise. • Soft relaxing music, a monotonous recitation (e.g., a mantra), or both, may help induce a state of deep ­relaxation and sleep. • Do not use sleep-inducing medications for napping ­because their after-effects last longer than needed. • Do not attempt to nap early in the evening. The gate for sleep is commonly closed 1 to 3 hours before the habitual bedtime (a “forbidden zone” for sleep) and late morning, unless there is considerable sleep debt. The best time to nap is during early afternoon. • For athletes who have a competition that starts early morning and ends on the same day (e.g., Tae Kwon Do in the Olympics), bring a mat, a blanket, perhaps some mittens, pillows (including some for the legs), eyeshades, earplugs, and a player loaded with relaxing music. It is important to create a personal space. • For athletes who travel frequently and have problems falling asleep in unfamiliar environments, especially with mattresses that do not fit their preferences, carrying a portable foam mattress to place on top of the hotel bed may prove beneficial. • If napping for longer than 20 minutes, make sure there is at least a 2-hour period between the end of the nap and the beginning of the competition. • Minimize and shorten sleep inertia after a long sleep episode with a hot shower, caffeinated drinks in ­moderation, bright light, and light exercise.

• Use

From Postolache TT, Hung T-M, Rosenthal RN, et al: Sports chronobiology consultation: From the lab to the arena. Clin Sports Med 24(2):415-456, 2005.

rather than reaction to the meal. Short naps and, even more so, naps combined with caffeine and bright light are somewhat effective in counteracting the postprandial dip.73 The second sleep gate is the nocturnal window that starts in the late evening when “biological night” sets in and that lasts for a longer duration and is more omnipresent than the postprandial dip (see Fig. 8D-15). Certain interventions, including bright light, caffeine, and temperature manipulations used individually or in various combinations, can decrease sleepiness during the biological night.27,65,74

Even if it is not apparent to the individual, performance level is at its minimum during these sleep gates. The athletes would be competing, often unknowingly, against circadian adversity if a performance occurred during a sleep gate, as in a typical afternoon football game or a late-night (e.g., extra inning) baseball game.75 As a precaution, if an upcoming sporting event is known beforehand to occur during one of the sleep gates, prior exposure treatment to timed bright light can shift the circadian rhythm in anticipation of the match toward the more favorable outcome of peak performance. Using stimulants, including modafinil, to improve performance during these adverse circadian conditions is unethical and illegal because such treatments not only restore but also enhance performance.

Refractory Periods for Falling Asleep The refractory periods for sleep occur in the late morning (after sleep inertia has dissipated) and late afternoon and evening. People are often more alert before their usual bedtime than when they awaken in the morning. Thus, the evening refractory period is paradoxical because it usually occurs 3 hours before the habitual time for bed when conditions are constant. These sleep-restricted periods during the late morning and evening, respectively called the morning alertness zone and the wake maintenance zone (see Fig. 8D-15), are associated with the stimulating effect of the circadian system as it compensates for the accumulated time awake and associated sleep pressure, as described by Stiller and Postolache.31 Psychomotor, cognitive, and physical functions are performed at or near their peak at this time.

Evidence That the Wake Maintenance Zone is Optimal for Sports Performance Late afternoon to early evening hours are usually when world records are broken, as seen in runners, weight throwers, and 100- and 200-meter swimmers.2,76 For instance, early evening consistently yielded the peak speed of reaction time and enhanced muscle strength. However, the morning rather than the afternoon tended to be better for fine motor control (i.e., accuracy without speed, as is necessary for golf, darts, and archery) and certain cognitive tasks such as mental arithmetic and short-term memory (e.g., recall of complex coaching instructions). Two peaks in isometric strength of knee extensors and grip strength were seen in both the late morning and early evening, with a drop in muscular strength between 1 pm and 2 pm, coinciding with the early afternoon dip, a period when reaction time in a sizeable proportion of athletes is also reduced. In addition, peaks in short-term power output were reported during the late afternoon and early evening, with longer possible work times at 10 pm than at 6:30 am, as well as peak long jumps in the late afternoon. It is important to differentiate the reports from diurnal variation (Fig. 8D-16) and circadian variation (Fig. 8D-17). The diurnal variation may include either augmentation or dampening from duration of prior wakefulness (usually dampening), ­nutritional status (fast overnight may result in poorer performance early in the day, environmental temperature, joint stiffness, and lack of muscle warm-up in the morning). The

Nutrition, Pharmacology, and Psychology in Sports

455

WMZ

MAZ 09.00

Bedtime

11.00 13.00

15.00

18.00

Wake-up 04.00

22.00

09.00 SI

PD BN

Figure 8D-15   Sleep gates and alertness peaks. Sleep gates are in blue below the baseline, and peaks of alertness are in yellow/ orange above the baseline. In general, sleep gates are associated with poorer performance, and peaks of alertness with better performance. A superior player during a sleep gate may perform worse than an inferior player during a peak of alertness. MAZ, morning alertness zone (late morning); WMZ, wake maintenance zone (late afternoon/early evening); PD, postprandial dip (noon, early afternoon); BN, biological night (starts 2 hours before regular sleep onset time); SI, sleep inertia. (Redrawn from Postolache TT, Hung T-M, Rosenthal RN, et al: Sports chronobiology consultation: From the lab to the arena. Clin Sports Med 24[2]:415-456, 2005.)

1.5 1.0 0.5

1.5 1.0 0.5 0.0 –0.5 –1.0 –11 to –9 –9 to –7 –7 to –5 –5 to –3 –3 to –1 –1 to +1 +1 to +3 +3 to +5 +5 to +7 +7 to +9 +9 to +11 +11 to +13 –11 to –9 –9 to –7 –7 to –5 –5 to –3 –3 to –1 –1 to +1 +1 to +3 +3 to +5 +5 to +7 +7 to +9 +9 to +11 +11 to +13

0.0

effects—including the morning trough, early afternoon dip, and the evening top performance—even after unmasking. These observations may be important when attempting to set records or personal bests, or for qualifying for major events by meeting certain performance standards. Swim Performance (Z-transformed)

Swim Performance (Z-transformed)

circadian effects involve very time-consuming laboratory designs to untangle them from the masking factors listed previously. Methods such as a constant routine, forced desynchrony, and ultrashort sleep-wake cycle have been used experimentally but have probably little use ­clinically with individual athletes. It is, however, very important to note, as ­demonstrated recently, that athletes performing an ­athletic task (i.e., swimmers swimming), rather ������������������ than a surrogate one (������������������������������������������������� nonspecific vigilance test) retain the circadian

–0.5 –1.0 0200 1800 1400 2000 0200 0800 1400 2000 Environmental Time of Day

Figure 8D-16    Swim performance as a function of environmental time of day. Means and standard error for z-transformed performance values are shown, with lower scores representing better performance. The cyclic nature of performance across time is evident, with repeatedly and significantly faster swimming occurring during the 1100, 1400, 1700, 2000, 2300 trial times compared with the earlier monitored times of 0200, 0500, and 0800. (Redrawn from Kline CE, Durstine JL, Davis JM, et al: Circadian variation in sports performance. J Appl Physiol 102:641-649, 2007.)

Time Relative3 to Tmin Figure 8D-17    Swim performance relative to the core body temperature minimum (Tmin). Means and standard errors of z-transformed swimming trial times with lower scores depicting faster trials. Time of performance corresponded significantly with proximity to Tmin, with trials running from −1 to +3 hr relative to Tmin receiving significantly worse times than the trials run from −9 to +13 hr. Diurnally (considering that Tmin in that study was at about 3:00 am—earlier than before), a peak performance diurnal equivalent occurred at about 9 to 11 pm, a trough at about 2 to 4 am, and an early afternoon dip between noon and 2 pm. (Redrawn from Kline CE, Durstine JL, Davis JM, et al: Circadian variation in sports performance. J Appl Physiol 102:641-649, 2007)

456

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Name:

Date:

Score:

The Postolache-Soriano Athlete’s Morningness-Eveningness Scale (AMES) Directions: This Scale is designed to help you identify your chronotype, that is, your tendency toward a morning (“lark”), mid-range, or evening (“owl”) performance pattern. To complete this Scale, first print out the document. Then, read each question and consider all of the responses carefully. Then, complete each of the four items on this scale as accurately as you can; circle only one response per item. 1. At what time in the evening do you usually start feeling tired and in need of sleep? (7) A. 8:00 PM–9:30 PM (6) B. 9:31 PM–10:45 PM (5) C. 10:46 PM–12:30 AM (4) D. 12:31 AM–1:45 AM (3) E. 1:46 PM–3:00 AM 2. Suppose that you were able to choose your own competition hours. For some athletes, it might be useful to think about the 3-hour block when there would be a greater chance of feeling “in the zone,” or performing “at peak.” Which one of the following 3-hour blocks would be your most preferred time? (8) A. 6:00 AM–9:00 AM (7) B. 9:00 AM–Noon (6) C. Noon–3:00 PM (5) D. 3:00 PM–6:00 PM (4) E. 6:00 PM–9:00 PM (3) F. 9:00 PM–Midnight

Calculate your sleep score by adding the values in parentheses beside your circled answers. Total Score:

3. One sometimes hears about “feeling best in the morning” or “feeling best in the evening” types of people. Which type do you consider yourself? (8) A. Definitely a “morning” type (6) B. More a “morning” than an “evening” type (3) C. More an “evening” than a “morning” type (1) D. Definitely an “evening” type 4. Suppose that you were able to choose your own training (practice) hours, and organize all other daily routines to protect those hours. Which one of the following 3-hour blocks would be your most preferred time? (8) A. 6:00 AM–9:00 AM (7) B. 9:00 AM–Noon (6) C. Noon–3:00 PM (5) D. 3:00 PM–6:00 PM (4) E. 6:00 PM–9:00 PM (3) F. 9:00 PM–Midnight

10 to 12 13 to 17 18 to 23 24 to 28 29 to 31

= = = = =

Extreme Evening Type Moderate Evening Type Midrange Moderate Morning Type Extreme Morning Type

Figure 8D-18    The Athlete’s Morningness-Eveningness Scale questionnaire. (From Postolache TT, Hung TM, Rosenthal RN, et al: Sports chronobiology consultation: From the lab to the arena. Clin Sports Med 24[2]:415-456, 2005; adapted from Horne JA, Ostberg O: A selfassessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol 4[2]:97-110, 1976.)

Morningness-Eveningness

Circadian Markers

Some people naturally tend to stay up very late at night, whereas others wake up very early. Although environmental factors play an important role, endogenous circadian rhythms have been shown to determine the chronotype, or the “morningness-eveningness” preference, in these so-called “morning larks” and “night owls.” The original Morningness-Eveningness Questionnaire was formulated by Horne and Ostberg78 and has been followed by ­several others, such as the Diurnal Type Scale by Torsvall and Akerstedt79 and the Morningness Composite Scale by Smith and collegues.80 The Postolache-Soriano Athlete’s ­ Morningness-Eveningness Scale63 (Fig. 8D-18) is a morningness-eveningness questionnaire adapted for ­athletes.56,63

When designing interventions for shifting circadian rhythms and for measuring the timing of circadian rhythms, certain physiologic markers are essential. One of the most important is the timing of the core body temperature minimum (Tmin), which is typically reached 2 hours before habitual wake time. Tmin should be measured, for this purpose, in constant conditions during extended wakefulness in near darkness with regular intake of water and food, using a rectal probe in a semireclined position. To eliminate “masking” by the sleep-wake cycle, food and water metabolism, and positional physiologic changes (e.g., orthostatic noradrenergic activation), the temperature reading should be monitored continuously. Except during well-designed circadian research study protocols, this

Nutrition, Pharmacology, and Psychology in Sports

“constant routine” procedure is essentially impractical for athletes. Even though rectal temperature is considered the gold standard in measuring core body temperature, recent chronobiology studies have made use of intra-aural devices capable of measuring temperature through the tympanic membrane.81 Intra-aural devices are considered more convenient and well tolerated but are not considered as precise as the traditional rectal temperature method. An ingestible pill telemetry system for monitoring core temperature is expected to gain a larger use in the years to come.

Hormonal Markers Predictable rhythmic changes of certain hormones are among the most distinctive markers of biological (“internal”) day-night alterations. Two of the most important are cortisol and melatonin. Cortisol levels are at their highest in the early morning, with a peak typically occurring between 6 am and 8 am, and the lowest levels of cortisol are usually detected at midnight. Melatonin is secreted during the biological (internal) night, a period defined by hormonal, electrophysiologic, and behavioral parameters, its timing and duration reflecting prior exposure to light and dark cycles. Its secretion is, thus, limited to the biological night and begins, in most cases, about 2 hours before habitual bedtime. The evening increase in melatonin secretion is associated with an increase in sleep propensity. Because melatonin levels are suppressed by bright light, saliva or blood samples taken to measure these levels should be collected in dim light. Dim light melatonin onset (DLMO) measured by saliva is recommended as a highly replicated surrogate measure because blood measurements are often inconvenient. There is a relatively abrupt and predictable onset of melatonin secretion preceding the onset of sleep, so the DLMO is a conveniently measured circadian marker. The procedures for the collection of saliva for circadian phase assessments using DLMO are as follows: • Subject must remain awake in dim light (<10 lux) during assessment (e.g., in a dark hotel room, a TV is allowed if seen on a small screen from at least 4 meters). • To prevent sample contamination, use of toothpaste or mouthwash during phase assessment is prohibited. Caffeine, chocolate, bananas, and use of lipstick must be avoided for 5 hours before testing. • Small snacks and fluids are allowed except for 10 minutes before samples taken. • A 2-mL saliva sample should be obtained every 30 ­minutes. • Saliva sample should be centrifuged, frozen, and sent to a laboratory for analysis.

Evidence of Circadian Regulation in Athletic Performance Time-of-day differences in sports performance may be positively or negatively affected by environmental masking factors such as ambient temperature and light and by behavioral masking factors like food intake, exercise, and sleep-wake cycle. Without properly controlling for them, a time-of-day rhythm in performance may either mask or spuriously suggest a circadian rhythm in performance. In an

457

effort to control for these potential confounding masking influences in sports performance, Kline and colleagues devised an unmasking, rigorous constant-routine protocol to expose the underlying circadian rhythmicity in swimming performance.77 In this recent study, 25 trained swimmers were observed for 50 to 55 hours with a constant ultrashort sleep-wake cycle of 1 hour of sleep in complete darkness followed by 2 hours of wakefulness in dim light. The swimmers, instructed to perform as if in a race, were randomly selected to swim 200-meter trials during six of the eight time intervals (2 am, 5 am, 8 am, 11 am, 2 pm, 5 pm, 8 pm, and 11 pm), with a 9-hour gap between performances. The results of the study showed performance was significantly higher for the 11 am to 11 pm intervals compared with 2 am to 8 am. The best performance timing occurred at 11 pm, which coincides with 5 to 7 hours before Tmin, whereas the worst performance occurred at 5 am, which coincides with Tmin ± 1 hour (see Figures 8D-16 and 8D-17). The athletes’ core temperatures were taken intra-aurally to determine circadian phase. With a mean swim trial of 169.5 seconds, the difference between the circadian peak and trough in the performance of the population sample was 5.84 seconds. These results are similar to other studies involving swim performance in that they too found improved performance in the afternoon and evening as opposed to the morning. This study showed a marked circadian variation in ­performance while controlling for environmental and behavioral masking factors. Reilly and associates evaluated various soccer proficiency skills at various times of the day, relative to Tmax, in an effort to detect diurnal variations in performance levels.82 Juggling skills were significantly affected by time of day and exhibited the highest rating at 4 pm, and wall­volley trials showed a peak in performance at 8 pm. In a separate study, dribbling performance peaks occurred at 8 pm, whereas chip test performance displayed greater accuracy at 4 pm compared with 8 am. The maximal body temperature was correlated with higher skill performance between 4 pm and 8 pm, coinciding with markers of peak physical performance. Self-rated high alertness and low fatigue were found to occur at 8 pm. In a study funded by Major League Baseball, an analysis was performed to determine the extent of circadian adversity in a retrospective analysis of more than 24,000 games spanning from 1997 to 2006.83 Circadian advantage was calculated as the number of time zones crossed while traveling eastward, whereas each western-traveled time zone counted as negative values. Home-field advantage was controlled for by including only the games for which the away team held a circadian advantage in the analysis. Results demonstrated a nominal winning percentage (.517) for teams with both a 1-hour and 2-hour circadian advantage; however, the 3-hour advantage teams showed a more pronounced winning record of .603. In contrast to other studies, direction of travel displayed no beneficial impact for either eastward- or westward-bound teams.

JET LAG Transcontinental travel is a common occurrence for college and professional athletes; however, it is potentially disruptive for sports performance. Chronobiology ­interventions

458

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

are critical for minimizing and shortening these disturbances. Changes in performance levels may result even from crossing only a few time zones. For example, Recht and colleagues84 showed that 1.24 more home runs were hit by the home team when playing against a visiting team that had traveled eastward for a Major League Baseball game. Although it usually represents a liability, when transmeridian travel helps in avoiding an alignment between a sleep gate and duration of the game or competition, it may offer a circadian advantage. Smith and associates85 showed that Monday Night Football games give West Coast teams an advantage in terms of playing during the wake maintenance zone (with its accompanied peak of performance) compared with East Coast teams. Their home advantage is enhanced playing at night on the West Coast but is eliminated for games played at night on the East Coast.85 This is consistent with the finding of Jehue and colleagues86 that there exists an advantage for West Coast teams over East Coast and Central teams playing night games, both at home and away, but a disadvantage for West Coast teams in day games played on the East Coast. In basketball, the home-court advantage was almost ­nullified against a team traveling west to east, with the visiting team scoring on average 4 more points87 than the team traveling east to west. This effect, similar to the findings for Monday Night Football games, may be caused by West Coast visitors to the East Coast playing night games at an earlier time according to their internal clock. For an in-depth discussion of jet lag, the reader is directed to Reilly and colleagues88 and Postolache and associates.63 Circadian rhythms are most strongly affected by light when the exposure occurs closely before or after Tmin. When light is received before Tmin, the circadian rhythm becomes phase delayed. Conversely, when light is received after Tmin, a phase advance occurs (Fig. 8D-19).

Bright Light Exposure and Avoidance for Jet Lag The principles of bright light application are based on the phase response to light89 and several laboratory simulation protocols and field studies.90 The method to synchronize the body clock most efficiently to a different time zone depends on the direction of travel; with eastward travel, phase advances are desirable, whereas phase delays are preferable with westward travel. Local time for the temperature minimum for the first day at travel destination(Tmin2), an essential marker separating phase delay from phase advancing effect of exposure to light (see Figure 18D-19), can be calculated from the timing of temperature minimum at (Tmin1): Eastward travel Tmin2 = Tmin1 + number of time zones crossed d Westward travel Tmin2 = Tmin1 − number of time zones crossed

One can design paradigms of light exposure and avoidance for jet lag, based on the formulas given previously.

3 2.32 Phase Advance (Hr)

2 1

1700

2100

0100

0900

1300

1700

1 2 2.7

Phase Delay (Hr)

3

Figure 8D-19    Phase response curve (PRC) to the bright light phase advances (positive values) and delays (negative values) are plotted against the timing of the center of the light exposure relative to the core body temperature. Units on the y-axis are hours. The maximal phase delay (2.7 hr) exceeds the maximal phase advance (2.3 hr). (Adapted from Khalsa SB, Jewett ME, Cajochen C, et al: A phase response curve to single bright light pulses in human subjects. J Physiol 2003;549[Pt 3]:945-52.)

To apply these principles, it is crucial to estimate Tmin because of the importance of the inflection point from when light phase-delays to when it phase-advances. It is difficult to determine Tmin in real-life situations, especially in elite athletes, because it refers to temperature minimum measured in constant conditions in the absence of sleep. Therefore, surrogate measures must be used to estimate Tmin based on the assumptions that Tmin will fall between 4 am and 5 am in most individuals, between 2 am and 4 am in morning types, and between 5 am and 7 am in evening types. If sleep onset and offset have been generally regular for the last week, an acceptable estimation of Tmin is to use a sleep-related measure, such as sleep duration midpoint, and add 1 to 2 hours. An even more precise estimation involves measuring DLMO and then adding 7 hours to the DLMO. Preflight adjustment to travel, along with several days of course-correcting shifting at the location of origin, is highly recommended to hasten circadian adaptation,90 although this is rarely feasible.

Room Light and Jet Lag Even lower intensities of light, such as those present in normal indoor illumination, have been shown to suppress melatonin and shift circadian rhythms.91 For instance, room light (<400 lux) was shown to alter circadian rhythms.92 Thus, exposure to room light in the hotel or at practice, if extended over several hours, should be considered as activating from a circadian standpoint and may either help or hurt the adjustment to the new time zone depending on whether it occurs during a favorable or unfavorable interval. A phase response curve (PRC) illustrates the relationship between phase shifts and circadian timing of a stimulus, for example, bright light. Brighter light should be used in the room during the favorable interval on the PRC, whereas during the unfavorable interval, lights should be dimmed or eliminated. Because short-wavelength light is a

Nutrition, Pharmacology, and Psychology in Sports 2

459

Melatonin and Jet Lag

Phase Advance (Hr) 1 A

B

2100

0900

1300

C 1700

2100

1 Phase Delay (Hr) 2 Figure 8D-20    Phase response curve (PRC) to melatonin phase advances (positive values) and delays (negative values) are plotted against the timing of the center of the light exposure relative to the core body temperature. It can be observed that the PRC to melatonin is relatively less ample than the one to light (not more than 1 hour maximum) and that overall the effects are nearly mirror imaging the PRC to light (i.e., late afternoon melatonin phase advances while light at the same period phase delays, and morning melatonin phase delays while light then phase advances). (Redrawn from Postolache TT, Hung TM, Rosenthal RN, et al: Sports chronobiology consultation: From the lab to the arena. Clin Sports Med 24[2]:415-456, 2005.)

more potent circadian shifter, during a potentially adverse time on the PRC (Fig. 8D-20) sunglasses with selectively low transmission in the blue range could be used indoors for reading or watching television. It has been recently reported that short-wavelength blue light suppresses ­melatonin production; thus, wearing blue-blocking glasses for a specified time before sleeping causes a delay in the secretion of melatonin resulting in a phase advance of the circadian rhythm.93 Table 8D-2 and Table 8D-3 provide general strategies for alleviating jet lag using light exposure and avoidance.

Melatonin, secreted during the biological night, is synthesized from serotonin in the pineal gland and suppressed by bright light to the eyes. Melatonin can be understood as a “dark pulse,” which indicates, in diurnal species, that it is time to rest and sleep. When melatonin is administered during the internal early morning, a phase delay occurs, and when ­administered during the internal late afternoon and early evening, the consequence is a phase advance of circadian rhythms (see Fig. 8D-19). Thus, the PRC to melatonin is a mirror image of the light PRC (see Fig. 8D-19).94 When melatonin and bright light are administered at phase-appropriate times, the combination is additive. For instance, evening light with morning melatonin will cause a strong phase delay, and morning light in combination with evening melatonin will cause a strong phase advance (see Figs. 8D-19 and 8D-20). Smaller doses of melatonin (0.5 to 1 mg) are recommended for shifting circadian rhythms, whereas larger doses (3 to 10 mg) are often used as a sleep inducer. The timing of melatonin administration is critical for the optimal treatment of jet lag. For details, the reader is directed to Postolache and colleagues.63 Melatonin may induce short-term reductions in mental and physical performance (e.g., psychomotor vigilance), possibly in association with its hypnotic and hypothermic effects.95 Although the decline in physical performance is generally short-lived, decrements in vigilance may persist for 3 to 5 hours or more after administration. However, the impairment in vigilance does not occur if the individual gets adequate sleep after the administration, as discussed by Atkinson and Drust.96 The day after taking melatonin, even in higher doses (5 mg), physical performance is not affected.97 Although melatonin has been used successfully to alleviate symptoms of jet lag, there are no convincing data that melatonin improves athletic performance, and the small amount of supportive evidence available is

  TABLE 8D-2  General Strategies for Light Exposure and Avoidance of Jet Lag: Eastward Travel* Interval First day of arrival: Desirable bright light avoidance Critical light avoidance Desirable bright light exposure Critical light exposure

Description

Start Range

End Range

12-hr time interval

Tmin1 + number time zones crossed - 12 Tmin1 + number time zones crossed - 9 Tmin1 + number time zones crossed

Tmin1 + number time zones crossed Tmin1 + number time zones crossed - 1 Tmin1 + number time zones crossed + 12

Tmin1 + number time zones crossed + 1

Tmin1 + number time zones crossed + 9

8-hr time interval At least 1 hr of light exposure within 12-hr interval (earlier exposure in range more beneficial) At least 1 hr of light exposure within 8-hr interval (earlier exposure in range more beneficial)

The following days: shift the time of bright light and dark exposure 2 hours earlier each day and stop the regimen of light avoidance when Tmin1 coincides with the desired wake-up time, continuing bright light exposure as early as possible after awakening. *The goal is to phase advance and to avoid phase delay. Tmin1=timing of temperature minimum at the point of origin, an essential marker for the effects of light exposure, separating circadian phase delays (before Tmin1) to advances (after Tmin1). Reprinted from Postolache TT, Hung TM, Rosenthal RN, et al: Sports chronobiology consultation: From the lab to the arena. Clin Sports Med 24(2): 415-456, 2005.

460

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 8D-3  General Strategies for Light Exposure and Avoidance of Jet Lag: Westward Travel* Interval First day of arrival: Desirable bright light avoidance Critical light avoidance Desirable bright light exposure Critical light exposure

Description

Start range

End range

12-hr time interval

Tmin - number time zones crossed

8-hr time interval

Tmin - number time zones crossed + 1

At least 1 hr light exposure within 12 hr interval (earlier exposure in range more beneficial) At least 1 hr light exposure within 8 hr interval (earlier exposure in range more beneficial)

Tmin - number time zones crossed - 12

Tmin - number time zones crossed + 12 Tmin - number time zones crossed +9 Tmin - number time zones crossed

Tmin - number time zones crossed - 9

Tmin - number time zones crossed - 1

The following days: shift the time of bright light and dark exposure 2 hours later every day, until the Tmin falls within 2 to 3 hours before wake-up time, then stop the regimen. *The goal is to phase delay and to avoid phase advance. Reprinted with permission from Postolache TT, Hung T-M, Rosenthal RN, Soriano JJ, Montes F, Stiller JW. Sports Chronobiology Consultation: From the Lab to the Arena. Clin Sports Med 2005;24(2):415-456.

c­ ontroversial. The most common mistakes made in melatonin use include the following63: • Wrong timing of melatonin administration. Indiscriminately taking melatonin as a sleep inducer can cause circadian shifts in an unwanted direction in certain circumstances. • The administered dose being too high. For phase shifting, dosages of 0.5 to 1 mg are sufficient. • Not checking the purity of the melatonin preparation. Although melatonin is not on the WADA prohibited drug list, the preparation of melatonin may contain impurities that could result in a positive doping test. Melatonin is marketed and sold as a food supplement in the United States. It is important to mention that it is not approved for the treatment of any disorder by the U.S. Food and Drug Administration.

Situations in Which It May Be Detrimental to Make a Circadian Phase Adjustment to the Destination Time Circadian shifts to the new time zone should be avoided when there is not enough time to comfortably readjust following a short trip away from home and when a more important competition or home game is scheduled a short time after the return. In fact, all efforts must be made to maintain as much as feasible the light-dark exposure, sleepwake activity, and meal schedule so as to be synchronized with the home rather than the destination’s schedule. Another situation in which a circadian shift to the new time zone should be avoided is during those rare situations in which transmeridian travel results in a better alignment between the time of the competition and the endogenous time of peak performance, especially when competition occurs during the opponent’s sleep gate. In this manner, the traveling athlete will have an advantage: the individual will be playing during an interval of time (either the morning alertness or wake maintenance zone) of maximal alertness and psychomotor functioning and the opposing athlete

will be competing in a time interval (one of the sleep gates) during which he or she is more prone to sleep and perform poorly on psychomotor skills. For instance, when a player who lives on the eastern shore of the United States travels for a match on the West Coast scheduled at 3 pm local time (falling within the postprandial dip for a local player), the East Coast player may have a circadian advantage because, if they have not adjusted to the new time zone, they will start play just as though it were 6 pm, which generally falls at the beginning of the wake maintenance zone. To some degree, this advantage may compensate for the effects of playing away from home and the accompanying travel fatigue. Similarly, a West Coast player may have an advantage when traveling from the West to the East Coast for a latenight competition. If not adjusted, he or she will play during the wake maintenance zone, whereas the East Coast athlete may experience impaired psychomotor vigilance associated with the onset of the body clock’s “internal night” during the competition.95

C

r i t i c a l

P

o i n t s

l Sleep quantity and quality are essential for health and performance. l Alternations between circadian peaks and troughs of performance occur naturally and are driven by the “body clock,” located in tiny bilateral nuclei of the hypothalamus, called the suprachiasmatic nucleus. l Adding sleep and chronobiology consultation to the arsenal of modern sports science may contribute to a more rapid and complete restoration of performance by preventing and correcting (a) sleep debt and (b) misalignments between the endogenous rhythms and the competitive demands. l Sleep apnea requires specialized diagnosis and treatment. l Timed short naps could be very important restorative factors. In addition to relaxation techniques, natural methods to promote napping and falling asleep easier are presented.

Nutrition, Pharmacology, and Psychology in Sports

l Timed bright light exposure and avoidance is a very effective method to shift circadian rhythms and has a very important place in managing jet lag. l The practice of chronobiology must compromise with issues of feasibility, logistics, and compliance, in contrast to the science of chronobiology, which, to a great degree, relies on precise measurements in highly controlled ­environments. l Naturally restoring peak performance with sleep and chornobiologic interventions could be important elements of a successful rehabilitation of athletes who have used illegal stimulants for performance enhancing, as well as in athletes with overtraining syndromes. l Ideally, assessments and interventions should be tailored to individual athletes, taking into account their previous performance at different times of the day, in different seasons, and in women in different menstrual phases. Also, important considerations should be made regarding the individual’s chronotype, competitive goals and schedules, sleep problems, total sleep requirement, and sleep debt, as well as response to transmeridian travel. Rehabilitation after physical injury could represent opportunities for chronobiological evaluations and individual interventions. l In addition to individual interventions, team level interventions (education of coaches, managers, trainers, team physicians, and sports psychologists) are important for planning practice/competition and rest schedules, reducing jet lag, anticipating, identifying, and addressing forthcoming chronobiologic adversities.

S U G G E S T E D

461

R E A D I N G S

Dunlap JC, Loros JJ, Decoursey PJ: Chronobiology: Biological Timekeeping. ­Sunderland, MA, Sinauer Associates, 2004. Lamberg L: Body Rhythms: Chronobiology and Peak Performance. Lincoln, NE, ASJA Press, 2003. Postolache TT: Sports chronobiology. Clin Sports Med 24(2), 2005 (entire issue). Waterhouse J, Minors D, Reilly T, Waterhouse M: Keeping in Time with Your Body Clock: A Guide to Maximising Your Mental and Physical Potential. New York, Oxford University Press, 2003.

R eferences Please see www.expertconsult.com

C H A P T E R   

9

The Young Athlete Mininder Kocher Sports injuries are being seen with increased frequency in the pediatric and adolescent athlete because of increased participation in higher competitive levels at younger ages, increased recognition of injuries in this age group, and the advent of arthroscopy and magnetic resonance imaging. The pediatric athlete differs from the adult athlete in terms of physiology, growth, psychology, and skills. Injury patterns are age and sport specific. An understanding of the special considerations of the pediatric athlete and the common injury patterns is necessary for the successful management of sports injuries in these patients.

EPIDEMIOLOGY Epidemiology of Pediatric Sports Participation During the past 30 years, there has been a significant increase in the number of children and adolescents participating in physical activity and team sports, with the largest increase among adolescent females.1 The overall trend has seen a shift from the largely unstructured, unsupervised free play of the early 20th century to the evolution of organized and highly structured youth sports activities.2 It is estimated that at present up to 30 million children and adolescents participate in organized sports in the United States. In 1995, reports indicated that 15 million 5- to 14-year-olds played baseball within the United States.3 The Youth Risk Behavior Survey (YRBS) was a large population-based study performed throughout the 1990s that enabled accurate assessment of the emerging trends in youth sports participation. Results from the 1997 survey reported that 62% of U.S. high school students participated on one or more sports teams, with the majority playing in a combination of both school and nonschool teams.4 The YRBS study highlighted a number of significant demographic differences when results were compared for age, gender, and ethnicity. Although the number of women participating on sports teams has increased ­fivefold during the past 30 years, a disparity continues to exist between genders according to the 1997 YRBS study.1 Although almost 70% of male high school students participate in sports, only 53% of similarly aged females exhibit the same level of sporting interest.1,4 This gender disparity was even more dramatic among ethnic ­ minorities, with only 40% of Hispanic and African American females participating, compared with 62% and 71% for males, respectively.4 Furthermore, progression into adolescence was associated with a reduction in the involvement of both males and females in vigorous sporting activities.1,4 In males, a reduction in vigorous exercise participation occurred from 81% in grade 9 to only 67% by grade 12.1 Vigorous exercise

was defined as activity causing shortness of breath, lasting at least 20 minutes, 3 days a week.1 As expected, this trend was even greater in females, with 61% of female 9th graders participating in vigorous exercise, compared with only 41% by 12th grade.1 The growth and increasing popularity of school and community youth sports programs have become an integral part of American youth culture with the potential to benefit the long-term physical and psychosocial health of those children and adolescents who participate.4

Epidemiology of Pediatric Sports Injury Increased youth participation in sports and physical activities has resulted in an increase in sports-related injuries secondary to trauma and overuse.2 The annual rate of sports injuries within the United States is estimated at about 3 million, with up to 70% of those resulting from youth sports activities.3 The financial costs of managing these injuries in 1996 was well in excess of $1 billion.3 Pediatric sports injuries are often unique in terms of not only the underlying pathology but also the challenges in managing these injuries. Many patients participate in multiple teams during a given season, the rest periods between seasons are short if existent, and there is increasing demand for sporting success from parents, schools, and sporting establishments.5 Pediatric sports injuries can be classified according to the age of the athlete, the type of injury, and the sport or activity responsible for an injury.6 From an epidemiology standpoint, these classifications assist in the identification of potential risk factors for injury and the implementation of prevention strategies and rehabilitation plans that are appropriate for the age of the patient and the sport. Several studies have identified a correlation between increased risk for sports-related injury and increased age of the pediatric athlete.6 A number of explanations to explain these findings have been postulated. These include greater opportunity for injury in the adolescent athlete due to longer game times and more frequent and intense practices.6 The provision of medical assistance at many high school and college games allows for greater reporting of injuries.6 It appears that anatomic factors such as the increased size of the athletes and the ­resultant increased force and speed of collisions play an insignificant role because the same trend was noted for both ­contact and noncontact sports.6 Sports injuries can be broadly divided into acute traumatic and overuse-type injuries according to their pathophysiology.6 Whereas many acute traumatic injuries are the result of random events, overuse injuries are often the result of entrenched training errors and therefore have greater potential for prevention.6 The difficulty lies in identifying these overuse injuries because initially they are 463

464

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

only subtly disabling when compared with an immediate fall to the ground following a sprain. It is important that an injury be viewed in context of the sport in which it occurred because an injury that may be functionally disabling for one sport may have no relevance in another sport.6 Furthermore, it is important that physicians recognize that time lost from sports participation is often more of a concern to athletes and their coaches than the nature of the injury itself.6 These perceived differences in injury severity inevitably affect management programs. Among school athletes, football has the highest rate of injury, with wrestling not far behind.3 The rate of injury in both males and females at the high school and college level are comparable, with the exception of knee injuries, which are slightly greater in females at a college level.7 Fortunately, fatal sports injuries are rare. In a study conducted by Mueller and colleagues, 160 nontraumatic deaths occurred in high school and college athletes in the United States between 1983 and 1993, with the primary cause being ­cardiac death and only a small number of heatrelated injuries.8 They also reported 53 traumatic deaths from 1982 to 1992 in football resulting primarily from head and neck trauma.8

EXERCISE PHYSIOLOGY Endurance Training The increased popularity of endurance sports, such as swimming, running, rowing, and cycling, among children and adolescents has heightened awareness of aerobic training as a means of maximizing performance.9,10 The beneficial effect of aerobic training in adults is now well established, with increases in maximal oxygen uptake (Vo2max) of up to 15% to 20% reported in the literature.10 The ability to enhance the aerobic capacity of children and adolescents through endurance training remains controversial, however, because many of the studies to date have been methodically flawed and largely neglected adolescents.9-11 Although there are several physiologic parameters by which to measure aerobic fitness, Vo2max is the most commonly used in studies involving adult endurance.9,10 The usefulness of this parameter in children was questioned because most children fail to ever reach the plateau consistent with V �o2max.9,10 As a result, Vo2max has been replaced with peak Vo2 in pediatric endurance studies, which instead measures the highest Vo2 level achieved before the point of voluntary exhaustion.9,10 Despite the traditional view that prepubescent children are incapable of improving their aerobic capacity through endurance training, evidence is now emerging in the literature to the contrary.9,10 A review ������������������������ of 22 studies ���������� by Baquet and colleagues demonstrated that a 5% to 6% increase in peak Vo2 among both children and adolescents is possible with appropriate aerobic training.9 The ability to achieve these increases is influenced by several factors, including baseline peak Vo2 levels, program design, maturity level, and genetics.9,10 The role of pubertal status on a child’s ability to enhance aerobic capacity through endurance training remains unclear because of a lack of quality longitudinal data.9,10

Early research indicated that for the same relative training intensity, greater gains in peak Vo2 were demonstrated for circumpubertal relative to prepubertal subjects.9,10 Two theories have been used to explain this. First, a so-called maturational threshold below which training-induced adaptations in aerobic fitness were physiologically limited, and second, the greater level of habitual activity among children, maintained their Vo2 closer to its maximum potential making additional increases in peak Vo2 more difficult to achieve.9,10 Although limited, evidence is slowly emerging to contradict these theories as a better understanding is gained of the role of genetic, environmental, and endocrine influences.9-11 High-quality longitudinal studies that document not only chronologic age but also maturity status are essential.10 Designing a program that incorporates appropriate levels of training duration, frequency, and intensity is essential to achieving the desired increase in aerobic capacity.9-11 Baquet’s literature review a found that three or four sessions per week, lasting 30 to 60 minutes in duration, were optimal.9 Interestingly, no clear relationship was found between the length of training program and peak Vo2 improvement.9 Training intensity is generally defined in terms of the percentage of maximal heart rate.9,10 Several studies have confirmed that a heart rate that exceeds 80% of maximum is required to obtain significant increases in peak Vo2.9 Comparison between continuous and interval training and the effect on peak Vo2 is limited to prepubertal children.9,11 Nine of the 16 studies reviewed by Baquet and colleagues demonstrated a significant increase in peak Vo2 following continuous training.9 However, only 3 of the 16 studies showed improvement when the heart rate was less than or equal to 80% of the maximum.9 The implementation of continuous training among children poses difficulties with regard to compliance and motivation.11 Interval training not only is easier to put into practice but also has more consistently positive results. Programs that combine continuous and intermittent exercises make results difficult to interpret.9 The increasingly competitive nature of sports has resulted in a reluctance by athletes to take adequate breaks from training and performing.12 The damaging effects of prolonged endurance training on skeletal muscle and function are well documented in the literature, as is the huge capacity that human skeletal muscle has for repair and adaptation, given adequate recovery time.12 A study by Grobler and associates demonstrated that although minor exerciseinduced muscle damage is a precursor for adaptation, the reparative capacity of skeletal muscle is limited, and the cumulative effects of repetitive trauma and injury to skeletal muscle may lead to reduced performance, especially in long-distance runners.12 Further research is needed to investigate the limits of skeletal muscle regenerative capacity after chronic injury.12

Flexibility Extremes of joint and ligament laxity have important implications for the pediatric athlete owing to the increased risk for both acute traumatic and overuse-type sporting injuries, in addition to a number of degenerative orthopaedic

The Young Athlete

conditions, many of which have long-term implications for sports participation and performance.13 Childhood is associated with a gradual reduction in flexibility, with the greatest loss occurring around puberty as a result of a growth-induced muscle-tendon imbalance.14 This loss of flexibility is less pronounced in females.14 Excessive tightness during this time of rapid growth is thought to play a major role in both acute and overusetype injuries, affecting in particular the lower back, pelvis, and knee.13 Slight improvements in flexibility are observed following the pubertal growth spurt in both males and females until early adulthood, at which point it plateaus and then starts to decline once again.14 Although only 4% to 7% of the general population meet all criteria for generalized ligament laxity, evaluation of flexibility still remains an essential component of the clinical assessment of a young athlete because it enables identification of those individuals at increased risk in addition to providing invaluable information for injury prevention and rehabilitation programs.13,15 Studies performed by Marshall and colleagues in 1980 demonstrated that increased flexibility was associated with a greater risk for sports-related injuries, particularly in those requiring rapid change of direction or acceleration.13 Several instrumented tests are available to test the flexibility of individual joints, but simple screening tests, such as the modified Marshall test devised in 1978, are more commonly used as a routine part of the clinical assessment of the young athlete.13 By measuring thumb-to-forearm apposition, the modified-Marshall test can quickly identify extremes of flexibility that warrant further, more in-depth investigation and assessment relevant to a given sporting interest.13

Strength Training Traditionally, strength training was discouraged among children because of the perceived risk for growth disturbances and other injuries.16 Research during the past 20 years, however, has demonstrated that strength training can not only be a safe and effective component of any comprehensive fitness program but can also provide clear health benefits to children and adolescents.16,17 These benefits include improved athletic performance as a result of increased coordination, muscle strength, and power, in addition to enhancement of long-term health because of increased cardiorespiratory fitness, reduced risk for injury, and improved bone mineral density and blood lipid profile.16-18 Research shows that expertly tailored strength training programs in children and adolescents are associated with increased muscle strength and performance advantages in sports such as football and weightlifting.18 Increases in strength 50% to 65% above baseline have been reported in the prepubescent athlete over a 2- to 3-month training period.19 In the preadolescent child, however, this increased strength occurs in the absence of muscle hypertrophy, highlighting the role of neurogenic adaptation as the likely cause. Neurogenic adaptation refers to the recruitment of increased motor neurons that can fire with each muscle contraction.18 Moreover, the loss of benefits after the program is discontinued for 6 weeks provides

465

f­ urther evidence for this hypothesis.16 In contrast, strength training during and after puberty is further enhanced by the hormonally induced increase in muscle growth that occurs in both males and females.18 Although the risk for injury associated with strength training is real, research shows that it is no greater than in any other sport when adult supervision is available to ensure proper technique and when safety precautions are taken.16,18 Data obtained by the National Electronic Injury Surveillance System between 1991 and 1996 estimated that strength training was responsible for more than 20,000 injuries annually in those younger than 21 years.20 The usefulness of these results are limited, however, by the lack of distinction between competitive and recreational injuries or comment regarding the quality of the equipment being used or the presence of adult supervision.18 Of note, 40% to 70% of those injuries were attributable to muscle strains, primarily within the lumbar area.18 Case reports indicate that children and adolescents participating in strength training may be at risk for specific lumbar injuries, including herniated intervertebral disks, paraspinous muscle sprains, spondylolisthesis, and pars interarticularis stress fractures.16

Thermoregulation and Heat-Related Injuries Heat-related illnesses are preventable.21 Despite this, heat stroke remains the third most common cause of exerciserelated death among high school athletes in the United States, after head injuries and cardiac disorders.22 There are several physiologic characteristics unique to the children that contribute to the thermoregulatory disadvantage they face in extreme climatic conditions; these include increased surface area–to–body mass ratio, reduced sweating capacity, greater generation of metabolic heat per mass unit, and slower rate of heat acclimatization.21,23 A large surface area–to–body mass ratio is advantageous in mild to moderate climates because of the increased convective surface it provides.22 In hot, humid weather, however, this provides a larger area for heat influx, thereby raising the core temperature and increasing the risk for heatinduced illnesses.22 Conversely, in cold climates, enhanced metabolic heat production and cutaneous vasoconstriction are often insufficient to overcome the heat lost from their vast surface area, particularly in cold water.24 Sweat glands play a central role in the pediatric athlete’s ability to thermoregulate. By 3 years of age, the number of sweat glands a person shall possess is fixed.22 Despite having a greater density of sweat glands per skin area than adults, the sweating capacity in children is restricted because of a lower sweating rate and a higher sweating threshold.25 As a result, their ability to dissipate body heat by evaporation is reduced until the transition is made to an adult sweating pattern in late puberty.21,23 The reluctance of children to drink during prolonged exercise further exacerbates this thermoregulatory disadvantage.26 The American Academy of Pediatrics recommends prehydration in addition to enforced periodic drinking during the course of prolonged exercise.21 Although water is readily available, flavored drinks are often easier for children to tolerate.21 Moreover, because

466

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the risk for dehydration is even greater in children with certain diseases or conditions such as cystic fibrosis, diabetes, and anorexia, the need for optimal fluid intake during exercise is essential.21

PSYCHOSOCIAL ASPECTS OF SPORTS PARTICIPATION Psychosocial Development Participation in sports activity is associated with a large number of health benefits that can influence both physical and psychosocial well-being. The social interaction associated with sports participation is instrumental in a child’s psychosocial development, including character development, self-discipline, emotion control, cooperation, empathy, and leadership skills.25 The acquisition of new skills aids in building confidence and self esteem.25 It also allows children to experiment with success and failure in a lowrisk environment.27 The YRBS study mentioned earlier was a nationally representative study conducted throughout the 1990s by the Centers for Disease Control and Prevention. It evaluated the new trends in sports participation, with particular focus on health behaviors.4 The study identified a strong positive trend between sports participation and several types of positive health behaviors in both white males and females, including consumption of fruit and vegetables as part of a healthy diet, reduced levels of smoking and illegal drug use, and a reduced risk for suicide.4 This trend was not found among ethnic minorities, and in fact, among Hispanics and African Americans, the risk for negative health behaviors actually increased with sports participation.4

Readiness for Sport Knowledge of cognitive and motor developmental milestones and of the factors that motivate children and adolescents to participate in sports is essential when designing sports activities that are both rewarding and beneficial.25,27 Motor development is a sequential process like any other developmental milestone, and the rate of progression varies among children.28 Participation in most sports require fundamental motor skills, including kicking, throwing, running, jumping, and catching.28 Most children acquire these skills through informal play, but mastery often requires more formal instruction and repetition.28 Although this process of acquisition and mastery can potentially be accelerated through intensive instruction and practice, research shows that it rarely speeds up motor development or leads to enhanced athletic performance.28 The principal motivating factors for young children to participate in sports activities are fun and enjoyment.25 For an activity to be viewed as enjoyable, there must be a certain level of excitement but ultimately a sense of personal achievement associated with the improvement or mastery of specific skills.25 We must acknowledge that although virtually all children have the ability to acquire new motor skills, the ease of acquisition and degree of mastery may vary between children.27 Research has shown that children who feel less competent with one particular skill are less

likely to continue with that sport in the long term.27 Therefore, it is important that young children are exposed to a range of sports that challenge and enable them to acquire a variety of fundamental motor skills.27 Progression into adolescence is associated not only with a number of physical changes resulting from the pubertal growth spurt but also with a shift in the motivational factors influencing sports participation.25,27 Cognitive and motor development is now sufficient to allow for the incorporation of strategy into sports such as football or basketball.28 The need for fun and excitement is overtaken by social factors such as interaction with friends and physical appearance, although mastery of skills still remains important.25,28 Differing rates of progression through puberty can result in inequality within and between genders.28 Those who experience earlier growth spurts may be temporarily taller, heavier, and stronger, which often leads to unrealistic expectations because of the erroneous conclusion that they are destined to become better athletes than their less mature peers.28-30

Adult Involvement The level of adult involvement has increased significantly with the evolution of organized sports. Although the traditional role of supervisor still exists, the nature of adult involvement in youth sports has also evolved. An increased level of sophistication has developed because of the advent of specialized coaches, sports psychologists, nutritionists, and personal trainers, all of whom undoubtedly affect the psychosocial development of the young athlete. Adults are vital for the enforcement of rules and the creation of a safe, controlled environment in which to impart their knowledge and assist children and adolescents in the acquisition of new skills and development of appropriate attitudes to sports.27,28 Their involvement in sports activities can also have a detrimental impact on psychosocial development through the expression of negative and unsportsman-like behavior, negative reinforcement, and the enforcement of demands and expectations that exceed the child’s abilities.25,27 In the early years of life, parental influence is instrumental in the development of lifelong core values and attitudes.25 By 12 years of age, a child’s attitudes to winning are already well established and often directly reflect the values held by their parents.25 These values and attitudes are often acquired through observation of parental behavior, and although extreme parental behavior is rare, the use of negative comments or reinforcements is frequent.25 Variation was found between sporting codes, with the greatest incidence among soccer and rugby players.25 Children of relaxed and supportive parents who positively reinforce their child’s performance not only are more self-confident but also are more likely to be successful athletes.25,27 As the child progresses to adolescence, the role of parents starts to diminish as the role of the coach increases.25 Coaches, through their provision of feedback and reinforcement, have a large impact on the confidence and selfperception of the young athlete.25 The increasingly competitive nature of sports has lead to a shift in goals that are largely adult oriented and focus on winning at any cost.3 Competitive behaviors start

The Young Athlete

to emerge at 3 to 4 years of age, and the potential exists to either enhance or exploit this trait through the use of sports.25 The danger arises when the demands and expectations placed on young athletes by their parents or coaches exceed their abilities.25 This can result in the development of unhealthy competitive behavior with serious antisocial interpersonal consequences or even problems such as burnout and chronic stress.25

NUTRITION The nutritional concerns of the pediatric athlete are complex and unique from their adult counterparts because they involve the interaction between normal growth and development and the optimization of athletic performance.31,32 During the 1980s, there was an erroneous belief that leanness correlated with enhanced athletic performance as a result of studies that demonstrated a positive correlation between running performance and percentage of body fat.33 In fact, not only is there a lack of scientific evidence to prove that reducing weight alone will improve athletic performance but it is also true that deliberate caloric restriction in children and adolescents is likely to have detrimental implications for athletic performance and for growth and development and general health.33 Unfortunately, these erroneous beliefs are perpetuated today by coaches with little or usually no training in athlete nutrition.33 In the case of school-based coaches, their employment is often dependant on the success of their teams, and controlling an athlete’s weight is often the easiest parameter by which a coach can try and ensure athletic success.33 In fact, by reducing the dietary fat contribution, it is possible that essential sources of protein, as well as minerals and vitamins such as calcium, magnesium, iron, zinc, B12, and other fat-soluble vitamins critical for growth, may also be eliminated from the diet.32 Diet should play an integral role in any comprehensive training program, with specific attention to energy requirements, including appropriate combinations of protein, carbohydrates, fat, vitamins, and minerals.32 These requirements are often subject to large interindividual variation between sporting codes and often within a given sport.32 Results gained from the YRBS study in the 1990s confirmed that children and adolescents involved in regular sporting activities not only maintain healthier diets consisting of greater amounts of fruit and vegetables but also are often less concerned with caloric intake and energy balance.34 For young athletes, the energy requirements must be sufficient to ensure normal growth and development but must also provide the additional calories to account for physical training.32 The recommendations for estimated energy requirements in young athletes set by the Food and Nutrition Board are based on age, height, weight, and physical activity classification.32 Protein is an essential part of a young athlete’s diet because it is required to build amino acids necessary for the growth and development of lean body mass and healthy bones. It is also an alternative to carbohydrates as a ������� source of energy��.32 There is a lack of research regarding the recommended daily protein intake for young athletes.32 For adults, 12% to 15% of dietary energy should come from

467

protein; however, in children, the demands are greater, especially when involved in competitive, intensive training during periods of rapid growth.32 Research shows that children and adolescents up to the age of 13 to 15 years have restricted glycolytic capacity, which questions the role of high carbohydrate diets.32 Regardless, nutritionists recommend that at least half of a young athlete’s diet consist of carbohydrate owing to the importance of this energy source during high-intensity training.33 A significant amount of research is needed with regard to optimal nutrition of the pediatric athlete.

PERFORMANCE-ENHANCING SUBSTANCES The use of performance-enhancing substances among children and adolescents is increasing as a result of media exposure, the availability of so-called natural supplements, the absence of formal drug testing in schools, and the increasingly competitive nature of youth sports.35 Pediatric athletes are at high risk because of increased susceptibility to societal pressures at a time when they are often dealing with complex developmental and ­psychosocial changes. The term ergogenic is derived from the Greek meaning “to make work” and refers to the inherent ability of many substances to enhance athletic power and endurance.35 In many cases, the ergogenic effects of a substance are actually secondary to their intended use.35 It is therefore essential that physicians dealing with athletes, especially those competing in high-level sports, have a working knowledge of substances that contain ergogenic properties because inappropriate prescribing or counseling may result in an athlete’s disqualification from a competition.36

Anabolic-Androgenic Steroids Although a wide range of performance-enhancing substances are available in the United States, anabolic­androgenic steroids are by far the most publicized and intensely studied. Anabolic-androgenic steroids are a synthetic analogue of the male hormone testosterone, and their use in the pediatric athlete for both performance and physique enhancement has been documented in the medical literature for well over 20 years.35 The use of androgenic steroids is widespread, with an estimated 4% to 12% of male adolescents and 0.5% to 2% of female adolescents using anabolic-androgenic steroids in the 1990s despite being banned by almost every major athletic-governing body.36 As the name suggests, anabolic-androgenic steroids have both masculinizing and tissue-building effects, such that when used in conjunction with adequate strength training and proper diet, they have the ability to increase muscle size and strength, enabling high-intensity workouts and possibly even a reduced recovery time following workouts.35 As a result, strength athletes (e.g., lifters, throwers, and football players) and those participating in sports such as swimming and running that require frequent, highintensity workouts are attracted to the substance.35 Research conducted by Kindlundh and associates in 1999 demonstrated a significant correlation between the

468

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

use of anabolic-androgenic steroids in adolescents and the abuse of other common drugs, such as alcohol, tobacco, cannabis, and opioids.36 Although the perceived performance-enhancing ­benefits appear high, the side effects of using anabolicandrogenic steroids are extensive and often irreversible.35 In addition to personality changes and psychological problems that are associated with steroid use, premature closure of epiphyseal plates with subsequent linear growth arrest, irreversible alopecia, gynecomastia, acne, and irreversible ­ masculinization of secondary sexual characteristics in females are just a few of the more dramatic and often psychologically ­devastating side effects of anabolic­androgenic steroid use.35

Regulation of Performance-Enhancing Substances Drug testing is both time consuming and expensive, making the widespread testing of young athletes virtually impossible.26 Despite this, many schools and youth organizations have implemented voluntary drug testing, which has a dual benefit of identifying and providing assistance for athletes with abuse problems as well as reducing the peer pressure to use drugs.24 With the introduction of the Dietary Supplement Health and Education Act in 1994, the role of the U.S. Food and Drug Administration in regulating natural supplements was eliminated.12 Since this time, natural agents such as creatine, androstenedione, and DHEA have been widely accessible through health stores and the Internet.12 This accessibility results in an erroneous perception that these substances are safe, even though the absence of regulatory control eliminates any legal requirement of manufacturers to declare all active ingredients and potential interactions and to fully test their products for short- and long-term effects.24 The use of performance-enhancing drugs among athletes of any age is unethical, unhealthy, and potentially life-threatening.26 As physicians, we have a responsibility to acquire and impart factual knowledge to young athletes contemplating the use of these substances. Although the effectiveness of using scare tactics that emphasize the negative effects of substance use has been questioned, there is a clear role for positive counseling with regard to healthy alternatives such as strength training and conditioning, nutrition, and skill acquisition through coaching and camps.26

ARTHROSCOPY IN CHILDREN The use of arthroscopy in the pediatric and adolescent population has dramatically expanded over the past decade as a result of increased youth participation in sports and the subsequent rise in sports-related injuries.5 With the advent of smaller, more sophisticated arthroscopic instruments over the past decade, the major obstacle to its application in children was overcome.5 In fact, Gross noted that after extensive experience, despite the difference in joint size, basic techniques of arthroscopy are largely the same in both children and adults.37 At present, arthroscopy is indicated in the management of several shoulder,

Figure 9-1   Little League shoulder. Widening of the proximal humeral physis associated with repetitive overuse.

elbow, wrist, hip, knee, and ankle injuries in the pediatric ­ opulation.5 Advantages to arthroscopy in this population p include reduced postoperative morbidity, smaller incisions, more rapid return to activities, decreased inflammatory response, and improved visualization of joint structures.5 Shoulder injuries in the pediatric athlete include acute fractures, overuse injuries such as Little League shoulder (Fig. 9-1), and shoulder instability (Fig. 9-2). Most major shoulder injuries requiring arthroscopy are related to instability and can be divided into two descriptive groups: traumatic anterior instability and multidirectional instability (MDI).5,37-43 The incidence of elbow injuries continues to increase as a result of the growing popularity of youth sports. Many of the elbow injuries are repetitive, overuse-type injuries, such as osteochondritis dissecans (OCD), which is prevalent in baseball, racket sports, and gymnastics.44 In fact, Little League elbow is now an accepted term for a common overuse injury in young throwing athletes, with etiologies including fragmented medial epicondyle (Fig. 9-3), OCD (Figs. 9-4 and 9-5), ulnar hypertrophy, and medial epicondylitis.43-46 Wrist arthroscopy is not a commonly practiced treatment modality among pediatric and adolescent patients because many injuries achieve successful healing nonoperatively and also because of the restricted size of the joint space.46 Kocher and colleagues note an increasing incidence of repetitive use injuries, such as triangular fibrocartilage injuries (Fig. 9-6), and believe arthroscopy is indicated for débridement or determination of the extent of ligamentous injury in those patients failing nonoperative therapies.47,48

The Young Athlete

469

A

B Figure 9-2   Traumatic anterior shoulder instability. A, Bankart lesion. B, Repair of Bankart lesion.

Figure 9-3   Medial epicondyle widening associated with Little League elbow.

Figure 9-4   Sagittal magnetic resonance image of the elbow demonstrating chondral defect of the capitellum associated with osteochondritis dissecans.

Although hip arthroscopy is a commonly used diagnostic and treatment modality for hip pathologies in adults, its application in children and adolescents is only beginning to increase. Indications in the pediatric population include isolated labral tears (Fig. 9-7), loose bodies, chondral­ ­injuries, and internal derangement associated with Perthes’ disease and epiphyseal dysplasias.49-53 The risk for complications, although low, includes pudendal nerve irritation and recurrent injury.5 Currently, the largest application of arthroscopy in children and adolescents is in the treatment of knee pathology, which is directly attributable to increased athletic ­activity.37 Key indications for knee arthroscopy include OCD (Figs. 9-8 and 9-9), discoid meniscus, tibial spine fractures (Figs. 9-10 and 9-11), and partial and complete anterior cruciate ligament (ACL) tears.54-63 At present, the use of ankle arthroscopy in children is restricted to a small number of conditions, including OCD, loose body removal, and triplane fracture repair, owing to technical challenges resulting from the size of the joint and the risk for neurovascular damage.37,64-67

470

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 9-7   Radial labral tear of the hip. Figure 9-5   Lateral radiograph of the elbow demonstrating a loose body in the anterior elbow.

A

B

Figure 9-6   Ulnar styloid fracture (A) associated with a triangular fibrocartilage tear (B).

The Young Athlete

471

Figure 9-8   Osteochondritis dissecans of the knee. Anteroposterior radiograph (A) and corresponding coronal magnetic resonance image (B).

A

B

Figure 9-9   Fixation of unstable osteochondritis dissecans lesion of the knee. Immediate postoperative anteroposterior radiograph (A) and 3-month postoperative radiograph (B) demonstrating lesion healing.

A

B

472

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 9-10   Suture fixation of tibial spine fracture. A, Guidewires brought through the tibial spine fragment. B, Suture fixation.

A

B

Figure 9-11   Epiphyseal cannulated screw fixation of tibial spine fracture. A, Displaced fracture. B, Screw fixation.

A

CONCLUSIONS Pediatric sports injuries are being seen with increased frequency. Just as the child is not a “little adult,” the pediatric athlete is not a “little adult athlete.” An understanding of the unique considerations of the pediatric athlete with

B

respect to epidemiology, endurance, flexibility, strength, thermoregulation, psychology, and nutrition is important background knowledge. Recognition of common injury patterns of the shoulder, elbow, wrist, hip, knee, and ankle is essential to effective management.

The Young Athlete

C

r i t i c a l

P

o i n t s

l Sports injuries are being seen with increased frequency in the pediatric and adolescent athlete because of increased participation in higher competitive levels at younger ages, increased recognition of injuries in this age group, and the advent of arthroscopy and magnetic resonance imaging. l The pediatric athlete differs from the adult athlete in terms of physiology, growth, psychology, and skills. l Injury patterns are age and sport specific. l An understanding of the special considerations of the pediatric athlete and the common injury patterns is necessary for the successful management of sports injuries in these patients. l Excessive tightness during this time of rapid growth is thought to play a major role in both acute and overusetype injuries, affecting in particular the lower back, pelvis, and knee. l Strength training can be a safe and effective component of any comprehensive fitness program, but it can also provide clear health benefits to the pediatric age group.

S U G G E S T E D

473

R E A D I N G S

Anderson SJ, Griesemer BA, Johnson MD, et al: Climatic heat stress and the exercising child and adolescent. Pediatrics 106(1):158-159, 2000. Baquet G, Praagh EV, Berthoin S: Endurance training and aerobic fitness in young people. Sports Med 33(15):1127-1143, 2003. Elsass W, Wingler I: Psychological aspects of sports in children and adolescents. In DeLee J, Drez D (eds): Orthopaedic Sports Medicine: Principles & Practice, 1st ed. Philadelphia, WB Saunders, 1994:687-702. Faigenbaum AD: Strength training for children and adolescents. Clin Sports Med 19:593-619, 2000.

l There are several physiologic characteristics unique to children and adolescents that contribute to the thermoregulatory disadvantage they face in extreme climatic conditions, including increased surface area–to–body mass ratio, reduced sweating capacity, greater generation of metabolic heat per mass unit, and slower rate of heat acclimatization. l The social interaction associated with sports participation is instrumental in a child’s psychosocial development, including character development, self-discipline, emotion control, cooperation, empathy, and leadership skills. l The nutritional concerns of the pediatric athlete are complex and unique from their adult counterparts because they involve the interaction between normal growth and development and the optimization of athletic performance. l The use of arthroscopy in children and adolescents has dramatically expanded over the past decade as a result of increased youth participation in sports and the subsequent rise in sports-related injuries.

Hergenroeder AC: Prevention of sports injuries. Pediatrics 101(6):1057-1063, 1998. Sullivan J, Anderson S (eds): Care of the Young Athlete. Rosemont, Ill, American Academy of Pediatrics and American Academy of Orthopaedic Surgeons, 2000, pp 33-34.

R E F E R E N C E S Please see www.expertconsult.com

C H A P T E R  

10

The Female Athlete Letha Y. Griffin, April Armstrong, and Marlene DeMaio “Males and females have different patterns of illnesses and different life spans…. Understanding the bases of these sexbased differences is important to developing new approaches to prevention, diagnosis and treatment.”1 This statement is as true for sports medicine as it is for other areas of medicine.

Before the 1970s, few women participated in organized sports. However, the passage of Title IX of the Educational Assistance Act of 1972,2 which required institutions receiving federal money to offer equal opportunities to both males and females in all programs including ­athletics, sparked a rapid growth not only in collegiate women’s sport opportunities but also in sport opportunities ­ available to high school and recreational female athletes (Table 10-1). As the number of women athletes grew, so did the demand for sports equipment. Initially for many sports, women wore smaller sized men’s sports gear. Now, sport shoes, sport clothes, and protective equipment, including braces, have all been designed specifically for women (Fig. 10-1). Since 1991, women have out-purchased men in athletic shoes and apparel.3 Before 1970, rarely were the results of women’s sports contests found in newspapers or within the pages of Sports Illustrated or on nightly TV news or radio; however, this trend is beginning to change (Fig. 10-2). In the past several years, Coach Pat Summitt’s Tennessee, Andy Landers’ University of Georgia, and Geno Auriemma’s University of Connecticut women’s basketball teams have all been headliners in USA Today, Sports Illustrated, and other sports publications. Women’s golf and tennis events are now aired on prime-time television. The world knows the names of Michelle Wie and Anna Kournikova. The Women’s National Basketball Association (WNBA) aired 47 games in 2002, which were viewed by 60 million people.4 Parallel, and perhaps one could argue, secondary to the increased emphasis on women’s sport participation, there has been an improvement in women’s sport performance in swimming and running events (Table 10-2), and the speed of basketball, soccer, tennis, and volleyball games has increased with the improvement in women’s skills and overall ­athleticism. In 2003, Annika Sorenstam was a ­competitive

Figure 10-1   Women’s sport gear is a big business.

player in a Professional Golfers’ Association (PGA) event, and the world was enthralled by Mia Hamm and the members of the United States’ gold medal–­winning women’s soccer team’s performance in the 2004 Olympic Games. With the increased emphasis on women’s sport participation and the market value of women’s athleticism, greater funding for research in women’s sport issues has been achieved. Funded studies have improved our ­understanding of the nutritional needs of women; of proper conditioning and rehabilitation techniques for women; and of prevention and treatment of musculoskeletal sport injuries, such as patellofemoral issues, stress fractures, and anterior cruciate ligament injuries, in women.

   TABLE 10-1  Growth in Women’s Participation in Sports Participation Level

1971-1972 (No. of ­Women Participating)

2000-2001 (No. of Women Participating)

High school College Olympic

294,015 29,972 1264 (15.5% of all athletes)

2,784,154 150,916 4935 (37.8% of all athletes)

Data from http://www.nfhs.org, http://www.ncaa.org, and http://www.olympic. org.

Figure 10-2   Women’s sports are newsworthy.

475

476

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Anatomic and Physiologic Parameters

   TABLE 10-2  Improvements in Women’s Sport ­Performance during the Past 30 Years

Improvement in Women’s Sport Performance Olympic Events

1972

1996

2004

59.44 4:19.89 8:59.69

54.50 4:07.25 8:27.89

53.52 4:05.34 8:24.54

800 m run 1500 m run

2:03 4:26.5

1:57.04 4:00.83

1:56.38 3:57.90

Collegiate Events

Sanctioned 1982

1996

2005

23.16 49.37 4:41.61 16:02.34

22.59 49.04 4:42.46 16:06.23

21.97 47.50 4:37.11 15:46.84

800 m 1500 m 10,000 m

2:05.22 4:17.90 33:36.51

2:03.27 4:17.92 32:56.63

2:02.84 4:11.37 33:02.21

High School

1972

1998

2005

800 m 3200 m

2:04.7 10:51.0

2:06.30 10.25.99

2.03.73 10.12.16

Boston Marathon

1972

1996

2006

3:10:26

2:27:12

2:23:38

Swimming

100 m free 400 m free 800 m free Track

Swimming

50 yd free 100 yd free 500 yd free 1650 yd free Track

Track

Moreover, since the National Institutes of Health (NIH) Revitalization Act of 1993,5 directing the NIH to establish guidelines for inclusion of women and minorities in clinical research, NIH-funded research studies have provided analyses by sex and gender—an approach that has led to greater understanding of the results of sexual dimorphism on disease and health.

CONCEPT OF SEX AND GENDER DIFFERENCES AND THEIR INFLUENCE ON HEALTH, DISEASE, AND SPORT PERFORMANCE Sex is biologic; it has been defined in the Institute of Medicine’s report on Exploring the Biologic Contributions to Human Health as, “The classification of living things, ­generally as male or female according to their reproductive organs and functions assigned by chromosomal component,” whereas gender is cultural and social based and is, “A person’s selfrepresentation as male or female or how that person is responded to by social institutions based on the individual’s gender presentation.”1 During the past several decades, scientists have realized that sexual dimorphism extends beyond the reproductive system, permeating each of the biologic systems, including the musculoskeletal system.6

There are distinct anatomic and physiologic differences between men and women (Table 10-3). These differences may affect the female athlete’s overall sport performance or preselect her for certain sports. Moreover, based on these differences, equal but parallel sport opportunities rather than mixed sex competition appears to be appropriate for many areas of sport. For example, it has been consistently reported that women have a greater percentage of body fat than men. ­Skinfold measurements in female athletes are typically thicker than those in their male counterparts.7-9 Because of their greater percentage of body fat, women are more buoyant and better insulated than men. This may give them an advantage in endurance swimming events. Most of the additional body fat is distributed in their hips and lower body, whereas most men carry additional subcutaneous fat in their abdomen and upper body. In addition to increased body fat in the lower body, the female has also a wider pelvis relative to thorax and shorter leg limb lengths, which lower her center of gravity. This theoretically provides her with better balance, thus increasing a woman’s ability in balance sports. Studies of the muscles of the upper and lower extremities have demonstrated smaller cross-sectional areas in women relative to men.10-12 The cross-sectional area changes become more apparent at 13 to 15 years of age.11 When cross-sectional area is expressed as a component of limb length, the gender differences are not as evident. In addition, little difference is found between males and females when comparing strength per cross-sectional area, but the absolute strength measurements are greater in males.12 It has been shown that female strength-training athletes can generate up to 82 kg of fat-free body mass, whereas the largest male strength-training athlete has been reported to generate 121 kg.10 Women generally have a lower oxygen carrying capacity than men, most likely secondary to a smaller stroke volume and a smaller heart size. To compensate for this, a female athlete must increase her heart rate for equivalent work of a male. Her lower hemoglobin levels compound the gender oxygen carrying capacity difference. Kang and colleagues used incremental cycling and treadmill exercise protocols to characterize the gender-related differences in the Vo2 to work with matched physical fitness levels.13 During cycling, they found that the rate at which absolute Vo2 increased was no different between men and women; however the absolute values were greater in men because they were heavier. When Vo2 was normalized to the subjects’ body mass, the slope of increase was greater in women, suggesting that women are less efficient and consume more oxygen per unit of ����������������������������� �������������������������� body mass than men. These researchers suggested that the greater increase in massspecific Vo2 may be attributable to the fact that female athletes are, in general, lighter than males. During treadmill exercise, no gender differences were observed when looking at relative Vo2 and heart rate, but a greater increase in absolute Vo2 was seen in men. It was hypothesized that the greater body weight of men during load-bearing exercise increased work rate. The results of this study would support that fitness tests measuring submaximal heart rate

The Female Athlete

477

   TABLE 10-3  Anatomic and Physiologic Gender Differences Parameter

Postpubertal Girls

Postpubertal Boys

Impact

Oxygen pulse (efficiency of cardiorespiratory system) Vo2 max (reflects level of aerobic fitness) Metabolism (basal metabolic rate) Thermoregulation

Lower

Higher

Lower

Higher

Higher oxygen pulse provides boys an advantage in aerobic activity. Boys have greater aerobic capacity.

6%-10% lower (when related to body surface area) Equals boys

6%-10% higher (when related to body surface area) Equals girls

Endocrine system   Testosterone   Estrogen

Girls need fewer calories to sustain same activity level as boys. Equal ability to adequately sweat in a hot environment to decrease core body temperature.

Lower Higher

Higher Lower

Height Weight Limb length

64.5 in 56.8 kg

68.5 in 70.0 kg Longer

Articular surface

Smaller

Larger

Body shape

Narrower shoulders Wider hips Legs 51.2% of height More fat in lower body

Wider shoulders Narrower hips Legs 52% of height More fat in upper body

Percentage of muscle/total-body weight* Percentage of fat/total-body weight

~36%

~����� 44.8%

~������� 22%-26%

~���������� 13% to 16%

Age at skeletal maturation

17-19 yr

21-22 yr

Cardiovascular system   Heart size   Heart volume   Systolic blood pressure

Boys have increased muscle size, strength, and aggressiveness. Unknown whether related to increase in ligamentous laxity or in rate of anterior cruciate ligament injuries Increased height and weight in boys give them structural advantages. Boys can achieve a greater force for hitting and kicking. May provide boys with greater joint stability; boys have greater surface area to dissipate impact force. Girls have lower center of gravity and therefore greater balance ability; girls have increased valgus angle at the knee that increases knee injuries; boys and girls have different running gaits. Boys have greater strength and greater speed. Girls are more buoyant and ­better insulated; they may be able to ­convert fatty acid for metabolism more rapidly. Girls develop adult body shape/form sooner than boys.

Smaller Smaller Lower

Larger Larger Higher

  Hemoglobin Pulmonary system   Chest size   Lung size   Vital capacity   Residual volume

10%-15% higher per 100 mL blood Smaller Smaller Smaller Smaller

Larger Larger Larger Larger

Stroke volume in girls is less, ­ necessitating an increased heart rate for a given submaximal cardiac output; cardiac output in girls is 30% less than in boys; the risk for ­hypertension may be less in girls. The oxygen carrying capacity of blood is greater in boys. Total lung capacity in boys is greater than in girls.

*There are no appreciable differences in these parameters before puberty; therefore, prepubertal boys and girls can compete on a fairly equal basis. From Yurko-Griffin LY, Harris S: Female athlete. In Sullivan JA, Anderson SJ (eds): Care of the Young Athlete. Rosemont, Ill, American Academy of Orthopaedic ­Surgeons, 2000, pp 138-148, with permission.

during incremental exercise should be gender specific. In later stages of life, Vo2max decreases more rapidly in men than in women, resulting in less gender-related differences in the later decades of life.14 It has also been suggested that women have a greater impairment in pulmonary gas exchange during exercise compared with men. However, Olfert and associates found no difference in gas exchange in women compared with men when subjects were matched for age, height, aerobic capacity, and lung size.15 They reported that fitness level and lung size played a more important role than gender.

Ozkaplan and colleagues also reported the same ­ pattern of recovery for both genders from exercise-induced ­inspiratory muscle fatigue.16

Nutrition One of the primary nutritional concerns for the female athlete is inadequate dietary intake resulting in inadequate energy for sport as well as deficiencies in iron, calcium, and other nutritional needs. Athletes participating in more esthetic sports (e.g., ballet, gymnastics, skating,

478

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

long-­distance running) appear to be most at risk.17-19 In one study of female dancers, dietary intake was less than 70% of the recommended daily intake.20 The frequency of disordered eating in female athletes has been reported to be 16% to 72%.18,21,22 It is not uncommon for female athletes to either fast, skip meals, or consume low-fat or low-caloric meals. Some engage in more extreme measures to control weight such as self-induced vomiting, diet pills, laxatives, or use of diuretics. Disordered eating has been associated with reproductive dysfunction and osteoporosis.18,23-31 Wade and colleagues described the “metabolic fuel hypothesis.” The athlete is challenged to balance energy intake with energy expenditure.32 If an athlete remains in a negative fuel balance, the body will sustain essential physiologic functions such as thermoregulation and locomotion, compromising less critical functions such as reproductive function, adipose tissue deposition, and growth. Amenorrhea is the absence of normal menstrual periods for 3 or more consecutive months. Athletic amenorrhea, the low-estrogen and amenorrhea state associated with disordered eating, has traditionally been thought to result in decreased bone mineral density secondary to the loss of estrogen’s protective effect on bone. Agostini and coworkers in 1993 used the term female athlete triad (Fig. 10-3) to refer to the association of disordered eating, amenorrhea, and osteoporosis.33 Recently it has been suggested that a negative energy balance may be the key factor leading to bone loss.24 Cobb and colleagues demonstrated that disordered eating was associated with low bone mineral density in athletes with and without menstrual irregularities.24 Although osteoporosis was the original term used in the triad, Khan and associates recently suggested that osteopenia replace osteoporosis as the third component because it is a far more common entity.34 Osteopenia is defined by the World Health Organization as bone mineral density (BMD) that is 1.0 to 2.5 standard deviations below the young adult reference mean as measured by dual-energy x-ray absorptiometry (DEXA), and osteoporosis is a BMD that is more than 2.5 standard deviations below the young adult average value. Treatment of the female triad starts with prevention through education. Athletes should receive counseling on the benefits of “fueling” one’s body for maximal sport performance. Athletes, in general, respond better to such

   TABLE 10-4  Examples of Calcium-Rich Foods Food

Calcium Content

Sardines (with bones) Macaroni and cheese Whole milk Yogurt Swiss cheese Calcium-fortified orange juice Broccoli Broccoli Cottage cheese Ice cream Dark green leafy vegetables Waffle American cheese Salmon (canned with bones) Tofu Oysters (raw) Shrimp (canned) Beans (dried, cooked) Eggs Orange Bread

370 mg/cup 360 mg/cup 300 mg/cup 270-350 mg/cup 270 mg/cup 200-250 mg/cup 200 g/cup 150 mg/large stalk 200 mg/cup 200 mg/cup 200 mg/cup 180 mg/waffle 170 mg/cup 170 mg/cup 150 mg/4 oz 110 mg/7-9 oz 100 mg/3 oz 90 mg/cup 50 mg/per 2 eggs 50 mg/medium-sized orange 25 mg/slice

a positive approach—that is, enhancing performance through proper nutrition rather than stressing the need to eat properly to avoid osteoporosis and stress fractures. To minimize chances of developing osteopenia or osteoporosis, sport medicine practitioners should remind female athletes of all ages of the value of calcium in their diet, particularly during the teens and early 20s. Women need to be banking adequate stores of calcium in their bones to avoid development of osteoporosis as they age. Table 10-4 lists calcium-rich foods and Table 10-5 provides recommendations for women regarding daily calcium needs. The role of estrogen replacement (oral contraceptive pills) to minimize or avoid osteopenia developing in hypoestrogenic amenorrheic athletes is unclear.35 Some studies report an increase in bone mass following their use,36 whereas others do not.37 Iron deficiency occurs typically in endurance athletes but is also present in other sports in which diet restriction is prevalent38-42 and can have a negative impact on sport performance, cognition, immune system, and gastrointestinal functions.43 Iron is important for the oxygen carrying capacity of hemoglobin-myoglobin and for important    TABLE 10-5  Daily Adequate Intake Recommendations for Dietary Calcium Age Group (yr)

DISORDERED EATING

AMENORRHEA

Figure 10-3   Female athlete triad.

OSTEOPOROSIS

1-3 4-8 9-18 19-50 51-70 >70 Amenorrheic athletes (all ages) Pregnant and lactating women

Suggested Intake (mg/day) 500 800 1300 1000 1200 1200 1500 1500

Based on recommendations of National Osteoporosis Foundation (http://www. NOS.org).

The Female Athlete

functions of many enzymes. Iron deficiency may limit the athlete’s endurance performance. True anemia is not as common in the female athlete as is nonanemic iron deficiency. Daily losses and loss through menstruation are considered the primary two causes of iron deficiency. Iron supplementation in anemic athletes can increase aerobic power and improve performance44,45; however, it is unclear whether iron supplementation in athletes who are iron deficient and not anemic has the same benefits. Such supplementation does have the theoretical benefit of increased oxygen support for contracting muscle. Athletes with iron deficiency may complain of impaired training and performance and difficulty recovering after exercise.46 Box 10-1 lists common foods that are iron rich. Vegetarian athletes are at particular risk for developing iron deficiency because they have no red meat in their diet and should be counseled to either include iron-rich foods in their diet that do not come from meat sources or add to their diet a vitaminmineral supplement that contains the daily requirement for iron.

Physical Examination The basic components of the preparticipation history and physical examination are the same for males and females. In women, the medical history should include not only a history of prior medical problems and surgical procedures but also a menstrual history and dietary history. The date of first menses should be documented, as well as the date of the last period, the average length of periods, the presence of dysmenorrhea, and the pregnancy history. Athletes typically have a delay in onset of menses; in nonathletes, menses begins at an average of 12.5 years, and in athletes, at 13.5 to 15.5 years. If an athlete has not had a period by the age of 16 years, she is considered to have primary amenorrhea and should be referred to her primary care physician for further evaluation. Secondary amenorrhea is a cessation of menses for 3 months. The normal menstrual cycle lasts on average 28 days with a range of 25 to 35 days. Cessation of menses or infrequent periods may be an early Box 10-1  Iron-Rich Foods Liver Lean red meats, including beef, pork, lamb Seafood, such as oysters, clams, tuna, salmon, shrimp Beans, including kidney, lima, navy, black, and pinto beans; soy beans; tofu; ������������������� and ������������� lentils Iron-fortified whole grains, including cereals, breads, rice, and pasta Greens, including collard greens, kale, mustard greens, spinach, and turnip greens Vegetables, including broccoli, Swiss chard, asparagus, parsley, watercress, Brussels sprouts Chicken and turkey Blackstrap molasses Nuts Egg yolks Dried fruits, such as raisins, prunes, dates, and apricots Curry powder, paprika, thyme

479

sign of a ­significant nutritional health problem and requires immediate attention. Athletes with secondary amenorrhea are at risk for developing the female triad described earlier. Dysmenorrhea (painful periods) is less common, but athletes with persistent dysmenorrhea or increasingly disabling pain should be referred for gynecologic assessment to rule out etiologies such as pelvic infection, ovarian cyst, or endometriosis. Breast discomfort may be reported by athletes. Proper support and, in some sports, breast padding may be recommended. The breast may be subjected to blunt trauma leading to contusions, hematomas, or abrasions. The nipple is the most prone to injury, and bloody discharge should be referred for further assessment because in some cases this could be a sign of intraductal neoplasm.47 Application of bandages or petroleum jelly to the nipple before sport activity may help to decrease nipple irritation. Idiopathic scoliosis is more prevalent in young girls than boys. Normally, idiopathic scoliosis does not cause pain. Any back pain in association with scoliosis requires further investigation because this could be associated with a syrinx, disk herniation, tethered cord, tumor, or spondylosis.47 The preparticipation history and physical examination should serve as a time to identify athletes at risk for certain musculoskeletal difficulties, such as patella and shoulder laxity issues, forefoot abnormalities, and anterior cruciate ligament (ACL) injury risk from posturing of the lower extremities. Counseling regarding exercise to minimize risk or shoe adaptation for forefoot abnormalities can be provided at this time.

CONDITIONING For female athletes, just like their male counterparts, proper conditioning is an important component in decreasing injury rates and improving performance.48-56 Before 1964, few advocated resistance training for females. Rumored myth was that women could not increase their strength to the extent that men could, even if they participated in a well-structured strengthening program. However, the success of the Eastern European women who engaged in weight-training programs in preparation for the 1964 Olympics drew attention to the potential benefits of such training for women. In the early 1970s, Wilmore and associates investigated alterations in strength and body composition in 47 women and 26 men after a 10-week weight-training program and found that both groups had similar gains in weight, but muscle hypertrophy was greater in the men than in the women.57 Moreover, results of a study comparing changes in skeletal muscle in seven men and eight women after 4 months of heavy resistance training performed 3 times a week demonstrated no difference between the two groups in alteration of body weight or fat-free weight.50 Muscle hypertrophy occurred in both men and women in the upper but not the lower extremities. The authors concluded that muscle hypertrophy, as a consequence of weight-training, was similar in men and women. Such results led coaches to structure strength and conditioning programs for women in a similar fashion to that for men. For many sports in organized athletics, offseason ­ activity typically includes aerobic muscular and

480

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

c­ ardiovascular conditioning activities such as biking, running, swimming, or circuit weight-training done at 70% to 80% of the maximal heart rate for 20 minutes 3 to 4 times a week.58,59 Also during the off season, weight-­training with free weights or machines is recommended. In the ­immediate preseason, aerobic conditioning and weight-training are still emphasized, but refining of sport-specific skills and sportspecific muscle training occupies the ­majority of time. Studies have also shown that older women (i.e., postmenopausal women) can also increase muscle strength and endurance through training, although considerable variability exists in the amount of improvement obtained (see “Menopause and Exercise”).60-62 In the past 5 to 10 years, following research efforts on prevention of ACL injuries, interest in the concept of neuromuscular control and proprioception conditioning, as part of women’s conditioning programs, has occurred. Strong flexible muscles are of little help if they do not react at the appropriate time. Women have been found to move differently from men. They jump, land, and pivot in a more upright position.63-66 Also, they tend to land with femoral internal rotation, apparent knee valgus, and foot pronation,67-70 a position often associated not only with ACL injuries but also with patellar pain, patellar subluxation, and patellar dislocation. Drills that teach women to land, jump, and cut in a position of hip and knee flexion, with the body centered over the lower extremities are now advocated and have been incorporated into alternative warm-up programs for women in basketball, soccer, and team handball.71-73 In the early days of women’s sport, stretching was not emphasized because women were thought to have excellent flexibility. However, current thought is that women’s and men’s stretching programs should be similar.74 Conditioning is associated with multiple benefits (Box 10-2), including decreasing the risk for cardiac disease, ­diabetes mellitus, asthma, and depression. Even in our high school and collegiate women, the benefits of exercise are many. For example, those who exercise are less likely to smoke, drink alcohol, use drugs, and have unwanted pregnancies. They remain in school longer and have a higher graduation rate than those who do not ­participate in sport.75 Box 10-2  Benefits for Women of Exercise   and Athletic Participation

• Athletic

participation improves self-confidence, selfe­ steem, and self-image • Athletic participation increases leadership skills and team-building skills • Exercise increases bone mineral density, cardiovascular fitness, and muscle tone • Exercise decreases incidence of asthma, diabetes, cancer, cardiovascular disease, depression and hypertension • Young women who participate in sport are more likely to graduate, learn value of goal setting, be able to deal effectively with success and failure • Young women who participate in sport are less likely to smoke, drink alcohol, use drugs, become pregnant

Box 10-3  Conditioning Tips for Recreational Athletes Emphasize core strength to minimize stress on lower ex­tremities. Emphasize strengthening of scapular stabilizers and mus­ cles involved in dynamic stabilization of the glenohumeral joint to minimize laxity issues of this joint. Emphasize vastus medialis oblique (VMO) strength when doing lower extremity strengthening exercises to improve patellar tracking. Minimize loading the patellofemoral joint in a fully flexed knee position, that is, consider short arc extension and leg press exercises in place of squats, lunges, and full arc extension exercises. Perform upper extremity strengthening exercises at shoulder height and below to minimize stress on the rotator cuff (e.g., pull-downs, overhead dumbbell press).

The novice and more mature athletes should use caution in structuring their conditioning programs. For example, it is safer in this age group or activity level to do shoulder strengthening exercises under shoulder height to help prevent irritation to the rotator cuff. Similarly, these athletes should do quadriceps strengthening exercises but take care not to do so at the expense of creating high forces across the patellofemoral joint, that is, do short arc instead of full arc extensions on a leg extension machine, and avoid lunges and squats (Box 10-3). Because overuse injuries of the knee and feet are more prevalent in women, women athletes should emphasize core strength in their weight-training programs to minimize stress on their lower extremities and should be encouraged to wear supportive shoes with good arches.

MENOPAUSE AND EXERCISE The decrease in ovarian hormones ends menstruation and defines menopause. The fall in estrogen level contributes to the development of osteoporosis and affects overall health. Exercise, a modifiable factor, affects them both. The risk of all-cause and cardiovascular mortality is greatest for sedentary adults.76 Vigorous activity declines as women age.77 Hu found that increased body mass index and decreased activity were robust independent predictors of death in women.78 Activity in postmenopausal women is associated with improved fitness, decreased percentage of body fat, and improved body composition.79-88 The proven benefits of regular low- and moderate-intensity aerobic exercise for women are prevention of obesity and cardiovascular disease,76,89,90 maintenance of bone density, and decreased mortality. The benefits of exercise during and after menopause pertain to bone density, breast cancer, body weight, the cardiovascular system, diabetes mellitus, and mental health.78,90-107 Suggested exercises for postmenopausal women are in Box 10-4. Because cardiovascular physiology is different for men and women and because cardiovascular risk increases with age, recent studies have evaluated only women.78,97,101

The Female Athlete

Box 10-4  Exercises for Postmenopausal Women* Daily: Low- to Moderate-Intensity Activity†‡ Dancing Gardening Housework Prescribed home exercise Swimming Walking Weight training§ Yard work§ Exercise 3 to 6 Times a Week Dynamic exercise of large muscles: 30-60 min Exercise 2 Days a Week: Moderate to High-Intensity Activity Resistance training: 8-10 exercise sets, 10-15 repetitions each *A

preparticipation examination is recommended for all women older than 50 years. †Low intensity is 40%-60% of maximal capacity. ‡Moderate intensity is 60%-75% of maximal capacity. §Yard work and weight training were independent and robust predictors for high bone density.

­ everal prospective studies with very large samples of S postmenopausal women showed that walking and vigorous exercise are associated with significant dose-dependent decreases in cardiovascular risk and events. 97,101 Other benefits included decreased carotid artery stiffness108-110 and decreased blood pressure in hypertensive111 and normotensive112 subjects.

Bone Density In the United States, there are 1.5 million osteoporotic fractures per year, with direct costs of about $18 billion.113 About 30% of postmenopausal women have osteoporosis.114 Because of the morbidity and mortality associated with osteoporotic fractures, promotion of bone mass and prevention of bone loss are critical.115 Bone loss is due to increased turnover, especially resorption. Perimenopausal bone loss is greatest in trabecular bone (wrist and vertebrae) and may be prevented by estrogen replacement, which has documented side effects, including an increased risk for breast cancer.116,117 Exercise increases bone density and prevents bone loss, primarily through its effect on bone density and secondarily by muscle strengthening82,118-125 and prevention of falls.126 Impact loading (weight-bearing exercise), like estrogen, directly stimulates the osteoblast. Walking is simple, requires no special equipment, and has high compliance.93 Progressive resistance loading (strength-­training and isometrics) increases muscle mass, which increases bone density. Water exercise also increases bone density,127 but not as much as weight-bearing exercise.128 A prospective randomized study found less bone loss in the Tai Chi Chun cohort.129 Reviews of trials have shown that exercise increases bone density with different exercises having varying effectiveness on specific bones.93,102,106,130,131 These

481

exercise effects on bone density may not hold true in overweight or obese postmenopausal women.132 For exercise to be an effective modifier of bone mineral density, diet, calcium,102,133-135 and vitamin D supplements should be optimized.90,121 Higher levels of physical activity have been associated with improved balance136 and less hip fracture due to falls.126 Decreased fall risk may be accomplished by agility training.137-144

Breast Cancer Physically active women have a decreased risk for breast cancer in most recent cohort studies94,99,100,103,104 and in a prospective twin study.145 The reduction in breast cancer risk is estimated to be 30% to 40% for women who are physically active throughout their lifetimes and is more strongly supported in studies of postmenopausal women.146 The proposed mechanism is the reduction of fat, leading to less peripheral aromatization of estrogen from androgen. The speculated association between higher bone density and breast cancer was thought to be a consequence of the exposure to estrogen that resulted in the higher bone density. Physical activity has been negatively associated with estrogen levels147 but not with urinary excretion.148 Postmenopausal women with high body mass index and low physical activity had the highest estrogen.147 Regular exercise resulted in significantly decreased risk in very large, multicenter prospective studies.103,104 Postmenopausal women who walked briskly for 1.25 to 2.5 hours/week had an 18% decrease in the risk for breast cancer.103 However, leisure physical activity was not associated with breast cancer incidence in 37,105 women.149 Replacement estrogen at the time of enrollment has been associated with increased risk compared with women not on replacement therapy.104 Of 27 cohort and case control studies, 16 were associated with a significant decrease in risk for breast cancer in women with increased activity; in 6 studies, there was a trend, and in 5, there was no effect.146

PREGNANCY AND EXERCISE The goals of exercise during pregnancy are to promote the health of the mother and the fetus, to help the body adapt to pregnancy, and to increase the ease of delivery. Numerous studies have evaluated the effects of exercise during pregnancy on the mother and her fetus. A meta-analysis (11 trials, 472 women), combined with direct contact with the researchers, revealed that regular aerobic exercise in healthy pregnant women appeared to improve or maintain physical fitness.150 Sedentary primigravida women significantly improved their fitness level with moderate exercise 3 times per week for 15 weeks during their pregnancy without any ill effects to the mother or neonate.151 High-volume and medium-volume intense exercise in elite competitive athletes maintained the level of fitness without risk to mother and fetus and was thought to promote rapid return to previous level of fitness in 41 women.152 Those exercising reported a high level of well-being and less frequency of somatic symptoms, anxiety, and insomnia.153 Above­average level of activity during the second and third trimesters in healthy women was associated with greater mood

482

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

s­ tability.154 Women who exercise during early and late pregnancy are also likely to avoid excessive weight gain. There is no association of moderate exercise with abnormal development, premature labor or delivery, fetal compromise, or abnormalities at birth.155,156 Rates of conception, spontaneous abortion, congenital abnormalities, premature labor, and premature rupture of membranes are not greater in well-conditioned women who regularly exercise during pregnancy. Regular exercise during pregnancy may be associated with a decreased incidence of preeclampsia,157 shorter labor, less incidence of cesarean birth, and less fetal compromise during labor,155 but the data are conflicting.156 Recently, the American College of Obstetrics and Gynecology published updated guidelines for exercise during pregnancy (Box 10-5).155 Many pregnant women do not meet basic physical activity recommendations. Two reports of cross-sectional telephone surveys of 1979 pregnant women and 44,657 nonpregnant women (18 to 44 years of age) described the activity level of those surveyed.158 Nonpregnant women were most likely to meet activity recommendations.158 Walking was the most common form of exercise for both groups.158 Other forms of activity were swimming laps, weight-lifting, gardening, and aerobics. Many types of activity may be prescribed during pregnancy (Box 10-6). The gravid uterus should be protected Box 10-5  Guidelines for Exercise During ­Pregnancy

• For all women of childbearing age, an accumulation of

30 minutes or more of moderate exercise a day should occur on most, if not all, days of the week. In the ­absence of either medical or obstetric complications, pregnant women should do so as well. • No exercise should be done in the supine position ­after the first trimester. Pregnant women should avoid ­prolonged periods of motionless standing. • Be aware of the changes and needs of pregnancy. Stop with fatigue; do not exercise to exhaustion. ­Modifications in exercise routine from weight-bearing (e.g., jogging) to non–weight-bearing (e.g., stationary cycling and swimming) activities in later pregnancy may help with continuing exercise on a regular basis. • When good balance is necessary or ­abdominal trauma is a possibility, exercise should be avoided. • Pregnancy requires intake of an extra 300 kcal/day. ­Additional caloric requirements will be based on ­exercise duration and intensity. • Exercise in pregnancy requires appropriate hydration, clothing, and environment to dissipate heat. • Resuming exercise in the postpartum period is determined on a case-by-case basis and by physical limitation. From Bathgate SL, Larsen JW, Chahine EB, Macri CJ: Exercise in pregnancy and beyond: Benefits for long-term health. In Garrick JG (ed): Orthopaedic Knowledge Update Sports Medicine 3. Rosemont, Ill, American Academy of Orthopaedic Surgeons, 2004, 379-387. Adapted from American College of Obstetricians and Gynecologist: ACOG Committee Opinion No. 267. Obstet Gynecol 99:171-173, 2002.)

Box 10-6  Suggested Exercises during Pregnancy Overall Fitness Impact activities: walking������������������������� ,�������������������� *������������������� jogging, running Low-impact activities: bicycle, elliptical trainer, crosscountry ski machine Aquatic: swimming����������������������� ,������������������ *����������������� water aerobics Stretching: upper and lower extremities and core Resistive: aerobic weight-training, light weights in the ­upright position† Exercise to Adapt to Pregnancy and Assist   with Delivery Core exercises (back and abdomen) Low back and hamstring stretching Pelvic floor (Kegel’s), buttock, and abdominal strength­ening *The

safest exercises during pregnancy.287 is scant published literature on the effect of weight training and pregnancy. However, it has not been associated with adverse outcome anecdotally287 or as part of a fitness program.152

†There

from trauma; collision sports are to be avoided. Because of the risk for decompression sickness, scuba diving is not advised. Adequate hydration, avoidance of saunas, and ­limiting exercise in very high temperatures and high humidity are required to avoid hemoconcentration. Box 10-7 reviews the relative and absolute contraindications to aerobic exercise during pregnancy.159 Exercise

Box 10-7  Relative and Absolute ­Contraindications to Aerobic ­Exercise during Pregnancy Relative Contraindications Severe anemia Unevaluated maternal cardiac arrhythmia Chronic bronchitis Poorly controlled type 1 diabetes Extreme morbid obesity Extreme underweight (body mass index < 12) History of extremely sedentary lifestyle Intrauterine growth restriction in current pregnancy Poorly controlled hypertension Orthopaedic limitations Poorly controlled seizure disorder Poorly controlled hyperthyroidism Heavy smoker Absolute Contraindications Hemodynamically significant heart disease Restrictive lung disease Incompetent cervix, cerclage Multiple gestation at risk for premature labor Persistent second- or third-trimester bleeding Placenta previa after 26 weeks of gestation Premature labor during the current pregnancy Ruptured membranes Preeclampsia, pregnancy-induced hypertension From American College of Obstetricians and Gynecologists: ACOG Committee Opinion No. 267. Obstet Gynecol 99:171-173, 2002.)

The Female Athlete

should be prescribed for the healthy pregnant woman after an evaluation by an obstetrics health provider before starting a regular program.

Box 10-8  Treatment of Stress Fractures   in Women

MUSCULOSKETAL INJURIES

• Treat the fracture • Address causative factors: estrogen deficits, menstrual

irregularities, nutritional concerns, shoe wear, training techniques, others

Stress Fractures Studies in the 1970s and 1980s, most of which were done on the military population, reported that women have a higher incidence of stress fractures than men.56,160-163 In contrast, recent studies in the athletic populations have shown a modest increased risk or no increased risk for stress fractures in women compared with men.164-168 Hame and colleagues in 2004, following their survey of 5900 Division I collegiate athletes, found that although the incidence of all fractures was similar for men and women (0.0438 and 0.0461, respectively), the incidence of stress fractures was double for women, and stress fractures were most common in track and field athletes (Fig. 10-4).31,169 The reason for this variability in study results is not clear. Most agree that women athletes with menstrual irregularities have a higher incidence of stress fractures than women athletes with regular menstrual cycles170-174; although two studies, albeit with small sample sizes, have reported no correlation between menstrual history and stress fractures.175,176 The reason for the relationship between abnormal menses and the increased incidence of stress fractures in athletic women is not clear, nor is the literature on the relationship between low bone mineral density in ­ amenorrheic athletes and stress fractures consistent. Although one study in athletes with similar training schedules reported that ­athletes with stress fractures were more likely to have lower ­femoral neck and spine bone density,177 ­others report no ­association between the incidence of stress fractures and

low bone ­mineral density.178 Both scenarios are reasonable because, in addition to hormonal variation, other variables that might influence the difference in the incidence of stress fractures between male and female athletes include bone geometric differences, the higher level of adipose tissue compared with lean body mass in women, and static and dynamic biomechanical differences between the sexes.179,180 As previously discussed in this chapter, a relationship has been defined among amenorrhea (loss of normal menstrual periods for more than 3 months), low bone mineral density, and disordered eating. Treatment of the female athlete with a stress fracture should include, in addition to treatment of her fracture, an investigation into her training habits, dietary habits, ­menstrual history, and bone mineral density (Box 10-8). In most women athletes with recurrent fractures, an evaluation of bone mineral density is recommended, even in the absence of reported menstrual abnormalities or unusual habits.

Anterior Cruciate Ligament Injuries Both prospective and retrospective studies have shown a disproportionate number of noncontact (NC) ACL injuries in females participating in sports involving pivoting, cutting and changing directions, or jumping.181-188

The Total Number of Primary Stress Fractures by Sport and Gender in a Study of 5900 Division I Athletes

30

Male Fractures Female Fractures

20 15 10 5

C

lo po

l er

is Te nn

g in

le yb al

Vo l

W at

im m

in g/

D

iv

ftb

al

l

er So

Sw

cc So

ym G

f G ol

al l tb Fo o

re w C

ba ll Ba ro sk ss et C ba ou ll nt ry /T ra ck

0

se

Number of Fractures

25

Ba

483

Figure 10-4   Total number of primary stress fractures by sport and gender in a study of 5900 Division I athletes.   (Redrawn from Hame SL, LaFemina J, McAllister DR, et al: Fractures in the collegiate athlete. Am J Sports Med 32:449, 2004����� ��������� .)

484

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Risk Factors Multiple risk factors have been proposed, and despite extensive study of these, more questions than answers remain. A list of the proposed risk factors, divided by the category classification (environmental, anatomic, hormonal and neuromuscular) is provided in Box 10-9. Shoes, surfaces, and shoe-surface interactions, easily modifiable environmental risk factors, are enticing to pursue, but to date, investigators have not found the ideal shoe, surface, or shoe-surface interaction that affords adequate traction for sport performance but does not put the knee at risk for injury. Much has been written about the association of hip varus, knee valgus, foot pronation, ACL size, size of femoral notch, and body mass index with NC ACL injuries. To date, no one anatomic factor has been conclusively demonstrated to result in or be a reliably precipitating factor of NC ACL injury.

Although there is evidence that ACL injuries are not randomly distributed throughout the menstrual month, how hormones might alter the ACL risk equation is still debatable. Alteration of ACL biology, increases in knee laxity (although the relationship of laxity to ACL injury has not been definitively associated), and effects on mood or agility have been suggested. Neuromuscular risk factors (i.e., factors that relate to how we move) are the most fruitful for yielding reproducible data on influencing injury risk. Most (75% to 85%) NC ACL injuries occur when landing a jump, stopping, or pivoting to change direction. The injury position is accepted to be at foot strike with the knee near full extension during deceleration or change of direction.189 Studies have shown that strength, muscle coordination, muscle recruitment patterns, muscle endurance, and knee stiffness are important in maintaining knee joint stability. Increased hamstring strength, stiffness, and endurance are associated with decreased anterior tibial translation, whereas high

Box 10-9  Summary of Risk Factors for Anterior Cruciate Ligament Injury Environmental • Harder ground (high evaporation, low rainfall) may increase the shoe traction interface.288 • Increasing the shoe-surface coefficient may increase risk for injury (direct risk) and alter movement (indirect risk).289-291 • The role of prophylactic knee braces is confusing, except in one prospective randomized study of collegiate intramural football that showed 3 times the knee injury rate in nonbraced participants.292 Anatomic • Q angle,293,294 knee valgus,292,295-297 and increased foot pronation298-301 have been associated with increased risk for ­anterior cruciate ligament (ACL) tear in some studies. • The role of body mass index is not clear,302,303 but higher indices appear to be associated304 • Femoral notch indices,305-307 ACL geometry,304-308 and tibial measurements are not definitively associated with increased risk for ACL tear. However, the smaller femoral notch in females, even when normalized to body weight, appears to be associated.304,309-313 Hormonal • Sex hormones affect knee ligament laxity, but the overall effect is not clear. • Injury rates have been noted to vary with the phase of the menstrual cycle.301,314 It appears that increased injury occurs during the early and late follicular phases.192,315-317 Hormonal intervention for ACL prevention is not justified. • There is no evidence to recommend activity modification or restriction with respect to the phase of menstrual cycle to prevent ACL injury. Biomechanical

• The ACL has different mechanical properties in women.194 • Neuromuscular factors appear to be the most critical difference between adult men and women and are modifiable.311,318 • The knee is part of a kinetic chain and as such is influenced by it. • Awkward landing, inability to recover from perturbed gait, and difficulty changing directions have been associated with increased risk for ACL injury.63,319,320 • Women exhibit less knee and hip flexion, increased knee valgus, increased internal rotation of the hip, and increased ­external rotation of the tibia when landing from a jump or changing direction.63,64,191,319-326 • Quadriceps-dominant activation patterns64,191,321,327-331 during deceleration (landing) and cutting has been noted in ­females and is thought to be associated with increased risk. • Decreased quadriceps stiffness and strength and decreased knee stiffness in females are thought to be associated with ­increased risk.192,332-335 • Fatigue, not exercise,336 is associated with loss of dynamic control of the lower extremity.337 • Imbalance in strength, flexibility, and coordination is thought to be associated with increased risk for ACL ­injury.295,296,338,339

Adapted from Griffin LY, Albohm MJ, Arendt EA, et al: Understanding and preventing non-contact ACL injuries: A review of the Hunt Valley II meeting, January 2005. Am J Sports Med 34(9):1512-1532, 2006, with permission.

The Female Athlete

levels of quadriceps activity with hamstring imbalance and inadequate stiffness are associated with greater amounts of anterior tibial translation, as is muscle fatigue.190 Women have significantly less muscle strength in the quadriceps and hamstrings than men, even when strength is normalized for body weight.191 Sex differences also exist in the ability to produce adequate muscle stiffness across the knee.192 Regarding muscle firing patterns, elite female athletes, unlike athletic men, prefer to contract their quadriceps first rather than their hamstrings as a response to anterior tibial translation (a quadriceps-dominant knee).191 These neuromuscular factors appear to put women at risk for NC ACL injuries.

Mechanism of Injury Motion patterns at the knee and the entire lower extremity kinetic chain influence the occurrence of ACL injury. Loading at the knee is affected by the center of gravity and postural adjustment. Abnormal loading that exceeds the ACL’s ultimate failure load results in ACL injury.193 Some of the factors that determine the load across the ACL include the applied load to the knee (ground reaction force, body weight, muscle loads), knee stiffness (muscle coactivation), and ACL material properties (size, shape, modulus).193 Poor postural adjustment of the center of gravity and certain positions (knee extension, increased valgus with landing, internal knee torque) increase the load on the ACL.193 Stress on a smaller ACL will be greater for a given load and the load to failure will be lower for an ACL with a smaller surface area, given equal material properties. A recent cadaveric study (20 knees, 10 males and 10 females) of the structural properties of the ACL found lower mechanical properties in the female ACL.194 The mean age for the male specimens was 39 years (range, 26 to 50 years) and for female specimens, 37.7 years (range, 17 to 50 years). After taking into account age and anthropomorphic measures (body and the ACL size), the female specimens had an 8.3% lower strain at failure, a 14.3% lower stress at failure, a 9.43% lower strain energy density at failure, and a 22.49% lower modulus of elasticity. Dynamic loading refers to the loads transmitted across a joint that change over time and with flexion angle; it is related to nerve-muscle interaction and can be modified by training. Hence, neuromuscular risk factors associated with injury may be potentially modifiable through training. These risk factors form the basis of most NC ACL injury prevention programs.

Anterior Cruciate Ligament Reconstruction Prospective and retrospective studies have compared the postoperative results between males and females and between graft types (Table 10-6). In all studies, females had improved knee function and had less instability after reconstruction. With respect to instrumented arthrometry and clinical examination, studies have demonstrated no difference between males and females with patellar bone-tendon-bone (BTB) grafts.195,196 Hamstring grafts in females were associated with significantly increased laxity and failure in only three studies, which were not rigidly controlled.195,197-199 The amount of postoperative anterior knee pain is variable but appears to be less after hamstring

485

reconstruction in both males and females. A review of the literature revealed no study evaluating quadriceps tendon grafts for ACL reconstruction specifically with respect to sex. Fixation of the graft and other surgical techniques are not dependent on the patient’s sex. Similarly, postoperative rehabilitation is prescribed independent of sex. Reconstruction does not prevent post-traumatic osteoarthritis.200,201 One study subjectively and objectively evaluated female soccer players (mean age, 31 years) 12 years after ACL injury.201 Of the 84 women answering questionnaires, 63 (75%) had knee-related symptoms that affected their quality of life. Of the 67 with radiographs, 34 (51%) had changes consistent with osteoarthritis. The ACL was reconstructed in just over 61% of the women. There was no significant difference in the rates of limited function or radiographic arthritis between females undergoing ACL reconstruction and those treated nonoperatively.201 One retrospective cohort study of men (91%) and women (9%) compared reconstruction (n = 3795) to nonoperative management (n = 2781) and found no effect of sex on the results. Reconstruction was significantly associated with less repeat surgery and less meniscal and chondral reinjury.202

Prevention of Noncontact Anterior Cruciate Ligament Injury Identification of modifiable neuromuscular risk factors for NC ACL injury has led to the development of specific training programs designed to decrease NC ACL injury incidence and rates. The goals of neuromuscular prevention programs are to decrease knee loading and to improve protective motions in the kinetic chain. Most programs have achieved this goal; in no reports did ACL injury rates increase. In four studies, there was no effect on the number of ACL injuries.203-206 The components of successful programs include traditional stretching and strengthening, aerobic exercise, agility drills, plyometrics, and risk awareness.189 Many programs have sport-specific drills to help athletes more safely respond to unanticipated movement. Other exercises thought to improve movement patterns include core strengthening, exercise to increase hamstring to quadriceps strength, and exercise to decrease fatigue. Measured benefits of these programs include decreased landing forces, decreased dynamic knee valgus, decreased varus and valgus moments, and increased muscle activation (Box 10-10). A meta-analysis of six studies identified common elements that effectively decreased ACL injury rates in the three effective studies: plyometric training; technique training (sport specific), including balance training; and strength training.67 The limitations of the studies analyzed included few number of studies, small sample populations, low exposures, lack of robust statistical analysis, and poor study design. The duration of the effects of the training programs and how often and what type of maintenance training is required are unclear. Studies evaluating prevention programs are summarized in Table 10-7.

Patellofemoral Joint Injuries Arendt has classified disorders of the patellofemoral joint in females into three categories: patellofemoral pain, dislocation, and arthritis.207 Although historical and anecdotal

486

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 10-6  Summary of Literature on Anterior Cruciate Ligament Reconstruction in Females Reference

Study Type

N

Graft

Mean Age (yr)

Mean Follow-up (mo)

Acute F 29 (16-54) M 29 (15-45) Chronic F, 26 (14-51) M, 24 (15-41)

26 (22-25)

23 (16-43) 26(16-47)

37(21-68) 39(21-68)

23.2 (14-47) 29.0 (13-48)

40.9 (24.81) 39.0 (24-60)

27.3 (15-55) 30.3 (15-56)

51.94 (24-102) 59 (24-113)

Conclusion

Studies Designed to Compare Males (M) and Females (F)

340

Matched cohort from random selection, 1/1991-9/1993

94 F, 47 M, 47

BTB

341

Retrospective, 1991-1994

429 F, 133 M, 296

BTB

199

Retrospective, 10/1991-10/1996

65 F, 39 M, 26

HS

342

Retrospective, 6/1987-9/1991

200 F, 63 M, 137

BTB

343

Prospective nonrandomized, 1993-1994

90 F, 42 M, 48 90 F, 43 M47

BTB

Average 25 (13-52)

60

HS

Average 24 (13-52)

60

344

Retrospective survey, 1991-1996

BTB

21.5 ± 7.7 26.5 ± 9.0

61 ± 18 56 ± 16

198

Prospective nonrandomized, 1994-2000

151 F, 77 M, 74 40 F, 14 M, 26 40 F, 18 M, 22 200 F, 100 M, 100

347

Retrospective consecutive series span 7 yr

BTB HS HS

26 28 30 28

36 (30-46) 36 (30-46) 12-84

No significant ­difference in ­outcomes or ­complications Overall no ­significant differences by IKDC; Lysholm, Tegner; functional tests; or subjective complaints. Increased laxity in ­females on KT 1000, Lachman, PS No difference in complications; Lachman, ­anterior drawer, and ­functional testing; x-ray; subjective complaints Similar outcomes; At 5 yr for IKDC, Lysholm, females had higher KT 1000 at 2 yr, but there was no difference at 5 yr BTB: greater risk for developing early signs of degenerative joint disease No significant ­difference in outcome by gender BTB, no sex ­difference Possible increased laxity in HS grafts in females No significant ­difference in graft failure, IKDC, activity level, or functional tests between ­females and males

Studies Evaluating BTB and HS in Females only

345

Retrospective consecutive, nonrandomized

22

BTB, 18 HS, 4

15-19

31 (12-67)

346

Prospective nonrandomized

76

BTB, 37 HS, 39

25.2 (13-52) 23.2 (14-47)

52.0 (24-58) 40.9 (24-81)

All patients had normal (40%) or nearly normal scores by IKDC. Higher failure rate in HS (23%) com­ pared with BTB (8%), but not statistically significant

The Female Athlete

487

   TABLE 10-6  Summary of Literature on Anterior Cruciate Ligament Reconstruction in Females—cont’d Reference

Study Type

N

Graft

Mean Age (yr)

Mean Follow-up (mo)

348

Prospective nonrandomized, 1998-2001

65

BTB, 22 HS, 43

28 29

45 (30-65) 42 (30-62)

349

Prospective consecutive, nonrandomized

59

BTB, 28 HS, 31

28 (16-50) 25 (13-53)

26 (23-30) 25 (23-31)

Conclusion Both grafts in females are satisfactory. HS grafts are associated with less morbidity, greater return to previous level, and higher quality of life. No difference between HS and BTB in females in postoperative ­laxity, Lysholm score, Tegner activity level

BTB, bone-tendon-bone; HS, hamstring; IKDC, International Knee Documentation Committee; PS, pivot shift.

reports concluded that patellofemoral conditions are more common in females, recent literature supports this observation with respect to osteoarthritis and recurrent patellar instability, but not with patellofemoral pain.

Patellofemoral Pain The literature is not clear regarding the incidence208-211 and prevalence212,213 of patellofemoral pain in females compared with males. Of two prospective studies, one showed no difference in incidence,208 and in the other, only a 3% difference was noted between boys and girls (7% of 151 boys developed patellofemoral pain compared with 10% of 131 girls).211 The incidence decreases with age in girls208,209 and boys.214 The etiology of patellofemoral pain is most likely multifactorial.215 In general terms, the inciting or associated factors may be classified as anatomic (malalignment, abnormal patellar height, shallow trochlear groove), dynamic (muscle imbalance or overuse), or specific (chondral, plica, or neuroma). Anterior knee pain is a common symptom that may reflect poor compensation to a change in activity or baseline knee dynamics (loss of homeostasis) rather than a discrete diagnosis.215 Because the etiology may be multifactorial and the differential diagnosis is extensive, a thorough history is essential. Questions should differentiate pain from instability. Pain characteristics, swelling, giving way, mechanical symptoms, and grinding should be documented. Inciting events, overuse, trauma, response to Box 10-10  Measured Benefits of Anterior ­Cruciate Ligament Injury   Prevention ­Programs Decreased landing forces Decreased dynamic knee valgus Decreased varus and valgus moments of the lower extremity Increased muscle activation Improved movement patterns

treatment, and medical history are important to note. The physical examination should focus on the lower extremity as part of the kinetic chain, and the hip and the knee (iliotibial band, hamstrings) should be evaluated for flexibility and contracture. Femoral version, Q angle, quadriceps bulk and tone, patellar tracking and tilt, patella location, patellar chondrosis, pes planus, and ligamentous stability must be assessed. In some athletes, however, physical findings may be subtle or even absent. Malalignment may not be present.215 Females may have pain resulting from an imbalance between the hamstrings and the quadriceps. In athletes, pain may be associated with quadriceps shortening, altered vastus medialis response time, and hypermobile patella.211 Increased cartilage stress may play a role. Women have smaller cartilage volumes and thickness than men, even when adjusted for body weight and height.216 Patellar ­ contact areas on magnetic resonance imaging (MRI) were the same for men and women when normalized for size,217 as were bulk T2 patellar cartilage values.218 Patellofemoral pain was associated with greater patellar cartilage ­thickness estimated by MRI in men but not in women.219 In postmenopausal women, less cartilage volume on T2-weighted, fat-saturated, sagittal gradient-echo MRI was noted in women with pain.220

Patellofemoral Dislocation The incidence of first-time patellar dislocations in women may not be as high as previously reported, and the overall incidence is not known. The rate was the same for men and women in at least one study.221 Recurrent patellar instability has been documented more often in females.222-224 A prospective study of 189 patients found that the risk for dislocation was highest in females aged 10 to 17 years and that patients with recurrent dislocation and a history of instability are most likely to be female (P < .05) and older than the first-time dislocators (P < .05).222 The risk for recurrent instability after dislocation was 7 times greater than for first-time dislocators.222 Patients who sustain dislocation are likely to be active, and the injury is likely to occur during sports.221

488

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 10-7  Summary of Prevention Programs Reference

Study Design

Intervention

350

Nonrandomized controlled: alpine skiing 1 yr, with 2 yr of historical controls Severe injuries reduced by 62% among trained skiers Prospective: soccer M: 300 intervention, 300 controls 87% decrease in NC ACL injury 1.15 rate in control 0.15 rate in intervention Prospective: high school soccer, basketball, ­volleyball F: 366 intervention, 463 controls M: 434 0.12 ACL injuries per 1000 exposures in trained 0.22 per 1000 exposures in untrained 72% decreased rate of injury 66% compliance Randomized controlled: handball F: 111 intervention, 126 control 0.34 ACL injuries per 1000 exposures in trained 1.171 per 1000 exposures of practice 4.68 ACL injuries per 1000 exposures in untrained 23.38 per 1000 exposures in untrained of games Prospective: high school soccer F: 42 intervention, 258 control 0.25 ACL injuries per 1000 exposures in trained 0.33 per 1000 exposures in untrained Compliance not reported Prospective, randomized controlled: professional soccer F: 62 intervention, 78 control 0.73 ACL injuries per 1000 exposures in trained 0.22 per 1000 exposures in untrained 32% compliance Prospective: handball Three consecutive seasons Intervention second and third seasons only F: 855 intervention second season, 850 third season 0.18 ACL injuries per 1000 exposures in trained (2000-2001) 0.28 per 1000 exposures in untrained (1998-1999)

Awareness

351

69

352

205

206

73

Reference

71

72 6 wk NM,*†‡§ 3 times/wk, for 60-90 min, 18 preseason sessions

NM†‡

353

7 wk NM, 13 tread­mill speed sessions, 2 times/wk; ­agility sessions, for 20 ­preseason sessions 203 Balance board ­training,† for 108 in-season sessions

NM*†‡, 4 wk with 21 preseason sessions, 41 in-season sessions, for 15 min, 5 phases

204

Study Design 36% decrease risk for ACL injury 71% compliance Prospective: soccer 1 yr F: 575 intervention, 854 control 0.04 per 1000 AE in trained 0.14 per 1000 AE in untrained 72% reduction in ACL injury 100% reduction of NC ACL injuries in last 6 wk of season Prospective controlled: soccer 14- to 18-year-olds over 2 yr 36 teams in season Year 1: 1041 intervention; 1905 age- and skill-matched controls Year 2: 844 intervention; 1913 age- and skill-matched controls Year 1: 0.05 ACL injuries per 1000 exposures in trained; 47 per 1000 in untrained Year 2: 0.13 ACL injuries per 1000 exposures in trained; 0.51 per 1000 in untrained 98% compliance Cluster randomized ���������������������� controlled trial: team handball 1837 females and males Exposure F: 808 intervention, 778 control M: 150 intervention, 101 control 81 injuries in control group 48 injuries in intervention group ? Compliance Prospective controlled: ­handball F: 134 intervention, 142 control 0.04 ACL injuries per 1000 exposures in trained 0.21 per 1000 exposures in untrained 83% compliance 80% decreased risk for ACL injury One ACL tear in the intervention group and five in the control group Prospective controlled: high school athletes 2 yr F: 577 intervention, F: 862 control 0.167 ACL injuries per 1000 exposures in trained 0.078 in the untrained

*Plyometrics. †Balance.

‡Strength

training. of movements and feedback to teach body positioning and technique. ACL, anterior cruciate ligament; AE, athletic exposures; F, females; M, males; NC, noncontact; NM, neuromuscular.

§Analysis

Intervention

NM*†‡§ Began day 1 of season 15-min alternative warm-up

NM*†‡§, 20-min warm-up

NM†

NM*, 24 preseason sessions, 41 in-season sessions

NM*, 2 times/wk, for 20 min, during season; NM†

The Female Athlete

The proposed etiologies of dislocation in women are multifactorial and based on anatomic and neuromuscular factors (Box 10-11). Increased Q angle,225-227 femoral anteversion,228-231 patella alta,207,225,232 trochlear dysplasia,207,225,233 and quadriceps dysplasia225 are anatomic ­factors that have been associated with patellar instability and may represent individual risk factors. Patella alta is more common in women and is associated with instability. One computed tomography study of men and women (59%) found trochlear dysplasia in 85% and quadriceps dysplasia seen as lateral tilt in 83% of patients with patellar ­instability.225 The treatment of patellar instability in women is the same as in men but with emphasis on the etiologic factors that are modifiable, particularly quadriceps strength and endurance and neuromuscular control.

Patellofemoral Arthritis Patellofemoral arthritis occurs more frequently in women and has been noted to occur at least twice as often as in men.234 Women older than 55 years have been noted to have more severe arthritis in the knee than men.235 The etiology of the arthritis has been described as idiopathic, patellar instability with objective signs, post-traumatic (blunt trauma), and chondral calcinosis.207 Other etiologic factors include estrogen, which has a receptor on human chondrocytes235-237 and a specific allele of the estrogen receptor alpha gene (haplotype PX). Homozygotes with the haplotype PX had an increased risk for knee ­arthritis in men and women238 and of generalized arthritis in women.237 Box 10-11  Proposed Etiologic Factors ­Associated with Patellofemoral ­Dislocation in Women Anatomic Factors Knee • Increased Q angle225-227 • Trochlear dysplasia207,225,233 • Patella alta207,225,232 • Tibial tuberosity–trochlear groove distance225 • Quadriceps dysplasia225 Hip • Femoral anteversion228-231 Neuromuscular Factors* • Landing patterns Knee • Valgus • Abduction • Increased external rotation Hip • Increased internal rotation • Quadriceps-dominant loading • Less recruitment and activation of the quadriceps *These factors have been studied with respect to noncontact anterior cruciate ligament injury but will also favor conditions of patellofemoral lateral dislocation.

489

Modifiable factors of importance include osteoporosis and body weight. Osteoporosis has been associated with lower than expected incidence of arthritis.239,240 Less bone mass accounts for less subchondral bone stiffness, leading to a decreased incidence of arthritis.235 One study of postmenopausal women documented radiographic progression of arthritis in those with higher bone mineral density.240 The risk for arthritis is increased in obese people of several nationalities.241-245 In women, weight loss may decrease the risk for symptomatic knee arthritis.241 The treatment of patellofemoral arthritis in women is the same as for men. However, when total knee arthroplasty is indicated, two anatomic features of the patellofemoral joint may affect the procedure in women: less patellar height246 and larger Q angle.225-227 The average patellar height was 25.3 mm in men and 22.5 mm in women, demanding a thinner patellar component so that the resected patella is 8 to 10 mm.246 Implant companies have recently introduced endoprostheses that account for the anthropomorphic differences between men and women (narrower distal femur in the anteroposterior dimension in women).

Shoulder Instability There is a common perception that multidirectional shoulder instability (MDI) is more prevalent in women; however, this has not been supported in the literature.247 It has been suggested that females may have more lax shoulders,248-250 but this does not translate to symptomatic MDI of the shoulder. Laxity refers to increased normal asymptomatic motion of a joint, whereas instability refers to increased symptomatic motion of a joint.247 Increased laxity of the shoulder in a female does not ­ necessarily predispose her to develop symptomatic shoulder instability. In fact, there is evidence that looseness traits, such as ­ passive ­ finger, thumb, elbow, or knee hyperextension, do not ­necessarily correlate with shoulder laxity or ­instability.251-253 There have also been no reported sexrelated differences in mechanism of injury, predisposition to injury, treatment approaches, or outcomes for treating MDI. There is no support in the literature that shoulder instability should be treated any ­differently based on gender.

Frozen Shoulder The frozen shoulder syndrome, otherwise known as adhesive capsulitis, is common in women. A 2% overall prevalence in the general population has been reported, with 55% to 70% of these individuals being women.254-256 It is not known why women are at more risk for developing adhesive capsulitis.254-256 Theories include hormonal fluctuations in the premenopausal years and autoimmune factors, but neither of these etiologies has been ­substantiated. Frozen shoulder refers to a global decrease in active and passive range of motion of the shoulder resulting in contracture. It can be classified into primary and secondary. Primary frozen shoulder is idiopathic, whereas secondary frozen shoulder is related to an intrinsic or extrinsic cause. Primary frozen shoulder is more common in the female gender, in those older than 40 years, and in a variety of medical conditions such as diabetes and thyroid disease.

490

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

It can develop in recreational women athletes after minor or no trauma and is of great concern to them because it can affect not only sport participation and upper body exercise routines but also activities of daily living. The contralateral shoulder may become involved in 20% to 30% of patients but usually presents in a delayed fashion after the affected shoulder has recovered.255 The pathophysiology of frozen shoulder is a subject of debate. The two working theories are that it is either an inflammatory condition254,257,258 or a fibrosing ­condition.259 Some evidence would suggest that it is a combination of both these processes.257,260-262 Regardless of the ­underlying pathologic process, it is generally agreed that the process is localized to the joint capsule. A typical course of frozen shoulder can last up to 2 years or even longer; one author reported up to 7 years.263 At the final stage of the disease process, some limitations in motion may still persist, but in most cases, these are not considered functionally limiting.264 The natural history of frozen shoulder follows a continuum of four stages (Table 10-8).256 Diagnosis of frozen shoulder is primarily dependent on history and physical examination in both females and males. It has been suggested that patients classified into stage 1 or 2 may respond better to nonoperative measures of ­treatment, whereas patients with a more chronic ­presentation of the disease will have a higher incidence of ­ultimate surgical management after conservative treatments have failed.254-256 A plain radiograph is also ­recommended to rule out an obvious intrinsic or extrinsic cause for the contracture. Other potential causes include glenohumeral osteoarthritis, calcific tendinitis, and rotator cuff disease. The treatment of frozen shoulder typically starts with nonoperative measures. There are no sex-related differences in the treatment of frozen shoulder. The initial goal of treatment is to control the patients’ pain. This can initially be achieved with activity modification, anti-inflammatory medications, and occasionally pain medications. Exercise and physical therapy are key to the treatment protocol to maintain and regain range of motion. An intra-articular steroid injection may also be beneficial to decrease pain so that the patient can tolerate a stretching program.265 It is suggested that corticosteroid injections hasten the resolution of pain in the early stages of the disease but do not change the overall final functional result.254,256,260,266-268    TABLE 10-8  Stages of Frozen Shoulder Syndrome Stage

Phase

1

Timing (mo)

Presentation

Treatment

0-3

Pain, decreased motion Pain (more chronic), increasing loss of motion Pain lessening, significant motion loss

NSAIDs, PT ± injection NSAIDs, PT ± injection

2

Freezing

3-9

3

Frozen

9-15

4

Thawing

15-24

Resolving

NSAIDs, PT ± injection; ­possible surgery Resolving

PT, physical therapy; NSAIDs, nonsteroidal anti-inflammatory drugs.

Carette and coworkers randomized 93 patients into four different treatment groups: (1) intra-articular corticosteroid injection performed under fluoroscopic guidance with 12 sessions of physical therapy, (2) steroid injection alone, (3) saline injection with physical therapy, and (4) saline injection alone.269 They concluded that of these treatment options, a single intra-articular injection of corticosteroid administered under fluoroscopy, combined with supervised physical therapy, provided faster improvement in shoulder range of motion. If nonoperative treatment fails (6 to 12 months), operative intervention may be considered. More typically, this is the case for an individual who presents in stage 3 of the disease with a chronic contracture of the shoulder. Operative intervention includes closed manipulation under anesthesia, arthroscopic capsular release and manipulation, or open capsular release. Closed manipulation of the shoulder is not commonly performed because of the increased risk for fracture. Most authors now prefer a controlled manipulation after arthroscopic capsular release.270-274 An arthroscopic capsular release includes a full release of the rotator interval and anterior inferior capsule down to the 5-o’clock position (beyond this increases the risk to the axillary nerve), with or without release of the posterior capsule (if restricted internal rotation with shoulder in abduction). The release is finalized with a gentle closed manipulation, which requires less force and less risk for injury to adjacent structures because of the previous intraarticular release. Open capsular release is rarely performed since arthroscopic techniques were developed. Multiple postoperative rehabilitation protocols have been suggested. The primary goal is to control pain. Frequent shoulder motion can be done with either daily physical therapy visits immediately postoperatively or continuous passive range of motion machines. Individuals should not wear a sling and should be mobilized as frequently as possible within pain tolerance. Recreational sporting activity may be resumed when the pain is controlled. Pain is the main limiting factor for activity, and typically no specific activity restrictions are prescribed.

Forefoot Problems in Women Athletes Traditionally, athletic women’s feet problems were attributed to women having to wear smaller sizes of footwear made for men.275 Studies have substantiated that women’s feet are proportionally different than those of men. Women have a wider forefoot, shorter arches and metatarsal length,275-278 and a shorter foot length per total body height than do men.277 Interestingly enough, women demonstrate no statistically significant increase in contact area or plantar pressure during gait.275,279 However, because women have shorter body length per body height than men, a woman’s foot strikes the ground more times than a man’s in covering the same distance.280 During the past 20 to 30 years, shoe manufactures have incorporated these intrinsic anatomic differences into athletic shoe lines for women and generated a market that even as early as 1994 yielded an estimated $5.4 billion in women’s athletic shoe sales, compared with $5.3 billion for the sale of men’s athletic shoes in the same year.275

The Female Athlete

Unfortunately, although women have had daily shoewear designed for them for centuries, such footwear, although not as sadistic as the Chinese tradition of foot-binding, has in many instances been styled for looks rather than function. High spiked heels with pointed toe boxes, often worn 1 to 1½ sizes smaller than the foot, play havoc on women’s feet and can be associated with bunionettes, bunions, neuromas, hammertoes, and metatarsalgia—all problems for which women (including athletic women) seek medical care far more frequently than men.277,281-284 For example, hallux valgus is 2 to 4 times more common in women than men in our society.285 However, Coughlin and Frey276 reported that in ancient times, when no shoes were worn, and in cultures in more recent years in which no footwear is worn (e.g., areas in the Belgium Congo, West Africa, and Primitive New Zealand), neither men nor women have bunions. Surgical approaches to athletic women’s forefoot problems need to take into account that function, not looks, is of prime importance. For example, if during correction of a bunion deformity in a ballerina or cross-country runner, the first ray is shortened, an increase in the mechanical load placed on the second or third metatarsals can result in recurrent stress fractures.286 Similarly, any surgery that results in restriction of dorsiflexion of the great toe in a ballerina could severely hamper her performance. Caution, therefore, is suggested when contemplating surgical correction of cosmetic forefoot deformities in women athletes. Even with functional deformities, orthotics, tapes, pads, and other more conservative measures should be initially used. Footwear suggestions for athletic women are provided in Box 10-12. Box 10-12  Footwear Suggestions   for Athletic Women

• Fit shoes at the end of the day and to the larger foot. • Avoid very high spiked heels except on rare social occasions. • Select shoes with wide toebox and a well-constructed arch. • When appropriate, replace athletic footwear halfway through the event. • Make certain that athletic shoes hold the heel firmly to prevent the foot from sliding forward in the shoe, increasing the chance of forefoot problems.

S U G G E S T E D

C

r i t i c a l

P

491

o i n t s

• Since 1972, the number of women athletes has increased, as has the market value of women as sport participants.

• A greater understanding of the results of sexual dimor-

phism on health and disease has occurred since 1993, when NIH-funded research studies began providing analyses of results by sex and gender. • The primary nutritional concern for the female athlete is inadequate dietary intake. • Regular exercise benefits overall health and fitness. • Regular moderate submaximal exercise during low-risk pregnancy is safe and is a modifiable factor promoting maternal fitness and sense of well-being. • The American College of Obstetricians and ­Gynecology has established guidelines for the indications and contraindications of exercise during pregnancy. • Treatment of the female athlete with a stress fracture should include an investigation into training habits, footwear, dietary habits, menstrual history, and bone mineral density. • Patellofemoral pain is multifactorial and may occur in the presence or absence of malalignment. • Women have a higher risk for recurrent patellar instability and dislocation than men. • Isolated arthritis of the patellofemoral joint is more common in women. • The treatment of patellofemoral disorders is the same for men and women. • The exact factors resulting in the increased risk for ACL injury in females are still unclear. • Females undergoing ACL reconstruction have improved knee function and less instability than before surgery. • Prevention of ACL injury is based on training to ­improve strength and response to perturbed gait and movement patterns during deceleration and changing directions. • Prospective studies evaluating ACL prevention programs have demonstrated that most prevention programs that incorporate aerobic conditioning, strengthening and flexibility exercises, agility activities, plyometric drills, or ACL injury awareness teachings decrease the ­incidence of ACL injury. • Frozen shoulder is more common in women. • There is no support in the literature that shoulder ­instability should be treated differently based on sex. • Women’s feet are proportionally different than men’s. • Surgical approaches to athletic women’s forefoot ­problems need to take into account that function, not looks, is of prime importance.

R E A D I N G S

American College of Obstetricians and Gynecologists: Exercise during pregnancy. Educational pamphlet AP119. Washington, DC, ACOG, 2003, pp 171-173. Bathgate SL, Larsen JW, Chahine EB, Macri CJ: Exercise in pregnancy and beyond: Benefits for long-term health. In Garrick JG (ed): Orthopaedic Knowledge Update Sports Medicine 3. Rosemont, Ill, American Academy of Orthopaedic Surgeons, 2004, pp 379-387. Drinkwater BL: Women in Sport. Volume VIII of the Encyclopaedia of Sports Medicine: An IOC Medical Committee Publication in collaboration with the International Federation of Sports Medicine. Oxford, UK, Blackwell Science– ­Osney Mead, 2000. Garrett WE, Lester GE, McGowan J, Kirkendall DT: Women’s Health in Sports and Exercise. American Academy of Orthopaedic Surgeons Symposium. Rosemont, Ill, American Academy of Orthopaedic Surgeons, 2001. Griffin LY, Albohm MJ, Arendt EA, et al: Understanding and preventing non­contact ACL injuries: A review of the Hunt Valley II meeting, January 2005. Am J Sports Med 34(9):1512-1532, 2006.

Griffin LY, Garrick, JG (eds): Women’s musculoskeletal health: Update for the new millennium. Clin Orthop 372:1-179, 2000. Hannafin JA, Chiaia TA: Adhesive capsulitis: A treatment approach. Clin Orthop 372:95-109, 2000. Ireland ML, Nattiv A: The Female Athlete. Philadelphia, Elsevier Science, 2002. Thomas DB, Taylor DC: The female athlete triad. In Garrick JG (ed): ­Orthopaedic Knowledge Update Sports Medicine 3. Rosemont, Ill, American Academy of Orthopaedic Surgeons, 2004, pp 345-352. Tosi L, Griffin LY, O’Connor MI (eds): Sexual dimorphism in musculoskeletal health. Orthop Clin North Am 37(4):531-540; 555-610, 2006.

R E F E R E N C E S Pleases see www.expertconsult.com

C H A P TE R  

11

Environmental Stress S e c t i o n

A

Heat Illness Jorge E. Gómez, Andrew J. Grove, and Wendy McBride

EPIDEMIOLOGY OF ­ HEAT-RELATED ILLNESS IN SPORTS During the period from 1999 to 2003, 3442 deaths due to heat illness were reported in the United States; 53% of these occurred in individuals 15 to 64 years of age and 40% in individuals older than 65 years. The lowest incidence of heat-related death occurred in children ages 5 to 14 years.1 During the unusually hot late summer of 2005, three deaths were reported in high school football players according to the National Center for Catastrophic Sports Injury Research.2 In the previous 2 years, no deaths were reported. Heat exhaustion and other milder forms of heat illness in athletes occur much more commonly, although the true incidence is not known. Fortunately, most forms of heat illness are preventable.

PHYSIOLOGIC MECHANISMS OF HEAT PRODUCTION AND DISSIPATION Metabolic processes at rest produce heat as a byproduct, with muscle contributing the largest amount of heat ­produced at rest. About 75% of metabolic energy produced by working muscle is converted to heat.3 Muscle work dramatically increases heat production.4 Fever is another source of heat production.5 Shivering is a reflex mechanism for heat production, normally in response to excessive heat loss due to low ambient temperatures. Shivering consists of rapid, low-amplitude, involuntary muscle contractions. Although adipose tissue serves to insulate the body in addition to being an energy store, it is not a very metabolically active tissue and therefore does not produce much heat. The control of body temperature is accomplished by endocrine, respiratory, circulatory, and neural mechanisms. Dissipation of excess heat is accomplished primarily by regulating blood flow to the skin. Flushing of the skin typically seen in individuals exercising in the heat occurs in response to increasing core temperature and represents maximal dilation of cutaneous blood vessels.

Blood ­ circulation brings heat generated within the body to the surface. There, heat dissipation may occur by three mechanisms; radiation, evaporation, or conduction. The potential to dissipate heat by each of these mechanisms is naturally dependent on the difference between the body temperature and the ambient temperature. Radiative heat transfer is the principal mechanism of heat dissipation by the body. Theoretically, radiative heat transfer may occur from anywhere within the body and is dependent on the temperature difference between the heat source and the outside, and the insulating capacity of the interposed tissues. Radiative heat transfer at the surface is impeded by any insulating materials, such as clothing and equipment. Radiative heat transfer from within the body is impeded by the insulating capacity of the adipose tissue. Blood flow represents a form of conductive heat transfer. The capacity of the blood to convey heat from the core to the surface is dependent on sympathetic control of cutaneous vascular tone, the circulating volume, and the cardiac output. Conductive heat transfer at the skin occurs when a medium cooler than the skin is in contact with the skin, such as a cool towel, water, an ice bag, or a cooling blanket. Sweating is the primary mechanism of evaporative heat transfer and is enhanced with any additional water applied to the skin.

PHYSIOLOGIC ADAPTATIONS TO EXERCISING IN THE HEAT Regular exposure to hot environments results in a number of physiologic adaptations that allow the exercising athlete to better resist the negative effects of heat.3 These adaptations collectively are referred to as acclimatization and include the following: • Decreased core body temperature at rest • Decreased heart rate during exercise • Increased perspiration rate • Initiation of sweating at a lower core temperature • Initiation of thirst drive at a lower serum osmolality • Decreased sodium losses in sweat and urine • Expanded plasma volume 493

494

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Contrary to popular belief, athletes who are acclimatized to the heat actually sweat more. However, the body is able to maintain plasma volume because of increased absorption of sodium from sweat before excretion. The expansion of the plasma volume is extremely important for maintaining cardiac output during heat stress, which ensures efficient transfer of heat from the core. Evidence suggests that the expanded plasma volume is likely a result of elevated plasma aldosterone levels, which increase sodium and water retention from the kidney. Acclimatization to the heat requires about 2 weeks of controlled exposure in healthy children and teens and about 7 to 10 days in healthy adults.

RISK FACTORS FOR HEAT ILLNESS AND PREVENTION The physician caring for athletes should be cognizant of important risk factors for heat illness, including the following: • Overweight • Poor conditioning • Previous history of heat illness • Current infection (e.g., common cold) • Predehydration (excessive use of caffeine or alcohol) • Use of illicit drugs • Use of certain prescription medications (see those listed under “Heat Stroke”) • Chronic disease, especially diseases of the lung, heart, kidneys, and blood • Sickle cell trait Prevention of heat illness begins with preparticipation screening, which includes medical history questions addressing the risk factors listed earlier, especially a previous history of heat illness, use of medications, and history of chronic diseases. Individuals with risk factors for heat illness should be cautioned about intense exercise in extremely hot and humid conditions and should be counseled about proper hydration. Signs of proper hydration include the absence of thirst or dry mouth, regular voiding of urine, clear to lightly colored urine, and soft stools. Individuals exercising in the heat should be instructed to drink regularly before, during, and after exercise. Generally, water is the fluid of choice for hydration during exercise. Fluids containing sodium and other electrolytes generally are not needed except in special circumstances, including a previous history of hyponatremic dehydration, evidence of ongoing hyponatremic dehydration (see later), continuous endurance exercise, and repeated bouts of intense exercise in a given day (e.g., weekend tournament play). Fluids containing carbohydrate have been found to be beneficial only for individuals performing continuous endurance exercise lasting an hour or more.6 Special consideration should be given to children exercising in the heat. Children have increased risk for heat illness because of the following reasons: • Children have greater surface area–to-mass ratio than adults, allowing greater heat transfer between the environment and the body.

• Children produce more heat per kilogram body weight during exercise than adults. • Children produce less sweat than adults. • The capacity for heat transfer from the core to the surface through the blood is limited in children because of their limited cardiac output.4 • The thirst drive is not as highly developed in children as in adults. Field studies of exercising children have shown that, left on their own, children exercising in the heat generally do not replenish fluids adequately, even when they have unlimited and unhindered access to water.7 Other studies of exercising children have shown that children will voluntarily hydrate more adequately if flavored beverages are available, compared with plain water.8 Adults supervising children engaging in sports activities in the heat should regularly remind children to drink and have flavored beverages available. Certain coaching innovations have been used to decrease the risk for heat illness in middle school and high school athletes. One method is to allow players unlimited access to water. Another approach used in football is to conduct the first week of practice in shorts and helmets only to allow players to acclimatize to the heat.

HEAT SYNCOPE Definition Heat syncope does not result from excess body heat. It is form of vasovagal syncope. As such, the typical individual who suffers heat syncope is a female with a previous history of syncope.

Presentation The typical scenario is a hot, humid day, and the individual is unaccustomed to the heat, has not eaten much during the day, has had recent upper respiratory infection symptoms, and is not drinking enough fluids.

Treatment and Prevention Treatment of heat syncope is the same as for vasovagal syncope. The ABCs of cardiopulmonary resuscitation should be followed. The legs should be elevated, the heart rate monitored, and blood pressure monitored if possible. Once the individual regains consciousness, he or she should be given fluids and carbohydrate (e.g., a light snack or sport drink) and instructed to refrain from activity for the rest of the day. Any underlying illness should be addressed.

HEAT EXHAUSTION Definition Heat exhaustion is primarily a dehydration phenomenon. It may occur in conditions of high temperature as well as during moderately high temperatures with high

Environmental Stress

495

humidity. The risk for heat illness is a function of both ambient temperature and humidity. Although core body temperature may become elevated above normal in heat exhaustion, the body continues to maintain heat regulatory functions, with the result that body temperature in heat exhaustion rarely rises above 40° C. Individuals with heat exhaustion typically have lost 10% or more of their body weight as water through perspiration during activity. Those at greatest risk for heat illness are elderly persons, individuals with hypertension, and those unacclimatized to the heat.

oral rehydration should be given sodium or sodium­containing fluids. If after another 30 minutes symptoms do not improve or worsen, the individual should be taken to a medical facility immediately for intravenous infusion of normal or half-normal saline, serum electrolyte determination, and close monitoring.

Presentation

The cause of heat cramps, or muscle cramps occurring with heavy exertion in hot, humid conditions, remains elusive. Studies have failed to implicate dietary habits, nutrient intake, or serum electrolyte disturbances as causative factors. Recent studies conducted with university football players have revealed that individuals prone to heat cramps generally have higher sweat sodium losses and higher sweat volume losses than do individuals not prone to cramping.10 The absence of serum electrolyte disturbances in athletes with heat cramps suggests that the problem is intracellular. Definite studies on the cause of heat cramps would therefore involve intracellular probes or muscle biopsy done during exercise at the time of cramping, which would be difficult to accomplish.

Typical symptoms of heat illness include severe fatigue, weakness, dizziness, muscle cramps, abdominal pain, headache, and nausea. Anxiety and hyperventilation may follow these unpleasant symptoms. Common physical signs include pallor, listlessness, vomiting, unsteadiness, collapse, and muscle rigidity. Minor mental status changes are also common, including confusion, irritability, combativeness, and emotional lability. These should be contrasted with the more severe mental status changes seen in heat stroke (see later). Physical examination of the individual with heat exhaustion reveals the prior physical findings. The individual may have dry skin if the dehydration is severe or may continue to sweat. Blood pressure is usually low, and pulse is elevated. In more severe dehydration, the skin is doughy, and tenting may be present.

Treatment Initial treatment of heat exhaustion includes removing the individual from physical activity and moving him or her to a cool or shady place. If the individual is able to tolerate fluids by mouth, water should be given in frequent small amounts. Heat illness is usually cured by oral rehydration with water because most cases involve isonatremic dehydration. Individuals with ongoing vomiting may require intravenous hydration.

Hyponatremic Dehydration Occasionally, certain individuals experience hyponatremic dehydration. Known risk factors for hyponatremic dehydration include female gender, continuous exercise lasting more than 1 hour, low sodium intake before exercise, and higher than usual sodium losses in the sweat. It should be noted that even individuals who experience hyponatremic dehydration have sweat sodium concentrations less than those seen in cystic fibrosis. In most cases of heat exhaustion, serum electrolyte determinations are not readily available. However, the clinical response to oral rehydration is often telling. Fatal cases of hyponatremic dehydration that have been reported in military trainees have revealed a pattern of lack of improvement of heat exhaustion symptoms despite water rehydration.9 In these cases, recruits developed abdominal distention and began producing urine but continued to have mental status changes and muscle cramps. Coma, seizures, and death ensued. As a general rule, individuals with symptoms of heat ­exhaustion whose symptoms do not respond within 30 minutes of ­monitored

HEAT CRAMPS Definition

Treatment Traditional cures for heat cramps, such as salt tablets, ­electrolyte-containing drinks, infusion of intravenous ­fluids, ice massage, and pickle juice, lack any evidence base for efficacy and anecdotally have been successful ­sporadically at best. A general approach to the treatment of muscle cramps includes removing the athlete from activity, gentle massage and passive stretching of the muscle, and administration of an electrolyte-containing drink. It appears that once cramps set in, there is little that can be done during the course of a game that will cause the cramps to abate.

Prevention Prevention should be aimed at identifying individuals who are prone to cramping. Adequate preseason conditioning in a sport-specific manner may prevent early season cramps. Proper attention to diet, including salt-containing foods, adequate pre-exercise hydration, and rest, may also help susceptible individuals avoid cramps. Although numerous placebo-control trials have not identified muscle cramping as a common side effect of creatine supplementation, anecdotally, many athletes have reported discontinuing creatine use because of muscle cramping. Interestingly, one recent study indicated that use of ­creatine monohydrate may actually enhance an athlete’s resistance to heat effects.11 Angiotensin-­converting enzyme inhibitors, which are being prescribed for more and more young athletes owing to the increased ­prevalence of ­hypertension, may cause muscle cramping with ­exertion. Individuals with recurrent episodes of muscle cramping despite reasonable preventive measures should be suspected of having a primary myopathy. The most common myopathy found in otherwise healthy-appearing ­ individuals is

496

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

McArdle’s disease (myophosphorylase deficiency). Mildly affected individuals often do not become symptomatic until adolescence, and then bouts of vigorous exercise cause muscle aches and cramping, sometimes lasting for hours. Muscle creatine phosphokinase is markedly elevated, and myoglobinuria may be present after an attack.12

HEAT STROKE Definition The pathophysiology of heat stroke involves breakdown of the homeothermic mechanisms, allowing core body temperature to rise above 40° C. Heat stroke may occur with or without exercise but is always seen in the setting of high ambient temperatures. Individuals at greatest risk for heat stroke are children younger than 4 years, adults older than 65 years, individuals who are overweight, and those who are ill or who take certain medications, including psychotropics (haloperidol [Haldol], chlorpromazine), medications for Parkinson’s disease (inhibit perspiration), tranquilizers (phenothiazines, butyrophenones, and thioxanthenes), and diuretics.13,14 A common victim is an elderly person living alone in a dwelling without air conditioning who succumbs after several days of unusually high temperatures. In athletes, heat stroke may occur after just an hour of exercise in extreme heat. The loss of thermoregulatory function may be sudden, with the athlete passing quickly from confusion and irritability to obtundation and coma. Seizures often ensue, and cardiovascular collapse and cardiac arrest may be imminent.

Presentation The individual with heat stroke will be lethargic or obtunded. The skin is hot, dry, and usually flushed. Early warning signs include lack of sweating, throbbing headache, dizziness, nausea, confusion, and a rapid, strong pulse. Once the victim becomes obtunded, breathing is often rapid, the pulse is rapid and weak, and the blood pressure is low. Body temperature will usually be above 40° C (104° F). Studies indicate that body surface thermometers and tympanic thermometers do not reliably measure core temperature in the exercising individual.15,16 Despite concerns over preserving modesty in the field, rectal temperature measurement is the most reliable method of monitoring core body temperature in victims of heat illness.

Treatment The first responder should follow the ABCs of cardiopulmonary resuscitation (CPR). Emergency Medical Service (EMS) should be activated. As soon as respiratory effort is adequate and a pulse is present, the patient must be cooled rapidly. Some authorities advise immersing the victim in ice water if possible. However, several fatalities have been reported using this method. In these cases, ice water or ice bath immersion was thought to have resulted in simultaneous cutaneous vasoconstriction, thereby decreasing heat loss from the surface, and shivering, which produced

a­ dditional body heat, with the result that core ­temperature continued to rise, leading to cardiac arrest. A less risky method for cooling involves placing ice bags in locations where large-bore blood vessels are near the skin surface, including the scalp, neck, axillae, groin, and popliteal fossae.17,18 At the same time, evaporative cooling is enhanced by wetting the entire skin surface with water and blowing air over the body. Infusion of cooled intravenous fluids, if available, should begin. Respirations, pulse, and blood pressure should be continually monitored until EMS arrives in case CPR needs to be initiated.

Prevention Although heat stroke is much less common than heat exhaustion, those responsible for athletic event coverage must be prepared for this possibility. This includes having ice and water available, a method of blowing air over the body, such as towels or a fan, and a means for transporting the victim to a cooler place until EMS arrives.

EXERTIONAL RHABDOMYOLYSIS Definition Exertional rhabdomyolysis represents necrosis of muscle cells due to excessive lactate buildup and acidosis during exercise. Excessive buildup of lactate and acid, normal byproducts of muscle work, occurs when physical work demands far exceed the ability of muscles to maintain work output. This can occur for one or more of three conditions: (1) the muscles are unaccustomed to a certain level of mechanical work; (2) the work demand exceeds the ability of even well-conditioned muscles to work; (3) other factors such as dehydration, acidosis, excess body temperature, excess ambient temperature, or hypoxia impair muscle function. Most cases of exertional rhabdomyolysis occur because of the first reason, that is, inadequate physical conditioning. Fortunately, most of these cases are relatively mild and result only in a condition known as exertional myositis, in which muscle pain and weakness are associated with an increase in serum creatine phosphokinase levels, typically less than 20,000 units/L and mild myoglobinuria without glomerular injury. A combination of high ambient temperature and humidity can lead to dehydration and severe acidosis, resulting in widespread breakdown of muscle tissue. Cellular breakdown products—most importantly, myoglobin, proteases, and inflammatory mediators—enter the circulation. Obligatory filtration of large amounts of myoglobin results in glomerular injury of varying degrees. High serum pH compromises visceral function, most notably of the heart, and proteases damage the respiratory epithelium, sometimes leading to adult respiratory distress syndrome.

Presentation The individual with exertional rhabdomyolysis presents with severe muscle pain and weakness and evidence of renal failure, commonly hypertension and edema, accompanied by a rise in serum creatinine.

Environmental Stress

Treatment Treatment of exertional rhabdomyolysis includes correction of acidosis, proper hydration to aid in the clearance of myoglobin without causing fluid overload, monitoring of renal function, and support of cardiac and pulmonary function.

Prevention Exertional rhabdomyolysis can be prevented by avoiding exercise vastly in excess of what the individual is accustomed to, ensuring adequate hydration of athletes exercising in hot and humid conditions, and ensuring adequate recovery between exercise bouts. In the young athlete, the most important measure for avoiding exertional ­rhabdomyolysis is careful and proper conditioning. Exertional rhabdomyolysis is most likely to occur in young athletes at the beginning of a season. It should be noted that exertional rhabdomyolysis has occurred in youngsters not competing in athletics but involved in other activities such as band, drill team, and Reserve Officers’ Training Corps (ROTC) in instances in which inordinate physical exercise is used as punishment. Use of exercise as punishment has several negative consequences, such as future avoidance of healthful physical activity and, in the most dire case, possibly lifethreatening exertional rhabdomyolysis. Exertional rhabdomyolysis occurs next most commonly in young and middle-aged adults either performing a single bout of exercise beyond the intensity to which they are accustomed, especially male “weekend warriors,” or after several bouts of intense training without adequate recovery between bouts. Individuals in this group can avoid exertional rhabdomyolysis by paying attention to fatigue and muscle pain as indicators to stop a given bout of exercise and to delayed-onset muscle soreness (diffuse muscle soreness 1 to 2 days after a bout of intense exercise) as an indicator for ensuring proper recovery between exercise bouts.

Exertional Rhabdomyolysis and Sickle Trait Several case reports have highlighted the increased risk for exertional rhabdomyolysis in individuals with sickle trait. In one report, military recruits with sickle trait were 30 times more likely to die during basic training than were recruits without the trait.19 Heat stress produces metabolic changes that foster sickling in individuals with sickle trait. These changes include severe hypoxemia, acidosis, hyperthermia, and red cell dehydration. Individuals with sickle trait appear to be at greatest risk with high-intensity exercise, specifically repeated sprints, that is performed first thing in the morning (when the body is relatively dehydrated) in high humidity. Sickling and the ensuing rhabdomyolysis may occur very rapidly. In contrast to the athlete with heat cramps, the individual experiencing a sickling crisis during exercise may succumb very soon after beginning exertion, have relatively little muscle pain and rigidity, and have pronounced muscle weakness. Prevention of exertional rhabdomyolysis associated with sickle trait focuses on knowing the hemoglobin status of each athlete and taking precautions in high-humidity conditions to avoid repeated long

497

sprints with little or no recovery time. Knowing the hemoglobin status of each athlete may not be practical in certain situations or locations. Many states perform hemoglobin testing as part of the screening for metabolic diseases in newborns. However, this information may not be readily available, especially to high school and collegiate athletes, or in states where hemoglobin status has not been part of the newborn screening program. Training guidelines for prevention of rhabdomyolysis from sickle trait include the following20: • Gradual buildup in intensity and duration of workouts • Year-round conditioning • Cessation of activity with onset of symptoms (cramping, pain, weakness) • Close monitoring of athletes with known sickle trait during exercise • Education of athletes with sickle trait about symptoms of rhabdomyolysis • Optimal control of asthma in affected athletes Although it is unlikely that organizations will mandate that the hemoglobin status of each athlete be known before clearance is allowed for athletics, adherence to the preceding general recommendations will likely prevent serious or life-threatening rhabdomyolysis in athletes who are unaware that they carry the trait.

C

r i t i c a l

P

o i n t s

l Most cases of heat illness are preventable. l Allowing a period of acclimatization early in a season may prevent heat illness. l Submerging the athlete with heat stroke in an ice bath may lead to cutaneous vasoconstriction and shivering, which may further increase core temperature. l The athlete with apparent heat exhaustion who does not improve clinically after 30 minutes of oral rehydration should be suspected of having hyponatremia. l Athletes with sickle trait must pay close attention to hydration. Their coaches must be educated about avoiding repeated short sprints with little rest in between.

S U G G E S T E D

R E A D I N G S

American Academy of Pediatrics, Committee on Sports Medicine and Fitness: Climatic heat stress and the exercising child and adolescent. Pediatrics 2000; 106:158-159, 2000. Armstrong LE, Casa DJ, Millard-Stafford M, et al: Exertional heat illness during training and competition. Med Sci Sports Exercise 556-572, 2007. Gatorade Sports Science Institute, Sports Science Library. www.gssiweb.com. National Athletic Trainers Association. Activity Health Tip #1. Heat Illnesses. www.nata.org Sawka MN, Burke LM, Eichner ER, et al: Exercise and fluid replacement. Med Sci Sports Exercise 337-390, 2007.

R eferences Please see www.expertconsult.com

498

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

B

Cold Injury Andrew J. Grove and Jorge E. Gómez

Cold injury has been described and detailed since early times. For the most part, this type of injury has been of most consequence in military situations. As early as 218 bc, Hannibal lost nearly 45% of his men while crossing the Alps.1 Napoleon experienced the loss of 250,000 troops from cold injury when invading Russia in the 19th century.2 In the winter of 1777 to 1778, 10% of George ­Washington’s army perished because of cold.3 During the 20th century, cold injury was responsible in both World War II and the Korean War for 10% of troop casualties.4 The military has provided the basis for much of present practical understanding of cold injury because of a long history of cold exposure experiences and documentation of these experiences in the historical record. More recently, a greater number of individuals have become involved in cold-weather sports and activities. Many of these activities, such as downhill and cross­country skiing, snowboarding, backpacking, mountaineering, snowmobiling, and hunting, often involve long exposure times to cold environments at high levels of exertion. As exposure time increases, the risk for cold injuries increases dramatically. Individuals involved in these activities account for most nonmilitary incidents of hypothermia, frostbite, and nonfreezing cold injury. Knowledge of presentation, basic management, and prevention of these conditions is important for sports medicine physicians because of the likelihood that they may encounter them in practice.

PHYSIOLOGY OF COLD INJURY Humans are unable to vary body temperature greatly on their own and can only avoid impairment in function by maintaining a core temperature in the 4°C range. To survive in cold environments, core temperature is ­maintained by two mechanisms: increasing heat production and decreasing heat loss. Heat loss is decreased through adequate insulation. Although skin temperature can vary greatly, core temperature remains nearly constant at about 37°C. This high core temperature is maintained by a high rate of metabolic heat production.5 In the thermoneutral range of temperature from 28°C to 34°C, the nude human body does not require insulation because it will not cool. Metabolic heat produced by the body is ­ sufficient to ­ maintain body temperature in this ­environment.

Heat Production As mentioned earlier, humans maintain a constant core temperature in a very narrow range. The production of heat, or thermogenesis, is accomplished through four major ways: (1) resting metabolic rate, (2) exercise-induced thermogenesis, (3) the thermic effect of food, and (4) thermoregulatory thermogenesis (particularly shivering).6 The resting (or basal) metabolic rate refers to energy expenditure at rest in a thermoneutral environment. About 1 kcal/kg per hour of heat is generated at resting metabolic rate. The body is not particularly efficient in its use of energy: 75% of energy is released as heat, whereas only 25% of energy is used for work. To use energy, it must be provided. Food is the source of energy, and it comes in the form of carbohydrate, protein, and fat. Fats are the most energy dense of the three forms of food energy, providing 9 kcal/g. Carbohydrates produce 4 kcal/g of energy, and protein produces a little less than this amount. The average 70-kg individual requires about 2400 kcal/day for regular daily activity. Increased activity or exercise produces heat above and beyond that generated by the resting metabolic rate. The amount of heat varies according to the intensity and duration of the activity, be it walking, running, jumping, or swimming. Athletes can require high amounts of energy to maintain muscle activity, even 6000 to 10,000 kcal/day. For military operations in cold environments, 3400 to 4300 kcal/day of energy is recommended, with up to 5000 kg/day for some training operations.7 Carbohydrates are the main source of readily available energy for athletes. High-­carbohydrate diets are thus recommended. Caloric intake should consist of about 60% of the total as carbohydrate, 25% as fat, and 15% as protein. The absorption, breakdown, and storage of ingested food also produce heat. This process is a part of what is referred to as the thermic effect of food.6 The thermic effect of food provides a very small proportion of total body heat production. Thermoregulatory thermogenesis refers to involuntary mechanisms by which the body increases heat production in order to maintain the precarious heat balance necessary to sustain core temperature. Shivering is the predominant form of thermoregulatory thermogenesis. It represents an involuntary pattern of repetitive, rhythmic muscle contractions.6-8 These muscle contractions can increase the

Environmental Stress

metabolic rate up to 5 times that of the ­ resting state.6,9 The amount of shivering is proportional to the amount of cold stress experienced.8 However, shivering can be maintained only as long as glycogen stores are available. Nonshivering thermogenesis provides another involuntary form of heat production through the effects of sympathetic discharge on exposure to cold environments. However, its contribution to overall heat production is negligible.

Heat Loss Heat produced by the body needs to be balanced by heat loss to maintain thermoneutrality. Heat is lost through the skin and lungs, with the skin accounting for the significant majority (about 90%). Heat loss through the skin can be controlled with vasoconstriction, which can reduce cutaneous blood flow by 100-fold. Loss of heat from the lungs is quite variable. Although respiration at rest is often under involuntary control, voluntary control comes into play with various types and intensities of physical activity and at altitude. Bodily heat loss occurs through four mechanisms: conduction, convection, evaporation, and radiation. Conduction occurs through direct contact between surfaces. Heat is lost through the transfer to the cooler object. Water is an extremely effective heat conductor, and significant heat loss can occur when an athlete is wet, either when immersed in water or covered in sweat. Water immersion can increase cooling 100 times faster than air cooling at the same temperature.10 Other solids such as rock and metal also are effective conductors. Conductive losses can be significant in cases of hypothermia, where individuals may be sleeping directly on cold ground.11 Trapping layers of air with insulation is a way to minimize heat loss that can occur. Convection loss occurs through the transfer of heat from the body to air or water flowing around it. The amount of loss is proportional to the velocity of the air or water traveling over the body surface. In cold environments, convective losses are potentially much more significant when wind velocity is high. The term wind chill was developed in an attempt to express potential for convective heat loss from wind. Typically, convective and conductive losses account for only 10% to 15% of total heat loss.12 However, in certain sports like cycling, running, wind surfing, and swimming, these mechanisms of heat loss can play a larger role. Evaporative heat loss occurs through the evaporation of sweat from the skin as well as from evaporation of water from the respiratory tract. It accounts for 25% to 30% of total heat loss.12 Evaporation occurs more quickly in dry air. Cold conditions can thus allow more evaporative losses from exposed skin. At higher altitudes, respiratory losses increase with the increasing respiratory rate. These in­creased losses, when coupled with increased skin losses through sweat, can render an individual extremely vulnerable to hypothermia and other cold injury. Radiation accounts for the largest amount of heat loss by the body and refers to the emission of infrared energy from the body. About 60% of heat loss occurs through this mechanism.12 It is related to body mass and surface area and occurs more quickly in individuals with a greater surface area–to-mass ratio, such as children.13

499

CONTRIBUTING FACTORS TO THERMAL BALANCE Underlying Conditions Certain underlying medical conditions can interfere with maintenance of thermal balance. Hypothyroidism, hypoadrenalism, hypopituitarism, diabetes, and hypoglycemia can impair the body’s ability to produce heat.8,12,14 Because the hypothalamus is the central thermoregulatory center, any condition that affects it can alter heat balance. Elderly individuals are more at risk for altered heat balance, owing to impairment in heat generation.8,12,14,15 Some of the inability for elderly people to maintain heat balance may be explained by decreased adiposity with age.15 With increasing numbers of elderly individuals participating in sports and exercise activities, consideration of age should be taken into account for recommendations regarding activity in cold environment. Other conditions, such as multiple sclerosis and Parkinson’s disease, can centrally impair thermoregulatory function.8 Skin conditions can alter the skin’s natural ability to limit heat loss through its barrier function. Psoriasis, ichthyosis, sunburn, and exfoliative dermatitis all alter this barrier and allow increased loss of heat. Injuries to the skin, such as burns or open wounds, have the same effect.8

Drugs Drugs can adversely affect thermal balance through a variety of mechanisms. Ethanol is a peripheral vasodilator, which leads to increased heat loss.12 It is the most common drug associated with hypothermia.16,17 Phenothiazines suppress shivering thermogenesis, making individuals unable to counteract heat loss, leading to a negative heat balance.18 Medications that decrease centrally mediated thermoregulation also increase risk for cold injury. These include benzodiazepines, barbiturates, and tricyclic ­antidepressants.19 Sports medicine physicians must ask active patients about medications when anticipating activity in cold conditions. Adequate assessment of risk will assist in emphasizing other measures to help the athlete mitigate cold injury risk.

Clothing For clothing to prevent heat loss, it must act first and foremost as an effective insulator. Multiple layers of clothing allow layers of warm air to be trapped close to the body. The most ideal materials for inner layers of clothing in cold weather are those that wick moisture away from the skin, such as lightweight polyester and polypropylene. The presence of excessive moisture on the skin can lead to increased conductive heat losses. Outer layers ideally allow moisture to transfer to the air, where it can evaporate without significant heat loss.8 For middle layers of insulation, materials that can remain warm even when wet are ideal. Wool and polyester fleece are typical appropriate choices. Down insulation is most commonly used for extremely cold and dry conditions because of its superior insulating ability. However, when down becomes wet, it loses its insulating ability and is therefore not a good choice in moist ­conditions.

500

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Many individuals who exercise in cold weather “overdress” with too many layers for a given amount of activity, failing to anticipate the normal rise in heat production with increased physical activity. By dressing down and removing layers as heat production increases, the athlete can avoid excessive sweat production and absorption of moisture into insulation. Conversely, as the athlete decreases activity level, he or she must add layers to account for decreased heat production and increased heat loss. Keeping the head covered with a knit or fleece cap can markedly diminish heat loss. In subzero conditions, up to 50% of total resting heat production can be lost through the head.20 Mittens provide more protection than gloves, owing to their decreased surface area. Adding increased insulation to gloves is counterproductive because the addition further increases the surface area. Thick but loose socks are essential. Any covering for the ­ extremities that is too tight may constrict blood flow. Care must be taken that the feet do not become too warm. Increased warmth in thick socks may increase perspiration and facilitate increased heat loss. Socks should be changed frequently during prolonged exposures so that they stay dry.8

Acclimatization and Heredity Vasoconstriction of peripheral vessels occurs upon exposure to cold, especially in the hands. This response is modulated by a phenomenon called cold-induced vasodilation (CIVD).21 With CIVD, peripheral vessels briefly dilate. The presence of increased CIVD response correlates with decreased risk for frostbite. CIVD is more pronounced in certain ethnic groups, such as Eskimos and Sherpa. Humans can acclimatize to cold conditions, mainly manifested by blunting of physiologic responses to the cold. However, although heat acclimatization can occur relatively rapidly and markedly, acclimatization to cold is not nearly as quick or pronounced. These altered physiologic responses are not as effective at preventing cold injury.8

Types of Cold Injury Cold injury can manifest in a handful of ways and occurs when heat production is surpassed by heat loss. The generalized form of cold injury is hypothermia. Localized cold injury can be divided into freezing and nonfreezing types. The freezing type is referred to as frostbite, whereas nonfreezing injury can take a couple of different forms.

HYPOTHERMIA Hypothermia is defined as a core body temperature of less than 35°C. Functionally, it may be defined as a drop in body temperature that prevents the body from generating enough heat to maintain physiologic function.12 From 1999 to 2002, the Centers for Disease Control reported 2622 deaths due to exposure to natural cold. Nearly half of those reported deaths occurred in individuals 65 years and older.22 It can be assumed that a large number of these cases were of secondary hypothermia, in which hypothermia is due to the presence of a systemic disease that alters thermoregulation. Primary hypothermia occurs as a direct result of overwhelming environmental cold stress.14 The athlete

in a cold environment is more at risk for primary hypothermia. When all severities of hypothermia are included, mortality has been reported at 17%, with ­ mortality rate increasing proportionately with the severity.23

Symptoms and Classification The diagnosis of hypothermia requires accurate and reliable measurement of the core temperature. Rectal or esophageal probes that can read temperatures below 34° C are required to make the diagnosis. Oral or axillary probes do not reflect core temperature and often can be much lower. The amount of decrease in core temperature determines the severity of the hypothermia. Mortality is directly related to severity. Mild hypothermia corresponds to a temperature of 32° to 35° C. Vasoconstriction is maximal, causing the extremities to be cool and pale. Fine movements of the hands are difficult. Increased urinary frequency and volume, called cold diuresis, occurs at this stage. The patient is at maximal levels of involuntary thermogenesis with uncontrollable shivering. Mental status is altered and manifested by listlessness, confusion, and disorientation. Speech may be dysarthric. Victims are often ambulatory, but often ataxic. Mild degrees of tachycardia, tachypnea, and hypertension may be present, but vital signs are ­stable. In moderate hypothermia, core temperature is in the range of 28° to 32° C. The condition of the victim is more serious. He or she will exhibit a markedly altered mental status, with slurred speech, apathy, and amnesia. Deterio­ ration of mental status may continue to stupor and coma. At this stage, the phenomenon of paradoxical undressing may occur in the field. The individual will inexplicably disrobe despite the conditions and worsening physiologic state. Deep tendon reflexes diminish and become absent. Vital signs become unstable: hypotension follows with worsening dehydration, and bradycardia emerges. Atrial dysrhythmias, namely atrial fibrillation, begin to occur. Shivering slows and eventually ceases. Without the ability to generate heat, the body may quickly begin to cool to the ambient temperature. Without treatment, death will become inevitable. Severe hypothermia refers to temperatures less than 28°C. At this profound stage, the victim is often comatose and may appear dead. Pupils may be fixed and dilated. Hypotension and bradycardia are severe. In fact, blood pressure and pulse readings may be difficult to obtain. Arrhythmias, such as ventricular fibrillation and asystole, are common. Respiratory rate decreases to the point of apnea. Muscles become rigid. Urine output falls precipitously, and oliguria occurs. Individuals with severe hypothermia have the highest mortality rate.

Pathophysiology Neurologic As core temperature drops, nerve conduction becomes delayed. This conduction delay can cause many of the observed neurologic symptoms, such as mental status changes, ataxia, and muscle stiffness.11 Hypothermia causes overall decreased cerebral metabolism and global ­depression

Environmental Stress

of function. Clinically, the electroencephalogram becomes abnormal below 33.5°C and silent below 20°C.19

Cardiovascular Hypothermia exerts many effects on the heart. It alters electrical activity in the pacemaker cells and ­myocardium itself. One end result of these alterations is resulting abnormal repolarization.24 Electrical conduction is slowed, and the myocardium is irritable. The electrical abnormalities are clinically manifested by electrocardiogram (ECG) changes. One well-known ECG finding in hypothermic individuals with core temperature 33° C or lower is the J wave. J waves are positive deflections seen at the QRS and ST-segment junction. The J wave is present in about 80% of hypothermic individuals with temperatures below 33° C.11,24 It is considered nearly pathognomonic of hypothermia. Other ECG changes can become notable as hypothermia progresses, reflecting the effects on the conduction system. PR interval becomes prolonged initially, then is followed by prolongation of the QRS and then the QT interval.14

Renal In early stages of hypothermia, peripheral vasoconstriction causes a relative central hypervolemia and then produces diuresis. This phenomenon, along with a multifactorial cold diuresis, serves to deplete intravascular volume. Cardiac output then decreases, reducing renal blood flow up to 50% in moderate to severe hypothermia. Glomerular filtration rate decreases, which can then lead to acute renal failure.12,19

Treatment In general, treatment depends on the severity of the hypothermia and the treatment setting. Methods of rewarming may be categorized as follows: passive rewarming, active external rewarming, active internal rewarming, and extracorporeal rewarming.24 Passive rewarming consists of insulating the patient and allowing thermogenesis, specifically shivering, to increase core temperature. Examples of methods for active external rewarming include chemical heat packs, heat lamps, warming blankets, and hot baths. Active internal rewarming has been attempted using warmed air for inhalation, warmed intravenous fluids, and warmed fluids instilled into the gastrointestinal tract. Extracorporeal rewarming refers specifically to shunting blood from the body, usually by cardiopulmonary bypass or venoarterial or venovenous shunt, to a machine that rewarms the blood and returns it to the body. After hypothermia is confirmed with an accurate temperature (rectal is best), reduction of further heat loss becomes a priority. All wet clothes should be immediately replaced with dry garments. However, individuals with markedly altered mental status should not have clothes removed, but rather have warm blankets placed over them.11 Care should be observed when moving ­ moderately to severely hypothermic patients so as to not precipitate ventricular fibrillation that can result from myocardial instability. For mild ­ hypothermia, passive rewarming is usually adequate.

501

Passive rewarming has been successfully accomplished in individuals with core temperatures as low as 32°C. Individuals with moderate hypothermia (28° to 32°C) require active external warming. Care must be taken in instituting active rewarming to avoid thermal burns to the skin. Another hazard of active external rewarming is the phenomenon known as afterdrop, in which core temperature drops precipitously after rewarming is started because of shunting of cold blood away from the periphery to the core. Rewarming the trunk before rewarming the limbs in moderate hypothermia is recommended to avoid afterdrop. Severe hypothermia (<28° C) requires active rewarming in an intensive care setting. Methods of active internal rewarming are invasive to varying degrees, from instillation of warm water into the gut to peritoneal or thoracic lavage. Extracorporeal rewarming is highly invasive and should be carried out only in centers with experience in these techniques. Cardiac arrest can occur either as a direct result of the hypothermia or during the rewarming process. Defibrillation is recommended in these circumstances to convert to a pulse-generating rhythm. Moderate and severe forms of hypothermia may induce several metabolic changes that can affect cardiac and other organ function, including acidosis, hyperkalemia, and intravascular coagulation. Patients must be monitored carefully for these changes.

Prevention The following measures can prevent hypothermia in athletes exercising in the cold: • Cold-weather competitions should be highly organized with numerous marshals using well-developed communication networks to attend quickly to victims. • Athletes should check weather conditions ahead of time and dress appropriately. Wearing too many layers may cause excessive sweating and increase heat loss through evaporation. • Athletes undertaking long training routes in remote areas should train with a partner and carry a cell phone. The cell phone should not substitute for having a partner because cell phone failures have occurred and left individuals stranded. • The winter athlete must pay particular attention to energy stores because muscle glycogen is used at a faster rate owing to increased thermogenesis. Increased intake of carbohydrate is indicated. • The diuresis associated with cold exposure can lead to hypovolemia, which can decrease the athlete’s resistance to hypothermia. Attention to adequate hydration is important. • Athletes planning for winter competition should keep in mind that adaptation to the cold requires weeks of living in a cold environment. • The winter athlete should be made well aware of the warning signs of hypothermia, including excessive fatigue, listlessness, local pain, numbness, and excessive shivering. • Winter athletes who find themselves lost in the cold should remember that chances for survival are better by

502

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

taking temporary shelter and conserving heat than by continuing to trek in the cold, worsening fatigue, energy depletion, and exposure.

C

r i t i c a l

P

o i n t s

l Humans can avoid impairment in function only by maintaining a core temperature in the 4° range of normal. l Certain underlying medical conditions can interfere with maintenance of thermal balance, including hypothyroidism, hypoadrenalism, hypopituitarism, diabetes, hypoglycemia, advanced age, multiple sclerosis, and Parkinson’s disease, as well as injury to the skin. l Mulitple layers of clothing allow layers of warm air to be trapped close to the body. The ideal materials for the inner layer are those that wick moisture away from the skin. l Active external rewarming is necessary for moderate hypothermia. Rewarming the trunk before rewarming the limbs is recommended to avoid afterdrop.

S U G G E S T E D

R E A D I N G S

Auerbach PS: Wilderness Medicine, 5th ed. Philadelphia, Mosby, 2007. Castellani JW, Young AJ, Ducharme MB, et al: Prevention of cold injuries during exercise. Med Sci Sports Exercise 2013-2029, 2006. Department of the Army: Prevention and Management of Cold-Weather Injuries. Washington, DC, Department of the Army, Technical Bulletin Medicine #508, 2005. www.army.mill/usapa/med/DR_pubs/dr_a/pdf/tbmed508.pdf Exercise and cold weather: Stay motivated, fit, and safe. www.mayoclinic.com. Gonzalez RR, Sawka MN: Biophysics of heat transfer and clothing considerations. In Pandolf KB, Sawka MM, ������������������������������������������������ Monzales RR (����������������������������������� eds): Human Performance Physiology and Environmental Medicine at Terrestrial Extremes. Indianapolis, Benchmark, 1988, pp 45-95.

R eferences Please see www.expertconsult.com

S e c t i o n

C

Altitude Wendy J. McBride and Jorge E. Gómez The athlete training or competing at high altitude is presented with unique alterations in the environment, including lower oxygen tension, lower air resistance, greater exposure to ultraviolet radiation, and generally colder temperatures. Common questions that arise regarding exercise capacity at high altitude include: When should I arrive before a competition at altitude? Is it best to train at the same altitude at which I will compete? How long does it take to acclimate? To best answer these questions, it is important to set forth some basic definitions of altitude and discuss the physiologic changes that occur at altitude.

DEFINITIONS OF ALTITUDE Altitude can be divided into three main classifications: high, very high, and extreme. Most competitive athletics take place in the first category, high altitude. High altitude ranges from 1500 to 3500 m (5000 to 11,000 feet) above sea level. In 1974, the International Federation of Sports Medicine banned competition above 3050 m (10,000 feet). Moreover, the governing body of soccer, FIFA, has now banned any competition at an altitude greater than 2500 m (8200 feet). It is therefore remarkable that the number of endurance athletes (such as ice-climbers and heli-­skiers) who participate at very high altitude (3500 to 5800 m or

10,000 to 19,000 feet) has grown steadily over the past decade. Finally, the extreme altitude classification covers the range above 5800 m (19,000 feet). This altitude requires considerable acclimatization to survive and is often referred to as the “death zone.”1

PHYSICAL FEATURES OF A HIGH-ALTITUDE ENVIRONMENT The atmospheric changes that occur are similar in each zone of altitude. At high altitude, the barometric pressure is reduced, with a parallel decrease in the inspired partial pressure of oxygen (Pio2). Although the decline in ­barometric pressure with altitude is linear, there is a sharp decline in Pio2, and therefore arterial partial pressure of oxygen (Pao2), at about 2500 m (i.e., 8200 feet or roughly the elevation of Park City, Utah). By 18,000 feet, the atmospheric pressure is reduced to almost half of its sea-level value, and baseline Pio2 is only 75 mm Hg.2 Another physical feature of high altitude is dry air, which can increase the risk for dehydration. One can expect an increase in ultraviolet radiation exposure of about 4% for each 300 m above sea level.3 Finally, a decrease in air

Environmental Stress

density and air resistance can be a factor in many sporting events (e.g., more homeruns at Coors Field in Denver). Although it is important to take into account the physical features of high altitude before competition, it is imperative to understand the physiologic effects of altitude on the body.

PHYSIOLOGIC EFFECTS OF A HIGH-ALTITUDE ENVIRONMENT Altitude affects almost every major organ system of the body. Hypoxia, however, is the most prominent physiologic manifestation of high-altitude effects. To best understand the body’s response to hypoxia, it is helpful to consider the oxygen cascade, which describes the process through which oxygen must pass. First, oxygen passes from the environment (determined by the oxygen partial pressure, which varies with altitude) into the alveoli, which is a function of ventilation. Oxygen then traverses the pulmonary capillaries to the cardiovascular system. Oxygen presentation to the tissues is determined by both cardiac output and hemoglobin concentration. The degree of oxygen extraction is determined by the oxygen difference between blood and tissue as well as the affinity of hemoglobin for oxygen. The availability of oxygen to skeletal muscle is dependent on the biochemical state of the muscle as well as the number and functional capacity of mitochondria.4 It is helpful to think of physiologic adaptations in relation to the oxygen cascade because effects can be seen at each level. Some adaptations are immediate, whereas others require prolonged altitude exposure. Immediately on arrival to a high altitude, chemoreceptors in the brain sense the lower oxygen. This stimulates the hypoxic ventilatory response, which allows an increased workload despite decreased oxygen availability. Hyperventilation produces a respiratory alkalosis, which stimulates renal excretion of bicarbonate that equilibrates over the first week of a highaltitude experience. Moreover, the sympathetic nervous system is activated acutely, and epinephrine levels rise. An increased resting heart rate is therefore seen in an attempt to maintain cardiac output.2 Some would argue that the main mechanism for improved sea-level performance after high-altitude exposure is an increase in erythrocyte volume and red cell mass because this directly increases maximal oxygen consumption. This is known as the erythropoietic paradigm.5 The lower partial pressure of oxygen at high altitude induces erythropoietin production by the kidneys. Erythropoietin begins to increase within the first few hours and peaks at about 48 hours. Thereafter, it declines to near baseline after about 1 week. Red cell mass continues to increase slowly, but it may take up to 2 years to reach the levels of high-altitude natives, and the threshold altitude for sustained increase in blood erythropoietin concentration is 2200 m.5 To further the advantages of increased red cell mass, as oxygen carrying capacity of the blood is increased, the red blood cells also become more proficient at delivering oxygen to muscle cells. This is due to increased concentrations of 2,3-biphosphoglcyerate, which decreases the affinity of hemoglobin for oxygen.2

503

Although the benefits of increased red cell production are immediate on arrival at a high altitude, initially athletes are subject to dyspnea and a premature sensation of fatigue. This decrement in performance is secondary to increased lactate concentrations in the blood. However, with chronic exposure, the buffer capacity of skeletal muscle is increased. This process requires two separate events: increased renal bicarbonate excretion and an increased number of carbonic anhydrase isoforms in glycolytic muscle fibers. Therefore, peak lactate concentrations are lower in individuals who are acclimatized to high altitude; this condition is known as the lactate paradox.6 Improved uptake of oxygen by skeletal muscle is increased in long-term exposure to high altitude by ultrastructural changes in the muscle, specifically increased capillary density, mitochondrial number, and tissue myoglobin concentration. Although most of the metabolic changes associated with high-altitude exposure described earlier are complete by 3 to 4 weeks, the ultrastructural changes may take months to develop.4

COMPETITION AT HIGH ALTITUDE As previously mentioned, high altitude–induced hypoxia reduces the amount of oxygen available to do work. Maximal aerobic power (Vo2max) is reduced by 1% to 2% for every 100 m above 1500 m (or for every 390 feet above 4900 feet.). The high-altitude environment naturally affects aerobic and anaerobic athletes differently. For events lasting longer than 2 minutes, high altitude–induced hypoxia produces a reduction in aerobic power. For short sprints or other anaerobic events, most of the adenosine triphosphate required for muscular contraction is obtained from fast twitch muscle fibers using glycolytic metabolism, which does not require oxygen input. The effects of altitude on anaerobic versus aerobic performance were first fully appreciated during the 1968 Mexico City Olympics, the first Summer Olympic Games to be conducted at high altitude. Mexico City has an altitude of 2100 m (6889 feet). Multiple short-distance records were broken owing in part to decreased air resistance. In endurance events, however, times were 7% to 10% longer than previous competitions among the same athletes at sea level.1 To avoid the deleterious effects of high altitude, athletes often question the appropriate amount of acclimatization time. To allow an athlete to compete at maximal capacity, a period of 3 to 6 weeks has been recommended by multiple sources. Many coaches and trainers believe that if adequate time for acclimatization is not possible, competing immediately after arrival to a high altitude may be best. This hypothesis, however, has not been scientifically tested at this point. Because cardiac output does not begin to fall until 72 hours after arrival, this, in theory, appears a reasonable approach.5 Competition at high altitude may exert a psychological effect on performance. Although many athletes who work and train at high altitude consider it an advantage, those who come from sea level are often intimidated by high-altitude competition. Often referred to as the milehigh effect, athletes can be unnecessarily stressed about

504

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

high-altitude competition. According to Dr. Randall ­Wilber, chief physiologist at the United States Olympic Training Center, without proper psychological preparation, performance can be appreciably hindered.1

TRAINING AT HIGH ALTITUDE Since the 1940s, many anecdotal reports have touted the benefits of training at high altitude. With the underscored belief that “if you can run fast at altitude where there is little oxygen, you can then certainly run faster where there is lots of oxygen at sea level.” For this reason, the first of three training facilities for the United States Olympic Team was built in Colorado Springs in 1978 (elevation 6100 feet). Over the past 40 years, research has shown that training at high altitude can improve performance at high altitude and translate to increased endurance at sea level. There are potential detrimental effects of high altitude, however. During high-intensity interval workouts, running speed and lactate threshold are all lower at high altitude. Hypoxia can lower myocardial contractility and therefore cardiac output. For undetermined reasons, athletes can experience decreased immune function and sleep disturbances at high attitude. Even at moderate elevations, there is sleep disturbance and even periodic breathing in some susceptible athletes. Sleep is essential to muscle recovery.7 Will the increased number of mitochondria, myoglobin chains, and red cell mass associated with high-altitude training compensate long term for the decreased training load at high altitude? To help determine the answer, a novel approach was concocted in the 1990s that allows ­athletes the benefit of increased workload at sea-level conditions and the metabolic adaptations seen with highaltitude dwellers. This is known as the live-high, train-low approach. Because it is likely not feasible for an athlete to train physically at sea level during the day and then work and sleep at high altitude, multiple apparatus have been developed to stimulate a high-altitude environment: the Colorado Altitude Training Hatch, the Hypoxico Tent System, and most famously, the Nike Oregon Project. Athletes who use the hypoxic systems typically train at close to sea level and then reside in a hypoxic apartment for 8 to 18 hours a day. The hypoxic home can simulate an altitude environment similar to 2000 to 3000 m (6560 to 9840 feet). Several studies suggest there is an improvement in serum erythropoietin levels, reticulocyte count, and red blood cell mass using the live high, train low method.5 A simpler version of the live-high, train-low theory allows athletes to live at high altitude but to use supplemental oxygen while training. Oxygen tubing can ­obviously be cumbersome while training, but normoxic training facilities have been established. The benefits of high-low training in runners can persist for up to 3 weeks after altitude training.3 First introduced by Levine and colleagues in 1991, the live-high, train-low apparatus shows promise, but more standardized testing is required. It is clearly not feasible for the amateur or unsponsored athlete. Of particular importance, the World Anti-Doping Agency (WADA) issued a statement in 2007 stating that consideration is being made to consider the live-high, train-low apparatus illegal in amateur sports.8

ADVERSE EFFECTS OF HIGH ALTITUDE Many individuals, professional athletes included, can experience varied degrees of high-altitude illness. Rate of ascent is important in the development of these infirmities. Rapid ascent is an ascent of minutes to hours in duration, fast ascent takes place over days, and slow ascent requires weeks. Often, symptoms of high-altitude distress can be associated with more than one syndrome. Along the spectrum from least to most severe altitude syndromes, this section addresses acute mountain sickness (AMS), high-altitude pulmonary edema (HAPE), and high-altitude cerebral edema (HACE). AMS is characterized by headache, nausea and vomiting, shortness of breath, lethargy, sleeplessness, and anorexia. The illness is quite common. At elevations of 8000 feet in Colorado, 12% of newcomers were noted to be affected.9 Symptoms are most severe after about 48 to 72 hours and are gone by 4 to 5 days. Symptoms are usually mild and self-limited. Rapid ascent can produce persistent symptoms that may progress to more severe illness. AMS can ultimately be completely avoided if a gradual ascent policy is used. For those with mild symptoms, rest, hydration, and analgesics can be beneficial. Avoidance of alcohol is ideal. For those with persistent symptoms, the U.S. Food and Drug Administration has approved the use of acetazolamide for both prophylaxis and treatment of AMS. Acetazolamide is a carbonic anhydrase inhibitor (i.e., it prevents the combination of water and carbon dioxide to form bicarbonate in the blood). Respiratory rate is therefore increased, and one receives more oxygen without raising the blood pH. Adult dosing is 250 mg twice daily. Those with a known history of AMS should start prophylaxis 1 day before ascent. A study by Greene and colleagues in 1981 found that climbers of Mount Kilimanjaro were able to reach a higher altitude and experienced a lower symptom score when taking acetazolamide. These climbers took 500 mg nightly for 5 nights before ascent and nightly during the first stage of ascent.10 HAPE is a severe form of altitude illness with a 10% mortality rate. HAPE is most commonly seen in young men and is associated with a too-high, too-fast ascent. The most frequent symptoms are dyspnea and cough in combination with elevated heart rate and respiratory rate. Cyanosis can be present as well as pulmonary overload and edema. Pulmonary edema in this case is likely secondary to pulmonary hypertension. Pulmonary capillaries become leaky, but left ventricular function remains normal. Rales are heard on examination, and there is often production of pink, frothy sputum. The clinical course is similar to acute respiratory distress syndrome. Chest radiographs are diagnostic. Treatment is immediate descent and oxygen therapy. There is some evidence to support the use of diuretics; however, most patients are quite dehydrated by this point. Diuretic use, therefore, carries a significant risk for hypotension. Some institutions use morphine for treatment, as with pulmonary edema of cardiac origin. However, respiratory depression is frequent. Although most cases resolve within 24 to 48 hours from descent, 30% require intubation, and 10% ultimately die.11 HACE is defined by severe, often fatal neurologic sequelae of untreated AMS. It rarely occurs at less than

Environmental Stress

12,000 feet and takes 2 to 3 days to develop. Symptoms are severe headache, ataxia, and altered mental status. Focal neurologic signs, such as paralysis and seizures, may occur. HACE is associated with cerebral spinal pressures greater than 300 mm H2O. Edema is confirmed by computed tomography. Papilledema is commonplace on physical examination. As in HAPE, treatment is immediate descent with oxygen therapy. Upon arrival to a medical facility, intravenous steroids should be initiated. Intubation and controlled hyperventilation may be useful.

CURRENT RECOMMENDATIONS To achieve the best possible exercise capacity while reducing the potential for adverse side effects from altitude, we can make the following general recommendations: 1. Diet. Food intakes are typically reduced 10% to 50% during acute high-altitude exposure, depending on the individual and rate of ascent. However, because of metabolic stimulation, caloric needs are increased to avoid negative nitrogen balance and resultant loss of muscle mass. Protein demands are therefore increased long term. Athletes spending brief stays at high altitude, however, will benefit from a high-carbohydrate diet that should start before and continue during the initial 3- to 4-day critical period of acute high-altitude exposure. A study by Lawless and associates in 1999 demonstrated that carbohydrate consumption significantly increased oxygen tension and oxyhemoglobin saturation in arterial blood of subjects during simulated altitude exposure. Moreover, the energy production per liter of oxygen uptake is greater when carbohydrate is the energy source compared with fat.12 A low-salt diet and avoidance of alcohol allow better adaptation and can deter AMS. Supplementation of diet generally is not necessary. However, because of increased erythropoiesis in long-term high-altitude exposure, women and vegetarians may benefit from iron supplementation. 2. Fluids. Inappropriate thirst and appetite, together with increased insensible water loss, transient diuresis and increased energy expenditures can lead to rapid dehydration and glycogen depletion at high altitude. Furthermore, dehydration may exacerbate AMS. The Wilderness Medical Society recommends that athletes “consume a minimum of 3 to 4 liters of fluid per day containing 200 to 300 g of carbohydrate. This should prevent dehydration, improve energy balance, improve the oxygen delivery capability of the circulatory system, replenish muscle glycogen, and conserve body protein levels.”12 3. Workout intensity. As previously mentioned in the section on physiologic adaptations, workload is necessarily lower until adaptation can occur. Pushing workouts too hard and too fast may increase injury or overtraining potential. If a prolonged stay at high altitude is planned, it is recommended that a journal be kept of workout activities. Record morning resting heart rate and weight, then track the workout intensity and rate of fatigue. For short-term stays at altitude, many coaches and trainers advocate arrival 24 to 48 hours before an isolated athletic event. Again, for anaerobic events (e.g., sprints,

505

football games), there is less detriment to high-altitude participation. Psychological preparation should always be directed to minimize the mile-high effect. Sports psychologists routinely travel with professional teams and rely on mental imagery to support their athletes. Likewise, the use of portable oxygen tanks on the sideline is of little physiologic support during brief athletic events after rapid ascent.8 The psychological crutch may make it a worthwhile expense, however. C

r i t i c a l

P

o i n t s

l The main injuries that can occur at higher altitudes include hypoxia, radiation exposure, and dehydration. l Immediate physiologic adaptations to exercise and high altitude consist of increased red-cell mass via increased production of erythropoietin; increased concentrations of 2,3-diphosphoglycerate, which decreases the affinity of hemoglobin for oxygen; lower peak lactate concentrations as a result of increased renal bicarbonate excretion; and increased muscle carbonic anhydrase. l Long-term exposure to high altitudes results in further physiologic adaptations, including increased muscle capillary density, increased mitochondrial number, and increased tissue myoglobin. l Immediate exercise upon arrival to high altitude may result in hypoxia, leading to decreased myocardial contractility and decreased cardiac output, and sleep disturbance. Psychological effects of training at altitude may also hinder performance. l Best training benefits from altitude occur when athletes “live high, train low”. l Acute mountain sickness (AMS) is the most common form of altitude sickness. It consists primarily of shortness of breath, lethargy, sleeplessness, and anorexia. Symptoms are usually mild and self-limited. Abstaining from alcohol, getting plenty of rest, and avoiding dehydration can help prevent AMS. l High-altitude pulmonary edema (HAPE) and high-­altitude cerebral edema (HACE) are associated with rapid ascent and altitudes above 12,000 feet. Both these conditions require immediate recognition and intensive ­hospital care.

S U G G E S T E D

R E A D I N G S

Bartsch P, Saltin B: General introduction to altitude adaptation and mountain sickness. Scandi J Med Sci Sports 18(Suppl 1):1-10, 2008. Curtis R: Outdoor Action Guide to High Altitude: Acclimatization and Illnesses. http://www.princeton.edu/∼ oa/safety/altitude.html DeFranco MJ, Baker CL 3rd, DaSilva JJ, Piasecki DP, Bach BR Jr: Environmental issues for team physicians. American Journal of Sports Medicine 36(11): 2226-2237, 2008. Dumont L, Mardirosoff C, Tramer MR: Efficacy and harm of pharmacological prevention of acute mountain sickness: Quantitative systematic review. BMJ 321(7256):267-272, 2000. Stream JO, Grissom CK: Update on high-altitude pulmonary edema: Pathogenesis, prevention, and treatment. Wilderness & Environemental Medicine 19(4): 293-303, 2008.

R eferences Please see www.expertconsult.com

C H A P T E R  

12

The Team Physician: Preparticipation Examination, On-Field Emergencies, and Ethical and Legal Issues Thomas M. DeBerardino and Brett D. Owens The concept of the team physician in the United States began as a grass roots phenomenon in the 1950s with community physicians volunteering to care for the athletes participating in their local athletic programs. Team physicians began their association with collegiate athletic teams as early as the 1960s. Most college athletic teams had organized team physician coverage by 1980. Orthopaedic surgeons formed the bulk of the ranks of team physicians in the early days, especially at the high school level. Family practice physicians, however, quickly became integral parts of the athletic health team at all levels of competition. The health care team, consisting of physicians, physical therapists, and athletic trainers, has become the essential working body of current sophisticated team coverage programs throughout the country. The role of any team physician is one of cooperation and support of the athletic trainers. The trainers are most familiar with the pulse of their team because they are among the players on a daily basis.

THE TEAM PHYSICIAN Definition A physician of any specialty with an unrestricted medical license can serve as a team physician. This person is responsible for coordinating the medical care and ultimately providing treatment for athletic team members.

Roles and Functions of the Team Physician A good team physician must address the physical, emotional, and spiritual needs of the athletes. The job requires a comprehensive approach performed within the context of the sport and the needs of the team.1-4 Success of the team physician requires a broad medical knowledge base, not only of the musculoskeletal system but also of cardiorespiratory function, dermatology, neurology, pharmacology, and the physical demands of specific sports. The team physician interacts with the players, trainers, and coaching staff in various venues. Sideline coverage during actual competitions is the front line of team coverage, but the bulk of treatment and physician-player interaction likely occurs in the venue of the training room.1,5,6 It is important

to remember that the team trainer serves as the facilitator of all athlete–physician interactions. Evaluating athletes in their own environment (the training room) during regularly scheduled, prearranged times rather than in a formal clinic or office setting is both efficient and respectful of the limited time that the athletes have for anything other than schoolwork and team activities. Athletes are more relaxed in their familiar setting and generally provide a better history and a higher level of compliance with ­ recommendations given there. On the other hand, team physicians often make special accommodations in their office schedules to squeeze in athletes from their teams who have urgent problems. The team physician or other appropriate designee (practice partner, fellow, physician assistant, or qualified nurse) should always be available, if only by phone, to the athletic trainer, athletes, and coaching staff.

QUALIFYING ATHLETES TO PLAY Decisions about qualifications to play are often contentious issues that become more significant as the age of the athlete and level of competition increase. With youngsters in community leagues, a conservative approach is warranted. At the high school level, the opportunity to compete often influences scholarship opportunities. For a few performers at the highest level, a scholarship may mean an opportunity to attend National Collegiate Athletic Association (NCAA) Division I schools. For most of these high school athletes, however, athletic scholarships may represent their only opportunity to afford an education at Division II or III NCAA schools as well as universities, colleges, and junior colleges with many other affiliations. For these athletes, the ability to play well may truly be their only option for financing a college education. College-level participation carries an even larger set of potential economic and social rewards. Although only a very small percentage of college athletes advance to the professional level, many more gain employment and social opportunities from their achievements in athletics. Notwithstanding these significant forces, the team physician must constantly remember that he or she must always consider the health and welfare of the athletes as the most important variable when making any decision regarding an athlete’s ability to compete. Medical decisions are very ­difficult in this context. 507

508

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

PREPARTICIPATION EXAMINATION Gone are the days of simply listening to the heart and lungs and checking for a hernia. The preparticipation examination of today is much more sophisticated in its purpose and in the global nature of the evaluation. The age of the athletes, the level of competition, and the nature of the sport dictate the comprehensiveness of the evaluation. At the junior high school level, this examination may require only a general health evaluation to rule out conditions that contraindicate participation in certain sports. The major challenges at this level are the wide range of musculoskeletal maturation that exists in peripubertal athletes and the possibility of discovering previously unidentified congenital defects that could pose serious health threats.7-11 By the time athletes reach high school and college level participation, the likelihood of finding serious congenital problems is diminished. Musculoskeletal injuries and potential prior orthopaedic surgery are common findings at this level of competition; therefore, a thorough musculoskeletal examination is paramount. The preparticipation examination can serve a multitude of functions. Besides providing an evaluation of the general health of the athlete, the examination may assess the level of aerobic conditioning. The physical maturity level of the younger athletes can also be assessed. Specific conditions that may limit or prevent participation in specific sports may also be uncovered. In 1997, the American Academy of Family Physicians, the American Academy of Pediatrics, the American Medical Society for Sports Medicine, the American Osteopathic Academy of Sports Medicine, and the American Orthopaedic Society for Sports Medicine endorsed and published a joint policy statement, guide, and forms for use in the preparticipation physical evaluation.12 The primary goal of any preparticipation physical examination is to help maintain the health and safety of the athlete. Table 12-1 delineates the basic objectives of the preparticipation examination.

Timing Ideally, the preparticipation examination should be scheduled several weeks before the sport season begins. This allows most deficiencies uncovered during the examination to be adequately addressed and corrected.

Frequency Despite the establishment of basic frequency guidelines adopted by the NCAA, many state and local governing authorities have yet to fully incorporate these guidelines into their own standard operating procedures for the administration of preparticipation examinations. The NCAA guidelines stipulate an initial full screening examination upon entry into the collegiate level of athletics. Interim health status questionnaires should be completed in subsequent years. The questionnaires provide guidance for a limited physical examination that is focused on any new areas of injury that have changed since the prior examination. To ensure compliance with any variations in local examination requirements, physicians are encouraged

   TABLE 12-1  Preparticipation Examination Objectives Objective Detect life-threatening or disabling ­conditions Detect conditions that might limit ­competition Detect conditions that might ­predispose the athlete to injury Meet certain legal and insurance ­requirements Determine general health status Undertake physician wellness ­counseling Assess physical maturity Evaluate fitness

Essential

Preferred

X X X X X X X X

to check with local authorities that oversee and sponsor specific sports in their area. The American Heart Association (AHA) recommends both a history and physical examination before participation in collegiate sports and that an interim history and blood pressure measurement be obtained in each of the subsequent 3 to 4 years after the initial comprehensive preparticipation examination. The AHA recommends repeat screening every 2 years and an interim history review in intervening years for high school athletes.13 Complete examination and possibly further testing are indicated if significant abnormalities or changes are detected during interim screens.

Logistics Examinations are often carried out in a large clinic space or training room that allows for the establishment of multiple examination stations. The athletes rotate through the various stations in order to complete the examination process. The alternative format is to perform the examinations in a standard individual or office-based setting. Table 12-2 lists the advantages of each examination format. A critical aspect of any station-based examination setup is to have an experienced head team physician available as the final station to review all of the data to determine ­clearance, and to make appropriate participation recommendations.

Station-Based Examination Setup Tips A variable number of stations can be used depending on the space and manpower available. A typical setup might include the stations shown in Table 12-3. Stations 1 through 8 are required. Stations 9 and 10, in addition to stations addressing nutrition, strength, speed, agility, power, endurance, and balance, are optional.12 Station 1 is a crucial administrative desk necessary to maintain an orderly flow of the athletes and provide accountability. Portions of the forms can and should be completed (by the athletes or by the parents of younger athletes) before the actual examination. The written health history is doublechecked by a health care professional at station 2. ­Stations 3, 4, and 5 are often grouped together as the athletes await an opening in the all-important and often backed-up ­station 6 (medical examination station). Station 6 is often

The Team Physician

History

   TABLE 12-2  Comparison of Office-Based  

and ­Station-Based Preparticipation Examinations Examination Advantages

Station-Based

Cost savings Time savings Physician workload savings Medical records availability Privacy Comprehensive health care

Office-Based

XX XX XX

XX XX XX

s­ ubdivided into multiple stations if sufficient manpower is on hand to help alleviate the potential congestion at this station. Only a portion of the medical examination is done at each of the substations. Station 7, the orthopaedic examination, may take various forms. If physicians without sports medicine training are performing the examination, or if time is a factor, a quick 13-point general orthopaedic screening examination may be used.12,14,15 If the evaluation is performed by orthopaedic surgeons or other sports physicians, they may wish to use joint-specific testing to go into much more detail, especially regarding the examination of ankles, knees, elbows, and shoulders. The team physician should man station 8, where all the collected data generated at the previous stations is reviewed and a decision is rendered as to whether the athlete may safely participate. The team physician is thereby responsible for the final clearance section on the preparticipation evaluation form. This involves not only clearing an athlete for play but also making recommendations for appropriate follow-up of any concerns that surfaced during the examination process. The team physician (and the ­physician’s support staff) is responsible for facilitating communication between other primary or consulting physicians, athletic trainers, coaches, and parents. This process can be enhanced by carefully documenting problems and specific recommendations in the clearance section of the preparticipation evaluation form.    TABLE 12-3  Stations for Preparticipation Examination Station Number

Function

Required Personnel

1 2

Sign-in Review history

3

Height and weight Visual acuity Vital signs

Team support staff Physician, physician ­assistant (PA), nurse, or senior athletic trainer Support staff or other volunteer Athletic trainer or nurse Athletic trainer, PA, or nurse Physician(s), PAs, nurse practitioners Physician(s), PAs, nurse practitioners Team physician

4 5 6 7 8 9 10

Medical examination Orthopaedic ­examination Review— reassessment Body composition Flexibility

509

Exercise physiologist, athletic trainer, or physical therapist Physical therapist or athletic trainer

A comprehensive history and physical examination make up the bulk of the preparticipation examination. A ­careful history is the most important part of the preparticipation evaluation. Special emphasis in the history should be placed on screening for recent illnesses, allergic reactions, cardiac or pulmonary problems, musculoskeletal problems, skin problems, previous head injuries and other neurologic problems, previous heat illness, medication problems, immunizations, and menstruation abnormalities in the female athlete. The specific set of questions found on the recommended preparticipation evaluation form is aimed at detecting these potential problems, and the ­questions should be reviewed carefully with each athlete (Fig. 12-1).12 Sudden death is one of the greatest tragedies in sport, and it deserves special mention. More than 95% of sudden deaths in athletes younger than 30 years of age are ­secondary to cardiovascular problems, the most common problem being hypertrophic cardiomyopathy. There are currently no universally accepted screening standards for high school and college athletes.13 Although preparticipation screening may be of limited value in detecting cardiovascular causes of sudden death, the AHA recommends that “some form of preparticipation cardiovascular screening for high school and collegiate athletes is justifiable and ­compelling, based on ethical, legal, and medical grounds.”13,16 Specific historical questions and examination components recommended by the AHA are listed in Table 12-4. A comprehensive cardiovascular screen is recommended for each athlete. If abnormalities are detected or suspected, a referral is made for further cardiac evaluation.

Physical Examination The physical examination focuses on the areas of greatest importance in sports participation, and it specifically addresses problems uncovered by the history. Standard components of the preparticipation examination can be found in Box 12-1.12

Height and Weight Height and weight should be recorded with each physical examination. Excessive weight change should alert physicians to possible underlying problems such as eating disorders and steroid use.

Head, Eyes, Ear, Nose, and Throat The eyes are the most important component of this part of the examination. Visual acuity testing using a Snellen eye chart, or equivalent automated devices, should be performed on all athletes. Athletes requiring eye protection before participation include those with a history of significant eye injury or surgery, absence of one eye, or best corrected vision less than 20/40.12 Protective eye wear should be evaluated, and the risks of participation in sports with glasses or contacts should be addressed. Anisocoria (unequal pupils) should be noted in the chart. Although pupil inequality may be physiologic, it is important to know the athlete’s baseline pupil size when assessing later head injuries.

510

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 12-1   Preparticipation physical evaluation. (From American Academy of Family Physicians, American Academy of Pediatrics, American Medical Society for Sports Medicine, American Orthopaedic Society for Sports Medicine, American Osteopathic Society for Sports Medicine: Preparticipation Physical Evaluation, 2nd ed [monograph]�� �������������. Physician ��������������������� Sportsmed, 1997.) ������

The Team Physician

Figure 12-1,  cont’d

511

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

512

   TABLE 12-4  American Heart Association

­ ecommendations for Cardiovascular Screening   R in the ­Preparticipation Examination History

Physical Examination

Sudden death in family member < 50 yr Prior heart disease in family Exertional dyspnea or chest pain Syncope Excessive fatigability Heart murmur Systemic hypertension Parental verification of history

Blood pressure Heart murmur* Peripheral (femoral) pulses Marfan stigmata†

*Precordial auscultation is recommended in both supine-sitting and standing ­positions to identify heart murmurs reflecting hypertrophic cardio­ myopathy. †Tall stature, arachnodactyly, kyphoscoliosis, anterior chest deformity, arm span greater than height, decreased upper body length–to–lower body length ratio, heart murmur or midsystolic click, ectopic lens, family history of Marfan syndrome. From American Heart Association: Cardiovascular preparticipation screening of competitive athletes. Circulation 94:850-856, 1996.

be further evaluated with electrocardiography or referred for further evaluation by a cardiac specialist.

Pulmonary Basic auscultation of the lungs should detect abnormal breath sounds, wheezes, crackles, or rubs. Any history of asthma or other airway disease is cause for physicians maintaining a high index of suspicion when evaluating an athlete for exercise-induced bronchospasm.17

Abdominal and Gastrointestinal Examine the athlete in a supine position. Abdominal tenderness, rigidity, mass, or organomegaly (particularly liver and spleen) requires further evaluation before sports participation. Pain or enlargement in the hypogastric area or pelvis in the female athlete should raise suspicion of pregnancy or other gynecologic problems and may warrant a pelvic examination in the appropriate setting.

Genitourinary Cardiovascular The AHA recommends checking blood pressure, auscultating for murmurs, palpating peripheral pulses (femoral), and assessing for stigmata of Marfan syndrome with each examination.13 Blood pressure should be checked with a proper-sized cuff and, if elevated, should be rechecked on at least three occasions. The heart should be auscultated with the athlete in both supine and standing positions.13 The murmur of hypertrophic cardiomyopathy, the most common cause of sudden cardiac death in athletes younger than 30 years, can worsen in the standing position or with a Valsalva maneuver. However, many patients with hypertrophic cardiomyopathy may exhibit no murmur with auscultation. Overall, heart murmurs requiring further evaluation include any systolic murmur grade 3/6 or more in severity, any diastolic murmur, and any murmur that gets louder with a Valsalva maneuver.12 Abnormal heart rhythms may

A thorough male testicular examination should be performed, and physicians can use the opportunity to counsel athletes about future self-examination to screen for testicular cancer. Undescended testes, masses, absence of one testicle, and inguinal hernias can be detected with a good examination. A female genitourinary examination is not part of the routine preparticipation examination. If warranted by history or other examination findings, it should be conducted in a more private setting.

Musculoskeletal Three methods used to evaluate the musculoskeletal system include a general screening examination, joint-specific testing, and a sports-specific examination. The method used is highly dependent on the given setting, time constraints, physician comfort and skill level, history of musculoskeletal injury or symptoms, and specific sport.

Box 12-1 Standard Components of the Preparticipation Examination

• Height • Weight • Eyes: visual acuity, pupil size • Oral cavity • Ears • Nose • Lungs • Cardiovascular system: blood pressure, pulses (femoral, radial), heart (rate, rhythm, murmurs) Abdomen: masses, ­tenderness, organomegaly • �������������������������������������������� • Genitalia (males only): single or undescended testicle, testicular mass, hernia • Skin: rashes, lesions (infectious) • Musculoskeletal system: contour, range of motion, and symmetry of neck, back, shoulder/arm, elbow/forearm, wrist/hand, hip/thigh, knee, leg/ankle, foot

Adapted from American Academy of Family Physicians, American Academy of Pediatrics, American Medical Society for Sports Medicine, American Orthopaedic Society for Sports Medicine, American Osteopathic Society for Sports Medicine: Preparticipation Physician Evaluation, 2nd ed (monograph). Physician Sportsmed, 1997.

The Team Physician

It is reasonable to use a general orthopaedic screen for asymptomatic athletes with no previous musculoskeletal injuries.18,19 Fields and Delaney recommend asking if the athlete has ever broken a bone, had to wear a cast, or had an injury to any joint. A positive answer to the question places the athlete in a special risk group that requires further examination.20 The entire joint-specific examination is more time­consuming than a general screen, but some sports physicians prefer this method because of its accuracy. A reasonable approach involves supplementing a general screen with focused joint-specific testing of problematic areas. Joint-specific screening involves inspection and range of motion testing of the spine and upper and lower extremities. Strength and stability testing of the shoulders, elbows, knees, and ankles is performed. Specific or specialized tests are applied as needed to potentially problematic joints. Independent of the examination method used, all positive findings should be well documented, and athletes should be directed to proper evaluation, treatment, and rehabilitation well before the season begins.

Neurologic Gross motor neurologic function is generally screened through the musculoskeletal examination. Unexplained strength deficiencies, paresthesias, history of recurrent stingers or burners, history of head injury or concussion, or other focal or generalized neurologic deficits warrant a more comprehensive neurologic examination.

Skin Athletes should be screened for rashes, infections, infestations, and other suspicious lesions. Transmissible infections and infestations should be adequately treated before athletes are allowed to participate in sports in which other athletes may be at risk.

Clearance for Participation Determining clearance is the most challenging and important decision in the preparticipation examination. An experienced physician familiar with the demands of specific sports best determines clearance for participation. In the station-based setting, this takes place at station 8, or the last medical station. Important information can be found in the answers to questions on the health history form about medications and the potential presence of cardiac disease. Questions about passing out, becoming dizzy, or experiencing shortness of breath during exercise, as well as a family history of premature heart disease or sudden death, may alert the physician to the possibility of potentially lethal cardiac lesions. Elevated blood pressure is the most frequent abnormality found in the physical examination. If the blood pressure falls outside the accepted age group limits, the measurement should be repeated, taking care to use a large enough cuff for the athlete’s arm size. Mild to moderate hypertension, not associated with underlying heart disease or target organ damage, should not limit participation. Blood

513

­ ressure should be monitored frequently to assess the effect p of exercise on the pressure over time. Severely hypertensive athletes should be restricted until their blood pressure is under good control.21 Disturbances of cardiac rhythm or nonphysiologic cardiac murmurs deserve further investigation before the athlete is approved for competition. A history of head injury or concussion may present team physicians, athletes, and parents with challenging decisions. Much controversy exists regarding the classification of concussion and return-to-play guidelines. Although current concussion grading systems and clearance lack a scientific basis, frequent usage dictates that team physicians should be familiar with them.22-28 The guidelines, however, should never replace the complex individualized clinical assessment of each athlete. Overall, an athlete who has been without symptoms for a period of time may be cleared for participation in all sports.12 Athletes with a history of burners or stingers may be cleared for all sports if they are without symptoms and have a normal physical examination.12 Recommendations involving convulsive disorders are reviewed in Table 12-5. Skin infections demand appropriate treatment before the athlete is allowed to participate in sports involving body contact, either with another competitor or with mats, as in wrestling. Liver, spleen, or kidney enlargement will be a limiting factor in contact and collision sports, and underlying causes should be determined. Exercise-induced bronchospasm is frequently overlooked. A decision should be made whether it affects clearance of the athlete. In all cases, it should be recognized and appropriately treated. The presence of a hernia by itself is not necessarily a disqualifying factor but requires further discussion with the athlete and parents. The loss of any one of a pair of organs (eye, kidney, testicle) has traditionally been a disqualifying feature for contact or collision sports or other activities in which the second of the paired organs may be in jeopardy. This traditional conservatism, however, has been challenged repeatedly in courts of law and has always been decided in favor of the athlete who wishes to participate. In this case, it is recommended that a formal document be drafted by the school, or its lawyer, to be signed by the athlete and parents or guardians indicating that they are completely informed about the specific potential consequences of participation. Other specific issues involving substance and supplement abuse, history of heat illness, blood-borne pathogens, and recent acute illness should be evaluated before determining clearance. Amenorrhea, disordered eating, and possible pregnancy should be assessed with female athletes. Evidence of previous orthopaedic injuries is increasingly common with the advancing age of the screened athletes. The most important recommendation stemming from the orthopaedic examination is often that adequate rehabilitation of previous orthopaedic injuries be accomplished before participation. These recommendations are provided to the athletic trainer, school nurse, or coaching staff so that all involved parties are working toward a common goal.

514

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 12-5  Medical Conditions and Sports Participation This table is designed to be understood by medical and nonmedical personnel. In the “Explanation” sections below, “needs evaluation” means that a physician with appropriate knowledge and experience should assess the safety of a given sport for an athlete with the listed medical condition. Unless otherwise noted, this is because of the variability of the severity of the disease or the risk for injury among the specific sports, or both. Condition

May Participate

Atlantoaxial instability* (instability of the joint between cervical vertebrae 1 and 2) Explanation: Athlete needs evaluation to assess risk for spinal cord injury during sports participation. Bleeding disorder* Explanation: Athlete needs evaluation. Cardiovascular diseases Carditis (inflammation of the heart) Explanation: Carditis may result in sudden death with exertion. Hypertension (high blood pressure) Explanation: Those with significant essential unexplained hypertension should avoid weightlifting and powerlifting, body building, and strength training. Those with secondary hypertension (hypertension caused by a previously identified disease) or severe essential hypertension need evaluation. Congenital heart disease (structural heart defects present at birth) Explanation: Those with mild forms may participate fully; those with moderate or severe forms, or who have undergone surgery, need evaluation.† Dysrhythmia (irregular heart rhythm) Explanation: Athlete needs evaluation because some types require therapy or make certain sports dangerous, or both. Mitral value prolapse (abnormal heart valve) Explanation: Those with symptoms (chest pain, symptoms of possible dysrhythmia) or evidence of mitral regurgitation ­(leaking) on physical examination need evaluation. All others may participate fully. Heart murmur Explanation: If the murmur is innocent (does not indicate heart disease), full participation is permitted. Otherwise, the athlete needs evaluation (see “Congenital heart disease” and “Mitral valve prolapse” above). Cerebral palsy* Explanation: Athlete needs evaluation. Diabetes mellitus*‡ Explanation: All sports can be played with proper attention to diet, hydration, and insulin therapy. Particular attention is needed for activities that last 30 minutes or longer. Diarrhea§ Explanation: Unless disease is mild, no participation is permitted because diarrhea may increase the risk for dehydration and heart illness. See “Fever” below. Eating disorders Anorexia nervosa, bulimia nervosa Explanation: These patients need both medical and psychiatric assessment before participation. Eyes Functionally one-eyed athlete, loss of an eye, detached retina, previous eye surgery, or serious eye injury Explanation: A functionally one-eyed athlete has a best corrected visual acuity of < 20/40 in the worse eye. These athletes would suffer significant disability if the better eye were seriously injured, as would those with loss of an eye. Some athletes who have previously undergone eye surgery or had a serious eye injury may have an increased risk for injury because of weakened eye tissue. Availability of eye guards approved by the American Society for Testing Materials (ASTM) and other protective equipment may allow participation in most sports, but this must be judged on an individual basis. Fever§ Explanation: Fever can increase cardiopulmonary effort, reduce maximum exercise capacity, make heat illness more likely, and increase orthostatic hypotension during exercise. Fever may rarely accompany myocarditis or other infections that may make exercise dangerous. Heat illness, history of Explanation: Because of the increased likelihood of recurrence, the athlete needs individual assessment to determine the ­presence of predisposing conditions and to arrange a prevention strategy. HIV infection§ Explanation: Because of the apparent minimal risk to others, all sports may be played that the state of health allows. In all athletes, skin lesions should be properly covered, and athletic personnel should use universal precautions when handling blood or body fluids with visible blood. Kidney, absence of one Explanation: Athlete needs individual assessment for contact/collision and limited contact sports. Liver, enlarged Explanation: If the liver is acutely enlarged, participation should be avoided because of risk of rupture. If the liver is ­chronically enlarged, individual assessment is needed before contact/collision or limited contact sports are played. Malignancy* Explanation: Athlete needs individual assessment. Musculoskeletal disorders Explanation: Athlete needs individual assessment. Neurologic History of serious head or spine trauma, severe or repeated concussions, or craniotomy Explanation: Athlete needs individual assessment for contact/collision or limited contact sports, and also for noncontact sports if there are deficits in judgment or cognition. Recent research supports a conservative approach to management of concussion.

Qualified Yes Qualified Yes No Qualified Yes

Qualified Yes Qualified Yes Qualified Yes Qualified Yes Qualified Yes Yes Qualified Yes Qualified Yes Qualified Yes

No

Qualified Yes Yes

Qualified Yes Qualified Yes Qualified Yes Qualified Yes Qualified Yes

The Team Physician

515

   TABLE 12-5  Medical Conditions and Sports Participation������� —cont’d Condition

May Participate

Neurologic—cont’d Convulsive disorder, well controlled Explanation: Risk for convulsion during participation is minimal. Convulsive disorder, poorly controlled Explanation: Athlete needs individual assessment for contact/collision or limited contact sports. Avoid the following ­noncontact sports: archery, riflery, swimming, weightlifting or powerlifting, strength training, or sports involving heights. In these sports, occurrence of a convulsion may be a risk to self or others. Obesity Explanation: Because of the risk for heat illness, obese persons need careful acclimatization and hydration. Organ transplant recipient* Explanation: Athlete needs individual assessment. Ovary: absence of one Explanation: Risk for severe injury to the remaining ovary is minimal. Respiratory Pulmonary compromise, including cystic fibrosis* Explanation: Athlete needs individual assessment, but generally all sports may be played if oxygenation remains satisfactory during a graded exercise test. Patients with cystic fibrosis need acclimatization and good hydration to reduce the risk for heat illness. Asthma Explanation: Upper respiratory obstruction may affect pulmonary function. Athlete needs individual assessment for all but mild disease. See “Fever” above. Acute upper respiratory infection Explanation: Upper respiratory obstruction may affect pulmonary function. Athlete needs individual assessment for all but mild disease. See “Fever” above. Sickle cell disease Explanation: Athlete needs individual assessment. In general, if status of the illness permits all but high exertion, contact/­ collision sports may be played. Overheating, dehydration, and chilling must be avoided. Sickle cell trait Explanation: It is unlikely that individuals with sickle cell trait (AS) have an increased risk for sudden death or other medical problems during athletic participation except under the most extreme conditions of heat, humidity, and possibly increased altitude. These individuals, like all athletes, should be carefully conditioned, acclimatized, and hydrated to reduce any ­possible risk. Skin Boils, herpes simplex, impetigo, scabies, molluscum contagiosum Explanation: While the patient is contagious, participation in gymnastics with mats, martial arts, wrestling, or other contact/collision or limited contact sports is not allowed. Herpes simplex virus probably is not transmitted via mats. Spleen, enlarged§ Explanation: Patients with an acutely enlarged spleen should avoid all sports because of risk for rupture. Those with a ­chronically enlarged spleen need individual assessment before playing contact/collision or limited contact sports. Testicle, absent or undescended Explanation: Certain sports may require a protective cup.

Yes Qualified Yes

Qualified Yes Qualified Yes Yes Qualified Yes

Yes Qualified Yes Qualified Yes Yes

Qualified Yes

Qualified Yes Yes

*Not discussed in text of the monograph. †Mild, moderate, and severe congenital heart disease are defined in 26th Bethesda Conference: Recommendations for determining eligibility for competition in athletes with cardiovascular abnormalities, January 6-7, 1994. Med Sci Sports Exerc 26(Suppl 10):246-253, 1994. ‡Well controlled. §AAP recommendation as indicated; see text for qualifications by other commentators. Reprinted with permission from American Academy of Family Physicians, American Academy of Pediatrics (AAP), American Medical Society for Sports Medicine, American Orthopaedic Society for Sports Medicine, American Osteopathic Academy of Sports Medicine: Preparticipation Physical Evaluation, 2nd ed (monograph). Physician Sportsmed, 1997; and American Academy of Pediatrics Committee of Sports Medicine and Fitness: Medical conditions affecting sports participation. ­Pediatrics 94:257-760, 1994.

Essentially three possible phrases can form the final decision regarding an athlete’s ability to participate: 1. Participate without restriction 2. Participate only after completing further evaluations or rehabilitation 3. Not participate in specific sports owing to certain disqualifying factors Scenarios whereby a team physician performs the preparticipation examination but has minimal control over the evaluation of specific problems uncovered in the evaluation due to various potential insurance coverage plans and rules are more commonplace in the age of managed care. Clear communication is essential in these situations, and there is

space near the end of the ­ recommended ­ preparticipation examination form (see Fig. 12-1) to write specific recommendations for further evaluation before clearance. In addition to good communication between physicians, the team physician should make sure that athletes, parents, trainers, and coaches fully understand any restrictions and need for further evaluation. A free flow of information occurs in many sports settings, and team physicians need to respect the confidentiality of the athlete. If an ­ athlete wishes to participate in a specific sport against medical ­recommendations, or the athlete has been cleared by a physician in disagreement with the team physician, it may be a good idea to have the athlete and his or her parents sign a waiver stating that they are fully aware of the risks of participation and that they assume those risks.

516

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

MEDICAL SUPERVISION OF ATHLETES The traditional function of a team physician has been medical supervision of the athlete. The team physician (when present) is responsible for medical coverage at the various athletic venues. Obviously, the presence of a physician is often limited to high-risk situations and high-risk sports. When a physician is not available for coverage, certified athletic trainers or other personnel trained in the prevention of injuries and in early evaluation and care of injured athletes should be present. When physicians and athletic trainers are not available at high school or higher levels of competitions, coaches should be certified in American Red Cross advanced first aid and cardiopulmonary resuscitation (CPR), or the equivalent. The team physician often treats psychological conditions. A team physician must be willing and able to counsel athletes and to refer them to consultants when necessary. Problems with alcohol and drug abuse, performance anxiety, failed emotional relationships, unwanted pregnancy, eating disorders, and deaths in the family are all common. The death of a team member, particularly as a result of competition, calls for the greatest degree of sensitivity and skill in dealing with the grieving process and with media attention and in helping the team back into competition.29 The team physician should be able and willing to evaluate the circumstances and either treat or refer to an appropriate specialist. The team physician is frequently asked for advice on the proper conditioning techniques to be used to prevent or rehabilitate injuries. The requests may be for both preseason and in-season conditioning as well as for general and sport-specific techniques. The advice may relate to endurance, strength, flexibility, agility, or nutrition. It often concerns issues of protective equipment. It is important to understand the scientific merits and liabilities of the equipment worn by athletes for practice and play. Optimally, the physician understands the considerations involved in the selection and fit of the equipment and in the prevention of injuries and reinjuries. Supervision at tournaments often involves coordination of a larger medical team to ensure that skilled care is available at all times and venues. Most injuries require significant skill in musculoskeletal evaluation. Nonmusculoskeletal injuries certainly occur as well. For example, head and facial injuries, eye trauma, heat illness, and abdominal injuries may also test the physician’s evaluating and decision-making abilities. For athletes with traumatic conditions, many of the more difficult decisions concern return-to-participation issues. Should the athlete be allowed back during the same contest? Should he or she be held out of competition? If so, for how long? What are the parameters for allowing return to play? Minimal evidencebased data exist regarding return-to-play issues. Should a team physician obey the basic science principles of soft tissue and bony healing, or should he or she allow an athlete to return to play sooner because functionally the athlete can? It is important that the team physician thoroughly explain the known risks and complications of an injury to athletes and, if needed, to parents and coaches. In addition to his or her medical opinion, the physician allows the athlete to make an informed decision about returning to play.

Team physicians must be careful not to influence athletes with must-play philosophies, and they must avoid making poor medical decisions motivated by pressures from nonmedical higher powers. This must be balanced by gentle reassurance of frightened athletes with obviously stable injuries that are clearly safe for full play. The team physician should develop a prearranged system of emergency care for each sporting event covered. Team physicians, together with the team trainers and coaching staff, should estimate the breadth and scope of medical coverage required in different competitions.30 Team physicians should ideally have advanced cardiac life support certification. It is the physician’s responsibility to make sure general medical supplies and emergency equipment are readily available at the athletic venue and in the training room. The assumption that the school’s, team’s, or paramedic’s equipment and supplies will cover all medical needs can be tragically misguided.31 At most high-risk and collision sporting events, an ambulance should be present to facilitate rapid emergency transportation of the injured athlete. If a patient needs to be transferred for medical care, a system should be in place to notify the hospital in advance while the patient is en route. A well-organized team physician and team trainer staff know needed phone numbers, appropriate tertiary care centers, and who will take charge in a catastrophic event. Drills should be conducted periodically to evaluate availability and access to proper medical equipment, to become familiar with the scope of the local emergency medical system, and to determine proper routes for mobilization of the emergency medical system.32

ON-FIELD EMERGENCIES The Physician’s Medical Bag Part of the team physician’s responsibility is to plan for all types of emergencies that may occur during athletic events. A well-organized, well-equipped medical bag is necessary for emergency care and care during travel. The term ­medical bag is a euphemism in sports medicine. It may refer to one or more bags, boxes, trunks, or duffels, depending on the sport, the viewers, the extent of a trip, and, most of all, equipment being provided by accompanying athletic trainers and equipment managers or by the venues being visited. Estimation of the breadth of medical coverage in the different athletic settings is helpful when putting together a medical bag.30 There is no perfect list of ideal medications or equipment, and individual preferences and circumstances will affect the contents of the medical bag. In addition, the training and comfort of each individual physician with emergency procedures and medications will determine how the physician is equipped.33 Knowledge of the risks and injuries associated with each particular sport is helpful when planning a medical bag. The need for specific medications and equipment can be dependent on the length of the journey, the destination, the age and condition of the participants, and the number of participants and spectators.30,31 Environmental conditions such as extreme temperatures or humidity should be taken into consideration. Specific medical needs of the ­ athletes and coaches can be established during the preseason while

The Team Physician

517

conducting preparticipation examinations and will help the team physician meet any special needs. Clear communication must take place with all athletic training and potential medical resources before an event. The team physician can contact the medical director of the local emergency medical system to determine its ­capability.34 Paramedic units will likely stock most emergency medications and equipment, but other resources can be arranged for backup. Portable emergency medication kits may be obtained on consignment from medical facilities in a team’s home location. The team physician’s personal medical bag should be relatively lightweight and portable. It should be large enough to store adequate emergency equipment and medications, and it should have multiple compartments that are easily accessible. Contents can be divided into well-labeled smaller kits that enhance access and maintain organization. Mayne has suggested using a large fishing tackle box.35 Tool boxes may also be useful. There are currently a variety of newer bags to choose from that will likely meet the requirements of most team physicians. The contents of the medical bag can be broken into two main categories—equipment and medications. An inventory list should be included in all bags to ensure adequate restocking. A notebook, or any portable digital recording device, can be handy for documenting injuries or medical problems and required treatment. Common medications used in routine medical care include antibiotics, decongestants, antihistamines, anti-inflammatory drugs, gastrointestinal drugs, ophthalmologic drugs, dermatologic drugs, and local anesthetics. Some physicians store stronger pain medicines, intravenous fluids, and muscle relaxants. ­Emergency medications are vital to the medical bag, and the team physician must be prepared to address cardiovascular, asthmatic, anaphylactic, airway, and spinal cord emergencies. Equipment required for CPR should be available. Automatic external defibrillators (AEDs) are available at some athletic events.36 Prior knowledge of the AED locations is essential to their successful employment during a cardiac emergency. Ambulances on the scene may provide cardiac monitor, defibrillator, spine board, stretcher, splint, and airway and suction equipment.34 Table 12-6 contains a suggested list of medications and equipment that will likely be helpful in most sports settings.31,33-35,37 The list can be modified according to the preferences and needs of individual physicians and the athletic events they cover. Most team physicians also stock medications required for routine medical care, especially when traveling with a team.

treatment, and disposition of injuries on the field or in other venues. When possible, this plan should be in writing. Knowledge of specific sports and their associated injury patterns and familiarity with the physical environment enable team physicians to anticipate potential problems. Although some general principles exist, emergency coverage is highly individualized because sports are vastly different in their potential for serious trauma and illness.38 A team leader and appropriate medical personnel should be available at all times on the sidelines of most organized sporting events. Optimally, the team leader is either the team physician or a senior athletic trainer, but in many situations with limited resources, it may be the coach. The team leader is responsible for designating responsibilities and supervising on-field management of injuries, for implementing rarely needed disaster protocols, for prearranging a network of referrals and emergency care, and for ultimately directing the treatment of all serious injuries.39 Sideline physicians should familiarize themselves with the equipment and training of the local paramedics/emergency medical system. It is helpful to arrive early at athletic events to receive updates on injured or ill athletes and to meet with health care personnel from visiting teams. Specific equipment requirements are unique to individual sporting events, but certain emergency medications and equipment should be available at most venues, especially at high-risk collision sporting events such as football and hockey. The level of competition may also demand that certain minimal emergency requirements be met. For example, direct access to a fully equipped, well-trained emergency medical system is more reasonable at a professional football game compared with a high school football game. Most schools or teams own a spine board. Splints, heart monitors, oxygen, blankets, and other cardiopulmonary resuscitation equipment are usually provided by the local paramedics, but it is the ultimate responsibility of the team physician to ensure access and availability at higher level and higher risk events. The team physician’s medical bag, combined with the athletic facility’s equipment, should provide adequate resources for most emergencies in the event that an ambulance and paramedics are not immediately available. Efficient and effective communication is readily available through cellular phones, radios, or nearby phones. Local emergency phone numbers should be readily accessible, and medical facilities should be notified of the athlete’s condition and expected time of arrival. Appropriate transportation must be planned in advance.

Preparation

A downed athlete should be assessed immediately and thoroughly. The recognition of early warning signs and early activation of the emergency medical system, basic life support (CPR), and defibrillation can maximize chances of survival.40 First, a primary survey is performed to screen for lifethreatening injury.21,40 The mnemonic ABCDE will aid in a stepwise assessment of the athlete: Airway, Breathing, Circulation, Disability, and Exposure. The airway (A) is assessed by looking, listening, and feeling for breathing. A face-down unconscious or confused athlete is assumed

Catastrophic injuries in sports medicine occur infrequently and usually make up fewer than 1 in 100,000 reported ­injuries.32 Life-threatening emergencies on the field, although rare in organized sports, rely on well-organized, well-trained medical personnel experienced in sports coverage. Adequate preparation and insightful anticipation on the part of the team physician are vital to good outcomes. Before the season, the team physicians and athletic trainers should establish a detailed plan for the ­evaluation,

General Injury Assessment

518

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 12-6  Contents of the Medical Bag System

Equipment

Medication

General

Notepad, Dictaphone Inventory card Portable phone and emergency numbers Prescription pad Thermometer Sharps container, biohazard bag Band-aids Examination gloves Cotton swabs Cold packs Tape measure Scissors (tape) Vision testing card Patch, eye shield Fluorescein strips Penlight Ophthalmoscope with ultraviolet light Cotton-tipped applicators Nasal tampon Silver nitrate sticks

Analgesics Acetaminophen Acetaminophen with codeine* Ketorolac (Toradol)* Meperidine (Demerol)*

Eye kit

Nose

Dental Airway

Ear Allergic

Cardiac

Gastrointestinal

Hank’s Balanced Saline Solution Airways (oral: sizes 3-6; nasal: sizes 26, 28, 30) Bag-valve mask Endotracheal tubes (sizes 6-8) with stylet Laryngoscope with straight and curved blades Syringe (10 and 12 mL) No. 12 or 14 over-the-needle catheter (emergency needle cricothyroidotomy) No. 11 scalpel blade Lubricant Heavy wire, bolt cutter (remove facemask) Tongue blades Mirror, indirect laryngoscope* Bulb suction syringe Otoscope Ear wax curette See Airway, Oral, and Respiratory

Blood pressure cuff Stethoscope Mouth barrier Paramedics or team physician Defibrillator Cardiac monitor (automatic external defibrillator most ­practical) Various syringes, needles, and intravenous catheters Lactated Ringer’s or normal saline (liter bags) and 50% ­dextrose ampules Intravenous tubing Intracatheter (angiocatheter) needles (18 gauge)

Endocrine Extremities

Neurologic

Athletic tape Elastic bandages Alumafoam finger splints Stockinette Webril Fiberglass casting tape or prefabricated splint Casting gloves Sling Orthoplast* Reflex hammer

Antibiotic ophthalmic drops* Tetracaine ophthalmic drops* Ketorolac (Acular) drops* Eye irrigant, wash Epistaxis Afrin nasal spray Neo-Synephrine Vaseline or K-Y jelly Oxygen (paramedics) Albuterol inhaler Decongestants, cough syrup*

Cortisporin otic suspension* Antihistamine (nonsedating)* Epinephrine (1:1000 0.3 mg automatic injection and 1:10,000 vial available for IV use) Diphenhydramine (oral, IV) Prednisone, methylprednisolone (oral, IV) Paramedics or team physician Epinephrine Atropine Lidocaine Sodium bicarbonate Nitroglycerine* β-Blocker* Furosemide*

Antacids* Antidiarrheals (Lomotil, Imodium)* Antiemetic (Tigan, Compazine)* Insulin* Ampules of 50% dextrose* Glucagon Anti-inflammatory agents NSAIDs (e.g., ibuprofen, naproxen)* Oral steroids (e.g., Medrol Dosepak)* Injectable steroids*

Diazepam*

The Team Physician

519

   TABLE 12-6  Contents of the Medical Bag������� —cont’d System

Equipment

Medication

Spine

Paramedics, school or team physician: Spine board Hard cervical collar Stretcher Suture kit Sutures (absorbent and nonabsorbent; 4-0, 5-0, 6-0) Sterile gauze pads (4 × 4 inches, 2 × 2 inches) Nonadherent sterile pads (Telfa) Sterile gloves (latex and nonlatex) Syringes and needles Scalpel blades (No. 11, 15) Sterile saline irrigant Alcohol swabs Betadine or other cleansing solution (Hibiclens) Steri-Strips Tincture of benzoin Moleskin Skin coverage (athletic trainer/team physician) Op-Site Bioclusive ProWrap Stretch tape Heavy towel clip (posterior clavicle dislocation)* Blankets* Sandbags* Crutches*

Methylprednisolone (Solu-Medrol) or ­dexamethasone (Decadron) (spinal injury)*

Skin

Infectious disease

Miscellaneous

Lidocaine 1% (± epinephrine) Antibiotic ointment* Hydrocortisone cream* Antifungal cream* Petroleum jelly*

Antibiotics* Amoxicillin Erythromycin Penicillin Cephalexin Acyclovir Bactroban ointment

*Optional. This is not meant to be an all-inclusive list. There are many sections in which equipment overlaps and is not relisted (e.g., cotton-tip applicators in the eye kit can be used for gross light-touch screening in the neurologic section).

to have a cervical spine injury and should be logrolled to a face-up position (Fig. 12-2). The team leader maintains in-line positioning of the head and neck while directing three members of the medical team (one at shoulders, hips, and knees) to roll the athlete. The chin lift or jaw thrust ­ maneuver should be applied to open the airway while ­ maintaining neck stabilization. The airway may be obstructed by the tongue, mouth guard, vomitus, or avulsed

Figure 12-2   The logroll. (From Halpern B, Cardone D: Injuries and emergencies on the field. In Mellion M, Walsh W, Shelton G [eds]: The Team Physician’s Handbook, 2nd ed. Philadelphia, Hanley & Belfus, 1997.)

teeth, and it should be cleared with a finger sweep, suction, or both. An oral or nasal airway can be used to maintain airway patency, and an oral airway should be inserted immediately in any unconscious patient. Helmets should be left in place, and football facemasks can be removed using a sharp knife, a screwdriver, or bolt or wire cutters. Hockey facemasks open without need for removal. Endotracheal intubation may be needed to secure and protect the airway. Intubation should be performed by properly trained personnel. Mandibular and maxillofacial fractures or laryngeal fracture and laryngeal edema are serious injuries and can seriously compromise the airway. Needle cricothyroidotomy can quickly reestablish the obstructed airway (Fig. 12-3). It is performed by palpating the cricothyroid membrane and puncturing the overlying skin with a 14-gauge catheter over the needle directed at a 45-degree caudad angle. Oxygen tubing, attached to a high-pressure oxygen source, can be hooked to the ­catheter needle hub, and intermittent ventilation can be provided by cutting an open hole in one side of the tubing and occluding the hole with a thumb for 1 second, then releasing it for 4 seconds and repeating the process. Breathing (B) may resume after the airway is opened, but artificial ventilation using mouth-to-mask, mouth-tomouth, or bag-mask techniques should be started if spontaneous respirations are absent despite an open airway. High concentrations (100%) of oxygen should be administered if available. An open pneumothorax, flail chest with associated pulmonary contusion, or a massive hemothorax or tension pneumothorax can severely impair ventilation and are important to ascertain during initial screening.

520

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 12-3   Needle cricothyroidotomy. (From Halpern B, Cardone D: Injuries and emergencies on the field. In Mellion M, Walsh W, Shelton G [eds]: The Team Physician’s Handbook, 2nd ed. Philadelphia, Hanley & Belfus, 1997.)

Circulation (C) is assessed by checking the carotid pulse. Systolic blood pressure can be quickly estimated by the presence of carotid pulses (>60 mm Hg) or radial pulses (>80 mm Hg). If the athlete is pulseless, immediately activate the emergency medical system. Early defibrillation with a standard defibrillator or an automatic external ­defibrillator may have the single greatest impact on survival.40-42 Hypotension following injury should be considered hypovolemia until proved otherwise, and the hemorrhage source should be sought immediately. Disability (D) is evaluated by performing a limited neurologic examination to establish level of consciousness and pupillary size and reaction. Level of consciousness may be described using the mnemonic AVPU: Alertness, response to Vocal stimuli, response to Painful stimuli, and Unresponsive.21 The Glasgow Coma Scale may also be used.43 Head injuries are the most common cause of disability in organized sporting events. The “D” may also stand for “defibrillation” or “drugs,” which may be required for successful cardiopulmonary resuscitation.

Exposure (E) should be adequately evaluated by inspecting the extremities and other body parts for bleeding, fractures, or severe contusions. Blood pressure may be checked at this time. The “E” can also stand for “environment.” Moving the athlete out of direct physical danger, or ­applying rewarming or cooling techniques, may be required. Airway protection, CPR, and other lifesaving measures may begin as soon as a problem is identified, even before the completion of the primary survey (Table 12-7). Once an identified problem is stabilized, the survey is resumed. In addition to CPR, large-caliber (18-gauge) intravenous catheters, oxygen, and electrocardiographic ­ monitoring should be used appropriately when available. Enough information is usually obtained during the primary survey and resuscitation to decide about immediate transfer to a referral facility. A secondary survey is performed in athletes whose injuries do not warrant immediate transfer. The secondary survey is a “head-to-toe” evaluation for injuries, and it includes assessment of the vital signs (including temperature) and continual reassessment of the ABCs. If the athlete makes it through the secondary survey without being transferred, decisions must ultimately be made about ­disposition and follow-up. The team physician must address the issues of whether it is safe to return to play. Figure 12-4 summarizes the primary survey–­resuscitation– secondary survey approach and the decision-making process regarding hospital transfer, follow-up, and return to play.14 The injuries listed in the figure all can be associated with life-threatening complications, and they should be diligently ruled out.

Specific Injuries and Illnesses Specific injuries will be identified during both primary and secondary surveys. Immediately life-threatening injuries compromising the airway, breathing, and circulation must be recognized during the primary survey and treated immediately to ensure a better chance for survival. Other life-threatening injuries, either because of a more delayed presentation or because they indirectly affect the ABCs, should be sought during the secondary survey. Less serious injuries can be dealt with only after the athlete is deemed stable.

   TABLE 12-7  Summary of Current Guidelines for Cardiopulmonary Resuscitation Adult (>8 yr) One Rescuer

Adult (>8 yr) Two Rescuers

Child (1-8 yr) One Rescuer

Infant (<1 yr) One Rescuer

2 hands 80-100 1½-2

2 hands 80-100 1½-2

Heel of 1 hand 100 1-1 ½-2

2 or 3 fingers 100+ ½-1

10-12 1/5-6

10-12 1/5-6

20 1/3

20 1/3

15:2

5:1

5:1

5:1

Chest Compression

Method Rate (per min) Depth (in) Rescue Breathing

Rate (per min) (breath/sec) Ratio

(Compressions:breaths)

The Team Physician Differential

Primary Survey A: Airway and Cervical Spine Logroll, remove facemask, chin lift or jaw thrust maneuver

Cervical spine injury (assume), laryngeal fracture/edema, foreign body (tongue), maxillofacial/oral trauma

B: Breathing “Look, listen, and feel,” artificially ventilate: 2 breaths assess chest rise/fall

Pneumothorax (tension), hemothorax, flail chest, pulmonary contusion, exercise-induced bronchospasm or anaphylaxis

C: Circulation Pulseless

Pulseless ventricular fibrillation until proved otherwise, shock (hypotension) secondary to hemorrhage

activate EMS, start CPR

D: Disability (Defibrillation, Drugs) Defibrillate! (AED) if required, AVPU (Alertness, response to Vocal or Painful stimuli, Unresponsive)

Unconscious/mental status change head injury, dysrhythmia, hypoxia, hypotension, seizure, hypoglycemia, heatstroke, hypothermia

E: Exposure (Environment) Undress appropriately to assess all injuries, remove from danger, remove from hot/cold environments

Other hemorrhaging injuries, heat/cold injuries

Resuscitation Stabilize ABCs, CPR, “IV-O2-Monitor” Oral/nasal airway, endotracheal intubation, 100% O2, needle cricothyroidotomy, appropriate fluids and ACLS drugs, consider oro/nasogastric tube and Foley catheter

Secondary Survey

• Complete vital signs • Continuous reassessment of ABCs Head and Neck

Differential

Further head and/or cervical spine injury

Eyes, Ears, Nose, and Throat

Ocular/globe injuries (any loss of vision), soft tissue or bony facial trauma, basilar skull fracture, scalp lacerations (step-off), nasal fracture or septal hematoma, dental injuries or intraoral hematoma, laryngeal/tracheal injuries, auricular hematoma

Thorax/Cardiovascular

Rib fractures, pneumothorax, hemothorax, lung contusion/laceration, EIB, myocardial contusion, shock, MI/dysrhythmia/other cardiac problem, posterior clavicle dislocation

Abdomen/Pelvis/Genitourinary

Muscle contusion; rectus sheath hematoma; liver, spleen, kidney, pancreas, small bowel, or vesicourethral injuries; ruptured testicle; hematocele; testicular torsion; vulvar hematoma

Extremities/Musculoskeletal

Fracture or dislocation (bone); ligament, muscle, tendon, vascular, nerve, or skin injuries; compartment syndromes; frostbite

Environmental/Miscellaneous

Hospital Transfer

Sideline/Re-evaluation

Safe to return? Risk further injury? Athlete comfortable?

Return to Play

Heat illness, hypothermia, anaphylaxis

Figure 12-4   Emergency injury assessment. ACLS, advanced cardiac life support; AED, automated external defibrillator; CPR, cardiopulmonary resuscitation; EIB, exercise-induced bronchospasm; EMS, emergency medical system; MI, myocardial infarction.

521

522

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Head and Neck Head and neck injuries are the most common cause of catastrophic athletic injuries. Loss or alteration of consciousness signifies head injury, and the athlete should be immediately assumed to have a cervical spine injury. Be aware that hypoxia, hypovolemia (shock), hypoglycemia, seizures, heatstroke, hypothermia, and therapeutic or abusive drugs (e.g., alcohol, cocaine) can cause alteration or loss of consciousness. Cerebral concussion is the most common head injury encountered by team physicians. Its hallmarks are confusion and amnesia.22 Other associated findings include headache, dizziness or vertigo, blurriness of vision, tinnitus, nausea or vomiting, gross incoordination, emotional lability, seizures, slowness to answer questions, easy distractibility, disorientation, slurred or incoherent speech, vacant staring, repetitive asking of questions, amnesia, and generalized lack of awareness of surroundings. Amnesia can be retrograde (no memory of events before injury) or anterograde (no memory of events after injury). Severe retrograde amnesia may indicate a more serious injury. Head injury evaluation is subdivided into on-field and sideline phases. The most important objective of the initial field examination is to make an accurate diagnosis of the athlete’s level of consciousness and to rule out the presence of associated injuries, particularly to the cervical spine.28 Observe for spontaneous movement when approaching a downed athlete. Total lack of motion signifies loss of consciousness or a cervical spine injury. Leave the helmet in place, and if the athlete is unconscious, logroll the athlete to the supine position and assess the need for facemask removal and CPR. If the athlete is conscious and breathing adequately, question his or her orientation to time, place, and person in the position in which he or she is found. Maddocks and Saling have proposed using recent memory questions such as what field are we on, which team are we playing, which quarter is it, and who scored last as an initial orientation screen.44 If the athlete is confused or disoriented, it is safest to assume he or she has a cervical spine injury and manage appropriately. When orientation is established, the athlete can be questioned about specific symptoms such as head or neck pain, neurologic symptoms, and other associated symptoms. After the athlete’s confusion and orientation improve to the point at which he or she can follow commands, and after cervical spine injury has been clinically ruled out, the athlete should be helped to the sitting position for a brief period. The athlete should then be slowly helped to his or her feet and then to the sidelines. If severe incoordination is noted, a stretcher or cart can be used. Once on the sidelines, the athlete undergoes a thorough history and a comprehensive physical examination that emphasizes complete neurologic, neck, musculoskeletal, and head and facial region (including funduscopic) examinations. Neuropsychological testing can be performed if available. Recent memory, new learning, and delayed recall can be assessed using short word recalls. Concentration can be assessed by having the athlete repeat the months in reverse or spans of digits backwards (e.g., 1-4-2, 6-9-3-1, and so on up to 6 or more digits). Periodic re-evaluation is extremely important, and all head-injured athletes should be observed on

the sideline for at least 15 to 20 minutes because frequently signs and symptoms worsen with time. Multiple concussion grading systems and return-toplay guidelines exist. They are all empirical, however, and they lack a scientific basis. An American Medical Society of Sports Medicine survey in 1994 showed that most team physician members of the American Medical Society of Sports Medicine were familiar with the Cantu guidelines,23 but that there was large variability in guideline adherence, and that the level of competition may influence decision making. In 1997, the American Academy of Neurology issued a practice parameter that introduced a new set of guidelines based on Kelly’s work and the concussion guidelines published by the Colorado Medical Society.22,27,45 The three sets of guidelines are summarized in Table 12-8. It is far beyond the scope of this chapter to discuss specifics in the management of concussion, but a few safe rules apply. The only universally agreed on management decision is that all symptomatic athletes should not return to play. It is also safe to assume that an athlete should not return to play the same day if he or she experienced loss of consciousness or symptoms lasting longer than 15 minutes. If it is decided that an asymptomatic athlete can return to play, provocative exertional testing should be performed in an attempt to reproduce concussive symptoms. An athlete should perform a 40-yard dash, five push-ups, five sit-ups, and five deep knee bends or jumping jacks; if signs or symptoms recur, the athlete should not be allowed to return and should be re-evaluated periodically. If an athlete does not become symptomatic with exertional testing and returns to play, he or she should be continually reassessed throughout the competition. Concussion grading systems and returnto-play guidelines do not replace personal experience and good medical decision making. If in doubt, be conservative. Signs or symptoms such as persistent vomiting; deteriorating state of consciousness; clear otorrhea, rhinorrhea, or other signs of basilar skull fracture; focal neurologic deficits; large scalp lacerations with step-off deformities; increasing headache; possible intoxication; high-risk conditions (e.g., hemophilia, blood-thinning medication, childhood); or inadequate postinjury supervision warrant urgent transfer and further evaluation. Seizures may occur as a result of concussion or from newly presenting uncontrolled generalized epilepsy. McCrory and Berkovic46 pointed out that concussive convulsions are likely a nonepileptic phenomenon of which all team physicians should be aware. A brief initial tonic stiffening is often followed by a short period of myoclonic jerks, after which the athlete may get up and walk to the sideline without experiencing a significant postictal phase. Concussive convulsions must be differentiated from posttraumatic seizures secondary to structural brain injury, convulsive syncope, drug-related epilepsy (e.g., cocaine), and idiopathic primary epilepsy. Athletes should be transported to the hospital for further evaluation. Concussive convulsions are likely benign, do not require specific antiepileptic drug treatment, and are not associated with a risk for long-term epilepsy; treatment should focus on proper management of the associated concussive injury.46 Generalized seizures can also result from newly presenting primary or uncontrolled epilepsy. Well-controlled primary epilepsy is not a contraindication to participating

The Team Physician

523

   TABLE 12-8  Current Concussion Guidelines Cantu23

American Academy of Neurology22

Colorado Medical Society27

1

No LOC PTA <30 min

Confusion without amnesia No LOC

2

LOC <5 min or PTA >30 min but <24 hr

3

LOC >5 min or PTA >24 hr

Transient confusion No LOC Sx/mental status abn Resolve <15 min Transient confusion No LOC Sx/mental status abn Resolve >15 min Any LOC, either brief (sec) or ­prolonged (min)

Grade Grading

Confusion with amnesia No LOC Any LOC

Return to Play First Concussion

1 2 3

Second Concussion

1 2 3

Third Concussion

1 2

When asx asx 1 wk Wait 1 mo, then asx 1 wk

*Normal at 15 min Asx 1 wk Brief LOC: asx 1 wk† Prolonged LOC: asx 2 wk†

*Normal at 20 min Asx 1 wk Wait 1 mo, then asx 2 wk, consider sooner if asx 2 wk

2 wk if asx 1 wk At least 1 mo, asx 1 wk, consider ­terminating season Terminate season, return next yr if asx

Asx 1 wk† Asx 2 wk† Asx 1 wk†

Terminate practice or contest for day Consider terminating season, but can RTP if asx 1 mo Terminate season, 3 mo if asx

Terminate season, return next yr if asx Terminate season, next yr if asx

Asx 1 wk† Asx 2 wk†

Terminate season, 3 mo if asx Terminate season, next yr if asx

*Mental status and postconcussive abnormalities must be clear. †No signs or symptoms at rest or with exertion. Abn, abnormality; asx, asymptomatic; LOC, loss of consciousness: PTA, post-traumatic amnesia; RTP, return to play; sx, symptoms.

in contact sports. Uncontrolled epilepsy is a contraindication, however. Most generalized seizures will be short (1 to 2 minutes) and followed by a postictal state of confusion. With status epilepticus, however, pharmacologic intervention may be needed. Treatment of status epilepticus involves protecting the athlete from further injury, placing the athlete on his or her side if possible, and administering antiepileptic medications such as diazepam (5 mg intravenous or 0.3 mg/kg rectally) or lorazepam (up to 0.1 mg/kg).47 The airway must be protected and maintained, and ventilatory support may be required. Other forms of serious head injury include cerebral contusions, subdural hematomas, intracerebral hematomas, epidural hematomas, and skull fractures. Have a high index of suspicion if an athlete has a depressed level of consciousness and a headache. All athletes with head injuries should be re-evaluated in 24 hours, and sooner if signs or symptoms worsen. If team physicians use periodic re-evaluation and arrange for good postinjury supervision, most catastrophes can be avoided. If suspicion for a serious head injury is present, the athlete should be immediately transferred by ambulance to the nearest hospital or tertiary care facility for computed tomography (CT) or magnetic resonance imaging and further evaluation and management. Epidural hematomas are associated with high morbidity and mortality, especially if the diagnosis is missed. ­Epidural hematomas most commonly result from tears of the middle meningeal artery, and they are associated with a skull fracture in 80% of cases.48 Bleeding occurs between the inner

layer of the skull and the dura mater. A high-velocity impact to the temporoparietal region, such as with a baseball, bat, or fall, can result in an epidural hematoma.49 Athletes often experience a brief loss of consciousness followed by a lucid interval. This period of lucidity, which may last several hours, can lead to false reassurance and often contributes to a missed diagnosis. Rapid neurologic deterioration to coma often follows the lucid interval. Specific signs include a fixed dilated pupil on the injured side, deteriorating level of consciousness, and hemiplegia or other focal neurologic deficits. Permanent neurologic injury or death can result without prompt surgical intervention. If an epidural hematoma is suspected, the athlete should be urgently transferred to a hospital, and a CT scan should be obtained. Spine injuries occur frequently in athletic participation and competition. Catastrophic neck injuries in sports are uncommon but are devastating when they occur.50 Injuries to the spinal cord, bony vertebral column, or supporting ligaments make up only 2% to 3% of all sports injuries.51,52 The most common sports associated with cervical spine trauma in the United States are football, wrestling, and gymnastics.53 A systematic and highly organized approach is needed to evaluate downed athletes with a suspected cervical spine injury. All athletes with altered consciousness, neck pain, or motor or sensory neurologic symptoms, no matter how transient, should be assumed to have a cervical spine injury until proved otherwise. When necessary, use a logroll technique to move the athlete to a supine position while

524

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� ABCDE Quick, comprehensive, complete, neurologic examination and history evaluation for paresthesias and weakness

Abnormal

Normal

Palpate for cervical tenderness

Yes

Immobilize and prohibit further activity X-ray evaluation needed

No

Test active, not passive, cervical range of motion by having the athlete touch the chin to chest, chin to right shoulder, chin to left shoulder, right ear to right shoulder, left ear to left shoulder

Abnormal (pain or restricted motion)

Within normal limits OK to return to play Repeat examination is needed Figure 12-5   Assessment of a neck injury. (From Halpern B, Cardone D: Injuries and emergencies on the field. In Mellion M, Walsh W, Shelton G [eds]: The Team Physician’s Handbook, 2nd ed. Philadelphia, Hanley & Belfus, 1997.)

maintaining in-line positioning of the head and neck with the trunk. After assessing and stabilizing the ABCs, use a stepwise approach to perform evaluation, as illustrated in Figure 12-5.39 The athlete’s helmet and chin strap should be left in place, and the facemask should be removed, if necessary, to protect the airway.54 With the assistance of properly trained personnel, the athlete should be placed on a spine board, with the head and neck carefully immobilized, and transported to the hospital for further ­evaluation. The logroll technique is once again used to place the athlete on the backboard, this time adding a fourth ­ person to manage the board. If a helmet is on, the shoulder pads should be left in place, and the head and neck should be immobilized on the spine board using tape, sandbags, and bolsters, the entire time being careful to prevent ­inadvertent neck ­ flexion. If there is no helmet, a simple hard collar can be used. An unstable cervical injury can be converted to a permanent neurologic injury if improperly handled. Health care personnel managing spine injuries should be adequately trained and well rehearsed. A few good suggestions should be considered when evaluating suspected cervical injuries. After stabilization of the ABCs, assess the athlete’s level of consciousness and perform a quick, comprehensive neurologic examination. Neurologic symptoms accompanying a suspected neck injury, especially bilateral or lower extremity symptoms, should be treated as an unstable cervical spine fracture or a cervical cord injury.50,52,55 Bilateral symptoms (e.g., burning hands syndrome), transient quadriparesis, or central

cord neurapraxia, should immediately alert the physician to cord injury. Spinal cord injuries must frequently be differentiated from burners or stingers. Burners and stingers are common. It is estimated that more than half of athletes in contact and collision sports have experienced them.56,57 The two most common mechanisms described involve a lateral blow to the neck, sometimes with rotation, which results in an ipsilateral brachial plexus stretch injury (upper cord) or a contralateral cervical nerve root “pinch” or compression.23 Signs and symptoms include muscle weakness and burning occurring most commonly in a C5 and C6 distribution. Burners and stingers typically are unilateral, resolve in 5 to 10 minutes, and are accompanied by relatively full, pain-free cervical motion.52 However, it can be extremely difficult to make a determination on the field. If there is any suspicion of a spine injury, the athlete’s neck should be immobilized. Ultimately, no matter how minor the neck injury, no athlete should return to play without demonstrating full upper extremity and neck muscle strength, full and painfree cervical range of motion, no complaints of pain with axial loading of the spine, and no signs or symptoms.52,58 Many of the same principles used with traumatic cervical spine injuries can be applied to traumatic thoracolumbar spine injuries. If an athlete experiences thoracolumbar pain, neuromuscular symptoms, or loss of motion, the spine must be immobilized and the athlete placed on a spine board and transferred for further evaluation.

The Team Physician

Eye, Ear, Nose, and Throat Most facial injuries sustained during athletic events consist of superficial soft tissue injuries and are minor in nature.59 Initial inspection for obvious swelling, facial asymmetry, or facial deformity will reveal specific trauma. The facial bones should be palpated carefully for fractures. Facial movements should be tested, and signs of basilar skull fracture such as periorbital ecchymosis (raccoon eyes) or mastoid area ecchymosis (Battle’s sign), should be sought. Visual acuity must be tested. Any loss of vision in the face of trauma precludes return to play and warrants immediate ophthalmologic evaluation. Other signs and symptoms of serious injury that require further evaluation include visual field cuts, eyelid or orbit asymmetry or protrusion, perception of flashing lights, abnormal extraocular muscle movement, pupillary asymmetry, corneal or scleral lacerations, sharp stabbing or deep throbbing pain, double vision, foreign body in the cornea, anterior chamber bleeding (hyphema), or a dark subconjunctival mass.37,59 A funduscopic examination should be performed in all cases of suspected eye injury to assess the chambers and retina. Most eye injuries should be covered by having the patient close his or her eye and by placing a sterile pad gently over the eye, followed by taping a protective hard shield over the pad. A pressure patch should almost never be applied in traumatic injuries, and all potentially serious eye injuries should be immediately evaluated by an ophthalmologist. Nasal fractures are almost always accompanied by profuse epistaxis. Deformities are frequently obvious and occasionally may be corrected in the acute setting by experienced practitioners. Athletes may not return to play unless there are no associated serious injuries and the nose can be adequately protected. Cerebrospinal fluid (CSF) leakage from the nose can occur with cribriform plate or sphenoid sinus fractures but is sometimes difficult to detect because of profuse bleeding. CSF may be distinguished from blood by a “ring” test, which is easily and quickly performed by placing a few drops of fluid from the nose on an open gauze pad. A positive test involves a clear ring of CSF diffusing far outside the centrally located blood.59 Athletes with CSF leakage should be immediately transferred to the hospital for further evaluation and treatment. The nasal septum must be closely examined for a septal hematoma, a bluish, tender, bulging submucosal mass on the nasal septum. These hematomas should be aspirated or incised immediately. A nasal tampon is used to control bleeding and to prevent recurrence. The oral cavity should be closely examined for fractured or avulsed teeth, hematomas, lacerations, and dental malocclusion. An avulsed tooth can be replaced in its original position, and the oriented athlete should be instructed to gently bite down to hold it in place. If not, the tooth can be placed in Hank’s Balanced Saline Solution (normal saline has the wrong pH), and the athlete should immediately seek professional dental care. If Hank’s Balanced Saline Solution is not available, cold whole milk is probably the next best medium, followed by normal saline–soaked gauze. Intraoral hematomas and lacerations can signal underlying maxillary or zygomatic fractures and should be thoroughly inspected.59 Dental malocclusion can be screened by asking the athlete if his or her teeth “fit together” correctly.59

525

Malocclusion requires closer evaluation for mandibular or maxillofacial fractures. The external ear should be closely inspected for hematoma formation. If present, the hematoma should be aspirated, and a collodion, plaster of Paris, or other firm cast should be applied immediately. An otoscopic examination screens for hemotympanum, bleeding behind the tympanic membrane that signifies a basilar skull fracture. Laryngeal fractures, laryngeal edema, and tracheal fractures can severely compromise the airway. If an athlete complains of hoarseness or anterior neck pain or exhibits stridor, bony crepitus, or subcutaneous emphysema, he or she must be monitored extremely closely and transferred emergently to the hospital. A needle cricothyroidotomy may be warranted if the airway collapses.

Thorax Thorax injuries generally result from rapid deceleration or high-energy impact and are divided into chest wall and intrathoracic injuries. They occur most frequently in highspeed, high-energy contact and high-altitude sports.60 Rib fractures result from blunt trauma. The fourth through ninth ribs near the midaxillary line are most frequently involved. Most rib fractures are uncomplicated, but life-threatening injuries such as pneumothorax, hemothorax, lung contusion or laceration, or injury to the liver, kidney, or spleen can accompany these injuries. Tenderness at the midclavicular line generally indicates a costochondral sprain rather than a fracture. Decreased or absent breath sounds, tachypnea, sudden dyspnea, and pleuritic chest pain suggest a pneumothorax. There are three types of pneumothorax: simple, tension, and open. Simple and tension pneumothoraces can occur spontaneously, and all three types may occur secondary to chest wall trauma or with accompanying rib fractures.60-62 A simple pneumothorax commonly presents with tachycardia, decreased breath sounds, decreased tactile fremitus, and hyperresonance over the affected area.63 A suspected simple pneumothorax should be confirmed with a posteroanterior and lateral chest radiograph. Expiratory films may be needed to detect small pneumothoraces. The size of the pneumothorax will help determine definitive treatment.63 Field management of simple pneumothoraces is supportive and includes oxygen and close monitoring. Clinical deterioration can indicate a large spontaneous pneumothorax (simple), bilateral pneumothoraces, or tension pneumothorax. A tension pneumothorax usually presents with tachycardia, neck vein distention, tracheal deviation away from the affected side, hyperresonance to percussion on the affected side, and ultimately hypotension. Tension pneumothorax occurs in 1% to 2% of patients with spontaneous pneumothorax.61 It is a true medical emergency that requires immediate needle aspiration. Often the patient is too unstable to wait for radiographic confirmation. A 14-gauge catheter-over-needle is inserted into the second intercostal space at the midclavicular line just above the third rib.39,61 The diagnosis is confirmed by a rapid gush of air coming through the needle. Open pneumothorax presents as a “sucking” chest wound. Affected athletes will need ventilatory support, and

526

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

a sterile occlusive dressing should be placed completely over the wound and taped on three sides.21 Massive hemothorax presents similarly to tension pneumothorax with decreased breath sounds and hypotension. In contrast, dullness with percussion is noted over the pooled blood in the chest cavity. Treatment involves ­ventilation, intravenous fluids, and emergent chest tube placement. A pulmonary contusion usually presents more insidiously, with cough, hemoptysis, and dyspnea. Athletes should be transferred to a hospital for immediate further evaluation. Pulmonary contusions generally resolve with time, but they have the potential of progressing to acute respiratory failure. Multiple rib fractures may cause a flail chest. Chest wall motion becomes asymmetric or paradoxical because a segment of the chest does not have bony continuity with the rest of the thoracic cage.21 The segment protrudes on inspiration and indents on expiration. Initial treatment involves ventilation, pain control, and intravenous fluids. Sternal fractures may also occur but are rare in sports. Exercise-induced bronchospasm may be confused with chest injury. It may present with dyspnea, wheezing, coughing, and anxiety. The athlete may be ­ gasping for breath, tachycardic, and using secondary muscles of respiration. A severe asthma attack can be fatal and demands recognition. If severe airway symptoms are present, administer ­subcutaneous epinephrine (up to 0.3 mL of 1:1000 solution). If the athlete can cooperate, two to three puffs of an albuterol or other β-agonist metered-dose inhaler, preferably with a spacer, or an albuterol nebulizer is ­indicated.64 Myocardial contusion may occur as a result of blunt chest trauma. It often presents as a dull ache in the chest that may be attributed to a chest wall contusion. A high index of suspicion should be maintained. If a myocardial contusion is suspected, an electrocardiogram should be performed. Patients with myocardial contusion are at risk for cardiac dysrhythmias and may warrant critical care unit admission for monitoring. Commotio cordis is another often-fatal chest wall injury that results in ventricular fibrillation.41,65 Instantaneous or near-instantaneous cardiac arrest occurs after a nonpenetrating chest blow, often from a ball or other small object. It has been reported most frequently in youth baseball, and more than 70% of victims are younger than 16 years of age.65 It may occur during competitive or recreational sports or on the playground. The athlete may perform an action after being struck, such as picking up a ball, and then suddenly collapse. Failure to make an immediate diagnosis may lead to lost resuscitation time; consequently, most of these injuries result in death. The availability of automated external defibrillators may prevent a fatal outcome.42,65 Myocardial infarction, cardiac dysrhythmias, and other cardiac problems may occur in athletes, especially with increasing age. In addition, spectators experiencing cardiac events may require the care of a team or event physician. A good knowledge of the principles of advanced cardiac life support is important. Posterior clavicle dislocation at the sternoclavicular joint may be caused by a direct blow to the proximal clavicle or forcible lateral shoulder compression. The athlete

may present with respiratory distress resulting from direct ­clavicular compression of the airway or from lung puncture. Injury to the great vessels and hemodynamic compromise may also occur. The diagnosis may be made by noting a flattening in the region of the proximal clavicle compared with the opposite side and by viewing the supine patient from a superior position. If dyspnea, stridor, dysphonia, choking, or coughing is noted, airway compromise should be suspected, and emergent reduction is warranted.

Abdominal, Pelvic, and Genitourinary Injuries Like thoracic injuries, abdominal injuries primarily occur as a result of rapid deceleration or high-energy blunt impact to the abdomen or flank. They are most often sustained in high-energy contact sports, high-speed sports, or high-altitude falls.60 A large force concentrated over a small area, such as a direct blow to the abdomen from a bicycle ­ handlebar, can injure the intra-abdominal organ located in the region of the blow.66 The initial presentation of abdominal injuries varies from dramatic to insidious, and over-­reliance on a single, initially benign abdominal examination may lead to missed diagnoses and poor ­outcomes.66 Many athletes with blunt abdominal trauma may have no external signs of injury. Presentation of complications may follow an insidious course, and physical examination has been estimated to be only 65% accurate in blunt abdominal trauma.67 Tenderness, rigidity, distention, guarding, flank pain, costovertebral angle tenderness, hematuria, inability to void, flank ecchymosis (Grey Turner’s sign), periumbilical ecchymosis (Cullen’s sign), referred shoulder pain (Kehr’s sign), lower rib fractures, or signs of shock should raise the team physician’s suspicion for serious intra-abdominal injury. Any doubt on the part of the team physician should lead to hospital transfer and an abdominal CT scan or diagnostic peritoneal lavage and a period of close observation. Common abdominal injuries include muscle contusion, rectus sheath hematoma, and injury to the liver, spleen, kidney, pancreas, or small bowel. Genitourinary problems include vesicourethral injury, vulvar hematoma, ruptured testicle, varicocele and hematocele, and testicular or scrotal contusion. Muscle contusions and rectus sheath hematomas must be differentiated because the latter problem may involve continued bleeding from the inferior epigastric artery and the need for surgical intervention.60,66,68 The abdominal wall should be carefully palpated while the athlete contracts the rectus by lifting his or her legs off of the table. Increased tenderness to palpation superficially with this maneuver may signal an abdominal wall injury. In contrast to abdominal wall injuries, athletes with visceral injuries may experience a decrease in pain while the abdominal wall is contracted because the contracted rectus “guards” the deeper visceral structures.66 Although athletes with a rectus sheath hematoma may have increased muscle guarding, nausea and vomiting, and a palpable mass when compared with a more simple rectus contusion, the differentiation can be difficult clinically, and any suspicion for a hematoma deserves hospital transfer, a CT scan, and close observation. The spleen is the most commonly injured organ in all sports-related abdominal traumas.69 Left upper quadrant

The Team Physician

tenderness, abrasion, or contusion, localized lower rib tenderness or fracture, left shoulder pain (Kehr’s sign), and hypotension can all result from a splenic injury, but physical examination is often unreliable, and athletes need ­ careful follow-up in suspicious cases.60 CT scan or ultrasound will often confirm the diagnosis. A recent history of mononucleosis, especially in the preceding 4 weeks, may predispose the spleen to injury. White blood cell counts may be elevated more than 20,000 in splenic trauma. Liver injuries present similarly to spleen injuries, but on the opposite side of the abdomen, the right upper quadrant. Pain may refer to the right shoulder if bleeding irritates the diaphragm. Injury should be suspected in the presence of lower right-sided rib fractures. Patients may be hemodynamically stable even with moderate to severe injury. Liver function tests may be elevated, but CT scan is the most reliable diagnostic tool. The most common kidney injuries are renal contusions. They may be occult or obvious. Gross or microscopic hematuria may occur. Flank pain, costovertebral angle tenderness, or gross hematuria may signal renal injury. The diagnosis is usually made by renal ultrasound, intravenous pyelogram, or CT scan. CT scanning should be performed if the athlete has gross hematuria or has microscopic hematuria associated with low blood pressure, and blood ­pressure should be closely monitored for hypotension.60 Close monitoring is warranted in the acute phase until imaging rules out further injuries such as kidney ­laceration, vesicoureteral disruption, or associated retroperitoneal ­bleeding.66,70 The pancreas and small bowel may be injured, especially when a large force concentrated over a small area affects these organs such as occurs with a hockey stick, handlebar, or helmet.66 Pancreatic injury usually presents insidiously after the blow. The initial abdominal examination may be benign. If any unexplainable periumbilical or upper abdominal pain is present, especially if it is associated with nausea, vomiting, or radiation to the back, the athlete should be transferred to the hospital, where a serum ­amylase level can be obtained and CT scan may be performed. Small bowel injuries also occur insidiously with little or no initial pain. Duodenal hematomas can present with intractable vomiting 12 hours to days after the initial injury.66,70 Urethral injuries occur with pelvic fractures and straddle injuries. They may be associated with penile, vulvar, or perineal hematomas. The prostate may be absent or “highriding” on examination.60 Blood at the urethral meatus indicates urethral trauma and warrants timely imaging with a retrograde urethrogram. Scrotal and testicular injuries result from direct blows to the groin. Pain and swelling may make it difficult to differentiate a simple contusion from a severe injury such as testicular rupture or hematocele. The degree of pain may be an unreliable indicator of severe testicular injury.71 A soft tissue contusion of the scrotum or testicle is most common. However, if the testicle is severely swollen, nonpalpable, or the scrotum does not transilluminate, prompt urologic referral is indicated to rule out testicular rupture or hematocele, which may require surgery.71 Testicular ultrasound and Doppler studies are often helpful.

527

Musculoskeletal and Extremity Injuries The most common injuries a team physician encounters on the sidelines are musculoskeletal trauma.14 Musculoskeletal injuries that compromise hemodynamic stability should be identified early in the primary survey, but most lesser injuries are identified during the secondary survey. After inspecting for obvious deformity, the initial extremity evaluation should concentrate on obvious swelling, skin and soft tissue lacerations, skin color (pallor), pain, loss of motion, joint instability, neurologic abnormalities (­sensory and motor), and presence and quality of pulses. Open fractures should be covered, skin lacerations repaired, and most dislocations reduced and splinted after assessing neurovascular status. Although most extremity injuries can be treated on the sideline with rest, ice, compression, and ­elevation, more serious injuries should be referred for appropriate further evaluation and management. An experienced physician can reduce most dislocations in the field. The one exception to this rule may be the hip. Neurovascular status must be carefully documented before and after reductions. Dislocations may be more easily reduced acutely before significant muscle spasm develops. A general rule for reduction is to apply axial traction and to reproduce the mechanism of injury, occasionally using one hand to gently guide a joint back to congruity. Reductions should be splinted, and in cases of suspected neurovascular injury, athletes should be transferred to the hospital for vascular imaging studies as indicated. Most reductions should be followed with radiographic evaluation to ­ confirm the reduction and rule out associated fractures. Fractures may be open or closed. Open fractures should be covered with a saline-soaked sterile dressing and splinted. Physicians should not attempt to replace soft tissue or bone, probe the wound, or reduce the fracture. With traumatic amputations, the physician should irrigate the proximal stump with Ringer’s solution and apply a pressure dressing with sterile gauze. The amputated part should be irrigated, wrapped in sterile gauze, placed in a plastic bag, and put on ice, being careful to keep the amputated extremity cool, but not allowing it to freeze.39,72 Peripheral vascular or nerve injuries should be suspected with joint dislocations, long-bone fractures, or crush injuries.72 Absent or reduced peripheral pulses, obvious external bleeding, unilateral edema, pallor or cyanosis, coolness, bruit, hematoma, hypoesthesia, or partial or complete paralysis should alert physicians to these injuries, which require urgent transfer and evaluation.39,72 Bleeding should be controlled with direct pressure. Effort-induced thrombosis is a serious vascular injury that should not be missed. It involves thrombosis of the subclavian or axillary veins and can present after trauma or insidiously with repetitive overhead or throwing activities.73 It may also occur as a result of heavy weightlifting. The condition can be associated with anatomic factors (e.g., first rib, pectoralis minor) or hypercoagulable states. Signs and symptoms include pain, circumferential swelling, generalized extremity weakness, and pallor. Urgent vascular studies such as Doppler ultrasound or venography are warranted in suspected cases. Acute compartment syndrome usually involves the lower leg or forearm but can occur in the thigh, foot, and

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

hand. Acute compartment syndrome can follow fractures, exercise, swelling, contusion, or muscle ruptures, or can be associated with constricting bandages (immobilization) or equipment.32 Compartment syndrome usually presents insidiously after trauma, sometimes as late as 72 hours after the injury.74 Pain out of proportion to the injury is the hallmark of a compartment syndrome. The affected compartment may be palpably tight. Inability to extend the fingers or toes may be an early clinical sign. Pulselessness, paresthesias, paralysis, and pallor are all late signs, and their presence may indicate some permanent damage to underlying soft tissues.32,74 If suspected, the extremity should be immediately elevated above the level of the heart, and the athlete should be emergently transferred for compartment pressure testing, fasciotomy, or both.

Environmental and Miscellaneous Emergencies Heat and cold injuries can pose serious problems for athletes. The best treatment for both is prevention through proper planning and good training principles. Adequate hydration, proper acclimatization and conditioning, daily prepractice weighing, adequate light-colored clothing, and close awareness of temperature and humidity are all important in preventing heat illness. If only ambient temperature and humidity are known, the heat stress danger chart in Figure 12-6 can be used. However, wet bulb globe temperature (WBGT) may provide the most useful index of potential for heat stress, and there are guidelines for exercise based on the calculated WBGT.75-77 The WBGT can be calculated using the formula: WBGT = 0.7 (WB) × 0.1 (DB) × 0.3 (BB), where WB = wet bulb, DB = dry bulb, and BB = black bulb. Cold injuries can be prevented by increasing body heat production and by decreasing body heat loss. Caloric intake, increasing muscular activity, adequate layering of clothing, adequate coverage and wind protection (head, face, extremities), staying dry, and proper warm-up can all help prevent cold injuries. Cold exposure may cause a range of injuries. Frostnip is a mild injury that develops slowly, and it most commonly affects the ears, nose, cheeks, fingertips, chin, and toes.78 Athletes complain of numbness associated with skin blanching and ice crystal formation on the skin’s surface. Treatment involves gentle warming by having the athlete place his or her hand in the axillary or groin areas, or by cradling the affected body part in a warm hand. Frostbite is a more severe cold injury, and it involves freezing of the superficial or deep soft tissues. Frostbite is classified into degrees of injury based on acute physical findings after freezing and rewarming.79 Superficial frostbite includes first- and second-degree injuries, and deep frostbite includes third- and fourth-degree injuries. First-degree frostbite exhibits localized burning and numbness, redness, and swelling. Second-degree frostbite is present if there is superficial blistering. Blisters are filled with a clear or milky fluid. Third-degree injury exhibits deep blistering, whereby injury has extended into the reticular dermis, and blisters are filled with a dark blood-tinged fluid. Fourth-degree frostbite extends into the underlying subcuticular tissues and can frequently involve muscle and bone.79 Fourth-degree frostbite may present with rigidity and a waxy yellow or mottled blue

HEAT STRESS DANGER CHART 100 Percent Relative Humidity

528

90 ZONE 3

80 ZO

70

NE

60

2

ZONE 1

50 40 65

70

75

80

85

90

Air Temperature (°F) Figure 12-6   Heat stress danger chart. Environmental conditions in zone 1 are fairly safe for participation. Normal heat stress precautions should be taken. In zone 2, moderate heat stress precautions should be taken. Workouts should be less intense, shorter, and with more frequent fluid breaks. More careful observation of individuals at increased risk is needed. In zone 3, heat stress danger is at its greatest. Workouts should be rescheduled to a cooler part of the day and should be relatively easy. Light clothing and a minimum of equipment should be worn. Extra fluids for everyone and close observation for early heat injury symptoms are essential. (From Mellion M, Shelton G: Thermoregulation, heat illness, and safe exercise in the heat. In Mellion M [ed]: Office Sports Medicine, 2nd ed. Philadelphia, Hanley & Belfus, 1996; adapted from Fox EL, Mathews DK: The Physiological Basis of Physical Education and Athletics, 3rd ed. Philadelphia, Saunders College Publishing, 1981.)

appearance.78 Tissue that is allowed to refreeze after thawing may be at risk for further damage. All grades of frostbite should be transferred to a hospital (if possible) and should undergo rapid rewarming using a whirlpool with water temperature of 40° to 42° C (104° to 108° F).80 The rewarming process can be extremely painful, and generous use of analgesics may be required. Rubbing or massaging frostbitten tissue or thawing frostbitten tissue using dry heat is absolutely contraindicated. Hypothermia may be mild (34° to 36° C), moderate (30° to 34° C), or severe (<30° C). Early signs and symptoms include cold hands and feet, shivering, chills, incoordination, dysarthria, tachycardia, and tachypnea. An athlete with a progression of these symptoms should be removed from the cold and windy environment, given dry clothes, and warmed. Warming in the field may be performed by placing the athlete in a sleeping bag with other warm ­bodies. Hot water bottles placed in the axilla, groin, and neck regions or special devices that deliver heated, humidified oxygen may also be used if available. Hot drinks are helpful, especially in the early stages. With moderate hypothermia, athletes develop increased fatigue, muscular weakness, finger and toe numbness, confusion, and incoordination. In severe hypothermia, shivering slows and then may stop, the respiratory rate slows, and blood pressure decreases. Cardiac arrhythmias such as sinus bradycardia, atrial fibrillation, and ventricular fibrillation may occur. As hypothermia progresses, the athlete becomes extremely confused and stuporous, markedly hypotensive, and bradycardic and the respiratory rate becomes markedly depressed.

The Team Physician

Athletes with moderate to severe hypothermia should be transferred immediately from the field. Further heat loss can be prevented by removal from the cold and wind, changing into dry clothes, and placement in a warm environment. ABCs should be monitored closely, intravenous access established early, and cardiac monitoring initiated. Do not administer intravenous fluids warmer than room temperature until in the hospital setting unless well trained and equipped. Confirm core temperature with a rectal probe. With moderate hypothermia, external rewarming using hot water bottles or heating pads placed in the axilla, groin, and neck regions can be attempted during transfer. The severely hypothermic patient, however, should be rewarmed only in a hospital setting because of the electrolyte, metabolic, and cardiovascular changes that can take place during rewarming.81 During rewarming, acid-base disturbances occur, the central circulation becomes overwhelmed by acidotic blood returning from the periphery, pulmonary edema often ensues, and the athlete is at extremely high risk for arrhythmias, especially fatal ventricular fibrillation. Handle the athlete gently and avoid physical jarring because it can trigger ventricular fibrillation. Pulses can be difficult to detect in severe hypothermia, and CPR should not be started prematurely because it can trigger dysrhythmias. Advanced cardiac life support proceeds in the standard fashion, with the exception that if ventricular fibrillation persists after three shocks, further shocks should not be provided until after rewarming to 30° C.40 Internal rewarming techniques such as heated and humidified oxygen, warmed intravenous fluids, peritoneal lavage, and esophageal rewarming tubes can be implemented in the hospital setting. Heat illness occurs on a continuum from heat stress to heat cramps, heat exhaustion, heatstroke, and sometimes death. Increased heart rate and blood pressure, mild dizziness, fatigue, and irritability are present with the mildest forms of heat injury. The athlete may experience cramps, muscle spasms, increasing weakness, fatigue, nausea, and vomiting. Core body temperature may remain normal. The athlete with heat stress, heat cramps, and mild heat exhaustion should be treated with mild cooling (e.g., stopping activity and retreating to a cool place), oral rehydration (e.g., sports drink with sodium; about 1 teaspoon salt to 1 quart water), and gentle stretching. These signs and symptoms may signal impending severe heat illness, and they should be respected. More severe heat exhaustion ensues with further sodium and water loss, and athletes may experience syncope, ­ dyspnea, piloerection, cutaneous flushing, decreased urine output, nausea and vomiting, and orthostasis. The athlete may be profusely sweating, and hypotension can develop. Core temperature is increased but is less than 105° F. Increasing irritability and headache are usually present, but there are commonly no serious mental status changes beyond mild confusion. Treatment involves more rapid cooling by removing the athlete to a cool environment, cool mist water sprays and fans, and oral rehydration using hypotonic fluids for several hours.81 Intravenous rehydration is sometimes required. Heatstroke is a true medical emergency. The affected athlete may collapse secondary to extreme hyperthermia and autoregulatory failure.82 Core temperature exceeds 105° F and is often much higher. Central nervous system

529

changes, including confusion, lethargy, seizures, and coma, are the hallmark of heat exhaustion. Classically, athletes have flushed, hot, dry skin, but some athletes retain their ability to sweat. Mortality rate increases with increasing temperature and duration of hyperthermia.83 There are often multiple associated electrolyte and metabolic abnormalities. The ABCs should be closely monitored. After determining the core body temperature, immediately cool the athlete by removal to a cool environment, using sprays of cool water mist and large fans to facilitate evaporation, and placing ice packs over major vessels in the axilla, groin, and neck. The athlete’s clothing should be stripped and he or she placed, if available, in an ice water bath or packed on ice (less effective).81,82 Place a large-bore intravenous line early, but avoid uncontrolled rapid fluid administration. Hypotensive patients may be challenged with 250to 500-mL boluses of normal saline. The airway should be closely monitored, and oxygen should be administered as required. Monitor core body temperature every few minutes, and discontinue external cooling when the temperature drops below 102° F and stabilizes. Overall, rapid cooling and early transport are the cornerstones of treating this condition, and they should occur concurrently, without letting one interfere with the other. Athletes will do best in a setting such as an intensive care unit that can provide close hemodynamic monitoring and support, continued cooling, proper airway support, and further evaluation of the multiple organ systems often involved. Hyponatremia occurs in 9% to 29% of ultraendurance athletes, and it has been reported in marathon runners.84-86 Hyponatremic collapse occurs in some of these athletes, but most athletes likely tolerate their electrolyte abnormality without significant symptoms or collapse. Athletes may experience confusion, disorientation, and slowed ­thinking. Severe cases may result in seizures or death. The most affected athletes may manifest a sodium depletion form of heat exhaustion, in which sodium depletion occurs rapidly over a short time.84 Although exact causes are unknown, water intoxication, sodium depletion, heat exhaustion, atrial natriuretic peptide, and antidiuretic hormone have been implicated.84-89 Water intoxication may occur when athletes consume excessive hypotonic fluids during an endurance event, and it is likely the most common cause of hyponatremia in endurance and ultraendurance events. Serum electrolytes should be checked whenever possible before giving intravenous fluids to athletes, especially confused athletes, affected by heat illness after ultraendurance events (e.g., more than 4 hours) and marathons. Fluids should not, however, be withheld from dehydrated hyperthermic heat exhaustion and heatstroke victims in need of volume replacement.82 Normal saline solution is likely the safest fluid to administer, but remember that cooling is the primary treatment of hyperthermic patients and that rapid, large-volume fluid administration in heatstroke victims is not warranted. If severely hyponatremic athletes are hemodynamically stable, fluids should be withheld, and they should be transferred for further evaluation and management. If seizures or other significant central nervous system symptoms occur, 3% normal saline can be administered slowly, but only in experienced hands. Heat syncope, also called exercise-induced syncope, is usually seen after an athlete stops exercising abruptly at the

530

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

end of an endurance event. Athletes are maximally vasodilated and may be dehydrated, and blood “pools” in the lower extremities. Venous return to the heart is decreased, which results in too little blood flow to the brain. Treatment involves having the athlete lie down in a cool place with his or her legs elevated, drink cold water, and rest.82 Heat edema occurs in unacclimated athletes exercising in hot environments. Affected athletes have increased aldos­ terone levels secondary to relative decreased plasma volumes (from profuse sweating) and peripheral vasodilation. Increased aldosterone results in sodium and water retention, which causes dependent edema in the hands and feet.82 The edema is usually transient, and it resolves over the first few days of heat exposure.82 Edema also may occur in waterintoxicated athletes after endurance and ultraendurance events. If mental status changes are noted in association with swollen extremities, water intoxication and associated hyponatremia should be ruled out until proved otherwise. Anaphylaxis can be idiopathic, triggered by an allergen (e.g., food, medication, Hymenoptera sting), or exerciseinduced. Bronchospasm, laryngospasm, and vascular collapse can occur in any combination or individually. Athletes may complain of sudden itching, hives, throat tightness, change in voice, and difficulty breathing. The reaction may result in generalized urticaria, severe bronchospasm and airway edema, vascular collapse, and gastrointestinal upset. The exact mechanism of exercise-induced anaphylaxis is not known, but an allergen such as a specific food ingested before exercise in hot, humid environments may play a role. Treatment involves rapid recognition, intravenous access, epinephrine subcutaneously (0.3 to 0.5 mL of 1:1000 solution), or intravenously (0.3 to 0.5 mL of 1:10,000 solution) with severe shock, diphenhydramine (25 to 50 mg intramuscular or intravenous), generous intravenous fluids as needed, and methylprednisolone or ­prednisone to prevent delayed signs and symptoms.47 The patient should be immediately transferred to a hospital for further monitoring.

ETHICAL AND LEGAL CONSIDERATIONS The Sports Medicine Team The team physician is not alone in caring for the athletes on the team. He is the key “player” on a sports medicine team that consists of the physician, the athlete, the coach, and the athletic trainer, when one is available. The care of athletes is a team effort in which members of the sports medicine team support each other for the benefit of the athlete and the athletic team. This team concept leads to the best possible care for the athlete.1,2,4,6,46 Each of the key players on the sports medicine team has a support system. The athlete’s support system includes teammates, family and significant others, friends, teachers, and the athletic trainer. The coach’s support system includes the athletic director, the school or league administration, the coaching staff and other coach colleagues, the equipment manager, and the athletic trainer. The team physician’s support staff is much more complex. It includes clinical support, research support, and the athletic

trainer. Elements of clinical support include other medical specialists, physical therapists, sports psychologists or psychiatrists, nutritionists, dentists, podiatrists, equipment managers, and health educators. The research support for a team physician includes medical researchers, exercise physiologists, sports psychologists, kinesiologists, nutritionists, physical educators, sociologists, and the athletic equipment industry. The athletic trainer occupies a unique position at the center of athletic health care. The athletic trainer is a therapist, counselor, and confidant for the athlete, an advisor and friend to the coach, and the “eyes and ears” for the team physician. In this last capacity, the athletic trainer provides triage and screening; supervises conditioning, care, and rehabilitation; and provides continuous functional evaluation of the athlete. Every high school and college athletic program should have a certified athletic trainer. Happy is the team physician who finds himself working with a talented, full-time, certified athletic trainer, who thereby lightens the physician’s burden immeasurably. The physician and athletic trainer can develop an excellent working relationship by establishing well-delineated lines of authority, responsibilities, procedures, and routines. Special relationships exist between physicians and coaches. These are as varied as human beings can make them and, ideally, are cordial and filled with mutual respect. The physician recognizes the goals and tasks confronting the coach, and the coach appreciates that the team physician is ultimately responsible for the health and safety of the athlete. In other words, both should work toward the common goal of safe participation and team success. One of the worst situations that can arise is an adversarial relationship between the coach and the medical staff.

Relationship of the Team Physician to Institution It is important for the physician to establish an explicit formal relationship with the school, league, or team.4 This agreement should include the job description, any fiscal arrangements, and a statement of expectations. When possible, especially if monetary arrangements are involved, the agreement should be in writing. If a formal contract is not appropriate, it is even more critical to discuss these items before the season begins. The team physician may be hired or solicited by a variety of people, including the athletic director, coach, athletic trainer, or business manager.6 Consequently, it is extremely important for the job description of the team physician to identify explicitly the person to whom the physician reports. The job description should also indicate what services are to be provided by the team physician both at home and away. Remuneration for services or travel and other benefits should be delineated. The job description should include all of the expectations that the institution has of the physician as well as those that the physician has of the institution.

Responsibilities of the Team Physician The ethical responsibilities of the team physician reflect the many relationships involved in the care of the team. Perhaps the foremost responsibility is to allow athletes

The Team Physician

to participate. In public and school-based programs, athletes have a right to participate if there is no valid medical contraindication against it. They even have the right to insist on participation by waiving the liability of the school or institution if they are injured or have a medical condition that increases risk. Consequently, the team physician should know both the sports medicine literature and the sport well enough to avoid disqualifying athletes from participating for insignificant or outdated reasons. On the other hand, the physician has an equally important responsibility to protect the athlete from injury, reinjury, and permanent disability. When there are valid reasons contraindicating participation, the athlete must be counseled and thoroughly informed about the risks and dangers. Sometimes it is necessary to protect the athlete from himself or herself. For example, it may be especially difficult to reason with an athlete who has a “participate at any cost” attitude. It is the team physician’s responsibility to provide optimal health care for the athlete. Superficial encounters are not acceptable. Good record-keeping is basic. Health care starts with the preparticipation physical examination, which should be thorough and timely. The physician is responsible for putting together a team of competent examiners who can meet with the athletes early enough before the competitive season to allow treatment and rehabilitation of the many defects that may be found. The physician’s responsibility for the athlete’s confidentiality is often a difficult problem in the sports setting. Generally, the physician has a relationship with the school or professional club that inhibits the normal strict rules governing doctor-patient confidentiality. The athlete should be given advance notice of any potential sharing of medical information, and it is best that this notice be in writing with the athlete signing an acknowledgment. Even with such an acknowledged release of information, the team physician must be sensitive to the issue of how widely medical information is disseminated. The physician, however, has a broader responsibility to contribute to the success of the team. The players, the coach, and the athletic trainer have all dedicated their time and effort to the sport. They place their trust in the physician to provide expert and compassionate care. Institutional responsibilities are also important for the team physician. The leagues, schools, and teams that have team physicians usually have made large commitments in personnel and finances. They may reasonably expect the team physician to be a contributor to the overall success of the athletic program. The primary job of the physician is to provide optimal care for the athletes. At the college and professional level, the physician is expected to prescreen potential scholarship and professional athletes to protect the institution from funding an athlete who is not definitely qualified to participate. Another major institutional responsibility of the physician is to protect the institution from unnecessary liability. Provision of high-quality, timely medical care is basic in this regard. Additionally, the preparticipation ­ evaluation may identify athletes who are at high risk for illness or injury and can then either provide successful treatment and rehabilitation or disqualify them from playing. Often the physician can provide safety

531

i­ nformation about the field of play or about dangerous athletic techniques—information that may help to prevent ­unnecessary injuries.

Legal and Medicolegal Considerations The team physician faces a variety of legal and medicolegal considerations. First among these is the issue of institutional and professional liability. When a physician functions as a team physician for pay or other tangible remuneration, the normal rules of medical liability apply. When the team physician, however, acts as a volunteer and treats an athlete without compensation, “good Samaritan” laws apply. Almost every state has a good Samaritan law, and more than three fourths have sports-specific laws. It is important to note that many good Samaritan laws require that the physician conform to the standard of care in the community. Preparticipation evaluations are not covered by good Samaritan laws even if they are performed without charge. Difficult issues occur when physicians travel with athletic teams outside of the states or countries in which they are licensed. For major athletic events, the host state or country generally passes legislation granting licenses to visiting team physicians. For routine competitions and tournaments, it is recommended that the traveling team physician work through the host team or tournament physician or a local physician in the host town. Good Samaritan laws apply in states that have them if the physician is not compensated. No suit has yet been brought against a team physician traveling with a team to another state over the issue of practicing without a license. Another item of legal significance is that of providing care to minors. Permission to provide care should be obtained from the responsible adults before the season, and appropriate forms should be carried with the team at all times.1,4 These considerations are magnified in the context of decision making about a return to competition after an injury has occurred. One must act prudently. On the sideline in the heat of battle, it is far better to err on the side of conservatism than to risk permanent injury. ­McKeag3 has suggested guidelines to help in making ­decisions in these return-toplay situations on the sidelines of sporting events: 1. A definite diagnosis has been made. 2. The injury cannot be worsened by continued play. 3. The athlete can compete fairly and protect himself. One other thorny legal issue is the athlete’s right to participate. During the past 2 decades, several judicial decisions have affirmed that an athlete has a right to participate in junior high school, high school, and college sports despite a variety of disqualifying conditions. All these cases have been decided in favor of the athlete, if he or she is ­willing to release the appropriate institution from ­liability. A recent case of national note involved a 17-year-old basketball player with hypertrophic cardiomyopathy who wanted to continue to play despite the fact that his older brother had died from the same disease in an exercise-related death. This case raises the question of taking reasonable judicial rulings to an unreasonable extreme.

532

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Team Physician Rewards Team physicians derive a variety of rewards from their service. Most find that the greatest reward is the immense personal satisfaction derived from providing a service to the community while working with young, highly motivated people. Rarely are tangible rewards able to match this element. Serving as a team physician is a labor of love. At anything less than the professional team level, most of the time spent by the physician will be as a “volunteer.” Above the high school level, some compensation may be part of the agreement. This will be extremely variable, however, ranging from pure volunteerism in some collegiate environments to a considerable retainer or fee-for-service arrangement with a professional club. Some colleges and universities provide team physician services through the student health organization or an affiliated medical school, but most of these arrangements include large amounts of “volunteer” time as well. Surgeons may receive fees for surgical procedures performed, but often these services are provided on a discount basis as well. The less tangible but equally important benefit of being a team physician is credibility, both in medicine and in the sport. Undoubtedly, an affiliation with a team, from high school to the professional level, enhances a physician’s prestige in the community and may contribute to practice building as well.

C

r i t i c a l

P

o i n t s

• The team physician serves as a key member of the health

care team and works with the athletic trainers and other providers to coordinate care of athletes. • The team physician is responsible for qualifying athletes to participate in athletic events, including the preparticipation examination. • The team physician is responsible for ensuring the proper preparation for and execution of emergency treatment protocols for athletes with acute injuries. • The team physician’s primary loyalty is to the health of the athlete. However, this will require discussion with the co­aches,­ trainers, team management, parents, and/or agents. • Although being a team physician can be demanding, this service benefits the athletes and teams served and can provide a great deal of personal satisfaction for the ­individual.

S U G G E S T E D

R E A D I N G S

Dunn WR, George MS, Churchill L, Spindler KP: Ethics in sports medicine. Am J Sports Med 35(5):840-844, 2007. Gomez JE: Sideline medical emergencies in the young athlete. Pediatr Ann 31(1): 50-58, 2002. Wingfield K, Matheson GO, Meeuwisse WH: Preparticipation evaluation: An ­evidence-based review. Clin J Sport Med 14(3):109-122, 2004.

R E F E R E N C E S Please see www.expertconsult.com

C H A P T E R  

 13

Basic Imaging Techniques S ecti o n

A

Basic Imaging Techniques in the Adult Jack Clement

Imaging plays a central role in the diagnosis and management of orthopaedic disorders and sports-related injuries. As people’s lives have become increasingly prolonged and physically active in our society, sports-related injuries have become more common among all age groups. Patients are increasingly educated with respect to their injuries and the imaging modalities used to diagnose them. Treating physicians have also become more adept at combining their clinical skills with imaging findings to effect better patient care. Finally, physicians have specialized in the fields of orthopaedic and musculoskeletal radiology to provide a higher level of service for their referring providers and their patients. Taken together, these events have resulted in an explosion in imaging ­utilization. The past several years have witnessed significant advancements in all imaging modalities. Magnetic resonance imaging (MRI), valued for its excellence in portraying differences in tissue contrast, has progressed to the level of 3.0 Tesla high-field strength magnets, significantly increasing spatial resolution. Computed tomography (CT) technology has also advanced, with multi-slice detectors now capable of generating 64, 128, and soon 256 slices through an anatomic space with a single gantry rotation, improving both spatial and temporal resolution. Conventional radiography has also witnessed its own revolution in the transition from film-based systems to full digital radiography, increasing the ability to visualize and diagnose injury and, perhaps more importantly, improving the rapidity of delivery of those images and their interpretation to the referring physician. Ultimately, patient care benefits from these advancements. In this chapter, we review basic imaging techniques, including radiography, ultrasonography, CT, nuclear medicine, and MRI. In the first section, we provide an overview of the techniques underlying image acquisition in the different technologies, with an extended discussion of more advanced modalities. Arthrography, fluoroscopy, and orthopaedic interventional procedures are also presented. In the second section, we address imaging strategies and

interpretative findings for several common sports-related injuries, including injuries to bone, ligaments, tendons, and cartilage. Strategies are discussed in terms of the modality most apt to provide a prompt and firm diagnosis while minimizing patient discomfort and cost. Finally, in the third section, we provide a brief discussion of the transition to digital imaging and the use of PACS (picture archive and communication system) for image acquisition, storage, and distribution.

IMAGING TECHNIQUES Conventional Radiography Routine radiographic evaluation remains the mainstay for the initial evaluation of the orthopaedic sports-related injury.1 Discrete protocols are well established for the evaluation of a particular anatomic location.2,3 At least two views, usually in orthogonal projections, are required to evaluate any bone or joint adequately. Fractures or dislocations easily identified on a single view may be frustratingly occult on a second view (Fig. 13A-1). Limiting examinations to a single view in pursuit of either speed or decreased patient dose will lead to missed pathology and increase the risk for poor outcome.4 Occasionally, specialized views may be needed to evaluate more subtle lesions or to examine specific anatomic locations. Examples are the scaphoid view in the wrist, Judet views in the hips, and either the sunrise or notch view in the knee. These views should be specifically requested when desired. Radiographs are obtained by passing a highly modified flux of x-ray photons through tissue. Differential transmission results in the radiograph. Bombarding a tungsten target with a beam of electrons and allowing only those emitted in the direction of the film to proceed create the x-rays. These events all take place within an x-ray tube. Image quality is highly dependent on patient positioning as well as on film exposure. Motion should be eliminated. 533

534

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13A-1     Carpometacarpal dislocation. A, Frontal radiograph of the right hand shows subtle malalignment of the third, fourth, and fifth carpometacarpal joints.   B, Lateral radiograph of the hand shows gross dislocation of the carpometacarpal joints with dorsal displacement.

A

Exposure is controlled by altering the number of incident photons or the energy of the incident x-ray photons, or both. The quantity of incident photons is governed by the strength of the current used to generate the electron beam, whereas the energy of the x-ray photons is governed by the strength of the voltage potential through which the electrons are accelerated before striking the tungsten target. The current is typically abbreviated as mAs (milliamperessec) and the voltage as kVp (kilovoltage potential). Once x-rays are formed, their individual energy cannot be modified. However, filters that provide differential absorption can modify the quality of the collective beam. These filters function to remove low-energy x-rays, which would contribute only to patient dose and not to image formation. As a general rule, thicker and denser structures generally require both more energetic and a greater of quantity of x-ray photons for a diagnostic image. The converse is true for thinner and less dense anatomic locations. The advent of digital imaging permits more latitude in film exposure, with images acquired over a larger range of exposure values remaining diagnostic. The radiograph is primarily used in evaluation of osseous injuries. However, changes in soft tissues can also be inferred by examining the changes in normal ­fat–soft tissue interfaces. Thus, soft tissue masses, joint effusions, and often tendon contours can be visualized (Fig. 13A-2). This may raise concern for soft tissue or osseous injury and predicate the need for further imaging. In the setting of fractures, serial radiographs are often used to evaluate the healing process. Additionally, plain film radiography is often diagnostic in the evaluation of bone neoplasms and plays a central role in the evaluation of arthritides.5,6

Conventional Tomography Conventional tomography is a radiographic technique used to improve visualization of structures in a particular focal plane while blurring structures above and below

B

the plane of interest. This is accomplished by moving the x-ray source and the detector in opposite directions during radiographic exposure. This technique is infrequently used in current practice, replaced largely by CT with multiplanar re-formations and MRI with multiplanar acquisitions. However, tomography may occasionally be requested in the evaluation of subtle fractures, to assess healing across a fracture plane, or to look for subtle mineralization in lytic bone lesions.7-9

Fluoroscopy Fluoroscopy is a technique in which radiographs are captured in real time and displayed on a monitor. Frame rates range from 4 to 12 per second. Higher frame rates result in a smoother picture but increase patient dose. Still frames can also be acquired and typically have higher spatial resolution than those of continuous imaging. The spatial resolution of a fluoroscopic image is significantly less than that of a radiograph. This is mainly to minimize patient radiation dose during fluoroscopy. Fluoroscopy has become an essential tool in the arsenal of both the radiologist and the orthopaedic surgeon. Musculoskeletal radiologists often use fluoroscopy for needle guidance into myriad joints for purposes of arthrography, arthrocentesis, and medication delivery. Fluoroscopy is also often used for precision spine procedures such as epidural steroid injections, facet joint injections, and nerve root blocks (Fig. 13A-3). Non–image-guided joint injections have failure rates up to 30%, even in large joints such as the shoulder. Failure rates are similar for non–imageguided interventional spine procedures.10 Fluoroscopy may also be used in the dynamic evaluation of joints to evaluate for instability.11 Orthopaedic surgeons often use intraoperative fluoroscopy to better visualize the underlying osseous anatomy. This is commonly employed in open fracture reduction and fixation, arthrodesis and spinal fusion, and joint replacement surgery.

Basic Imaging Techniques

A

535

B

Figure 13A-2   Elbow joint effusion with occult radial head fracture. A, Lateral radiograph of the elbow demonstrates large elbow joint effusion with displacement of anterior and posterior fat pads (arrows). Discrete fracture not visualized. B, Sagittal T2-weighted fat-saturated magnetic resonance image of the same elbow confirms large elbow joint effusion and demonstrates a nondisplaced fracture to the anterior aspect of the radial head (arrow).

Arthrography

Figure 13A-3   Left L4 nerve root block (transforaminal epidural steroid injection). Posteroanterior fluoroscopic spot image of the lower lumbar spine with curved-tip 22-gauge needle in the left L4-L5 neural foramen and left L4 neurogram after the injection of 0.5 mL of iodinated contrast. This is followed by injection of a steroid and anesthetic preparation along the nerve root with central flow along the epidural space.

Arthrography is a technique in which a contrast agent is injected into a joint to provide better visualization of the structures that stabilize the joint as well as potential damage to those structures. Before the advent of CT and MRI, the radiographic arthrogram was used to diagnose ligamentous and tendinous injury by noting abnormal contrast patterns and inferring the subsequent injury.12 However, the combination of intra-articular contrast and either CT or MRI now allows exceptional depiction of intra-articular anatomy and pathology. Ligamentous and tendinous injuries need no longer be inferred by abnormal arthrograms; they can be directly visualized and treatment precisely planned. As a technique, arthrography is minimally invasive, usually performed with either 25- or 22-gauge needles, and is well tolerated by the patient. It can routinely be performed in less than 5 minutes and is coordinated with the patient’s subsequent cross-sectional imaging examination. An iodinated contrast material is injected, occasionally with air, if the subsequent examination is with CT. For MRI examinations, a very dilute gadolinium mixture is injected into the joint (Fig. 13A-4). Often, bupivacaine (Marcaine), an ­ anesthetic, is admixed with the contrast to minimize discomfort. Imaging should be performed within about 30 to 45 minutes after the injection because the vascular synovium begins absorbing the injected contrast immediately after the procedure. A small amount of epinephrine may be included with the injectate to constrict synovial vessels and prolong resorption.13 Shoulder arthrography has been used for many years in the diagnosis of rotator cuff tears by noting abnormal

536

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13A-4   Shoulder arthrogram. Anteroposterior   fluoro­scopic image of the right shoulder demonstrating a   25-gauge needle in the glenohumeral joint with iodinated contrast flow in the joint.

contrast transit from the glenohumeral joint into the subacromial subdeltoid bursa. However, arthrography currently is infrequently performed without subsequent cross-sectional imaging to provide further anatomic data. CT arthrography using air and iodinated contrast materials and magnetic resonance (MR) arthrography using dilute gadolinium agents are used to improve detection of rotator cuff, labral, capsular, cartilage, and osseous abnormalities. MR arthrography is the examination of choice for evaluation of the shoulder (Fig. 13A-5). However, CT arthrography finds common use in patients who cannot undergo MR examinations (claustrophobia, pacemaker, aneurysm clips). Arthrography of the shoulder may also be

A

B

performed in the treatment of adhesive capsulitis, in which a combination of lidocaine, saline, and a steroid agent is injected into the shoulder to effect distention, often with capsular rupture.14 This procedure is termed shoulder brisement and often improves patients symptoms. Knee arthrography was historically used in the detection of meniscal tears. However, MR has replaced conventional knee arthrography as the imaging modality for evaluation of the menisci. In patients who have persistent pain after meniscal repair, however, MR arthrography of the knee is commonly used in the differentiation of recurrent meniscal tears from the postoperative meniscus.15 Additionally, MR arthrography is also commonly used in the assessment of fragment stability in osteochondral lesions.16 CT arthrography of the knee is often performed in patients who cannot undergo MR assessment. Additionally, CT arthrography with coronal and sagittal re-formation provides excellent delineation of articular surface cartilage (Fig. 13A-6). Wrist arthrography is commonly used in the ­­diagnosis of intercarpal ligamentous injury. Based on the pattern of contrast flow from one wrist compartment to the next, deficiencies in the ligamentous structures that define the boundaries of the compartment are inferred.17 Injuries may be traumatic or degenerative in origin. MR arthrography of the wrist adds significant diagnostic accuracy in the assessment of intercarpal ligaments.18 The presence of joint contrast distends the joint and outlines the ligamentous structures. Additionally, MR affords assessment of adjacent cartilage, bone, and tendinous structures. With advancements in MRI, especially with high-field 3.0 Tesla magnets and dedicated wrist coils, MR arthrography of the wrist should continue to improve in diagnostic accuracy. MR arthrography of the hip is commonly performed in younger patients in whom acetabular labral tears are suspected. By distending the joint, contrast is forced into the tear, rendering it more conspicuous than in nonarthrographic images. In patients with suspected femoral­acetabular impingement, MR arthrography provides a complete diagnostic evaluation of the acetabular labrum, acetabular cartilage, and femoral morphology.19 Additionally, MR arthrography demonstrates intraosseous and

C

Figure 13A-5   Normal magnetic resonance arthrographic images of the shoulder in the axial (A), coronal (B), and sagittal   (C) planes. All images are T1 weighted. The coronal and sagittal images also are fat saturated. Note that the contrast within the joint is the only bright structure on the fat-saturated magnetic resonance images, whereas fat is also bright on the non–fat-saturated image.

Basic Imaging Techniques

A

B

537

C

Figure 13A-6   Computed tomographic arthrogram of the knee. Computed tomographic images of the knee after injection of iodinated contrast in the axial (A), coronal (B), and sagittal (C) re-formation projections. In the axial projection, the patellofemoral cartilage is well visualized, with contrast outlining a large cartilage flap along the lateral patellar facet (arrow). Sagittal and coronal   re-formation images demonstrate areas of full-thickness cartilage loss along the posterior weight-bearing surface of the medial femoral condyle with subchondral sclerosis and cystic change (arrows).

extra-­articular pathology such as bursitis and tendon tears that may also be a source of hip pain. The primary use of ankle arthrography is in assessment of lateral ankle pain and instability after ankle sprain.20 Tears of the lateral ligaments are well visualized with MR arthrography, and their extension into the syndesmosis can be assessed. Additionally, osteochondral lesions of the talar dome are evaluated with respect to size and stability of the osteochondral fragment.21 Finally, MR arthrography also provides a simultaneous assessment of the remaining tendons, ligaments, and joints about the ankle.

Ultrasonography Ultrasound is an imaging modality that uses high-­frequency pressure waves to generate images. Ultrasound probes are made from piezoelectric crystals that vibrate when an electrical current is placed through them. When vibratory pressure is reapplied to the crystal through the action of reflected sound waves, an electrical pulse is produced and detected. Thus, the ultrasound probe serves as both transducer and receiver. As pressure waves pass through subjacent tissues, a portion of the wave is reflected at acoustical interfaces between soft tissue structures. This reflected wave is detected, amplified, and used to create the image. Ultrasound probes operate in the 2.5- to 10-MHz range. The frequency of the transducer is directly proportional to image resolution while inversely proportional to imaging depth. Thus, high-frequency transducers commonly used for musculoskeletal applications have high spatial resolution but limited depth of penetration to only a few centimeters.22,23 Deeper imaging requires lower frequency transducers that produce images with diminished resolution. Additionally, ultrasound probes can be used to assess arterial and venous waveforms in tissues using color Doppler techniques. This plays an important role in differentiating cystic lesions from vascular lesions.

Ultrasound is an underutilized imaging modality for musculoskeletal pathology, largely supplanted by MRI, which provides a more global assessment of underlying anatomy. Additionally, ultrasound is comparatively more operator dependent, with the best results obtained from ultrasound technologists who perform musculo­skeletal imaging routinely. Many musculoskeletal applications are particularly suited to ultrasonography.24,25 This includes not only diagnostic imaging but also ultrasound-guided interventions such as steroid injections, fluid drainages, and biopsies. Importantly, requests for musculoskeletal ultrasound must be tailored to assessment of a specific anatomic structure and not requested as a survey ­technique. Ultrasound is particularly useful in the differentiation of solid from cystic lesions. In the knee, it is commonly used to assess palpable or painful masses in the popliteal fossa. Most commonly, this represents a Baker’s cyst. However, more sinister lesions such as a tumor or popliteal artery aneurysm are other possibilities that can be differentiated by ultrasound. Baker’s cyst, also termed popliteal cyst, is a fluid-filled potential joint space along the posteromedial knee between the semimembranosus and medial gastrocnemius tendons. These cysts may be a source of pain, especially if they rupture or hemorrhage, often extending far down into the calf. These cysts look like most other simple fluid collections, uniformly anechoic (black), well defined, and with prominent posterior acoustic enhancement (soft tissues behind the cyst look brighter because no echoes are reflected within the cyst) (Fig. 13A-7). Masses usually have internal echoes and internal blood flow, distinguishable by the use of color Doppler. Aneurysms demonstrate arterial flow unless they have thrombosed. Ultrasound may also be used in the assessment of other musculoskeletal cystic structures such as parameniscal cysts around the knee, paralabral cysts in the shoulder or hip, and ganglion cysts,

538

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B Figure 13A-7   Baker’s cyst aspiration. A, Ultrasound image of the posteromedial knee demonstrates elongated, well-defined, anechoic (black) fluid collection consistent with a Baker’s cyst. B, A needle has been advanced into the cyst under ultrasound guidance (arrow). Reverberation artifact is seen in the tissues deep to the metallic needle. C, The cyst has been completely decompressed with no demonstrable fluid remaining. This is usually followed by an injection of a steroid and anesthetic preparation.

C

bursal fluid collections, or soft tissue fluid collections. More complex cysts with either hemorrhagic or synovial debris have variable amounts of internal echoes. Ultrasound proves uniquely valuable in sonographic real-time guidance of needles into fluid collections to accomplish drainage or medication delivery (usually steroids). Using imaging guidance provides documented proof of technical success and minimizes patient discomfort by allowing direct visualization of the procedure in real time, thus minimizing needle manipulation. Almost any tendon can be visualized with ultrasound. Commonly imaged structures include the Achilles tendon, infrapatellar tendon, and rotator cuff tendons. A normal tendon appears as a relative echogenic (bright) structure with longitudinally oriented strands. There is typically no demonstrable fluid (anechoic or hypoechoic) around the tendon. Tendinopathy appears as thickening or thinning of the tendon with blurring and distortion of the normal longitudinal strands. If tenosynovitis is present, there is usually fluid extending longitudinally along the tendon sheath (Fig. 13A-8). This may be injected under ultrasound guidance. Tendon tears usually appear as hypoechoic (dark) areas within the tendon. In partial-thickness tears, some of the longitudinal fibers remain intact. In full-thickness tears, there is a complete hypoechoic gap, often with tendon retraction. If there is granulation tissue in the tear, it may appear heterogeneously hyperechoic.

Ultrasound also plays an occasional role in foreign body localization.26 Although metal and glass can usually be seen radiographically, wood is infrequently detected because its attenuation is similar to that of soft tissues. Ultrasound detects all three materials, providing confirmation and localization to aid removal (Fig. 13A-9).

Figure 13A-8   Tenosynovitis of the posterior tibial tendon. Ultrasound image of the posterior tibial tendon demonstrates normal echogenic tendon with longitudinally oriented fibers. There is irregular hypoechoic and anechoic fluid around the tendon indicating tenosynovitis (arrows).

Basic Imaging Techniques

Figure 13A-9   Soft tissue foreign body. Ultrasound image of the index finger demonstrates a small echogenic foreign body in the subcutaneous tissues surrounded by hypoechoic fluid (arrow). This corresponded to a deep wood splinter.

Computed Tomography CT was developed through the work of Sir Godfrey ­Hounsfield and has revolutionized imaging since its introduction about 35 years ago. It provides superb spatial and contrast resolution, displaying anatomy in two-­dimensional slices, which are then visualized in stacked sequence to create a three-dimensional representation of the underlying anatomy. Musculoskeletal CT applications abound, but it is most commonly used to evaluate complex fractures and radiographically occult fractures. CT images are acquired by rotating an x-ray tube and an opposing detector quickly around a patient. The basic components are housed in a toroid-shaped gantry. The x-ray tube is designed to emit a very narrow x-ray beam that passes through the patient and is picked up by the detector. The x-ray beam emits continuously as it encircles the patient, with one full rotation needed to acquire the data for a single slice. Differences in x-ray absorption (attenuation) as the beam rotates around the patient are measured by the detector, and the tomographic image is then backcomputed mathematically (using a Fourier transform). Early CT scanners acquired a single slice, then the patient was incrementally moved a few millimeters and the second slice was acquired, and so on until the anatomic area of interest was imaged. This required a large amount of time for image acquisition and resulted in significant image degradation due to motion artifacts. In the past 10 years, technical advances in CT engineering have led to continuous helical CT scanning.27 With this technique, the x-ray tube and detector are continuously rotated as the patient simultaneously smoothly moves through the scanner, creating a helically shaped pattern of data acquisition and resulting in a significant reduction in imaging time. More recent advances have led to multi-slice acquisition techniques in which multiple detectors are used in an array to acquire multiple slices with each rotation of the gantry. The gantry is capable of rotating at speeds of up to 330 milliseconds per rotation. Currently, 64 detector arrays are commercially available, able to acquire 64 slices with each rotation, and larger arrays are in development. The ­advantages are very

539

fast acquisition times and very thin collimation, resulting in exquisitely detailed anatomic images. CT image quality is influenced by several factors. Protocols are designed for specific anatomic locations to maximize image quality and minimize patient dose. Musculoskeletal applications usually use narrower 1- to 3-mm collimation to improve image quality and spatial resolution, especially in small body parts such as the hands or feet. Larger body parts are often scanned with larger collimation of 3 to 5 mm.28,29 Once the helical data set is obtained, images can be reconstructed in any plane, usually axial, sagittal, and coronal. Reconstruction parameters significantly influence the final image quality. Finally, filtering algorithms are applied to the data set to maximize resolution of bone or soft tissue, depending on clinical interest. CT images are visualized on a computer monitor, allow­ ing for image manipulation. Each pixel of the image is assigned a Hounsfield unit (HU) value, which reflects the attenuation of the x-ray beam at that unit of the image. More dense tissues (bone) have higher HU values than do less dense tissues such as fat or air. Dense tissues appear whiter, whereas less dense tissues appear blacker. Thus, grayscale CT images display anatomic data reflecting un­der­lying tissue composition. The displayed image can be manipulated by selecting the range of HU the grayscale map is applied to (window width) and what value should be assigned to the middle of the grayscale map (window level). For visualizing bone trabeculae, a high window width of 1000 to 2000 HU is used, and the window level is set to about 250 HU. For good soft tissue contrast, a narrower window width of 400 to 600 HU is used, and a window level of about 50 HU is used (Fig. 13A-10). With current workstations, window width and window level are changed dynamically by the x- and y-axes of the computer mouse. The subsequent changes to the image are displayed in real time until a satisfactorily windowed image is displayed. CT plays an essential role in the evaluation of traumatic injury.30 Although conventional radiographs remain the initial imaging study for the assessment of skeletal trauma, CT provides a more detailed and therefore diagnostically sensitive examination. In general, the more complex the joint, the more likely that a CT examination will provide additional diagnostic information and refine clinical management. Traumatic assessments of the head, cervical spine, chest, abdomen, and pelvis can be acquired in minutes. Datasets acquired from the axial skeleton can be used to reconstruct thoracic and lumbar spine images in sagittal and coronal planes. In the acute trauma patient, CT allows for the assessment of both osseous and soft tissue injury, providing a more comprehensive patient evaluation for operative planning.

Spine In the acute trauma patient, CT is commonly used for assessment of the cervical spine. Historically, the three-view cervical spine series (lateral, anteroposterior, and odontoid) was used in this regard. However, these examinations are often limited by patient habitus, lack of cooperation, or technical factors. Several studies have demonstrated the superiority of cross-sectional imaging in the detection of spinal fractures.31 Despite higher costs, most trauma

540

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-10   Effect of windowing on computed tomographic appearance of tissues. A, Axial computed tomographic image of the knee windowed for soft tissue evaluation in a patient with a tibial plateau fracture. There is a moderate suprapatellar effusion demonstrating a layering hematocrit effect. Fat reflecting fracture extension into the marrow floats on top, serum (fluid) in the middle, and slightly more dense blood products more dependent (arrows). B, The same image as in A, windowed for evaluation of bone detail.

c­ enters have replaced conventional radiography with CT as the initial cervical spine screening study in patients with high risk for injury.32 CT provides several advantages, including improved diagnostic accuracy, decreased imaging time, and minimal movement of the traumatized patient.33 Additionally, the examination can be performed in conjunction with assessment of other anatomic regions. In the cervical spine, helical images are obtained from the skull base through the first thoracic vertebra in the axial plane. These are then reformatted into two-­dimensional sagittal and coronal multiplanar images, similar to the

A

B

anteroposterior and lateral radiographic views. All three planes must be evaluated because pathology easily seen in one plane may be completely occult in another. Fractures parallel to the plane of imaging are often occult within that plane. For example, a transverse fracture to the base of the odontoid may be invisible on axial images, whereas in the sagittal and coronal re-formation planes, the fracture is clearly visible, and displacement can be assessed (Fig. 13A-11). Axial images are useful in assessment of fracture extension into the vertebral foramina, thus predicating the need for arterial assessment. The axial data acquired during

C

Figure 13A-11   Odontoid fracture. Axial (A) computed tomographic image of the cervical spine at the level of the odontoid process with sagittal (B) and coronal (C) re-formations. There is a nonunited subacute fracture to the base of the odontoid, which is easily visible on the re-formation images (arrows). The fracture is not detectable on the axial image because it lies along the plane of image acquisition.

Basic Imaging Techniques

the initial scan may be used to generate three-dimensional rotational images to provide the clinician with a complete spatial representation of the underlying anatomy. However, the three-dimensional images typically are not used for diagnostic purposes because there is too much structure overlap, which obscures subtle underlying pathology. CT is less commonly used as the initial screening examination in the thoracic spine and lumbar spine. However, when screening radiographs reveal the possibility of underlying fracture, CT is used to further characterize the underlying injury. Often, CT will reveal not only the initial injury but also adjacent injuries not seen on the initial radiograph. As with the cervical spine, axial, sagittal, and coronal images are used for complete assessment. CT plays a crucial role in the evaluation of the fracture pattern, fragment retropulsion, and possible impingement on the spinal canal and spinal cord (Fig. 13A-12). In the nontrauma patient, CT is often used to assess spinal anatomy. Lytic bone lesions are reliably evaluated with respect to degree of osseous destruction, internal mineralization, soft tissue extension, and spinal canal compromise.34 In some cases, a definitive diagnosis can be made by the CT appearance. If the lesion appears aggressive, however, CT can then be used to guide biopsy of such lesions. In patients with back pain unable to undergo MR examinations, CT is usually performed after myelography.35 This provides a detailed assessment of disk and facet pathology and subsequent spinal canal and neural foraminal compromise. In patients with persistent pain after fusion, CT can be used to assess for hardware loosening or malpositioning, integrity of the fusion bone mass, spinal canal stenosis, and pseudarthrosis formation.36,37

Pelvis Plain radiographs are the initial means of evaluating the patient with a suspected pelvic fracture. Usually, this consists of a single anteroposterior view, often obtained in the

A

B

541

trauma room. Supplemental views such as inlet and outlet views, as well as Judet oblique views of the acetabula, are performed infrequently. CT scan is essential in the diagnosis and preoperative planning of pelvic, acetabular, and sacral fractures. Studies have shown that plain films miss up to one third of pelvic fractures, including almost two thirds of acetabular rim fractures.38 CT is indicated in acetabular fractures, hip dislocations, suspected sacral or sacroiliac joint injuries, and questions of pelvic ring stability.39 Images are acquired in the axial plane, usually in combination with imaging of abdominal structures. Reconstructions in the sagittal and coronal planes can be acquired from the raw data and provide a better picture of fracture anatomy and fragment displacement. In the setting of pelvic fractures, much attention focuses on preventing hemorrhage into the retroperitoneal space. Such hemorrhage can be life-threatening and difficult to control. Control of arterial hemorrhage and fracture reduc­ tion are important management issues. Venous hemorrhage is often well controlled with pelvic stabilization and pelvic volume reductions. CT performed with intravenous contrast allows for assessment of soft tissue structures and arterial injury.40 Arterial injuries demonstrate contrast extravasation and pseudoaneurysm formation. The CT scan provides important diagnostic and anatomic data and guides percutaneous angiographic intervention. The apophyses of the pelvis are traction epiphyses that act as the attachments of muscles and tendons. Compared with epiphyseal centers, apophyseal ossification and fusion occurs later, usually during or just after the adolescent years of increasing sports-related activities. Thus, apophyses are common sites of both acute and chronic avulsive injuries. Although conventional radiography is often all that is needed in the evaluation of these injuries, CT is often useful in assessing the size of the avulsed fragment as well as displacement (Fig. 13A-13). Additionally, healing avulsion injuries often demonstrate both areas of bony lysis and sclerosis with abundant periosteal new bone formation,

C

Figure 13A-12   Lumbar spine burst fracture. Computed tomographic images of the lumbar spine in axial (A), coronal reformatted (B), and sagittal reformatted (C) projections show a comminuted burst fracture of the L2 vertebral body. There is significant fragment retropulsion into the spinal canal with greater than 50% narrowing of the canal, best seen on the axial and sagittal images (arrows).

542

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 13A-13   Popliteus tendon avulsion. Axial (A) and coronal reformatted (B) computed tomographic images of the knee demonstrate an avulsion fracture arising from the lateral femoral condyle posteriorly. The insertion of the popliteus tendon into the fragment is confirmed on the coronal T2-weighted fat-saturated magnetic resonance image (C).

occasionally mimicking neoplastic bone. CT is often useful in more definitively characterizing these lesions as healing apophyseal avulsion injuries.41

Peripheral Joints Injuries to the appendicular skeleton are usually assessed using conventional radiography. Exceptions occur in the setting of occult fractures as well as complex fractures for which detailed cross-sectional imaging provides further diagnostic information. As a general rule, the more complex the anatomy in a particular location, the more likely that CT will provide diagnostic information beyond that obtained with the radiograph. In the setting of acute trauma, the benefits of CT include its ability to delineate fracture extent and to characterize the complexity of a fracture. This includes assessment of fracture comminution, fragment displacement and rotation, intra-­articular extension and articular surface incongruity, physeal extension, osteochondral intra-articular loose bodies, and concomitant soft tissue damage. In the healing phase, CT provides an assessment of the healing process, including fragment alignment; the presence of bridging callus; and the integrity and positioning of internal fixation hardware. Three-dimensional reconstructions, although not usually helpful in the initial diagnosis, often serve as a useful surgical planning tool to improve the understanding of complex fractures and anatomic relationships. CT also plays an important role in the assessment of arthroplasty devices, including alignment, subtle fractures, polyethylene wear, and particle disease (Fig. 13A-14). In the lower extremities, nondisplaced hip fractures are often radiographically occult because of the presence of osteopenia. MRI is usually the imaging study of choice in this setting. However, patients unable to undergo MR examinations can be assessed with CT to improve ­diagnostic sensitivity.42 CT provides a better assessment of degenerative changes in the hip that may change management from internal fixation to arthroplasty placement. Fractures of

the knee are usually completely assessed radiographically. However, CT is used in the setting of tibial plateau fractures to assess degree of comminution, articular surface depression, and fragment displacement.43 Coronal and sagittal reconstructions are useful in assessing the integrity of the plateau (Fig. 13A-15). In the ankle and foot, CT is helpful in the evaluation of comminuted calcaneal fractures.44 In the midfoot, CT is often helpful in assessing fractures difficult to characterize because of overlapping structures on radiographs. This includes Lisfranc’s joint, in which CT can demonstrate subtle avulsion injuries to the base of the

A

B

Figure 13A-14   Particle disease. Computed tomographic coronal (A) and sagittal (B) re-formation images of the right hip showing total hip arthroplasty in place. There is asymmetric polyethylene wear as well as several scalloped lucencies within the bony acetabulum secondary to lysis from particle disease (arrows).

Basic Imaging Techniques

A

543

B Figure 13A-15   Tibial plateau fracture. Axial (A), coronal (B), and sagittal (C) computed tomographic images of the knee demonstrate a comminuted lateral tibial plateau fracture with significant depression of the articular surface.

C

second metatarsal, and reconstruction images provide an assessment of Lisfranc’s joint alignment.45 In the upper extremities, shoulder dislocations are often associated with osseous injuries. CT is often useful in detecting Hill-Sachs impaction fractures and Bankart’s glenoid fractures. Combined with arthrography, CT can be used to assess for cartilaginous and labral injuries in the dislocated shoulder as well as intra-articular loose bodies.46 The elbow is the third most dislocated joint after the glenohumeral joint and interphalangeal joints. It is the most dislocated joint in a child. CT is often used in the characterization of complex elbow fracture-dislocations. It provides excellent depiction of fracture comminution, fragment displacement and rotation, and the presence of displaced intra-articular fragments that could lead to impingement and limited range of motion. In the wrist, CT has an established role in the assessment of occult fractures, including nondisplaced scaphoid fractures, hook of hamate fractures, and triquetral avulsion injuries (Fig. 13A-16). In the hand, CT may allow detection of subtle avulsion injuries along the insertions of the flexor and extensor tendons. This typically requires thin sections and

strict attention to image acquisition and reconstruction technique.

Infection The diagnosis of osteomyelitis is more readily made using nuclear medicine and MRI. However, CT will show changes reflecting underlying infection, including an increase in marrow fat attenuation secondary to underlying infection. In severe infection, bone dissolution may be seen. In the setting of chronic osteomyelitis, CT may be used to identify sequestra, bone abscesses, soft tissue abscesses, and sinus tracts.47

Nuclear Medicine Nuclear medicine examinations differ from other imaging modalities in that they reflect functional information as opposed to detailed anatomic information. Anatomic resolution is actually quite poor. However, the benefits of skeletal scintigraphy are its exquisite sensitivity to changes in bone turnover that may result from a number of ­pathologic

544

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-16   Hook of hamate fracture. Computed tomographic axial (A) and sagittal (B) images of the wrist demonstrate a nondisplaced fracture along the base of the hook of hamate (arrows).

entities, including fractures, infections, tumors, and arthritides. Additionally, skeletal scintigraphy offers the possibility of imaging the entire skeleton simultaneously, providing an overall skeletal survey that is often useful in the assessment of occult fractures, polyarthritis, and osseous metastatic disease. Unlike radiographic and CT imaging examinations that depend on transmitted photons from an exogenous source, nuclear medicine studies rely on imaging gamma photons that originate from an administered radioactive agent. Radiopharmaceuticals may be administered orally, intravenously, or subdermally in the case of lymph studies, or they may be inhaled in ventilation imaging. Skeletal scintigraphic examinations, also termed bone scans, rely on the intravenous administration of bisphosphonates linked to technetium-99m, either technetium-99m methylene diphosphonate or technetium-99m hydroxymethylene diphosphonate. These agents are phosphate analogues that are incorporated into newly formed hydroxyapatite ­crystals in bone.48 Because bone is a very dynamic tissue, this incorporation can be imaged in as little as 15 to 30 minutes after radiotracer injection. However, the presence of significant soft tissue background precludes rapid image acquisition. Degree of bone uptake is affected by blood flow and osteoblastic activity. Unincorporated radiotracer is excreted through the kidneys with a half-life of about 4 hours. Patients are instructed to drink a lot of fluid to facilitate background clearing with images obtained 2 to 4 hours after injection. Bone scans may be performed using a three-phase imaging protocol. This includes blood flow images, blood pool images, and delayed images. Flow images are obtained, not of the whole body but of a particular area of pathologic interest such as a diabetic foot. Images are obtained for about 1 minute after injection and reflect only regional blood flow. Blood flow may be increased (hyperemia) or decreased depending on the pathologic state and is often compared with the contralateral extremity. Flow is

­ sually increased in the setting of inflammation, infection, u or injury. Blood pool images are static images obtained immediately after first-pass flow images. These images reflect soft tissue accumulation of radiotracer secondary to leakage from blood vessels. In the setting of soft tissue inflammation and neovascularization, there is increased blood pool activity. Finally, delayed-phase imaging is performed 2 to 4 hours after injection to allow background activity to clear and provide an unobscured representation of bone activity. Whole-body images are obtained in anterior and posterior projections because there is significant absorption of the gamma photons by the body. Often, spot images of a particular anatomic location are obtained to achieve better anatomic resolution and diagnostic sensitivity (Fig. 13A-17). Patients undergoing bone scans undergo imaging with a gamma camera. This consists of a sodium iodide crystal that converts the gamma energy to light, which is then detected by a photomultiplier tube and converted into a voltage to generate an image. Lead collimators are placed between the patient and the crystal to select only those photons from a single direction and absorb scattered photons that would degrade image quality. Greater anatomic detail can be obtained by using higher resolution collimators, with maximal resolution of 6 to 7 mm.49 Scintigraphic images may be obtained in planar or CT fashion. In planar imaging, the images are displayed like a radiograph, with data superimposed as if it arose from a single plane in the patient. Overlapping structures may obscure pathology. Using the single-photon emission computed tomography (SPECT) technique, 360-degree acquisitions are obtained with the gamma camera and images reconstructed as in x-ray CT using filtered back­projection. This minimizes superimposed radioactivity, resulting in improved image contrast and diagnostic sensitivity.50 SPECT images require a longer period of acquisition than do planar images and are more sensitive to patient motion during acquisition.

Basic Imaging Techniques

545

Figure 13A-17   Normal bone scan. A, Anterior and posterior whole-body projections obtained 4 hours after injection of technetium99m methylene diphosphonate demonstrate normal distribution of radiotracer activity throughout the skeleton. There is normal renal and bladder activity. B, Spot image of the pelvis obtained to assess for osteitis pubis demonstrates normal activity flanking the pubic symphysis and no scintigraphic evidence for osteitis pubis.

A

B

Trauma Nuclear medicine bone scans play an important role in the assessment of pain in the sports medicine athlete.51 Specifically, bone scans provide a very sensitive method for detecting fractures that are radiographically occult, and for assessment of overuse injuries in bone.52 A negative bone scan virtually eliminates the possibility of fracture or significant osseous injury. Overuse injuries in the athlete comprise both periostitis and stress fractures. Stress periostitis, termed shin splints ��� in the calf, results ������������������������������������������������� from repetitive stresses from muscle and tendon attachments to bone with tearing of Sharpey’s fibers.53 Bone scans demonstrate elongated areas of mildly, and occasionally patchy, cortical increased uptake on delayed images in the posteromedial or anterolateral tibia. Flow and blood pool images are usually normal, and these are not routinely required in the setting of suspected shin splints (Fig. 13A-18). More focal, intense areas of uptake along the tibial shaft indicate the presence of a stress fracture. Stress fractures

occur in normal bone when repetitive forces overwhelm reparative mechanisms. Early stress fractures are often radiographically occult, and often the only manifestation is very subtle bone resorption, faint sclerosis, or early periostitis. Radiographic examinations are only about 15% sensitive to the presence of a stress fracture, whereas skeletal scintigraphic sensitivity approaches 100%.54 Stress fractures often show increased uptake on flow and blood pool images and demonstrate intense activity on 4-hour delayed images (Fig. 13A-19). Note that the bone scan does not

R

L

R

L

Anterior

Posterior

Rt Lateral

Lt Lateral

R

Anterior

Lateral

Figure 13A-18   Shin splints. Four-hour delayed bone scintigraphic images of the bilateral calves in the anterior and lateral projections show elongated areas of increased uptake along the posteromedial and anterolateral tibias.

Figure 13A-19   Tibial stress fracture. Anterior, posterior, and lateral 4-hour delayed bone scintigraphic images of the calves demonstrate focal area of intense uptake along the anteromedial aspect of the middle third of the right tibial diaphysis corresponding to a tibial stress fracture. Diffuse uptake along the midshaft of the left tibia corresponds to shin splints.

546

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

reliably demonstrate a stress fracture from a frank fracture; thus, correlation with plain film radiographs should always be performed. In young athletes, stress fractures often occur in the posterior elements of the lower lumber spine; this is termed spondylolysis. This is a defect in the pars interarticularis, the bridge of bone between upper and lower facet joint articulations in the posterior aspect of the neural arch. These fractures are thought to occur secondary to repetitive trauma and are often a source of low back pain in the athlete. Plain film radiographs are insensitive to the presence of early spondylolysis. Sensitivity is increased substantially using planar skeletal scintigraphic imaging. However, SPECT imaging is the modality of choice because it provides a 50% increase in lesion detection compared with planar imaging.55 SPECT scans demonstrate increased uptake in the posterior elements of the involved vertebrae. Usually, the fractures are bilateral and confined to one level, but they may be unilateral and occasionally multilevel. A positive scintigraphic examination is often followed by CT examination to assess the morphology of the pars defect and any subsequent spondylolisthesis that may affect therapy (Fig. 13A-20). In a patient with a known spondylolytic defect, a negative SPECT scan indicates a healed lesion, often through fibrous union, and it is unlikely that the spondylolysis is a source of the back pain.56,57 If this remains of concern, a directed steroid and anesthetic injection into the pars defect can be performed. Bone scans are often useful in the assessment of clinically suspected occult fractures. An acute fracture demonstrates

increased flow, blood pool activity, and delayed-phase activity at the site of the fracture. In general, bone scans show focal abnormalities as soon as 24 hours after injury and almost certainly by 72 hours. Occult fractures in the wrist are often imaged using scintigraphy. This includes occult fractures to the scaphoid, hamate, and triquetrum. Studies have demonstrated that bone scans are more sensitive than MRI in the setting of radiographically occult nondisplaced scaphoid fractures. A negative bone scan virtually eliminates the possibility of an underlying fracture. A positive bone scan may predicate the need for anatomic imaging, either MRI or CT, to further define fracture extent and assess fracture character. As fractures heal, degree of bone uptake decreases over time but may persist over months to years as the fracture continues to remodel. Bone scans are often used in the setting of osteoporotic compression fractures in elderly patients. These scans are sensitive to the acute compression fracture and are often positive within 24 hours of the antecedent event. Skeletal scintigraphy may be used to establish the chronicity of a radiographically detected compression fracture. Compression fractures that demonstrate increased uptake on bone scans are considered either acute or subacute and healing (Fig. 13A-21). The presence of activity is often used to determine the need for vertebroplasty or kyphoplasty.58 MRI is also often used in this regard. In elderly and osteoporotic patients, early imaging of a compression fracture may lead to a false-negative diagnosis because the reparative process tends to ramp up more slowly and is less robust.

A

C

D

B

Posterior

Figure 13A-20   Pars stress fractures. A, Four-hour delayed bone single-photon emission computed tomographic images of the lumbar spine demonstrating increased uptake in the bilateral L3 posterior elements and the right L4 posterior element corresponding to active pars interarticularis stress fractures (arrows). B, The abnormalities are barely visible on the posterior planar image of the lumbar spine. C, Sagittal re-formation computed tomographic image demonstrates linear lucencies in the right L3 and L4 pars interarticularis corresponding to areas of abnormal uptake (arrows). D, Axial computed tomographic image demonstrates left-sided pars defect at L3 (arrow).

Basic Imaging Techniques

A

B

C

547

Posterior

Figure 13A-21   Spine compression fracture. Coronal (A) and sagittal (B) bone single-photon emission computed tomographic images of the lower thoracic and lumbar spine as well as a posterior spot (C) image of the thoracolumbar spine demonstrate band-like areas of intense activity in the T11 and L2 vertebral bodies corresponding to acute compression fractures.

Orthopaedic Infections Bone scintigraphy plays an important role in the evaluation of bone inflammation and bone infection.59 Although radiographs are usually the first imaging modality in the setting of osteomyelitis, they are notoriously insensitive to the changes of early osteomyelitis, the time when diagnosis needs to be made and therapy instituted. Radiographic findings of osseous destruction and reparative periostitis usually take about 10 to 14 days to develop after symptoms ensue. Three-phase bone scans are often used in the setting of suspected osteomyelitis to differentiate it from cellulitis. Both cellulitis and osteomyelitis show increased uptake on flow and blood pool images. However, in the setting of cellulitis, there is relatively normal activity on 4-hour delayed images, whereas osteomyelitis demonstrates intense activity on the delayed images. Occasionally, 24-hour delayed images must be obtained, during which there is further clearing of soft tissue activity in cellulitis, whereas continued intense activity is seen in osteomyelitis. Using these parameters, the sensitivity of bone scintigraphy in the detection of osteomyelitis is nearly 95%. Osteomyelitis may also be imaged using radiolabeled leukocytes.60 This is usually performed when bone has undergone recent surgery or trauma. A three-phase bone scan performed in this setting would be positive because of the changes from the surgery or trauma. Leukocytes are labeled with indium-111 or technetium-99m hexamethylpropylene amine oxine (HMPAO). The leukocytes migrate to sites of infection in addition to normal hematopoietic marrow. An area of focally increased activity beyond that of normal marrow activity indicates infection. Leukocyte imaging may be combined with marrow imaging using technetium-99m sulfur colloid.61 Although both demonstrate uptake in normal hematopoietic marrow, only the leukocytes show uptake in infected bone. Thus, discordant leukocyte activity with absent sulfur colloid activity indicates osteomyelitis. This technique has a reported sensitivity of about 90% in the detection of infection.

Positron emission tomography (PET) is a functional diagnostic imaging technique that uses a positron-emitting glucose analogue, F-18 fluoro-2-deoxy-d-glucose (FDG). The FDG uptake in cells is directly proportional to glucose metabolism, which is typically increased in the setting of infection and inflammation. PET scanning has been shown to be sensitive and specific in the setting of orthopaedic infections and is an emerging modality in the evaluation of both soft tissue infections and osteomyelitis.62

Bone Tumors Skeletal scintigraphy may be used in the evaluation of the aggressiveness of a bone lesion. It is not capable of determining whether a bone lesion is benign or malignant. When combined with anatomic features obtained on radiographic, CT, and MRI examinations, a definitive diagnosis can often be made and the need for an invasive biopsy avoided. Additionally, bone scans allow for assessment of the entire skeleton in the detection of a potentially polyostotic process. This includes evaluation of entities such as fibrous dysplasia, Langerhans cell histiocytosis, Paget’s disease, and metastatic disease. One of the most common applications of bone scintigraphy is in the screening for osseous metastatic disease. Most malignancies metastatic to bone demonstrate increased uptake due to the bone reparative process (Fig. 13A-22). The only notable exceptions are multiple myeloma and renal cell carcinoma, which are not infrequently cold on bone scintigrams. PET scanning is a well-established imaging modality for the initial staging and management of many cancers, including lung, breast, and lymphoma. Its use in sarcomas is less well established but is an active area of investigation.63,64

Magnetic Resonance Imaging Since its inception in the 1980s, MRI has revolutionized the evaluation of sports-related injuries. MRI provides superb soft tissue contrast compared with all other imaging modalities, providing the means to evaluate muscle,

548

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Physics

Figure 13A-22   Metastatic disease to bone. Anterior and posterior whole-body skeletal scintigraphic images demonstrate multiple areas of abnormal radiotracer accumulation in the axial and proximal appendicular skeleton in a pattern consistent with extensive osseous metastatic disease. This patient had metastatic breast carcinoma.

tendons, ligaments, cartilage, and bone marrow pathology with exquisite sensitivity. Images can be obtained in any plane, allowing protocols to be tailored to a specific pathologic question or anatomic entity. MRI has continued to advance in diagnostic prowess, with high-field–strength magnets now available, leading to improved anatomic resolution and more rapid image acquisition. The physics of MRI acquisition are complex and are the subject of dedicated texts. Fortunately, protocols used in musculoskeletal MRI are relatively straightforward and reproducible from patient to patient. Some pulse sequences may yield excellent anatomic detail but poor sensitivity to underlying pathology, whereas others are very sensitive to underlying pathology, but anatomic detail is diminished. MR examinations are interpreted by synthesizing the data obtained from the different pulse sequences. This section provides a basic understanding of MR principles and the language used to describe the sequences obtained. Sequences are introduced with respect to the appearance of the imaged tissues and their contribution to diagnosis.

MRI represents a completely different method of imaging tissue than CT, ultrasound, and conventional radiography. No ionizing radiation is used. The energy employed in MRI is the same frequency as in radio waves, which are ubiquitous in the surroundings. MRI is based on exciting and receiving signals from protons (hydrogen nuclei). The exact molecular environment in which these protons reside has a profound effect on the MR signals generated and subsequent images.65-67 MRI begins with a strong, relatively homogeneous magnetic field. Some MRI magnets use a fixed (permanent) magnetic field, limited in field strength to about 0.5 Tesla (500 Gauss, or about 500 times the earth’s magnetic field), whereas superconducting magnets have commercially available magnetic field strengths of 3.0 Tesla (30,000 times the earth’s magnetic field). In general, the stronger the magnetic field, the better the quality of the images that may be obtained. When protons (the most abundant paramagnetic element in the body) are placed in the magnetic field, they align with the field, some antiparallel, a few more parallel. This results in an equilibrium net longitudinal magnetization. The protons spin around an axis that is aligned with the main magnetic field. This spinning of protons is termed precession. The frequency of precession is directly proportional to magnetic field strength. This frequency is termed the resonant frequency and is equal to the radiofrequency (RF) that will induce resonance (cause energy absorption) in the protons. The resonant frequency of protons differs depending on its chemical composition. It may be a part of a water molecule, a fat molecule, or other biologic molecule. The net longitudinal magnetization cannot be measured directly because it is obscured by the main magnetic field. Therefore, a radio pulse equal to the resonant frequency is used to disturb the longitudinal magnetization and change its alignment to the main magnetic field. The recovery of net longitudinal magnetization as the protons realign with the main field reflects a T1 property of tissue and can be imaged. Free water (cerebrospinal fluid, cysts) requires a longer period of time to recover (and therefore appears dark on T1 images), complex water (tissue water) has an intermediate time to recover, and fat recovers very quickly (and is thereby bright on T1 images). When net longitudinal magnetization is rotated by the excitation pulse, the spins of the rotated protons are in phase, meaning they are all precessing together. Several factors will cause the spins to rapidly dephase, including heterogeneity of the magnetic field, the chemical environment of the proton (fat or water proton), and exogenously applied magnetic field gradients. This dephasing is a T2 property of the tissue. Importantly, as the spins dephase, a refocusing pulse may be applied, causing the dephased protons to rephase over time. In addition to the main magnetic field, MR systems also require gradient coils. These coils superimpose a weak magnetic field gradient onto the main magnetic field. Thus, a free water proton at one end of the magnet will have a slightly different resonant frequency than a free water proton at the other end of the magnet. Such gradients are the basis for three-dimensional localization of acquired data.

Basic Imaging Techniques

MRI magnets operate in the RF range. The room they are placed in is termed a Faraday cage and precludes outside RF noise. RF coils are basically antennae used to both transmit and receive RF pulses. Pulse transmission alters the net longitudinal magnetization. As the protons realign with the magnetic field, they emit energy as RF waves that are detected by the same RF coil. RF coils are usually placed as close to the patient as possible to maximize the signal. There are dedicated coils for knee, wrist, and spine, and surface coils for shoulder and hip. Use of the appropriate coil is crucial for good image quality.

Pulse Sequences Pulse sequences refer to the parameters of the RF pulses and gradient changes that are used to give rise to the MR image. Each results in different realignment of protons whose subsequent relaxation gives rise to emitted radio waves that are detected by the surface coils. Three basic types of pulse sequences are used in musculoskeletal radiology: spin echo, gradient echo, and inversion recovery. Several parameters within each of these pulse sequences may be altered to maximize contrast resolution in the subsequent images.

Spin Echo The spin echo pulse sequence is the most basic MR sequence. It begins with a 90-degree pulse to excite the protons and rotate the net longitudinal magnetization. This is followed by a 180-degree pulse to refocus the dephasing protons and produce an echo that is detected by the coil. Both T1- and T2-weighted images can be obtained in this fashion, depending on when one waits for the echo. This time length, in milliseconds, is termed the TE (time to echo) and represents the time interval between the initial RF pulse and the time when the receiver coil listens for the returning signal. In T1-weighted images, a relatively short TE is used, whereas the longer the TE, the more T2 weighted the image is. After the echo is detected, the ­90-degree pulse may be repeated for the next set of images. The time between initial 90-degree pulses is the TR (time to recovery) and is proportional to the total imaging time. T1-weighted images use a short TR, whereas T2 images use a longer TR, allowing a more complete return to net magnetization. During spin echo imaging, after the 180-degree refocusing pulse and the resulting echo, the spins begin to dephase again. A second 180-degree refocusing pulse may be performed to refocus the spins a second time. As the spins rephase, a second echo is produced and may be detected, although its amplitude is not as great as the first echo. These repeat 180-degree refocusing pulses may be performed several times, each producing an echo with gradually decreasing amplitude. In musculoskeletal MRI, 8 to 16 refocusing pulses may be applied after the initial 90-degree RF pulse. This is termed the echo train length. Detecting this many echoes after a single RF pulse significantly decreases imaging time. Longer echo train lengths result in lower signal-to-noise ratios and image blurring with subsequent diminished diagnostic sensitivity. Spin echo pulse sequences performed with multiple 180-degree refocusing pulses are termed fast spin echo sequences.

549

Gradient Echo Gradient echo images differ from spin echo sequences in that the initial RF pulse is less than 90 degrees. This is termed a reduction in the flip angle. A lower flip angle results in more T2 weighting, whereas a larger flip angle results in more T1 weighting to the subsequent images. Instead of a 180-degree refocusing pulse, there is a reversal of the gradients, which results in the detected echo. Gradient echo sequences typically have short TE and TR values, enabling fast image acquisition. Gradient echo images play two important roles in musculoskeletal imaging. The first is the ability to perform three-dimensional volume acquisitions, which then allows for very high spatial resolution in the resulting images. This is important for evaluation of small anatomic structures such as the intercarpal wrist ligaments. Second, because of the lack of the 180-degree refocusing pulse, gradient echo images result in susceptibility, or blooming, artifacts. These artifacts result from magnetic field inhomogeneity at the interface of any two substances with different magnetic susceptibilities (water/ fat, calcium/soft tissue, metal/soft tissue). Calcium or metal within the soft tissues appears much more prominent on gradient echo images as a result of the blooming artifact. This can often be helpful in the detection of loose bodies or subtle mineralization in lesions.

Inversion Recovery Inversion recovery pulse sequences begin with a 180-degree inversion pulse followed by a period before the 90-degree excitation pulse is applied. The time interval between these two pulses is termed the TI (inversion time). All tissues have a TI at which they are crossing baseline and cannot be excited. If the 90-degree excitation pulse is applied at that time, the image data after a subsequent 180-degree refocusing pulse will not contain data reflecting that particular tissue, termed suppression. Inversion times are most commonly used to suppress fat signal from tissues. Note that the inversion time is dependent on magnet field strength.

T1-Weighted Images T1-weighted images are usually acquired using spin echo pulse sequences. T1 pulse sequences have a short TE of less than 30 msec and a short TR of 400 to 800 msec. In musculoskeletal radiology, T1-weighted images are deemed anatomic sequences because they provide excellent delineation of osseous structures as well as soft tissue and fat plane interfaces. On T1-weighted images, fat is very bright, water and muscle are intermediate (gray) in signal intensity, and bone, tendons, and fibrous tissue are dark. Because fat is so bright on T1-weighted images, marrow pathology is best detected on these sequences.

T2-Weighted Images A T2-weighted image has a long TE of greater than 70 msec and a long TR of greater than 2000 msec. Because of the long TR, T2 spin echo sequences are quite lengthy; therefore, T2 images are usually obtained using fast spin echo (FSE) techniques. The signal-to-noise ratio in a

550

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

T2-weighted image is less than a T1-weighted image; therefore, spatial and anatomic resolution are diminished as compared with a T1-weighted image. However, water and fluid are very bright on T2-weighted images. Since many pathologic processes result in edema and an increase in extracellular water, T2-weighted images are deemed pathology sequences by musculoskeletal radiologists.

Proton Density Images Proton density (PD) images are obtained using spin echo or fast spin echo technique. They are an intermediate sequence with a short TE and a long TR. Fat is bright in signal, muscle intermediate, and calcium, tendons, and fibrocartilage dark. However, water is brighter in signal than in a T1-weighted image, although not as bright as in a T2-weighted image. Proton density images have the highest signal-to-noise ratios of all the spin echo sequences and, therefore, the best anatomic and spatial resolution. Proton density sequences are most commonly used for the detection of meniscal pathology in the knee. Because water and fat are bright, proton density sequences are less sensitive to marrow pathology as compared with T1-weighted images.

STIR Images STIR (short T1 inversion recovery) images result from an inversion recovery sequence designed to eliminate the signal from protons in fat. In that regard, it is a fat suppression technique aimed at rendering musculoskeletal pathology much more conspicuous. STIR sequences are the most sensitive for detection of several pathologic processes, ­including infections, bone marrow, and muscle edema, as well as several types of tumors. STIR images, however, have a few shortcomings. First, it is a relatively long sequence to acquire, which leads to image degradation secondary to patient motion. Fast spin echo techniques can be applied to STIR images to speed up image acquisition. Second, although STIR images demonstrate superb contrast resolution between normal and pathologic tissue, the anatomic resolution of STIR images is quite poor. Correlation with a similarly acquired T1 or PD sequence is often necessary to clarify the nature of the pathologic process first detected on the STIR images.

Fat Suppression Fat suppression is a crucial component of musculoskeletal imaging. By suppressing the signal from fat, which is bright on T1, T2, and PD sequences, pathology is rendered more conspicuous. Fat suppression is performed using two discrete techniques. Inversion recovery sequences suppress signal from fat by timing the 90-degree excitation pulse (after the initial 180-degree inversion pulse) at the inversion time during which fat protons cannot be excited. This inversion time is dependent on field strength, becoming longer in higher field magnets, leading to longer acquisition times. The second type of fat suppression used is termed fat saturation. This is a technique in which an RF pulse is applied that is specific for the resonant frequency of

­ rotons in lipids, thus nullifying their signal. Fat saturap tion techniques do not provide as uniform a level of fat suppression as do STIR images. This is because local magnetic field variations slightly alter the resonant frequency of the precessing lipid protons. These effects are more prominent in the periphery of the magnet, where joints are commonly imaged. Fat saturation is typically applied to T2 fast spin echo sequences because fat is typically bright on these sequences. This creates an image similar to a STIR, but with better anatomic resolution. T2 fast spin echo fatsaturated images increase the conspicuity of soft tissue and marrow edema, but not to the same degree as STIR images. Fat saturation is also commonly used with T1-weighted images. This is usually performed in the setting of administered contrast agent (gadolinium), either intravascularly or intra-articularly. Fat-saturated gadolinium-enhanced T1-weighted images do not suppress the signal from gadolinium, and this usually remains the only bright entity on the subsequent images. Exceptions include blood products and melanin, but these are less frequently encountered in the imaging of the athlete. For the occasional hematoma, a non–fat-saturated T1-weighted image, obtained in nearly all MR protocols, will confirm the bright signal components of an acute hematoma.

Contrast Agents Intravenous contrast agents are used in a wide variety of MRI examinations. However, musculoskeletal examinations tend not to require intravenous contrast agents routinely. MRI contrast agents consist of a chelate of gadolinium. These agents work by decreasing relaxation times on both T1-weighted and T2-weighted images along the distribution of the gadolinium. Thus, the tissues and compartments into which the gadolinium distributes appear bright on both T1-weighted and T2-weighted images. T1weighted images with fat saturation are usually obtained after contrast administration because of superior spatial resolution compared with T2-weighted images. Gadolinium contrast agents may be administered intraarticularly or intravascularly. Although the indications for intravenous gadolinium are infrequent in the injured athlete, several pathologic processes may be best imaged after gadolinium administration. This may necessitate the patient returning for further imaging after an initial imaging study has been performed. Intravenous gadolinium is often useful in the characterization of mass lesions. Cystic lesions such as ganglia demonstrate a thin rim of smooth enhancement but no internal enhancement. Solid lesions show internal enhancement, sometimes with areas of nonenhancement corresponding to either necrosis or cystic components. Intravenous gadolinium is also commonly employed in the assessment of osteomyelitis, in which inflamed, infected bone demonstrates avid enhancement. If there is bone necrosis or abscess formation, these areas will not enhance. Avascular necrosis of bone, seen commonly in the hip and wrist, also does not demonstrate enhancement of the necrotic bone. Neighboring areas of viable reparative bone, however, demonstrate enhancement. MR arthrography is an increasingly used imaging procedure. Arthrograms may be performed either directly

Basic Imaging Techniques

A

B

551

C

Figure 13A-23   Knee magnetic resonance arthrogram. A, Anteroposterior fluoroscopic spot image of the knee demonstrating a 25-gauge needle in the medial aspect of the joint with iodinated contrast flowing superiorly into the suprapatellar pouch. This is followed by injection of 10 to 20 mL of dilute gadolinium. B and C, Coronal and sagittal T1-weighted fat-saturated MR images of the same knee after injection with gadolinium. Fluid in the joint is bright and outlines adjacent structures, including the menisci and articular surface cartilage.

(a needle is placed in the joint and a dilute gadolinium mixture is injected) or indirectly (gadolinium is injected intravenously and the joint of interest exercised before the MR examination). The principle of indirect arthrography is that the vascular synovium allows gadolinium to diffuse from the vascular compartment into the joint. However, the concentration of gadolinium in a joint after indirect arthrography is much less optimal when compared with direct arthrography, especially in large joints for which arthrography is commonly requested. Direct arthrography also distends the joint, forcing contrast into articular surface tears, which may be not be visualized in a nondistended joint. Additionally, direct arthrography can be performed in minutes and is usually less painful than a vascular injection (Fig. 13A-23). The most common joint to be imaged with MR arthrography is the shoulder. MRI allows visualization of subtle labral pathology that may lead to instability. Additionally, MR arthrography allows improved visualization of articular surface tears of the rotator cuff. Some clinicians require arthrography for all their shoulder MR examinations because of its heightened sensitivity for shoulder pathology. MR arthrography may be performed for other joints. In the knee, it is commonly used to look for re-tears of the postoperative meniscus. In the hip, MR arthrography is used in the assessment of acetabular labral tears and femoral-acetabular impingement. In the ankle and elbow, arthrography may be used to assess osteochondral lesion stability as well as medial and lateral stabilizing ligaments. MR arthrograms of the wrist are commonly used to assess integrity of the triangular fibrocartilage complex as well as the interosseous ligaments.

Image Quality There are several factors that affect the quality of MR images. In general, the signal-to-noise ratio is the most important factor to affect image quality. However, spatial resolution and contrast-to-noise ratios are also important considerations in obtaining high-quality diagnostic images. Signal to noise is a ratio of the amount of signal derived for each pixel of the image relative to the amount of noise within that pixel. High signal is good only relative to diminished noise. Decreases in signal may be desirous if there is a more profound decrease in relative noise. One method to simply increase signal to noise is to repeat the image acquisition parameters, referred to as the number of excitations (NEX). Instead of a single NEX per image slice, two or more excitations may be performed and the resulting data combined to increase signal to noise. The tradeoff is imaging time. Each additional NEX adds incrementally to imaging time. Signal to noise also increases linearly with field strength. Thus, a 3.0 Tesla high-field magnet has theoretically twice the signal to noise as a 1.5 Tesla magnet and 6 times the signal to noise of a 0.5 Tesla open-field extremity magnet. The increased signal to noise can be used to increase imaging speed (decrease NEX) or to increase image quality by increasing spatial and contrast resolution. Spatial resolution reflects the ability to distinguish two adjacent objects. It is generally considered the smallest object that can be detected on an image. For MR examinations, spatial resolution is typically determined by the size of individual pixels. The size of the pixel is determined by the field of view (FOV) and the image matrix size by the equation: pixel size = FOV/matrix size. Spatial

552

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

r­ esolution may be increased by shrinking the field of view or increasing the matrix size. However, the smaller the pixel, the fewer protons in that pixel to provide signal. Thus, there is a concomitant decrease in signal to noise. With higher magnetic field strength of the magnet, a smaller pixel size can be used. This leads to significant increases in spatial and contrast resolution and measurable increases in diagnostic accuracy. MR protocols are designed to balance spatial resolution requirements with signal to noise to produce the best diagnostic images possible for that ­magnet.

Picture Archiving and Communication System Perhaps no advancement has changed the practice of medical imaging as profoundly as PACS. Medical imaging centers on information acquisition, interpretation, and disbursement. The timeliness of each of these events directly affects patient care. PACS creates an all-digital imaging environment in which there is nearly instantaneous access to images and imaging reports from multiple locations. Workflow and speed are significantly improved with the elimination of film and film libraries. Radiologists may view images of a patient while they remain in the scanner, determining the need for alternative sequences or intravenous contrast. This may be performed at a location far remote from where the images are actually being acquired. Digital systems allow more efficient routing of complex examinations to the specialists trained to interpret them. Referring clinicians are capable of reviewing imaging studies at their office locations, in the operating room, or even at the patient’s bedside. Images are stored in vast digital archives and can be retrieved rapidly. Thus, radiologists and clinicians have access to the patient’s imaging history, often solving diagnostic problems and precluding further examinations.

IMAGING OF SPECIFIC SPORTS-RELATED INJURIES Imaging of Bone Plain film radiography remains the initial imaging modal­ ity of choice in the evaluation of bone pathology. This in­cludes traumatic injury, infection, tumors, dysplasias, met­abolic bone disease, and arthropathies. Radiography is fast and inexpensive. Additionally, there are several pathologic bone entities that are diagnosed best on plain film radiographs when advanced CT and MRI remain nonspecific. However, advanced imaging modalities have an important role in evaluating bone. CT is helpful in characterizing occult as well as complex fractures.68 Similarly, MR examinations allow visualization of some of the earliest changes in bone marrow that reflect underlying inflammatory, neoplastic, or overuse conditions. These pathologic entities are often occult on radiographs. Skeletal scintigraphy is also often exquisitely sensitive to early changes in pathologic bone. Exceptions occur in extremely aggressive processes, which do not permit a bone reparative response.69

Fractures Most fractures are diagnosed and classified using conventional radiography. A minimum of two views are required for the evaluation of any bone. Performing abbreviated examinations leads to missed fractures. All suspected fractures should undergo radiography to illustrate ­ fracture type, fragment displacement, intra-articular extension, and the possibility of an underlying predisposing lesion (Fig. 13A-24). As a rule, the joints on either side of a bone should also be included on the radiograph. With complex anatomy, more views may be obtained to provide a more complete assessment. Alternatively, CT may be required to detect occult or subtle fractures. CT is especially useful in the detection of fractures in complex anatomic regions such as the spine, face, extremities, and pelvis. Quite commonly, CT detects fractures beyond those that are seen on the conventional radiograph. Bone scintigraphy is also an effective modality in the detection of occult fracture. It is a highly sensitive modality but lacks specificity. A normal bone scan virtually eliminates the possibility of fracture. However, a positive bone scan could reflect fracture, infection, tumor, or arthropathy. Clinical and radiographic data are used to refine the scintigraphic findings. Finally, MRI is also exquisitely sensitive to the detection of bone marrow edema, an invariable finding in the setting of a fracture (Fig. 13A-25). However, MRI may not readily differentiate bone contusions from true cortical fractures, especially in the case of avulsion injuries and small chip fractures. These fracture fragments are better delineated with CT. MRI also provides important data concerning concomitant soft tissue injury, which is often a more important finding than the fracture itself.

Stress Fractures Stress fractures are defined as fractures in normal bone secondary to overuse. This is in contrast to insufficiency fractures that occur in abnormal bone with normal use. Stress fractures are common in the athlete during periods of increased activity. The athlete usually complains of pain with activity that usually resolves with rest. Stress fractures begin as tiny microfractures in trabecular bone subjected to repetitive stresses.70 Over time, this may evolve to complete cortical disruption. Early stress responses and stress fractures are invisible on most radiographic examinations. When visible, stress fractures may demonstrate a thin rim of sclerosis perpendicular to the major trabecular lines, an area of lucency in the cortex, or an area of periosteal reaction (Fig. 13A-26). Typical locations of stress fractures include the medial femoral neck and diaphysis, anterior and posterior tibial mid-diaphysis, sacrum, calcaneus, metatarsal shafts, pars interarticularis, and sesamoid bones. Stress fractures are uncommon in the upper extremity, but they do occur.71,72 Early diagnosis of stress fractures is best made with MRI or scintigraphy, both of which are nearly 100% sensitive.73-75 Bone scans show increased uptake in the affected bone. The location and pattern of uptake, combined with the clinical presentation, are usually diagnostic. MRI examinations demonstrate bone marrow edema in the affected bone. There is often a low signal line perpendicular to the

Basic Imaging Techniques

A

553

B

Figure 13A-24   Solitary bone cyst with pathologic fracture. A, Single view of the left humerus demonstrating a lytic expansile lesion in the proximal humeral metaphysis compatible with a bone cyst. B, Radiograph of the left humerus obtained 1 week later after trauma demonstrating a nondisplaced pathologic fracture through the bone cyst with characteristic fallen bone fragment in the base of the cyst (arrow).

main trabecular lines on T1-weighted images corresponding to the stress fracture plane (Figs. 13A-27 and 13A-28). Increased periosteal edema may also be observed. Overall, MRI is the favored imaging modality because it is a more expeditious examination, is more specific, and provides an overall assessment of both bone and soft tissue anatomy.76

Avulsion Injuries Avulsion injures are common in the adolescent athlete. These fractures result when tensile forces result in a piece of bone or cartilage being pulled away from the host bone. The injury may result from an acute muscular contraction, tensile forces on ligaments, or chronic repetitive traction stresses. Avulsion fractures may occur at any age but are more common in children and, especially, adolescents in whom muscle strength is greater than the comparatively weaker unfused apophyses. In the adult, the myotendinous unit tends to be the weakest link and is most subject to injury with excessive tensile forces. The diagnosis of avulsion injuries is usually made with conventional radiography.77 These injuries occur in

­ redictable locations at the insertions of ligaments, tenp dons, and joint capsules (Fig. 13A-29). Osseous fragments of various sizes are seen at the site of attachment. Difficulty arises in the differentiation of normal, developing epiphyseal centers from avulsion fractures. Knowledge of the appearance of the developing skeleton is essential for proper diagnosis. CT may aid in diagnosis of subtle or obscured avulsion injures. Two-dimensional reconstruction images are extremely helpful in identifying small, minimally displaced avulsion fracture fragments. MRI is not typically used in the diagnosis of avulsion injuries. The small avulsed cortical bone fragments are not well visualized on MRI (Fig. 13A-30). Additionally, avulsion fractures do not produce a significant amount of marrow edema compared with other fractures. These factors combine to decrease the sensitivity of MRI to the detection of avulsive injuries. MRI, however, does play a role in evaluating other potential soft tissue pathology, which may mimic an avulsive injury. Avulsion fractures are most common around the pelvis, where large muscles in the lower extremities have their attachments.78 Frequent sites of injury include the

554

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B Figure 13A-25   Occult scaphoid fracture. A, Radiograph of the right wrist demonstrates no evidence of scaphoid fracture.   B and C, Coronal T1-weighted and coronal T2-weighted   fat-saturated MR images of the right wrist demonstrate abnormal signal along the waist of the scaphoid consistent with a nondisplaced scaphoid fracture.

C

Figure 13A-26   Femoral neck stress fracture. A, Coronal computed tomographic   re-formation image of the right hip in a runner demonstrating sclerosis with periosteal reaction around a subtle area of cortical lucency along the medial aspect of the right femoral neck (arrow). B, Coronal T2-weighted fat-saturation MR image of the right hip in the same patient demonstrating a large amount of bone marrow edema surrounding an irregular area of low signal corresponding to the stress fracture.

A

B

Basic Imaging Techniques

555

B

A Figure 13A-27   Distal femoral stress fracture. A, Anteroposterior radiograph of the right knee demonstrating no radiographic abnormality. B and C, Coronal T1-weighted and T2-weighted fat-saturated MR images of the knee demonstrate a large amount of marrow edema in the metaphysis and distal diaphysis of the femur. There is a linear area of abnormal signal transversely along the distal diaphysis corresponding to an incomplete stress fracture line. Prominent periosteal edema is also present around the distal femur.

C

a­ ttachment of the sartorius along the anterosuperior iliac spine, rectus femoris along the anteroinferior iliac spine, hamstring insertion on the ischial tuberosity, and the abdominal muscular attachment to the iliac crests. In the proximal femur, avulsion injuries to the greater trochanter (gluteal tendons) and lesser trochanter (iliopsoas tendon) may occur. Avulsion injuries to the lesser trochanter usually indicate underlying bone pathology, of which metastatic disease is the primary concern.79 In the lower extremities, avulsion fractures occur at ligamentous, capsular, and tendinous insertions. This includes the cruciate ligaments in the knee, the popliteal tendon, and the collateral ligaments. In the ankle, the medial

malleolus is often avulsed with eversion injuries. The attachments of numerous tendons and ligaments in the foot, as well as capsular attachments around the myriad small joints, lead to a number of avulsive injuries. In these small joints, CT is the most sensitive modality for detection of small avulsed fragments. In the upper extremity, greater tuberosity avulsive fractures secondary to action of the rotator cuff musculature may occur. Additionally, with an anterior dislocation event, the anterior band of the inferior glenohumeral ligament may avulse a piece of bone from the proximal humerus, which is termed a bony humeral avulsion of the glenohumeral ligament (BHAGL) lesion.80

556

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-28   Calcaneal stress fracture. Sagittal T1-weighted (A) and fat-saturated T2-weighted (B) MR images of the ankle demonstrating an area of irregular low signal transverse to the normal trabeculae of the calcaneus with surrounding bone marrow edema compatible with calcaneal stress fracture.

B Figure 13A-29   Anteroinferior iliac spine avulsion fracture. Anteroposterior (A) and frog-leg (B) radiographs of the right hip in a young patient show an acute avulsion fracture to the anteroinferior iliac spine with small displaced cortical fragment (arrows). This lies at the attachment of rectus femoris.

A

Basic Imaging Techniques

A

557

B

Figure 13A-30   Anterior cruciate ligament avulsion. Coronal (A) and sagittal (B) fat-saturated T2-weighted magnetic resonance images of the knee showing an acute avulsion of the tibial attachment of the anterior cruciate ligament (arrows). The ligament itself remains intact. Note the paucity of bone marrow edema associated with the avulsion injury.

In the chronic phase, avulsion injuries often lead to extensive hypertrophic bone formation as part of the attempted healing process. The appearance on radiographs may be quite sinister and mimic an osseous neoplasm.81 Further imaging with CT and MR may be helpful, but often the clinical history is diagnostic.

Occult Fractures Occult fractures are defined as fractures that cannot be detected on radiographs. Several factors may contribute to nonvisualization, including fragment nondisplacement, complex overlapping anatomy, or osteopenia. Historically, skeletal scintigraphy has been the imaging modality of choice in the setting of suspected occult fractures.82 Bone scintigraphic examinations demonstrate increased uptake at the fracture site corresponding to increased osteoblastic activity. In the hyperacute phase, bone scans may be negative in the setting of occult fractures because the ­reparative process has not yet begun. Scintigraphic examinations should be delayed for at least 48 hours after the acute injury in young patients and at least 72 hours in elderly patients. MRI has a similar sensitivity for detection of occult fractures as does nuclear scintigraphy.83 Advantages of MRI include its ability to detect fractures in the hyperacute phase. Thus, evaluation can be performed immediately after injury, allowing for expeditious diagnosis and treatment. Additionally, MRI allows better characterization of both the fracture and other soft tissue injuries.84 Fractures on MRI appear as irregular linear areas of low signal on T1-weighted images surrounded by bone marrow edema. Often, edema is first detected

on T2-weighted or STIR sequences, and the fracture line is then identified on the T1-weighted images (see Fig. 13A-25). Occult fractures may occur anywhere. In elderly patients, they are common in the sacrum and hip. ­ Osteopenia severely limits their detection radiographically and occasionally with CT examinations. Occult fractures of carpal bones are also common, especially the scaphoid bone, hook of hamate, and triquetrum.85 CT is often helpful in the evaluation of occult fractures in the hands and feet (see Fig. 13A-16).

Imaging of Muscle MRI is typically the imaging modality of choice in the evaluation of muscle injury. Normal muscle is intermediate in signal on all MRI pulse sequences, with other structures generally described as having high or low signal with respect to normal muscle signal. On T1-weighted images, normal muscle demonstrates a marbled appearance owing to the presence of fat interposed between adjacent muscles and muscle fibers. Low signal tendons course through the muscles and may be sheet-like along the myotendinous junctions. On T2-weighted images, muscle maintains its intermediate signal intensity, and there is no fluid signal between normal muscles. Several pathologic processes, including trauma, tumors, denervation injury, infections, inflammation, ischemia, and dystrophies, can affect muscle. MRI, especially STIR sequences, is exquisitely sensitive to muscle pathology, but is often nonspecific. Entities that produce muscle inflammation usually result in muscle edema. This manifests as high signal on T2-weighted and STIR fluid-­sensitive

558

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

sequences. Muscle atrophy with fatty replacement re­sults in muscle with high signal on non–fat-saturated T1-weighted images. Muscle strains are either incomplete or complete muscle tears. They usually occur as sudden events and are one of the most frequent sports-related injuries.86 Muscles that span two joints, such as the hamstrings, biceps femoris, triceps, biceps brachii, rectus femoris, and gastrocnemius, are at the highest risk for injury. The myotendinous junction is the weakest link in the muscle unit.87 Grade I strains pre­ sent as mild muscle swelling with feathery interstitial edema and high signal on T2-weighted images, ­particularly near the myotendinous junction. Grade II strains demonstrate more prominent feathery edema, also with interfascial fluid and focal defects in the muscle filled with hemorrhagic fluid. Grade III strains indicate complete muscle rupture. This manifests as discontinuity of muscle and tendon traversing the muscle with a retracted and wavy appearance to the torn ends (Fig. 13A-31). A large hematoma usually forms between the torn fragments.88 Direct muscle contusions also demonstrate feathery edema within muscle with high signal on T2-weighted images. There is focal muscle enlargement but no focal muscle defects. Muscle fibers are always seen coursing through the edema. An intramuscular hematoma may form when blood dissects between muscle fibers. The appearance of a hematoma depends on its age. Importantly, acute hematomas are usually bright on T1-weighted images. Over time, blunt trauma to muscle may result in an ossified mass of granulation tissue termed myositis ossificans. This is best evaluated with CT to demonstrate the characteristic peripheral ossification. The diagnosis is best made with imaging because pathologic evaluation of myositis ossificans is often confused with more sinister lesions such as osteosarcoma.89 Muscle denervation results in predictable MRI changes in muscle signal over time. Acutely, muscle signal is normal. Over 10 to 14 days, the muscle becomes edematous with intramuscular diffuse high signal on fluid-sensitive sequences. Changes are confined to muscles supplied by the affected nerve. Over several months to years, denervated muscles atrophy, diminishing in bulk and becoming fatty. This is easily visualized on T1-weighted images.90 Soft tissue tumors, many of which arise in muscle, are best imaged using MRI. Contrast is often helpful to distinguish solid enhancing components from necrotic debris or hemorrhagic products. Typically, these lesions are not painful and present as a palpated abnormality. Common intramuscular tumors include benign entities such as hemangioma, lipoma, neurofibroma, and myxoma, as well as malignant entities such as sarcoma, lymphoma, and metastasis (Fig. 13A-32). Hemangiomas and lipomas are usually readily diagnosed on MRI with characteristic imaging features. Other entities are less reliably distinguished, often necessitating biopsy. This is usually done with CT or ultrasound guidance. Rhabdomyolysis is the massive destruction of muscle with elevated levels of creatine kinase in serum. Causes include excessive exercise, massive trauma, hypokalemia, vascular ischemia, and drug or alcohol overdose. MRI demonstrates extensive edema, hemorrhage, and necrosis in involved muscle with diffusely increased signal on

T2-weighted images.91 Although imaging findings are relatively nonspecific, the clinical history and laboratory findings usually establish the diagnosis.

Imaging of Tendons Sports-related tendon injuries are commonly encountered in the athlete. Tendon injuries may result from an acute traumatic event but usually arise secondary to repetitive stress and microtrauma. MRI is the modality of choice for evaluating tendon pathology because of its multiplanar capabilities as well as excellent soft tissue contrast. Tendon pathology includes tendinopathy, tenosynovitis, partial and complete tears, and tendon dislocations. Normal tendons contain very few mobile protons, so they are uniformly dark on all imaging sequences. Notable exceptions include the triceps tendon and quadriceps tendon, which have a normal striated appearance.92 Additionally, as tendons fan out at their osseous insertions, they may also show slightly increased signal intensity. A third reason for abnormally high signal in a tendon is the magic angle phenomenon. This results when tendons are oriented at 55 degrees to the bore of the magnet during T1-weighted and PD sequences. The phenomenon does not occur on T2-weighted sequences, which helps differentiate the finding from true tendinous pathology.93 T1-weighted sequences are excellent for evaluating tendon anatomy and morphology, including course and caliber. They demonstrate tendon subluxations and dislocations as well as tendon thickening or thinning.94 However, they lack sensitivity for intratendinous pathology. Fluid-sensitive sequences with fat saturation (T2 with fat saturation or STIR) are quite sensitive to tendon abnormalities because most tendon pathology results in edema, swelling, and increased tendon water content. Tendons are best evaluated in the axial plane with respect to the tendon itself. Tendons that curve necessitate different planes of imaging. Tendons are typically of uniform caliber throughout their course. In tendon degeneration, variably termed tendinosis or tendinopathy, the tendon initially becomes thickened and high in signal because of increased water content from degeneration and inflammation. This results in intermediate signal intensity on T1- and T2-weighted images. With chronic tendinopathy, the tendon may become thinned with or without abnormal intrasubstance signal. Partial-thickness tears demonstrate focal discontinuity of tendon fibers with a fluid-filled gap. Some fibers remain intact, and the tendon remains taut. With full-thickness tears, bright signal traverses the entirety of the tendon. The torn tendon end may be blunted or frayed and is often proximally retracted and undulating. Many tendons are invested by a synovial tendon sheath. Fluid completely surrounding the tendon suggests tendon sheath inflammation, termed tenosynovitis. The underlying tendon may be either normal or abnormal. Tenosynovitis may result from chronic overuse, synovial inflammation, or underlying infection. Some tendons, such as the flexor hallucis longus and the long head of biceps, may normally have noticeable fluid in their tendon sheath because they communicate with the ankle and shoulder joints, ­respectively.

Basic Imaging Techniques

A

C

B

D

Figure 13A-31   Hamstring rupture. Axial proton density images at the level of the ischial tuberosity (A) and slightly more inferior (B), with corresponding T2-weighted fat-saturated MR images (C and D), demonstrating absence of the hamstring insertion onto the right ischial tuberosity (arrows). The tendon is retracted, and there is prominent perimuscular and peritendinous hemorrhage and edema. Coronal STIR MR image (E) confirms right hamstring tear with prominent edema and allows comparison to the opposite normal hamstring insertion.

E

559

560

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13A-32   Malignant fibrous histiocytoma. Axial (A) and sagittal (B) T1weighted fat-saturated MR images of the knee after the intravenous injection of gadolinium demonstrate a large, homogeneously enhancing, intramuscular mass lesion in the posterior knee just adjacent to the popliteal artery. CT-guided biopsy revealed a malignant fibrous histiocytoma.

A

Knee Within the knee, the infrapatellar tendon is a commonly injured structure. It is subject to repetitive injury in sports involving use of the quadriceps musculature, including jumping, kicking, and running. Patellar tendinopathy usually presents as anterior knee pain and is termed jumper’s knee. The normal infrapatellar tendon is best evaluated in the sagittal plane and is typically dark on all pulse sequences.95 There may be a small amount of striation as it inserts into either the tibia or inferior patellar pole. Patellar tendinopathy presents as thickening of the tendon with intrasubstance high signal on fluid-sensitive sequences. Tendinopathy usually begins proximally and progresses distally as it increases in severity. Additionally, there may be stress-related edematous changes in either the inferior patellar pole or the tibial tuberosity. Full-thickness rupture usually occurs proximally as an acute event, often superimposed on chronic tendinopathy (Fig. 13A-33).96 The quadriceps tendon is injured less frequently than the infrapatellar tendon. This tendon consists of four tendon slips arising from the rectus femoris and vastus medialis, intermedius, and lateralis. It is characteristically striated in appearance but of uniform caliber. Injuries to the quadriceps tendon usually occur in the setting of a tendon weakened by degeneration. This may occur in the setting of chronic steroid use or underlying connective ­tissue disorder. Tendinopathy manifests as thickening and increased signal in the tendon. Full-thickness disruption demonstrates tendon retraction with a fluid-filled gap. The patella often falls inferiorly resulting in a redundant, undulating infrapatellar tendon.97 Other tendons around the knee include the popliteal tendon, origins of the medial and lateral gastrocnemius, the insertions of the hamstrings, and the insertion of the iliotibial band. Injuries to these structures parallel those of any other tendon, with tendon strains demonstrating swelling and high signal and frank tears manifesting fiber discontinuity and fluid-filled gaps. With repetitive motion, the iliotibial band may irritate the subjacent soft tissues along the lateral aspect of the knee, giving rise to iliotibial

B

band friction syndrome. This manifests as soft tissue edema between the iliotibial band and the lateral femoral condyle. The iliotibial band itself may become slightly thickened (Fig. 13A-34)

Ankle There are numerous tendons that traverse the ankle joint, and several are at risk for injury in the athlete. Additionally, most of these tendons must change course along their length, passing through fibro-osseous tunnels as well as passing over one another before arriving at their insertions. This leads to a high incidence of ankle tendon pathology. MRI is the imaging modality of choice for evaluation of ankle tendon pathology.98 For purposes of evaluation, the tendons of the ankle are divided into four compartments based on their location in the ankle: anterior, posterior, medial, and lateral. Four tendons are found in the anterior compartment. From medial to lateral, these are the anterior tibial, extensor hallucis longus, extensor digitorum longus, and peroneus tertius. These serve to dorsiflex the ankle and foot and are best evaluated in the axial and sagittal planes. Injury to these structures is relatively infrequent compared with tendons in the other three compartments. Of the anterior tendons, the anterior tibial tendon is the most commonly injured. Such injury may be seen in older patients or in individuals who participate in hill running. Partial or complete tears of this tendon often present as a sinister mass lesion on the anterior ankle rather than with symptoms typical of a tendon abnormality (Fig. 13A-35). Lateral compartment tendons include the peroneal longus and brevis tendons. These act as lateral stabilizers of the ankle and serve to evert the foot. These tendons pass posterior to the lateral malleolus within the retromalleolar groove and are held in place by the peroneal retinaculum. The tendons share a common tendon sheath proximally, with peroneal longus lying superficial to peroneal brevis. The peroneal brevis tendon attaches to the base of the fifth metatarsal, whereas the longus crosses under the cuboid,

Basic Imaging Techniques

A

561

B Figure 13A-33   Intrapatellar tendinopathy. Sagittal proton density (A), sagittal T2-weighted fat-saturated (B), and axial gradient echo (C) MR images of the knee demonstrate thickening of the infrapatellar tendon near its insertion into the inferior patellar pole. There is focal high signal in the proximal central aspect of the tendon (arrows). The tendon, however, remains intact.

C

passing across the plantar aspect of the foot to attach at the base of the first metatarsal and medial cuneiform. The peroneal brevis tendon may develop a longitudinal split tear during repeated dorsiflexion events when it becomes impinged between the lateral malleolus and the overlying peroneal longus. Split peroneal brevis tears demonstrate three tendons

in the retromalleolar groove, as opposed to the expected two (Fig. 13A-36). There is often concomitant peritendinous fluid and inflammatory change indicating tenosynovitis. Disruption of the peroneal retinaculum may allow for ­lateral subluxation of the peroneal tendons. A shallow retromalleolar groove predisposes to such subluxations.99

562

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

Figure 13A-34   Iliotibial band friction syndrome. Coronal T1-weighted (A) and T2-weighted fat-saturated (B) MR images of the knee demonstrate thickening of the distal aspect of the iliotibial band at the level of the lateral femoral condyle (arrows). There is subjacent soft tissue edema, seen as low signal on the T1-weighted images and high signal on the T2-weighted images.

Within the medial compartment are the three flexor tendons of the ankle: tibialis posterior, flexor digitorum longus, and flexor hallucis longus. The tibialis posterior is the largest of the three, about twice the size of the other two. It is the primary invertor of the ankle as well as a crucial stabilizer of the arch of the foot. This

A

B

tendon passes under the medial malleolus and has a broad base of attachment to the medial navicular, cuneiforms, and bases of the first through fourth metatarsals. At its attachment to the navicular, it often fans out, with the appearance of a thickened tendon with high signal within it. Tears in the posterior tibial tendon are often

C

Figure 13A-35   Anterior tibial tendon tear. Axial T2-weighted (A), sagittal T1-weighted (B), and sagittal fat-saturated T2-weighted (C) MR images of the ankle show a full-thickness tear of the anterior tibial tendon with discontinuous tendon fibers, tendon retraction, and prominent edema and inflammatory changes.

Basic Imaging Techniques

Figure 13A-36   Peroneal brevis split tear. Axial   fat-saturated T2-weighted image at the level of the distal fibula demonstrates marked flattening with central split of the peroneal brevis tendon by the overlying peroneal longus tendon (arrow).

seen in middle-aged women with flat-foot deformity and degenerative arthritis.100 Tears typically occur where the tendon arches under the medial malleolus rather than at its insertion. MRI shows thickening and intrasubstance high signal (Fig. 13A-37). Tenosynovitis manifests as

563

peritendinous fluid, which is best appreciated on fluidsensitive sequences. The flexor hallucis longus tendon is the most lateral of the medial ankle tendons. It arches under the sustentaculum tali, between the sesamoids, to insert at the base of the distal phalanx of the great toe. The tendon sheath of flexor hallucis longus often communicates with the ankle joint, and it is common to see peritendinous fluid around the tendon. Tendinopathy of the flexor hallucis longus is infrequent and is seen in ballet dancers. However, tenosynovitis is relatively more common and manifests as peritendinous loculated fluid within the tendon sheath out of proportion to fluid in the ankle joint. Rupture of the tendon may rarely occur, demonstrating a retracted, undulating tendon with peritendinous edema (Fig. 13A-38). The posterior compartment of the ankle contains two tendons: the Achilles tendon and plantaris tendon. The Achilles tendon is the most commonly injured tendon in the ankle. It is also one of the longest tendons in the body, arising from the gastrocnemius, soleus, and plantaris muscles. The Achilles tendon does not have its own tendon sheath. Instead, there is a paratenon just posterior to the tendon and partially encompassing it. Fluid within the paratenon is termed paratenonitis and is best seen on fluid-sensitive sequences. Achilles tendinopathy usually is ­secondary to chronic overuse. The tendon becomes thickened and ovoid as opposed to its normal flat or concave anterior margin. There is intrasubstance high signal on fluid-sensitive MRI sequences (Fig. 13A-39). Degeneration, partial, and complete tears usually occur 4 to 6 cm above the calcaneal insertion of the Achilles. Full-­thickness rupture usually occurs in middle-aged, unconditioned athletes and is best seen on MRI as a fluidfilled gap on T2-weighted images in the sagittal plane (Fig. 13A-40).101

Figure 13A-37   Posterior tibial tendon tendinopathy. Axial T2-weighted MR images, without (A) and with (B) fat saturation, demonstrate swelling and increased signal in the posterior tibial tendon with peritendinous inflammatory changes (arrows).

A

B

564

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-38   Flexor hallucis longus tendon rupture. Axial   T2-weighted image (A) of the ankle demonstrates no tendon in the expected location of the flexor hallucis longus along the posterior ankle (arrow). Sagittal T1-weighted (B) and T2-weighted fatsaturated (C) MR images confirm fullthickness flexor hallucis longus rupture with tendon retraction into the distal calf. The torn tendon end is mildly swollen and tendinopathic (arrows). There is moderate edema in the soft tissues of the posterior ankle.

C

Shoulder The shoulder joint has the largest range of motion of any joint in the body. It is an inherently unstable joint, supported by a large cuff of tendons termed the rotator cuff. Injuries to the rotator cuff are common and are best evaluated using MRI.102 The addition of intra-articular contrast often improves diagnostic accuracy for both tendon tears and other shoulder pathology. The rotator cuff tendons comprise the supraspinatus superiorly, the infraspinatus and teres minor posteriorly, and the subscapularis tendon anteriorly. These muscles act as dynamic stabilizers of the shoulder joint throughout its range of motion. Injury to the rotator cuff is common in the athlete and may lead

to significant loss of function. Finally, a fifth tendon, the long head of the biceps, courses along the bicipital groove before entering the shoulder joint and terminating on the supraglenoid tubercle. Like most tendinous injuries, single acute traumatic events are a less common cause of rotator cuff tendon tears. Rather, repeated microtrauma with tendon degeneration leads to chronic tendinopathy and finally partial- and fullthickness tears. MRI demonstrates the entire range of rotator cuff pathology as well as changes in the coracoacromial outlet that may predispose to impingement. Careful attention must be given to imaging. Axial images from the top of the acromion through the joint are acquired first. Oblique

Basic Imaging Techniques

A

565

C

B

Figure 13A-39   Achilles tendinosis. Axial T2-weighted (A), sagittal T1-weighted (B), and sagittal T2-weighted fat-saturated (C) MR images demonstrate a markedly swollen Achilles tendon along the posterior ankle with a few focal areas of intrasubstance increased signal (arrows). The tendon remains intact.

coronal and oblique sagittal images are then acquired along the plane of the scapula. Higher-field MRI provides more detailed image quality and allows for more accurate and sensitive depiction of pathology. Rotator cuff tendons are evaluated in all three planes. Normal tendons are dark on all pulse sequences. Magic angle is commonly observed on T1 or PD sequences in the coronal plane and should not be confused with tendon pathology. Normal tendons are low in signal on T2-weighted images. Rotator cuff tendinopathy appears as thickening of the tendon with intrasubstance intermediate signal. Partial tears appear as a fluid-filled gap in the tendon. This may occur on either the bursal or articular side of the tendon. Full-thickness tears demonstrate a

fluid-filled gap in the tendon, often with tendon retraction. The size of the tear and the degree of tendon retraction should be quantified for purposes of therapy. Tears usually begin along the anterior insertional fibers of supraspinatus (Fig. 13A-41). With chronic tears, the involved muscle begins to atrophy and eventually becomes fatty.103 Arthrographic MRI demonstrates fluid traversing the full-thickness tear from the glenohumeral joint into the subacromial subdeltoid bursa. Arthrography is often helpful in the evaluation of the postoperative shoulder to assess the integrity of repair. Normal postoperative changes in an intact repaired tendon may appear similar to a recurrent tear. The addition of arthrographic gadolinium allows differentiation of postoperative changes from a

Figure 13A-40   Achilles tendon full-thickness tear. Sagittal T1-weighted (A) and sagittal T2-weighted   fat-saturated (B) MR images of the ankle show rupture of the Achilles tendon with fluid-filled gap. Note that rupture usually occurs several centimeters from its calcaneal insertion.

A

B

566

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-41   Full-thickness supraspinatus tear. Fat-saturated T2-weighted MR images of the shoulder in oblique coronal (A) and oblique sagittal (B) orientations demonstrate a full-thickness tear involving anterior insertional fibers of supraspinatus. There is a fluid-filled gap at the site of the tear (arrows)

r­ ecurrent tear by demonstrating contrast extension into the tear (Fig. 13A-42).104 Calcific tendinopathy is less reliably detected on MRI but appears as globular areas of low signal (calcium) on all pulse sequences within the tendon or adjacent bursa. With active inflammation related to calcific tendinopathy, there may be a large amount of peritendinous feathery edema and inflammatory changes (Fig. 13A-43). Tears of the teres minor are rare, even in the setting of massive rotator cuff tears. Subscapularis tears are also less common but may occur in the setting of anterior dislocation of the shoulder. Partial tears of the subscapularis may allow for dislocation of the long head of the biceps from the bicipital groove. The tendon may dislocate intra-­articularly deep to the subscapularis, within the substance of the subscapularis tendon itself, or less commonly anterior to the subscapularis.105 Axial images are useful in tracing the course of the long head of the biceps tendon (Fig. 13A-44).

Elbow Tendons of the elbow are commonly affected by tendinopathy in the athlete. Similar to the ankle, tendons are compartmentalized into anterior, posterior, medial, and lateral groups. Medially is the common flexor tendon, which inserts into the medial epicondyle and gives rise to the flexor-pronator group. Laterally is the common extensor tendon, which inserts into the lateral epicondyle and gives rise to the extensor-supinator group. The normal tendons

are dark on all pulse sequences. Injuries to the medial ­common flexor tendon and lateral extensor tendon generally result from repeated valgus or varus stress, respectively. Medially, this is usually termed medial epicondylitis (golfer’s elbow), and laterally, lateral epicondylitis (tennis elbow). In the strictest sense, epicondylitis is a misnomer because the condyle itself is not inflamed. Rather, pathology is ­ confined to the insertions of the tendons, which become thickened and contain foci of high signal compatible with partial tearing. Complete tears demonstrate fluidfilled gaps. Epicondylitis is best seen on coronal and sagittal fluid-­sensitive sequences of the elbow (Fig. 13A-45). With recurrent stress, tearing may extend into the medial and lateral collateral stabilizing ligamentous complexes. The anterior compartment contains the biceps tendon and the brachialis tendon. The biceps tendon inserts onto the radial tuberosity, and the brachialis tendon inserts onto the proximal ulna at the ulnar tuberosity. The distal biceps insertion may have a mildly striated appearance as it fans out along its insertion. Distal biceps tendinopathy may result from repetitive overuse and microtrauma. The tendon may be thickened, thinned, or surrounded by mild inflammatory changes. The biceps tendon more commonly ruptures completely when injured. This typically occurs in middle-aged men. MRI demonstrates a fluidfilled gap in the expected location of the distal biceps tendon. The tear usually occurs near the insertion onto the radial tuberosity, and the torn tendon end often retracts several centimeters into the arm (Fig. 13A-46). Imaging

Basic Imaging Techniques

A

567

B Figure 13A-42   Recurrent rotator cuff tear. T1-weighted   fat-saturated MR arthrographic images of the shoulder in the axial (A) and oblique coronal (B) planes, as well as a   T2-weighted fat-saturated sequence in the oblique coronal plane (C), demonstrate a recurrent supraspinatus tear with contrast extending from the glenohumeral joint into the subacromialsubdeltoid bursa through the tendon defect. In this case, the suture anchors have retracted out of the anterior aspect of the greater tuberosity (arrows).

C

must be tailored to find the retracted tendon in the arm. Brachialis tendon injuries are infrequent even in the setting of a biceps rupture. The posterior compartment of the elbow comprises the triceps tendon. This tendon often has a normal striated appearance along its length as it proceeds to insert onto the olecranon. Tendinopathy is seen with repetitive forceful extension and demonstrates thickening and high signal on MRI. Full-thickness tears are quite rare. The tendon is best evaluated in the axial and sagittal planes.

Hip The hip is a common source of pain in the athlete. MRI is the examination of choice for evaluating the structures of the hip. Tendon and myotendinous injuries may range from mild partial tearing to full-thickness tendon ruptures. The gluteus medius and minimus tendons are a common site of tendon degeneration and partial tearing. This is well seen on T2-weighted MRI as high signal both within and around the tendons as they insert into the greater ­trochanter. There is often a greater trochanteric bursal

568

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-43   Calcific tendinopathy of supraspinatus. Sagittal T1-weighted (A) and T2-weighted fat-saturated (B) magnetic resonance images of the shoulder demonstrate a curvilinear area of low signal in the substance of the distal supraspinatus tendon corresponding to an area of calcific tendinopathy (arrow). There is only minimal inflammation around the calcific deposit at the current time.

fluid collection. The hamstrings include the biceps femoris, semimembranosus, and semitendinosus. They originate on the ischial tuberosity and are a common site of tendon tears. Coronal and axial images are best for evaluating integrity of the hamstrings in the diagnosis of partialand full-thickness tears (see Fig. 13A-31).

A

Wrist There are several tendons that traverse the wrist both dorsally and volarly. These tendons are best visualized on MRI, specifically axial and sagittal views. As with most other tendons, they are low signal on all pulse sequences.

B

Figure 13A-44   Dislocated biceps long head tendon. T1-weighted fat-saturated MR arthrographic images of the shoulder in the axial (A) and oblique coronal (B) planes demonstrate intrasubstance dislocation of the long head of the biceps tendon (arrows).

Basic Imaging Techniques

569

Figure 13A-45   Lateral epicondylitis. Coronal gradient echo (A) and sagittal proton density with fat saturation (B) MR images of the elbow demonstrate a fluid-filled gap along the insertion of the common extensor origin at the lateral epicondyle (arrows). The underlying radial collateral ligamentous complex is intact.

A

B

Tenosynovitis usually results from repetitive overuse and demonstrates high signal fluid within the tendon sheath. The tendon may be enlarged or normal. Typical tendons affected include the extensor pollicis longus and abductor pollicis brevis tendons, termed de Quervain’s tenosynovitis (Fig. 13A-47). Flexor carpi radialis and flexor carpi ulnaris are also frequently affected. The extensor carpi ulnaris, located along a groove on the dorsum of the ulna, is one of a few tendons that may sublux or dislocate. This is appreciated best on axial T1-weighted anatomic images.

Imaging of Ligaments As with tendons, ligaments are soft tissue structures that are best evaluated with MRI. The diagnostic accuracy of MRI with respect to ligamentous injury is not matched by any other imaging modality. Imaging protocols for a specific joint are designed for optimal visualization of commonly injured ligaments in that joint. This includes both acquisition planes and the specific types of sequences obtained. As with most musculoskeletal MRI, T1-weighted images provide excellent anatomic detail, whereas T2-weighted images provide excellent sensitivity for ligamentous injury. In comparison to tendons, ligaments have a more heterogeneous appearance. They may be low signal, intermediate signal, or striated. They may be round, flat, or ovoid. Knowledge of the normal appearance of a specific ligament is essential to diagnosing ligamentous pathology.

Knee There are several commonly injured ligaments in the knee. Ligamentous injury is the second most common indication for a knee MRI, second only to meniscal pathology. The anterior cruciate ligament (ACL) is one of the most commonly injured structures in the athlete. It is a taut ligament that runs parallel to the intercondylar notch of the femur. It has a striated appearance and fans out slightly as it inserts

into the femur. MRI has a reported accuracy of 95% to 100% in the diagnosis of ACL tears.106-108 The ACL is best visualized on oblique sagittal images. Suspected tears should be confirmed on coronal images. When the ACL tears, it often explodes, with no identifiable normal fibers on MRI. Focal full-thickness tears of the ACL are also common and manifest as a low-lying tendon or a focal fluid-filled gap in the tendon.109 Several findings are commonly associated with ACL tears, including pivot shift bone marrow contusions along the weight-bearing aspect of the lateral femoral condyle and the posterior rim of the lateral tibial plateau, deepening of the lateral femoral sulcus, Segond’s fracture (avulsion fracture of the lateral tibia from the lateral capsular ligament), and anterior subluxation of the tibia with respect to the femur, with a buckled appearance to the posterior cruciate ligament (PCL) (Fig. 13A-48). The value of MRI in ACL injury is not only in the confirmation of a clinically suspected ACL tear but also in the detection of associated injuries, including meniscal tears, collateral ligament injuries, and osteochondral fractures (Fig. 13A-49).110 The PCL is normally seen as a thick, gently curving, low signal structure in the posterior aspect of the intercondylar notch. It is best visualized on sagittal MRI. The PCL is less commonly torn compared with the ACL and is less frequently repaired. PCL tears usually manifest as overstretching of the ligament with thickening and intermediate signal within the normally dark tendon.111 Focal tears are also possible with fluid-filled defects in the tendon. Occasionally, the PCL may avulse a piece of bone from its tibial attachment. This commonly occurs in younger patients but can affect those of any age (Fig. 13A-50). Collateral ligamentous injuries are well visualized on coronal MRI of the knee. Axial images are supportive in that they demonstrate the cross-sectional integrity of the ligaments. MRI is commonly used to grade collateral ligament injuries, with the medial collateral ligament much more commonly injured than the fibular collateral ligament.112,113 Grade I injuries demonstrate high signal edema

570

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

Figure 13A-46   Ruptured biceps tendon. Axial proton density (A) and axial T2-weighted (B) images of the elbow demonstrate torn biceps tendon with tendon swelling, intratendinous abnormal signal, and peritendinous edema (arrows). Sagittal fat-saturated proton density image (C) shows retraction of the blunt torn tendon into the soft tissues of the antecubital region with adjacent edema.

C

Figure 13A-47   De Quervain’s tenosynovitis. Axial T2-weighted (A) and coronal fat-saturated T2-weighted (B) images demonstrate thickened extensor pollicis longus and abductor pollicis brevis tendons along the radial aspect of the wrist (arrows). There is peritendinous edema but no full-thickness tendon rupture.

A

B

Basic Imaging Techniques

A

571

B Figure 13A-48   Anterior cruciate ligament tear with Segond’s fracture. Sagittal proton density magnetic resonance image (A) demonstrates acute full-thickness anterior cruciate ligament tear with no identifiable fibers remaining in the intercondylar notch. Coronal T1-weighted (B) and T2-weighted fat-saturated (C) MR images show an avulsion fracture from the lateral cortex of the tibia along the attachment of the lateral capsular ligament, termed Segond’s fracture (arrows).

C

superficial to the ligament. The ligament itself is intact. Grade II injuries show edema within the ligament itself as well as flanking the ligament, yet the ligament remains taut. Grade III injuries indicate full-thickness disruption with high signal completely traversing the ligament. There is often ligament retraction and undulation (Fig. 13A-51).

Ankle MRI is not typically used in the setting of acute ankle sprains, a common event for the athlete. Diagnosis is usually made on the basis of clinical assessment, and

patients typically respond to conservative management. MRI is used in patients who fail conservative therapy and have persistent pain or instability.114 With excellent soft tissue contrast and multiplanar capabilities, MRI is the imaging modality of choice in the evaluation of not only ligamentous injury but also tendinous and osteochondral lesions. In general, the ankle ligaments are taut, thin, low signal structures on all sequences. Occasionally, the ligaments may have a mildly striated appearance. Acute tears demonstrate a heterogeneous ligament with adjacent edema and hemorrhage, occasionally with a fluid-filled defect. Chronic tears often

572

A

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

Figure 13A-49   Anterior cruciate ligament tear with bone contusions and lateral meniscal tear. A, Sagittal T2-weighted ­  fat-saturated MR image demonstrates full-thickness tear of the anterior cruciate ligament. B, More lateral image shows typical pivot shift bone contusions along the weight-bearing aspect of the lateral femoral condyle and the posterior rim of the tibial plateau (arrows). There is also a bucket handle tear of the body and posterior horn of the lateral meniscus with a fragment flipped anteriorly adjacent to the anterior horn of the lateral meniscus (small arrow).

A

B

Figure 13A-50   Posterior cruciate ligament avulsion. Sagittal proton density (A) and T2-weighted fat-saturated (B) MR images of the knee demonstrate an avulsion of the tibial attachment of the posterior cruciate ligament with small, slightly displaced fragment of bone (arrow). The ligament remains intact. Note that there is only mild bone marrow edema associated with the fracture, consistent with an avulsive mechanism.

Basic Imaging Techniques

A

573

B

Figure 13A-51   Medial collateral ligament sprain. Coronal T1-weighted (A) and coronal T2-weighted fat-saturated (B) MR images show a thickened and undulating medial collateral ligament that has torn from its distal tibial attachment. There is high signal edema flanking the torn medial collateral ligament. The appearance is consistent with a grade III sprain.

show an ­irregular, possibly thickened or thinned, mildly heterogeneous ligament. The lateral ankle ligaments are affected in 90% of all ankle ligament injuries. The anterior and posterior tibiofibular ligaments are superior and make up the most inferior aspect of the tibiofibular syndesmosis. These ligaments are best appreciated on axial images at the level of the ankle joint. Disruption of the anterior tibiofibular ligament is termed a high ankle sprain. The anterior and ­posterior ­talofibular ligaments, as well as the calcaneofibular ligament, are more inferior along the lateral ankle. The talofibular ligaments are best seen on axial images, whereas the calcaneofibular ligament is seen best on a coronal view. The anterior talofibular ligament is the most commonly torn ligament in the ankle. Partial tears demonstrate thickening and heterogeneity of the ligament with a joint effusion. Full-thickness tears demonstrate a frank fluid defect in the ligament, with joint fluid leaking out of the anterolateral joint space. If the lateral ankle sprain is more severe, the calcaneofibular ligament is the next to tear, followed by the posterior talofibular ligament (Fig. 13A-52). Associated complications of lateral ankle sprains include osteochondral fractures, instability, sinus tarsi syndrome, anterolateral impingement syndrome, and peroneal brevis split tears. All these entities are readily identifiably on ankle MRI.115,116 The medial ankle ligaments include the tibiotalar, tibiocalcaneal, talonavicular, and spring ligament. Coronal MRI is best for examining the deltoid ligament complex, with the tibiotalar and tibiocalcaneal making up the bulk

of the visualized ligaments.117 The deep tibiotalar component typically has a more striated appearance. Injuries to the deltoid ligament are relatively infrequent compared with lateral ankle sprains. However, the deltoid ligament complex may suffer a crush injury between the medial talus and medial malleolus during inversion injury events.

Shoulder In combination with the rotator cuff and glenoid labrum, ligaments of the shoulder also aid in stabilizing the shoulder joint.118 MR arthrography is the diagnostic examination of choice for evaluating the glenohumeral ligaments in addition to providing superb visualization of both labral and rotator cuff pathology.119 Arthrographic images are obtained in axial, oblique coronal, and oblique sagittal planes. T1-weighted images with fat saturation are typically obtained in all three planes because only intra­articular gadolinium remains high in signal. Distention of the joint will allow separation of the glenohumeral ligaments from adjacent structures and better visualization. Additionally, an ABER (abduction external rotation) view is often obtained. This tightens anterior capsular structures and relaxes the rotator cuff tendons, allowing better definition of tears in both of these locations.120,121 There are three glenohumeral ligaments. The superior glenohumeral ligament is the smallest and arises from the superior glenoid tubercle, extending anteriorly to merge with the coracohumeral ligament. It is not typically

574

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

D

C Figure 13A-52   Lateral ankle sprain. A, Axial T2-weighted MR image of the ankle after inversion injury demonstrates abnormal morphology and signal within the anterior talofibular ligament. The ligament is thinned anteriorly and thickened posteriorly (arrow). It is heterogeneously increased in signal throughout. B, Axial T2-weighted MR image slightly more superior at the level of the tibial plafond demonstrates focal disruption of the anterior tibiofibular ligament or syndesmotic ligament (arrow). C, Coronal T2-weighted MR image shows focal disruption of the calcaneofibular ligament (arrow). D, Sagittal T2-weighted MR image with fat saturation confirms a moderate ankle joint effusion.

Basic Imaging Techniques

thought to be an important stabilizer of the glenohumeral joint.122 This structure is best seen on axial MR arthrographic images at the level of the coracoid process. The middle glenohumeral ligament typically arises from the anterior superior labrum and extends inferiorly and laterally to insert onto the lesser tuberosity. It may be absent in some patients. In others, it may be thickened and cordlike. If in combination with a diminutive or absent anterior superior labrum, it is termed a Buford complex.123 The middle glenohumeral ligament is also best characterized on axial MR arthrographic images. Its intra-articular location should not be confused with a dislocated long head of biceps tendon. The inferior glenohumeral ligament is the largest of the ligaments and contributes the most to shoulder stability.124 It is composed of a thick anterior band, a posterior band, and the intervening axillary pouch. The anterior and posterior bands arise from the anterior and posterior aspects of the inferior labrum, respectively, and insert along the surgical neck of the humerus.125 The inferior glenohumeral ligaments are well visualized in all three MR arthrographic planes. The ABER view, in which axial images are taken with the patient in the abducted and externally rotated position, provides the best assessment of the anteroinferior labral ligamentous complex, a common site for lesions that result in chronic anterior instability.126 Shoulder dislocation events may result in Bankart’s lesion affecting the anterior labral ligamentous complex. Alternatively, anterior dislocation may cause the anterior band of the inferior glenohumeral ligament to tear from its humeral attachment, termed a humeral avulsion of the glenohumeral ligament (HAGL) lesion (Fig. 13A-53).

A

575

Wrist The ligaments of the wrist are divided into intrinsic and extrinsic ligaments. The extrinsic ligaments of the wrist connect the carpal bones to the distal radius and are poorly visualized on MRI. The intrinsic ligaments of the wrist interconnect the carpal bones and are better visualized on MRI. These ligaments are best visualized in the coronal plane. The quality of the MRI is crucial for generating diagnostic images. High-field magnets, preferably 3.0 Tesla, as well as use of dedicated wrist coils, produce exquisite images. The administration of intraarticular gadolinium into the radiocarpal joint further enhances the detection of intrinsic wrist ligamentous injury.127 The scapholunate and lunatotriquetral ligaments are the two intrinsic ligaments of greatest clinical significance. Both these ligaments are triangular structures that attach to the proximal aspect of their respective carpal bones. Tears manifest as high signal fluid traversing the ligament (Fig. 13A-54). There may be separation of the carpal bones. MR arthrographic images demonstrate contrast transit into the midcarpal joint in the case of either scapholunate or lunatotriquetral ligament tears.128 Also of note is the triangular fibrocartilage complex (TFCC), which stabilizes the distal radioulnar joint. This is a biconcave structure that attaches to the cartilage of the lateral radius as well as to the base of the ulnar styloid. Extension of contrast into the distal radioulnar joint during radiocarpal joint arthrography is diagnostic of a TFCC tear. Tears in the TFCC, whether peripheral or central, are best seen on high-field MR arthrographic images in the coronal plane.129

B

Figure 13A-53   Humeral avulsion of the inferior glenohumeral ligament. Axial (A) and coronal (B) T1-weighted fat-saturated MR images of the shoulder after injection of intra-articular gadolinium demonstrate avulsion of the humeral attachment of the anterior band of the inferior glenohumeral ligament after an anterior shoulder dislocation event. There is a contrast-filled defect where the ligament should attach to the proximal humerus (arrows).

576

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13A-54     Scapholunate ligament tear. Coronal gradient echo   (A) and coronal T2-weighted fat-saturated (B) MR images of the wrist demonstrate gross disruption of the scapholunate ligament with widening of the scapholunate distance and a fluid-filled gap (arrows). There is also bone marrow edema in the scaphoid and capitate carpal bones related to contusions.

A

Elbow The elbow is stabilized medially and laterally by the ulnar collateral ligamentous complex and the radial collateral ligamentous complex, respectively. Note that the radial head floats in the elbow joint, rotating around the axis of the ulna, and is devoid of ligamentous attachments. The radial collateral ligamentous complex is composed of four structures: radial collateral ligament (RCL), lateral ulnar collateral ligament (LUCL), annular ligament, and accessory collateral ligament.130 The annular ligament arises from the anterior and posterior margins of the lesser sigmoid notch of the ulna and encircles the radial head. The RCL arises from the lateral epicondyle deep to the common extensor origin and inserts into the annular ligament. These two ligaments are the primary lateral stabilizers of the elbow in the face of varus stress. The LUCL also arises from the lateral epicondyle posterior to the RCL, passes deep to the radial head, and inserts onto the supinator crest of the lateral ulna. It is a prominent posterolateral stabilizer of the elbow. The RCL and LUCL are well seen on coronal MRI of the elbow and are typically low signal on all pulse sequences.131 The annular ligament is best visualized on axial MRI of the elbow. Injury to the radial collateral ligamentous complex usually results from chronic repetitive microtrauma and is associated with degeneration and tearing of the common extensor origin, commonly termed lateral epicondylitis. Acute varus stress may lead to rupture of the radial collateral ligamentous complex, manifested by a fluid signal bisecting the ligament and widening of the radial capitellar joint space (Fig. 13A-55). Injuries to the LUCL may lead to posterolateral rotatory instability with subluxation of the radial head.132 The medial ulnar collateral ligamentous complex is composed of anterior, posterior, and transverse bundles. The anterior bundle is the largest and most important stabilizer of the medial aspect of the joint. It extends from the medial epicondyle to the medial aspect of the coronoid

B

process and is best seen on coronal MRI.133 It typically flares as it inserts onto the medial epicondyle deep to the common flexor origin and tapers distally. It may show mild high signal in the flared portion. This ligament provides the primary restraint to valgus stress. Injuries to the ligament are often due to repetitive microtrauma and degeneration, often associated with medial epicondylitis and partial tears of the common flexor origin. Abrupt acute full-thickness UCL tears occur more commonly than RCL acute tears. Abnormal signal is seen in the expected location of the UCL on coronal fluid-sensitive MRI. Torn ligament ends are often retracted, frayed, and undulating in appearance (Fig. 13A-56). The posterior bundle arises from the medial epicondyle and extends inferiorly to the medial olecranon. The transverse bundle extends from the medial olecranon to the medial aspect of the coronoid. These two ligaments make up the floor of the cubital tunnel in which the ulnar nerve passes. The ligaments are poorly visualized on MRI and are not major contributors to medial stability.

Imaging of Cartilage Cartilage imaging includes detection of abnormalities of both fibrocartilage and hyaline or articular cartilage. Important fibrocartilage structures include the glenoid labrum in the shoulder, the acetabular labrum in the hip, and the medial and lateral menisci of the knee. These structures all serve important joint-stabilizing and forceredistribution roles within their respective joints. Tears in these structures result in abnormal mobility and force distribution, leading to premature joint degeneration and pain. Composed of soft tissue, the menisci and labra are best visualized using MRI.134,135 The administration of intra-articular gadolinium improves visualization of tears in all three structures, especially in the shoulder and knee. For patients unable to undergo MRI, high-resolution CT arthrography is also sensitive for detection of fibrocartilage tears.136

Basic Imaging Techniques

577

Figure 13A-55   Radial collateral ligament tear of the elbow. Coronal proton density fat-saturated (A) and coronal T1-weighted (B) MR images of the elbow demonstrate fullthickness rupture of the radial collateral ligamentous complex from the lateral epicondyle of the elbow just below the insertion of the common extensor tendon. The ligament is retracted slightly, and there is a fluid-filled defect where it attaches to the lateral humerus (arrow). Also note slight widening of the radiocapitate articulation with an elbow joint effusion.

A

B

Hyaline cartilage is an avascular structure present along all synovial articular surfaces. It is a supportive, low-friction structure designed to bear mechanical stress. It is thickest in larger joints that undergo significant stress, such as the knee. Cartilage degradation is the hallmark event

A

B

Figure 13A-56   Ulnar collateral ligament sprain of the elbow. A, Coronal gradient echo MR image of the elbow demonstrates thickening and high signal along the humeral attachment of the ulnar collateral ligament compatible with ligament sprain (arrow). The ligament remains intact. B, Fluid along the attachment is confirmed in the sagittal fat-saturated proton density MR image (arrow).

in the evolution of osteoarthritis. Early detection of cartilage injury and degradation is important to preserving joint integrity and slowing the progression to debilitating arthritis. Improved detection and characterization of hyaline cartilage abnormalities is an active area of imaging development. Cartilage abnormalities may mimic other intraarticular abnormalities such as ligamentous injury or meniscal tears. Additionally, early detection of cartilage abnormalities allows for early institution of therapy to prevent further cartilage degeneration and subsequent osteoarthritis. Recent advances in CT and especially MRI have significantly improved the ability to detect cartilage abnormalities. This has been driven by the development of several relatively new orthopaedic procedures for the treatment of cartilage abnormalities. These include microfracture, chondrocyte transplantation, articular cartilage transplantation, osteochondral allografting, and osteochondral plug transfers.137,138 Accurate depiction of cartilage abnormalities allows appropriate preoperative planning to maximize therapeutic and procedural success. Cartilage is not directly visualized on radiographic examinations. Its thickness can be inferred by examining joint spacing. Additionally, radiographic changes of osteoarthritis, including osteophytes, subchondral sclerosis, and subchondral cyst formation, are seen with more advanced stages of cartilage loss. More complex imaging modalities, such as CT and MRI, are required for a more detailed assessment of cartilage integrity.139 CT arthrography using intra-articular iodinated contrast provides excellent depiction of the cartilage surface. Current generation CT scanners provide better spatial resolution than MRI, so subtle cartilage surface fissuring and flaps are better visualized. However, cartilage abnormalities deep to the

578

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

articular surface, such as softening and separation from the subchondral bone plate, are not detected. In this regard, MRI provides an assessment of hyaline cartilage surface integrity and deeper cartilage lesions. Additionally, MRI provides excellent depiction of surrounding osseous and soft tissue structures to provide a more complete assessment of the joint. For this reason, MRI has evolved into the imaging modality of choice in the assessment of chondral lesions.140 Various MRI sequences may be used to image cartilage abnormalities. Typically, fast spin echo T2-weighted images or PD images with fat saturation are employed. Fast spin echo sequences are superior to routine spin echo sequences due to a higher contrast between joint fluid and articular cartilage.141 Spoiled gradient echo recall (SPGR) imaging is a three-dimensional gradient echo technique that has also been used to evaluate cartilage and has ­historically

A

D

been superior at depicting the internal trilaminar appearance of cartilage.142 MRI of articular cartilage has improved as the spatial resolution of MR systems has improved. Thus, more subtle surface fissuring and fibrillation, as well as cartilage flaps and frank defects, are better depicted with stronger MR magnets. Displaced cartilage fragments are also more readily locatable (Fig. 13A-57). Highfield 3.0-Tesla magnets provide for excellent spatial depiction of cartilage abnormalities, better than that of 1.5-Tesla systems and significantly better than 1.0-Tesla and weaker dedicated extremity systems. Additionally, 3.0-Tesla imaging allows visualization of the internal architecture of cartilage and abnormalities of its trilaminar structure.143 MRI also plays an important role in the assessment of osteochondral lesions and cartilage repair. MR

C

B

E

Figure 13A-57   Osteochondral fracture. Coronal T2-weighted fat-saturated MR image (A) demonstrates a free intra-articular osteochondral fragment in the lateral aspect of the intercondylar notch (arrow). The donor site along the lateral femoral trochlea is well seen on the sagittal proton density (B) and T2-weighted fat-saturated MR (C) images (arrows). Sagittal proton density (D) and   T2-weighted fat-saturated (E) MR images obtained 2 months after arthroscopic repair demonstrate the cartilage fragment tacked down in situ. There remains fluid signal undercutting the repaired osteochondral fragment, indicating that it has not completely incorporated into the underlying bone (arrow).

Basic Imaging Techniques

A

B

579

C

Figure 13A-58   Lateral meniscal tear with parameniscal cysts. Sagittal proton density (A) and T2-weighted fat-saturated (B) MR images demonstrate a small horizontal tear in the anterior horn of the lateral meniscus. There is an adjacent, slightly loculated parameniscal cyst along the anterior horn of the lateral meniscus, best seen in the coronal T2-weighted fat-saturated MR image (C) (arrow).

a­ rthrography using dilute gadolinium allows the differentiation of loose in situ cartilage fragments from attached fragments. Additionally, arthrographic MRI is also helpful in assessment of cartilage ingrowth after surgical intervention and, in combination with clinical improvement, may help guide the return to activity.144

Knee Imaging of meniscal tears is the most common indication for an extremity MRI. MRI provides superb depiction of meniscal fibrocartilage tissue and has a greater than 95% sensitivity for the detection of meniscal tears.145 Sagittal PD sequences have been shown to be the most sensitive for meniscal pathology, whereas sagittal T2-weighted fatsaturation images are the most specific. Coronal images are also important for assessing the body of the medial and lateral menisci. The menisci are low signal structures on all sequences. There may be intrasubstance signal within the periphery of the menisci. In older patients this is interpreted as myxoid degeneration, whereas in younger patients it usually represents peripheral penetrating blood vessels. As long as this abnormal signal does not extend to a meniscal surface, it is not deemed a tear. Meniscal tears are defined as abnormal intermediate or high signal that extends to a meniscal surface.146 The orientation of the abnormal ­ signal is usually used to describe the tear (horizontal, vertical, oblique, complex). Fluid may traverse the tear and gather under pressure in adjacent parameniscal cysts, which are readily identifiable on fluid-sensitive sequences (Fig. 13A-58). More complex tears may demonstrate displaced fragments. Circumferential vertical tears often demonstrate large segments of meniscal tissue displaced into the intercondylar region in what is termed a bucket handle tear (Fig. 13A-59).147 Oblique vertical tears

may show meniscal ­ tissue displaced into the meniscal recesses. MRI provides an accurate depiction of meniscal morphology, displaced fragments, and other associated injuries.148 Much progress has been made in the imaging of hyaline cartilage abnormalities of the knee. Knee articular cartilage is quite thick and therefore more amenable to detailed cartilage imaging. Additionally, it is a common site of early cartilage degeneration, termed chondromalacia, and osteoarthritis. Chondromalacia is typically graded at four levels, with grade I changes representing only abnormal signal within the cartilage and no surface irregularity. Grade II changes indicate surface cartilage loss and irregularity, whereas grade III changes reflect deeper fibrillation defects and cartilage loss. Grade IV changes indicate full-thickness cartilage defects with subchondral bone changes, such as edema or cysts (Fig. 13A-60).149,150 In most cases, it is best to describe the morphology of the hyaline cartilage abnormality in addition to providing a grade classification. Cartilage must be evaluated in multiple planes. In the knee, the axial, sagittal, and coronal images each highlight different portions of the articular cartilage, typically cartilage that is oriented orthogonal to the imaging plane. Defects easily identifiable in one plane may be obscured in a second plane. Hyaline cartilage has a zonal structure. The thin superficial or tangential zone contains densely packed collagen fibrils, which are orientated parallel to the articular surface. Below the superficial zone is a transitional zone containing obliquely oriented collagen fibrils, which gradually change with depth to become perpendicular. The deepest and largest zone is the radial zone, in the upper two thirds of which the collagen fibrils are perpendicularly oriented, whereas in the lower third, there are also numerous curved, obliquely oriented fibrils. The water concentration differs

580

A

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

C

Figure 13A-59   Medial meniscus locked bucket handle tear. Coronal T2-weighted fat-saturated MR image (A) demonstrates truncated appearance to the body of the medial meniscus with a small amount of adjacent bone marrow edema within the medial femoral condyle. In addition to the anterior and posterior cruciate ligaments, there is a third abnormal low signal structure in the intercondylar notch (arrow). Sagittal proton density (B) and T2-weighted fat-saturated (C) MR images confirm a curvilinear low signal structure inferior to the posterior cruciate ligament corresponding to the handle of a bucket handle tear of the medial meniscus with fragment flipped into the intercondylar notch (arrows). This appearance is termed the double posterior cruciate ligament sign.

slightly between the zones, being slightly higher in the superficial zone than the radial zone. With early cartilage degeneration and subsequent repair, the orientation of collagen fibrils becomes less uniform. This leads to alterations in the water content of the cartilage, with more degenerated cartilage having a more disoriented collagen structure and thereby a higher water content. High-resolution T2 imaging of cartilage may visually demonstrate such alterations in water content and thereby provide an assessment of cartilage quality within the joint (Fig. 13A-61).151,152

A

B

Shoulder Labral tears are a common source of pain and instability in the throwing athlete. The labrum is a fibrocartilaginous structure best imaged using MR techniques.153 The shoulder joint does not typically contain a large amount of intra-articular fluid, and the detection of labral tears is diminished using nonarthrographic MRI. Occasionally, labral tears may only be inferred by the detection of small paralabral cysts on T2-weighted fluid-sensitive images

C

Figure 13A-60   Chondral defect with displaced fragment. Coronal (A) and sagittal (B) T2-weighted fat-saturated MR images of the knee demonstrate a focal full-thickness cartilage defect along the posterior weight-bearing aspect of the lateral femoral condyle (arrows). The cartilage fragment is displaced into the posterior aspect of the joint adjacent to the posterior cruciate ligament, seen in the T2-weighted fat-saturated sagittal image (C; arrow).

Basic Imaging Techniques

A

B

C

D

581

Figure 13A-61   Cartilage degeneration. A, Coronal T2-weighted fat-saturated MR image of the knee in a 16-year-old demonstrates grade III chondromalacia changes along the medial femoral condyle (arrow). B, The corresponding T2-weighted cartilage map. Red and orange areas indicate normal cartilage water content, whereas green and blue areas indicate increased cartilage water content (high T2 values), reflecting cartilage degeneration. The thinned cartilage along the medial femoral condyle shows diffusely abnormal high T2 values, indicating poor cartilage quality. Medial and lateral tibial plateau cartilage is normal. C, Sagittal T2-weighted fatsaturated MR image of the same knee demonstrates normal thick cartilage along the medial femoral trochlea and adjacent patella. However, there is abnormal signal within the cartilage (arrow). D, This is reflected in the corresponding T2 map, which demonstrates diffusely high T2 values in the patellofemoral cartilage, indicating early cartilage degeneration.

(Fig. 13A-62). The injection of 10 mL of intra-articular gadolinium distends the shoulder joint and forces contrast into labral tears, increasing their detection. MR arthrography is the examination of choice for detection of labral pathology.154 Superior labral tears are often characterized with respect to their involvement of the biceps anchor, termed SLAP (superior labrum, anterior to posterior) lesions (Fig. 13A-63).155 Anteroinferior labral tears may become more evident on ABER views, during which the anterior labral ligamentous complex is tightened. The glenohumeral joint is a common location of cartilage loss and osteoarthritis. However, cartilage thickness in the shoulder is about half that in the knee. Additionally, the shoulder is a deeper joint than the knee, and glenohumeral cartilage is therefore farther from the receiving surface shoulder coil, decreasing spatial resolution. These

factors diminish the sensitivity for detection of hyaline cartilage lesions in the shoulder. Use of high-field magnets, good surface coils, and arthrographic technique increases the sensitivity for glenohumeral chondromalacia, osteochondral lesions, and intra-articular cartilage fragments (Fig. 13A-64).

Hip The acetabular labrum is an increasingly recognized source of hip pain in the young athlete. Labral tears may be secondary to trauma, chronic overuse, impingement, or underling hip dysplasia. Tears are best imaged using MRI, with intra-articular gadolinium essential for diagnosis.156 Like the glenoid labrum, the acetabular labrum is a low signal triangular structure composed of fibrocartilage. Tears demonstrate abnormal contrast extension into the

582

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-62   Shoulder paralabral cyst. Axial (A) and oblique coronal   (B) T1-weighted fat-saturated high-field (3.0 Tesla) MR images of the shoulder after gadolinium arthrography confirm a rounded contrast collection adjacent to the posterior superior glenoid labrum consistent with a paralabral cyst that communicates with the joint (arrows). The cystic nature of the lesion is confirmed on the fluid-sensitive oblique coronal T2-weighted fat-saturated image (C). Note that the labral defect cannot be directly seen on the images.

C

labrum, usually between the labrum and the adjacent bone. Paralabral cysts occasionally develop and are well demonstrated on T2-weighted images. Hip articular cartilage is poorly visualized using MR techniques. The hip joint is the deepest joint imaged, which decreases the spatial resolution obtainable. Highfield technique and use of intra-articular contrast increase sensitivity for detection of chondral abnormalities, including deep fissuring, chondral defects, and intra-articular fragments. More advanced findings of osteoarthritis, such as subchondral cyst formation and bone marrow edema, are readily seen on fat-saturated fluid-sensitive sequences such as STIR and T2-weighted images. Additionally, MRI provides a global assessment of the hip, often identifying other sources of pain such as occult fractures and myotendinous injuries.

ORTHOPAEDIC INTERVENTIONS Radiologists play an important role in image-directed therapy. This includes peripheral joint injections, spine injections, bursal injections, fluid aspiration, vertebroplasty, and biopsy. Familiarity with imaging anatomy

allows the radiologist to participate in diagnosis, procedural planning, and therapy. Procedures are not performed blindly, but in a highly directed manner. This results in a high degree of therapeutic success. Procedures are typically documented with imaging, demonstrating arthrograms for joint injections, fluid removal in the case of cyst aspirations, and epidurograms in the case of spine procedures. Biopsy procedures typically document the location in the tumor at which the biopsy sample was taken and indicate the needle tract used to gain access to the lesion. This may have important implications in oncologic surgical planning. Additionally, in the event of therapeutic failure in the setting of an imageguided injection, procedural technical failure is usually not the cause.

Peripheral Joints Nearly any joint may be injected under imaging guidance. This includes small joints in the hand, wrist, and feet as well as larger joints such as the hip, knee, and shoulder. Fluoroscopy is required for real-time needle visualization. Use of a C-arm provides more flexibility in procedural

Basic Imaging Techniques

A

583

B

Figure 13A-63   SLAP (superior labrum, anterior to posterior) tear. A and B, Coronal T1-weighted fat-saturated MR images of the shoulder after intra-articular injection of gadolinium demonstrate abnormal contrast extension into the superior labrum extending anterior to posterior along the biceps anchor (arrows). The pattern of contrast extension in the superior labrum is consistent with a type II SLAP tear.

planning and allows the operator to change the view while the procedure is under way. Peripheral injections typically use 22- or 25-gauge needles of variable length. The 25-gauge needles pierce the joint capsule better and provide better intra-articular flow, but are slightly more

A

difficult to guide. Superficial lidocaine is used at the discretion of the operator. The needle is advanced to the joint, often touching bone. A small amount of iodinated contrast is injected to confirm an intra-articular location; this is followed by the pharmaceutical agent, which is

B

Figure 13A-64   Intra-articular cartilage fragment in the shoulder. Axial T1-weighted fat-saturated (A) and oblique coronal T2weighted fat-saturated (B) high-field (3.0 Tesla) images of the shoulder after arthrography demonstrate a small free cartilage fragment within the inferior aspect of the joint (arrows).

584

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

pain may return. Steroids usually begin working in a few days, with the full effect in about a week. Small joints usually respond to injections more quickly than larger joints (Fig. 13A-65).

Spine Procedures

Figure 13A-65   Fluoroscopically guided hip injection. Fluoroscopic spot image of the left hip in a patient with hip dysplasia and severe degenerative cartilage loss. A 25-gauge needle has been advanced into the joint, confirmed by injected iodinated contrast (darker material) spreading out within the joint. This is followed by injection of a steroid and anesthetic preparation, which further disperses and dilutes the contrast.

usually a mixture of a long- and short-acting steroid and a long-acting anesthetic. Patients may experience pain during the injection. This usually resolves rapidly once the needle is removed and the pressure within the joint diminishes. Patients often report significant improvement in symptoms immediately after the injection because of the effects of the anesthetic. As these effects wears off,

A

B

Imaging plays a crucial role in directed spine injections. The combination of clinical assessment and imaging often allows the site of a patient’s pain to be determined with a high degree of accuracy. Spine interventions performed with imaging allow highly directed injections at the site of pathology.157 Spine procedures performed with imaging include nonselective paralaminar midline epidural steroid injections, selective transforaminal epidural steroid injections (nerve root blocks), and facet joint injections and synovial cyst rupture.158 As with peripheral joint injections, a small amount of iodinated contrast is injected in the appropriate space to confirm proper needle location. In the case of epidural steroid injections, an epidurogram or neurogram should be documented (Fig. 13A-66). This is followed by injection of a steroid and anesthetic mixture. During the injection, pain may be reproduced, and it may be severe. This invariably resolves as soon as the injection is complete. Facet joint injections and sacroiliac joint injections are performed similarly to peripheral joint injections. Percutaneous diskography is an established diagnostic procedure that assesses disk integrity in patients with back pain. It is a procedure meant to elicit pain to determine which disks are responsible for a patient’s back pain or radiculopathy.159 Diskography is usually performed in the posterolateral oblique approach. After skin and deep local anesthesia, 22-gauge needles are advanced into the anterior thirds of the lumbar disks. Contrast is then injected into the disk under pressure. Disk morphology and integrity are assessed, as is the patient’s pain response during the injection. Usually no more than 2 mL is injected into a disk. Multiple disks are usually assessed in a single

C

Figure 13A-66   Lumbar epidural steroid injection. A, Posterior oblique fluoroscopic spot image of the lumbar spine demonstrating 22-gauge spinal needle advanced toward the sublaminar epidural space using a left paralaminar approach. B, Lateral image demonstrating epidurogram after the injection of 0.5 mL of iodinated contrast. C, Lateral fluoroscopic spot image obtained after injection of steroid and anesthetic mixture demonstrates dispersion and dilution of injected contrast. Note grade I degenerative listhesis at L4-L5, which is the cause of this patient’s back pain (arrow).

Basic Imaging Techniques

A

585

C

B

Figure 13A-67   Lumbar kyphoplasty. A, Sagittal T2-weighted MR image of the lumbar spine demonstrates a superior end-plate compression fracture of L2 with about 30% loss of vertebral body height. The fracture line is indicated by the area of irregular linear signal (arrow). There is no retropulsion of fracture fragments. B and C, Anteroposterior and lateral intraoperative fluoroscopic spot images after kyphoplasty demonstrate dense bone cement within the right and left aspects of the vertebral body along the fracture plane. The patient had no fracture-related pain immediately after the procedure.

­ rocedure. CT is often performed after the procedure to p provide a more detailed anatomic evaluation of disk architecture, as well as degenerative changes in the remainder of the lumbar spine. Sacral insufficiency fractures and osteoporotic compression fractures are a common occurrence in the elderly ­population. Vertebral fractures occur in 25% of women older than 50 years and in 40% of women older than 80 years. These fractures are often clinically silent but may present with significant pain and debilitation. Vertebroplasty and kyphoplasty can be effective at controlling pain, hastening the healing process, and minimizing debilitation.160 These procedures involve image-guided insertion of large-caliber needles in the vertebral body using a transpedicular approach. Methacrylate cement is then injected into the vertebral body, filling the fracture plane and cementing the fragments together. In the case of kyphoplasty, a cavity in the vertebral body is first created using a small balloon. This helps to restore vertebral height and prevent progressive kyphosis.161 The cavity and fracture are then filled with cement. Cement is injected under direct fluoroscopic visualization to ensure flow within the vertebral body and not within the spinal canal. Patients usually report significant pain relief immediately after the procedure. These procedures are usually performed in a hospital setting because they are not without potential life-threatening complications (Fig. 13A-67).

Cyst Aspirations Image-guided fluid aspiration is a commonly performed interventional orthopaedic procedure. Ultrasound provides excellent depiction of cystic structures, including popliteal cysts, ganglion cysts, parameniscal cysts, ­ paralabral cysts, and bursal fluid collections. Cysts are punctured under direct sonographic visualization, fluid is drained, and an anti-inflammatory steroid and anesthetic mixture is injected. If the fluid is too thick to be aspirated, the cyst is ruptured with the steroid mixture. Ultrasound is also useful in peritendinous injections in the case of tendinopathy and tenosynovitis.

Biopsies CT and MRI play a central role in the detection and characterization of both soft tissue and osseous mass lesions. However, the small bore of the MR magnet and the tremendous magnetic field render MRI-guided interventions and biopsies impractical. In these cases, CT is usually used to direct the biopsy. CT fluoroscopy, in which real-time CT images can be obtained using a foot pedal at the side of the gantry, is often used to guide fine-needle aspirations and core biopsies. Procedures can be performed accurately and expeditiously with minimal patient discomfort. Biopsy trajectory and lesion sampling should be documented as part of the examination (Fig. 13A-68).

586

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13A-68   CT-directed biopsy. A and B, Axial CT images of the right hip. There is a large lytic lesion that has eroded through the medial aspect of the right acetabulum and extends into the right hemipelvis. A, The needle is directed toward the lesion. B, The needle has been advanced into the lesion. Aspiration revealed bloody synovial fluid related to particle disease from patient’s right hip arthroplasty.

c r i t i c a l

P

o i n t s

l Imaging plays a central role in the daignosis and management of orthopaedic disorders and sports-related ­injuries. l Imaging studies must be chosen with consideration of cost and likelihood of establishing a firm diagnosis. l Plain film radiography remains the mainstay for initial imaging evaluation of most orthopedic conditions. l Significant advancements have been made in the quality of advanced imaging modalities, specifically CT and MRI, and their use in orthopaedic practices has increased accordingly. l MRI provides the best soft tissue contrast of all imaging modalities and is central to imaging soft tissue pathology. l For the orthopaedic physician, CT plays a central role in evaluating osseous anatomy. l Higher-field MR magnets provide better spatial and temporal resolution than lower-field magnets, thus improving diagnostic accuracy and therapeutic outcomes. l Orthopaedic and sports medicine practitioners should expect high-quality musculoskeletal imaging examinations, including interpretation by subspecialty trained orthopaedic radiologists as well as electronic image availability on PACS stations.

S U G G E S T E D

R E A D I N G S

Helms CA, Major NM, Andersen MW, et al: Musculoskeletal MRI, 2nd ed. ­Philadelphia, Saunders, 2008. Manaster BJ, May DA, Disler DG: Musculoskeletal Imaging: The Requisities, 3rd ed. Philadelphia, Mosby, 2004. Resnick D, Kang HS, Pretterklieber ML: Internal Derangements of Joints, 2nd ed. Philadelphia, Saunders, 2006. Resnick D, Kransdorf M: Bone and Joint Imaging, 3rd ed. Philadelphia, Saunders, 2004. Stoller D, Tirman P, Bredella M: Diagnostic Imaging: Orthopaedics. AMIRSYS, 2003.

R eferences Please see www.expertconsult.com

Basic Imaging Techniques

S ecti o n

587

B

Imaging Considerations in the Skeletally   Immature Patient Jack Clement

Sports-related injuries in the pediatric patient have increased in incidence during the past decade. This has been the result of both an increasing number of active children and an increase in the activity intensity of those children who are sports oriented. Although traumatic injuries remain common, overuse and stress-related injuries have become more frequent in the pediatric population, even appearing in toddlers. Many of these conditions mimic their adult counterparts; however, several considerations unique to immature skeletons influence imaging diagnoses and subsequent therapies. Nontraumatic entities such as tumors, infections, and developmental variations may be a source of symptoms. This section discusses various imaging modalities and their implementation in imaging the skeletally immature athlete. Several common normal skeletal variations are presented along with traumatic and pathologic processes particular to the pediatric patient.

SKELETAL MATURATION Bone develops by either intramembranous mesenchymal ossification or endochondral ossification. Intramembranous ossification is a process by which primitive mesenchymal cells directly differentiate into membranous bone. It is responsible for development of bones such as the skull, clavicle, and facial bones. Endochondral ossification reflects a process by which the bones are first cartilaginous and then transformed into bone. The long bones, vertebrae, and skull base form by the process of endochondral ossification. The process of skeletal maturation proceeds in a fairly predictable manner, with ossification centers appearing in their cartilaginous precursors in a relatively defined sequence.1 The process may be delayed or accelerated in some children. Radiographs of the hand and wrist are often used to evaluate the bone age of a patient by comparing the ossification centers with compiled female and male standards.2 Epiphyseal and apophyseal ossification centers have variable appearances depending on the particular bone and the degree of ossification. Ossification centers may appear fragmented or sclerotic, and they may be mistaken for pathologic processes (Fig. 13B-1). Comparison with atlases of normal developmental variants or with the opposite extremity often resolves confusion. Developing long bones have four discrete parts defined by their location with respect to the physis, or growth plate. The epiphysis and metaphysis flank the growth plate, whereas the diaphysis comprises the shaft of the

bone, merging into the metaphysis in a region termed the metadiaphysis. Short tubular bones have a physis on only one side of the bone, at the site of greatest joint motion. Apophyses are ossification centers protruding from a bone that does not articulate with a movable joint and therefore do not contribute to linear bone growth. They often comprise ligament and tendon attachment sites to bone (Fig. 13B-2). The primary ossification center is the first to form, usually in the diaphysis of the chondral precursor. Early diaphyseal ossification occurs around the main diaphyseal nutrient vessel, which splits and extends toward the metaphyses. The secondary ossification centers form in the epiphyses and appear in a predictable pattern. Their formation is triggered by the invasion of juxta-articular blood vessels. During the first 18 months of life, an anastomotic communication of vessels develops across the physis, allowing processes such as infection and tumor to cross readily. From 18 months until physeal closure, these anastomoses do not exist, and the physis serves as a relative barrier to further extension of metaphyseal or epiphyseal pathologic processes.3 The physis is the growth center of long bones and is responsible for bone lengthening. It is composed of parallel columns of chondrocytes in various stages of differentiation into bone. The chondrocytes divide in the germinal zone, along the epiphyseal side of the physis. They continue to grow in the adjacent proliferative zone, further swelling in size in the hypertrophic zone. These cells give rise to the cartilage matrix that eventually ossifies by the process of endochondral ossification within the zone of provisional calcification. Ossification subsequently occurs along the metaphyseal portion of the physis. As skeletal maturity evolves, the physis become thinner and mildly more irregular, finally closing as the bone reaches maturity. Bone widening occurs through osteogenic activity of the surface periosteum. The physis, specifically the junction of the metaphyseal bone with the physeal cartilage, is the weakest element of growing bone. Additionally, diaphyseal cortex is relatively thick and strong, whereas metaphyseal cortex is comparably weaker. Fractures, therefore, often occur at the metaphyseal-physeal junction. The perichondrium is a ring of tissue that is tightly attached to the bone at this location. In contrast, the periosteum enveloping the diaphyseal cortex is loosely attached. Subperiosteal processes (infection, hematoma, tumor) may extend up the diaphysis by elevating this loosely attached periosteum (Fig. 13B-3).

588

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13B-1   Normal knee epiphyseal ossification. Radiographs of the left knee in the anteroposterior (A) and lateral (B) projections demonstrate a fragmented and irregular appearance of the medial aspect of the distal femoral epiphyseal ossification center. This is a normal variation and should not be mistaken for fracture or reactive bone formation.

A

However, the perichondrium provides a relative barrier to further extension into the joint.

IMAGING CONSIDERATIONS Imaging in the pediatric patient presents challenges infrequently encountered in the adult patient. Foremost, children often are not willingly cooperative. This manifests as motion and not only degrades image quality but also may lead to a completely nondiagnostic examination. In the worst case, motion artifact may result in an erroneous diagnosis. Additionally, some examinations require

Figure 13B-2   Supine anteroposterior radiograph of the left hip demonstrates the normal greater trochanteric apophysis, with physeal radiolucent cartilage (arrow) separating it from the subjacent intertrochanteric femur.

B

c­ ontrast injections that may introduce discomfort, leading to motion. Sedation is often required in younger patients to obtain adequate quality images, usually in the setting of longer duration examinations such as magnetic resonance imaging (MRI). Radiographs and computed tomography (CT) can usually be performed without sedation, although some views may need to be repeated. The rapidity of image acquisition afforded by multidetector computed tomographic scanners has diminished the need for sedation ­during CT.4 A second feature of pediatric imaging that is often overlooked is the size of the anatomy being imaged. The pediatric knee is much smaller than the adult knee. Spatial resolution in radiography, CT, nuclear medicine, and MRI is relatively maximized and cannot be increased further when applied to smaller pediatric anatomy. Therefore, by definition, detail of a particular anatomic structure is less well visualized in the pediatric patient than in the adult patient. Magnification makes anatomy look bigger, but does not increase the underlying spatial resolution. This is especially relevant in MRI, in which motion, combined with the smallness of the underlying anatomy, often leads to interpretative challenges. A final consideration in pediatric imaging is radiation dose. Adequate genital shielding should be provided to all patients undergoing radiography and CT. With radiography, dosages are decreased as dictated by the parameters used to obtain the images. With CT, however, incorrect imaging parameters generate perfectly diagnostic examinations but may impose substantial patient radiation dosages. Imaging facilities should be conversant with pediatric CT protocols designed to minimize patient dosage yet provide adequately diagnostic images.5 As an ordering physician, it is imperative to consider as many diagnostic scenarios as possible so that the appropriate imaging procedure may be performed, obviating the need for further imaging and additive radiation dose. The radiologist is at your service in this regard.

Basic Imaging Techniques

A

C

589

B

D

Figure 13B-3   Subperiosteal abscess. Axial (A) and coronal (B) T2-weighted magnetic resonance (MR) images of the ankle demonstrate a T2 bright subperiosteal fluid collection extending nearly circumferentially around the metadiaphyseal region of the distal tibia. The fluid collection extends to the physis but not beyond it (arrows), also seen in the T2-weighted fat-saturated sagittal MR image (C). The abscess is secondary to distal tibial metaphyseal osteomyelitis, evidenced by the abnormal signal in the marrow, high signal on the T2-weighted fat-saturated MR image, and heterogeneous low signal on the T1-weighted sagittal MR image (D).

590

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13B-4   Triplane fracture with threedimensional reconstructions. A, Sagittal re-formatted computed tomographic scan of the left ankle demonstrates a displaced triplane fracture. B, Threedimensional reconstruction of the same ankle allows better visualization of fragment relationships. An angulated elongated fracture of the left distal fibula is also evident.

A

IMAGING TECHNIQUES Radiography Conventional radiography remains the mainstay of initial evaluation of musculoskeletal processes in the pediatric patient. As with adult radiography, two views of any anatomic location are required for adequate assessment. Occasionally, comparison views of the opposite extremity are obtained to differentiate a suspected abnormality from a normal or developmental variation.6 In some cases, this may be done with a single view. Motion may occasionally interfere with image quality either by blurring the image or by leading to suboptimal positioning for assessing a particular anatomic structure. The technologist should review all images for adequacy and repeat views as necessary. Optimally, this is done in conjunction with the interpreting physician so that repeat views or comparison views may be obtained expeditiously.

B

for image acquisition. Cooperative patients can often be imaged without intravenous sedation, but younger patients may require medicinal assistance to remain still for adequate imaging.

Ultrasonography Musculoskeletal ultrasonography is used for selected purposes in the pediatric patient. It is an excellent modality for visualizing nonmineralized structures. It is most commonly used in assessment of the neonatal hip (Fig. 13B-5). Other uses include the assessment of large tendons and soft tissue masses and the presence or absence of joint effusions. As with adult patients, ultrasound may be used to guide aspiration, injection, or biopsy of soft tissue pathologic entities in children.

Computed Tomography CT is performed to provide a more detailed anatomic assessment of soft tissue and osseous injuries. It is often used in preoperative and postoperative assessment of alignment and healing. Two-dimensional re-formations and threedimensional reconstructions often provide superior visualization of injuries (Fig. 13B-4). With current-generation multidetector scanners, sedation is rarely needed. CT is also commonly used in the assessment of bone lesions because it allows for visualization of the internal architecture of the lesion, often permitting a definitive diagnosis.7 CT may also be used to guide biopsy of such lesions.

Magnetic Resonance Imaging As in adult imaging, MRI in children provides superb soft tissue contrast resolution. It is an excellent modality for imaging bone marrow, joints, cartilage, tendons, and soft tissues. Unfortunately, MRI requires a relatively long time

Figure 13B-5   Neonatal hip ultrasound. Coronal ultrasound view of the left hip demonstrates the stippled-appearing, cartilaginous femoral head (arrow) lying within the acetabulum, with normal coverage of the femoral head.

Basic Imaging Techniques

Scintigraphy Bone scintigraphy is performed with technetium-99m methylene diphosphonate (MDP). Pediatric scans are performed similarly to those in adult patients. Immediate flow, blood pool, and delayed-phase images may be obtained. Additionally, single-photon emission computed tomography (SPECT) may also be acquired. SPECT images are especially useful in evaluation of the pediatric spine for the presence of occult pars stress fractures.8 Compared with adult scans, pediatric bone scans are typically more difficult to interpret. First, spatial resolution is diminished because of the smaller size of the imaged structures. This does not usually produce significant diagnostic difficulties. Second, the physeal growth centers are metabolically active and take up a tremendous amount of radiotracer in children (Fig. 13B-6). Thus, pathologic processes (e.g., fractures, tumors, infection) in close proximity to an active physis may be obscured. Spot images may be acquired in select cases to distinguish periphyseal pathologic processes from activity in the adjacent physis.

IMAGING SPECIFIC PEDIATRIC STRUCTURES Bone Cortical bone appears similar on both pediatric and adult imaging studies, including radiography, CT, and MRI, with the only difference being the relatively thinner appearance of pediatric bone. On the other hand, medullary bone has a different appearance in the pediatric skeleton compared with the adult skeleton. Although the medullary bone is similar, the bone marrow it contains is quite different.

A

591

Pediatric patients have a pattern of red and white marrow distribution that evolves with age and is different from that in the adult patient.9 Although red and yellow marrow are indistinguishable on radiography, CT and MRI are both capable of differentiating them, and MRI is far superior. Fatty marrow is bright on T1-weighted images and intermediate in signal on T2-weighted images. It is low in signal on fat-suppressed images. Red marrow is intermediate to low in signal intensity on T1- and T2-weighted images. The pattern of red and yellow marrow distribution in the skeleton and within a bone changes with age. At birth, nearly the entire osseous skeleton is composed of red marrow. When epiphyses and apophyses ossify, they contain red marrow only transiently before converting to yellow marrow. Conversion from red to yellow marrow proceeds from the extremities to the axial skeleton, beginning in the distal bones of the hands and feet and proceeding proximally in a relatively symmetrical manner. Within a long bone, epiphyses and apophyses initially convert, followed by the center of the diaphysis, distal metaphysis, and finally the proximal metaphysis. If there is an increased demand for hematopoiesis (chronic anemia, leukemia), ­reconversion may also occur. It generally proceeds in the reverse pattern of conversion. There are natural variations in red and yellow marrow distribution from patient to patient. Small differences are normal, but significant asymmetries are worrisome for an underlying marrow infiltrative process. The appearance of the physis varies depending on the age of the patient, the specific bone imaged, and the imaging modality employed. With radiography, the physis appears as an area of transverse lucency at the junction of the metaphysis and epiphysis. If the epiphysis has not yet ossified, it is not defined as a discrete linear structure but rather merges with the epiphyseal cartilage. The physis may be wide, narrow, smooth, or irregular, depending on the specific bone imaged (Fig. 13B-7). Comparison views

B

Figure 13B-6   Normal pediatric bone scan. A, Whole-body bone scan in a 13-year-old boy shows physiologic uptake throughout the skeleton. There is normal physeal uptake, seen particularly around the knees (arrow), ankles, and wrists. B, Bone scintigraphic spot images of the hands in the same patient show normal uptake in the distal radial and ulnar growth plates.

592

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13B-7   Normal growth plates. A, Anteroposterior radiograph of the knee shows a normal lucent appearance of the distal fibular, proximal tibial, and proximal fibular growth plates in a 12-year-old girl. B, Anteroposterior internal rotation view of the right shoulder shows a normal appearance of the growth plates of the tip of the acromion, tip of the coracoid, and proximal humerus. In internal rotation, the proximal humerus demonstrates a double lucency that should not be mistaken for fracture (arrows). C, Anteroposterior view of the left hand shows normal distal radial and ulnar physes. Additionally, growth plates along the distal aspect of the metacarpals are well visualized. No growth plate is seen along the proximal aspect of the second to fifth metacarpals, which is normal. The proximal aspect of the first metacarpal, adjacent to a mobile joint, has its own growth plate.

C

may be helpful in difficult cases. With age, the physis gradually narrows, finally disappearing in the mature skeleton. On CT, the physis has a similar appearance to that seen radiographically. Using two-dimensional re-formations, CT affords a more detailed examination of the structure of the physis. MRI demonstrates the cartilaginous physis as well as the adjacent metaphysis and epiphysis. The physis typically has a high signal on fat-suppressed T2-weighted and proton density images, reflecting water content of the physeal cartilage. It is higher in signal than adjacent epiphyseal cartilage (Fig. 13B-8). As the physis matures, the high signal thins as the thickness of cartilage within the physis decreases. As described earlier, scintigraphy demonstrates

intense radiotracer uptake at the physis proportional to its osteoblastic activity. Maturing physes demonstrate proportionately less activity, whereas the closed and fused growth plate is quiescent.

Cartilage Normal cartilage is soft tissue attenuation in both pediatric and adult patients. It is not readily visualized on either radiographs or computed tomographic scans. Loss of cartilage can be inferred by the assessment of relative joint space narrowing. Cartilage can be visualized on ultrasound. It is typically hypoechoic with through-­transmission of the ultrasound beam. In young pediatric patients with

Basic Imaging Techniques

A

B

C

D

large unossified cartilage structures such as the femoral or humeral heads, bright spicular reflectors seen within the cartilage correspond to vascular channels. Ultrasound is often used in evaluation of developmental dysplasia of the neonatal hip.10 Like adult cartilage, pediatric cartilage is best visualized using magnetic resonance techniques. Cartilage is uniformly hypointense on T1-weighted images and intermediate in signal on T2-weighted fat-saturated images. T2-weighted images provide excellent contrast of cartilage with adjacent high-signal joint fluid and low-signal subchondral bone. High-field magnets permit greater spatial resolution and therefore better morphologic assessment of cartilage integrity. Growth plate cartilage is typically higher in signal on T2-weighted images than is epiphyseal cartilage.11 Hyaline cartilage and epiphyseal unossified cartilage do not demonstrate increased uptake on radionuclide bone scans. However, the growth plates along epiphyseal and apophyseal growth centers demonstrate intense radiotracer uptake.

593

Figure 13B-8   Magnetic resonance (MR) images of normal physis. A and B, Sagittal T1-weighted and fat-saturated T2-weighted MR images of the knee in a 14-year-old girl demonstrate a normal appearance of the distal femoral and proximal tibial physeal cartilage. Nonossified cartilage has low signal on T1-weighted MR images and high signal on T2-weighted MR images. The tibial physis is slightly thinner than the femoral physis. C and D, Sagittal T1-weighted and fat-saturated T2-weighted MR images of the same knee obtained 2 years later demonstrate progressive physeal closure. The tibial physis has completely closed, whereas the femoral physis remains open, although it is thinner than it appears in B.

Soft Tissues Soft tissues in the pediatric musculoskeletal system have a similar appearance to their adult counterparts across imaging modalities. These tissues include tendons, ligaments, menisci, labra, bursae, and muscles. For the most part, all these structures are best visualized using the multiplanar capabilities and excellent soft tissue contrast afforded by MRI techniques. Ultrasound is often useful in the differentiation of solid and cystic mass lesions and is occasionally used in the assessment of tendons of the extremities.

IMAGING PEDIATRIC MUSCULOSKELETAL PROCESSES Trauma In pediatric patients, dislocations and ligamentous injuries are relatively less common than osseous injuries.12 Additionally, diagnosis of fractures in children is more difficult

594

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13B-9   Displaced Salter-Harris I fracture. A, Anteroposterior externally rotated image of the left shoulder demonstrates abnormal alignment of the proximal humeral ossification center with the humeral metaphysis without cortical fracture. There is diffuse soft tissue swelling. Periosteal reaction in the distal humerus (arrow) is secondary to a subacute, healing fracture in this patient who suffered recurrent child abuse. B, Coronal T2-weighted fat-saturated MR image of the same shoulder demonstrates a medially displaced epiphysis consistent with a displaced Salter-Harris I fracture. The periosteum is stripped from the medial aspect of the proximal humeral shaft (arrow), and there is a large joint effusion and soft tissue swelling.

because of the smaller size of the bones, the presence of the radiolucent growth plate, and the variable amounts of epiphyseal ossification. MRI often provides significantly more information with respect to the extent of a childhood fracture than does plain film radiography (Fig. 13B-9). The pattern of fracture in the child depends on the stage of skeletal maturation. Compared with the adult skeleton, pediatric bone is more pliable and may actually bend without completely fracturing. This is more commonly seen in infants and toddlers. As a general rule, childhood fractures heal more quickly than their adult counterparts, and nonunion is decidedly infrequent. The periosteum and endosteum are capable of

exuberant callus formation, much more so than in adults. Fracture healing also occurs in the setting of the constantly remodeling and growing pediatric bone. This may lead to alterations of growth, eventually manifesting as bone deformity. Compared with those in adults, childhood fractures are capable of greater bony remodeling after healing, leading to correction of significant angular deformities and fragment offset. Angular deformities are more readily remodeled and corrected in the plane of adjacent joints than perpendicular to those joints. In young children, incomplete fractures are common and reflect the types of activity these children are engaged in. A bowing fracture reflects a bend in the bone without a

Figure 13B-10   Buckle fracture of the radius and ulna. Anteroposterior (A) and lateral (B) radiographic images of the left wrist and distal forearm demonstrate buckle fractures of the distal radius and ulna (arrows), with minimal angulation. There is diffuse soft tissue swelling.

A

B

Basic Imaging Techniques

A

B

595

C Figure 13B-11   Salter-Harris IV fracture of the knee. Anteroposterior (A) and lateral (B) radiographic images of the right knee demonstrate a linear fracture line through the mid aspect of the epiphysis extending through the physis and exiting from the medial aspect of the metaphysis. The fracture is nondisplaced. There is a joint effusion present on the lateral view (arrow). The fracture line is also well visualized on the coronal T1-weighted (C) and T2weighted fat-saturated (D) (MR) images (arrows). Sagittal T2-weighted fat-saturated MR image (E) also demonstrates a large joint effusion and a fluid-fluid level (arrow), confirming an intra-articular fracture.

D

E

cortical break. A greenstick fracture is an incomplete fracture with a cortical break along the tensile (convex) side of the cortex but not on the compressive (concave) side. When the site of cortical failure occurs along the compressive side of the bone, a buckle fracture results, with a protuberance of bone at the site of deformity. In the wrist, along the distal metaphysis of the radius, this is commonly called a torus fracture (Fig. 13B-10). The physis lies at the center of traumatic injuries to the immature skeleton. It is the weakest part of the developing bone. Childhood fractures are often characterized by their involvement of the physis according to the Salter-Harris classification system. A Salter-Harris I fracture represents a separation of the physis. Salter-Harris II fracture has a fracture plane through the physis and metaphysis, whereas a

Salter-Harris III fracture involves the physis and epiphysis. A Salter-Harris IV fracture involves the metaphysis, physis, and epiphysis (Fig. 13B-11). Finally, a Salter-Harris V fracture represents a crush injury to the physis. All these injuries may result in growth abnormalities from physeal damage, with Salter-Harris V injures having the greatest likelihood. The most important fracture not to be overlooked in the pediatric patient is the nonaccidental fracture. Nonaccidental trauma is a significant cause of morbidity and mortality in children.13 It is an entity that crosses all gender and socioeconomic boundaries. By law, the interpreting physician is required to report findings compatible with child abuse. The diagnosis of child abuse depends on the evaluation of a particular fracture with respect to the expected

596

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13B-12   Metaphyseal corner fracture. Lateral (A) and oblique (B) images of the lower left leg in a toddler show a circumferential metaphyseal corner fracture of the distal tibia. When circumferential around the metaphysis, it is termed a bucket handle fracture. It is nearly pathognomonic of child abuse.

A

a­ ctivities of the child, the skeletal maturation of the child, and the fracture pattern. There are several characteristic findings in nonaccidental trauma. Metaphyseal corner fractures, also called bucket handle fractures, are commonly seen in the distal long bones, including the femur, humerus, radius, and tibia. They are considered pathognomonic of child abuse (Fig. 13B-12). Spiral fractures in the long bones of infants and toddlers are also suspicious. Unusual fractures, including fractures of the metacarpals, posterior rib fractures, skull fractures, scapular fractures, or spinous process fractures, may be seen. Finally, fractures in various stages of healing are also highly suspicious for child abuse. To this end, a radiographic skeletal survey or a bone scan is often performed in young children to screen for further occult trauma.14

Figure 13B-13   Clavicular greenstick fracture. Frontal view of the right clavicle and shoulder demonstrates a focal cortical break along the superior aspect of the middle third of the clavicle with minimal angulation. The inferior cortex is intact. Findings are consistent with a greenstick fracture of the clavicle.

B

Extremity Fractures In the upper extremities, the most common fracture in the pediatric patient involves the clavicle, usually occurring in the middle third. Greenstick fractures are common and these fractures usually heal completely (Fig. 13B-13). The second most common site is the surgical neck of the humerus. The growth plate, corresponding to the anatomic neck of the humerus, should not be mistaken for a proximal humeral fracture, especially on internal rotation views where the growth plate appears at two levels (see Fig. 13B-7). The elbow is a common site of fracture. The presence of an elbow joint effusion, delineated by distention of the joint and displacement of adjacent anterior and posterior fat

Basic Imaging Techniques

597

Figure 13B-14     Supracondylar fracture. Anteroposterior (A) and lateral (B) radiographic images of the left elbow demonstrate a supracondylar fracture of the distal humerus with minimal posterior angulation. There is a large elbow joint effusion, well seen on the lateral view by displacement of the normal anterior and posterior fat pads (arrows).

A

B

pads, indicates a high likelihood of fracture.15 In younger children, a supracondylar distal humeral fracture is most common, whereas in adolescents, radial head fractures are more common (Fig. 13B-14). Medial condyle avulsion fractures (little leaguer’s elbow) result from the action of the flexor carpi ulnaris. The avulsed fragment may displace, becoming trapped in the joint and leading to mechanical symptoms. Knowledge of the normal sequential appearance of the ossification centers of the elbow is important to distinguish a displaced medial condylar fragment from a normal trochlear ossification center. Elbow dislocations are common in the child and are invariably posterior. Avulsion of the medial epicondyle is a common concomitant injury. In the infant, nursemaid’s elbow results from forceful traction on the elbow resulting in the radial head dislocating from the annular ligament. The dislocation is usually readily reduced by supination. Distal radial and ulnar fractures are also common in children. These usually take the form of buckle or torus fractures of the distal radius and occasionally the ulna. Bowing deformities without fracture occur in the midshaft of the radius and ulna. The both-bones forearm fracture refers to fractures to the midshaft of the radius and ulna, usually transverse and nondisplaced. Physeal injuries to the distal radius are also common. More distal carpal and metacarpal fractures are uncommon in children but increase in frequency in the teenage years and adulthood. In the lower extremities, fractures of the hip and ankle are most common in children. In the hip, a slipped capital femoral epiphysis describes a fracture of the proximal femoral growth plate. It is more common in obese adolescent boys, particularly African Americans.16 The femoral epiphysis usually slips posteriorly and inferomedially. The fracture differs slightly from a classic Salter-Harris I fracture in that it occurs between the proliferative and hypertrophic zones of the physis as opposed to through the hypertrophic zone only. Both mechanical and ­endocrine conditions have been implicated as potential causes (Fig. 13B-15).

Fractures around the knee are relatively uncommon. They typically take the form of Salter-Harris II or III fractures to either the distal femur or proximal tibia. The proximal tibial physis is the most common site of Salter-Harris V fractures, occurring in the setting of impaction injuries. Patella fractures are uncommon and are usually transverse in orientation. Developmental patellar variations, such as a bipartite or tripartite patella, should not be confused with patellar fractures, although these variations are occasionally a source of knee pain. Toddlers occasionally develop a spiral fracture to the distal tibial shaft, best diagnosed using oblique radiographs. In the ankle, there is an 18-month window in older childhood and adolescence during which the distal tibial physis closes asymmetrically. Generally, the physis closes from medial to lateral and posterior to anterior, leaving the anterolateral physis vulnerable to injury. Triplane fractures are Salter-Harris IV fractures, named for fracture extension in coronal, transverse, and sagittal planes. These fractures occur secondary to external rotation of the foot with respect to the distal tibia. The fracture extends through the epiphysis in the sagittal plane, travels horizontally through the unfused lateral aspect of the physis, and finally exits the metaphysis in the coronal plane. Juvenile Tillaux fractures are Salter-Harris III fractures similar to triplane fractures but without the metaphyseal component. The Tillaux fracture is secondary to avulsive forces of the anterior talofibular ligament along the anterolateral aspect of the distal tibial epiphysis during external rotation of the foot (Fig. 13B-16).17 Avulsive injuries are common in adolescent athletes. These often occur at apophyses, the attachments of tendons to bone.18 Avulsion fractures usually present after a forceful muscle contraction with subsequent pain and limited motion. Abnormal forces across joints may lead to avulsion injuries along the attachments of ligamentous and ­capsular structures. These injuries may be subtle, with only a small fleck of cortical bone avulsed. Chronic microstresses

598

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 13B-15   Slipped capital femoral epiphysis. A, Anteroposterior radiographic view of the left hip shows malalignment of the femoral epiphysis with the subjacent femoral metaphysis in an adolescent. The patient has a large body habitus, as indicated by the generous soft tissues on the film. B and C, Coronal and sagittal T2-weighted fat-saturated MR images of the same hip confirm malalignment of the femoral epiphysis with slippage posteriorly and inferomedially. Minimal edema in the physis (arrows) and a small joint effusion are also seen.

A

B Figure 13B-16   Juvenile Tillaux fracture. Coronal (A) and sagittal (B) computed tomographic re-formations of the ankle demonstrate a minimally displaced fracture of the anterolateral tibial epiphysis. The medial aspect of the physis is fused, whereas the lateral aspect of the physis is widened (arrow). The minimally displaced anterolateral Tillaux fragment is well seen in the axial   (C) computed tomographic scan.

C

Basic Imaging Techniques

may also lead to avulsion fractures or stress changes in the underlying bone, termed apophysitis. In adolescents, the pelvis is the most common site of avulsion injuries. Typical locations include the rectus femoris attachment along the anteroinferior iliac spine, sartorius attachment to the anterosuperior iliac spine, and the hamstring insertion to the ischial tuberosity. Other sites include the knee, ankle, foot, shoulder, and wrist. The presence of some avulsion fractures is highly associated with other ligamentous ­injuries. Radiography is usually diagnostic in the evaluation of avulsion fractures. CT provides a more detailed cross­sectional view, allowing more accurate assessment of avulsed fragment size and displacement. CT is especially valuable in detection of subtle avulsion injuries in complex structures such as the hand, wrist, and foot. MRI is less sensitive in the detection of avulsion fractures than CT. Avulsion injuries generally produce far less marrow edema than linear fractures.19 Additionally, tiny avulsed cortical bone fragments are devoid of signal on MRI and may be overlooked. Conversely, stress apophysitis secondary to chronic avulsive forces is well visualized using MRI, demonstrating prominent bone marrow edema. Healing avulsion injuries may produce a diagnostic dilemma. In adolescent athletes, avulsion injuries often heal with exuberant hypertrophic bone formation, often mimicking a more sinister malignant lesion on radiographs. In these cases, CT or MRI plays a role in evaluation, allowing visualization of the healing avulsion fracture and excluding the presence of a destructive lesion.20

Osteochondroses Osteochondroses are defined as noninflammatory and noninfectious derangements of bone growth that occur at bone growth centers during periods of activity. Disordered bone growth results in morphologic alterations to the epiphysis or apophysis. Although the exact cause of osteochondroses is unknown, trauma is thought to play a significant role with subsequent compromise of blood supply, bone necrosis, and finally bone regrowth. Osteochondritis dissecans most commonly affects teenagers and young adults. It is likely post-traumatic in origin, secondary to impaction injuries.21 The condition is most common in the lateral aspect of the medial femoral condyle followed by the medial talar dome. It is occasionally bilateral. Radiographs demonstrate a linear lucency in the subchondral bone (Fig. 13B-17). If the fragment is free floating, there may be joint locking. MRI plays an important role in the assessment of the osteochondral fragment.22 If the fragment remains in situ with intact overlying cartilage, surgical intervention is not deemed necessary. ­Magnetic resonance arthrography is often useful in assessment of potentially loose osteochondral fragments. Avascular necrosis of the femoral capital epiphysis, termed Legg-Calvé-Perthes disease, is a common cause of hip pain with maximal incidence in the 5- to 10-yearold patient. It is more common in boys by a 5:1 margin and is usually unilateral. Patients present with slowly progressive joint pain and gait disturbances. Early diagnosis may be made with MRI that shows typical findings of bone necrosis.23 Radiography demonstrates more

599

progressive ­ findings, including subchondral fractures, fragmentation, and flattening of the femoral head with mixed lucency and sclerosis, and increased joint spacing due to swollen epiphyseal cartilage and joint effusion (Fig. 13B-18). Osgood-Schlatter disease is osteochondrosis of the tibial tubercle. It is frequently bilateral, commonly seen in teenagers, with boys outnumbering girls 3:1. Clinical findings most commonly involve pain over the tibial tubercle. This entity, also termed tibial tubercle stress apophysitis, results from chronic repetitive injury to the tibial attachment of the infrapatellar tendon, often a result of jumping sports.24 This leads to a fragmented appearance of the tibial tubercle with ossification in the distal infrapatellar tendon. Radiographs are supportive of the diagnosis. MRI may demonstrate bone marrow edema in the tibial tubercle as well as edema in the attachment of the infrapatellar tendon. Sinding-Larsen–Johansson disease is osteochondritis of the infrapatellar tendon attachment to the inferior patellar pole. It demonstrates changes similar to those seen in Osgood-Schlatter disease but along the inferior margin of the patella.25 Köhler’s disease is a relatively rare osteochondrosis affecting the tarsal navicular. It affects children in the 3to 5-year age range, typically boys. Patients complain of midfoot pain with gait disturbances. Radiographs demonstrate an irregular, sclerotic, and occasionally flattened navicular bone. The condition is usually self-limited, with the navicular resuming a normal shape 2 to 4 years after presentation.26 Freiberg’s infraction usually affects the head of the second metatarsal. It presents most commonly in teenagers with foot pain. Radiographs demonstrate flattening of the head of the second metatarsal. Over time, degenerative arthritis of the second metatarsophalangeal joint ensues. Blount’s disease, also termed osteochondrosis deformans tibiae, is a developmental deformity of the proximal tibia. Three forms have been described: infantile, juvenile, and adolescent.27 All have in common abnormal stress placed on the posteromedial tibia, leading to growth suppression of the posteromedial physis. The decrease in longitudinal growth that ensues leads to further varus angulation and progressive worsening of the condition. Radiographic findings include sloping and fragmentation of the medial epiphyseal ossification center, widening of the growth plate, and “beaking” of the medial metaphysis. MRI is often helpful in the evaluation of the growth plate, especially in the case of adolescent Blount’s disease, which is often due to premature fusion of the medial portion of the proximal tibial growth plate.28

Infection and Inflammation Osteomyelitis In children, osteomyelitis is usually hematogenous in origin, either from transient asymptomatic bacteremia or sepsis. It occurs most commonly in the metaphyses, which contain slow-flowing venous sinusoids at the terminal segments of medullary vessels. The physis represents a relative barrier to epiphyseal spread. However, transphyseal spread may occur in infants younger than 18 months owing

600

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 13B-17   Osteochondritis dissecans. Anteroposterior (A) and lateral (B) views of the left knee demonstrate a subtle area of lucency in the lateral aspect of the anterior weight-bearing portion of the medial femoral condyle (arrows). Coronal T1-weighted (C) and sagittal T2-weighted fat-saturated (D) MR images confirm the presence of an osteochondral lesion at this location. The fragment remains in situ, and the overlying cartilage is intact.

A

B

C

D

Figure 13B-18   LeggCalvé-Perthes disease. Axial T1-weighted (A) and coronal T2-weighted fat-saturated   (B) MR images of the right hip after intra-articular gadolinium injection demonstrate coxa magna with a flattened, widened femoral epiphysis, findings consistent with longstanding Legg-Calvé-Perthes disease.

A

B

Basic Imaging Techniques

A

601

B Figure 13B-19   Pubic osteomyelitis. Axial T2-weighted fat-saturated (A) and T1-weighted fat-saturated gadoliniumenhanced MR images in the axial (B) and coronal (C) projections demonstrate edema in the right pubis with prominent enhancement of the bone and adjacent soft tissues. Findings are consistent with pubic osteomyelitis. A small, peripherally enhancing soft tissue abscess (arrow on B) is present on the postgadolinium images.

C

to the presence of transphyseal vessels. This may allow the ­development of subsequent septic arthritis. Increased intramedullary pressure secondary to infection may lead to bone necrosis. Alternatively, increased pressure often results in transcortical extension of infection. The relatively loose attachment of the periosteum along the metaphysis and diaphyseal regions allows the longitudinal dissection of subperiosteal abscesses along the shaft of bone (see Fig. 13B-3). Further extension through the periosteum may lead to soft tissue involvement. The firm attachment of the periosteum at the physis often prevents extension into the joint, except in the case of the proximal femur, proximal humerus, distal tibia and fibula, and proximal radius, where the metaphyses are intra-articular, thus permitting direct seeding of these joints. The most common sites of osteomyelitis are the metaphyses of actively growing large bones. This includes the proximal and distal ends of the femur, proximal tibia and humerus, and distal radius. Flat bones, specifically

the ilium, vertebrae, and calcaneus, are also commonly involved. The most common organism in the pediatric population is Staphylococcus aureus. Group B streptococci are also a common cause in the neonatal population. Radiographs are insensitive to the early changes of osteomyelitis. Bone destruction may take up to 10 days to become evident. Soft tissue changes such as soft tissue swelling and obliteration of normal fat planes, although subtle, are detected earlier than bone changes. The diagnosis of osteomyelitis can be made much earlier with MRI, which shows changes of bone marrow edema in the involved metaphysis. MRI is also capable of showing complications of osteomyelitis such as intraosseous abscess formation, subperiosteal abscess formation, and soft tissue extension (Fig. 13B-19).29 Three-phase scintigraphy is also sensitive to the detection of osteomyelitis, although interpretation may be confounded by activity in the developing physis adjacent to an area of metaphyseal osteomyelitis (Fig. 13B-20).

602

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 13B-20   Humeral osteomyelitis. A, Frontal radiograph of the right humerus in a young child shows abnormal periosteal new bone formation along the distal diaphysis and metaphysis of the right humerus. B, Bone scan demonstrates asymmetrically increased uptake along the distal metaphysis of the right humerus ��(arrows) ������������������������������������������������ compared with the left humerus. Biopsy revealed osteomyelitis.

Septic Arthritis Septic arthritis is a relatively common clinical emergency in pediatric patients. It may develop hematogenously, directly seeded from synovial blood vessels, by contiguous extension from adjacent metaphyseal osteomyelitis or soft tissue cellulitis, or rarely from a direct puncture wound. Once the joint is infected, it effectively becomes an abscess. Increased pressure in the joint leads to compromised blood supply to the epiphysis. In combination with active infection, this leads to rapid bone and cartilage destruction and dissolution. Imaging findings in septic arthritis are nonspecific. Joint effusions are common, although they may be absent in small joints in which the relatively thin joint capsule allows decompression into the adjacent tissues.30 Imaging, especially MRI, allows for the detection of joint effusions as well as other potential causes such as osteomyelitis, soft tissue cellulitis, tumors, or even occult trauma. Transient toxic synovitis is a self-limited condition that may mimic septic arthritis. It is thought to be viral in origin and most often involves the pediatric hip. It often produces a joint effusion, but patients are likely to have less pronounced symptoms than with a truly septic joint. In general, however, there should be a very low threshold for arthrocentesis of a suspected infected joint.

in the first few months of the disease. Extra-articular manifestations include uveitis, iritis, and iridocyclitis. Polyarthritis affects five or more joints and is often symmetrical in distribution. It has a greater predisposition for small joints than does oligoarthritis, which tends to affect larger joints. Systemic arthritis, also termed Still’s disease, is a systemic disease that affects joints as well as internal organs. Patients have a fever and rash. Eye involvement is rare. A fourth category of juvenile idiopathic arthritis is the psoriatic variant, in which patients with an inflammatory arthropathy also have typical psoriatic skin and nail manifestations. Finally, enthesitis-related arthritis affects the spine, hips, and entheses (tendon attachments to bone). It is more common in males and in families with a history of spondyloarthropathies. Eye involvement is also common. The diagnosis of juvenile idiopathic arthritis may be difficult. It is often a diagnosis of exclusion, arrived on after other entities have been excluded. Imaging studies are nonspecific, demonstrating joint effusions and synovial proliferation. MRI may aid in demonstrating the condition of

Inflammatory Arthritis Juvenile idiopathic arthritis, formerly juvenile rheumatoid arthritis, is the newest terminology for a group of poorly understood arthropathies that affect children.31 It is a chronic inflammatory arthropathy in which the immune system targets the synovium, leading to synovial hypertrophy and hyperemia with joint effusions. In addition to bone and cartilage destruction that result from chronic joint inflammation, the chronic hyperemic state also leads to growth disturbances, including epiphyseal overgrowth and thin gracile diaphyses. This may lead to wildly deformed joints depending on the severity of the disease (Fig. 13B-21). Juvenile idiopathic arthritis has been subdivided into five subtypes depending on symptoms and number of joints involved. Oligoarthritis affects less than five joints

Figure 13B-21   Juvenile idiopathic arthritis. In a child, anteroposterior radiographs of the bilateral hands show severe deformity as well as osteopenia, joint space narrowing, and carpal destruction with collapse. Growth disturbances manifest as ballooning epiphyses and shortened metacarpals.

the joints, specifically the articular cartilage. Radiographs

Basic Imaging Techniques

C

A

B

603

D

are also helpful in following progression of the disease and assessing bone growth and deformity.32

Benign Lesions Several benign bone conditions may affect the pediatric patient. Some of these are painful and require surgical intervention. They may mimic sports-related injuries such as stress fractures. Other benign bone conditions are entirely incidental findings that may be seen on radiographs obtained for other reasons. These should not be mistaken for malignant lesions. Finally, some lesions, although benign, may predispose the patient to pathologic fracture. Preventive intervention should be considered with respect to lesion morphology and the child’s activity level.

Figure 13B-22   Osteoid osteoma. Axial (A) and coronal re-formation (B) computed tomographic images of the left forearm demonstrate circumferential marked periosteal new bone formation along the shaft of the ulna with a central lucent nidus (arrows) corresponding to an osteoid osteoma. Corresponding axial (C) and coronal   (D) T2-weighted fat-saturated magnetic resonance images show both periosteal and cortical high signal edema along the involved portion of the bone.

Osteoid osteomas are benign bone-forming lesions often encountered in children and young adults. Patients classically complain of pain that is worse at night and relieved by aspirin. Lesions may be seen anywhere but are most common in the femur, tibia, and posterior elements of the spine. A central lucent nidus, usually smaller than 1 cm, may be intramedullary, cortical, or periosteal. The nidus often has a small focus of central mineralization. These lesions are extremely inflammatory. There is often an exuberant, thick, benign-appearing periosteal reaction and sclerosis. If intra-articular, there may be a joint effusion. MRI demonstrates both periosteal and soft tissue edema around the lesion, best seen on T2-weighted fat-suppressed or STIR sequences (Fig. 13B-22). CT is the examination of choice for demonstrating the nidus and the adjacent periosteal reaction. CT may also be used to guide a radiofrequency probe into the

604

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

C Figure 13B-23   Chondroblastoma. Anteroposterior (A) and lateral (B) radiographs of the right knee demonstrate a subtle lytic lesion in the posteromedial aspect of the proximal tibial epiphysis in a young patient with knee pain and limp (arrows). Sagital STIR (C), T1-weighted (D), and   T1-weighted fat-saturated gadolinium-enhanced   (E) MR images of the same knee confirm the presence of a cartilaginous lesion (arrow) and show extensive adjacent epiphyseal enhancement and edema consistent with the robust inflammatory response elicited by a chondroblastoma.

D

E

Basic Imaging Techniques

A

B

605

C

Figure 13B-24   Giant cell tumor. Coronal T1-weighted (A), coronal (B), and sagittal (C) T2-weighted fat-saturated magnetic resonance images of the knee demonstrate an eccentric low-signal lesion in the lateral femoral condyle extending to the subchondral cortex. Biopsy confirmed giant cell tumor. Note that the physes are closed, as is typical with these lesions.

Figure 13B-25   Nonossifying fibroma. Anteroposterior radiograph of the knee demonstrates an eccentric, lobular lesion along the lateral aspect of the distal femoral diaphysis with a well-marginated sclerotic border corresponding to a benign nonossifying fibroma (arrow).

lesion to effect ablation, which has a high percentage of therapeutic ­success.33 Chondroblastomas are infrequent benign cartilaginous epiphyseal or apophyseal lesions seen most commonly in teenagers. They are typically seen around the knee as well as the distal humerus. Patients present with pain, tenderness, swelling, and occasionally a joint effusion. Plain film radiographs demonstrate a round or oval, well-marginated, lucent lesion eccentrically in the epiphysis. Most lesions have a small amount of internal matrix, although a large percentage are uniformly lucent. Occasionally, a benign periosteal reaction is seen. Chondroblastomas are demonstrated well by MRI; they are low in signal on T1-weighted images and heterogeneous in signal on T2-weighted images owing to the variable internal cartilage stroma and mineralization. Commonly, adjacent inflammatory changes manifest as bone marrow edema well out of proportion to the size of the actual tumor (Fig. 13B-23). Giant cell tumors are uncommon in young children, rarely appearing in a bone before physeal closure. Most occur in young adults, who present with either low-grade pain or pathologic fracture. These tumors are usually eccentric, lytic, and located along the subarticular end of the bone. They typically lack a sclerotic rim. MRI readily displays the tumor and allows assessment of cortical and subchondral breakthrough (Fig. 13B-24). Benign fibrous cortical defects, also termed nonossifying fibromas, are benign asymptomatic lesions of bone often seen incidentally. They are cortical lesions, often seen eccentrically in the metadiaphyseal regions of the long bones, particularly around the knee. Radiographically, they are well marginated and lobular and have a well-defined sclerotic margin (Fig. 13B-25). Larger nonossifying fibromas may lead to pathologic fracture. Over time, the lesions involute, with bone filling the fibrous lucent defect. The lesions are low in signal on both T1- and T2-weighted MRI, with a thick low signal rim corresponding to peripheral sclerosis.

606

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B Figure 13B-26   Unicameral bone cyst with pathologic fracture. A, Anteroposterior view of the left humerus demonstrates an expansile, elongated, lytic lesion in the proximal metaphysis of the left humerus. A few thin internal bony septations are seen. B, Frontal view of the left shoulder in the same patient 4 days later shows a pathologic fracture through the lesion with a fallen fragment of bone in the dependent aspect of the bone cyst (arrow).

MRI is not necessary in their evaluation because radiographs are usually diagnostic. Intraosseous hemangiomas are commonly seen in the spine, skull, and facial bones. Typical radiographic and CT findings are prominent thickened vertical trabeculations. There may be mild bone expansion. MRI often demonstrates macroscopic fat within the lesion. Rarely, these lesions are symptomatic.

Solitary or unicameral bone cysts are common bone lesions usually found in the proximal metaphysis of the humerus or femur. Radiographically, they appear as expansile, well-marginated lucent lesions with internal bony septations. They are asymptomatic unless a pathologic fracture has resulted from the weakened bone. A “fallen fragment”, corresponding to a free osseous fragment in the dependent portion of the lesion, is seen in a fair number of cases of pathologic fracture (Fig. 13B-26). MRI demonstrates the cystic nature of the lesion, uniformly bright on T2-weighted images. Solitary bone cysts differ from aneurysmal bone cysts in that the latter contain blood-filled cavernous spaces, are reactive in origin, and are often symptomatic. Aneurysmal bone cysts are often expansile and may break through the cortex. On MRI, they commonly demonstrate dependent fluid-fluid levels. Osteochondromas, also termed exostoses, are a common form of bone dysplasia. The lesion forms from a metaphyseal outgrowth of cartilage, usually in a tubular bone such as the femur, tibia, or rib. The exact cause is unknown. The island of cartilage grows as an epiphysis grows, eventually ossifying. By definition, the marrow cavity of the lesion is contiguous with the marrow of the subjacent parent metaphysis, and there is a variable thickness cartilage cap that thins with age. The lesions may be sessile, with a broad base of attachment, or exophytic. The latter lesions may extend several centimeters into adjacent soft tissues and result in mechanical symptoms such as myositis, adventitial bursal formation, or nerve impingement (Fig. 13B-27). They are also prone to fracture. These lesions are typically seen incidentally on imaging examinations. If painless, no further work-up is required. If painful, MRI is indicated to assess soft tissue changes as well as the morphology of the lesion, in particular the cartilage cap, which in rare cases may degenerate into chondrosarcoma.34 Rare patients may have numerous osteochondromas, termed hereditary multiple exostoses. These patients have a higher incidence of malignant degeneration than patients with a solitary osteochondroma and should be monitored accordingly. Fibrous dysplasia is a common benign disorder in which bone is replaced by fibrous tissue. It is usually monostotic, although polyostotic forms exist, usually associated with a number of endocrine abnormalities. Fibrous dysplasia typically arises in growing bones of older children and adults. Common locations include the proximal femur, tibia, ribs, and craniofacial bones. Affected bone is predisposed to pathologic fracture, and chronic deformity may result. Lesions are usually asymptomatic. Radiographically, fibrous dysplasia is a well-circumscribed, lytic lesion in a long bone, often with a hazy ground-glass appearance of the internal matrix. It is well marginated, often with a thick sclerotic border. It may be slightly expansile with scalloping of the endosteal cortex, but does not demonstrate cortical breakthrough or a soft tissue mass (Fig. 13B-28). Lesions usually demonstrate increased uptake on bone scintigraphy. This modality is used to assess for polyostotic disease.

Aggressive Lesions Osteosarcoma is the most common primary malignancy of bone in the 10- to 25-year-old group, constituting more than half the malignant bone lesions that occur in

Basic Imaging Techniques

A

607

B

Figure 13B-27   Exophytic osteochondroma. Coronal (A) and sagittal (B) T2-weighted fat-saturated MR images of the knee in an adolescent with medial knee pain and a palpable, painful mass. There is an exophytic osteochondroma arising from the medial metaphysis of the distal femur. Its internal marrow is contiguous with the marrow of the distal femur. Inflammatory changes in the soft tissues are seen surrounding the osteochondroma, with formation of a tiny adventitial bursa (arrow in A).

the first two decades of life. The most common location is the metaphyses of long bones, with greater than 50% of lesions occurring around the knee. Tumors may, however, arise in any bone. Males have a slightly higher incidence than females. The appearance of the lesion varies with the degree of matrix produced by the lesion. Some types are quite osteoblastic with extensive new bone formation. Chondroblastic subtypes produce cartilaginous matrix, whereas others produce predominately fibroblastic stroma.

Figure 13B-28   Fibrous dysplasia. Anteroposterior radiograph of the pelvis demonstrates an expansile, mixed lytic and sclerotic lesion in the proximal left femur corresponding to an area of fibrous dysplasia.

Parosteal osteosarcoma originates from the periosteum and often wraps around the diaphysis, growing outside the bone. This subtype is uncommon in children, with a peak incidence in the third decade. Telangiectatic osteosarcoma is entirely lytic, often appearing cystic. It occurs most commonly around the knee. On radiographs, a typical osteosarcoma appears as a destructive eccentric metaphyseal lesion with mixed lytic and sclerotic regions. The tumor often penetrates the cortex, leading to formation of malignant-appearing periosteal new bone in a sunburst configuration. Radiographs are usually diagnostic. MRI is useful in mapping the true extent of the tumor, within both the bone marrow and adjacent soft tissues (Fig. 13B-29). Tumors are typically of low signal on T1-weighted images, displacing normal fatty marrow. Osteosarcomas are heterogeneously high signal on T2-weighted images and demonstrate heterogenous enhancement. Areas of mineralization appear as signal voids within the tumor mass. Osteosarcomas typically avidly take up radiotracer on bone scintigraphic examination. CT of the chest is often used in the detection of metastatic lesions. Ossifying metastatic lesions may show uptake on bone scans. Ewing’s sarcoma is the most common bone tumor in the first decade of life, usually occurring after 5 years of age. It is twofold more common in males but rare in Asians and African Americans. Children often present with systemic symptoms such as fever, pain, and an elevated white blood cell count. The most common location is the femoral diaphysis. However, lesions are also common in the flat bones of the pelvis, ribs, and spine. Radiographically, Ewing’s sarcoma appears as a predominantly lytic, permeative lesion in the diaphysis of a long bone. It has poorly

608

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B

D

Figure 13B-29   Osteosarcoma. Anteroposterior (A) and lateral (B) radiographs of the right knee demonstrate a lytic lesion in the right femoral metaphysis that shows areas of internal mineralized matrix. The lesion has eroded through the posterior cortex of the femur, extending into the adjacent soft tissues and lifting the periosteum to form Codman’s triangle (arrow in B). Coronal (C) and sagittal (D) T1-weighted MR images reflect the true extent of the tumor within the marrow cavity, with a low-signal tumor replacing the high-signal fatty marrow. Destroyed posterior femoral cortex with soft tissue extension can also be appreciated on the sagittal view.

Basic Imaging Techniques

A

609

B

C

D Figure 13B-30   Ewing’s sarcoma. Anteroposterior (A) and lateral (B) radiographs of the left knee demonstrate a subtle permeative lesion in the distal diaphysis of the femur with mild periosteal reaction. Tissue planes are distorted, indicating a large soft tissue component (arrows). Axial (C) and coronal (D) T1-weighted fat-saturated gadolinium-enhanced MR images of the femur confirm the presence of an elongated mass in the diaphysis of the femur extending into the distal metaphysis. A large soft tissue component extends circumferentially around the femur. Biopsy revealed Ewing’s sarcoma.

610

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

defined margins. Periosteal reaction is common, often appearing either sunburst in configuration or lamellated like an onion skin. Rarely, benign, thick, wavy periostitis is seen. There is often a large soft tissue component to Ewing’s sarcoma. CT and MRI are often helpful in demonstrating the entire osseous and soft tissue extent of the tumor (Fig. 13B-30). Bone scintigraphy demonstrates both the primary lesion and bone metastatic lesions, common in Ewing’s sarcoma. Several other malignant tumors may affect bone. In children younger than 12 months, neuroblastomas are the most common cause of bony destructive lesions. Like Ewing’s sarcoma, these are small blue cell tumors that infiltrate marrow. They are most common in the chest, where they are termed Askin’s tumor. They are often associated with a pleural effusion and have a poor prognosis. Leukemia also commonly involves the skeleton, demonstrating osteopenia and focal osteolytic lesions on radiographs. MRI

C l

r i t i c a l

P

o i n t s

 ediatric patients are imaged similarly to adult patients P but present technical challenges to high-quality image acquisition. l Radiography remains the mainstay for diagnosis and treatment of pediatric orthopaedic disorders. l Radiation dosage in pediatric patients should be closely monitored and protocols adjusted to minimize dosage while maximizing diagnostic quality. l Pain in the pediatric patient often reflects underlying injury, but attention should be given to a host of developmental and neoplastic conditions. l The pediatric skeleton is a unique structure that is constantly changing in appearance as it matures to the adult skeleton. Familiarity with the appearance of these changes is important to distinguish normal developmental processes from pathologic processes. l The physeal growth plate cartilage plays a central role in the appearance of pediatric pathologic processes, including bone fracture patterns, tumor occurrence, and spread of infections.

is sensitive to marrow infiltration from leukemia. Neither leukemia nor neuroblastomas produce tumor matrix.

CONCLUSION Imaging of the pediatric skeleton, although similar to that of the adult skeleton in many respects, often presents both unique technical and diagnostic challenges. At the center of these challenges are the growth plate and epiphyseal ossification centers. Their continued evolution with age is reflected across their appearance on radiography, CT, scintigraphy, and MRI. Familiarity with normal variations is crucial to accurate diagnoses. Pediatric traumatic injuries differ from adult injuries in patterns of injury and healing. Additionally, the complaint of pain in children and adolescents is often not associated with trauma but rather reflects a host of conditions particular to the growing skeleton, including both benign and aggressive entities.

S U G G E S T E D

R E A D I N G S

Baert AL, Johnson KJ, Bache E (eds): Imagining in Pediatric Skeletal Trauma: Techniques and Applications. New York, Springer, 2007. Helms CA, Major NM, Andersen MW, et al: Musculoskeletal MRI, 2nd ed. Philadelphia, Saunders, 2008. Manaster BJ, May DA, Disler DG: Musculoskeletal Imaging: The Requisities, 3rd ed. Philadelphia, Mosby, 2004. Resnick D, Kang HS, Pretterklieber ML: Internal Derangements of Joints, 2nd ed. Philadelphia, Saunders, 2006. Resnick D, Kransdorf M: Bone and Joint Imaging, 3rd ed. Philadelphia, Saunders, 2004. Stoller D, Tirman P, Bredella M: Diagnostic Imaging: Orthopaedics. AMIRSYS, 2003.

R eferences Please see www.expertconsult.com

C H A P T E R  

14

Overuse Injuries Luke Choi Overuse is the leading cause of about half of all sports injuries. The frequency of overuse injuries evaluated in ­primary care sports medicine clinics is even greater, reportedly up to twice the frequency of acute injuries. Most injuries evaluated in running injury clinics are related to overuse, and about half of these involve the lower leg (20%), ankle (15%), and foot (15%). Repetitive microtrauma due to overuse injuries leads to local tissue damage in the form of cellular and extracellular degeneration and is most likely to occur when an athlete changes the mode, intensity, or duration of training— a phenomenon described as the “principle of ­transition.”1 Physical training uses prescribed periods of intense activity to induce the desired goal of “super compensation” or performance improvement. A mismatch between overload and recovery can lead to breakdown on a cellular level, and repetitive overload on tissues that fail to adapt to new or increased demands can lead to tissue breakdown and overuse injury. In theory, however, this subclinical tissue damage can accumulate for some time before the ­ person experiences pain and becomes symptomatic. On the ­ systemic level, rapid increases in training load without adequate recovery may cause a global “overtraining syndrome.” Strong predictors of overuse musculoskeletal injury include a previous history of injury as well as walking or running more than 20 miles per week.2 Both intrinsic and extrinsic factors contribute to overuse injuries. Intrinsic factors are biomechanical abnormalities unique to a particular athlete and include such features as malalignments, muscle imbalance, inflexibility, weakness, and instability. High arches, for example, have been demonstrated to predispose to a greater risk for musculoskeletal overuse injury than low arches (“flat feet”) in military recruits.3 Extrinsic (avoidable) factors that commonly contribute to overload include poor technique, improper equipment, and improper changes in the duration or frequency of activity. These improper changes in activity duration and frequency of “training errors” are the most common causes of overuse injuries in recreational athletes. Vulnerability to extrinsic overload varies with the intrinsic risk factors of an individual athlete. Sports-acquired deficiencies, categorized as an extrinsic risk factor, actually represent the product of biomechanical abnormalities and training errors. Because sports activity can overload an athlete’s musculoskeletal system in predictable ways, athletic repetition without proper conditioning can propagate muscular imbalance and flexibility deficits. Injuries are often related to biomechanical abnormalities removed from the specific site of injury, underscoring the importance of evaluation of the entire kinetic chain. Common overuse injuries include tendinopathies, stress fractures, chronic exertional compartment syndrome, and shin splints.

TENDINOPATHIES Tendinopathy is a clinical condition characterized by activity-related pain, focal tendon tenderness, and intratendinous imaging changes. It represents a common and significant problem, with a prevalence of 14% in elite athletes, and requires a recovery time of 3 to 6 months with first-line conservative management. Despite its clinical significance, only recently have strides been made in understanding the pathology underlying tendinopathy. Historically, it was thought to be one of inflammation, and consequently, the condition was labeled ­ tendinitis. However, recent histopathologic studies have shown the underlying pathology to be primarily one of tendon degeneration (tendinosis).4,5 In contrast to acute traumatic tendon injury, sport-related injuries most often involve repetitive submaximal loading of the tissues, resulting in repetitive microtrauma. An understanding of the anatomic pathophysiologic basis of these conditions is critical to their diagnosis and management (Table 14-1). Although the pathologic label tendinosis has been in use for more than 25 years to describe collagen ­degeneration in tendinopathy, many clinicians still use the term ­tendinitis to describe painful chronic overuse injury, implying that the fundamental problem is ­ inflammatory. Maffulli and colleagues advocate the use of the term ­tendinopathy as a generic descriptor of clinical conditions such as pain, swelling, and impaired performance in and around ­ tendons ­ arising from overuse, with the labels tendinosis and tendinitis most appropriately applied after histopathologic examination.6 This nomenclature separates chronic ­ degeneration of tendons from acute and mainly inflammatory processes, with implications for treatment and ­management (Table 14-2). Tendinosis has been described as a failure of cell matrix adaptation to trauma because of an imbalance between matrix degeneration and synthesis. The classic pathology is a loss of the normal collagenous architecture and replacement with an amorphous mucinous material that lacks the parallel, longitudinal architecture of normal tendon.7 Histologic examination reveals intratendinous collagen degeneration with fiber disorientation and thinning, hypercellularity, scattered vascular ingrowth, increases in the amount of ground substance and the proteoglycan concentration of ground substance, and a decrease in the ratio of type I to type III collagen.8 Any inflammatory response or the presence of inflammatory cells is notably lacking in tissue samples of tendinopathy, differentiating tendinosis or chronic overuse pathology from acute injury and tendinitis. Astrom and Rausing described the major lesion in chronic Achilles tendinopathy as “a degenerative process characterized by a curious absence of inflammatory cells and a poor healing response.”9 Similar histopathologic 611

612

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 14-1  Bonar’s Modification of Clancy’s Classification of Tendinopathies Pathologic Diagnosis

Concept (Macroscopic Pathology)

Histologic Appearance

Tendinosis

Intratendinous degeneration (commonly caused by aging, microtrauma, and vascular ­compromise)

Tendinitis, partial rupture

Symptomatic degeneration of the tendon with ­vascular disruption and inflammatory repair ­response Inflammation of the outer layer of the tendon (paratenon) alone, regardless of whether the paratenon is lined by synovium Paratenonitis associated with intratendinous ­degeneration

Collagen disorientation, disorganization, and fiber separation with an increase in mucoid ground substance, increased prominence of cells and vascular spaces with or without neovascularization, and focal necrosis or calcification Degenerative changes as noted above with superimposed evidence of tear, including fibroblastic and myofibroblastic proliferation, hemorrhage, and organizing granulation tissue Mucoid degeneration in the areolar tissue is seen. A scattered, mild, mononuclear infiltrate with or without focal fibrin deposition and fibrinous exudate is also seen. Degenerative changes as noted for tendinosis with mucoid ­degeneration with or without fibrosis and scattered inflammatory cells in the paratenon alveolar tissue

Paratenonitis Paratenonitis with tendinosis

Adapted from Clancy WGJ: Tendon trauma and overuse injuries. In Leadbetter WB, Buckwalter JA, Gordon SL (eds): Sports-Induced Inflammation: Clinical and Basic ­Science Concepts. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1990, pp 609-618.

findings in posterior tibial tendon dysfunction of a degenerative tendinosis with mucinous degeneration, fibroblast hypercellularity and neovascularization,10 and higher proportions of collagen type III at the expense of collagen type I11 support the notion of a common disease process in overuse tendon injury, leading ultimately to tendon degeneration and an insufficient repair response. The term tennis elbow was introduced in the 1880s, but it was not until 1979 that pathology of the extensor carpi radialis brevis tendon was associated with lateral tennis elbow. At surgery, Nirschl found the extensor carpi radialis brevis tendon to contain disrupted collagen fibers, increased cellularity, and neovascularization in more than 600 patients with lateral tennis elbow12 (Fig. 14-1). Acute inflammatory cells were almost always absent from the tendon, but chronic inflammatory cells were occasionally present in small numbers in supportive or adjacent tissues. When chronic inflammatory cells were present, their presence resulted from the repair of partial tears. Although Nirschl coined the term angiofibroblastic hyperplasia for the histology seen in elbow tendinosis—presumably to ­ emphasize the neovascularization (angio) and increased cellularity

   TABLE 14-2  Comparison of Overuse Tendinosis and Overuse Tendinitis Comparison Factors

Tendinosis

Tendinitis

Prevalence Time for full recovery (initial) Time for full recovery (chronic) Likelihood of full recovery Focus of conservative therapy

Common 2-3 mo

Uncommon 2-3 days

3-6 mo

4-6 wk

About 80%

99%

Encourage collagen synthesis, maturation, and strength Excise abnormal tissue 70% to 85% 4-6 mo

Anti-inflammatory therapy and drugs

Role of surgery Prognosis of surgery Surgical recovery

Not known 95% 3-4 wk

Adapted from Khan K, Cook J: The painful nonruptured tendon: Clinical ­aspects. Clin Sports Med 22:715, 2003

(fibroblastic)—these features are both typical of the wellrecognized pathologic entity of tendinosis. Histopathology of symptomatic rotator cuff tendons reveals mucoid degeneration and fibrocartilaginous metaplasia13 as well as cellular distortion and necrosis, calcium deposition, fibrinoid thickening, hyalinization, fibrillation, and microtears. There is loss of the characteristic crimped pattern of tendon, and parallel bundles of collagen separate and become disorganized. As early as the 1930s, Codman described rotator cuff “rim rent,” which is localized disruption of the innermost fibers of the supraspinatus tendon attaching most closely to the articular surface of the humeral head.14 In the 1940s, Wilson and Duff reported degenerative processes to be the basis of rotator cuff tendinopathy.15 Later that decade, McLaughlin observed that insertional tendinopathy of the external rotators of the shoulder consisted of calcification, hyaline degeneration (characterized by the hypocellularity, vacuolation, nuclear pyknosis, more homogeneous matrix, and decreased eosinophilia—all of which probably represent mucoid degeneration), and microtears without inflammation in the tendon tissue.16 Fibrocartilaginous metaplasia was occasionally present. Functionally, whereas the healing response to an acute tendon injury involves an organized triphasic response of inflammation, proliferation, and maturation, the response to an overuse injury involves an inadequate, incomplete, and disorganized repair mechanism resulting in a substantially defective “repaired”’ tendon lacking in extracellular tissue organization, with decreased resistive strength and more vulnerability to further injury. Although the exact role of overuse in the pathogenesis of chronic tendon injuries and disorders is not completely understood, it is speculated that fatigued tendon loses its basal reparative ability with intensive repetitive activity, often eccentric in nature, leading to cumulative microtrauma that further weakens the collagen cross-linking and noncollagenous matrix and disturbs the microvasculature and macrovasculature of the tendon.17 Ensuing local tissue hypoxia and impaired nutrition and energy metabolism likely play an important role in the sequence of events leading to tendon degeneration. Leadbetter has called this the “tendinosis cycle.”1 One of the first animal models of tendinopathy, ­developed

Overuse Injuries

A

B

C

D

E

F

613

Figure 14-1   A, Photomicrograph of a specimen of normal tendon, showing parallel bundles of uniform-appearing collagen oriented along the long axis of the tendon. The matrix, which is composed primarily of proteoglycans, glycosaminoglycans, and water, is stained evenly. No vascular structures are apparent within the tendon (hematoxylin and eosin stain ×10). B, Photomicrograph demonstrating tendinosis of the extensor carpi radialis brevis tendon. The entire specimen appears to be hypercellular, with focal areas that are densely cellular. Some of the hypercellular regions are parallel to the tendon fibers (solid arrow), whereas others are not (open arrow). There is no evidence of an inflammatory response, as indicated by the absence of polymorphonuclear leukocytes, lymphocytes, and macrophages (hematoxylin and eosin stain ×20). C, Photomicrograph showing an enlarged view of the area around the open arrow in B.The area in which angiofibroblastic hyperplasia meets normal tendon contains active fibroblasts that are randomly oriented and appear to be infiltrating the surrounding tissue. Disorganized collagen abuts normal-appearing collagen. The matrix within the pathologic areas is loose and pale in appearance (hematoxylin and eosin ×100). Continued

614

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 14-1—cont’d  D, Photomicrograph showing that the maturing vessel walls in areas of tendinosis stain darkly for elastin. The areas of fat (open arrow) within tendon fascicles are an abnormal finding. The infiltrative appearance (double solid arrow) of tendinosis and the distinct boundary of normal tendon suggest a reparative and regenerative process (×40). E, Photomicrograph of a specimen from a patient who had severe tendinosis. The pale blue regions indicate abnormal collagen and matrix production. Densely cellular regions (indicated by the red stain) reveal angiofibroblastic hyperplasia permeating the tendon in linear clefts and clusters.  F, Photomicrograph demonstrating the permeative nature of tendinosis. This finding is indicated by the presence of widespread cellularity and randomly oriented vascular hyperplasia (vimentin stain ×40). (From Kraushaar BS, Nirschl RP: Current concepts review— tendinosis of the elbow [tennis elbow]. Clinical features and findings of histological, immunohistochemical, and electron microscopy studies. J Bone Joint Surg Am 81:259-278, 1999.)

by ­Soslowsky and colleagues, has shown persistent microscopic changes of tendinosis in rat rotator cuff supraspinatus tendon after exposure to multiple factors, including impingement and overuse.18,19 Neither impingement nor overuse alone produced the same degree of changes, implying a multifactorial etiology for the pathologic effects of overuse on rotator cuff tendon. Current thinking supports the belief that a spontaneous tendon rupture is a typical end-state manifestation of degenerative processes in the tendon tissue, with partial macroscopic tears as a stage in the continuum of tendon degeneration. Analysis of surgical specimens of Achilles tendons reveals that, although ruptured and tendinopathic tendons are histologically significantly more degenerated than control tendons, the general pattern of degeneration seen is common to the ruptured and tendinopathic tendons, suggesting the possibility of a common, as yet unidentified, pathologic mechanism acting on both tendon populations.20 Laboratory and molecular analyses of tendinopathy have begun to reveal strategies that may guide future clinical management of overuse tendon injury. It has been hypothesized in the past that tendon degeneration may be preceded by acute and then chronic phases of inflammatory “tendonitis.”21,22 Although no inflammatory infiltration has been observed in multiple studies of biopsy specimens of tendinopathic tendons, recent in vitro work demonstrates that a “molecular inflammation cascade” mediated by interleukin-1β in human tendon cells can induce connective tissue cell expression of cytokines that further induces known matrix destructive enzymes such as matrix metalloproteinases (MMP-1 and MMP-3).23 Clinically, the activity of metalloproteinases in tendon destruction and degeneration is the target of the use of injectable aprotinin, a metalloproteinase inhibitor, in the setting of patellar and Achilles tendinopathy as an alternative to corticosteroid therapy.24,25 Apoptosis, mediated by overuse-induced, stress-activated protein kinases, may also play a role in tendon degeneration and weakening, presenting another set of molecular targets for future therapies aimed at preventing or treating tendinopathy more effectively.26,27

Shoulder Overuse Injuries Rotator Cuff Disorders The most common cause of shoulder pain in older adults is rotator cuff pathology, particularly with overhead athletes. Rotator cuff dysfunction leads to a muscular imbalance, allowing the unopposed deltoid to cause upward migration

of the humeral head toward the acromion and coracoacromial arch. In general, younger individuals (<40 years) usually have instability as the root cause of impingement pain, whereas older individuals have a structural cause. Both groups classically complain of anterolateral shoulder pain during abduction (overhead activity) and pain on lying on an adducted shoulder. The most common location for rotator cuff tendinopathy occurs on the articular surface of the supraspinatus, about 2 cm from the distal insertion. This region is hypovascular, predisposing this location to breakdown. Adduction of the arm, such as when sleeping on one’s side, causes a “wringing out” effect, essentially decreasing the rotator cuff’s vascular supply.28 Impingement syndromes have been categorized by their etiology into external (subacromial) primary and secondary syndromes, and a new category, internal impingement, has recently been advocated to incorporate injury to the posterior superior glenoid. Primary external impingement syndrome, which is typically seen in older people (>40 years) is the classic scenario described by Neer.29 He divided primary external impingement syndromes into three progressive stages: rotator cuff edema, tendinitis, and frank tears. He postulated that a hooked or curved acromion, an osteophyte extending off the acromioclavicular joint, or a thickened coracoacromial ligament could lead to rotator cuff impingement during abduction of the arm. Interestingly, recent studies have shown that the acromion’s shape is acquired rather than congenital.30 Secondary external impingement is seen in younger patients with instability. An angled scapula from weak scapular stabilizing muscles, posterior capsule tightness, or lax anterosuperior ligaments can lead to the putative impingement of the rotator cuff. Furthermore, associated posterior capsule tightness causes anterosuperior translation of the humeral head when the shoulder is flexed. Internal impingement is significantly less common and represents impingement of the supraspinatus and infraspinatus tendon against the superoposterior aspect of the glenoid with repeated abduction and external rotation and accompanying instability. This usually pre­ sents as posterior shoulder pain provoked during the cocking phase of overhead activity and as instability; it can also present as positive impingement tests during examination. Physical examination of the shoulder requires skill and practice. Impingement tests have been developed by Neer and Hawkins. The most sensitive of these appears to be the modified Hawkins test. A positive empty-can test is a nonspecific indicator of rotator cuff pathology. Various tests for shoulder instability have been recently advocated. Abduction and external rotation tests of the shoulder have been the mainstay of identifying anterior shoulder

Overuse Injuries

i­nstability. The relocation maneuver is a useful adjunct to these tests. The active apprehension test developed by O’Brien appears to have high sensitivity and specificity for identifying tears of the superior lateral glenoid labrum.31 It is important to distinguish between rotator cuff ­tendinitis and tear because the treatments may differ markedly. Rotator cuff tendinitis and tear are two ends of a continuum. Progression of cuff fiber failure leads to upward displacement of the humeral head. Initially, rotator cuff edema and tendinitis are found. Over time, an acromial traction spur in the coracoacromial ligament develops. Eventually, the rotator cuff tears, typically at the supraspinatus attachment to the greater tuberosity 2 cm proximal to the distal insertion. Partial-thickness tears, on the articular or the bursal surface, quickly become full­thickness tears. Finally, chronic full-thickness tears lead to so-called cuff tear arthropathy. Many believe that a distinction between tendinitis and partial-thickness tears cannot be made because they likely represent the same pathology. However, tendinitis and partial-thickness tears can be distinguished from full-thickness tears. Tendinitis and partial tears may actually be more painful than full-thickness tears. Because tears are usually a cumulative injury, onset of presentation does not differentiate between ­ tendinitis and tear. The physical examination is notorious for its inability to screen for rotator cuff tears, yet it is more useful than the history. Marked weakness seen with isometric testing of the supraspinatus (empty-can test), subscapularis (Gerber push-off test), and infraspinatus (resisted external rotation with arm at the side or appositive external rotation lag sign) can lead to a diagnosis of a full-thickness tear. Likewise, a positive drop arm test, in which the patient is unable to control the downward descent of an abducted shoulder, typically indicates a diagnosis of a full-thickness rotator cuff tear. Individuals with supraspinatus and infraspinatus gross muscular atrophy should be suspected to have massive rotator cuff tear, a cervical radiculopathy, or a suprascapular nerve injury. Injections of local anesthetics can also be used to clarify diagnoses. For example, anesthetic phase response to an injection into the subacromial space highly suggests pain emanating from the rotator cuff or the subacromial bursa. Because the history and physical examination findings are often unclear, imaging becomes the mainstay of diagnosis. Plain radiographs, although limited in the imaging of shoulder anatomy, can provide indirect evidence of full-thickness tears and possible predisposing factors. Rotator cuff tears may manifest as avulsed fragments or cystic changes of the greater tuberosity, subacromial sclerosis, acromial traction spurs, and superior migration of the humeral head. Cuff tear arthropathy radiographic changes have already been described. Some clinics are able to perform diagnostic ultrasound as an immediate adjunct to the physical examination. To date, ultrasound for diagnoses of rotator cuff pathology has been limited to tertiary academic centers and specialty clinics. If clinical suspicion is high, magnetic resonance imaging (MRI) is the criterion standard for identifying rotator cuff tears. Magnetic resonance arthrography is even more sensitive for undersurface rotator cuff tears. Treatment protocols for rotator cuff disease start with proper diagnosis and identification of ­biomechanical

615

Box 14-1  Shoulder Rehabilitation Guidelines and Protocol Initial Phase (Pain Control) Reduce pain and inflammation; promote tissue healing Relative rest Modalities Medications (nonsteroidal anti-inflammatory drugs, ­steroid injection) Surgery Reestablish passive range of motion Pendulum exercises Manual treatments Posterior capsule stretching and mobilization Retard muscle atrophy and promote scapular control Isometric to rotator cuff and scapular stabilizers (scapular pinch) Closed kinetic chain exercises Maintain fitness of the rest of the kinetic chain Reactivation Phase (Correct Imbalance   in Flexibility and Strength) Reestablish active range of motion Active-assisted using wand Promote scapular control and kinetic chain of upper extremity Proprioceptive neuromuscular facilitation Advance closed kinetic chain exercises Modified push-ups Promote force generation Plyometrics Open kinetic chain Integrated exercise with lower limbs Maintenance Phase (Functional Adaptations) Add additional planes of movement (e.g., diagonals) Add high-level plyometrics Promote sports-specific activity From Kibler WB, McMullen J, Uhl T: Shoulder rehabilitation strategies, guidelines, and practice. Orthop Clin North Am 32:527-538, 2000.

­ eficits. Patients with partial-thickness tears should be d treated as if they have chronic tendinitis before they are introduced to aggressive surgical intervention. Bursal side tears appear to do well with a nonsurgical course because they have excellent vascularity and potential healing.32 In contrast, most full-thickness tears of the rotator cuff should be referred for surgical evaluation. Nonsurgical approaches are attempted for most of the rotator cuff spectrum, beginning with activity modification, oral analgesics and anti-inflammatories, and a trial of physical therapy33 (Box 14-1). Steroid injections should be limited because there are reports of intratendinous steroid injections propagating rotator cuff tears.34 The posterolateral approach is ­ relatively straightforward. A 21- to 25-gauge needle containing 1 mL of long-acting steroid and 4 mL of local anesthetic can be inserted beneath the posterolateral acromion and aimed toward the coracoid process. Moreover, modalities, including injections, should be ­ performed in isolation.

616

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Multidirectional Instability The tremendous mobility in the shoulder often leads to instability. The shoulder capsule is a naturally capacious structure, and, in reality, two humeral heads can fit into one shoulder capsule. Unidirectional instability is typically anterior and often occurs after a traumatic dislocation. Further, common “bad posture” patterns (rounded, protracted shoulders with internally rotated arm, thoracic kyphosis, forward head attitude) predispose to anterior instability. On the other hand, multidirectional instability is often observed in people with generalized ligamentous laxity. Swimmers, in particular, appear to be predisposed to this condition. The Beighton scale35 can be used to identify people with hypermobile shoulders or generalized ligamentous laxity. This scale is a 10-point scale that allows 1 point each for the ability to have the thumb touch the forearm (tested bilaterally); 5th metacarpophalangeal joint extension greater than 90 degrees (tested bilaterally); elbow extension greater than 10 degrees (tested bilaterally); and lumbar flexion with knee fully extended such that palms reach the floor. A grade of 4 to 9 is defined to be lax. Multidirectional instability arises in the absence of trauma. It has a multifactorial origin including labral insufficiency, weak cuff compressors, scapulohumeral dyskinesis, developmental dysplasia, and excessive capsular redundancy or compliance. The term AMBRI (atraumatic, multidirectional, bilateral, rehabilitation, inferior capsular shift) syndrome was coined to describe this pathology. Patients with AMBRI lesions may have several sequelae: repeated dislocations, cysts, and the dead arm syndrome. The glenoid labrum is vulnerable to injury with either type of shoulder instability. The labrum is a cartilaginous ring covering of the glenoid. It provides an increased surface area for situation of the humeral head and acts as a stop block to prevent subluxation. Injuries to the labrum include so-called SLAP lesions (superior labrum, anteroposterior [AP] tears) and Bankart lesions. A Bankart lesion is a tear of the anteroinferior labrum; a “bony” Bankart lesion also involves injury to the glenoid. Examination findings include excessive inferior translation, revealed by the sulcus test, excessive AP translation, evidenced by the drawer test, and generalized ligamentous laxity. Multidirectional instability is primarily treated with aggressive rehabilitation similar to rotator cuff rehabilitation. McMahon and colleagues showed that there is a low level of rotator cuff activity when the humeral head subluxes or dislocates.36 Therefore, rotator cuff strengthening is recommended for multidirectional instability. Some surgeons have attempted thermal shrinkage of the shoulder capsule to tighten the capsule. However, the effects of tightening appear to be short-lived.

Scapular Winging Shoulder pain in the young overhead athlete is secondary to many underlying issues, including shoulder instability, structural abnormalities, muscle imbalance, and nerve injury. As previously discussed, instability is most common in the anterior direction. On occasion, multidirectional shoulder instability is noted in people with generalized ligamentous laxity. Unilateral or multidirectional ­ instability

may be accompanied by inhibition weakness (weakness due to tight antagonist muscles) of the scapular stabilizer musculature. Inhibition weakness of the serratus, trapezius, rhomboids, levator, and latissimus muscles results in scapular dyskinesis and winging. Although most cases of scapular winging are caused by compensation for pain, scapular winging in athletes should be thoroughly investigated for true neurogenic or structural etiologies. Muscles that attach to the scapula act in various force couples. If any of these force couples are disrupted by true nerve injury, structural abnormalities, secondary problems, or various types of scapular winging can result. Scapular winging can be divided into four different presentations that have subtle differences. Patients should be observed for the general placement of the scapula, with either medial or lateral translocation identified. Further, the medial border and the inferior angle of the scapula should be noted for prominence and rotation. Shoulder abduction and forward flexion can either accentuate or improve the scapular winging. Evidence of shoulder drooping can also help determine the cause of scapular winging. In general, patients with scapular winging caused by serratus anterior weakness from long thoracic nerve injury are described as having “the scapula off the thorax.” Long thoracic nerve palsy is often seen in patients with neuralgic amyotrophy (Parsonage-Turner syndrome). With this syndrome, exquisite pain precedes weakness and scapular winging. On the other hand, scapular winging occurring because of trapezius weakness from spinal accessory nerve injury is described as a “winging with rotation.” Various bony conditions (e.g., scoliosis, scapular fracture nonunions) and neuromuscular disorders (e.g., muscular dystrophies) can also cause scapular winging. Most athletes, such as volleyball players, present with scapular winging because of weak scapular stabilizers (mixed weakness). The lateral scapular slide test is an excellent way to identify weak periscapular musculature.

Nerve Entrapment Volleyball players with gross muscle atrophy of the posterior shoulder musculature, namely of the supraspinatus and infraspinatus, should be suspected to have a suprascapular nerve injury. The suprascapular nerve comes off the brachial plexus at the level of the upper cervical trunk (C5 derivation) and then traverses to the posterior shoulder. Injury may occur as the suprascapular nerve passes underneath the superior transverse scapular ligament at the scapular notch. If the nerve is injured at this level, both supraspinatus and infraspinatus muscles may be denervated. Denervation atrophy of the supraspinatus is more difficult to observe than that of the infraspinatus because of the overlying trapezius. Nerve entrapment at the scapular notch is rare and should be investigated for a space-occupying lesion or traction etiologies. The more common entrapment site—under the inferior transverse scapular ligament at the spinoglenoid notch—results in isolated infraspinatus denervation. ­Volleyball players famously develop ­anterior shoulder instability and resultant labral lesions from serves.37 Twenty percent of elite volleyball players have been noted to have infraspinatus atrophy because of traction on the suprascapular nerve at the spinoglenoid notch.38

Overuse Injuries

The repeated use of the “float serve,” in which rapid shoulder acceleration is followed by eccentric deceleration at ball impact, is a predisposing factor. Glenohumeral ganglion cysts, usually from labral pathology, may entrap the nerve at this level, because of its close proximity to the joint. A sling effect is also created across the spine of the scapula, making the suprascapular nerve susceptible to compression and traction injuries. Partial denervation of the infraspinatus is observed if one of the terminal branches is spared or in cases of neurapraxia rather than axonal injury. If serratus anterior type of scapular winging is present, neuralgic amyotrophy should be suspected. Glenohumeral synovial cysts can be seen on MRI and more sensitively with MRA. Electrodiagnostic tests can find the degree and location of suprascapular nerve injury. Needle electromyography remains the cornerstone of diagnosis, with spontaneous abnormal activity observed in the infraspinatus and normal examination in the cervical paraspinals and other C5 and upper trunk–­innervated muscles. The infraspinatus should be sampled at ­ various sites to pick up nerve branch injury. Spontaneous abnormal activity in the supraspinatus, sampled with a ­tangential approach to the level of the periosteum, can determine whether suprascapular nerve injury is at the scapular notch or the spinoglenoid notch. Nerve conduction study (NCS) of the suprascapular nerve is fraught with the technical difficulties observed in other proximal NCSs. However, side-to-side NCS comparisons can be useful. Dual-channel needle recording of the supraspinatus and infraspinatus muscles while stimulating at Erb’s point is the favored technique.39 Side-to-side latency differences greater than 0.4 msec indicate pathology.

Lateral Elbow Overuse Injuries Lateral Epicondylitis Tennis elbow represents an array of diagnoses, with the most common being lateral epicondylitis. More properly, lateral epicondylitis should be termed wrist extensor ­tendinosis. The musculotendinous structures about the lateral epicondyle of the elbow are those of the common extensor origin, including the extensor carpi radialis longus, the extensor carpi radialis brevis, the extensor digitorum communis, and the extensor carpi ulnaris. The extensor brevis, which is most commonly involved in lateral epicondylitis, lies beneath the extensor longus. The complex origin of the extensor brevis includes the common extensor tendon at the lateral epicondyle, the lateral collateral and annular ligaments, the investing fascia, and the intermuscular ­septum. The normal biomechanics of the lateral epicondylar structures during sport have been most thoroughly described for tennis. Morris and associates evaluated the muscle activity about the elbow during tennis strokes in healthy professional and collegiate players using an electromyographic (EMG) technique.40 The greatest muscle activity during the groundstrokes was noted in those muscles stabilizing the wrist, specifically, the extensor carpi radialis brevis, the extensor carpi radialis longus, and the extensor digitorum communis. The extensor carpi radialis

617

brevis was noted to have the greatest activity of all muscles tested; this occurred during the acceleration and early follow-through phases. The authors suggested that these muscles provide optimal stability for these phases of the groundstroke by maintaining the position of the wrist in extension and radial deviation. Lateral epicondylitis typically occurs in the fourth and fifth decades, although it has been identified in patients ranging in age from 12 to 80 years. The male and female prevalence rates appear equal. Three fourths of patients experience symptoms in their dominant arm. Morris’ initial implication of racket sports as the primary cause of epicondylitis led to various studies reviewing the possible epidemiologic and etiologic factors in tennis. It is estimated that 10% to 50% of persons who play tennis regularly will experience symptoms of tennis elbow at some point during their careers. In 1979, Gruchow and ­Pelletier noted an association between playing time and the incidence of tennis elbow in club players.41 The risk for developing symptoms consistent with tennis elbow was 2.0 to 3.5 times greater in players with more than 2 hours of racket time per week than in those who played tennis less than 2 hours per week. Compared with younger players, male and female players older than 40 years had a fourfold and twofold greater incidence of tennis elbow, respectively. Several specific technique, equipment, and playing surface factors have been implicated in the development of lateral epicondylitis. A higher incidence of poor stroke mechanics, such as leading with a flexed elbow and striking the ball off center on the racket, has been identified in affected players. Improper grip size, racket weight, and racket stringing generate higher loads in the lateral muscle-tendon unit. Also, harder court surfaces impart greater momentum to the ball and subsequently increase the force transmitted through the racket to the extensor mass. Numerous other sports and occupational activities that require forceful or repetitive forearm use have also led to lateral epicondylitis. A wide spectrum of theories on the pathophysiology of lateral epicondylitis has been proposed. The current consensus based on clinical and surgical evidence suggests that lateral epicondylitis is initiated as a microtear, most often within the origin of the extensor carpi radialis brevis. This process may originate in the extensor digitorum communis or extensor carpi radialis longus tendon as well. The affected tendon usually contains grayish, homogeneous, edematous, friable tissue. Tendon fibers may appear fibrillated, with an apparent sinus tract extending from the elbow joint. In the series of 88 surgically treated elbows reported by Nirschl and Pettrone, 97% demonstrated varying amounts of this gross pathologic tissue at the origin of the extensor carpi radialis brevis tendon.42 Of those elbows with macroscopic pathologic tissue within the extensor carpi radialis brevis, 35% also demonstrated gross tendon rupture. Nirschl has described a characteristic microscopic appearance of “angiofibroblastic hyperplasia” of the involved tissue. The normal parallel orientation of collagen fibers is disrupted by an invasion of fibroblasts and vascular granulation-like tissue without an acute or chronic inflammatory ­component. Lateral epicondylitis is characterized by pain at the lateral epicondyle, which often radiates into the forearm and is typically insidious in its onset. A history of ­ repetitive

618

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

activity or overuse can often be elicited. A diagnosis of wrist extensor tendinopathy can be confirmed by the middle finger resistance test and Mill’s test (pain reproduced with wrist extension, radial deviation, and pronation). Maximal tenderness should be over the extensor carpi radialis brevis tendon 1 to 2 cm distal and anterior to the lateral epicondyle rather than the supinator or the posterior elbow. Cervical radiculopathy (C6 or C7) can occasionally mimic tennis elbow. Thus, an upper limb neurologic examination, cervical range of motion testing, and provocative tests, such as the Spurling maneuver, should be performed to discern radicular pain. Rarely, injuries to the radial nerve can cause refractory lateral elbow pain. Location of maximal tenderness is at the leading edge of the supinator rather than at the lateral epicondyle. The arcade of Frohse is an electrophysiologic entrapment site of the radial nerve at the elbow. If crepitus, clicking, joint effusion, or elbow flexion contractures are observed, elbow joint or ligament pathology (e.g., radiocapitellar chondromalacia and synovitis, osteochondritis dissecans (OCD), lateral collateral ligament incompetency) should be suspected. Unlike throwing athletes, ligamentous pathology is rarely seen in tennis players. Radiographs of the affected elbow are usually normal, but 22% to 25% of patients may have calcification within the soft tissues about the lateral epicondyle. The calcification appears to have no prognostic implications and may disappear after treatment. Nonoperative treatments for lateral epicondylitis have been very successful. The common objectives of all conservative measures are relief of the pain and reduction of inflammation followed by guided rehabilitation. Relief of pain and inflammation is the primary goal of the first phase of nonsurgical treatment. Cessation of the offending activity is required initially, but complete inactivity or immobilization is avoided because this may lead to disuse atrophy, which compromises later rehabilitation. Ice is recommended for its local vasoconstrictive and analgesic effects. An oral anti-inflammatory medication should be administered for a 10- to 14-day period if the patient has no medical contraindication to use of such a drug. Those patients who demonstrate some improvement without complete relief may require a second course of medication after a brief period of abstinence. If the patient does not respond to these initial therapeutic measures, a corticosteroid injection should be considered. The choice and dose of steroid preparation has remained arbitrary, however, because carefully controlled prospective comparisons of commonly used agents have not been done. Care should be taken to instill the mixture deep to the extensor carpi radialis brevis, anterior and distal to the lateral epicondyle, into the fatty subaponeurotic recess. Injection of the mixture superficially may result in subcutaneous atrophy, whereas intratendinous injection may lead to adverse permanent changes within the tendon ultrastructure. The physical therapy modalities of ultrasound and highvoltage galvanic stimulation have been used with variable success. However, there are no prospective, randomized, controlled studies to demonstrate their efficacy. Upon relief of initial pain and inflammation, the second phase of nonsurgical treatment is begun. This phase emphasizes continued tissue healing through avoidance of the abusive

aspects of the causative activity and guided rehabilitation. If the patient uses aberrant techniques in sports or occupational activities, these should be identified and corrected. For example, in tennis, the forehand stroke should allow the player to hit the ball in front of the body with the wrist and elbow extended. This allows the torso and upper arm to provide most of the stroke power, rather than the wrist extensors solely. The two-handed backhand stroke allows a distribution of force between the upper extremities, and thus greatly diminishes force at the leading lateral ­epicondyle. Proper equipment, especially in the racket sports, is essential to preventing lateral epicondylitis. Proper racket grip size is assessed by measuring from the proximal palmar crease to the tip of the ring finger, along its radial border. Lighter rackets, though providing less momentum, allow ease of positioning for impact. Frames of low-vibration materials, such as graphite and epoxies, dampen impact forces imparted to the extensor origin. Using rackets that are less tightly strung or that have a higher string count per unit area and playing on “slower” surfaces, such as clay courts, will diminish the loads transmitted to the elbow. Counterforce bracing was introduced by Ilfeld in 1965 (Fig. 14-2). Theoretically, this type of brace inhibits full muscular expansion and thus decreases the force experienced by sensitive or injured muscular tissue proximal to the band. Groppel and Nirschl demonstrated with threedimensional cinematography and surface electromyography that lower extensor muscle activity was produced by the use of counterforce bracing during the tennis serve and one-handed backhand.43 Snyder-Mackler and Epler, employing the more sensitive indwelling EMG technique, noted significantly reduced muscle activity in the extensor carpi radialis brevis and extensor digitorum communis of healthy subjects during maximal voluntary isometric

Figure 14-2   Lateral elbow counterforce brace.   (From Morrey BF: The elbow and its disorders. Philadelphia, WB Saunders, 1985.)

Overuse Injuries

c­ ontraction while using an air-bladder type of counterforce brace.44 Counterforce bracing may be used during the early rehabilitation period; if pain recurs, the first-phase treatments may be reinitiated. The rehabilitative program begins with wrist extensor stretching and progressive isometric exercises. Initially, these exercises may be done with the elbow flexed to minimize the pain; then, as the symptoms allow, the exercises are done with the elbow in full extension. As strength, endurance, and flexibility improve, eccentric and concentric resistive exercises are performed. When the patient is capable of sprint repetitions to fatigue without significant elbow symptoms, a sport or job simulation is staged. If it is successfully completed, the patient is encouraged to return to normal activity and to gradually increase the duration and intensity of exposure. Although most authors report that most patients with lateral epicondylitis respond to nonoperative care, there are few studies on the long-term outcome of nonsurgical treatment. The available literature suggests that 5% to 15% of patients will suffer a recurrence of symptoms, but most of these patients with relapses will not have been fully rehabilitated or will have prematurely discontinued preventive measures. In a prospective review of nonoperative treatment, Binder and Hazleman noted that 26% of patients had a recurrence of symptoms and that more than 40% had prolonged minor discomfort.45 However, most clinical reports agree that nonoperative management remains the mainstay of treatment for lateral epicondylitis and that surgical treatment is infrequently necessary. The indications for surgical treatment of lateral epicondylitis include persistent debilitating pain at the lateral epicondyle unresponsive to a well-managed nonoperative program spanning a minimum of 6 to 12 months, after the exclusion of other pathologic causes for the pain. The history of surgical treatment for lateral epicondylitis spans nearly three quarters of a century and includes a host of techniques of varying popularity. In general, four main approaches have been employed: (1) extra-articular procedures that involve the common extensor origin; (2) intraarticular procedures that excise the synovial fringe and a portion of the orbicular ligament; (3) extra-articular procedures that lengthen the extensor carpi radialis brevis tendon distally; and (4) extra-articular procedures that excise the pathologic tendon and then reattach the origin. Currently, most advocate an extra-articular technique wherein the pathologic portion of the extensor tendon origin is excised, the defect is repaired, and the origin is reattached to the epicondyle.

Medial Elbow Overuse Injuries The elbow is subjected to tremendous valgus stresses during overhead activities, which result in specific injury patterns unique to the throwing athlete. The forces generated as the result of repetitive throwing are primarily concentrated on the medial structures of the elbow. Consequently, medial elbow problems predominate in the athlete engaged in overhead activities. Although acute traumatic injuries to the osseous, musculotendinous, and ligamentous structures of the elbow may occur, most are chronic overuse injuries resulting from repetitive intrinsic and extrinsic overload.

619

Baseball players are the athletes most commonly affected; medial elbow symptoms account for up to 97% of elbow complaints in pitchers. However, athletes who participate in other sports that require similar overhead motion, such as football, volleyball, tennis, and javelin throwing, can be likewise affected.

Valgus Instability Injury to the ulnar collateral ligament (UCL), initially recognized in javelin throwers, has been reported to occur with increasing frequency in other types of overhead athletes as well. Microtears of the UCL occur once the valgus forces generated during the cocking and acceleration phases of throwing exceed the intrinsic tensile strength of the UCL. Improper throwing mechanics, poor flexibility, and inadequate conditioning result in additional cumulative stress transmission to the UCL complex, leading to attenuation and eventual rupture of the UCL. The diagnosis of valgus instability is based on the athlete’s history, physical examination, and radiographic studies. Patients with acute UCL injury usually experience the sudden onset of pain after throwing, with or without an associated popping sensation, and are unable to continue throwing. Patients with chronic injury usually describe a gradual onset of localized medial elbow pain during the late-cocking or acceleration phase of throwing. Athletes may also describe pain after an episode of heavy throwing that results in the inability to subsequently throw at more than 50% to 75% of their normal level. Patients with chronic instability also commonly present with ulnar nerve symptoms. This is due to local inflammation of the ligamentous complex, which produces secondary irritation of the ulnar nerve within the cubital tunnel.46 Physical examination of the elbow for valgus instability is best performed with the patient seated and the wrist secured between the examiner’s forearm and trunk. The patient’s elbow is flexed between 20 and 30 degrees to unlock the olecranon from its fossa as a valgus stress is applied (Fig. 14-3A). This maneuver stresses the anterior band of the anterior bundle of the UCL. It is important to palpate the UCL along its course from the medial epicondyle toward the proximal ulna as valgus stress testing is performed. Valgus laxity is manifested by increased medial joint-space opening as compared with the contralateral extremity. Comparison with the uninvolved elbow should always be performed to differentiate between physiologic and pathologic laxity. Loss of a firm end point, coupled with increased medial joint-space opening with valgus stress, is consistent with an attenuated or incompetent UCL. Testing of the functionally more important posterior band of the anterior bundle can be accomplished by the milking maneuver, which is performed by pulling on the patient’s thumb with the patient’s forearm supinated, shoulder extended, and elbow flexed beyond 90 degrees This maneuver generates a valgus stress on (Fig. 14-3B).�������������������������������������������� the flexed elbow; a subjective feeling of apprehension and instability, in addition to localized medial-side elbow pain, is indicative of UCL injury. Point tenderness and swelling over the UCL vary with the amount of inflammation and edema present. The absence of increased pain with wrist flexion, combined with

620

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 14-3  Elbow examination for medial instability. A, The examination for valgus stability is done with the elbow flexed 25 to 30 degrees (to unlock the olecranon), testing the anterior band of the anterior bundle of the ulnar collateral ligament. The examiner firmly grasps the patient’s elbow and forearm applying varus-valgus stress while palpating the UCL. B, The milking maneuver tests the posterior band of the anterior bundle of the UCL. The maneuver is performed by applying downward and valgus stress with the forearm supinated and the elbow flexed more than 90 degrees.  (Adapted from Kvitne RS, Jobe FW: Ligamentous and posterior compartment injuries of the elbow. In Jobe FW [ed]: Operative Techniques in Upper Extremity Sports lnjuries. St. Louis, Mosby, 1996, p 415.)

pain localization slightly posterior to the common flexor origin, differentiates UCL injury from flexor-­pronator muscle injury.47 Decreased range of motion (loss of terminal extension) secondary to flexion contractures (which develop as a result of the repeated attempts at healing and stabilization) may also be present in cases of chronic valgus instability. Overall performance by the athlete, however, may not be significantly compromised because the throwing motion does not require full elbow extension and can be accomplished with a flexion arc between 20 and 120 degrees.48 Routine radiographs may show changes consistent with chronic instability, such as calcification and occasionally ossification of the ligament. Stress radiographs can be used to confirm instability, especially in apprehensive patients and in patients in whom the clinical findings are equivocal. Medial joint opening greater than 3 mm is consistent with instability. MRI is useful in evaluating ligamentous avulsions, partial ligamentous injuries, mid-substance tears, and the status of the surrounding soft tissues. Computed tomography (CT) arthrography has also been reported to be useful in the evaluation of the UCL complex. Specific treatment programs may be implemented after the diagnosis of valgus instability is made. Initially, a nonoperative treatment protocol is instituted to reduce inflammation and pain. A brief period of rest (2 to 4 weeks) is recommended, coupled with use of nonsteroidal antiinflammatory medications (NSAIDs) and local physical therapy modalities. Corticosteroid injections are not recommended because further ligamentous attenuation may occur. When the acute inflammation has subsided, a supervised flexibility and strengthening program is instituted, aimed at restoring muscle tone, strength, and endurance to provide dynamic elbow stability. The pronator teres, flexor carpi ulnaris (FCU), and flexor digitorum superficialis should be targeted because they are potentially important secondary dynamic stabilizers of the elbow. EMG analysis has shown maximal activity of the flexor-pronator mass during the acceleration phase of the pitching cycle in healthy athletes; however, in athletes with valgus instability, a paradoxical decrease in activity has been observed in these muscle

groups.49 This finding may be a reflection of the primary disorder predisposing the elbow to instability, or may be attributable to muscular inhibition through a painful feedback loop arising from injury to the UCL complex. This situation is similar to that observed in overhead athletes with anterior shoulder instability in which the subscapularis (a dynamic stabilizer of the shoulder) has been shown to have decreased activity. Strengthening and conditioning of the flexor-pronator mass may potentially enhance performance by increasing valgus stabilization and theoretically increasing functional protection of the UCL. A well-supervised throwing and conditioning program is begun at 3 months, after the athlete has regained full range of motion and strength. In addition, an evaluation of the athlete’s throwing motion is essential to identify and correct improper mechanics. Nonoperative management instituted at an early stage has been shown to arrest the progression of instability and functional impairment, with as many as 50% of athletes being able to return to their preinjury level of throwing. Surgical intervention is indicated for competitive athletes with acute complete ruptures of the UCL or chronic symptoms secondary to instability that have not significantly improved after at least 3 to 6 months of nonoperative management. Operative treatment consists of either repair or reconstruction of the UCL. The goals of surgery are to reestablish stability of the elbow and to allow the athlete to return to maximal functional levels.

Medial Epicondylitis Medial epicondylitis is much rarer than its lateral counterpart, the latter occurring from 7 to 20 times more frequently. It also occurs within the fourth and fifth decades, with apparently equal male and female prevalence rates. Although termed golfer’s elbow, medial epicondylitis occurs often in baseball pitchers and in those who participate in a variety of other sports and occupational activities that ­create valgus force at the elbow.50 The musculotendinous structures about the medial epicondyle include the flexor-pronator muscle mass origin.

Overuse Injuries

From proximal to distal, this includes the pronator teres, the flexor carpi radialis (FCR), the palmaris longus, the flexor digitorum superficialis, and the FCU. The pronator teres and FCR, which are most commonly involved in medial epicondylitis, both arise from the medial supracondylar ridge. The biomechanics of the medial elbow have been most thoroughly defined by the pitching mechanism. Peak angular velocity and valgus forces exceeding the tensile strength of the medial musculotendinous and ligamentous structures may be produced primarily during the acceleration phase, which extends from the point at which forward velocity of the ball is essentially zero to ball release. These valgus forces are transmitted initially to the flexor-pronator musculature at the medial epicondyle and subsequently to the deeper medial collateral ligament. The pronator teres and FCR have been identified as the most common sites of pathologic change. In an EMG evaluation of the tennis serve, Morris and associates corroborated the biomechanical theories of the baseball pitch.40 They noted that the highest muscle activity occurred during the acceleration phase and was seen in the pronator teres of the flexorpronator mass. They suggested that during this phase the pronator is providing optimal forearm positioning while transferring momentum and power to the ball. Medial epicondylitis is characterized by pain along the medial elbow that is worsened by resisted forearm pronation or wrist flexion. This medial pain is often insidious in onset. Tenderness is usually distal and lateral to the medial epicondyle, most often over the pronator teres and FCR. The range of motion of the elbow and that of the wrist are usually complete. Normal strength and sensation are typically noted in the extremity. If, however, concomitant ulnar neuropathy exists, varying degrees of diminished sensibility in the ring and little fingers, as well as a Tinel’s sign at the elbow, may be present. When evaluating the patient with suspected medial epicondylitis, it is essential to consider primary ligamentous instability or primary ulnar neuropathy in the differential diagnosis. Valgus stress testing with the wrist flexed and the forearm pronated will produce pain and laxity if collateral instability is present. Maximal elbow flexion and wrist extension for 3 minutes (elbow flexion test) will produce pain and numbness if ulnar neuropathy is present. Plain radiographs of the elbow are most often normal. Throwing athletes, however, may have medial ulnar traction spurs and medial collateral ligament calcification. Medial epicondylitis, like lateral epicondylitis, is most often successfully treated with a nonsurgical regimen. The basic principles of nonsurgical treatment for lateral epicondylitis apply to medial epicondylitis as well. Phase 1 consists of rest from the offending activity for initial relief of pain and inflammation. The use of NSAIDs, galvanic stimulation, and possibly corticosteroid injection may provide adjuvant benefit. The second phase includes technique enhancement, equipment modification, possibly counterforce bracing, and a rehabilitation program. The rehabilitation program begins with wrist flexor and forearm pronator stretching and progressive isometric exercises. As flexibility, strength, and endurance improve, eccentric and concentric resistive exercises are included. A sport or job simulation is then performed, followed

621

by a gradual return to normal activity. The indications for ­ surgical treatment of medial epicondylitis include ­persistent pain at the medial elbow unresponsive to a wellmanaged nonoperative program for a minimum of 6 to 12 months, after exclusion of any other pathologic causes for the pain.

Valgus Extension Overload Medial tension overload secondary to repetitive valgus stress can also result in injury to the surrounding structures of the elbow. Microtrauma and inflammation of the UCL occur, with eventual attenuation and insufficiency of the ligamentous complex. The elbow becomes subluxated in valgus during extension, leading to excessive force transmission to the lateral aspect of the elbow, as well as extension overload within the posterior compartment. Compressive and rotatory forces are increased within the radio-­capitellar articulation, leading to synovitis and the development of osteochondral lesions (OCD and osteochondral fractures) that can fragment and become loose bodies.51,52 Athletes may report symptoms of catching or locking when loose bodies develop. Medial tension overload resulting in valgus instability also leads to extension overload of the posterior compartment. The extension forces generated during the acceleration and follow-through phases of the throwing motion, which are normally absorbed by the ligamentous, capsular, and muscular structures of the elbow, are excessively transmitted to the posterior compartment. Repeated impaction of the posteromedial ­olecranon in the olecranon fossa leads to chondromalacia and subsequent hypertrophic spur and osteophyte formation, especially in the medial aspect of the ulnar notch. Posteromedial impingement secondary to encroachment on the olecranon fossa by osteophytes and scar tissue results in pain during the late acceleration and follow-through phases of throwing. These hypertrophic osteophytes and traction spurs can frequently be observed on plain radiographs, especially on the axial olecranon view. Loose ­bodies and osteochondral lesions may occasionally be seen as well. Nonoperative treatment consists of an initial period of rest, ice, and NSAIDs to alleviate pain and inflammation, followed by functional strengthening of the elbow and forearm. Stretching, isotonic, isokinetic, and isometric strengthening and conditioning exercises of the forearm are implemented. As strength improves, the athlete may begin plyometric exercises concentrating on the flexor-pronator musculature, as well as an interval throwing program. Surgical intervention is recommended for patients who have failed nonoperative therapy or who have symptomatic traction spurs or loose bodies. There is a wide spectrum of underlying medial elbow instability; athletes who have failed conservative therapy and have persistent symptoms attributable to chronic valgus instability may also be candidates for operative management. Elbow arthroscopy has replaced formal arthrotomy as the surgical procedure of choice for joint débridement and has been shown to have good results with low complication rates in symptomatic patients.53 Chondromalacia of the ulnohumeral or radio-capitellar joint may be treated with débridement or drilling. Loose bodies and osteochondritic lesions can also be addressed. Débridement of hypertrophic

622

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

synovium or scar tissue can be performed as well. Osteophytes and hypertrophic spurs in the posterior and medial aspects of the olecranon can be débrided to decompress the olecranon fossa. Reconstruction of the UCL is reserved for athletes with recalcitrant symptoms associated with chronic valgus instability for whom nonoperative management and less invasive procedures have failed. These athletes usually have medial elbow instability that potentiates symptoms of posteromedial impingement if left unaddressed. Postoperative rehabilitation is begun early to maintain range of motion as well as to strengthen the elbow gradually. Athletes usually progress through a graduated throwing program that allows them to return to full activity within 3 months.

Flexor-Pronator Injuries and Ruptures The flexor-pronator musculature provides dynamic stability to the medial elbow and may be injured with repetitive valgus stress. Continued activity and throwing beyond the limits of muscle fatigue may lead to injury and, occasionally, rupture of the flexor-pronator musculature.54 These injuries usually occur during the acceleration and followthrough stages of the throwing motion, when forceful extension of the elbow and pronation of the forearm occur. Patients generally present with pain and swelling along the medial aspect of the elbow. On examination, there is usually tenderness at the medial epicondylar origin with pain that may be exacerbated by wrist flexion and elbow extension. It is important to evaluate for concurrent UCL injury. Minor partial injuries of the flexor-pronator musculature may be treated with rest, ice, and NSAIDs. More severe injuries and complete ruptures that compromise elbow stability require surgical repair. Hypertrophy of the pronator teres secondary to repetitive activity may result in compression of the median nerve and the development of pronator syndrome. Patients usually present with fatigue-like pain in the proximal volar aspect of the forearm that gradually worsens with continued activity. Symptoms are usually exacerbated by ­resistance to pronation of the forearm combined with wrist flexion. Surgical exploration, including elevation and division of the superficial head of the pronator teres, may be necessary to decompress the median nerve and provide symptomatic relief. Although uncommon, compartment syndrome as a result of hypertrophy of the flexor-pronator musculature has also been reported. Patients describe pain localized to the medial elbow and proximal forearm that is worsened by increased throwing and activity, typically forcing pitchers to stop throwing after only a few innings. This condition can usually be prevented by adequate warm-up and proper timing of pitching to ensure adequate rest between ­workouts.

Ulnar Neuropathy Symptoms involving the ulnar nerve are very common in throwing athletes because of its superficial location, making it susceptible to injury. More than 40% of athletes with valgus instability develop ulnar neuritis secondary to irritation from inflammation of the UCL, and as many as 60% of throwers with medial epicondylitis also have concomitant ulnar nerve symptoms.55

Ulnar nerve entrapment results from both pathologic and physiologic responses to repetitive trauma. Mechanical factors include compression, traction, and irritation of the nerve. Compression of the ulnar nerve proximal to the cubital tunnel may be due to a tight structure (arcade of Struthers or intermuscular septum) or to hypertrophy of an adjacent muscle (anconeus epitrochlearis or medial head of the triceps). Compression at the level of the cubital tunnel may result from osteophytes, loose bodies, synovitis, or a thickened retinaculum (Osborne’s lesion). Compression may also occur distal to the cubital tunnel at the FCU aponeurosis or at the deep flexor-pronator aponeurosis after the ulnar nerve passes between the two heads of the FCU. The pressure within the ulnar nerve in the flexed elbow and extended wrist has been shown to be elevated to more than 3 times the resting level.56 This has been attributed to nerve compression as well as to physiologic stretching of the nerve (the ulnar nerve normally moves 7 mm medially and elongates 4 to 7 mm during elbow flexion).57,58 As the elbow flexes, increased tension on the arcuate ligament and the UCL also increases tunnel pressures. During the throwing motion, with further elbow flexion and wrist extension combined with shoulder abduction, the intraneural pressure may be elevated to as much as 6 times the resting level. Any tethering of the nerve secondary to chronic changes associated with valgus overload (e.g., scar tissue, calcification of the UCL, traction spurs, degenerative changes in the ulnar groove) further increases intraneural pressures. Traction on the nerve may also result from restriction of its normal mobility. Additional friction on the nerve may be caused by dislocation, present in up to 16% of the population.59 As a result, the cumulative effects of prolonged and repeated pressure elevations produce nerve fibrosis and ischemia. Athletes with ulnar neuropathy usually present with intermittent medial elbow pain that may occasionally radiate down the medial aspect of the forearm into the hand. As inflammation progresses, they may also describe ­clumsiness or heaviness of the fingers on the involved side, as well as numbness and paresthesias in the little and ring fingers. Typically, these symptoms resolve with rest and are exacerbated by throwing or overhead activity. Athletes generally do not complain of weakness in the extremity, a late finding in ulnar neuropathy, because their performance is usually affected in the early stages before the development of motor changes. Painful popping or snapping sensations may also be experienced by patients with recurrent nerve subluxations or dislocations. A careful neurologic evaluation of the neck and upper extremity is mandatory to rule out more proximal causes of neuropathy. Palpation of the ulnar nerve in its groove through a full range of motion should be performed to examine for subluxation or dislocation. The nerve may feel “doughy” or thickened. Patients usually exhibit a positive Tinel sign at the cubital tunnel as well as a positive elbow flexion test (i.e., reproduction of pain, numbness, and paresthesias in the ulnar nerve distribution with maintained maximal elbow flexion and wrist extension for at least 1 minute). The earliest sensory changes are noted with vibrometry or monofilament threshold tests. Nerve-­ending density tests (e.g., two-point discrimination) become positive later as the condition progresses. Motor weakness, if observed, is seen

Overuse Injuries

earliest in the intrinsic hand muscles, such as the ­abductor digiti quinti and adductor pollicis, because the intrinsic motor fibers lie more superficial within the ulnar nerve in the cubital tunnel and are thus more susceptible to injury. Extrinsic weakness involving the flexor digitorum profundus and FCU is usually associated with more severe and advanced compression because the extrinsic motor fibers lie deep within the nerve and are better protected. Plain radiographs of the elbow, especially the cubital tunnel view, may be helpful in determining the presence of any associated pathologic changes in the bones. MRI may be used to identify the presence of soft tissue masses that may be compressing the ulnar nerve, as well as to evaluate the status of surrounding soft tissue structures. Electrodiagnostic studies may be used as an adjunctive diagnostic tool, usually depending on the severity of the patient’s condition. A negative electrodiagnostic study, however, does not rule out the diagnosis of ulnar neuritis. Nerve conduction velocities across the elbow are usually decreased only in cases of advanced or chronic nerve entrapment. A dynamic electromyogram may be more helpful when the diagnosis is equivocal and may aid in differentiating cervical, elbow, and more distal nerve involvement. Thermography is currently being investigated as a diagnostic test for ulnar neuropathy. However, no conclusive evidence has yet been reported. Nonoperative management of ulnar neuropathy usually begins with rest, ice, and NSAIDs. Immobilization of the elbow for a brief period (2 to 3 weeks) may be necessary, especially in cases of ulnar nerve subluxation or dislocation. Local corticosteroid injections are not recommended. Although nonoperative treatment has had high success rates in the general population, many athletes, especially those with associated valgus instability, experience a recurrence of symptoms on resumption of throwing and ultimately require surgical intervention. Indications for surgery include failed nonoperative management, persistent ulnar nerve subluxation, symptomatic tension neurapraxia, and concomitant medial elbow problems (e.g., valgus instability) that require surgery. Surgical options include simple decompression, medial epicondylectomy, and anterior subcutaneous and submuscular transpositions. Simple decompression and medial epicondylectomy have been shown to have poor results in the overhead athlete and are thus not recommended.

Pediatric Elbow Overuse injuries of the elbow are common, particularly among young throwing athletes. The differential diagnosis is different for adults and children. In general, the weak musculoskeletal links in adults usually are the musculotendinous junction and ligaments, such as in ulnar collateral ligament tears. The weak links in children are the bony growth plates (physes) and growth centers (the epiphyses). Injury of these sites can lead to medial epicondyle fragmentation. At the elbow, epiphyses are several secondary ossification centers, all of which close at different ages. These centers should not be confused with loose bodies. Traction apophysitis may occur because of pull along the epiphyseal growth center (e.g., medial epicondylar avulsion or epicondylitis).

623

Elbow pain in the pediatric population has often been referred to as “little leaguer’s elbow,” observed in about 25% of youth league pitchers. This term consists of a variety of diagnoses, all caused by repetitive valgus extension overload. Little leaguer’s elbow presents with elbow discomfort, a decrease in throwing effectiveness, and occasionally a loss of elbow extension. Little leaguer’s elbow includes medial epicondylar avulsion, medial humeral overgrowth (when chronic), ulnar neuritis, and OCD of the capitellum. Tensile forces across the medial elbow can cause avulsion injuries of the medial epicondyle and ulnar neuritis. If the elbow pain has been chronic, medial humeral ­overgrowth can occur. Typically, the pediatric athlete experiences overload from repeated throwing, both in games and ­ during practice with coaches and parents. Sidearm ­ throwing appears to predispose to elbow injury more so than classic overhead pitching. With throwing biomechanics, different stresses are created at different phases of throwing. During the cocking and acceleration phases, tremendous valgus traction and lateral compression stresses are created (Fig. 14-4). During the follow-through phase, ­ extension stress is created. When throwing a curveball, as the wrist supinates before ball release, eccentric activity of the flexor-­pronator group is needed in the acceleration and follow-through phases.60 This action causes further stress to the medial elbow. Compressive forces to the lateral elbow can lead to OCD of the capitellum and radial head. The OCD etiology is likely associated with vascular insufficiency in combination with compressive forces. Surgery is often needed if osteochondral fragments become displaced. OCD should not be confused with Panner’s disease (developmental osteochondrosis of the capitellum), which is a self-limited condition occurring indolently in a younger age group. Unlike OCD, loose bodies are rarely found in Panner’s disease. Radiographs can be quite helpful in discerning the diagnosis (Figs. 14-5 and 14-6). Contralateral elbow films may be taken for side-to-side comparisons of the ossification centers (as opposed to loose bodies). Evidence of an anterior fat pad (sail sign) or a posterior fat pad on radiographs suggests an elbow effusion and intra-articular pathology. Treatment of Little Leaguer’s elbow is directed at ceasing the offending activity, controlling pain, and addressing the kinetic chain. Arthroscopic referral is important in cases with chondral injury, for removal of loose bodies, and for management of OCD through drilling or fragment fixation. Open reduction is necessary for displaced OCD fractures. Foremost, kinetic chain assessment must be accomplished. Children who have improper ­biomechanics at the shoulder or trunk can begin to substitute elbow musculature to generate higher throwing velocity. Young overhead athletes may also develop growth overload to the shoulder. These factors may go unrecognized in the younger patient because they often have referred pain rather than localized shoulder pain. The “Little Leaguer’s shoulder” is a stress fracture to the proximal humeral physis. Cessation of throwing is recommended for 6 weeks to 6 months to prevent growth plate arrest. To prevent these pediatric injuries, several organizations have imposed recommendations and rules on throwing. Suggested recommendations include participating in

624

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A Cocking B Acceleration C Deceleration Figure 14-4   Valgus compressive stresses of the elbow commonly occur during late cocking (A) and early acceleration (B) in the arm of the throwing athlete. C, Deceleration of the arm. (Modified from Miller MD, Cooper DE, Warner JJP: Review of Sports Medicine and Arthroscopy. Philadelphia, WB Saunders, 1995, p 123.) less than six innings of pitching per week, throwing fewer than 100 pitches per week either in practice or game, and discouraging breaking balls (curveballs, sliders, forkballs, sinkers, splitters) until after puberty. Student athletes are advised to weight-train only after reaching puberty.

Wrist Overuse Injuries De Quervain’s Syndrome Stenosing tenosynovitis of the first dorsal compartment (de Quervain’s syndrome) is the most common tendinitis about the wrist in the athlete. It is the result of shear microtrauma resulting from repetitive gliding of the first dorsal compartment tendons (abductor pollicis longus [APL] and extensor pollicis brevis [EPB]) beneath the sheath of the first compartment over the radial styloid. Activities requiring forceful grasping coupled with ulnar deviation

A

B

C

or repetitive use of the thumb predispose the athlete to this condition.61 These include golf, fly fishing, and certain racquet sports (squash and badminton). The athlete presents with radial wrist pain and tenderness over the first dorsal compartment. Finkelstein’s test is pathognomonic in making the diagnosis: the patient flexes the thumb into the palm while the examiner ulnarly deviates the wrist, producing the patient’s symptoms.62 Treatment is dictated by the stage of the disease. Rest and immobilization may be helpful in very early stages (25% to 72% cure rates reported); however, corticosteroid injection into the first dorsal compartment has provided favorable results.63 Additional injections may be indicated, and if no progress occurs, surgical release of the first dorsal compartment may be performed. Patients with de Quervain’s disease have a high incidence of EPB tendon traveling in a separate compartment, and this must be decompressed as well as that of the APL. During the

D

Figure 14-5   Anteroposterior (A) and lateral (B) radiographs of a 14-year-old girl with osteochondritis dissecans of the right capitellum demonstrating a crescent sign (arrow on A) of the anterior capitellum. Anteroposterior (C) and lateral (D) radiographs of the same patient 3 years later demonstrating flattening of the capitellum and loose bodies in the joint. (From Kobayashi K, Burton KJ, Rodner C, et al: Lateral compression injuries in the pediatric elbow: Panner’s disease and osteochondritis dissecans of the capitellum. Am Acad Orthop Surg 12:246-254, 2004.)

Overuse Injuries

A

625

racquet sports, weight-training, and other sports requiring repetitive wrist extension. Physical examination reveals tenderness and swelling at the intersection point, and frequently crepitus is noted as the wrist is actively extended and flexed (hence, the term squeakers). Symptoms normally respond to a course of rest, splinting, and NSAIDs, with or without injection. In rare cases that do not respond, surgical decompression of the second-compartment tendons, release of the fascia of the APL and EPB muscles, and débridement of the bursa are indicated.66,67 A graduated return to weight-training and sport is recommended after symptoms are relieved.

B

Extensor Carpi Ulnaris Tendinitis

C

D

Extensor carpi ulnaris (ECU) tendinitis is second to de Quervain’s in frequency in the athlete.68 ECU tendinitis is seen in rowing and racquet sports and is quite common in the nondominant wrist of tennis players caused by the twohanded backhand. Biomechanical studies have shown that the wrist in tennis is in ulnar deviation for most shots and that the nondominant wrist is in extensive ulnar deviation during the two-handed stroke. ECU tendinitis may be the result of underlying ulnar wrist pathology such as triangular fibrocartilage complex (TFCC) injury. Treatment of ECU tendinitis involves splinting, rest, NSAIDs, occasional steroid injection of the sheath, and attention to technique, with modification to avoid recurrence. Failure to respond to this treatment regimen may indicate underlying pathology, and a further work-up is recommended.

Subluxation of the Extensor Carpi Ulnaris E

F

Figure 14-6   Anteroposterior (A) and lateral (B) radiographs of a 6-year-old boy with Panner’s disease of the right elbow, taken after a fall. Arrows indicate irregular ossification of the capitellum. An effusion of the right elbow joint is seen on both T2-weighted coronal magnetic resonance imaging (MRI) (C) and fast spin-echo sagittal MRI (D). Anteroposterior (E) and lateral (F) radiographs of the same child 1 year later show that healing has occurred, but flattening is clearly present. (From Kobayashi K, Burton KJ, Rodner C, et al: Lateral compression injuries in the pediatric elbow: Panner’s disease and osteochondritis dissecans of the capitellum. Am Acad Orthop Surg 12:246-254, 2004.)

surgery, care must be taken to protect the branches of the superficial and radial nerves that dorsally overlie the first dorsal compartment.

Intersection Syndrome Intersection syndrome is an inflammatory condition located at the crossing point of the first dorsal compartment muscles (APL and EPB) and the radial wrist extensors (extensor carpi radialis longus and extensor carpi radialis brevis [ECRB]), which lies 4 to 6 cm proximal to the radial carpal joint.64,65 This entity is frequently seen in oarsmen,

Although not truly an overuse syndrome, subluxation of the ECU tendon should be considered in the diagnosis of the athlete with ulnar wrist pain. Subluxation of the ECU results from rupture or attenuation of the ECU subsheath, usually due to a sudden volar flexion ulnar deviation stress such as hitting a low forehand in tennis. It has also been reported in golfers, weightlifters, and rodeo bronco ­riders. The anatomy has been well described by Taleisnik and involves ­rupture of the medial wall of the subsheath, which is separate from the overlying supratendinous retinaculum.69 Diagnosis may be made by having the athlete actively ulnarly deviate the wrist in full supination, observing the ECU tendon subluxing ulnarward over the styloid.69 Diagnosis may be confirmed by injecting lidocaine into the ECU sheath, which should result in complete relief of pain. Underlying pathology such as TFCC injuries should be considered. In acute injuries, some authors recommend casting for 6 weeks with the wrist pronated and dorsiflexed. Rowland recommended open repair in acute injuries for a more predictable outcome.70 If performed, a more aggressive rehabilitation program may be recommended. In chronic cases, reconstruction of the subsheath may be performed.

Wrist Flexor Tenosynovitis FCR tendinitis is rare in the athlete, but the treating physician should be aware of the condition. Before insertion at the base of the second metacarpal, the FCR passes through

626

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

a tunnel formed by the transverse carpal ligament, scaphoid tuberosity, trapezial ridge, and radial margin of the FCR tunnel. The condition usually responds to rest and splinting, although injection into the tunnel may be indicated. FCU tendinitis is more common and has been reported in golf and racquet sports such as badminton and squash.71 Pisotriquetral compression syndrome may be an accompanying condition because the pisiform is a sesamoid bone within the substance of the FCU. Pisotriquetral arthritis is best visualized on a lateral radiograph of the wrist in slight supination and mild extension. Treatment of rest, splinting in 25 degrees of volar wrist flexion, and corticosteroid injection into the sheath or pisotriquetral joint results in resolution of symptoms in 35% to 40% of cases according to Palmieri.72 In refractory cases, pisiform excision with or without Z-plasty lengthening of the FCU is usually curative, with return to racquet sports in 6 to 8 weeks. Trigger finger may be produced by direct pressure on the metacarpophalangeal joint flexion crease from repeated racquet use and is occasionally seen in the athlete. Cortisone injection is the preferred treatment in nondiabetic patients, with cure rates that range from 36% to 91%.72,73 Open release of the A-1 pulley under local anesthetic is indicated in refractory cases.

Carpal Tunnel Syndrome The young athlete will occasionally present with acute carpal tunnel syndrome due to significant tenosynovitis of the digital flexors secondary to repetitive digital flexion activities. In most of these cases, symptoms will resolve with rest, immobilization, and NSAIDs, with or without steroid injection. A short course of oral corticosteroids may be indicated if initial treatment fails. The EMG is usually normal, and it is rare that the athlete requires carpal tunnel release.74

Knee Overuse Injuries Patellar Tendinopathy Patellar tendinopathy is a common and significant syndrome encountered in sports medicine (Fig. 14-7). Referring to a clinical condition characterized by activity-related anterior knee pain associated with focal patellar-tendon tenderness, patellar tendinopathy is believed to result from repeated loading of the knee extensor mechanism and is thus most prevalent in sports involving some form of jumping.75 In recognition of this association with jumping, patellar tendinopathy was first described as and is commonly referred to as “jumper’s knee.” This term is misleading, however, because the condition is found in a wide variety of sports participants, including those who do not participate in sports involving jumping. Another traditionally popular term to describe the clinical condition is “patellar tendinitis.” Histopathologic studies, however, have consistently shown the pathology underlying patellar tendinopathy to be degenerative rather than inflammatory; thus the suffix -itis, implying the presence of inflammation, is inaccurate.76 To describe the histopathologic presentation of the condition, the term tendinosis is preferred. This distinction is important because the correct labeling and

Figure 14-7   Ultrasound appearance of patellar tendinopathy. On the left side of the figure, the right tendon shows a large area of decreased echogenicity (blackened area) associated with tendon thickening. On the right side, the left tendon shows a normal appearance of the patellar tendon.   (From Warden SJ: Patellar tendinopathy. Clin Sports Med 22: 743-759, 2003.)

understanding of the pathology has repercussions for management and the outcome expectations of both the clinician and athlete. To clarify the terminology surrounding the syndrome, it has been advocated that the term patellar tendinopathy be used clinically to describe overuse conditions of the patellar tendon. Alternative terms such as tendinosis and tendinitis should only be applied after pathologist examination of tissue biopsies because these refer to distinct histopathologic conditions that cannot be assessed clinically. In line with these recommendations, the term patellar tendinopathy is used in the following discussion, which overviews current understanding of the pathology, pathophysiology, and pathogenesis of patellar tendinopathy, and discusses aspects of its clinical examination, imaging, and ­management. The pathology underlying patellar tendinopathy only recently has been defined, reflecting the confusion ­resulting from differences in nomenclature, rather than a paucity of data. Macroscopically, the patellar tendon of patients with jumper’s knee contains soft, yellow-brown, and disorganized tissue, commonly labeled mucoid degeneration, and some investigators report hyaline degeneration. At light microscopy, the pathologic tendon tissue shows the presence of abnormal collagen, tenocytes, and abundant abnormal small vessel ingrowth. The amorphous and disorganized collagen bundles show degenerative and necrotic tendon tissue replacing collagen. Clefts in the collagen suggest microtears, which may be interpreted as microscopic partial ruptures.77 The characteristic reflective polarized light appearance of normal collagen is lost, tenocytes lose their fine spindle shape, and nuclei appear more rounded. A major feature is the absence of inflammatory cells in excision biopsies, even at the periphery of abnormal tissue and in patients who have only had symptoms for 4 months. Therefore, the tendons of patients suffering from patellar tendinopathy appear to have ­tendinosis, a degenerative condition, rather than an inflammatory condition. Sudden repeated tensile overload may cause microtearing and fraying of tendon fibers followed by focal degeneration,

Overuse Injuries

particularly at the tendon attachment to the patella.78 In patients with spontaneous patellar tendon rupture, a variety of intratendinous pathologies have been noted, including hypoxic degenerative change, mucoid degeneration, tendolipomatosis, and calcifying tendinopathy; or, a combination of these pathologies are present, even though the patients never experience symptoms of patellar tendinopathy.79 Therefore, asymptomatic tendinosis can precede acute tendon rupture.80 The classic site of patellar tendon overuse lesions is in the lower pole of the patella; these lesions occur because of repeated overloads on the extensor mechanism. It is more frequent in athletes whose sports involve activities requiring sudden maximal muscle-tendon unit exertion, such as occurs in jumping. The diagnosis of patellar tendinopathy is based on patients’ subjective reports of pain related to activity levels. The often insidious onset usually relates to an increase in frequency or intensity of rapid, repetitive ballistic movements of the knee joint, with dull ache in the anterior aspect of the knee after strenuous activity. Continuing to play with these symptoms eventually results in pain that interferes with performance. Other common complaints include pain after sitting for long periods and when going up and down stairs. Tissue alterations are observed more frequently close to the patellar junction than elsewhere in the tendon, probably because of the high concentration of mechanical stresses in this area. However, distal alterations can also appear as the tendon narrows at the tibial tuberosity. The main physical findings in patellar tendinopathy are tenderness and swelling in chronic cases at the inferior pole of the patella (tibial tuberosity in distal jumper’s knees) or in the main body of the tendon with the knee fully extended and the quadriceps relaxed. With the knee flexed to 90 degrees, thus causing tension in the patellar tendon, tenderness significantly decreases.81 The main differential diagnosis is with the patellofemoral syndrome, and the two conditions can coexist. High-resolution realtime ultrasonography and MRI are the imaging modalities of choice in patients with patellar tendon disorders. CT is used but does not offer any advantage over imaging methods not using ionizing radiation. Conservative management regimens usually are based on the patient’s subjective report of pain. Treatments include correction of predisposing factors, relative or absolute rest from jumping activities or from sport, stretching and strengthening, physical therapy modalities, ice, massage, NSAIDs, and corticosteroids by injection or ­ iontophoresis. Surgery generally is performed when patients do not improve after 3 to 6 months of conservative management. A variety of surgical methods for treatment of jumper’s knee are described.82 These include drilling of the inferior pole of the patella, resection of the tibial attachment of the patellar tendon with realignment, excision of degenerated areas, arthroscopic débridement, repair of macroscopic defects, multiple longitudinal tenotomies (known in the Mediterranean countries as scarifications), percutaneous needling, and percutaneous longitudinal tenotomy. Excision of the damaged tissue and insertion of a quadriceps tendon graft can be performed as a salvage procedure in extensive partial ruptures.

627

Iliotibial Band Friction Band Syndrome Although less common than anterior knee pain, lateral knee pain is frequently seen in sports that require repetitive knee flexion and extension, such as distance running and cycling. Iliotibial band (ITB) friction syndrome is the most common cause.83 ITB friction syndrome results from excessive friction between the ITB and the lateral femoral epicondyle. Most frequent in distance runners and military recruits, the condition can occur with any activity requiring repetitive knee flexion and ­extension. The iliotibial tract originates proximally from the confluence of the fascia from the tensor fascia lata, the gluteus maximus, and the gluteus medius. Proximally, the posterior aspect of the ITB attaches to the lateral intermuscular septum, connecting the ITB to the linea aspera on the posterior femur. At the knee, the ITB has an anterior expansion, the iliopatellar band, and a posterior expansion to the biceps femoris. The iliopatellar band attaches the anterior aspect of the iliotibial tract to the patella, acting as a stabilizer of the patella against medially directed forces. The iliotibial tract crosses the knee to insert on Gerdy’s tubercle just lateral and proximal to the tibial tubercle. The function of the ITB depends on the knee position. At less than 20 degrees of knee flexion, the ITB lies anterior to the lateral femoral epicondyle, and assists in knee extension. At greater than 30 degrees of knee flexion, the ITB lies posterior to the lateral femoral epicondyle. The cause of ITB syndrome is attributed to friction between the deep layer of the ITB and the lateral femoral epicondyle. At 20 to 30 degrees of knee flexion, the ITB rubs against the lateral femoral epicondyle. In runners, impingement occurs near foot strike, predominantly in the foot contact phase of the gait cycle or in the deceleration phase,84 and the repetitive microstrains in the ITB may lead to degenerative changes or chronic bursitis. Although it is not proved scientifically, it often is stated that running on roads may cause excessive pronation of the foot on the high side, causing injury.85 Although the ITB syndrome is most common in distance runners, it also is reported in cyclists and athletes involved in sports requiring repetitive knee flexion exercises. Genu varum, excessive pronation with internal rotation of the tibia, a lateral condylar spur, or leg-length discrepancy can increase the tension in the ITB or create friction against the epicondyle. It is unknown whether increased thickness of the ITB is a risk factor or a secondary phenomenon. Training errors also may be responsible for predisposing to ITB syndrome. Experienced runners who abruptly increase their weekly mileage have a higher incidence of ITB syndrome. A more recent theory proposes that hip abduction weakness leads to ITB syndrome. Fatigued runners with hip abduction weakness are prone to increased thigh adduction and tension on the ITB. ITB syndrome presents with pain or burning over the lateral aspect of the knee, about 3 cm proximal to the lateral joint line. Occasionally, the pain radiates proximally along the ITB. Activities such as distance running always on the same side of the street or running downhill may aggravate the symptoms. Initially, the symptoms subside shortly after running but recur after subsequent exercise

628

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

bouts. As the condition progresses, the symptoms may persist even during daily activities. Physical examination reveals tenderness over the lateral femoral epicondyle, greater with the knee at 30 degrees of flexion. A rubbing or snapping sensation may be palpated as the ITB passes over the lateral femoral epicondyle. In the compression test, the knee is flexed to 90 degrees. Pressure is applied to the lateral femoral epicondyle, and the knee is gradually extended (Fig. 14-8). The test is positive if the patient reports pain at 30 degrees of knee flexion similar to the pain experienced when exercising. Tightness of the ITB should be evaluated by Ober’s test (Fig. 14-9). A tight ITB is present if the hip remains abducted and does not drop passively below an imaginary horizontal line. In isolated ITB syndrome, there should be no knee effusion, instability, or positive McMurray’s test. Radiographs mostly are normal (with the exception of very prominent epicondyle or spur), and the differential diagnosis includes tears of the lateral meniscus, tendinopathy of the biceps femoris or the popliteus, early degenerative joint disease of the knee, stress fractures, and lumbar disk pathology. An injection of local anesthetic helps to differentiate extra-articular from intra-articular derangement pathology. MRI confirms the diagnosis of ITB syndrome in patients considered for surgery86 and excludes other diagnoses, such as a lateral meniscal tear. Patients with ITB syndrome have a significantly thicker ITB over the lateral femoral epicondyle. Also, most patients have a small bursa deep to the ITB in the region of the lateral femoral epicondyle. Most patients with ITB syndrome respond to nonoperative measures, with activity modification, NSAIDs, stretching, gluteus strengthening exercises (Fig. 14-10), and orthotics used to correct excessive foot pronation. After a short period of avoidance of running or cycling, ­stretching is undertaken and should be continued after return to activity. Most patients’ symptoms improve within 3 to 6 weeks. A corticosteroid injection into the underlying bursa can be considered in refractory cases. Training errors need to be evaluated, and this may involve decreasing mileage, altering stride length, avoiding hills, or periodically changing direction when running on a sloped surface. In cyclists, the seat height or the foot position may need to be changed. Surgery should be considered after at least 6 months of nonoperative management. After arthroscopy to exclude intraarticular pathology, surgical excision of the affected ITB is performed through a longitudinal incision at the level of the lateral femoral condyle with the knee at 30 degrees of flexion. A triangular piece of the posterior half of the ITB is resected, and the leg is moved through a full range of motion to ascertain that there is no further impingement.

Foot and Ankle Overuse Injuries Achilles Tendon Disorders Achilles tendon disorders occur most often in athletes, and most often in those involved in running sports. An annual incidence of Achilles tendon disorders of 7% to 9% in top-level runners has been reported. Commonly and ­ inappropriately generalized as “Achilles tendinitis” by many clinicians, posterior heel pain in the setting of an

Figure 14-8   The Noble compression test is often positive in patients with ITBFS. This test is performed with the patient supine. The thumb of the examiner is placed over the lateral femoral condyle, and active knee flexion-extension is performed. Maximal pain will be at 30 degrees of knee flexion.

overuse injury of the foot and ankle actually encompasses a spectrum of distinct and often coexistent pathologic disorders with both inflammatory and degenerative etiologies. The classification system set forth by Puddu and colleagues separates degenerative conditions of the tendon itself (tendinosis with or without partial rupture) from inflammation of the paratenon (paratenonitis), inflammation of the tendon substance at its insertion (insertional tendinitis), and inflammation of the commonly afflicted bursa anterior to the insertion of the Achilles tendon on the calcaneus (retrocalcaneal bursitis) from complete tears caused by acute injury.87 Many intrinsic and extrinsic etiologic factors have been proposed to account for the development of Achilles tendon disorders. Common intrinsic etiologies invoked include various alignment and biomechanical faults, including hyperpronation of the foot, limited mobility of subtalar joints and limited range of motion of the ankle joint, leg-length discrepancy, varus deformity of the forefoot and increased hindfoot inversion, decreased ankle dorsiflexion with the knee in extension, poor vascularity, genetic makeup, and gender, age, endocrine, or metabolic factors. Changes in training pattern, poor technique, monotonous, asymmetric, and specialized training, previous injuries, footwear, and environmental factors such as training on hard, slippery, or slanting surfaces have been cited by many authors as extrinsic factors that may predispose the athlete to tendinopathy. Training errors have been reported to be involved in 60% to 80% of runners who have tendon overuse injuries. Training errors cited include running a distance that is too long, running at an intensity that is too high, increasing distance too greatly or intensity too rapidly, and performing too much uphill or downhill work. Training errors, alignment, biomechanics, and extrinsic factors such as footwear and training surfaces can create microtrauma resulting from nonuniform stress within the ­Achilles tendon from different individual force

Overuse Injuries

A

629

To test for contraction of the fascia lata.

B C A negative Ober A positive Ober Figure 14-9   The Ober test. (From Miller MD, Sekiya JK: Sports Medicine: Core Knowledge in Orthopaedics. Philadelphia, Mosby, 2006.)

c­ ontributions of the gastrocnemius and soleus, ­producing abnormal load concentrations within the tendon and frictional forces between the fibrils, and leading to localized fiber damage.88 In this manner, excessive motion of the hindfoot in the frontal plane, especially a heel-strike with excessive compensatory pronation, is thought to cause a “whipping action” on the Achilles tendon and predispose it to tendinopathy.89 History and physical examination play a key role in the diagnosis of Achilles tendon injury. The onset of pain, its duration, and aggravating factors should be documented. A classic history involves an insidious and gradual increase in pain located 2 to 6 cm proximal to the insertion of the tendon and felt after exercise within days of a change in activity levels or training techniques. Rest often relieves symptoms, but return to activity reactivates the pain, generally within a few training sessions. In patients with advanced tendinopathy, pain may occur during exercise and, when severe, may interfere with the activities of daily living. Runners typically experience pain at the beginning

and at the end of a training session, with a period of diminished discomfort in between. Clinical examination should start by the exposure of both legs from above the knees, and the patient should be examined standing and prone. Careful inspection should reveal malalignment, deformity, areas of swelling, obvious asymmetry in the size of the tendon, localized thickening, erythema, and any previous scars. Palpation should document contours of the tendons, tenderness, thickening, palpable tendon nodules or defects, crepitation, and warmth. Biomechanics of the foot, ankle, and leg during walking and running, including slow motion analysis, should be evaluated in athletes. All patients should be examined for evidence of ankle instability. The “painful arc” sign may help to distinguish between lesions of the tendon and paratendon. Although, peritendinitis is characterized by crepitus, exquisite tenderness, and swelling that does not move with tendon action, chronic Achilles tendinopathy is notable for absence of crepitation and swelling, with focal tender nodules that move as the ankle is dorsiflexed and

630

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 14-10   Relative effectiveness of three iliotibial band stretches. (From Fredericson M, White JJ, Macmahon JM, Andriacchi TP: Quantitative analysis of the relative effectiveness of 3 iliotibial band stretches. Arch Phys Med Rehabil 83:589-592, 2002.)

plantar flexed. The VISA-A scale is a subjective, quantitative scale of symptoms and dysfunction in the Achilles tendon and may be a useful tool to assess and follow symptomatology over time.90 Both ultrasound and MRI play a role in the diagnosis of Achilles tendon disorders. Ultrasonography provides an inexpensive, sensitive analysis of the pathology of the Achilles tendon, with data regarding tendon width, water content within the tendon and peritendon, and collagen integrity. Abnormal tendons may have increased tendon diameter, focal hypoechoic intratendinous areas (areas of increased water content that at surgery have been shown to be degenerated tissue), localized tendon swelling and thickening, collagen discontinuity, and tendon sheath swelling or calcifications. In the acute phase, ultrasound examination may reveal fluid surrounding the tendon. In the chronic phase, thickening of the hypoechoic paratenon may be seen, although ultrasonography has not been shown to reliably differentiate focal tendinosis from partial rupture.91 Abnormalities detected by ultrasonography in an asymptomatic Achilles tendon can accurately presage the development of Achilles tendinopathy. MRI provides extensive information on the internal morphology of the tendon and the surrounding structures and is used often for evaluation before surgical intervention. MRI can also help characterize retrocalcaneal bursitis and insertional tendinitis. In patients with chronic tendinopathy, MRI often reveals tendon thickening and increased signal within the Achilles tendon. Areas of mucoid degeneration are shown on MRI as a zone of high signal intensity on T1- and T2-weighted images. The goals of treatment in Achilles tendinopathy are threefold: (1) to minimize the pain, (2) to prevent further degeneration, and (3) to allow return to baseline activity. In athletes, an additional demand is that the recovery time should be as short as possible. Initial conservative

­ anagement aims to relieve symptoms and correct factors m causing load imbalance and repetitive strain on the tendon and surrounding structures. This includes a combination of strategies aimed at controlling inflammation and correcting training errors, limb malalignment, decreased flexibility, and muscle weakness and the use of appropriate equipment during sports.92 The role of anti-inflammatory therapy such as oral NSAIDs or steroid pain relievers to control inflammation remains controversial. Although no inflammatory infiltrate has been documented in histologic analyses of tendinopathic samples, anti-inflammatory medication does help to diminish pain and facilitate rehabilitation in cases of chronic tendinopathy and most certainly has a place in the management of retrocalcaneal bursitis and insertional tendinitis.93 Cryotherapy has also been shown to be useful to help control inflammation and facilitate therapy in tendinopathy.94 Occasionally, complete rest or cessation of the training that caused the symptoms may be required for a short time to settle severe symptoms. Because the repair and remodeling of collagen fibers are stimulated by loading of the tendon, only very short courses of complete rest should be prescribed. Recently, Ohberg and Alfredson have described successful ultrasound-guided injection of polidocanol, a sclerosing agent, to decrease the neovascularization and symptomatology of chronic midportion Achilles tendinosis.95 Appropriate and progressive exercises using eccentric exercise programs targeting specific muscle hypertrophy, speed, strength, and endurance requirements represent the gold standard for Achilles tendon rehabilitation and appear to be effective in most athletes. Mafi and associates have shown prospectively that a program of eccentric calf ­muscle training was superior to concentric training in Achilles tendinosis.96 Correction of biomechanical imperfections is clinically important, even if their effects on ­tendinitis are unclear. Interventions improving ­ flexibility of the ankle

Overuse Injuries

joint, flexibility of calf muscles, amount or speed of foot pronation, and foot and ankle mechanics (with orthotics) have been implicated in ameliorating symptoms of ­tendinopathies. Operative treatment is recommended for patients who do not respond adequately to a 3- to 6-month trial of appropriate conservative treatment. To date, no prospective randomized controlled studies comparing operative and conservative treatment of Achilles tendinopathy have been published. Surgery for overuse tendinopathies usually involves excision of fibrotic adhesions and degenerated nodules, or decompression of the tendon by longitudinal tenotomies. Reconstructive procedures may be necessary if large nodules and lesions are excised. Some authors have used open or percutaneous multiple longitudinal incisions of the tendon.97 In most studies, satisfactory results in 75% to 100% of the patients have been reported after operative intervention of Achilles tendinopathy. Generally, long-standing tendinopathies are associated with poorer surgical outcomes. Recently, an overall complication rate of 11% was documented in a series of 432 consecutive patients.98

Posterior Tibial Tendon Dysfunction Posterior tibial tendon dysfunction (PTTD) is a common cause of painful acquired flatfoot deformity in adults and is associated with substantial functional problems resulting in significant morbidity. These patients typically have a loss of hindfoot inversion, inability to negotiate uneven ground, climb, and descend stairs. As acquired flatfoot syndrome advances, progressive collapse of the medial longitudinal arch, hindfoot valgus, and forefoot abduction abnormalities are noted. Shoe fitting is difficult. Pain and instability in the hindfoot have significant impact on daily routines. Many etiologic factors have been proposed for PTTD, including trauma and anatomic, mechanical, and ischemic processes. None has been specifically proved a causative agent. Hypotheses on the association between preexisting pes planus and PTTD suggest that the chronic stress placed on the posterior tibial tendon because of the flexible planovalgus foot and a tight heel cord could lead to an overuse injury, resulting in repetitive microtrauma and degeneration with time. Johnson and Strom have described three distinct stages of PTTD.99 In stage I, the patient has pain and swelling along the course of the tendon. Because the length of the tendon is normal, the patient is able to perform single heel raise. The flatfoot deformity is minimal, the alignment of the hindfoot-forefoot complex is normal, and the subtalar joint remains flexible. In stage II, the patient is unable to perform single heel raise because of attenuation or disruption of the posterior tibial tendon. The tendon is enlarged and elongated and functionally incompetent. The foot has adopted a pes planovalgus position with collapse of the medial longitudinal arch, hindfoot valgus and subtalar joint eversion, and forefoot abduction through the talonavicular joint. The subtalar joint remains flexible, and the hallmark of this stage is that with the ankle in equinus, the talonavicular joint can be reduced. Stage III disease presents with the patient unable to perform a single heel raise and a more severe flatfoot deformity. In stage III disease, the pes planovalgus

631

deformity is fixed, and the laterally subluxed navicular ­cannot be reduced. The histopathology of PTTD reveals a degenerative intratendinous process similar to that seen in Achilles tendinopathy. Mosier and associates studied 15 normal cadaveric tendons and 15 surgical specimens from patients with stage II PTTD.100 Four types of histopathology were ­present in the disease samples: (1) increased mucin content, (2) hypercellularity, (3) neovascularization, and (4) chondroid metaplasia. Disruption of the normal array of collagen bundles represented a degenerative tendinosis with a nonspecific reparative response to tissue injury. Treatment of PTTD depends on many factors. During the time that the foot remains flexible, treatment is possible with a corrective orthosis such as the University of ­California Biomechanics Laboratory brace, molded anklefoot orthosis, articulated molded ankle-foot orthosis, or Marzano brace. The goal of nonoperative treatment in flexible flatfoot deformities is to control the progressive valgus of the calcaneus. If a rigid deformity of the foot develops, the orthosis should be accommodative to bony deformity and help prevent progression of the deformity. If nonoperative or conservative treatment fails, surgery is indicated because the progression of dysfunction may be rapid and disabling. When conservative treatment fails in the early stages of posterior tibial dysfunction, soft tissue surgical procedures such as tenosynovectomy and tendon débridement may halt the progression of disease. Once flatfoot deformity develops, surgical procedures involving osteotomies and arthrodesis are necessary.

Plantar Fasciitis Plantar fasciitis is a common overuse condition of the plantar fascia, typically at its attachment to the medial tubercle of the calcaneus. The pain can be mild to severe and occurs more often in the morning. Palpation at the medial tubercle and along the plantar fascia can be tender. Excessive pronation is a risk factor for plantar fasciitis. Plain radiographs may demonstrate a heel spur, which is thought to be a result of chronic inflammation and calcification. There is evidence that a heel spur may be present in symptomatic heels.101 Stretching of the gastroc-soleus complex, foot orthotics, and local therapeutic modalities are commonly ordered. Local extracorporeal shock wave therapy has also been considered, but long-term efficacy is still unclear.102

Tarsal Tunnel Syndrome Tarsal tunnel syndrome is a condition in which the tibial nerve is mechanically stretched or entrapped as it courses around the medial malleolus. Direct trauma, supinationinversion ankle sprains, and excessive pronation have all been cited as risk factors.103 Patients will complain of paresthesias in the heel and midfoot and even into the toes with prolonged weight-bearing activities. Forced pronation and the Tinel test (percussion of the tibial nerve posterior to the medial malleolus) may reproduce symptoms. Nerve conduction studies can help confirm the diagnosis, as well as differentiate it from entrapment of the medial or lateral plantar nerves. The medial calcaneal nerve, which branches off the tibial nerve before it travels around the

632

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

medial malleolus, can be entrapped and cause pain along the medial aspect of the calcaneus.104

STRESS FRACTURES Stress fractures are common overuse injuries that are frequently seen in athletes and military recruits. Stress fractures can be defined as a partial or complete bone fracture that results from repeated application of a stress lower than the stress required to fracture the bone in a single ­loading.105,106 Stress fractures account for 0.7% to 20% of all injuries presenting to sports medicine clinics. Track and field athletes have the highest incidence of stress fractures when compared with athletes in other sports such as football, basketball, soccer, and rowing. In a review of 320 athletes with stress fractures, the tibia was the most commonly involved bone (49.1%), followed by the tarsals (25.3%) and the metatarsals (8.8%).107 Bilateral stress fractures occurred in 16.6% of cases. The site of stress fractures also appears to vary from sport to sport. Among track athletes, navicular stress fractures predominate; tibial stress fractures are most common in distance runners; and ­ metatarsal stress fractures predominate in dancers. Stress fractures have been documented to be more common in female military recruits than in their male counterparts.108 Although this trend may also be true in the civilian athletic population, the data are more controversial. With the increasing participation in athletics by elderly people, stress fractures should not be overlooked in this population. Although stress fractures have been described in nearly every bone of the human body, they are more common in the lower extremity weight-bearing bones. Certain stress fractures may be associated with a specific sport, such as the humerus in throwing sports, the ribs in golfers and rowers, the spine in gymnastics, the lower extremity in running activities, and the foot in gymnastics and basketball. Although there is site-specific variability in time to healing and the propensity for delayed union, nonunion, and complete fracture, stress fractures may be broadly classified as low-risk or high-risk injuries. Low-risk stress fractures have a favorable prognosis when treated with activity restriction. In contrast, high-risk stress fractures are prone to delayed union or nonunion, especially if the diagnosis is delayed.109 Stress fractures develop when bone fails to adapt adequately to the mechanical load experienced during physical activity. An abrupt increase in the duration, intensity, or frequency of physical activity without adequate periods of rest may lead to an escalation in osteoclast activity. Ground reaction forces and muscular contraction result in bone strain. Bone normally responds to strain by increasing the rate of remodeling. In this process, lamellar bone is resorbed by osteoclasts, creating resorption cavities, which are subsequently replaced with denser bone by osteoblasts. Because there is a lag between increased activity of the osteoclasts and osteoblasts, bone is weakened during this time. If sufficient recovery time is allowed, bone mass eventually increases. If loading continues, however, microdamage may accumulate at the weakened region. Remodeling is thought to repair normally occurring microdamage. The process of microdamage accumulation and bone remodeling, both resulting from bone strain, plays an important

part in the development of stress fractures. If microdamage accumulates, repetitive loading continues, and remodeling cannot maintain the integrity of the bone; a stress fracture may result.110 The exact mechanical phenomenon responsible for initiating stress fractures remains unclear. One theory holds that excessive forces are transmitted to bone when the surrounding muscles become fatigued.111,112 A reduction in ultimate strength has been demonstrated in bone specimens subjected to hyperphysiologic loading regimens, rendering the bone susceptible to microfractures. Under continued loading conditions, these microfractures may propagate and coalesce into stress fractures. A second theory holds that muscles may contribute to stress fractures by concentrating forces across a localized area of bone, thus causing mechanical insults stronger than the stress-bearing capacity of the bone.113 The origins of stress fractures are most likely site-specific and depend on the bone density and geometry, the direction of the load, the vascular supply to the bone, the surrounding muscular attachments, the skeletal alignment, and the type of athletic activity. Biomechanical factors may predispose to stress fractures by creating areas of stress concentration in bone or promoting muscle fatigue. High arches may predispose to increased risk for femoral and tibial stress fractures, whereas pes planus may predispose to metatarsal stress fractures.114-116 Leg-length inequality has been postulated as a risk factor. Friberg reported a higher incidence of tibial, metatarsal, and femoral fractures in the longer leg and a higher incidence of fibular stress fractures in the shorter leg in military recruits.117 A leg-length discrepancy has also been reported to be associated with a higher incidence of stress fractures in athletes.118,119 In particular, a leg-length inequality of greater than 0.5 cm was reported in 70% of women with stress fractures, compared with only 36% of women without stress fractures. Other biomechanical variables linked to increase stress fracture risk have included hip external rotation of greater than 65 degrees,120 greater forefoot varus, restricted ankle joint dorsiflexion, narrow transverse diameter of the tibial diaphysis, and smaller calf circumference measurement.121,122 No studies have reported the effects of such physiologic factors of muscle mass or muscle strength on predisposition to stress fracture. Additionally, no consistent relationships have been observed between body size or composition and stress fracture risk. In addition to mechanical influences, systemic factors, including nutritional deficiencies, hormonal imbalances, collagen abnormalities, and metabolic bone disorders, can contribute to the development of stress fractures. One important risk factor is that of a history of a previous stress fracture. A family history of osteoporosis is considered to be a risk factor for low bone density and osteoporosis in both females and males, but it is not clear that this necessarily predisposes to stress fractures in athletes. Nutritional status, in particular low calcium intake, may contribute to stress fracture. Other dietary factors such as fiber, protein, alcohol, and caffeine intake may play a role but have not been as well studied. A high incidence of stress fractures has been reported in women.123 Therefore, it is especially important to investigate intrinsic abnormalities in female athletes.

Overuse Injuries

The female athlete triad refers to a female athlete with an eating disorder, amenorrhea, and osteoporosis. Women participating in high-level figure skating and gymnastics are particularly prone to this triad. In an effort to minimize body fat and maintain high athletic performance, many of these athletes develop eating disorders during puberty. A high incidence of amenorrhea and oligomenorrhea has also been described in competitive female distance runners. Infrequent or absent menstrual periods lead to an estrogen-deficient state, resulting in decreased bone mineral density and an increased risk for stress fractures. Male endurance athletes are also predisposed to stress fractures because of abnormally low sex hormones.124,125 Testosterone levels may decrease up to 25% within 2 days of vigorous training. Testosterone inhibits interleukin-6, a cytokine responsible for enhancing osteoclast development.126 Therefore, low levels of testosterone in endurance athletes lead to increased osteoclast production and bone resorption. Anecdotal observation and clinical case series suggest that a transition in training, in particular increasing ­mileage, as well as a higher absolute volume of training can predispose to stress fractures in athletes, although little controlled research has examined this aspect. Additionally, although no data exist on how stress fracture risk is specifically affected by training surface, it may be prudent to advise athletes to minimize the time they spend training on hard, uneven surfaces. A higher incidence of stress fracture was reported in military recruits using older or worn running shoes.127 It is unclear whether this is a direct result of decreased shock absorption or perhaps decreased mechanical support. Reports are conflicting about whether or not insoles can prevent stress fractures. From a practical standpoint, it is important for individuals to train in shoes appropriate for their foot type. Athletes with high-arched rigid feet should select cushioned shoes. Athletes with low arches should select shoes providing stability and motion control. Early diagnosis is essential for avoiding complications and returning the athlete to competition as soon as possible. A patient with a stress fracture typically presents with a history of insidious onset of activity-related pain that progressively worsens over time. A comprehensive history should include a review of the training program, in particular any recent increases in activity level. In addition, the patient’s general health, medications, diet, occupation, related activity, past injuries, and menstrual history in women should be assessed. The most consistent finding on physical examination is localized pain to palpation for superficial bones. For bones that are deep, such as the pelvis or femoral shaft, pain may be elicited through gentle range of motion or specific diagnostic tests. Evaluation of limb biomechanics for leglength discrepancies, malalignment, muscle imbalance, or weakness should be performed. The differential diagnosis is extensive and includes stress reaction, muscle strain, nerve entrapment, neoplasm, and infection. Stress reactions represent prefracture areas of bone remodeling in which the bone is weakened but not physically disrupted.

633

Although stress fractures are usually suspected based on a thorough history and physical examination, various imaging modalities may be helpful in confirming the diagnosis. Imaging studies also provide information on the extent of injury and the predicted time to recovery. Radiographs are initially normal for the first 2 to 3 weeks of symptoms and may reveal no findings for several months. In cortical bone, periosteal reaction, cortical lucency, or a fracture line may be appreciated on later films. In cancellous bone, the findings are more subtle and consist of a band-like area of focal sclerosis without periosteal reaction. Radionuclide imaging has traditionally been the standard means for confirming clinically suspected stress fractures in patients with negative bone radiographs. In the early stages of a stress fracture, before any changes on plain radiographic films, bone scans are highly sensitive for detecting stress injuries.128 Acute stress fractures reveal discrete, localized areas of increased uptake on all three phases of a technetium-99m diphosphonate bone scan. Soft tissue injuries are characterized by increased uptake in the first two phases only, and shin splints are typically positive only on delayed images. As healing of the stress fracture occurs, the flow or angiographic phase (phase I), followed by the blood pool or soft tissue phase (phase II), reverts to normal.129 The intensity of activity on delayed images (phase III) decreases over 3 to 18 months as the bone remodels, often lagging behind the clinical resolution of symptoms. Therefore, bone scans should not be used to monitor healing and return to activity. The disadvantage of nuclear imaging is its lack of specificity. Increased focal uptake may be consistent with a stress fracture, osteoid osteoma, osteomyelitis, or other orthopaedic conditions. In addition, nuclear scintigraphy may be overly sensitive in detecting bone strain. In one study, 12 of 26 bone scans (46%) revealed positive findings at asymptomatic sites.130 These areas of increased uptake may represent subclinical foci of bone remodeling or stress reactions and are classified as pre–grade I lesions with increased uptake in all three phases on the superficial or periosteal layer of bone. The clinical significance of these lesions is unclear. Although many of these lesions progress to stress fractures if untreated, other lesions resolve despite continuous training.131 A brief rest period is appropriate treatment for these lesions. The development of single-photon emission computed tomography (SPECT) has enhanced the contrast resolution of scintigraphic images by eliminating the surrounding soft tissue. This has resulted in improved detection and localization of small stress fractures, especially in the spine and pelvis. MRI is a valuable tool in identifying stress fractures when the clinical diagnosis is in doubt. Besides having a higher specificity than scintigraphy for distinguishing bone in­volvement from soft tissue injuries, MRI is helpful in grading the stage of certain stress fractures and, therefore, predicting the time to recovery.132 MRI avoids exposing the patient to radiation and requires less imaging time than three-phase bone scintigraphy. The disadvantage of MRI is the added cost. A classification system for grading stress fractures by nuclear scintigraphy and MRI has been proposed.133 Based on the stage of the stress fractures, the approximate time to healing can be predicted (Table 14-3).

634

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 14-3  Radiologic Grading System for Stress Fractures Grade

Radiograph

Bone Scan

Magnetic Resonance Imaging

Treatment

1 2

Normal Normal Discrete line (±); periosteal reaction (±) Fracture or periosteal reaction

Positive STIR image Positive STIR and T2-weighted images Positive T1- and T2-weighted images Positive T1- and T2-weighted images

Rest for 3 wk Rest for 3-6 wk

3

Mild uptake confined to one cortex Moderate activity; larger lesion confined to unicortical area Increased activity; no definite cortical (>50% width of bone) break More intense fracture line; bicortical uptake

4

Rest for 12-16 wk Rest for 16+ wk

STIR, short T1 inversion recovery. Adapted from Arendt EA, Griffiths HJ: The use of MR imaging in the assessment and clinical management of stress reactions of bone in high-performance athletes. Clin Sports Med 16:291-306, 1997.

The first step in treating stress fractures is identifying and correcting any predisposing factors. The most common cause of stress fractures is a sudden increase in the quantity or intensity of training. Intrinsic factors such as nutritional, hormonal, and medical abnormalities also need to be assessed. Medical evaluation is considered for any patient with risk factors such as female athletes, athletes with recurrent or multiple stress fractures, and athletes with a history of stress fractures with delayed healing times. In amenorrheic or oligomenorrheic athletes, a return to normal menstrual function by decreasing the training regimen may take months to years. Replacement therapy with oral contraceptives or estrogen can help hasten the return to a normal menstrual state, thereby improving bone mineral density. After a thorough work-up is completed, definitive treatment is commenced. Treatment of stress fractures depends on whether the injury has a low risk (noncritical) or high risk (critical) for complications. Most low-risk stress fractures can be successfully treated with rest followed by a gradual resumption of activity. For lower extremity low-risk stress fractures, a rest period of 2 to 6 weeks of limited weight-bearing progressing to full weight-bearing may be necessary. This is followed by a phase of low-impact activities, such as biking, swimming, or pool running.134 When the patient can perform low-impact activities for prolonged periods without pain, high-impact exercises may be initiated. Typically, the athlete commences a program of increasing jogging mileage followed by a return to sport-specific activities. High-risk stress fractures can present treatment challenges and often require surgical intervention. These problematic stress fractures include fractures of the femoral neck, patella, anterior cortex of the tibia, medial malleolus, talus, tarsal navicular, fifth metatarsal, base of the second metatarsal, and great toe sesamoids. For most patients, a trial of nonoperative therapy is recommended. In the high-performance athlete whose livelihood depends on early return to activity, surgical intervention may be appropriate. Patients with a displaced stress fracture or stress fractures with chronic radiographic findings, such as intramedullary sclerosis or cystic changes, may require open reduction with internal fixation (ORIF) with, or without, bone grafting. Stress fractures are best managed through prevention. Training errors are the most frequent culprits and need to be corrected. New activities, such as hill running or running on a hard surface, may be contributing factors.

The condition of the running shoes also needs to be assessed. In military personnel, improvement in boots, such as the use of viscoelastic insoles, may help reduce the incidence of lower extremity stress fractures.135 Athletes, coaches, military personnel, and parents should be educated about the deleterious effects of overtraining and the importance of periodic rest days. In addition, female athletes and their coaches need to be alerted to the adverse effects of eating disorders and hormonal abnormalities.

Upper Extremity Stress Fractures Stress fractures are far less common in the upper extremity than in the lower extremity, but fractures have been described in athletes participating in throwing sports, tennis, and swimming. Unlike the lower limb, where weightbearing forces play a dominant role, upper limb stress fractures are predominantly caused by muscle forces on the bone. Stress fractures of the upper limb have been described in the clavicle, scapula, humerus, olecranon, ulna, radius, scaphoid, and metacarpals, with the humerus being the most frequently reported.136-142 These injuries should be suspected whenever there is a history of overuse activity and localized bony tenderness is present. All upper extremity stress fractures are considered low risk and respond favorably to rest from the aggravating activity. Training errors and technique problems should be addressed before return to sports activity is allowed.

Humerus Stress fractures of the humerus may occur in the shaft in adults or through the proximal growth plate in skeletally immature athletes (Little Leaguer’s shoulder).143 Repetitive torsional forces and opposing muscular contractions during the throwing maneuver are the likely causative factors. In skeletally immature athletes, the rotator cuff muscles attach proximal to the physis, and the deltoid, pectoralis major, and triceps muscles attach distally, creating torsional and tension forces on the growth plate. The chief complaint is an insidious onset of pain that is exacerbated by hard throwing. Physical examination most commonly reveals focal tenderness over the stress fracture and discomfort with manual resistance to shoulder abduction and internal rotation. Occasionally, mild weakness may be detected, especially in adolescent athletes. Although early radiographs of the humerus are unremarkable, in cases of

Overuse Injuries

A

635

B

Figure 14-11   Comparison radiographs of Little Leaguer’s shoulder in a 15-year-old baseball player. Note the widening of the proximal humeral epiphysis in the right shoulder (A) compared with the left shoulder (B). (From Cassas KJ, Cassettari-Wayhs A: Childhood and adolescent sports-related overuse injuries. Am Fam Physician 73:1014-1022, 2006.)

chronic stress fractures, humeral shaft radiographs may demonstrate cortical thickening along the midthird of the medial cortex. In adolescent pitchers, widening of the lateral portion of the physis may be present on external rotation AP radiographs (Fig. 14-11). The widening may be associated with lateral fragmentation, sclerosis, or cystic changes. Treatment of humeral shaft and proximal humeral growth plate stress fractures is avoidance of the precipitating factor, usually throwing or weightlifting. Early diagnosis and proper treatment are essential because continued forces may result in a spiral fracture of the humerus or premature closure of the physis.144 In most cases, symptoms resolve with activity restriction of 8 weeks in adults and of 12 weeks in adolescents. When the patient is asymptomatic, a gradual return to throwing is permitted. Adherence to Little League rules regarding the number of innings a player can pitch per week and adequate rest days between performances are the best preventive strategies. Additionally, common sense must be used when throwing during practice sessions or at home.

Spine Stress Fractures Stress reaction, stress fractures, and spondylolysis and spondylolisthesis represent a continuum of injury to the pars interarticularis of the lumbar spine. Of the five types of spondylolysis described (dysplastic, isthmic, degenerative, traumatic, and pathologic), the isthmic variety occurs most commonly in athletes.145 The most common ­vertebral level affected is the L5 segment followed by the L4 and L3 levels. Spondylolysis most often originates in children between the ages of 5 and 10 years or in athletic adolescents and has an overall incidence in the general population of about 5%.146 Although a hereditary predisposition may play a role in the development of spondylolysis, physical forces are the major contributing factor. Repetitive hyperextension or extension and rotation place the spine at risk in sports such

as gymnastics, diving, wrestling, and weightlifting.147,148 The pars is susceptible to injury because of its small size, making it the weak link of the vertebral ­segment. Most patients with spondylolysis are asymptomatic. For athletes who develop symptoms, the complaints are described as unilateral low back pain that is exacerbated by activity. The one-leg lumbar hyperextension test, in which the patient is asked to hyperextend the back while standing on the ipsilateral leg of the affected side, usually reproduces the pain.149 Occasionally, the pain may radiate into the buttocks. Hamstring muscle spasm is a compensatory reflex to stabilize the painful segment. Neurologic symptoms are rare except in patients with high-degree slips. For L5-S1 spondylolisthesis, the most common neurologic deficit is an L5 radiculopathy. Radiographic assessment should include AP, lateral, and oblique views. The oblique view is necessary to demonstrate the defect in the pars, which has been referred to as the “fractured neck in the Scotty dog” (Fig. 14-12). Many stress fractures are not visualized on plain radiographs, especially in the acute phases, and require nuclear ­imaging. Bone scanning can also help determine whether the changes in the pars are recent or long term. SPECT is more sensitive than traditional planar technetium scans in localizing stress fractures and distinguishing soft tissue from bone lesions150 (Fig. 14-13). Athletes with asymptomatic spondylolysis require no treatment. For symptomatic patients with acute spondylolysis, treatment consists of restriction of athletic competition. The length of activity restriction is controversial, with recommendations varying from 6 weeks to 6 months, but generally should continue until the athlete is asymptomatic.151 Favorable prognostic factors include early diagnosis and lesions located close to the vertebral body.152 In patients with a bilateral pars defect, spondylolisthesis (Fig. 14-14) may occur, leading to a less ­favorable ­prog­nosis. A lumbosacral antilordotic orthosis that prevents extension can be helpful in relieving symptoms during the

636

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

SPONDYLOLYSIS

Figure 14-12   Radiographic evidence (A) and illustration (B) of spondylolysis. (A, From Helms CA: Fundamentals of Skeletal Radiology. Philadelphia, WB Saunders, 1989.)

painful period.153,154 Patients with more severe or lasting symptoms may require temporary recumbency. Resolution of symptoms usually occurs in 6 weeks to 6 months and depends on the duration of symptoms before diagnosis. Although only 30% to 50% of stress fractures demonstrate bony healing, most athletes become asymptomatic and develop no long-term sequelae. When the athlete is pain free, he or she may begin a graduated rehabilitation program emphasizing a spinal stabilization program consisting of abdominal and paraspinal muscle strengthening.

Rib Stress Fractures Stress fractures of the ribs have been described in athletes performing repetitive upper extremity activity, in people with occupations involving protracted overhead activity, and in patients with a chronic cough.155 Rib stress fractures occur most commonly at the first rib or ribs 4 through 9. The most plausible explanation for rib stress fractures is the repetitive contraction of muscles, which act in opposing directions on the ribs.

First Rib

Figure 14-13   Pars interarticularis stress fracture demonstrated on an axial single-photon emission computed tomography scan. (From Boden BP, Osbahr DC, Jimenez C: Low-risk stress fractures. Am J Sports Med 29:100, 2001.)

Throwing athletes and patients who perform persistent overhead activities, such as paper hangers and auto assemblers, are prone to first rib stress fractures. Stress fractures of the first rib have been reported in basketball players (“rebounder’s rib”) who repeatedly dunk the basketball, and in weightlifters who perform upper body “flies” against heavy resistance. Stress fractures of the first rib commonly occur at the groove for the subclavian artery, just posterior to the scalene tubercle. This foci of relative weakness lies between the anterior scalene muscle (superior-directed force) and the serratus anterior and intercostal muscles (inferior-directed force).156 It has been postulated that the opposing muscular contractions can precipitate the injury.157 Patients with first rib stress fractures typically report a gradual onset of scapular, shoulder, or clavicular pain that is aggravated by overhead activity, deep ­inspiration, and coughing. Less commonly, patients are seen with acute pain when a stress fracture progresses to a complete fracture. Physical findings are usually limited

Overuse Injuries 0

I

II

III

637

IV

Figure 14-14   Grading system for spondylolisthesis. (From Borenstein DG, Wiesel SW: Low Back Pain: Medical Diagnosis and Comprehensive Management. Philadelphia, WB Saunders, 1989.)

to local tenderness at the first rib. The stress fracture is often visible on an AP chest radiograph (Fig. 14-15), but it can be confirmed by a bone scan. Most first rib stress fractures heal uneventfully after a 4-week period of relative rest. In cases when an acute fracture develops, there is a risk for delayed union with healing often requiring 6 to 12 months of activity restriction.

Middle Ribs Stress fractures of the middle ribs typically occur in golfers and elite rowers but have also been reported in gymnasts, tennis players, and swimmers. In a retrospective review of 19 golfers with rib stress fractures, Lord and coauthors found the injury to be prevalent in beginners who had dramatically increased their playing time.158

Figure 14-15   First rib stress fracture in a collegiate basketball player (arrows). Note the abundant callus formation surrounding the healing stress fracture. (From Boden BP, Osbahr DC, Jimenez C: Low-risk stress fractures. Am J Sports Med 29:100, 2001.)

All the fractures were located along the posterolateral aspect of the ribs, with most occurring on the leading side of the trunk. The fourth to sixth ribs were most commonly involved. In rowers, the injury has a propensity for developing in elite, female scullers (rowers who use two oars) at the rib’s posterolateral segments. In the middle ribs, muscle action of the serratus anterior and external oblique muscles can cause stress fractures from repetitive, opposing bending forces.159 It has been suggested that bending forces peak during exhalation. During this phase, the external oblique muscles are under maximal tension and pull the ribs inward and down. Simultaneously, the serratus anterior muscle is contracting eccentrically to produce an upward and outward force on the rib. During rowing, the harmful position is believed to occur at the end of the drive phase, in which the hands are drawn toward the body with shoulder extension and scapular retraction. The diagnosis of rib stress fractures is often confused with intercostal muscle strains, leading to a delay in diagnosis. Point tenderness over the lateral ribs may help differentiate the disorders. Radiographs, including oblique views, should be performed and often can confirm the diagnosis. Because of the obliquity of the lateral ribs and the absence of early radiographic findings, nuclear scintigraphy is particularly helpful for identifying middle rib stress fractures. Rib stress fractures respond favorably to activity modification, with complete healing seen within 4 to 6 weeks. No known cases of delayed union or nonunion have been reported for middle rib stress fractures. The rehabilitation program should include endurance exercises for the serratus anterior muscle, as well as a review of possible training errors or faulty body mechanics. For rowers, a modified finish position may reduce the incidence of recurrence. This involves avoiding extremes of the layback position or reducing trunk and shoulder extension (less reach and pull-through).

638

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Pelvic Stress Fractures Stress fractures may occur at multiple locations in the pelvis, including the sacrum and pubic rami. In addition, stress injuries may occur at pelvic joints such as the symphysis pubis (osteitis pubis) and the sacroiliac joints. Because the pelvis is a ring structure, athletes with stress injuries in the pubic rami may develop a second abnormality in the sacrum, sacroiliac joints, or symphysis pubis or a more benign condition such as muscle strains, bursitis, or low back pain.

Sacrum Most sacral stress fractures may be classified as insufficiency fractures occurring in elderly women with osteoporosis. Stress fractures of the sacrum in young athletes are uncommon but may be underdiagnosed.160 When a thorough history is obtained, many of these athletes, often runners or weightlifters, may be found to be oligomenorrheic or amenorrheic with a subnormal bone density.161 Sacral stress fractures are thought to be due to repetitive cyclic loading of the vertebral column on the sacrum. Athletes with sacral stress fractures typically report the insidious onset of low back and sacral pain that may radiate into the buttocks. There are no specific diagnostic tests, but many patients have localized tenderness over the sacrum. Because of the curvature of the sacrum and overlying bowel gas, plain films are rarely positive. The most sensitive imaging modality for diagnosing sacral stress fractures is a SPECT scan. The lesions typically heal with a 4-week period of rest.

Pubic Rami Stress fractures of the pubic rami are relatively rare, but they have been reported in long-distance runners and military recruits. Women are more susceptible to pubic rami stress fractures than are men. This fact may be due to differences in pelvic geometry, gait biomechanics, or a low estrogen state. In female military recruits, maintaining the same stride length as their male counterparts may place increased forces on the pubic rami.162 The most common symptom is pain in the groin that is exacerbated by prolonged activity. Less frequently reported symptoms include pain in the buttock or thigh. On physical examination, tenderness may be present at the pubic rami. Hip range of motion is usually full. A positive standing test in which the athlete experiences pain or is unable to stand unsupported on the affected leg is suggestive of a pubic stress fracture.163 Stress fractures of the pubic rami usually occur on the inferior rami, adjacent to the symphysis pubis. Tensile forces occur at this site from the symphysis pubis medially and from the hip extensors laterally. Most fractures can be diagnosed from a nondisplaced fracture line on radiographs. Nuclear scanning may be helpful for early detection or to confirm the diagnosis (Fig. 14-16). Stress fractures of the pubic rami have a high healing rate after 6 to 10 weeks of rest. A small risk for nonunion or refracture is present if an adequate rest period is not enforced. Prevention of pubic rami stress fractures is best achieved through training modification and reduction of the stride length in female military recruits.

Figure 14-16   Technetium-99m diphosphonate bone scan showing a superior pubic ramus stress fracture with associated increased uptake adjacent to the symphysis (arrows). (From Boden BP, Osbahr DC, Jimenez C: Low-risk stress fractures. Am J Sports Med 29:100, 2001.)

Lower Extremity Stress Fractures Femoral Neck Stress fractures of the femoral neck, although uncommon, have a high complication rate if the diagnosis is missed or the patient is improperly treated. These fractures may develop as the hip musculature becomes fatigued with prolonged activity and subsequently loses its protective shock-absorbing effect. Intrinsic factors, such as coxa vara and osteopenia, also may predispose the femoral neck to injury. The primary presenting symptom is pain in the anterior groin region. The pain may occur only with activity and weight-bearing and may be accompanied by an antalgic gait. Alternatively, the pain may present as an aching sensation that intensifies during or shortly after activity and ceases with rest. Strenuous activity may intensify the pain enough to prohibit further activity. Some patients experience little or no discomfort and present only after fracture completion. Patients often give a history of having participated in an activity with the characteristic triad of being new, strenuous, and highly repetitious. Physical examination occasionally reveals tenderness to palpation over the anterior aspect of the hip. Limitation of the extremes of hip motion is the most frequent finding. Active straight leg raising and logrolling of the thigh may accentuate hip pain. Heel-strike and other percussive tests have demonstrated poor correlation with femoral neck fatigue fractures.164 Examination of the lower lumbar and sacral spine, the lower extremity, and the contralateral hip is essential to rule out other causes of hip pain. The pain may also be related to radiculopathies, sacroiliac or

Overuse Injuries

knee disorders, referred pain from another lower extremity fatigue fracture, or a fatigue fracture in the contralateral hip. The differential diagnosis of an athlete with hip pain includes not only fatigue fractures of the femoral neck but also soft tissue abnormalities (e.g., synovitis, synovial herniation pits,164 muscle and tendon injuries), neoplasm, infection, avascular necrosis, and transient osteoporosis of the hip. It is important to rule out these causes of hip pain before instituting treatment for a presumed femoral neck fatigue fracture. Plain radiographs, including an AP view of the pelvis (including the proximal femora) and a surgical lateral view of the proximal femora, are usually the first images obtained. Plain radiographs are initially negative in as many as two thirds of patients with femoral neck fatigue fractures. Within the first week after the onset of pain, plain radiographs demonstrate changes in fewer than 10% of patients. Studies suggest that fewer than 55% of patients ever have radiographic evidence of osseous changes.165,166 Plain radiographs typically depict late changes, often with periosteal and endosteal bone formation, and only occasionally demonstrate a radiolucent fracture line (Fig. 14-17). The application of MRI technology to the diagnosis of fatigue fractures was first reported in 1986. Since then, several studies have reported the characteristic findings of femoral neck fatigue and insufficiency fractures on MRI: decreased signal on T1-weighted images and increased signal on T2-weighted images and short T1 inversion recovery (STIR) sequences.167,168 These findings are believed to represent edema secondary to microscopic trabecular fractures. Compared with scintigraphy, MRI demonstrated greater accuracy, sensitivity, and specificity. MRI is also useful in differentiating between soft tissue injury, femoral head disorders, and fatigue fractures of the femoral neck. Fullerton and Snowdy reported the first prospective evaluation of femoral neck fatigue fractures analyzed with both plain radiography and radionuclide imaging.169 This new classification system divides fractures into three types: compression side, tension side, and displaced (Fig. 14-18). The compression-side fracture most commonly demonstrates sclerosis on the compression side of the femoral neck, but a variety of radiographic changes are possible, ranging from the combination of negative radiographs and a positive bone scan of the compression side of the neck to sclerosis on the compression side with a cortical break (Figs. 14-19 and 14-20). In a tension-side fracture, there is callus or overt tension-side cortical disruption. Tensionside fractures also demonstrate a wide spectrum of changes, ranging from the combination of normal radiographs and tension-side uptake on a bone scan to an overt fracture line on the tension side without displacement (Figs. 14-21 and 14-22). The final category is that of displaced fractures (Fig. 14-23). In general, treatment of compression-side femoral neck fatigue fractures is based on the inherent stability of the injury. Although most can be treated nonoperatively, some have found that MRI helps to identify the minority that warrant internal fixation.170 For a compression-side injury without a fatigue line identifiable on MRI or one with a fatigue line less than 50% of the width of the femoral neck, most recommend conservative treatment with strict maintenance of non–weight-bearing status with use of crutches

639

until the patient is asymptomatic. Athletic activity is then restricted for 4 to 6 weeks before allowing a return to athletic activity. For compression-side injuries with a fatigue line greater than or equal to 50% of the width of the femoral neck, percutaneous fixation with three 6.5- or 7.0-mm cannulated screws is performed, followed by weight-­bearing as tolerated (Fig. 14-24). For patients treated ­nonoperatively, careful follow-up with weekly plain radiographs is essential to determine whether the fracture is progressing. If progression is noted, internal fixation is the preferred method of treatment. Displaced femoral neck fatigue fractures are urgent surgical situations. Most agree that expeditious ORIF of this type of fracture is required to prevent further complications. Despite rigid internal fixation with a sliding compression hip screw with a side plate or cannulated screws, the prognosis in the young athlete is generally poor. Postoperative management precludes weight-bearing for 8 to 12 weeks, after which weight-bearing is permitted as tolerated. Long-term follow-up is recommended to identify the complications associated with these fractures. The treatment of nondisplaced tension-side femoral neck fatigue fractures remains controversial. Despite the unfavorable biomechanical forces on the tension side of the femoral neck and the high theoretical likelihood of this type of fracture progressing and ultimately displacing, few examples of this occurrence have been reported in patients treated conservatively. Successful treatment of tensionside femoral neck fatigue fractures with strict bed rest for about 3 weeks, followed by strict non–weight-bearing status and then partial weight-bearing with crutches for up to 14 weeks has been demonstrated.171 However, treatment of tension-side fractures is still controversial, and there are no prospective studies comparing conservative and operative treatment. Even though there are reports of successful conservative treatment of tension-side fractures, most authors, mindful of the potential for serious complications should this fracture displace, view tension-side fractures with a great deal of respect and recommend either extremely cautious conservative treatment or internal fixation. Severe and extremely disabling complications following the completion and displacement of fatigue fractures of the femoral neck have been reported. Serious sequelae include nonunion, varus malunion, osteonecrosis, and subsequent arthritic changes in young adulthood. Nondisplaced fractures with a visible fracture line across the neck have a high propensity to settle in the varus position when treated nonoperatively. Despite emergent ORIF, displaced fractures are highly likely to develop complications. Johansson and colleagues followed up 7 elite and 16 recreational athletes with femoral neck fatigue fractures and found a 30% complication rate overall.172 Of the 23 fractures, 10 were complete and displaced. Nonunion, osteonecrosis, or refracture occurred in half of those 10 fractures despite surgical treatment. Osteonecrosis in 3 patients with displaced fractures necessitated hip arthroplasty in 2 and arthrodesis in 1. No elite athlete returned to the prior activity level regardless of the type of fracture sustained. Beyond the inability to return to high-level athletic endeavors, however, the lifelong consequences of a severe disability in a young adult must be considered the most serious complication.

640

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E Figure 14-17   During unsupervised recreational athlete training for a marathon, this 29-year-old man ran 10 miles a week. He experienced insidious onset of right groin pain starting 1 month ago that was not relieved by rest. Magnetic resonance imaging (MRI) with coronal short T1 inversion recovery (STIR) sequence (A) and coronal T1-weighted sequence (B), show edema (arrowheads) and overt fracture line (arrows), making this a grade 4 stress fracture. Radiograph obtained the same day (C), shows hardly discernible sclerotic line at the stress fracture site (arrow). Follow-up MRI at 1 month of relative rest shows remarkable decrease of edema and no discernible fracture line on T1-weighted (D) and STIR sequences (E). (From Berger FH, de Jonge MC, Maas M: Stress fractures in the lower extremity: The importance of increasing awareness amongst radiologists. Eur J Radiol 62:16-26, 2007.)

Overuse Injuries

641

Figure 14-18   Fullerton and Snowdy classification of femoral neck stress fractures. A, Tension-side   fracture. B, Compression-side fracture. C, Displaced-side fracture. (From Fullerton LR, Snowdy HA: Femoral neck stress fractures. Am J Sports Med 16:365, 1988.)

A

B

C

Femoral Shaft The femoral shaft is particularly susceptible to repetitive stresses and the subsequent formation of a stress fracture on the medial compression side of the femur at the junction of the proximal and middle thirds of the shaft.173 This location serves as the origin of the vastus medialis muscle and the insertion of the adductor brevis muscle, both of which have been implicated as causative factors in the development of stress fractures at this site. The typical symptom of stress fracture of the femoral shaft is insidious onset of pain located in the groin and thigh region. Because of the overlying musculature, the site of injury is often difficult to localize. The fulcrum test may aid in making the diagnosis of femoral shaft stress fractures. With the patient seated on the edge of the examining table, gentle pressure is applied to the dorsum of the knee

Figure 14-19   Anteroposterior radiograph of the proximal femur of a 20-year-old U.S. Navy Sea, Air, and Land trainee illustrates a compression-side femoral neck fatigue fracture. Periosteal new bone formation is present in the inferior femoral neck, and healing endosteal callus formation is present in the inferior half of the femoral neck. (From Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293-302, 1997.)

with the clinician’s hand. The opposite arm is used as a fulcrum under the patient’s thigh and is moved from distal to proximal. When the arm is placed under the site of the stress fracture, the patient experiences pain and apprehension. This test may also be useful in assessing the healing response. Most femoral shaft stress fractures are related to training errors and can be successfully treated by a period of rest followed by gradual resumption of activity. The rest period consists of 1 to 4 weeks of toe-touch weight-bearing progressing to full weight-bearing. Most patients are able to resume light jogging by 6 weeks and full sport-specific activity by 12 weeks.

Patella Stress fractures of the patella have been reported in athletes and in cerebral palsy patients with knee-flexion contractures. They may also be a complication of total knee replacement.174 An apparent bipartite patella that is symptomatic and is visualized as increased uptake on

Figure 14-20   Planar technetium-99m methylene diphosphate radionuclide bone scan of a 19-year-old U.S. Marine recruit reveals markedly increased activity in the right inferior femoral neck, consistent with a compression-side femoral neck fatigue fracture. (From Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293-302, 1997.)

642

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 14-21   A, Anteroposterior radiograph of the proximal femur of a 19-year-old U.S. Marine recruit reveals a complete femoral neck fatigue fracture in which there is overt tension-side cortical disruption with propagation of the fracture across the femoral neck. B, Technetium-99m methylene diphosphate radionuclide bone scan of a 20-year-old U.S. Navy Sea, Air, and Land trainee demonstrates moderate uptake at the superior femoral neck of the right hip consistent with a tension-side femoral neck fatigue fracture. (From Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293-302, 1997.)

A

nuclear scanning may actually be an acute or chronic stress ­fracture. In athletes, stress fractures may occur in either a longitudinal or a transverse direction. It has been postulated that transverse stress fractures initiate on the anterior surface of the patella due to repeated tension forces from the quadriceps and patellar tendons with the knee in flexion. During stance, the quadriceps force required to stabilize the knee is greater than 200% of body weight. Transverse patellar stress fractures have also been identified after anterior cruciate ligament reconstruction. The cause may be related to stress risers created at the bone

Figure 14-22   T2-weighted coronal magnetic resonance image of a 29-year-old U.S. Marine depicts increased signal intensity along the superior aspect of the left femoral neck, consistent with a tension-side femoral neck fatigue fracture. This subtle finding is characteristic of tension-side injuries. (From Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293-302, 1997.)

B

plug defect after harvesting of a bone-patellar tendon-bone graft. In addition, any postoperative knee-flexion contracture may result in increased patellofemoral forces predisposing to a stress fracture. Stress reactions in the patella that are scintigraphically positive but are not evident on plain films may be an underappreciated clinical entity causing patellofemoral pain after reconstruction of the anterior cruciate ligament. Untreated or misdiagnosed patellar stress fractures may develop into complete fractures during athletic activity. Because the overall incidence of patellar stress fractures is low and the number of reported series is small, it is difficult to make definitive treatment recommendations. Nonetheless, restriction of activity is appropriate when nondisplaced fractures are identified on scintigraphy and are not seen on radiographs. When the stress fracture is

Figure 14-23   Anteroposterior radiograph of the proximal femur in a 19-year-old U.S. Navy Sea, Air, and Land trainee shows a complete displaced (Fullerton and Snowdy type III) femoral neck fatigue fracture. (From Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293-302, 1997.)

Overuse Injuries

643

air cast brace is believed to unload the tibia by compressing the lower leg, thus redistributing forces and decreasing the amount of tibial bowing.177 Alternatively, the pneumatic brace may act as a venous tourniquet, shifting electrolytes into the interstitial fluid space. This creates an electronegative charge that stimulates osteoblastic bone formation.

Anterior Tibia

Figure 14-24   Anteroposterior radiograph of a complete nondisplaced compression-side femoral neck fatigue fracture after percutaneous internal fixation with three cannulated 6.5-mm screws. (From Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293-302, 1997.)

evident on radiographs, treatment should be individualized to the patient. Nonoperative treatment with careful observation is recommended for the patient who does not require an immediate return to activity. For high-demand athletes and those with displaced fractures and nonunions, ORIF is appropriate. A standard tension-band wiring technique with Kirschner wires or cannulated compression screws provides excellent fixation.

Medial Tibia In athletes, especially runners, the tibial shaft is the most common site of stress fractures. Tibial stress fractures may occur at any location along the shaft of the bone but are most frequently encountered in the posteromedial cortex or compression side. Biomechanical examination may reveal either a rigid, high-arched foot that is incapable of absorbing load, or an excessively flat foot that causes muscle fatigue. Most tibial stress fractures are transverse in orientation, but longitudinal stress fractures have also been reported.175 Longitudinal stress fractures often have an atypical presentation and require MRI for diagnosis.176 Physical examination reveals tenderness at the site of the stress fracture. Occasionally, swelling is noted, and in the later stages, callus formation may be palpable. The differential diagnosis should include infection, tumors, medial tibial stress syndrome (shin splints), and exertional compartment syndrome. Treatment involves relative rest until the pain resolves. Most tibial stress fractures heal within 4 to 8 weeks, with an average time to return to full sports of 8 to 12 weeks. After the acute pain subsides, the athlete can maintain aerobic fitness with low-impact exercises, such as stationary bicycling, swimming, and stair-climbing machines. On resolution of pain, a gradual resumption of impact exercises is initiated. Athletes may also benefit from a short-period use of either a walking boot or pneumatic brace (an air cast), which may be removed for nonimpact cross-training. The

Stress fractures of the anterior cortex of the midshaft of the tibia are among the critical stress fractures because they are prone to delayed union, nonunion, and complete fracture. Findings on full-length tibial radiographs are subtle, often leading to a delay in diagnosis. Therefore, careful scrutiny of the plain films with a magnifying glass, especially the lateral view, is essential when a patient presents with pain in the middle third of the tibia. Radiographs focusing on the middle third may reveal an anterior tibial stress fracture. It is believed that the middle-anterior cortex of the tibia is vulnerable to nonunion because of poor vascularity and increased tension because of morphologic ­bowing of the tibia formed by constant tension from posterior muscle forces.178 Histopathologic examination of chronic anterior cortex tibial stress fractures has revealed fibrotic infiltrations, local osteonecrosis, and limited or no bone­remodeling activity, consistent with pseudarthrosis. In contrast to compression tibial stress fractures (posteromedial), which usually occur in distance runners, tension tibial stress injuries (anterior) typically occur in athletes performing repetitive jumping and leaping activities. Patients present with point tenderness over the anterior aspect of the central third of the tibia. These fractures have the potential to progress to complete fractures. Radiographs initially are often normal but subsequently develop a characteristic V-shaped (or wedge) defect in the anterior cortex, with the open end of the V being directed anteriorly.179 Callus formation is generally absent. The radiographic appearance has also been referred to as the “dreaded black line” because of the prolonged healing time (Fig. 14-25). This appearance is due to bony resorption and indicates nonunion. At this late stage, bone scanning often is normal, and patients may only have minimal symptoms and thus may be fully participating in sports. After the cortex becomes hypertrophied and the fissure widens, the healing capacity is extremely limited. The differential diagnosis should include infection, tumors, medial tibial stress syndrome, and exertional compartment syndrome. Treatment programs have included prolonged periods of rest and immobilization (up to 4 to 6 months), bone stimulation, and surgery. Initial treatment of stress fractures of the anterior midportion of the tibia is generally a trial of rest, with or without immobilization, for a minimum of 4 to 6 months. If the radiographs reveal chronic changes, such as a wide fissure, surgical intervention becomes necessary. Numerous treatments have been proposed for stress fractures that display delayed union. Prompt healing has been reported after excision and bone grafting of the lesion. In a series in which a regimen of rest and external electrical stimulation was evaluated, seven of eight patients showed complete healing after an average of 8.7 months of treatment. Other authors have described less ­favorable results with electromagnetic stimulation. Chang and

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

644

A

B

Figure 14-25   This 26-year-old male professional ballet dancer had complaints of progressive bilateral anterior tibia pain during jumping exercises for the preceding 8 weeks. Conventional lateral radiographs of both tibiae show multiple, wedge-shaped fractures of the anterior cortex of the midtibia, the so-called dreaded black lines, indicating high-risk tensionside stress fractures (arrows). (From Berger FH, de Jonge MC, Maas M: Stress fractures in the lower extremity: The importance of increasing awareness amongst radiologists. Eur J Radiol 62:16-26, 2007.)

­ arris reported good to excellent results in five patients H with recalcitrant stress fractures treated with reamed unlocked tibial nails.180 Intramedullary fixation has become the favored approach for recalcitrant anterior cortex tibial stress fractures. Athletes do not return to activity until evidence exists of cortical bridging on radiography. If after 4 to 6 months, there is no evidence of healing either clinically or radiologically, surgical intervention is indicated.

Proximal Fibula Stress fractures of the fibula are much less common in the proximal bone than in the distal third.181 Because the fibula has a limited role in weight-bearing, the cause of fracture is likely the result of a combination of muscle traction and torsional forces. In one report, several soldiers developed proximal fibular stress fractures after performing intense jumping exercises from a squatting position.182 Affected patients report diffuse proximal and lateral leg pain that is often exacerbated by exercise or knee range of motion. Instability of the proximal tibiofibular joint should be excluded on examination. Plain films may reveal a transverse or oblique fracture line at the neck of the proximal fibula with periosteal new bone formation developing during the healing phase. Radionuclide imaging is a useful diagnostic tool in uncertain cases. ­Resolution of symptoms

Figure 14-26   Healing distal fibular stress fracture in a 53-year-old woman. The stress fracture was caused by intense exercise during a “boot camp” to lose weight. (From Boden BP, Osbahr DC, Jimenez C: Low-risk stress fractures. Am J Sports Med 29:100, 2001.)

usually occurs within 3 to 6 weeks of reduction of activity. When the patient is pain free and there is no local tenderness, a graduated return to activity may be commenced.

Distal Fibula Although stress fractures may occur at any site in the fibula, the most common location is the distal third of the bone just proximal to the inferior tibiofibular ligaments at the junction of cortical and cancellous bone.183 This injury predominantly occurs in distance runners who train on hard surfaces. Micromotions of the fibula, caused by rhythmic contraction of the long toe flexors, may be responsible for producing distal fibular stress fractures.184 Patients report pain and occasionally swelling over the distal fibula and ankle. Symptoms usually develop insidiously over several days to weeks, but onset may be abrupt when a cortical fracture occurs. The symptoms are aggravated by physical activity and relieved by rest. Pain is elicited by pressing the fibula toward the tibia. As healing progresses, callus formation may be palpable. In the early stages, radiographic findings are subtle or absent. The earliest change is a hazy patch of new bone in the fibula just proximal to the ankle joint. With progression of the lesion, callus or a fracture line may be visualized (Fig. 14-26). Stress fractures of the distal fibula have an excellent prognosis when diagnosed early and treated with a 3- to 6-week period of rest. Without treatment, symptoms may persist for 3 to 6 months.

Medial Malleolus The medial malleolus is a relatively uncommon site for stress fractures, but they can occur in athletes participating in running and jumping activities. Repetitive impingement of the talus on the medial malleolus during ankle ­dorsiflexion

Overuse Injuries

and tibial rotation may result in a medial malleolar stress fracture. Medial malleolar stress fractures generally present with a several-week history of mild discomfort followed by an acute episode that results in seeking medical attention. Patients may have pain during athletic activities for several weeks before an acute episode. The pain increases with activity and is relieved by rest. Although the fracture line is frequently vertical from the junction of the tibial plafond and the medial malleolus, it may run obliquely from the junction to the distal tibial metaphysis (Fig. 14-27). For individuals with negative radiographs and a positive bone scan or an incomplete fracture visualized on MRI, treatment is individualized on the basis of the level of athletic activity. Most patients are successfully treated with cast immobilization or ankle bracing and avoidance of impact activities. Athletes desiring early return to competition may be treated with percutaneous drilling and immobilization or internal fixation. Both surgical and ­nonsurgical treatments usually result in a full return to activity; however, resolution of symptoms may take 4 to 5 months with nonoperative therapy.185 Internal fixation with malleolar screws is advocated for patients with a complete fracture line on radiographs.186 Because of the high shear forces exerted at the fracture site, nonunion may develop. In this circumstance, ORIF with two cancellous screws is required. Bone grafting is indicated when there is fracture displacement and, in chronic cases, when sclerosis is present at the fracture site.

Talus Talar stress fractures most commonly involve the lateral body near the junction of the body with the lateral process of the talus. Talar neck stress fractures have been

A

B

645

reported but are considered rare.187 Athletes may present with prolonged lateral ankle or sinus tarsi pain (for several months) following an ankle sprain despite rehabilitation. Excessive subtalar pronation and plantar flexion may predispose athletes to injury as the lateral process of the calcaneus impinges on the concave posterolateral corner of the talus. Alternatively, a supinated foot may concentrate forces on the lateral process of the talus. Plain films often fail to reveal the stress fracture. CT scans are helpful in identifying the lesion at the posterolateral border of the talus. The stress fracture often extends into the subtalar joint, which explains the symptoms in the region of the sinus tarsi. Outcomes after early return to activity are poor. Therefore, a 6-week trial of non–weight-bearing cast immobilization is recommended, followed by rehabilitation and use of an orthosis to correct any excess pronation. Nonunion fractures respond well to surgical excision of the lateral process.

Tarsal Navicular Tarsal navicular stress fractures occur primarily in active athletes involved in sprinting and jumping sports.188 The presentation typically involves an insidious onset of nondescript pain in the medial arch area that is aggravated by activity. Findings on examination are usually limited to tenderness over the proximal dorsal portion of the navicular, sometimes referred to as the N spot, with occasional limitation of subtalar motion or dorsiflexion of the ankle. Navicular stress fractures occur in the sagittal plane in the central third of the bone or at the junction of the central and lateral thirds of the navicular.189 This site corresponds to the zone of maximal shear stress on the navicular from the surrounding bones. The lesion begins at the ­proximal

C

Figure 14-27   This 26-year-old male professional soccer player has a known medial malleolus stress fracture, anteromedial impingement, and loose body of the left ankle. A, Radiograph shows medial malleolus stress fracture at 12-months’ follow-up, originating at the classic anatomic landmark (junction of tibial plafond and medial malleolus lines, arrows). Magnetic resonance imaging and computed tomography (CT) were performed on the same day. B, Axial PD-weighted image shows a hypointense fracture line (arrow). C, CT scan shows the fracture line with bordering sclerosis (arrow). (From Berger FH, de Jonge MC, Maas M: Stress fractures in the lower extremity: The importance of increasing awareness amongst radiologists. Eur J Radiol 62:16-26, 2007.)

646

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

dorsal articular surface and propagates in a distal and plantar direction, resulting in a partial or complete injury. Microangiographic studies have demonstrated that the navicular is supplied by peripheral, medial, and lateral vessels, leaving the central third relatively avascular. The diagnosis of navicular stress fracture is often missed on routine radiographs because the tarsal navicular lies in an oblique direction. When the diagnosis is suspected, an anatomic anteroposterior radiograph should be obtained with the foot inverted. Additional radiologic studies may include bone scanning, tomography, CT, and MRI (Fig. 14-28). Bone scans are sensitive for detecting navicular stress fractures, but are not specific for distinguishing stress reaction from stress fracture. CT with reconstructions and MRI are more sensitive than bone scanning and provide information on the extent of the lesion. Patients who have an early diagnosis of partial or complete navicular stress fracture have a high union rate if treated for 6 to 8 weeks with non–weight-bearing cast immobilization. When weight-bearing is permitted initially, the risk for delayed union, nonunion, or recurrence is dramatically higher. Minimally displaced navicular stress fractures may be treated with cast immobilization or ORIF. Displaced fractures, delayed unions, and nonunions are best treated with ORIF and bone grafting. Fixation is accomplished by means of one or two compression screws placed across the fracture. A semirigid molded orthosis is recommended for arch support ­ during the rehabilitation phase and after return to athletic activity.

Calcaneal stress fractures require a high index of suspicion to avoid misdiagnosis. The lesions usually occur in longdistance runners and military recruits, or as insufficiency fractures in older, osteoporotic people.191 Calcaneal stress fractures present with localized tenderness over the medial or lateral aspects of the calcaneus. The most common site is the upper posterior margin, just anterior to the apophyseal plate and at a right angle to the normal trabecular pattern (Fig. 14-29). Two weeks after the onset of symptoms, radiographs typically demonstrate a sclerotic appearance on lateral radiograph parallel to the posterior margin of the calcaneus. A less common location for calcaneal stress fractures is adjacent to the medial tuberosity. If the diagnosis is uncertain, a bone scan or MRI can help differentiate stress fractures from Achilles tendinosis, retrocalcaneal bursitis, or plantar fasciitis (Fig. 14-30). Treatment is achieved with 6 to 8 weeks of weight-bearing rest with the use of a soft heel cushion. Joint mobilization and flexibility of the calf muscles are indicated when appropriate. Orthotics may be prescribed to control excessive pronation. Running is ­usually resumed after 6 weeks.

Cuboid and Cuneiform

Metatarsals

Stress fractures of cuboid and cuneiform bones are rare. These are generally considered noncritical stress fractures that may be treated with weight-bearing rest until bony tenderness resolves, after which a gradual return to sport activity is commenced. As for other ­ noncritical

After the tibia, the metatarsal bones are the most common site of stress fractures. Most metatarsal stress fractures occur in the second, third, and fourth metatarsals, with fractures of the second metatarsal neck being the most prevalent. Less common locations include the base of the

A

B

stress ­fractures, a short period in a walking boot may provide comfort. One report did propose a period of non– weight-bearing on crutches for 4 weeks for a cuboid stress fracture, followed by progressive weight-bearing and return to sport activity at 8 weeks.190

Calcaneus

C

Figure 14-28   A, Standard anteroposterior radiograph of a professional basketball player with midfoot pain does not reveal a stress fracture. B, Bone scan reveals increased uptake in the navicular bone. C, Computed tomographic scan demonstrates a complete navicular stress fracture. (From Boden BP, Osbahr DC: High-risk stress fractures: Evaluation and treatment. J Am Acad Orthop Surg 8: 344-353, 2000.)

Overuse Injuries

26%

647

79% 18% 56%

A

21%

B

Figure 14-29   Schematic drawings showing the anatomic distribution (according to percentage) of stress injuries in the anterior, middle, and posterior parts of the bone (A) and in the upper and lower regions of the bone (B). (From Sormaala MJ, Niva MH, Kiuru MJ, et al: Stress injuries of the calcaneus detected with magnetic resonance imaging in military recruits. J Bone Joint Surg Am 88:2237-2242, 2006.)

fifth metatarsal, the base of the second metatarsal in female ballet dancers, and the metatarsal heads.192 Metatarsal shaft stress fractures were initially described in military personnel and were referred to as “march foot” and subsequently as “march fracture.”193 In soldiers, an abrupt increase in training exercises on a hard surface predisposes to injury. Modification of shoes to a viscoelastic insole and training on grass or a soft surface can dramatically decrease the incidence of stress fractures.194 In addition to military personnel, metatarsal shaft stress fractures frequently occur in distance runners and ballet dancers, or iatrogenically after dorsal malunion of a first metatarsal osteotomy in which the weight is transferred to the second metatarsal. Distance runners who train more than 20 miles

A

per week are particularly prone to metatarsal stress injuries. During running, forces are highest under the ­second metatarsal, with bending strain being 6.9 times greater than that of the first metatarsal.195 Although the common denominator in metatarsal stress fractures is excessive force transmitted to the bone without adequate rest periods, the exact origin of these fractures has yet to be clearly defined. Whereas the first metatarsal has some mobility at its proximal articulation, the second metatarsal is more firmly secured at the base, thereby transmitting more force to the bone. A short first metatarsal was originally implicated as a predisposing factor for metatarsal stress fractures, but this has been disproved in two separate reports.196,197 In one study, the metatarsal lengths from

B

Figure 14-30   Plain radiograph (A) and T1-weighted sagittal magnetic resonance image (B) of the ankle in a 19-year-old recruit who had two stress injuries in the calcaneus. Both fracture lines are seen on the radiograph and on the magnetic resonance image. (From Sormaala MJ, Niva MH, Kiuru MJ, et al: Stress injuries of the calcaneus detected with magnetic resonance imaging in military recruits. J Bone Joint Surg Am 88:2237-2242, 2006.)

648

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 14-31   This 45-yearold woman had insidious onset of forefoot pain. At presentation, a conventional radiograph was negative (A). After 16 days, periosteal reaction (arrowhead) and an overt fracture line (arrow) are visible at the distal shaft of the second metatarsal (B). After 7 weeks, prominent callus formation shows healing progress of stress fracture (C). (From Berger FH, de Jonge MC, Maas M: Stress fractures in the lower extremity: The importance of increasing awareness amongst radiologists. Eur J Radiol 62:16-26, 2007.)

A

radiographs of patients with metatarsal stress ­fractures were ­compared with those of a control group.198 The authors found no difference between the length of the first metatarsal in the fracture group and in the control group. An alternative hypothesis holds that fatigue of the plantar flexors of the foot, especially the flexor digitorum longus, results in increased strain in the metatarsal, predisposing it to stress fractures.199 Patients with metatarsal shaft stress fractures complain of forefoot pain that is exacerbated by running, jumping, or dancing activity. The most consistent finding on examination is localized tenderness at the fracture site. Occasionally, the patient may walk with a limp or have swelling on the dorsum of the foot. Radiographs performed within 10 days of the onset of symptoms are usually negative, but thereafter a subtle fracture line with surrounding callus ­formation may be present (Fig. 14-31). The injury responds well to rest and a short-leg walking boot. Occasionally, a below-the-knee walking cast may be necessary if healing is delayed or the patient is experiencing severe pain. As healing progresses, the use of the boot can be discontinued and the patient can use a stiff-soled shoe or a carbon-reinforced bar inserted into the sole of the shoe. For a dorsal malunion of the first metatarsal, an orthosis extending under the hallux and a metatarsal pad proximal to the second metatarsal head may be curative. Most metatarsal stress fractures heal after 4 weeks of activity modification and immobilization. Patients are observed clinically and allowed to commence a graduated activity program once the symptoms have resolved and there is no local tenderness.

Fifth Metatarsal Stress fractures of the fifth metatarsal occur at the proximal diaphysis of the bone just distal to the tuberosity and the ligamentous structures. This injury has a high incidence in basketball players. Fifth metatarsal stress fractures have a propensity for delayed union or nonunion and have a high risk for refracture after nonoperative treatment.200 An acute injury is often preceded by a 2- to 3-week history of discomfort over the lateral aspect of the foot. On examination, point tenderness is present at the site of the stress fracture.

B

C

Pain is exacerbated by inversion of the foot. Radiographs reveal a radiolucent line with variable degrees of periosteal reaction and intramedullary sclerosis (Fig. 14-32). Treatment depends on the stage of the lesion. For the patient with prodromal symptoms but negative radiographs, avoidance of weight-bearing activity and a ­semirigid metatarsal functional brace that unloads the fifth metatarsal can be curative.201 If symptoms persist for more than 3 weeks or if radiographs reveal a stress fracture, treatment options include non–weight-bearing cast immobilization for 6 weeks or intramedullary screw fixation. For highdemand athletes, internal fixation with a compression screw provides good results and faster return to activity.202 For patients with a delayed union and medullary sclerosis on radiographs, intramedullary fixation with curettage is recommended. To avoid reinjury, a functional metatarsal brace should be used for at least 1 month after surgery. Careful preoperative assessment of the radiographs to determine the width and length of the screw is critical to avoid intraoperative complications, such as iatrogenic fracture of the metatarsal. In an average-sized adult, a cannulated 4.5-mm lag screw is preferred. Drilling the intramedullary canal before screw placement to débride the intramedullary fibrous tissue is recommended. The screw should be countersunk to avoid skin irritation at the base of the bone. Axial screw fixation is performed without opening the fracture site and risking further damage to the blood supply. Full return to competition can usually be achieved after 8 to 10 weeks.

Great Toe Sesamoids The function of the great toe sesamoids is to diminish the pressure on the metatarsal head and to provide a mechanical advantage to the flexor hallucis brevis. Sesamoid stress fractures are uncommon but may lead to prolonged disability if misdiagnosed or treated incorrectly.203 The injury has a slight predominance at the medial sesamoid, which lies directly under the head of the first metatarsal. Repeated dorsiflexion of the great toe during running and jumping

Overuse Injuries

A

B

649

C

Figure 14-32   This 29-year-old man, with no recollection of trauma, is an active recreational soccer player presenting with increasing pain over the lateral margin of the foot, present even at rest. A fracture line of the fifth metatarsal, distal to the tuberosity, was found, consistent with Jones’ fracture ��(�A, arrow). Follow-up after 2 weeks (B�) and 6 weeks (C) showed hardly any tendency of healing, a known complication of this tension-side type of stress fracture (arrows). (From Berger FH, de Jonge MC, Maas M: Stress fractures in the lower extremity: The importance of increasing awareness amongst radiologists. Eur J Radiol 62:16-26, 2007.)

can result in tensile forces on the sesamoid sufficient to cause a transverse stress fracture. The clinical diagnosis is suggested by tenderness directly over the plantar aspect of the first metatarsophalangeal joint, discomfort with maximal dorsiflexion of the first toe, or push-off disability. Radiographs should include weight-bearing anteroposterior and lateral views as well as an axial view centered on the sesamoids (Fig. 14-33). Serial images may be helpful in distinguishing a stress fracture from an acute fracture. The reported incidence of bipartite sesamoid in the general population ranges from 5% to 30%, and the incidence of bilaterality is about 80%. Most bipartite sesamoids occur in the medial sesamoid. Radiographically, the bipartite ­sesamoid has smooth edges and is larger than the undivided sesamoid. Findings suggestive of a stress ­ fracture

include a transverse fracture line with jagged margins. Nuclear scanning can help differentiate an acute fracture or a stress fracture from a bipartite sesamoid. Acute stress fractures are treated for 6 weeks with a non–weight-bearing cast that extends to the distal tip of the toe to prevent dorsiflexion. Because the incidence of delayed union, nonunion, or recurrence is so high, early surgical intervention is appropriate for selected patients. In chronic cases, histologic specimens have demonstrated fibrous nonunion or pseudarthrosis with large resorptive cavities; therefore, the threshold for operative treatment is lowered. Surgery involves complete excision of the sesamoid with careful dissection to avoid disruption of the flexor hallucis brevis. Bone grafting without excision also ­ provides

Figure 14-33   Anteroposterior (A�) and lateral (B) radiographs of the foot of a soccer player show a medial sesamoid stress fracture. (From Boden BP, Osbahr DC: High-risk stress fractures: Evaluation and treatment. J Am Acad Orthop Surg 8:344-353, 2000.)

A

B

650

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

satisfactory results. In the rare case of concomitant medial and lateral sesamoid stress fractures, partial sesamoidectomy is preferred over complete excision to avoid a cockup ­deformity. Postoperatively, a cast is applied for 2 to 3 weeks, followed by gradual resumption of activities.

Exertional Compartment Syndrome Chronic exertional compartment syndrome is defined as reversible ischemia secondary to a noncompliant osseofascial compartment that is unresponsive to the expansion of muscle volume that occurs with exercise. Most commonly seen in the lower leg, exertional compartment ­ syndrome in athletes has also been described in the thigh and medial compartment of the foot.204 It presents as recurrent episodes of leg discomfort experienced at a given distance or intensity of running. Although a characteristic history is highly suggestive of exertional compartment syndrome, no physical examination finding can firmly establish the diagnosis. Diagnosis based solely on clinical presentation can lead to misdiagnosis and inappropriate therapy or delay of proper therapy. An exercise challenge and documentation of elevated compartment pressure in one or more of the compartments of the leg confirms the diagnosis. The characteristic presenting complaint of patients with chronic exertional compartment syndrome is recurrent exercise-induced leg discomfort that occurs at a welldefined and reproducible point in the run and increases if the training persists. The quality of pain is usually described as a tight, cramp-like, or squeezing ache over a specific compartment of the leg. Athletes can reliably predict at what intensity or what distances the discomfort will occur as well as how long pain will last, depending on the ­intensity and distance run. Relief of symptoms only occurs with discontinuation of activity.205 Examination may or may not demonstrate fascial hernias. In some cases, the classic exertional component is not as evident, and patients complain of pain at rest or with daily activities as well. Women may be more susceptible to chronic lower leg compartment syndrome than men, and women may also, for unclear reasons, respond less well than men to operative fasciotomy.206 Chronic compartment syndrome, left untreated, can develop into an acute syndrome.207 Several factors are believed to contribute to an increase in intracompartmental pressure during exercise. These are enclosure of compartmental contents in an inelastic fascial sheath, increased volume of the skeletal muscle with exertion due to blood flow and edema, muscle hypertrophy as a response to exercise, and dynamic contraction factors due to the gait cycle. It has also been proposed that myofiber damage as a result of eccentric exercise causes a release of protein-bound ions and a subsequent increase in osmotic pressure within the compartment. The increase in osmotic pressure increases capillary relaxation pressure, thus decreasing the blood flow.208 Development of symptoms may be more common at the beginning of a running season due to muscle hypertrophy, which decreases the volume in the compartment. Rapid increases in muscle size due to fluid retention are also believed to play a role in the development of chronic exertional compartment syndrome in athletes taking the popular supplement creatine monohydrate.209

A neurologic and vascular examination should also be performed with reproduction of the symptoms. Understanding the distribution of nerves and functions of muscles in relation to symptoms can help identify the affected compartment in cases in which the pain is not well localized to one specific compartment, or it may help determine which compartments are more severely affected in cases in which more than one compartment is involved. There are four major compartments in the leg (Fig. 14-34). Each is bound by bone and fascia, and each contains a major nerve. The anterior compartment contains the extensor hallucis longus, extensor digitorum longus, peroneus tertius, and anterior tibialis muscles, as well as the deep peroneal nerve. The lateral compartment contains the peroneus longus and brevis as well as the superficial peroneal nerve. Posteriorly, there are two compartments: the superficial posterior and the deep posterior compartments. The superficial compartment contains the ­ gastrocnemius and soleus muscles and the sural nerve. The deep posterior compartment contains the flexor hallucis longus, flexor digitorum longus, and posterior tibialis muscles, as well as the posterior tibial nerve. Some authors believe that the posterior tibialis should be considered a separate compartment because it is surrounded by its own fascia.210 Anterior compartment syndrome is most common (45%), followed by the deep posterior compartment (40%), lateral ­ compartment (10%), and superficial posterior ­compartments (5%).211 If the anterior compartment is affected, the patient may display weakness of dorsiflexion or toe extension and paresthesias over the dorsum of the foot, numbness in the first web space, or even transient or persistent foot drop. ­Paresthesias in the plantar aspect of the foot and weakness of toe flexion and foot inversion may be revealed when the deep posterior compartment is involved, whereas dorsolateral foot hypoesthesia and plantar flexion weakness may be present if the superficial posterior compartment is affected. Lateral compartment pressure elevation with compression of the superficial peroneal nerve can induce sensory changes over the anterolateral aspect of the leg and weakness of ankle eversion. An inversion and equinus deformity may also be present. Several techniques have been described in the literature for measuring both static and dynamic intramuscular pressures. These techniques include the needle manometer, the wick catheter, slit catheter, continuous infusion, and a solid-state transducer intracompartmental catheter. Each of these techniques offers several advantages and disadvantages. All are time-consuming, however, and require some degree of skill and experience to set up and perform. The most preferred method for measurement of ­compa­rtmental pressures is with a battery-operated, handheld, digital fluid pressure monitor. The Stryker Intra­compartmental Pressure Monitor (Stryker Corporation, ­Kalamazoo, Mich) is a convenient and easy-to-use measuring device for use in the clinical setting. This device has been found to be more accurate, versatile, convenient, and much less time-consuming to use than the needle manometer method.212 Measurements were also found to be more reproducible among different examiners with the Stryker instrument. The usefulness of pressure measurement and maintenance of patient safety with this invasive technique relies

Overuse Injuries Tibia Anterior compartment Lateral compartment

Anterolateral incision

Tibialis posterior (fifth compartment) Fibula

651

Posteromedial incision 1 2

4

Deep posterior compartment

3 Superficial posterior compartment A B Figure 14-34   A, Cross-sectional anatomy of the lower extremity. B, Cross-sectional view of the lower extremity compartments shows location of the anterolateral and posteromedial incisions that allows access to the anterior and lateral compartments   (1 and 2) and the superficial and deep posterior compartments (3 and 4). (From Miller MD, Sekiya JK: Sports Medicine: Core Knowledge in Orthopaedics. Philadelphia, Mosby, 2006.)

upon a thorough knowledge of the anatomy of the leg. Before attempting to measure compartment pressures, the ­physician should thoroughly study the anatomic structures in each compartment to avoid damaging neurovascular structures. Generally accepted criteria for the diagnosis of chronic exertional compartment syndrome (CECS) are described by Pedowitz and colleagues.213 One or more of the following pressure criteria must be met in addition to a history and physical examination that is consistent with the diagnosis of CECS: Pre-exercise pressure greater than or equal to 15 mm Hg; 1 minute postexercise pressure greater than or equal to 30 mm Hg; or 5 ­minutes postexercise pressure greater than or equal to 20 mm Hg. ­Clinicians should also be aware that standard exercise protocols often used in the clinical setting may or may not be adequate to raise intracompartmental pressure and diagnosis may require the sport-specific activity to induce symptoms and raise intracompartmental pressure. Recent interest has focused on the use of noninvasive tools in the diagnosis of chronic compartment syndrome: triple-phase bone scan, MRI, near-infrared spectroscopy, and technetium-99m methoxyisobutyl isonitrile (MIBI) perfusion imaging. The dynamic bone scan may support the diagnosis based on specific tracer uptake patterns. The ­characteristic appearance is that of decreased radionuclide concentration in the vicinity of the area of increased pressure, with increased soft tissue concentration both superior and inferior to the abnormality. The area of decreased uptake is believed to be due to the increased pressure and decreased blood flow to the region.214 On MRI, exercise changes are characterized by swelling within a compartment, which manifests as intramuscular diffuse high signal intensity on T2-weighted images. Failure of the edematous muscle to return to baseline normal appearance by 25 minutes after completion of exercise is diagnostic. The triple-phase bone scan and MRI offer alternatives to direct intracompartmental pressure measurements in cases in which the athlete is averse to repeated needle sticks or in which the results of pressure monitoring may be borderline.215 Near-infrared spectroscopy measures tissue deoxygenation of skeletal muscle caused by elevated intramuscular pressure during exercise.216 MIBI perfusion imaging is a technique that assesses the uptake

of an intravenously injected radiopharmaceutical, MIBI, by peripheral muscles. The uptake of the radiopharmaceutical is largely determined by muscle perfusion, but hypoxia also inhibits uptake of MIBI, enhancing its potential for ­detecting muscle ischemia. Cases have been reported in which visually detectable decreases in MIBI uptake in one or more compartments were noted during exercise when compared with studies taken at rest.217 Treatment of chronic exertional compartment syndrome can include both conservative and surgical intervention. Conservative measures include relative rest (limiting activity to the level that avoids any more than minimal symptoms), NSAIDs, stretching and strengthening of the involved muscles, and orthotics (particularly in cases of excessive pronation). Some athletes will simply choose to refrain from the causative activity, which is a viable option provided they remain neurovascularly intact. In cases in which symptoms persist despite 6 to 12 weeks of conservative care, or in cases of extreme pressure elevation, surgical remediation (fasciotomy of the involved compartments with or without fasciectomy) should be undertaken. Single- and double-incision, as well as endoscopic, techniques have been described. Regardless of the technique, any fascial hernias must be included in the fascial incision. Because of a high rate of coexistence, some authors advocate release of the lateral compartment whenever a procedure for anterior compartment syndrome is performed. Others have stated that this dual release may not be necessary if clinical evaluation and compartment pressure testing fail to demonstrate lateral compartment involvement.218 When performing a deep posterior compartment release, attention must be given to adequate decompression of the tibialis posterior. Postoperatively, a compressive dressing is applied. Drains are not normally necessary. Crutches are used for comfort for a few days, but the patient begins active and passive motion immediately. After the wound is healed, walking and bicycling are encouraged. Patients may begin a light jog in 2 weeks and resume run training at 6 weeks. It usually takes 3 months for full rehabilitation, but patients with deep posterior compartment fasciotomies may need longer.219

652

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Medial Tibial Stress Syndrome Shin splints, or medial tibial stress syndrome, can be described as a clinical entity characterized by diffuse tenderness over the posteromedial aspect of the distal third of the tibia.220 In mild cases, pain is present only with exercise; in more severe cases, rest pain is present. Shin splints have been reported to account for 12% to 18% of running injuries and to occur in 4% of all military recruits in basic training.221,222 Women appear more frequently affected than men.223,224 Medial tibial stress syndrome is to be differentiated from stress fracture and exertional compartment ­syndrome (Table 14-4). Although different entities, they may coexist. Plain films are negative (except in cases of previous or coexistent stress fracture). Bone scans will demonstrate characteristic vertical linear increased activity along the ­ tibial periosteum, which differs from the more focal fusiform increased radiotracer uptake exhibited by stress ­fractures.225 Medial tibial stress syndrome is thought by most to represent a periostalgia or tendinopathy along the tibial attachment of the tibialis posterior or soleus muscles.226 Other proposed etiologies have included posterior compartment syndrome and fascial inflammation.227 Detmer proposed a classification scheme for medial tibial stress syndrome based on etiology. Type 1 included local stress fractures, type 2 periostitis/periostalgia, and type 3 was due to deep posterior compartment syndrome.228 Increased valgus forces on the rear foot and excessive pronation that result in increased eccentric contraction of the soleus and posterior tibial muscles are often contributing causes. Intrinsic factors that may increase valgus forces and pronation include femoral anteversion, genu varum, tibia or forefoot varus, and an excessive Q angle. Other intrinsic factors linked to medial tibial stress syndrome include ­ excessive planus or cavus, tarsal coalition, leglength inequality, and muscle imbalances.229,230 Extrinsic risk factors include improper shoewear, a rapid transition in training, inadequate warm-up, running on uneven or hard surfaces, running in cold weather, and low calcium intake.231 Treatment of medial tibial stress syndrome includes relative rest and the correction of any recent transition

   TABLE 14-4  Medial Tibial Stress Syndrome (MTSS) and Tibial Stress Fracture Pain

MTSS

Onset Insidious With activity Painful at onset Improves with activity Worsens on cessation Location Posteromedial tibia (soft tissue) Junction of the middle and distal thirds Quality Diffuse Pronation None with percussion Pain with terminal heel rise

Stress Fracture Insidious Painful at onset Worsens with activity Improves on cessation Medial tibia (bone) Junction of middle and distal thirds Pin point Pain with percussion Pain with tuning fork Painful hop test

Adapted from Chou LH, Akuthota V, Drake DF, et al: Sports and performing arts medicine. 3. Lower-limb injuries in endurance sports. Arch Phys Med Rehabil 85:59-66, 2004.

in training. Hill running and running on uneven surfaces should be avoided. Proper shoewear is essential to minimize rear foot valgus and to correct excessive pronation, pes planus, or pes cavus. Orthotics are useful in cases that cannot be controlled by shoewear alone. NSAIDs and anti-inflammatory modalities (i.e., ionto­ phoresis and ultrasound) can be useful adjuncts in the ­rehabilitation of medial tibial stress syndrome. A strengthening and flexibility program should be initiated with the goal of correcting any muscle imbalances. Flexibility of the gastroc-soleus should be emphasized, as well as strengthening (concentric and eccentric), including the foot intrinsics, dorsiflexors, plantar flexors, invertors, evertors, and gluteals. All deficits within the kinetic chain should be corrected. A compressive sleeve may provide symptomatic relief. Operative therapy (posterior fasciotomy) has been described for the athlete with severe limitations of physical activity, frequent recurrence, or no response to available therapy.232 Surgical treatment for periostalgia has not been uniformly successful and should be reserved for recalcitrant symptoms that have not responded to a well-documented treatment program of at least 6 months.

Overuse Injuries

C

r i t i c a l

P

o i n t s

l Repetitive microtrauma due to overuse injuries leads to

l­ocal tissue damage in the form of cellular and ­extracellular degeneration and is most likely to occur when an athlete changes the mode, intensity, or duration of training— a phenomenon described as the “principle of transition.” l Both intrinsic and extrinsic factors contribute to overuse injuries. Intrinsic factors are biomechanical abnormalities unique to a particular athlete. Extrinsic (avoidable) factors that commonly contribute to overload include poor technique, improper equipment, and improper changes in the duration or frequency of activity. l Tendinopathy is a clinical condition characterized by activity-related pain, focal tendon tenderness, and intratendinous imaging changes. Historically, it was thought to be one of inflammation and, consequently, the condition was labeled tendinitis. However, recent histopathologic studies have shown the underlying pathology to be primarily one of tendon degeneration (tendinosis). l Although Nirschl coined the term angiofibroblastic hyperplasia for the histologic features seen in elbow tendinosis— presumably to emphasize the neovascularization ­ (angio) and increased cellularity (fibroblastic)—these ­features are both typical of the well-recognized pathologic entity of tendinosis. l For most overuse injuries, relative rest, modification of activities, icing, and NSAIDs (for pain relief rather than to speed recovery) are recommended. Symptom duration and severity and patient response to treatment should help guide specific interventions.

S U G G E S T E D

653

R E A D I N G S

Barrow GW, Saha S: Menstrual irregularity and stress fractures in collegiate female distance runners. Am J Sports Med 16:209–216, 1988. Boden BP, Osbahr DC: High-risk stress fractures. J Am Acad Orthop Surg 8:344– 353, 2000. Dameron TB Jr: Fractures of the proximal fifth metatarsal: selecting the best treatment option. J Am Acad Orthop Surg 3:110–114, 1995. Fulcher SM, Kiefhaber TR, Stern PJ: Upper-extremity tendinitis and overuse ­syndromes in the athlete. Clin Sports Med 17:433–448, 1998. Khan KM, Cook JL, Bonar F, et al: Histopathology of common tendinopathies. ­Update and implications for clinical management. Sports Med 27:393–408, 1999. Maffulli N, Khan KM, Puddu G: Overuse tendon conditions: time to change a ­confusing terminology. Arthroscopy 14:840–843, 1998.

fractures can be defined as a partial or complete bone fracture that results from repeated application of a stress lower than the stress required to fracture the bone in a single loading. l A high incidence of stress fractures has been reported in women. Therefore, it is especially important to investigate intrinsic abnormalities in female athletes. The ­ “female athlete triad” refers to a female athlete with an eating disorder, amenorrhea, and osteoporosis. l Treatment of stress fractures depends on whether the injury has a low risk (non-critical) or high risk (critical) for complications. Most low-risk stress fractures can be successfully treated with rest followed by a gradual resumption of activity. For lower extremity low-risk stress fractures, a rest period of 2 to 6 weeks of limited weight-bearing progressing to full weight-bearing may be necessary. This is followed by a phase of low-impact activities, such as biking, swimming, or pool running. Once the ­patient can perform low-impact activities for prolonged periods without pain, high-impact exercises may be initiated. Typically, the athlete commences a program of increasing jogging mileage followed by a return to sportspecific activities. l High-risk stress fractures can present treatment challenges and often require surgical intervention. These problematic stress fractures include fractures of the femoral neck, patella, anterior cortex of the tibia, medial malleolus, talus, tarsal navicular, fifth metatarsal, base of the second metatarsal, and great toe sesamoids. l Stress

Matheson GO, Clement DB, McKenzie DC, et al: Stress fractures in athletes: a study of 320 cases. Am J Sports Med 15:46–58, 1987. Nirschl RP, Pettrone FA: Tennis elbow: the surgical treatment of lateral epicondylitis. J Bone Joint Surg Am 61:832–839, 1979. Shin AY, Gillingham BL: Fatigue fractures of the femoral neck in athletes. J Am Acad Orthop Surg 5:293–302, 1997.

R E F E R E N C E S Please see www.expertconsult.com

C H A P T E R  

15

Head Injuries Robert C. Cantu and Robert V. Cantu

RELEVANT ANATOMY AND BIOMECHANICS The three main components of the brain are the cerebrum, the cerebellum, and the brainstem. The cerebrum consists of right and left hemispheres, divided into the frontal, parietal, occipital, and temporal lobes. The right hemisphere controls the majority of functions on the left side of the body, and the left hemisphere controls functions on the right side, owing to crossing of nerve fibers in the brainstem. In most people, the left hemisphere is slightly more developed, or dominant, and is the area in which written and spoken language is organized. This is true for more than 95% of right-handed people and even most lefthanded people.1 The cerebrum consists of two main layers. The cerebral cortex or gray matter is the outer layer of the brain (about 2 cm in width). The cortex contains the centers of cognition and personality and helps to coordinate complicated movement. The cortex is highly organized into areas that control specific functions. The inner layer of the brain, or white matter, consists of a network of myelinated nerves, which allow different areas of the brain to communicate with one another. The cerebellum is the area of the brain that controls balance and helps coordinate movement. The brainstem connects the brain to the spinal cord. The brainstem controls various autonomic functions such as respiration, heart rate, blood pressure, and wakefulness. Knowledge of the forces that produce skull and brain injuries requires an understanding of the following ­principles: • A forceful blow to the resting, moveable head usually produces maximal brain injury beneath the point of cranial impact (coup injury). • A moving head colliding against an unyielding object usually produces maximal brain injury opposite the site of cranial impact (contrecoup injury). Such lesions are most common at the tip and undersurfaces of the frontal and temporal lobes (Fig. 15-1). • If a skull fracture is present, the preceding two principles do not pertain because the bone itself, whether it is transiently (linear skull fracture) or permanently (depressed skull fracture) displaced at the moment of impact, may directly injure brain tissue. In discussing brain injuries, it is essential to realize that three types of stresses can be generated by an applied force: compressive, tensile (the opposite of compressive, sometimes called negative pressure), and shearing (a force applied parallel to a surface). Uniform compressive stresses are tolerated fairly well by neural tissue, but shearing stresses are tolerated poorly.

Figure 15-1   Example of coup and contracoup injury showing bleeding in both temporal lobes.

The cerebrospinal fluid acts as a shock absorber, cushioning and protecting the brain by converting focally applied external stresses to a more uniform compressive stress. This is accomplished because the fluid follows the contours of the sulci, thus more equally distributing damaging shearing forces. Despite the presence of cerebrospinal fluid, shearing stresses may still be imparted to the brain. If rotational forces are applied to the head, shearing forces will occur at those sites where rotational gliding is hindered. These areas are characterized by (1) rough, irregular surface contacts between the brain and skull, hindering smooth movement; (2) dissipation of the cerebrospinal fluid between the brain and skull; and (3) impedance of brain motion by dura mater brain attachments. The first condition is most prominent in the frontal and temporal regions and explains why major brain contusions occur at these sites. The second condition explains the coup and the contrecoup injuries. When the head is accelerated before impact, the brain lags toward the trailing surface, thus squeezing away protective cerebrospinal fluid and allowing the shearing forces to be maximal at this site. Brain lag actually thickens the layer of cerebrospinal fluid under the point of impact, which explains the lack of coup injury in a moving head injury. In contrast, when the head is stationary before impact, there is neither brain lag nor disproportionate distribution of cerebrospinal fluid, accounting for the absence of contrecoup injury and the presence of coup injury. 657

658

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The scalp has energy-absorbing properties; 10 times more force is required to produce a skull fracture in a cadaveric head with the scalp in place than in one with the scalp removed. In addition, Newton’s law—force equals mass times acceleration—must be appreciated. An athlete’s head can sustain far greater forces without brain injury if the neck muscles are tensed at the moment of impact. In the relaxed state, the mass of the head is essentially its own weight. With the neck rigidly tensed, however, the mass of the head approximates the mass of the body.

CLASSIFICATION Concussion The most common head injury sustained by an athlete is a concussion. Multiple definitions of concussion exist. The word itself is derived from the Latin verb ­concussus, which means “to shake violently.”2 The Committee on Head Injury Nomenclature of the Congress of Neurological Surgeons defines concussion as “a clinical syndrome characterized by immediate and transient posttraumatic impairment of neural function, such as alteration of consciousness, disturbance of vision, equilibrium, etc, due to brainstem involvement.”3 As Kelly has stated, a concussion is a “trauma-induced alteration in mental status that may or may not involve loss of consciousness.”3 The American Orthopaedic Society for Sports Medicine defines concussion as “any alteration in cerebral function caused by a direct or indirect (rotation) force transmitted to the head resulting in one or more of the following acute signs and symptoms: a brief loss of consciousness, lightheadedness, vertigo, cognitive and memory dysfunction, tinnitus, blurred vision, difficulty concentrating, amnesia, headache, nausea, vomiting, photophobia, or balance disturbance. Delayed signs and symptoms may also include sleep irregularities, fatigue, personality changes, inability to perform usual daily activities, depression, or lethargy” (Box 15-1).3 Although some have suggested that a concussion is a physiologic disturbance without structural damage, animal and human data have shown that neurochemical and structural changes with loss of brain cells can occur. A neurochemical cascade begins within minutes following a concussion and can continue for days. It is during this period that neurons remain in a vulnerable state, susceptible to minor changes in cerebral blood flow, increases in intracranial pressure, and anoxia.4 Animal studies have shown that during this susceptible period, a decrease in cerebral blood flow that normally would have little consequence can produce extensive neuronal cell death.4 Several attempts have been made to classify concussions based on their severity, with guidelines for return to play. The most commonly used classifications have three grades, with grade 1 described as mild, grade 2 as moderate, and grade 3 as severe. The classification schemes vary somewhat, but are all based on clinical presentation of the athlete, and especially with the Cantu classification, the duration of symptoms5 (Table 15-1).

BOX 15-1  Postconcussion Signs  and Symptoms “Bell rung” Depression “Dinged” Dizziness Excess sleep Fatigue Feeling “in a fog” Feeling “slowed down” Headache Inappropriate emotions or personality changes Loss of consciousness Loss of orientation Memory problems Nausea Nervousness Numbness and tingling Poor balance and coordination Poor concentration, easily distracted Ringing in the ears Sadness Seeing stars Sensitivity to light Sensitivity to noise Sleep disturbance Vacant stare, glassy eyed Vomiting

Evaluation Clinical Presentation and History The grade 1, mild concussion (based on the Cantu grading scale) is the most difficult to recognize and judge. The patient does not lose consciousness but suffers impaired intellectual function, especially in remembering recent events and in assimilating and interpreting new information. Grade 1 concussion occurs most frequently (>90% of concussions) and often escapes medical attention. Players commonly are “dinged” or have their “bell rung” and continue to play. Dave Meggyesy, an author and former professional football player, described this condition: “Your memory’s affected, although you can still walk around and sometimes continue playing. If you don’t feel pain, the only way others know that you have been ‘dinged’ is when they realize you can’t remember the plays.”6 Grade 2 involves either more protracted postconcussion signs and symptoms beyond 30 minutes or a brief loss of consciousness up to 1 minute. It is therefore more easily identified than a grade 1 concussion. If the period of unconsciousness is brief and if the athlete has no neck problems after regaining consciousness, clinical judgment may dictate that removal on a fracture board is not necessary. The athlete should be removed from the game and evaluated by a neurologist at a medical facility. Grade 3 concussions involve loss of consciousness for longer than1 minute. Initial treatment should follow the advanced trauma life support (ATLS) guidelines for trauma

Head Injuries

659

  TABLE 15-1  Evidence-Based Classification Schemes for Concussion5 Mild: Grade 1

Moderate: Grade 2

Severe: Grade 3

Cantu

No LOC; PTA, PCCS < 30 min

American Academy of Neurology

Transient confusion; no LOC; symptoms or abnormalities resolve in < 15 min

LOC < 1 min, PTA > 30 min but < 24 hr, PCCS > 30 min but < 7 days Transient confusion; no LOC; symptoms or abnormalities last > 15 min

LOC > 1 min, PTA > 24 hr, PCCS > 7 days Any LOC

LOC, loss of consciousness; PTA, post-traumatic amnesia (anterograde/retrograde); PCCS, postconcussion signs/symptoms other than amnesia. Data from Cantu RC: Posttraumatic retrograde and anterograde amnesia: Pathophysiology and implications in grading and safe return to play. J Athl Train 36:244-248, 2001.

resuscitation. The athlete should be suspected to have sustained a cervical spine fracture and should be transported on a fracture board, with the head and neck immobilized, to a hospital with neurosurgery service. All athletes with severe concussion should be evaluated for possible intracranial bleeding.

Postconcussion Syndrome The postconcussion syndrome consists of headache (especially with exertion), labyrinthine disturbance, fatigue, irritability, and impaired memory and concentration. The true incidence of this syndrome is not known. Persistence of symptoms reflects altered neurotransmitter function and usually correlates well with the duration of post-traumatic amnesia, and it suggests that the athlete should be evaluated by computed tomography (CT) and neuropsychiatric testing. Before an athlete is allowed to return to play after a head injury, the criteria in Table 15-2 should be met. Otherwise, the athlete risks cumulative brain injury as well as the second-impact syndrome.   TABLE 15-2  Criteria for Return to Play after Concussion Grade 1

Grade 2

Grade 3

Athlete may return to play in 2 wk if asymptomatic at rest and with ­exertion for 7 days

Athlete may return to play in 1 mo if asymptomatic at rest and exertion for 7 days

Minimum of 1 mo; may return to play then if asymptomatic for 1 wk; consider terminating season

Terminate season; may return to play next season if asymptomatic

First Concussion

Athlete may return to play that day in select situations if clinical examination results are normal at rest and with exertion; otherwise return to play in 1 week Second Concussion

Return to play in 2 wk if asymptomatic for 1 wk

Third Concussion

Terminate season; may return to play next season if ­asymptomatic

Terminate season; may return to play next season if asymptomatic

Second-Impact Syndrome Second-impact syndrome is defined as a rapid brain swelling and herniation after a second head injury in a still symptomatic athlete.7,8 Between 1980 and 1998, the National Center for Catastrophic Sports Injury Research in Chapel Hill, North Carolina, identified 35 probable cases of second-impact syndrome in football players alone. Autopsy or surgery and MRI findings confirmed 17 of these cases. An additional 18 cases, although not conclusively documented with autopsy findings, most likely resulted from secondimpact syndrome. The syndrome, first described by Schneider in 1973, occurs when the athlete has had a head injury—often a concussion or worse, such as a cerebral contusion—and sustains a second injury before the symptoms associated with the first have cleared.9 The initial symptoms are typically postconcussive and may include visual, motor, or sensory changes as well as difficulty with thought and memory processes. Before these symptoms resolve, which may take days or weeks, the athlete returns to competition and receives a second blow to the head.

Physical Examination and Testing Although concussion is the most common athletic head injury, the leading cause of death from athletic head injury is intracranial hemorrhage. There are four types of ­ hemorrhage, of which every trainer and team physician must be aware: epidural, subdural, intracerebral, and subarachnoid. Because all four types of intracranial hemorrhage may be fatal, rapid and accurate initial assessment and appropriate follow-up are mandatory after an athletic head injury.

On-Field Evaluation Initial evaluation of a “down” athlete should include assessment of level of consciousness. If the athlete is unconscious, evaluation and treatment should follow the ABCs of ATLS resuscitation. If the patient is face down or lying on the side, he or she should be carefully logrolled into the supine position. Evaluation of the airway and assessment of respirations is the first concern. Whenever possible, the helmet should be left on and the facemask removed for airway access, with the neck manually immobilized. The helmet should be removed only in the rare instance that facemask removal

660

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

does not provide adequate airway access. The helmet, when left on, can be taped onto a fracture board to help stabilize the upper spine. Pupils should be assessed for size and symmetry. Response to verbal and noxious stimulus should be evaluated. The patient with prolonged loss of consciousness should be transported to a medical facility with neurosurgical services. CT scanning should be performed to evaluate possible intracranial hemorrhage. Contusion or injury of the brain, as seen with any intracerebral hematoma, usually causes headache and, often, an associated neurologic deficit, depending on the area of the brain involved. The injury may also precipitate a seizure. If a seizure occurs, it is important to logroll the patient onto the side. By this maneuver, any blood or saliva rolls out of the mouth and nose, and the tongue cannot fall back to obstruct the airway. A padded tongue depressor or oral airway can be inserted between the teeth. Fingers should not be inserted into the mouth of an athlete having a seizure because amputation can easily result. Traumatic seizures typically last 1 to 2 minutes. The athlete will then relax, and transportation to the nearest medical facility can occur. The awake athlete should be evaluated on the sidelines for continued signs and symptoms of head injury. The athlete’s level of consciousness, memory, speech, coordination, reflexes, vision, concentration, hearing, and gait should be evaluated. Symptoms such as headache, nausea and vomiting, confusion, dizziness, sensitivity to light, irritability, and amnesia should be evaluated. Symptoms should be assessed both at rest and with exertion. The athlete with persistent symptoms should not return to competition. The athlete with worsening headache, nausea and vomiting, or decreased level of consciousness should be transported to a medical facility for further evaluation.

Imaging If intracranial hemorrhage is suspected, CT scan is the modality of choice. CT scan will rapidly diagnose the location, type, and extent of bleeding and assist with operative decision making (Fig. 15-2). If there is a concern for carotid dissection or stroke, MRI with magnetic resonance angiogram can be obtained. Positron-emission tomography has been used to evaluate patients with persistent symptoms.

Treatment Options Nonoperative Initial treatment of a mild concussion requires the player to be removed from the game and observed on the bench. After a sufficient time (as short as 15 to 30 minutes), if the athlete has no headache, dizziness, or impaired concentration (including orientation to person, place, and time and full recall of events that occurred just before the injury), return to game may be considered. Before returning to the game, the player should be asymptomatic at rest and demonstrate movement with the usual dexterity and speed during exertion. If an athlete has any symptoms during rest or exertion, continued neurologic observation is essential. Various neuropsychological tests and tests of balance and coordination have been developed to help determine whether the athlete may safely return to competition.

These tests may play a role at the professional and perhaps collegiate level, but their complexity and expense may preclude routine use (see Table 15-2).

Operative Surgical intervention may be required following an athletic head injury that results in intracranial bleeding. A full discussion of operative indications is beyond the scope of this chapter, but it is important to know the types of injuries for which surgery may be necessary. The four general types of intracranial hematomas include epidural, subdural, intracerebral, and subarachnoid. Epidural Hematoma

An epidural hematoma is usually the most rapidly progressing intracranial hematoma. It is frequently associated with a fracture of the temporal bone and results from a tear of the middle meningeal artery supplying the cover (dura) of the brain. This hematoma accumulates between the skull and the covering of the brain and may reach a fatal size in 30 to 60 minutes. The athlete typically has a loss of consciousness followed by a lucid interval, but this does not always occur. Thus, the athlete may initially remain conscious or regain consciousness after the head trauma and then experience an increasing headache and a progressive decline in level of consciousness. This occurs as the clot accumulates and the intracranial pressure increases. If an epidural hematoma is present, it will almost always declare itself within 1 or 2 hours from the time of the injury. The brain substance is usually free from direct injury; thus, if the clot is promptly evacuated, full recovery is to be expected. Because this lesion may be rapidly fatal if missed, all athletes receiving a major head injury must be observed closely and frequently, preferably during the ensuing 24 hours. This observation should be done at a facility where full neurosurgical services are immediately ­available. Subdural Hematoma

With a subdural hematoma, the athlete usually does not regain consciousness, and the need for immediate neurosurgical evaluation is obvious. A subdural hematoma occurs between the brain surface and the dura, and so it is located directly on the brain. It is the most common fatal athletic head injury. A subdural hematoma usually results from a torn vein running from the surface of the brain to the dura or from diffuse injury to the surface of the brain. It may also result from a torn venous sinus or even a small artery on the surface of the brain. Unlike an epidural hematoma, a subdural hematoma is often associated with injury to the brain tissue. If the symptoms of a subdural hematoma are severe enough to necessitate emergent surgery, the mortality rate is high, not because of the clot itself, but because of the associated brain injury. Intracerebral Hematoma

Intracerebral hematoma is a third type of intracranial hemorrhage seen after head trauma. In this instance, the bleeding is into the brain substance itself, usually from a torn artery. Bleeding may also result from rupture of a congenital vascular lesion, such as an aneurysm or arteriovenous malformation. Intracerebral hematomas are not

Head Injuries

661

Figure 15-2   Computed tomographic scans of intracranial hemorrhage. A, Epidural hematoma,   B, subdural hematoma,   C, subarachnoid hematoma,   D, intracerebral hematoma.

A

B

C

D

usually associated with a lucid interval and may be rapidly progressive. Death occasionally occurs before the injured athlete can be transported to a hospital. Because of the intense reaction such a tragic event precipitates among fellow athletes, family, students, and community, it is imperative to obtain a complete autopsy to clarify the causative factors. The autopsy often reveals a congenital lesion that may indicate the cause of death was other than presumed and potentially unavoidable. Subarachnoid Hemorrhage

The final type of intracranial hemorrhage is subarachnoid, confined to the cerebrospinal fluid space along the surface of the brain. After head trauma, such bleeding is usually

the result of disruption of the tiny surface brain vessels and is analogous to a bruise, but it can also result from a ruptured cerebral aneurysm or arteriovenous malformation. As with the intracerebral hematoma, there is often brain swelling. Because bleeding is superficial, surgery is not usually required unless a congenital vascular anomaly is present. After intracranial hemorrhage, prophylactic anticonvulsant therapy is usually given for 1 week, and a long course may be indicated if the patient actually experienced a seizure. Because the chance of post-traumatic epilepsy is less than 10% with a concussion or contusion, anticonvulsant therapy is given only if late epilepsy ­actually occurs.

662

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

WEIGHING THE EVIDENCE Sports-related head injuries receive substantial public attention and are responsible for 70% of traumatic athletic deaths and 20% of permanent disability related to sports.10 Mild to moderate concussion has been recognized as an epidemic in sports, and it can have an impact on the scholastic performance of the athlete. Even a single, mild sports-related concussion can temporarily affect neuropsychological test responses.11 According to the National Center for Catastrophic Sports Injury Research, sports with the highest risk for head injury per 100,000 participants are football, gymnastics, and ice hockey. The absolute numbers of severe head injuries are highest from football because more than 10 times as many people play football as each of the other two sports. Box 15-2 lists some of the most hazardous sports for head and spine injury. Concussion is a common injury at the professional level. In the National Football League, it is estimated that 100 to 120 concussions occur per year, or about one every two to three games.12 In professional soccer, 52% of players have reported at least one concussion in their career.13 The National Hockey League has seen an increase in the number of concussions, with the rate from 1997 to 2002 more than triple that of the preceding decade.14 Multiple theories have been proposed for the increase, such as bigger, faster players, new equipment, and harder boards and glass. Fortunately the rate has reached a plateau since 1997, which was the year the National Hockey League instituted its concussion program. There is some evidence to suggest that younger athletes may be at higher risk for concussion. For example, the concussion rate for collegiate soccer players is estimated at 1 per 3000 athletic exposures, whereas the rate for high school players has been reported at 1 per 2000 exposures. Among Canadian amateur hockey players aged 15 to 20 years, 60% have reported sustaining a concussion during either a practice or a game. Authors have expressed

BOX 15-2  Sports with the Highest Risk  for Head Injury Auto racing Boxing Cycling Equestrian sports Football Gymnastics Hang-gliding Ice hockey Lacrosse Martial arts Motorcycling Parachuting Rugby Skiing Soccer (goalie) Track (pole vaulting)

A u t h o r ’ s P r e f e r r e d M e t h o d Athletes who have sustained a prolonged loss of consciousness or exhibit neurologic deficits should be triaged to a medical center for further evaluation. Abbreviated neurologic examinations such as the Glasgow Coma Scale are useful in predicting outcome after a severe head injury (Table 15-3).16 On arrival at a medical center, a full neurologic examination should be performed, including assessment of mental status, speech, memory, motor and sensory function, cranial nerve function, and reflexes (normal and abnormal). CT scan is helpful in evaluation of potential intracranial hemorrhage and skull fracture. MRI study will identify more diffuse injury such as diffuse axonal shear.17   TABLE 15-3  Glasgow Coma Scale Sign

Evaluation

Score

Eye opening (E)

Spontaneous To speech To pain None Obeys Localizes Withdraws Decorticate Decerebrate None Oriented Confused conversation Inappropriate words Incomprehensible sounds None

4 3 2 1 6 5 4 3 2 1 5 4 3 2

Best motor   response (M)

Verbal response (V)

Total EMV score by   adding best response   in each category.   Range from 3-15.

1

concern that injury during adolescence may impair the plasticity of the developing brain.15

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Postoperative care following a traumatic athletic head injury is the same as for any trauma patient. Whether an athlete should be allowed to return to competition after intracranial surgery is an area of controversy and beyond the scope of this chapter. The ultimate goal when discussing athletic head injuries is prevention. The increased attention on head injuries in sports has made athletes, trainers, physicians, and fans more aware of the potential dangers of brain injury in sports. Football was one of the first sports to focus on prevention of head and neck injury, as evidenced by the rule change in 1976 prohibiting initial contact with the head (spear tackling). The number of fatalities due to head injuries in football has declined from a high of 162 during the 10-year span of 1965 to 1974 to a low of 32 during the 10 years of 1985 to 1994.18

Head Injuries

BOX 15-3  Return to Play in Head Injury Because the subject does not lend itself to prospective, randomized studies, the guidelines for return to play after a traumatic brain injury are based largely on retrospective analysis and judgment. A primary goal is to avoid secondary injury such as a more severe concussion or worse yet a second-impact syndrome. It is for this reason that athletes who remain symptomatic after even a grade 1 concussion should not return to play. Another goal is to prevent the long-term effects of multiple minor traumatic brain injuries that can lead to permanent changes such as the classic description of dementia pugilistica seen in boxers. It is with these aims in mind that the guidelines in Table 15-2 were developed. Factors other than the concussion severity must be weighed in the return-to-play decision. The athlete’s concussion history, including total number, time between injuries, and severity of the blow causing the concussion are important factors. When making the return-to-play decision, one should consider all pieces of the concussion puzzle and when in doubt, err on the side of caution: “If in doubt, sit them out.”

S U G G E S T E D

R E A D I N G S

Cantu RC: Guidelines for return to contact sports after a cerebral concussion. Phys Sportsmed 14:76-79, 1986. Cantu RC, Mueller FO: Brain injury-related fatalities in American football, 19451999. Neurosurgery 52(4):846-853, 2003. Ghiselli G, Schaadt G, McAllister DR: On-the-field evaluation of an athlete with a head or neck injury. Clin Sports Med 22(3):445-465, 2003. Grindel SH: Epidemiology and pathophysiology of minor traumatic brain injury. Curr Sports Med Rep 2:18-23, 2003.

C

r i t i c a l

P

663

o i n t s

l Sporting activities account for about 20% of head injuries per year in the United States. l About 70% of traumatic athletic deaths result from head injury. l A concussion can occur without a loss of consciousness (grade 1). l Repetitive minor brain injury (concussion) can lead to permanent neurologic changes. l An athlete who returns to competition still symptomatic from a prior concussion may risk suffering a secondimpact syndrome. l The four types of intracranial hemorrhage include epidural, subdural, intracerebral, and subarachnoid. l Athletes with an epidural hematoma may have a brief lucid interval following initial loss of consciousness. l For the unconscious athlete, the ABCs of trauma care should be followed. l Rule changes have had a positive impact on the incidence of head and neck injuries in football.

Okonkwo DO, Stone JR: Basic science of closed head injuries and spinal cord injuries. Clin Sports Med 22(3):467-481, 2003. Schneider RC: Head and neck injuries in football: Mechanisms, treatment, and prevention. Baltimore, Williams & Wilkins, 1973.

R eferences Please see www.expertconsult.com

C H A P T E R

16

  

Spinal Injuries S ecti o n

A

Cervical Spine Injuries 1. Cervical Spine Injuries in the Adult Joseph S. Torg

This chapter presents guidelines for classification, evaluation, management, and return to play criteria for injuries that occur to the cervical spine and related neural structures as a result of participation in competitive and recreational activities. Although all athletic injuries require careful attention, the evaluation and management of cervical spine injuries should proceed with particular caution. The actual or potential involvement of the nervous system creates a high-risk situation in which the margin for error is low. An accurate diagnosis is imperative, but the clinical picture is not always representative of the seriousness of the injury at hand. An intracranial hemorrhage may present initially with minimal symptoms yet follow a precipitous downhill course, whereas a less severe injury, such as neurapraxia of the cervical spinal cord associated with alarming paresthesias and paralysis, will resolve swiftly and allow a return to activity. Although the more severe injuries are rather infrequent, this low incidence coincidentally results in little, if any, management experience for the on-site medical staff.

EMERGENCY MANAGEMENT There are several principles that should be considered by individuals responsible for athletes who may sustain injuries to the cervical spine.1,2 1. The team physician or the trainer should be designated as the person responsible for supervising on-the-field management of the potentially serious injury. This person is the “captain” of the medical team. 2. Previous planning must ensure the availability of all necessary emergency equipment at the site of potential injury. At a minimum, this should include a spine board, a stretcher, and equipment necessary for helmet removal and the initiation and maintenance of cardiopulmonary resuscitation. 3. Previous planning must ensure the availability of a properly equipped ambulance as well as a hospital equipped and staffed to handle emergency neurologic problems.

4. Previous planning must ensure the immediate availability of a telephone for communicating with the hospital emergency room, ambulance, and other responsible individuals in case of an emergency. Managing the unconscious or spine-injured athlete is a process that should not be done hastily or haphazardly. Being prepared to handle this situation is the best way to prevent actions that could convert a repairable injury into a catastrophe. Be sure that all the necessary equipment is readily accessible and in good operating condition and that all assisting personnel have been trained to use it properly. Onthe-job training in an emergency situation is inefficient at best. Everyone should know what must be done beforehand so that on a signal, the game plan can be put into effect. A means of transporting the athlete must be immediately available in a high-risk sport such as football and “on call” in other sports. The medical facility must be alerted to the athlete’s condition and estimated time of arrival so that adequate preparation can be made. The availability of the proper equipment is essential! A spine board is necessary and is the best means of supporting the body in a rigid position. It is essentially a fullbody splint. By splinting the body, the risk for aggravating a spinal cord injury, which must always be suspected in the unconscious athlete, is reduced. In football, appropriate instruments are also essential if it becomes necessary to remove the facemask. A telephone must be available to call for assistance and to notify the medical facility. Oxygen should be available and is usually carried by ambulance and rescue squads, although it is rarely required in an athletic setting. Rigid cervical collars and other external immobilization devices can be helpful if properly used. Manual stabilization of the head and neck is recommended if other means are not available. Properly trained personnel must know, first of all, who is in charge. Everyone should know how to perform cardiopulmonary resuscitation and how to move and ­transport the athlete. They should know where emergency ­ equipment 665

666

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

is located, how to use it, and the procedure for activating the emergency support system. Individuals should be assigned specific tasks beforehand, if possible, to prevent duplication of effort. Being well prepared helps alleviate indecisiveness and second-guessing. Prevention of further injury is the single most important objective. Do not take any action that could possibly cause further injury. The first step should be to immobilize the head and neck by supporting them in a stable position (Fig. 16A1-1). Then, in the following order, check for breathing, pulse, and level of consciousness. If the victim is breathing, simply remove the mouth guard, if present, and maintain the airway. It is necessary to remove the facemask only if the respiratory situation is threatened or unstable or if the athlete remains unconscious for a prolonged period. Leave the chin strap on. After it is established that the athlete is breathing and has a pulse, evaluate the neurologic status. The level of consciousness, the response to pain, the pupillary response, and any unusual posturing, flaccidity, rigidity, or weakness should be noted. At this point, simply maintain the situation until transportation is available or until the athlete regains consciousness. If the athlete is face down when the ambulance arrives, change his or her position to face up by logrolling the athlete onto a spine board. Gentle longitudinal traction should be exerted to support the head without attempting to correct alignment. Make no attempt to move the injured person except to transport him or her or to perform cardiopulmonary resuscitation if it becomes necessary. If the athlete is not breathing or stops breathing, the airway must be established. If face down, he or she must be turned to a face-up position. The safest and easiest way to

accomplish this is to logroll the athlete. In an ideal situation, the medical support team is made up of five members: the leader, who controls the head and gives the commands; three members to roll; and another to help lift and carry when it becomes necessary. If time permits and the spine board is on the scene, the athlete should be rolled directly onto it. Breathing and circulation are much more important at this point, however. With all medical support team members in position, the athlete is rolled toward the assistants—one at the shoulders, one at the hips, and one at the knees. They must maintain the body in line with the head and spine during the roll. The leader maintains immobilization of the head by applying slight traction and by using the crossed-arm technique. This technique allows the arms to unwind during the roll (Fig. 16A1-2).

A

B

A

C B Figure 16A1-1   A, Athletes with suspected cervical spine injury may or may not be unconscious. All who are unconscious, however, should be managed as though they had a significant neck injury. B, Immediate manual immobilization of the head and neck unit. First check for breathing. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

Figure 16A1-2   A, Logroll to a spine board. This maneuver requires four individuals: the leader to immobilize the head and neck and command the medical-support team, and the remaining three individuals positioned at the shoulders, hips, and lower legs. B, Logroll. The leader uses the crossed-arm technique to immobilize the head. This technique allows the leader’s arms to unwind as the three assistants roll the athlete onto the spine board. C, Logroll. The three assistants maintain body alignment during the roll. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

Spinal Injuries

A

667

B

Figure 16A1-3   Head and helmet must be securely immobilized. A, Remove cage-type masks by cutting the plastic loops with Dura shears, EMT scissors, or Trainer’s Angel. Make the cut on the side of the loop away from the face. B, Remove the entire mask from the helmet so that it does not interfere with further resuscitation efforts. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

The following guidelines are outlined by Kleiner and colleagues3: Heavy persons, including many athletes, can be handled more efficiently with a six-plus-person lift; this is also preferred for suspected spine injuries. The Inter-Association Task Force recommends that the six-plus-person lift be used along with a scoop stretcher whenever possible. In the athletic arena, there are usually a sufficient number of certified athletic trainers, physicians, and EMS personnel on hand to effectively administer the six-plus-person lift. For the six-plus-person lift, rescuer 1 immobilizes the neck. The rescuer’s hands are placed on the athlete’s shoulders (under the shoulder pads) with the thumbs pointed away from the athlete’s face. The athlete’s head will then be resting between the rescuer’s forearms. The other six rescuers position themselves along the athlete’s sides: one on each side of the chest, pelvis, and legs. The hands are slid under the athlete and equipment, if any, to provide a firm, coordinated lift. To lift, rescuer 1 gives the command “prepare to lift; lift. ” The assistants lift the athlete 4 to 6 inches off the ground. It is imperative to maintain a coordinated lift and to prevent any movement of the spine. One of the rescuers at the thigh level must control the legs with his or her arms toward the feet so the splint can be slid into place from the foot end. After the splint is in place, while positions are maintained, rescuer 1 gives the command “prepare to lower; lower, ” and the athlete is lowered onto the splint. In the case of larger athletes, as many as 10 individuals should participate in the lift, with one on each side of the chest and pelvis, two at the legs, one at the head, and one with the splint. The Inter-Association Task Force does not recommend the use of fewer than fourplus-persons to lift athletes suspected of having a spinal injury, even smaller athletes and children, in part due to the weight of the athlete while wearing protective equipment.3

The facemask should be removed from the helmet as quickly as possible any time a player is suspected of having a spinal injury, even if still conscious and regardless of respiratory status. The type of mask that is attached to the helmet determines the method of removal. Bolt cutters are used with the older single- and double-bar masks. The newer masks that are attached with plastic loops should be removed by cutting the loops with an instrument capable of cutting through the newer loop straps made of harder plastic, such as Dura shears, EMT scissors, or the Trainer’s Angel. Remove the entire mask so that it does not interfere with further rescue efforts (Fig. 16A1-3). Once the mask has been removed, initiate rescue breathing following the current standards of the American Heart Association. After the athlete has been moved to a face-up position, quickly evaluate breathing and pulse. If there is still no breathing or if breathing has stopped, the airway must be established. The jaw thrust technique is the safest first approach to opening the airway of a victim who has a ­ suspected neck injury because in most cases it can be accomplished by the rescuer grasping the angles of the victim’s lower jaw and lifting with both hands, one on each side, displacing the mandible forward while tilting the head backward. The rescuer’s elbows should rest on the surface on which the victim is lying (Fig. 16A1-4). If the jaw thrust is not adequate, the head tilt–jaw lift should be substituted. Care must be exercised not to overextend the neck. The fingers of one hand are placed under the lower jaw on the bony part near the chin and are lifted to bring the chin forward, supporting the jaw and helping tilt the head back. The fingers must not compress the soft tissue under the chin, which might obstruct the airway. The other hand presses on the victim’s forehead to tilt the head back (Fig. 16A1-5). The transportation team should be familiar with handling a victim with a cervical spine injury, and they should be receptive to taking orders from the team physician or

668

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16A1-4   Jaw thrust maneuver for opening the airway of a victim with a suspected cervical spine injury.

the trainer. It is extremely important not to lose control of the care of the athlete; therefore, be familiar with the transportation crew that is used. In an athletic situation, arrangements with an ambulance service should be made ahead of time. An appreciation of the controversy that currently exists between emergency medicine physicians and technicians on one hand and team physicians and athletic trainers on the other regarding helmet removal is in order. Existing emergency medical services guidelines mandate removal of protective headgear before transport of an individual suspected of having a cervical spine injury to a fixed medical installation. These guidelines were implemented with motorcycle helmets in mind to facilitate both airway accessibility and application of cervical spine immobilizing devices. Clearly, such a procedure contradicts the longstanding principle adhered to by team physicians and athletic trainers of leaving the helmet in place on the football player suspected of having a cervical spine injury until he or she is transported to a definitive medical facility. It must be emphasized that this particular problem is of more than academic interest. Specifically, there have been occasions in which emergency medical technicians under the direction of emergency room physicians unfamiliar with the nuances of the relationship between helmet, shoulder pads, and the injured cervical spine have precipitated turf battles by refusing to move the injured player before helmet removal. Again, such episodes represent more than an honest difference of opinion. Such episodes are clearly detrimental to the health and wellbeing of the injured player. It is my view that removal of the football helmet and shoulder pads on site exposes the potentially injured spine to both unnecessary and awkward manipulation and disruption of the immobilizing capacity of the helmet and shoulder pads. Also, removal of the helmet alone subjects a potentially unstable spine to hyperlordotic deformity. I agree with the National Collegiate Athletic Association Guidelines for helmet removal.4 Unless there are

Figure 16A1-5   Head tilt—jaw lift maneuver for opening the airway. This is used if jaw thrust is inadequate or if a helmet is being worn.

special circumstances, such as respiratory distress coupled with an inability to access the airway, the helmet should never be removed during the prehospital care of the athlete with a potential head and neck injury unless the following conditions are present: 1. The helmet does not hold the head securely, such that immobilization of the helmet does not immobilize the head. 2. The design of the sport helmet is such that even after removal of the facemask, the airway cannot be controlled and ventilation cannot be provided. 3. After a reasonable time, the facemask cannot be removed. 4. The helmet prevents immobilization for transportation in an appropriate position. When helmet removal is necessary in any setting, it should be performed only by personnel trained in this procedure. If removal of the helmet is needed to initiate ­treatment or to obtain special radiographic studies, a specific protocol needs to be followed. With the head, the neck, and the helmet manually stabilized, the chin strap can be cut. While stability is being maintained, the cheek pads can be removed by slipping the flat blade of a screwdriver or bandage scissors under the pad snaps and above the inner surface of the shell. One individual provides manual stability of the chin and the neck, and the other workers stabilize the head by placing their thumbs or index fingers into the ear holes on both sides. By pulling both laterally and longitudinally, the helmet shell can be spread and eased off. If a rocking motion is necessary to loosen the helmet, the head and neck unit must not be allowed to move. Those individuals participating in this important maneuver must proceed with caution and must coordinate every move.

Spinal Injuries

A

669

B

Figure 16A1-6   A, Four members of the medical support team lift the athlete on the command of the leader. B, The leader maintains manual immobilization of the head. The spine board is not recommended as a stretcher. An additional stretcher should be used for transporting the patient over long distances. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

Supporting the concept of leaving the helmet on is the work of Swenson and associates5 and Gastel and colleagues.6 Swenson and associates studied sagittal cervical alignment in live subjects with various combinations of helmets, shoulder pads, and no equipment. They concluded that football players with a potential cervical spine injury should be immobilized for transport with both helmet and shoulder pads left in place, thereby maintaining the neck in a position most closely approximating normal. Gastel and colleagues performed a similar study using stable and surgically destabilized cadaver spines. They concluded that to maintain a neutral position and minimize secondary injury to the cervical neural elements, both the helmet and the shoulder pads should be either left on or removed in the emergency setting. Lifting and carrying the athlete requires five individuals: four to lift, and a leader to maintain immobilization of the head. The leader initiates all actions with clear, loud verbal commands (Fig. 16A1-6). The same guidelines apply to the choice of a medical facility as to the choice of an ambulance: Be sure it is equipped and staffed to handle an emergency head or neck injury. There should be a neurosurgeon and an orthopaedic surgeon to meet the athlete on arrival. Radiographic facilities should be standing by. Once the athlete is in a medical facility and permanent immobilization measures have been instituted, the helmet can be removed. The chin strap may now be unfastened and discarded. The following protocol is recommended3,4: The helmet should be removed in a controlled environment after radiographs have been obtained and only by qualified medical personnel with training in equipment removal. Helmet removal should never be attempted without thorough communication among all involved parties. One person should stabilize the head, neck, and helmet while another person cuts the chin strap. Accessible internal helmet padding, such as cheek pads, should be removed, and air padding should be deflated before removal of the helmet, while a second assistant manually stabilizes the chin and back of the neck, in a cephalad direction, making sure to maintain the athlete’s position. The pads are removed

through the insertion of a tongue depressor or a similar stiff, flat-bladed object between the snaps and helmet shell to pry the cheek pads away from their snap attachment. If an air cell–padding system is present, deflate the air inflation system by releasing the air at the external port with an inflation needle or large-gauge hypodermic needle. The helmet should slide off the occiput with slight forward rotation of the helmet. In the event the helmet does not move, slight traction can be applied to the helmet, which can then be gently maneuvered anteriorly and posteriorly, although the head-neck unit must not be allowed to move. The helmet should not be spread apart by the ear holes3,4 because this maneuver only serves to tighten the helmet on the forehead and occiput region (Fig. 16A1-7). Despite the advent of such high-technology imaging modalities as computed tomography (CT) and magnetic resonance imaging (MRI), the initial radiographic examination of a patient with suspected or actual cervical spine trauma remains a routine radiographic examination. The preliminary study, performed while immobilization of the head, the neck, and the trunk is maintained, includes an anteroposterior and lateral examination of vertebrae C1 to C7. If major fracture, subluxation, dislocation, or evidence of instability is not evident, the remainder of the routine examination, including open mouth and oblique views, should be obtained. Depending on the neurologic and comfort status of the patient, lateral flexion and extension views should be obtained at some point. CT and MRI may provide more detailed information; however, horizontally oriented fractures and subtle subluxations are best identified on the routine radiographs. The choice of imaging technique depends on the results of the routine examination, the neurologic status of the patient, the preference of the responsible physician, and the availability of the imaging modalities.

THE AMBULATORY PATIENT Fortunately, it is the rare athlete with a cervical spine injury who presents with neurologic impairment. Important physical findings indicative of injury in an individual

670

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 16A1-7   A, The helmet should be removed only when permanent immobilization can be instituted. The helmet may be removed by detaching the chin strap, removing cheek pads, deflating air padding, and gently pulling the helmet off in a straight line with the cervical spine. B, The head must be supported under the occiput during and after removal of the helmet. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

­ ithout neurologic findings are (1) presence of a wry neck w or torticollis posture; (2) limitation of cervical motion; and (3) presence of paravertebral muscle atrophy in subacute and chronic cases. An individual presenting with a history of trauma and one or more of these findings requires a careful neurologic examination and appropriate imaging studies.

CERVICAL SPINE INJURIES Athletic injuries to the cervical spine may involve the bony vertebrae, the intervertebral disks, the ligamentous supporting structures, the spinal cord, the roots, the peripheral nerves, or any combination of these structures. The panorama of injuries observed runs the spectrum from the “cervical sprain syndrome ” to fracture-dislocations with permanent quadriplegia. Fortunately, severe injuries with neural involvement occur infrequently. Those responsible for the emergency and subsequent care of the athlete with a cervical spine injury should possess a basic understanding of the variety of problems that can occur. The various athletic injuries to the cervical spine and related structures are the following:   1. Nerve root–brachial plexus injury   2. Stable cervical sprain   3. Muscular strain   4. Nerve root–brachial plexus axonotmesis   5. Intervertebral disk injury (narrowing-herniation) ­without neurologic deficit   6. Stable cervical fractures without neurologic deficit   7. Subluxations without neurologic deficit   8. Unstable fractures without neurologic deficit   9. Dislocations without neurologic deficit 10. Intervertebral disk herniation with neurologic deficit 11. Unstable fracture with neurologic deficit 12. Dislocation with neurologic deficit 13. Quadriplegia 14. Death

Criteria for return to contact activities after congenital and traumatic problems of the cervical spine are included at the end of this chapter.

Nerve Root–Brachial Plexus Injury The most common and poorly understood cervical injury is pinch-stretch neurapraxia of the nerve roots and the brachial plexus (Fig. 16A1-8).7 Typically, after an impact involving the head, neck, or shoulder, a sharp burning pain is experienced in the neck on the involved side that may radiate into the shoulder and down the arm to the hand. There may be associated weakness and paresthesia in the involved extremity lasting several seconds to several minutes. Characteristically, there is weakness of shoulder abduction (deltoid), elbow flexion (biceps), and external humeral rotation (spinatus). The key to the nature of this lesion is its short duration and the presence of a full, pain-free range of neck motion. Although most of these injuries are short lived, they are worrisome because of the occasional plexus axonotmesis that occurs. The youngster whose paresthesia completely abates and who demonstrates full muscle strength in the intrinsic muscles of the shoulder and upper extremities, and who, most important, has a full, pain-free range of cervical motion may return to his or her activity.8,9 The “burner syndrome ” has been attributed to different mechanisms. Bateman10 believed that root lesions were rarely involved in injuries to athletes, whereas peripheral nerve lesions were common. He recognized that direct blows as well as other mechanisms could result in varying peripheral nerve injuries about the shoulder. Chrisman11 described lateral neck flexion away from the involved side resulting in cervical sprain and traction injury to the cervical nerve roots. Rockett12 reported operative findings in patients with persistent burners. He noted scarring of the C5 and C6 nerve roots at their point of emergence from the vertebra between the anterior and middle scalene. He suggested that repetitive tightening of the scalenes causes trauma to the nerve roots with resultant scarring.

Spinal Injuries

671

C4 C5 Dorsal scapular n. (Rhomboid minor and major)

C5 To longus and scalene m.

NK

C5-C6 Suprascapular n. (Supraspinatus infraspinatus)

ER

P UP

C5-C6 Upper subscapular n. (Subscapularis)

C7

to n. vius la c K b su RUN T

LE

DD

MI

ER

LA TE

RA L

W LO R

C

D

R

O

Phrenic n. (diaphragm) C8

K

N

CO

RD

C5-C6 Lower subscapular n. (Subscapularis teres major)

C5-C6 Musculocutaneous n. (Coracobrachialis biceps brachialis)

C6

U TR

U TR

T1

Long thoracic n. (Serratus anterior)

IO

ER

ST

C5-C6 Axillary n. (Deltoid teres minor)

Lateral pectoral n. (Pectoralis major)

D

PO

L

IA

ED

R CO

M

Medial pectoral n. (Pectoralis major and minor) Thoracodorsal n. (Latissimus dorsi) (C6-C8)

Median n. C6-T1 (C7-T1)

Medial brachial cutaneous Medial antebrachial cutaneous

no motors

Ulnar n. Radial n. (C6-T1) Figure 16A1-8   Diagram of the brachial plexus demonstrating the location of Erb’s point (arrow). Presumably, brachial plexus stretch injuries result from traction of the plexus at this point. (From Torg JS: Athletic Injuries to the Head, Neck and Face, 2nd ed. St. Louis, Mosby-Year Book, 1991.)

Clancy13,14 has suggested a more distal injury. He believes that burners are brachial plexus injuries. Electrodiagnostic evidence indicated that plexus axonotmesis involved only the upper trunk. He noted different mechanisms of injury and suggested that the point of plexus injury may vary depending on neck, arm, and shoulder position. Neck hyperextension, shoulder depression, neck hypertension with lateral bend to the side of injury, and contralateral neck flexion with ipsilateral shoulder depression were some mechanisms reported. Clancy13,14 also recommended classifying these injuries based on the staging system of Seddon.15 Neurapraxia, the mildest form of injury, represents a reversible aberration in axonal function. Focal demyelinization can occur, producing an electrophysiologic conduction block or conduction slowing. Complete recovery usually occurs immediately or within a maximum of 2 weeks. Axonotmesis is an injury in which the axon and the myelin sheath are disrupted, but the epineurium remains intact. Wallerian degeneration occurs distal to the point of injury; functional recovery may occur, but it can be incomplete and unpredictable. The most severe injury, neurotmesis, is rarely seen in athletes

and results in complete disruption of the nerve. Prognosis is poor, and generally recovery does not occur. Clancy13,14 has defined cervical nerve pinch syndrome as those injuries that recover within 2 weeks; most likely, these represent neurapraxia. The term brachial plexus injury is reserved for injuries with weakness or sensory changes lasting longer than 2 weeks. In our opinion, the term brachial plexus injury should be reserved for anatomic localization. Many of these injuries are mixed lesions, and classification by Seddon’s system15 serves mainly to aid in describing a potential recovery course and prognosis. Robertson16 indicated that brachial plexus injury at Erb’s point is most likely to be a stretch injury. He reported that all patients became symptomatic after contact, causing ipsilateral shoulder depression and lateral neck flexion to the opposite side. Kelly and coworkers17,18 investigated the relationship between burners and cervical stenosis in younger patients aged 15 to 18 years. A review of 69 cervical spine radiographs demonstrated a significant decrease in the Torg ratio (the ratio of the anteroposterior diameter of the spinal cord to the anteroposterior diameter of the vertebral body)

672

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

as compared with the control group. They hypothesized that developmental cervical stenosis predisposes an athlete to experience burners as a result of concomitant foraminal narrowing with nerve root compression. Meyer19 studied 40 patients with “stingers” at the University of Iowa from 1987 to 1991. The mechanism was reported as extension-compression in 34 subjects and brachial plexus stretch in 6 subjects. Cervical spine radiographs were analyzed and compared with a control group of asymptomatic football players. Players with stingers demonstrated a statistically significant incidence of spinal stenosis of at least one cervical segment as defined by a Torg ratio of less than 0.8. For the stinger group, 47.5% had a ratio less than 0.8 for at least one level, compared with 25.1% in the asymptomatic group. No patient in the group with brachial plexus stretch injuries had a Torg ratio of less than 0.8. Meyer concluded that there is a relationship between cervical stenosis and the occurrence of nonparalyzing extension-compression injuries. It is clear that the typical burner can be caused by various injury mechanisms. History, physical examination, and appropriate diagnostic studies help differentiate between brachial plexus and cervical nerve root lesions. Brachial plexus injuries are more likely to occur in younger patients with less well-developed neck musculature. Usually, these are traction injuries resulting from lateral neck flexion away from the involved area and shoulder depression to the side of involvement. Neck pain can be present but is usually not a prominent feature. When pain is present, cervical spine radiographs are necessary. Typically, pain and paresthesias involving the arm and the shoulder are transient. On examination, Spurling’s test result is negative (Fig. 16A1-9). Weakness typically involves the deltoid, the spinati, and the biceps and might not be evident initially on clinical examination, making a follow-up visit necessary.

Figure 16A1-9   Spurling’s maneuver. The examiner applies pressure to the head, forcing the cervical spine into extension and lateral flexion toward the symptomatic side. This reproduces the pathomechanics of those injuries resulting from compression of the cervical nerve roots or the dorsal root ganglion in the involved intervertebral foramen.

Root lesions result from compression of the nerve root or dorsal root ganglion in the intervertebral foramen and are generally associated with radiologic evidence of cervical disk disease and developmental stenosis. In football players, these injuries usually occur when the player reaches the college or professional level. Hyperextension with lateral neck flexion is the common mechanism of injury. Neck pain and a decreased cervical range of motion may be present. Spurling’s test result is positive. Plain radiographic findings may be normal or may demonstrate loss of normal cervical lordosis and the changes of degenerative disk disease. MRI is indicated in patients with a persistent neurologic deficit and prolonged or recurrent symptoms and will demonstrate either acute disk herniation or degenerative disk disease with asymmetric disk bulging. In our experience, patients often have developmental spinal stenosis, degenerative disk disease, and asymmetric disk bulging that results in root irritation with cervical hyperextension. Persistence of paresthesia, weakness, or limitation of cervical motion requires that the individual be protected from further exposure and that he or she undergo neurologic, electromyographic, and roentgenographic evaluation. Persistent or recurrent episodes require a complete neurologic and radiographic or other imaging work-up. If routine roentgenographic films of the cervical spine are negative and a preganglionic root lesion is suspected, MRI, plain myelography, or CT myelography should be considered. Disk herniation, foraminal narrowing, and extradural intraspinal masses should be considered in the differential diagnosis. A complete electromyographic examination, including both nerve conduction studies and a needle electrode examination, may be helpful. These studies should be delayed for 3 to 4 weeks from the time of the initial injury. Nerve conduction studies should include both routine conduction and sensory nerve action potential evaluations. Electrode evaluation of the cervical spine musculature will differentiate between preganglionic root injuries and plexus disorders. Initial management must be directed at evaluation of the cervical spine, the shoulder girdle, the affected upper extremity, and the peripheral nervous system. The first obligation of the physician is to rule out a serious cervical spine injury. A patient history of bilateral symptoms or symptoms including the lower extremities should alert the physician to the possibility of cord neurapraxia, cervical spine fracture, or ligamentous injury. In this instance, the spine should be immobilized until injury is ruled out. If a player complains of neck pain, a complete cervical spine evaluation is mandatory, including radiographic examination. Characteristically, the signs and symptoms are transient and resolve within minutes. In athletes whose pain and paresthesias abate, a normal neurologic examination is required, and most important, a full pain-free range of ­ cervical motion must be seen before return to activity is allowed. Also, players must demonstrate normal strength on clinical examination before they return to participation. Those patients who have recurrent symptoms without weakness require careful follow-up; continued symptoms associated with weakness preclude further ­athletic ­participation.

Spinal Injuries

Figure 16A1-10   Crucial in the effective management of the athlete with recurrent burners is the implementation of an aggressive, year-round neck and shoulder muscle strengthening program. Effective are variable-resistance isotonic “neck machines.”

In brachial plexus injuries, prevention is based on an aggressive neck and shoulder strengthening program (Fig. 16A1-10). Neck rolls, or devices such as the cowboy collar, and high-profile shoulder pads also help prevent injuries by limiting the extent of lateral flexion and extension (Fig. 16A1-11).

A

673

Electrodiagnostic studies may be helpful but are not mandatory in the management of burners secondary to brachial plexus injury. Speer20 demonstrated that although there was no correlation between initial physical findings and the results of electrodiagnostic testing, evidence of muscular weakness 72 hours after the injury did correlate with positive electromyographic results. Bergfeld21 reported that electromyographic changes continue to appear long after weakness has apparently been resolved according to clinical examinations; ­ therefore, abnormal electromyographic findings should not be used as a criterion for exclusion from athletic participation. In summary, the burner pain syndrome or stinger results from either of two distinct injury patterns: traction to the brachial plexus or compression of the cervical nerve roots. Brachial plexus injuries are typically traction neurapraxias occurring in younger athletes as a result of shoulder depression and lateral neck flexion away from the side of injury. Cervical root injuries typically occur in older players. They are hyperextension injuries and are associated with degenerative disk changes, often in combination with developmental cervical stenosis.22 Criteria for return to athletic participation include absence of symptoms, normal strength, and painless, full range of motion of the cervical spine. Players who experience one or more burners should wear appropriate neck rolls or a cowboy collar to prevent extreme hyperextension and lateral bending of the cervical spine (Fig. 16A1-12). A year-round neck and shoulder muscle strengthening program will aid in the prevention of the burner syndrome.12

Acute Cervical Sprain Syndrome An acute cervical sprain is a collision injury frequently seen in contact sports. The patient complains of having “jammed” his or her neck, with subsequent pain localized to the cervical area. Characteristically, the patient pre­ sents with limitation of cervical spine motion but without radiation of pain or paresthesia. Neurologic examination is negative, and roentgenograms are normal.

B

Figure 16A1-11   Frontal (A) and lateral (B) views of the cowboy collar. This device, which is worn under the shoulder pads, effectively limits the extremes of extension and lateral bending of the cervical spine.

674

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16A1-12   The combined action of the football helmet, the cowboy collar, and the shoulder pads effectively limits the extremes of lateral bend of the neck.

Stable cervical sprains and strains eventually resolve with or without treatment. Initially, the presence of a serious injury should be ruled out by performing a thorough neurologic examination and determining the range of cervical motion. Range of motion is evaluated by having athletes perform the following actions: actively nod their head, touch their chin to their chest, extend their neck maximally, touch their chin to their left shoulder, touch their chin to their right shoulder, touch their left ear to their left shoulder, and touch their right ear to their right shoulder. If the patient is unwilling or unable to perform these maneuvers actively while standing erect, proceed no further. The athlete with less than a full, pain-free range of cervical motion, persistent paresthesia, or weakness should be protected and excluded from activity. Subsequent evaluation should include appropriate roentgenographic studies, including flexion and extension views to demonstrate fractures or instability. In general, treatment of athletes with “cervical sprains” should be tailored to the severity of the injury. Immobilizing the neck in a soft collar and using analgesics and antiinflammatory agents until there is a full, spasm-free range of neck motion is appropriate. Individuals with a history of collision injury, pain, and limited cervical motion should have routine cervical spine roentgenograms. Also, lateral flexion and extension roentgenograms are indicated after the acute symptoms have subsided. Marked limitation of cervical motion, persistent pain, or radicular symptoms or findings may require MRI to rule out intervertebral disk injury.

Intervertebral Disk Injuries Acute herniation of a cervical intervertebral disk associated with neurologic findings and occurring as an isolated entity is rare in the athlete. Acute onset of transient quadriplegia in an athlete who has sustained head impact but has negative cervical spine roentgenographic findings

should prompt consideration of an acute rupture of a cervical intervertebral disk. The syndrome of acute anterior spinal cord injury, as described by Schneider,23,24 may be observed in individuals with instability associated with acute disk herniation: “The acute anterior cervical spinal cord injury syndrome may be characterized as an immediate acute paralysis of all four extremities with a loss of pain and temperature to the level of the lesion, but with preservation of posterior column sensation of motion, position, vibration and part of touch.”23,24 The pressure of the disk is exerted on the anterior and lateral columns, whereas the posterior columns are protected by the denticulate ligaments. MRI or a CT myelogram should be performed to substantiate the diagnosis. Anterior diskectomy and interbody fusion for a patient with neurologic involvement or persistent disability because of pain should be considered. Albright and colleagues25 studied 75 University of Iowa freshmen football recruits who had had roentgenograms of their cervical spines after playing football in high school but before playing in college. Of this group, 32% had one or more of the following: “occult” fracture, vertebral body compression fracture, intervertebral disk-space narrowing, or other degenerative changes. Of this group, only 13% admitted having a positive history of neck symptoms. The development of early degenerative changes or intervertebral disk-space narrowing in this group was attributed to the effect of repetitive loading on the cervical spine as a result of head impact from blocking and tackling. Acute and chronic cervical intervertebral disk injury without frank herniation or neurologic findings occurs with considerable frequency in the athlete. Associated with a history of injury are neck pain and limited cervical spine motion. Roentgenograms may demonstrate disk-space narrowing and marginal osteophytes. Magnetic resonance imaging frequently demonstrates disk bulge without herniation. In general, management is conservative; permission to engage in activity is withheld until the youngster is asymptomatic and has a full range of cervical spine motion.

Cervical Vertebral Subluxation without Fracture Axial compression-flexion injuries incurred by striking an object with the top of the helmet can result in disruption of the posterior soft tissue supporting elements with angulation and anterior translation of the superior cervical vertebrae. Fractures of the bony elements are not demonstrated on roentgenograms, and the patient has no neurologic deficit. Flexion-extension roentgenograms demonstrate instability of the cervical spine at the involved level, manifested by motion, anterior intervertebral disk-space narrowing, anterior angulation and displacement of the vertebral body, and fanning of the spinous processes. Demonstrable instability on lateral flexion-extension roentgenograms in young, vigorous individuals requires aggressive treatment. When soft tissue disruption occurs without an associated fracture, it is likely that instability will result despite conservative treatment. When anterior subluxation greater than 20% of the vertebral body is due to disruption of the posterior supporting structures, posterior cervical fusion is recommended.

Spinal Injuries

Cervical Fractures or Dislocations: General Principles Fractures or dislocations of the cervical spine may be stable or unstable and may or may not be associated with neurologic deficit. When fracture or disruption of the soft tissue supporting structure immediately violates or threatens to violate the integrity of the spinal cord, implementation of certain management and treatment principles is imperative. These include the following: 1. Protection of the cord from further injury 2. Expeditious reduction 3. Attainment of rapid and secure stability 4. Implementation of an early rehabilitation program The first goal is to protect the spinal cord and the nerve roots from injury through mismanagement. It has been estimated that many neurologic deficits occur after the initial injury. That is, if a patient with an unstable lesion is carelessly manipulated during transportation to a medical facility or subsequently managed inappropriately, further encroachment on the spinal cord can occur. Second, once appropriate roentgenograms have been obtained and qualified orthopaedic and neurosurgical personnel are available, the malalignment of the cervical spine should be reduced as quickly and gently as possible. This will effectively decompress the spinal cord. When dislocation or anterior angulation and translation are demonstrated roentgenographically, immediate reduction is attempted with skull traction using Gardner-Wells tongs. These tongs can be easily and rapidly applied under local anesthesia without shaving the head in the emergency room or in the patient’s bed. Because these tongs are spring-loaded, it is not necessary to drill the outer table of the skull for their application. The tongs are attached to a cervical traction pulley, and weight is added at a rate of 5 pounds per disk space, or 25 to 40 pounds for a lower cervical injury. Reduction is attempted by adding 5 pounds every 15 to 20 minutes and is monitored by lateral roentgenograms. Unilateral facet dislocations, particularly at the C3-C4 level, are not always reducible by using skeletal traction. In such instances, closed skeletal or manipulative reduction under nasotracheal anesthesia may be necessary. The expediency of early reduction of cervical dislocations must be emphasized.26 It has been proposed that the presence of a bulbocavernous reflex indicates that spinal shock has worn off and that except for recovery of an occasional root at the site of the injury, neither motor nor sensory paralysis will be resolved regardless of treatment. The bulbocavernous reflex is produced by pulling on the urethral catheter. This stimulates the trigone of the bladder, producing a reflex contraction of the anal sphincter around the examiner’s gloved finger. Although the presence of a bulbocavernous reflex is generally a sign that there will be no further neurologic recovery below the level of injury, this is not always true. The presence of this reflex does not give the clinician license to handle the situation in an elective fashion. Malalignments and dislocations of the cervical spine associated with quadriparesis should be reduced as quickly as possible, by

675

whatever means necessary, if maximal recovery is to be expected. In most instances in which a vertebral body burst fracture is associated with anterior compression of the cord, decompression is logically effected through an anterior approach with interbody fusion. Likewise, intervertebral disk herniation with cord involvement is best managed through anterior diskectomy and interbody fusion. In patients with cervical fractures and dislocations, posterior cervical laminectomy is indicated only rarely when excision of foreign bodies or bony alignment of the spine is the most effective method of decompression of the cervical cord. Indications for surgical decompression of the spinal cord have been delineated. A documented increase in neurologic signs is the clearest mandate for surgical decompression. Further observation, expectant waiting, and procrastination in this situation are contraindicated. Persistent partial cord or root signs, with objective evidence of mechanical compression, are also an indication for surgical intervention. The use of parenteral corticosteroids to decrease the inflammatory reactions of the injured cord and the surrounding soft tissue structures is indicated in the management of acute cervical spinal cord injuries. The efficacy of methylprednisolone in improving neurologic recovery when given in the first 8 hours has been recently demonstrated. The recommended regimen is a bolus of 30 mg/kg of body weight of methylprednisolone administered intravenously followed by an infusion of 5.4 mg/kg/hour for 23 hours. It is essential that the regimen be started within 8 hours of the injury. Although the results of the National Acute Spinal Cord Studies have been critical of this regimen, it remains in keeping with the standard of care.27,28 The third goal in managing fractures and dislocations of the cervical spine is to effect rapid and secure stability to prevent residual deformity and instability with associated pain as well as to prevent the possibility of further trauma to the neural elements. White and coworkers recognized that the literature is neither always clear nor consistent in describing what constitutes an unstable cervical spine.29,30 Using fresh cadaver specimens, they performed load displacement studies on sectioned and unsectioned two-level cervical spine segments to determine the horizontal translation and rotation that occurred in the sagittal plane after each ligament was transected. The experiments constituted a quantitative biomechanical analysis of the effects of destroying ligaments and facets on the stability of the cervical spine below C2 in an attempt to determine cervical stability. The express purpose of the study was to establish indications for surgical treatment to stabilize the spine. Although the intent of the study was to define clinical instability to formulate treatment standards and was not intended to establish criteria for a return to contact athletics, it appears that their findings are relevant to the latter issue. White and colleagues described clinical stability as the ability of the spine to limit its patterns of displacement of physiologic loads to prevent damage or irritation of the spinal cord or the nerve roots.29,30 They delineated four important findings. First, in sectioning the ligaments, small increments of change in stability occur, followed without warning by sudden, complete disruption of the spine under stress. Second, removal of the facets alters the motion

676

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

s­ egment such that in flexion there is less angular displacement and more horizontal displacement. Third, the anterior ligaments contribute more to stability in extension than the posterior ligaments, and in flexion, the posterior ligaments contribute more than the anterior ligaments. The most relevant finding from the standpoint of criteria for return to contact sports is the fourth one. The adult cervical spine is unstable, or on the brink of instability, when one or more of the following conditions is present: (1) all the anterior or all the posterior elements are destroyed or are unable to function; (2) more than 3.5 mm of horizontal displacement of one vertebra exists in relation to an adjacent vertebra, measured on lateral roentgenograms (resting or flexionextension); or (3) there is more than 11 degrees of rotation difference compared with that of either adjacent vertebra measured on a resting lateral or flexion-extension roentgenogram (Figs. 16A1-13 and 16A1-14). The method of immobilization depends on the postreduction status of the injury. The literature has delineated the indications for using nonsurgical and surgical methods to achieve stability.31,32 These concepts for managing ­cervical spine fractures and dislocations may be summarized as follows: 1. Patients with stable compression fractures of the vertebral body, undisplaced fractures of the lamina or lateral masses, or soft tissue injuries without detectable

­ eurologic deficit can be adequately treated with tracn tion and subsequent protection with a cervical brace until healing occurs. 2. Stable, reduced facet dislocation without neurologic deficit can also be treated conservatively with a halo jacket brace until healing has been demonstrated by negative lateral flexion-extension roentgenograms. 3. Unstable cervical spine fractures or fracture-dislocations without neurologic deficit may require either surgical or nonsurgical methods to ensure stability. 4. Absolute indications for surgical stabilization of an unstable injury without neurologic deficits are late instability after closed treatment and flexion-rotation injuries with unreduced locked facets. 5. Relative indications for surgical stabilization in patients with unstable injuries without neurologic deficit are anterior subluxation greater than 20%, certain atlantoaxial fractures or dislocations, and unreduced vertical compression injuries with neck flexion. 6. Cervical spine fractures with complete cord lesions require reduction followed by stabilization by closed or open means, as indicated. 7. Cervical spine fractures with incomplete cord lesions require reduction followed by careful evaluation for surgical intervention.

4

5 –2° 6

+20° >3.5 mm –4°

7

ABNORMAL ANGLE

= 20 – (–2) = 22 = 20 – (–4) = 24

>11°

Figure 16A1-13   Abnormal angulation between two vertebrae at any one interspace is determined by comparing the angle formed by the projection of the inferior vertebral body borders with that of either the vertebral body above or the vertebral body below. If the angle at the interspace in question is 11 degrees or greater than that of either adjacent interspace, it is considered by White and associates to be clinical instability. (From White AA, Johnson RM, Punjabi MM, et al: Biomechanical analysis of clinical stability in the cervical spine. Clin Orthop 109:85, 1975.)

Figure 16A1-14   The method for determining translatory displacement, as described by White and colleagues. Using the posteroinferior angle of the superior vertebral body as one point of reference and the posterosuperior angle of the vertebral body below, the distance between the two in the sagittal plane is measured. A distance of 3.5 mm or greater is suggestive of clinical instability. (From White AA, Johnson RM, Punjabi MM, et al: Biomechanical analysis of clinical stability in the cervical spine. Clin Orthop 109:85, 1975.)

Spinal Injuries

677

Alar ligaments

Odontoid

TAL A

B

Figure 16A1-15   A, The atlantoaxial complex as seen from above. B, The disruption of the transverse ligament (TAL) with intact alar ligaments results in C1-C2 instability without cord compression. (Redrawn from Hensinger RN: Congenital anomalies of the atlantoaxial joint. In The Cervical Spine Research Society Editorial Committee: The Cervical Spine, 2nd ed. Philadelphia, JB Lippincott, 1989, p 242.)

The fourth and final goal of treatment is rapid and effective rehabilitation started early in the treatment process. A more specific categorization of athletic injuries to the cervical spine can be made. Specifically, these injuries can be divided into upper cervical spine, midcervical spine, and lower cervical spine injuries.

Upper Cervical Spine Fractures and Dislocations Upper cervical spine lesions involve C1 through C3. Although they rarely occur in sports, several specific injuries that can occur to the upper cervical vertebrae deserve mention. The transverse and alar ligaments are responsible for atlantoaxial stability (Fig. 16A1-15). If these structures are ruptured from a flexion injury with translation of C1 anteriorly, the spinal cord can be impinged between the posterior aspect of the odontoid process and the posterior rim of C1. The patient gives a history of head trauma and complains of neck pain, particularly with nodding, and may or may not present with cord signs. Roentgenographically, lateral views of the C1-C2 articulation demonstrate increase of the atlantodens interval. This interval is normally 3 mm in the adult. With transverse ligament rupture, it may increase up to 10 to 12 mm depending on the status of the alar and accessory ligaments. Note that increase in the atlantodens interval may be seen only when the neck is flexed. Fielding states that atlantoaxial fusion may be the “conservative” treatment for this lesion.33 He recommends posterior C1-C2 fusion using wire fixation and an iliac bone graft. Fractures of the atlas were described by Jefferson in 1920.34 These may be of two types: posterior arch fractures or burst fractures. Posterior arch fractures are more common, and with a brace support, they go on to achieve satisfactory fibrous or bony union. Burst fractures result from an axial load transmitted to the occipital condyles, which then disrupt the integrity of both the anterior and posterior arches of the atlas (Fig. 16A1-16). Roentgenograms

demonstrate bilateral symmetrical overhang of the lateral masses of the atlas in relation to the axis, with an increase in the paraodontoid space on the open mouth view. Clinically, the patient characteristically has pain and imitates the nodding motion. These fractures are considered stable when the combined lateral overhang of the atlas measures less than 7 mm. When the transverse diameter of the atlas is 7 mm greater than that of the axis, a transverse ligament rupture should be suspected (Fig. 16A1-17). Treatment, as recommended by Fielding, includes head-halter traction until muscle spasm resolves, followed by a brace support.33 If flexion-extension roentgenograms subsequently demonstrate significant instability, fusion may be indicated. Fractures of the odontoid have been classified into three types by Anderson and D’Alonzo.35 Type I is an avulsion of the tip of the odontoid at the site of the attachment of the alar ligament and is a rare and stable lesion. Type II is a fracture through the base at or just below the level of the superior articular processes. Type III involves a fracture of

Figure 16A1-16   Schematic representation of the four-part comminuted burst fracture of the atlas as seen from above. (Redrawn from Jefferson G: Fracture of the atlas vertebra. Br J Surg 7:407–422, 1919. By permission of the Publishers ButterworthHeinemann Ltd.)

678

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16A1-17   Illustration of a comminuted Jefferson fracture with both the transverse ligament intact (stable configuration) and a transverse ligament rupture (unstable configuration). (Redrawn from White AA, Punjabi MM: Clinical Biomechanics of the Spine. Philadelphia, JB Lippincott, 1978, p 204.) Y

X

Stable

the body of the axis (Fig. 16A1-18). When the odontoid is not displaced, planograms may be required to identify the lesion. The mechanism of odontoid fractures has not been clearly delineated; however, they appear to be due to head impact. All routine cervical spine roentgenographic studies should include the open mouth view to identify lesions involving the odontoid as well as the atlas. If these are ­negative, and if a lesion in this area is suspected,

Type I

Type II

Type III

Figure 16A1-18   Illustration of the three types of odontoid fractures in both the anteroposterior and lateral planes. Type I is an oblique avulsion fracture from the upper portion of the odontoid. Type II is a fracture of the odontoid process at its base. Type III is an odontoid fracture through the body of C2. (Redrawn from Anderson LD, D’Alonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg Am 56:1663–1674, 1974.)

X + Y ≥7 mm Unstable

planograms or bending films may further delineate pathologic changes in this area. Managing type II fractures is a problem. It has been reported that 36% to 50% of these lesions treated initially with plaster casts or reinforced cervical braces fail to unite. Cloward has reported that 85% of his patients heal within 3 months when treated with the halo brace.36 Current management involves more aggressive surgical management with early posterior stabilization by means of wiring and fusion. It is necessary to stabilize fibrous unions or nonunited fractures of the odontoid surgically if they are demonstrated to be unstable on flexion and extension views. Stabilization may be effected through either posterior C1-C2 wire fixation and fusion or anterior odontoid screw fixation. Fractures through the arch of the axis are also known as traumatic spondylolistheses of C2, or hangman’s fractures. These are relatively rare lesions. The mechanism of injury is generally recognized to be hyperextension. This injury is inherently unstable, but it has been shown to heal with predictable regularity without surgical intervention.

Midcervical Spine Fractures and Dislocations Acute traumatic lesions of the cervical spine at the C3-C4 level are rare and are generally not associated with fractures. These lesions are classified as follows: (1) acute rupture of the C3-C4 intervertebral disk; (2) anterior subluxation of C3 on C4; (3) unilateral dislocation of the joint between the articular processes; and (4) bilateral dislocation of the joint between the articular processes.26,37,38 An episode of transient quadriplegia in an athlete who has sustained head impact but has a cervical spine roentgenogram with negative findings suggests acute rupture of the C3-C4 intervertebral disk. The syndrome of acute anterior spinal cord injury, as described by Schneider and associates,23,24 may be observed. A cervical myelogram or an MRI will substantiate the diagnosis. Anterior diskectomy and interbody fusion may be the most effective treatment of this lesion. Anterior subluxation of C3 on C4 is a result of a shearing force through the intervertebral disk space that disrupts the interspinous ligament as well as the posterior ­ supporting structure. Roentgenograms demonstrate narrowing of the intervertebral disk space, anterior angulation and translation of C3 on C4, an increase in the distance between

Spinal Injuries

Figure 16A1-19   Roentgenogram demonstrates C3-C4 subluxation as manifested by anterior intervertebral disk space narrowing, anterior angulation, displacement of the superior vertebral body, and fanning of the spinous processes. (From Torg JS, Sennett B, Vegso JJ, et al: Axial loading injuries to the middle cervical spine: Analysis and classification. Am J Sports Med 19:17–25, 1991.)

Figure 16A1-20   Unilateral C3-C4 facet dislocation resulting in complete motor and sensory deficit distal to the lesion. There is fanning of the spinous processes of C3 and C4 and more than 20% anterior displacement of the body of C3 on C4 (arrow). (From Torg JS, Sennett B, Vegso JJ, et al: Axial loading injuries to the middle cervical spine: Analysis and classification. Am J Sports Med 19:17–25, 1991.)

679

the spinous processes of the two vertebrae, and instability without fracture of the bony elements (Fig. 16A1-19). Spinal fusion may be necessary for adequate stabilization in such cases, in contrast to cervical spine instability caused by fracture, in which adequate reduction and subsequent bony healing result in stability. When the patient has posterior instability, posterior fusion is preferable to an anterior interbody fusion. Unilateral facet dislocation at C3-C4 may result in immediate quadriparesis. This injury involves the intervertebral disk space, the interspinous ligament, the posterior ligamentous supporting structures, and the one facet with resulting rotatory dislocation of C3 on C4 without fracture (Fig. 16A1-20). At this level, strong skeletal traction does not usually yield a successful reduction, and closed manipulation under general anesthesia is necessary to disengage the locked joint between the articular ­processes. Bilateral facet dislocation at the C3-C4 level is a grave lesion (Fig. 16A1-21). Skeletal traction may not reduce the lesion, and the prognosis for this injury is poor.

Lower Cervical Spine Fractures and Dislocations Lower cervical spine fractures or dislocations are those involving C4 through C7. In injuries resulting from various athletic endeavors, most fractures or dislocations of the

Figure 16A1-21   Bilateral facet dislocation at the C3-C4 level demonstrates anterior angulation as well as translation greater than 50% of the width of the vertebral body associated with spinous fanning. The lesion resulted in quadriplegia. (From Torg JS, Sennett B, Vegso JJ, et al: Axial loading injuries to the middle cervical spine: Analysis and classification. Am J Sports Med 19:17–25, 1991.)

680

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

cervical spine, with or without neurologic involvement, involve this segment. Although unilateral and bilateral facet dislocations occur, they are relatively rare. Most severe, athletically incurred cervical spine injuries are fractures of the vertebral body with varying degrees of compression or comminution.39

Unilateral Facet Dislocations Unilateral facet dislocations are the result of axial loading, flexion-rotation types of mechanisms. The lesion may be truly ligamentous without an associated vertebral fracture. In such instances, the facet dislocation is stable and is usually associated with neurologic involvement. Roentgenograms demonstrate less than 50% anterior shift of the superior vertebra on the inferior vertebra. Attempts should be made to reduce the facet dislocation by means of skeletal traction. As with similar lesions described at the C3-C4 level, it may not be possible to effect a closed reduction. In this instance, open reduction under direct vision through a posterior approach with supplemental posterior element bone grafting should be performed.

Bilateral Facet Dislocations Bilateral facet dislocations are unstable and are almost always associated with neurologic involvement. These injuries are associated with a high incidence of ­ quadriplegia. Lateral roentgenograms demonstrate greater than 50% anterior displacement of the superior vertebral body on the inferior vertebral body. Immediate treatment, as described previously, consists of closed reduction with skeletal traction. Such lesions are generally reducible by means of skeletal traction and are then treated by halo-brace ­ stabilization

A

and posterior fusion. It should be noted that instability is directly related to the ease with which the lesion is reduced because the easier it is to reduce, the easier it is to re-dislocate. If skeletal traction is unsuccessful, either manipulative reduction under sedation or general anesthesia or open reduction under direct vision is recommended. When the dislocation is reduced closed and the reduction is maintained, immobilization should be effected by use of the halo brace for 8 to 12 weeks. Corrective bracing should continue for an additional 4 weeks.

Vertebral Body Compression Fractures Compression fractures of the vertebral body are a result of axial loading. Vertebral body fractures of the cervical spine can be classified into five types.40

Type I Simple wedge or vertebral end-plate compression fractures of the cervical vertebrae are common injuries that respond to conservative management and rarely are associated with neurologic involvement (Fig. 16A1-22). It is important to differentiate these lesions from compression fractures that are associated with disruption of the posterior element soft tissue supporting structures. The latter lesions are unstable and are frequently associated with neurologic involvement, including quadriplegia.

Type II An isolated anteroinferior vertebral body or “teardrop” fracture is without displacement, has intact posterior elements, and is not associated with neurologic involvement

B

Figure 16A1-22   A and B, Type I vertebral body end plate compression fracture involving the superior aspect of C6 (arrows). Extension and flexion views demonstrate absence of evidence of instability. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

Spinal Injuries

681

Figure 16A1-23   Type II anteroinferior vertebral body fracture demonstrates the characteristic isolated teardrop fracture. Lateral roentgenograms of the lesion demonstrate maintenance of adjacent disk space height as well as a lack of subluxation or spinous process fanning. If there is no disruption of the posterior elements, this is a relatively stable lesion. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

Figure 16A1-24   Type III comminuted burst fracture of C4 with displacement of fragments into the vertebral canal (arrow). (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

(Fig. 16A1-23). This is a relatively stable fracture and may be treated conservatively.28,41

Cervical Spinal Stenosis with Cord Neurapraxia and Transient Quadriplegia

Type III

Characteristically, the clinical picture of cervical cord neurapraxia (CCN) with transient quadriplegia involves an athlete who sustains an acute transient neurologic episode of cervical cord origin with sensory changes that may be associated with motor paresis involving both arms, both legs, or all four extremities after forced hyperextension, hyperflexion, or axial loading of the cervical spine.47,48 Sensory changes include burning pain, numbness, tingling, or loss of sensation; motor changes consist of weakness or complete paralysis. The episodes are transient, and complete recovery usually occurs in 10 to 15 minutes, although in some cases gradual resolution does not occur for 36 to 48 hours. Except for burning paresthesia, neck pain is not present at the time of injury. There is complete return of motor function and full, pain-free cervical motion. Routine roentgenograms of the cervical spine show no evidence of fracture or dislocation, but a demonstrable degree of cervical spinal stenosis is ­present.

Comminuted burst vertebral body fractures have intact posterior elements, but displacement of bony fragments into the vertebral canal may place the cord in jeopardy. Late settling of the fracture with deformity can occur. Surgical stabilization is recommended (Fig. 16A1-24).

Type IV The axial load three-part–two-plane vertebral body fracture consists of three fracture parts: (1) an anteroinferior teardrop; (2) a sagittal vertebral body fracture; and (3) disruption of the posterior neural arch.42-46 This lesion is unstable and is almost always associated with quadriplegia. Careful evaluation of the routine anteroposterior roentgenogram or CT scan is necessary to appreciate the sagittal vertebral body fracture, a finding that portends a grave prognosis (Fig. 16A1-25).

Type V This is a vertebral body three-part–two-plane compression fracture associated with disruption of posterior elements of an adjacent vertebra. This is an extremely unstable fracture (Fig. 16A1-26).

Determination of Spinal Stenosis: Method of Measurement To identify cervical stenosis, a method of measurement is needed. The standard method, the one most commonly employed for determining the sagittal diameter of

682

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B

D

Figure 16A1-25   A, A three-part–two-plane fracture of C6. Lateral view demonstrates prevertebral soft tissue swelling and an anteroinferior fracture fragment of C6 involving the entire vertebral body height and one third of the vertebral body width. There is about 1 mm of posterior displacement of the inferior aspect of the posterior vertebral body. The C6-C7 intervertebral disk space is minimally narrowed posteriorly with associated capsular disruption and “fanning.” B, Frontal view demonstrates a faint, linear radiolucency through the C6 vertebral body indicating a sagittal vertebral body fracture (arrow). There is mild lateral mass displacement. C, Computed tomographic examination demonstrates the sagittal fracture extending completely through the vertebral body with disruption of the lamina on the right. D, Diagrammatic representation of the three-part–two-plane vertebral body compression fracture demonstrates the anteroinferior teardrop as well as the sagittal vertebral body fractures and associated fracture through the lamina. (A, From Torg JS, Pavlov H, O’Neill MJ, et al: The axial load teardrop fracture. Am J Sports Med 19:355–364, 1991.)

the spinal canal, involves measuring the distance between the middle of the posterior surface of the vertebral body and the nearest point on the spinolaminar line. Using this technique, Boden and colleagues49 reported that the average sagittal diameter of the spinal canal from the fourth to

the sixth cervical vertebra in 200 healthy individuals was 18.5 mm (range, 14.2 to 23 mm). The target distance he used was 1.4 m. Others have noted that values of less than 14 mm are uncommon and fall below the standard deviation for any cervical segment. Other measurements

Spinal Injuries

A

683

B

Figure 16A1-26   A, Type V compression fracture of the vertebral body associated with fractures of the neural arch of an adjacent vertebra. Settling and posterior displacement of the superior vertebral segment occur. B, Distraction of the superior vertebral segment with skeletal traction permits visualization of fractures through the pedicles (arrow) of C6 in addition to the three-part–two-plane fracture of the body of C5. (From Torg JS [ed]: Athletic Injuries to the Head, Neck and Face. Philadelphia, Lea & Febiger, 1982.)

reported in the literature vary greatly. The variations in the landmarks and the methods used to determine the sagittal distance, as well as the use of different target distances for roentgenography, have resulted in inconsistencies in the so-called normal values. Therefore, the standard method of measurement for spinal stenosis is a questionable one.

On the basis of these observations, it may be concluded that the factor that explains the described neurologic picture of CCN is diminution of the anteroposterior diameter of the spinal canal, either as an isolated observation or in association with intervertebral disk herniation, ­ degenerative changes, post-traumatic instability, or

The Ratio Method An alternative way to determine the sagittal diameter of the spinal canal was devised by Pavlov and colleagues and is called the ratio method.50 It compares the standard method of measurement of the canal with the anteroposterior width of the vertebral body at the midpoint of the corresponding vertebral body (Fig. 16A1-27). The actual measurement of the sagittal diameter in millimeters, as determined by the conventional method, is misleading both as reported in the literature and in actual practice because of variations in the target distances used for roentgenography and in the landmarks used for obtaining the measurement. The ratio method compensates for variations in roentgenographic technique because the sagittal diameter of both the canal and the vertebral body is affected similarly by magnification factors. The ratio method is independent of variations in technique, and the results are statistically significant. Using the ratio method of determining the dimension of the canal, a ratio of the spinal canal to the vertebral body of less than 0.80 is indicative of cervical stenosis. I believe that the ratio of the anteroposterior diameter of the spinal canal to that of the vertebral body (SC/VB ratio) is a consistent and reliable way to determine cervical stenosis (Fig. 16A1-28) in those individuals who have experienced episodes of CCN. The ratio has a very low predictive value, however, and should not be used as a screening tool.

a b

ratio =

a b

Figure 16A1-27   The ratio of the spinal canal to the vertebral body is the distance from the midpoint of the posterior aspect of the vertebral body to the nearest point on the corresponding spinolaminar line (a) divided by the anteroposterior width of the vertebral body (b). (From Torg JS, Pavlov H, Gennario SE, et al: Neurapraxia of the cervical spinal cord with transient quadriplegia. J Bone Joint Surg Am 68:1354–1370, 1986.)

684

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 16A1-28   A and B, A comparison between the ratio of the spinal canal to the vertebral body of a stenotic patient and the same ratio in a control subject is demonstrated on lateral roentgenograms of the cervical spine. The ratio is about 1:2 (0.50) in the stenotic patient compared with 1:1 (1.00) in the control subject. (From Torg JS, Pavlov H, Gennario SE, et al: Neurapraxia of the cervical spinal cord with transient quadriplegia. J Bone Joint Surg Am 68:1354–1370, 1986.)

c­ ongenital anomalies. In instances of developmental cervical stenosis, forced hyperflexion, or hyperextension of the cervical spine, the caliber of an already narrow canal is further decreased, as explained by the pincer mechanism of Penning (Fig. 16A1-29).51 In patients in whom stenosis is associated with osteophytes or a herniated disk, direct pressure can occur, again when the spine is forced in the extremes of flexion and extension. It is further postulated that with an abrupt but brief decrease in the anteroposterior diameter of the spinal canal, the cervical cord is mechanically compressed, causing transient interruption of either motor or sensory function, or both, distal to the lesion. The neurologic aberration that results is transient and completely reversible. A review of the literature has revealed few reported cases of transient quadriplegia occurring in athletes. Attempts to establish the incidence indicate that the problem is more prevalent than may be expected. Specifically, in a population of 39,377 exposed participants, the reported incidence of transient paresthesia in all four extremities was 6 per 10,000, whereas the reported incidence of paresthesia associated with transient quadriplegia was 1.3 per 10,000 in the one football season surveyed. From these data, it may be concluded that the prevalence of this problem is relatively

Figure 16A1-29   The pincers mechanism, as described by Penning, occurs when the distance between the posteroinferior margin of the superior vertebral body and the anterosuperior aspect of the spinolaminar line of the subjacent vertebra decreases with hyperextension, resulting in compression of the cord. With hyperflexion, the anterosuperior aspect of the spinolaminar line of the superior vertebra and the posterosuperior margin of the inferior vertebra would be the “pincers.” (Redrawn from Torg JS, Pavlov H, Gennario SE, et al: Neurapraxia of the cervical spinal cord with transient quadriplegia. J Bone Joint Surg Am 68:13541–370, 1986.)

Spinal Injuries

4 5 1 2

1 4 2 5

5 4 1 2

2 4 1 5

17

1 College 2 Professional 3 Transient 4 Quadriplegic 5 Control

16

Canal

18

19

football. None of the 77 quadriplegic individuals (cohort 4) had an episode of neurapraxia of the spinal cord before the catastrophic injury. Also, none of the 45 high school, college, and professional players who had an episode of transient neurapraxia (cohort 3) became quadriplegic. These data, in combination with the absence of developmental narrowing of the cervical canal in the quadriplegic group (cohort 4), provide evidence that transient neurapraxia of the cervical cord and an injury associated with permanent catastrophic neurologic sequelae are unrelated (Fig. 16A1-30). Therefore, developmental narrowing of the cervical canal in a spine that has no evidence of instability is neither a harbinger of nor a predisposing factor for permanent neurologic injury. The data did not reveal an association between developmental narrowing of the cervical canal and quadriplegia. The major factor in the occurrence of cervical quadriplegia in football players is a tackling technique in which the head is used as the primary point of contact, with resulting transmission of axial energy to, and subsequent failure of, the cervical spine. The findings of this study demonstrated the high sensitivity, low specificity, and low predictive value of the ratio of the diameter of the cervical spinal canal to that of the vertebral body, precluding its use as a screening mechanism for determining the suitability of an individual for participation in contact sports. Developmental narrowing of the cervical canal without associated instability does not predispose an individual to permanent catastrophic neurologic injury and therefore should not preclude an athlete from participation in contact sports. More recently, a group of 110 patients with CCN has been studied. In this report, a classification system of CCN was developed, and MRI data were analyzed using a new computerized measurement technique.53 CCN was classified according to the type of neurologic deficit: plegia for episodes with complete paralysis; paresis for ­ episodes

3 3

15

high and that an awareness of the causes, manifestations, and appropriate principles of management is warranted.47 Characteristically, after an episode of CCN with or without transient quadriplegia, the first question raised concerns the advisability of restricting activity. In an attempt to address this problem, 117 young athletes have been interviewed who sustained cervical spine injuries associated with complete permanent quadriplegia while playing football between the years 1971 and 1984. None of these patients recalled a prodromal experience of transient motor paresis. Conversely, none of the patients in this series who had experienced transient neurologic episodes subsequently sustained an injury that resulted in permanent neurologic injury. On the basis of these data, it was concluded that a young patient who has had an episode of CCN with or without an episode of transient quadriplegia is not predisposed to permanent neurologic injury because of it.47 Subsequent to the description of neurapraxia of the cervical cord with transient quadriplegia, a number of issues concerning the disorder have arisen. An epidemiologic study evaluating 45 athletes who had an episode of transient neurapraxia of the cervical spinal cord revealed the consistent finding of developmental narrowing of the cervical spinal canal. The purpose of the study was to determine the relationship, if any, between a developmentally narrowed cervical canal and reversible and irreversible injury of the cervical cord with use of various cohorts of football players and a large control group. Cohort 1 was composed of 227 college football players who were asymptomatic and had no known history of transient neurapraxia of the cervical cord. Cohort 2 consisted of 97 professional football players who also were asymptomatic and had no known history of transient neurapraxia of the cervical cord. Cohort 3 was a group of 45 high school, college, and professional football players who had at least one episode of transient neurapraxia of the cervical cord. Cohort 4 was composed of 77 individuals who were permanently quadriplegic as a result of an injury while playing high school or college football. Cohort 5 consisted of a control group of 105 male subjects who were not athletes and had no history of a major injury of the cervical spine, an episode of transient neurapraxia, or neurologic symptoms.52 The mean and standard deviation of the diameter of the spinal canal, the diameter of the vertebral body, and the ratio of the diameter of the spinal canal to that of the vertebral body were determined for the third through sixth cervical levels on the radiographs of each cohort. In addition, sensitivity, specificity, and positive predictive value of the ratio of the diameter of the spinal canal to that of the vertebral body of 0.90 or less were evaluated. The findings of this study demonstrated that a ratio of 0.80 or less had a high sensitivity (93%) for transient neurapraxia. These findings also support the concept that the symptoms result from a transient reversible deformation of the spinal cord in a developmentally narrowed osseous canal. The low predictive value of the ratio (0.2%), however, precludes its use as a screening mechanism for determining the suitability of an athlete for participation in contact activities. Axial load, degree of instability, and the period of time from injury to reduction have been implicated as factors in permanent neurologic injury in athletes who play tackle

685

3 3

C3

C4

C5

C6

Cervical Spine Level Figure 16A1-30   Profile plot of the mean diameter of the spinal canal, demonstrating a significantly smaller value in cohort 3 compared with that in all other cohorts (P < .05). With the numbers available, no significant difference was found among cohorts 1, 2, 4, and 5. (From Torg JS, Naranja RJ Jr, Pavlov H, et al: The relationship of developmental narrowing of the cervical spinal canal to reversible and irreversible injury of the cervical spinal cord in football players. J Bone Joint Surg Am 78:1308-1314, 1996.)

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� .9

.9

.8

.8

.7

.7 Probability of Recurrence

Probability of Recurrence

686

.6 .5 .4 .3

.5 .4 .3

.2

.2

.1

.1 0

0 6

A

.6

7

8 9 10 11 12 13 MRI Disk Level Canal Diameter (mm)

.3

14 15

B

.4 .5 .6 .7 .8 .9 1 1.1 1.2 1.3 X-ray Spinal Canal/Vertebral Body Ratio

Figure 16A1-31   Graphs developed using regression analysis in which the risk for recurrence can be plotted as a function of the disk-level diameter measured on magnetic resonance imaging (A) and the spinal canal–to–vertebral body ratio calculated on the basis of x-ray films (B). The construction of these plots is based on the result that increased risk for recurrence is inversely correlated with canal diameter. Future patients with cervical cord neurapraxia can be counseled regarding their individual risk for recurrence based on the particular size of their spinal canal.

with motor weakness; and paresthesia for episodes that involve only sensory changes without motor involvement. The CCN grade was defined by the length of time that the neurologic symptoms persisted: grade I, less than 15 minutes; grade II, greater than 15 minutes but less than 24 hours; and grade III, greater than 24 hours. The CCN pattern was defined by the anatomic distribution of all the neurologic symptoms: quad, episodes that involve all four extremities; upper, episodes involving both arms; lower, episodes involving both legs; and hemi, episodes involving an ipsilateral arm and leg. Using this classification system, the incidence of CCN type was plegia in 44 cases (40%), paresis in 28 (25%), and paresthesia in 38 cases (35%). CCN was grade I in 81 cases (74%), grade II in 17 cases (15%), and grade III in 12 cases (11%). The pattern was quad in 88 cases (80%), upper in 17 cases (15%), lower in 2 cases (2%), and hemi in 3 cases (3%). To study the relationship of the spinal cord to the spinal canal, a computerized system was developed to analyze magnetic resonance images. This system consisted of a personal computer and a color scanner with a transparency adapter. Images were digitized on the scanner and then uploaded using an imaging software package. The midsagittal T1and T2-weighted images were digitized. Using a graphics digitizer pad with a resolution of 0.01 mm, the following measurements were made at levels C3-C7: To quantify spondylitic narrowing, the disk level canal diameter was measured as the shortest distance between the intervertebral disk and the body posterior elements; the cord diameter was determined by measuring the transverse diameter of the spinal cord at the appropriate level; and the space available for the cord was calculated by subtracting the spinal cord diameter from the disk-level canal diameter. Follow-up evaluation was obtained by questionnaire, telephone interview, or office evaluation and was available for 105 of the 110 cases. Sixty-three patients (57%) returned to contact activities after their first episode of CCN. Of this group, 35 patients (56%) experienced a second episode of CCN. Once again, there were no permanent or catastrophic neurologic injuries related to the occurrence of CCN. Patients returning to football had a

higher ­recurrence rate than those returning to other sports. Thirty-two of 52 football players (62%) who returned to the sport experienced a recurrence, compared with 3 of 11 players (27%) who returned to other sports. All radiologic measurements except spinal cord diameter were predictive or recurrent. The patients’ age, level of sports participation, radiographic findings, MRI findings, clinical CCN classification, and radiologic classification did not have predictive value in determining which patients were at risk for recurrence. The presence of disk herniation, cord compression, degenerative disk disease, or any other finding was not an indicator of whether patients would suffer future episodes of CCN. Based on the finding that narrowing of the canal is a causative factor of CCN, the recurrence and diameter data were analyzed and correlated. Graphic plots were constructed using logistic regression analysis of the percentage risk for recurrence versus the disk-level canal diameter and the SC/VB ratio. The plots demonstrated a strong inverse correlation between the risk for recurrence and the disklevel canal diameter and SC/VB ratio (Fig. 16A1-31). Of note, there is one reported case of a professional football player who had congential stenosis and sustained a partial cervical spinal cord injury manifested by mild upper extremity dysesthesias and mild weakness of the wrist extensors and biceps.96

PREVENTION Athletic injuries to the cervical spine that result in injury to the spinal cord are infrequent but catastrophic events. Accurate descriptions of the mechanism or mechanisms responsible for a particular injury transcend simple academic interest. Before preventive measures can be developed and implemented, identification of the mechanisms involved in the production of the particular injury is necessary. Because the nervous system is unable to recover significant function after severe trauma, prevention assumes a most important role when considering these injuries. Injuries resulting in spinal cord damage have been associated with football,49,54-58 water sports,59-63 wrestling,64 rugby,65-69 trampolining,70-72 and ice hockey.73,74 The use

Spinal Injuries

of epidemiologic data, biomechanical evidence, and cinematographic analysis has (1) defined and supported the involvement of axial load forces in cervical spine injuries occurring in football; (2) demonstrated the success of appropriate rule changes in the prevention of these injuries; and (3) emphasized the need for employment of epidemiologic methods to prevent cervical spine and similar severe injuries in other high-risk athletic activities. Identification of the cause of football-related cervical quadriplegia and its prevention center on four areas: (1) the role of the helmet-facemask protective system; (2) the concept of the axial loading mechanism of injury; (3) the effect of the 1976 rule changes banning spearing and the use of the top of the helmet as the initial point of contact in tackling; and (4) the necessity for continued research, education, and enforcement of rules.

A

C

687

The protective capabilities provided by the modern football helmet have resulted in the advent of playing techniques that have placed the cervical spine at risk for injury, with associated catastrophic neurologic sequelae. Available cinematographic and epidemiologic data clearly indicate that cervical spine injuries associated with quadriplegia occurring as a result of football are not hyperflexion accidents. Instead, they are due to purposeful axial loading of the cervical spine as a result of spearing and head-first playing techniques (Fig. 16A1-32). As a causative factor, the modern helmet-facemask system is secondary, contributing to these injuries because of its protective capabilities that have permitted the head to be used as a battering ram, thus exposing the cervical spine to injury. Classically, hyperflexion has been emphasized as playing a role in cervical spine trauma, whether the injury was

B

D

Figure 16A1-32   A, With the advent of the polycarbonate helmet-facemask protective device, use of the top or crown of the helmet as the initial point of contact in blocking and tackling became prevalent. Contact is made, the head abruptly stops, the momentum of the body continues, and the cervical spine is literally crushed between the two. In this instance, the fracture-dislocation transected the spinal cord. B, The injured player collapses, having been rendered quadriplegic. C, Further collapse is noted. D, The player is evacuated on a spine board and stretcher. (Photos courtesy of Randy Green, Vanderbilt Student Communications.)

688

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

due to a diving accident, trampolining, rugby, or American football. Epidemiologic and cinematographic analyses have established that most cases of cervical spine quadriplegia that occur in football have resulted from axial loading. Far from being an accident or an untoward event, techniques are deliberately used that place the cervical spine at risk for catastrophic injury (Fig. 16A1-33). Recent laboratory observations also indicate that athletically induced cervical spine trauma results from axial loading.56-58,75 Mertz and colleagues,76 Hodgson and Thomas,77 and Sances and colleagues78 measured stresses and strains within the cervical spine when axial impulses were applied to helmeted cadaver head-spine-trunk specimens. They were able to produce fractures of the lower cervical spine when the impulse was applied to the crown of the helmet. Hodgson and Thomas determined that direct vertex impact imparted a larger force to the cervical vertebrae than forces applied further forward on the skull. Gosch and associates79 investigated three different injury modes (hyperflexion, hyperextension, and axial compression) in anesthetized monkeys and concluded that axial compression produced cervical spine fractures and dislocations. Maiman and coworkers,80

Roaf,81-83 and White and associates29 demonstrated vertebral body fractures in the lower cervical spine caused by the axial loading of isolated spinal units. Roaf subjected spinal units to forces differing in direction and magnitude and concluded that hyperflexion of the cervical spine was an anatomic impossibility. In contrast, he was able to produce almost every variety of spinal injury with a combination of compression and rotation. Bauze and Ardran84 postulated that axial loads were responsible for cervical spine dislocations as well as fractures. They demonstrated that failure of the facet joints and posterior ligaments occurred when axial loads were applied to cadaveric spines. When the lower portion of the spine was flexed and fixed, and the upper part extended and free to move forward, vertical compression produced bilateral dislocation to the facet joints without fracture. If lateral tilt or axial rotation occurred as well, a unilateral dislocation was produced. The forces observed were all less than those required for bony failure and allowed facet dislocation without associated bony injury. Nightingale and associates85 analyzed the relationships among head motion, local deformations of the cervical

A

B

C

D

Figure 16A1-33   A, Subject (No. 37, foreground) lines up in front of ball carrier in preparation for tackling. B, Preimpact position shows tackler about to ram ball carrier with the crown of his helmet. C, At impact, contact is made with the top of the helmet. Although the neck is slightly flexed, it is clearly not hyperflexed. The major force vector is transmitted along the axial alignment of the cervical spine. D, The tackler recoils following impact.

Spinal Injuries

A

B

Figure 16A1-34   A, When the neck is in a normal upright anatomic position, the cervical spine is slightly extended because of the natural cervical lordosis. B, When the neck is flexed slightly, to about 30 degrees, the cervical spine is straightened and converted into a segmented column.

spine, and injury mechanisms using a cadaver head and neck model that had experienced impact in an anatomically neutral position. They observed that classic concepts of flexion and extension of the cervical spine do not apply as a mechanism of injury to a vertically impacted head. They further concluded that straightening of the cervical spine before injury may be a necessary element of the compressive flexion mechanism. In the course of a collision activity, such as tackle football, most energy inputs to the cervical spine are effectively dissipated by the energy-absorbing capabilities of the cervical musculature through controlled lateral bending, flexion, or extension motion. The bones, the disks, and the ligamentous structures can be injured when contact occurs on the top of the helmet when head, neck, and trunk are

A

B

689

positioned in such a way that forces are transmitted along the longitudinal axis of the cervical spine. When the neck is in the anatomic position, the cervical spine is extended as a result of normal cervical lordosis. When the neck is flexed to 30 degrees, the cervical spine straightens (Fig. 16A1-34). In axial loading injuries, the neck is slightly flexed, and normal cervical lordosis is eliminated, thereby converting the spine into a straight segmented column. Assuming head, neck, and trunk components to be in motion, rapid deceleration of the head occurs when it strikes another object, such as another player, trampoline bed, or lake bottom. This results in the cervical spine being compressed between the rapidly decelerated head and the force of the oncoming trunk. When the maximal amount of vertical compression is reached, the straightened cervical spine fails in a flexion mode, and fracture, subluxation, or unilateral or bilateral facet dislocation can occur (Fig. 16A1-35).86 Refutation of the “freak accident” concept with the more logical principle of cause and effect has been most rewarding in dealing with problems of football-induced cervical quadriplegia. Definition of the axial loading mechanism—in which football players, usually defensive backs, make a tackle by striking an opponent with the top of their helmet—has been a key element in this process. Implementation of rule changes and coaching techniques eliminating the use of the head as a battering ram has resulted in a dramatic reduction in the incidence of quadriplegia since 1976. Data on cervical spine injuries resulting from participation in football have been compiled by a national registry since 1971.56,58 Analysis of the epidemiologic data and cinematographic documentation clearly demonstrates that most cervical fractures and dislocations were due to axial loading. On the basis of these observations, rule changes banning both deliberate “spearing” and the use of the top of the helmet as the initial point of

C

D

E

Figure 16A1-35   Biomechanically, a straight cervical spine responds to an axial load force like a segmented column. Axial loading of the cervical spine first results in compressive deformation of the intervertebral disks (A and B). As the energy input continues and maximal compressive deformation is reached, angular deformation and buckling occur. The spine fails in a flexion mode (C), with resulting fracture, subluxation, or dislocation (D and E). Compressive deformation to failure with a resultant fracture, dislocation, or subluxation occurs in as little as 8.4 msec. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injuries. Clin J Sport Med 1:12–27, 1991.)

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

690 35

34

Quadriplegia

30 32 25 20 15 10 5 0

18 16

17 13

14

11 11 10

8

11 5

6

13

11 1

3 4 2 4

1995 1994 1993 1992 1991 1990 1989 1988 1987 1986 1985 1984 1983 1982 1981 1980 1979 1978 1977 1976 1975 Year Figure 16A1-36   The yearly incidence of permanent cervical quadriplegia for all levels of participation (1975 to 1995) decreased dramatically in 1977 following initiation of the rule changes prohibiting the use of head-first tackling and blocking techniques.

contact in making a tackle were implemented at the high school and college levels. Subsequently, a marked decrease in cervical spine injury rates has occurred. The incidence of permanent cervical quadriplegia decreased from 34 cases in 1976 to 5 in the 1984 season (Fig. 16A1-36). I believe that most athletic injuries to the cervical spine associated with quadriplegia also occur as a result of axial loading. Tator and colleagues studied 38 acute spinal cord injuries caused by diving accidents. They observed, “In most cases the cervical spine was fractured and the spinal cord crushed. The top of the head struck the bottom of the lake or pool.”87 Scher, reporting on vertex impact and cervical dislocation in rugby players, observed, “When the neck is slightly flexed, the spine is straight. If significant force is applied to the vertex when the spine is straight, the force is transmitted down the long axis of the spine. When the force exceeds the energy-absorbing capacity of the structures involved, cervical spine flexion and dislocation will result.”67 Tator and Edmonds73 reported the results of a national questionnaire by the Canadian Committee on the Prevention of Spinal Injuries due to Hockey, which recorded 28 injuries involving the spinal cord, 17 of which resulted in complete paralysis. They noted that in this series, the most common mechanism involved was a check, in which injured players struck the boards “with the top of their heads, while their necks were slightly flexed.” Reports in the recent literature on the mechanism of injury involved in cervical spine injuries resulting from water sports (diving), rugby, and ice hockey support our thesis.

PATHOPHYSIOLOGY OF CERVICAL CORD INJURY AS IT RELATES TO THE PRINCIPLES OF CORD RESUSCITATION Cervical cord injuries have resulted in reversible, incompletely reversible, and irreversible neurologic deficits. An explanation for this variable response to injury has been obtained from the study of the histochemical responses of a squid axon injury model to mechanical deformation.88 The spinal cord is considered an element with a low ­modulus of

rigidity in which compressive macroscopic loads applied to the cord result in localized tension within the tissue. Various macroscopic deformations result in local elongation. With axial elongation of the cord, all elements experience stretch. With extension or flexion, the tension in the cord will vary across the diameter. Highly localized loading, such as shearing from subluxation of the vertebral elements, or focal compressions, such as a weight-drop experiment, result in elongation of the elements in the direction of the long axis of the cord. The effects of mechanical deformation of the axon membrane lead to an alteration in membrane permeability as a result of the development of nonspecific defects in the membrane. This allows calcium to flow into the cell and results in depolarization of the membrane. The giant axon of the squid was used as the tissue model to determine the effects of high strain and uniaxial tension to various degrees of stretch in concert with the neurophysiologic changes of the single axon. These experiments showed that the degree of mechanical injury to the axon influences the magnitude of the calcium insult and the time course of the recovery phase. A low rate of deformation produces only a small reversible depolarization. The axon responds to the increased intracellular calcium by pumping it extracellularly with no residual deficit. As the rate of loading is increased, the magnitude of the ­depolarization and the recovery time to the original resting potential increase in a nonlinear fashion. The axon may or may not fully recover depending on the ability of the cell to pump calcium. With a large influx of calcium, intracellular calcium pumps may be overwhelmed, resulting in irreversible injury. The excess intracellular calcium results in activation of calcium-activated neutral proteases, which lead to cytoskeletal depolymerization and the accumulation of proteins intracellularly. The resulting increased osmotic pressure causes the cell to swell and eventually rupture (Fig. 16A1-37). In addition to the immediate and direct effect of mechanical deformation on the cytosolic calcium concentration within the axon, it has been shown that high strain rate elongation of isolated venous specimens elicits a spontaneous constriction. This mechanically induced vasospasm has the effect of altering blood flow in various regions as a function of the level of vessel stretch. Ultimately, the outcome for the neural tissue will depend synergistically on the level of calcium introduced into the cytosol and the degree to which the metabolic machinery of the cell may be compromised by regional reduction in blood flow.88

Clinical Correlation The clinical evidence of varying degrees of recovery to cervical spine injury correlates with the squid axon model. Cord neurapraxia and transient quadriplegia, a completely reversible lesion, are associated with developmental narrowing of the cervical spine. Cord deformation occurs rapidly and is attributable to a hyperflexion or hyperextension mechanism. Disruption of cell membrane permeability leads to a small increase in intracellular calcium, but spinal stability and cell anatomy are not disturbed, and the deleterious effects of local anoxia secondary to venous spasm do not impede recovery of axonal function.

Spinal Injuries 1o Resting axon 1o+∆1 Stimulus controlled strain and strain rate Ca++

Ca++ 1o

Elevated intracellular calcium

Excess calcium produces CANP and cytoskeletal depolymerization Accumulation of vesicles and elevated protein solution Increased osmotic pressure with cell swelling

Elevated intracellular hydrostatic pressure and axolemma rupture Figure 16A1-37   Schematic representation of the effects of elevated intracellular calcium concentration on cell viability. Specifically, elevated cytosolic-free calcium in excess of 50 m will result in calcium-activated neutral protease (CANP), which can damage protein structures of the cell.

Cervical cord lesions with incomplete reversibility are often associated with instability, such as is seen with subluxation or unilateral facet dislocation, whereby the cord undergoes maximal elastic deformation. It is proposed that lack of full recovery is attributable to prolonged duration of deformity with local anoxia inhibiting cell membrane function and a reduction of intracellular calcium concentrations. Irreversible cord injury with permanent quadriplegia results from an axial load mechanism, which causes a fracture or dislocation that renders the spine markedly unstable. The cord undergoes functional plastic deformation with anatomic disruption of axonal integrity.

Management Implications: Principles of Cervical Cord Resuscitation These observations support the concept that acute spinal cord injury with concomitant subluxation and dislocation should be reduced promptly. This approach contradicts previous approaches that recommended gradual reductions of cervical dislocations over a prolonged period of time.

691

Recent studies have documented the efficacy of methylprednisolone in the management of acute spinal cord injuries. These observations suggest the possible efficacy of other pharmacologic agents that would increase vasodilation and local blood flow and counteract the effects of local cord anoxia or enhance the removal of intracellular calcium. Correlation of both reversible and irreversible spinal cord injury with the effect of neuronal and small vessel deformation has clearly indicated the potential for neurologic recovery by reversing the effects of increased intracellular calcium ion concentration and tissue anoxia. Presumably, these observations suggest that it is secondary cord injury caused by hypoxia and aberration in cell membrane potential that are largely responsible for irreversible neurologic deficits. Thus, the concept of spinal cord resuscitation is proposed as an attempt to reverse secondary changes that occur to obtain maximal neurologic recovery. Such measures would include prompt relief of cord deformation, administration of intravenous corticosteroid, measures to facilitate spinal cord perfusion, and pharmacologic agents to facilitate the return of the calcium pump mechanism.89

CRITERIA USED TO GAUGE RETURN TO CONTACT ACTIVITIES AFTER CERVICAL SPINE INJURY Injury to the cervical spine and associated structures as a result of participation in competitive athletic and recreational activities is not uncommon. It appears that the frequency of these various injuries is inversely proportional to their severity. Whereas Albright and colleagues25 have reported that 32% of college football recruits sustained “moderate” injuries while in high school, catastrophic injuries with associated quadriplegia occur in fewer than 1 in 100,000 participants per season at the high school level. As indicated, the variety of possible lesions is considerable and the severity variable. The literature dealing with diagnosis and treatment of these problems is considerable. However, conspicuously absent is a comprehensive set of standards or guidelines for establishing criteria for permitting or prohibiting return to contact sports (boxing, football, ice hockey, lacrosse, rugby, wrestling) following injury to the cervical spinal structures. The explanation for this void appears to be twofold. First, the combination of a litigious society and the potential for great harm should things go wrong makes “no” the easiest and perhaps most reasonable advice. Second and perhaps most important, with the exception of the matter of transient quadriplegia, there is a lack of credible data pertaining to postinjury risk factors. Despite a lack of credible data, this chapter will attempt to establish guidelines to assist the clinician as well as the patient and his or her parents in the decision-making process.90 Cervical spine conditions requiring a decision as to whether or not participation in contact activities is advisable and safe can be divided into two categories: (1) congenital or developmental conditions, and (2) post-­traumatic conditions. Each condition has been determined to pre­ sent either no contraindication, relative contraindications, or an absolute contraindication on the basis of a variety of parameters. Information compiled from more than

692

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

1200 cervical spine injuries documented by the National Football Head and Neck Injury Registry has provided insight into whether various conditions may or may not predispose to more serious injury.56-58 A review of the literature in several instances provides significant data for a limited number of specific conditions. Analysis of many conditions predicated on an understanding of recognized injury mechanisms has permitted categorization on the basis of “educated” conjecture. And last, much reliance has been placed on personal experience that must be regarded as anecdotal. The structure and mechanics of the cervical spine enable it to perform three important functions. First, it supports the head as well as the variety of soft tissue structures of the neck. Second, by virtue of segmentation and configuration, it permits multiplanar motion of the head. Third, and most important, it serves as a protective conduit for the spinal cord and cervical nerve roots. A condition that impedes or prevents the performance of any of the three functions in a pain-free manner either immediately or in the future is unacceptable and constitutes a contraindication to participation in contact sports. The following proposed criteria for return to contact activities in the presence of cervical spine abnormalities or following injury are intended as guidelines only. It is fully acknowledged that for the most part they are at best

A

­ redicated on anecdotal experience, and no responsibility p can be assumed for their implementation. Critical to the application of these guidelines is the implementation of coaching and playing techniques that preclude the use of the head as the initial point of contact in a collision situation. Exposure of the cervical spine to axial loading is an invitation to disaster and makes all safety standards meaningless.

Congenital Conditions Odontoid Anomalies Hensinger91 has stated that “patients with congenital anomalies of the odontoid are leading a precarious existence. The concern is that a trivial insult superimposed on already weakened or compromised structure may be catastrophic.” This concern became a reality during the 1989 football season when an 18-year-old high school player was rendered a respiratory-dependent quadriplegic while making a head tackle that was vividly demonstrated on the game video. Postinjury roentgenograms revealed an os odontoideum with marked C1-C2 instability (Fig. 16A1-38). Thus, odontoid agenesis, odontoid hypoplasia, and os odontoideum are all absolute contraindications to participation in contact activities.

B

Figure 16A1-38   Inherent instability at C1 in a patient with an os odontoideum. This condition resulted in respiratory-dependent quadriplegia following a spear tackle by this 18-year-old high school football player. The reduction in the space available for the cord is vividly demonstrated by the lateral extension (A) and flexion (B) postinjury views. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injuries. Clin J Sport Med 1:12–27, 1991.)

Spinal Injuries

Spina Bifida Occulta This is a rare, incidental roentgenographic finding that presents no contraindication.

Atlanto-occipital Fusion This rare condition is characterized by partial or complete congenital fusion of the bony ring of the atlas to the base of the occiput. Signs and symptoms are referable to the posterior columns as a result of cord compression by the posterior lip of the foramen magnum and usually occur in the third or fourth decade. They usually begin insidiously and progress slowly, but sudden onset or instant death has been reported. Atlanto-occipital fusion as an isolated entity or coexisting with other abnormalities constitutes an absolute contraindication to participation in contact activities.

Klippel-Feil Anomaly This eponym is applied to congenital fusion of two or more cervical vertebrae. For purposes of this discussion, the variety of abnormalities can be divided into two groups: type I—mass fusion of the cervical and upper thoracic vertebrae (Fig. 16A1-39); and type II—fusion of only one or two interspaces (Fig. 16A1-40). To be noted, a variety of congenital problems have been associated with ­congenital fusion of the cervical vertebrae and include pulmonary, cardiovascular, and urogenital problems. Pizzutillo92 has pointed out that “children with congenital fusion of the

Figure 16A1-39   Type I Klippel-Feil deformity with multiple-level fusions and deformities as demonstrated on the lateral roentgenogram. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injuries. Clin J Sport Med 1:12–27, 1991.)

693

c­ ervical spine rarely develop neurologic problems or signs of instability. However, he further states that “the literature reveals more than 90 cases of neurologic problems … that developed as a consequence of occipital cervical anomalies, late instability, disk disease, or degenerative joint disease.” These reports included cervical radiculopathy, spasticity, pain, quadriplegia, and sudden death. Also, “more than two thirds of the neurologically involved patients had single level fusion of the upper area, whereas many cervical patients with extension fusions of five to seven levels had no associated neurologic loss.” Despite this, a type I lesion, a mass fusion, constitutes an absolute contraindication to participation in contact sports. A type II lesion with fusion of one or two interspaces with associated limited motion or associated occipitocervical anomalies, involvement of C2, instability, disk disease, or degenerative changes also constitutes an absolute contraindication to participation. On the other hand, a type II lesion involving fusion of one or two interspaces at C3 and below in an individual with a full cervical range of motion and an absence of occipitocervical anomalies, instability, disk disease, or degenerative changes should present no contraindication.

Developmental Conditions Developmental Narrowing (Stenosis) of the Cervical Spinal Canal This condition and its association with CCN and transient quadriplegia has been well defined.47,52,88 The definition of narrowing or stenosis as a cervical segment with one

Figure 16A1-40   Type II Klippel-Feil deformity with a   one-level congenital fusion at C3-C4 involving both the vertebral bodies and the lateral masses. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

694

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

or more vertebrae that have a SC/VB ratio of 0.8 or less is predicated on the fact that 100% of all reported clinical cases have fallen below this value at one or more levels. To be noted, 12% of asymptomatic controls also fell below the 0.8 level, as did 32% of asymptomatic professional and 34% of asymptomatic college players. In the group of reported symptomatic players, there was in every instance complete return of neurologic function, and in those who continued with contact activities, recurrence was not ­predictable. Clearly, the presence of developmental narrowing of the cervical spinal canal does not predispose to permanent neurologic injury. Eismont and colleagues have indicated, on the basis of experience of cervical fractures or dislocations resulting from automobile crashes, that the degree

of neurologic impairment was inversely related to the anteroposterior diameter of the canal.93 As a result of the all-or-nothing pattern of axial load football spine injuries, this phenomenon has not been observed in sports-related injuries. The presence of a SC/VB ratio of 0.8 or less is not a contraindication to participation in contact activities in asymptomatic individuals. I further recommend against preparticipation screening roentgenograms in asympto­ matic players. Such studies will not contribute to safety, are not cost effective, and will only contribute to the hysteria surrounding this issue. In individuals with a ratio of 0.8 or less who experience either motor or sensory manifestations of CCN, there is a relative contraindication to return to contact activities. In these instances, each case must be determined on an individual basis depending on the understanding of the player and his or her parents and their willingness to accept any presumed theoretical risk (Fig. 16A1-41). An absolute contraindication to continued participation applies to those individuals who experience a documented episode of CCN associated with ligamentous instability, MRI evidence of cord defects or swelling, symptoms or positive neurologic findings lasting more than 36 hours, or more than one recurrence.

Spear Tackler’s Spine

Figure 16A1-41   The ratio of the spinal canal to the vertebral body is the distance from the midpoint of the posterior aspect of the vertebral body to the nearest point on the corresponding spinolaminar line divided by the anteroposterior width of the vertebral body. A ratio of less than 0.8 indicates the presence of developmental narrowing (stenosis). Lateral roentgenogram of a 20-year-old intercollegiate football player who had one episode of transient quadriplegia that lasted 10 minutes following a hyperflexion injury. The canal “to” vertebral body ratios are narrow from C3 through C7. Specifically, the ratio at C4 measures 0.6. This player returned to active playing for two seasons without a recurrence. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Analysis of material more recently received by the National Football Head and Neck Injury Registry has identified a subset of football players with “spear tackler’s spine.”94 The entity consists of (1) developmental narrowing (stenosis) of the cervical canal; (2) persistent straightening or reversal of the normal cervical lordotic curve on an erect lateral roentgenogram obtained in the neutral position; (3) concomitant preexisting post-traumatic roentgenographic abnormalities of the cervical spine; and (4) documentation of the individual employing spear tackling technique (Fig. 16A1-42). In two instances in which preinjury roentgenograms and video documentation of axial loading of the spine due to spear tackling were available, a C3-C4 bilateral facet dislocation resulted in one and C4-C5 fracture-dislocation in the other, both players being rendered quadriplegic. It is postulated that the straightened “segmented column” alignment of the cervical spine, combined with head-first tackling techniques, predisposed these individuals to an axial loading injury of the cervical segment. Thus, this combination of factors constitutes an absolute contraindication to further participation in collision sports.

Traumatic Conditions of the Upper Cervical Spine (C1-C2) The anatomy and mechanics of the C1-C2 segments of the cervical spine differ markedly from those of the middle or lower segments.95 Lesions with any degree of occipital or atlantoaxial instability portend a potentially grave prognosis (Fig. 16A1-43). Thus, almost all injuries involving the upper cervical segment that involve a fracture or ligamentous laxity are an absolute contraindication to further participation in contact activities (Fig. 16A1-44). Healed,

Spinal Injuries

A

695

B

Figure 16A1-42   Roentgenograms of a 19-year-old intercollegiate linebacker with spear tackler’s spine. A, On the anteroposterior view, the cervical spine is noted to be tilted toward his left. This represents a wry neck attitude frequently seen in those with either acute or chronic cervical injury. B, The lateral view demonstrates several manifestations of spear tackler’s spine: (1) a cervical kyphosis, (2) developmental narrowing of the cervical canal, and (3) an old compression injury of C5. The kyphotic deformity was fixed in both flexion and extension. He subsequently sustained a bilateral C3-C4 facet dislocation and was rendered quadriplegic as a result of spear tackling.

A

B

Figure 16A1-43   The atlas-dens interval (ADI) is the distance on the lateral roentgenogram between the anterior aspect of the dens and the posterior aspect of the anterior ring of the atlas. In children, the ADI should not exceed 4.0 mm, whereas the upper limit in the normal adult is less than 3.0 mm. C1-C2 instability is vividly demonstrated in the extension (A) and flexion (B) views. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

696

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 16A1-44   A, Lateral roentgenogram of the cervical spine in the erect neutral position of a 21-year-old college football player demonstrates anterior translation of C6 on C7 of greater than 3.5 mm (arrows). B, A computed tomographic scan of C6 in the axial plane demonstrates a fracture through the lateral mass (arrow). Persistent displacement despite healing of the fracture constitutes an absolute contraindication to further participation in contact sports. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

nondisplaced Jefferson fractures, healed type I and type II odontoid fractures, and healed lateral mass fractures of C2 constitute relative contraindications provided the patient is pain free, has a full range of cervical motion, and has no neurologic findings.35 Because of the uncertainty of the results of cervical fusion, the gracile configuration of C1, and the importance of the alar and transverse odontoid ligaments, fusion for instability of the upper segment constitutes an absolute contraindication regardless of how successful the fusion appears roentgenographically.

Traumatic Conditions of the Middle and Lower Cervical Spine Ligamentous Injuries The criteria of White, Punjabi, and colleagues for defining clinical instability were intended to help establish indications for surgical stabilization (see Figs. 16A1-13 and 16A1-14).29,30 However, although the limits of displacement and angulation correlated with disruption of known structures, no one determinant was considered absolute. In view of the observations of Albright and colleagues that 10% (7 of 75) of the college freshmen in their study demonstrated abnormal motion, “as well as on the basis of our own experience, it appears that in many instances some degree of minor instability” exists in populations of both high school and college football players without apparently leading to adverse effects. The question, of course, is, what are the upper limits of “minor” instability? Unfortunately, there are no data available relating this question to

Figure 16A1-45   Lateral roentgenogram of the cervical spine taken in the erect neutral position demonstrates an anterosuperior compression defect in the vertebral body of C5 (arrow). This limbus deformity resulted from a previous compression injury to the ring epiphysis. There is no evidence of angulation, displacement, or instability of the spine. Such a radiographic finding would not constitute a contraindication to further participation. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Spinal Injuries

the clinical situation that allow reliable standards. Clearly, however, lateral roentgenograms that demonstrate more than 3.5 mm of horizontal displacement of either vertebra in relation to another or more than 11 degrees of rotation than either adjacent vertebra represent an absolute contraindication to further participation in contact activities. With regard to lesser degrees of displacement or rotation, further participation in sports enters the realm of “trial by battle,” and such situations can be considered a relative contraindication depending on such factors as level of performance, physical habits, position played (i.e., interior lineman or defensive back), and so on.

2. A healed stable end-plate fracture without a sagittal component on anteroposterior roentgenograms or involvement of the posterior or bony ligamentous structures (see Fig. 16A1-22) 3. Healed spinous process “clay shoveler” fractures Relative contraindications apply to the following healed stable fractures in individuals who are asymptomatic and neurologically normal and have a full pain-free range of cervical motion: 1. Stable undisplaced vertebral body compression fractures without a sagittal component on anteroposterior roentgenograms: The propensity for these fractures to settle, causing increased deformity, must be considered and carefully followed (Fig. 16A1-46). 2. Healed stable fractures involving the elements of the posterior neural ring in individuals who are asympto­ matic, neurologically normal, and have a full pain-free range of cervical motion (Fig. 16A1-47): In evaluating radiographic and imaging studies to find the location and subsequent healing of a posterior neural ring fracture, one must understand that a rigid ring cannot break in one location.73 Thus, healing of paired fractures of the ring must be demonstrated.

Fractures The following healed stable fractures in an asymptomatic patient who is neurologically normal and has a full range of cervical motion can be considered to present no contraindication to participation in contact activities: 1. Stable compression fractures of the vertebral body without a sagittal component on anteroposterior roentgenograms and without involvement of either the ligamentous or the posterior bony structures (Fig. 16A1-45)

A

697

B

Figure 16A1-46   A, Lateral roentgenogram of the cervical spine taken while the patient was in a cervical brace demonstrates undisplaced compression fracture of the vertebral body of C5. Notable is the fact that there is no associated angulation, displacement, intervertebral disk space narrowing, facet incongruity, or fanning of the spinous processes. B, Lateral flexion view demonstrates pathologic angulation as defined by White and associates. There is no translation, disk space narrowing, facet incongruity, or fanning of the spinous processes, suggesting a stable lesion. The increased angulation is attributed to the deformity of the vertebral body. Assuming that no progression of the deformity or evidence of instability occurred and that the patient had a pain-free neck with a normal range of motion, this situation would constitute a relative contraindication to participation in contact activities depending on the player’s level, position, and willingness to accept risk for reinjury. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

698

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 16A1-47   A, Computed tomographic scan of a vertebral neural arch in the transverse plane demonstrating a hairline fracture through the lateral mass (open arrow) as well as a more evident nondisplaced fracture through the ipsilateral lamina (closed arrow). B, The patient was treated in a halo brace with satisfactory evidence of healing as demonstrated on computed tomographic scan. Following immobilization and the return of normal pain-free motion, he was permitted to return to contact activity after rehabilitation was fully effected and pain-free cervical range of motion and paravertebral muscle strength returned. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Absolute contraindications to further participation in contact activities exist in the presence of the following fractures: 1. An acute fracture of either the body or posterior elements with or without associated ligamentous laxity 2. Vertebral body fracture with a sagittal component (see Fig. 16A1-25) 3. Fracture of the vertebral body with or without displacement with associated posterior arch fractures or ­ligamentous laxity (Fig. 16A1-48) 4. Comminuted fractures of the vertebral body with displacement into the spinal canal 5. Any healed fracture of either the vertebral body or the posterior components with associated pain, neurologic findings, and limitation of normal cervical motion 6. Healed displaced fractures involving the lateral masses with resulting facet incongruity

Intervertebral Disk Injury There is no contraindication to participation in contact activities in individuals with a healed anterior or lateral disk herniation treated conservatively (Fig. 16A1-49) or in those requiring an intervertebral diskectomy and interbody fusion for a lateral or central herniation who have a solid fusion, are asymptomatic and neurologically negative, and have a full pain-free range of motion. A relative contraindication exists in individuals with either conservatively or surgically treated disk disease with residual facet instability. An absolute contraindication exists in those with an acute or chronic “hard disk”

Figure 16A1-48   Lateral roentgenograms of the cervical spine in the erect neutral position demonstrate an anterosuperior compression defect in the vertebral body of C6. In addition, there is fanning of the C5-C6 spinous process, indicating posterior instability due to disruption of the intraspinous and posterior longitudinal ligaments (arrows). This situation constitutes an absolute contraindication to contact sports. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Spinal Injuries

699

­ erniation with associated neurologic findings, pain, or h significant limitation of cervical motion (Fig. 16A1-50).

Status after Cervical Spine Fusion A stable one-level anterior or posterior fusion in a patient who is asymptomatic, neurologically negative, and pain free and has a normal range of cervical motion presents no contraindication to continued participation in contact activities (Fig. 16A1-51). Individuals with a stable two- or three-level fusion who are asymptomatic and neurologically negative and have a pain-free full range of cervical motion have a relative contraindication. Because of the presumed increased stresses at the articulations of the adjacent uninvolved vertebrae and the propensity for development of degenerative changes at these levels, it appears that this patient is the rare exception who should be permitted to continue contact activities (Fig. 16A1-52). In individuals with more than a three-level anterior or posterior fusion, an absolute contraindication exists as far as continued participation in contact activities (Fig. 16A1-53).

Figure 16A1-49   Magnetic resonance sagittal image of the cervical spine in a 17-year-old high school football player with a history of neck injury. An anterior intervertebral disk herniation with disk space changes ����������������������������������������� is seen ��������������������������������� at the C5-C6 level. At the time of follow-up examination, the youngster was asymptomatic and neurologically normal and had a pain-free range of cervical motion. He was permitted to return to contact activities. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Figure 16A1-50   A sagittal magnetic resonance image of the cervical spine of a 17-year-old high school football player who complained of posterior neck pain while butt blocking as well as a right unilateral transient radiculopathy or “burner.”   Visualized are intervertebral disk herniations at C4-C5 and   C5-C6 that are indenting the spinal cord at both levels (arrows). Although the neurologic examination was normal, the presence of a wry neck attitude, limited neck extension, congenital canal narrowing (stenosis), and reversal of the normal cervical lordosis on roentgenogram precluded the individual from participation in contact sports. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Figure 16A1-51   Lateral roentgenogram of a 28-year-old professional ice hockey player who underwent a successful onelevel interbody fusion at C5-C6 for instability. He subsequently played 2 years without a problem. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

700

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

C l Managing

Figure 16A1-52   Lateral roentgenogram of the cervical spine of a 28-year-old former professional football player who had undergone a C4-C5-C6 posterior fusion for a posttraumatic instability. He subsequently returned to play 2 years of professional football; however, he developed stiffness, neck discomfort, and limited motion. The individual who elects to return to contact activities following more than a two-level fusion must understand that the probability of symptoms resulting from degenerative changes at the articulations above and below the fusion is increased. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

Figure 16A1-53   Lateral roentgenogram of an 18-yearold who had injured his neck playing football when he was 13 years old. At that time, a three-level posterior fusion and wiring was performed; however, it appears that periosteal stripping of adjacent vertebrae above and below resulted in a five-level fusion. Such a situation is an absolute contraindication to participation in contact activities. (From Torg JS, Glasgow SG: Criteria for return to contact activities following cervical spine injury. Clin J Sport Med 1:12–27, 1991.)

r i t i c a l

P

o i n t s

the unconscious or spine-injured athlete requires planning and preparation. All necessary equipment should be readily accessible and in good operating condition and all personnel trained in its proper use. Onthe-job training in an emergency situation is inefficient at best and a potential catastrophe at worst. l The “burner” pain syndrome or “stinger” results from two distinct injury patterns. Brachial plexus traction injuries, commonly seen in younger athletes, result from ipsilateral shoulder depression and contralateral neck flexion. Cervical root injuries, typically occurring in older players, are due to cervical hyperextension associated with degenerative disk changes, often in combination with developmental cervical stenosis. Criteria for return to athletic participation include absence of symptoms, normal strength, and painless full range of cervical spine motion. l Intervertebral disk injury without frank herniation and neurologic findings associated with a history of injury, neck pain, and limited cervical motion are frequently seen in collision activities. MRI frequently demonstrates disk bulge without herniation. Management is conservative, with return to activity after the individual is asymptomatic and has a full range of pain-free cervical motion. l The principles of spinal cord resuscitation include (1) protection from further injury; (2) administration of methyl prednisolone; (3) expeditious reduction; (4) rapid and secure spinal stabilization; and (5) implementation of an early rehabilitation program. l Anatomic variation and response to injury best define the cervical spine into upper (C1-C2), middle (C3-C4), and lower (C4-C7) segments. l Clues indicative of a serious cervical spine injury in the ambulatory patient are presence of a wry neck or torticollis posture, limitation of volitional cervical motion, and presence of paravertebral muscle atrophy. An individual presenting with a history of trauma and one or more of these findings requires a careful neurologic examination and appropriate imaging studies. l Cervical spinal stenosis with cervical cord neurapraxia is an acute transient neurologic episode of cervical cord origin with sensory changes that may be associated with motor paresis involving both arms, both legs, or all four extremities after forced hyperextension, hyperflexion, or axial loading of the cervical spine. Such episodes are transient, usually lasting 10 to 15 minutes, and except for paresthesia, neck pain is not present. There is complete return of motor function and full pain-free cervical motion. Routine radiographs characteristically demonstrate cervical spinal stenosis. l Available data indicate that individuals who have an episode of CCN with or without an episode of transient quadriplegia are not predisposed to permanent neurologic injury. The problem, however, is the propensity for reoccurrence, which is predictable. l The primary mechanism for athletic injuries of the cervical spine is axial loading, which occurs in American football when an individual strikes an opponent with the top or crown of the helmet. The implementation of rules changes and coaching techniques eliminating the use of

Spinal Injuries

the head as a battering ram has resulted in a dramatic reduction in the incidence of quadriplegia. l Criteria for return to athletic activity have been categorized as no contraindication, relative contraindication, and absolute contraindication depending on availability or lack of available risk data. As a general rule, return to activity requires that the individual have a stable cervical spine, be pain free and neurologically negative, and have a full range of cervical motion.

S U G G E S T E D

R E A D I N G S

Boden BP, Tacchetti RL, Cantu RC, et al: Catastrophic cervical spine injuries in high school and college football players. Am J Sports Med 34��������������������������������� -8:1223-1232, 2006. Koffler KM, Kelly JD: Neurovascular trauma in athletes. Orthop Clin N Am 33:523-534, 2002. Kwon BK, Vaccaro AR, Grauer JN, et al: Subaxial cervical spine trauma. J Am Acad Orthop Surg 14:78-79, 2006.

701

Spivak JM, Connolly PJ (eds): Orthopedic Knowledge Update, 3rd ed. Rosemont, Ill, American Academy of Orthopedic Surgeons, 2006. Torg JS, Corcoran TA, Thibault LE, et al: Cervical cord neurapraxia: Classification, pathomechanics, morbidity, and management guidelines. J Neurosurg 87:743850, 1997. Torg JS, Guille JT, Jaffe S: Current concept review: Injuries to the cervical spine in American football players. J Bone Joint Surg Am 84:112-122, 2002. Torg JS, Thibault L, Sennett B, et al: The pathomechanics and pathophysiology of reversible, incompletely reversible and irreversible cervical spinal cord injury. Clin Orthop 321:259-269, 1995. Torg JS, Vegso JJ O’Neill J, Sennett B: The epidemiologic, pathologic, biomechanical and cinematographic analysis of football-induced cervical spine trauma. Am J Sports Med 18:50-57, 1990.

R eferences Please see www.expertconsult.com

S ecti o n

A

Cervical Spine Injuries 2. Cervical Spine Injuries in the Child Peter D. Pizzutillo and Martin J. Herman

The pursuit of excellence in athletic endeavors has resulted in the development of highly effective training programs for adolescent athletes that have made them bigger, stronger, and faster than young athletes of a generation ago. The emphasis on winning is generated by pressures within the athlete and is enhanced by peers, parents, coaches, and society in general. Society likes “a winner.” Tremendous pressures are placed on young athletes to perform well not only for the satisfaction enjoyed in sports but also as a passport to a lifetime of success. This milieu has led to significant medical problems in athletes with the popularization of steroids, psychological turmoil due to unrealistic expectations, and the “burnout” phenomenon. Competitive behavior has become more aggressive and physical. Even a highly skilled finesse sport such as basketball is now played as a contact sport. Football and ice hockey have evolved from contact sports to the level of collision sports. The result is an increased incidence of injury in young athletes. Forty-four percent of injuries sustained in students 14 years of age and older are due to sports activity.1 In a high school survey conducted by Paulson,1 80 of 100 participants in football sustained an injury

during the playing season. This compares with 75 of 100 participants in wrestling, 44 of 100 participants in softball, 40 of 100 female participants in gymnastics, 28 of 100 male participants in gymnastics, 35 females and 29 males of 100 participants in track and cross-country, 31 of 100 male participants in basketball, 30 of 100 participants in soccer, and 18 of 100 male participants in baseball. This survey reflects all levels of severity; however, it is significant that 7% of high school teenagers were hospitalized as the result of sports injuries. Football injuries accounted for 20% of these cases, and basketball injuries accounted for 17.4%. Spinal injuries occur during sports activities but are less common than other musculoskeletal injuries. Although catastrophic injuries do occur, most injuries are minor soft tissue contusions, sprains, and paraspinal muscle strains. In a review of all spinal injuries in 406 children admitted to a level-one pediatric trauma center, sports was the most common mechanism of injury for children aged 10 to 14 years2; 68% of children had paravertebral soft tissue injuries. In a review of consecutive children treated at another pediatric level-one trauma center,3 27% of cervical spine injuries were sports-related; most of these ­injuries

702

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

occurred in adolescent boys and were isolated injuries. Analysis of the National Pediatric Trauma Registry from 1994 to 1999 for all cervical fractures, dislocations, and spinal cord injuries without radiographic abnormalities (SCIWORA) revealed that 66 of 408 children sustained these injuries during sports.4 Catastrophic spinal cord injury from cervical spine trauma is rare in children and adolescents. Although motor vehicle crashes are the most common mechanism, spinal cord injuries occur during sports activities,3,4 particularly collision sports and diving. A review of patients with spinal cord injury at several regional rehabilitation centers revealed that 10% to 20% of spinal cord injuries were due to sports-related accidents.5,6 From 1950 to 1989, 90% of fatal football injuries involved head or neck injuries.7 The modern helmet and facemask were developed in the 1950s and 1960s, and these devices led to increased use of the head in blocking and tackling techniques, resulting in an increase in deaths from head injury.8 In the 1970s, helmets were improved to protect the brain. A subsequent decrease in fatalities due to head injury followed, but the level of cervical spine injuries was sustained, primarily due to “spearing” techniques in tackling.9 In January 1976, the National Collegiate Athletic Association (NCAA) and the National Federation of State High School Associations (NFSHSA) formally adopted high school, college, and coaching rules that prohibited tackling or blocking with a helmeted head because of the vulnerability of the cervical spine to injury in this position. This single step has resulted in a significant decrease in serious cervical spine injury.10 Since 1977, there has been one fatality for every 10 million athlete exposures in football at the high school, college, and professional levels. The incidence of nonfatal but catastrophic injuries is difficult to report reliably because of incomplete recording of injuries.11,12 The National Football Head and Neck Injury Registry has been functional since 1955 and has provided much of the information that is used today for analysis. The decreasing rate of cervical spine injury in football is attributed to improvements in equipment, changes in game rules that better protect the athlete, more effective conditioning of the athlete, and better coaching of basic playing techniques, especially blocking and tackling. Similar developments are necessary to reverse the increasing incidence of cervical spine injury in other sports. Severe head and spinal injuries occur in other collision sports as well as in noncontact sports activities. Ice hockey is another collision sport in which severe head and spinal injuries occur.13,14 Between 1943 and 1999, 271 major spinal injuries were reported in Canadian ice hockey players, with 49% of these occurring in players 16 to 20 years of age.15 The most common mechanisms were impact with the boards and checking or pushing from behind. Education of coaches and players and rules changes by organized hockey have reduced the incidence of these injuries in recent years. From 1978 to 1982, the National Registry of Gymnastic Catastrophic Injuries documented 20 incidents of injury, including 17 patients with permanent quadriplegia and 3 deaths.16 These injuries occurred in skilled performers during practice settings. Analysis of this group revealed that permanent spinal cord injury was closely associated with use of the trampoline,17 especially when attempting

to perform a somersault. In many states, the trampoline has now been banned from physical education classes and is used only with spotters and physical restraints in teaching new skills in gymnastics. Catastrophic neurotrauma involving the cervical spinal cord has also occurred in rugby,16,18,19 wrestling20 and diving.21-23 Diving injuries account for 4% to 14% of spinal cord injuries in young patients24; most of these occur outside of organized programs.21 Downhill skiing had a mortality rate of 1.7% in 430 patients reported from Lake Tahoe in a 14-year study.25 Thirteen patients in this group had permanent radiculopathy, and 4 had permanent myelopathy. Cervical spine injury in skiers is frequently associated with concurrent head injury. The martial arts have contributed to cervical spine fractures and dislocations, usually as the result of a forceful foot strike to the head or a fall onto the head and neck area. At least 17 deaths have been reported in judo and karate as a result of this mechanism.26 Interestingly, soccer26 and boxing27 have not been associated with a high incidence of cervical spine injury.

ANATOMY OF THE CERVICAL SPINE Cervical spine injury in children younger than 8 years of age is uncommon and differs from injury in older adolescents and adults by virtue of site and mechanism of injury.28,29 The problem of evaluation of the cervical spine in childhood is complicated by the fact that much of the cervical spine is unossified and is undergoing progressive radiographic changes as ossification and growth proceed. By 8 years of age, the cervical spine has developed the adult configuration. Before the age of 1 year, the anterior ring of C1 is unossified, and it may be difficult to determine whether the upper cervical spine is unstable (Fig. 16A2-1). Between 3 and 6 years of age, the basilar synchondrosis becomes visible and may be mistaken for fracture at the base of the odontoid. By 6 years of age, the inner diameter of the spinal canal of the entire cervical spine has reached the adult level. In the child younger than 8 years, extension of the spine causes a spurious impression of subluxation of the anterior arch of the atlas over the superior aspect of the dens, which is not yet ossified. From infancy to 8 years of age, lateral neutral radiographs of the cervical spine reveal an increase in the angulation of the facet joint from 30 degrees to 60 degrees. In the younger child with a facet joint angle of 30 degrees, a greater degree of freedom in flexion and extension exists that may contribute to the appearance of pseudosubluxation commonly seen at the C2-C3 and C3-C4 levels. In the first decade of life, flexion-extension lateral radiographs of the cervical spine may reveal an atlantodens interval up to 5 mm, whereas the adult interval should not exceed 3.5 mm. Incomplete ossification of the cervical spine creates the appearance both of a truncated odontoid, until its tip ossifies at 10 to 12 years of age, and of apparent wedging of the vertebral bodies on lateral radiographs until ossification is more complete at 10 years of age. Whereas most fractures in adults occur in the lower cervical spine, the upper cervical spine is involved in up to 70% of cervical spine fractures in children. The relatively

Spinal Injuries

Figure 16A2-1   Lateral radiograph of the immature cervical spine reveals absence of the anterior arch of C1, presence of the basilar synchondrosis, and apparent wedging due to unossified vertebral bodies.

large size of the child’s skull may be a significant factor in injury of the upper cervical spine in this age group. A marked differential in elasticity between the spinal column and the spinal cord has been identified in the young child. The clinical expression of this differential in elasticity has been popularized by Pang and Wilberger30 in their report of SCIWORA in children. Pang and Wilberger’s study documented the presence of serious neurologic damage of the upper cervical cord in the absence of cervical spine osseous damage on initial radiographic evaluation in children younger than 8 years.

RECOGNITION AND PRIMARY TREATMENT The lifelong consequences of catastrophic spinal cord injury are of such magnitude that it is imperative that personnel dealing with athletes on a regular basis be well educated about the possibility of injury during practice and game conditions. Education alone can increase awareness of this problem and may indeed spare the injured athlete from further neural damage caused by mismanagement on the field. There is considerable difficulty in maintaining a high index of suspicion for spinal injury because severe neck injury does not frequently occur. The very mention of spinal cord injury creates an immediate emotional response among athletes, coaching staff, and the athletes’ families. In competitive conditions, whether in practice or in actual game situations, it can be quite difficult to evaluate the injured athlete adequately

703

and manage problems on the field. Preparation and education of everyone involved in the athlete’s care in the event of nonfatal catastrophic spinal cord injury is of the utmost importance. A properly equipped ambulance with attendants trained in safe transport of individuals with neurologic damage and identification of hospital facilities that have the capability of dealing with catastrophic neurologic injury must exist well in advance of injury in order to provide the optimal environment for the injured athlete’s treatment. The development of regional spinal cord injury centers has provided a network of experienced staff throughout the country who can provide the very best of care for the spinal cord–injured athlete. Education of emergency medical technicians has significantly decreased the incidence of secondary neural injury that was formerly caused by improper immobilization during transport. Ongoing educational efforts are needed to maintain current proficiency and to improve our existing level of care. The management of the athlete with a severe spinal injury requires rapid assessment with protection of vital structures.31 If the athlete is unconscious, quadriparetic, or quadriplegic, or has significant paresthesias or dysesthesias involving the upper and lower extremities, the cervical spine must be considered unstable and must be protected.32 The head and neck should immediately be immobilized in a neutral position. If the patient is in the prone position, an organized logroll maneuver may be performed in which the head and neck are turned as one unit with the patient’s trunk. This can be managed by having one member of the emergency team control the head and neck while grasping the shoulder area in order to prevent changes in flexion or extension. In addition to level of consciousness, it is important to determine the patient’s respiratory and circulatory status. If the athlete uses a mouthpiece during sports activity, the mouthpiece should be removed. Football players should have their facemasks removed, but the helmet and chin strap should be left in place until the athlete is evaluated neurologically. If the patient is not breathing, it is important to position the jaw in an appropriate forward attitude to open the airway without overextending the neck. If the jaw thrust maneuver is not successful in restoring breathing, rescue breathing must be initiated. The athlete must be transported in an expedient manner but under safe conditions to an appropriately identified medical facility capable of dealing with these problems. In athletes younger than 8 years of age, care must be taken to avoid the forced flexion of the cervical spine that occurs on a flat spine board because of the relative increased size of the head in relation to the size of the chest.33 A standard flat board can be used with a towel roll beneath the shoulders to create a more neutral position of the cervical spine. The minimal components provided by the evaluating facility should include a neurosurgeon, an orthopaedic surgeon, and adequate radiographic capability.

ACUTE SOFT TISSUE INJURY Acute soft tissue injury of the cervical spine may involve the disk, ligaments, muscle, and fascia. Typically, these injuries are the result of a collision or fall onto the head and neck

704

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

complex. The athlete usually complains of neck pain or neck and shoulder pain without distal radiation of pain or paresthesias. Physical examination reveals a limited range of motion of the cervical spine, usually in the presence of mild to moderate paraspinal muscle spasms, with no evidence of motor, sensory, or reflex changes in the upper or lower extremities. Radiographs of the cervical spine are normal and show no evidence of subluxation or dislocation but may reveal straightening of the cervical lordosis. Acute injury involving the fascia, muscle, or ligaments of the neck without disruption and instability should be treated symptomatically. A rehabilitation program comprising range of motion exercises and restitution of strength of the neck and shoulder girdle is important before gradual resumption of sports activity. The painful phase of soft tissue injury usually does not last more than 5 to 10 days and will allow the athlete to resume sports activity. If rehabilitation is not included as an integral part of the return-to-play program, the athlete will demonstrate a chronic decrease in range of motion of the cervical spine and diminished strength of the neck, especially in the flexor muscle group. Limitation of motion and weakness of the neck may lead to secondary injury with low-grade fascial, muscular, or ligamentous injury that perpetuates a vicious cycle of disability. Effective treatment of chronic cervical sprains and strains, therefore, includes a rehabilitation program designed to stretch out contracture of the cervical soft tissues and reconstitute the strength of the surrounding cervical and shoulder musculature. It is extremely important that children and adolescents who have sustained apparent innocuous injury to the cervical spine be re-evaluated on a serial basis. Herkowitz and Rothman34 reported development of instability of the cervical spine in individuals who initially demonstrated no radiographic evidence of bony or soft tissue abnormality. Subacute instability of the cervical spine is due to elastic and plastic deformation of the ligamentous and disk structures and may result in neurologic deficits in individuals who were initially neurologically normal. Children in the first decade of life who sustain neck injuries but appear to be normal by radiographic and neurologic testing need careful follow-up. Pang and Wilberger’s report30 primarily involved victims of vehicular injury and included only four sports injuries, but it demonstrated that 52% of patients with SCIWORA experienced the onset of serious neurologic problems an average of 4 days after their initial injury. Pollack and colleagues35 reviewed 42 children with spinal cord injury and found that within 10 weeks of the first injury, 8 children had a second spinal cord injury with more serious neurologic consequences; central or partial cord injury occurred in all 8, and 3 patients had severe quadriparesis or paraparesis. Pollack and colleagues proposed an arbitrary protocol that includes immobilization of the cervical spine in a brace for 3 months with no sports activity, close clinical follow-up, and repeat somatosensory evoked potentials (SSEPs) at 6 weeks. If dynamic radiographic studies and physical examination of the cervical spine are normal at the 3-month follow-up examination, the individual is ready to begin the return to sports. Full range of motion of the neck with demonstrated stability of the cervical spine on flexion-extension lateral radiographs and the absence of sensory or motor loss are required

before the athlete is allowed to return to competitive sports activity. Acute herniation of the cervical nucleus pulposus has been reported in adult sports activity36 but is rare in the child or adolescent athlete. Its presence can result in catastrophic neurologic injury with compromise of the anterior spinal cord.37 These patients experience a sudden onset of neck pain with radiation to both shoulders, arms, and hands, and they tend to hold the head tilted to the side of the disk lesion. Interscapular pain is commonly reported. When the head is tilted to the side of the lesion and then extended, there is an increase in pain. Herniation of the cervical disk most commonly occurs at the C5-C6 and C6-C7 levels. The immediate concern is to differentiate the acute herniated disk from the “burner lesion” that results in searing pain in a radicular distribution. A detailed neurologic assessment and an appropriate radiographic evaluation including computed tomography (CT) and magnetic resonance imaging (MRI) are usually required. Treatment of acute disk herniation in adolescent athletes requires decompression of the spinal canal. Repetitive axial compression of the cervical spine may result in chronic changes involving the disk. Albright and colleagues38,39 reported radiographic evidence of neck injury in 32% of freshman college football players in their preseason evaluations. Half of this study group had a past medical history of neck pain and showed abnormal radiographic findings involving the cervical spine. Linebackers and defensive backs were most commonly involved; running backs and wide receivers were at greater risk than linemen. Among athletes in whom the preseason physical examination or past medical history suggested a cervical spine problem, half demonstrated radiographic abnormalities of the cervical spine involving disk degeneration. Most of the athletes were unaware of any significant neck problems and had not sought prior medical evaluation. Axial loading appears to be the most important injury of the cervical spine. Torg and colleagues40 demonstrated that axial forces transmitted to the cervical spine in slight extension are dissipated primarily by the cervical muscles. When the neck is flexed 30 degrees, it becomes a straight segmented column. Axial forces applied under these conditions are transmitted directly to the vertebrae, ligaments, and disk rather than being dissipated by muscle. These observations have led to improvements in tackling and blocking techniques to reduce the frequency of cervical spine injury.

FRACTURES AND DISLOCATIONS Atlanto-occipital Instability Atlanto-occipital instability is usually the result of violent forces and is frequently fatal.41-43 Although atlanto-occipital instability has not been reported in sports injuries, it is conceivable that its incidence may be higher than suspected because appropriate diagnostic tests for spinal stability are not always conducted in acute fatalities, and there has not been much attention directed at the atlanto-occipital junction in the past.

Spinal Injuries

Athletes with Down syndrome are of special concern. The recent observation that individuals with Down syndrome may have atlanto-occipital hypermobility that excludes them from contact sports and from axial loading sports is important. In addition to the more commonly reported atlantoaxial instability, atlanto-occipital instability must be ruled out before athletes with Down syndrome can be medically cleared for sports activities (Fig. 16A2-2).

Jefferson Fracture When a high axial load is delivered from the apex of the skull to the cervical spine, tremendous forces are generated at the junction of the occipital condyle and the ring of the atlas.44 Low-level forces result in fractures of the posterior atlantal arch that are stable and can be successfully treated with immobilization by orthotics. When severe force is applied, a burst fracture of the atlas involving disruption of both anterior and posterior arches allows progressive displacement of the lateral masses of the atlas, producing consequent vascular and neurologic compromise. Most patients who have sustained injuries to the ring of the atlas are neurologically intact and must be evaluated in an expedient manner to avoid delay in diagnosis and secondary neurologic compromise. Patients with an injury of the ring of the atlas complain of neck pain and have severely restricted motion of the cervical spine in flexion, extension, lateral side bending, and lateral rotation. In the presence of a normal neurologic examination, a high index of suspicion of fracture of

Figure 16A2-2   Lateral radiograph of the upper cervical spine reveals significant anterior translation (arrows) of the occiput on the cervical axis.

705

the atlas is required when evaluating patients with a skull fracture or severe laceration of the scalp that suggests axial loading. Routine radiographs of the upper cervical spine are difficult to interpret, especially when the head is tilted in response to paravertebral muscle spasm. Detailed inspection of the ring of the atlas on both lateral and open mouth anteroposterior radiographs is required to determine the relationship between the lateral masses of the atlas and the axis. When the open mouth anteroposterior radiograph reveals combined overhang of the lateral masses of the atlas on the axis of more than 7 mm, instability and disruption of the transverse ligament must be assumed. computed tomographic imaging is a routine part of care of athletes with a suspected cervical spine fracture, especially when there is difficulty in obtaining adequate information from routine radiographs. Computed tomographic scans provide excellent detail in evaluation of an injury of the atlas (Fig. 16A2-3). Fractures of the anterior and posterior arches of the atlas, as well as the relationship of the odontoid to the anterior arch of the atlas, can be precisely evaluated on computed tomographic scans. Although most fractures of the atlas heal by nonoperative immobilization techniques, such as a halo brace, there is an occasional need for surgical fusion.45

Acute Atlantoaxial Instability Acute atlantoaxial instability is usually the result of severe flexion forces imposed on the cervical spine. If the atlantodens interval is greater than 5 mm in children, the transverse ligament is compromised and instability is present (Fig. 16A2-4). Posterior fusion of the atlas and axis is required to avoid spinal cord compression. Special concern exists about individuals with Down syndrome who are athletically active. The standard radiographic parameters of stability of the cervical spine in individuals without Down syndrome are not appropriate for judging stability in individuals with Down syndrome. Natural history studies indicate that one third of adults with Down syndrome demonstrate a radiographic appearance of instability at all levels of the cervical spine, but only 3% of these individuals experience neurologic problems. Caution must be exercised in evaluating individuals with Down syndrome to avoid undertreatment or overtreatment. Many children and adolescents with Down syndrome are actively involved in sports activities such as basketball, swimming, and horseback riding. Like other athletes, these individuals and their families derive a great deal of satisfaction, pride, and joy in their athletic accomplishments. A blanket prescription against sports involvement needlessly deprives these athletes of the sense of accomplishment that ­accompanies athletic endeavor and diminishes their selfesteem. On the other hand, children and adolescents with Down syndrome who demonstrate radiographic evidence of cervical instability should be advised against participation in sports activities that potentially endanger neural function, such as diving and gymnastics (Box 16A2-1). In the presence of neurologic dysfunction and radiographic cervical instability, surgical stabilization of the cervical spine is necessary to preserve existing neural function and to prevent progressive loss.

706

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

C Figure 16A2-3   A, Lateral radiograph reveals a break in the cortex (arrowhead) of the posterior arch of C1. B, Open mouth view reveals bilateral symmetrical overhang of the lateral masses of C1 on C2 (arrowheads) C, Computed tomography reveals a disruption in the ring of C1 in both the anterior and posterior arches.

Rotary Atlantoaxial Subluxation Children may demonstrate the insidious onset of wry neck deformity in association with posterior pharyngeal inflammation. Clinical examination of the involved child reveals a “cock-robin” attitude of the head with tilt of the head toward one shoulder and rotation of the chin toward the opposite shoulder. The patient demonstrates mild to moderate limitation of flexion and extension and nearly full lateral rotation to the side opposite the head tilt. In contrast, there is minimal lateral rotation toward the side of the head tilt. Cervical muscle spasm or localized soft tissue tenderness is usually absent. In addition, there is no prominence of the sternocleidomastoid muscle on the side of the head tilt, as is seen in congenital muscular torticollis. Most involved children are neurologically intact. Routine radiographs of the cervical spine should be supplemented by open mouth and lateral flexion-extension views of the upper cervical spine. Adequate lateral evaluation may be difficult to obtain if the radiology technician aligns the patient in the standard fashion for a lateral view

of the cervical spine because of the rotation and tilt of the skull and atlas. To avoid the confusing features caused by rotation and lateral tilt, the technician should be instructed to obtain a lateral radiograph of the skull to include the upper cervical spine. A lateral view of the skull using this technique will reveal a true lateral view of the atlas and permit more reliable interpretation of the relationship between the atlas and the axis. In the presence of malrotation, the lateral radiograph may spuriously suggest instability at the atlanto-occipital junction as well as at the atlantoaxial junction. In addition, the lateral mass of the atlas may appear anteriorly as a triangular wedge, the so-called sail sign (Fig. 16A2-5A). Lateral flexion-extension radiographs in neutral rotation are needed to reliably evaluate the degree of stability of the upper cervical spine as well as the existence of fixed rotary displacement between the atlas and axis. Computed tomographic scans have been extremely helpful in documenting the degree of displacement of the lateral mass of the atlas in relationship to the axis, the spatial relationship of the odontoid to the anterior arch of the atlas,

Spinal Injuries

707

Box 16A2-1 Recommendations for Activity Restrictions for the Athlete with Down Syndrome Based on Atlanto-Dens Interval*

1. ADI < 5 mm: no restrictions 2. ADI  =  5 mm to <10 mm and

physical examination ­ ormal: restrict from high-risk sports, including diving, n gymnastics, and equestrian events 3. ADI >10 mm: spinal fusion: no collision sports

*Measured on high-quality “true” lateral cervical spine radiograph or physician-directed fluoroscopic examination.

s­ tability is proved, the patient may be weaned to a soft cervical collar and begun on gentle range of motion exercises as well as isometric strengthening exercises. Repeat lateral flexion-extension radiographs of the cervical spine should be performed 6 weeks later to rule out recurrent instability. If instability persists following immobilization or if recurrent instability develops, posterior atlantoaxial surgical fusion is indicated.

Fracture of the Odontoid Figure 16A2-4   Lateral flexion radiograph of the cervical spine reveals significant anterior displacement of the ring of C1 from the odontoid in a patient with Down syndrome with hypoplastic odontoid.

and the space available for the cord dorsal to the odontoid (see Fig. 16A2-5B). Parke and colleagues46 demonstrated a rich network of sinusoidal vessels draining directly from the posterior pharynx to the soft tissues about the atlas and axis. During inflammatory states, such as those occurring with upper respiratory tract infection, hyperemia results in dissolution of the attachment of the transverse ligament to the anterior arch of the atlas. With progressive dissolution, gross instability occurs with loss of orientation of the ­ lateral masses of the atlas and axis. Treatment initiated before 4 weeks of clinical expression is successful in resolving rotary subluxation of the atlas and axis by nonsurgical methods. After 4 weeks, surgical stabilization is frequently required to maintain stability even when anatomic alignment can be regained by traction techniques (see Fig. 16A2-5C).47 Children and adolescents who present with a mild degree of rotary subluxation of the atlas and axis should be placed in a cervical collar and prohibited from recreational and sports activity. With more severe degrees of subluxation or fixed rotary subluxation, the patient should be protected and treated as an inpatient. Patients are initially treated by halter cervical spine traction or by halo traction in mild hyperextension and longitudinal traction. Once anatomic reduction has been obtained, the patient is immobilized either in a halo vest or a Minerva cast. After a 6- to 8-week period of immobilization, lateral flexion-extension radiographs of the cervical spine out of the cast or brace are needed to document stability. If

Fractures of the odontoid in children may be difficult to assess, especially in the presence of an unossified basilar synchondrosis. Acute separation of the odontoid through the basilar synchondrosis can occur in children younger than 7 years of age. Spontaneous reduction may occur, but marked widening of the retropharyngeal space is usually observed on radiographic evaluation of these patients.41 MRI evaluation may also reveal occult injury. With ossification of the ossiculum terminale, avulsion of the tip of the odontoid may be inadvertently suspected on radiographs. Lateral flexion-extension radiographs document the presence or absence of stability. In children older than 7 years, a type II odontoid fracture is more common and may be associated with nonunion as it is in adults. Most type II odontoid fractures heal through the use of nonsurgical techniques of immobilization such as halo vest stabilization but may require surgical fusion.48

Hangman’s Fracture The term hangman’s fracture refers to fractures involving the pedicle of the second cervical vertebra. These fractures are frequently the result of motor vehicle crashes or falls and do occur in children and adolescents.49 The most common mechanism of injury is that of extension and distraction, although other mechanisms have been suggested, including axial loading in extension and flexion. With bilateral disruption of the axial pedicles, the atlas and the anterior elements of the axis move as a single unit in flexion and extension. Schneider described anterior displacement of the “cervicocranium” with enlargement of the upper cervical spinal canal in flexion that spares the spinal cord from injury.8 The patient with a hangman’s fracture may present with neck pain and cradling of the head in the absence

708

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B Figure 16A2-5   A, Lateral radiograph of the cervical spine in a patient with rotary subluxation of C1-C2 presents a triangular wedged appearance of the anterior arch of C1. B, Computed tomography of C1-C2 reveals malalignment of the axis of the vertebrae with anterior translation of the lateral mass of C1 on C2. C, Lateral flexion radiograph reveals solid fusion between C1 and C2 without evidence of displacement.

of objective neurologic abnormalities. With persistent anterior displacement of the cervicocranium, neurologic deficits eventually develop. Early identification is of paramount importance. In addition to routine radiographs, computed tomographic scans of the cervical spine have allowed precise delineation of fracture patterns. Most patients with hangman’s fracture may be treated with gentle traction followed by halo vest or cast for 3 months. Lateral flexion-extension radiographs are necessary to demonstrate osseous healing and intersegmental stability. In the presence of nonunion or disruption of the C2-C3 disk, surgical fusion is indicated either by anterior fusion of C2 to C3 or posterior cervical fusion involving C1, C2, and C3.

Fracture of the Subaxial Cervical Spine Injury to the lower cervical spine from C3 to C7 may involve injury to the anterior elements of the spinal column, to the posterior elements, to the lateral elements, or

to a combination of sites. Clinical problems include facet dislocation with or without fracture, lamina fractures, and avulsion fractures of the spinous processes. Lateral mass fractures and pedicle fractures are uncommon in the subaxial spine compared with the incidence of injury in the upper cervical spine. The anterior elements of the spinal column are usually injured in flexion, with resultant compression fractures of the vertebral body and injury to the disk. Although disruption of the posterior longitudinal ligament is not common in athletic injuries, disruption may occur, especially with a flexion-distraction mechanism associated with significant intersegmental instability and the potential for neurologic catastrophe. As with injury at other levels, immediate immobilization of the spine is of extreme importance to prevent additional loss of neurologic function. Fracture at the C3-C4 vertebral level is rare.43 Athletic injuries most commonly result in injury at vertebral levels ranging from C4 to C7.14,22,35,50-55 Facet dislocation may be unilateral or bilateral and may occur with or without

Spinal Injuries

associated fracture. Unilateral facet dislocation is usually the result of axial loading in combination with flexion and rotation and does not usually result in neurologic damage. In the absence of facet fracture, the injury is primarily ligamentous and capsular, and the spine maintains its stability. Lateral radiographs of the cervical spine with unilateral facet dislocation reveal anterior translation of one vertebra on another of about 25%. Reduction of facet dislocation is obtained by closed traction techniques; occasionally, inability to reduce facet dislocation by closed methods necessitates open reduction with posterior fusion of the involved levels. Bilateral facet dislocation is a much more serious injury and occurs primarily through a mechanism of flexion. The spine is unstable in this situation and is associated with severe neurologic deficit including quadriplegia. Lateral radiographs of the cervical spine with bilateral facet dislocation reveal translation of more than 50% of one vertebra on another. These dislocations can usually be reduced by traction, immobilized in a halo cast, and stabilized by posterior fusion of the involved cervical vertebrae. Lamina fractures are difficult to diagnose on routine radiographs owing to the obliquity of the lamina in relation to the axis of the x-ray. CT is more reliable in identifi­ cation of lamina fractures. Fractures of the lamina do not usually participate in compression of neural tissue and heal with immobilization. Avulsion fractures of the spinous process are the result of vigorous exertion and are termed the clay shoveler’s fracture. The spinous process of the seventh cervical vertebra is most frequently involved. Treatment of the clay shoveler’s fracture is symptomatic because no subsequent instability results from this avulsion fracture. Injury of the anterior elements of the spinal column primarily involves axial loading resulting in compression fracture of the vertebral body. The extent of injury varies from a wedge fracture, which is stable, to the severely comminuted burst fracture, which is unstable and involves intrusion of bony elements into the spinal canal. The wedge fracture is common and is not associated with neurologic compromise. The posterior elements, including the ligamentous structures, are intact, and the spinal column remains stable. If disruption of the posterior elements is associated with anterior compression fracture of the vertebral body, stability is most likely compromised, and surgical stabilization is necessary. When progressive escalation of forces is experienced by the neck with axial loading, more severe injury of the vertebral body occurs, ranging from nondisplaced fracture fragments to wide displacement of bone and compromise of the spinal canal. Disruption of the posterior elements is more frequent with severe flexion and distraction forces and creates an extremely unstable clinical situation with severe neurologic compromise, including quadriplegia. Anterior decompression of the spinal canal with fusion is required, followed by posterior spinal fusion. Patients with facet dislocation or moderate to severe degrees of compression fracture require evaluation of the spinal canal to eliminate the possibility of concomitant extruded disk material (Fig. 16A2-6). Neurologically intact individuals with facet dislocation have been rendered quadriplegic following closed reduction owing to compromise of the anterior spinal cord by extruded disk material. If a disk is extruded, anterior surgical decompression of the

709

Figure 16A2-6   Patients with a severe compression fracture and disruption of the posterior soft tissues require further evaluation to rule out disk extrusion into the spinal canal.

disk should be performed, followed by reduction of facets with anterior and posterior spinal stabilization. Subacute or late instability of the cervical spine should be suspected in the presence of facet dislocation. ­Herkowitz and Rothman34 reported on six neurologically intact patients with no bone or soft tissue abnormalities on initial radiographs. Four patients had unilateral facet dislocations; one had a perched facet at C5-C6; and one had subluxation at C4-C5. Each patient subsequently developed radiographic changes indicating intersegmental instability with attendant neurologic compromise. It is important to perform repeat physical examinations and radiographic studies within 3 weeks of injury to rule out the existence of subacute instability. Once instability has been identified, surgical stabilization is required.

SPINAL CORD INJURY Spinal cord injury results from violent forces imposed on the spinal column. Injury may be direct, as in complete transection or bony compression of the spinal cord, or indirect as a result of hemorrhage, swelling, or secondary ischemia. With the clinical presentation of complete motor and sensory loss, transection of the spinal cord is likely and is irreversible. Incomplete lesions of the spinal cord usually present as mixtures of described syndromes. When severe forces are imposed by axial load on the cervical spine, a burst fracture of the vertebral body may result, producing bony impingement on the anterior spinal artery

710

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

with motor loss below the injury level and loss of sensations of pain and temperature. These deficits are usually permanent, and the degree of loss is equal in both upper and lower extremities. Severe flexion-extension moments applied to the cervical spine in the presence of spinal stenosis or secondary degenerative changes, which may occur in high school athletes, result in central cord hemorrhage and ischemia with primary involvement of the corticospinal tracts. Nonspecific sensory loss may be observed in the presence of incomplete motor loss involving all extremities. The upper extremities are usually significantly weaker than the lower extremities. Central cord involvement has a relatively favorable prognosis with varying degrees of recovery. Hemisection of the spinal cord results in loss of ipsilateral motor function and contralateral pain and temperature and is designated Brown-Séquard syndrome. Posterior spinal cord syndrome is a rare lesion in sports, with ischemia of the posterior spinal artery resulting in loss of dorsal column function and preservation of anterior cord function. These clinical syndromes do not usually appear in pure form but rather as parts of more complex lesions, most commonly involving a combination of central cord injury and BrownSéquard syndrome. The “burning hands syndrome” was first described by Maroon in 1977 as severe burning dysesthesias and paresthesias of both hands due to injury to the central fibers of the spinal tract.56 The injury is usually the result of ischemia. Wilburger and colleagues57 used MRI and SSEP to demonstrate that the burning hands syndrome was a reversible insult to the sensory pathways of the spinal cord. Vascular insults may result in thrombosis or embolization. In 1970, Schneider and associates58 reported seven cases of cervicomedullary injury in football players that resulted from “spearing.” Five of the athletes had no radiographic evidence of fracture or dislocation, although two showed evidence of atlantoaxial instability. Schneider and coworkers postulated that vertebrobasilar insufficiency with hypoperfusion of either the vertebral or basilar arteries could result in intramedullary cavitation and hemorrhage. A second possible mechanism of injury suggested by these authors involved acute arterial or venous obstruction from the brain due to uncal herniation through the tentorial notch. The final mechanism postulated by Schneider and colleagues involved high-velocity impact to the top of the head such that the brain interacts with the cervicomedullary junction, which is tethered by the dentate ligaments, resulting in secondary hemorrhages of the cervicomedullary junction. Fortunately, vascular injury of this sort in athletic events is uncommon. Of great concern is the problem of transient quadriplegia. Torg and colleagues55,59 described this problem as acute but transient episodes of sensory changes that may be associated with motor paresis in either both arms, both legs, or all four limbs following a forced hyperextension, hyperflexion, or axial load to the cervical spine. Complete recovery usually occurs in less than 15 minutes. Of interest is Torg’s report of 32 athletes with transient quadriparesis and associated developmental cervical spine stenosis.59 The degree of canal stenosis may be enhanced in flexion and extension by the “pincer mechanism” described by Penning60 or by infolding of the lamina ligaments, which

are capable of narrowing the spinal canal by 30% in hyperextension. Torg noted that 17 of the reported 32 athletes demonstrated developmental spinal stenosis. Only 4 of the 17 were able to return to play without permanent problems. Of the remaining 15 athletes without stenosis, 5 had congenital cervical fusions, and only 1 of these returned to play; 4 athletes had evidence of cervical instability, and 1 of these returned to competition; and of 6 athletes with degenerative disk changes, none returned to sports without problems. Therefore, of the group of 32 patients reported by Torg and associates, only 6 were able to return to play without problems. Although Torg and associates implied that athletes who have sustained transient quadriplegia with coincident developmental spinal stenosis should be discouraged from returning to competition, they conclude that athletes with transient quadriplegia and no demonstrated stenosis should be able to return to sports activity without a predisposition to permanent neurologic injury. The subset sample size is small in this study and does not allow formulation of a firm conclusion about the safety of return to competition. Transient quadriplegia in young athletes demands a detailed orthopaedic, neurologic, and imaging evaluation to rule out factors that may prohibit continued sports participation.61,62 In one review of 13 children between the ages of 7 to 15 years who had an episode of transient quadriplegia, all from sports activities, none had evidence of congenital stenosis based on measurements of the spinal canal diameter or the Torg ratio.63 The authors concluded that hypermobility of the cervical spine and not stenosis was the reason for the episode of transient quadriplegia. Evaluation of larger study groups of involved athletes is required before strong recommendations can be formulated about return to competition. The skeletally immature athlete who sustains this injury, even when clinical recovery is full and imaging of the cervical spine is normal, in the authors’ opinion is most safely treated by ­restriction from contact sports and other activities with risks for recurrent injury. “Burners” are described as episodes of searing pain in the upper extremities that follow the radicular distribution.7,40,64-67 Burners tend to occur after acute extension of the neck or a lateral stretch of the neck to the side opposite the painful arm with depression of the shoulder, as in a tackling maneuver. The symptoms usually last a few seconds in the initial episodes. The involved athlete allows his arm to hang limply at the side and then shakes or rubs the hand or arm vigorously to diminish the unpleasant searing pain. Numbness tends to last longer than the weakness; however, with repeated episodes, progressive residual weakness is observed. College football players have reported that burners last longer with increased frequency of the episodes, and occasionally persistent weakness, sensory loss, and pain are experienced whenever the arm is used. Rockett’s66 observations during surgical exploration of patients with burners document scarring at the C5-C6 nerve roots as they emerge between the anterior and posterior lamellae of the transverse processes. He subsequently recommended decompression of the nerve roots with lysis of nerve adhesions. Poindexter and Johnson65 performed electromyographic (EMG) evaluation of burners and suggested that they are the result of C6 radiculopathy rather than stretch of the brachial plexus.

Spinal Injuries

711

The initial complaints of athletes with burners suggest the diagnosis of acute herniated nucleus pulposus; however, with burners, the range of motion of the cervical spine remains normal, and symptoms are short lived. Burners, or “stingers” as they are also known, have been reported at least once in the careers of more than 50% of football players.64 During on-field evaluation, the affected player holds his head in a forward, stiff position to avoid extension and rotation of the neck. The presence of motor or sensory loss or arm or neck pain precludes return to play during that game until further evaluation is performed. In players who have sustained repeated injuries, full range of motion of the cervical spine and normal strength of the neck and shoulder girdle should be present before the athlete is permitted to return to competition.68 If weakness persists despite rest and rehabilitation, radiographic evaluation, EMG analysis, and MRI are needed to rule out less common lesions such as a herniated nucleus pulposus. EMG changes may persist in the absence of objective neurologic deficits for several years and cannot be used as a parameter for determining return to competition. Preventive measures that have been recommended to decrease the frequency of burners include neck and shoulder strengthening exercises, increasing the thickness of shoulder pads, and using neck rolls.

CONGENITAL ANOMALIES Congenital anomalies of the cervical spine primarily involve failure of formation or failure of segmentation of the vertebrae. Guidelines for participation in sports activities for young athletes with congenital anomalies have been based mostly on expert opinions, case reports, and biomechanical analyses because of the small number of children with known abnormalities who participate in these activities.69 Occipitalization of the atlas has been associated in ­neurosurgical literature with neurologic compromise; however, occipitalization is not usually associated with stenosis at the foramen magnum or with instability. The exception is the patient with occipitalization of the atlas and congenital fusion of C2-C3 in whom secondary hypermobility and instability frequently develop at the atlantoaxial junction. Instability has also been reported in individuals with hypoplasia of the odontoid in the presence of occipitalization of the atlas. Instability at this level requires posterior fusion of the occiput to the axis. Congenital absence of the posterior arch of the atlas is a rare congenital anomaly that is not usually associated with instability (Fig. 16A2-7). Lateral flexion-extension radiographs of the cervical spine as well as MRI evaluation are helpful to rule out cervical spine instability and chronic spinal cord impingement. In the absence of the posterior arch of the atlas, it is the author’s recommendation that athletes refrain from high-impact loading activities such as contact or collision sports and diving. Os odontoideum may be the result of nonunion or fracture through the body of the odontoid or congenital deformity. Lateral flexion-extension radiographs are required to document stability. Athletes with a stable os odontoideum should avoid impact-loading sports, including contact and collision sports. Individuals with an unstable os odontoideum require posterior surgical stabilization of the atlas

Figure 16A2-7   Lateral radiograph of the cervical spine reveals a complete absence of the posterior arch of C1 with no instability at C1-C2.

and axis. The normal spine that has undergone single-level spinal fusion should not be considered normal, and such an individual should not be allowed to return to full sports activity. There are no scientific data on the response of the surgically fused spine to forces imposed by sudden motion and the forces experienced in athletic activity. High-impact loading in the form of contact and collision sports and high diving should be avoided by this patient population. Congenital absence of the pedicles is a rare congenital anomaly that is usually alarming when viewed radiographically. If lateral flexion-extension radiographs demonstrate stability, however, there is no known reason to restrict the athlete’s activity. Congenital scoliosis of the cervical spine is not associated with instability. Mixed bony lesions may be noted with widening of the interpedicular distances suggestive of intraspinal lesions such as diastematomyelia. If appropriate radiographic studies and MRI eliminate the existence of intraspinal lesions and instability, involved athletes should be permitted to participate in all sports activities. Congenital fusion of the cervical spine, referred to as the Klippel-Feil syndrome, presents with a host of patterns that span the spectrum from single-level fusion to multiple levels of fusion to complete fusion from C2 to C7 (Fig. 16A2-8). It is extremely important to document the integrity of the occipitocervical junction in patients with Klippel-Feil syndrome to rule out instability. In the subaxial cervical spine, lateral flexion-extension radiographs may demonstrate anteroposterior translation of vertebrae as well as anterior gaping of open disk spaces.

712

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16A2-8   Lateral radiograph of the cervical spine reveals congenital fusion of C1 and C2 and also of C3, C4, C5, and C6.

In the absence of progressive radiographic changes in stability and of neurologic deficits, individuals with KlippelFeil syndrome should be observed; however, progressive translation or angular deformation at an open disk space should lead to exclusion of these patients from contact and collision sports and possibly to surgical stabilization of the unstable segment. The high association of renal anomalies in those with congenital scoliosis or congenital fusion of the cervical spine demands evaluation of the renal system by ultrasound to rule out clinically important anomalies. Unilateral absence of a renal system is the most common anomaly and is a significant factor in restricting involved individuals from contact sports (Fig. 16A2-9).The authors’ recommendations for play restrictions of collision sports for children with congenital cervical anomalies are listed in Box 16A2-2.

CONCLUSION Neglect of injury to the cervical spine can result in catastrophic neurologic damage as well as death. During the past decade, the study of mechanisms of sports injuries involving the neck has resulted in a significant decrease in the incidence of catastrophic and fatal injuries involving the cervical spine by means of alterations in competitive rules and the education of athletes and coaches in safe techniques of play. Comprehensive conditioning programs

Figure 16A2-9   Intravenous pyelogram demonstrates complete absence of one renal system with hydronephrosis of the remaining system.

Box 16A2-2 Contraindications to Collision Sports for ­Pediatric Athletes with Congenital Cervical Spine Anomalies (Authors, ­Recommendations) Absolute Contraindications •��� Occipitalization of C1 combined with C2-C3 fusion or odontoid hypoplasia •��� Os odontoideum •��� Subaxial fusions with instability* of the segments cephalad or caudad to the fusion mass •��� One episode of cervical cord neurapraxia (transient quadriplegia) •��� Congenital cervical spondylolisthesis •��� Atraumatic occiput–C1 instability Relative Contraindications •��� Occipitalization of C1 without segmental instability •��� Congenital absence of the posterior arch of C1 without segmental instability •��� Subaxial fusions without instability of the segments ­cephalad or caudad to the fusion mass *Stability determined on lateral flexion-extension cervical radiographs by anteroposterior translation of vertebrae as well as anterior gaping of open disk spaces.

Spinal Injuries

t­ ailored to the neck and shoulder girdle improve the athlete’s ability to resist damaging forces. Equipment deficits have been responsibly addressed by manufacturers with design improvement in such items as football pads, cervical rolls, and helmets. The “unexpected” contributes to the excitement experienced in sports. Unfortunately, it is a limiting factor that precludes the reduction of serious or fatal injury to zero. Most serious cervical spine injuries can be eliminated by strict adherence of coaches and officials to the rules of competition, use of effective equipment, instruction of athletes in safe techniques, and identification of high-risk athletes, combined with subsequent conditioning before competition.31 Team orthopaedic surgeons should educate the coaching staff about the serious nature of injury to the cervical spine. Injury to the soft or hard tissues of the neck requires attention to treatment guidelines and a comprehensive rehabilitation program that fosters full range of motion of the neck as well as normal strength. There have been no rigorous studies designed to prove that proper conditioning and preparation for competition decreases the incidence of injury of the cervical spine; however, the uncertainty of sports demands that the competitive athlete be in optimal physical condition during competition. Safe and effective competition requires appropriate mental and psychological preparation as well as physical conditioning to complement the state-of-the-art equipment available. It is only by adherence to a disciplined program that the incidence of serious and catastrophic cervical spine injuries can be lowered.

C

r i t i c a l

P

o i n t s

l Due to a low incidence of spinal injury in the young ­athlete, a high index of suspicion is needed to expedite early diagnosis and treatment. l Radiographs of the cervical spine of individuals in the first decade of life are difficult to evaluate due to incomplete ossification of vertebral elements and increased mobility compared with the adult spine.

713

l The team physician and coaches should identify local medical facilities experienced in the care of spinal injury prior to initiation of practice or game schedule. l Recurrent “burners” result in limited motion of the neck, with weakness of the neck, upper back, and shoulder girdle; these issues must be resolved prior to return to play. l While most individuals with Down syndrome may participate in sports without restrictions, those with ADI greater than 5mm are restricted from collision sports. l Latent instability of the cervical spine must be suspected in the presence of facet dislocation. l Athletes with transient quadriplegia associated with spinal stenosis are permanently restricted from collision sports.

S U G G E S T E D

R E A D I N G S

Frankel HL, Montero FA, Penny PT: Spinal cord injuries due to diving. Paraplegia 18:118–122, 1980. Schneider RC, Livingston KE, Cave AJE, et al: Hangman’s fracture of the cervical spine. J Neurosurg 22:141, 1965. Steinbruck D, Paeslack V: Analysis of 139 spinal cord injuries due to accidents in water sports. Paraplegia 18:86–93, 1980. Williams P, McKibbin B: Unstable cervical spine injuries in rugby: A 20-year ­review. Injury 18:329–332, 1987. Wroble RR, Albright JP: Neck and low back injuries in wrestling. Clin Sports Med 5(2):295–325, 1986. Wu WQ, Lewis RC: Injuries of the cervical spine in high school wrestling. Surg Neurol 23:143–147, 1985.

R eferences Please see www.expertconsult.com

714

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ecti o n

B

Thoracolumbar Injuries 1. Thoracolumbar Spine Injuries in the Adult William Dillin, Frank J. Eismont, and Scott Kitchel

Consider who we, as physicians, might see in the bend of time. Will she be 14 years old, suspended in the rings above the floor mat, each bodily gyration, each release and catch the animation of physical grace? Will he pose before the last hole, straddling his tee shot, knowing full well his best hole is the 19th, and strike with all the twisting force that his middle-aged trunk can consider, the longest distance of the weekend? At 90, will she walk at dawn, chasing the last wisp of darkness, arms pumping, feet striking asphalt, her spine battling load and gravity? Will he crash into the last defender with will and force, tumbling the last few feet into the end zone? Will her head be down, her upper back cringing, her arms exhausted as she rolls her chair towards the finish line? Will he clutch the undersurface of his board, spinning away from the upper edge of the half pipe, with earth and ice waiting his return? Will she glide past them, as they reach for her, ascending in a single mellifluous leap towards the rim? Will he hear nature’s footsteps in the rising sea, leap freely from his prone position, and tightrope top and bottom turns before the impact zone? Will she see yet another batter in the endless weekend of games, winding once again and curling spin and speed with bodily deception? Will he soar like a swan in flight from water, his kite 100 feet above, his body engaged in twist and twirls before inevitable return to the ocean’s side? Will she plant her pole in the steep slope in the outback and wipe the sweat from her eyes of her weekend tour? Will his racquet reach the apex of his toss, slicing the soft round ball in chosen direction? Will she look upon the course of trees and feel the blur of nature with each heartbeat as she runs? Will the world seem tame to his address, the bat that once was on his shoulder, the ball lost beyond the fence? Will she pirouette and point and unfurl her arms to sounds she cannot hear? At dusk, will he gather himself for a final water start onto his board, his sail arced forward to catch the wind, slipping silently across the water at nature’s speed—a single fin? Will he or she be one of us? (Fig. 16B1-1). And why might we be there? Is it the test of opposition? Is it the struggle to the end? Is it contest? Is it measurement? Is it fulfillment? Is it the dream? Is it our nature? “Exercise ferments the Humours, casts them into their proper Channels, throws off Redundancies, and helps Nature in those secret Distributions, without which the Body cannot subsist in its Vigour, nor the Soul act with Cheerfulness.1 So who are we? Can we distinguish between an exerciser and athlete? Is the competitive long-distance runner the athlete and the daily jogger, who burns a routine distance

and time, not for a race but for health, an exerciser? Will the same injury have separate and unequal meanings to the two of them? “The Greeks understood that mind and body must develop in harmonious proportions to produce a creative intelligence. And so did the most brilliant intelligence of our earliest days—Thomas Jefferson—when he said, not less than two hours a day should be devoted to exercise. If the man who wrote the Declaration of Independence, was Secretary of State, and twice President, could give it two hours, our children can give it ten or fifteen minutes.”2 As we integrate mind and body and twin them to exertion, what can we do with injury? Who will our future injured be? As the scientific evidence compounds promoting quality of life, and longevity of life is linked to weekly duration of exercise, health maintenance may be derived from consistency. And consistency may come from the successful management of injury that avoids deconditioning. “Those who think they have not time for bodily exercise will sooner or later have to find time for illness.”3 For every professional athlete, for every college athlete, for every high school athlete engaged in performance sports, we will encounter many from the group of “everyman and everywoman” slugging though their exercise routine. What are we to consider in the spectrum of injury? Is there a difference between achieving a zone of activity and the continuous striving for performance enhancement? Can injury be separated from the individual goal of exercise or the insistent demand of an individual in a dedicated sport? Who will we see in the factory of biology? We will see the young and the old and everyone in between—their biologic potentials for injury having some common elements and age-related discrepancies. We will see the male and female, the able and disabled, the reluctant, the resigned, and the conditioned zealot. We will see failure of biologic material exposed to continuous, repetitive load over time, or failure with exposure to sudden force in an instant. We will see the unprepared reach for goals way beyond their training and capacity and wonder why an injury occurs. We will see the fully trained to peak and perform, hurt, and wonder why this injury at this time. We will see the rates and types of injury vary among the chosen sports. We will see the common injuries of daily life sustained during exercise or derived from daily life and amplified and delimited in the presence of exercise. We will encounter the various motivations that surround a particular injury, and to cure the biology, we will also treat the psychology. We will see the full fluctuation of what biology reveals in

Spinal Injuries

715

lliocostalis thoracis

Spinalis thoracis Figure 16B1-1   Professional and amateur windsurfers share the north shore Maui waters at Kanaha.

failure, and we may be the guiding hosts for its recovery. We may be called on to provide advice for prevention and search a meager science for verifiable answers. No injury exists without context, and for us, knowing the context is vital to treating the injury. To our forefathers we turn for advice. “Medicine is not only a science; it is also an art. It does not consist of compounding pills and plasters; it deals with the very processes of life, which must be understood before they may be guided.”4 To the fundamental process of life we now turn as we examine injury to the thoracic and lumbar spine during exercise.

ANATOMY The anatomies of the thoracic and lumbar spines are sufficiently different that they are considered separately. Anatomy is discussed to correlate its practical application with common injury patterns.

Thoracic Spine The thoracic vertebral column is responsible for the support of the thorax and protection of the thoracic spinal cord and provides the origin for the ribs. There are 12 thoracic vertebrae that articulate with one another through the diarthrodial facet joints as well as the intervertebral disks. The typical thoracic vertebra is roughly heart shaped and midsize between the cervical and lumbar vertebrae. The size of these vertebrae increases gradually from the 1st to the 12th in a caudad direction. The pedicles are placed toward the upper end of the bodies, with the laminae arising from these in a more superoinferior direction, allowing them to overlap one another. The spinous processes are typically long and slender with a slight caudad direction so that they overlap the succeeding vertebrae. The thoracic vertebral arch is a small, round spinal canal that protects the thoracic spinal cord. A distinguishing factor of the thoracic vertebrae is that they articulate with the ribs and bear facets for these articulations. The thoracic vertebrae articulate with the head and the tubercle of the rib arising from them as well as the head of the rib below their own segmental level.

Longissimus thoracis

Figure 16B1-2   Musculature of the thoracic spine.

These articulations between the thoracic vertebrae represent a three-joint complex including the intervertebral disk as well as the facet joints. The stability of these articulations is strengthened further by the rib cage; this has two consequences for the thoracic spine. First, it is an extremely stable region because of these articulations and the supporting rib cage. Second, because of the frontal plane orientation of the joints and the rib cage, the thoracic spine is the least mobile of the spine regions. Ligamentous support is provided to the articulations of the thoracic spine from many locations. Ventrally, there is a strong anterior longitudinal ligament that runs continuously along the entire length of the thoracic spine. The posterior longitudinal ligament similarly runs in continuity over the length of the thoracic spine along the dorsal aspect of the vertebral bodies, forming the anterior wall of the vertebral canal. The facet joints in the thoracic spine have capsular ligaments, which provide stability and limit the excursion of these joints. Dorsally, the ligamentum flavum is less developed in the thoracic spine than its cervical or lumbar counterparts. The supraspinous and intraspinous ligaments are membranous sets of fibers that run between spinous processes. In the thoracic spine, the interspinous ligament provides only minimal stability. The individual fibers of the supraspinous ligament are arranged so that the more superficial fibers extend over several spinal segments, whereas the deeper, shorter fibers bridge only half spines. The musculature of the thoracic spine and thorax contributes a portion of its stability (Fig. 16B1-2). The muscles of the thoracic spine generally run in a longitudinal fashion. They have multiple origins and insertions arising from

716

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Dura Dorsal root

Spinal nerve Dorsal ramus

Ventral root Figure 16B1-3   Segmental spinal nerve innervations of spinal muscles.

many consecutive vertebrae or ribs and inserting into many consecutive vertebrae or ribs more caudad. These muscles seem particularly complex because of the great amount of fusion that takes place between adjacent segmental muscles to form longer ones and the tangential splitting that results in a number of superimposed layers. For the purposes of this chapter, it is not important to know the specific muscle names but merely that they can be grouped broadly according to the general direction of the muscle bundles and their approximate lengths. The longer muscles of the back can be categorized into the splenius, the erector spinae, and the transversospinalis groups. The splenius muscles arise from the midline and run laterally as well as cephalad. They are responsible for spine rotation as well as extension. The erector spinae group is the largest muscular mass of the back. These muscles arise along the midline and run nearly directly longitudinally in the thoracic spine. They are the predominant extensors of the thoracic spine. The transversospinalis muscles lie deep to the erector spinae and are much shorter. These muscles arise from the transverse processes and run cephalad to the spinous processes. They are predominantly rotators of the spine but also participate in extension.

A

Deep to these major muscle groups are a series of small segmental muscles that run between spinous processes or between transverse processes. These muscles include the interspinalis and intertransversarial. Both of these muscle groups are functionally less important in the thoracic spine. The muscles of the back are covered posteriorly by a dense fascia that separates them from the overlying structures. The fascia is continuous, with connections from the cervical spine through the sacrum. In the thoracic spine, this fascia is thin and transparent except where it blends into the muscle origin. The innervation of these back muscles is by the dorsal rami of the segmental spinal nerves (Fig. 16B1-3). The dorsal rami typically slant inferiorly as they proceed through the muscles, supplying muscle caudad to the level of the origin of each nerve. A few of the more lateral deep muscles may be innervated by ventral rami of the spinal nerves. The blood supply of the thoracic spine is from the segmental arteries (Fig. 16B1-4). Each thoracic vertebra is related to a pair of segmental intercostal arteries from the aorta. Each segmental vessel gives off twigs to the anterolateral aspect of the vertebral body, and larger vessels enter the vertebra from the vertebral canal. The dorsal branches of the segmental arteries are distributed mostly to the musculature of the back but give off spinal branches as they pass through the intervertebral foramen. These spinal branches enter the epidural space as well as anastomosing above and below corresponding vessels to form the major supply of the dura, nerve roots, and spinal cord. The vessels to each vertebral body come from the segmental arteries entering above and below the body so that each body typically receives nourishment from four arteries. These arteries pass between the posterior longitudinal ligament and the vertebral body, usually penetrating the bone by a common opening. The venous drainage of the vertebral column parallels the arterial supply and enters the internal vertebral plexus that surrounds the spinal cord. There is significant

B

Figure 16B1-4   Blood supply of thoracic spine. A, Lateral view. B, Posterior view.

Spinal Injuries

anterolateral drainage from the vertebra directly into the segmental veins, which ultimately drain into the vena cava.

Lumbar Spine The lumbar vertebrae are the last five vertebrae of the vertebral column. They are particularly large and heavy when compared with the vertebrae of the cervical or thoracic spine. Their bodies are wider transversely than anteroposteriorly. The pedicles of the lumbar spine are short and heavy, arising from the upper part of the body. Compared with the thoracic spine, the transverse processes pro­ject more laterally and ventrally. From the posterior surfaces of the superior articular processes, there are marked enlargements, called the mammillary process. The laminae are shorter vertically than the bodies; this causes a gap between the lamina at each level, which is bridged only by ligaments. The spinous processes are broader and stronger than those in the thoracic spine; they project in a dorsal direction with little caudad angulation. The articulations in the lumbar spine are the same three-joint complex. The joints are oriented in a more sagittal plane. This orientation allows the lumbar spine to have relatively more flexion and extension than its thoracic counterpart but significantly less rotation. This joint alignment also allows for lateral flexion in the lumbar spine. The same basic ligamentous structures are present in the lumbar spine as in the thoracic spine. The anterior longitudinal ligament is relatively thicker in the lumbar spine. The ligamentum flavum is much stronger than its thoracic counterpart. This increased strength is in part due to the fact that it serves as a bridge between adjacent laminae where there is no bony overlapping. The facet joint capsules of the lumbar spine are thicker and stronger in the lumbar spine, as are the supraspinous and infraspinous ligaments. The stability of the lumbar spine is related much more directly to the ligamentous structures than the thoracic spine because of the loss of stability added by the rib articulations and rib cage. The musculature of the lumbar spine is organized in the same pattern as that of the thoracic spine. As one moves more caudally into the lumbar area, the muscles of the superficial groups tend to become larger and stronger. These muscles provide some stability to the lumbar

717

spine as well as being the primary extensors of this region. The enveloping fascia in the lumbar spine is thicker and stronger than its thoracic counterpart. This fascia has been divided into three distinct layers, which provide stability as well as compartmentalization in the lumbar spine. The nerve and blood supply of the lumbar spine is functionally identical to that of the thoracic spine. The segmental vessels and the nerves run courses similar to the thoracic spine and provide similar function.

Intervertebral Disk The intervertebral disk is the fibrocartilaginous structure that forms the articulation between adjacent vertebrae. It provides a strong union, while allowing the degree of intervertebral motion necessary for function. Disks of the various portions of the spinal region differ considerably in size but are basically identical in their organization. They all consist of two components: the outer, laminar fibrous container (or anulus), and the inner, semifluid mass (the nucleus pulposus) (Fig. 16B1-5). The anulus fibrosus is a concentric series of fibrous lamellae. Its major function is to withstand tension from the torsional stresses of the vertebral column as well as the horizontal extensions of the compressed nucleus that it contains. The anulus is attached to the vertebral body through a blending of the fibers with the vertebral periosteum as well as the longitudinal ligaments. The nucleus pulposus occupies a concentric position within the confines of the anulus. Its major function is that of a shock absorber. The nucleus pulposus exhibits viscoelastic properties under applied pressure, responding with elastic rebound. There is no definite structural interface between the nucleus and the anulus. The two tissues blend imperceptibly. The disks make up about one fourth of the height of the entire spinal column. Moving from cephalad to caudad, the disks become larger in their cross-sectional area as well as thicker when measured from one vertebral end plate to the next. The thoracic disks are heart shaped compared with the more oval form seen in the lumbar spine. The blood supply and nutrition of the intervertebral disk is achieved primarily by diffusion from the adjacent vertebral end plates. The anulus is penetrated by capillaries for only a few millimeters. The disk is not inert; the normal disk tissue has a high rate of metabolic turnover. The disk itself has no direct innervation. Sensory fibers are abundant, however, in the adjacent longitudinal ligaments. The pain attributable to disk disease most likely is carried through those fibers.

Spinal Cord and Cauda Equina

Figure 16B1-5   Anatomy of the thoracic disk.

In the thoracolumbar spine, it is important to differentiate between spinal cord and cauda equina. The spinal cord typically ends at the thoracolumbar junction, and caudad to that level is the cauda equina. The cauda equina is merely a collection of nerve roots that are traversing the spinal canal until they exit at their appropriate foramen. This differentiation is important because of the differences in structural anatomy and the responses to injury of the spinal cord and cauda equina.

718

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The thoracic spinal cord is generally wider laterally than it is deep in the anteroposterior direction. The cord is smaller in the thoracic region than its cervical counterpart. The average anteroposterior depth found by Elliott5 was about 9 mm. An average anteroposterior vertebral canal diameter is 17 mm. The cord occupies about half of the space available within the vertebral column. The major function of the thoracic spinal cord is the transmission of nerve impulses, from the periphery to the brain and from the brain to the peripheral muscles. The tapered lower end of the spinal cord is the conus medullaris. From this point caudad, only individual nerve roots exist and are grouped together as the cauda equina. These nerve roots pass caudad until they reach the appropriate level, where they pass out through the vertebral ­foramen. The response of the spinal cord to injury is different from the response of the cauda equina. The nerve roots in the cauda equina recover from injury in the same fashion as a peripheral nerve. Injury to the spinal cord is generally irreversible, however, and has permanent consequences. This fact is important when considering injuries of the thoracolumbar spine with associated neurologic deficits.

BIOMECHANICS The thoracolumbar spine is a complex, three-dimensional structure with coupled motion characteristics.6 The thoracolumbar spine is capable of flexion, extension, lateral flexion, and rotation. The total range of motion is the result of a summation of the limited movements permitted between the individual vertebrae. The musculature and ligaments have key roles in the initiation and control of movements as well as in supporting the bone structures. The individual motions vary considerably in the different vertebral regions. Although all thoracolumbar vertebrae are united in the three-joint system of the intervertebral disk and the two zygapophyseal articulations, the size and shape of the intervertebral disk as well as the shape and orientation of the articular joints determine the types and range of motion available at an individual intervertebral articulation. The most common movement of the vertebral column is flexion. Flexion requires an anterior compression of the intervertebral disk, along with a gliding separation of the articular facets at the zygapophyseal joint. This movement is limited by the posterior ligamentous complex and the dorsal musculature. Extension is a more limited motion, producing posterior compression of the disk along with gliding motion of the zygapophyseal joint. Extension is limited by the anterior longitudinal ligament as well as the ventral musculature. The laminae and spinous processes limit extension by direct apposition. Lateral flexion necessarily is accompanied by some degree of rotation. It involves lateral compression of the intervertebral disk, along with a sliding separation of the zygapophyseal joint on the convex side, whereas an overriding of this joint occurs on the concave side. Lateral flexion is limited by the intertransverse ligament as well as the extension of the ribs. Rotation is related most directly to the thickness of the intervertebral disk. It involves compression of the ­ anulus

fibrosus fibers. Rotation also is limited directly by the geometry of the zygapophyseal joints. The disk limits rotation by resistance to compression in the anulus. Normal range of motion for the thoracolumbar spine cannot be considered without also considering the synchronous motions of the cervical spine. The entire vertebral column moves as a whole in all planes of motion. The column can rotate about 90 degrees to either side of the sagittal plane. Most of this rotation is accomplished in the cervical and thoracic sections. Flexion of 90 degrees is possible, using cervical, thoracic, and lumbar regions. About 90 degrees of extension is also possible, but this occurs primarily in a combination of the cervical and lumbar regions. Lateral flexion, which must be accompanied by some rotation, is allowed to nearly 60 degrees. This is primarily a cervical and lumbar function. The mobility of the thoracolumbar region is not uniform throughout any of its segments. The upper thoracic spine is impaired greatly in its motion by the rib cage. The articular facets in this region are oriented in the frontal plane. The lower thoracic region allows more flexion and extension because the disk and vertebral bodies progressively increase in size. Also, in the lower thoracic spine, the articular facet joints begin to turn more toward the sagittal plane, permitting greater flexion and extension but limiting rotation. The lumbar region is oriented to allow significant amounts of flexion, extension, and lateral flexion. The zygapophyseal joints are oriented in the sagittal plane, however, locking them against rotation (Fig. 16B1-6). This orientation allows a gliding action of the joints that permits the neural arches to separate and approximate during flexion and extension. The lumbosacral joints change their orientation so that they are midrange between frontal and sagittal planes. This alignment allows some rotation; however, this is limited by the iliolumbar ligaments. The essential function of the lumbosacral joints is to buttress the fifth lumbar vertebra in relation to the sacrum. Each region of the spine has its own characteristic curvature. These curves allow upright posture while maintaining the center of gravity over the major weight-supporting structures of the pelvis and lower limbs. The normal thoracic kyphosis places the thoracic vertebrae posterior to the center of gravity. This kyphosis compensates for the normal cervical lordosis, which allows the head to be held in its erect position. The lumbar lordosis brings the middle of the lumbar spine anterior to the center of gravity, allowing erect posture. The transitional vertebrae between each major spinal segment intersect the center of gravity and appear to be the most unstable regions of the spine. This fact is emphasized by the high incidence of fractures and dislocations in the transitional regions. The biomechanics of the intervertebral disk emphasize its functional competency. The anulus fibrosis receives most forces transmitted from one vertebral body to the other. It is constructed best to resist tension and shear. This resistance is accomplished by the radial alignment of the progressive lamellae of the anulus. Experimental analysis has shown that different portions of the anulus respond differently to the same degree of tension. It appears that the peripheral anulus has the greatest recovery, whereas the medial sections are more distensible.

Spinal Injuries

719

Spinous process of T5 Supraspinous ligament

Articular process of T8

Interspinous ligament

Interspinous ligament Articular process of L3 Articular process of L2

Capsule Articular process of T9

A

Ligamentum flavum

Ligamentum flavum Intervertebral disk

B

Figure 16B1-6   Zygapophyseal joints of thoracic (A) and lumbar (B) spine.

The nucleus pulposus is designed best to resist compression forces. It receives primarily vertical forces from the vertebral bodies and redistributes them in a radial fashion to a horizontal plane. The internal pressure of the nucleus distorts the anulus, which, with its resiliency, allows recovery from the pressure. The tension of the intervertebral ligaments and anulus preloads the disk. This preloading increases stability in the spine. Through this function, the disk must dissipate and transfer the axial thrust necessary to permit erect posture and motion.7 The spine must act as a flexible boom. The spine is the fulcrum of a first-class lever system.

CLINICAL EVALUATION Evaluation of the thoracolumbar spine in athletes must follow the same sound principles as any clinical evaluation. The essentials include an accurate, problem-oriented history; spine and lower extremity evaluation; physical examination; and appropriate and specialized diagnostic studies based on the history and physical examination. It is essential that the clinician develop a reproducible standardized method of obtaining the evaluation so that there are no omissions or missing clinical information that would prevent an accurate diagnosis. The injured athlete is highly motivated to return to preinjury activity. The evaluation must lead to a working diagnosis so that treatment may begin and return the athlete to participation as soon as possible.

History One of the keys to an accurate diagnosis of any thoracolumbar spine problem is a carefully taken history. The athlete generally understates his or her complaints and generally omits any past spine problems or injuries. The history should include the patient’s chief complaint, a discussion of the present illness, a past history of any spinal or general orthopaedic problems, and a brief discussion of any family history of back problems. A commonsense approach is essential in this process. A good clinician allows the athlete to render the story in his

or her own words but elicits the important information. History taking varies widely in extent and length of time as the clinical situation dictates. Often, after an acute injury, only a brief history is obtained, but the clinician should return at a later time to obtain a more complete history of any previous problem or injury predisposition. Spending a little extra time obtaining a complete history provides great dividends in understanding the athlete’s problem and working toward a correct diagnosis. The chief complaint must be provided by the athlete. This should key the clinician into a certain line of history taking related to that specific body part. The history of the present illness enlarges on this chief complaint. This history can be obtained from the injured athlete and from other participants who may have witnessed the injury or who were present at the time. The most common initial complaint in spine problems is pain. It is important to attempt to localize and characterize the pain. The onset of the pain, as related to the time of injury, may provide important information regarding the diagnosis. Diffuse aching pain in the lumbar spine that began the day after an extended workout would lead the clinician into checking for a musculoligamentous type of overuse injury. The most important part of the history is the present illness. This part of the history should work toward a chronologic development of the spine pain, its character, and any improvement between its onset and the examination of the patient. The temporal onset of the pain gives the clinician a clue to the correct diagnosis. Mechanical causes of back pain have a sudden, acute onset. Often the athlete reports that the onset of pain is associated with a specific activity. The pain generally starts immediately or within a few hours. A more insidious onset of pain should alert the clinician to consider medical causes of low back pain. A history of the duration and frequency of the pain is essential. The clinician should ascertain whether the pain is episodic or more persistent. Most mechanical pain of the thoracolumbar spine is intermittent. The frequency of the episodes varies depending on the exposure of the athlete.

720

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The duration of the pain can lead the clinician to an accurate diagnosis. Most muscular strains are relieved within a week. Disk problems generally require longer to resolve. A history of the quality of the pain itself must be obtained. The quality of the pain and its intensity can be helpful in identifying its source. The patient should be allowed to describe the pain in his or her own words, ­without being led by the examiner. The intensity of the pain must be ascertained. It is often difficult for the athlete to describe pain intensity; this may be facilitated by asking the athlete to rank the pain on a scale of 0 to 10, with 0 being absence of pain and 10 being the worst pain ever felt. Localizing the pain to a specific area in the thoracolumbar spine is helpful. The pain may be localized to a specific midline structure or may radiate out from there into the adjacent soft tissues. A key to the evaluation of thoracolumbar spine discomfort is any radiation of the pain. It is especially important to differentiate strictly back pain from pain that may radiate into the leg or foot. Pain that radiates into the lower leg is suggestive of nerve root impingement, such as occurs with disk herniation. The athlete should be questioned carefully about any aggravating or alleviating factors. Generally, mechanical lesions of the thoracolumbar spine improve with supine positioning and worsen with increased activity. Not all thoracolumbar spine movements necessarily exacerbate muscle strain pain. A careful history defines which movements appear to help or aggravate the pain. Questioning the athlete regarding specific variances in pain related to time of day helps to sort out mechanical disorders from underlying inflammatory problems. In the typical muscle strain injury, the athlete notes that the pain is worse at the end of the day, after he or she has been active. Medical disorders, such as inflammatory arthropathies, cause stiffness in the morning when attempting to get out of bed. During the day, the stiffness lessens. Underlying tumors of the spine or spinal cord generally cause increased pain at night because the pain is increased with recumbency and is more noticeable with the loss of other sensory input while in bed. A past history of spinal problems is a key to accurate diagnosis. The patient should be asked at several different times in several different ways if he or she has had any previous spine problems. Questions about childhood and adolescence should be included. A preexisting condition of the spine may lead the clinician to an accurate diagnosis because a previous problem may recur or may predispose to a new injury. In this portion of the history, a brief time should be spent discussing any family history of spine problems with the athlete. Familial predisposition to back problems is characteristic of many medical illnesses, including disk degenerative conditions, ankylosing spondylitis, Reiter’s syndrome, and other spondyloarthropathies. As more older athletes become involved with highdemand workout schedules, it is important to remember that these athletes have other social and occupational activities that may contribute to their spine problem. In such an athlete, it is essential to obtain an occupational history to determine what tasks the athlete may be performing at work that may contribute to spine problems. Individuals required to do heavy lifting at their jobs are at risk for developing mechanical low back pain. Review of ­

leisure-time activities is important to look for predisposing causes of the spine problem. The patient history is the essential foundation on which the remainder of the diagnostic process is constructed. By taking a little extra time and listening carefully to the athlete’s description of the chief complaint, the clinician should be able to generate a list of potential diagnoses to direct the remainder of the history taking and physical examination.

Physical Examination After carefully obtaining the history, a problem-oriented physical examination is the next step in the diagnostic ­process. Physical examination takes a variety of forms, depending on when the examination is performed. The clinician should never feel limited by time in evaluating acute injuries of the thoracolumbar spine. In the acute setting of an injury to the spine, enough time must be taken to rule out injuries that could produce instability or threaten the neurologic structures before moving the patient. On the field, management of acute spine injuries must be approached with caution because of the potential neurologic sequelae of inadequate immobilization or transportation of the unstable spine. If there is any question of a spinal column injury with neurologic symptoms, it is important to immobilize the athlete in the position in which he or she was found and not attempt to move the athlete. No attempt should be made to remove equipment, such as a football helmet or part of the uniform. The athlete may be immobilized on a spine board or immobilized without change in body position by a scoop, such as is used by emergency medical personnel. This process takes time and should not be rushed so that there is a minimal risk for increased neurologic injury or manipulation of the spine. The athlete and the provider are better served by overimmobilizing the injured athlete than by attempting to move him or her in a hurry to allow completion of the athletic event. A spine board always should be available for transporting the spine-injured athlete. This board often is available because emergency medical personnel are at the scene of the athletic contest. When no medical emergency personnel are on hand, the institution should have a spine board available for transporting these athletes. All athletes with evidence of injury to the spine and neurologic signs or symptoms should be transported immobilized on a spine board. Athletes who have significant pain in the spine secondary to a high-velocity injury should be transported immobilized on a spine board because of the potential for spinal instability, which could lead to neurologic sequelae. Any athlete who is in too much pain to allow mobilization simply on the basis of muscle spasm should be transported on a spine board. It cannot be overemphasized that adequate immobilization and careful transportation prevent possible catastrophic neurologic injury in athletes with spinal column injuries. The clinician should not hesitate to take all the time needed and should have adequate equipment available before any attempt is made to transport the injured athlete. The essentials of the physical examination of the thoracolumbar spine are no different from those for ­examination

Spinal Injuries

of any other region of the body. The objective of the examination is to show the physical abnormalities that sort out the possible pathologic conditions elicited during the history taking. Essentials include inspection, palpation, rangeof-motion testing, and neurologic examination. Inspection of the patient as a whole and of the thoracolumbar spine in particular should be carried out initially with the patient in the standing position and disrobed sufficiently to allow the spine to be seen. The patient should be viewed from behind as well as laterally and anteriorly. From behind, the level of the shoulder should be noted as well as any lateral curvature or cervical scoliosis. The patient should stand with the head centered over the pelvis and feet. Any deviation at one location in the spine must be compensated for by an opposite deviation elsewhere if the patient is standing erect. A list occurs if the thoracic vertebrae are not centered over the sacrum. The list may be measured by dropping a perpendicular line from the first thoracic vertebra and measuring how far this falls to the right or left of the gluteal cleft. When viewing the patient in the lateral plane, any exaggeration or decrease of the normal spine curvature should be noted. Particularly, any increase in thoracic kyphosis or decrease in lumbar lordosis is significant. The lower extremities should be viewed in the lateral plane. Particular attention should be paid to any flexion or extension deformities of the hips and knees. Inspection from the front of the patient should include the position of the head and level of the shoulders. It is generally easy to view the iliac wings, and these should be of equal height. There should be no tilt to the pelvis. The skin about the thoracolumbar spine should be inspected, noting the superficial structures. Particular attention should be paid to any skin lesion, such as café-au-lait spots or a tuft of hair over the spine. After adequate inspection of the thoracolumbar spine, the next step in a complete examination is palpation of the area of tenderness. The primary area of tenderness should be palpated, but because this will cause the patient maximal discomfort, it is wise to palpate certain other anatomic landmarks first. Tenderness should be assessed over the spinous process at each level. The paraspinal muscles should be palpated looking for tenderness as well as muscle spasm. The sacroiliac joints and the sciatic notches should be palpated for tenderness, and deep palpation of the posterior thighs should be performed. When an area of maximal tenderness is identified, palpation should be carried out in an attempt to identify the primary structure that is tender at that level. The tenderness may be superficial, such as that seen with the spinous processes or dorsal musculature, or it may be more deep and diffuse, such as that related to a fracture or disk injury. There may be no area of point tenderness in many musculoligamentous-type injuries of the thoracolumbar spine. Assessing the range of motion of the thoracolumbar spine is important to identify the problem adequately. The absolute range of motion is not of major significance because there is a great deal of individual variance. Range of motion of the lumbar spine should be assessed for flexion-extension and lateral flexion. The reported average range of forward flexion is 40 to 60 degrees. Forward flexion is a complex motion of the lumbar spine, sacroiliac

721

joints, and hip joints. It may be significantly influenced by tightness of the hamstrings if the gauge is the athlete’s ability to touch the toes. Extension is considered average at 20 to 35 degrees. There should be about 20 degrees of side bending, right and left. Rotation of the lumbar spine is limited and difficult to assess because it occurs in symmetry with the thoracic spine. The major thoracic motion to be evaluated is rotation. With the feet in place, the patient is rotated at the shoulder level. Rotation to nearly 90 degrees can be achieved in the average athlete. While assessing the range of motion, the patient should be asked to squat in place. This tests not only the general muscle strength of the lower extremities but also joint function. If the patient is unable to squat, it should be assessed whether this is secondary to pain or some specific decreased function in the lower extremities. When an athlete with an injury to one of the functional units of the spine attempts to bend or rotate, this motion is inhibited by protective muscle spasm. The lumbar spine may be observed not to have a normal curve in the erect position and not to reverse its lordosis with attempts at flexion. This observation is highly suggestive of protective muscle spasm. If the protective spasm is unilateral and predominantly affects the tissue on one side of the spine, a scoliosis may develop. Scoliosis also may develop from nerve root irritation on one side of the spine, such as occurs with disk herniation. After the patient has flexed forward fully, it is helpful to observe how he or she regains the erect posture. This gives clues to tissue injury as well as muscle integrity. Normally, the return to the erect position is accomplished by a derotation of the pelvis without changes of spine curvature until the patient has come up to 45 degrees. During the terminal 45 degrees, the low back resumes its lordosis. Pain caused by any motion should be noted. Pain precipitated by flexion is a nonspecific finding and may be related to many pathologic conditions. Pain with extension generally is related to an increase of the lordosis forces across the facet joints or stresses in the pars interarticularis; this in turn narrows the foramen where the nerve roots exit the spine and compresses the posterior disk. Pain with hyperextension generally can be related to a pathologic condition involving the facet joints, pars interarticularis, posterior disk structures, or neuroforamen. The pain may be back pain, leg pain, or both. No special significance should be placed on pain with lateral bending. Lateral bending causes ligamentous or muscular stretching and may be restricted in many conditions. In some cases, pain that increases with flexion to the ipsilateral side may be related to articular facet disease or lateral disk protrusion. This should be considered if radicular pain is elicited with lateral bending. Similarly, limitations of rotation are nonspecific. These limitations may be secondary to muscle spasm within the thoracic spine or simple increases in pain with this motion. A more helpful way of checking rotation is to seat the patient, stabilizing the pelvis and hips. This not only limits the rotation of the spine but also gives a more specific view of spine rotation, eliminating that from the hips and pelvis. A neurologic examination of the thoracic and lumbar spine must include sensory and motor evaluation of the thorax and the lower extremities. In the thoracic spine, this

722

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 16B1-1  Muscle Testing of the Lower Extremities Nerve Root

Muscle Group

L1 L2 L3 L4 L5 S1

Hip flexion Hip flexion Knee extension Foot dorsiflexion-knee extension Big toe extension-foot eversion Foot plantar flexors-knee flexion

L4 L5

Reflex L1 Knee jerk Posterior tibial Ankle jerk

S2

L2

S2

L3 L4

L3

evaluation is limited by sensory overlapping and the multiple levels of innervation. Sensation can be assessed most sensitively by light-touch examination of the thorax. Motor examination of the thoracic and paraspinal musculature is difficult in the athlete with thoracolumbar spine problems because of pain and attendant muscle spasm. Neurologic examination of the lower extremities in the patient with thoracolumbar spine pain is particularly important. Neurologic examination is essential in a patient who complains of any leg pain, numbness, or weakness. This evaluation must include motor strength testing, lighttouch sensation, reflex testing, sciatic and femoral nerve tension signs, and assessment of sacral motor and sensory function. Initial assessment of lower extremity strength can be made in rough terms by asking the patient to squat, then return to an erect position. This rough motor examination can be supplemented by asking the patient to walk first on the heels, then on the toes. Any weakness seen in toe walking is suggestive of weakness in the triceps sura musculature. Difficulty with heel walking is consistent with ankle dorsiflexion weakness. When this evaluation has been completed, the patient is seated on the table with the legs dangling off the side. Muscle testing in the lower extremities is performed, including the hip flexors, adductors, quadriceps, hamstrings, tibialis anterior, foot everters, extensor hallucis longus, and foot plantar flexors (Table 16B1-1). This information is recorded according to the standard nomenclature on a 0-to-5 scale. Any specific deficits are noted. A sensory examination of the lower extremities is carried out by light-touch testing (Fig. 16B1-7). This test is performed most easily by simple light stroking of the thighs and legs in all different dermatomes. Any specific deficit is noted and retested. We do not routinely advocate sharp-dull discrimination or position-sense testing unless other deficits have been found. Reflex examination of the lower extremities is carried out in the sitting position. This examination should include an evaluation of knee jerks and ankle jerks. The knee jerk reflex is mediated primarily through the L4 nerve root. The ankle jerk is mediated by the S1 nerve root. These reflexes are recorded in the standard 0-to-4 nomenclature. The patient is asked to lie supine on the examining table for evaluation of nerve root tension signs. The classic test of sciatic nerve irritation is the straight leg raising test.8 The intent of this test is to stretch the dura and nerve roots, reproducing leg pain. The patient experiences pain along the anatomic course of the sciatic nerve into the lower leg, ankle, and foot. Symptoms should not be produced until the leg is raised to at least 30 to 35 degrees. When the leg

S1

S5 S4 S3

L5 L4

S1

S1 S2

L4 L5

L5 S1

S1 L5

Figure 16B1-7   Dermatomal innervation of the lower extremities.

has been elevated beyond 70 degrees, no further stretching of the nerve roots and dura occurs. To be considered positive, the test must reproduce the patient’s radicular symptoms. Production of back pain does not indicate a positive result (Fig. 16B1-8). Many other sciatic nerve root tension tests have been described. In Lasègue’s test,9 the patient lies supine with the hip flexed to 90 degrees. The knee is extended slowly until the radicular pain is reproduced. This test is likely less specific than the straight leg raising test because hip and knee joints are moved. The bowstring sign10 is performed with the knee flexed to 90 degrees and the body bent forward to lengthen the course of the sciatic nerve. The examiner’s finger is pressed into the popliteal space to increase further the tension on the sciatic nerve. A positive test occurs if the patient’s pain increases down the leg. Milgram’s test11 is another sciatic nerve tension sign. The patient lies in the supine position, then raises both extended legs several inches above the examining table. This movement increases intra-abdominal pressure and intrathecal pressure. The patient is asked to hold the position for 30 seconds. The test is positive if the maneuver re-creates the radicular leg pain. Nerve tension signs of the femoral nerve have been described. The most popular names for these tests are reverse straight leg raising and femoral nerve tension sign. This testing is performed with the patient in the prone position. The knee is flexed to 90 degrees. The hip is extended with the pelvis fixed to the table. Recreation of anterior thigh radicular pain is considered a positive test. Sacral sensory and motor function is not checked routinely in the athlete unless it is indicated by the history or other portions of the physical examination. To assess this function adequately, evaluation of perianal sensation,

Spinal Injuries

723

A

B

C

Figure 16B1-8   A, Straight leg raising test. B, Femoral stretch test. C, Milgram’s test.

sphincter tone, contractility of the anal sphincter, and the superficial anal reflex is required. This reflex is mediated by the S2, S3, and S4 nerve roots. Touching the perianal skin should cause contraction of the anal sphincter and external anal muscles. The history and physical examination can be modified depending on the circumstances and the individual. The complete history and physical examination cannot be carried out on the field of competition while the competition is delayed. The clinician should never feel rushed when evaluating a patient with a thoracolumbar spine injury, however, and should not consent to move the patient until he or she is convinced there is no evidence of serious injury. Only a small portion of this evaluation may be carried out acutely, and the remainder of it should be done as soon as possible. The importance of a careful history and physical examination cannot be overemphasized. When this portion of the

evaluation is completed, a working diagnosis or differential diagnosis should be established. This diagnosis guides the clinician through the remainder of the evaluation process, including the use of diagnostic testing.

Diagnostic Testing Evaluation of thoracolumbar spine pain in the athlete frequently includes many radiographic techniques used to visualize the structures of the thoracolumbar spine. These tests can be extremely helpful in establishing the diagnosis and treatment plan. It is essential to obtain the radiographic images in a timely fashion and to interpret them properly to allow determination of the true diagnosis. The underlying problem that complicates the relationship between thoracolumbar spine pain and radiographs is the progressive anatomic changes that occur naturally in the thoracolumbar spine over time. This is generally not a problem in

724

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

young athletes, but as more older athletes pursue competition, it becomes more of an issue. By the age of 50 years, 95% of adults who come to autopsy show evidence of disk space narrowing, calcification, or marginal sclerosis.12 In a similar series of living patients, degenerative changes are present in 87%.13 Conversely, only 5% of people younger than age 20 years have evidence of abnormal radiographic findings without symptoms. A radiograph of the thoracolumbar spine is the initial film made in the radiographic diagnosis of spine problems.14 The advantages are availability, low exposure of tissue to radiation, speed, and relatively low cost. Tangential radiographs offer good resolution of bone structures but do not show the soft tissue structures. It is not possible to image the entire thoracic and lumbar spine in one series of radiographs. It is appropriate to center the x-ray beam over the area of identified pathology. In general, only anteroposterior and lateral views are indicated initially. Oblique views provide limited information in specialized settings but should not be included routinely. The normal radiographic anatomy of the thoracic spine is shown in Figure 16B1-9. In this anteroposterior projection, the spinous processes, transverse processes, pedicles, facet joints, and laminae can be visualized. Each of these structures should be visualized at every level, with particular attention paid to the levels of the patient’s pain or discomfort. The normal lateral radiographic anatomy of the thoracic spine is shown in Figure 16B1-10. The bodies of the vertebrae, pedicles, spinous processes, and intervertebral

disk spaces are seen best in the lateral projection. The normal thoracic kyphosis can be seen to form a smooth curve. The thoracic intervertebral foramina can be visualized in the lateral view. The normal radiographic anatomy of the lumbar spine in the anteroposterior projection is shown in Figure 16B1-11. As in the thoracic spine, this projection best visualizes the transverse processes, spinous processes, pedicles, facet joints, and laminae. Particular attention should be paid to the alignment of the spinous processes and any lateral movement of the lumbar spine or rotation appearing as a deviation in the alignment. Particular attention should be paid to the transverse processes, which are clearly visible on this view. Often, direct blows to the lumbar spine result in fractures of the transverse processes. The facet joints in the lumbar spine run in a vertical orientation and are adjacent to the pedicles. Changes within the facet joint may be suggestive of degenerative disease. The soft tissue shadow of the psoas muscle can be seen in the anteroposterior projection. Any asymmetry of the psoas shadow should be taken only in its clinical context because this may be affected by positioning, spine rotation, or muscle contraction. The lateral radiographic anatomy of the lumbar spine is shown in Figure 16B1-12. In this projection, the bodies of the vertebrae, intervertebral disk spaces, pedicles, and spinous processes are well seen. One should be able to observe the normal lumbar lordosis with the posterior aspects of the bodies lining up to form a smooth curve. Movement of a single vertebral body in the horizontal plane, forward or backward, causes a disruption of this curve. The disk

Figure 16B1-9   Normal anteroposterior anatomy of the thoracic spine.

Figure 16B1-10   Normal lateral anatomy of the thoracic spine.

Spinal Injuries

Figure 16B1-11   Normal anteroposterior anatomy of the lumbar spine.

Figure 16B1-12   Normal lateral anatomy of the lumbar spine.

725

Figure 16B1-13   Oblique projection of the lumbar spine.

spaces generally should increase in size from L1 to L4. The lumbar intervertebral foramina are seen best on this view. Oblique projections of the lumbar spine are particularly useful in showing pathology within the facet joints or pars interarticularis.15 These projections should be obtained when pathology within one of these two regions is suspected. They are not of value in every series of spine radiographs. An oblique view of the lumbar spine is shown in Figure 16B1-13. In this projection, the facet joint is seen in profile and can be examined for asymmetry or degenerative change. The pars interarticularis also is well visualized. In visualizing these familiar Scottie dogs, the neck of the dog is represented by the pars interarticularis; the nose, the transverse process; the eye, the pedicle; the ear, the superior articular process; and the front legs, the inferior articular process. A collar, or disruption of the neck, is suggestive of spondylolysis (Fig. 16B1-14). Other specialized views of the thoracic or lumbar spine may be indicated in specific pathologic conditions. Lateral flexion-extension views may supplement the evaluation of either the thoracic or the lumbar spine when instability is suspected.8 There is great margin for error in lateral flexion-extension views based on the ability to reproduce the position and the interpretation of the examiner. It is important to remember the limitations of plain radiographs.16 Radiographs do not visualize the contents of the spinal canal, such as the spinal cord, dural structures, or spinal ligaments. They do not visualize the disk itself but merely the space that it is occupying. Significant destruction of bone may exist that does not show up on plain films. Rauschning15 showed that 50% of the medullary bone of a

726

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B1-14   Spondylolysis of the lumbar spine. Figure 16B1-16   Bone scan of stress fracture of pars interarticularis.

Figure 16B1-15   Normal bone scan.

vertebral body must be destroyed before it can be seen on plain radiographs. Radionuclide imaging, or bone scanning, is a sensitive technique for the detection of bone abnormalities. A small dose of radioisotope, generally technetium, is injected intravenously and allowed to circulate through the entire body through the bloodstream. Any process that disturbs

the normal balance of bone production and resorption ­produces abnormalities on bone scans. Increased osteoblastic activity is associated with increased concentration of the radionuclide tracer. Interruption of metabolic activities of bone results in a decreased amount of visible tracer. Radionuclide imaging depends on blood flow to bone. It also depends on the rate of bone turnover. In young athletes, the epiphyseal and metaphyseal bone plates are areas of increased bone activity and show up as areas of increased radionuclide concentration. Bone scan images are obtained with a scintillation camera that detects the emission of gamma radiation. A large-field camera may be used to cover the entire skeleton, or spot views may be taken in the area of maximal interest. Projections may be obtained in anteroposterior and lateral planes to allow localization of the increased radionuclide tracer. Images can be taken at different times after injection of the radionuclides; this allows differentiation between blood flow abnormalities and bone metabolism abnormalities. A normal bone scan mirrors the response of normal bones to mechanical pressures (Fig. 16B1-15). Radionuclide imaging is useful in circumstances in which radiographic changes lag behind increased bone activity17; this is particularly true in the detection of stress reactions of the pars interarticularis (Fig. 16B1-16). The major limitation of radionuclide bone scanning is its low specificity. Although it is extremely sensitive in picking up any abnormality of the bone blood flow or metabolic activity, it is extremely nonspecific. In the evaluation of thoracolumbar spine pain in the athlete, bone scanning should

Spinal Injuries

727

not be considered until after a routine series of radiographs has been obtained. If an area of point tenderness persists that does not have correlating radiographic abnormalities, a bone scan should be considered. Bone scans are indicated early in the evaluation of stress fractures or stress reactions of the pars interarticularis, such as occur in young gymnasts. Bone scans should never be used as a first-line test and should be reserved for specific indications. Computed tomography (CT) is useful for evaluating abnormalities of the thoracolumbar spine because of its complex three-dimensional, spatial anatomy.18 CT images not only re-create anteroposterior and lateral radiographs but also allow cross-sectional imaging in all three planes. CT images best assess bony configuration and structure and show graded shadings of soft tissue, such as ligaments, disks, nerve roots, and fat. CT allows excellent visualization of the paraspinal soft tissues. CT scanning should be used to confirm clinical findings derived from the history and physical examination. CT is not a tool for primary diagnosis, but rather one for confirmation of this diagnosis when a primary bony abnormality is considered. Many studies19,20 have shown that routine CT scanning for thoracolumbar pain is not useful or costeffective. CT scanning should be directed to the area of pathology and should not be used as a shotgun technique to evaluate the whole spine. A CT section of the lumbar spine contains different anatomic structures depending on the level of the cross section. Each CT cut is able to assess only one slice of the skeleton. Abnormalities that are not contained in that plane are not viewed by the CT scanner. Figure 16B1-17 shows a typical CT scan cross section through the intervertebral disk. CT scanning of the thoracolumbar spine in the athlete is particularly useful for the diagnosis of bony abnormalities.21 This examination generally is most useful in the evaluation of significant trauma involving fractures of the thoracolumbar spine in which the question of spinal canal impingement needs to be answered. CT scanning is helpful in the evaluation of tumors of the thoracolumbar spine for localizing the lesion and determining its extent.

The usefulness of CT scanning alone to identify significant disk herniations continues to be questioned. Many clinicians still believe that myelography must be used with CT scans to obtain adequate information about disk pathology. Magnetic resonance imaging (MRI) appears to have advantages that supersede those of myelography and CT scanning in the diagnosis of disk pathology. MRI allows excellent visualization of the soft tissues. As the imaging technology has advanced, MRI has become progressively more useful in the diagnosis of problems of the thoracolumbar spine. MRI is now the imaging modality of choice for all soft tissue injuries of the thoracolumbar spine. MRI also has a role in evaluation of bone injuries when they are causing significant impingement of soft tissue structures, such as the nerve roots. The principle behind MRI involves the generation of a magnetic field by protons of hydrogen atoms. Hydrogen atoms are the major constituent of water, which is found in varying amounts in all structures of the body. MRI allows visualization of all body structures and is not limited by direct changes in the density of tissues, as are other radiographic techniques. A normal MRI image of the lumbar spine is shown in Figure 16B1-18. MRI allows visualization of the vertebral column, intervertebral disk, and spinal canal. The axial views show the paravertebral soft tissues, the spinal canal, and the disk or vertebral body. At this time, MRI is an excellent technique for viewing the spinal canal and soft tissues about the spine, including the nerve roots and intervertebral disks.22-24 In injuries commonly seen in the athletic spine, MRI is a superb technique for visualizing disk pathology, nerve root compression, and ligamentous injury associated with hemorrhage. MRI is an excellent technique for characterizing primary changes occurring within the spinal cord or nerve roots, intramedullary tumors, and syringomyelia.

Figure 16B1-17   Computed tomographic cross section through intervertebral disk.

Figure 16B1-18   Normal magnetic resonance imaging scan of the lumbar spine.

728

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Two other radiographic techniques bear discussion in the evaluation of thoracolumbar spine injuries in the ­athlete—tomography and diskography. In a limited number of patients, tomography presents a useful alternative to CT or MRI, particularly if CT or MRI is not available. Tomographic slices through the vertebral bodies and posterior elements may define bony abnormalities. Diskography is performed by inserting a spinal needle into a disk space, then injecting radiopaque dye.25,26 Information is obtained by the radiographic appearance of the dye, the injection pressure used, and the reproduction of the patient’s pain during the test. There continues to be controversy at present concerning the validity of diskography. Radiographically, it is a sensitive technique to image the internal architecture of the disk when it is combined with CT scanning. The clinical significance of diskography must be assessed in conjunction with the patient’s pain response when the disk is injected. Diskograms rarely are indicated in the management of thoracolumbar spine pain in athletes. Instrumented kinetic muscle testing of the thoracolumbar spine is used infrequently in athletes. At this time, there are no studies documenting the ability of this technique to evaluate the thoracolumbar spine in athletes. With time and further studies, it is hoped that instrumented kinetic muscle testing may prove to be successful in the rehabilitation of the athletic spine as well as in predicting certain deficiencies that may predispose the athlete to injury.

REHABILIATION CYCLE The acutely injured athlete often doesn’t follow a linear route from injury to return to play (Fig. 16B1-19); ironically, re-expression of clinical symptoms or true recurrence may complicate rehabilitation. In addition, evolution of a biologic process, such as the transition from a lumbar anular tear to a lumbar disk herniation, may cause different clinical symptoms. Often the rehabilitation restores physical capabilities but may not change the underlying biologic process. Rehabilitation should be an ongoing restoration process with the ability to be adaptable in the face of acute change in function. The reduction or resolution of clinical spinal symptoms may occur by either conservative or surgical methodologies. In some instances, reduction of activity may entail immobilization. Activation is the beginning of movement and load as it applies to the spine. The activation process may be more sequential and gradual if it follows a period

Injury

Rehabilitation

Recovery

Figure 16B1-19   Linear route from injury to return to play that the acutely injured athlete often does not follow.

Reduction/ Resolution of symptoms

Return to play

Sportsspecific functionaltiy

Activation

Functional restoration

Figure 16B1-20   Diagram showing the cycle of rehabilitation

of immobilization. If possible, the activation process may occur simultaneously with reduction or resolution of symptoms. Functional restoration follows the activation process. Functional restoration is the restoration of strength, flexibility, balance, coordination, and aerobic and anaerobic conditioning. After functional restoration is achieved, and the athlete has symptomatic improvement, sport-specific functionality can be restored, which may include not only sport-specific drills but also other unique conditioning factors relevant to that sport. This process is often superimposed on, and continues with, general and spinal functional restoration. Once an athlete demonstrates sport-specific functional capacity and symptom control, he or she may be ready to return to play. The return-to-play decision is made in the context of multiple factors: the nature of the injury, the nature of the treatment, the individual athlete, and the exact requirements of the sport applied to the athlete. A cyclic rehabilitation program (Fig. 16B1-20) has been proposed that is both conceptual and specific and allows for transitioning up and down a ladder of activity and restriction based on creating certain rehabilitation goals.27 Unfortunately there is currently no universal, evidencebased standard of rehabilitation that applies to either specific spinal pathologies, specific sports, framework of injury, recovery, or return to play.

LUMBAR SPINE STABILIZATION (CORE STRENGTHENING) A recent analysis of literature related to core strengthening found multiple synonyms that included lumbar stabilization, dynamic stabilization, neuromuscular retraining, neutral spine control, muscular fusion, and trunk stabilization.28,29 Core strengthening has an historical past. Joseph Pilates, for example, focused on girdle strength by ­recruiting the

Spinal Injuries

deep trunk muscles.28 Core stability has been defined as an active support mechanism generated from intra-abdominal pressure and tensioning of the thoracolumbar fascia and deep lumbar stabilizers.30 It therefore requires the core strength of musculature. Core strength includes the entire trunk region, including pelvis, lumbar spine, and scapulothoracic region, with those regions providing a solid base of support for extremity movement.28 From the sports perspective, core strength is translated to core stability, which facilitates precise maintenance of lumbar and pelvic posture. Conversely, inadequate core strength can lead to poor core stability and may decrease biomechanical efficiency and increased risk for injury.28 Although theoretically sound, the diversity of what constitutes core strengthening programs and its lack of adequate objective measurement have hampered obtaining evidence-based data. It is important to investigate and understand the mechanisms and actions by which strength contributes to stability and how stability is achieved and maintained during static and dynamic tasks.28 Lumbar stabilization is the ability of the spine to be subjected to active and passive loads without failure. Stability is a dynamic process that includes both static positions and controlled movement.31 Panjabi has described a concept of stability focused on three components: bone and ligaments, muscle function, and neural control.32 Segmental stability is achieved by the interplay and interdependence of these three factors. Accordingly, instability can result from tissue damage, which makes the segment more difficult to stabilize, from insufficient muscle strength or endurance, or from poor muscular control. He believes that instability is usually a combination of all three factors.31 There is an essential relationship between stiffness and movement of the motion segment in order to dissipate forces and minimize energy expenditure.31 The neuromuscular system is responsible for modulating stiffness and movement to match the demands of internal and external forces.31 Spinal musculature may be considered both segmental and global. The lumbar spine consists of motion segments that are interrelated and coordinated with coupled functionality that contributes to normal physiologic curves (lordosis) and permits global movement of the entire unit. There are both segmental and global groups of muscles. The deep muscles in the lumbar spine control the intersegmental motion and the superficial muscles control more global motion. The deep muscles originate or insert on the lumbar vertebrae and are largely responsible for the control of stiffness and relationship between the vertebrae. The large superficial muscles of the trunk generate torque for spinal motion, are responsible for global motion, and handle external loads applied to the spine.31,33 There are theories about the role of activation of spinal musculature and the potential role of global muscles substituting for impaired deep muscle when the deep muscles are dysfunctional.31 Proper sequencing of muscle activation and appropriate response to movement and load may be functionally important and is a focus of further research. Interestingly, complex lumbar spine function may also include more generalized functions such as lumbopelvic stability, which requires control of whole-body equilibrium.31 Some

729

authors feel there is a close link between lumbar stabilization, posture, balance, and proprioception.31 One thorough review of lumbar stabilization raised five clinical concerns31: (1) whether or not exercise could reverse the changes seen in muscle mass, fiber type, strength, and endurance; (2) whether or not exercise could change neural firing patterns so that patients with low back pain could recruit their muscles in the same way as patients without back problems; (3) whether or not exercise could improve the proprioceptive and balance problems in patients with back pain; (4) whether or not patients with back pain or other spinal damage could participate in a stabilization program; and (5) whether or not lumbar stabilization could improve clinical outcome of patients with back pain.31 From a sports medicine perspective, the task is not just activities of daily living but the mechanical demands of a particular sport. Clearly, the functions of stability (load) and movement must be addressed to allow the spine to meet these demands as it passes along its age-related degenerative cascade and to function adequately. One recent study evaluated patients with various abnormalities on lumbar MRI scans.34 Back pain and disability were assessed both before and up to 12 months after therapy, and any improvements were compared to the baseline MRI findings. Eighty-nine percent of patients had one or more of the following findings: disk bulging, high intensity zones within the disk, or end-plate or bone marrow changes in at least one lumbar segment. Only 11% patients had none of these changes at any level.34 The two most salient points of the study were, not surprisingly, that MRI abnormalities showed minimal association with baseline symptoms and had no significant negative influence on the outcome after therapy.34 As in other studies, there is often little or no correlation between function and morphology. The relationship between specific pathologies, core strengthening, and eventual outcome has been investigated. In two separate studies with homogenous pathologies, patients benefited from a core strengthening program. Specific stabilizing exercises were applied to symptomatic patients with low back pain and radiographic evidence of spondylolisthesis and spondylolysis. Low back pain improved with exercises targeted to the deep abdominal and lumbar multifidi muscles.35 Postoperatively, patients who had dynamic lumbar stabilization had a better outcome than those who had no exercise or had a nonspecific exercise program.36 One study tried to predict treatment response to a stabilization exercise program for patients with low back pain based on clinical parameters.37 This study concluded that the clinical examination (history and physical) could elicit the factors that predicted success or failure with a trunk stabilization program. Another study compared outcomes of patients with low back pain who received treatments that were either matched or unmatched to subgroups that were based on the patient’s initial clinical presentation.38 The subgroups were based on the type of treatment believed most likely to benefit the patient (i.e., manipulation, stabilization exercise, or specific exercise). The patients were also randomly assigned to one of the three treatment protocols. This study concluded that nonspecific low back pain was not a homogenous condition. Outcomes could be improved

730

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

when patient ­ grouping was used to help guide treatment decision making.38 The clinical implication of that study was that subgroups of patients with low back patients exist and that there is a difference in their response to conservative, exercise-based treatment. Clinical, mechanical, and observational features of both acute and chronic low back pain may yield a better classification system for active therapeutic management than does imaging alone. One study evaluated stratification of load during exercise and attempted to quantify spine stability and loading by comparing eight stabilization exercises.39 The purpose of the study was to determine the relationship between strengthening and load, thereby allowing a continuum of lumbar spine rehabilitation decision making. That study attempted to provide clinicians with a better sense of the loads imposed on different spinal tissues and the resulting lumbar spine stability that occurred while performing commonly prescribed stabilization exercises.39 The graphical representation of exercise compared stability with compression and abdominal training with extension (Fig. 16B1-21). Matching clinical subgroups with directed and specific exercise programs depends partially on the mechanical demands on the spine that are anticipated. The level of endurance of the different trunk muscles and the production of certain motor patterns are linked with lumbar spine health.38-42 In some patients, the predominant goal for therapy is the need to minimize compressive load penalties and avoid certain deviated spine postures.39 Further research is required to permit the clinician to choose the appropriate matched variables involving stability and compression load with the appropriate subgroup of patients. One recent study demonstrated that patients with an acute first episode of low back pain treated with a lumbar stabilization program had a reduced risk for pain recurrence than a control group.43 The exercise group had a recurrence rate of 30% at 1 year and 35% in the 2- to 3-year follow-up period.43 The control group, on the other hand, had a recurrence rate of 84% at 1 year and 75% at 2 to 3 years.43

There are some challenges to the concept of core strengthening in treating recurrent back pain. A randomized, controlled trial comparing spinal stabilization exercises with conventional physiotherapy for recurrent low back pain concluded that patients with low back pain had similar improvement with both treatments. There appeared to be no additional benefit to specific spinal stabilization exercises over a conventional physiotherapy package for patients with recurrent low back pain.44 Other critics have stated that conventional physical therapy is as effective as core stabilization in treating low back pain. A systematic review of randomized controlled trials evaluating a segmental stabilizing program for low back pain concluded that segmental stabilizing exercises were no more effective than other physiotherapy ­interventions.45 For the management of low back pain in sports-related circumstances, exercise programs are an important adjunct in back rehabilitation. It makes sense that the components of a core strengthening or lumbar spine stabilization program would be important for return to sports, where significant loads are imposed on the lumbar spine. Lumbar stabilization exercises directed at the following four areas are thought to be important by some authors31,46: (1) deep musculature (such as the multifidi) that provide intersegmental lumbar vertebral control; (2) muscles (such as the transverse abdominis, diaphragm and pelvic floor) that increase intra-abdominal pressure to increase lumbar stability; (3) global muscles (such as the latissimus dorsi, quadratus lumborum, and superficial spine flexors and extensors) that control trunk movement and co-contraction during activities such as walking and lifting; and (4) precise neural control of these muscles.31,46 In conclusion, it is tempting to speculate that future work may bring a refinement to the evidence regarding core strengthening and low back pain. By creating accurate subgroups of patients with low back pain and matching objective load and stability data to the various subgroups, the clinician may have the tools to direct individual patients toward appropriate exercise pathways.

Higher Compression–Higher Stability Fpn_arm/leg Emphasis on Abdominals Abdcurl SideBridge Higher Compression-Moderate Stability Abdcurl SideBridge Bridge_leg

Lower Compression–Moderate Stability Fpn_leg Bridge

Emphasis on Extensers Fpn_leg Fpn_arm/leg Sitting on a chair Bridge Sitting on a ball Bridge_leg Lower Compression–Lower Stability Figure 16B1-21   Exercises recommended for specific goals. (Redrawn from Kavcic N, Grenier S, McGill SM: Quantifying tissue loads and spine stability while performing commonly prescribed low back stabilization exercises. Spine 29:2319-2329, 2004.)

Spinal Injuries

COLD THERAPY AND HEAT THERAPY There is little evidence suggesting that either heat or cold therapy is better for many spinal conditions. A Cochrane review addressed the issue of heat or cold for low back pain.47 This review concluded that there was limited evidence to support superficial heat and cold therapy for low back pain and that there was a need for better randomized controlled trials. In a small number of trials, there was moderate evidence that heat wrap therapy provided a slight short-term reduction in pain and disability in patients with a mix of acute and subacute low back pain. The addition of exercise was found to further reduce pain and improve function, but there was insufficient evidence to evaluate the effects of cold for low back pain, and there was conflicting evidence for any differences between heat and cold for low back pain.47

MEDICATIONS A review of 50 randomized controlled clinical trials assessing medications and low back pain found that there was evidence to support the effectiveness of nonselective nonsteroidal anti-inflammatory drugs in acute and chronic low back pain. Similarly, there was evidence to support muscle relaxants in acute low back pain and antidepressants in chronic low back pain.48 A Cochrane review of nonsteroidal anti-inflammatory medications also supported their use in acute low back pain but did not state whether one was more effective than others.49 Corticosteroids have been advocated in radiculopathy but have a less clear indication in the treatment of low back pain.50 It is generally recommended to use a proton pump inhibitor ����������������������������������������������� concomitantly ��������������������������������� for patients taking nonsteroidal anti-inflammatory medications to reduce gastrointestinal complications.51-60

INJECTION: DIAGNOSTIC AND THERAPEUTIC IN THE THORACIC AND LUMBAR SPINE The injured athlete may resolve an acute clinical syndrome involving the thoracic or lumbar spine regions by the natural history of the clinical problem, medication management, alteration of activity followed by active rehabilitation, and ultimately return to sports participation. Clinical subsets may include athletes who experience acute recurrent selflimited episodes, acute recurrent episodes with chronic smoldering clinical symptoms, or chronic persistent symptoms that may interfere with performance. Under these circumstances, questions naturally arise regarding a different level of diagnostic and therapeutic intervention. The logical trail for potential anatomic pain generators is the level of evidence assumed to establish the source. The diagnostic assessment and therapeutic modulation of these potential sources of pain has been summarized by Boswell and colleagues.61 The contained information reveals the changing nature of evidence criteria and its anticipated change with time. The definition of a structure capable of generating axial or referred pain in the thoracic or lumbar spine has some

731

  TABLE 16B1-2  Ability to Identify the Source of Back Pain Percentage of Patients Pain Source

Pang et al (1998)77

Facet joint Facet joint and nerve root Facet joint and sacroiliac joint Lumbar nerve root irritation Internal disk disorder Sacroiliac joint Sympathetic dystrophy No cause identified Total

24 24 4 20 7 6 2 13 100

Manchikanti et al (2001)78

40 13 26 2 19 100

From Boswell MV, Trescot AM, Datta S: Interventional techniques: Evidence-based practice guidelines in the management of chronic spinal pain. Pain Physician 10:7-111, 2007.

precepts. The structure should have a nerve supply, be capable of causing pain similar to that seen clinically, be susceptible to diseases or injuries known to be painful, and be shown to be a source of pain in patients, using reliable and valid diagnostic techniques.61,62 The source of pain identification is coupled to the strategy of managing that painful source. In some instances, it may not really matter whether the pain source is identified because the therapeutic option is either unpalatable or unconfirmed or disproportionate to the level of symptoms. Even the accuracy of identification of specific anatomic sites and the interpretation of testing are challenged by some authors.63-76 Two studies using diagnostic blocks as the strategy for determining pain source in the absence of other noninvasive evidence produced the results shown in Table 16B1-2.77,78 The anatomic arguments and literature sources supporting the facet (zygapophyseal) joints, the intervertebral disk, dorsal root ganglion, sacroiliac joint, postlaminectomy syndrome, and spinal stenosis as potential the pain-­generators are well documented by Boswell and colleagues.61 Interventional techniques use chemical means such as local anesthetics and steroids. For example, local anesthetics interrupt the pain-spasm cycle and nociceptor transmission, whereas corticosteroids reduce inflammation.61 Diagnostic interventional techniques may either suppress pain from an anatomic structure or stimulate pain from a suspected anatomic structure. Alleviation of pain from an anatomic source by diagnostic blockade makes inherent sense, but there is variability in accuracy of injection and in sensitivity and specificity for each structure and invasive test. Pain stimulation from an anatomic structure may demand an even greater degree of difficulty in interpretation because the patient is asked to separate a potentially painful experience inherent in the test itself from the usual experience of their pain, known as concordant pain. The balance between the degree of intervention, level of disability, and often a season-based time constraint for return to sports participation requires experience and judgment. Familiarity with the spinal pathologies in the nonsports environment is helpful in assessing the context of routine activities of daily living versus performance-oriented

732

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

sports. Added to the texture of decision making are potential career choices in the professional setting and physical achievement with a variety of goals in the amateur environment. Interventional techniques may be used for their short-term value in relieving pain, serving as a symptom suppression vehicle to facilitate transition into functional rehabilitation or other therapeutic considerations. The evidence reviewed by Boswell and associates summarizes and characterizes the long-term relief from interventional procedures as being moderate to limited.61 Further discussion of the various injection techniques and level of evidence is beyond the scope of this chapter. Interested readers are referred to the work by Boswell and colleagues.61

TRIGGER POINT INJECTION A recent review summarized the role of trigger point injections in the nonoperative management of acute and chronic low back pain.79 This study recommended against using trigger point injections in nonspecific categories of acute or chronic low back pain.79 It concluded that injections could be considered for patients with back pain secondary to suspected myofascial syndrome, thought to be to the result of hyperirritable foci of taut muscle bands.79,80 Trigger point injections are often used in the context of a more comprehensive program of physical therapy and medication and the number of injections should be limited.79 Other studies have also concluded that myofascial trigger point injections should only be used within the framework of other treatment modalities.81 The choice of chemical agents for trigger point injections varies widely among clinicians. For example, one recent study compared botulinum toxin A to bupivacaine trigger point injections for the treatment of myofascial pain syndrome and concluded there was no outcome difference between the two and thus recommended the more economical bupivacaine.82 Critics of trigger point injections point out the lack of scientific evidence supporting its use. One study found no benefit to botulinum toxin A trigger point injections for cervicothoracic myofascial pain.83 Another study recommended against the use of trigger point injections generally because of lack of sufficient evidence to support their usage.84

THORACOLUMBAR SPRAINS AND STRAINS The sports medicine physician will be confronted by patients with soft tissue injury in the thoracic and lumbar spine sustained during participation in sports activities or precipitated in the nonsports environment but with symptoms interfering with performance and activity. Since 80% to 90% of the adult population will experience nonspecific low back pain during their lifetime, at least transient interference in sports is common. In general, recovering from soft tissue injuries of the thoracic and lumbar spine should have a favorable course. The biggest issue is the timing of injury and lost playing time, especially in competitive athletics. Bono cited three different studies showing the influence of low back pain and playing time: 30% of college football players lost time,

38% of professional tennis players, and 90% of professional golfers.85-88 A sprain may be defined as an injury to a ligament that may affect individual fibers but does not disrupt the continuity of the structure.85 The muscle fibers may be disrupted at the junction of muscle and tendon or within the muscle itself.85

Thoracic The key points in the evaluation of soft tissue injuries in the thoracic spine are an accurate history and physical examination. It is important to determine whether or not there was a direct blow or twisting mechanism that caused soft tissue injury. Pain location and onset are often helpful in evaluation of soft tissue strains and injuries, but specific recall may not always be possible but may be represented by a change in activity or alteration of training. The most common injury of the thoracic spine in athletes involves the soft tissues. Soft tissue injuries are either musculoligamentous strains or sprains or contusions related to a direct blow. In the differential diagnosis of thoracic region pain and athletes, Karlson linked anatomic location with injury probability. Pain located in the superior thorax was correlated with first rib stress fractures; pain located in the mid-thorax anteriorly is most likely costochondritis; pain located in the mid-thorax anterolaterally to posteriorly was most likely a rib stress fracture; pain located in the mid-thorax posteriorly was considered most likely rib subluxation; pain located in the inferior thorax was most likely a slipping rib.89 Other muscular injuries include the intercostal muscles, the serratus anterior muscle, and miscellaneous muscle avulsions.89 Physical examination findings of importance include the location and the size of the area of tenderness. Consideration of the differential diagnosis of thoracic area symptoms may lead to further diagnostic testing, or confidence in history and physical examination may allow the clinician to confirm a soft tissue diagnosis. Most strains and sprains of the thoracic region are treated like other areas of the body. Controlling the inflammatory process by local measures (e.g., cold therapy) and nonsteroidal anti-inflammatory medicine may facilitate reduction of clinical symptoms. A program of rehabilitation, reconditioning, and sport-specific training will allow a gradual return to full participation. The time required for this process to occur is often a function of the extent of injury and the nature of the aggravating factor related to a specific sport that induced the injury. Stretching to allow return to normal range of motion and muscle strengthening theoretically reduce the risk for repeat injury. Contusions of the thoracic spine are usually related to a direct blow to the bony elements or the paraspinal musculature. Muscle spasm, loss of range of motion, and considerable pain accompany this type of injury. Reduction in inflammation is important as the initial phase of treatment, which may include local modalities as well as medication. Once pain has subsided, the next goal is achieving normal range of motion and normal function. Normal range of motion includes rotation, lateral flexion, and extension of the thoracic region. Muscle conditioning and stretching should be targeted to these functions. Initially, isometric

Spinal Injuries

muscle contraction is begun while there is still significant pain and loss of motion. When pain and muscle spasm subside, stretching of the thoracic spine is instituted, again primarily involving lateral flexion and rotation motions. Ballistic stretching is absolutely contraindicated. Since individual flexibility among athletes varies, no specific guideline can be given as to the amount of lateral flexion and rotation that can be achieved. The return to sports requires sufficient range of motion and pain control so that the athlete is not risking further injury by splinting of the thoracic spine.

Lumbar One recent study found that muscle strains were the most frequent injury and that acute back injuries were significantly more common than either overuse injuries or injuries associated with preexisting conditions.90 It is important to describe the event leading to pain from a lumbar strain as well as the provocative and palliative factors. It is also important to determine whether there is an underlying chronic condition contributing to the problem. Clinical observation of muscle spasm, restricted range of motion, and localized tenderness should be sought. X-rays are usually not helpful but may be valuable in determining whether underlying structural abnormalities are present. After an acute injury, a short period of cold therapy to limit localized tissue inflammation and edema should be instituted for about 48 hours. In general, the most debilitating portion of the musculoligamentous injury is the associated muscle spasm. This spasm can be controlled with the use of antispasmodic medications. A short trial of a lightweight lumbosacral corset may also help to control muscle spasm and may make the athlete feel more comfortable. When the initial period of spasm and pain has been controlled, treatment is directed toward rehabilitation, which includes strengthening and regaining normal range of motion. There have been attempts to classify low back pain in athletes that has no radicular component.91 Exercise within a pain-free zone of motion and restriction of painful postures are fundamental principles.91

Anterior column

733

Return to sports participation after lumbar ligamentous strains and sprains requires an individualized program for the athlete based on the particular demands of the sport.

THORACIC AND LUMBAR FRACTURES The potential injuries derived from the wide variety of sports activities means that almost every thoracic and lumbar fracture pattern may be encountered. Clearly some sports are inherently riskier than others in this regard. Hang gliding, mountaineering, racecar driving, skiing, and snowboarding would presume a different risk profile than sports such as golf, tennis, and long-distance running. Some authors have classified the incidence of thoraciclumbar spine injuries by anatomic zones: T11-L1 (52%), L1-L5 (32%), T1-T10 (16%).92 Important considerations in the evaluation of thoracolumbar trauma include neurologic status, spinal stability, and associated internal organ injury. In the more mature athlete, consideration for preexisting bone quality may be a factor in susceptibility to fracture. The history of the mechanism and force of injury may also be important to lead the astute clinician to search for more occult spinal injury. The neurologic status of the patient is paramount and is a critical consideration in terms of potential nonoperative versus operative treatment and ultimate return to play. The anatomic level and severity of neurologic involvement may produce a variety of neurologic patterns. Decision making for surgical treatment of thoracolumbar fractures depends on three factors: injury mechanism and pathology, neurologic status, and posterior ligament integrity.93 Although there are multiple classifications systems to assess mechanism of injury and inherent spinal stability, the three-column classification of spine injury is commonly used (Fig. 16B1-22).94-98 The spine is divided into three distinct columns, and injuries to these columns become the basis of classification.98 The anterior column is represented by the anterior longitudinal ligament and the anterior portion of the vertebral body.98 The middle column comprises

Middle column

Posterior column

Figure 16B1-22   The commonly used three-column classification system to assess mechanism of injury and inherent spinal stability.

734

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the remaining part of the vertebral body and the posterior longitudinal ligaments.98 The posterior boundary of the middle column is the origin of the pedicles.98 The posterior column is composed of the remaining posterior elements, including pedicle, lamina, facet joint, spinous process, and posterior soft tissues.98 Use of this classification system forms the basis for evaluating whether a fracture is stable or unstable and influences treatment decisions. If there is a suspected fracture at the time of injury, immobilization on a spine board should be performed if there is any question regarding spinal stability or the risk for additional neurologic impairment. The injured athlete is immobilized and then can then be safely transferred to where further assessment of stability and neurologic status can be performed. The mechanism and other important historical features of the injury should be recorded. This includes not only the description of the event but the immediate and subsequent symptoms experienced by the athlete. Exact location and description of axial pain, presence of numbness, weakness, or pain in the lower extremities, even if only transient, should be sought. Prior history of spine-related problems should be obtained. Physical examination may be limited because of pain, but inspection and palpation and a thorough neurologic examination should be performed. Imaging should include anterior-posterior and lateral radiographs. Since the three-column classification system is used to assess spinal stability, particular attention should be paid to the lateral x-ray. Dynamic x-rays in flexion and extension are often not useful initially because of limited motion from pain. Visualization of the x-ray pattern may suggest compromise to the spinal canal or potential instability secondary to ligamentous injury. Ligamentous injury in the posterior column may be suspected when there is abnormal spacing between two adjacent spinous processes compared with other levels. The spinal canal can be assessed by CT scanning, which can provide axial, sagittal, coronal, and three-dimensional imaging. In the absence of neurologic injury, a CT scan is generally more valuable than an MRI in providing information about bony anatomy and fracture classification. MRI provides superior information regarding soft tissues, ligament, disk, and neural structures.92 In the injured athlete, treatment determinations in traumatic spinal injuries are focused on fracture configuration, spine stability, and neurologic status.

r­ etroperitoneal or intra-abdominal process. The presence of hematuria should raise a red flag for possible renal injury and should warrant further evaluation. Some authors have suggested CT scanning if plain radiographs reveal lumbar transverse process fractures to make sure there are no other concomitant spinal fractures.101 In a review of professional football players with transverse ­ process fractures “associated visceral injuries were rare and the time lost from sports is only an average of 3.5 weeks.”102 Treatment is directed at pain reduction and allowing adequate time for biologic healing of the fracture to occur, although healing of the transverse process often does not occur because of distraction of the fracture fragment from muscle pull. Healing is often judged by x-ray evaluation and by direct palpation of the site of injury. Once painless range of motion is achieved and mobilization and strengthening of the trunk accomplished, return to active sports is permitted. In noncontact sports, equipment changes may not be necessary to protect the area, but in contact sports, padded equipment modification may be helpful in reducing the risk for reinjury.

Thoracic Compression Fractures The most common fracture in the thoracic spine region is the compression fracture (Fig. 16B1-23). Compression fractures result from failure of the anterior bone column in compression as a flexion moment occurs. There is usually a significant history of trauma in the younger athlete to generate the force required to produce a thoracic compression fracture. Diminishing bone mass from osteoporosis may predispose to fracture in the older athlete. Although the middle column is intact with a compression fracture, and canal ­compromise

Lumbar Transverse Process Fractures Direct trauma to the lumbar area may cause fractures of the transverse processes or other posterior elements. These fractures are stable and are rarely associated with neurologic injury. Intra-abdominal injuries can occur in association with fractures of the transverse process or with blunt trauma to the area of the costovertebral angle. A high index of suspicion for retroperitoneal and intra-abdominal processes is warranted in athletes who have a fracture of the transverse process.99,100 Examination of the abdomen may reveal guarding or tenderness, and tenderness in the costovertebral angle should raise concern about possible abdominal injury, particularly to the kidney. Further consultation or directed assessment may be necessary to ­ diagnose a

Figure 16B1-23   Thoracic compression fracture not visualized on lateral radiograph in mature recreational walker but identified on bone scan.

Spinal Injuries

is therefore unlikely, it is important to be sure that the neurologic examination is normal and that there is no associated transient neurologic abnormality. Physical examination may reveal point tenderness and guarded range of motion. Usually the diagnosis of compression fractures of the thoracic spine is made by anteroposterior and lateral x-rays. The lateral x-ray is usually diagnostic of compression fracture. The amount of compression can be determined by comparing the height of the anterior and posterior aspects of the vertebral body. Careful evaluation of other vertebral bodies may reveal more than one compression fracture. Most compression fractures in athletes will show less than 25% compression of the anterior vertebral body compared with the posterior vertebral body. If anterior vertebral body compression is more than 50% of the posterior vertebral body height or adjacent intervertebral bodies, CT should be considered to assess the spinal canal. Compression fractures with less than 25% compression deformity are treated symptomatically with analgesia and possibly immobilization in a thoracic orthosis and exclusion from sports. The purpose of bracing is to provide immobilization for pain relief and to prevent further ­flexion of the thoracic spine, which could increase the deformity. The duration of bracing depends on pain relief and the underlying bone quality. The two most common forms of braces are the Jewett extension orthosis and a molded polypropylene thoracolumbar spine ­ orthosis (TLSO). Cement augmentation techniques (e.g., kyphoplasty and vertebroplasty) involve the injection of cement into the compression fracture and have become more popular techniques for the treatment of the osteoporotic compression fracture.103-105 Vertebroplasty involves the percutaneous instillation of liquid cement into the fractured vertebra, whereas kyphoplasty expands the vertebral body through inflation of a balloon, which creates a void for the placement of viscous cement (RF12-F14). Their role in the treatment of vertebral compression fractures in younger individuals, including young athletes, is unknown.

735

Thoracolumbar Burst Fractures Burst fractures of the lumbar spine are caused by direct axial loading of the spine or by a combination of flexion and axial loading. This injury involves not only the anterior column but also the middle column in compression and is frequently associated with bony retropulsion into the spinal canal and neurologic involvement. Thoracolumbar burst fractures without neurologic involvement and with mild to moderate canal narrowing can be managed conservatively (Fig. 16B1-24).106,107 Burst fractures involving more than 50% of the vertebral body, producing more than 50% bony impingement of the spinal canal, or having more than 20 degrees of kyphosis at the level of fracture have traditionally been considered indications for surgery.108 Surgical goals include restoring anatomic height of the vertebral body, providing spinal stability, and achieving spinal canal decompression by either direct or indirect decompression. The surgical approach may be anterior, posterior, or combined. Anterior spinal surgery involves a retroperitoneal approach to the affected segment, corpectomy with removal of bony fragments from the spinal canal, and vertebral body reconstruction of the corpectomy defect and stabilization. Spinal canal restoration can also be accomplished with posterolateral decompression of the bony fragments in the canal and vertebral height restoration with indirect decompression by distraction-lordosis and instrumentation. Combined anterior and posterior approaches can also be used with the anterior approach ensuring adequate spinal canal decompression and middle column reconstruction with biomaterials, followed by posterior instrumentation fusion to ensure adequate stability for early mobilization of the patient. Whether an anterior, posterior, or combined approach is used, the patient is typically mobilized in a TLSO for 3 months.

Thoracolumbar Fractures Summary Other injury patterns not described in this chapter include distraction-flexion injuries, fracture-dislocations, and ­ distraction-extension injuries. Characteristics and Figure 16B1-24   A 38-year-old dentist involved in a dune buggy “hard landing” sustained a burst fracture of L1 without neurologic deficit and was treated with a thoracolumbar spine orthosis. She is asymptomatic and has returned to normal sports activities but not dune buggy riding.

736

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B1-25   Sacral stress fracture in a female collegiate basketball player with complaints of back and leg pain. She successfully returned to collegiate basketball after conservative treatment.

t­ reatment of these injuries is beyond the scope of this chapter but have been described elsewhere.92,109 In the context of acute thoracolumbar injury evaluation, appropriate immobilization and transportation without precipitation or aggravation of existing neurologic injury forms the initial principle of management. History, physical examination, and diagnostic imaging (plain x-ray, MRI, CT) form the cornerstone of assessment and in deciding on surgical or nonsurgical treatment of these injuries. Decision making for the surgical treatment of thoracolumbar fractures concentrates on three factors: injury morphology, neurologic status, and posterior ligament integrity.93

SACRAL STRESS FRACTURE In 1989, Volpin and associates reported on stress fractures of the sacrum following strenuous activity.110,111 In their series of three cases, sacral stress fractures were identified in a military population.111 The concept of fracture in the sacrum without direct trauma is either due to insufficiency in bone mass (osteoporosis) and subsequent failure or repetitive microtrauma from overuse leading to fatigue fracture. Sacral stress fractures result from concentration of the body forces that are dissipated from the spine to the sacrum and ala.110,112 The usual causes of athletic stress fractures, generally, is an increase in the intensity and duration of a sports activity or a change in the way the athletic activity is performed, resulting in concentrated load on the affected anatomic structure. Sacral stress fractures should be considered in the differential diagnosis of low back and leg pain in the athlete (Fig. 16B1-25).

An underlying metabolic bone disease may lead to insufficiency fracture in the athlete. Older athletes may have osteoporosis as a contributing factor to insufficiency fracture. The typical sacral stress fracture in the athlete may begin with either an acute history or the development of insidious unilateral buttock, upper thigh, or low back pain. There are no pathognomonic physical examination findings diagnostic of sacral stress fractures. Various physical maneuvers designed to load the area have been described, such as the hopping test.110,113 MRI scan of the pelvis and single-photon emission computed tomography (SPECT) and bone scan typically reveal the stress fracture. A three-phase protocol to diagnose and manage pelvic stress injuries in the athlete has been described.114 Phase I involves stopping the painful activity, including substituting non–weight-bearing activities such as swimming or cycling for weight-bearing activities in order to maintain aerobic capacity. Phase II is initiated after a 3- to 5-day pain-free interval. This phase employs light-weighted exercises to facilitate strength and correct strength imbalances. Sport-specific rehabilitation is initiated at this time. Phase III focuses on return-to-sport specific activity with gradual progression every other day to normal activity.114 This process of recovery can take from 3 to 18 weeks.115

Kyphosis and Scoliosis Spinal deformity may be encountered in adolescents and young adults as an intrinsic structure rather than a derivative of sports participation. Scoliosis may be defined as a coronal plane pathology and kyphosis as abnormal curvature in the sagittal plane.

Spinal Injuries

The classification of scoliosis can be related to its underlying generation: congenital (present at birth), neuromuscular (deformity associated with impaired neuromuscular states), and idiopathic (undefined). The typical athlete with scoliosis most likely will have congenital or idiopathic scoliosis.116-118 The classification of kyphosis is also related to presumptive underlying cause: congenital (present at birth), neuromuscular (sagittal plane deformity associated with impaired neuromuscular states), postural (related to posture), Scheuermann’s disease (vertebral body wedging and associated end plate changes).119 In the sports medicine context, spinal deformity is usually an incidental finding, rather than a result of specific injury. Inspection and palpation may reveal the underlying deformity: asymmetry of shoulder and pelvic height, abnormalities in spinal contour expressed as prominent asymmetric thoracic chest wall or paralumbar musculature, round back posture, or leg-length inequality associated with the spinal deformity. Exclusion of neurologic abnormalities is important in assigning spinal deformity. Radiographic evaluation of suspected spinal deformity includes standing posteroanterior and lateral x-ray views. These views allow the quantitative measurement of the projected curves, which is a critical factor in deciding treatment. There are four general patterns of curvature: thoracic, lumbar, thoracolumbar, and double major curves. The rationale for full spine standing anteroposterior and lateral x-rays is the ability to assess all the potential curves that might be associated with deformity. These curve patterns have been classified by King and colleagues and more recently by Lenke and associates.120,121 Plain x-ray coronal plane measurements include the determination of the central sacral vertical line, the end vertebrae, the apical vertebra, and the stable vertebra. Kyphosis is a normal physiologic curve in the thoracic spine and has a normal range of 20 to 40 degrees. Scheuermann’s kyphosis is defined by the following characteristics: anterior wedging greater than 5 degrees involving three or more consecutive vertebral bodies, kyphosis greater than 45 degrees on the standing lateral x-ray, irregularity of the vertebral end plates, and Schmorl’s nodes, which are depressions in the vertebral bodies from ballooning of the disk into the vertebral end plate.122 Other abnormalities may be seen in patients with Scheuermann’s kyphosis, including mild scoliosis and spondylolisthesis.122 Curves less than 25 degrees can be observed for progression.122 Bracing should be considered in the skeletally immature patient for curves measuring 30 to 45 degrees, or for curves greater than 25 degrees with documented progression of more than 5 degrees.122 Conservative management of Scheuermann’s kyphosis consists of bracing at 50 degrees in addition to an exercise program.122 Round back deformity (not possessing the radiographic criteria of Scheuermann’s) is treated with thoracic extension exercises. In general, indications for surgical intervention for scoliosis include curve progression despite bracing, curves greater than 40 degrees in the skeletally immature, and curves greater than 50 degrees with a mature skeleton.122 Indications for surgery in Scheuermann’s kyphosis include curves greater than 70 degrees, intractable pain,

737

curve progression, neurologic compression, cardiopulmonary compromise, and significant trunk deformity.122,123

Adult Scoliosis Scoliosis may be encountered in the adult years as a derivative of preexisting adolescent scoliosis or as a new-onset deformity. The patterns of deformity, the amount of concomitant degenerative changes in the spine, the natural history of deformity progression, and the clinical presentation are different in the adult as compared with the adolescent with scoliosis.124 Indeed, it is often the symptoms from the associated degenerative change that results in clinical presentation in the adult with scoliosis. These changes include spinal stenosis, spondylolisthesis, rotational subluxation, lumbar hyperlordosis, and rigidity of the curve.124 The Scoliosis Research Society has recently defined a classification system for adult spinal deformity that is different from the adolescent. This classification system is a radiographic classification as opposed to clinical classification, in which there are six major coronal deformities and a single sagittal plane deformity without associated thoracic or lumbar coronal deformities that would meet the requirements of the primary coronal deformity.124 The goal of this classification system is the recognition that there are unique challenges related to adult scoliosis that are inherently different from adolescent growth period issues. The specifics of this classification are beyond the scope of this chapter and are described elsewhere.124-131

Degenerative Lumbar Scoliosis Degenerative lumbar scoliosis presents a difficult combination of multiplanar deformity. A radiographic study of degenerative lumbar scoliosis tried to define x-ray parameters and canal measurements.132 Associated degenerative changes such as hypertrophy of the ligamentum flavum, posterior disk bulging, and bony overgrowth are thought to be more likely to contribute to stenosis regardless of the scoliosis.132 Degenerative scoliosis may exist without neurologic symptoms, or it may occur in association with spinal stenosis with symptomatic back and leg symptoms. A quantitative radiographic and clinical analysis demonstrated that lateral listhesis, L3 and L4 obliquity, lumbar lordosis, and thoracolumbar kyphosis were significantly correlated with pain.133 Degenerative lumbar scoliosis in association with lumbar spinal stenosis usually involves the L3 and L4 nerve roots, which are foraminally compressed in the concave side of the curve, and the L5 and S1 nerve roots, which are compressed in the lateral recess on the convex side of the curve.134 Sports medicine considerations for management of the active older athlete with adult scoliosis typically involve traditional nonoperative measures.135 There will be active individuals with some measure of deformity who wish to maintain their levels of performance. In addition, many individuals with superimposed osteoporosis and adult scoliosis will need to strike a balance between desired level of performance and symptom overload. The complexity of surgical decision making, as it applies to adult scoliosis, is beyond the scope of this discussion.

738

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

THORACIC DISK HERNIATION Thoracic disk herniation can occur spontaneously without a significant history of trauma or specific athletic causation. The most commonly involved levels are T9-10, T10-11 and T11-12.136 A retrospective review of thoracic disk herniations found that 27% of the patients had surgery and 73% were treated conservatively.137 This study concluded that thoracic disk herniations only occasionally lead to surgery and that many patients return to active lifestyles without the need for surgery.137 Another study reviewed the natural history of asymptomatic thoracic disk herniations and concluded that the patients rarely developed symptoms during the period of review.138 In a study of serial thoracic MRI scans assessing degenerative changes, thoracic disk herniation was present in 10% of subjects on the initial scans, and 27% of patients improved radiographically.139 An additional 1.5% of subjects were found to have another thoracic disk herniation at the time of follow-up MRI.139 In another review of thoracic disk herniations, initial symptoms included pain in 57% of patients, sensory disturbance in 24%, motor involvement in 17%, and bowel and bladder dysfunction in 2% of patients.140,141 The presenting signs and symptoms on the initial evaluation involved motor or sensory symptoms in 61% of patients and bowel and bladder dysfunction in 30%.140,141 Chest wall or upper abdominal pain may result from intercostal nerve root involvement from posterolateral thoracic disk herniations.142-145 Central disk herniations may be painless but can present with lower extremity neurologic dysfunction. Lower extremity paresis or paralysis may be a presenting finding.142,146 Symptoms mimicking a C8 radiculopathy can occur with T1-T2 herniated disk involving T1 nerve root.147 Physical examination may reveal dermatomal numbness in the distribution of an involved intercostal nerve from a posterolateral herniation. Spinal cord compression from a central disk herniation may present with spastic paraparesis and long tract findings, such as hyperreflexia, clonus, and upgoing plantar response (Babinski sign).148,149 Abdominal and cremasteric reflexes should be sought as part of the examination.150 Diagnostic studies include MRI and myelography followed by CT (CT-myelography). MRI directly visualizes the thoracic spine, including the disk and the neural elements, and reveals any spinal cord compression. Thoracic CT-myelography may also be helpful in delineating anatomic defects and location of a thoracic disk herniation when MRI is inconclusive. The treatment of thoracic disk herniation depends on the clinical syndrome and location of the disk herniation. A posterolateral herniation that does not compress the spinal cord and only involves an isolated nerve root producing intercostal nerve root pain may be treated conservatively. If the pain is severe and does not resolve, surgical ­intervention can be considered. With more central thoracic disk herniations leading to neurologic signs and symptoms, surgery may be required. Surgical options include anterior excision through a transthoracic approach for disk spaces T4 through T12140-151 or minimally invasive video-assisted

thoracoscopic surgery for central herniations.152-154 Posterior costotransversectomy may be used for paracentral and lateral disk herniations with the approach from the side of the disk herniation.140,155,156 The costotransversectomy involves removal of pedicle, hemilamina, facet joint, and medial rib to allow access to the canal.140 Posterior transpedicular strategies involve unilateral removal of pedicle and facet to gain access to the thoracic spinal floor while avoiding the spinal cord.157-159 Alternatively, a transfacet approach can be used that provides limited exposure for access to the central canal but can provide good exposure for lateral disk herniations.140,160,161 Return to full sports activity is dependent on the clinical circumstance and the nature of the treatment. For conservatively treated patients without neurologic deficits, return to full sports activity without restriction can be anticipated after an adequate rehabilitation program is completed. For thoracic disk herniations surgically decompressed without fusion and having complete recovery, the decision for return to sport participation should be individualized. Following thoracotomy with disk excision and fusion, many surgeons would not allow contact sports, although some would consider this on an individualized basis (Fig. 16B1-26). A solid bony fusion should be achieved, and neurologic recovery must be complete for the patient to return to sports after successful rehabilitation.

THORACIC SPINAL STENOSIS Thoracic spinal stenosis is less common than either cervical or lumbar spinal stenosis because of the relative lack of motion of the thoracic spine. Thoracic spinal stenosis can occur from calcified disk protrusions, posterior vertebral body osteophytes, ossification of the posterior longitudinal ligament, facet joint or lamina hypertrophy, or ossification of the ligamentum flavum.162-166 Congenital narrowing of the spinal canal may also exist and can be a contributor to stenosis. If a critical diminution of space available for the spinal cord results, then neurologic findings may ensue. The most common presentation for symptomatic spinal stenosis involves a gait disorder characterized by spastic gait. Lower extremity numbness with a defined sensory level at or below the level of spinal cord compression may also be present. Lower extremity weakness and bowel and bladder symptoms may also occur. The full spectrum of upper motor neuron findings may be elicited, including lower extremity hyperreflexia, clonus, and abnormal plantar response (Babinski sign). From an athletic perspective, thoracic spinal stenosis is most likely to present in older patients with acquired (degenerative) narrowing of the spinal canal. Younger patients may become symptomatic from a combination of a preexisting congenital narrow canal with superimposed degenerative changes. Asymptomatic thoracic spinal stenosis does not require surgical intervention. Return to full participation in sports in the asymptomatic individual with thoracic spinal canal stenosis depends on the nature of the sport, the degree of narrowing, and the clinician’s opinion as to the potential for injury. There are no unequivocal guidelines for these decisions.

Spinal Injuries

739

Figure 16B1-26   A 35-year-old recreational athlete with the delayed onset of long tract lower extremity symptoms with thoracic disk herniation treated with transthoracic decompression and interspace fusion with rib graft. Neurologic symptoms resolved, and he is in active rehabilitation in preparation for return to recreational sports.

If clinical findings of spinal cord compression are present, surgical treatment is indicated. The choice of surgical approach is often dictated by the location of spinal cord compression (Fig. 16B1-27).167 For predominantly anterior compression, an anterior approach would be preferred. For predominant posterior pathology or posterolateral compression of the spinal cord, a posterior decompressive laminectomy may be performed.167

Recent studies have shown satisfactory results for thoracic laminectomy for thoracic spinal stenosis.168-170 However, one study noted a 14.5% neurologic complication rate with posterior decompression for thoracic spinal stenosis.171 After resolution of neurologic symptoms following posterior thoracic decompressive laminectomy, return to sports participation can be considered, although collision sports should probably be avoided.

Figure 16B1-27   Middle-aged recreational walker with subtle onset of gait disorder rapidly progressing to extreme difficulty walking. Posterior thoracic decompression resolved the clinical symptoms in this patient with a combination of congenital narrowing and facet joint encroachment into the spinal canal.

740

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

LUMBAR DEGENERATIVE DISK DISEASE Lumbar degenerative disk disease is a universal age-related condition consisting of morphologic changes in the lumbar motion segment. These changes may be symptomatic or asymptomatic. Disk pathology, including focal disk protrusions and anular tears, are common in the asymptomatic population.172 Degenerative changes in the lumbar spine are present in virtually everyone by middle age.173 These changes consist of disk degeneration and marginal osteophyte formation of the vertebral bodies with remodeling changes in the facet joints. Disk degeneration is generally thought to be the primary pathologic process, with facet joint changes occurring secondarily. Because about 80% of the population has been estimated to experience transient back pain at some point in time, there is clearly an interface where degenerative changes may contribute to an active symptomatic state and periods of time where the architectural changes are asymptomatic. It should be remembered that back pain is a symptom and not a diagnosis, and both spinal and nonspinal causes of low back pain must be considered. Spinal causes of back pain are numerous and can include the disk, facet joints, bones, ligaments, muscles, and nerves. Extraspinal causes of low back pain include both intra-abdominal and intrapelvic conditions, such as ulcer disease, abdominal aortic aneurysm, renal conditions, endometriosis, and many other conditions. The natural history of low back pain is relatively benign, with up to 90% of patients having resolution within 6 weeks and only 1% having low back pain for longer than 1 year.174 Because symptoms may resolve quickly, imaging is usually not indicated for acute low back pain if noninvasive conservative treatment is planned. Recurrence of low back pain is common, with subsequent episodes occurring in 20% to 67% of patients.175,176 It can be difficult to identify the cause of low back pain. The degenerative changes commonly seen on imaging studies may not be the source of pain.177,178 MRI has not been found to be a reliable predictor of future low back pain in asymptomatic patients.179 In addition, MRI is typically static imaging, which may not predict what actually occurs during dynamic activity and under loading. It has been found that structural abnormalities on MRI do not preclude successful lumbar spine rehabilitation.180 If it is important to identify the source of pain, invasive diagnostic testing is indicated. Usually this involves either the stimulation of pain or the suppression of pain from a particular anatomic site. Although theoretically logical, invasive testing has variable accuracy in its ability to localize the pain generator.181 In addition, invasive testing cannot reliably localize all potential pain generators, such as the disk, ligament, muscle, tendon, or facet joint. In general, there is little value in attempting to identify the source of low back pain unless such identification has therapeutic implications. If identifying a specific anatomic pain generator can lead to specific treatment, for example, surgery, then such testing is potentially valuable. If the patient is going to continue with conservative treatment, such as physical therapy or oral medications, then the information obtained from invasive testing is unlikely to be valuable.

Figure 16B1-28   Spine pain: loaded with sensors.

The degenerative disk cascade was developed by Kirkaldy-Willis.182-184 This model presumes the interplay of the three-joint complex: the intervertebral disk and the paired posterior facet joints.184 This tripod forms the basic building block of the spine: the motion segment. The lumbar spine is thus composed of stacked motion segments that are interconnected to form a physiologic posture (lordosis in the lumbar spine and kyphosis in the thoracic spine). The degenerative disk disease cascade is characterized by periods of mechanical dysfunction, instability, and restabilization.182 An athlete may be seen in any phase of the degenerative cascade. The question often arises about whether the injury occurred because of some predisposition to injury from the degenerative cascade process or whether the injury caused the morphologic architectural degenerative changes. This is usually a question that cannot be answered definitively (Fig. 16B1-28).

Anular Tear Within the degenerative cascade, radial tears of the disk occur. The anulus fibrosus can be a source of pain because of its sensory innervation in its outer layers.185-188 The intervertebral disk is a syndesmosis, which is essentially a fibrous union. Anatomic changes in the anulus consisting of fissuring or frank tears occur and may be a potential source of pain.186-188 Controversy exists about whether MRI can identify a disk with a radial tear producing an area of bright signal on T2-weighted imaging (a high-intensity zone) as being symptomatic or asymptomatic.189-191 The hallmark invasive diagnostic test to determine whether the disk is the source of a patient’s back pain is lumbar diskography. Diskography involves injection of a dye into the disk and recording the patient’s pain response as either painless, concordant (reproducing the patient’s typical pain), or discordant (producing atypical pain). A less important feature of diskography is the radiographic image that it produces, either normal or abnormal. ­Diskography

Spinal Injuries

makes theoretical sense but is often challenged on an ­evidence-based basis.192-203 Some authors think that pressure-controlled manometric diskography can distinguish between asymptomatic and symptomatic anular tears, although this is controversial.204

Facet Joint The paired posterior facet joints are capable of producing pain based on their neural innervation.205-207 The facet joint is a diarthrodial joint with synovium and cartilaginous surfaces. The diagnosis of the facet joint as a source of pain is based on injection of the joints under x-ray control and producing pain relief.206,207 However, there are many challenges to the accuracy of facet injection as a diagnostic tool, and its utility is not universally accepted.206

Degenerative Disk Disease in the Athlete There is controversy surrounding the question of whether athletes have a higher incidence of low back pain and more degenerative lumbar changes than the general population. One study of Olympic athletes suggested a higher prevalence of and more lumbar disk degeneration than the normal population when using historical controls.208 Back pain and plain x-ray changes in the thoracolumbar spine were reviewed in elite wrestlers, gymnasts, and soccer and tennis players 14 to 25 years of age.209 Symptomatic back pain was reported in 50% to 85% of athletes, and x-ray abnormalities were found in 36% to 55% of athletes.209 The authors concluded that such high-demand athletes were subjected to increased symptoms.209 In a landmark study comparing elite athletes to control subjects, back pain was found to be less common in athletes than in controls, and there were no significant differences in hospitalizations or pensions.210 Not surprisingly, weightlifters and soccer players had more degenerative changes than runners and shooters.210 In a follow-up study, athletes did not report a higher frequency of back pain than nonathletes despite having significantly more radiologic abnormalities.211 Some anatomic changes, such as progressive loss of disk height or new-onset disk space narrowing, correlated with back pain.211 Another review of the epidemiology of low back pain in athletes suggested that, although common, episodes of low back pain were often short-lived.212 Because different sports produce different stresses on the spine, back strengthening programs might theoretically reduce the incidence of low back pain in athletes.212 The effect of mechanical load on low back pain in athletes is unclear. In a study of intercollegiate rowers, 32% experienced low back pain during college, and those who were symptomatic had a greater incidence of low back pain later in life compared with those who remained asymptomatic.213 The lifetime prevalence of back pain in the rowers did not differ from that of the general population, although the asymptomatic intercollegiate rowers had a lower incidence of low back pain than the general population as the study progressed. In another study, elite athletes competing in cross-country skiing, rowing, and orienteering were matched to nonathletic controls.214 The elite athletes had more low back pain during periods of training

741

and ­competition.214 The implication of these studies is that elite athletes in these endurance sports may have increased mechanical demand on the lumbar spine and more symptoms. For the nonelite athlete or casual exerciser, it is desirable to find a mechanical corridor that allows endurance and load training but keeps symptoms to a minimum.215 One recent study showed that two primary risk factors for long-term spinal problems were sports-related injuries and overuse in triathletes.216 An unanswered question is whether or not future scientific evidence will result in a change in training methods to reduce the mechanical load on the thoracolumbar spine, thereby reducing the incidence of low back pain.

Surgery for Degenerative Disk Disease Lumbar surgery in the athlete typically involves surgery for disk herniation, spinal stenosis, spondylolisthesis, or instability. The goals of surgery for instability in the athlete involve either motion segment elimination (fusion) or motion segment preservation (artificial disk replacement or dynamic stabilization techniques). A recent review comparing fusion to nonoperative care for chronic low back pain concluded that lumbar fusion may be more efficacious than unstructured, heterogeneous nonoperative care but that fusion may not be better than a structured rehabilitation program that includes cognitive-behavior therapy (Figs. 16B1-29 and 16B1-30).217 A 1-year follow-up study comparing fusion surgery with a conservative physical therapy ������������ regimen for ���� degenerative disk disease showed no advantage for surgery.218,219 A randomized trial comparing fusion to conservative treatment for chronic low back pain showed superior outcome in the fusion group.220-224 No particular fusion technique was found to be superior to the other techniques.222,224 Dynamic neutralization stabilization has been advocated as an alternative to fusion, but the clinical success of such techniques and their applicability to the athlete remain to be proved.225-228 Motion preservation with artificial disk replacement is a newer technology whose long-term outcome and complications are unknown.229-240 One study demonstrated the superiority of a particular type of lumbar artificial disk replacement compared with circumferential fusion for one-level degenerative disk disease.240 Although motion preservation is an appealing concept for the athlete, the issues of durability and long-term clinical outcome remain unanswered. Furthermore, the theoretical advantage of reducing adjacent segment degeneration bordering interbody surgery remains unproved.

LUMBAR DISK HERNIATION Lumbar disk herniation is produced by protrusion of the nucleus pulposus through the outer covering of the disk, known as the anulus fibrosus. This protrusion of nuclear material may take multiple anatomic forms, commonly described as a protrusion, an extrusion, or a sequestration.241 This deformation may be sufficient in size and location to produce mechanical compression of the surrounding neurologic structures. An extruded disk disrupts the anulus fibrosus but maintains some continuity with the

742

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B1-29   Middle-aged female runner treated with nonsteroidal antiinflammatory medications for low back pain. She currently runs 18 miles per week.

parent disk. A sequestered fragment, commonly known as a free fragment, is displaced from the disk space and ­isolated from it. From a practical point of view, the net effect is mechanical compression of the spinal nerve root or roots by the disk material. The clinical onset of a symptomatic lumbar disk herniation may vary widely. Crescendo back or buttock pain may be the precursor to radicular leg pain. This may occur suddenly or gradually, may occur without preceding back or buttock symptoms, and may present with sudden dramatic

Figure 16B1-30   Lumbar degenerative disk disease in a female professional basketball player who returned to professional basketball after conservative treatment.

radicular pain.242 It is common for a patient to be able to describe the day and time of the acute onset of radicular pain but not necessarily be able to relate it to a specific activity. A lumbar disk herniation is not always associated with a specific event, even in the sports population. The level and the anatomic location of the herniation (central, posterolateral, foraminal, or extraforaminal) determine which spinal nerve may be affected.243 In addition, a herniated disk may produce a wide variety of clinical signs and symptoms such as numbness, weakness, and pain.

Spinal Injuries

Lumbar disk herniations most commonly produce calf pain because the most common location of a herniation is at the L4-5 and L5-S1 levels, producing mechanical compression of either the L5 or S1 nerve roots, respectively. Proximal anterior thigh pain may result from involvement of the L2, L3, or L4 spinal nerve roots. There are three clinical patterns related to lumbar disk herniation that can influence treatment implications: cauda equina syndrome, progressive neurologic deficit, and radicular pain. Cauda equina syndrome can result from a large lumbar disk herniation that affects the sacral nerve roots (S2, S3, and S4). The cauda equina syndrome is characterized by a loss of control of bowel and bladder function. The treatment of this condition is urgent surgical decompression to maximize the chance of bowel and bladder recovery.241 It is ideal for surgery to take place in the first 24 to 72 hours of this clinical syndrome.244-248 However, some authors have not shown a temporal relationship between onset of cauda equina symptoms and functional bowel and bladder recovery.249 The second clinical condition, which may also accompany cauda equina syndrome, is progressive lower extremity neurologic deficit. A progressive neurologic deficit constitutes an absolute indication for urgent or emergent surgical intervention.241 Controversy exists regarding established, nonprogressive lower extremity neurologic deficits secondary to lumbar disk herniation. Some clinicians advocate conservative care based on the fact that most studies have not shown a difference in outcome and ultimate return of strength between surgical and nonsurgical treatment for nonprogressive lower extremity motor weakness.250-252 Recent studies have shown a correlation among functional motor recovery, the degree of preoperative lower extremity weakness, and the timing of surgery.253,254 Clearly, in the athletic population in which performance is a major concern, surgical intervention is a consideration in the presence of even nonprogressive lower extremity motor weakness. Radicular pain derived from a lumbar herniated disk may be sufficiently intense to warrant surgical intervention during the early phase of clinical presentation. This is, however, an exception to the general principle of trying conservative treatment for a period of time for at least 4 to 6 weeks because about 80% of patients with radicular pain secondary to lumbar disk herniation significantly improve without surgical intervention.255-257

Clinical History and Physical Examination The key clinical questions to be asked of the patient are about the onset, duration, and location of pain; the presence or absence of normal bowel and bladder function; and the presence and distribution of any numbness or weakness in the lower extremities. Inspection of the lumbar spine may demonstrate a truncal list either toward or away from the anatomic location of the herniation; the side of the list appears to be independent of the location of the disk herniation.258 Range of motion of the lumbar spine may be restricted because of back pain, and leg pain may be produced by either back extension or flexion, suggesting a dynamic component to neurologic compression. Neurologic ­examination

743

involves reflex, sensory, and motor examination of the lower extremities and a rectal examination, if there is any suspicion of cauda equina syndrome. A careful clinical examination may suggest which nerve root is affected by a lumbar disk herniation and allow correlation with imaging ­studies.258-264 Tension signs are an important indicator of nerve root inflammation. These signs include the straight leg raising test, contralateral straight leg raising test, femoral stretch test, and Lasègue’s sign. The straight leg raising test is considered positive if the patient’s typical leg pain is reproduced in the arc between 0 to 70 degrees of leg elevation. A contralateral straight leg raising test is positive when the asymptomatic leg is elevated and produces opposite (symptomatic) leg pain. Foot dorsiflexion of the symptomatic leg during performance of the straight leg raising test may produce radicular pain (Lasègue’s sign). Of note, the straight leg raising test reflects compression and inflammation primarily involving the L5 and S1 nerve roots. Mechanical pressure and inflammation involving the L2, L3, and L4 nerve roots is elicited during the femoral nerve stretch test, which is either performed with the patient prone while the involved hip is extended or is performed with the patient in the lateral decubitus position while the upper (symptomatic) hip is extended. The test is positive if the patient’s pain is reproduced in the femoral nerve distribution (anterior thigh).

Diagnostic Imaging The gold standard for imaging of a lumbar disk herniation is MRI. In most patients, MRI defines the relevant spinal anatomy, including a lumbar disk herniation and neural compression. If the MRI is of suboptimal quality, if the patient cannot tolerate the procedure, if there are contraindications to its use (e.g., the presence of a cardiac pacemaker), or if the suspected pathology is not clearly visualized, other diagnostic tests should be performed. In most instances, this would be a traditional lumbar myelogram with a postcontrast CT scan. In rare circumstances, CT diskography may be helpful in further identifying questionable pathologies such as a foraminal or extraforaminal lumbar disk herniation.

Conservative Treatment In the absence of a cauda equina syndrome or a progressive neurologic deficit, conservative (nonoperative) management of lumbar disk herniation is the initial standard of treatment because most lumbar disk herniations do not require surgery. Conservative measures include activity restriction, physical therapy, medication management, and spinal injections (epidural steroid injections or selective nerve root blocks). In a retrospective study of lumbar disk herniations, Hakelius demonstrated that 38% of patients with a lumbar disk herniation were improved by 1 month, 52% by 2 months, and 73% by 3 months.257 Two subsequent studies suggested that the optimal timing of surgical intervention was within 3 months.255-256 Clearly there is a point at which the optimal timing of surgery needs to be balanced with giving the patient an adequate trial of conservative management.265-273

744

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 16B1-1 Anatomic Zones for Stenosis Central canal stenosis Lateral recess stenosis Foraminal stenosis Extraforaminal stenosis

Figure 16B1-31   Lumbar disk herniation in a 38-year-old professional basketball player with preoperative weakness in the left anterior tibial muscle and left leg pain treated with lumbar microdiskectomy and eventual return to professional basketball (same season).

Surgery The goal of surgery is the removal of mechanical compression from the involved lumbar spinal nerves. The goal of surgery is resolution or lessening of leg pain, not back pain. Surgery is less predictable in resolving motor or sensory deficits than in improving radicular leg pain.255,259,260,261,274,275 The gold standard surgical procedure is subtotal lumbar diskectomy.276-280 The anatomic location of a disk herniation and other associated pathology may mandate additional bone, ligament, or facet joint resection to adequately decompress the nerve. There is no proven advantage of microsurgical decompression to standard diskectomy.281,282 A recent Cochrane update suggested that diskectomy provides quicker relief from acute sciatica from lumbar disk herniation than does nonoperative treatment. Furthermore, microdiskectomy gave comparable results to those of traditional diskectomy, with little evidence regarding outcome from other newer and less invasive techniques.282

Return to Sports Return to sports participation after symptomatic lumbar disk herniation, whether treated operatively or nonoperatively, depends on the amount of back or leg pain, neurologic status of the patient, and demands of the particular sport. A study of professional and Olympic athletes operated on for lumbar herniated disk with a microscopic diskectomy technique showed an average return to sports at 5.2 months, with 88% achieving successful return to play at their previous level (Fig. 16B1-31).283 In another study of elite athletes and return to play, 90% returned to prior intercollegiate sports following a single-level microdiskectomy, but none of the three athletes undergoing two-level diskectomies, nor the single athlete undergoing percutaneous diskectomy, returned to his or her sport.284

LUMBAR SPINAL STENOSIS Lumbar spinal stenosis, or neurogenic claudication, is broadly defined as narrowing of the central or foraminal canal through which the neurologic elements pass. The

Spinal Alignment Normal Spondylolisthesis Scoliosis Kyphosis Multidirectional malalignment

spinal nerve roots emerge from the spinal cord as the cauda equina and then course though the central spinal canal, exiting through the neural foramen bilaterally at each level of the lumbar spine. At each level of the spine, the nerve root passes below the pedicle of that same level; for example, the L5 nerve root passes below the L5 pedicle. The L1 nerve roots exit most proximally and the sacral nerve roots most distally. The spinal nerve roots may be crowded together centrally within the dural sac from extrinsic compression as the spinal canal concentrically narrows, or a single nerve root may be compressed individually as it exits the spinal canal. Central spinal stenosis is a global compression of the dural sac. One study comparing stenotic canals with normal canals found that the mean transverse area of the dural sac in the stenotic canals was 89.6 mm2 ± 35.1 mm2, compared with the normal canals, where the mean transverse area was 178 mm2 ± 50 mm2.285 This study concluded that constriction of the transverse area of the cauda equina to less than 75 mm2 will cause increased pressure within the nerve roots.285 A good correlation was found between reduced cross-sectional area and narrowed anteroposterior diameter of the dural sac.285 Structures contributing to neural compression included hypertrophic ligamentum flavum, disk protrusion, hypertrophic facet joints and laminae, and olisthesis. In acquired stenosis, the minimum cross-sectional area often occurs at the interfacet level.285 These age-related degenerative changes contribute to typical central lumbar spinal stenosis. At some point, the neural elements may become crowded together by the progressive degenerative process and may reach a critical level such that symptoms develop (Box 16B1-1).285 Lateral lumbar spinal canal stenosis may occur from neural compression at various anatomic zones.286,287 The lateral recess of the spinal canal has been divided into three zones based on the trajectory of the spinal nerve root: the entrance, the midzone, and the exit zone.286,287 Entrance zone lateral recess stenosis is located medial to the superior articular process of the facet joint where the nerve could be compressed between the anterior disk and the posterior facet joint.286,287 Midzone lateral recess stenosis occurs at the pars interarticularis and inferior to the pedicle. Exit zone lateral recess stenosis involves the intervertebral foramen. Although various studies have reported on the minimal cross-sectional area of the cauda equina required to ­produce

Spinal Injuries

symptoms of spinal stenosis, they are of little practical and clinical value and are not commonly used.288 Other studies have reported that no statistically significant correlation was found between the severity of symptoms of spinal stenosis and dural cross-sectional area.289 In general, imaging cannot differentiate symptomatic from asymptomatic individuals.290 It is therefore clear that imaging alone cannot predict symptomatic lumbar spinal stenosis. The underlying premorbid structure of the spine clearly plays a role in the potential for future development of symptoms. The presence of asymptomatic underlying congenital stenosis or structural abnormalities such as spondylolisthesis, scoliosis, kyphosis, or multidirectional malalignment can predispose to symptoms if normal agerelated changes, such as facet arthrosis, minor disk bulging, or herniation, occur. What role does spinal stenosis play in the athletic population? With people living longer and wanting to remain athletically active, clinical spinal stenosis will likely be encountered by many sports medicine physicians. In a review of a 5-year hospital admission cohort with a diagnosis of lumbar spinal stenosis, it was found that 9.8% were younger than 51 years old.291 The clinical onset of spinal stenosis is usually gradual but relentless as the critical space available for the neural elements is reduced. Sudden and dramatic nerve root symptoms may result from a long-standing severely obstructed spinal canal or from a sudden worsening of an underlying stenotic canal from a concomitant disk herniation. Most patients, however, describe a gradual and progressive increase in clinical symptoms, which can lead to a significant reduction in their activities in order to control their symptoms. A common complaint of golfers with symptomatic spinal stenosis, for example, is leg symptoms with walking, even if they use a riding cart. One study compared 100 patients with lumbar herniated disks, 100 patients with lateral recess stenosis, and 100 patients with central canal stenosis to determine symptoms and physical examination features of these disorders.292 The duration of symptoms before surgery and analgesic use was found to be significantly shorter in patients with disk herniations than in patients with stenosis. Tension signs (positive straight leg raising or femoral nerve stretch test) were more common with disk herniation than lateral stenosis and were uncommon with central stenosis.292 Because central spinal canal stenosis may occur at any intervertebral level, multiple nerve roots may be affected, and diffuse leg symptoms may be present unilaterally or bilaterally. Central spinal canal stenosis may produce only severe back pain and bilateral buttock pain without radicular leg symptoms, whereas foraminal stenosis typically produces radicular leg symptoms. Although spinal stenosis may produce the cauda equina syndrome, or progressive measurable neurologic deficits, it most commonly presents with few objective neurologic findings. Neurogenic claudication is characterized by buttocks pain or by leg pain, weakness, or numbness originating from spinal nerve compression elicited during standing or walking.293-295 The distribution of symptoms in the lower extremities may conform to sciatic or femoral nerve patterns or may simultaneously involve spinal nerve roots with both a sciatic and femoral nerve distribution. The

745

  TABLE 16B1-3  Vascular versus Neurogenic Claudication Evaluation

Vascular

Neurogenic

Walking distance Palliative factors Provocative factors Walking uphill Bicycle test Pulses Skin Weakness Back pain Back motion Pain character

Fixed Standing Walking Painful Positive (painful) Absent Loss of hair, shiny Rarely Occasionally Normal Cramping, distal to proximal Uncommon

Variable Sitting, bending Walking, standing Painless Negative (painless) Present Normal Occasionally Commonly Limited Numbness, aching, proximal to distal Occasionally

Atrophy

common presenting history is one of buttocks pain or leg pain, numbness, or weakness exacerbated by walking and standing and relieved by sitting or bending forward.293-295 The reason for this is that the spinal canal is narrower when standing and walking than when sitting. Patients with stenosis therefore characteristically assume a flexed or sitting posture to produce canal enlargement and relieve their leg or buttocks symptoms. Neurogenic claudication must be distinguished from vascular claudication, which has a different etiology and slightly different clinical features (Table 16B1-3).296 One distinguishing feature of vascular claudication is the production of leg symptoms while in a position of lumbar spine flexion, as in riding a bicycle.297 This position of flexion while riding a bicycle typically does not produce symptoms in a patient with spinal stenosis because sitting increases the size of the spinal canal and reduces spinal nerve compression.297 Some authors have used the combination of a treadmill test and a bicycle test to help distinguish vascular from neurogenic claudication.298 From a diagnostic perspective, normal arterial Doppler studies essentially rule out vascular claudication. There are few studies on the natural history of lumbar spinal stenosis. One study of 32 patients observed over a 4-year period found that 70% of the patients remained unchanged, 15% improved, and only 15% worsened.299 This study supports the concept that the surgical decisions are related to quality-of-life factors and not simply to the anatomic configuration of the spinal canal.299 In general, surgical treatment of spinal stenosis involves decompression of the stenotic segments (Fig. 16B1-32). Whether only symptomatic areas should be decompressed or all stenotic segments, both asymptomatic and symptomatic, should be decompressed is an area of controversy and beyond the scope of this discussion. Similarly, the decision of whether to perform a concomitant fusion with the surgical decompression is often controversial and not pertinent to this discussion. Early surgical intervention is mandated for stenotic patients with cauda equina syndrome or with progressive neurologic deficit. For most patients with spinal stenosis, however, the decision about surgery rests on the degree to which symptoms interfere with the patient’s quality of life.

746

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B1-32   Middle-aged recreational walker with an iatrogenic spinal stenosis and associated spondylolisthesis treated with anterior-posterior fusion and decompression. Postoperatively, the patient walks 40 minutes   5 times per week.

Physical Examination Inspection of the lumbar spine may be normal or may demonstrate postural abnormalities such as loss of lumbar lordosis, forward posturing, scoliosis, or kyphosis. Range of motion of the lumbar spine may be restricted, and back or leg symptoms may be exacerbated by back extension. Tension signs, such as the straight leg raising test and the femoral nerve stretch test, are often negative. It is important to examine both hip joints for painful or restricted range of motion because of the potential of having concomitant hip disease with spinal stenosis. Neurologic examination of the lower extremities is usually normal or nonfocal. A rectal examination should be performed if there is any suspicion of cauda equina syndrome. Because lumbar spinal stenosis often coexists with cervical stenosis (tandem stenosis), an upper extremity neurologic examination should also be performed to look for evidence of long tract signs that might indicate the presence of cervical myelopathy.300,301

of activity, physical therapy, medication management, and spinal injections (epidural steroid injections or selective nerve root blocks).302,303

Surgery

The lumbar MRI is the standard initial diagnostic imaging test for suspected lumbar spinal stenosis. However, lumbar myelography with a postcontrast CT are also valuable for surgical planning and for identification of pathologic levels in patients with spinal stenosis. CT is commonly used for patients who cannot tolerate MRI, in patients with contraindications to MRI (e.g., presence of a cardiac pacemaker), in patients with spinal deformity, or in patients who have had prior spinal instrumentation in whom metal artifact might obscure visualization with MRI.

The goal of surgery is spinal nerve decompression and restoration of functional activity by reduction or resolution of the leg symptoms provoked by walking or standing. The surgical strategy for lumbar spinal stenosis involves neural decompression and possible correction of any associated deformity or instability by fusion (Fig. 16B1-33). Current surgical techniques include traditional decompression or less invasive surgery. The role of such minimally invasive surgery remains to be proved.304-306 Both traditional and minimally invasive decompression should be able to address the full spectrum of compressive pathology and preserve stability.307-328 Interspinous distraction devices have recently been approved for surgical treatment of spinal stenosis.329-332 The theory behind such devices is that they produce a focal kyphosis at the stenotic segment, thereby opening the spinal canal and producing relief of symptoms arising from that level. Although conceptually attractive, their intermediate and long-term outcome is unknown. It is unlikely, however, that they would play a significant role in the older athletic population. Finally, it is important to ­realize that stenosis is generally a condition of older age and is therefore often associated with other age-related comorbidities. Surgical outcome is heavily influenced by multiple factors, including the number and types of such ­comorbidities.333-345

Conservative Treatment

Congenital Spinal Stenosis

Assuming that there is no evidence of a cauda equina syndrome or a progressive neurologic deficit, most patients with symptomatic spinal stenosis undergo a course of conservative management. Because lumbar stenosis is generally slowly progressive, this nonoperative phase may continue for some time. Conservative treatments include restriction

Congenital lumbar spinal stenosis is a type of stenosis in which there is a congenital narrowing of the spinal canal.346 Under such circumstances, minimal additional age-related structural changes can produce significant incremental compression of neurologic structures within the spinal canal. Although typical acquired lumbar spinal

Diagnostic Imaging

Spinal Injuries

747

Figure 16B1-33   Recreational golfer treated with laminectomy, facet fusion, and facet-pedicle vertebral fixation with 4.5 36-mm shaft screws for symptomatic lumbar spinal stenosis and associated degenerative lumbar spondylolisthesis.

stenosis is associated with an older population, congenital spinal stenosis is likely to be symptomatic much earlier in life and often manifests in the prime of athletic careers (Figs. 16B1-34 and 16B1-35). Congenitally short pedicles are a common recognizable radiographic feature of congenital spinal stenosis. One radiographic analysis of lumbar congenital spinal stenosis noted a significantly smaller cross-sectional area of the canal at all measured lumbar levels. Pedicles were markedly shorter in the congenital stenosis group at each lumbar level. In addition, anteroposterior canal diameter was significantly smaller than in control patients.347 The congenital spinal stenosis patient was found to be symptomatic at a younger age, with fewer degenerative changes and with multiple levels of involvement compared with degenerative spinal stenosis.347 Congential narrowing of the canal is a risk factor for additional pathology, such as a herniated disk, to become symptomatic.348 Congenital lumbar spinal stenosis is likely to be encountered in the younger athletic population.

Degenerative Lumbar Spondylolisthesis Degenerative spondylolisthesis, also known as spondylolisthesis with an intact neural arch, differs from isthmic spondylolisthesis in which there is a defect in the pars interarticularis (Table 16B1-4). Because of the intact neural arch, degenerative spondylolisthesis results in central canal narrowing as a result of the anterolisthesis of the affected vertebra. Isthmic spondylolisthesis does not result in central canal narrowing because the posterior elements remain in their normal position as the vertebral body slips forward. Vertebral translation in degenerative spondylolisthesis occurs as a result of both disk degeneration and facet arthrosis.349 Intersegmental instability results from inability of the lumbar facet joints to withstand shear forces, resulting in the slippage.350 The amount of forward slipping is typically 30% or less of the superior vertebra on the inferior vertebra.351 The most common location of degenerative spondylolisthesis is at L4-5, in contrast to isthmic spondylolisthesis, in which the L5-S1 level is most commonly involved.351

Figure 16B1-34   A 17-year-old football player with the sudden onset of progressive motor weakness in both lower extremities treated with lumbar decompression and diskectomy for symptomatic congenital lumbar spinal stenosis with a central disk herniation. He experienced a full neurologic recovery but decided not to return to football.

748

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B1-35   A 26-year-old professional basketball player with congenital spinal stenosis and symptomatic lumbar disk herniation treated with a lumbar microdiskectomy. He returned to professional basketball during the same season as his surgery.

Degenerative spondylolisthesis is a condition of older age, rarely occurring before 40 years of age, in contrast to isthmic spondylolisthesis, which occurs in youth, generally around the age of 10 years. Its clinical presentation is usually that of typical lumbar spinal stenosis. Anatomically, both the central and lateral recess zones become stenotic, and the traversing nerve root is usually more commonly affected than the exiting nerve root. For example, a degenerative spondylolisthesis at the L4-L5 level would more commonly affect the traversing L5 nerve root than the exiting L4 nerve root. Therefore, lateral calf pain (L5 dermatomal distribution) is more commonly the presenting symptom than anterior thigh pain (L4 dermatomal distribution). The diagnosis of degenerative lumbar spondylolisthesis should be made from a standing lateral x-ray because supine views may not reveal a slippage.352 Some authors have noted that spinal motion may affect the sagittal alignment of spondylolisthesis, with flexion typically exacerbating the slip and extension often reducing it.353 Dynamic axial loading during MRI has been described and advocated to demonstrate whether the slip is dynamic and changes with positioning.354,355 The initial treatment of degenerative spondylolisthesis is conservative and is the same as with typical lumbar spinal stenosis without spondylolisthesis. A recent study showed

  TABLE 16B1-4  Degenerative versus Isthmic Spondylolisthesis

Age of onset Level Pars interarticularis Central canal Foraminal canal Degree of slip

Degenerative ­Spondylolisthesis

Isthmic ­Spondylolisthesis

Older than 40 yr Any level; most commonly at L4-5 Intact Narrowed Narrowed Less than 30% of inferior vertebral body

About 10 yr Any level; most commonly L5-S1 Lysis Patent Narrowed No limit

surgery was superior to conservative care in both pain and function in symptomatic degenerative spondylolisthesis and spinal stenosis.356 The most commonly performed surgical procedure for lumbar degenerative spondylolisthesis is fusion with or without instrumentation, although there may be some select instances in which decompression without fusion is performed.349,357-361 A recent analysis of hospital data showed that decompression with fusion was less likely to require additional surgery compared with decompression alone.361 In addition, the presence of a solid fusion was associated with a better long-term outcome than if a pseudarthrosis was present.360

Isthmic Spondylolisthesis Isthmic spondylolisthesis and spondylolysis result from a stress fracture of the pars interarticularis. There are some anatomic variants, which include an acute fracture involving the pars and fractures of either the pedicle, facet, or lamina.362 Fracture of the pars without an associated slip (anterolisthesis) is termed spondylolysis. Spondylolisthesis refers to the presence of an associated anterior subluxation of the involved vertebra on the subjacent vertebra.363 It is classified (graded) by the degree of anterior displacement of the upper vertebra on the lower vertebra (Fig. 16B1-36). Other radiographic parameters have been described but are of limited clinical value.364-367 Stress fractures of the pars commonly occur during the childhood and adolescent growth period when the spine is exposed to repetitive stresses. Spondylolysis is thought to represent a fatigue fracture, usually as a result of repetitive mechanical stresses, or occasionally as a result of a single load of sufficient force to cause failure.362 Single-photon emission computed tomography (SPECT) or MRI may detect the stress reaction in bone consistent with a potentially evolving stress fracture (Fig. 16B1-37). The morphology of the pars fracture can be seen best by computed tomography (CT). Pars defects may be either unilateral or bilateral, and at a single vertebral level or at multiple levels, although most commonly at L5-S1. In

Spinal Injuries

Figure 16B1-36   Classification of spondylolisthesis: grade I, 0% to 25% slip; grade II, 25% to 50% slip; grade III, 50% to 75% slip; grade IV, 75% to 100% slip; grade V, spondyloptosis (complete displacement of upper vertebra in front of lower vertebra). (From Wiltse LL, Winter RB: Terminology and measurement of spondylolisthesis. Clin Orthop 117:23-29, 1976.)

some patients with a unilateral stress fracture, the contralateral pars or pedicle may show a stress reaction which can be detected on either SPECT or MRI. Most low-grade (grade I or II) isthmic spondylolistheses do not progress after the age of 18 years.368 Progressive

749

slipping may occur in adults as a result of associated degenerative changes in the disk at the slip level.369 The degree of forward slipping does not necessarily correlate with the amount of pain. For example, a highgrade slip may be asymptomatic, whereas a low-grade slip may be symptomatic. Isthmic spondylolisthesis may vary in presentation and clinical effect. Clinical symptoms include back pain, leg pain, leg numbness, or leg weakness. In the teenage years, back pain and hamstring spasm may dominate the clinical picture, with the back pain exacerbated with activity. In middle age, with the onset of degenerative disk space narrowing, the clinical symptoms may suggest stenosis, with leg pain predominating and worsened by walking and standing. This is due to compression of the exiting nerve root by a combination of foraminal narrowing produced by collapse of the disk and involvement of the nerve from the overlying fibrocartilaginous pars defect. Most adolescents with acute spondylolysis can be successfully managed by conservative measures, such as activity restriction, bracing, and lumbar trunk strengthening. Most are capable of eventually returning to their particular sport. Patients with chronic spondylotic defects and low-grade spondylolisthesis can usually be managed with a conservative care regimen also. Injection of the pars under fluoroscopic guidance can indicate whether the pars is the source of the patient’s back pain. Some authors have advocated pars defect repair in patients with minimal or no slip and a normal disk by MRI, but the mainstay of surgical treatment is fusion of the affected segment. As individuals age, degeneration within the affected disk can lead to low back pain and leg pain.370 Under these circumstances, surgical intervention may be warranted and may include a variety of decompressive and fusion options. The symptoms of acute spondylolysis may resolve, and patients may return to their sport when they are symptom free and can handle increasing spinal loads and the demands of their particular sport. Treatment usually entails a progressive program that begins with achieving

Figure 16B1-37   Single-photon emission computed tomography (SPECT) imaging with increased tracer uptake and matching magnetic resonance imaging with edema in pars area.

750

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B1-38   Middle-aged marathon runner who returned to sport after an anterior-posterior fusion for adult isthmic spondylolisthesis.

pain-free motion. This is followed by attaining the ability to perform their usual daily activities, followed by the ability to run and to strengthen supporting musculature, ultimately progressing to sport-specific drills. Return to play with isthmic spondylolisthesis is often correlated with the grade and progression of the slip. Patients with nonprogressive low-grade spondylolisthesis may return to play after progressive and pain-free reintroduction of loads of daily activities, trunk strengthening, and specific sports drills. Patients with high-grade slips are more problematic and require more individualized decision making. The outcome following conservative and surgical treatment of adult isthmic spondylolisthesis has been examined. Some studies have demonstrated improved function and more pain relief with surgery than with exercise.371 A randomized, controlled, 9-year follow-up study comparing a conservatively treated group of patients undergoing physical therapy with a surgical group undergoing posterolateral fusion, either with or without instrumentation, showed that patients who underwent fusion reported better outcome than conservatively treated patients.372 The authors thought that the long-term outcome of conservatively treated patients likely reflected the natural course of the condition and that no significant long-term improvement should be expected in adult patients with ­sympto­matic isthmic spondylolisthesis. It was thought that most patients would continue to have substantial pain, functional disability, and a reduced quality of life over many years.372 In another study, the preoperative and postoperative SF-36 (Short Form-36) scores of adult patients with isthmic spondylolisthesis were compared.373 Significant improvement in six of eight scores was noted, and 55% of scores were within the normal range.373 The best operation among different surgical strategies for adult isthmic spondylolisthesis remains unclear. In adult patients with unstable low-grade isthmic ­spondylolisthesis,

posterior instrumented fusion was compared with combined anterior-posterior fusion.374 The 2-year outcomes following combined anterior-posterior surgery were statistically superior to posterior fusion for unstable spondylolisthesis, although the differences between the two groups lessened after 6 months. The benefits of combined fusion must be balanced against the morbidity and costs associated with the additional surgery.374 Other authors also believe that combined anteriorposterior fusion procedures might be preferred for adult isthmic spondylolisthesis (Fig. 16B1-38), although the literature does not provide a definitive answer.375 A review of the literature on fusion for low-grade adult isthmic spondylolisthesis was unable to determine the best surgical technique for fusion (posterior lumbar fusion, posterior lumbar interbody fusion, anterior lumbar interbody fusion, use of instrumentation or not). The outcomes of fusion are generally good, but reports vary widely.376 Instrumentation is thought to play a beneficial role in achieving reduction and fusion for low-grade isthmic spondylolisthesis, although this perception has yet to be conclusively proved. For the sports medicine physician, an asymptomatic adolescent with isthmic spondylolisthesis may develop symptoms in middle age with age-related degeneration of the spondylolisthetic segment. Conservative treatment remains the first line of treatment, but surgery often achieves an excellent outcome with return to active sports participation.

RETURN-TO-PLAY DECISIONS One area of controversy for the sports medicine specialist and spine subspecialist is the decision about return to play after thoracolumbar spine injury. Often, philosophy more than science dictates the decision about return to play because there are no prospective, randomized, ­ double-blind studies on which to base many of the

Spinal Injuries

751

Figure 16B1-39   A 38-year-old   professional radio host and high-level recreational hockey player with successful index lumbar microdiskectomy and two repeat additional microdiskectomies for recurrent lumbar disk herniation. After surgery, the patient is pain free, and neurologically intact. Whether the patient returned to contact sports is unknown.

decisions. Therefore, the nature of the injury, the clinical signs and symptoms, and the type of treatment are key ingredients in the decision-making process.377 It might be assumed that decisions regarding return to play after cervical spine injuries would generate consensus of opinion. A 2001 survey of spine surgeons, however, demonstrated lack of consensus in the ability and level of return to play based on clinical scenarios of 10 cervical spine–injured patients.378 There was wide variability in the recommendations made for return to play. Spine subspecialists tended to recommend return to play at a higher impact level of play than sports medicine subspecialists. Clinicians who had been in practice longer selected a return to play at a lower impact level than clinicians who were more recently in practice. Categorization may help conceptualize sports in terms of potential risk categories, but there is no prospective study using these categories for determining return to play. Risk categorization is therefore a balance between the potential for recurrence of symptoms and the potential for catastrophic injury. Therefore, even in the analysis of risk categories of the cervical spine, there is more assumption than science, and no concrete data exist to help the clinician with many decisions. Most opinions regarding return to play are based on certain common features: being asymptomatic or having minimal symptoms, having normal or near-normal active range of motion, and achieving functional restoration of strength and endurance.377-381 If these are achieved, the athlete may progress to sport-specific drills and conditioning and ultimately to full competitive play. There will always, however, be assumptions about the potential for

recurrent injury based on the nature of the sport and the position the athlete plays. Repeat surgery at the same level of the spinal region may influence the decision to return an athlete to full performance.379 Some authors believe that repeat or additional surgery in the same spinal region is a contraindication for return to contact sports (Fig. 16B1-39).379 Reasons for this include potential instability produced by multiple procedures at the same level. Regional anatomic considerations play a role in making return-to-play decisions. Despite little surplus space surrounding the spinal canal in the thoracic spine, the relative immobility of the thoracic spine provides stability. The bony elements of the rib cage, together with the paraspinal musculature and sternum, increase thoracic stability by 20% to 40%.379,382 Three zones of the thoracic spine have been described: a midzone, proximal, and distal junctional zone.379 The bony structures change when transitioning from the cervical spine to the thoracic spine. In addition, the facet orientation changes from coronal in the upper thoracic spine to sagittal in the lower thoracic spine. Furthermore, the size of the thoracic vertebral bodies increases from proximal to distal in the thoracic spine.379 The cervicothoracic and thoracolumbar junctions are areas of potential instability and risk for injury because of the transition from a fixed segmental structure in the mid-thoracic spine to the more mobile cervical and lumbar spine. There is debate about whether spinal instrumentation should stop at the cervicothoracic junction or should continue to the more stable thoracic spine. One recommendation is that athletes be restricted from contact play if they undergo an instrumented fusion that crosses the cervicothoracic or

752

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

thoracolumbar junction, or if a fusion terminates at either junctional transition area.379 This restriction did not apply to athletes fully recovered from thoracic decompression and fusion that did not involve the transitional zones, assuming that they fulfilled all other criteria for return to play.379 There are two common lumbar conditions that frequently affect athletes: lumbar disk herniation and lumbar spondylolisthesis. Successful conservative or surgical treatment of lumbar disk herniation is not a contraindication to return to full contact sports as long as all other criteria for return to play are met. Return to play following surgery for the athlete with spondylolysis or spondylolisthesis is less clear. Return to play following posterolateral fusion, with or without instrumentation or direct repair of the pars interarticularis, is permitted provided that the athlete is pain free and neurologically intact and that the spine is structurally sound.379 Because the spinal cord typically terminates at the L1-2 level, only spinal nerves pass through the lumbar canal, and nerve root compression is much better tolerated than spinal cord compression. Because of the capacious nature of the lumbar canal relative to the size of the neural elements, the lumbar spine has a degree of tolerance to stenosis not present in any other region of the spine. In the lumbar spine, therefore, more anatomic narrowing and compression are needed to produce neurologic deficits than in other regions.379 Recommendations for return to contact sports after spinal surgery have been summarized by Burnett and Sonntag (Table 16B1-5).379 A survey of surgeons found significant variations in surgeon opinion regarding the timing and level of return to sports in children and adolescents after spine surgery.381 Patients treated surgically for scoliosis and low-grade spondylolisthesis (Meyerding grade I and II) were generally allowed to resume low-impact noncontact sports and return to gym class by about 6 months after surgery. For higher-grade spondylolisthesis (grades III and IV), many surgeons withheld noncontact sports for slightly longer, and most surgeons recommended restricting contact sports for a full year. Most surgeons recommended that patients undergoing fusion for scoliosis and high-grade spondylolisthesis never return to collision sports. Return to play after nonoperative and operative treatment for thoracic and lumbar spine injuries revealed a wide range of expert opinion.

CONCLUSION Reconsider then who we, as physicians, do see in the bend of time. Will she be 14 years old, suspended in the rings above the floor mat, each bodily gyration, each release and catch the animation of physical grace—acute ­ spondylolysis? Will he pose before the last hole, straddling his tee shot, knowing full well his best hole is the 19th, and strike with all the twisting force that his middle-aged trunk can consider, the longest distance of the weekend—lumbar anular tear? At 90, will she walk at dawn, chasing the last wisp of darkness, arms pumping, feet striking asphalt, her spine battling load and gravity—thoracic compression fracture? Will he crash into the last defender with will and force, tumbling the last few feet into the end zone—lumbar transverse process fracture? Will her head be down, her upper back cringing,

  Table 16B1-5  Recommendations for Return to Contact Sports after Spinal Surgery*

Op Location/Procedure Cervical    Occiput–C2 region    Subaxial region    Posterior      foraminotomy      single-level      multilevel      laminectomy (with or without fusion)/laminoplasty       single level       2 level       >2 level    Anterior     diskectomy fusion/arthroplasty       single level       2 level       >2 level      foraminotomy       single level       multilevel      corpectomy       single level       multilevel Thoracic    Cervicothoracic junction zone    Midthoracic     with deformity     without deformity  ��������������������������������������   ������������������������������������ Thoracolumbar junction zone Lumbar    diskectomy/laminectomy/laminoplasty     single level     multilevel    anterior or posterior fusion/arthoplasty     single level     multilevel

Return to Contact Sports No

Yes Yes Yes Yes No

Yes Yes Yes No No Yes Yes No Yes Yes Yes Yes

*To be considered eligible for a full return to activity, patients must be pain free, neurologically intact, and have completed an uneventful rehabilitative course. From Burnett MG, Sonntag VKH: Return to contact sports after spinal surgery. Neurosurg Focus 21 (4): E5, 2006.

her arms exhausted as she rolls her chair toward the finish line—thoracic muscle strain? Will he clutch the undersurface of his board, spinning away from the upper edge of the half pipe, with earth and ice waiting his return—lumbar burst fracture—on impact? Will she glide past them, as they reach for her, ascending in a single mellifluous leap towards the rim—asymptomatic thoracic scoliosis? Will he hear nature’s footsteps in the rising sea, leap freely from his prone position, and tightrope top and bottom turns before the impact zone—congenital lumbar stenosis with sudden back pain? Will she see yet another batter in the endless weekend of games, winding once again and curling spin and speed with bodily deception—lumbar muscle sprain? Will he soar like a swan in flight from water, his kite 100 feet above, his body engaged in twist and twirls before inevitable return to the ocean’s side—lumbar strain with radiographic evidence of Scheuermann’s disease? Will she plant her pole in the steep slope in the outback and wipe the sweat from her eyes of her weekend tour—structural but asymptomatic lumbar spinal stenosis? Will his racquet reach the apex of his toss, ­slicing

Spinal Injuries

the soft round ball in chosen direction—acute lumbar herniated disk? Will she look on the course of trees and feel the blur of nature with each heartbeat as she runs—sacral stress fracture? Will the world seem tame to his address, the bat that once was on his shoulder, the ball lost beyond the fence—adult isthmic spondylolisthesis? Will she pirouette and point and unfurl her arms to sounds she cannot hear— grade II asymptomatic isthmic spondylolisthesis? At dusk, will he gather himself for a final water start onto his board, his sail arced forward to catch the wind, slipping silently across the water at nature’s speed—lumbar degenerative disk disease? He or she will be one of us.

C l Thoracic

r i t i c a l

P

o i n t s

and lumbar fractures should be evaluated for spinal stability, neurologic status, and associated injury. Spinal stability and neurologic status govern nonoperative and operative treatment choices. The three-column theory of spinal stability by Denis is often used to assess stability. Minor fractures involve the spinous process, transverse process, pars interarticularis, and facet joints. Major fractures involve compression fractures, burst fractures, fracture-dislocations, and flexion-­distraction injuries. Immobilization before movement of the injured athlete should be performed if there is any question regarding spinal stability or neurologic involvement. Sacral stress fracture may be due to overuse with repetitive microtrauma, a fatigue fracture, inadequate bone mass, or an insufficiency fracture. Imaging is likely to make the diagnosis: pelvis MRI, SPECT scan, or bone scan. l Scoliosis is a coronal plane deformity, and kyphosis is a sagittal plane deformity. Most athletes with scoliosis have congenital or idiopathic types. Most athletes with kyphosis have congenital, postural, or Scheuermann’s types. Spinal deformity can be measured on full spine, standing anteroposterior, and lateral x-rays. Most athletes with spinal deformity are diagnosed because of other clinical events and not sports injuries. Most athletes with spinal deformity treated conservatively can participate in sports within the context of the conservative treatment. Most athletes with spinal deformity that is surgically treated will not be precluded from noncontact sports. l Thoracic disk herniations may present with unilateral chest wall or abdominal pain. Central thoracic disk herniations may compress the spinal cord, causing long tract symptoms and clinical findings: weakness and numbness in the legs and bowel and bladder dysfunction. Surgery is indicated for neurologic findings. Anterior transthoracic surgery provides access to the spinal canal for spinal cord decompression and thoracic diskectomy and interspace fusion. Alternative posterior approaches are the costotransversectomy and transpedicular approaches to canal decompression and diskectomy, but these have certain anatomic and technical limitations. l Thoracic spinal stenosis may occur as the result of degenerative changes narrowing the spinal canal. Congenital narrowing with superimposed degenerative changes may

753

lead to earlier presentation of symptoms. Thoracic-level spinal cord compression may be expressed clinically as a range from a subtle myelopathic gait disorder on one extreme to paralysis on the other. Surgical intervention should be directed by an approach that addresses the most compressed aspect of the spinal cord. l Imaging evidence of degenerative disk disease is common in athletes but is not necessarily related to symptoms. The natural history of degenerative disk disease and the symptom of low back pain have a relatively benign course. Invasive diagnostic testing in the lumbar spine would ideally be reserved for anticipated probable change in treatment based on the clinical condition of the patient and the likelihood of finding the pain generator. Low back pain in athletes associated with degenerative disk disease is usually treated conservatively. Eighty percent of lumbar ­herniated disks resolve within 3 months. Imaging must correlate with clinical findings (look at the patient and then the MRI scan). Loss of bowel or bladder function or progressive motor deficit is a surgical condition. Static neurological deficit and leg pain are not mandatory surgical conditions (requires judgment and evaluation of clinical effect and clinical course). The final common denominator of surgical value is spinal nerve decompression and not the title of the operation. l Spinal stenosis is a structural definition. The clinical expression of spinal stenosis involves back pain, leg pain, leg numbness, or leg weakness associated with walking or standing. Neurogenic claudication is the term used for dynamic leg symptoms associated with structural lumbar spinal stenosis. Stenotic patients usually experience worsening symptoms when their lumbar spine is extended and improvement when their lumbar spine is in flexion. l Isthmic spondylolisthesis is defined as a forward slippage of the cephalad vertebra on the caudad vertebra due to defects in the pars interarticularis. Isthmic spondylolisthesis can be graded based on the degree of slipping of the cephalad vertebra on the caudad vertebra. Most people with spondylolysis and low-grade spondylolisthesis respond to conservative treatment and successfully return to their particular sport. Occasional adolescent and young adult patients with isthmic spondylolisthesis will evolve into candidates for pars repair or segmental lumbar fusion. Older adults with isthmic spondylolisthesis may become symptomatic with superimposed degenerative changes in the affected motion segment. Some symptomatic adults with isthmic spondylolisthesis respond to surgical intervention involving decompression and fusion.

R E F E R E N C E S Please see www.expertconsult.com

754

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ecti o n

B

Thoracolumbar Injuries 2. Thoracolumbar Spine Injuries in the Child William C. Lauerman and John A. Zavala

Injuries to the thoracolumbar spine and associated back pain represent a relatively small proportion of sports injuries. About 10% of all sports injuries involve the spine, with certain sports (e.g., gymnastics, football, rowing) representing greater risks, particularly to the low back.1 In our experience, it is not uncommon to see skeletally immature athletes presenting with back pain; indeed, among the children and adolescents in our practice, thoracolumbar spine injuries likely represent 20% to 25% of sports injuries with pain persisting for greater than 1 month. When considering the pediatric athlete with back pain, it is helpful to be familiar with the prevalence of back pain in a general population of children and adolescents. Although the prevalence of low back pain in children younger than the age of 10 years has been reported to be as low as 1% to 3%, it increases in teenagers, varying among different researchers but reported to be as high as 20% to 25%.2 Relatively few children or adolescents present for treatment. When involved in organized sports requiring daily training and regular competition, however, the athlete with a complaint of persistent back pain will often visit the trainer or the physician. These complaints may stem from either isolated or repetitive trauma; in many cases, they may not be related to sports participation at all. It is essential, however, to address the athletic component of the pain complaint simultaneously, including timely return to participation, while bearing in mind the prevalence, the differential diagnosis, and the expected response to treatment of pediatric back pain in general. Appropriate management of the skeletally immature athlete with a back injury or persistent back pain often requires persistence, diplomacy, and luck. Increasingly, young children are participating in organized sports with schedules for training and competition and training techniques that may not match the physiology of the growing skeleton. In addition, attention to appropriate training and stretching regimens, including stretching of the neck and the low back, may be lacking; this is particularly true in elementary school children. Other factors to be considered when evaluating the underage athlete with persistent back pain are the desires and the motivation of the athlete. Although most athletes at all levels enjoy their chosen sport, we occasionally see grade school and, in particular, high school athletes with persistent complaints of back pain who subconsciously or consciously seem to be signaling a desire to gracefully avoid that sport and move on to a different activity. Sometimes

a limited desire of the child or adolescent to participate in sports does not match the enthusiasm of the parents to have the child play. Back pain can provide an acceptable exit strategy for the child to deal with what may represent unrealistic expectations on the part of the parent. Pain can be a mechanism for the child to cope with parental stresses. Factoring this into the management of the pediatric athlete with back pain is important but requires finesse, and suspected lack of enthusiasm should never represent a reason not to evaluate a complaint of pain.3 Finally, the reaction of the parents to what is all too often a protracted course of low back pain with attendant activity restrictions may add to the clinician’s challenge. Many parents find it unfathomable that a child can have back pain, let alone pain that eludes diagnosis and treatment. Reassuring the family of the relatively common nature of the problem being addressed is frequently quite helpful.

RELEVANT ANATOMY AND BIOMECHANICS The thoracolumbar spine comprises most of the vertebral column and contrasts the relatively rigid thoracic spine with the increased mobility of the lumbar spine. The thoracic spine receives its inherent stability from the thoracic rib cage. The erector spinae muscles function bilaterally to produce back extension and unilaterally to produce lateral bending. Deep to the erector spinae is the transversospinal musculature, which contributes to extension, lateral bending, and rotation of the vertebral column. Muscular strains involving the erector spinae and transversospinalis muscles are common in the athletic population. Important ligamentous structures include the anterior and posterior longitudinal ligaments, interspinous ligaments, and the supraspinous ligaments. The lumbosacral articulation has been demonstrated to be exposed to the highest forces in the thoracolumbar spine. The oblique inclination of the sacrum subjects the neural arch to both anterior compressive and shear forces. The erector spinae also place increased stress on the neural arch in the erect stance. There are a number of factors predisposing to back injuries and back pain in the growing child and adolescent, including the presence of a skeletally immature spine composed of the disk–vertebral body complex, the ligaments, and the musculotendinous unit. The adolescent growth

Spinal Injuries

755

spurt places the soft tissue structures of the spine at particular risk, because of a difference in maturation between the osseous components of the spine and the soft tissue structures. Full maturation of the bony pars interarticularis does not occur until early in the third decade. Tight lumbodorsal fascia and hamstring muscles may impose increased stress on the spine, thereby predisposing to back pain. The disk–vertebral body complex is also unique in the growing child. Before skeletal maturity, the cartilaginous and bony end plates are densely adherent to one another, but the ring apophysis, which extends peripherally along the margins of the vertebral body and is intimately adherent to the anulus, is at risk for injury.4 Such injuries to the ring apophysis can present clinically and radiographically as pediatric disk herniations.

CLASSIFICATION Thoracolumbar spine injuries in pediatric athletes can be broken down into two broad categories: those related to acute trauma and those involving chronic pain, possibly owing to chronic repetitive trauma (Table 16B2-1). Athletes with a history of acute trauma often describe a fairly significant and acute history of back pain. This may include either a significant traumatic event, or a relatively minor event that unmasks an underlying minor pain and brings these symptoms to clinical attention. In attempting to sort out the potential diagnoses, it is helpful to determine whether the athlete was truly asymptomatic before the significant symptoms began and whether a significant trauma occurred at the time of the onset of symptoms. Several manifestations of acute trauma may be seen. Fracture or dislocation of the anterior column of the thoracolumbar spine is unusual. When it occurs, the usual mechanism is an axial load, typically with the patient describing a fall on the buttocks either from a height or at a relatively rapid speed (Fig. 16B2-1). More commonly the injury is a relatively benign compression fracture, often occurring in the lower thoracic spine or at the thoracolumbar junction. Some compression fractures at the thoracolumbar junction are not easily seen on plain radiographs. When significant pain or tenderness is present at this location, plain radiographs of the thoracolumbar junction must

  Table 16B2-1  Injury Classification of Thoracolumbar Injuries Acute Fracture-dislocation

Chronic

Isthmic spondylolysis and ­spondylolisthesis (type IIA and IIB) Acute pars fracture (type IIC) Lumbar disc injury, diskogenic pain Disk herniation Apophyseal injury Disk degeneration Acute traumatic disk herniation Overuse syndrome Infection Spinal deformity Spinal tumor Infection

Figure 16B2-1   A 14-year-old boy was thrown from his bike during a competition, landed on his head and neck, and experienced the sudden onset of mid-thoracic pain. He was neurologically normal. Plain lateral radiography demonstrated mild wedging at T6 and T8 and a more significant wedge compression fracture of T7 (arrowheads). The absence of endplate irregularity or Schmorl’s node formation differentiates this from Scheuermann’s kyphosis, which would also involve multilevel wedging.

be carefully scrutinized. If radiographss are inconclusive or if a neurologic deficit is present, even if transiently, computed tomography (CT) or magnetic resonance imaging (MRI) should be performed. Acute traumatic fractures of the pars interarticularis have been reported but are exceedingly rare.5 The acuity of the pars fracture itself can be difficult to determine based merely on the presence of sudden onset of back pain. It can also be difficult to radiographically distinguish an acute fracture of the pars from a subacute or chronic fracture. Initial evaluation consists of plain lateral radiographs and, when in doubt, should include oblique radiographs and single-photon emission computed tomography (SPECT). CT is also useful in defining the defect at the pars interarticularis,6 although it can still be difficult to be certain that the radiographic abnormality is truly acute. MRI is the best study to diagnose an acute pars fracture by marrow changes at the pars. Most patients who have back pain and are found to have a pars fracture have had a preexisting spondylolysis rather than an acute fracture. Appropriate counseling regarding the goals of treatment is essential in this setting. Disk injuries may also occur as a result of acute trauma, although at least 50% of symptomatic disk herniations in both adults and adolescents present with the insidious onset of back pain or leg pain, or both.7-9 In contrast to adult patients with a symptomatic disk herniation, disk

756

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

herniation in the pediatric athlete may or may not involve radiculopathy. Back pain is the most common complaint in the child or adolescent with disk herniation, with less than half of all patients initially describing radicular symptoms. It is relatively rare to see a functionally significant, or worsening, neurologic deficit, but the presence of subtle neurologic abnormalities should be determined. The possibility of cauda equina syndrome, including urinary retention, should also be borne in mind. A positive tension sign (straight leg raising test or bowstring sign) is present in almost all pediatric patients with a clinically significant disk herniation. Chronic back pain in a pediatric athlete should, for the most part, be approached as one would approach back pain in the pediatric population at large. The differential diagnosis is varied, but in the athlete, the symptoms are much more likely to be due to repetitive trauma or an overuse syndrome. Although the diagnosis of an overuse syndrome may be considered initially, other diagnoses should be considered in those patients who fail to respond to treatment. Spondylolysis and associated spondylolisthesis are common causes of low back pain, particularly in children younger than 10 years and in adolescents.10 Occult spondylolysis is the most common provisional diagnosis in adolescents referred to us with recalcitrant back pain. It is noteworthy that this diagnosis is frequently found to be erroneous in the adolescent population. It is also worth noting that spondylolysis is present in 5% to 6% of the normal population at skeletal maturity and is frequently asymptomatic. Fredrickson and colleagues noted spondylolisthesis in 75% of a group of school-aged children with spondylolysis.11 Spondylolysis is more common in males and in certain ethnic groups, such as Aleutian Eskimos.12 Ninety percent or more of the defects are found at L5, and although they are virtually never present in newborns, 75% of pars defects occur—and can be radiographically documented—by age 6 years.11 The remaining 25%, occurring in later childhood and adolescence, likely account for the well-known association between spondylolisthesis and certain high-risk sports such as gymnastics, football (especially in interior linemen), and weightlifting.13,14 It has been demonstrated that football players with spondylolysis have a higher incidence of back pain than those with disk abnormalities and spinal instability.15 The acute onset of back pain, therefore, even in the presence of a defect in the pars interarticularis, cannot always be assumed to be caused by this defect (Fig. 16B2-2). Spondylolisthesis has been classified based upon the cause of the slippage (Box 16B2-1).16,17 The most common form of spondylolisthesis in all age groups is type II (isthmic), and this is the most prevalent by far in the pediatric population. Isthmic spondylolisthesis involves a defect in the pars interarticularis. The possible types of disorder of the pars include fatigue fracture of the pars (type IIA), elongation of the pars (type IIB), and acute fracture of the pars (type IIC). The concept of fatigue fracture of the pars interarticularis as the underlying disorder in spondylolysis was first popularized by Wiltse.17,18 It is believed that in certain individuals, the pars is susceptible to repetitive hyperextension stresses and that there is therefore a hereditary predisposition to this susceptibility. Elongation of the pars interarticularis (type IIB) may be seen without frank

Figure 16B2-2   A 16-year-old football player was struck in the back during a game and suffered acute onset of low back pain. Plain lateral radiography demonstrates isthmic spondylolisthesis at L5-S1. By the time of his orthopaedic evaluation, he was asymptomatic, and he has been symptom free for more than 2 years.

fracture and may permit the development of anterolisthesis. Acute traumatic fracture of the pars (type IV), although rare, may also be seen.5 Diskogenic back pain can be seen infrequently in the pediatric athlete. The patient typically presents with a chronic history of nonspecific back pain, which occasionally radiates into the buttocks and the thighs. The potential causes of such pain include disk herniation, apophyseal injury, and disk degeneration. The latter includes socalled juvenile diskogenic disease, which appears on MRI as decreased water content on T2-weighted imaging, Schmorl’s nodes, and end-plate irregularities. This can be

Box 16B2-1 Classification of Spondylolysis Wiltse-Newman I. Dysplastic II. Isthmic A. Lytic-fatigue fracture of pars B. Elongated pars C. Acute pars fracture III. Degenerative IV. Traumatic V. Pathologic From Wiltse LL, Newman PH, Macnab I: Classification of spondylolysis and spondylolisthesis. Clin Orthop 117:23-29, 1976.

Spinal Injuries

seen in either the thoracolumbar or lumbar spine with vertebral body flattening or wedging. In the lumbar spine, this condition has been referred to as lumbar Scheuermann’s disease.19 This condition is commonly seen in association with back pain in young adults and is occasionally seen in the pediatric athlete. Chronic overuse syndrome is common in the adolescent athlete with persistent low back pain. This should, however, be a diagnosis of exclusion, and a thorough radiographic work-up should be performed before making this diagnosis. Axial low back pain is the typical complaint, with minimal radiation into the buttocks, the thighs, or the legs. Relief with rest or activity restrictions is typical. Most pediatric athletes have back injuries or back pain that falls into one of the above categories. The remaining few can have back pain from less common conditions in the pediatric population, such as spinal deformities. Adolescent idiopathic scoliosis, although not typically painful, can be associated with mild to moderate pain (Fig. 16B2-3). This condition should only rarely require activity limitation, and the athlete who voluntarily restricts sports participation should be thoroughly evaluated for other causes of pain. Causes of painful scoliosis include tumors such as osteoid osteoma or osteoblastoma as well as intraspinal lesions. Hyperkyphosis is another cause of back pain in the adolescent, and pain is a relatively common presenting complaint of juvenile patients with Scheuermann’s kyphosis.

Figure 16B2-3   A 14-year-old swimmer had a 6-month history of mild low back pain that did not interfere with her participation in sports. Physical examination suggested scoliosis, which was confirmed on this plain anteroposterior radiograph. Her pain improved with a stretching and strengthening regimen, and the scoliosis, which progressed, was treated with nighttime bracing.

757

A number of tumors of the spine, although all quite rare, may present with back pain in the child or the adolescent. These include lesions of the posterior elements, such as osteoid osteoma or osteoblastoma; lesions of the anterior column, including eosinophilic granuloma, lymphoma, or primary malignancy such as Ewing’s sarcoma; and lesions of the spinal cord or cauda equina, such as ependymoma, astrocytoma, or neuroblastoma (Fig. 16B2-4). The presence of severe pain, pain at rest, and neurologic findings should trigger a more aggressive diagnostic work-up. Infectious conditions of the spine, such as disk space infection and vertebral body osteomyelitis, can occur in the pediatric population, particularly in younger children, in whom the mean age at presentation is 6 to 7 years. A history of constitutional symptoms such as fever, chills, or weight loss is common, and a history of a recent infection, such as an upper respiratory or skin infection, is frequent. The presence of pain at rest is also relatively common. Finally, metabolic abnormalities, generalized systemic malignancies such as leukemia or lymphoma, and visceral disorders should be considered.

EVALUATION Clinical Presentation and History Taking a thorough history is the key to the diagnosis and management of the pediatric athlete with back pain. The diagnostic and therapeutic goal, which should be clearly expressed to the patient and the parents, is return to full activity with minimal or no symptoms. Often, a concise diagnosis cannot be made, and overemphasis on obtaining a concrete diagnosis can lead to frustration, unnecessary imaging studies, and needless limitation of activity. History taking begins with inquiry about possible trauma. Although it is helpful to try to identify a traumatic episode, in our experience, this is uncommon. The patient should be asked about the onset of pain, whether it is acute or insidious, and whether there was any preexisting pain before the event or the day in which it was noticed. Frequently, a relatively long history of gradually worsening pain, culminating in a single day or event when the pain was noticed, can be elicited. When a traumatic episode can be identified, it is important to inquire about the presence of any transient neurologic signs or symptoms at the time of onset (e.g., inability to move an extremity, extremity or whole-body numbness, or loss of bowel or bladder function). Although these are extremely rare with sports injuries to the thoracolumbar spine, their occurrence, if present, should be elicited. Many pain complaints are chronic and of insidious onset. A helpful mnemonic to aid in the history is CLEAR: C represents the character of the pain complaint (e.g., burning, stabbing). L represents the location of the pain. Clinical terms such as back, buttocks, hips, or spine frequently mean different things to different people, and it is essential to have the patient accurately define where he or she experiences the pain. This can be difficult, particularly with younger children, but is essential in the evaluation of any persistent complaint. E represents exacerbation: What makes the pain worse? This is usually activity, but certain

758

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A Figure 16B2-4   A 14-year-old girl who presented after injuring her back playing soccer complained of diffuse back pain and tingling in her feet. A, On plain anteroposterior radiography, there is collapse of T10, irregularity of both pedicles, and a prominent soft tissue mass. B, Sagittal magnetic resonance images demonstrate a pathologic fracture, proved on biopsy to be caused by lymphoma.

activities and positions may be particularly painful and should be determined. A stands for amelioration: What can the patient do to lessen the pain? In general, rest reduces or relieves most musculoskeletal back pain. The child or the adolescent who clearly describes pain that occurs at night and awakens him or her needs to be evaluated much more aggressively than the typical teenager with activityinduced back pain. Finally, R stands for radiation: Where does the pain go? True radicular pain (typically pain that radiates below the knee) is seen in only 1% to 2% of children or adolescents with back complaints but, when present, should be identified.9 It is important to determine what the appropriate evaluation is, based on the severity of the patient’s complaints. Pain is truly subjective and therefore impossible to quantify, but certain factors provide insight into the magnitude of the problem. This is particularly true in children and adolescents who, when given a choice, will nearly always pursue their preferred activities and rarely have secondary gain issues. It is important to inquire about any interference with daily activities. Is the child missing school? Is he or she participating in sports? Are any medications

being used to help with the pain? Anything other than occasional interference with preferred activities represents a potentially more serious problem that mandates further evaluation. The presence or absence of neurologic symptoms should be determined. The presence of leg pain or paresthesias may signify nerve root compression or irritation. The presence of L’hermitte’s sign, a radiating electric sensation down the back and into the legs with forward flexion of the neck, is rare but may represent an underlying intraspinal abnormality masquerading as a sports injury. Bowel or bladder dysfunction is unusual, but it is important to inquire about its presence to avoid missing a cauda equina syndrome. The presence of new-onset enuresis should also be determined. Although the treatment of back pain in this population is fairly standardized, it is important to inquire about any previous treatment and the patient’s response to it. Since activity restriction is often the initial treatment, its prescription, and the response to it, should be identified. It is also important to determine whether the patient was compliant with any prescribed restrictions. The clinician

Spinal Injuries

should inquire as to whether or not physical therapy was done and what it entailed. The patient should be asked about whether physical modalities or an active back exercise program was prescribed. It is important to define which exercises were pursued and the completeness or incompleteness of compliance. Any history of medication usage and the response to it should also be sought. Finally, the previous use of bracing or casting should be explored. Identifying the type of brace, the regimen prescribed, patient compliance, and whether any sports participation was attempted while in the brace provides a foundation for further treatment options. With children, it is generally important to obtain a thorough general medical history. This includes getting a birth history, determining the presence of any developmental milestone delays, and taking both a general family history and a history of any spinal disorders such as scoliosis or spondylolisthesis. A review of systems should be obtained, and the presence of any constitutional symptoms such as fever, chills, or weight loss should be noted. Finally, the presence of any psychosocial factors such as depression should be identified.

759

Box 16B2-2 Typical History and Physical ­Examination Findings Spondylolysis and Spondylolisthesis •��� Back pain with back extension maneuvers •��� Usually acute exacerbation of chronic low grade pain •��� Limited forward flexion due to hamstring tightness •��� Palpable step-off of lumbar spinous processes (for highgrade slip) •��� Tenderness with pressure on and movement of affected spinous process Lumbar Disk Herniation

•��� Pain with sitting or any maneuver that increases intraabdominal pressure (e.g., cough or sneeze)

•��� Often of sudden onset •��� Significant back pain and stiffness with postural abnor-

mality (sciatic list) abnormalities often absent, but may have weakness or reflex change •��� High incidence of positive tension sign (straight leg raising sign, bowstring sign, or femoral nerve stretch test)

•��� Neurologic

Physical Examination and Testing Examination of the patient with a low back complaint begins with inspection, which cannot be performed adequately unless the patient is in an examining gown that opens in back, is disrobed down to underpants and bra, and has shoes and socks off. Inspection of the skin is performed, and the presence of any skin lesions, such as caféau lait spots or hairy patches, is noted. Asymmetry of the shoulders, the pelvis, the scapulae, and the skin creases is noted. Adam’s forward bend test—having the child bend forward to touch his or her toes and noting asymmetry of the ribs or the flank—is suggestive of scoliosis and should be performed. The child’s gait is then observed, including having the child walk on the heels and toes. A broad-based gait may suggest an underlying disorder such as myelopathy. This may be further evaluated by having the patient perform tandem gait (having the patient heel-toe walk as in a sobriety test). Palpation of the back is then performed. Palpation helps define exactly where the painful area is, or was, located. The exact site of tenderness is noted, whether in the midline, in the paraspinal region, in the buttocks, or over the trochanteric bursae. Palpation may also identify a stepoff in the lower lumbar spine, suggesting a high-grade spondylolisthesis. Range of motion of the spine should be tested. Limitation of forward flexion may represent disk disease, although hamstring tightness or spasm, commonly seen with spondylolysis and spondylolisthesis, may also limit forward flexion. Pain on extension of the lumbar spine, particularly with a painful catch, is commonly seen with spondylolysis or spondylolisthesis. A positive single-legged hyperextension test has been thought to signify the presence of an active spondylolysis, but a recent report suggests that this test is a poor predictor of active spondylolysis in adolescent athletes with low back pain.20 Although neurologic abnormalities are exceedingly rare in pediatric patients with back pain, a neurologic

e­ xamination should be performed. The presence of leg atrophy or asymmetry should be noted. Light-touch sensory testing, motor strength assessment, and testing of the reflexes should be performed. Deep tendon reflexes may be normally brisk in this age group. The presence of upper motor neuron findings such as spasticity, hyperreflexia, clonus, extensor plantar response (Babinski’s sign), or asymmetry of the superficial abdominal reflexes should also be noted. Tension signs, such as the straight leg raising sign, the bowstring sign, and the femoral nerve stretch test, should also be assessed because they are highly sensitive for disk herniation in this population.3,9 The history and physical examination features of, and distinction between, spondylolysis or spondylolisthesis and disk herniation are listed in Box 16B2-2. Examination of associated nonspinal areas is also important, including the abdominal contents, the hips, and when indicated, the sacroiliac area (by testing for Faber’s or Gaenslen’s sign). Because back pain may be due to nonspinal causes, the clinician should examine the abdomen and hips, when indicated.

Imaging Radiographic imaging studies are costly, time-consuming, and sometimes inaccurate in identifying the source of back pain in all age groups, including children and adolescents. Furthermore, it is relatively uncommon for a radiographic abnormality to alter treatment in the early phase of back pain. It is essential to bear in mind that a radiographic diagnosis does not need to be made in most cases at the time of initial assessment. This includes conditions such as spondylolisthesis and disk herniation. The decision for imaging in the pediatric athlete with back pain is determined, therefore, by several factors, including the severity and duration of symptoms and the response to previous treatment. Although uncommon, a history of

760

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 16B2-5   Lateral radiograph from an 11-year-old swimmer with a 2-month history of back pain. Six weeks earlier, he had stopped participating in swimming, and he had been housebound for 1 week before this image was taken. This radiograph demonstrates end-plate erosion and vertebral body involvement at T9-T10 (arrowhead) consistent with diskitis and vertebral osteomyelitis.

acute trauma, particularly associated with neurologic signs or symptoms, constitutes an indication for imaging. There is no universally accepted algorithm for imaging in the pediatric patient with back pain, whether an athlete or not. In the acute setting, we frequently find it helpful to institute treatment before obtaining radiographss and to assess and monitor the child’s response to the treatment. If the patient does not improve, plain radiographs are obtained, and treatment may be altered. If no further improvement is seen, more sophisticated imaging, such as MRI, may be obtained and treatment adjusted accordingly. Plain radiographs are the usual initial imaging test. They are indicated in a pediatric athlete with a history of acute trauma in whom fracture is suspected. We typically obtain plain radiographs in adolescents with significant symptoms lasting more than 3 to 4 weeks or in children or adolescents with severe pain or those who are unable to attend school (Fig. 16B2-5). The standard lumbar radiographic series includes standing anteroposterior (AP) and lateral views of the lumbar spine as well as a spot lateral (coned down) view of L5-S1. Plain lateral radiographs can identify 80% of pars defects (Fig. 16B2-6) and essentially all cases of spondylolisthesis.21 These views also identify virtually all significant fractures, such as compression fracture of the vertebral body. If a pars defect is suspected but not identified on the plain AP and lateral views, oblique views of the lumbar spine are ordered (Fig. 16B2-7), but these are not routine in our practice because of the significant increase in radiation they produce. Standing posteroanterior (PA) and lateral scoliosis radiographs are ordered if truncal asymmetry or hyperkyphosis is present on physical examination (see Fig. 16B2-3). It is important to recognize that hyperkyphosis, with or without mild secondary scoliosis, is frequently misinterpreted by primary care practitioners as scoliosis; therefore, all initial scoliosis evaluations should include a lateral radiograph.

Figure 16B2-6   A plain lateral radiograph of the lumbar spine from a 15-year-old soccer player with low back pain for 8 months demonstrates lysis of the pars interarticularis at L5 (arrowhead) without spondylolisthesis.

Figure 16B2-7   A defect in the pars interarticularis of L5 (arrowhead) is seen on this oblique radiograph. The normal bony continuity of the pars (the “neck of the Scotty dog”) of L3 and L4 can also be appreciated.

Spinal Injuries

A

761

B

Figure 16B2-8   Coronal (A) and sagittal (B) single-photon emission computed tomographic images from a 15-year-old boy with uptake in the region of the L5 pars interarticularis, consistent with spondylolysis.

A higher frequency of radiographic thoracolumbar abnormalities is present in the adolescent athlete compared with controls. Repetitive hyperflexion of the immature spine is thought to result in anterior disk herniation (marginal Schmorl’s nodes), whereas tension shear is thought to result in abnormalities of the vertebral ring apophysis.22 A higher frequency of anterior end-plate lesions have been demonstrated in competitive adolescent skiers, wrestlers, and gymnasts.23 Nuclear imaging with technetium-99m bone scanning can be helpful in identifying occult lesions of the spine not seen on plain radiographs. The sensitivity of technetium is enhanced with SPECT, which provides tomographic images of the radiotracer in the lumbar spine ­(Fig. 16B2-8). SPECT scanning is the preferred imaging modality for identifying occult pars interarticularis lesions.24 It is also sensitive, although nonspecific, for identifying other bony lesions such as tumors of the posterior elements, apophyseal fractures, and diskitis. Some of these lesions may be missed on MRI. Although technetium scanning is sensitive in identifying occult cases of spondylolysis, it is less useful in determining the acuity of an injury because a bone scan may be abnormal for up to 18 months after fracture. CT is also helpful in certain conditions, particularly in the evaluation of spondylolysis.6 Although not as sensitive as SPECT scanning, CT using 3-mm parallel cuts is a very sensitive modality for identifying pars defects. It is also very specific for differentiating pars defects from sclerosis of the pars or the pedicle, from tumors such as osteoid osteoma or osteoblastoma, or from apophyseal injuries. In addition, CT is useful for monitoring healing of a pars defect. CT

provides the best definition of the bony anatomy of the posterior elements of the lumbar spine. MRI is commonly employed, although it may not provide significant therapeutically useful information in the pediatric or adolescent athlete with acute pain. MRI in a pediatric athlete with acute pain is of limited usefulness and should be restricted until after a trial of conservative treatment has been tried. Despite its limitations, we do employ MRI initially for patients with clear-cut neurologic signs or symptoms and for patients who are clinically deteriorating. Similarly, a history of constitutional symptoms merits early evaluation with MRI. We also typically use MRI in the pediatric athlete with functionally disabling pain for longer than 3 months. Although 3 months may still be relatively early in the disease process, patient and parent expectations generally mandate its use. The obvious advantages of MRI are that it is noninvasive, does not deliver any ionizing radiation, and provides orthogonal imaging of the entire lumbar spine. It is the test of choice for herniated nucleus pulposus.25 The clinician should be aware that disk abnormalities are common in asymptomatic individuals, even adolescents. Besides identifying disk herniations, MRI also is ideal for defining other changes in the disk, such as disk degeneration or infection. MRI is thought by some clinicians to be less sensitive and less specific than either SPECT or CT in identifying pars defects. The usefulness of MRI in spondylolisthesis, however, is in its ability to identify the presence or absence of foraminal stenosis on the parasagittal images. Such foraminal stenosis is a common cause of radicular leg pain in the patient with spondylolisthesis, particularly high-grade slips.

762

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

In cases in which surgery is elected in the pediatric patient with spondylolysis or spondylolisthesis, we routinely perform MRI to exclude other sources of pain. MRI is also helpful in identifying and defining rare, but ominous, causes of back pain in children and adolescents. These include intraspinal diseases such as tumors, intraspinal lipoma, and tethered cord, which usually present with neurologic findings. Tumors of the anterior column, although rare, and infections of the spine are also optimally imaged with MRI. On the other hand, MRI is inferior to CT for defining bony lesions of the posterior elements.

TREATMENT OPTIONS Initial Nonoperative Treatment Most pediatric athletes presenting with new-onset back pain respond quickly to a series of general measures (Box 16B2-3). Therefore, it is rarely essential to make a specific diagnosis early in the course of the patient’s complaints. Furthermore, it is rarely cost-effective to order sophisticated imaging of children or adults with recent-onset back pain except in unusual circumstances. Most pediatric athletes present with a more chronic pain of insidious onset. These patients should be evaluated for so-called red flags, which are potential indicators of a more serious underlying condition such has fracture, infection, or tumor. These red flags include night pain, fever, weight loss, neurologic symptoms or signs and constitutional symptoms. The presence of incapacitating pain causing the patient to stay home from school or to be bedridden should lead to a more aggressive evaluation. Similarly, a history of constitutional symptoms such as unexplained weight loss, fever, or chills or the presence of significant neurologic signs or symptoms suggests the need for prompt imaging.3 Unless one or more red flags are present, most patients are placed on activity restrictions and are started on generic treatments such as moist heat, stretching, and over-the-counter medications such as acetaminophen or ibuprofen. The goal of this approach is to rule out more serious underlying disease and to return the patient to full activity with minimal or no pain. This may require relatively frequent re-evaluation. We typically bring the pediatric athlete with back pain back for re-evaluation on a weekly basis for the first 2 or 3 weeks and then every 2 or 3 weeks thereafter until acceptable resolution of the pain has occurred. In most cases, improvement will occur. Depending on the duration of the patient’s symptoms, their severity, and the rapidity of improvement, the athlete is allowed to gradually return to his or her daily activities and is then started on a program of back stretching and strengthening exercises under the guidance of a physical therapist. Once the exercise program has been mastered without recurrence of pain, gradual resumption of athletic activities is allowed. Because pain is a strictly subjective complaint, it is difficult to define objective guidelines for when the patient can resume specific activities. We find that the most useful guideline is the presence of minimal or no pain, full range of motion, and a normal neurological examination.

Another aspect of the management of the pediatric athlete with ongoing back pain is counseling. The athlete and, in particular, the parents often do not understand that back pain, even in a child, is a relatively common occurrence. It is imperative that the physician reinforce the fact that this condition may occasionally involve a prolonged recovery. It also needs to be emphasized that, although it is exceedingly frustrating for the family, the exact cause of the pain may not be determined. The appropriate role of imaging studies is to guide treatment; they should therefore be obtained only when they can reasonably be expected to have a significant impact on therapeutic decision making. In general, imaging is rarely helpful, and therefore not indicated, in the first 2 to 4 weeks of symptoms. We typically wait 2 to 4 weeks before obtaining plain radiographs for those individuals whose back pain fails to significantly improve. If plain radiographic findings are negative and the pain persists, we consider an MRI after 6 to 12 weeks of symptoms. When there is particular concern about the presence of a spondylolysis, we often proceed to SPECT before getting an MRI. If the MRI is normal and symptoms persist for 6 months or longer, we often recommend technetium bone scanning.

Diagnosis and Nonoperative Treatment of Spondylolysis and Spondylolisthesis Spondylolysis with or without spondylolisthesis (slippage) is a chronic fatigue fracture of the pars interarticularis. It results from both hereditary and environmental factors (nature and nurture). The child or adolescent with spondylolysis or spondylolisthesis typically presents with a history of chronic low-grade back pain, sometimes with an acute exacerbation. The degree of slippage is based on Meyerding’s classification, whereby the superior aspect of the sacrum is divided into quarters and the slip is described as the relationship of the L5 vertebral body to the sacrum (S1).26 Grade I represents a slippage of 0% to 25% of L5 on S1; grade II is 25% to 50% slip; grade III is 50% to 75% slip, and grade IV is more than 75% anterolisthesis. Spondyloptosis refers to the situation in which the body of L5 sits anterior to the sacrum—a slip of more than 100%. Most children with spondylolysis have either no slippage or a low-grade (grade I or II) spondylolisthesis, and therefore many of the more dramatic associated physical findings seen with high-grade slips (e.g., flattening of the buttocks, a transverse abdominal crease, or gait alterations) are absent. In the thin patient, a palpable step-off of the lower lumbar spinous processes may be appreciated, and there is frequently painful limitation of extension associated with a painful catch on back extension. Hamstring spasm is frequently present, and forward flexion may be limited because of the tight hamstrings. Diagnostic studies start with plain radiography. A standing spot lateral view identifies 80% of pars defects but may be supplemented with oblique views if needed. SPECT is the most sensitive study for identifying occult pars defects that are not seen on plain radiographs. CT is helpful to more clearly define the pars fracture (Fig. 16B2-9). CT can help differentiate an acute fracture of the pars from a chronic fracture.

Spinal Injuries

763

Box 16B2-3 Treatment Options Acute Lumbar Strain •��� Limited period of activity restriction •��� Moist heat and tissue massage •��� Nonsteroidal anti-inflammatory drugs (NSAIDs) •��� Stretching and strengthening program •��� Gradual return to play Spondylolysis and Spondylolisthesis

•��� Activity restriction with or without bracing •��� NSAIDs •��� Physical therapy with core strengthening (stabilization program) and stretching stabilization if pain > 1 year or high-grade (grade III or IV) spondylolisthesis

•��� Surgical

Lumbar Disk Herniation

•��� Initial period of restricted activity •��� NSAIDs and moist heat •��� McKenzie’s back program (generally an extension program)

•��� Surgical treatment if unacceptable pain for 6-12 weeks (physical and radiographic findings must correlate) or progressive weakness or cauda equina syndrome

Overuse Syndrome •��� Plain radiography, magnetic resonance imaging, and bone scan correlate (must rule out other conditions) •��� Activity restriction, NSAIDs, moist heat, and physical therapy program •��� Alternative treatments may be considered (chiropractic, acupuncture) •��� Gradual resumption of activities •��� If recurrence: change to less competitive level of athletic participation or change sport

The primary goal of treatment is relief of back pain and the resumption of full activities, including sports. The goal is not to heal a pars fracture or to reduce a slippage because most individuals with spondylolysis or spondylolisthesis are asymptomatic, or minimally symptomatic, even with these abnormalities. The goal of treatment is to render the patient asymptomatic or minimally symptomatic. It is important to stress this from the outset to the athlete and the parents. Most patients with symptomatic spondylolysis or spondylolisthesis are started on an initial trial of nonsteroidal anti-inflammatory drugs (NSAIDs), moist heat, and activity limitation. Depending on the severity and the duration of the patient’s symptoms, athletic participation may be either restricted or eliminated. As the symptoms resolve, athletic participation is gradually resumed, and the patient is weaned off medication and is usually started on a lumbar stabilization program. The athlete is encouraged to continue these exercises, even if he or she is asymptomatic, for at least one season and frequently longer. Unfortunately, some patients either fail to respond or suffer a recurrence of pain. In these circumstances, we typically repeat the course of treatment and try to ensure that compliance has been adequate. Another option in this ­

Figure 16B2-9   Axial computed tomographic image of a 17year-old football player demonstrating unilateral spondylolysis with sclerosis of the contralateral pars.

setting is the use of bracing; a number of patients respond to a Boston antilordotic brace worn on a full-time basis initially, with gradual weaning over a 6- to 12-week period.27 In general, the goal of bracing is to render the patient asymptomatic rather than to promote healing of a pars fracture. Fracture healing is very unlikely, and when it does occur, it is most likely to do so in a young child with a recent fracture and with no slippage. Physical therapy modalities such as ultrasound, electrical stimulation, and massage are only rarely used but may be helpful to provide temporary relief of symptoms.27 Electrical stimulation has also been tried for spondylolysis when an initial trial of conservative treatment has failed. A few small retrospective case series have demonstrated some success with electrical stimulation. However, these results may not be generalizable to the entire population, and additional clinical research is required.28 Acute fracture of the pars interarticularis (type IV) is unusual, but when it is identified early, fracture healing can sometimes be achieved. The most reliable method of immobilizing the lumbosacral junction is a pantaloon spica cast, although some authors report an acceptable healing rate with other types of bracing.27 Serial CT can be useful in following the progression of the defect to healing.

Operative Treatment of Spondylolisthesis Although most patients with spondylolysis or spondylolisthesis are either asymptomatic or respond to nonsurgical treatment, surgery is occasionally required. Intractable pain after 1 year of appropriate treatment is the most common indication for surgery. A therapeutic dilemma arises in the patient who achieves acceptable symptomatic relief by activity restriction but is unable to resume athletic activity without symptoms. Surgery may be considered under such circumstances, but return to high-level sports competition following surgery is unpredictable. Other less common indications for surgery include a significant or worsening neurologic deficit or the presence of cauda

764

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 16B2-10   A and B, Serial lateral radiographs over a 15-month period demonstrating dramatic progression of isthmic spondylolisthesis in an adolescent athlete. Progression of this type is somewhat uncommon and is usually associated with pain, as it was in this boy, whose symptoms quickly resolved after L5-S1 fusion and cast immobilization.

equina syndrome. These are unlikely except in a highgrade slip or spondyloptosis. In addition, a progressive slip (Fig. 16B2-10) or a slip of greater than 50% in a skeletally immature child generally mandates prophylactic stabilization, regardless of symptoms. The traditional gold standard surgical treatment is a posterior spinal fusion. For an L5 spondylolysis or lowgrade L5-S1 slip, fusion from L5 to the sacrum is typically performed. In situ noninstrumented posterolateral fusion with autogenous iliac crest bone graft and cast immobilization has a very high success rate with minimal morbidity.21 Laminectomy or nerve root decompression is rarely necessary in the pediatric population, even in the presence of leg pain. The role of instrumentation as an adjunct to posterolateral fusion for the child with spondylolisthesis is a controversial issue. In the smaller, skeletally immature patient (typically <10 years), the fusion rate with noninstrumented techniques is sufficiently high that the riskto-benefit ratio for transpedicular instrumentation in the developing spine would appear to be excessive. Fusion in skeletally mature teenagers, particularly in those with a high-grade (grade III or IV) slippage, is more commonly performed with segmental pedicle screw instrumentation. The role of reduction of spondylolisthesis is beyond the scope of this discussion but is sometimes attempted. It is typically reserved for patients with a slippage of greater than 50%, those with associated lumbosacral kyphosis, and those with a cosmetically objectionable appearance.29,30 Reduction of a high-grade slip, such as a spondyloptosis,

is associated with a significant risk for neurologic injury, particularly to the L5 nerve root. Direct repair of the pars defect may be performed but is generally reserved for patients with minimal or no slip, for patients without chronic pars changes, and for patients with normal disk by MRI at the level of the spondylolysis.31 Repair may be performed by a tension band wiring technique, by a direct repair across the fracture with a screw, or with compression using a pedicle screw with a hook and rod.

Nonoperative Treatment of Lumbar Disk Disease In North America, only 1% to 2% of clinically significant disk herniations occur in children younger than 18 years, and the incidence is equal in boys and girls. About 50% of patients report a history of trauma. Low back pain, stiffness, abnormal posture, and limping are the most common presenting complaints, and leg pain, paresthesias, and subjective weaknesses are present in about 25%. Many authors report a prolonged duration of symptoms before diagnosis, lasting from several months to a year.9,32 Three clinical features are typical of adolescents with a lumbar disk herniation. First is the frequent presence of significant back stiffness, spasm, and postural abnormality such as a sciatic list. Second is the frequent absence of any neurologic abnormality. Third is the presence of a positive straight leg raising sign.33 The latter is both highly sensitive

Spinal Injuries

Figure 16B2-11   Computed tomographic myelogram through the L4-L5 interspace in an 18-year-old wrestler with a 3-month history of pain in his back and left buttock.   A characteristic posterolateral herniation, in the presence of a somewhat narrow spinal canal, leads to marked thecal sac and nerve root compression.

and specific for disk herniation in the pediatric population and should be performed in all patients with a suspected disk herniation. Diagnostic imaging should be considered after 6 weeks of unsuccessful conservative treatment, in the presence of a significant or worsening motor deficit, or if there is any concern about bowel or bladder dysfunction from a cauda equina syndrome (Fig. 16B2-11). MRI is the diagnostic test of choice and identifies most disk herniations.25 Disk abnormalities, including herniation, can occur in asymptomatic individuals; therefore, the patient’s signs and symptoms must correlate with the radiographic findings before surgical intervention is considered. MRI is not as accurate as CT in the diagnosis of an apophyseal endplate fracture.34 This variant of disk injury is less common than true disk herniation but is seen occasionally in young patients. The signs and symptoms of apophyseal fracture are similar to those of a disk herniation; when this condition is suspected and the MRI is nondiagnostic, CT should be performed. Occasionally, a posteriorly displaced fleck of bone at the disk–vertebral body interface can be identified on the plain lateral radiograph and should alert the clinician to the possibility of an apophyseal end-plate fracture. This should be confirmed by CT.33 The initial treatment of the pediatric patient with a lumbar disk herniation is conservative. Our approach is similar to that used for adults. An initial period of restricted activity is initiated, and NSAIDs are often prescribed. Moist heat is also sometimes helpful for back pain in the early symptomatic period. Once the patient is able to walk without a limp and has a near-normal range of back motion, physical therapy is initiated. We generally prescribe a McKenzie back exercise program under the supervision of a physical therapist and allow the patient to resume normal activities

765

gradually, including sports, as the symptoms subside. Most children and adolescents respond to such treatment within several weeks, and a good prognosis and return to sports can usually be anticipated. In those patients who fail to respond to conservative therapy, several options are available. Continued or more stringent activity restrictions can be considered. A trial of epidural steroid injections may also be undertaken; this is often successful, at least in the short term, in relieving radicular pain. As a final option, surgery may be considered. In addition to frank disk herniation, symptomatic degenerative disease of one or more lumbar disks is occasionally seen in the adolescent. This condition has been referred to as juvenile disk disease or lumbar Scheuermann’s disease. Radiographic findings are similar to those seen in the thoracic spine in Scheuermann’s kyphosis. These include end-plate irregularity, Schmorl’s nodes, and flattening or wedging of one or more vertebral bodies.4 Actual kyphosis is rarely seen, but loss of normal lumbar lordosis is common. On MRI, disk desiccation at one or more levels is identified in addition to the previously mentioned findings. The prevalence of juvenile disk disease has not been adequately defined, but it is more common in boys than in girls and is an occasional, although not a common, cause of low back pain in adolescents. Its presence is a common finding in young adults with low back pain (Fig. 16B2-12). Children may present with a primary complaint of axial back pain with occasional radiation into the buttocks and not associated with disk herniation. Such pain is referred to as referred pain rather than radicular pain. Pain radiating distally into the posterior thigh or down the leg is uncommon unless a herniation is present. Physical findings are generally nonspecific but include paraspinal spasm and loss of motion. The neurologic examination is usually normal, and the straight leg raising sign is typically negative, in contrast to disk herniation, in which the straight leg raising sign or bowstring sign is almost always positive. Treatment of a patient with juvenile disk disease depends on the stage of presentation. Acutely, the standard regimen of activity restriction, NSAIDs, and moist heat is usually successful. An aggressive back rehabilitation program is usually initiated. Unfortunately, many patients with symptomatic juvenile disk disease develop low back pain in early adulthood, and we therefore encourage these individuals to continue with a program of aggressive back stretching and strengthening well beyond their recovery time frame. We also counsel and encourage long-term participation in sports activities that are likely to be less stressful on the lumbar spine than collision sports.

Operative Treatment of Lumbar Disk Injuries The usual indication for surgery in the adolescent with a lumbar disk herniation is persistent unacceptable pain despite appropriate nonoperative treatment for a minimum of 6 to 12 weeks.35 Although adult surgical candidates almost always have distal leg pain (radicular pain), adolescent patients often complain primarily of pain in the buttocks and the posterior thigh (referred pain). As with the adult with a symptomatic disk herniation, an ­adolescent

766

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 16B2-12   Plain lateral radiograph (A) and sagittal T2-weighted magnetic resonance image (B) of a 41-year-old man with chronic low back pain and lumbar Scheuermann’s disease.

should have a positive straight leg raising sign to be considered a good surgical candidate. The MRI is the preferred confirmatory diagnostic test to identify a disk herniation and must correlate with the patient’s neurologic examination and tension sign (straight leg raising test or bowstring test). Surgery typically involves a hemilaminotomy and partial diskectomy. Fusion is rarely, if ever, indicated for an adolescent with a herniation and no prior surgery. The outcome of surgical diskectomy in adolescents is generally not as good as in adults. Although greater than 90% of adults experience significant long-term relief of their sciatica following uncomplicated diskectomy, only 75% to 80% of adolescents have similar relief.7,36

Overuse Syndrome One of the more difficult problems encountered by physicians treating adolescents with low back pain, particularly adolescent athletes, is the patient with ongoing resistant low back pain without an apparent cause. Although overuse syndrome should be considered a diagnosis of exclusion, this condition represents a common cause of chronic low back pain in adolescents. It is not, however, a common cause of low back pain in the preadolescent child, and should be considered only after excluding other causes. In this age group, a definitive cause of low back pain can usually be found. If the physical examination is benign and imaging studies, including MRI and bone scanning, are normal, a chronic strain or overuse syndrome can be assumed. Typically, these patients present with a prolonged history of backache and exhibit a high level of frustration because of the ongoing pain and not having a clear diagnosis. Evaluation of these patients begins with a thorough history and physical examination. As noted previously, the diagnosis of overuse syndrome should be considered only

if plain radiography, MRI, and bone scanning are normal. In the adolescent with persistent symptoms, the radiographs are typically repeated every 6 to 12 months, and the MRI is repeated every year until the symptoms resolve. In most cases, symptoms resolve over a period of several weeks. However, some cases of recalcitrant low back pain may persist for a year or longer. Most patients with overuse syndrome respond well to a regimen of activity restriction, NSAIDs, moist heat, and a program of stretching and strengthening exercises followed by gradual resumption of activity. There is rarely a role for bracing in individuals with undiagnosed chronic low back pain. Trigger point injections with a local anesthetic and corticosteroid are occasionally useful, but epidural steroid injections are reserved for patients with radicular pain from a disk herniation. Finally, it is important to consider the athletic goals of the child or adolescent. Some adolescent athletes may be reasonably comfortable with most activities but may experience recurrence of symptoms when they return to competition. It may be beneficial for these individuals to consider another sport that may be less stressful on the lumbar spine.

WEIGHING THE EVIDENCE Much of what is written on thoracolumbar injuries in the adolescent athlete focuses on the treatment of spondylolysis and spondylolisthesis. Better outcome is generally achieved with a period of up to 3 months of activity restriction, with or without bracing, compared with patients who continue with athletic participation.13,37 Unilateral pars lesions have a significantly higher rate of bony healing than bilateral lesions. Good and excellent long-term outcome is achieved in about 90% of patients after conservative treatment of early spondylolysis.38,39 Long-term follow-up of

Spinal Injuries

patients with unilateral spondylolysis has demonstrated no progressive slippage. In patients with bilateral spondylolysis, there appears to be a 5% incidence of symptomatic slip progression. Patients with a higher degree of slippage have been shown to have more long-term disk degeneration at the listhetic level.40 There are few data on long-term outcome following surgical treatment of pediatric disk herniations. As noted previously, about 75% to 80% of such patients experience long-term resolution of their symptoms.8,36

A u t h o r s ’ P r e f e rr e d M e t h o d In assessing the pediatric athlete with back pain without a history of significant trauma, infection, tumor, or metabolic disease processes, it is important to review the type and effect of previous treatments. In general, activity restriction is the first phase of treatment if this has not been previously employed. Over-the-counter NSAIDs and acetaminophen, along with moist heat and stretching, are then added. The patient is then re-evaluated after about 2 or 3 weeks, and plain radiographs are obtained if there has been no improvement in symptoms. If spondylolysis is suspected and radiographs are negative, SPECT or possibly MRI is performed. If SPECT is positive, CT is performed to further define the bony anatomy, particularly if repair of the pars defect is being considered. In the patient in whom spondylolysis is identified, activity restriction is continued until symptoms subside, after which participation in athletics is gradually resumed. If a spondylolysis is not present, an MRI is typically obtained after 2 to 3 months of symptoms. In patients in whom a lumbar disk herniation is suspected, an MRI is obtained if symptoms persist beyond 6 weeks. If a disk herniation is demonstrated by MRI, and there is no progressive weakness, NSAIDS are prescribed and activity is restricted, particularly strenuous activities and prolonged sitting, both of which generate increased intradiskal pressure and can aggravate symptoms. As the acute symptoms subside, a McKenzie extension-type low back exercise program is initiated. Patients generally respond well to this regimen and return to sport within 8 to 12 weeks. As noted previously, surgical intervention is reserved for patients with unacceptable pain for 6 to 12 weeks. Patients suffering from chronic overuse syndrome present a particular challenge. Most children suffering from chronic overuse generally have improvement of symptoms with activity restriction, NSAIDs, and supervised therapy. If athletic participation is resumed too quickly, the pain may return.

CRITERIA FOR RETURN TO PLAY The question often arises about whether asymptomatic children and adolescents with spondylolysis or spondylolisthesis should restrict their activities (Box 16B2-4). We do not recommend activity restrictions for the asymptomatic pediatric athlete with slippage of up to 25% (­Meyerding

767

Box 16B2-4 Return-to-Play Criteria Spondylolysis •��� Nontender to palpation •��� Minimal pain with activity •��� Full painless range of motion Spondylolisthesis

•��� Grade I: no activity restriction •��� Grade II: no participation in high-risk sports •��� Grade III/IV: surgical stabilization Lumbar Disk Herniation

•��� Nontender with full range of motion • Neurologically normal with negative straight leg raise test grade I). This includes allowing participation in sports such as gymnastics and football. These patients should be followed with serial standing lateral radiographs of the lumbosacral spine, at 6-month intervals for the first 1 or 2 years after diagnosis and then annually until skeletal maturity. Skeletally immature individuals with 25% to 50% slippage (Meyerding grade II) are advised not to participate in high-risk collision and contact sports such as gymnastics, football, and weightlifting but are allowed to participate in other noncontact athletic pursuits.41 Skeletally immature individuals with spondylolisthesis of greater than 50% (Meyerding grade III, IV, and spondyloptosis) are typically advised to undergo surgical stabilization because of the risk for progression. In our experience, such patients are usually symptomatic. After skeletal maturity, activity restrictions should be based solely on the patient’s symptoms because significant slip progression is uncommon in the adult population, particularly with low-grade slips (grade I and II). The optimal timing of return to play following spinal fusion is unclear. A survey of the Scoliosis Research Society demonstrated that half of the surgeons who responded to the survey permitted return to noncontact sports 6 months after surgery. Return to contact sports (basketball, soccer) was permitted after 1 year by one third of respondents for both low-grade and high-grade slips. However, return to collision sports such as football and hockey after fusion for low-grade slips was permitted by only 11% at 6 months and by 36% at 1 year.42 It is important when assessing guidelines for return to play to consider both the magnitude of the slip, the individual athlete, the sport, and the position played. The decision about when a pediatric athlete with a disk herniation can return to sports is also unclear. We allow such patients to return to athletics when their symptoms have resolved, when they have a full range of back motion, and when they are neurologically normal. Most patients are significantly better by 4 to 6 weeks and are able to return to at least noncontact sports at that time. Return to a collision sport such as football may take significantly longer. After uncomplicated hemilaminotomy and diskectomy, patients are allowed to return to swimming and limited exercise cycling at 2 to 3 weeks, are allowed to begin running at 4 to 6 weeks, and can resume contact sports at about 4 to 6 months, if their symptoms permit and after they have

768

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

completed appropriate rehabilitation. Most patients are eventually able to return to full athletic participation.

CONCLUSION Back pain in the pediatric athlete is a relatively common condition seen by a variety of medical specialists. The symptoms are rarely disabling, although the parents are often worried and anxious, particularly when symptoms and functional limitations persist for an extended period. The pediatric athlete with back pain only rarely reports a history of an acute traumatic episode, although a precise onset of pain is described by many. A stepwise approach to evaluation and management will result in symptomatic resolution in most patients. The presence of significant structural abnormalities such as disk herniation or spondylolisthesis must be identified. Although rare, the clinician needs to be aware of the possibility of more serious conditions such as tumor or infection in the pediatric athlete with back pain. Fortunately, most athletes in this age range can expect a satisfactory resolution of symptoms, although it may take longer than expected by the physician, the patient, and the family. C

r i t i c a l

P

o i n t s

l Most pediatric and adolescent back pain can be grouped into lumbar strain, spondylolysis or spondylolisthesis, lumbar disk injuries, and overuse syndrome l Judicious use of radiographic imaging in initial evaluation if no red flags are present. l Initial trial of activity restriction, NSAIDs, and moist heat can be undertaken in all benign cases. l SPECT is done in cases of normal radiographs if there is a high suspicion of spondylolysis; CT is performed if SPECT is positive to further demonstrate bony anatomy of defect. l Bony healing of spondylolysis is not required for return to sport; be guided by symptoms, not radiographs.

l Lumbar

disk disease includes a high incidence of axial back pain, and the pain stays above knee (referred pain). l Lumbar disk herniation includes radicular leg pain; neurological examination may be normal; and positive tension signs (straight leg raising sign or bowstring sign) nearly always are present. l Long-term results of surgical treatment of lumbar disk disease are not as good in children as in adults. l Patients with spondylolysis, lumbar strain, and lumbar disk herniation may return to sport when the following criteria are met: nontender to palpation, full range of motion, and minimal pain with activity.

S U G G E S T E D

R E A D I N G S

Beutler WJ, Fredrickson BE, Murtland A: The natural history of spondylolysis and spondylolisthesis: 45-year follow-up evaluation. Spine 28(10):1027-1035, 2003. Bradford DS, Hy SS: Spondylolysis and spondylolisthesis. In Weinstein SL (ed): Principles and Practices. New York, Raven Press, 1994, Spine: The Pediatric ������������������������������������������������������������������������� pp 585-601. Epstein JA, Epstein NE, Marc J, et al: Lumbar intervertebral disk herniation in teenage children: Recognition and management of associated anomalies. Spine 9:427-432, 1984. PrinciKing H: Back pain in children. In Weinstein SL (ed): The Pediatric Spine: ������� ples and Practices. New York, Raven Press, 1994. Jackson DW, Wiltse LL, Dingeman RD: Stress reactions involving the pars interarticularis in young athletes. Am J Sports Med 9(5):304-312, 1981. Miller SF, Congeni J, Swanson K: Long-term functional and anatomical follow-up of early detected spondylolysis in young athletes. Am J Sports Med 32(4):928933, 2004. Newman PH: The etiology of spondylolisthesis. J Bone Joint Surg Br 45:39-59, 1963. Parisi P, Di Silvestre M, Greggi T: Lumbar disc excision in children and adolescents. Spine 26(18):1997-2000, 2001. Rubery PT, Bradford DS: Athletic activity after spine surgery in children and adolescents. Spine 21(4):423-437, 2002. Wiltse LL: The etiology of spondylolisthesis. J Bone Joint Surg Am 44:539-559, 1962.

R eferences Please see www.expertconsult.com

C H A P T E R�  

 �� 17

Shoulder S ect i o n

A

Anatomy and Biomechanics 1. Functional Anatomy and Biomechanics of the Adult Shoulder Patrick J. McMahon and Richard E. Debski

The goal in treating the individual with an injured shoulder is not only to eliminate pain but also to restore normal function. Today’s knowledge of anatomy, pathoanatomy, and biomechanics has resulted in dramatic improvements in the treatment of shoulder injuries. There is less bony stability at the shoulder than at other diarthrodial joints, so soft tissues guide and limit motion. Shoulder motion necessitates the coordinated actions of four separate articulations. With many activities, including rigorous athletics, there is appreciable demand on the shoulder. Injury resulting in pain and loss of function occurs when the physiologic limits of the tissues are exceeded.

instability results from large, symptomatic translations and may manifest clinically as dislocation, the articular surfaces of the joint no longer being in contact. In the shoulder, ­anterior dislocation results in the humeral head being not only ­ anterior but also inferior to its normal position on the glenoid. Subluxation is less severe instability, defined as abnormal, partial contact of the articular surfaces. ­Unfortunately, we lack biomechanical studies defining normal joint translations and clinical tests for glenohumeral joint stability, so clinicians are often left without objective findings to diagnose subluxation. Instead they must use subjective clinical complaints to distinguish between laxity and instability; glenohumeral laxity does not elicit ­complaints

DEFINITIONS Even though the term shoulder is often used to refer to the glenohumeral joint, normal function of the shoulder requires the coordinated function of four articulations: the sternoclavicular, the acromioclavicular, the glenohumeral, and the scapulothoracic (Fig. 17A1-1). Furthermore, the shoulder is composed of nearly 30 muscles and three bones (humerus, clavicle, and scapula) in addition to the upper thorax. In this chapter, shoulder motion refers to the complex interaction of all these structures. The actions (either translations or rotations) between the humerus and the glenoid are described as follows. Small linear movements that take place between the articular surfaces of the humeral head and the glenoid are termed translations. Large angular rotations also occur between the humerus and the scapula. In the normal shoulder, motion is composed of large angular rotations and small glenohumeral translations.1,2 We define the three possible translations as anterior/posterior, superior/inferior, and medial/lateral. The three rotations are internal/external, adduction/abduction in the scapular plane, and adduction/ abduction in the horizontal plane (Fig. 17A1-2). Laxity is defined as normal motions, often called joint play, and has wide variability among individuals. Joint

Shoulder Acromioclavicular joint Sternoclavicular joint

Glenohumeral joint

Figure 17A1-1   Normal function of the shoulder requires the coordinated function of scapulothoracic, sternoclavicular, acromioclavicular, and glenohumeral joints.

769

770

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Y

the Y-shaped view of the supraspinatus fossa (coined the “tangent sign”) on sagittal-oblique magnetic resonance imaging (MRI).6 The anterior portion is more than twice as large as the posterior.7 It passes under the acromion and through the supraspinatus fossa and inserts on the greater tuberosity with an extended attachment of fibrocartilage. The tendon has parallel independent collagen fascicles that probably allow differential excursion of portions of the tendon8 needed for the enormous range of shoulder mobility. Shoulder position affects the morphology of the tendon. External and internal rotation twists the myotendinous junction of the abducted shoulder and elongates the anterior and posterior portions, respectively.9 It inserts onto the greater tuberosity over a distance of about 15 mm and less than 1 mm from the articular cartilage of the humeral head.10 The supraspinatus is active during the entire arc of scapular plane abduction; paralysis of the suprascapular nerve results in about a 50% loss of abduction torque.11,12 Biomechanical studies have found that internal degeneration of the supraspinatus tendon and partial-thickness tears correlate with decreased ultimate tensile strength.13,14 In an animal model, changes consistent with tendinopathy alone were associated with diminished biomechanical properties.15,16 The articular side was more easily ruptured than the bursal side,17 and the tear acts as a stress riser. ­Tendon tears have adverse effects on the muscle of the rotator cuff as well. Atrophy occurs and fat, which can be quantified with computed tomography,18 accumulates within the muscle and unfortunately may be irreversible, even after healing of a tendon repair. This finding in an infraspinatus muscle with an intact tendon, when there is also a supraspinatus tendon tear, is a particularly poor prognosticator of outcome after repair.4 It indicates that over time, the rotator cuff as a whole has been functioning poorly. The infraspinatus and the teres minor muscles originate on the posterior scapula, inferior to the scapular spine, and insert on the posterior aspect of the greater tuberosity. There is a posterior rotator interval consisting of the glenohumeral joint capsule medially, which fuses with the supraspinatus and infraspinatus tendons laterally.19 The insertion of the teres minor tendon can be distinguished from the others by a small separate tubercle that can be palpated on the most posteroinferior aspect of the greater tuberosity. These muscles function together to externally

X

3 2

Z

1

1. Abduction 2. External rotation 3. Horizontal abduction Figure 17A1-2   To describe rotation at the shoulder, only three axes are needed. We suggest the use of internal/external, adduction/abduction in the scapular plane, and adduction/ abduction in the horizontal plane. Description of rotation about other axes, such as flexion and extension, causes confusion.

of pain or instability. Because of wide variability among individuals in their perception of pain, secondary gain, and inconsistency in clinical tests for glenohumeral joint stability, it is often difficult to distinguish laxity and instability.

GLENOHUMERAL JOINT ANATOMY Glenohumeral Muscles Four muscles compose the rotator cuff: the supraspinatus, subscapularis, infraspinatus, and teres minor (Fig. 17A1-3). Normally, there is no fat or, at most, a few small streaks of fat within these muscles.3,4 The supraspinatus has its origin on the posterior-superior scapula, superior to the scapular spine, and fills the supraspinatus fossa. Atrophy is defined as the muscle occupying less than half of the fossa5 or the entire supraspinatus muscle being below a line between the coracoid and the apex of the scapular spine on

Rotator cuff Supraspinatus

Infraspinatus

Teres minor

Subscapularis

A

B

Figure 17A1-3   A and B, The supraspinatus, subscapularis, infraspinatus, and teres minor are the four muscles of the rotator cuff.

Shoulder

rotate and extend the humerus. Both account for about 80% of external rotation strength in the adducted ­position. The infraspinatus is more active with the arm at the side; the teres minor activates mainly with the shoulder in 90 degrees of elevation. The subscapularis muscle arises from the anterior scapula, and its tendon is the only one to insert on the lesser tuberosity. It is the anterior component of the rotator cuff and functions to internally rotate and flex the humerus. At its superior margin is the rotator interval between the subscapularis and supraspinatus. Medially, the rotator cuff tendons are absent altogether, separated by the coracoid, a bony extension of the scapula with an inferiorly directed tip that overlies part of the subscapularis. Some individuals may be prone to subscapularis tendon tears from a shortened distance between the humerus and the coracoid.20,21 Caudad to the coracoid lays the subscapularis bursa, a fluidfilled sac that communicates with the glenohumeral joint and where loose bodies may accumulate. The anatomy and function of the rotator interval is described in further detail later in this chapter. The tendinous insertion of the subscapularis is continuous with the anterior capsule, and these two structures are responsible for providing anterior glenohumeral stability.22,23 In the superior two thirds of the muscle, there is a consistent pattern of tendinous bands evenly dispersed in the midportion of the muscle that condense laterally into a single large, flat tendon.24 The inferior third of the subscapularis is muscular from origin to insertion. Below the subscapularis is the latissimus dorsi; the superior margin of the latissimus dorsi tendon is a good reference for locating the inferior margin of the subscapularis when detaching the subscapularis in its entirety from the humerus. The deltoid is the largest of the glenohumeral muscles and covers the proximal humerus from its tripennate origin at the clavicle, acromion, and scapular spine to its insertion midway on the humerus at the deltoid tubercle. Abduction of the joint results from activity of the anterior and middle portions. The anterior portion is also a forward flexor. The posterior portion does not abduct the joint but instead adducts25 and extends the humerus. The deltoid is active throughout the entire arc of glenohumeral abduction; paralysis of the axillary nerve results in a halving of abduction torque.26 The deltoid muscle can fully abduct the glenohumeral joint with the supraspinatus muscle ­inactive. The teres major muscle originates from the inferior angle of the scapula and inserts on the medial lip of the bicipital (i.e., intertubercular) groove of the humerus, posterior to the insertion of the latissimus dorsi. The axillary nerve and the posterior humeral circumflex artery pass superior to it through the quadrilateral space that is also bordered by the teres minor, the triceps, and the humerus. Because of its proximity to the rotator cuff, the tendon of the teres major muscle, like the latissimus dorsi,27 can be used in unipolar transfer for rotator cuff reconstruction.28 These procedures must be done with care because the axillary and radial nerves and the neurovascular pedicles are at risk.29 The teres major muscle contracts with the latissimus dorsi muscle,30 and they function as a unit in humeral extension, internal rotation, and adduction.

771

Passive Stability of the Glenohumeral Joint Bony Articulation of the Glenohumeral Joint The size of the glenohumeral articular surfaces correlates directly with differences in size and height of both men and women and therefore is extremely variable.31 The humeral head represents about one third of a 45-mm diameter sphere and is spherical32 or slightly elliptical, larger in the vertical dimension than in the horizontal.31 It has a mean thickness of 19 mm.31,33 The articular surface spans an arc of 150 to 160 degrees,34 and the angle between its inferior margin and the shaft (the head-shaft angle) is a mean of 45 degrees.31 The superior-most point on the humeral head articular surface is a mean of about 8 mm above the greater tuberosity.31,35 Lateral humeral offset, the distance from the base of the coracoid to the lateral-most point of the greater tuberosity, is a mean of 56 mm.31 The lateral humeral offset correlates with the radius of curvature and the humeral head diameter but not with the neck shaft angle.36 There is both medial and posterior humeral offset relative to the shaft, that is, the center of the humeral head is about 3 mm posterior37 and nearly 10 mm medial to a line through the middle of the humeral shaft.35 Precise replication of humeral head height is important in shoulder arthroplasty; small malalignments in humeral offset are acceptable.38 The amount of malalignment critical to a satisfactory outcome is unknown. The proximal humerus is generally separated anatomically into four parts: the articular surface, the greater tuberosity, the lesser tuberosity, and the diaphyseal humeral shaft. The humeral head is angulated medially 45 degrees to the long axis of the humeral shaft and retroverted 30 degrees relative to the transcondylar axis of the distal humerus (Fig. 17A1-4).35,39 Retroversion may be determined more accurately with computed tomography than by palpating the epicondylar axis or using the forearm as a reference.40 Between the two tuberosities is the bicipital groove, in which lies the tendon of the long head of the biceps brachii muscle. The bicipital groove has been used in the past as a landmark for correct placement of the humeral head prosthesis in shoulder arthroplasty, but it may be unreliable because there is considerable variation

Distal end of humerus, 0°

Proximal end of humerus 30° retrograde

Figure 17A1-4   Relative to the shaft, the humeral head is angulated medially 45 degrees to the long axis of the humeral shaft and retroverted 30 degrees relative to the transcondylar axis of the distal humerus.

772

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

in its orientation.41,42 The tendon is held in place by the coracohumeral ligament and the transverse humeral ligament, which is made up primarily of fibers of the tendon of the subscapularis tendon that extend laterally and form a sling covering the bicipital groove.43 Dislocations of the tendon of the long head of the biceps brachii disrupt these fibers and result in the biceps tendon lying either under or within the subscapularis tendon. During abduction of the glenohumeral joint, the proximal humerus slides on the tendon of the long head of the biceps brachii. If the tendon ruptures, translation of the humeral head increases.44 The mean size of the glenoid is about 35 mm in vertical diameter and 25 mm in horizontal diameter.15,31,34,45,46 Relative to the plane of the scapula, the fossa is angled slightly superior and posterior,2,47 offering little bony support to inferior instability with the arm at the side (Fig. 17A1-5). There is variation in the glenoid along its rim so that the available bone for insertion of an anchor, as is done in instability repair, depends on the position of the glenoid rim.48 In the first 2 years of life, the glenoid is normally retroverted about 6 degrees, but by the end of the first decade of life, it reaches adult retroversion of 2 to 6 degrees.49 There has been considerable controversy about the conformity of the two articular surfaces,50,51 partly because the bony glenoid is relatively flat, whereas the articular surface is not. Articular cartilage is thicker at the periphery of the glenoid than at the center; conversely, the articular surface of the humeral head is thicker at the center than at the periphery. Unfortunately, radiographs lead to the mistaken impression that the articular surface of the glenoid is flatter than that of the humeral head. Direct measurement of the articular surfaces found the articular surfaces to be matched nearly perfectly with similar radii of curvature, and during joint motion, the humeral head maintained uniform contact with the glenoid.32 This point is controversial, however, and others have found the surfaces to be noncongruent, with the radius of curvature of the articular surface of the glenoid being slightly larger than that of the humerus.31 This is important because glenohumeral translations possibly occur at the extremes of motion1 and certainly occur in unstable joints.

The position of the glenohumeral articular surfaces contributes to joint stability. Because the glenoid is concave and tilted slightly superior when it is viewed in an anteroposterior direction, inferior joint translation is coupled to lateral translation of the humeral head. The lateral translation tightens the superior capsular structures and the supraspinatus, maximizing inferior joint stability.52 Knowledge of normal glenohumeral joint anatomy is important to maximize outcome after surgical repair. Nonunion may occur if the tuberosities are not positioned in contact with the humeral shaft in treatment of four-part proximal humerus fractures. A greater tuberosity malpositioned above the articular surface of the humeral head may lead to pain from abnormal contact with the coracoacromial arch. Failure to reproduce humeral thickness and offset may lead to weakness and diminished range of motion. Abnormal soft tissue tensioning and glenoid impingement may be responsible. Excessive glenoid retroversion may also lead to joint instability. Such alterations from the normal articular surface geometry are unusual unless they are the result of glenoid hypoplasia or trauma. After shoulder arthroplasty, however, alterations of joint geometry are not unusual. Although limitations of prosthetic design may sometimes make reconstruction of the glenohumeral joint difficult,33 surgeons remain responsible for precisely positioning head and glenoid components.

Glenohumeral Joint Capsule The thin redundant joint capsule has almost twice the surface area of the humeral head (Fig. 17A1-6).53 This, in part, allows the greatest range of joint motion of any single diarthrodial joint. Different regions of the joint capsule provide passive stability, and their contribution is dependent on the position of the shoulder. With the arm at the side, the superior portion of the capsule is taut, and the inferior portion is lax. With overhead elevation, this relationship reverses. There are folds or thickenings visible on the inside of the capsule with the shoulder at the side, which have been termed glenohumeral ligaments. Although this ­terminology

Opening into subscapularis bursa

5°

Infraspinatus

Biceps Sup. glenohumeral ligament Middle glenohumeral ligament

Glenoid

Teres minor

Anterior inferior glenohumeral ligament Labrum

Figure 17A1-5   Relative to the plane of the scapula, the fossa is angled slightly inferior and posterior, offering little bony support to inferior instability with the arm at the side.

Figure 17A1-6   The thin redundant joint capsule has almost twice the surface area of the humeral head, allowing   a tremendous range of joint motion.

Shoulder

is generally accepted, it needs some clarification. Ligaments are soft tissue structures that connect bone to bone. They are most commonly band-like, with parallel collagen fibers running between their insertion sites, and have clearly defined edges, such as the anterior cruciate ligament of the knee. The glenohumeral capsule as a whole may be considered a sheet-like ligament connecting the humerus and scapula, but the collagen fibers are not organized in a parallel fashion,54,55 and the margins of the folds are indistinct. This may be the reason they have been described with variable prevalence; different authors have had varying success in identifying them.56-58 Also, with the shoulder in abduction and external rotation, even the most consistently reported fold in the anteroinferior capsule, the anterior band of the inferior glenohumeral ligament, seems to disappear. Current surgical treatment for glenohumeral joint instability includes tightening the capsule not only between but also perpendicular to the insertion sites. Biomechanical studies that either applied an anterior load or translation to the humerus indicated that capsular strains are not ­ simply oriented along the length of the folds as is commonly found with band-like ligaments.59,60 Also, the mechanical properties of the capsule are not consistent with a band-like ligament resisting loads along its length.61,62 Although terminology may be currently causing confusion in anatomic, biomechanical, and clinical studies, there is little doubt that different regions of the capsule have differing roles in joint function, and this understanding has resulted in improved outcomes for shoulder injuries. ­Further study of the glenohumeral joint capsule is needed; the following descriptions, however, detail the anatomy and function of the aforementioned glenohumeral ­ligaments. The superior glenohumeral ligament originates anterior to the tendon of the long head of the biceps brachii. If the glenoid had the markings of a clock, with the 12-o’clock position superior and the 3-o’clock position anterior, the origin of the superior glenohumeral ligament would correspond to the area from the 12-o’clock to the 2-o’clock positions. The superior glenohumeral ligament runs inferior and laterally and inserts into the fovea capitis on the humerus, just superior to the lesser tuberosity. The superior glenohumeral ligament is important in resisting inferior subluxation with the arm at the side.63 The middle glenohumeral ligament arises from the neck of the glenoid just inferior to the origin of the superior glenohumeral ligament and inserts on the humerus just medial to the lesser tuberosity. Viewed from inside the joint, it crosses over the superior margin of the subscapularis tendon. The presence of the middle glenohumeral ligament is the most variable.56,58 It may be a distinct, band-like structure, it may be confluent with the inferior glenohumeral ligament, or it may be poorly defined. The middle glenohumeral ligament limits anterior translation of the humeral head with the arm in moderate (45 degrees) abduction and external rotation.64 The inferior capsule has been identified as the inferior glenohumeral ligament, a broad and complex structure with closely packed collagen fiber bundles.65 It has been likened to a hammock, with an anterior band, a posterior band, and an axillary pouch in between.66 With abduction

773

and external rotation, the anterior band fans out, and the posterior band becomes cord-like. Likewise, with internal rotation, the posterior band fans out, and the anterior band appears cord-like. The anterior band of the inferior glenohumeral ligament arises from various areas corresponding to the 2-o’clock to 4-o’clock positions on the glenoid. The posterior band originates at the 7-o’clock to 9-o’clock positions and inserts on the posterior articular margin of the humerus. With the arm at the side, both the anterior and posterior bands pass through a 90-degree arc and insert on the humerus. In addition to resisting anterior translation with the shoulder abducted and externally rotated, the anterior band of the inferior glenohumeral ligament is a secondary restraint to inferior instability with the arm at the side.67 The posterior band is probably important in resisting inferior subluxation in the joint abducted more than 45 degrees.63 The importance of the anterior capsuloligamentous structures in resisting posterior glenohumeral instability has been implied.68,69 Whether it occurs alone or in combination with a Bankart lesion, small permanent stretching of the capsule contributes to anterior instability.55,70-74 Tearing of the capsule can also occur, and afterward the anterior band may heal, albeit in a lengthened state. In this way, the manner of tearing rather than permanent stretching of the capsuloligamentous structures may be responsible for the spectrum of glenohumeral instability that is observed clinically.55 Permanent stretching of the capsuloligamentous soft tissues at the time of ligament failure occurs not only in the medial to lateral direction but also in the superior to inferior direction.59,60 This concurs with the clinical finding that permanent lengthening of the capsule occurs not only along the anatomic alignment of the glenohumeral ligaments but also in the superior to inferior direction, and plication of the capsule in both directions leads to an acceptable outcome.75,76 The optimal amount of ligament plication after repetitive episodes of instability remains unknown and may be different in individuals of different ages at the time of injury.77 Findings of biomechanical studies indicate, however, that the amount of plication necessary after an initial instability episode is small, probably less than a few millimeters.55,71-73

Rotator Interval and Coracoacromial Ligament The rotator interval consists of the superior glenohumeral ligament and the coracohumeral ligament. The coracohumeral ligament originates on the base of the coracoid and passes between the supraspinatus and subscapularis tendons to blend with the insertion of the subscapularis tendon on the lesser tuberosity. It helps to support the dependent arm and acts as a restraint to external rotation in this shoulder position.78 The coracohumeral ligament has anterior and posterior bands originating at the coracoid. The anterior band inserts onto the lesser tuberosity and becomes most taut with external rotation when the arm is at the side. The posterior band inserts onto the greater tuberosity and becomes most taut with internal rotation when the arm is at the side. Injury to the rotator interval results in posteroinferior instability,79 and tightening of these tissues significantly

774

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

decreases posterior and inferior translation of the joint.79 Posterior dislocation may not occur, even after tearing of the entire posterior capsule, unless the rotator interval also tears.80 Because the coracoacromial ligament was thought to have little functional importance, its release has traditionally been performed liberally in treatment of rotator cuff injuries. However, it has roles in normal shoulder function.81,82 The coracohumeral ligament is a barrier to anterosuperior translation of the humeral head, especially after rotator cuff rupture. Detachment of the coracoacromial ligament also results in increased anterior and inferior translation of the internally and externally rotated glenohumeral joint because of impairment of interaction between the coracoacromial ligament and the coracohumeral ligament.83

Labrum The labrum surrounds the periphery of the glenoid and is composed of dense fibrous connective tissue; it is the insertion of the capsuloligamentous structures. The anatomy of the anteroinferior region, which is the glenoid insertion site of the anterior band of the inferior glenohumeral ligament, is more complex. The insertion site of this ligament has two attachments, one to the glenoid labrum and the other directly to the anterior neck of the glenoid.84 The size of each attachment relative to the other is variable, but both are likely to be important in resisting anterior translation of the humeral head. The labral attachment demonstrates similarities to direct ligament or tendon insertions, whereas the glenoid neck attachment near the 3-o’clock position is similar to indirect ligament insertions, such as the tibial insertion of the medial collateral ligament. There are superficial fibers that run parallel to the bone surface, blending in with the periosteum, and deep fibers that attach straight to the bone. At the 6-o’clock position, as the collagen bundles approach the glenoid, they appear disorganized, but this finding probably represents a change in direction of the collagen bundles, that is, the collagen bundles of the mid-substance of the anteroinferior capsule are primarily oriented in a radial fashion, but they change direction near the glenoid and form the labrum as a circular system of fibrous, collagen bundles.65 The glenoid labrum acts not only as an attachment site for the capsuloligamentous structures but also as an extension of the articular cavity, increasing the surface area of the glenoid85 by nearly 50%.86 Removal of the labrum decreases joint stability to translational forces by 20%.87

Intracapsular Pressure Anatomic studies, surgical findings, and MRI studies confirm that a layer of synovial fluid less than 1 mm in thickness is present in the normal glenohumeral joint.88 Because joint volume is finite, a negative intracapsular pressure is generated with attempts to distract the glenohumeral joint. In this way, slightly negative intracapsular pressure, which is present in the normal joint,89 aids in centering the humeral head.90 The negative intracapsular pressure is likely to be of particular importance in limiting inferior subluxation of the humeral head.67

Active Stability (Concavity-Compression) of the Glenohumeral Joint Large forces are generated in the shoulder muscles to move the shoulder against gravity. In the normal shoulder, this contributes meaningfully to joint stability through the application of a compression force. This is the component of the glenohumeral joint force that acts perpendicular to the glenoid fossa such that the concave humeral head is compressed into the glenoid fossa. Termed concav-­ ity-­compression,87 this action was initially reported to be important in maintaining joint stability at the midranges of shoulder elevation87 when the passive restraints are lax. The shoulder muscles that are active in elevation of the arm91 include the rotator cuff and are directed toward the glenoid fossa. The anterior capsuloligamentous structures are lax, so the compression force maintains stability. In addition to the rotator cuff muscles, any muscle that crosses the glenohumeral joint can contribute to concavity-compression from the component of its force acting perpendicular to the glenoid. The muscle orientation along with the magnitude of the muscle force determines how much a muscle contributes to concavity-compression. Concavity-compression is also important at the end range of elevation.92-94 Because muscle forces are larger with the arm in 90 degrees of elevation than at lower levels, concavity-compression is also larger, as long as the humeral head remains centered on the glenoid. It is also important to note that recent studies indicate that muscles, specifically the pectoralis major and the latissimus dorsi, may in certain circumstances contribute to glenohumeral instability.94-97 When the shoulder is forced into horizontal abduction, similar to the apprehension position, fibers of these muscles have ideal orientation to pull, possibly with passive tension, the humeral head anteroinferiorly, resulting in dislocation.94

Interplay of Passive and Active Stability of the Glenohumeral Joint In the apprehension position, precise interplay of the active and passive restraints results in anterior joint stability. The way in which the anterior band of the inferior glenohumeral ligament acts with concavity-compression can be likened to the manner by which a rider controls a powerful horse. In this example, the horse represents concavity-­compression force, and the reins represent the anterior band of the inferior glenohumeral ligament. When the horse is running straight ahead, little control on the reins is needed to maintain it on a path. If the horse begins to turn, a light tug on the opposite rein redirects the horse onto the path. In the same way, the anterior band of the inferior glenohumeral ligament, which alone has biomechanical properties insufficient to resist the large force of dislocation,55,72,73,77 can maintain the humeral head in the glenoid. Concavity-compression is not significantly altered after injury to the anteroinferior capsule alone.98 In fact, the compression force, that is, the component oriented perpendicular to the glenoid, is large even after anteroinferior passive restraint injury. As in the example, if the horse is running straight ahead, the rider can drop the reins, and the horse will not initially deviate from the path.

Shoulder

Unfortunately, in the shoulder with pathologic changes, other joint forces that contribute to instability can offset concavity-compression. For example, in the shoulder with a supraspinatus tendon tear, the glenohumeral joint force is oriented superior rather than perpendicular to the glenoid as in the normal shoulder. Over time, this results in erosion of the superior rim of the glenoid, and the humeral head can sublux a small amount superiorly.99 This serves only to direct the glenohumeral joint force more superiorly. Eventually, the entire superior rim of the glenoid becomes worn, and the humeral head ends up abutting the acromion. In a similar manner, if a traumatic anterior dislocation results in damage to the anteroinferior glenoid, the humeral head subluxes a small amount anteroinferiorly. The glenohumeral joint force is no longer directed perpendicular to the glenoid but instead is directed a small amount anteroinferiorly. The passive restraints fail in time, and the instability worsens.

KINEMATICS Kinematics is the study of joint motion, including rotations and translations. Loosely, this means the study of motion and stability. Normal shoulder function requires the coordinated function of four articulations. In the following paragraphs, we detail the kinematics of each.

Sternoclavicular Joint Kinematics The sternoclavicular joint is the only true synovial joint connecting the upper extremity and the axial skeleton. Although it lacks bony stability, four ligaments and the intra-articular disk stabilize it. The interclavicular ligament provides restraint to superior joint motion and is taut when the shoulder is at the side.100 Anterior and posterior motion is prevented by the anterior and posterior capsular structures, the anterior being the stronger of the two.101 The anterior structures also resist superior motion. The joint is also stabilized by the costoclavicular ligaments, which run obliquely and laterally from the first rib to the inferior surface of the clavicle. The intraarticular disk restrains medial displacement of the clavicle (Fig. 17A1-7). Sternoclavicular joint Anterior sternoclavicular Articular ligament disk Interclavicular ligament

Costoclavicular ligament Figure 17A1-7   The sternoclavicular joint is stabilized by the interclavicular ligament, anterior and posterior capsular structures, costoclavicular ligaments, and the intra-articular disk.

775

Acromioclavicular Joint Kinematics In the coronal plane, the acromioclavicular joint is inclined 20 to 50 degrees,53,102 which may affect joint stability. Unlike the sternoclavicular joint, there is meaningful stability from structures other than the joint capsule. The clavicle is stabilized to superior motion at the acromioclavicular joint by the conoid and trapezoid coracoclavicular ligaments, whereas the joint capsule is a restraint to anterior and posterior translations. Acromioclavicular ligaments are thickenings of the capsule. The inferior ligament is the primary restraint to anterior translation.103 The trapezoid coracoclavicular ligament is the primary restraint to posterior translation (Fig. 17A1-8).103 These functions were determined during constrained joint motion, with change based on the constraints placed on the acromioclavicular joint.104 During 6 degrees-of-freedom (DOF) joint motion, the translation of the clavicle increased almost 50% compared with previous studies that allowed only 3 DOF motion. Therefore, the kinematic constraints placed on the acromioclavicular joint during loading affect not only the resulting joint motion in the primary direction of loading but also the magnitude and direction of force in each ligament and the coupled motions. The motion at the acromioclavicular joint also changes with joint pathology and surgical procedures. Shoulders with injury to the acromioclavicular joint capsule were subjected to three loading conditions, and differences were found between the force in the trapezoid and conoid, with each loading condition suggesting that these ligaments should not be treated as one structure when surgical treatment is considered.105 Furthermore, the coracoclavicular ligaments could not compensate for the loss of capsular function during anterior-posterior loading as occurs after acromioclavicular joint injuries. Most recently, the effect of acromioplasty and distal clavicle resection on joint kinematics and in situ forces in response to external loads was determined.106 Acromioplasty alone did not significantly affect acromioclavicular joint motion or the in situ forces in each ligament. However, distal clavicle resection significantly affected the motion of the acromioclavicular joint in response to posterior loading. Some of these changes in joint motion and in situ forces could increase the potential for further degeneration at the distal clavicle and lead to loss of function. These procedures could also affect the high compressive loads that are transmitted across the acromioclavicular joint to the axial skeleton during activities of daily living. Joint compression was shown to decrease the posterior translation in response to a posterior load.107 The application of joint compression to the capsule-transected acromioclavicular joint decreased the amount of posterior and superior translation during posterior and superior loading, respectively, while increasing the coupled translations (anteriorposterior, superior-inferior or proximal-distal). The joint contact force also increased. Therefore, common surgical techniques such as distal clavicle resection, which initially reduce painful joint contact, may cause unusually high loads to be supported by the soft tissue structures at the acromioclavicular joint. In addition, the compressive loads transmitted across a capsule-transected acromioclavicular joint could be concentrated over a smaller area because of the increased coupled motion and joint contact force.

776

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17A1-8   The conoid and trapezoid coracoclavicular ligaments and the joint capsule stabilize the acromioclavicular joint. Acromioclavicular ligaments are thickenings of the capsule.

Acromioclavicular joint Coracoclavicular ligament Trapezoid ligament Conoid ligament

Coraco-acromial ligament

Acromioclavicular ligament

Transverse humeral ligament Synovial sheath of biceps

The clavicle moves with shoulder motion, but how much it moves was reported differently in the past.39,100 During the past decade, three-dimensional in vivo kinematics of the acromioclavicular joint and clavicle during arm elevation have been examined using MRI and electromagnetic tracking devices. Images from a vertically open MRI of 14 shoulders during arm elevation revealed that the clavicle translated nearly 2 mm posteriorly at 90 degrees of abduction, and nearly 2 mm anteriorly at maximal abduction.108 The clavicle also translated nearly 1 mm in the superior direction. The scapula rotated about 35 degrees through an axis passing generally through the insertions of both the acromioclavicular and the coracoclavicular ligaments on the coracoid process during arm elevation. Clavicular motion with respect to the thorax in subjects with or without shoulder pathology was found to include elevation (range, 11 to 15 degrees), retraction (range, 15 to 29 degrees), and posterior long-axis rotation (range, 15 to 31 degrees), with variability between subjects and planes of motion during arm elevation.109

Scapulothoracic Kinematics Motion between the scapula and the thorax, the scapulothoracic articulation, is an integral part of normal shoulder function. Other than muscular attachments, only the acromioclavicular joint and the coracoclavicular ligaments support the scapula. The scapula is mobile in many directions because of this unique articulation. The scapulothoracic muscles include the trapezius (upper, middle, and lower portions), levator scapulae, serratus anterior, pectoralis minor, and rhomboid muscles. They all act to position the scapula in the proper orientation on the thorax during shoulder motions. The levator scapulae and the upper trapezius provide postural support. The middle trapezius and rhomboids retract the scapula, whereas the serratus anterior protracts the scapula. The trapezius and serratus anterior rotate the lateral part of the scapula upward. The upper trapezius and levator scapulae elevate the scapula.

This scapula positioning provides maximal stability at the glenohumeral joint while maintaining a large range of motion. The motion that takes place between the thorax, the scapula, and the humerus is extremely complex. Kinematic studies of the entire shoulder, including both the glenohumeral and scapulothoracic motion, probably should be performed using only live subjects until enough information is available to enable these complex motions to be simulated in cadaveric studies. To simplify this complex motion, basic science studies have focused on isolated motions at the glenohumeral joint with the scapula fixed. The relative motion between the scapulothoracic articulation and the glenohumeral joint during abduction is termed the scapulothoracic rhythm. Past study has found that for the first 30 degrees of abduction, glenohumeral motion is much greater than scapulothoracic motion, the ratio of motions being reported to range from 4:1 to 7:1.2,110 Thereafter, both joints move about the same amount.2,110 Recently, significant efforts have been made to better quantify the three-dimensional motion at the scapulothoracic articulation.111-113 Three rotations are used to describe the orientation of the scapula on the thorax: (1) upward and downward rotation, (2) external and internal rotation, and (3) posterior and anterior tilting.111 The scapula has been shown to rotate upward, rotate externally, and tilt posteriorly 113,114 during arm elevation in asymptomatic individuals under passive114 and active conditions.111 The effects of external rotation, fatigue, impingement syndrome and growth and development on scapulothoracic motion have also been assessed. Interestingly, differences were observed between children and adults during elevation of the arm in the scapular plane.115 Children had a greater contribution from the scapulothoracic joint, most notably upward rotation. In addition, scapular rotation was different between active and passive arm elevation in normal adults.116 Increased upward rotation and external rotation of the scapula were observed during active arm elevation. These findings suggest that the muscles (upper

Shoulder

and lower trapezius and serratus anterior) have an important role during arm elevation and that the glenohumeral capsule and passive muscle tension contribute to this motion. Fatigue due to an external rotation task was found to alter the resting position of the scapula and the posterior tilting during the early range of scapular plane elevation during arm elevation.117 These kinematic changes might affect the subacromial space and facilitate impingement of the humerus on the acromion. A 6-week rehabilitation program for patients with shoulder impingement was also assessed.118 No changes in scapular kinematics were found following the program, even though range of motion increased. Therefore, scapular kinematics was not responsible for improvements in function with these patients. The resting length of the pectoralis minor has also been shown to alter scapular kinematics during arm elevation.119 Subjects with a short pectoralis minor had scapular kinematics that were similar to those exhibited in earlier studies by subjects with shoulder impingement. These results suggest that a short pectoralis minor may be a potential mechanism for subacromial impingement. Subjects with symptoms of shoulder impingement were found to have different scapulothoracic motion.112 Decreased scapular upward rotation, increased anterior tipping, and increased scapular medial rotation were typical during work activities. Therefore, these motion changes are important during rehabilitation of patients with symptoms of shoulder impingement related to occupational exposure to overhead work. When comparing subjects with impingement syndrome and subjects without known pathology or impairments that were matched by age, sex, and hand dominance, the impingement group demonstrated slightly greater scapular upward rotation during flexion and slightly greater scapular posterior tilt during scapular plane elevation.120 All these differences in scapular kinematics could be compensatory strategies for glenohumeral weakness or motion loss in subjects with impingement syndrome.

Glenohumeral Joint Kinematics The normal glenohumeral joint is lax in all directions. There is passive glenohumeral translation of nearly 12 mm in the anterior and posterior directions in cadaveric shoulders.121 In normal subjects, passive humeral translation has been shown to be 8 mm anteriorly, 9 mm posteriorly, and 11 mm inferiorly. Because there is a wide variability of laxity among individuals, these values can reach up to 20 mm in subjects without shoulder instability. It is important to note that these translations do not represent in vivo kinematics occurring when muscle forces are present and create joint compression. The amount of anterior-posterior translation observed during loading experiments using a robotic testing system122 supports the commonly described function of the glenohumeral joint capsule121,123-125 in restricting motion of the humerus more at the limits of motion rather than the midranges. Translations under maximal loading were smaller at 0 and 90 degrees of abduction than at 30 and 60 degrees of abduction. However, posterior translation was restricted in some cases by contact between the acromion

777

and humerus, whereas anterior translation was limited by contact with the coracoid at all abduction angles. Because the 6-DOF kinematics of clinical tests is commonly used to diagnose shoulder injuries, knowledge of glenohumeral joint translations during simulated simple translation tests is important.126 At 60 degrees of glenohumeral abduction and 0 degrees of flexion/extension, a clinician applied anterior and posterior loads to the humerus at 0, 30, and 60 degrees of external rotation until a manual maximum was achieved. Before each test, the reference position of the humerus shifted posteriorly 1.8 ± 2.0 and 4.1 ± 3.8 mm at 30 and 60 degrees of external rotation, respectively. Anterior translation was significantly greater (18.2 ±5.3 mm at 0 degrees of external rotation) than at 30 degrees (15.5 ± 5.1 mm) and 60 degrees (9.9 ± 5.5 mm), respectively. Coupled translations at 0, 30, and 60 degrees of external rotation occurred in the inferior direction during these tests. Therefore, clinicians performing a simulated simple translation test cause coupled inferior translations and anterior translations that are a function of external rotation. Glenohumeral translation has also been assessed in vivo. An electromagnetic position sensor was used to compare anteroposterior laxity of the shoulders in female athletes who were either swimmers or soccer players.127 At 90 degrees of abduction and neutral rotation, the glenohumeral translation for the soccer players was about 10 mm in both the dominant and nondominant shoulder. The translations in the swimmers were significantly greater at 12.4 mm in the dominant and 13.8 mm in the nondominant shoulders. These results suggest that athletic activity might influence joint laxity, or there may be selection bias for different activities because of joint laxity. In vivo glenohumeral joint laxity was also studied using an instrumented shoulder arthrometer.128 Analysis of recreational athletes with no history of shoulder injury revealed no differences in glenohumeral joint laxity between right and left shoulders. Subsequently, the amount of glenohumeral joint translation as a function of force to reach a capsular end point was found to be different between loading directions.129 The anterior-directed translations required a greater magnitude of applied force to reach capsular endpoint than inferior-directed translations, even though the magnitude of translation was not different. Much less glenohumeral translation occurs during arm elevation and rotation in vivo when muscle forces are present and create joint compression than during simple translation tests.121 Radiographic analysis of normal volunteer subjects has shown that centering within 1 mm of the humeral head on the glenoid is maintained in all positions except when the arm is in maximal horizontal extension and external rotation.1 In this position, about 4 mm of posterior translation occurs. In cadaver shoulders with a rigidly fixed scapula, passive forward elevation of the humerus up to 55 degrees and passive extension to 35 degrees in the ­sagittal plane correlated with no measurable glenohumeral translation. Small translations began to occur beyond 55 degrees of flexion or 35 degrees of extension. In another study, throughout humeral elevation in the scapular plane to 90 degrees, essentially pure rotation between the humerus and glenoid occurred130,131 when muscle forces were simulated to move cadaver shoulders. Other investigators also

778

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

found this ball-and-socket motion, even with different internal-external rotation positions of the humerus.132 The in vivo three-dimensional kinematics of the shoulder joint during active abduction has been studied to assess the effects of joint pathology.133 In 25 subjects experiencing shoulder symptoms for more than 18 months and without full-thickness rotator cuff tears, three rotations were examined: abduction/adduction, internal/external rotation, and flexion/extension. Their abduction was not different from controls, and translation occurred in the medial, proximal, and anterior directions. However, in the subjects with shoulder symptoms, proximal translation was increased. Rotator cuff tears are usually associated with excessive superior translation at the glenohumeral joint. Therefore, in vivo translations at the glenohumeral joint were quantified by open MRI techniques during active shoulder abduction with or without infraspinatus and supraspinatus muscle paralysis.134 The humeral head always remained centered in the glenoid fossa during active abduction, suggesting that a structurally intact muscle-tendon-bone unit can prevent superior translation even though the muscles are not functioning properly. This in vivo analysis is supported by experiments that simulate active abduction of the upper extremity.131 Internal and external rotations of the humerus are also important for shoulder motion and are often related to instability and injury mechanisms at the glenohumeral joint. The rotational range of motion at 45 degrees of abduction with an applied 4-N-m moment about the long axis of the humerus in subjects without shoulder pathology was found to be 139 ± 41 degrees in normal subjects.135 The neutral zone laxity was 78 ± 46 degrees. These values could be used to assess normal joint laxity or outcomes after surgery or rehabilitation.

SUMMARY Treatment of shoulder injuries requires understanding of shoulder anatomy, function, and biomechanics. We must continue to study anatomy because advances in treatment are accompanied by new ways to injure the shoulder, as for example use of the arthroscope required precise knowledge of the axillary nerve location in relation to commonly used portals. Findings continue to demonstrate how the passive stabilizers guide the shoulder muscles for optimal function. Our treatments for shoulder injuries are better than those of the past, but full restoration of shoulder function is sometimes difficult. In the future, cooperative research into the anatomy, biology, and biomechanics of both the normal and injured shoulder offer best chances for superb outcomes.

C

r i t i c a l

P

o i n t s

l The rotator cuff muscles act together with the other shoulder muscles to result in normal motion and ­stability. l The three portions of the deltoid muscle do not act ­similarly during shoulder motion. l Rotator cuff tendon injury may have adverse effects on the muscle as well. l Most surgery of the proximal humerus should closely ­replicate the normal anatomy. l Portions of the glenohumeral capsule have specific function for its stability; the anteroinferior portion is vital to prevent anterior instability. l The anatomy of the glenohumeral labrum varies at ­different positions along the rim of the glenoid. l Force on the glenohumeral joint resulting from the ­muscles during shoulder motion is very important to joint ­stability. l Joint stability results from complex interplay of many components, both passive and active. l Shoulder kinematics is complex, partly because of the coordinated actions at four different articulations. l Clinicians should strive to distinguish the normal kinematics of joint laxity from the abnormal so that instability can be treated.

S U G G E S T E D

R E A D I N G S

Eberly VC, McMahon PJ, Lee TQ: Variation in the glenoid origin of the IGHL anterior band: Implications for repair of the Bankart lesion. Clin Orthop 400: 26-31, 2002. Gohlke F, Essigkrug B, Schmitz F: The patterns of the collagen fiber bundles of the capsule of the glenohumeral joint. J Shoulder Elbow Surg 3(3):111-128, 1994. Itamura J, Dietrick T, Roidis N, et al: Analysis of the bicipital groove as a landmark for humeral head replacement. J Shoulder Elbow Surg 11(4):322-326, 2002. McMahon PJ, Tibone JE, Cawley PW, et al: The anterior band of the inferior glenohumeral ligament: Biomechanical properties from tensile testing in the ­position of apprehension. J Shoulder Elbow Surg 7(5):467-471, 1998. Steinbeck J, Lilenquist U, Jerosch J: The anatomy of the glenohumeral ligamentous complex and its contribution to anterior shoulder stability. J Shoulder Elbow Surg 7(2):122-126, 1998.

R eferences Please see www.expertconsult.com

Shoulder

S ect i o n

779

A

Anatomy and Biomechanics 2. Anatomy, Biomechanics, and Kinesiology of the Child’s Shoulder Ralph J. Curtis Jr.

The shoulder is a complex series of four joints—the ­glenohumeral joint, the acromioclavicular joint, the sternoclavicular joint, and the scapulothoracic articulation— that function together to accomplish the primary function of prehensile use of the upper extremity. The shoulder is well suited for this purpose because of the minimal bony constraints and elaborate soft tissue attachments that allow a large degree of freedom and multiplanar range of motion at the joint. This premium on shoulder motion is accomplished by sacrificing inherent stability, which explains why instability is a common feature of shoulder patho­l­ ogy. This chapter endeavors to outline the basic features of shoulder bone and soft tissue anatomy, biomechanical function, and kinematics that make this structure unique. The specific differences between the child and the adult are ­emphasized.

SPECIAL FEATURES OF PEDIATRIC ANATOMY The unique aspects of pediatric anatomy are based on the body’s need for continued growth through the period of life from birth to adulthood. In the skeletally immature athlete, open growth plates include primary physes and secondary apophyses, which are characteristic of growing bone. The physis itself is an area of rapidly growing cartilage that has much less tensile strength than the surrounding epiphysis or metaphysis. Therefore, the physis is more vulnerable to injury by excessive compressive and shearing forces. Injuries to the growth plate are characterized by rapid healing and remodeling but also by the potential for growth arrest or disturbance. The histologic appearance of the growth plate is well recognized.1 It has a typical columnar orientation of cartilage cells with progressive cellular hypertrophy, provisional calcification, and finally endochondral ossification. The zone of hypertrophy and the zone of provisional calcification have most often been implicated as the anatomic sites of growth plate fractures (Fig. 17A2-1). Long bones in skeletally immature individuals are characterized by a primary ossification center within the diaphysis (shaft), a physis (growth plate), and an epiphysis or secondary ossification center at either end. In children, the diaphysis is surrounded by a thick periosteal sleeve that provides appositional growth and remodeling. The

growth plate or physis is the primary site of longitudinal bone growth that occurs by endochondral ossification. The epiphysis or secondary ossification center begins as a completely cartilaginous anlagen at the ends of long bones. As maturity approaches, this structure is progressively ossified, leaving only the articular surface as cartilage. The external surface of the epiphysis or perichondrium serves as a direct attachment point for muscles, tendons, and ligaments. This anatomic arrangement increases the vulnerability of the physeal plate to injury, such as fracture. Injuries to the growth plate have been classified in detail by Salter and Harris, Rang, and more recently Ogden.2 The Salter-­Harris classification, the most commonly used in clinical practice, describes five types of injuries involving the physeal plate (Fig. 17A2-2). The plasticity of young bone is greater than that in an adult. This relates to the thick periosteal tube that surrounds immature bone. Bone tends to get stiffer and more brittle with age. Immature bone is susceptible to plastic deformation characterized by the greenstick type of fracture. In addition to injury associated with acute macrotrauma, the athletic pediatric population is especially susceptible to

EPIPHYSIS

PHYSIS

germinal proliferating palisading

ZONE OF GROWTH

hypertrophy calcification degeneration

ZONE OF CARTILAGE TRANSFORMATION

vascular entry osteogenesis

ZONE OF OSSIFICATION

remodeling

METAPHYSIS

Metaphyseal vessels METAPHYSIS Figure 17A2-1   The physis of the proximal humerus. Note that fractures through the growth plate often occur through the zone of hypertrophy and the zone of provisional calcification.

780

DeLee & Drez’s Orthopaedic Sports Medicine

1

2

4

3

5

Figure 17A2-2   The Salter-Harris classification as applied to the proximal humerus. Type 1 is a transverse fracture through the physis. Type 2 is a transverse fracture through the physis with a small metaphyseal fragment. Type 3 is a transverse fracture through the physis including a fracture through the epiphysis. Type 4 is a longitudinal fracture across the epiphysis and the metaphysis through the physis. Type 5 is a crushingtype injury to the physis involving a central portion.

repetitive overuse microtrauma-type activities. Repetitive stress in a tensile fashion across the epiphysis or apophysis can lead to stress-related injuries, such as osteochondroses and stress fractures.

DEVELOPMENTAL ANATOMY Prenatal human development is divided into two stages: the embryonic period that extends from conception to the eighth week of development and the fetal period that continues from the eighth week until birth.3 The embryonic period is characterized by rapid formation and differentiation of the upper limb buds from the primitive ectoderm and mesoderm into an adult-like extremity.4 During the fourth week of gestation, the upper limb buds develop as small elevations on the ventrolateral body wall opposite the lower six cervical and first and second thoracic segments. The upper limb bud begins as a sac of ectoderm filled with mesoderm. It is from the mesodermal mesenchymal cells that the musculoskeletal tissues of the upper extremity develop.4 During the fifth week, peripheral nerves grow into the mesenchyme of the upper limb buds from the brachial plexus, stimulating the development of limb musculature. Simultaneously, the central core of the humerus begins to chondrify and the scapula appears, positioned at the level of C4 and C5. The precursor to the shoulder joint, called the interzone, appears between the humerus and scapula at this stage. Also during the fifth week of gestation, the clavicle begins to ossify through intramembranous ossification from two centers; it is one of the first bones to ossify in the human embryo.5,6 During the sixth week, the hands develop from the distal mesenchymal tissue. Bending occurs at the elbow because of the asymmetric growth rates between volar and dorsal

t­ issues. The interzone of the shoulder develops a true ­layered configuration with a chondrified layer on either side of a loose layer of central cells. Bone formation occurs in the primary ossification center within the diaphysis of the humerus. The scapula enlarges and extends from C4 to T7.3 In the seventh week, the upper limbs rotate laterally through 90 degrees on their longitudinal axis with the elbows and the extensor muscles facing laterally and posteriorly while the radius assumes a lateral position. The shoulder joint is now well formed. The scapula descends to its position between the first and fifth ribs.7 By the eighth week of gestation, the musculature of the upper extremity has developed, and the shoulder joint has progressed into a fully formed adult-like glenohumeral joint with distinct capsular ligamentous thickenings.6 It is during this embryonic period that the rapidly differentiating upper limbs are most vulnerable to the effects of certain toxins and environmental agents. Exposure to these factors can lead to congenital deformities of the upper extremity and shoulder joint.8,9 The fetal period, the time from 8 weeks of gestation to birth, mainly involves the enlargement of the structures differentiated and developed during the embryonic period. The primary center of ossification for most long bones appears in the diaphysis of the bone between the 7th and 12th weeks of development.2 From the 12th to the 16th weeks, the physes and epiphyses appear and continue to develop throughout the fetal period. The muscular, tendinous, and ligamentous structures around the shoulder become distinct by the 13th week and continue to mature throughout the gestational period.6,10-12 The shoulder joint complex is fully formed by the time of birth. Postnatal development through childhood consists of further growth and enlargement with maturation of the bony, cartilaginous, musculotendinous, and ligamentous tissues. Several factors involving this rapid growth of the musculoskeletal system distinguish the pediatric athlete from the adult.13

ANATOMY Osseous Anatomy Clavicle The clavicle forms by intramembranous ossification during the fifth gestational week from two different areas in the diaphyseal portion of the bone. The medial physis of the clavicle is the most important, providing up to 80% of the remaining longitudinal growth of this bone (Fig. 17A2-3). Interestingly enough, the medial epiphysis is one of the last to ossify, with appearance of its secondary ossification center between 12 and 19 years of age and fusion to the shaft of the clavicle at age 22 to 25 years. The lateral clavicular epiphysis is usually inapparent radiographically, appearing, ossifying, and then fusing during a period of a few months at about 19 years of age.11,14-16 This is significant during evaluation and treatment of distal clavicular injuries in the adolescent because these injuries are ­usually physeal fractures as opposed to true acromioclavicular ­dislocations.17 The clavicle is subcutaneous in position and extends from the sternoclavicular joint medially to the ­acromioclavicular

Shoulder

781

Coracoid tip Multiple acromial centers

Common glenoid-coracoid

Primary Secondary Figure 17A2-3   Radiograph of the clavicle in a child with inapparent physes medially and laterally.

joint laterally. It is S shaped when viewed from above with a more cylindrical configuration medially and a somewhat flattened and narrow shape laterally. The clavicle provides attachment for many of the major shoulder girdle muscles, including the trapezius, deltoid, sternocleidomastoid, and pectoralis major muscles. It also provides a bony roof over the thoracic outlet protecting the axillary vessels and brachial plexus. The clavicle is capable of motion in multiple planes. Most of this motion occurs through the sternoclavicular joint and includes rotation, translation, and an ability to pivot anterior to posterior as well as superior to inferior. When the shoulder is taken through a full range of motion, the clavicle rotates about its longitudinal axis about 50 degrees and is elevated upward about 30 degrees. Only a small amount of motion occurs laterally through the acromioclavicular joint.18

Scapula The scapula is a large, flattened, triangular bone positioned at the posterolateral aspect of the bony thorax situated about between the third and ninth ribs. The scapula provides a framework for attachment of many of the major muscles about the shoulder and a mobile base for the glenohumeral joint at the glenoid. It has five major components—the body, neck, spine, glenoid, and coracoid. The scapula first appears as chondrified anlagen in the fifth gestational week. It begins at the level of C4-C5, and then after formation of the shoulder joint in the seventh week, the scapula descends from the cervical area to its more adult-like position overlying the first through fifth ribs. Failure of the scapula to descend to its normal adult position results in Sprengel’s deformity.8 The body of the scapula forms by intramembranous ossification throughout its primary center, which is usually completely ossified by birth. The remaining multiple ossification centers are highly variable in terms of number and position (Fig. 17A2-4). At about 1 year, an ossification center for the coracoid process appears. By 10 years of age, a common physis appears for the base of the coracoid and upper glenoid. A third, somewhat variable ossification center can appear at puberty at the tip of the coracoid and may be misidentified as an avulsion fracture. By the age of

Scapula Figure 17A2-4   The multiple ossification centers of the scapula.

15 to 16 years, these three centers usually coalesce. The acromion ossifies by forming between two and five ossification centers. These usually appear by puberty and are completely fused by the age of 22 years. Failure of fusion of one of the acromial physes results in an unfused os acromiale, which may have clinical implications in impingement (Fig. 17A2-5).19-21 At puberty, the center for the vertebral border and inferior angle of the scapula and a horseshoe-shaped epiphysis for the lower three fourths of the glenoid appear. They fuse to the remaining scapula by the 22nd year. Because of the multitude of centers of ossification, many anomalies of the scapula have been described, including bipartite coracoid, duplication of the acromion process, dysplasia of the glenoid, and scapular clefts.11,14,22 The body of the scapula is oriented at a 30- to 45-degree angle to the coronal plane of the body. It is somewhat

Figure 17A2-5   Axillary radiograph of the shoulder demonstrating an unfused os acromiale.

782

DeLee & Drez’s Orthopaedic Sports Medicine

c­ oncave on its costal surface with slight convexity on the dorsal surface. Dorsally, the body of the scapula is divided by a thin rigid bone known as the spine of the scapula. It separates the dorsal aspect of the scapula into the supraspinatus and infraspinatus fossae. At the lateral edge of the scapular spine, these two fossae communicate by way of the spinoglenoid notch. The acromion is an extension of the spine of the scapula that rotates to form a flattened roof above the shoulder joint. The acromion is a flattened structure with some variability in terms of its angle of inclination and morphologic features.23 Bigliani has described this morphologic appearance as one of three types.24,25 The type I acromion has a relatively high angle and flat undersurface. The type II acromion has a downward curve and a decreased angle of inclination. The type III acromion has a hooked configuration along the anterior portion and a further reduction of the angle of inclination. The lateral portion of the scapular body narrows to form the scapular neck, which supports the glenoid fossa. The glenoid is a concave, pear-shaped structure covered with articular cartilage that is oriented laterally at approximate right angles to the long axis of the scapular body. The glenoid has an average of 5 degrees of superior tilt and retroversion of 3 to 9 degrees in relationship to the long axis of the scapula.26-28 The coracoid process is a bony projection off the anterior surface of the scapula just medial to the scapular neck. It projects anteriorly and laterally and has a hooked configuration. It serves as the origin of several muscles and ligaments that provide stability at the acromioclavicular joint. Superiorly and medial to the coracoid is the supraspinous notch, which contains the suprascapular nerve.

Figure 17A2-6   Radiograph demonstrating the single proximal humeral ossification center. This physis usually closes between 19 and 22 years of age.

known as the greater tuberosity. Anterior and medial to the bicipital groove is a smaller prominence of bone known as the lesser tuberosity. The tuberosities serve as the insertion for the tendons of the rotator cuff muscles. The anatomic neck is that space between the articular cartilage and the ligamentous and tendinous attachments on the tuberosities. The surgical neck is that portion of the proximal humerus that lies below the tuberosities and above the metaphysis.27,30-32

Articulations

Proximal Humerus

Glenohumeral Joint

The proximal humerus consists of four main components: humeral head, greater tuberosity, lesser tuberosity, and metaphyseal portion of the shaft. At birth, the humerus is completely ossified throughout its diaphysis and metaphyseal portions. The secondary ossification center for the humeral head is rarely ossified until after the first 6 postnatal months.2 The secondary ossification center for the greater tuberosity appears between the seventh month and third year of age, whereas the secondary center for the lesser tuberosity appears about 2 years after the appearance of the greater tuberosity. By the age of 5 to 7 years, the three proximal ossification centers of the humeral head, greater tuberosity, and lesser tuberosity coalesce to become a single proximal ossification center (Fig. 17A2-6). This proximal humeral physis usually closes between 19 and 22 years of age and accounts for about 80% of longitu­dinal growth of the humerus postnatally.11,14 The humeral head has a large convex oval shape covered with hyaline cartilage that articulates with the glenoid. The head forms an upward head-shaft angle between 130 and 140 degrees and is in 25 to 30 degrees of retroversion as it relates to the humeral epicondyles.29 With the arm in the anatomic position, the intertubercular or bicipital groove is anterior and allows access for the long head of the biceps tendon into the shoulder joint. Immediately distal to the articular surface and posterior to the intertubercular groove is a prominent projection of bone

The glenohumeral articulation is a true synovial joint between the humeral head and the glenoid. It is the most mobile major joint in the body. There are several unique anatomic features of this joint that accommodate motion while sacrificing inherent stability. First, there is minimal bony constraint because the articular surface area of the humeral head is up to 3 times greater than the relatively small articular surface of the glenoid. At any point in time, only 25% to 30% of the humeral head is in contact with the glenoid articular surface.33 Second, the radius of curvature of the glenoid is greater than that of the humeral head. The glenoid labrum and the thicker glenoid peripheral articular cartilage do deepen the socket and provide some stability for the joint by decreasing this relative difference in radius of curvature.34 However, this mismatch still allows for inherent increased translation of the humeral head on the glenoid.35 The capsule of the glenohumeral joint is reinforced by thickened areas known as the glenohumeral ligaments.36 These primary static stabilizers attach to the glenoid circumferentially through the labrum. The labrum is a wedge-shaped structure composed of densely packed ­collagen tissue attached at the periphery of the glenoid.34 The humeral attachment of the capsular ligaments occurs along the region of the anatomic neck except medially, where the attachment extends distally along the shaft. The proximal humeral physis lies in an extracapsular position

Shoulder Capsular attachment

Acromioclavicular ligaments Thick periosteal tube

Acromion C

Physeal plate

Figure 17A2-7   The glenohumeral joint capsule and its relationship to the proximal humeral physis. Note that the physis is predominantly extra-articular except on the medial side, where it becomes an intra-articular structure.

except along this medial side, where it is intra-articular (Fig. 17A2-7). These capsular ligamentous bands have been more precisely defined in the anterior and posterior capsule as the superior, middle, and inferior glenohumeral ligaments. Together they form a complex functional unit that provides static restraint for the glenohumeral joint. With the arm adducted at the side, the inferior capsule and ligaments are highly redundant, but as the arm is taken through a range of abduction, elevation, flexion, or extension, these ligaments sequentially tighten to provide stability for the joint.37-40 With the arm in the anatomic position, the intertubercular groove between the greater tuberosity and the lesser tuberosity is positioned about 1 cm lateral to the midline.41 This groove is covered by a transverse ligament and permits access for the long head of the biceps into the joint. The long head of the biceps attaches superiorly to the glenoid rim and labrum as a prominent intra-articular structure. The biceps may function to help stabilize the joint by tensioning the labrum to resist superior translation of the humeral head. The rotator cuff tendons form a sleeve of thickened tissue that covers the joint anteriorly, posteriorly, and superiorly. The tendinous contributions from the subscapularis muscle attach to the lesser tuberosity; the supraspinatus, infraspinatus, and teres minor tendons coalesce to form a posterior and superior sleeve of tissue that attaches to the greater tuberosity. The rotator cuff inserts immediately adjacent to the insertion of the capsular ligaments and functions in both static and dynamic stability.

Acromioclavicular Joint The acromioclavicular joint is formed by the lateral end of the clavicle and its articulation with the medial aspect of the acromion. The clavicle becomes more flattened in its outer third and is surrounded by an extremely thick periosteal tube. This periosteum is continuous laterally with the acromioclavicular ligaments that span the acromioclavicular joint and is continuous inferiorly with the coracoclavicular ligaments that provide stability for the distal clavicle

783

Coracoclavicular ligaments

HH

Figure 17A2-8   The distal clavicle and its relationship to the acromioclavicular joint in the immature patient. Note the thickened periosteal tube surrounding the distal clavicle, which is continuous with the acromioclavicular and the coracoclavicular ligaments. C, coracoid; HH, humeral head.

(Fig. 17A2-8). The acromioclavicular joint is a diarthrodial joint stabilized primarily by the strong coracoclavicular ligaments that extend from the coracoid to the undersurface of the distal third of the clavicle. There are two portions to the coracoclavicular ligament, the conoid and trapezoid, that begin at the level of the acromioclavicular joint laterally and extend medially for about 3 cm along the clavicle.42 In the mature individual, an intra-articular disk covers the end of the distal clavicle and protects the two adjacent incongruous surfaces.43,44 In addition, the acromioclavicular joint is stabilized and protected by strong muscular attachments of both the deltoid and trapezius muscles. The deltoid attaches all along the anterior aspect of the distal clavicle and anterior acromion, whereas the trapezius muscle inserts posteriorly on the distal clavicle. These multiple muscular and ligamentous attachments provide relative protection to the acromioclavicular joint and distal clavicle. In children, fracture occurs more commonly through the exposed clavicular shaft as opposed to the distal portion. Also in children and adolescents, injury to the acromioclavicular joint is actually fracture through the distal physis with splitting of the periosteal tube rather than a true dislocation.45-47

Sternoclavicular Joint The sternoclavicular joint is a diarthrodial joint composed of the large medial end of the clavicle, the sternum, and the first rib (Fig. 17A2-9). The joint surfaces are relatively flat and extremely incongruous, with little inherent bony stability. A fibrocartilaginous disk provides further cushioning and stability for this joint, which is surrounded by a strong series of ligaments.48 The anterior and posterior capsular ligaments provide the major support. The posterior portion of the capsular ligament is stronger and heavier and provides the primary support against downward displacement of the lateral clavicle. These strong ligaments attach primarily to the epiphysis of the medial clavicle, which helps to explain why medial clavicular physeal injuries in children are more common than true sternoclavicular dislocations.49 In addition to the capsular ligaments, the

784

DeLee & Drez’s Orthopaedic Sports Medicine Medial physis

Costoclavicular ligament

Cla vicl

Interclavicular ligament

e

Clavicle Spine of scapula

1st rib

Sternum

Deltoid

Anterior sternoclavicular ligament

Figure 17A2-9   The medial clavicle and its relationship to the sternum and the first rib through the sternoclavicular joints. Note the complex series of ligaments and the extra-articular position of the medial physis of the clavicle.

intra-articular disk ligament is a dense fibrous structure extending from the first rib through the joint and attaching through the disk to the strong capsular ligaments. Further stability for the joint is provided by the interclavicular ligament and costoclavicular ligaments. The interclavicular ligament runs from clavicle to clavicle, attaching to the superior aspect of the manubrium. The costoclavicular ligaments run from the first rib to the inferior surface of the medial clavicle. They help suspend the clavicle much like the boom of a crane. Most clavicular motion occurs through the sternoclavicular joint. This joint has the capability of allowing 30 to 35 degrees of upward clavicular elevation (pivot), 35 degrees of anterior to posterior glide (translation), and 45 to 50 degrees of rotation about the long axis of the clavicle. This motion is extremely important for normal shoulder function as the scapula rotates to allow normal abduction and elevation of the arm. In addition to its important function in mobility, this joint provides the only true bony articulation between the upper extremity and the axial skeleton. Its position anterior to the mediastinum also gives it a protective function. In posteriorly displaced fractures and dislocations around the medial clavicle and sternoclavicular joint, injury can occur to the major neurovascular structures exiting the mediastinum.47

Scapulothoracic Articulation The scapulothoracic articulation does not represent a true joint in the strictest sense. This articulation between the scapula and the thorax is surrounded by heavy musculature that allows a gliding action of the scapula on the posterior thorax. It is through this action that the scapula provides a mobile base for the glenohumeral joint to complete the motion of elevation.50,51

Muscles Deltoid The deltoid is the large muscle that defines the external contour of the shoulder. It is one of the most important of the glenohumeral muscles, providing the primary motor power for the glenohumeral joint. It originates from the lateral third of the clavicle, the acromion, and the scapula

Acromion

Tendons of origin

Tendons of insertion

Shaft of humerus

Figure 17A2-10   The deltoid muscle, with its origin on the clavicle, the acromion, and the scapula and its insertion into the deltoid tuberosity distally along the anterolateral aspect of the humerus.

and inserts into the deltoid tuberosity along the anterolateral aspect of the proximal humerus (Fig. 17A2-10). The deltoid receives its innervation from the axillary nerve that approaches the muscle posteriorly through the quadrilateral space to track on the underside of the muscle from posterior to anterior. The deltoid functions in elevation in the scapular plane through the action of the anterior and middle thirds. Abduction in the coronal plane decreases the contribution of the anterior third and increases the contribution of the posterior third. Flexion is a product of the anterior and middle thirds of the deltoid, whereas extension involves the posterior and middle thirds. The deltoid is active in any form of elevation, therefore, and loss of deltoid function is significant.1

Rotator Cuff The rotator cuff consists of four muscles: the supraspinatus, the infraspinatus, the teres minor, and the subscapularis (Fig. 17A2-11). They originate on the scapula and insert through a combined tendinous insertion into the greater and lesser tuberosities of the humerus. The rotator cuff is important in providing dynamic stability for the glenohumeral joint by resisting shear forces produced by the action of other muscles.1,52,53

Shoulder THE ROTATOR CUFF Supraspinatus Infraspinatus Teres minor

785

Clavicular portion

Pectoralis major muscle

Subscapularis

Sternal portion

Back Costal portion

Front Figure 17A2-11   The muscles and tendons of the rotator cuff. These muscles include the supraspinatus, the infraspinatus, the teres minor, and the subscapularis.

The supraspinatus muscle arises from the supraspinatus fossa and passes laterally under the coracoacromial arch, inserting into the greater tuberosity just posterior to the bicipital groove.54 It is innervated by the suprascapular nerve. This muscle is active in any motion involving elevation at the shoulder. It exerts maximal effort at about 30 degrees of elevation and is the center for the headdepressing effect of the rotator cuff.55-57 The infraspinatus muscle arises from the infraspinous fossa and travels laterally to insert on the posterior aspect of the greater tuberosity.54 It is innervated by the suprascapular nerve. It is one of the two main external rotators of the humerus and accounts for as much as 60% of external rotation force. It also functions as a glenohumeral head depressor through its common attachment with the remaining rotator cuff. The teres minor originates from the middle portion of the lateral border of the scapula and the dense fascia of the infraspinatus.54 It inserts into the lower posterior aspect of the greater tuberosity. It is innervated by a posterior branch of the axillary nerve. The teres minor also functions as an external rotator at the shoulder. The subscapularis muscle arises from the costal surface of the scapula and converges to insert on the lesser tuberosity of the humerus.54,58 It is innervated by the subscapular nerve. It functions as an internal rotator and passive stabilizer against anterior subluxation and serves in its upper fibers to depress the humeral head.

Pectoralis Major The pectoralis major consists of three portions. The upper portion originates on the medial one half to two thirds of the clavicle and inserts along the lateral lip of the bicipital groove distal to the subscapularis. The middle portion originates from the manubrium and upper two thirds of the body of the sternum and ribs two through four and inserts directly behind the clavicular portion. The inferior portion originates from the distal body of the sternum, the fifth and sixth ribs, and the external oblique muscle fascia and inserts along with the other two portions into the humerus by rotating through 90 degrees (Fig. 17A2-12).

Figure 17A2-12   The relationship of the three portions of the pectoralis major: the clavicular portion, the sternal portion, and the costal portion.

The pectoralis major is innervated by the lateral pectoral nerve (C5, C6, and C7) to the clavicular portion and by the medial pectoral nerve (C8 and T1) to the remaining portion. The muscle is a powerful adductor of the shoulder but also participates in flexion, extension, and internal rotation, depending on arm position.1

Latissimus Dorsi The latissimus dorsi muscle is a very large and powerful muscle that originates through the lumbar aponeurosis to the spinous process of T6 to the sacrum, the posterior iliac crest, and the lower four ribs. It passes upward and deep to the humerus to attach along the bottom of the intertubercular groove of the humerus. The muscle is innervated by the thoracodorsal nerve. The latissimus dorsi muscle extends, adducts, and rotates the arm medially, drawing the shoulder downward and backward. It is important in throwing and climbing activities.1

Biceps Brachii The biceps originates proximally by two heads: the long head from the supraglenoid tubercle and labrum within the shoulder joint, and the short head from the tip of the coracoid with the coracobrachialis. It inserts distally into the bicipital tuberosity of the radius. Innervation of the biceps is through the musculocutaneous nerve (C5, C6). The biceps functions primarily in flexion and supination at the elbow. It has a secondary humeral head-depressing effect at the shoulder.1

Scapular Stabilizers Trapezius This muscle originates from the spinous processes of the C7 through T12 vertebrae, the ligamentum nuchae, and the external occipital protuberance. Insertion is over the distal third of the clavicle, the acromion, and the spine of the scapula (Fig. 17A2-13). It is innervated by the ­accessory

786

DeLee & Drez’s Orthopaedic Sports Medicine

BIOMECHANICS Trapezius muscle

Levator scapulae muscle Rhomboideus minor muscle Rhomboideus major muscle

The primary function of the shoulder is to position the arm and hand in space to carry out specific sports activity. A multiplanar range of motion with strength and stability is required to accomplish this task. To accomplish successful function of the shoulder girdle, the integrated motion of all four joints—sternoclavicular, acromioclavicular, glenohumeral, and scapulothoracic—is necessary. The acromioclavicular and sternoclavicular joints provide a stable strut that supports the action of the more mobile glenohumeral and scapulothoracic articulations. This complex interaction of four joints depends on precise function of articular surfaces, ligaments, and muscle-tendon units to provide coordination for shoulder mobility.59

Sternoclavicular Joint Figure 17A2-13   The posterior periscapular musculature, including the trapezius, the levator scapulae, and the rhomboids.

spinal nerve (cranial nerve XI). The nerve runs parallel and medial to the vertebral border of the scapula in the medial 50% of the muscle. The trapezius acts as a scapular ­retractor.1

Rhomboids The rhomboids originate from the lower ligamentum nuchae, C7 and T1 for the rhomboid minor and T2 through T5 for the rhomboid major. They insert on the posterior portion of the medial base of the spine of the scapula and into the posterior surface of the medial border to the inferior angle of the scapula (see Fig. 17A2-13). Innervation to the rhomboid muscles is by the dorsal scapular nerve (C5). The action of the rhomboids is retraction of the scapula with some participation in elevation of the scapula.

Levator Scapulae This muscle originates from the posterior tubercle to the transverse processes of the first through fourth cervical vertebrae and inserts into the superior angle of the scapula and along the medial border of the scapula to about the level of the scapular spine (see Fig. 17A2-13). The innervation of this muscle is from the cervical plexus and occasionally the dorsal scapular nerve. The levator acts to elevate the superior angle of the scapula, and in conjunction with the serratus anterior, it produces upward rotation of the scapula.1

Serratus Anterior The serratus anterior originates from the outer surface of the first eight ribs and follows the curvature of the ribs to insert along the medial aspect of the scapula on its costal surface. It is innervated by the long thoracic nerve. The serratus protracts the scapula and participates in upward rotation. Absence of serratus activity produces a winging of the scapula with forward flexion of the arm.1

The sternoclavicular joint is a shallow, relatively incongruous joint supported by a strong ligamentous complex. The sternoclavicular joint is oriented somewhat posterior, lateral, and upward. It contains an intra-articular disk or meniscus that helps provide congruity and stability.60 Four main ligamentous groups provide support for the sternoclavicular joint. The anterior capsule is supported by anterior sternoclavicular ligaments, which provide support against anterior translation of the joint. Posteriorly, the capsule is supported by the posterior sternoclavicular ligament and the costoclavicular ligament that extends from the first rib to the clavicle. The posterior capsular ligament is the most important in supporting the shoulder against downward-oriented forces. The interclavicular ligaments extend from clavicle to clavicle across the superior manubrium and provide superior constraint. The costoclavicular ligaments run from the first rib to the inferior surface of the medial clavicle.61 With the arm adducted to the side, these ligaments are tightened; with the arm elevated, they are lax. Most clavicular motion occurs through the sternoclavicular joint. Motion at the sternoclavicular joint includes about 30 to 35 degrees of upward rotation (pivot), 35 degrees of anterior to posterior glide (translation), and up to 45 to 50 degrees of axial rotation. Stability at the joint is provided by tightening of ligaments opposite the direction of motion (Fig. 17A2-14).

Acromioclavicular Joint The articular surfaces of the acromioclavicular joint are not perfectly congruent and are supported by an intra­articular disk or meniscus.60 This “plane-type” joint is oriented somewhat posterior to the perpendicular from the coronal plane. Ligamentous stability is provided by the acromioclavicular capsular ligaments and the more important coracoclavicular ligaments.43 The coracoclavicular ligaments consist of two components, the conoid ligament and the trapezoid ligament. The clavicle moves at the acromioclavicular joint through about 30 degrees of elevation (see Fig. 17A2-14). Joint motion includes anterior to posterior translation, inferior to superior rotation, and compression. The coracoclavicular ligaments are the primary restraints to displacement in both an anterior to posterior and a superior

Shoulder

AC

Upward rotation 30°-35°

Elevation 30° SC

Anterior to posterior glide 35°

Humeral head

787

Glenoid

Labrum

Axial rotation 45°-50°

Figure 17A2-16   The humeral head and its relationship to the glenoid. The glenoid socket is deepened by the glenoid labrum, which provides both congruity and increase in contact area.

Figure 17A2-14   Motion of the clavicle occurs predominantly through the sternoclavicular (SC) joint, including 30 to 35 degrees of upward rotation, 35 degrees of anteriorto-posterior glide, and 45 to 50 degrees of axial rotation. The clavicle moves at the acromioclavicular (AC) joint through an arc of about 30 degrees of elevation.

to inferior direction. Compared with the sternoclavicular joint, motion at the acromioclavicular joint provides only a small contribution to overall shoulder motion.61,62

Glenohumeral and Scapulothoracic Joint Motion The humeral head is composed of 120 degrees of the arc of a sphere. It is inclined superiorly in relation to the shaft, with a neck-shaft angle of 130 to 140 degrees. The articular surface is retroverted in relation to the transepicondylar axis of the humerus about 30 degrees (Fig. 17A2-15).63 The glenoid surface has been described as pear shaped. In the superior to inferior plane, it represents 75 degrees of an arc with a length of about 3.5 to 4 cm. In an anterior to posterior plane, it describes a 50-degree arc with a length of 2.5 to 3 cm. It is retroverted 7 degrees in relation to the scapular axis, and there is a 5-degree superior tilt. The glenoid labrum attaches circumferentially to the rim of the

35°

Figure 17A2-15   The relationship between the glenoid, the humeral head, and the coronal axis of the humerus. The humeral head is retroverted about 35 degrees in relation to the transepicondylar axis of the humerus.

glenoid and deepens the socket. It has been found that the labrum along with thicker peripheral articular cartilage increases the relative contact of the humeral head with the glenoid from about one fourth to one third.34,64 Saha has described the glenohumeral ratio as 0.8 in the coronal plane and 0.6 in the horizontal plane (Fig. 17A2-16).65 The range of motion at the shoulder complex is greater than at any other joint in the body. By convention, any upward motion of the shoulder, whether in forward flexion, abduction, or extension, is defined as elevation. The shoulder can be carried through an arc of elevation from 0 to 180 degrees. In addition, there is an internal rotation–to–external rotation arc of about 150 degrees. There is a horizontal plane range of motion in adduction and abduction of about 170 degrees. Although complementary motion in the sternoclavicular and acromioclavicular joints is necessary, most shoulder motion occurs through the glenohumeral and scapulothoracic joints.63,66,67 The relative contribution of the glenohumeral joint and scapulothoracic articulation to overall shoulder motion has been well studied.68,69 This motion is described as scapulothoracic rhythm. The average ratio of glenohumeralto-scapulothoracic motion is about 2:1. There is general agreement that in the first 30 degrees of abduction, there is a much greater contribution from the glenohumeral joint to this range of motion. There is a variably decreasing contribution from the glenohumeral joint beyond 30 degrees of abduction and an almost equal contribution between glenohumeral and scapulothoracic joints during the last 60 degrees of elevation (Fig. 17A2-17). An obligatory external rotation of the humerus occurs to allow complete elevation. This external rotation of the humerus clears the greater tuberosity from the coracoacromial arch and loosens the inferior capsular ligamentous structures to allow full elevation.63,70-74 The glenohumeral joint itself undergoes three types of motion at the joint surface. Spinning is rotation of the humeral head at a single instant center of rotation on the glenoid. Sliding is pure translation of the humeral head on the glenoid with a change in the instant center of rotation. Rolling is motion between the rotating humeral head and glenoid in which the instant center also changes. At any given time, only about one third of the humeral head is in contact with the glenoid. This is maximized

788

DeLee & Drez’s Orthopaedic Sports Medicine 180°

180°

120° GH 90°

60° GH 60° GH

30° 0° ST

30° ST

0°

A

B

ST

C

Figure 17A2-17   A-C, The average ratio of glenohumeral (GH) to scapulothoracic (ST) motion is about 2:1. For the first 30 degrees of abduction, it is all glenohumeral motion. In the last 60 degrees of elevation, there is an almost equal contribution between the glenohumeral and the scapulothoracic joints.

with the ­shoulder at a functional position between 60 and 120 degrees of elevation.63,75,76 The instant center of rotation within the humeral head has a small amount of upward translation in the first 30 degrees of elevation equal to about 3 mm. Above 30 degrees, there is a further increase in elevation of about only 1 mm. With regard to translation in an anterior to posterior plane, this is relatively small, measuring 1 to 2 mm through normal shoulder motion. Pathologic states such as cuff deficiency increase superior translation, and in cases of shoulder instability, there is increased anterior to posterior translation.63,77,78

Static Glenohumeral Stability Complementary version of the humeral head and glenoid contributes to the static stability of the glenohumeral joint. The 30-degree retroversion of the humeral head in relation to the humeral transepicondylar axis matches the amount of glenoid version with the scapula positioned at 30 to 45 degrees on the thorax in the coronal plane.79 In addition, the 5-degree superior tilt of the glenoid has also been recognized as a static stabilizer against inferior instability. Although the glenohumeral joint is a relatively shallow ball-and-socket type of joint, the conformity of the articular surfaces does provide some static stability. The glenohumeral articulation has been traditionally described as a mismatch, with the radius of curvature at the glenoid surface being greater than the radius of curvature at the humeral head. Recently, however, it has been recognized that the functional radius of curvature of the glenoidlabral complex is nearly equal to that of the humeral head. The labrum functions as a static stabilizer by deepening the socket, providing as much as 50% of its depth, and by increasing the surface area for contact with the humeral head. The labrum also functions as a buttress to humeral

head translation and provides an attachment point for the glenohumeral ligaments and long head of biceps tendon.34,80-82 This is more consistent with studies that reveal only a small degree of anterior translation with forward flexion and a small degree of posterior translation with extension.69 Another factor with regard to static stability is the ­normal negative intra-articular pressure that is present within the closed glenohumeral joint capsule. Many studies have shown that venting the capsule leads to an immediate increased translation both in the anterior to posterior plane and in the superior to inferior plane. The magnitude of this effect appears to depend on arm position.83,84 The glenohumeral joint capsule provides the primary static constraint for the shoulder joint. It is a continuous layered capsule of collagen fiber bundles with various thicknesses and orientations. With the advent of arthroscopy, better definition of the anatomy of the glenohumeral joint capsule has been obtained (Fig. 17A2-18).85 Four major ligamentous structures are found within the capsule: the coracohumeral ligament, superior glenohumeral ligament, middle glenohumeral ligament, and the important inferior glenohumeral ligament complex.86,87 The coracohumeral ligament extends from the lateral base of the coracoid as two separate bands attaching to the greater tuberosity and lesser tuberosity in the region of the rotator interval. In the adducted arm, it provides a primary restraint to inferior translation as well as external rotation.88 When the arm is adducted, flexed, and internally rotated, it also provides restraint to posterior translation.89 The superior glenohumeral ligament is somewhat variable in its size and relative contribution. It runs parallel to the much larger extra-articular coracohumeral ligament from the superior rim of the glenoid adjacent to the biceps insertion to the lesser tuberosity. It functions as a primary restraint to external rotation as well as inferior translation in the adducted arm.88 Similar to the coracohumeral

Shoulder

789

labrum anteriorly, inferiorly, and posteriorly and runs laterally to the humeral head, attaching between the subscapularis and the triceps. It is characterized by substantially thicker areas anteriorly and posteriorly called the superior bands. The inferior glenohumeral ligament is the primary stabilizer of the shoulder joint against anterior and posterior translation with the arm abducted above 60 degrees.90 In the abducted shoulder, external rotation places the inferior glenohumeral ligament in a more anterior position to resist anterior translation; in the internally rotated shoulder, the inferior glenohumeral ligament is positioned posteriorly to resist posterior translation (Fig. 17A2-19).91-94

Dynamic Glenohumeral Stability

Figure 17A2-18   Arthroscopic view of the glenohumeral ligaments. G, glenoid; H, humeral head; I, inferior glenohumeral ligament; M, middle glenohumeral ligament; S, subscapularis tendon.

l­igament, it is a secondary restraint to posterior translation in the adducted, flexed, and internally rotated arm.89 The middle glenohumeral ligament is highly variable, with its origin along the anterior-superior labrum, the scapular neck, or the supraglenoid tubercle and insertion into the lesser tuberosity along with the subscapularis tendon. This ligament can be either a distinct cord-like band or a sheet-like structure that blends with the inferior glenohumeral ligament. In ligament cutting studies, it has been shown to be a primary stabilizer against anterior translation when the arm is abducted up to 45 degrees. It is also thought to be important in limiting external rotation above 60 degrees of abduction. The inferior glenohumeral ligament complex is an important structure that has been described as having a hammock-like appearance. It extends from the glenoid

Dynamic glenohumeral stability is a product of several factors that combine to enhance the static stability of the joint. Active contraction of the shoulder musculature compresses the humeral head into the glenoid, providing increased joint reactive forces that resist translation. This so-called concavity-compression mechanism produced by the muscular activity on congruent articular surfaces may be as important in the midrange of motion as any of the static capsular constraints in stabilizing the glenohumeral joint.95-97 The dynamic effect of the rotator cuff as an active counterbalancing restraint against the shear forces resultant to the muscular power of the deltoid, pectoralis, and latissimus dorsi has been well demonstrated (Fig. 17A220).95,98,99 The action of the cuff forms an effective forcecouple relationship with the larger muscles to center the humeral head during shoulder motion.100 Additionally, the muscles of the rotator cuff may also serve in a dynamic way to pretension the glenohumeral ligament complex through the common insertion of glenohumeral joint capsule and rotator cuff at the humerus.101,102 Specialized nerve endings in the rotator cuff and glenohumeral joint capsule provide proprioception, which is another component of dynamic stabilization of the glenohumeral joint. This proprioceptive function allows for effective muscular sequential action during joint motion.103,104

Tight AIGH when abducted and externally rotated Figure 17A2-19   With the shoulder abducted and externally rotated, the inferior glenohumeral ligament becomes the primary restraint to anterior translation of the glenohumeral joint. AIGH, anterior-inferior glenohumeral ligament.

Figure 17A2-20   The rotator cuff acts as a dynamic stabilizer resisting the shear forces exerted on the glenohumeral joint by the larger muscles, including the deltoid, the pectoralis major, and the latissimus dorsi.

790

DeLee & Drez’s Orthopaedic Sports Medicine

The biceps tendon may also play a role in stability. The biceps can increase joint compression, and it has been demonstrated that sectioning of the long head of the biceps increases force across the inferior glenohumeral ligament complex in the abducted and externally rotated arm.105-108 In summary, the glenohumeral joint has a great degree of stable motion because of the complex interrelationship between the articular surfaces, capsular ligamentous structures, and dynamic muscle stabilizers. The glenohumeral ligament complex, particularly the inferior component, is the primary static restraint against anterior and posterior displacement. Translation is also limited by the geometry of the articular surfaces, the labrum, and the normal negative intra-articular pressure. The rotator cuff and biceps exert a dynamic element to stability by counterbalancing the forces of the larger power muscles acting at the ­shoulder.

A

Wind-up/cocking phase

B

Acceleration phase

KINESIOLOGY The primary function of the shoulder is to position the hand in space to carry out prehensile function. The shoulder has evolved into an efficient anatomic structure to produce the wide range of multiplanar motion required to accomplish this task. Unfortunately, stability is sacrificed anatomically to achieve this goal of motion. When the shoulder is taken through a range of elevation, motion occurs primarily at the glenohumeral and scapulothoracic articulations. Secondary supportive motion occurs at the sternoclavicular joint and to a lesser degree at the acromioclavicular joint. Stability of the glenohumeral joint has both static and dynamic components.109 The humeral head is normally located within a 6-mm area on the center of the glenoid in a superior to inferior plane while remaining nearly neutral in an anterior to posterior translation plane during motion. Synergy between the rotator cuff and the deltoid is a key to normal function of the shoulder. The deltoid has a mobile point of origin on the acromion, clavicle, and scapula that allows the muscle to always function within its most efficient length-tension curve. At the initiation of abduction, the long lever arm of the deltoid applies a force across the shoulder with a tendency to cause upward displacement or shear force across the humeral head. This force exerted by the deltoid becomes a more compressive force once the shoulder is abducted above about 45 degrees. All four components of the rotator cuff work in concert to help improve deltoid function by serving as a humeral head depressor and stabilizer against this superior shear force that is created. The rotator cuff is therefore extremely important in providing dynamic stability in its force-couple relationship with the deltoid.110,111 The biomechanics and kinesiology of the throwing motion have been studied extensively by high-speed cinematography and by electromyographic analysis, providing a tremendous amount of information concerning the kinematics of the shoulder.112-117 The demand on the shoulder during throwing is intense in terms of both the extremes in range of motion required and the large muscle forces involved (Fig. 17A2-21). The kinesiology of the shoulder during the throwing motion shows many similarities to kinematics found in other overhand sports activities, such as the tennis serve, volleyball spike, javelin throw, and

C

Follow-through phase

Figure 17A2-21   The baseball throw has been studied extensively and has become a model for the study of biomechanics and kinesiology at the shoulder. The three phases of pitching include wind-up/cocking phase (A), acceleration phase (B), and follow-through phase (C).

f­ reestyle swimming stroke, and it can therefore be used as a model.118,119 The pitch has been broken down into three major phases of action: the wind-up and cocking phase, the acceleration phase, and the follow-through phase. The wind-up and cocking phase begins as the arm is elevated to about 90 degrees and then is horizontally extended or abducted. With these motions, the shoulder is externally rotated, maximally tightening the anterior-inferior glenohumeral ligament complex. The subscapularis tendon begins as both a dynamic and static stabilizer, but as the wind-up and cocking phase is completed, it is elevated into a more superior position, decreasing its effectiveness. This phase is complete when the shoulder reaches its maximal point of external rotation. Electromyographic studies have shown that deltoid activity is high during this phase. The posterior deltoid serves as the prime source of extension, whereas the anterior and lateral components of the deltoid provide elevation and some external rotation. The supraspinatus, infraspinatus, and teres minor act sequentially throughout external rotation and abduction. With their combined stabilizing effect, they provide relief from anterior to posterior translation across the joint. Adequate strength in the rotator cuff musculature in this phase is of critical importance.120 The acceleration phase consists of a rapid unwinding of the potential energy stored in the soft tissues as they are positioned in the wind-up and cocking phase. Initiation of the acceleration phase occurs as the subscapularis, pectoralis major, and latissimus dorsi begin internal rotation. This phase is brief, lasting only 0.01 second. Peak rates of internal rotation of up to 7000 degrees per second

Shoulder

have been measured. Although this represents a relatively passive activity from the standpoint of electromyographic activity, tremendous rotational stress is applied to the glenohumeral joint soft tissues.120 The follow-through phase consists of the shoulder’s final continuation of internal rotation and horizontal adduction after the ball leaves the hand. Many muscle groups are active during this phase, including the posterior deltoid, rotator cuff, biceps brachii, latissimus dorsi, and pectoralis major. These muscles fire sequentially in an attempt to decelerate the arm during this phase. Stress across the posterior capsule is at its peak.120 The pitching cycle has helped to reveal the complex function and interaction of joints, muscles, and tendons required to support normal shoulder activity. Further study of highly sport-specific activities, such as pitching, should enable us to continue to learn more about the kinesiology of the shoulder. This will help in the design and implementation of programs for both prevention and rehabilitation of injuries.

C

r i t i c a l

P

o i n t s

l The rapidly growing physis of the proximal humerus distinguishes the pediatric shoulder from the adult ­shoulder. l The proximal humeral physis is vulnerable to injury through direct and indirect forces applied to the shoulder owing to its predominantly extracapsular position. l The shoulder joint represents a series of three articulations (glenohumeral, acromioclavicular, scapulothoracic) that function together to allow for a wide range of multiplanar motion.

791

l The soft tissue envelope around the glenohumeral joint, including the glenohumeral capsular ligaments and the rotator cuff tendons, provides the primary source of ­stability for the shoulder. l Dynamic shoulder stability involves an effective forcecouple between the rotator cuff and the larger power muscles, including deltoid, pectoralis, and latissimus.

S U G G E S T E D

R E A D I N G S

Fealy S, Rodeo SA, Dicarlo EF, O’Brien SJ: The developmental anatomy of the neonatal glenohumeral joint. J Shoulder Elbow Surg 3:217-222, 2000. Jobe FW, Moynes DR, Tibone JE, Perry J: An analysis of the shoulder in pitching: A second report. Am J Sports Med 12:218-220, 1984. Lee SB, Kim KJ, O’Driscoll SW, et al: Dynamic glenohumeral stability provided by the rotator cuff muscles in the mid-range and end-range of motion: A study in cadavera. J Bone Joint Surg Am 82:849-857, 2000. Lippitt SB, Vanderhooft JE, Harris SL, et al: Glenohumeral stability from ­concavitycompression: A quantitative analysis. J Shoulder Elbow Surg 2:27-35, 1993. O’Brien SJ, Schwartz RS, Warren RF, Torzilli PA: Capsular restraints to anteriorposterior motion of the abducted shoulder: A biomechanical study. J Shoulder Elbow Surg 4:298-308, 1995. Pagnani MJ, Deng XH, Warren RF, et al: Role of the long head of the biceps brachii in glenohumeral stability: A biomechanical study in cadavera. J Shoulder Elbow Surg 5:255-262, 1996. Renfree KJ, Wright TW: Anatomy and biomechanics of the acromioclavicular and sternoclavicular joints. Clin Sports Med 22:219-237, 2003. Roh MS, Wang VM, April EW, et al: Anterior and posterior musculotendinous anatomy of the supraspinatus. J Shoulder Elbow Surg 9:436-440, 2000. Samuelson RL: Congenital and developmental anomalies of the shoulder girdle. Orthop Clin North Am 11:219-231, 1980. Steinbeck J, Liljenqvist U, Jerosch J: The anatomy of the glenohumeral ligamentous complex and its contribution to anterior shoulder stability. J Shoulder Elbow Surg 7:122-126, 1998.

R eferences Please see www.expertconsult.com

S ect i o n

B

Injuries to the Sternoclavicular Joint   in the Adult and Child Charles E. Rosipal, Michael A. Wirth, Igor Cezar da Silva Leitao, and Charles A. Rockwood, Jr.

A review of the early literature on this subject indicates that in the 19th century, dislocations of the sternoclavicular joint were managed essentially the same way as fractures of the medial part of the clavicle.1,2 It was recommended by Sir Astley Cooper in his 1824 text1 that the injury be treated not only with a clavicle bandage but also with a sling “which through the medium of the os humeri and

scapula supports it and prevents the clavicle from being drawn down by the weight of the arm.” Cooper reported that he had never seen an isolated traumatic posterior dislocation of the sternoclavicular joint but suggested that it might occur as a result of excessive force.1 He did, however, describe a posterior dislocation of the sternoclavicular joint in a patient who had such severe

792

DeLee & Drez’s Orthopaedic Sports Medicine

scoliosis that as the scapula advanced laterally around the chest wall, it forced the medial end of the clavicle behind the sternum. Davie, a surgeon in Suffolk, resected the medial end of the clavicle after so much pressure developed on the esophagus that the patient had significant difficulty with swallowing. He must have been a excellent surgeon because in 1824 he resected 1 inch of the medial clavicle with a saw! He protected the vital structures in the area from the saw by placing “a piece of well beaten shoe leather under the bone whilst he divided it.” This case probably represents the first resection of the medial end of the clavicle, either for trauma or arthritis. Rodrigues3 most likely published the first case of traumatic posterior dislocation of the sternoclavicular joint, “a case of dislocation inward of the internal end of the clavicle.” The patient’s left shoulder was against a wall when the right side of the chest and thorax were compressed almost to the midline by a cart. Immediately, the patient experienced shortness of breath, which persisted for 3 weeks. When first seen by the physician, the patient appeared to be suffocating, and his face was blue. The left shoulder was swollen and painful, and there was “a depression on the left side of the superior extremity of the sternum.” Pressure on the depression greatly increased the sensation of suffocation. Rodrigues observed that when the outer end of the shoulder was displaced backward, the inner end of the clavicle was displaced forward, which relieved the asphyxia. Therefore, treatment consisted of binding the injured shoulder backward with a cushion between the two scapulae, but only after the patient had been bled twice within the first 24 hours. Rodrigues may have seen other cases of posterior dislocation because he stated that the patient “retained a slight depression of the internal extremity of the clavicle; such, however, is the ordinary fate of the patients who present this form of dislocation.” In the late 19th century, a number of articles appeared from England, Germany, and France, but it was not until the 1930s that articles by Duggan,4 Howard,5 and ­Lowman6 appeared in the American literature.

Coracoclavicular lig. Subclavius m. Costoclavicular lig. Ant. Sternoclavicular lig.

Tendon of Subclavius m.

Figure 17B-1   Normal anatomy around the sternoclavicular and acromioclavicular joints. Note that the tendon of the subclavius muscle arises in the vicinity of the costoclavicular ligament from the first rib and has a long tendon structure. (Redrawn from Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

Ligaments of the Sternoclavicular Joint There is so much joint incongruity that the integrity has to come from its surrounding ligaments: the intra-articular disk ligament, the extra-articular costoclavicular ligament (rhomboid ligament), the capsular ligament, and the interclavicular ligament.

Intra-articular Disk Ligament The intra-articular disk ligament is a very dense, fibrous structure that arises from the synchondral junction of the first rib to the sternum and passes through the sternoclavicular joint, dividing the joint into two separate joint spaces (Fig. 17B-2).7,8 The upper attachment is on the superior and posterior aspects of the medial clavicle. DePalma10 has

SURGICAL ANATOMY The sternoclavicular joint is a diarthrodial type of joint and is the only true articulation between the clavicle of the upper extremity and the axial skeleton (Fig. 17B-1). The articular surface of the clavicle is much larger than that of the sternum, and in the adult, both are covered with fibrocartilage. The enlarged, bulbous medial clavicle presents a saddle-type joint with the clavicular notch of the sternum.7,8 The clavicular notch of the sternum is curved, and the joint surfaces are not congruent. Cave9 has demonstrated that in 2.5% of patients, the inferior aspect of the medial clavicle has a small facet that articulates with the superior aspect of the first rib at its synchondral junction with the sternum. Because less than half of the medial clavicle articulates with the upper angle of the sternum, the sternoclavicular joint has the distinction of having the least amount of bony stability of any major joint in the body. If a finger is placed in the superior sternal notch, one can feel that with motion of the upper extremity, a large part of the medial clavicle is completely above the superior margin of the sternum.

Figure 17B-2   Cadaveric dissection shows how the intraarticular disk ligament (held in forceps) divides the joint into two separate joint spaces.

Shoulder

793

Interclavicular lig. Ant. sternoclavicular lig.

Joint cavity

Articular disk Synchondrosis of manubrium and Ist rib

Costoclavicular lig. (Rhomboid)

A Articular disk lig.

B Figure 17B-3   A, Normal anatomy around the sternoclavicular joint. Note that the articular disk ligament divides the sternoclavicular joint cavity into two separate spaces and inserts onto the superior and posterior aspects of the medial clavicle. B, The articular disk ligament acts as a checkrein for a medial displacement of the proximal clavicle. (Redrawn from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

shown that the disk is perforated only rarely; such perforation allows free communication between the two joint compartments. Anteriorly and posteriorly, the disk blends into the fibers of the capsular ligament. The disk acts as a checkrein against medial displacement of the inner clavicle (Fig. 17B-3).

Costoclavicular Ligament The costoclavicular ligament, also called the rhomboid liga-­ ment, is short and strong and consists of an anterior and a posterior fasciculus (see Fig. 17B-1).8,9,11 Cave9 reported that the average length is 1.3 cm, with a maximal width of 1.9 cm, and it is 1.3 cm thick. Bearn11 has shown that a bursa is always present between the two components of the ligament. The two different parts of the ligament have a twisted appearance.8 The costoclavicular ligament is attached below to the upper surface of the first rib and at the adjacent part of the synchondral junction with the sternum and above to the margins of the impression on the inferior surface of the medial end of the clavicle, sometimes known as the rhomboid fossa (Fig. 17B-4)7,8 Cave9 has shown from a study of 153 clavicles that the attachment of the costoclavicular ligament to the clavicle can be any of three types: (1) a depression, the rhomboid fossa (30%); (2) flat (60%); and (3) an elevation (10%). The fibers of the anterior fasciculus arise from the anterior medial surface of the first rib and are directed upward and laterally. The fibers of the posterior fasciculus are shorter; they arise lateral to the anterior fibers on the rib and are directed upward and medially. The fibers of

Figure 17B-4   Cadaveric dissection showing the costoclavicular ligament connecting the upper surface of the first rib to the inferior surface of the medial end of the clavicle.

the anterior and posterior components cross to allow for stability of the joint during rotation and elevation of the clavicle. The two-part costoclavicular ligament is in many ways similar to the two-part configuration of the coracoclavicular ligament on the outer end of the clavicle. Bearn11 has shown experimentally that the anterior fibers resist excessive upward rotation of the clavicle and that the posterior resist excessive downward rotation. Specifically, the anterior fibers also resist lateral displacement, and the posterior fibers resist medial displacement.

Interclavicular Ligament The interclavicular ligament connects the superomedial aspects of each clavicle with the capsular ligaments and the upper sternum (see Fig. 17B-3). According to Grant,7 this band may be homologous with the wishbone of birds. This ligament assists the capsular ligaments to produce “shoulder poise,” that is, to hold up the shoulder. This function can be tested by putting a finger in the superior sternal notch; with elevation of the arm, the ligament is lax, but as soon as both arms hang at the sides, the ligament becomes tight.

Capsular Ligament The capsular ligament covers the anterosuperior and posterior aspects of the joint and represents thickenings of the joint capsule (Fig. 17B-5; see Fig. 17B-3). The posterior portion of the capsular ligament is heavier and stronger than the anterior portion and is the most important restraint to anterior and posterior translation of the joint.12 According to the original work of Bearn,11 this may be the strongest ligament of the sternoclavicular joint, and it is the first line of defense against the upward displacement of the inner clavicle caused by a downward force on the distal end of the shoulder. The clavicular attachment of the ligament is primarily onto the epiphysis of the medial clavicle, with some secondary blending of the fibers into the metaphysis. The senior author has demonstrated such

794

DeLee & Drez’s Orthopaedic Sports Medicine

The clavicle, and therefore the sternoclavicular joint, in normal shoulder motion is capable of 30 degrees to 35 degrees of upward elevation, 35 degrees of combined forward and backward movement, and 45 degrees to 50 degrees of rotation around its long axis (Fig. 17B-7). It is likely to be the most frequently moved joint of the long bones in the body because almost any motion of the upper extremity is transferred proximally to the sternoclavicular joint.

Epiphysis of the Medial Clavicle

Figure 17B-5   Cadaveric dissection showing the anterior fibers of the capsular ligament of the sternoclavicular joint.

attachment, as have Poland,13 Denham and Dingley,14 and Brooks and Henning.15 Although some authors report that the intra-articular disk ligament greatly assists the costoclavicular ligament in preventing upward displacement of the medial clavicle, Bearn11 has shown that the capsular ligament is the most important structure in preventing upward displacement of the medial clavicle. In experimental postmortem studies, he evaluated the strength and the role of each of the ligaments at the sternoclavicular joint to see which one would prevent a downward displacement of the outer clavicle. He attributed the lateral “poise of the shoulder” (i.e., the force that holds the shoulder up) to a locking mechanism of the ligaments of the sternoclavicular joint (Fig. 17B-6). To accomplish his experiments, Bearn11dissected all the muscles attaching onto the clavicle, the sternum, and the first rib and left all the ligaments attached. He secured the sternum to a block in a vise. He then loaded the outer end of the clavicle with a 10 to 20 pounds of weight and cut the ligaments of the sternoclavicular joint, one at a time and in various combinations, to determine each ligament’s effect on maintaining poise of the clavicle, that is, which ligament was most important in holding the lateral end of the shoulder up or, on the contrary, which ligament would rupture first when force was applied to the outer end of the clavicle. Bearn determined, after cutting the costoclavicular, intra-articular disk, and interclavicular ligaments, that they had no effect on clavicle poise. However, division of the capsular ligament alone resulted in downward depression on the distal end of the clavicle. He also noted that the intra-articular disk ligament tore under 5 pounds of weight once the capsular ligament had been cut. Bearn’s article has many clinical implications for the mechanisms of injury of the sternoclavicular joint.

Range of Motion of the Sternoclavicular Joint The sternoclavicular joint is freely movable and functions almost like a ball-and-socket joint in that the joint has motion in almost all planes, including rotation.16,17

Although the clavicle is the first long bone of the body to ossify (5th intrauterine week), the epiphysis at the medial end of the clavicle is the last of the long bones in the body to appear and the last epiphysis to close (Fig. 17B-8).7,8,13 The medial clavicular epiphysis does not ossify until the 18th to 20th year, and it fuses with the shaft of the clavicle at about the 23rd to 25th year.7,8,13 Webb and Suchey,18 in an extensive study of the physis of the medial clavicle in 605 males and 254 females at autopsy, reported that complete unions may not be present until 31 years of age. This knowledge of the epiphysis is important because we believe that many of the so-called sternoclavicular dislocations are fractures through the physeal plate.

Applied Surgical Anatomy The surgeon who is planning an operative procedure on or near the sternoclavicular joint should be completely knowledgeable about the vast array of anatomic structures immediate posterior to the sternoclavicular joint. There is a “curtain” of muscles, comprising the sternohyoid, sternothyroid, and scaleni, that is posterior to the sternoclavicular joint and the inner third of the clavicle, and this curtain blocks the view of the vital structures. Some of these vital structures include the innominate artery, the innominate vein, the vagus nerve, the phrenic nerve, the internal jugular vein, the trachea, and the esophagus (Fig. 17B-9). If one is considering the possibility of stabilizing the sternoclavicular joint by running a pin down from the clavicle and into the sternum, it is important to remember that the arch of the aorta, the superior vena cava, and the right pulmonary artery are also very close at hand. Another structure to be aware of is the anterior jugular vein, which is between the clavicle and the curtain of muscles. The anatomy books state that this vein can be quite variable in size. We have seen it as large as 1.5 cm in ­diameter. This vein has no valves, and when it is nicked, it looks as though someone has opened up the flood gates.

MECHANISM OF INJURY Because the sternoclavicular joint is subject to practically every motion of the upper extremity and because the joint is so small and incongruous, one would think that it would be the most commonly dislocated joint in the body. However, the ligamentous supporting structure is so strong and so designed that it is one of the least commonly ­dislocated joints in the body. Traumatic dislocation of the

Shoulder

A

795

B

C

D

E Figure 17B-6   The importance of the various ligaments around the sternoclavicular joint in maintaining normal shoulder poise. A, The lateral end of the clavicle is maintained in an elevated position through the sternoclavicular ligaments. The arrow indicates the fulcrum. B, When the capsule is divided completely, the lateral end of the clavicle descends under its own weight without any loading. The clavicle will appear to be supported by the intra-articular disk ligament. C, After division of the capsular ligament, it was determined that a weight of less than 5 lb was enough to tear the intra-articular disk ligament from its attachment on the costal cartilage junction of the first rib. The fulcrum was transferred laterally so that the medial end of the clavicle hinged over the first rib in the vicinity of the costoclavicular ligament. D, After division of the costoclavicular ligament and the intra-articular disk ligament, the lateral end of the clavicle could not be depressed, as long as the capsular ligament was intact. E, After resection of the medial first costal cartilage along with the costoclavicular ligament, there was no effect on the poise of the lateral end of the clavicle, as long as the capsular ligament was intact. (Redrawn from Beam JG: Direct observation on the function of the capsule of the sternoclavicular joint in clavicular support. J Anat 101:159-170, 1967.)

s­ ternoclavicular joint usually occurs only after tremendous forces, either direct or indirect, have been applied to the ­shoulder.

Direct Force When a force is applied directly to the anteromedial aspect of the clavicle, the clavicle is pushed posteriorly behind the sternum and into the mediastinum. This may occur in a variety of ways: an athlete lying on his back on the ground is jumped on, and the knee of the jumper lands directly on the medial end of the clavicle; a kick is delivered to the front of the medial clavicle; a person is run over by a vehicle; or a person is pinned between a vehicle and a wall. Because of our anatomy, it would be most unusual for a direct force to produce an anterior sternoclavicular dislocation.

Indirect Force A force can be applied indirectly to the sternoclavicular joint from the anterolateral or posterolateral aspect of the shoulder. This is the most common mechanism of injury to the sternoclavicular joint. Mehta and coworkers19 reported that three of four posterior sternoclavicular dislocations were produced by indirect force, and Heinig20 reported that indirect force was responsible for eight of nine cases of posterior sternoclavicular dislocation. It was the most common mechanism of injury in our series of 168 patients. If the shoulder is compressed and rolled forward, an ipsilateral posterior dislocation results; if the shoulder is compressed and rolled backward, an ipsilateral anterior dislocation results (Fig. 17B-10). One of the most common causes that we have seen is a pile-on in a football game. In

796

DeLee & Drez’s Orthopaedic Sports Medicine

35° 45°

35°

A

B

C

Figure 17B-7   Motions of the clavicle and the sternoclavicular joint. A, With full overhead elevation, the clavicle elevates 35 degrees. B, With adduction and extension, the clavicle displaces anteriorly and posteriorly 35 degrees. C, The clavicle rotates on its long axis 45 degrees as the arm is elevated to the full overhead position. (Modified from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

this instance, a player falls on the ground, landing on the lateral shoulder; before he can get out of the way, several players pile on top of his opposite shoulder, which applies significant compressive force on the clavicle down toward the sternum. If, during the compression, the shoulder is rolled forward, the force directed down the clavicle produces a posterior dislocation of the sternoclavicular joint. If the shoulder is compressed and rolled backward, the force directed down the clavicle produces an anterior dislocation of the sternoclavicular joint. Other types of indirect force that can produce sternoclavicular dislocation are a cave-in

Figure 17B-8   Tomogram demonstrating the thin, wafer-like disk of the epiphysis of the medial clavicle. (From Rockwood CA, and Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

on a ditch digger, with lateral compression of the shoulders by the falling dirt; lateral compressive force on the shoulder when a person is pinned between a vehicle and a wall; and a fall on the outstretched abducted arm, which drives the shoulder medially in the same manner as lateral compression on the shoulder.

Most Common Cause of Injury to the Sternoclavicular Joint The most common cause of dislocation of the sternoclavicular joint is vehicular crashes; the second is an injury sustained during participation in sports.21-23 Omer,22 in his review of patients from 14 military hospitals, accumulated 82 cases of dislocation to the sternoclavicular joint. He reported that almost 80% of these cases occurred as the result of vehicular crashes (47%) and athletics (31%). We reviewed 19 patients with posterior sternoclavicular joint injuries managed at the University of Texas Health Science Center at San Antonio. Seventy-nine percent of these injuries were the result of motor vehicle crash or sportsrelated trauma.24 Probably the youngest patient to have a traumatic sternoclavicular dislocation was a 7-month-old girl with an anterior dislocation reported by Wheeler and associates.25 The injury occurred when she was lying on her left side and her older brother accidentally fell on her, compressing her shoulders together. The closed reduction was unstable, and the child was immobilized in a figure-of-eight bandage for 5 weeks. At 10 weeks, the child had a full range of motion, and there was no evidence of instability. The senior author has seen an anterior injury in a 3-year-old patient that occurred as a result of an automobile crash (Fig. 17B-11).

Shoulder

797

Posterior Dislocation of the Clavicle TRACHEA ESOPHAGUS

STERNOTHYROID STERNOHYOID VAGUS N. ANT. JUGULAR V. CLAVICLE (dislocated) SUBCLAVIAN A. &. V. PHRENIC N. PULMONARY A. & V.

Sternoclavicular joint

STERNUM

Clavicle Right innominate artery Right subclavian vein Lymph node

TRACHEA

Esophagus

HEART

Left common carotid artery Left subclavian vein Left subclavian artery Left lung

Right lung VERTEBRA

B

A R. ant. jugular v. R. common carotid a. R. int. jugular v. R. ext. jugular v. R. vagus n. R. subclavian a. R. subclavian v.

L. ant. jugular v. L. common carotid a. L. int. jugular v. L. ext. jugular v. L. subclavian a. Thoracic duct L. subclavian v.

Innominate a. R. brachiocephalic v.

L. vagus n. Aortic arch

Sup. Vena Cava

Pulmonary a.

C

D Figure 17B-9   Applied anatomy of the vital structures posterior to the sternoclavicular joint. A and B, Sagittal views in cross section demonstrating the structures posterior to the sternoclavicular joint. C, A diagram demonstrates the close proximity of the major vessels that are posterior to the sternoclavicular joint. D, A cadaveric dissection showing the relationship of the medial end of the right clavicle to the innominate artery in the mediastinum. (A and C, Redrawn from Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984; B redrawn from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

798

DeLee & Drez’s Orthopaedic Sports Medicine

A

Figure 17B-11   Radiograph of a 3-year-old child with traumatic anterior dislocation of the left sternoclavicular joint. The chest film demonstrates that the left clavicle is superior to the right, suggesting an anterior displacement of the left medial clavicle. (From Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

Anterior Dislocation The anterior dislocation is the most common type of sternoclavicular dislocation. The medial end of the clavicle is displaced anteriorly or anterosuperiorly to the anterior margin of the sternum (Fig. 17B-12).

Posterior Dislocation B Figure 17B-10   Mechanisms that produce anterior or posterior dislocations of the sternoclavicular joint. A, If the patient is lying on the ground and a compression force is applied to the posterior-lateral aspect of the shoulder, the medial end of the clavicle will be displaced posteriorly. B, When the lateral compression force is directed from the anterior position, the medial end of the clavicle is dislocated posteriorly. (Redrawn from Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

CLASSIFICATION OF PROBLEMS OF THE STERNOCLAVICULAR JOINT There are two types of classifications. One is based on the cause of the dislocation, and the other on the anatomic position that the dislocation assumes.

Classification Based on Anatomy Detailed classifications are confusing and difficult to remember, and the following classification is suggested.

Posterior sternoclavicular dislocation is uncommon. The medial end of the clavicle is displaced posteriorly or posterosuperiorly with respect to the posterior margin of the sternum (Figs. 17B-13 and 17B-14).

Classification Based on Cause Traumatic Injury Sprain or Subluxation Acute sprains to the sternoclavicular joint can be classified as mild, moderate, or severe. In a mild sprain, all the ligaments are intact, and the joint is stable. In a moderate sprain, there is subluxation of the sternoclavicular joint. The capsular, intra-articular disk, and costoclavicular ligaments may be partially disrupted. The subluxation may be anterior or posterior. In a severe sprain, there is complete disruption of the sternoclavicular ligaments, and the dislocation may be anterior or posterior.

Acute Dislocation In a dislocated sternoclavicular joint, the capsular and intra-articular ligaments are ruptured. Occasionally, the costoclavicular ligament is intact but stretched out enough to allow the dislocation.

Shoulder

A

799

B

Figure 17B-12   A, Clinically, there is an evident anterior dislocation of the left sternoclavicular joint (arrow). B, When the clavicles are viewed from down around the level of the patient’s knees, it is apparent that the right clavicle is dislocated anteriorly. (From Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

Recurrent Dislocation

Congenital or Developmental Subluxation or Dislocation

If the initial acute traumatic dislocation does not heal, mild to moderate forces may produce recurrent dislocations. This is a rare entity.

Newlin26 reported a case of a 25-year-old man who had bilateral congenital posterior dislocation of the medial ends of the clavicle that simulated an intrathoracic mass. Guerin27 first reported congenital luxations of the sternoclavicular joint in 1841. Congenital defects with loss of bone substance on either side of the joint can predispose to subluxation or dislocation. Cooper1 described a patient with scoliosis so severe that the shoulder was displaced forward enough to dislocate the clavicle posteriorly behind the sternum.

Unreduced Dislocation The original dislocation may go unrecognized, it may be irreducible, or the physician may decide not to reduce certain dislocations.

Atraumatic Problems For a variety of nontraumatic reasons, the sternoclavicular joint may sublux or enlarge.

INCIDENCE OF INJURY TO THE STERNOCLAVICULAR JOINT

Spontaneous Subluxation or Dislocation

Sternoclavicular injuries are rare, and many of the authors apologize for reporting only three or four cases. Attesting to this rarity is the fact that some orthopaedists have never treated or seen a dislocation of the sternoclavicular joint.28,29

One or both of the sternoclavicular joints may spontaneously sublux or dislocate anteriorly during overhead motion. The problem is usually not painful (Fig. 17B-15).

A

B

Figure 17B-13   Interpretation of the cephalic tilt radiographs of the sternoclavicular joints. A, In a patient with posterior dislocation of the left sternoclavicular joint, the medial half of the left clavicle is projected below a horizontal line drawn tangential to the superior aspect of the normal left medial clavicle. B, After manual reduction is performed in the same patient, the medial aspects of both clavicles appear on the same horizontal line.

800

DeLee & Drez’s Orthopaedic Sports Medicine

A B

C

D

E

F

Figure 17B-14   Posterior dislocation of the left sternoclavicular joint. A, A 19-year-old man has a 24-hour-old posterior displacement of the left medial clavicle that occurred from direct trauma to the anterior chest wall. He noted the immediate onset of difficulty in swallowing and some hoarseness. Note that the left medial prominence of the clavicle is lost. B, After a failed attempt at closed reduction under general anesthesia, a sterile towel clip is placed percutaneously around the medial left clavicle after preparing the area with povidone-iodine (Betadine). C, While assistants provide countertraction on the torso and traction on the ipsilateral extremity, the surgeon provides a lateral and anterior force on the clavicle with the towel clip. This results in an audible “pop” as the left sternoclavicular joint reduces. D, After reduction, the left medial clavicular prominence can easily be seen. E and F, Clinical photographs 4 months after reduction demonstrate a stable, healed, left sternoclavicular joint. The patient was asymptomatic with normal range of motion and strength.

Shoulder

A

801

B

Figure 17B-15   Spontaneous anterior subluxation of the sternoclavicular joint. A, With the right arm in the overhead position, the medial end of the right clavicle spontaneously subluxes out anteriorly without any trauma. B, When the arm is brought back down to the side, the medial end of the clavicle spontaneously reduces. This is usually associated with no significant discomfort.

The incidence of sternoclavicular dislocation, based on the series of 1603 injuries of the shoulder girdle reported by Cave,30 is 3%. (The total incidence in the study was glenohumeral dislocations, 85%; acromioclavicular, 12%; and sternoclavicular, 3%.) In the series by Cave and in our experience, dislocation of the sternoclavicular joint is not as rare as posterior dislocation of the glenohumeral joint.

Ratio of Anterior to Posterior Injuries Undoubtedly, anterior dislocations of the sternoclavicular joint are much more common than is the posterior type. However, the ratio of anterior to posterior dislocations is only rarely reported. Theoretically, one could survey the literature and develop the ratio of anterior ­dislocations to posterior dislocations, but most of the published material on sternoclavicular dislocations is on the rare posterior dislocation. Of the references listed at the end of this subchapter that deal with injuries of the sternoclavicular joint, more than 60% discuss only the rare posterior dislocation of the sternoclavicular joint and the various complications associated with it. The largest series from a single institution is reported by Nettles and Linscheid,21 who studied 60 patients with sternoclavicular dislocations (57 anterior and 3 posterior). This gives about a 20:1 ratio of anterior dislocations to posterior dislocations of the sternoclavicular joint. Waskowitz23 reviewed 18 cases of sternoclavicular dislocations, none of which was posterior. However, in our series of 185 traumatic injuries, there have been 135 patients with anterior dislocation and 50 patients with posterior dislocation. In 1986, Hotchkiss31 reported bilateral traumatic dislocation of the sternoclavicular joint. A 28-year-old man was run over by a cart and sustained an anterior dislocation of the right shoulder and a posterior dislocation of the left shoulder. Rockwood has had experience in treating four cases of bilateral sternoclavicular dislocation.

Dislocations of Both Ends of the Clavicle To our knowledge, the first case of dislocation of both ends of the clavicle was reported by Porral32 in 1831. In 1923, Beckman33 reported a single case and reviewed the literature on 15 cases that had previously been reported. With the exception of this patient, all patients had been treated conservatively and had acceptable function. In one patient, a brachial plexus neuropathy developed and was treated by excision of a portion of the clavicle. Until recently, only four additional cases have been reported to our knowledge.34-37 In 1990, Sanders and associates (including Rockwood)38 reported six patients who sustained a dislocation of both ends of the clavicle (anterior dislocation of the sternoclavicular joint and posterior dislocation of the acromioclavicular joint). Two patients, who had fewer demands on the shoulder, did well with only minor symptoms after nonoperative management. The other four patients had persistent symptoms that were localized to the acromioclavicular joint. Each of these patients underwent reconstruction of the acromioclavicular joint, which resulted in painless, full range of motion and return to normal activity.

Combinations of Sternoclavicular Fractures and Dislocations of the Clavicle Elliot39 reported a tripartite injury around the clavicular region in which the patient sustained an anterior subluxation of the right sternoclavicular joint, a type II injury to the right acromioclavicular joint, and a fracture of the right mid-clavicle. Tanlin,40 Arenas and colleagues,41 and Friedl and Fritz42 reported a patient with an anterior dislocation of the sternoclavicular joint and a fracture of the mid-­clavicle. Velutini and Tarazona43 reported a bizarre case of posterior dislocation of the left medial clavicle, first rib, and a section of the manubrium. All these injuries involving the sternoclavicular joint and the clavicle were associated with severe trauma to the shoulder region— the involved shoulder struck an immovable object or was

802

DeLee & Drez’s Orthopaedic Sports Medicine

severely ­ compressed. Neurovascular injuries must always be ruled out with severe trauma to the shoulder.

SIGNS AND SYMPTOMS OF INJURIES TO THE STERNOCLAVICULAR JOINT Mild Sprain In a mild sprain, the ligaments of the joint are intact. The patient complains of a mild to moderate amount of pain, particularly with movement of the upper extremity. The joint may be slightly swollen and tender to palpation, but instability is not noted.

Moderate Sprain (Subluxation) A moderate sprain results in a subluxation of the sternoclavicular joint. The ligaments are either partially disrupted or severely stretched. Swelling is noted, and pain is marked, particularly with any movement of the arm. Anterior or posterior subluxation may be obvious to the examiner when the injured joint is compared with the normal sternoclavicular joint.

Severe Sprain (Dislocation) A severe sprain is analogous to a joint dislocation. The dislocation may be anterior or posterior. The capsular ligament and the intra-articular disk ligament are ruptured. Regardless of whether the dislocation is anterior or posterior, there are characteristic clinical findings of sternoclavicular joint dislocation.

Signs Common to Both Anterior and Posterior Injuries The patient has severe pain that is increased with any movement of the arm, particularly when the shoulders are pressed together by a lateral force. The patient usually supports the injured arm across the trunk with the normal arm. The affected shoulder appears to be shortened and thrust forward when compared with the normal shoulder. The head may be tilted toward the side of the dislocated joint. The patient’s discomfort increases when he or she is placed into the supine position, at which time it will be noted that the involved shoulder will not lie back flat on the table.

Signs and Symptoms of Anterior Injury The medial end of the clavicle is visibly prominent anterior to the sternum (see Fig. 17B-12) and can be palpated anterior to the sternum. It may be fixed anteriorly or may be quite mobile.

Signs and Symptoms of Posterior Injury The patient with a posterior dislocation has more pain than a patient with anterior sternoclavicular ­ dislocation. ­Stankler44 reported two patients with unrecognized

­ osterior dislocations in which venous engorgement of the p ipsilateral arm developed. The anterosuperior fullness of the chest produced by the posteriorly dislocated clavicle is less prominent and visible compared with the normal side. The usually palpable medial end of the clavicle is displaced posteriorly (see Fig. 17B-14). The corner of the sternum is easily palpated compared with the normal sternoclavicular joint. Venous congestion may be present in the neck or in the upper extremity. Breathing difficulties, shortness of breath, or a choking sensation may be noted. Circulation to the ipsilateral arm may be decreased. The patient may have difficulty swallowing or a tight feeling in the throat, may be in a state of complex shock, or may have a pneumothorax. We have seen a number of patients who clinically appeared to have an anterior dislocation of the sternoclavicular joint but, by radiography, were shown to have ­complete posterior dislocation. The point is that one cannot always rely on the clinical findings of observing and palpating the joint to make a distinction between the ­anterior and posterior dislocations.

RADIOGRAPHIC EVALUATION OF INJURIES TO THE STERNOCLAVICULAR JOINT Anteroposterior Views The older literature reflects that routine radiographs of the sternoclavicular joint, regardless of the special views, are difficult to interpret. Special oblique views of the chest have been recommended, but because of the distortion of one of the clavicles over the other, interpretation is difficult. The older literature on dislocation of the sternoclavicular joint indicates that the diagnosis is best made from a clinical examination of the patient and not from the radiographs. However, it goes on to say that tomography offers more detailed information, often showing small fractures in the vicinity of the sternoclavicular joint. Occasionally, routine anteroposterior or posteroanterior radiographs of the chest or sternoclavicular joint suggest that something is wrong with one of the sternoclavicular joints. It would be ideal to take a view at right angles to the anteroposterior plane, but because of our anatomy, it is impossible to take a true 90-degree cephalic-to-caudal lateral view. Lateral radiographs of the chest are at right angles to the anteroposterior plane, but they cannot be interpreted because of the density of the chest and the overlap of the medial clavicles with the first rib and the sternum.

Special Projected Views Kattan45 has recommended a special projection, as have Ritvo and Ritvo,46 Schmitt,47 Fedoseev,48 and Féry and Leáonard.49 Kurzbauer50 has recommended special lateral projections. Hobbs,51 in 1968, proposed a view that comes close to being a 90-degree cephalocaudal lateral view of the sternoclavicular joints. In the same year, Heinig20 recommended an x-ray projection of the sternoclavicular joint that resembles a “swimmer’s view” of the cervical spine.

Shoulder

A

803

B

Figure 17B-16   A, Positioning of the patient for radiographic evaluation of the sternoclavicular joint, as described by Heinig.   B, Heinig view demonstrating a normal relationship between the medial end of the clavicle (C) and the manubrium (M). (Redrawn from Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996.)

Heinig View With the patient is a supine position, the x-ray tube is placed about 30 inches from the involved sternoclavicular joint, and the central ray is directed tangential to the joint and parallel to the opposite clavicle. The cassette is placed against the opposite shoulder and centered on the manubrium (Fig. 17B-16).

Hobbs View In the Hobbs view, the patient is seated at the x-ray table, high enough to lean forward over the table. The cassette is on the table, and the lower anterior rib cage is against the cassette (Fig. 17B-17). The patient leans forward so that the nape of the flexed neck is almost parallel to the table. The flexed elbows straddle the cassette and support the head and neck. The x-ray source is above the nape of the neck, and the beam passes through the cervical spine to project the sternoclavicular joints onto the cassette.

For children, the distance from the tube to the cassette is 45 inches; for adults, whose anteroposterior chest diameter is greater, the distance should be 60 inches. The technical setting of the machine is essentially the same as for a posteroanterior view of the chest. For example, imagine that your eyes are down at the level of the patient’s knees and you are looking up toward his clavicles at a 40-degree angle. If the right sternoclavicular joint were dislocated anteriorly, the right clavicle would appear to be displaced more anteriorly or riding higher on an imaginary horizontal line compared with the normal left clavicle (see Fig. 17B-12). The reverse is true if the left sternoclavicular joint is dislocated posteriorly (i.e., the left

Serendipity View This view is rightfully called the serendipity technique because that is the way it was developed. Accidentally, the senior author found that the next best thing to having a true cephalocaudal lateral view of the sternoclavicular joint is a 40-degree cephalic tilt view. The patient is positioned on the back squarely and in the center of the x-ray table. The tube is tilted at a 40-degree angle off the vertical and is centered directly on the sternum (Fig. 17B-18). A nongrid 11×14 inch cassette is placed squarely on the table and under the patient’s upper shoulders and neck so that the beam aimed at the sternum will project both clavicles onto the film. The tube is adjusted so that the medial half of both clavicles is projected onto the film. It is important to note that the cassette should be placed squarely on the x-ray table (i.e., not angulated or rotated) and that the patient should be positioned squarely on top of the cassette.

Figure 17B-17   Positioning of the patient for radiographic evaluation of the sternoclavicular joint, as recommended by Hobbs. (Modified from Hobbs DW: The sternoclavicular joint: A new axial radiographic view. Radiology 90:801, 1968.)

804

DeLee & Drez’s Orthopaedic Sports Medicine

In 1959, Baker52 recommended the use of tomography, which was developed in the late 1920s, and said it was much more valuable than routine films and the fingertips of the examining physician. Morag and Shahin,53 in 1975, reported on the value of tomography, which they used in a series of 20 patients, and recommended that it be used routinely to evaluate problems of the sternoclavicular joint. From a study of normal sternoclavicular joints, they pointed out the variation in the radiographic appearance in different age groups.

Magnetic Resonance Imaging

clavicle would be displaced inferiorly or riding lower on an imaginary horizontal plane than the normal right clavicle) (see Fig. 17B-13). The idea, then, is to take a 40-degree cephalic tilt radiograph of both medial clavicles and compare the relationship of the injured clavicle to the normal clavicle.

Because computed tomography (CT) has been the imaging study of choice, few data have been collected on the use of magnetic resonance imaging (MRI) to study disorders of the sternoclavicular joint. Klein and colleagues54 used MRI to study cadavers, healthy volunteers, and patients with known disorders of the sternoclavicular joint. They pointed out that the small size of the sternoclavicular joint is poorly imaged with the body coil. The surface coil and the sternoclavicular joint move with breathing of the patient, thereby causing artifacts; even vascular pulsations and swallowing can produce artifacts. Brossmann55 and associates correlated MRI with anatomic sections in 14 sternoclavicular joints from elderly cadavers. They concluded that MRI did depict the anatomy of the sternoclavicular joint and surrounding soft tissues. T2-weighted images were superior to T1-weighted images in depicting the intra-articular disk. Magnetic resonance arthrography allowed delineation of intra-articular disk.

Special Techniques

Computed Tomography

Tomograms

Without question, the computed tomographic scan is the best technique to study any or all problems of the sternoclavicular joint (Fig. 17B-20). It clearly distinguishes injuries of the joint from fractures of the medial clavicle and defines minor subluxations of the joint. The orthopaedist must remember to ask for comoputed tomographic scans of both sternoclavicular joints and the medial half of both clavicles so that the injured side can

Figure 17B-18   Positioning of the patient to take the “serendipity” view of the sternoclavicular joints. The x-ray tube is tilted 40 degrees from the vertical position and is aimed directly at the manubrium. The nongrid cassette should be large enough to receive the projected images of the medial halves of both clavicles. In children, the tube distance from the patient should be 45 inches; in thicker-chested adults, the distance should be 60 inches.

Tomograms can be very helpful in distinguishing between a sternoclavicular dislocation and a fracture of the medial clavicle. They are also helpful in questionable anterior and posterior dislocation of the sternoclavicular joint to distinguish fractures from dislocations and to evaluate arthritic changes (Fig. 17B-19).

Figure 17B-19   Tomogram demonstrating a fracture of the left medial clavicle. The clinical preradiographic diagnosis was an anterior dislocation of the left sternoclavicular joint. (From Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

Shoulder

A

B

C

D

805

Figure 17B-20   Computed tomographic scans of the sternoclavicular joint demonstrating various types of injuries. A, A posterior dislocation of the left clavicle compressing the great vessels and producing swelling of the left arm. B, A fracture of the medial clavicle that does not involve the articular surface. C, A fragment of bone displaced posteriorly into the great vessel. D, A fracture of the medial clavicle into the sternoclavicular joint.

be compared with the normal side. The patient should lie flat in the supine position. If one requests a study of the right sternoclavicular joint, the x-ray technician may rotate the patient to the affected side and provide views of only the one joint. Hartman and Dunnagan56 reported on the use of CT arthrography to demonstrate capsular disruption in a patient after a traumatic injury to the joint.

strain to both sternoclavicular joints. Her bra size, over a period of 4 weeks, had jumped from a 36B to a 38EE. The increase in weight depressed both shoulders and produced pain while upright in both sternoclavicular joints. The discomfort was completely relieved by pushing both elbows up, thus elevating the distal clavicles, which, in turn, took the strain off her sternoclavicular joints. She was referred to her gynecologist and surgeon to determine the quickest way to reduce the size of her breasts.

Treatment of TRAUMATIC INJURIES

Moderate Sprain (Subluxation)

Mild Sprain The joint is stable but painful. Application of ice for the first 12 to 24 hours followed by heat is helpful. The upper extremity should be immobilized in a sling for 3 to 4 days, and then, gradually, the patient can regain use of the arm in everyday activities. The senior author had a fascinating case of a young mother who had a semiacute or semichronic

For subluxation of the sternoclavicular joint, application of ice is recommended for the first 12 hours, followed by heat for the next 24 to 48 hours. The joint may be subluxed anteriorly or posteriorly, and the subluxation may be reduced by drawing the shoulders backward as if reducing and holding a fracture of the clavicle. A clavicle strap can be used to hold the reduction. A sling and swath should also be used to hold up the shoulder and to prevent motion of the arm. The patient should be protected from further possible injury for 4 to 6 weeks.

806

DeLee & Drez’s Orthopaedic Sports Medicine

DePalma57 suggested a plaster figure-of-eight dressing, and ­McLaughlin58 ­recommended the same type of treatment that would be used for fracture of the clavicle, with the addition of a sling to support the arm. Allman59 prefers the use of a soft figure-of-eight bandage with a sling and occasionally uses adhesive strapping over the medial end of the clavicle. In certain circumstances in which the subluxation cannot be reduced, some authors57,60 have recommended repair of the ligaments and temporary internal fixation of the sternoclavicular joint with pins drilled from the clavicle into the sternum. Postoperatively, DePalma61 applies a plaster figure-of-eight cast and, in addition, supports it with a sling and swathe. The pins and the cast are removed after 6 weeks. As pointed out later in the section on the authors’ preferred method of treatment, the use of pins across the sternoclavicular joint has too many serious complications. Occasionally, after conservative treatment of a type II injury, the pain lingers, or the symptoms of popping and grating persist. This situation may require joint exploration. Bateman60 has commented on the possibility of finding a tear of the intra-articular disk, which should be excised. Duggan4 reported a case in which the patient still had popping of the joint several weeks after an injury to the sternoclavicular joint. It has been our experience that if a patient fails a prolonged course of conservative treatment, excision of the symptomatic intra-articular disk, along with the portion of the clavicle that is medial to the costoclavicular (rhomboid) ligaments, can improve these symptoms of popping and grating (Fig. 17B-21).

Severe Sprain (Dislocation) The dislocation of the sternoclavicular joint may be anterior or posterior.

Nonoperative Treatment Anterior Dislocation There is still some controversy regarding the treatment of acute or chronic anterior dislocation of the sternoclavicular joint. In 1990, deJong and Sukui62 reported long-term

A

follow-up results in 10 patients with traumatic anterior sternoclavicular dislocations. All patients were treated nonoperatively with analgesics and immobilization. The results of treatment were good in seven patients, fair in two patients, and poor in one patient at an average follow-up of 5 years. Most acute anterior dislocations are unstable after reduction, and many operative procedures have been described to repair or reconstruct the joint. Technique of Closed Reduction

Closed reduction of the sternoclavicular joint may be accomplished with local or general anesthesia or, in stoic patients, without anesthesia. Most authors recommend the use of narcotics or muscle relaxants. The patient is placed supine on the table, lying on a 3- to 4-inch thick pad between the shoulders. In this position, the clavicle may reduce with direct gentle pressure over the anteriorly displaced clavicle. However, when the pressure is released, the clavicle usually dislocates again. Occasionally, the clavicle will remain reduced. Sometimes, the physician will need to push both shoulders back to the table while an assistant applies pressure to the anteriorly displaced clavicle. Laidlaw63 treated an interesting case of a patient who had a dislocated clavicle. The sternoclavicular joint was dislocated anteriorly and was mildly symptomatic. The acromioclavicular joint was most symptomatic and was treated by excision of the distal clavicle. Surprisingly, the anteriorly dislocated sternoclavicular joint reduced and became pain free. More recently, Thomas and Friedman64 reported an ipsilateral anterior sternoclavicular dislocation and clavicle shaft fracture. Closed reduction of the dislocation was possible only after open reduction and internal fixation of the clavicle fracture. Unfortunately, the anterior sternoclavicular dislocation was diagnosed solely by clinical examination without the aid of CT, tomograms, or special projections. Postreduction Care

If, with the shoulders held back, the sternoclavicular joint remains reduced, the shoulders can be stabilized with a soft figure-of-eight dressing, a commercial clavicle strap harness, or a plaster figure-of-eight cast. Some authors recommend a bulky pressure pad over the anterior medial clavicle that is held in place with elastic tape. A sling might be used

B

Figure 17B-21   A, An operative photograph showing excision of a symptomatic intra-articular disk in a 39-year-old patient who failed conservative treatment of a type II injury. B, Surgical specimen after excision.

Shoulder

because it holds up the shoulder and prevents motion of the arm. Immobilization should be maintained at least 6 weeks, and then the arm should be protected for another 2 weeks before strenuous activities are undertaken. If the sternoclavicular joint again dislocates when the reduction pressure is released, as it usually does, a figure-of-eight dressing or a sling can be used until the patient’s symptoms subside. Most anterior closed reductions of the sternoclavicular joint are unstable, and even with the shoulders held back, the joint is unstable. Although some authors have recommended operative repair of anterior dislocations of the sternoclavicular joint, we believe that the ­operative complications are too great and the end results too unsatisfactory to consider an open reduction. Certainly in children, in whom many if not most of the injuries are physeal fractures, a nonoperative approach should be strongly ­considered.

Posterior Dislocation  A careful examination of the patient is extremely important. Complications are common with posterior dislocation of the sternoclavicular joint, and the patient should receive prompt attention. A careful history and physical examination should be done to rule out damage to the pulmonary and vascular systems. The sternoclavicular joint must be carefully evaluated by all available radiographic techniques, including, when indicated, combined aortogram and computed tomographic scan for potential vascular injuries (Fig. 17B-22). If specific complications are noted, appropriate consultants should be called in before reduction is performed. Worman and Leagus65 reported a posterior dislocation of the sternoclavicular joint in which it was noted at surgery that the displaced clavicle had put a hole in the right pulmonary artery. The

A

807

clavicle had prevented exsanguination because the vessel was still impaled by the clavicle. If a closed reduction had been performed in the emergency department, the result could have been disastrous. Cooper and coworkers66 reported a posterior sternoclavicular dislocation that transected the internal mammary artery and lacerated the brachiocephalic vein. The vascular injuries were associated with fractures of the anterior ends of the first to third ribs and marked posterior instability of the medial end of the clavicle. The brachiocephalic vein was repaired, but the posteriorly displaced medial clavicle impinged on the suture line. To maintain reduction of the sternoclavicular joint, a novel method of stabilization was employed using an external fixator. General anesthesia is usually required for reduction of a posterior dislocation of the sternoclavicular joint because the patient has so much pain and muscle spasm. However, for the stoic patient, some authors have performed the reduction under intravenous narcotics and muscle relaxants. Heinig20 successfully used local anesthesia to reduce a posterior dislocation. From a review of the earlier literature, it would appear that the treatment of choice for posterior sternoclavicular dislocation was by operative means. However, since the 1950s, the treatment of choice of posterior sternoclavicular dislocation is closed reduction.20,58,67-75 Some authors75,76 who had previously done open reductions reported that they were amazed at how easily the dislocation reduced under direct vision, and thereafter, they used closed reductions with complete success. Techniques of Closed Reduction

Many different techniques have been described for closed reduction of a posterior dislocation of the sternoclavicular joint.

B

Figure 17B-22   A, Computed tomographic scan revealing a posterior fracture-dislocation of the sternoclavicular joint with significant soft tissue swelling and compromise of the hilar structures. B, Duplex ultrasound study revealing a large pseudoaneurysm of the right subclavian artery. Note the large neck of the pseudoaneurysm, which measured about 1 cm in diameter (arrow). (From Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996.)

808

DeLee & Drez’s Orthopaedic Sports Medicine

A

Sand bag between shoulders

B Figure 17B-24   Clinical photograph showing a patient with a posterior sternoclavicular dislocation positioned with a roll of towels between the scapulae to extend the shoulder before attempting a closed reduction maneuver.

C

Figure 17B-23   Technique of closed reduction of the sternoclavicular joint. A, The patient is positioned supine with a sandbag placed between the two shoulders. Traction is then applied to the arm against countertraction in an abducted and slightly extended position. In anterior dislocations, direct pressure over the medial end of the clavicle may reduce the joint. B, In posterior dislocations, in addition to the traction, it may be necessary to manipulate the medial end of the clavicle with the fingers to dislodge the clavicle from behind the manubrium. C, In stubborn cases of posterior dislocations, it may be necessary to sterilely prepare the medial end of the clavicle and use a towel clip to grasp around the medial clavicle to lift it back into position. (Redrawn from Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

downward pressure is exerted on the shoulders (Fig. 17B-25). The clavicle is levered over the first rib into its normal position. Buckerfield and Castle78 reported that this technique has succeeded when the abduction traction technique has failed. Butterworth and Kirk67 used a similar adducted position, except they applied lateral traction on the upper humerus. Additional Techniques

Heinig20 and Elting79 both reported that they accomplished reduction by placing the patient supine on the table with three or four folded towels between the two shoulders. Forward pressure was then applied on both shoulders, which accomplished the reduction. Other authors have put their knee between the shoulders of the seated patient and, by pulling back on both shoulders, accomplished a reduction. Stein75 used skin traction on the abducted and extended arm to accomplish the reduction gently and ­ gradually.

Abduction Traction Technique

For the abduction traction technique,10,29,69,72,73,77 the patient is placed on his or her back with the dislocated shoulder near the edge of the table. A 3- to 4-inch thick sandbag is placed between the patient’s shoulders (Fig. 17B-23). Lateral traction is applied to the abducted arm, which is then gradually brought back into extension. This may be all that is necessary to accomplish the reduction. The clavicle usually reduces with an audible snap or pop, and it is almost always stable. Too much extension can bind the anterior surface of the dislocated medial clavicle on the back of the manubrium. Occasionally, it may be necessary to grasp the medial clavicle with one’s fingers to dislodge it from behind the sternum. If this fails, the skin is prepared, and a sterile towel clip is used to grasp the medial clavicle to apply lateral and anterior traction (Fig. 17B-14). Adduction Traction Technique

In this technique,78 the patient is supine on the table with a 3- to 4-inch bolster between the shoulders (Fig. 17B-24). Traction is then applied to the arm in adduction, while a

Figure 17B-25   Clinical photograph of a patient with a posterior sternoclavicular dislocation showing tractioncountertraction technique while gradually bringing the abducted arm into extension.

Shoulder

Many authors have reported that closed reduction usually cannot be accomplished after 48 hours. However, others78,80 have reported closed reduction as late as 4 and 5 days after the injury. Postreduction Care

After reduction, to allow ligament healing, the shoulders should be held back for 4 to 6 weeks with a figure-of-eight dressing or one of the commercially available figure-ofeight straps used to treat fractures of the clavicle. If closed maneuvers fail in the adult, an operative procedure should be performed because most adult patients cannot tolerate the posterior displacement of the clavicle into the mediastinum. Gangahar and Flogaites81 reported a case of late thoracic outlet syndrome following an unreduced posterior dislocation, and Borrero82 reported late and significant vascular problems. The senior author was asked to evaluate a patient who, following a significant injury, complained of swelling and bluish coloration of his left arm after any type of physical activity. He really did not have many local sternoclavicular joint symptoms, but by physical examination, the left clavicle was displaced ­posteriorly. The computed tomographic scan demonstrated a major posterior displacement of the left clavicle (see Fig. 17B-38). Because of the marked displacement and the vascular compromise, arteriography combined with the computed tomographic scan was performed, which did not reveal any vascular leak. With the help of the chest surgeon, the clavicle was removed from the mediastinum, the medial 11⁄2 inches were removed, and the shaft was ­stabilized to the first rib. The greatest displacement that we have seen was in a patient with a posteroinferior dislocation of the medial clavicle down into an intrathoracic position. However, Louw and Louw83 reported a case in which a 30-year-old patient with a T3-T4 paraplegia and posterior dislocation of the left sternoclavicular joint had essentially no problems. He underwent an extensive rehabilitation program and could lower himself from his wheelchair to the floor and back again without assistance, and he could transfer from his wheelchair to the bed, bath, and car without difficulty. If children and young adults younger than 23 to 25 years of age have symptoms from the pressure of the posteriorly displaced clavicle into the mediastinum, an operative procedure should likewise be performed. However, children may have no symptoms, and the physician can wait and watch to see whether the physeal plate remodeling process removes the posteriorly displaced bone.84 Indeed, as with other childhood fractures, the potential for remodeling is significant and may extend until the 23rd to 25th year. The senior author85 has demonstrated a similar mechanism to support conservative treatment of adolescent acromioclavicular joint injuries or “pseudodislocation,” in which there is a partial tear of the periosteal tube containing the distal clavicle. The coracoclavicular ligaments remain secured to the periosteal tube. Because of its high osteogenic potential, spontaneous healing and remodeling to the preinjury “reduced” position occur within this periosteal conduit. Zaslav and associates86 have reported successful treatment of a posteriorly displaced medial clavicle physeal injury in an adolescent athlete with

809

CT documentation of remodeling, most probably within an intact periosteal tube. Similarly, Hsu and associates87 reported successful treatment of a posterior epiphyseal fracture dislocation of the medial end of the clavicle in a 15-year-old patient.

Technique of Operative Treatment The operative procedure should be performed in a manner that disturbs as few of the anterior ligament structures as possible. If the procedure can be performed with the anterior ligaments intact, the reduction may be stable with the shoulders held back in a figure-of-eight dressing. If all the ligaments are disrupted, a significant decision has to be made to try to stabilize the sternoclavicular joint or to resect the medial 1 to 11⁄2 inches of the medial clavicle and stabilize the remaining clavicle to the first rib. Some of the older literature of the 1960s and 1970s recommended stabilization of the sternoclavicular joint with pins. Elting79 used Kirschner wires to stabilize the joint and supplemented ligament repairs with a short toe extensor tendon. Denham and Dingley14 and Brooks and Henning15 used Kirschner wires. DePalma57 and Brown88 recommended a repair of the ligaments and stabilized the sternoclavicular joint with one or two Steinmann pins. Habernek and Hertz,89 Nutz,90 Pfister and Weller,91 Kennedy,92 Tagliabue and Riva,93 Hartman and Dunnagan,56 Bankart,94 Ecke,95 and Stein75 avoided the use of pins across the sternoclavicular joint and used loops of various types of suture wires across the joint. Burri and Neugebauer96 recommended the use of a figure-of-eight loop of carbon fiber. Maguire,97 Booth and Roper,98 Barth and Hagen,99 and Lunseth and coworkers100 reconstructed the sternoclavicular joint using local tendons of the sternocleidomastoid, subclavius, or pectoralis major tendons for repair. Haug101 reported on the use of a special plate to stabilize the joint. The complications of fixation of the sternoclavicular joint with Kirschner wires or Steinmann pins are horrendous and are discussed in the section on complications.

Recurrent or Unreduced Dislocation Nonoperative Approach

It has been stressed by many authors that most patients who have recurrent long-standing sternoclavicular joint dislocation are asymptomatic and require no treatment. Holm­ dahl,28 Louw and Louw,83 and Borrero82 have reported complications after unreduced posterior dislocations. Surgical Reconstructions

There are several basic procedures to maintain the medial end of the clavicle in its normal articulation with the sternum. Fascia lata, suture, internal fixation across the joint, subclavius tendons, osteotomy of the medial clavicle, and resection of the medial end of the clavicle have been advocated. More recently, Spencer and Kuhn102 performed a biomechanical analysis of sternoclavicular reconstructions and reported a new technique involving a semitendinosus graft in a figure-of-eight fashion through the clavicle and manubrium that provided improved peak-to-load failure results (Fig. 17B-26).

810

DeLee & Drez’s Orthopaedic Sports Medicine

A

C

Fascia Lata

Bankart94 and Milch103 used fascia lata between the clavicle and the sternum. Lowman6 used a loop of fascia in and through the sternoclavicular joint so that it acts like the ligamentum teres in the hip. Speed104 and Key and Conwell71 reported on the use of a fascial loop between the clavicle and the first rib. Allen105 used fascia lata to reconstruct a new sternoclavicular ligament. Subclavius Tendon

Burrows106 recommended that the subclavius tendon be used to reconstruct a new costoclavicular ligament. The origin of the subclavius muscle is from the first rib just 6 mm lateral and 1.3 mm anterior to the attachment of the costoclavicular ligament.7,8 The insertion of the tendon is to the inferior surface of the junction of the middle third with the outer third of the clavicle, and the muscle fibers arising from the tendon insert into the inferior surface of the middle third of the clavicle. The muscle fibers coming off the tendon look like feathers on a bird’s wing. Burrows detaches the muscle fiber from the tendon, does not disturb the origin of the tendon, and then passes the tendon through drill holes in the anterior aspect of the proximal end of the clavicle. When comparing his operation with the use of free strips of fascia, Burrows said that it is “safer and easier to pick up a mooring than to drop anchor; the obvious mooring is the

B

Figure 17B-26   A, Clinical photograph showing a patient with a symptomatic chronic posterior sternoclavicular dislocation on the right. B, The proposed skin incision after the patient has been prepared and draped. C, Intraoperative photograph showing a semitendinosus graft being positioned for a figure-of-eight reconstruction.

tendon of the subclavius separated from its muscle fiber and suitably realigned.” Lunseth and associates100 reported a modified Burrows procedure with the additional use of a threaded Steinmann pin across the joint. Osteotomy of the Medial Clavicle

As previously described, Omer,22 following repair or reconstruction of the ligaments, creates a step-cut osteotomy lateral to the joint and detaches the clavicular head of the sternocleidomastoid muscle from the proximal fragment. Resection of the Medial End of the Clavicle

McLaughlin,58 Breitner and Wirth,107 Pridie,108 ­Bateman,60 and Milch103 all have recommended excision of the medial clavicle when degenerative changes are noted in the joint. If the medial end of the clavicle is to be removed because of degenerative changes, the surgeon should be careful not to damage the costoclavicular ligament. Arthrodesis

Arthrodesis was once reported109 in the treatment of a habitual dislocation of the sternoclavicular joint. However, this procedure should not be done because it prevents the previously described normal elevation, depression, and rotation of the clavicle. The end result would be a severe restriction of shoulder movement.

Shoulder

PHYSEAL INJURIES As has been described earlier in the chapter, the epiphysis on the medial end of the clavicle is the last epiphysis in the body to appear on radiography and the last one to close. The epiphysis on the medial end of the clavicle does not appear on radiographs until about the 18th year and does not unite with the clavicle until the 23rd to 25th year (see Fig. 17B-8).7,8,18 This is important information to remember because many of the “dislocations of the sternoclavicular joint” are not dislocations but rather physeal injuries. Most of these injuries will heal with time, without surgical ­intervention. In time, the remodeling process eliminates any bone deformity or displacement. Anterior physeal injuries can certainly be left alone without problem. Posterior physeal injuries should be reduced. If the posterior dislocation cannot be reduced and the patient is having no significant symptoms, the displacement can be observed while remodeling occurs. If the posterior displacement is symptomatic and cannot be reduced closed, the displacement must be reduced during surgery. In 1967, Denham and Dingley14 reported three cases of medial clavicle physeal injury in patients aged 14 to 16 years. They demonstrated at surgery that the pathology was indeed a physeal fracture of the medial clavicle. In 1972, Brooks and Henning15 presented a paper in which they concluded from a review of nine cases that many “sternoclavicular dislocations” and “fractures of the medial clavicle” were indeed medial clavicle physeal injuries. In 1984, Lemire and Rosman110 reported a case of “double fracture” of the medial clavicle in a 15-year-old patient. One fracture was through the physis and the other through the medial third of the clavicle. They surgically restored the clavicle into the periosteal tube, and the patient returned to normal activity without any further problems. Winter and associates111 reported two cases of posterior physeal injuries in children, one 9 years old and the other 12 years old. One had displacement against the trachea. The other, as seen on the computed tomographic scan, had compression of the subclavian vein, which was associated with a carotid artery bruit. Krenzien112 reported on luxations of the sternoclavicular joint but did not comment on whether they were physeal injuries.

Anterior Displacement of the Medial Clavicle If the physeal injury is recognized, or if the patient is younger than 25 years, closed reduction, as has been described for anterior dislocation of the sternoclavicular joint, should be performed. The shoulders should be held back in a ­clavicular strap or figure-of-eight dressing for 3 to 4 weeks, even if the reduction is unstable. Healing is prompt, and remodeling will occur at the site of the ­deformity.

Posterior Displacement of the Medial Clavicle Closed reduction of this injury should be performed in the manner described for posterior dislocation of the sternoclavicular joint. The reduction is usually stable, with

811

shoulders held back in a figure-of-eight dressing or strap. Immobilization should continue for 3 to 4 weeks. If the posterior physeal injury cannot be reduced, if the patient is not having symptoms, and if the patient is a child or an adult younger than 23 years, one could wait and watch to see whether remodeling eliminates the posteriorly ­displaced clavicle.

treatment of ATRAUMATIC PROBLEMS Spontaneous Subluxation or Dislocation As with glenohumeral joint instability, the importance of distinguishing between traumatic and atraumatic instability of the sternoclavicular joint must be recognized if complications are to be avoided. Rowe113 described several patients who have undergone one or more unsuccessful attempts to stabilize the sternoclavicular joint. In all cases, the patient was able to voluntarily dislocate the clavicle after surgery. In addition, he has described several young patients who were able to “flip the clavicle out and back in,” without elevation of the arms.114 In our experience, spontaneous subluxations and dislocations of the sternoclavicular joint are seen most often in patients younger than 20 years and more often in females. Without significant trauma, one or both of the medial clavicles spontaneously displace anteriorly during abduction or flexion to the overhead position (see Fig. 17B-15). The clavicle reduces when the arm is returned to the side. This is usually associated with laxity in other joints of the extremities. Rockwood and Odor reviewed 37 cases and found it to be a self-limiting condition.115 Surgical reconstruction should not be attempted because the joint will continue to subluxate or dislocate and surgery may indeed cause more pain, discomfort, and an unsightly scar. The condition should be carefully explained to the patient and the family, and they should be told that ultimately it will not be a problem and the symptoms may disappear.

Congenital or Developmental Conditions Congenital or developmental problems (e.g., absence or partial absence of bone or muscles) can produce ­subluxation or dislocation of the sternoclavicular joint. Specific rehabilitation or surgical procedures are not usually ­necessary.26,27

Arthritis The management of patients with osteoarthritis can usually be done with conservative nonoperative treatment; that is, heat, anti-inflammatory agents, and rest.116,117 However, the patient must be thoroughly evaluated to rule out other conditions that mimic the changes in the sternoclavicular joint (e.g., tumor, metabolic, infectious, or collagen disorders). Patients with post-traumatic arthritic changes in the sternoclavicular joint, which follow fractures of the sternoclavicular joint and previous attempts at reconstruction, may require a formal arthroplasty of the

812

DeLee & Drez’s Orthopaedic Sports Medicine

joint and careful stabilization of the remaining clavicle to the first rib. Pingsmann and coauthors118 reported on the midterm results in eight women who did not respond to conservative therapy and underwent surgical treatment for primary arthritis of the sternoclavicular joint. After limited resection of the medial clavicle and preservation of the coracoclavicular and interclavicular ligaments, they reported improvement in the Constant and Rockwood score and concluded that the arthroplasty was effective and safe for the treatment of chronic, symptomatic degenerative arthritis of the sternoclavicular joint. Patients with collagen disorders, such as rheumatoid arthritis, and some patients with condensing osteitis of the medial clavicle may require an arthroplasty. In operating on the sternoclavicular joint, care must be taken to evaluate the residual stability of the medial clavicle. It is the same analogy as used when resecting the distal clavicle for a

Authors’ Preferred Method T r a u ma t i c P r o b l e m s

for

complete old acromioclavicular joint problem. If the coracoclavicular ligaments are intact, an excision of the distal clavicle is indicated. If the coracoclavicular ligaments are gone, then, in addition to the excision of the distal clavicle, the coracoclavicular ligaments must be reconstructed. If the costoclavicular ligaments are intact, the clavicle medial to the ligaments should be resected and beveled smooth. If the ligaments are gone, the clavicle must be stabilized to the first rib. If too much clavicle is resected, or if the clavicle is not stabilized to the first rib, an increase in symptoms can occur. Patients with sternocostoclavicular hyperostosis have a difficult problem to manage.119-127 The condition cannot be arrested with drugs. Treatment is mainly dependent on analgesic and anti-inflammatory medications and physical therapy. Occasionally, surgical excision of the bony mass to allow an increase in function of the upper extremity is indicated.

Treatment

Type I Injury (Mild Sprain)

For mild sprains, we recommend the use of cold packs for the first 12 to 24 hours and a sling to rest the joint. Ordinarily, after 5 to 7 days, the patient can use the arm for everyday living activities. Type II Injury (Subluxation)

In addition to the cold pack, we may use a soft, padded figureof-eight clavicle strap to hold the shoulder back gently to allow the sternoclavicular joint to rest. The figure-of-eight harness can be removed after a week or so, and then the patient either uses a sling for a week or so or is allowed to gradually return to everyday living activities. Type III Injury (Dislocation)

In general, we manage almost all dislocations of the sternoclavicular joint in children and in adults by either a closed reduction or a nonoperative “skillful neglect” form of treatment. The acute traumatic posterior dislocations are reduced closed and become stable when the shoulders are held back in a figure-of-eight dressing. Most of the anterior dislocations are unstable, but we accept the deformity because it is less of a problem than the potential problems of operative repair and internal fixation. Anterior Dislocation

In most instances, knowing that the anterior dislocation will be unstable, we will still try to reduce the anterior displacement. Muscle relaxants and narcotics are administered intravenously, and the patient is placed supine on the table with a stack of three or four towels between the shoulder blades. While an assistant gently applies downward pressure on the anterior aspect of both shoulders, the medial end of the clavicle is pushed backward where it belongs. On some occasions, rare as they may be, the anterior displacement

of

may stay adjacent to the sternum. However, in most cases, either with the shoulders still held back or when they are relaxed, an anterior displacement promptly recurs. We explain to the patient that the joint is unstable and that the hazards of internal fixation are too great, and we prescribe a sling for a couple of weeks and allow the patient to begin using the arm as soon as the discomfort improves. Most of the anterior injuries that we have treated in children and in adults younger than 25 years of age are not dislocations of the sternoclavicular joint but type I or II physeal injuries, which heal and remodel without operative treatment. Patients older than 23 to 25 years of age with anterior dislocations of the sternoclavicular joint do have persistent prominence of the anterior clavicle. However, this does not appear to interfere with usual activities and, in some cases, has not even interfered with heavy manual labor. We wish to re-emphasize that we do not recommend open reduction of the joint and would never recommend transfixing pins across the sternoclavicular joint. If the reduction happens to be stable, we place the patient in either a figure-of-eight dressing or in whatever device or position the clavicle is most stable. If the reduction is unstable, the arm is placed into a sling for a week or so, and then the patient can begin to use the arm for gentle daily activities. Posterior Dislocation

It is important to take a careful history and to perform a careful physical examination. The physician should obtain radiographs, tomograms or computed tomographic scans, or angiogram computed tomographic scans to document whether there is any compression of the great vessels in the neck or arm or any difficulty in swallowing or breathing. It is also important to determine whether the patient has any feeling of choking or hoarseness. If any of these symptoms are present, indicating pressure on the mediastinum, the appropriate specialist should be consulted.

Shoulder

813

A u t h o r s ’ P r e f e r r e d M e t h o d f o r ­T r e a t m e n t o f T r a u ma t i c P r o b l e m s — c o n t ’ d We do not believe that operative techniques are usually required to reduce the acute posterior sternoclavicular joint dislocation. Furthermore, once the joint has been reduced closed, it is usually stable (Fig. 17B-27). Although we used to make the diagnosis of anterior or posterior injury of the sternoclavicular joint on physical examination, we know now that one cannot rely on the anterior swelling and firmness as being diagnostic of an anterior injury. We have been fooled on several occasions

when, from physical examination, the patient appeared to have an anterior dislocation, but radiographs documented a posterior problem. Therefore, we recommend that the clinical impression always be documented with appropriate radiographs before any decision to treat or not to treat is made. For a closed reduction, the patient is placed in the supine position with a 3- to 4-inch thick sandbag or three to four folded towels between the scapulae to extend the shoulders.

A

B

C

D

Figure 17B-27   Posterior dislocation or type I epiphyseal separation of the left sternoclavicular joint in a 12-year-old boy.   A, The 40-degree cephalic tilt serendipity radiograph reveals that the left clavicle is significantly lower on the horizontal plane than the normal right clavicle. B, Before reduction, the medial end of the left clavicle was displaced posteriorly compared with the normal right clavicle. The only remaining prominence of the left sternoclavicular joint was the prominence of the superomedial corner of the manubrium. C, Under general anesthesia, a closed reduction was performed by traction on the arm out into abduction and extension. The clavicle reduced with an audible pop back into position. It was restored to the same horizontal level as the normal right clavicle. D, Clinically both clavicles were palpable at the same level after reduction. (From Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.) Continued

814

DeLee & Drez’s Orthopaedic Sports Medicine

A u t h o r s ’ P r e f e r r e d M e t h o d f o r ­T r e a t m e n t o f T r a u ma t i c P r o b l e m s — c o n t ’ d The dislocated shoulder should be over toward the edge of the table so that the arm and the shoulder can be abducted and extended. If the patient is having extreme pain and muscle spasm and is quite anxious, we use general anesthesia; otherwise, narcotics, muscle relaxants, or tranquilizers are given through an established intravenous route in the normal arm. First, gentle traction is applied on the abducted arm in line with the clavicle while countertraction is applied by an assistant who steadies the patient on the table. The traction on the abducted arm is gradually increased while the arm is brought into extension. Reduction of an acute injury usually occurs with an audible pop or snap, and the relocation can be noted visibly. If the traction in abduction and extension is not successful, an assistant grasps or pushes down on the clavicle in an effort to dislodge it from behind the sternum. Occasionally, in a stubborn case, especially in a thick-chested person or a patient with extensive swelling, it is impossible to obtain a secure grasp on the clavicle with the assistant’s fingers. The skin should then be surgically prepared and a sterile towel clip should be used to gain purchase on the medial clavicle percutaneously (see Fig. 17B-14). The towel clip is used to grasp completely around the shaft of the clavicle. The dense cortical bone prevents the purchase of the towel clip into the clavicle. Then the combined traction through the arm plus the anterior lifting force on the towel clip will reduce the dislocation. After the reduction, the sternoclavicular joint is stable, even with the patient’s arms at the side. However, we always hold the shoulders back in a well-padded figure-ofeight clavicle strap for 3 to 4 weeks to allow for soft tissue and ligamentous healing. We have not used the technique described by Butterworth and Kirk67 involving traction on the arm in adduction, but we believe this technique is valid and will use it if the traction abduction technique fails. The complications of an unreduced posterior dislocation are numerous; for example, thoracic outlet syndrome81 and vascular compromise82 and erosion of the medial clavicle into any one of the many vital structures that lie posterior to the sternoclavicular joint. Therefore, in adults, if closed reduction fails, an open reduction should be performed. For an open reduction, the patient is supine on the ­table, and three to four towels or a small sandbag is placed ­between the scapulae. The upper extremity should be draped out free so that lateral traction can be applied ­during the open reduction. In addition, a folded sheet around the patient’s thorax should be left in place so that it could be used for ­countertraction during the traction on the ­involved extremity. An anterior incision is used that parallels the superior border of the medial 3 to 4 inches of the clavicle, and then extends downward over the sternum just medial to the involved sternoclavicular joint (Fig. 17B-28). As previously described, this should usually be done with a thoracic surgeon. The trick is to remove sufficient soft tissues to expose the joint but to leave the anterior capsular ligament intact. The reduction can usually be accomplished with traction

and countertraction while lifting up anteriorly with a clamp around the medial clavicle. Along with the traction and countertraction, it may be necessary to use an elevator to pry the clavicle back to its articulation with the sternum. When the reduction has been obtained, and with the shoulders held back, the reduction will be stable because the anterior capsule has been left intact. If the anterior capsule is damaged or is insufficient to prevent anterior displacement of the medial end of the clavicle, we recommend an excision of the medial 1 to 11⁄2 inches of the clavicle and securing the residual clavicle to the first rib with 1-mm Dacron tape. The medial clavicle is exposed by careful subperiosteal dissection (Fig. 17B-29). When possible, any remnant of the capsular or intra-articular disk ligaments should be identified and preserved because these structures can be used to stabilize the medial clavicle (Fig. 17B-30). The capsular ligament covers the anterosuperior and posterior aspects of the joint and represents thickenings of the joint capsule. This ligament is primarily attached to the epiphysis of the medial clavicle and is usually avulsed from this structure with posterior sternoclavicular dislocations. Similarly, the intraarticular disk ligament is usually intact where it arises from the synchondral junction of the first rib and sternum and is avulsed from its attachment site on the medial clavicle. If the sternal attachment site of these structures is intact, a nonabsorbable No. 1 cottony Dacron suture is woven back and forth through one or both ligaments so that the ends of the suture exit through the avulsed free end of the tissue. The medial end of the clavicle is resected (Fig. 17B-31), being careful to protect the underlying structures (see Fig. 17B-9D). The medullary canal of the medial clavicle is drilled out and curetted to receive the transferred capsular or intraarticular disk ligament (Fig. 17B-32). Two small drill holes are then placed in the superior cortex of the medial clavicle, about 1 cm lateral to the site of resection (Fig. 17B-33). These holes communicate with the medullary canal and are used to secure the suture in the transferred ligament. The free ends of the suture are passed into the medullary canal

Figure 17B-28   Proposed skin incision for open reduction of a posterior sternoclavicular dislocation. (Modified from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

Shoulder

815

A u t h o r s ’ P r e f e r r e d M e t h o d f o r ­T r e a t m e n t o f T r a u ma t i c P r o b l e m s � — c o n t ’ d� of the medial clavicle and out the two small drill holes in the superior cortex of the clavicle (Fig. 17B-34). While the clavicle is held in a reduced anteroposterior position in relationship to the first rib and sternum, the sutures are used to pull the ligament tightly into the medullary canal of the clavicle. The suture is tied, thus securing the transferred ligament

to the clavicle (Fig. 17B-35). The stabilization procedure is completed by passing several 1-mm cottony Dacron sutures around the medial end of the remaining clavicle and securing the periosteal sleeve of the clavicle to the costoclavicular ligament (Fig. 17B-36). Postoperatively, the patient should be held with the shoulders back in a figure-of-eight dressing for 4 to 6 weeks to allow for healing of the soft tissues. Note: We do not recommend the use of Kirschner wires or Steinmann pins or any other type of metallic pins to stabilize the sternoclavicular joint. The complications are horrendous; see the section on complications for details. Recurrent Traumatic ­ ternoclavicular Joint S

Dislocation

of

the

Recurrent anterior or posterior dislocation of the sternoclavicular joint after an acute injury is extremely rare. We have seen only one patient with recurrent posterior dislocation after a traumatic injury. Usually the joint is stable following reduction, or it remains permanently anteriorly or ­posteriorly displaced. This entity should not be confused with the problem of spontaneous subluxation or dislocation. Unreduced Traumatic Sternoclavicular Joint

A

Dislocation

of

the

As previously described, most patients with an unreduced and permanent anterior dislocation of the sternoclavicular joint are not very symptomatic, have almost a complete

B Figure 17B-29   A and B, Subperiosteal exposure of the medial clavicle. Note the posteriorly displaced medial end of the clavicle. (Redrawn from Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996.)

Figure 17B-30   Forceps holding the anterior portion of the sternoclavicular ligament, which was avulsed from its attachment on the medial clavicle. The sternal attachment site of this ligament was intact. (From Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.) Continued

816

DeLee & Drez’s Orthopaedic Sports Medicine

A u t h o r s ’ P r e f e r r e d M e t h o d f o r ­T r e a t m e n t o f T r a u ma t i c P r o b l e m s — c o n t ’ d

A

Figure 17B-32   The medullary canal of the medial clavicle is curetted in preparation for receiving the transferred sternoclavicular capsular ligament. (Modified from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

B Figure 17B-31   A and B, Excision of the medial clavicle is facilitated by creating drill holes at the intended site of osteotomy. (Redrawn from Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

range of motion, and can work and even perform manual labor without many problems. Because the joint is so small and incongruous and because the results we have seen in patients who have had attempted reconstructions are so ­miserable, we usually recommend a nonoperative skillful

neglect type of treatment. In patients who have had a previous failed sternoclavicular operative procedure, we perform a repeat arthroplasty with a resection of the medial clavicle in an attempt to reduce and stabilize the joint with suture, fascia, tendons, and so on. If the patient has persistent symptoms of traumatic arthritis for 6 to 12 months after a dislocation, and if the symptoms can be completely relieved by injection of local anesthesia into the sternoclavicular joint region, we would perform an arthroplasty of the sternoclavicular joint. This would include a resection of the medial 1 inch of the clavicle with a beveling of the superoanterior corner for cosmetic purposes, débridement of the intra-articular disk ligament, and stabilization of the remaining clavicle to the first rib with either 1- or 3-mm cotton Dacron tape. If the costoclavicular ligaments do not stabilize the medial clavicle, it is essential to reconstruct a ligament-like structure between the clavicle and the first rib. We also detach the clavicular head of the sternocleidomastoid to temporarily resist the upward pull of the clavicle by this muscle. In the adult with unreduced posterior dislocation, because of the potential problems that can be associated with the clavicle remaining displaced posteriorly into the mediastinum, an open reduction is usually indicated and requires excision of the medial 1 inch of the clavicle and stabilization to the first rib as described earlier.

Shoulder

817

A u t h o r s ’ P r e f e r r e d M e t h o d f o r ­T r e a t m e n t o f T r a u ma t i c P r o b l e m s — c o n t ’ d

A

A

B

B Figure 17B-33   A and B, Drill holes are placed in the superior cortex of the clavicle, about 1 cm lateral to the osteotomy site. (A, Redrawn from Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996.)

Figure 17B-34   A and B, The free ends of the suture are passed into the medullary canal and out the two holes in the superior cortex. (A, Redrawn from Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996; B, from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

Continued

818

DeLee & Drez’s Orthopaedic Sports Medicine

A u t h o r s ’ P r e f e r r e d M e t h o d f o r ­T r e a t m e n t o f T r a u ma t i c P r o b l e m s — c o n t ’ d

A A

B Figure 17B-35   A and B, The transferred capsular ligament is secured into the medial clavicle by tying the sutures exiting from the superior cortex of the clavicle.   (A, Redrawn from Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996; B, from Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998.)

B Figure 17B-36   A and B, Closure of the periosteal sleeve around the medial clavicle and secure fixation of these structures to the costoclavicular ligament. (A, Redrawn from Rockwood CA, Green DP, Bucholz RW, Heckman JD [eds]: Fractures in the Adult. Philadelphia, JB Lippincott, 1996.)

Shoulder

Authors’ Preferred Method for Treatment Injuries of the Medial Clavicle We believe that many of the anterior and posterior injuries of the sternoclavicular joint in patients younger than 25 years, which are thought to be “dislocations” of the sternoclavicular joint, are injuries to the medial physis of the clavicle. Many authors have observed at the time of surgery that the intra-articular disk ligament stays with the sternum, and we agree with them. In addition, we submit that the unossified or ossified epiphyseal disk, depending on the age of the patient, also stays with the sternum. Anatomically, the epiphysis is lateral to the articular disk ligament, and it is held in place by the capsular ligament and can be mistaken for the intra-articular disk ligament. As previously described, the medial epiphysis does not ossify and thus does not appear on the radiograph until the 18th year. Therefore, the diagnosis cannot be “documented” until after ossification occurs. However, we would still perform the closed reduction maneuvers as described earlier for a suspected anterior or posterior injury. Open reduction of the physeal injury is seldom indicated except for the possibility of an irreducible posterior displacement in a patient who was having significant symptoms of compression of the vital structures in the mediastinum. After reduction, the shoulders are held back with a figure-of-eight strap or dressing for 3 to 4 weeks. In 1966, the senior author first treated a 16-year-old boy with a fracture of the medial one third of the clavicle and a “dislocation” of the sternoclavicular joint that turned out to be a type I epiphyseal injury. Because no references on this problem could be located, it was thought to be an original observation. The case report was prepared for publication, but a good friend of the senior author, Lee Rogers, M.D. (who was later to become Professor and Chairman of the Department of Radiology at Northwestern Medical School), recommended review of a text by John Poland that had been written in 1898. As happens with most “new ideas” in orthopaedics, the observation was not new. Poland,13 in his text entitled Traumatic Separation of Epiphyses of the Upper Extremity, described the entity in detail. He reviewed several French articles that evaluated more than 60 cases of fracture of the medial epiphysis of the clavicle that clinically appeared as sternoclavicular dislocations. He discussed the anatomy of the joint and the classifications of the injury and described methods of treatment. He ­described in detail an article by Verchere, written in 1886, which probably represents the first published report of a death caused by posterior dislocation of the ­sternoclavicular joint:

A 20-year-old man, following a severe crushing injury, died on the seventh day, of subcutaneous emphysema. Autopsy revealed that the inner third of the right clavicle was detached of its periosteum, and its smooth rounded end was displaced posteriorly and had produced a perforation of the pleural sac about the size of a 2-franc piece. The hole was in the left lung, whereas the ipsilateral lung escaped injury. The report very clearly described that the sternoclavicular joint was not

of

819

Physeal

injured and that the 5-mm-wide epiphyseal plate was held firmly in place by the ligaments about the joint. The specimen of the separated epiphysis was placed in the Dupuytren Museum. Poland went on to describe that the capsular ligament is primarily attached to the epiphysis of the medial ­clavicle and stated that with injury, the epiphysis is held by the capsular ligament and stays with its articulation with the sternum. In treating the 16-year-old boy with the fracture of the medial clavicle and the “dislocation” of the sternoclavicular joint, the senior author figured that he needed to, at least, line up the clavicle and put the sternoclavicular joint in better approximation. The fragment was 3.7 cm long and had rotated 90 degrees from the long axis of the clavicle. In eagerness to explore the anatomy, the fragment of clavicle, which had been completely stripped of its periosteum, fell onto the floor. However, the sternal end of the fragment did not have a smooth cartilaginous articular surface. It was rough and had the appearance of the end of a chicken leg when your last bite took off the epiphysis. The costoclavicular ligaments were intact to the periosteum inferiorly, and in the most medial corner of the periosteal tube was a dense structure that could be taken for the intra-articular disk ligament—or it could have been the unossified epiphysis. Treatment consisted of closing the periosteal tube and hoping that the epiphysis was still present adjacent to the sternum. Later, microscopic studies of the most medial end of the fragment revealed the provisional zone of calcification of the metaphysis, indicating that indeed there had been a separation through the physeal plate and that it had occurred through the zone of hypertrophy. Serial radiographs revealed a gradual replacement of the medial clavicle, and after 18 months, the entire defect had been replaced with bone. The patient returned to his former job as a manual laborer. Since 1966, in other cases of “dislocation of the sternoclavicular joint,” we have been able to document with the 40-degree cephalic tilt radiograph and computed tomographic scans that the injury really was a physeal fracture because the thin wafer-like disk remained in its normal articulation with the sternum, whereas the metaphysis and shaft were displaced. Some of the physeal fractures have been type II injuries, with a small fragment of the metaphysis remaining with the epiphysis. Obviously, before the epiphysis ossifies at the age of 18 years, one cannot be sure whether a displacement around the sternoclavicular joint is a dislocation of the sternoclavicular joint or a fracture through the physeal plate. Despite the fact that there is significant displacement of the shaft with either a type I or type II physeal fracture, the periosteal tube remains in its anatomic position, and the attaching ligaments—the costoclavicular ligaments inferiorly and the capsular and inter-articular disk ligaments ­medially—are intact to the periosteum.

820

DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method o f A t r a u ma t i c D i s o r d e r s

for

Treatment

Spontaneous Subluxation or ­ Dislocation of the Sternoclavicular Joint

We have seen 39 patients with this problem, and about the only symptom that they have is that the medial end of the clavicle subluxes or dislocates anteriorly when they raise their arms over their heads.128 This occurs spontaneously and without any significant trauma. It might be considered a voluntary or an involuntary problem because it occurs whenever the patient raises the arms to the overhead position. Some patients seen for another shoulder problem are completely unaware that with the overhead motion, the medial end of the clavicle subluxes or dislocates. We have never seen a posterior spontaneous subluxation of the sternoclavicular joint. Only occasionally does the patient with atraumatic anterior displacement complain of pain during the displacement. Because it is difficult to stabilize the joint and prevent the subluxation or dislocation and end up with a pain-free range of motion, we manage the problem with nonoperative skillful neglect. The anatomy of the problem is explained to the patient and the family. We explain further that surgery is of little benefit, that they should discontinue the voluntary aspect of the dislocation, and that in time the symptoms will either disappear or they will completely forget that the dislocation is a problem. In the review by Rockwood and Odor of 37 patients with spontaneous atraumatic subluxation, 29 were managed without surgery and 8 were treated with a surgical reconstruction elsewhere.128 With an average follow-up of more than 8 years, all of the 29 nonoperated patients were doing just fine without limitations of activity or lifestyle. The eight patients who were treated with surgery had increased pain, limitation of activity, alteration of lifestyle, persistent instability, and a significant scar. In many instances, the patient, before a previous reconstruction or resection, had minimal discomfort, an excellent range of motion, and complained only of the “bump” that slipped in and out of place with certain motions. Postoperatively, the patient still had the bump, along with a scar and a painful range of motion. Arthritis

Most patients with simple degenerative arthritis or postmenopausal osteoarthritis can be managed with rest, moist heat, and anti-inflammatory medications. As previously described, it is important to do a good work-up of the patient with arthritis of the sternoclavicular joint to rule out tumor, arthropathies, condensing osteitis of the medial clavicle, and sternocostoclavicular hyperostosis. Patients with post-traumatic osteoarthritis or sternocostoclavicular hyperostosis may require resection of the medial clavicle. Care must be taken to remove sufficient, but not too much, bone of the medial clavicle. Operative Management of Atraumatic Sternoclavicular Conditions

When operating on the sternoclavicular joint, care must be taken to evaluate the residual stability of the medial clavicle. It is the same analogy as used when resecting the distal

clavicle for an old acromioclavicular joint problem. If the coracoclavicular ligaments are intact, an excision of the distal clavicle is indicated. If the coracoclavicular ligaments are gone, then in addition to excision of the distal clavicle, you must reconstruct the coracoclavicular ligaments. If the costoclavicular ligaments are intact, the clavicle medial to the ligaments should be resected and beveled smooth. If the ligaments are gone, the clavicle must be stabilized to the first rib. If too much clavicle is resected, or if the clavicle is not stabilized to the first rib, an increase in symptoms can ­occur. With the patient supine on the operating table, place three to four towels or a sandbag between the scapulae. Make an anterior incision that parallels the superior border of the medial 7 to 8 cm of the clavicle. The incision then extends downward over the sternum just medial to the involved sternoclavicular joint. Incise the fascia and periosteum of the medial clavicle in line with the skin incision. The clavicular head of the sternocleidomastoid muscle and the medial clavicular origin of the pectoralis major muscle are reflected subperiosteally to expose the sternoclavicular joint. When satisfactory exposure has been obtained, incise the anterior capsule and inspect the sternoclavicular joint. This is facilitated by débridement of the intra-articular disk ligament, which is often frayed and degenerative in appearance (see Fig. 17B-21). Ordinarily, the intra-articular disk ligament is a very dense, fibrous structure that arises from the synchondral junction of the first rib to the sternum and passes through the sternoclavicular joint, dividing the joint into two separate spaces. The upper attachment of the ligament is on the superior and posterior aspects of the medial clavicle. Anteriorly and posteriorly, the disk blends into the fibers of the capsular ligament and acts as a checkrein against medial displacement of the inner clavicle. After the ligament is excised, the joint is carefully re-examined. If further inspection reveals degenerative changes of the sternoclavicular articular joint surfaces, excise the medial 1.5 to 2 cm of the clavicle, being careful not to damage the costoclavicular (rhomboid) ligament or the vascular structures that are located posteriorly to the medial clavicle and sternoclavicular joint. These vital structures are protected by passing a curved Crego or ribbon retractor around the posterior aspect of the medial clavicle, which isolates them from the operative field during the bony resection. Excision of the medial clavicle is facilitated by creating drill holes through both cortices of the clavicle at the intended site of clavicular osteotomy. Following this step, an air drill with a side-cutting burr or an osteotome is used to complete the osteotomy. The anterior and superior corners of the clavicle are beveled with an air bur for cosmetic purposes, and the periosteum is closed around the remaining clavicle. In 1997, Rockwood and associates129 reported a series of 23 patients who had undergone a resection of the medial end of the clavicle. The patients were divided into two groups: group I, those who underwent resection of the medial end of the clavicle with maintenance or reconstruction of the costoclavicular ligament; and group II, those who had a resection

Shoulder

821

Authors’ Preferred Method for Treatment o f A t r a u ma t i c D i s o r d e r s � —cont’d without maintaining or reconstructing the costoclavicular ligament. The outcome in all except one of the seven patients in group II was poor, with persistence or worsening of preoperative symptoms. The only patient in this group with a successful result suffered a posterior epiphyseal separation in which the costoclavicular ligament remains attached to the periosteum, thus preventing instability. All of the eight

COMPLICATIONS OF INJURIES TO THE STERNOCLAVICULAR JOINT A physician treating sternoclavicular instability faces many challenges. Complications of the injury itself, improper selection of treatment, potential intraoperative misadventures, and postoperative problems such as migration of hardware and iatrogenic instability can all threaten the surgical outcome and, at times, even the patient’s life. Thorough knowledge of the potential pitfalls is necessary if they are to be avoided.130 Serious complications that occur at the time of dislocation of the sternoclavicular joint are primarily limited to the posterior injuries. About the only complication that occurs with the anterior dislocation of the sternoclavicular joint is a “cosmetic bump” or late degenerative changes.131-133 Many complications have been reported secondary to the retrosternal dislocation: pneumothorax and laceration of the superior vena cava,134 venous congestion in the neck, rupture of the esophagus with abscess and osteomyelitis of the clavicle,135 pressure on the subclavian artery late in the patient who was untreated,44,136 compression of the right common carotid artery by a fracture-dislocation of the sternoclavicular joint,136 brachial plexus compression,72 and hoarseness of the voice, onset of snoring, and voice changes from normal to falsetto with movement of the arm (Figs. 17B-37 and 17B-38).29,73,135,137,138 Wasylenko

patients in group I who had a primary surgical resection of the medial end of the clavicle with maintenance of the costoclavicular ligaments had an excellent result. When the operation was performed as a revision of a previous procedure with reconstruction of the costoclavicular ligaments, the results were less successful, but only one of seven patients was not satisfied with the outcome of treatment.

and Busse139 reported a posterior dislocation of the medial clavicle that caused a fatal tracheoesophageal fistula. Gangahar and Flogaites81 reported a case of posterior dislocation of the clavicle that produced a severe thoracic outlet syndrome with swelling and cyanosis of the upper extremity. Nakayama and associates140 reported on a 17-year-old female with tracheal stenosis caused by a posterior dislocation of the right sternoclavicular joint. Rayan141 reported a thoracic outlet syndrome and brachial plexopathy caused by a chronic posterior sternoclavicular dislocation of 4 years’ duration. The injury was presumably missed owing to an associated dislocation of the glenohumeral joint. The patient experienced paresthesias and decreased sensation and at one time unknowingly burned her medial forearm with a hot iron. One year after medial clavicle excision and stabilization of the remaining clavicle to the first rib, she was asymptomatic except for mild discomfort related to the glenohumeral joint. Gardner and Bidstrup142 reported on three patients who had severe great vessel injuries following blunt chest trauma and posterior dislocation of the sternoclavicular joint. Two cases involved the innominate artery, and one case involved the carotid and subclavian arteries. Gale and associates143 reported a retrosternal dislocation of the clavicle associated with stridor and dysphagia. Several of our patients have had unusual complications that resulted from traumatic injuries to the sternoclavicular joint. One patient, as the result of a posterior dislocation

Figure 17B-37   Computed tomographic scan demonstrates posterior dislocation of the clavicle back into the mediastinum, displacing the trachea.

822

DeLee & Drez’s Orthopaedic Sports Medicine

and rupture of the trachea, developed massive subcutaneous emphysema (Fig. 17B-39). Another had an anterior dislocation on the right and a posterior dislocation on the left. When first seen, his blood pressure was very low. After reduction of the posterior dislocation, his blood pressure, as recorded on his monitor, instantly returned to normal

(Fig. 17B-40). It was theorized that the posteriorly ­displaced clavicle was irritating some of the vital structures of the mediastinum. Another patient, who had vascular compromise, is described in the section discussing the authors’ preferred method for treatment of posterior dislocations (see Fig. 17B-38). Another patient had a ­traumatic injury

A

B

C

D

E Figure 17B-38   Open reduction of a posterior dislocation of the left sternoclavicular joint causing compression of the great vessels in the mediastinum and resultant swelling in the patient’s left arm. A, Chest film does not suggest any serious problem with the left medial clavicle. B, Clinically, there was a depressed medial end of the left clavicle compared with the right. C, The computed tomographic scan reveals posterior displacement of the medial clavicle back into the mediastinum, compressing the great vessels and slightly displacing the trachea. D, The patient was carefully prepared for a surgical repair in cooperation with a cardiovascular surgeon. The patient was prepared from the base of the neck down to the umbilicus so that we could manage any type of vascular problem or complication. Open reduction was accomplished without any vascular incident. The medial end of the clavicle was totally unstable, so the medial 2 cm was resected and the remaining clavicle stabilized to the first rib. E, Four months after surgery, the patient’s slight anterior displacement of the clavicle was essentially asymptomatic, and the remaining clavicle was stable.

Shoulder

Figure 17B-39   Complications of sternoclavicular dislocation. As a result of posterior dislocation of the sternoclavicular joint, the patient had a lacerated trachea and developed massive subcutaneous emphysema. (From Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

to both sternoclavicular joints—the left was posterior, and the right was anterior. After reduction of the left posterior dislocation, the right side remained unstable. However, he eventually had painless full range of motion. Worman and Leagus,65 in an excellent review of the complications associated with posterior dislocations of the sternoclavicular joint, reported that 16 of 60 patients reviewed from the literature had suffered complications of the trachea, esophagus, or great vessels. We should point out that even though the incidence of complications was 25%, only four deaths have been reported as a result of the injury.13,92,144

COMPLICATIONS OF OPERATIVE PROCEDURES Because of the degree of motion at the sternoclavicular joint, tremendous leverage force is applied to pins that cross the sternoclavicular joint; such force causes migration of hardware and fatigue breakage of the pins. Through 2000, eight deaths,21,29,145-149 four near deaths,65,88,150,151 and six additional cases of intrathoracic K-wire migration12 had been reported from complications of transfixing the ­sternoclavicular joint with Kirschner wires or Steinmann pins. The pins, either intact or broken, migrated to the heart, pulmonary artery, innominate artery, or aorta. All but one of the seven deaths or near deaths were reported in the 1960s.21,29,65,88,147,150 One each was reported in 1974145 and 1992.149 To our knowledge, there were no deaths reported that occurred as a result of migrating pins from the sternoclavicular joint until the report in 1984 by Gerlach and colleagues146 from West Germany. They reported two deaths that resulted from migrating nails that

823

Figure 17B-40   Complications of sternoclavicular joint dislocation. This patient had an anterior dislocation on the right and a posterior dislocation on the left. As a result of the posterior dislocation, he had sufficient pressure on the mediastinal structures to cause significant hypotension. When the posterior dislocation was reduced, the blood pressure on the continuous monitor promptly returned to normal. (From Rockwood CA, Green DP [eds]: Fractures [3 vols], 2nd ed. Philadelphia, JB Lippincott, 1984.)

caused cardiac tamponade. The physicians were charged with manslaughter by negligence. We do not recommend any type of transfixing pins—large or small—across the sternocla-­ vicular joint. Brown88 has reported an incidence of three complications in 10 operative cases: two from broken pins that had to be removed via a window in the sternum, and one, a near death, in which the pin penetrated the back of the sternum and entered into the right pulmonary artery. Nordback and Markkula152 removed a pin that migrated completely inside the aorta. Jelesijevic and associates,153 Pate and Wilhite,150 Rubenstein and colleagues,154 and Schechter and Gilbert155 reported cases in which the pin migrated into the heart. Leonard and Gifford147 and Liu and coworkers156 reported on migration to the pulmonary artery. Sethi and Scott157 reported on migration of the pin to lacerate the subclavian artery. Ferrandez and associates158 described two cases of Kirschner wire migration into the mediastinum. Clark and associates145; S. Gaston, in the report by Nettles and Linscheid21; and Salvatore29 reported migration of pins into the aorta and resultant death. Grabski76 reported on the migration of the pin to the opposite breast in a 37-year-old woman. In addition, the senior author has treated patients in whom the pin has migrated into the chest and up into the base of the neck. Omer,22 in a review of 14 military hospitals, reported on 15 patients who had elective surgery for reduction and reconstruction of the sternoclavicular joint. Eight patients were followed by the same house staff for more than 6 months with the following complications: of the five patients who had internal fixation with metal, two developed osteomyelitis, two had fracture of the pin with recurrent dislocation, and one had migration of the pin into

824

DeLee & Drez’s Orthopaedic Sports Medicine

the mediastinum with recurrent dislocation. Of the three patients who had soft tissue reconstructions, one ­developed recurrent dislocation with drainage, one developed recurrent dislocation, and the third developed arthritis and extremity weakness and was discharged from military service. Omer commented on this series of complications: “It would seem that complications are common in this rare surgical problem.” To Omer’s comment we can only add, amen.

New bone with remodeling of clavicle

Sternum

Iatrogenic Instability Failure to preserve the coracoclavicular ligament when it is intact and failure to reconstruct it when it is deficient both severely compromise the surgical result. As noted in the previous discussion of surgical results, both Rockwood and Eskola159 noted vastly inferior results when the residual medial clavicle was not stabilized to the first rib, as well as an inability to obtain equivalent results when the rhomboid was reconstructed in delayed fashion. The other potential cause of iatrogenic instability is excessive resection that removes bone to a point lateral to the rhomboid ligament. This problem is an extremely difficult one that is best avoided. However, in these difficult cases, we have occasionally performed subtotal claviculectomy to a point just medial to the coracoclavicular ligaments. Although this technique leaves the extremity without a “strut” connecting it to the thorax, it can produce substantial relief of pain and improvement in motion and activity.130

SPECIAL CONSIDERATIONS IN THE PEDIATRIC ATHLETE Incidence The true incidence of injuries to the medial end of the clavicle in children is unknown. Knowledge of these injuries is contained in case reports and in small series that define a variety of treatments with short-term followup.78,96,106119,160-165 Rowe reports that fractures and dislocations of the medial clavicle constitute less than 6% of all injuries of the clavicle for all age groups.166 Rang estimates that injuries to the medial clavicle make up only 1% of all clavicle injuries in children.167 Nordquist and Petersson found that Allman type II injuries constitute 3% of all clavicle fractures.168

Classification Extraphyseal fractures of the medial clavicle in children have been reported,94,169 but most repaired injuries to this region in children are Salter-Harris type I and type II injuries (Fig. 17B-41).96,106,119,160,162,164,165 The ­ displacement can be anterior or posterior. Anterior displacements of sternoclavicular joint injuries occur twice as often as do posterior displacements15,107 in large series, but most case reports in the literature are for retrosternal displacements owing to the morbidity associated with these injuries. True dislocation of the sternoclavicular joint in children is rare. Most reported cases of younger patients are part of a larger

Figure 17B-41   Diagram depicting a Salter-Harris type I injury to the medial clavicular physis. Healing is by periosteal new bone formation with significant potential for remodeling.

series of adults and do not exclude Salter-Harris injuries as a possibility.88,163,170

Treatment As has been described earlier in this section, the epiphysis on the medial end of the clavicle is the last epiphysis in the body to appear on radiograph and the last one to close. The epiphysis on the medial end of the clavicle does not appear on radiographs until about the 18th year and does not unite with the clavicle until the 23rd to 25th year.7,8,18 If children and adults younger than 23 to 25 years of age have symptoms from the pressure of the posteriorly displaced clavicle into the mediastinum, an operative procedure should be performed. Children, however, may have no symptoms, and the physician can wait and watch to see whether the physeal plate remodeling process removes the posteriorly displaced bone.171 Indeed, as with other childhood fractures, the potential for remodeling is significant and may extend until the 23rd to 25th year. The senior author85 has demonstrated a similar mechanism to support conservative treatment of adolescent acromioclavicular joint injuries, or pseudodislocations, in which there is a partial tear of the periosteal tube containing the distal clavicle. The coracoclavicular ligaments remain secured to the periosteal tube. Because of its high ­osteogenic potential, spontaneous healing and remodeling to the preinjury “reduced” position occur within this periosteal conduit. Zaslav and associates86 have reported successful treatment of a posteriorly displaced medial ­clavicle physeal injury in an adolescent athlete with CT ­ documentation of remodeling, most probably within an intact periosteal tube. Similarly, Hsu and associates87 reported successful treatment of a posterior epiphyseal fracture-­dislocation of the medial end of the clavicle in a 15-year-old patient.

Anterior Displacement of the Medial Clavicle If the physeal injury is recognized, or if the patient is younger than 25 years, closed reduction, as has been described for anterior dislocation of the sternoclavicular

Shoulder

joint, should be performed. The shoulders should be held back in a ­ clavicular strap or figure-of-eight dressing for 3 to 4 weeks, even if the reduction is unstable. ­ Healing is prompt, and remodeling will occur at the site of the ­deformity.

Posterior Displacement of the Medial Clavicle Closed reduction of this injury should be performed in the manner described for posterior dislocation of the sternoclavicular joint. The reduction is usually stable with the shoulders held back in a figure-of-eight dressing or strap. Immobilization should continue for 3 to 4 weeks. If the posterior physeal injury cannot be reduced and if the patient is not having symptoms, one can treat the injury expectantly while remodeling occurs. Open reduction of the physeal injury is seldom indicated except for the irreducible posterior displacement in a patient with symptoms of compression of the vital mediastinal structures.

825

l Careful examination of the patient with a posterior sternoclavicular dislocation is important because complications are common. If specific complications are noted, appropriate consultants should be called in before reduction is performed. l Complications from a posterior sternoclavicular dislocation can happen early after the acute event to very late in the course of the injury. l Children and young adults have a tremendous capacity for remodeling of the physeal plate; the period for remodeling may extend to the 23rd to 25th year of life. l The portion of the clavicle medial to the costoclavicular (rhomboid) ligament should be removed when performing a sternoclavicular joint reconstruction or when excising a symptomatic intra-articular disk. l Transfixing pins across the sternoclavicular joint is never recommended.

CRITERIA FOR RETURN TO SPORT Return to noncontact sports can be attempted 6 weeks after injury or when the athlete has regained full, painless range of motion and the fracture site and medial clavicle are nontender. Contact sports should be avoided for an additional 4 to 6 weeks to allow remodeling and strengthening of the medial clavicle. The true incidence of medial clavicular physeal injury is unknown. Because the medial clavicular epiphysis does not ossify until 18 years of age, the diagnosis is difficult to verify radiographically. Because the capsule attaches mainly onto the epiphysis, however, it is conceivable that most “dislocations” in patients younger than 25 years of age are actually physeal separations rather than true dislocations. This distinction is clinically relevant because residual displacement in this injury can potentially diminish with time as remodeling occurs.

C l The

r i t i c a l

P

S U G G E S T E D

R E A D I N G S

Nettles JL, Linscheid R: Sternoclavicular dislocations. J Trauma 8:158-164, 1968. Omer GE: Osteotomy of the clavicle in surgical reduction of anterior sternoclavicular dislocation. J Trauma 7:584-590, 1967. Poland J: Separation of the epiphyses of the clavicle. Traumatic Separation of Epiphyses of the Upper Extremity. London, Smith, Elder, 1898. Rockwood CA Jr, Odor JM: Spontaneous atraumatic anterior subluxations of the sternoclavicular joint in young adults: Report of 37 cases. Orthop Trans 12:557, 1988. Rockwood CA, Odor JM: Spontaneous anterior subluxation of the sternoclavicular joint. J Bone Joint Surg Am 71:1280-1288, 1989. Rockwood CA, Groh GI, Wirth MA, Grassi FA: Resection arthroplasty of the sternoclavicular joint. J Bone Joint Surg Am 79:387-393, 1997. Spencer EE, Kuhn JE, Huston LJ, et al: Ligamentous restraints to anterior and posterior translation of the sternoclavicular joint. J Shoulder Elbow Surg 11(1):4347, 2002. Waskowitz WJ: Disruption of the sternoclavicular joint: An analysis and review. Am J Orthop 3:176-179, 1961. Wirth MA, Rockwood CA: Complications following repair of the sternoclavicular joint. In Bigliani LU (ed): Complications of the Shoulder. Baltimore, Williams & Wilkins, 1993, pp 139-153. Wirth MA, Rockwood CA: Acute and chronic traumatic injuries of the sternoclavicular joint. J Am Acad Orthop Surg 4(5):268-278, 1996.

o i n t s

posterior portion of the capsular ligament is heavier and stronger than the anterior portion and is the most important restraint to anterior and posterior translation of the sternoclavicular joint. l CT is the best technique to study any or all problems of the sternoclavicular joint. It clearly distinguishes injuries of the joint from fractures of the medial clavicle and defines minor subluxations of the joint.

R eferences Please see www.expertconsult.com

826

DeLee & Drez’s Orthopaedic Sports Medicine

S ect i o n

C

Injuries to the Acromioclavicular   Joint in Adults and Children Matthew T. Provencher, Augustus D. Mazzocca, and Anthony A. Romeo

The acromioclavicular (AC) joint anchors the clavicle to the scapula. About 9% of shoulder girdle injuries involve damage to the AC joint. Studies have shown that most AC joint injuries occur in adults in their 20s (43.5%), that AC ­dislocation is overwhelmingly more common in males (5 to 1), and that these dislocations are more often incomplete than complete (2 to 1).1 This section addresses injuries to the AC joint as well as other entities such as degenerative AC joint disease, atraumatic osteolysis of the distal clavicle, and intra-articular distal clavicle fractures as they pertain to athletics. The anatomy and design of the AC joint make it a resilient joint that can accept a significant amount of force before disruption. Numerous procedures and protocols have been devised to treat the AC joint. It is this multitude of research with various conflicting outcomes that can make choosing an appropriate treatment confusing. For this reason, it is important to understand the anatomy and biomechanics of the joint so that basic principles can be applied. An understanding of these basic principles allows for evaluation of certain clinical situations and application of a multitude of treatments to specific patient disorders and needs. This section defines and explains the anatomy and biomechanics of the AC joint and describes evaluation, diagnosis, and nonoperative and operative treatment of various disorders of the AC joint.

trapezius muscles. Fibers from the superior AC ligament blend with the fascia of the trapezius and the deltoid, adding stability to the joint when they contract or stretch. The AC joint capsule and the capsular ligaments are the primary restraints of the distal clavicle to anteriorto-posterior translation (Fig. 17C-2).5 These ligaments insert an average of 16.7 mm posterosuperiorly on the capsule and 12.8 mm anterosuperiorly. The maximal average distance is 16.1 mm from the joint line, and the greatest is 20 mm from the joint line. These numbers should be kept in mind during distal clavicle resection. These ligaments are primarily involved in horizontal stability.6 Posterior horizontal instability of the distal clavicle can cause abutment of the posterolateral portion of the clavicle into the spine of the scapula. Clinically, resistance to posterior translation is critical to avoid painful horizontal instability of the AC joint with abutment of the posterolateral end of the clavicle onto the spine of the scapula.7 Serial sectioning of the AC joint ligaments revealed that the superior ligament contributed 56% of the resistance to posterior displacement of the clavicle, with the posterior ligament contributing 25%. Consequently, surgical

RELEVANT ANATOMY AND BIOMECHANICS The AC joint is diarthrodial and has 6 degrees of freedom moving in the anteroposterior (AP) as well as the superoinferior planes. It is surrounded by a joint capsule that has synovium and an articular surface that is made up of hyaline cartilage containing an intra-articular meniscus-type structure (Fig. 17C-1). This intra-articular disk has tremendous variation in size and shape. DePalma, Petersson, and Salter and their colleagues have demonstrated that with age, this meniscal homologue undergoes rapid degeneration, until it is no longer functional beyond the fourth decade.2-4 Its actual function in the joint is negligible. Both static and dynamic structures stabilize the AC joint. The static stabilizers include the AC ligaments (superior, inferior, anterior, and posterior), the coracoclavicular ligaments (trapezoid and conoid), and the coracoacromial ligament. The dynamic stabilizers include the deltoid and the

Figure 17C-1   Normal anatomy of the AC joint. (From Rockwood CA Jr, Williams GR Jr, Young DC: Disorders of the acromioclavicular joint, In Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB [eds]: The Shoulder, 3rd ed. Philadelphia, Saunders, 2004, Fig. 12-3)

Shoulder

Figure 17C-2   Resection or injury of the AC ligaments causes horizontal instability and, if in excess, can cause abutment of the posterior clavicle into the anterior portion of the scapular spine.

t­ reatment of the AC joint should be designed to avoid rendering the superior and posterior ligaments incompetent. The coracoclavicular ligament’s main contribution is in vertical stability, preventing superior and inferior translation of the clavicle (Fig. 17C-3). This complex is made up of two structures: the trapezoid and the conoid ligaments. These ligaments span the space (1.3 cm) between the coracoid and the clavicle.8 The trapezoid is anterior and lateral to the conoid, and both the trapezoid and the

Figure 17C-3   Injury to both the coracoclavicular ligaments frequently occurs in the face of AC ligament injury and causes an inferior translation of the scapulohumeral complex from the clavicle. It is important to note in this illustration that the clavicle stays in its normal anatomic position, tethered by the sternoclavicular joint, and the scapulohumeral complex subluxates inferiorly.

827

conoid are posterior to the pectoralis minor attachment on the ­ coracoid. The coracoid insertion of the conoid ligaments is highly variable, and the distance from the lateral trapezoid ligament to the distal clavicle is about 15 mm. Bearden and colleagues9 reported a range of values for the ­coracoclavicular space of 1.1 to 1.3 cm. This distance becomes ­clinically important when differentiating between incomplete and complete AC joint separations. The larger the distance between the coracoid and the clavicle, the more likely it is that a complete dislocation has occurred. The ­coracoclavicular ligaments perform two major ­ functions: (1) they mediate synchronous scapulohumeral motion by attaching the clavicle to the scapula, and (2) they strengthen the AC articulation. Fakuda and coworkers6 reported that with small displacements, the AC ligaments are the primary restraints to posterior (89%) and superior (68%) translation of the clavicle. With larger displacement, the conoid ligament was found to be the primary restraint (62%) to superior translation, whereas the AC ligaments were still the primary restraints to posterior translation. The trapezoid ligament was found to be the primary restraint to compression of the AC joint at both small and large displacements. Fakuda and coworkers6 stated, “If maximum strength of healing after an injury to the AC joint is the goal, all ligaments should be allowed to participate in the healing process.” Urist10 determined that the AC ligament was the primary restraint to anterior and posterior displacement, and the coracoclavicular ligament, specifically the conoid, resulted in an overall superior displacement or an inferior displacement of the entire scapulohumeral complex. One of the largest concerns after distal clavicle resection, even if performed arthroscopically, is the potential for residual instability of the distal clavicle, especially in the posterior direction. Corteen and Teitge11 investigated the effect of ligament augmentation on posterior translation of the clavicle after a 1-cm resection. Pure resection without repair yielded a 32% increase in posterior translation (5.6 to 7.4 mm) from the intact state, which was not improved after capsular repair but did decrease with both capsular repair and coracoacromial ligament augmentation. It has been hypothesized that residual pain after a distal clavicle excision may be due to resultant instability, and protection of the soft tissue envelope should not be overlooked.7 Recent studies have shown that the AC and, especially, the coracoclavicular ligaments do not act in isolation. When the AC ligaments have been resected, the coracoclavicular ligaments have been shown to take an increased load, especially with anterior and posterior displacement. The conoid increases its force significantly with an anterior load.12 Finally Debski and colleagues,12 advancing on past research, recommended that the conoid and trapezoid ligaments should not be considered as one structure when surgical treatment is considered, and that capsular damage resulted in a shift of load to the coracoclavicular ligaments. They also reported that the intact coracoclavicular ligaments could not compensate for the loss of capsular function during AP loading, as occurs in type II AC joint injuries. Klimkiewicz and associates7 confirmed and reported that the superior and posterior AC capsule ligaments are the most important in preventing posterior translation of the clavicle to the scapula.

828

DeLee & Drez’s Orthopaedic Sports Medicine

35°

45°

35°

A

B

C

Figure 17C-4   Motions of the clavicle and the sternoclavicular joint. A, With full overhead elevation, the clavicle is elevated 35 degrees. B, With adduction and extension, the clavicle displaces anteriorly and posteriorly 35 degrees. C, The clavicle rotates on its long axis 45 degrees as the arm is elevated to full overhead position. (From Rockwood CA, Green DP [eds]: Fractures in Adults, 2nd ed. Philadelphia, JB Lippincott, 1984.)

The coracoacromial ligament is important as a secondary glenohumeral stabilizer to prevent anterosuperior displacement of the humeral head in longstanding massive rotator cuff disease (cuff tear arthropathy).

Motion of the Acromioclavicular Joint Rockwood and colleagues reported that there is about 5 to 8 degrees of motion detected at the AC joint, with forward elevation and abduction to 180 degrees.1 The clavicle rotates between 40 and 50 degrees during full overhead elevation (Fig. 17C-4). This motion is combined with scapular rotation rather than occurring through the AC joint. This synchronous motion between the clavicle, which is rotating upward as the scapula rotates downward during abduction, and forward elevation was described by Codman13 as synchronous scapular-clavicular rotation. The coracoclavicular ligaments coordinate this. The motion of the AC joint is important clinically in that both fusion of the AC joint and implantation of a coracoclavicular screw to stabilize the clavicle to the scapula allow full forward elevation in abduction. This motion has also caused these screws and this hardware to migrate as well as to break over time.

CLASSIFICATION OF ACROMIOCLAVICULAR DISLOCATION The pathologic process of AC joint dislocations was originally described by Cadenat14; it involves sequential injury, beginning with the AC ligaments, extending to the coracoclavicular ligaments, and finally affecting the deltoid

and trapezial muscles and fascia. Tossy and colleagues later classified the injury as types I, II, or III.15 Rockwood expanded the classification in 1984 to include types IV, V, and VI (Fig. 17C-5).1 The expanded classification recognized a variety of complete AC joint dislocations. These classifications correlate with increasing soft tissue injury and are shown in Box 17C-1.

EVALUATION Clinical Presentation and History The mechanism of most AC joint injuries and distal clavicle fractures is simply direct trauma. The subcutaneous position of this joint, which does not have large amounts of muscle protecting it, theoretically puts it at an increased risk for injury. Direct trauma is caused by a fall or blow to the acromion with the arm adducted. The stability of the sternoclavicular joint transfers the energy to the AC and coracoclavicular ligaments (Fig. 17C-6). Indirect injury can result from a fall onto an adducted outstretched hand or elbow, which causes the humerus to translocate superiorly, driving the humeral head into the acromion and causing damage.

Physical Examination and Testing Pain originating from the superoanterior aspect of the shoulder may be challenging to localize to one specific structure. A likely explanation of this phenomenon is the innervation of the AC joint and the superior aspect of the glenohumeral joint. The lateral pectoral nerve provides sensation to the anterior aspect of the shoulder. Gerber

Shoulder

829

NORMAL

TYPE I

TYPE IV

TYPE II

TYPE V

Conjoined tendon of biceps and coracobrachialis

TYPE III

TYPE VI

Figure 17C-5   The classification of the ligamentous injuries that can occur to the AC joint. In a type I injury, a mild force applied to the point of the shoulder does not disrupt either the AC or coracoclavicular ligament. In a type II injury, a moderate to heavy force applied to the point of the shoulder disrupts the AC ligaments, but the coracoclavicular ligaments remain intact. In a type III injury, when a severe force is applied to the point of the shoulder, both the AC and the coracoclavicular ligaments are disrupted. In a type IV injury, not only are the ligaments disrupted, but the distal end of the clavicle is displaced posteriorly into or through the trapezius muscle. In a type V injury, a violent force applied to the point of the shoulder not only ruptures the AC and coracoclavicular ligaments but also disrupts the muscle attachments and creates a major separation between the clavicle and the acromion. A type VI injury is an inferior dislocation of the distal clavicle in which the clavicle is inferior to the coracoid process and posterior to the biceps and coracobrachialis tendons. The acromioclavicular and coracoclavicular ligaments have also been disrupted. (From Rockwood CA Jr, Williams GR Jr, Young DC: Disorders of the acromioclavicular joint, In Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB [eds]: The Shoulder, 3rd ed. Philadelphia, Saunders, 2004, Fig. 12-14).

and colleagues16 evaluated patterns of pain and found that irritation to the AC joint produced pain over the AC joint, in the anterolateral neck, and in the region of the anterolateral deltoid (Fig. 17C-7). Irritation of the subacromial space produced pain in the region of the lateral acromion and the lateral deltoid muscle but did not produce pain in the neck or the trapezius region.

The history and the mechanism of injury are important in making an accurate diagnosis (Box 17C-2). A direct blow to the AC joint or a fall on the elbow forcing the head of the humerus into the AC joint is the mechanism associated with an AC separation. A painful symptomatic AC joint without a history of discrete injury resulting in separation of the joint can generally be termed an arthrosis

830

DeLee & Drez’s Orthopaedic Sports Medicine

Box 17C-1 Classification Type I: AC ligament sprain with the AC joint intact Type II: AC ligament tear, coracoclavicular ligament intact; AC joint subluxated Type III: AC and coracoclavicular ligaments torn; 100% dislocation of joint Type IV: Complete dislocation with posterior ­displacement of distal clavicle into or through the trapezius muscle Type V: Superior dislocation of the joint of 100% to 300%, increasing the coracoclavicular ligament distance 2 to 3 times, including disruption of the deltotrapezial fascia Type VI: Complete dislocation with inferior displacement of distal clavicle into a subacromial or subcoracoid ­position

(degenerative condition). A triad of point tenderness, positive pain at the AC joint with cross-arm adduction, and relief of symptoms by injection of a local anesthetic agent identify pathologic processes of the AC joint. The crossarm adduction test is performed with the arm elevated to 90 degrees and then adducted across the chest with the elbow bent at about 90 degrees. This cross-arm adduction produces pain specifically at the AC joint (Fig. 17C-8A). It may sometimes produce pain in the posterior aspect of the shoulder associated with a tight posterior capsule or at the lateral aspect of the shoulder, which can also be associated with rotator cuff disease. The reason that the cross-arm adduction test causes pain at the AC joint specifically is compression across the AC joint with that motion. O’Brien (see Fig. 17C-8B and C) recommended the active compression test for diagnosis of AC joint abnormalities. O’Brien’s test may be particularly helpful when

A

attempting to differentiate symptoms of AC joint arthrosis from intra-articular disease, especially with lesions of the superior glenoid labrum. The test is performed with the arm elevated to 90 degrees, the elbow in extension, adduction of 10 to 15 degrees, and maximal pronation of the forearm with obligate internal rotation of the arm. The examiner applies a downward force resisted by the patient. Symptoms referred to the top of the shoulder and confirmed by examiner palpation of the AC joint indicate damage to this structure. Symptoms referred to the anterior glenohumeral joint suggest labral or biceps disease. Walton and colleagues17 recently defined the accuracy for determining AC arthrosis. They describe using Paxinos’ test (thumb pressure at the posterior AC joint) and a bone scan to accurately assess damage to the AC joint. O’Brien recommended the active compression test for diagnosis of AC joint abnormalities. The O’Brien test may be particularly helpful when attempting to differentiate symptoms of AC joint arthrosis from intra-articular pathology, especially with lesions of the superior glenoid labrum. The test is performed with the arm elevated to 90 degrees, elbow in extension, adduction of 10 to 15 degrees, and a maximal pronation of the forearm with obligate internal rotation of the arm. The examiner applies a downward force resisted by the patient. Symptoms referred to the top of the shoulder and confirmed by examiner palpation of the AC joint indicate damage to this structure. Chronopoulos and associates18 found that the cross-body adduction test and AC resisted extension test (arm flexed 90 degrees, elbow bent 90 degrees, and patient asked to extend arm against resistance—pain in AC joint is positive) had the greatest sensitivity, whereas the active compression test had the greatest specificity and highest overall accuracy. Positive combinations of these tests increased the accuracy. Other conditions associated with AC joint pain include pseudogout and synovial ­chondromatosis. ­Aseptic

B

Figure 17C-6   Mechanism of AC joint injury. A, Direct trauma as result of a fall or blow to the acromion with the arm adducted.   B, Indirect injury caused by falling on an adducted outstretched hand or elbow, causing the humerus to translocate superiorly.

Shoulder

831

Box 17C-2 Typical Findings Suprascapular n.

C5 C6 C7 C8 T1

Lateral pectoral n.

A

B Figure 17C-7   A, Innervation of the AC joint in the superior aspect of the glenohumeral joint is by two nerves. The lateral pectoral nerve provides sensation to the anterior aspect of the shoulder. The suprascapular nerve provides innervation of the posterior aspect of the AC joint and posterior structures.   B, Superficial pain pattern produced by irritation of the AC joint and irritation of the subacromial space. (From Gerber CR, Galantay R, Hersche O: The pattern of pain produced by irritation of the acromioclavicular joint and the subacromial space. J Shoulder Elbow Surg 7[4]:352-355, 1998.)

i­nflammation in the joint has been reported in Crohn’s disease. AC joint cysts have been associated with glenohumeral arthritis, and there is significant involvement of the AC joint with rheumatoid arthritis. The following are the basic mechanisms and the radiographic and clinical examination findings for the six types of AC joint injuries (Table 17C-1).

Type I Direct force to the shoulder produces a strain to the AC ligament. The coracoclavicular and AC ligaments are intact, and the radiographic examination is normal.

Mechanism of Injury Direct trauma • Fall or blow to acromion of adducted arm • Stability of sternoclavicular joint transfers energy to AC and coracoclavicular ligaments Indirect trauma • Fall on adducted outstretched hand or elbow • Humerus translocates superiorly—driven into ­acromion Nontraumatic or chronic overuse • Suggests AC arthrosis—weightlifting, repetitive overhead or throwing activities Physical Examination Diffuse shoulder pain—anterolateral neck, AC joint, anterolateral deltoid Point tenderness at AC joint Positive cross-arm adduction test (arm flexed 90 degrees, adducted across chest) produces compression and pain in AC joint O’Brien’s active compression test may localize pain to AC joint Paxinos test (thumb pressure at the posterior AC joint) Diagnostic injection into AC joint relieves pain Positive combinations of above tests are additive and increase accuracy of diagnosis Radiographic Findings Zanca view best to delineate AC joint (AP shoulder view usually overpenetrates the joint) X-ray beam 10 to 15 degrees to head and using 50% of the AP penetration strength Stress views may be helpful in chronic situation Five pounds placed on bilateral wrists (adducted arms), one radiograph to compare to contralateral side Useful to differentiate between type II and III injuries Not necessary with obvious AC injury and higher order deformities (type IV, V or VI)

Type II Increased force to the point of the shoulder is severe enough to rupture the AC joint but not severe enough to rupture or affect the coracoclavicular ligaments. There is pain with motion, and the distal clavicle is unstable in the horizontal plane. On radiographic examination, the lateral end of the clavicle may be slightly elevated; however, stress views fail to demonstrate a 100% separation of the clavicle and acromion (Fig. 17C-9).

Type III This injury is a complete disruption of both the AC and the coracoclavicular ligaments without significant disruption of the deltoid or trapezial fascia. The upper extremity is usually held in an adducted position with the acromion depressed while the clavicle appears “high riding.” The clavicle is unstable in both the horizontal and vertical planes, and stress views on radiographic examination are

832

DeLee & Drez’s Orthopaedic Sports Medicine

10°

90°

A

B

10°

90°

C Figure 17C-8   A, Cross-arm adduction test is performed with the arm flexed to 90 degrees in adduction across the body with a finger placed on the AC joint, indicating pain at that spot only. It is important to understand that this test may produce pain posteriorly if there is a tight posterior capsule decreasing internal rotation or in the glenohumeral joint if there is glenohumeral arthritis. It is positive for AC joint disease only if cross-arm adduction produces pain at the AC joint itself. B and C, O’Brien’s test is performed with the arm flexed to 90 degrees with the elbow in extension and adducted 10 to 15 degrees with maximal supination; it is then performed again in maximal pronation. Symptoms referred to the AC joint with either of these maneuvers or with the arm in supination indicate more of an AC joint disorder, whereas symptoms referred to the anterior glenohumeral joint that are increased in maximal pronation indicate more of a superior labral disorder.

abnormal. Although the clavicle appears high on the radiograph, in reality the acromion and the remainder of the upper extremity are displaced inferior to the horizontal plane of the lateral clavicle (Fig. 17C-10). Pain with motion is severe typically for the first 1 to 3 weeks.

Type IV The distal clavicle is posteriorly displaced into the trapezius muscle and may tent the posterior skin. The posteriorly displaced clavicle is easily seen on an axillary radiograph. It is important to evaluate the sternoclavicular joint because there can be an anterior dislocation of the sternoclavicular joint and posterior dislocation of the AC joint (Fig. 17C-11).

Type V This is a more severe form of a type III injury, with the entire trapezial and deltoid fascia being stripped off the acromion as well as the clavicle. It is manifested by a twofold to threefold increase in the coracoclavicular distance or a 100% to 300% increase in the clavicle-to-acromion radiographic distance. The shoulder manifests as a severe droop secondary to downward displacement of the scapula and humerus with stabilization of the clavicle (Fig. 17C-12). An effective physical examination marker to differentiate type III from type V can be noted by having the patient shrug the shoulders. If the AC joint reduces, this is not a type V injury; however, if it does not reduce, then it is,

Shoulder

833

  TABLE 17C-1  Type of Acromioclavicular Joint Injury and Associated Findings Type

AC Ligament Injury

CC Ligament Injury

Deltotrapezial fascia

Clinical Findings

Radiographic Findings

I II

Intact Ruptured

Intact Intact

Intact Intact

III

Ruptured

Ruptured

Mild injury

IV

Ruptured

Ruptured

V

Ruptured

Ruptured

Injured as clavicle is posteriorly displaced Injured and stripped off clavicle

Normal Lateral end of clavicle slightly elevated. Stress views < 100% separation Plain films and stress radiographs abnormal—100% separation. In reality, acromion and upper extremity are displaced inferior to lateral clavicle Clavicle displaced posteriorly on axillary view 100% to 300% increase in clavicle to acromion distance

VI

Ruptured

Ruptured

AC tenderness Pain with motion, clavicle unstable in horizontal plane Clavicle unstable in both horizontal and vertical planes, extremity adducted and acromion depressed relative to clavicle Possible skin tenting and posterior fullness More severe type III injury, shoulder with severe droop; if shoulder shrug does not reduce then type V injury Rare inferior dislocation of distal clavicle; accompanied by other severe injuries; transient paresthesias

Possible injury

Clavicle lodged behind intact conjoined tendon

AC, acromioclavicular; CC, coracoclavicular.

indicating that the trapezial and deltoid fasciae have been stripped off the clavicle.

Type VI This is an inferior dislocation of distal clavicle. Gerber and Rockwood have reported three cases.19 This injury is associated with severe trauma and frequently accompanied

A

by multiple other injuries. The mechanism is thought to be severe hyperabduction and external rotation of the arm combined with retraction of the scapula. The distal clavicle goes in one of two directions, either subacromial or subcoracoid. With the subcoracoid dislocation, the clavicle becomes lodged behind the intact conjoined tendon. In a type VI injury, transient paresthesias are present in most patients before reduction and subside afterward.

B

Figure 17C-9   Radiographic appearance of a type II AC joint injury to the right shoulder (A). With stress, the coracoclavicular distance in both shoulders measures 1.5 cm. In the injured right shoulder, however, the AC joint is widened compared with the normal left shoulder (B). (From Rockwood CA, Green DP [eds]: Fractures in Adults, 2nd ed. Philadelphia, JB Lippincott, 1984.)

834

DeLee & Drez’s Orthopaedic Sports Medicine

A

B

Figure 17C-10   Radiographic appearance of a type III injury to the right shoulder. Stress radiographs were made to compare the right (A) with the left (B) shoulder. Not only is the right AC joint displaced compared with the left, but also, and more significantly, there is a great increase in the coracoclavicular interspace in the injured right shoulder compared with the normal left shoulder. (From Rockwood CA, Green DP [eds]: Fractures in Adults, 2nd ed. Philadelphia, JB Lippincott, 1984.)

B

A

C

D

Figure 17C-11   Type IV dislocation of the AC joint. A, The AP radiograph reveals obvious deformity of the AC joint. The distal end of the clavicle appears to be inferior to the acromion. B, The axillary view confirms that the clavicle is displaced posteriorly away from the acromion process. C, Computed tomographic scan reveals that the left clavicle is in its normal position adjacent to the acromion. Note that the right clavicle is completely absent. D, A lower cut demonstrates the posterior displacement and also confirms the fact that the clavicle is inferior to the acromion. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, Saunders, 1990.)

Shoulder

835

Figure 17C-12   A, This is a type V injury and is differentiated from a type III injury by the twofold to threefold increase in the coracoclavicular distance as well as a stripping of the deltoid and trapezial fascia from the clavicle. This can be clinically differentiated from a type III injury by having the patient shrug his or her shoulders. If the AC joint reduces and the deformity diminishes, this indicates a type III injury. If the deformity persists and the AC joint does not reduce, this indicates a type V injury. This phenomenon is due to the stripping of the trapezial and deltoid fascia. B and C, Clinical photographs of patient with type V AC dislocation. A severe upward displacement of the right clavicle has occurred into the base of this patient’s neck (B). Note the severe upward displacement of the clavicle in this patient’s right shoulder   (C). There was so much tension on the skin that it was becoming necrotic. ( B and C, From Rockwood CA, Green DP [eds]: Fractures in Adults, 2nd ed. Philadelphia, JB Lippincott, 1984.)

A

Imaging and Radiologic Evaluation An important aspect of radiologic evaluation of the AC joint is that it requires one third to one half of the radiographic penetration required for the denser glenohumeral joint. This is why in a standard AP view of the shoulder, the AC joint will be overpenetrated (dark), and small or subtle disease may be overlooked. When the history and physical examination indicate possible disease in the AC joint, specific directions must be given to the radiology technician to perform the appropriate view. AP and lateral views are standard for the shoulder; however, a Zanca view is the most accurate in looking at the AC joint. This view is performed by tilting the radiologic beam 10 to 15 degrees toward the head and using only 50% of the standard shoulder AP penetration strength (Fig. 17C-13).

Stress View

B

Stress views are performed with 5 pounds placed on the wrists. A comparison is made on the AP radiograph between the coracoclavicular distance on the normal side and that on the investigative side. These are mainly used to differentiate between type II and type III injuries. The literature has shown that the usefulness of these views does not outweigh the added cost, patient discomfort, and time consumption. Patients who present with a clinically obvious AC injury and deformities suggestive of complete dislocation (types III, IV, V, or VI) often demonstrate maximal coracoclavicular interspace widening on routine AP views (Fig. 17C-14). When there is a normal coracoclavicular interspace but a complete dislocation of the AC joint, a coracoid fracture should be suspected. A Stryker notch view is helpful in diagnosing this condition (Fig. 17C-15).

Normal Radiologic Findings

C

The configuration of the AC joint on AP radiographs varies significantly. Zanca20 reported that the AC joint width is normally between 1 and 3 mm. Petersson3 reported that the AC joint space diminishes with increasing age, so a joint space of 0.5 mm is normal in 60-year-old patients. The coracoclavicular interspace can also exhibit variability. Bosworth21 stated that the average distance between the

836

DeLee & Drez’s Orthopaedic Sports Medicine

A

B

10°

X-ray

C

D Figure 17C-13   Explanation of why the AC joint is poorly visualized on routine shoulder radiographs. A, This routine AP view of the shoulder shows the glenohumeral joint well. The AC joint is too dark to be interpreted, however, because that area of the anatomy has been overpenetrated by the x-rays. B, When the usual exposure for the shoulder films is decreased by two thirds, the AC joint is well visualized. The inferior corner of the AC joint, however, is superimposed on the acromion process. C, Tilting the tube 15 degrees upward provides a clear view of the AC joint. D, Position of the patient for the Zanca view: a 10- to 15-degree cephalic tilt of the x-ray tube for visualizing the AC joint. (From Rockwood CA Jr, Williams GR Jr, Young DC: Disorders of the acromioclavicular joint. In Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB [eds]: The Shoulder, 3rd ed. Philadelphia, Saunders, 2004, Figs. 12-21 [A to C] and 12-22 [D].)

clavicle and the coracoid process is usually between 1.1 and 1.3 cm. Bearden and colleagues9 reported that an increase in the coracoclavicular distance of 25% to 50% more than on the normal side indicated complete coracoclavicular ligament disruption.

TREATMENT A multitude of opinions exist in the literature regarding the treatment of AC joint injuries (Table 17C-2). In most AC joint separations, incomplete injuries (types I and II) are treated nonoperatively with a sling, ice, and a brief period of immobilization, typically lasting 3 to 7 days. Complete AC joint injuries (types IV, V, and VI) are usually treated operatively owing to the significant morbidity associated with persistently dislocated joints and severe soft tissue disruption.

Treatment of type III injuries remains controversial, with a trend toward initial nonoperative treatment in most cases. This controversy is in part a result of early literature reports that evaluated all AC joint injuries using a type I through III classification system. Type III injuries included what are now considered types IV, V, and VI injuries by Rockwood’s addition.1 Rockwood and colleagues1 reported that type III injuries are usually treated nonoperatively, particularly in patients who participate in contact sports (football, hockey, soccer, and lacrosse), in which the risk for reinjury is high. There is a subset of patients who have persistent pain and are unable to return to their sport or job with nonoperative treatment. In these cases, successful surgical stabilization has allowed return to sport or work. Evidence supporting nonoperative treatment of type III AC dislocations has been provided by a meta-analysis.22

Shoulder

837

A

B Figure 17C-14   Technique of obtaining stress radiographs of the AC joint. A, Anteroposterior radiographs are made of both AC joints with 10 to 15 pounds of weight hanging from the wrists. B, The distance between the superior aspect of the coracoid and the undersurface of the clavicle is measured to determine whether the coracoclavicular ligaments have been disrupted. One large horizontally positioned 14- by 17-inch x-ray cassette can be used in small patients to visualize both shoulders on the same film. In large patients, it is better to use two horizontally placed smaller cassettes and take two separate films to obtain the measurement. The arrows indicate the inferior subluxation of the scapulohumeral complex. (From Rockwood CA Jr, Williams GR Jr, Young DC: Disorders of the acromioclavicular joint. In Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB [eds]: The Shoulder, 3rd ed. Philadelphia, Saunders, 2004, Figs. 12-24 [A] and 12-25 [B].)

In a review of 1172 patients, 88% of the operatively treated and 87% of the nonoperatively treated patients had satisfactory outcomes. Complications included the need for further surgery (59% operative versus 6% nonoperative), infection (6% versus 1%), and deformity (3% versus 37%). Pain and range of motion were not significantly affected.

The authors did not recommend surgery for type III AC joint injuries in young patients. In 1997, McFarland and associates23 published the results of a survey of major league baseball team physicians evaluating treatment modalities for a type III injury in a pitcher. Sixty-nine percent reported that they would opt

838

DeLee & Drez’s Orthopaedic Sports Medicine

X-ray beam

  TABLE 17C-2  Treatment Options

Treatment ­Classification AC ligament repair

10°

Figure 17C-15   Stryker notch view.

for nonoperative treatment. Of the 32 patients with type III injuries, 20 were treated nonoperatively and 12 operatively. There was complete pain relief and normal function in 80% of the nonoperatively treated and 91% in the operatively treated patients. Larson and Hede24 prospectively compared nonoperative and operative treatment with similar rates of persistent symptoms (2/25 [8%] operative versus 3/29 [10%] nonoperative).

Nonoperative Treatment Most type I and type II AC joint separations are treated in a nonoperative fashion, and type III injuries are usually evaluated on a case-by-case basis, taking into account hand dominance, occupation, heavy labor, position/sport requirements (quarterbacks, pitchers), scapulothoracic dysfunction, and a risk for reinjury. Types IV, V, and VI are generally treated operatively. There is some literature information to support reduction of the clavicle in types IV, V, and VI injuries, turning them into a type III injury and then treating them conservatively.25 The main goals of treatment, whether surgical or nonsurgical, are to achieve a pain-free shoulder with full range of motion, strength, and no limitations in activities. The demands on the shoulder will differ from patient to patient, and these demands should be taken into account during the initial evaluation. Gladstone and colleagues26 discussed a four-phase rehabilitation program for athletes. The four phases are as follows: (1) pain control, immediate protective range of motion, and isometric exercises; (2) strengthening exercises using isotonic contractions; (3) unrestricted functional participation with the goal of increasing strength, power, endurance, and neuromuscular control; and (4) return to activity with sports-specific functional drills.

Essentials of Repair

AC ligament is repaired with reinforcing pin(s), screw, or plate Dynamic Transfer of short muscle head of biceps transfer with or without coracobrachialis CA ligament Transfer of CA transfer ligament alone or in concert with other procedures CC ligament Traditionally repair Bosworth screw technique—wires, suture loops, and grafts have been described Distal clavicle Classically, the resection distal clavicle is with CC excised and CC is reconstruction reconstructed using CA ligament40

Clinical and Operative ­Considerations

Level of Evidence

Usually implant is removed

IV

Partial transfer of structures—may alter shoulder mechanics Preserve length of CA ligament

IV

IV

Usually requires second procedure for hardware removal

IV

Can also be a salvage procedure for persistent pain after AC dislocation (especially type I and II injuries)

IV

Distal clavicle See Table 17C-3 resection without CC reconstruction Arthroscopic Repair or Technical reports repair and reconstruction have described reconstruction arthroscopically efficacy of viewed from the procedure subacromial space of CC ligaments Anatomic Reconstruction of Potential reconstruction CC ligaments advantages of the CC using soft of improved ligaments tissue grafts to horizontal plane reapproximate stability the conoid and trapezoid ligaments

IV

VI

IV

AC, acromioclavicular, CA, coracoacromial; CC, coracoclavicular.

Phase One The first phase of nonoperative treatment (Fig. 17C-16) is to decrease pain, allowing early range of motion to nourish the cartilage and to maintain maximal soft tissue function. Ice and some short-term immobilization can be used in this phase to decrease pain and reduce inflammation. Active-assisted range of motion is begun as early as possible for shoulder internal-external rotation and ­elevationdepression of the arm in the plane of the scapula (30 to 45 degrees of abduction, 30 to 40 degrees of forward flexion). It is important for the patient to reach the range of motion where pain begins but not go beyond this. Arm elevation in abduction allows the clavicle to rotate upward, which stresses the AC ligament. This can further increase pain and inflammation, so the athlete is instructed not to ­perform this motion. Other motions to decrease the ­atrophy

Shoulder

839

exercises, allowing up to full forward flexion and internal and external rotation, are performed with 90 degrees of shoulder abduction as well as with the arm at the patient’s side. Strengthening exercises are directed toward the deltoid, trapezius, and rotator cuff. Press maneuvers such as the bench press or the military shoulder press are limited because they increase the stress in the AC joint. Gladstone and coworkers’ criteria26 for advancing from phase two to phase three are a nonpainful range of motion, no pain or tenderness on palpation, and strength that is 75% of the contralateral side.

Phase Three

Figure 17C-16   Phase 1: Nonoperative treatment. Activeassisted range of motion for external rotation.

of the surrounding muscular groups in the shoulder are shoulder flexion and internal and external rotation. These exercises are done in an isometric fashion so as not to cause the clavicle to rotate. The patient or athlete is transitioned to the second phase when the range of motion and forward elevation are relatively pain free or with minimal pain up to 140 degrees of flexion and maximal external rotation compared with the contralateral arm. The criteria to advance to phase two are (1) 75% full range of motion, (2) minimal pain and tenderness on palpation of the AC joint, and (3) a manual muscle test grade of 4 out of 5 for the anterior deltoid, middle deltoid, and upper trapezius.

The main goal of phase three (Fig. 17C-18) is to increase strength of the entire shoulder complex musculature. Specific exercises emphasized during this phase are isotonic dumbbell shoulder flexion, abduction, shrugs, and bench press. Wilke and coworkers27 described a complete list of upper extremity plyometric drills that also start during phase three (Fig. 17C-19). Transition to phase four, which is the last rehabilitative stage, involving sports-­specific exercises, is allowed when the patient achieves (1) full range of the motion, (2) no pain or tenderness, (3) satisfactory clinical examination, and (4) if available, isokinetic test data with close to 100% of strength and range of motion compared with the contralateral uninjured side. These isokinetic tests are performed at 180 degrees per second and 300 degrees per second. It is important to emphasize that patients with type III injuries treated nonoperatively versus operatively demonstrate no difference in strength at 2 years’ ­

Phase Two The main goal of phase two (Fig. 17C-17) is to advance a patient to full painless range of motion and to increase the strength in an isotonic arc. Active-assisted motion

Figure 17C-17   Phase 2: Advance to active-assisted motion exercises allowing up to full forward flexion. Internal and external rotation is performed with 90 degrees of shoulder abduction as well as with the arm at the patient’s side. The T bar is used for active-assisted range of motion, allowing the patient to participate in his or her care.

Figure 17C-18   Phase 3: Increased strength and endurance of both the scapula stabilizers and specific rotator cuff muscles are attained using Thera-Bands and isometrics.

840

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17C-19   Phase 4: The last rehabilitative stage involves sport-specific exercise and allows throwing.

follow-up.28 Schlegel and colleagues29 found that there was some strength deficit, especially with forward flexion, and that this did affect function in heavy laborers. If symptoms persist, including increased instability, impingement due to scapular dyskinesia, decreased strength, inability to get the arm into a cocking position in throwing, and pain, especially posterior instability with the clavicle abutting the anterior portion of the spine of the scapula, then operative procedures may be indicated.

Operative Techniques Despite the prevalence and the success of nonoperative management of AC joint injuries, much of the literature has focused on surgical treatment. Operative treatment of types IV, V, and VI is generally recommended because of morbidity associated with persistent marked displacement of the distal clavicle, although there are reports of good results with conservative management.30 A closed reduction maneuver should be attempted because these types of dislocations can sometimes be reduced into a position that mimics a type III injury, then treated nonoperatively. The literature is replete with surgical techniques to address complete AC dislocations, including primary repair of the coracoclavicular ligaments, augmentation with autogenous tissue (coracoacromial ligament), and augmentation with absorbable and nonabsorbable suture as well as prosthetic material, and has included coracoclavicular stabilization with metallic screws.1,10,21,31-41 The WeaverDunn technique using transfer of the coracoacromial ligament has been the most popular procedure in the acute and chronic injury.40 Several more recent reports have described good results with modifications of the Weaver-Dunn

t­ echnique.38,41,42 However, in two independent studies by Tienen and Weinstein and their colleagues, compromised results were observed in patients who had residual subluxation or dislocation after surgery.38,41 Ammon and associates43 performed a biomechanical study comparing the Bosworth screw to a poly-l lactic acid (PLLA) bioabsorbable screw and found that the Bosworth screw provided superior strength (native ligament, 340 N; PLLA screw, 272 N; and Bosworth screw, 367 N). From a biomechanical perspective, the importance of the coracoclavicular ligaments and AC ligaments in controlling superior and horizontal translations has been elucidated.6,7,12,44,45 In fact, failure to surgically reproduce the conoid, trapezoid, and AC ligament function with current techniques may explain the observed incidence of recurrent instability and pain.7,45 Several authors have advocated using a separate and potentially more robust graft source to improve surgical results.33,46 The use of a free autogenous or allograft tendon has been further supported biomechanically.35 Other reconstruction grafts, such as the lateral half of the conjoined tendon,37 have also been described. Anatomic reconstruction of the coracoclavicular ligaments has been shown to be biomechanically superior than previous surgical constructs.47 Other types of fixation have been biomechanically evaluated, including suture cerclage and suture anchors.48 Although none of these techniques fully restored native AC joint stability, they were all found to be superior to the Weaver-Dunn procedure. There have been at least eight basic types of surgery performed for repair or reconstruction of the AC joint: (1) AC ligament repair, (2) dynamic muscle transfer, (3) coraco­ acromial liga­ment transfer, (4) coracoclavicular ­ ligament repair, (5) distal clavicle resection with coracoclavicular reconstruction, (6) distal clavicle resection without coraco­ clavicular reconstruction, (7) anatomic reconstruction of the coracoclavicular ligaments, and (8) arthroscopic repairs. In addition, some authors have advocated combinations of these procedures.

Acromioclavicular Ligament Repair (Level IV Evidence) Sage and Salvatore49 advocated AC ligament repair and reinforcement of the superior AC ligament with joint meniscus (Fig. 17C-20). Many have recommended transarticular smooth or threaded pins to supplement repair.9,31,49-59 In a comparison of smooth pins, threaded pins, and a cortical screw by Eskola and associates,60 13 of 86 patients available for follow-up had symptomatic osteolysis, and 8 of these 13 patients were among the 25 who had been treated with a Bosworth screw. Other authors have reported on the use of an AC joint plate for complete separations.61-65 Results have been good or excellent in 60% to 94% of patients. Broos and colleagues61 compared the Wolter plate with the Bosworth screw and found no significant difference in outcome.

Dynamic Muscle Transfer (Level IV Evidence) Transfer of the short head of the biceps with or without the coracobrachialis has been described,66-72 usually with acceptable results (Fig. 17C-21). However, Skjeldal and

Shoulder

Figure 17C-20   Acromioclavicular ligament repair. The AC joint is fixed internally with two unthreaded Kirschner wires. The wires are generally removed about 8 weeks after surgery. (From Justis EJ Jr: Traumatic disorders. In Canale ST [ed]: Campbell’s Operative Orthopedics, 7th ed, vol 3. St. Louis, Mosby, 1987.)

colleagues72 reported 10 complications in 17 patients, including coracoid fragmentation, infection, and pain.

Coracoclavicular Ligament Repair (Level IV and Level V Evidence) Coracoclavicular ligament repair (Fig. 17C-22) was introduced by Bosworth21 in 1941, who referred to it as a screw suspension procedure, which he performed percutaneously. Tsou39 reported on 53 patients in 1989 who underwent percutaneous cannulated screw coracoclavicular fixation and found a 32% technical failure rate. In 1968, Kennedy73 reported on use of a coracoclavicular screw with AC débridement and trapeziodeltoid repair. Jay and Monnet74 reported on 31 patients who ­underwent ­coracoclavicular ligament repair and Bosworth screw Coracoacrominal ligament

­ xation with ­ deltotrapezial repair. Lowe and Fogarty75 fi used a similar technique in 21 patients. Bearden and associates9 and Albrecht76 recommended using wire loops around the clavicle and the coracoid. Many others have used loops of other material.77-81 Bunnell in 192882 and Lom in 198883 used fascia lata to reconstruct the coracoclavicular ligaments. There have been numerous recent reports of coracoclavicular ligament repair by polydioxanone (PDS) suture or cerclage.84-90 Clayer and colleagues91 found that a PDS coracoclavicular sling did not maintain reduction, but good results were obtained in 6 patients. Gohring and coworkers84 and Pfahler and associates90 separately compared PDS cerclage with other techniques. Gohring compared surgical treatment of 64 complete AC joint dislocations with three techniques: tension band, Wolter hook plate, or PDS cord (braided). Early postoperative complications occurred in 43% of patients treated by use of a tension band, 58% of those treated by use of a hook plate, and 17% of those treated by use of a PDS cord. AC joint instability at a 35-month average follow-up was seen in 32% with a tension band, 50% with the plate, and 24% with the PDS cord. The authors recommended limiting surgery to younger, athletic patients.

Coracoacromial Ligament Transfer (Level IV Evidence) Neviaser58 introduced coracoacromial ligament transfer without coracoclavicular ligament repair. Variations on this principle have been reported.14,31,55,57,92-95 Several authors have emphasized imbrication of the deltotrapezial fascia as part of any surgical treatment.51,52,96 De la Caffiniere and colleagues97 felt that transfer of the coracoacromial ligament, which they attributed to Cadenat,14 is usually too weak and too short for the treatment of AC dislocation. They used a reinforcement flap made by lateral supraclavicular detachment of the superior fibrous capsular sheath. Of 26 patients, all 19 who had undergone

Pectoralis minor

Biceps and coracobrachialis Figure 17C-21   Transfer of the short head of the biceps with or without the coracobrachialis, shown here as Dewar’s technique. The coracoid process is transferred with the attached muscles by screw fixation to the undersurface of the clavicle. (Adapted from Dewar FP, Barrington TW: The treatment of chronic acromioclavicular dislocation. J Bone Joint Surg Br 47:32, 1965.)

841

Figure 17C-22   Coracoclavicular ligament repair and Bosworth screw fixation with deltoid trapezial repair.

842

DeLee & Drez’s Orthopaedic Sports Medicine

reinforced repair had no recurrence of dislocation, whereas all 7 without reinforcement had recurrence. Kumar and colleagues98 treated 14 AC dislocations with coracoacromial ligament transfer and coracoclavicular fixation with a screw. All 14 patients had excellent or good results. Guy and associates99 treated 23 chronic separations with coracoacromial ligament transfer and a Bosworth screw. Nineteen of 23 showed good to excellent results. The 4 patients with fair or poor results had a previous distal ­clavicle resection. At the time of a complete AC dislocation, the coracoclavicular ligaments may avulse the coracoid instead of tearing. Hak and Johnson100 reported one case of coracoid avulsion in association with AC dislocation, which they treated nonoperatively with a favorable outcome. Previous reports indicated similar outcomes with either surgical or nonsurgical treatment. Hak and Johnson100 recommended treating an AC dislocation with coracoid avulsion as if it were an isolated type III injury. Eyres and colleagues101 reported treating 12 coracoid fractures that were not associated with AC dislocation. The 10 patients without extension into the glenoid were treated nonoperatively, with good results. Results after more severe injuries may not be as good. Verhaven and associates102 achieved a 71% good or excellent outcome in 28 patients treated surgically for type V injuries. Outcome was unrelated to reduction of the joint, osteolysis, or calcifications. Athletes involved in throwing or contact sports are sometimes considered a special case. Some argue that throwing requires an anatomic reduction of the AC joint. Recent reports of successful nonoperative treatment of major league baseball pitchers suggest that this is not the case. Therefore, the preferred treatment of type III injuries remains nonoperative, with surgical treatment reserved for those patients who present with persistent symptoms after 3 to 6 months, even in high-level athletes.

Distal Clavicle Resection and Coracoclavicular Ligament Reconstruction (Level IV Evidence) Distal clavicle resection is undertaken as a salvage procedure for persistent pain after AC dislocation, especially type I or II injuries, or as treatment of degenerative or osteolytic AC joint arthrosis. In either case, reports ­indicate that a high rate of success can be expected, although patients with fractures or instability do not demonstrate the same outcome. Distal clavicle resection was reported separately in 1941 by Mumford103 and by Gurd.104 Mumford excised the distal clavicle in patients with persistent subluxation and degenerative changes and emphasized the need for coracoclavicular ligament reconstruction when the distal clavicle was noted to be tender. In general, when the distal clavicle is unstable, distal clavicle resection is accompanied by coracoclavicular ligament reconstruction with or without augmentation. In a classic article by Weaver and Dunn40 published in 1972, 15 patients with type III injuries were treated with distal clavicle resection and coracoclavicular ligament reconstruction using the coracoacromial ligament (Fig. 17C-23). Rauschningn and coworkers105 reported that 18 patients all had stable painless shoulders after

this procedure. Kawabe and colleagues106 and Shoji and ­associates107 transferred the coracoacromial ligament with an acromial bone block to the distal clavicle and fixed it with a screw.

Modified Weaver-Dunn with Augmentation (Level IV Evidence) Operative treatment begins with a diagnostic glenohumeral and subacromial arthroscopy. Berg has shown that some AC pathology can manifest or have concurrent superior labral problems. These can be definitively diagnosed and treated with arthroscopy. The coracoacromial ligament is released, preserving its overall length. A suture can be placed in the end of the ligament and brought out the anterior portal. An arthroscopic distal clavicle excision is completed. Typically, this involves removing 2 to 3 mm from the medial edge of the acromion and 7 to 8 mm from the lateral edge of the clavicle for a space greater than 1 cm. A small sabertype incision is made, starting slightly medial and posterior to the AC joint and extending to just above the coracoid. A horizontal incision is made in the deltotrapezial fascia across the AC joint. The joint is completely exposed with an anterior and posterior ­ subperiosteal ­ dissection using a ­needle-tip bovie. Care is taken to maintain the strength of the periosteum and ­ deltotrapezial fascia to allow a secure ­anatomic closure. Two drill holes with a 1.6-mm drill are made 5 mm medial to the distal end of the clavicle. The coracoacromial ligament is dissected out, and a No. 2 permanent suture is placed into its end. This is then placed into the end of the clavicle and tied over the holes. Downward stabilization of the clavicle and upward reduction of the scapula are maintained during this procedure. A gracilis or semitendinosus autograft is then harvested in the usual fashion. Of note, if a previous procedure has taken these tendons, a section of fascia lata, palmaris longus, or hamstring allograft can be used. A drill hole is placed in the coracoid, and using a loop of suture or a Hughson suture passer, the autograft is passed though the hole, twisted in a figure-of-eight fashion, and tied to itself with permanent suture. Breslow and colleagues108 have shown biomechanically that similar stability can be provided by placing a suture around the base of the coracoid or placing suture anchors in the coracoid itself for augmentation.

Distal Clavicle Resection without Ligament Reconstruction (Level IV Evidence) Much of the recent literature involves the development of the technique for arthroscopic distal clavicle resection and comparison of classic open and new arthroscopic ­methods. A summary of recent literature regarding resection of the distal clavicle is presented in Table 17C-3. Snyder and associates109 and Levine and colleagues110 reported results of arthroscopic resection; their combined results were good or excellent in 92%. Many authors have ­ contributed to the development of techniques for arthroscopic distal clav­ icle resection.109,111-113 Eskola,60 Flatow,114 and Levine110 and their associates have all reported worse ­outcomes for

Shoulder

843

Coracoacromial Lig.

Incision in Langer’s Lines

Deltoid Capsule

A

B

D

C

E

F

Figure 17C-23   Dr. Charles Rockwood’s method of reconstructing a chronic type III, IV, V, or VI AC dislocation. A, The incision is made in Langer’s lines. B, The distal end of the clavicle is excised. C, The medullary canal is drilled out and curetted to receive the transferred coracoacromial ligament. D, Two small drill holes are made through the superior cortex of the distal clavicle. The coracoacromial ligament is carefully detached from the acromion process. E, With the coracoacromial ligament detached from the acromion, a heavy nonabsorbable suture is woven through the ligament. F, The ends of the suture are passed out through the two small drill holes in the distal end of the clavicle. The coracoclavicular lag screw is inserted, and when the clavicle is reduced down to its normal position, the sutures used to pull the ligament snugly up into the canal are tied. (From Rockwood CA Jr, Williams GR Jr, Young DC: Disorders of the acromioclavicular joint. In Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB [eds]: The Shoulder, 3rd ed. Philadelphia, Saunders, 2004, Fig. 12-60.)

r­ esection in patients with instability of the lateral clavicle. A stabilization procedure in addition to resection is ­indicated for patients with AC joint arthrosis and instability. Eskola reported poorer results for patients with a history of fracture of the distal clavicle.

Excessive posterior translation after distal clavicle resection can be associated with pain, and the AC joint capsule helps restrain this motion. Blazar and coworkers115 looked at translation of the clavicle after distal clavicle resection. Motion in the AP direction was 8.7 mm, compared with

  TABLE 17C-3  Clinical Indications, Treatment, and Results of Distal Clavicle Resection (DCR) Injury

Treatment

Outcome

Auge (1998)92 10 Cook & Tibone (1988)119 17 Eskola et al (1987)60 73

Study

No. of Patients

Osteolysis in weightlifters Type II in athletes Mixed

Arthroscopic DCR Open resection Open resection

Flatow et al (1995)114

41

Arthrosis vs instability

Arthroscopic resection

Levine et al (1998)110

24

Arthrosis

Arthroscopic resection

Novak et al (1995)117

23

Arthrosis

Open resection

Petchell et al (1995)118 Snyder et al (1995)109

18/39 50

Arthrosis without instability Arthrosis

Open resection Arthroscopic resection

Returned to sport after average of 3.2 mo 16 of 17 returned to sport 21 good, 29 satisfactory, 23 poor; poor result more common in fractures 27 of 29 good or excel for arthrosis; 7 of 12 instability 71% excellent, 16.5% good, 12.5% failures; failures in stability 18 of 23 good or excellent with normal motion and strength All satisfied 47 good or excellent

844

DeLee & Drez’s Orthopaedic Sports Medicine

3.2 mm on the contralateral side. Visual analog pain scores correlated with the amount of translation. Translation and pain did not correlate with amount of apparent joint space after surgery. Klimkiewicz and colleagues7 used a cadaver model to evaluate the contributions of the superior, inferior, anterior, and posterior portions of the capsule. Sectioning of the anterior and inferior capsular ligaments had no significant effect on posterior translation; however, sectioning of the superior and posterior ligaments had significant effects (56% and 25%, respectively). To avoid posterior translation, techniques that spare the posterior and superior capsular ligaments should be used. Branch and associates116 demonstrated in a cadaver model that only a 5-mm resection of the distal clavicle is required to ensure that no bone contact between the distal clavicle and the acromion occurs with elevation. They found no difference between removal of the superior or the inferior ligament for joint access. Also in a cadaver model, Matthews and coworkers88 compared arthroscopic with open distal clavicle resection. No significant differences were found between the two methods in terms of displacement.

Distal Clavicle Resection in Association with Acromioclavicular Ligament Injury (Level IV Evidence) There is conflicting evidence regarding the effect of resection on strength and range of motion. Auge92 reported that all 10 of their patients who underwent resection of the distal clavicle were able to return to their previous sports. Novak and colleagues117 reported no clinically perceptible loss of motion or strength in 18 of 23 patients, although objective strength testing was not performed using a calibrated measuring device. Petchell and associates118 found that motion and strength were not restored in their patients who underwent resection for arthrosis without instability. Although all their patients self-reported that they were satisfied, more than 50% had ongoing difficulties with activities of daily living, sleeping, and working. In addition, 29% of patients were unable to participate in their previous sports activities. Cook and Tibone119 reported on open resection in 17 athletes with type II AC joint separations and chronic pain. Sixteen of 17 returned to their previous level of activity, although some complained of decreased strength, which was seen on Cybex testing at low speed but not high speed.

Failed Distal Clavicle Resection (Level IV to Level V Evidence) The clinical evaluation of failed distal clavicle excision is often subtle and difficult to evaluate. Nicholson, in an evaluation of 28 patients with unstable clavicles secondary to an aggressive distal clavicle excision, found a painful click or pinch at the posterior AC joint with forward elevation at

Authors’ Preferred Method

of

and above 90 degrees.5 He also found reproduction of pain with forced posterior clavicle translation, trapezius spasm, and a manual AP translation of the distal clavicle of more than 1 cm. There remains some controversy regarding the development of AC joint symptoms after arthroscopic partial distal clavicle resections. Neer120 recommended the removal of any osteophytes from the inferior aspect of the distal clavicle when performing open subacromial decompressions. He opined that these osteophytes could contribute to narrowing of the space available for the rotator cuff. With the development of arthroscopic techniques for subacromial decompression, some surgeons have suggested removing “osteophytes” from the inferior clavicle. However, after arthroscopic acromioplasty, part of the native distal clavicle is exposed, and certainly some techniques have included removal of this inferior aspect of the clavicle in the “coplaning” ­procedure. Fischer and associates121 reviewed 183 subacromial decompressions and divided them into three groups. The group in which the distal clavicle was not co-planed and the group that had a formal arthroscopic distal clavicle resection had no postoperative symptoms referable to the AC joint. However, the group that included a partial distal clavicle resection (co-planing) along with the subacromial decompression showed a high incidence of postoperative AC joint symptoms (14 of 36, or 39%). Because of these results, an “all or none” philosophy has developed. In other words, the distal clavicle is left alone for routine subacromial decompression, or a formal distal clavicle resection is performed if the patient has significant AC pathology.

Arthroscopic Repairs (Level VI Evidence) Several technical reports122-125 have described the ability to reconstruct an AC joint dislocation arthroscopically, as first depicted by Wolf and Pennington.125 The basic premise is that the procedure is performed through arthroscopic visualization in the subacromial space to facilitate viewing of graft and suture material into the clavicle. The purported advantages122 are the ability to preserve the capsule of the distal clavicle (preserving the capsular ligaments), the fact that a staged procedure for screw removal is not required, and the use of smaller, more cosmetic incisions. Overall, the procedure allows for a transcoracoid-transclavicular loop technique122,123; however, this has yet to be clinically validated.

WEIGHING THE EVIDENCE Most evidence for treatment of AC joint injury has typically been level IV and above. Each section of treatment has been highlighted with relative levels of evidence to aid the reader in treatment recommendations to patients.

Treatment

Treatment of AC joint disease (Fig. 17C-24) always involves a detailed history, physical examination, and understanding of the patient’s goals and needs. Overall, type I and II

i­njuries are treated with the previously described nonoperative protocol. These athletes show a return to sports in 1 to 4 weeks depending on the amount of contact or collision

Shoulder

Authors’ Preferred Method

of

845

Treatment—cont’d

in the sport. Type III injuries are treated nonoperatively; however, close follow-up and scrutiny of stability and pain are important. Nonsurgical treatment for type III injuries is generally recommended126; however, after 3 to 6 months of persistent symptoms, pain, and limitations, operative intervention is warranted. Type IV through VI injuries are treated operatively with an anticipated return to sport 6 months after surgery. When discussing return to play, the type of sport, position, and level must be considered. It has been shown127,128 that reconstruction of the clavicle, with appropriate anatomic positioning of drill holes representing the conoid and trapezoid ligaments, results in a stronger construct, especially in anterior127,128 and posterior translation. Baker and colleagues127 demonstrated that as the drill hole in the clavicle moved more forward (anterosuperior to anteroinferior drilling vector), the overall stability was improved, especially in the anterior direction.

The use of grafts for reconstruction of the AC joint was first reported by Abbott and colleagues.129 In their study, an autologous semitendinosus graft was used to reconstruct the AC joint. Lee and associates35,130 found that semitendinosus graft stiffness reapproximated the native coracoclavicular ligament stiffness better than that of the AC ligament transfer. They also found no statistical difference in load-to­failure among the three tendon grafts tested (i.e., gracilis, toe extensors, and semitendinosus). They also found that stiffness after the suture and tape repairs was not significantly different from that after the tendon graft reconstruction. Others131,132 have reported on the use of allograft after failed surgical repair. Mazzocca and associates128 investigated the effect of anatomic coracoclavicular ligament reconstruction in a ­ cadaveric model by comparing it with a traditional ­Weaver-Dunn procedure or an arthroscopic reconstruction.

A

B

C

D

Figure 17C-24   A, Saber-type incision (dashed line) starting slightly medially and posterior to the AC joint and extended to just above the coracoid. This is accomplished after diagnostic glenohumeral arthroscopy and an arthroscopic distal clavicle excision have been performed. B, A horizontal incision is made in the deltoid trapezial fascia across the AC joint. The joint is completely exposed with careful anterior and posterior subperiosteal dissection to ensure full thickness of the flaps. C, The coracoacromial ligament is dissected off the acromion, and a No. 2 permanent Ethibond suture is placed in a Krakow-type manner. The clavicle is secured. D, A curet is used to hollow out a trough for the coracoacromial ligament in the distal clavicle, and a large hole is made in the clavicle about 1 cm from the edge using a 4.5-mm drill. It is then made larger either with a curet or using the drill as a reamer. Two smaller holes are made on either side of this with a 3.2-mm drill. Continued

846

DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method

of

Treatment—cont’d

They ­ demonstrated that the Weaver-Dunn procedure had significantly greater laxity than an anatomic coracoclavicular reconstruction or arthroscopic reconstruction. Additionally, they showed that anterior and posterior translation was markedly improved with the anatomic techniques, even though superior displacement was not significantly different

E

among the groups. The clinical and biomechanical success of using semitendinosus autografts or allografts allowed us to modify existing techniques in the literature and reconstruct the trapezoid and conoid ligaments in an “anatomic” ­position. In the remaining sections, we describe the ­anatomic coracoclavicular ligament reconstruction.

F

G Figure 17C-24—cont’d   E, The two limbs of the Ethibond suture tied around the coracoacromial ligament are then placed through the small holes, and the ligament is placed into the bone tunnel. (Inset shows this with magnification.) Upward pressure is placed on the scapulohumeral complex to reduce the coracoclavicular distance. A suture can be placed around the coracoclavicular ligaments if possible. F, A tenaculum clamp is used to hold the reduction that has been accomplished with upper traction on the scapulohumeral complex. This facilitates easy and tight fitting of the coracoacromial ligament into its bone tunnel and passage of the semitendinosus graft. G, Placement of the semitendinosus graft and the braided Ethibond suture in a figure-of-eight pattern around the base of the coracoid and through the large hole in the distal clavicle, with concomitant fixation of the coracoacromial ligament in the bone tunnel. Of note, and not shown in these illustrations, is repair of the deltotrapezial fascia with nonabsorbable sutures, which is extremely important.

Anatomic Coracoclavicular Ligament Reconstruction (Level V Evidence) Based on past biomechanical studies, in addition to anatomic and clinical observations, an operative procedure was devised that would recreate both the conoid and trapezoid coracoclavicular ligaments individually as well as augmentation of any remaining superior and posterior AC ligaments.

Approach A patient with a type V AC separation is shown (Fig. 17C25). In the osteologic analysis of 118 clavicles, the mean length from the end of the clavicle, or the AC joint, to

the ­coracoclavicular ligaments was 46.3 ± 5 mm; the distance between the trapezoid laterally and conoid medially was 21.4 ± 4.2 mm. Thus, we center our incision roughly 3.5 cm from the distal clavicle or AC joint and make it ­curvilinear along Langer’s lines toward the coracoid process (Figs. 17C-26 and 17C-27). Control of the superficial skin bleeders down to the fascia of the deltoid is accomplished with a needle-tip bovie. Once the entire clavicle is palpated, full-­thickness flaps are made from the midline of the clavicle both posteriorly and anteriorly, skeletonizing the clavicle. This is done in the area of the coracoclavicular ligament, making the osteologic measurements listed ­earlier ­important.

Shoulder

A

C

847

B

Figure 17C-25   Typical findings after a type V AC joint injury, anterior view (A) and lateral view (B). There is a twofold to threefold increase in the coracoclavicular distance as well as a stripping of the deltoid and trapezial fascia from the clavicle. This patient was unable to reduce the AC joint with an active shoulder shrug, indicating a true type V injury instead of a type III injury. C, Anteroposterior radiographic view demonstrates widening, evidence of a type V injury.

There remains debate regarding the addition of a distal clavicle excision to an AC joint reconstruction ­procedure. In the acute situation when the presence of AC joint arthrosis is usually minimal, or in situations in which stability is of paramount concern, the AC reconstruction may be performed with retention of the distal clavicle. However, patients who have a chronic AC joint injury or preexisting AC joint arthrosis may benefit from a distal clavicle ­ excision. In these situations, about 5 mm of distal clavicle is removed with the use of a sagittal saw (Fig. 17C-28). The posterior edge of the clavicle is ensured to be smooth and free of abutment using a sagittal saw and rasp (Fig. 17C-29).

Graft Preparation Depending on surgeon preference, semitendinosus alloautograft or anterior tibialis allograft can be used for this procedure (Fig. 17C-30). Lee and associates35,130 found no difference in peak load-to-failure between semitendinosus, toe extensors, and gracilis tendons for reconstruction of

Figure 17C-26   Initial exposure for access to the AC joint. The skin incision is made in line with Langer’s lines, from anterior to posterior. Once the deltotrapezial fascia is encountered, this layer is incised sharply over the midline to develop full-thickness fascial flaps in a medial to lateral direction. Repair of this layer during closure is an important aspect of the case.

848

DeLee & Drez’s Orthopaedic Sports Medicine

5 mm

35 mm

m

15 m

Figure 17C-28   In some chronic situations, and in patients with preexisting AC joint arthrosis, a distal clavicle excision may be considered. About 10 mm of the distal end of the clavicle is removed with a small sagittal saw. Ensure that the cut is made parallel to the joint, which is slightly oblique in the coronal plane. This is not performed in all cases, and it remains debatable whether to perform a distal clavicle excision concomitantly with AC joint reconstruction. Figure 17C-27   Distances for the distal clavicle excision and bone tunnels are shown. The conoid bone tunnel should start about 45 mm from the AC joint, in the posterior one third of the clavicle. The trapezoid bone tunnel should be positioned 15 mm anteromedial to the conoid bone tunnel.

the AC joint. In this technique, there are two options for handling the fixation to the coracoid process. One option involves a bone tunnel interference screw type fixation, and the other option involves looping the ligament around the coracoid process. For interference screw fixation to the coracoid process, the graft is folded in its middle, and a No. 2 FiberWire (Arthrex, Naples, Fla) or a No. 2 nonabsorbable suture is placed through the doubled-over tendon graft in a Krakow manner. One to two Krakow sutures are placed in the remaining two free ends of the graft. The graft is placed on the table in a moist sponge until the bone tunnels are prepared.

coracoid is reamed to a depth of 15 to 17 mm. Copious irrigation is used to remove any excess bone shavings from the reaming.

Graft Fixation to Coracoid Process—Interference Fit Technique The Biotenodesis driver is assembled, and a 5.5 × 15 mm bioabsorbable Biotenodesis screw is placed on the end. A Nitinol wire is placed through the center cannulation of the screwdriver and used to shuttle the suture from the graft through the driver. The long end of the Krakow suture attached to the graft should be placed through the cannulated portion of the Biotenodesis driver. The tenodesis driver is advanced to touch the tendon graft, and the

Bone Tunnel Preparation in the Coracoid The diameter of the doubled-over portion of the graft is measured with a standard tendon-measuring device or using the handle of the Biotenodesis system (Arthrex, Naples, Fla) to determine graft size (Fig. 17C-30). The appropriate cannulated reamer is chosen (usually, 6 or 7 mm). For fixation of graft to the base of the coracoid process, finger palpation of both lateral and medial portions of the coracoid process and drilling into the coracoid base under direct visualization with a cannulated reamer guide pin is completed. A coracoid drilling guide (Arthrex, Naples, Fla) can also be used, which acts as a pipe-fitting device, fitting over the base of the coracoid process and angling the surgeon in the correct position. When the guide pin has been inserted and confirmed by digital palpation not to be out of the coracoid process, the cannulated reamer of the specific graft size is used and the

Figure 17C-29   If the distal clavicle is excised, care is taken to ensure that the posterior aspect of the distal clavicle is beveled, using the small sagittal saw and a rasp, in order to avoid posterior abutment.

Shoulder

849

A

B

Figure 17C-30   Graft preparation. Here, a tibialis anterior allograft is used, and the two ends of the tendon are stitched in Krakow fashion using No. 2 FiberWire (Arthrex, Naples, Fla). The graft is then doubled over and measured using graft sizers from the Biotenodesis System (Arthrex). The doubled-over graft size is usually 6 to 7 mm. If larger, the graft should be trimmed (especially in an allograft situation). The graft is then placed under tension on the back table in preparation for implantation.

Figure 17C-31   Coracoid graft fixation technique with bioabsorbable screw. Fixation of the graft in the coracoid bone tunnel using Biotenodesis (Arthrex, Naples, Fla) screw fixation. The two free limbs are evenly distributed from the coracoid tunnel. A nonabsorbable high-strength suture (No. 5 FiberWire, Arthrex) is passed under the coracoid and will serve as nonbiologic supplemental fixation, in addition to the soft tissue graft.

entire tendon, driver, and screw complex is placed into the ­coracoid bone tunnel (Fig. 17C-31). The Krakow suture, ­usually ­ measured to be about 15 mm in length running up and down the doubled portion of the tendon graft, should disappear from view when the tendon, driver, and screw complex is placed into the coracoid bone tunnel. A clamp is placed on the back portion of the teardrop handle of the driver to lock the suture in place. The screw is advanced over the driver-tendon complex, creating a secure interference screw fit. The clamp is released, and the screwdriver complex is removed from the screw. Digital palpation confirms that the screw is flush with the bone tunnel, and by pulling on the two ends of the tendon, it is confirmed that there is a secure fit. The sutures from the graft are tied together over the existing interference screw, giving both interference screw and suture anchor advantages.

around the base of the coracoid. This will ­ eventually become the nonbiologic fixation, reducing the clavicle to the scapula.

Graft Fixation to Coracoid Process—Loop Technique To avoid placing a bone tunnel in the coracoid process, the graft can be looped around its base (Fig. 17C-32). Looping the graft around the base of the coracoid process can be facilitated by the use of a curved aortic cross-clamp (Satinsky clamp) and a suture-passing device. At the same time that the graft is passed, a No. 5 FiberWire is also passed

Bone Tunnels in the Clavicle It is important to make the bone tunnels in as accurate a position as possible to recreate the coracoclavicular ligament (Fig. 17C-33). The complex osteologic measurements provided are to aid the surgeon in finding the insertions of the conoid and the trapezoid and not meant as absolute numbers. A cannulated reamer guide pin is used for placement of the tunnels. The first tunnel, for the conoid process, is roughly 45 mm away from the distal end of the clavicle. The footprint of the conoid ligament is extremely posterior, along the entire posterior edge of the clavicle, and this is why making this bone tunnel as ­posterior as ­possible (i.e., in the posterior half of the clavicle) is extremely important. The guide pin is also angled about 45 degrees from the direct perpendicular of the clavicle to re-create the oblique nature of the ligament. A 6-or 7-mm reamer is used to create the tunnel with careful attention confirming that the tunnel is as posterior as possible in the clavicle without “blowing out” the posterior cortical rim (Fig. 17C-34). After this is confirmed, a 15- to 16-mm length bone tunnel is created.

850

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17C-32   Coracoid graft fixation with soft tissue passed under A B the coracoid. Alternatively, nonabsorbable suture may be passed around the coracoid (A) and used to shuttle the soft tissue graft under the coracoid, instead of using the Biotenodesis screw for graft fixation into the coracoid. At the same time, a high-strength No. 5 suture (FiberWire, Arthrex, Naples, Fla) is used for nonbiologic fixation and left in place once the graft is shuttle underneath the coracoid (B).

The same procedure is repeated for the trapezoid ligament, which is a more anterior structure than the conoid and is usually placed in the center point of the clavicle, about 15 mm away from the center portion of the ­previous tunnel. Two guide pins are used before reaming to confirm the accurate placement of the tunnels. After the tunnel reaming, copious irrigation follows to remove any bone fragments.

Interference Screw Fixation of Graft to Clavicle One limb of the biologic graft is taken and placed through the posterior bone tunnel, re-creating the conoid ligament (Fig. 17C-35). The other limb is passed through the ­anterior bone tunnel, re-creating the trapezoid ligament. At

the same time that the graft is brought through (Fig. 17C36), each limb of the No. 5 FiberWire should be brought through the respective bone tunnels as well (Fig. 17C-37). Upper displacement of the scapulohumeral complex by the assistant reduces the AC joint. A large point-of-reduction forceps placed on the coracoid process and the clavicle can assist while securing the tendon grafts. The AC joint should be over-reduced during initial fixation because of an inevitable amount of creep in the tendon graft. With complete upper displacement on the graft, ensuring its tautness, a 5.5 × 15 mm bioabsorbable interference screw is placed in either the posterior or midline bone tunnel. The No. 5 FiberWire is brought up through the cannulated process of this screw, and after assessment that this is done successfully, the second bioabsorbable interference screw is placed in the other,

Figure 17C-33   The guide pins are placed into the distal clavicle, A Bat points which approximate the attachment sites of the conoid and trapezoid ligaments. The conoid ligament attaches more posteriorly and more medially than the trapezoid ligament. The first tunnel, for the conoid process, which is extremely posterior, is roughly 45 mm away from the distal end of the clavicle, along the entire posterior edge of the clavicle (A). The guide pin is also angled about 45 degrees from the direct perpendicular of the clavicle to recreate the oblique nature of the ligament. A 6- or 7-mm reamer is used to create the tunnel with careful attention confirming that the tunnel is as posterior as possible in the clavicle without “blowing out” the posterior cortical rim. Once this is confirmed, a 15- to 16-mm bone tunnel is created. The same procedure is repeated for the trapezoid ligament, which is a more anterior structure than the conoid and is usually placed in the center point of the clavicle, about 15 mm away from the center portion of the previous tunnel. Two guide pins are used before reaming to confirm the accurate placement of the tunnels (B).

Shoulder

851

Figure 17C-34   After the tunnel reaming of the conoid (A) and trapezoid (B) tunnels, copious irrigation follows to remove any A B bone fragments.

unfilled, bone tunnel. With both grafts secured using interference screw fixation, the No. 5 FiberWire is tied over the top of the clavicle, becoming a nonbiologic fixation for the over-reduced AC joint (Figs. 17C-38 and 17C-39).

Closure One of the most important concepts with AC or coracoclavicular joint reconstruction is the closure of the deltotrapezial fascial flaps from the dissection. A nonabsorbable suture, in a modified Mason-Allen–type fashion is placed through the deltoid fascia. Six or seven sutures are used, and the knots are tied on the posterior aspect of the trapezius. This closure of the deltotrapezial fascia should completely obscure the grafts as well as the clavicle. If there is any concern regarding the fascial repair, the deltoid should be repaired through small drill holes in the anterior cortex of the clavicle. Some worry exists, though, that with two 6-mm bone tunnels in the clavicle, further smaller defects in the clavicle for deltoid fixation could lead to an iatrogenic fracture. In unpublished three-point bending data accomplished at the University of Connecticut biomechanics laboratory, as long as interference screws were in the reamed holes of the clavicle, there was not a significant difference in the load-to-failure between the modified Weaver-Dunn technique and the interference screw fixation method. The subdermal skin is closed with 2-0 or 3-0 absorbable sutures, and the skin itself is closed with either a 2-0 running or interrupted nylon suture, everting the skin edges. A compression dressing is applied, and the patient is placed in a supportive sling with external rotation to 0 degrees and an upward force on the arm.

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS A variety of factors affect postoperative management of surgery involving the AC joint. If the procedure includes only a distal clavicle resection, then a short (1 to 3 days)

period of immobilization is followed by range of motion activities. Strengthening begins at 4 to 6 weeks. Heavy weight training can begin at 3 months, but power athletes often require 6 to 12 months to return to peak strength. After a coracoclavicular ligament reconstruction, the arm is supported with an external device such as a sling and immobilizer. Gentle range of motion activities in the supine position can begin after 7 to 10 days. Range of motion with the arm unsupported in an upright position should be delayed until the reconstruction has had time to develop early biologic stability. For an acute repair, this takes 4 to 6 weeks. A chronic repair with severe soft tissue involvement, for example, a type V separation, may take up to 6 to 12 weeks before unsupported range of motion is allowed. An emphasis at this point should be placed on strengthening the scapular stabilizers. These muscles decrease the load on the joint by keeping the scapula in a relatively retracted position. Strengthening in an acute repair begins at 6 to 12 weeks, with weight training started at 3 to 4.5 months. Strengthening in a chronic repair is appropriately delayed. We use this protocol for the anatomic AC reconstructions. Historically, motion is limited until pins are removed after 6 to 8 weeks. After coracoclavicular screw fixation, range of motion begins when pain subsides. Bosworth8 recommended no heavy activity for 8 weeks. Alldredge133 recommended no immobilization, Bearden and ­colleagues9 a sling for 10 to 14 days, Jay and Monnet74 a sling for 4 weeks, and Gollwitzer85 a Velpeau cast for 4 weeks. Recommendations regarding hardware removal have varied.9,21,73,74,133,134 After coracoid transfer, Brunelli and Brunelli69 recommended 90 degrees of elbow flexion, with gradual straightening starting on day 5 to reduce the AC joint, and protected activities for 6 to 8 weeks.

Complications Hardware migration, the most serious complication of AC joint injury, is associated with surgical treatment of dislocations. The frequency of pin migration and seriousness

852

DeLee & Drez’s Orthopaedic Sports Medicine

A

B Figure 17C-35   A, Fixation of the conoid and trapezoid graft limbs to the reduced clavicle using Biotenodesis screw (Arthrex, Naples, Fla) fixation. One limb of the biologic graft is taken and placed through the posterior bone tunnel, re-creating the conoid ligament. The other limb is passed through the anterior bone tunnel, re-creating the trapezoid ligament. B, As an alternative to a screw in the coracoid, the authors prefer that the graft be placed under the base of the coracoid.

Figure 17C-36   A limb of No. 5 FiberWire (Arthrex, Naples, Fla) is also brought out each of the respective bone tunnels with the soft tissue graft. Upper displacement of the scapulohumeral complex by the assistant reduces the AC joint. The AC joint should be over-reduced during initial fixation because of an inevi­ table amount of creep in the tendon graft. With complete upper displacement on the graft, ensuring its tautness, a 5.5 ×   15 mm bioabsorbable interference screw is placed in either the posterior or midline bone tunnel. The No. 5 FiberWire is brought up through the cannulated process of this screw, and after assessment that this is done successfully, the second bioabsorbable interference screw is placed in the other, unfilled, bone tunnel.

of potential complications have prompted most surgeons to abandon their use, especially the use of smooth pins. Those who still use pins check their position with frequent radiographs and remove them after some interval of healing. Pin migration into the lung and spinal canal has been reported.135,136 Lindsey and Gutowski137 reported migration into a patient’s neck posterior to the carotid sheath. Eaton and Serletti138 and Urban and Jaskiewicz139 reported migration into the pleural cavity. Sethi and Scott140 reported laceration of the subclavian artery by a migrated pin. ­ Grauthoff and Kalmmer141 reported five cases of migration into the aorta, subclavian artery, or lung. Loss of reduction of the AC joint is not uncommon. The weight of the arm and scapula places a tremendous static force on the coracoclavicular reconstruction. Younger patients have a tendency to discontinue efforts at supporting the arm for the first 6 weeks, which is necessary to protect the reconstruction. Efforts at augmentation of the repair and reconstruction have helped to reduce the incidence of complete failure, but partial loss of reduction remains common. Mayr and associates142 reported a lost reduction rate of 28%, with a less satisfactory outcome in these patients. Other surgical complications include infection, ­aseptic reaction to the reconstruction, calcifications, and ­ erosion through the clavicle from nonabsorbable materials used to augment the repair and reconstruction, fracture of the coracoid, osteolysis, and persistent pain. Reported rates of infection range from 0% to 9%, with an average of 6% taking into account numerous reports.85,86,88,143 Colosimo and colleagues144 reported

Shoulder

853

A

Figure 17C-38   Final view of the anatomic coracoclavicular ligament reconstruction. Note the position of the clavicular bone tunnels in relation to the center line of the clavicle and the interference screw fixation backed up with a nonbiologic No. 5 FiberWire suture through the cannulated holes of the interference screws.

B

C Figure 17C-37   A, B, With both grafts secured using interference screw fixation, the No. 5 FiberWire (Arthrex, Naples, Fla) is tied over the top of the clavicle, becoming a nonbiologic fixation for the over-reduced AC joint. C, Alternatively, the authors prefer that only the graft is utilized under the coracoid.

an aseptic ­­foreign-body ­reaction to Dacron graft used to reconstruct the coracoclavicular ligaments. Calcification in the reconstructed ligament has been noted but does not appear to affect results.86 In fact, when the reduction is maintained and calcification occurs, the stability of the reconstruction appears to be enhanced. Erosion of cerclage material through the clavicle or coracoid is a well-documented complication.78,145,146 A modification of the cerclage technique to place material through an osseous tunnel in the clavicle rather than completely around it decreases the severity of this complication because erosion does not create a complete discontinuity between the medial and lateral clavicle. Fracture of the coracoid may occur with placement of a coracoid screw.147

Osteolysis associated with AC fixation has been reported.60 Smith and Stewart95 recommended ­ resection of the distal clavicle at the time of surgical reduction to avoid this complication. The complication of late AC joint arthrosis is avoided, and therefore distal clavicle ­resection has become an integral part of any AC instability ­reconstruction. Chronic pain after surgical treatment of AC instability can be another challenging complication. Many possible causes need to be considered, including horizontal instability (anterior to posterior) of the clavicle, subacromial disease, and neurologic injury. Neurologic injury can occur with the initial trauma or with the surgical procedure. For example, suprascapular neuropathy may occur after distal clavicle resection and has been associated with resections of greater than 1 cm.148

Figure 17C-39   Final AP radiograph demonstrating reduction of the AC joint with the two 5.5-mm clavicular tunnels.

854

DeLee & Drez’s Orthopaedic Sports Medicine

  TABLE 17C-4  Return to Play and Postoperative Regimen Strengthening—Start Scapula Exercises, then Progress

Sport Training

When pain allows (within 3 days)

4-6 wk

3 mo

7-10 days, gentle ROM

4-6 wk, longer if type V or severe III

4-6 wk, delay if hardware needs to be removed

3-4 mo

7-10 days, gentle ROM

4-6 wk

4-6 wk

3-4 mo

Procedure (­Postoperative Time)

Protected Time (in Sling)

Passive ROM

Distal clavicle excision alone

1-3 days

Immediately

CC reconstruction

3-6 wk

Anatomic CC reconstruction

4-6 wk

Active and ­Active­Assisted ROM

Full Unrestricted Activity In power athletes may take up to 6-12 mo 5-6 mo, no contact sports for 6 mo, may take 9-12 mo to reach peak strength 5-6 mo, no contact sports for 6 mo, may take 9-12 mo to reach peak strength

CC, coracoclavicular; ROM, range of motion.

CRITERIA FOR RETURN TO PLAY Power athletes and heavy physical demand workers are the longest to rehabilitate and generally take 9 to 12 months to reach peak strength, especially with pressing activities or lifting from the floor (i.e., dead lift). After an anatomic AC reconstruction, we limit return to contact sports for a minimum of 6 months. Before this, the therapist should work with the athlete on sport-specific training exercises and also ensure that the Cybex upper extremity power testing is within 10% to 15% of the contralateral normal limb (Table 17C-4).

SPECIAL POPULATIONS AND SPECIAL INJURIES TO THE ACROMIOCLAVICULAR JOINT Osteolysis of the Distal Clavicle Atraumatic osteolysis of the distal clavicle (Fig. 17C-40) is a stress overload syndrome of the distal clavicle. It occurs predominantly in young athletes who have a history of intense weight training associated with bench press activities and military shoulder press.149-152 Scavenius and Iversen153 found a link between weightlifting and atraumatic osteolysis of the distal clavicle. They reported in 25 elite weightlifters compared with agematched controls that 28% of the weightlifters exhibited radiographic evidence of atraumatic osteolysis of the distal clavicle. On average, this group tended to be active, lifting heavy weights twice as long, and subjects were older as well as heavier when compared with the controls.154 Typically, the athlete presents with a slow onset of pain in the area of the AC joint, with occasional radiation to the surrounding deltoid muscle or superior border of the trapezius. The pain is intensified by activity, especially bench press on a flat bench, typically using greater than 200 pounds (>90 kg). Slawski and Cahill155 reported a 79% bilateral involvement with weightlifters once they present to the physician with symptoms.

Radiographically, a Zanca view (cephalic tilt of 10 to 15 degrees), when taking AP radiographs of the AC joint, provides the best image of the distal clavicle. The changes are represented by a loss of subchondral bone detail at the distal clavicle, cystic appearance in the subchondral area and osteoporosis to the distal third of the clavicle. Late manifestations include a distinct widening of the AC joint with cysts and lucencies at the clavicular end of the acromion. A bone scan is useful in assessing the biologic activity at the distal clavicle and can be used to support the diagnosis of distal clavicle osteolysis. Pitchford and Cahill’s154 conservative treatment of atraumatic osteolysis of the distal clavicle is directed toward eliminating the provocative maneuvers causing it. Unfortunately, many power athletes require the weight training to maintain and increase strength as well as their overall body mass if they want to remain competitive. Pitchford and Cahill154 suggested the following surgical indications for the treatment of distal clavicle osteolysis: (1) a confirmed diagnosis of atraumatic osteolysis, and (2) an unwillingness on the part of the athlete to accept a lower level of performance. The surgical procedure is a distal clavicle resection that can be performed with open or arthroscopic techniques. In summary, atraumatic osteolysis of the distal clavicle is a cumulative stress on the distal clavicle caused by ­muscular forces across the AC joint with press-type maneuvers. This entity predominantly occurs in weightlifters but may be seen in athletes who use weightlifting for strength and conditioning. The clinical history and examination, combined with appropriate radiographs, allows for the correct diagnosis. The diagnosis is supported by a positive bone scan. The treatment of choice is avoidance of the activities associated with increased symptoms. When this fails, distal clavicle resection may be indicated.

Intra-articular Acromioclavicular Joint Fractures Minimally displaced distal clavicle fractures or acromial fractures are relatively stable owing to the ligamentous stability provided by the AC, coracoclavicular, and ­coracoacromial

Shoulder

A

B

C

D

855

Figure 17C-40   A, Radiograph of the AC joint demonstrating widening of the AC joint and loss of subchondral cortical detail of the distal clavicle (arrow). B, Radiograph of the AC joint demonstrating osteoporosis of the distal clavicle and cortical thinning.   C, Radiograph of the AC joint demonstrating cystic changes (arrowheads) and loss of subchondral integrity. D, Radiograph of AC joint demonstrating cystic changes of the clavicular portion of the acromion (arrowheads) and widening of the AC joint. (From Pitchford MR, Cahill BR: Osteolysis of the distal clavicle in the overhead athlete. Operative Tech Sports Med 5:74, 1997.)

ligaments. These are generally treated nonoperatively with success unless there are extenuating effects (open injury, neurovascular compromise). A sling for comfort, then early range of motion and shoulder strengthening exercises when pain permits, is recommended. Some of these fractures may predispose the joint to early post-­traumatic arthrosis. When indicated by the persistence of pain unrelieved with nonoperative treatment, a distal clavicle resection is performed.

Acromioclavicular Injuries in the Child The developmental anatomy of the shoulder provides insight regarding the type of injuries that occur in a skeletally immature athlete. At 1 year of age, there is an ossification center seen at the tip of the coracoid. At age 10 years, the base of the coracoid and upper fourth of the glenoid have ossified, and these fuse to the scapula by the age 15 years. Near puberty, the acromion forms between two and five ossification centers that fuse by the age of 22 years. Failure of the acromion ossifications to fuse can occur without any loss of shoulder function. In children, the classification of AC injuries is based on the position of the clavicle with respect to the periosteal sleeve and intact ligaments. This classification system has been reported by Curtis and colleagues and is organized in a similar progression as the adult or skeletally mature patient classification scale (Fig. 17C-41).156

Figure 17C-41   Acromioclavicular ligament injuries. Displacement of the distal clavicle occurs through a tear in the periosteal tube. This occurs in children who sustain a severe force to the shoulder. The AC and costoclavicular ligaments remain intact through the periosteal tube. (From Beim GM, Warner JP: Clinical and radiographic evaluation of the acromioclavicular joint. Oper Tech Sports Med 5:68, 1997.)

856

DeLee & Drez’s Orthopaedic Sports Medicine

Type I injury: sprain of AC ligaments with the periosteal sleeve intact Type II injury: partial disruption of the periosteal sleeve with slight winding of the AC joint Type III injury: periosteal tube disrupted with instability of the distal clavicle, superior displacement 25% to 100% of the distal clavicle on the AP radiograph Type IV injury: periosteal tube disrupted, distal clavicle displaced posteriorly through or into trapezius muscle seen on axial lateral radiograph Type V injury: periosteal tube disrupted, deltoid and trapezial detachment, clavicle displaced subcutaneously greater than 100% of the normal Type VI injury: inferior displacement of the clavicle behind the coracoid process Fracture of the coracoid through the common growth plate with the upper glenoid fossa may mimic an AC injury, but the coracoclavicular interspace remains intact.118 The axillary view best demonstrates a fracture to the coracoid. This should be suspected in AC injuries within the first three decades of life. A computed tomographic scan is indicated if there is any concern of a displaced fracture involving the glenohumeral joint.

l Capsule

and capsular ligaments are primary restraints to AP translation. The main contribution of the coracoclavicular ligaments is vertical stability—they mediate synchronous scapulohumeral motion and strengthen AC articulation. l Distal clavicle resection of even less than 1 cm may have residual posterior instability. l AC injuries present clinically as acute (direct or indirect injury; see Box 17C-2) or chronic (arthrosis, degenerative AC joint); diagnostic tests and injections help to differentiate the causes of shoulder pain. l AC injuries are suspected in weightlifters, overhead repetitive lifters, and some throwers, in whom the subcutaneous joint with fascial covering is easily injured. l Physical examination and radiographs determine the neurovascular status of the upper extremity and evaluate the neck and sternoclavicular joint for associated injuries; attempt to localize pain to AC joint (AC compression, cross-body adduction, and other tests); and assess skin condition for tenting (type IV if posterior tenting). l Types I and II AC injuries are treated nonoperatively. Treatment of type III injuries remains controversial. Types IV, V, and VI injuries are usually treated ­operatively.

Treatment in Children The treatment of AC joint injuries in skeletally immature athletes is typically nonoperative because the most common injuries are type I, II, and III. Nonoperative treatment consists of a sling for pain control over the first 3 to 7 days, ice, nonsteroidal anti-inflammatory medications, and mild analgesics as needed. Most athletes in this age group do not require physiotherapy, but if limitations in range of motion are present after 2 to 4 weeks, a short course of physiotherapy is beneficial. Eidman and colleagues62 reported that conservative treatment of these injuries have gone on to heal without clinically relevant sequelae. Nuber and Bowman25 report in their chapter that surgical treatment of type IV, V, and VI injuries is successful. Replacement of the clavicle into its periosteal sleeve with suturing the sleeve closed and then fixation of the coracoclavicular lag screw or transacromial fixation is recommended. The fixation is then removed after 4 to 6 weeks before physical therapy is started in the child.

C l

r i t i c a l

P

S U G G E S T E D

R E A D I N G S

Chronopoulos E, Kim TK, Park HB, et al: Diagnostic value of physical tests for isolated chronic acromioclavicular lesions. Am J Sports Med 32:655-661, 2004. Debski RE, Parsons IM, Woo SL, Fu FH: Effect of capsular injury on acromioclavicular joint mechanics. J Bone Joint Surg Am 83:1344-1351, 2001. Fakuda K, Craig E, AN K, et al: Biomechanical study of the ligament systems of the acromioclavicular joint. J Bone Joint Surg Am 68:434-439, 1986. Mazzocca AD, Santangelo SA, Johnson ST, et al: A biomechanical evaluation of an anatomical coracoclavicular ligament reconstruction. Am J Sports Med 34:236246, 2006. Rockwood CJ, Williams G, Young DC: Disorders of the acromioclavicular joint. In Rockwood CJ, Matsen F (eds): The Shoulder, 3rd edition. Philadelphia, Saunders, 2004, pp 521-595. Schlegel TF, Burks RT, Marcus RL, Dunn HK: A prospective evaluation of untreated acute grade III acromioclavicular separations. Am J Sports Med 29:699703, 2001. Tienen TG, Oyen JF, Eggen PJ: A modified technique of reconstruction for complete acromioclavicular dislocation: A prospective study. Am J Sports Med 31:655-659, 2003. Walton J, Mahajan S, Paxinos A, et al: Diagnostic values of tests for acromioclavicular joint pain. J Bone Joint Surg Am 86:807-812, 2004. Weaver J, Dunn H: Treatment of acromioclavicular injuries, especially complete acromioclavicular separation. J Bone Joint Surg Am 54:1187-1194, 1972.

o i n t s

 he AC joint is diarthrodial, providing 6 degrees of freeT dom. The collagenous disk between joints predictably degenerates with age. l The shoulder is stabilized by static (AC, CC, and CA ligament, and capsule) and dynamic (deltoid, trapezius) ­structures.

R eferences Please see www.expertconsult.com

Shoulder

S ect i o n

857

D

Injuries to the Glenoid, Scapula, and Coracoid 1. Glenoid and Scapula Fractures in Adults and Children Allen Deutsch, Jason A. Craft, and Gerald R. Williams, Jr.

The scapula is intimately linked to shoulder function and mobility. It links the appendicular and axial skeletons through the clavicle and the acromioclavicular, sternoclavicular, and glenohumeral joints and presents a stable platform for the upper extremity. Injury to the scapula may disrupt normal shoulder function. The incidence of scapular fracture has been reported to be 3% to 5% of shoulder girdle injuries1,2 and 0.4% to 1% of all fractures.3,4 The low incidence of scapular fracture is due to its protected position along the rib cage, the enveloping musculature, and its relative mobility, which permits dissipation of forces. Scapular fractures most commonly involve the scapular body (49% to 89%), the glenoid neck (10% to 60%), and the glenoid cavity (10%).5-10 Scapular fractures are usually sustained as the result of severe trauma. Most series report motor vehicle or motorcycle crashes as the cause of injury in more than half of cases.5,6,8-10 Associated injuries are common, including rib fracture, pneumothorax, and head injury. Rowe11 reported that 71% of the patients in his series of scapular fractures had other associated injuries: 45% had fracture of other bones, including the ribs, sternum, and spine; 3% sustained a pneumothorax; 4% sustained brachial plexus injuries; and 19% sustained other shoulder girdle dislocations. A recent multicenter trauma review of scapular fractures found that they were associated with 1.1% of all blunt trauma admissions. Despite being relatively rare, 99% had associated injuries, with rib fractures (43%) and lower and upper extremity fractures (36% and 33%) being the most common. However, 28% and 27% had associated intrathoracic trauma and head trauma, respectively.12 These numbers highlight the large amount of force required to obtain these fractures and reinforce the need to fully evaluate that patient for associated injuries. Several classification systems for scapular fractures have been reported in the literature. Zdravkovic and Damholt13 divided scapular fractures into three types: type I fractures, or fractures of the body; type II fractures, or fractures of the apophyses (including the coracoid and acromion); and type III fractures, or fractures of the superior lateral angle (i.e., scapular neck and glenoid). Zdravkovic and Damholt13 considered the type III fracture to be the most difficult to treat; these represented only 6% of their series. Thompson and coworkers14 presented a classification system that divided these fractures according to the likelihood that associated injuries would be present. Their cases resulted from blunt trauma. Class I fractures included fractures of the coracoid and acromion process and small

fractures of the body. Class II fractures included glenoid and scapular neck fractures. Class III fractures included major scapular body fractures. Thompson and colleagues14 reported that class II and class III fractures were much more likely to have associated injuries. Wilber and Evans10 described 40 patients with 52 scapular fractures. The patients were divided into two groups on the basis of fracture location: group I, which included patients with fractures of the scapular body, neck, and spine; and group II, which included patients with fractures of the acromion process, coracoid process, or glenoid. They reported unsatisfactory results of treatment of patients in group II because of residual pain and loss of glenohumeral motion. Ideberg15,16 devised a classification system of five types of scapular fracture with an associated intra-articular glenoid component. This system was modified by Goss17,18 with inclusion of six types. Type I fractures involve the glenoid rim and are subdivided into Ia, anterior rim; and Ib, posterior rim. Type II to V fractures extend from the glenoid fossa to various exit points along the scapula. Type II fractures exit the lateral border of the scapula, below the infraglenoid tubercle. Type III fractures exit the superior border and typically extend medial to the base of the coracoid. Type IV fractures extend directly across the scapula to the medial border and usually exit superior to the scapular spine. Type V fractures are combinations of types II to IV. Type VI fractures encompass glenoid fractures with extensive intra-articular comminution (Fig. 17D1-1).17 Goss19 described the superior shoulder suspensory complex, consisting of the glenoid, coracoid, acromion, distal clavicle, coracoclavicular ligaments, and acromioclavicular ligaments. This bone–soft tissue ring maintains the normal, stable relationship between the upper extremity and the axial skeleton. Single disruptions of the superior shoulder suspensory complex, such as an isolated scapular neck fracture, are usually anatomically stable because the integrity of the complex is preserved, and nonoperative management yields good functional results. When the complex is disrupted in two places (double disruption), such as a scapular neck fracture with an acromioclavicular joint disruption, a potentially unstable anatomic situation is created. Because the superior shoulder suspensory complex includes the glenoid, acromion, and coracoid, many double disruption injuries involve the scapula. Open reduction is indicated for double disruptions that are accompanied by significant displacement, which may lead to delayed union, malunion, or nonunion as well as long-term functional deficits.

858

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17D1-1   Classification of fractures of the glenoid cavity. Type Ia, anterior rim fracture; type Ib, posterior rim fracture; type II, fracture line through the glenoid fossa exiting at the lateral scapular border; type III, fracture line through the glenoid fossa exiting at the superior scapular border; type IV, fracture line through the glenoid fossa exiting at the medial scapular border; type Va, combination of types II and IV; type Vb, combination of types III and IV; type Vc, combination of types II, III, and IV; type VI, comminuted fracture. (Modified from Goss TP: Scapular fractures and dislocations: Diagnosis and treatment. J Acad Am Orthop Surg 3:22-33, 1995.)

Ia

II

Ib

III

Va

Vb

IV

Vc

VI

ANATOMY The scapula is enveloped by multiple layers of muscles. The anterior surface provides attachment for the subscapularis, the serratus anterior, the omohyoid, the pectoralis minor, the conjoined tendon of the coracobrachialis and short head of the biceps, the long head of the biceps, and the long head of the triceps (Fig. 17D1-2). The posterior surface of the scapula provides muscle attachment sites for the levator scapulae, the rhomboid major, the rhomboid minor, the latissimus dorsi, the teres major, the teres minor, a portion of the long head of the triceps, the deltoid, the trapezius, the supraspinatus, the infraspinatus, and a portion of the omohyoid (Fig. 17D1-3). The intramuscular position of the scapula provides it with great mobility and a protective cushion that are no doubt responsible for the low incidence of scapular injury. The proximity of neurovascular structures to the scapula places them at risk for injury. The pectoralis minor tendon inserts at the base of the coracoid process and the lateral border of the suprascapular notch. The brachial plexus and axillary artery travel posterior to the pectoralis minor tendon. The suprascapular nerve passes under the transverse scapular ligament as it passes through the suprascapular notch to innervate the supraspinatus muscle, whereas the suprascapular artery passes over the ligament. The suprascapular nerve continues through the spinoglenoid notch, or the junction between the base of the acromion and the

Coracobrachialis and short head of biceps Pectoralis minor Omohyoid

Biceps, long head

Triceps

Serratus anterior

Subscapularis

Figure 17D1-2   The muscle attachments to the anterior surface of the scapula. (Modified from Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

Shoulder Omohyoid m. Supraspinatus m.

Trapezius m.

Biceps m.

Levator scapulae m. Deltoid m.

Rhomboideus minor m.

Triceps m. Infraspinatus m.

Rhomboideus major m.

Teres minor m.

Teres major m. Latissimus dorsi m.

Figure 17D1-3   The muscle attachments to the posterior surface of the scapula. (Modified from Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

neck of the scapula, to innervate the infraspinatus muscle. At the medial border of the scapula, the dorsal scapular and spinal accessory nerves course along with the branches of the transverse cervical artery. The osseous components of the scapula, which consist of the body and spine, the coracoid process, the acromion process, the glenoid, and the inferior angle, arise from several ossification centers.20-22 At birth, the body and spine form one ossified mass. The coracoid process, the acromion process, the glenoid, and the inferior angle are cartilaginous, however. The coracoid process is a coalescence of four or five centers of ossification. The center of ossification for the midportion of the coracoid appears at the age of 3 to 18 months and may be bipolar. The ossification center for the base of the coracoid, which includes the upper third of the glenoid, appears at 7 to 10 years. Two ossification centers appear at the age of 14 to 16 years: a center for the tip and a shell-like center at the medial apex of the coracoid process. The ossification centers for the

859

base and the midportion of the coracoid coalesce during adolescence at 14 to 16 years of age. The other ossification centers fuse at 18 to 25 years of age (Figs. 17D1-4 and 17D1-5). The acromion is a coalescence of two or three centers of ossification that appear between the ages of 14 and 16 years, coalesce at the age of 19 years, and fuse to the spine at the age of 20 to 25 years. Failure of the anterior ­acromion ossification center to fuse to the spine gives rise to the os acromiale. This unfused apophysis is present in 2.7% of random patients and is bilateral in 60% of cases.23 The size of the os acromiale depends on which of the four ossification centers of the acromion have failed to fuse (Fig. 17D1-6). The most common site of nonunion is between the meso-acromion and the meta-acromion, which corresponds to the mid-acromioclavicular joint level. An axillary lateral radiograph clearly demonstrates the lesion (Fig. 17D1-7). Norris24 has reported that the os acromiale has been mistaken for fracture and that there is an association between the os acromiale and a rotator cuff tear. The inferior angle of the scapula arises from an ossification center that appears at the age of 15 years and fuses with the remainder of the scapula at the age of 20 years. The vertebral border arises from an ossification center that appears at 16 to 18 years of age and fuses by the 25th year. The glenoid fossa ossifies from four sources: (1) the coracoid base (including the upper third of the glenoid), (2) the deep portion of the coracoid process, (3) the body, and (4) the lower pole, which joins with the remainder of the body of the scapula at 20 to 25 years of age. Because many athletes are adolescents and because many of the apophyses do not fuse until the age of 25 years, ­caution must be exercised in interpreting radiographs of the scapula. The os acromiale is the most frequently quoted unfused apophysis and can be confused with fracture.23,24 In addition, the physes at the base of the coracoid and the tip of the coracoid process can be difficult to distinguish from fracture. In the appropriate setting, a radiograph of the contralateral scapula is useful in determining whether a radiographic “line” is truly a fracture or an unfused ­apophysis.

Figure 17D1-4   A normal ossification pattern at the base of the coracoid. A crescent-shaped center is seen at the apex of the coracoid. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

860

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17D1-7   Os acromiale. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

CLINICAL EVALUATION Figure 17D1-5   An epiphyseal line is seen across the upper third of the glenoid because this portion of the glenoid ossifies in common with the base of the coracoid. This may be confused with a fracture and is the precise location of most type III glenoid fractures. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

PA MTA PA = Pre-acromion

History Scapular fractures in athletes can result from either direct or indirect mechanisms of injury. Although most athletes are not subjected to the high-energy trauma associated with motor vehicle crashes, direct blows to the scapula can occur with enough force to cause fracture in contact sports such as hockey and football. Direct blows to the acromion can cause either acromion fracture or acromioclavicular separation. In addition, direct blows to the scapula or to the lateral aspect of the shoulder can cause scapular body fractures or glenoid fractures. Alternatively, glenoid fractures can be the result of indirect trauma incurred during a violent glenohumeral dislocation or a fall on an outstretched arm.

MSA = Meso-acromion

MSA BA

MTA = Meta-acromion

Physical Examination

BA = Basi-acromion

The athlete with a scapular fracture typically presents with the arm adducted and protected from all movements. Abduction is especially painful. Although ecchymosis is less than expected from the degree of bone injury present, severe local tenderness is a reliable finding.25 Athletes with scapular body fractures or coracoid process fractures often complain of increasing pain with deep inspiration secondary to the pull of the pectoralis minor or serratus anterior muscles. Frequently, rotator cuff function is extremely painful and weak secondary to inhibition from intramuscular hemorrhage. This has been described as a pseudorup-­ ture of the rotator cuff26 and usually resolves within a few weeks. Scapular fracture is often associated with other injuries that need more urgent treatment. Significant associated injuries have been reported to occur in 35% to 98% of all patients with scapular fractures.25 The highest incidence of serious associated injuries occurs in fractures sustained during high-speed motor vehicle crashes.6,8-10,14,27 McLennen and Ungersma28 reported 16 pneumothoraces in 30 patients who presented with fractured scapulae.

A

B Figure 17D1-6   A, The diagram represents the ossification centers of the acromion. B, The most common site of failure of ossification lies between the meso-acromion and the metaacromion. (Modified from Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

Shoulder

861

Of the 16 pneumothoraces, 10 were delayed in onset from 1 to 3 days. The authors recommended a follow-up chest radiograph, physical examination, and blood gas determination for all patients with scapular fractures. Other series have reported an overall incidence of pneumothorax associated with scapular fracture of between 11% and 38%.6,14,27 Ipsilateral rib fractures,14 pulmonary contusion,14,27 arterial injury,27 and brachial plexus injury6,8,9,14,27 have also been reported in association with scapular fractures. Physical examination should be directed toward detecting any of these possible associated injuries.

RADIOGRAPHIC EVALUATION Most scapular fractures can be adequately visualized on routine radiographic views. A true anteroposterior view of the scapula, combined with an axillary or true scapular lateral view, demonstrates most scapular body or spine fractures, glenoid neck fractures, and acromion fractures (Figs. 17D1-8 through 17D1-10). “Special” views may be required in selected circumstances. The Stryker notch view, as described later in this chapter section, is useful for coracoid fractures25 (Fig. 17D1-11). The apical oblique view described by Garth and colleagues29 and the West Point lateral view30,31 are useful for evaluating anterior glenoid rim fractures. Computed tomography (CT) is a useful adjunct in evaluating intra-articular glenoid fractures. The contralateral normal shoulder, as well as the involved shoulder, may be scanned to provide a means for comparison of the pathologic findings noted in the involved shoulder, especially in adolescents.25 CT allows confirmation of the size, location, and degree of displacement of fracture fragments and detects the presence of instability. Three-dimensional images can be generated as well (Fig. 17D1-12). These three-dimensional computed tomographic reconstructions are useful in surgical planning. With appropriate software, the humeral head can be subtracted from the image so that an unobstructed view of the scapula and glenoid can be obtained if needed. Glenoid rim fractures associated with glenohumeral instability pose perhaps the most difficult decisions for treatment of fractures of the scapula among athletes. The

Figure 17D1-8   An anteroposterior view of the glenoid showing an anterior-inferior glenoid fracture.

Figure 17D1-9   A tangential scapular lateral view (trauma series lateral view) showing a displaced scapular body fracture with a bayonet position.

athlete does not always relay a history of glenohumeral dislocation in association with the initial injury. Because the decision regarding operative or nonoperative treatment of these glenoid rim fractures depends on whether they are associated with instability,25 the physician should make every attempt to verify the presence or absence of instability. In this regard, stress views with or without fluoroscopic control or an examination under anesthesia may be ­helpful.

Figure 17D1-10   The fractured base of the acromion with a posterosuperior humeral head dislocation is well seen on a tangential scapular lateral view.

862

DeLee & Drez’s Orthopaedic Sports Medicine

Fracture of the neck of the scapula is the second most common scapular fracture.25 The glenoid articular surface is intact, and the fracture line most often extends from the

suprascapular notch area across the neck of the scapula to its lateral border inferior to the glenoid. The glenoid and coracoid may be separated or may remain as an intact unit. Although glenoid neck fractures are often affected, their displacement is limited by an intact clavicle and by the acromioclavicular and coracoclavicular ligaments.1 Displacement, however, does occasionally occur in these fractures. It is often measured in terms of displacement percentage, amount of medialization of the glenoid, and the glenopolar angle compared with the normal side. The glenopolar angle,39 a measure of glenoid rotational displacement, is defined as the angle between the plane of the glenoid and a line running from the superior glenoid to the inferior pole of the scapula as seen on an anteroposterior radiograph. Optimal treatment of these fractures is debatable; most series report good functional results in patients with glenoid neck fractures regardless of the method of treatment7,23,40,41 (Fig. 17D1-13). Hitzrot and Bolling42 in 1916 stated that manipulation and traction had no effect on displaced glenoid neck fractures and that the results were so satisfactory without reduction that attempts to achieve reduction were unnecessary. Armstrong and Vanderspuy6 reported that six of seven of their patients with glenoid neck fractures had some residual stiffness, but no patient had a functional disability. Zdravkovic and Damholt13 came to the same conclusion in their report, in which patients had an average of 9 years of follow-up. A large meta-analysis by Zlowodzki and associates43 that reviewed 520 fractures in 22 case series found that 88% of fractures (7 of 8) that involved only the neck and were treated surgically had a good or excellent outcome. However, 77% (80 of 104) of those treated conservatively had a good or excellent outcome as well. Most authors recommend closed treatment of glenoid neck fractures.32-34,36 For displaced fractures, DePalma33 recommended closed reduction and olecranon pin traction for 3 weeks, and Bateman32 favors closed reduction

Figure 17D1-12   Three-dimensional computed tomographic scan showing the amount of medialization and caudal displacement of the glenoid component of the glenoid neck fracture.

Figure 17D1-13   A healed glenoid neck fracture with marked medial displacement and full range of motion. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

Figure 17D1-11   A fracture of the base of the coracoid, seen best on a Stryker notch view. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

TREATMENT OPTIONS IN ADULTS The recommended treatment of specific types of scapular fracture varies according to whether the fracture is intraarticular or extra-articular. Most extra-articular fractures (i.e., glenoid neck, scapular body or spine, acromion, and coracoid fractures) are managed nonoperatively.32-36 Intraarticular fractures, particularly those associated with glenohumeral instability, are managed operatively.11,15,33,37,38

Extra-articular Fractures Glenoid Neck Fracture

Shoulder

and a shoulder spica cast for 6 to 8 weeks in cases in which “shortening of the neck is sufficient to favor subluxation or interfere with abduction.” However, McLaughlin35 casts doubt on the usefulness of closed reduction in these fractures because most of them are affected and difficult to move. Lindholm and Leven34 found that all fractures healed in the position displayed at the time of the original injury (i.e., without additional displacement). Recently, van Noort and van Kampen44 reported on their results of nonoperative treatment of glenoid neck fractures. Of 13 patients, four had greater than 1 cm displacement, but no correlation could be made between displacement and outcomes. In fact, SF-36 scores were similar to agematched controls. Other authors1,45 have recommended open reduction of glenoid neck fractures (Fig. 17D1-14). Gagey and colleagues45 reported only one good result among 12 displaced glenoid neck fractures with closed treatment. They recommended open reduction and internal fixation (ORIF) because the displaced glenoid would “disorganize the coracoacromial arch.” These poor results were recently corroborated in a study by Pace and coworkers.46 They followed nine patients with scapular neck fractures for at least 2 years. None of the fractures was displaced more than 8 mm medially, 40 degrees angulated, or 100% translated. Despite these minimal displacements, all patients had some pain in the shoulder at rest or with work, and 33% had functional, measurable weakness. They attributed this pain and weakness to magnetic resonance imaging (MRI)– documented rotator cuff tendinopathy and subacromial bursitis caused by the glenoid malunion changing the cuff forces from compression to shear. Surgery and nerve exploration should also be considered for those fractures associated with a suprascapular nerve palsy confirmed by ­electromyography.7,47 Ada and Miller5 described 24 patients with displaced scapular neck fractures. Of the 16 treated conservatively, 50% complained of night pain, 40% had weakness of abduction, and 20% had decreased range of motion. ORIF was used to treat eight patients with scapular neck fractures having greater than 40 degrees of angulation or more than 1 cm of medial displacement of the glenoid surface. None of these patients complained of night pain, and all regained at least 85% of abduction. Romero and colleagues48 found that a glenopolar angle of less than 20 degrees (normal, 30 to 45 degrees) correlated to more severe pain, loss of function, and motion. Goss19 introduced the concept of the superior shoulder suspensory complex composed of a clavicle–­ acromioclavicular joint strut, the coracoclavicular ligament linkage, and the superolateral scapula (Fig. 17D1-15). He thought that injuries at two of these sites rendered an unstable “floating shoulder” and required operative stabilization. However, many authors noted that fractures of the glenoid neck and clavicle were frequently minimally displaced. The concept of an unstable floating shoulder was clarified in a biomechanical study by Williams and associates.49 They proved that in order to get significant displacement with ipsilateral scapular neck and clavicle fractures, there must also be injures to the coracoclavicular or acromioclavicular ligaments. These injuries have been managed both conservatively

863

and operatively, with good functional results reported for both treatment methods. Ramos and associates50 reported 92% good or excellent results in 16 patients treated conservatively with aggressive rehabilitation. Williams and associates51 correlated functional outcome with medial glenoid displacement in nine patients treated nonoperatively. Six patients with 2.2 cm or less of medial displacement had good or excellent results. They recommended that nonoperative management be considered in fractures with less than 3 cm of medial displacement of the glenoid. Edwards and ­coworkers52 obtained excellent clinical results with nonoperative treatment of 20 patients with less than 5 mm of scapular displacement. Van Wellen and colleagues53 described successful treatment of ipsilateral displaced glenoid neck and clavicle fracture with balanced traction. Hardegger and coworkers1 recommended open reduction and scapular fixation for displaced glenoid neck fractures associated with a fracture of the clavicle or disruption of the coracoclavicular ligaments (Fig. 17D1-16). They postulated that a severe displacement in these injuries would result in “functional imbalance” of the shoulder mechanism. Leung and Lam54 managed these injuries with open reduction of both the clavicle and scapula fractures because of “loss of the normal lever arm of the cuff.” They reported good or excellent functional results in all but 1 of the 15 patients they treated. Herscovici and associates55 believed that this injury disrupts the suspensory structures, leading to anteromedial displacement of the glenoid and ptosis of the shoulder due to muscle forces and the weight of the arm. They used open reduction of the clavicle in seven patients with excellent functional results. However, a study by Labler and colleagues,56 in which they reviewed their treatment of 17 ipsilateral clavicle and scapular neck fractures, underscores the fact that not all these injuries need ORIF and that the decision should be individualized depending on displacement. In their series, nine patients underwent ORIF of both fractures if possible, but at least of the clavicle, and eight were treated conservatively. The patients undergoing ORIF had more displaced fractures than the nonoperatively treated patients. Nevertheless, an equal number in each group obtained good and excellent results. The investigators thought that much of the outcome in their series was related to associated injuries. They recommended ORIF if displacement of the neck was more than 25 mm or the glenopolar angle was reduced 30 degrees compared with the opposite side. This algorithm is supported by van Noort and colleagues,57 who organized a multicenter study of floating shoulder injuries. Of the patients treated conservatively, those with a minimally displaced glenoid had average Constant scores of 85, but those patients with caudal dislocation of the glenoid had average Constant scores of 42 (Box 17D1-1).

Scapular Body Fracture Fracture of the body of the scapula is the most common type of scapular fracture and is correlated with the highest incidence of associated injury.25 When injury is the result of high-energy trauma, these fractures may be comminuted and displaced. Cain and Hamilton40 reported five scapular fractures in professional football players that were

864

DeLee & Drez’s Orthopaedic Sports Medicine

A

C

E

B

D

Figure 17D1-14   A, Anteroposterior radiograph of the scapula shows the difficulty determining fracture orientation on some radiographs.   B, Three-dimensional CT better illustrates the fracture configuration.   C-E, Postoperative radiographs show fixation of the scapular neck and body fractures, restoring the normal lever arm of the shoulder joint.

Shoulder

865

Box 17d1-1 Glenoid Neck Fractures 1. Isolated glenoid neck fracture a. Nonoperative treatment i. <1 cm medialized ii.  <20 degrees glenopolar angle difference iii.  Early passive range of motion with progressive active-assisted range of motion b. Operative treatment i. >1 cm medialized ii. >20 degrees glenopolar angle difference iii. Suprascapular nerve palsy—with exploration iv������������������������������������������������������� . ����������������������������������������������������� Intact lateral border fragment that impinges on humeral head in external rotation 2.��������������������������������������������������  ������������������������������������������������� Glenoid neck fracture with clavicle fracture a.����������������������������  ��������������������������� Nonoperative treatment i. <1 cm medialized ii������������������������������������������������ . ���������������������������������������������� <20 degrees glenopolar angle difference iii������������������������������������������� . ����������������������������������������� <1 cm clavicle fracture shortening iv����������������������������������������� . ��������������������������������������� Early passive range of motion b.�������������������������  ������������������������ Operative treatment i������������������������������������������������������ . ���������������������������������������������������� Clavicle open reduction with internal fixation (ORIF) only 1������������������������������������� . ����������������������������������� Minimal glenoid displacement 2������������������������������������� . ����������������������������������� >1 cm clavicle shortening ii�������������������������������� . ������������������������������ Clavicle and scapula ORIF 1���������������������������������� . �������������������������������� >1 cm clavicle shortening 2������������������������������������������������ . ���������������������������������������������� No glenoid reduction with clavicle ORIF 3������������������������������������ . ���������������������������������� >1 cm glenoid medialization 4���������������������������������������������������� . �������������������������������������������������� >20 degrees glenopolar angle difference

Figure 17D1-15   Diagram showing the linkage that is important in the superior shoulder suspensory complex.

Figure 17D1-16   Postoperative anteroposterior radiograph showing techniques of fixation of a “floating shoulder” injury.

the result of direct blows to the shoulder. The musculature surrounding the scapula makes nonunion a rare ­occurrence. Scapular malunion is rarely associated with clinical symptoms.2,11,33,36 Consequently, most authors favor a sling, ice, and supportive measures until the initial pain subsides2,11,35,36 Neer58 and Bateman32 reported immobilization using cross-strapping with adhesive moleskin in a nonambulatory patient with a scapular body fracture. This type of immobilization, however, has occasionally been associated with residual shoulder stiffness.35 On occasion, scapular malunion results in painful crepitus interfering with range of motion that may require removal of a bone prominence.59 Nordqvist and Petersson60 found poor long-term results in some patients with more than 10 mm of displacement. Cole61 includes 100% translation of the lateral border of the scapula in his indications for fixation of extra-articular scapula fractures. Usually these fractures are associated with spine or acromion fractures or with glenoid fractures. In these instances, the treatment of the body fracture is dictated by the associated fractures. Unusual causes of scapular body fractures include indirect injury, low-energy injury, and stress fracture. These injuries are treated nonoperatively with early mobilization. Wyrsch and coworkers62 reported a scapular body fracture in a professional boxer who sustained the injury during an attempted punch that completely missed the opponent. This injury was caused by a voluntary muscle contraction and was treated successfully with progressive active range

866

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17D1-18   Axial magnetic resonance imaging in a patient with superiorly displaced acromion fracture, showing complete disruption of posterior rotator cuff with defect in posterior deltoid. Also, the subscapularis is ruptured, and the biceps tendon has displaced from the groove. Figure 17D1-17   Scapular Y radiograph showing superiorly displaced acromion fracture with posterosuperior humeral head dislocation.

of motion and physical therapy. Deltoff and Bressler63 described the case of a scapular fracture sustained while a man was performing push-ups. McAtee64 reported an isolated scapular body fracture in a 41-year-old man playing touch football. Scapular body stress fracture has been reported in an elderly woman who had used a cane for ambulation.65 All these fractures were treated with a short period of protection and progressive range of motion.

Acromion Fracture Acromion fractures make up only 8% to 10% of all scapula fractures,1,9,10,66 and although fractures of the acromion are rare, when they do occur, it is usually the result of one of two mechanisms. First, an acromion fracture can result from a downward blow directly applied to the superior aspect of the acromion. Second, acromion fracture can result from superior displacement or dislocation of the humeral head (Fig. 17D1-17). When the injury is a result of a downward blow to the acromion, acromioclavicular dislocation is much more common than acromion fracture. Caution should be used in distinguishing this minimally displaced fracture from an os acromiale. Os acromiale usually has a more sclerotic border than an acute fracture, and in questionable cases, a radiograph of the contralateral side may be helpful because the os acromiale is bilateral in 60% of cases.23 The supraspinatus outlet view may be useful in estimating the amount of displacement if any is present.67 Significant displacement of an acromion fracture resulting from a downward blow to the acromion should alert the clinician to possible associated brachial plexus avulsions.25,68 Significant superior displacement of the acromion associated with superior displacement or dislocation of the humeral head should alert the clinician to possible injury

to the rotator cuff (Fig. 17D1-18).68 Fortunately, when a fracture does occur, it is usually minimally displaced. Less common mechanisms of acromion fractures include avulsions of the acromion from muscular over-pull and stress fractures from overuse. Most acromion fractures, because they are minimally displaced, should be treated closed.2,10,11,36,68 McLaughlin35 stated that “bony union is the rule, despite the presence or absence of immobilization, provided the fragments are in apposition.” Neer68 recommended symptomatic treatment only. Wilber and Evans,10 on the other hand, reported residual stiffness in patients with acromion fractures. They recommended cast immobilization in 60 degrees of abduction, 25 degrees of flexion, and 25 degrees of external rotation for 6 weeks. Most authors recommend ORIF for markedly displaced acromion fractures to reduce the acromioclavicular joint and prevent nonunion, malunion, and secondary impingement (Figs. 17D1-19 and 17D1-20).11,58,68 In a large meta-analysis study, Zlowodzki and associates43 examined data from many scapular fracture studies. Those that treated acromion fractures were combined and showed 70% good results with operative treatment and 82% good results with nonoperative treatment. There was no way to examine the indications for fixation across the different studies, and some of the studies included fractures not isolated to the acromion. Gorczyca and associates69 highlight the importance of looking for these injuries when the mechanism is suggestive. Their patient, with an unrecognized, nondisplaced acromion fracture, was allowed to bear weight with a platform walker. The acromion fracture then displaced and required operative treatment. Their surgical indications included displacement enough to effect deltoid function and lead to impingement in a manual laborer. Kuhn and associates70 recommended a classification system to help determine the need for operative intervention. Type I fractures with minimal displacement and type II displaced fractures without a decrease in the subacromial

Shoulder

867

shots and contraction of the posterior deltoid as the club struck the golf ball. Stress fractures have also been reported after subacromial decompression secondary to thinning of the acromion.74-76 However, if after nonoperative trials the stress fractures have not healed and are symptomatic, good results can be obtained with ORIF (Box 17D1-2).77

Glenoid (Intra-articular) Fractures

Figure 17D1-19   Postoperative anteroposterior radiograph showing open reduction with internal fixation of acromion and suture anchor repair of rotator cuff.

space warrant nonoperative treatment. In type II displaced fractures in which the subacromial space is diminished by the inferior pull of the deltoid on the acromial fragment; ORIF is required to prevent secondary impingement. All the fractures were treated conservatively in this study, and 5 of 19 (26%) went on to nonunion. Stress fractures of the acromion may occur in athletes. These injuries should be managed nonoperatively. Ward and colleagues71 reported a stress fracture at the base of the acromion in a professional football player resulting from repeated microstresses secondary to weightlifting and blocking assignments. Veluvolu and associates72 reported a case of an acromial stress fracture in a jogger using arm weights while running. Hall and Calvert73 described a woman golfer who sustained a stress fracture at the base of the acromion as a result of repetitive stress by repeated

Historically, intra-articular glenoid fractures—in the absence of associated glenohumeral instability—have been managed nonoperatively.10,25 Intact glenohumeral ligaments prevent gross displacement of the fracture and maintain a stable shoulder that many authors have had ­success treating nonoperatively11,15,36,37 (Figs. 17D1-21 and 17D1-22). Surgical intervention was initiated only for glenoid fractures associated with glenohumeral instability.78,79 These reports were limited because standardized outcome measures were lacking and the incidence of late glenohumeral arthritis was unknown. Surgical treatment of glenoid fractures has received greater attention recently.17,18,41,54,66,80-87 Indications for ORIF of intra-­articular glenoid fractures depend on fragment size, fracture ­ displacement, and stability of the glenohumeral joint.11,15,17,18,33,37,38,41,54,66,80-87 Glenoid rim fractures (type I) are usually sustained during traumatic glenohumeral subluxation or dislocation. In the setting of recurrent anterior instability, Rowe and coworkers11,38 recommended excision of an anterior rim fragment of up to 25% of the articular surface with repair of the capsule back to the remainder of the glenoid. DePalma33 believed that glenohumeral instability will result if the fragment is greater than 25% of the anterior rim, greater than 33% of the posterior rim, or displaced more than 10 mm. He recommended immediate open reduction for these situations.33 Rockwood37 recommended open reduction of the fragment with screw fixation if the fracture involves at least 25% of the ­glenoid

Box 17D1-2 A  cromion Fractures Mechanisms   1. Downward blow on acromion (rule out brachial plexus injures)   2. Superior blow by humeral head (rule out rotator cuff injures)   3.  Avulsions   4. Stress fractures (numbers 3 and 4 are managed conservatively unless symptomatic nonunion or displacement occurs)

Figure 17D1-20   Axillary lateral radiograph showing open reduction with internal fixation of acromion and reduction of the glenohumeral joint.

Kuhn Classification   1. I—Minimally displaced (nonoperative; early range of motion, minimal deltoid use until healed)   2. IIa—Displaced without subacromial space decrease (nonoperative)   3. IIb—Displaced with subacromial space decrease (open reduction and internal fixation) From Kuhn JE, Blasier RB, Carpenter JE: Fractures of the acromion process: A proposed classification system. J Orthop Trauma 8:6-13, 1994.

868

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17D1-21   Combined glenoid articular fracture with satisfactory position. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

Figure 17D1-22   This fracture healed well without problems  and with good preservation of the joint surface. (From Rockwood CA, Matsen FA [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

A

C

B

D

Figure 17D1-23   Anteroposterior radiograph (A) and computed tomographic scan (B) of an anterior glenoid rim fracture (type Ia) involving about 33% of the articular surface, resulting in anterior instability of the shoulder. The fracture was repaired with five suture anchors through an anterior deltopectoral approach. Postoperative radiographs show anatomic reduction (C) and a reduced glenohumeral joint (D).

Shoulder

B

A

C

869

D

E

F

Figure 17D1-24    Anteroposterior (A) and axillary (B) radiographs show a large displaced anteroinferior glenoid rim fracture, type Ia. Its position and displacement are better evaluated with sagittal (C) and axial (D) images. E and F, Postoperative radiographs showing excellent reduction of the fragment and glenohumeral joint using a headless compression screw system.

and is associated with instability. A biomechanical study by Gerber and Nyffeler88 showed that an anterior rim defect that was 50% of the maximal anteroposterior diameter of the glenoid would lead to 30% loss of anterior stability. Although debate exists among ­ surgeons about the amount of displacement and the size of the fragment that are acceptable, it is well accepted that rim fractures associated with persistent or recurrent instability should undergo ORIF.1,15,25,33,37,78,86,87 The size of the glenoid rim fragment and the quality of the bone determine the method of fixation to stabilize the fragment. Anterior glenoid rim fragments that are too small to accommodate a screw can be reduced and stabilized with suture anchors in the glenoid and sutures passed through the fragment and

tied (Fig. 17D1-23). Screw fixation can be used for large enough fragments with good bone quality (Figs. 17D1-24 and 17D1-25). Good results have been obtained with arthroscopically assisted reduction and percutaneous fixation of a displaced intra-articular glenoid fractures using cannulated screws,81 K-wires,41 or transglenoid sutures.89 Careful monitoring of fluid extravasation and assessment for compartment syndrome are recommended, as is having a low threshold to convert to open reduction if adequate reduction cannot be obtained arthroscopically. If the rim fracture is comminuted, the fragment can be excised, and a tricortical graft harvested from the iliac crest can be ­internally fixed to the glenoid rim.17 The results of operative fixation of glenoid rim fractures have been successful;

870

DeLee & Drez’s Orthopaedic Sports Medicine

A

B

C Figure 17D1-25   Anteroposterior radiograph (A) and magnetic resonance image (B) demonstrate a posterior glenoid fracture involving about 40% of the articular surface, leading to posterior subluxation of the humeral head. Note the comminution extending down into the articular surface and lateral scapular body. This fracture was repaired by limited internal fixation to restore the posterior glenoid rim and glenohumeral stability (C).

Shoulder

B

A

C

E

871

D

Figure 17D1-26   A, Anteroposterior radiograph shows an intra-articular glenoid fracture involving the superior half of the glenoid (type III).   B, Computed tomography demonstrates involvement of the coracoid process. C and D, The fracture was reduced and fixed with a superior-­toinferior lag screw and an anterior plate. E, The patient regained full forward elevation and had an excellent result.

872

DeLee & Drez’s Orthopaedic Sports Medicine

most authors reported restoration of glenohumeral stability and good functional results.25,41,78,80,81,86,87,89,90 For glenoid fossa fractures of types II to V, the amount of articular displacement and the degree of comminution determine the need for ORIF (Fig. 17D1-26). Goss17,18 recommended ORIF for an articular step-off of 5 mm, fracture separation of enough distance to preclude reliable healing, and subluxation of the humeral head out of the center of the glenoid. Kavanagh and associates82 described successful surgical treatment of displaced (4 to 8 mm) intra-articular fractures of the glenoid fossa. They emphasized that uncertainty still remained with regard to the amount of glenoid articular incongruity that could be accepted without risking long-term pain, stiffness, and post-traumatic degenerative arthritis. Poppen and Walker91 demonstrated that transarticular forces of 0.9 times body weight can be generated at the glenohumeral joint by lifting a 5-kg mass to shoulder height with the elbow extended. This suggests that disruption of articular congruence probably leads to unacceptable joint contact stress. Soslowsky and coworkers92 demonstrated that the maximal thickness of glenoid articular cartilage is 5 mm. On the basis of this information, several surgeons have adopted displacement of 5 mm or more as the indication for ­reduction and stabilization.66,78,79,82,93 The results of operative fixation of glenoid fossa fractures have been less predictable. Bauer and colleagues66 reported greater than 70% good or very good functional results for patients treated surgically for grossly displaced fractures of the glenoid rim, neck, fossa, and acromion. Leung and associates79 reported nine ­excellent and five good results in 14 patients treated surgically for displaced intra-articular glenoid fractures. Mayo and associates84 reviewed 27 displaced glenoid fossa fractures treated with ORIF at 43 months of follow-up and found that 82% of patients had good or excellent results. Three patients had articular incongruities measuring 2 mm or less. Ruedi and Chapman93 maintain that glenoid fractures that result in incongruity and instability benefit from ORIF to prevent arthritic changes. Recently, Schandelmaier and associates94 found Constant scores averaging 94% of the uninjured side in a series of operatively treated glenoid fossa fractures evaluated from 5 to 23 years ­postoperatively. Worse results were related to brachial plexus injures and infection. Overall, the large meta-analysis by Zlowodzki and associates43 found that 82% (45 of 55) of operatively treated glenoid fractures obtained good and excellent results, whereas only 67% (6 of 9) of those treated nonoperatively obtained equal outcomes. For type V and VI fractures, the degree of glenoid comminution determines treatment. ORIF is employed if the degree of comminution is minor and will allow stable fixation. If comminution is too severe to permit fixation of all fragments, limited reduction and stabilization of the articular segment are performed. Caution should be used in approaching these comminuted fractures because failed attempts at fixation disrupt the limited remaining soft tissue envelope. In selected cases with humeral head ­subluxation, partial fixation of larger fragments may be used to allow reduction of the humeral head, but any fixation used should

Box 17D1-3 Intra-Articular glenoid fractures 1. Type I (Ia—anterior, Ib—posterior) a. No shoulder instability i. Early range of motion ii. Strengthening b. Instability i. Ia—Anterior approach; Ib—posterior approach 1. Fragment fixation 2. Large fragment a. Screws 3. Small, comminuted fragment a. Suture anchors 2. Types II to IV a. Nonoperative i. <5 mm displacement ii. Centered humeral head b.���������������  �������������� Operative i.��������������������������  ������������������������� >5 mm displacement ii.������������������������������������  ����������������������������������� Humeral head subluxation 3.���������������������������������  �������������������������������� Types V and VI (comminuted) a.�����������������������������������������������  ���������������������������������������������� Nonoperative—sling, early range of motion i.��������������������������  ������������������������� Severe comminution ii.��������������������������������  ������������������������������� Centered humeral head b.���������������  �������������� Operative i.������������������������������������������������������  ����������������������������������������������������� Large fragments to accept fixation partial open reduction with internal fixation, with humeral head subluxation

allow early passive range of motion exercises. If fracture comminution prevents even limited fixation, nonoperative management is employed. Techniques of nonoperative management include sling and swathe immobilization and early range of motion, immobilization in an abduction splint and range of motion above the splint if possible, or traction and range of motion as allowed by the traction17 (Box 17D1-3).

TREATMENT OPTIONS IN CHILDREN The criteria for operative treatment of scapular and glenoid fractures in children and adolescents are not known. These types of injuries are extremely uncommon in children and adolescents. In addition, the capacity for remodeling in children makes the amount of tolerable displacement less certain. The amount of glenoid or scapular remodeling possible as a function of age is not known. Therefore, treatment recommendations must be made intuitively, based on the age and projected growth remaining. Nonoperative treatment is recommended for most scapular fractures in children. The only potential exceptions to this rule are displaced glenoid fossa fractures, floating shoulders with greater than 3 cm of displacement of the glenoid fragment, and glenoid rim fractures that are associated with recurrent or persistent glenohumeral instability.

Shoulder

Authors’ Preferred Method

o f ­T r e a t m e n t i n

Glenoid Neck Fracture

Most glenoid neck fractures are impacted and stable and do not require any reduction to obtain a good clinical result. Symptomatic local care, followed by passive exercises, will result in a rapid return of motion. Strengthening exercises may be instituted at 4 to 6 weeks. In the case of a double disruption of the superior shoulder suspensory complex, including the glenoid neck and another portion of the superior shoulder suspensory complex, surgical management becomes necessary if displacement at one or both sites in unacceptable. Reducing and stabilizing one of the disruptions will usually indirectly reduce and stabilize the other, if the reduction is performed acutely (within 1 week of injury). Ipsilateral glenoid neck and clavicle fractures are managed operatively if medial displacement of the glenoid neck fracture is more than 3 cm or if the intact ­lateral border fragment is positioned in such a way that it would impinge on the humeral head in external rotation. In most acute cases, reduction and stabilization of the clavicle will reduce and stabilize the glenoid neck. If clavicular fixation fails to reduce the medial displacement of the glenoid neck, reduction and fixation of the glenoid neck are carried out. Criteria for Return to Athletics. Healing is normally complete after about 6 weeks. Return to sports should be delayed, however, until range of motion has returned to normal and the strength of the shoulder is 90% of that of the uninvolved extremity. This normally takes 3 to 4 months. Body and Spine Fractures

Assuming that serious associated injuries have been ruled out, symptomatic treatment is indicated for virtually all patients with this type of fracture. Ice and sling immobilization are used initially. Within 1 to 2 weeks, passive range of motion and stretching exercises can be instituted. As pain and swelling subside, active range of motion and progressiveresistance exercises can be instituted. Criteria for Return to Athletics. The athlete should be withheld from competition until the fracture has healed and there is a full range of motion (typically 6 to 12 weeks). Foam padding over the posterior aspect of the scapula helps cushion blows that may be encountered when the athlete returns to contact competition. Acromion Fractures

Most acromion fractures are stable and are minimally displaced. Therefore, a sling is required for only 3 to 5 days. When pain diminishes enough to permit exercise, active and passive range of motion exercises are begun. Resisted deltoid exercises are avoided for 6 weeks to allow fracture union. In the instance of a displaced acromion fracture, ORIF are performed with tension band or compression screw fixation for distal fractures or a 3.5-mm malleable reconstruction plate for more proximal injuries. Caution should be exercised to rule out the presence of os acromiale, rotator cuff tear, or brachial plexus injury. Criteria for Return to Athletics. The athlete with an acromion fracture, regardless of whether it was displaced

873

Adults

and required fixation, should be withheld from competition until fracture union is complete and range of motion is pain free. This normally requires between 6 and 12 weeks. If the fracture was accompanied by a complication, such as a ­rotator cuff tear or brachial plexus injury, appropriate treatment should be instituted and return to sport delayed ­accordingly. Glenoid Fracture: Stable Glenohumeral Joint

Glenoid fossa fractures with less than 5 mm of displacement that are not associated with instability are treated symptomatically, with a sling for immobilization, until pain permits range of motion exercises (7 to 10 days). Glenoid rim or fossa fractures involving 20% or more of the articular surface that are displaced more than 5 mm are treated surgically. Types II to V glenoid fossa fractures with displacement of 5 mm or more are treated surgically. The approach depends on whether extensile exposure of the lateral angle of the scapula is required. Extensile exposure of the anterior aspect of the lateral scapular border is limited because of the axillary nerve. The posterior approach is the most utilitarian and is used for most type II, IV, and V fractures. An extensile exposure may be obtained by reflecting the posterior deltoid origin from its attachment on the scapular spine (Fig. 17D1-27). Displaced type III glenoid fossa fractures usually do not require exposure of the lateral scapular border and can frequently be stabilized through an anterior deltopectoral approach. If the posterior cortex is comminuted, a posterior deltoid-splitting approach may be used in place of an anterior approach. Rehabilitation is similar for operatively and nonoperatively treated glenoid fossa fractures without glenohumeral instability because both are stable. Passive mobilization is instituted within the first week of injury or surgery. Activeassisted range of motion is added at 4 weeks. Strengthening exercises are added as range of motion is restored and absence of pain permits (typically 6 to 8 weeks). When nonoperative management is undertaken (i.e., less than 5 mm of displacement), close follow-up with radiographs and physical examination is necessary to document maintenance of glenohumeral stability. Criteria for Return to Athletics. Return to sport is possible after fracture union has occurred, range of motion has reached its maximal level, and strength has returned to within 90% of that of the opposite extremity (typically 3 to 4 months, depending on the type of fracture). The athlete should be warned about the possibility of the development of glenohumeral arthritis, particularly if he or she is involved in a sport that places a large demand on the shoulder. Glenoid Fracture: Unstable Glenohumeral Joint

Anterior glenoid fractures (type Ia) associated with glenohumeral instability are best treated with surgical repair. If the fragment is of good quality and large enough to accept a small fragment screw, ORIF through an anterior ­ deltopectoral Continued

874

DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method

o f ­T r e a t m e n t i n

Adults—cont’d

Figure 17D1-27   Posterior extensile exposure of the glenoid. The skin incision begins at the base of the scapular spine and extends laterally, parallel to the scapular spine. At the posterolateral corner of the acromion, the incision turns distally and medially and parallels the lateral border of the scapula (A). The posterior deltoid is reflected from its origin on the scapular spine from its most medial attachment to the posterolateral corner of the acromion. The interval between the infraspinatus and teres minor is indicated by the curved arrow (B). The infraspinatus insertion is sharply divided and retracted medially, thereby protecting the supraspinatus nerve at the supraspinatus notch (C). The teres minor and the axillary nerve are retracted laterally, providing excellent exposure to the posterior scapula (D). A plate is applied to the posterior aspect of the glenoid (E).

approach is performed. The fragment is reduced and held provisionally with a temporary wire, and definitive fixation is performed with a standard 3.5-mm cortical screw, a partially threaded cancellous screw, or a variable pitch headless compression screw. Smaller fractures may be fixed with suture

anchors placed in the glenoid and sutures passed through the fragment. If the fragment is too small to accommodate a screw or suture, it is excised, and the anteroinferior capsule is repaired to the raw surface of the remaining glenoid. Pendulum exercises are begun immediately postoperatively.

Shoulder

Authors’ Preferred Method

o f ­T r e a t m e n t i n

Two to 3 weeks after surgery, the patient is encouraged to use the arm for everyday living activities, and gentle passive flexion and external rotation exercises are begun. At 6 to 8 weeks, stretching and strengthening exercises are instituted. Posterior glenoid fractures (type Ib) with significant displacement or posterior glenohumeral instability require operative intervention through a posterior approach. A deltoid-splitting approach is usually adequate. If the fragment is not too large, this may be combined with a muscle-splitting, internervous approach between the infraspinatus and teres minor or an infraspinatus-splitting approach. The more ­distal interval (i.e., between infraspinatus and teres minor) is preferred if the fragment involves the inferior third of the glenoid or if the fragment is large. This exposure can be made extensile by reflecting the infraspinatus and elevating the dorsal portion of the teres minor origin. If exposure of the midportion of the posterior glenoid is required, an infraspinatus-splitting approach may be more appropriate. If the fragment represents 25% or more of the articular surface,

Authors’ Preferred Method

of

Adults—cont’d

it is reduced and stabilized with a screw. Smaller fragments are excised, and the capsule is reattached to the remaining glenoid. Criteria for Return to Athletics.  Regardless of whether the fragment has been excised or fixed, the presence of glenohumeral instability dictates postponement of a return to competition. Rehabilitative efforts should concentrate on passive range of motion until fracture union has occurred. Further therapy should emphasize restoration of glenohumeral stability through rotator cuff and scapular stabilizer muscle exercises. Return to sport is possible after glenohumeral stability has been achieved, range of motion is restored, and strength has returned to within 90% of that of the opposite extremity (typically 4 to 6 months for anterior instability and 6 to 9 months for posterior instability). The athlete should be warned about the possibility of the development of glenohumeral arthritis, particularly if he or she is involved in a sport that places a large demand on the shoulder.

Treatment

All extra-articular scapular fractures in children who are 12 years and younger are treated nonoperatively initially. This recommendation is based on the belief that most of these fractures will heal and that the potential for remodeling is significant. The only exception to this rule is a displaced (>3 cm) glenoid neck fracture in combination with a displaced, ipsilateral clavicle shaft fracture (i.e., a floating shoulder). Under these circumstances, the clavicle is reduced anatomically and fixed rigidly with a plate and screws. Glenoid rim fractures in children who are 12 years and younger are not treated operatively unless they are associated with recurrent or persistent glenohumeral instability or are displaced 1 cm or greater. The indications in children older than the age of 12 years for operative treatment are the same as for adults. The surgical techniques for ­stabilization

in

Children

of glenoid rim fractures in children are the same as those described for adults. In children who are 12 years and younger, the technique of placing suture anchors within the remaining, uninvolved glenoid and passing sutures through the osseocartilaginous glenoid rim is preferred. Anchors with long-term absorbable (i.e., 3 months) sutures are preferred. Glenoid fossa fractures in patients older than 12 years are treated similarly to glenoid fossa fractures in adults. The potential for remodeling is greater in children who are 12 years and younger. The amount of residual displacement that is capable of being remodeled is not known. ORIF is currently performed in glenoid fossa fractures with 1 cm or greater of articular displacement. The surgical techniques, postoperative rehabilitation, and timing for return to athletics are the same as for adults.

S U G G E S T E D C

r i t i c a l

P

o i n t s

l Scapular fractures are relatively rare.

l The muscular envelope provides protection and dissipation of forces. l Usually, large direct forces are required for fracture, so the orthopaedist needs to evaluate for associated injuries. l Multiple ossification centers fuse at various times in the adolescent scapula, making comparison radiographs of the opposite scapula important when evaluating injuries in these patients.

875

R E A D I N G S

Cole P: Scapula fractures. Orthop Clin North Am 33(1):1-18, 2002. Goss TP: Double disruptions of the superior shoulder complex. J Orthop Trauma 7:99-106, 1993. Goss TP: Scapular fractures and dislocations: Diagnosis and treatment. J Acad Am Surg 3:22-33, 1995. Niggebrugge AH, van Hesden HA, Bode PJ, van Vugt AB: Dislocated intraarticular fractures of the anterior rim of the glenoid treated by open reduction and internal fixation. Injury 24:130-131, 1993. Ramos L, Mencia R, Alonso A, Ferrandez L: Conservative treatment of ipsilateral fractures of the scapula and clavicle. J Trauma 42:239-242, 1997.

R eferences Please see www.expertconsult.com

876

DeLee & Drez’s Orthopaedic Sports Medicine

S ect i o n

D

Injuries to the Glenoid, Scapula, and Coracoid 2. Fractures of the Coracoid in Adults and Children Allen Deutsch, Jason A. Craft, and Gerald R. Williams, Jr.

Fractures of the coracoid process of the scapula are uncommon and have received especially little attention in the sports medicine literature. Injury may occur as an apparently isolated phenomenon or in association with other injuries around the shoulder girdle. The coracoid process of the scapula serves as an anchoring point for the attachment of multiple ligaments and muscles. Ligaments include the coracohumeral and coracoacromial ligaments and the conoid and trapezoid components of the coracoclavicular ligaments. These last two ligaments perform an essentially suspensory function, exerting a static upward force on the scapula through the coracoid process.1 On the contrary, the muscular attachments exert a dynamic, active, and largely inferior force on the coracoid. The muscular origins comprise the pectoralis minor from the body and tip of the coracoid and the conjoined tendon from the tip. The conjoined tendon consists of the coracobrachialis and the short head of the biceps brachii. Consideration of these ligamentous and muscular attachments to the coracoid will give some insight into the proposed mechanisms of coracoid fracture. The location of the coracoid fracture (i.e., tip, body, or base in relation to the musculoligamentous structures) also determines the stability of the fracture and hence the propensity for displacement.

MECHANISM OF INJURY Mariani2 has suggested that direct and indirect mechanisms cause acute coracoid fracture. The direct type of injury appears to be a relatively rare phenomenon, probably because of the coracoid’s deep-seated, sheltered anatomic location. Therefore, a direct external blow to the coracoid severe enough to result in fracture usually involves massive trauma more common to motor vehicle crashes than to sporting endeavors. Direct trauma to the coracoid from the interior may arise in two circumstances, however. Anterior translation of the humeral head in subcoracoid glenohumeral dislocations may result in a direct coracoid impact that is sufficient to cause fracture.3-6 This too must be considered an uncommon injury, but it has been proposed that the combination of glenohumeral dislocation with coracoid fracture may be underdiagnosed. As discussed later, this shortcoming may be related to the difficulty of obtaining a good axillary lateral radiograph in an acutely painful shoulder or to the widespread practice of relying on the more difficult to interpret and less readily reproducible scapular lateral radiograph.

McLaughlin7 considered glenohumeral dislocation the most common cause of coracoid fracture. It has also been suggested that an undetected coracoid fracture might account for occasional cases of prolonged convalescence after glenohumeral dislocation3 and may conceivably be confused with recurrent anterior instability or rotator cuff disease. Wong-Chung and Quinlan described fracture of the coracoid tip that prevented closed reduction of an anterior glenohumeral dislocation.8 The other direct mechanism of coracoid fracture would in theory involve a blow to the lateral clavicle causing inferior displacement and impact with the coracoid.2 This would result in acromioclavicular ligamentous disruption but would preserve the coracoclavicular ligaments. Although this scenario appears not to have received specific attention in the literature, it is possible that some apparently isolated, undisplaced coracoid fractures might arise in this manner. The stress radiograph, discussed later, would be of particular relevance in this situation. So-called indirect mechanisms probably account for most fractures of the coracoid process. An indirect mechanism is probably most often responsible for isolated coracoid fractures.2,9 Smith10 described this mechanism in terms of a sudden, violent, and resisted contraction of the conjoined tendon and pectoralis minor. Wyrsch and associates described a case of an extra-articular scapula fracture with extension into the coracoid process in a professional boxer resulting from a violent muscle contraction.11 Mariani2 concluded that the coracoid is especially vulnerable to the stress of muscular action when the arm is in the position of abduction and extension. Benton and Nelson12 also drew attention to the stress placed on the coracoid when the arm is in this position. Another indirect mechanism of coracoid fracture involves a direct blow or fall onto the point of the shoulder,13 causing superior subluxation or dislocation of the lateral clavicle. Rather than the more common rupture of the coracoclavicular ligaments, the coracoid (proximal to the coracoclavicular ligaments) may fail.10,14-18 The pain accompanying the coracoid fracture may overshadow the acromioclavicular disruption so that, again, this injury may be misinterpreted as an isolated coracoid fracture if stress radiographs are not obtained.2,10 More commonly, however, the acromioclavicular dislocation is recognized, and the coracoid process fracture is unappreciated. Avulsion fractures of the coracoid resulting from strong traction forces have also been reported.19

Shoulder

Box 17D2-1 Coracoid Fracture Basics A. Rare injury B. Mechanisms 1. Direct a. External source (e.g., motor vehicle collision) b. Anterior humeral head dislocation c. Inferiorly displaced distal clavicle with acromioclavicular (AC) separation 2. Indirect a. Inferior: conjoined tendon and pectoralis minor ­avulsion b. Superior: superior AC separation with failure through coracoid instead of coracoclavicular ligaments c. Rare i. Suture tied under coracoid for AC ­reconstruction ii. Stress fractures

In addition to these direct and indirect mechanisms involving significant trauma, repetitive forces of a lesser degree have been reported by a number of authors to be a source of coracoid stress fracture. Boyer9 and Sandrock20 in separate reports described a fracture of the coracoid base in a young female trap shooter. The position of the gun butt directly over the coracoid tip was confirmed radiologically. Symptoms resolved, and the fracture healed when shooting was stopped. This appears to be an example of repetitive direct trauma resulting in stress fracture. A case of indirect trauma resulting from repetitive muscular action and leading to coracoid stress fracture in its distal half was reported by Benton and Nelson.12 They described a 19-year-old tennis player with a 4-year history of shoulder pain. It had been of insidious onset and was aggravated during serving. This patient eventually required excision of the distal fragment and reattachment of the conjoined tendon. Nontraumatic causes of coracoid fractures have also been described secondary to nonresorbable coracoclavicular cerclage fixation during acromioclavicular reconstruction, as well as in association with massive rotator cuff tears (Box 17D2-1).21

877

­ reviously mentioned case report by Benton and Nelson12 p of a distal stress fracture in a young tennis player. These problems of union and displacement and hence the possibility of persistent symptoms appear to be related to the location of the fracture in relation to the coracoclavicular ligaments.1 A fracture within the broad area of the attachment of these ligaments is likely to be splinted and minimally displaced. As the fracture line moves toward the tip and hence beyond the attachments of the conoid and trapezoid ligaments, however, the coracoid tip becomes increasingly subject to the displacing action of the pectoralis minor and the conjoined tendon.4,12 Coracoid fracture has been described in association with acromioclavicular separation and glenohumeral dislocation. Montgomery and Loyd16 described two adolescents with coracoid apophyseal avulsion at the site of attachment of the coracoclavicular ligament. Combalia and colleagues described a case of a 12-year-old boy who sustained an acromioclavicular dislocation with epiphyseal separation of the coracoid process as a result of a fall during a soccer match.25 Certain other associations have also been recognized. Zilberman and Rejovitzky24 encountered coracoid fractures in conjunction with clavicular shaft and acromion fractures. Wolf and colleagues5 described a combination of coracoid base fracture and avulsion of a thin spicule from the superior border of the scapula medial to the coracoid. This pattern of injury may be due to the fact that the fragments are connected by the suprascapular ligament. Acromioclavicular separation with coracoclavicular ligament disruption and coracoid fracture has also been described.26,27 Neurologic injuries may also be seen with coracoid fractures.13 The brachial plexus deep to the coracoid and pectoralis minor may be contused, resulting in either specific or subtle patchy neurologic deficits. Basal coracoid fractures may especially result in suprascapular nerve entrapment and may be confused with rotator cuff tears (Fig. 17D2-1).13

CLINICAL FEATURES OF CORACOID FRACTURES History

PATTERN OF CORACOID FRACTURE The pattern of coracoid fracture is variable. Most such fractures occur through the base of the process,2,3,9,22,23 and this is almost invariably the case with associated acromioclavicular injuries. Basal coracoid fractures may very rarely involve a significant portion of the superior portion of the glenoid articular surface. Fractures of the body or tip of the coracoid process of the scapula without acromioclavicular injury appear to be related more to violent muscular action1,4,12,24 but have been associated with anterior glenohumeral dislocation.6 These more distal fractures also appear to be more troublesome in terms of delayed union or nonunion and the related problem of displacement.4 Both these problems are well demonstrated in the

The history of coracoid fractures caused by shoulder injury is not particularly specific. About one third of reports of coracoid fractures attribute the injury to a motor vehicle crash.14 The next most common history obtained is that of a fall onto the point of the shoulder or a direct blow to the shoulder. It is of interest that football injuries involving both these mechanisms appear regularly in the literature. Typical football injuries have included a fall onto the shoulder, a direct blow from running into a goalpost, and apparently impact sustained during a rugby scrum.2 A fall backward onto the extended abducted and externally rotated arm also appears to be a well-established mechanism1,4 and may be brought out in the history. As previously discussed, coracoid stress fracture appears to be a real entity in sportsmen.4,9,20 A history of acute injury is conspicuously absent, and recalcitrant symptoms

878

DeLee & Drez’s Orthopaedic Sports Medicine

A

B

Figure 17D2-1   A, The coracoid process serves as the insertion of the pectoralis minor and the origin of the conjoined tendon of the coracobrachialis and the short head of the biceps. Reflection of the pectoralis reveals the proximity of the underlying brachial plexus. B, The suprascapular nerve is at risk with fractures of the coracoid base that extend into the superior scapular notch.

of insidious onset may pose a diagnostic dilemma. Pain is an invariable complaint with acute fracture of the coracoid but may be poorly localized at the front of the shoulder. Pain may be aggravated by arm movements that exert muscular forces onto the coracoid. The patient may recognize these particular movements, which include elbow flexion,28 shoulder flexion with the elbow extended,2,12 and combined shoulder abduction, extension, and external rotation.28 Similarly, the patient may volunteer the fact that the pain is aggravated by deep inspiration owing to pectoralis minor activation.2,12 As with all shoulder girdle injuries, neurologic ­symptoms13 may be prominent, especially in the form of transient paresthesias. This is not surprising in view of the intimate relationship of the major neurovascular structures to the coracoid and pectoralis minor.

Physical Examination Unless there has been an associated injury to the acromioclavicular joint or a glenohumeral dislocation, there will usually be no striking external abnormality. Falls onto or direct blows to the shoulder may, of course, result in localized areas of contusion or abrasion. Despite the deep-seated location of the coracoid process, swelling or loss of definition in the deltopectoral interval may be detected.28 Marked localized tenderness on palpation is a key finding. Specific stress tests2,12,28 such as resisted elbow flexion, resisted straight-arm raising, and coughing may also sharply localize discomfort to the coracoid region. Whenever suspicion of a coracoid injury is raised, attention should be specifically directed to the acromioclavicular joint and vice versa. Local acromioclavicular tenderness, swelling, subluxation, or obvious superior dislocation of the lateral clavicle may be apparent. As always, comparison with the normal shoulder may be of considerable assistance. Neurologic examination is mandatory in cases of coracoid fracture. Because deficits may be patchy and subtle, a thorough brachial plexus assessment is essential. Special emphasis should be placed on suprascapular nerve evaluation because of the risk for entrapment.13

Diagnostic Studies The anteroposterior view should be part of the routine shoulder series and is of particular relevance in detecting associated acromioclavicular injuries or fractures (Fig. 17D2-2).5,24 Although it is possible to diagnose some coracoid fractures with a plain anteroposterior radiograph,28 it is likely that most will be overlooked without additional views (Fig. 17D2-3).12 This is because the coracoid process is foreshortened and projected over the acromion and the spine of the scapula in this view.29 Many authors have stressed the value of the axillary lateral view in diagnosing coracoid fractures,3,6,9,12-14 but even this view may fail to demonstrate a basal coracoid fracture (Fig. 17D2-4).29 Froimson28 has also pointed out that the abduction required for a good axillary lateral view may be difficult to obtain because of the pain it provokes in patients with acute coracoid fractures. A much better profile of the coracoid, including the base, can be obtained by tilting the x-ray beam in a cephalic direction. This obviates the need to move the patient’s arm. Most authors recommend a supine position

Figure 17D2-2   Combined coracoid process fracture and acromioclavicular dislocation as demonstrated on routine anteroposterior radiograph.

Shoulder

A

879

B

Figure 17D2-3   A, A coracoid process fracture that is well visualized on a routine anteroposterior radiograph. B, Fractures of the base of the coracoid process can be difficult to visualize on routine anteroposterior views.

with a 30- to 35-degree cephalic tilt.2,6,17,24,29 Froimson found that a cephalic tilt of as much as 45 to 60 degrees was quite useful. Although the Stryker notch view was originally intended to identify the Hill-Sachs lesion characteristic of anterior ­glenohumeral dislocations, the authors have found this technique especial­ly useful in studying coracoid fractures (Fig. 17D2-5).30 Kopecky and colleagues22 reported on the value of computed tomographic scanning when doubt exists. The authors have also found that this modality is helpful in clarifying coracoid fracture morphology (Fig. 17D2-6). Specific attention should be paid to the acromioclavicular joint in the presence of coracoid fracture. Despite acromioclavicular dislocation, the coracoclavicular distance will be maintained (Fig. 17D2-7).2,10,14 Erect anteroposterior films of both shoulders are necessary to compare the coracoclavicular distance on the injured side with the uninjured side. Although stress views may be helpful in some cases, most often they are not required.

A

Normal coracoid epiphyses or apophyses should not be confused with fractures.12,30 The coracoid process forms from two ossification centers. The basal ossification center also forms the upper third of the glenoid, whereas the other forms the main body of the coracoid. The basal epiphyseal plate fuses at puberty. Smaller accessory ossification centers that are shell-like and rounded may be seen medial to the coracoid base or at its very tip (Fig. 17D2-8). Cottalorda and colleagues described a case of a 15-year-old boy who suffered a displaced epiphyseal separation as a result of a direct fall onto the shoulder while participating in judo.31 The only other diagnostic study apart from radiography that may be necessary is electromyography, which is indicated if suprascapular nerve entrapment is suspected.13 This most often occurs subacutely when the pain ­ associated with the acute fracture has subsided and strength of the supraspinatus and infraspinatus can be accurately assessed.

B

Figure 17D2-4   A, Fractures of the tip of the coracoid process are frequently well visualized on an axillary lateral radiograph.   B, Basal fractures of the coracoid process can be difficult to demonstrate, even on an axillary lateral radiograph.

880

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17D2-7   An anteroposterior stress radiograph in patient with a combined coracoid process fracture and an acromioclavicular dislocation reveals that the coracoclavicular interspace has been maintained.

Figure 17D2-5   The Stryker notch view is helpful in demonstrating basal coracoid process fractures that are difficult to visualize on other routine views.

Treatment Options Acute isolated fracture of the coracoid base is almost invariably nondisplaced and is treated conservatively with the expectation of a good result.1 If the acromioclavicular joint is sound, the basal fracture is splinted by the coracoclavicular ligaments, and displacement is minimal. Fracture surfaces are relatively large and predominantly cancellous. Prompt union is generally anticipated. Nonunion is infrequent, may be related to premature return to vigorous activities, and may require bone grafting and screw fixation (Fig. 17D2-9). Essentially, the basic treatment ought to be symptomatic—resting the affected arm in a sling,

Figure 17D2-6   Computed tomographic scanning is occasionally useful for clarifying the morphology of certain coracoid fractures.

a­ dministering analgesia for the initially severe pain, and gradually mobilizing the shoulder as symptoms regress and radiographic healing occurs. Basal coracoid fracture with suprascapular nerve palsy is a rare indication for early operative exploration, especially if there is any displacement of the fracture with narrowing of the suprascapular notch (Fig. 17D2-10). The prognosis for recovery from suprascapular entrapment appears to be poor once cancellous bone has formed in the region of the suprascapular notch.13 As the location of an isolated coracoid fracture approaches the tip of the coracoid process, opinions about treatment diverge. The closer the fracture gets to the tip of the coracoid process, the smaller is the stabilizing effect of the coracoclavicular ligaments and the greater is the propensity for displacement or nonunion exerted by the muscular attachments at the tip. Rowe32 recommends simple approximation of fragments with nonresorbable sutures in this situation, whereas McLaughlin7 considers that fibrous union is not uncommon and is rarely accompanied by any residual symptoms (Fig. 17D2-11). Benton and Nelson described late surgical treatment of a displaced coracoid fracture that irritated the surrounding soft tissue structures.12 Marked displacement and delayed union of these more distal fractures may significantly delay recovery.1,4,12 Moreover, surgical management of significantly displaced fractures frequently yields satisfactory results. When surgical management is selected, the choice is between fixation of the fragment (Fig. 17D2-12) and, if it is especially distal, excision of the fragment and reattachment of the pectoralis minor and conjoined tendon to the residual coracoid stump. Similarly, for a combined coracoid fracture and acromioclavicular dislocation, there appears to be no single best line of management. Martin-Herrero and coworkers33 reviewed seven patients with coracoid fractures. Four had associated acromioclavicular joint injuries, but all had “good or very good” results at final follow-up with conservative treatment. Bernard and associates,14 in a comprehensive review of this dual injury, found that surgical and nonsurgical methods of treatment appear to offer equally favorable results. It is also of interest that coracoid nonunion appeared to be no more common with this injury. Should nonunion arise, its combination with complete acromioclavicular dislocation appears to be compatible with a functional pain-free result.15

Shoulder

A

881

B

Figure 17D2-8   A, The normal basal coracoid physis as demonstrated on the Stryker notch view. B, The normal ossification center at the tip of the coracoid process.

A

B

Figure 17D2-9   A, Stryker notch view demonstrating nonunion of the base of the coracoid process in an 18-year-old football player. B, The nonunion was fixed through an anterior deltopectoral approach with an interfragmentary screw and bone graft.

882

DeLee & Drez’s Orthopaedic Sports Medicine

Figure 17D2-10   The scapula viewed from posterior, with the spine of the scapula and acromion removed. The basilar coracoid fracture is displaced and impinging on the suprascapular nerve through the suprascapular notch.

Acromion

Displaced Superior transverse scapular ligament

Humeral head

Entrapped suprascapular nerve

Cut end of scapular spine

Scapula

Posterior view acromion removal

Eyres and coworkers recently described a classification system for coracoid fractures that is based on the size of the coracoid fragment and whether a clavicle fracture or an acromioclavicular separation is present.34 They recommend surgical treatment of displaced fractures that extend into the glenoid or scapular body, fractures associated with clavicle fracture or acromioclavicular separation, and fractures with interposed ­ coracoid fragments preventing reduction of glenohumeral joint dislocation. Ogawa and colleagues described their experience with treatment of 67 coracoid fractures.35 Most patients had associated injuries, including acromioclavicular separations, clavicle fractures, and scapula fractures, that disturbed the link between the scapula and the clavicle. They recommended open reduction and internal fixation of

coracoid fractures that were posterior to the attachment of coracoclavicular ligaments to restore the scapuloclavicular connection to permit early therapy. When one is dealing with athletes with combined acromioclavicular dislocation and a displaced coracoid base fracture, the unusual physical demands of their sport may influence treatment more toward anatomic restoration. If this course is taken, the fracture location itself may preclude coracoclavicular fixation techniques, and trans­ articular pins may become necessary (Fig. 17D2-13).10 Screw fixation of the coracoid back to the body of the scapula is technically difficult but may be indicated if the fracture is not fixed acutely. Under these circumstances, direct ­ fracture exposure is necessary to ­ accomplish ­reduction.

Figure 17D2-11   A, Axillary lateral view demonstrating an acute fracture of the coracoid in a 42-year-old woman who sustained a glenohumeral dislocation. This patient also suffered concomitant suprascapular nerve palsy and was managed nonoperatively. Continued

A

Shoulder

883

Figure 17D2-11, cont’d   B and C, At 2 years’ follow-up, the suprascapular nerve palsy had resolved, the patient had returned to all previous activities, and physical examination revealed small deficits in forward elevation and internal rotation.

B

C

Figure 17D2-12     Anteroposterior (A) and axillary lateral (B) images show excellent reduction and fixation of an unstable distal coracoid fracture.

A

B

Figure 17D2-13   Combined coracoid process fracture and acromioclavicular dislocation treated with reduction and acromioclavicular wires.

884

DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method

of

Treatment

Without Acromioclavicular Dislocation

Fractures of the Distal Coracoid Process

Because the coracoclavicular and acromioclavicular ligaments remain intact in this injury, the fracture is stable. Therefore, treatment with a sling for comfort is sufficient. Pendulum exercises should be encouraged to prevent loss of motion in the shoulder; however, overhead elevation is restricted for 4 to 6 weeks to allow healing to occur at the base of the coracoid process. The athlete may return to competition after complete healing of the fracture and return of a full, painless range of motion. This usually requires 6 to 10 weeks.

Treatment of these fractures is symptomatic. Most patients become asymptomatic in 8 to 10 weeks regardless of the presence or absence of nonunion. Should a symptomatic nonunion result, excision of the fragment and reattachment of the pectoralis minor or conjoined tendons to the residual coracoid stump are curative. The athlete can return to competition when pain allows a full range of motion. This normally occurs about 4 to 6 weeks after injury (Box 17D2-2).

With Acromioclavicular Dislocation

When a coracoid process fracture is accompanied by a ­severely displaced acromioclavicular dislocation, open ­reduction with internal fixation is usually indicated. The displacement criteria for fractures of the coracoid base that involve the articular surface of the glenoid are more stringent than the criteria for fractures without articular involvement. Articular displacement of 5 mm or greater is treated with open reduction and internal fixation. Fixation options are limited because of the fracture of the coracoid process. Reduction and fixation of the acromioclavicular joint must be ­accompanied by reduction of the coracoid base and intraarticular glenoid component. Therefore, despite the small risk for acromioclavicular arthritis, fixation with transarticular smooth pins is indicated. The pins are removed after 6 to 8 weeks when radiographs reveal healing of the fracture. Before the pins are removed, the patient is not permitted to raise his or her arm overhead. If the fracture is not reduced within 1 week, reduction of the acromioclavicular dislocation will most likely not result in simultaneous reduction of the coracoid fracture. Under these circumstances, direct exposure of the coracoid base fracture is required. Interfragmentary screw fixation is performed. After pin removal, fracture healing, and restoration of a full, pain-free range of motion, the athlete may return to competition. It is not necessary to wait for screw removal in those treated with interfragmentary screw fixation. Return to competition is encouraged after restoration of a full, painfree range of motion. This normally requires 8 to 12 weeks.

POSTOPERATIVE MANAGEMENT AND REHABILITATION The postoperative rehabilitation of coracoid process fractures is the same as the nonoperative management of fractures that do not require surgery. This is because fractures selected for nonoperative management are either minimally displaced and stable or displaced without the need for anatomic reduction. Likewise, operatively treated ­fractures have been rendered stable. Pendulum exercises are instituted within 7 to 10 days of injury or surgery. Supine passive flexion to 90 degrees or less and passive external rotation are added 3 to 4 weeks after injury or surgery.

Box 17D2-2 C  oracoid Fracture ­Treatment Outline A.  Basal fractures 1.  Nonoperative treatment a.  Nondisplaced basal fractures i. Splinted by intact coracoclavicular ­ligaments



ii.  Cancellous surface allows reliable ­healing b.  Sling for comfort c.  Progressive range of motion as healing ­allows 2.  Operative treatment a.  Associated suprascapular nerve palsy b.  Extension into glenoid articular surface c. Associated operative acromioclavicular joint separation B.  Nonbasal fractures 1. More distal fractures lose splinting of ­coracoclavicular ligaments 2. Conjoined tendon and pectoralis minor place distracting forces at fracture 3.  Operative treatment a.  More than 1 cm displacement b.  Symptomatic nonunion c. Displacement precluding ­glenohumeral ­reduction after anterior shoulder ­dislocation

External rotation is particularly important in coracoid base fractures because scarring at the coracohumeral ligament attachment site may result in loss of passive external rotation. Passive and active-assisted overhead elevation is permitted 6 weeks after injury or surgery. An overhead pulley is useful in allowing the athlete to improve overhead elevation progressively. Rotator cuff and deltoid strengthening exercises are added 6 to 8 weeks after surgery. As these exercises are progressed, the important scapular rotators are also strengthened. Return to athletic competition is determined individually based on attainment of a pain-free range of motion and strength that approaches 90% of the uninjured side.

Shoulder

C

r i t i c a l

P

o i n t s

l Suspicion is the key to diagnosis, especially with direct blow to the shoulder or with an AC joint separation. l Pain is experienced with use of muscles attached to ­coracoid (pectoralis minor—deep inspiration; elbow ­flexion—short head biceps tendon; shoulder flexion with ­elbow ­extended—coracobrachialis). l A thorough brachial plexus examination is necessary ­because of proximity (especially axillary, musculocutaneous nerves).

S U G G E S T E D

885

R E A D I N G S

Garcia-Elias M, Salo JM: Nonunion of a fractured coracoid process after dislocation of the shoulder. J Bone Joint Surg [Br] 67:722, 1985. Gil JF, Haydar A: Isolated injury of the coracoid process: Case report. J Trauma 31:1696, 1991. Martin-Herrero T, Rodriquez-Merchan C, Munera-Martinez L: Fracture of the coracoid process: presentation of seven cases and reveiw of the literature. J Trauma 30:1597-1599, 1990. Ogawa K, Yoshida A, Takahashi M, Ui M: Fractures of the coracoid process. J Bone Joint Surg [Br] 79:17-19, 1997. Wong-Chung J, Quinlan W: Coracoid process preventing closed reduction of ­anterior dislocation of the shoulder. Injury 20(5):296-297, 1989.

R eferences Please see www.expertconsult.com

S E C T I O N

E

Scapulothoracic Disorders in Athletes John E. Kuhn

Scapulothoracic disorders include crepitus and bursitis, two related conditions that are not infrequently seen in the athletic population; scapular winging; and scapulothoracic dyskinesis. These conditions are often related to alterations in normal scapulothoracic kinematics. Before describing these conditions, it is important to understand the anatomy of the scapulothoracic articulation.

ANATOMY AND BIOMECHANICS OF THE SCAPULOTHORACIC ARTICULATION Seventeen muscles have their origin or insertion on the scapula (Box 17E-1, Fig. 17E-1), making it the cornerstone for coordinated upper extremity activity. These muscles include the rhomboideus major and minor, the levator scapulae, the serratus anterior, the trapezius, the omohyoid, and the pectoralis minor. Scapular winging or scapulothoracic dyskinesis may occur as a result of dysfunction of these muscles. The rotator cuff muscles (supraspinatus, infraspinatus, subscapularis, and teres minor) contribute to control activities of the glenohumeral articulation. Disorders of these muscles are common in athletes and are covered in other sections of this text. The scapulohumeral muscles provide power to the humerus and include the deltoid, the long head of the biceps, the short head of the biceps, the coracobrachialis, the long head of the triceps, and the teres major. Almost every functional upper extremity movement has components of scapulothoracic and glenohumeral motion.

Box 17E-1  Muscles with Origins or ­Insertions on the Scapula Scapulohumeral Muscles Long head of biceps Short head of biceps Deltoid Coracobrachialis Teres major Long head of triceps Scapulothoracic Muscles Levator scapulae Omohyoid Rhomboid major Rhomboid minor Serratus anterior Trapezius Pectoralis minor Rotator Cuff Muscles Supraspinatus Infraspinatus Subscapularis Teres minor From Kuhn JE: The scapulothoracic articulation: Anatomy, biomechanics, ­pathology and management. In Iannotti JP, Williams GR Jr (eds): ­Disorders of the Shoulder: Diagnosis and Management. Philadelphia, Lippincott ­Williams &Wilkins, 1999, pp 817-845.

886

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Conjoined tendon of coracobrachialis and biceps

Trapezius muscle

Pectoralis minor muscle Scapular notch Omohyoid muscle

Omohyoid Coracoid muscle Superior angle Levator scapular muscle Supraspinatus muscle Rhomboid minor muscle Infraspinatus muscle Medial border

Acromion Deltold muscle Glenoid fossa Scapula neck Triceps(long head)

Teres minor muscle

Serratus anterior muscle Subscapularis muscle

Lateral border

Rhomboid major muscle Teres major muscle Inferior angle Latissimus dorsi muscle Figure 17E-1   Muscles with origins or insertions on the scapula. Anterior and posterior views of the scapula demonstrate the multiple attachment sites for muscles of the scapula, making it the center for coordinated upper extremity motion. (Redrawn from Kuhn JE: The scapulothoracic articulation: Anatomy, biomechanics, pathology and management. In Iannotti JP, Williams GR Jr [eds]: Disorders of the Shoulder: Diagnosis and Management. Philadelphia, Lippincott Williams &Wilkins, 1999, pp 817-845.)

While at rest, the scapula is anteriorly rotated relative to the trunk about 30 degrees.1,2 The medial border of the scapula is also rotated, with the inferior pole diverging away from the spine about 3 degrees. The scapula is also tilted forward about 20 degrees in the sagittal plane when viewed from the side.1 It is thought by some that deviations in this normal alignment may contribute to glenohumeral instability, and likely contribute to scapulothoracic crepitus and bursitis.

Bursae around the Scapula The normal smooth gliding motion of the scapula on the chest wall occurs as a result of multiple scapulothoracic bursae. Two major, or anatomic, bursae and four minor, or adventitial, bursae have been described for the scapulothoracic articulation (Box 17E-2, Fig. 17E-2). The major bursae are easily and reproducibly found,3,4 whereas the adventitial bursae are not. The first major bursa is found in the space between the serratus anterior muscle and the chest wall. The second major bursa is located between the subscapularis and the serratus anterior muscles.3,5,6 The superomedial angle and the inferior angle of the scapula appear to be the two anatomic regions involved in patients with scapulothoracic bursitis. When symptomatic, these areas tend to develop inflamed bursae; however, these bursae may be adventitious because they are not found ­ reliably.3,7,8 When scapulothoracic bursitis affects the inferior angle of the scapula, most authors agree that the inflamed bursa lies between the serratus anterior muscle and the chest wall.6,9,10 This bursa has been called the infraserratus bursa6 and the bursa mucosa serrata.10,11 The second and more common site of scapulothoracic ­bursitis

occurs at the superomedial angle of the scapula. Codman believed that the inflamed superomedial angle bursa was also an infraserratus bursa, lying between the upper and anterior portion of the scapula and the back of the first three ribs.6 O’Donoghue also believed that the bursa between the serratus anterior and the chest wall was the involved bursa in athletes with pain and crepitus.12 Von Gruber, on the other hand, identified a bursa in this region between the subscapularis and the serratus anticus muscles, which he called the bursa mucosa angulae superioris scapulae.13 ­Williams and colleagues identified a third major bursa, the scapulotrapezial bursa, which lies between the superomedial scapula and the trapezius muscle.4 This bursa contains the spinal accessory nerve but is not thought to be a source of scapulothoracic crepitus or bursitis. A third minor or adventitial bursa, which Codman believed was the site of painful crepitus in scapulothoracic crepitus, was called the trapezoid bursa, and is found over the triangular surface at the medial base of the spine of the scapula under the trapezius muscle.6 Some believe that these minor bursae are adventitial and develop in response to abnormal pathomechanics of the scapulothoracic articulation.3,7,8 It would not be surprising, then, to find these ­bursae inconsistently or in different soft tissue planes.

SCAPULOTHORACIC CREPITUS Pathophysiology Symptomatic scapulothoracic crepitus has been described by a number of different authors and has been called the snapping scapula,9 the washboard syndrome,14 the ­scapulothoracic

Shoulder

887

Box 17E-2  Bursae Around the Scapula

Box 17E-3  Causes of Scapulothoracic Crepitus

Major/Anatomic Bursae Infraserratus bursae: between serratus anterior and chest wall Supraserratus bursae: between subscapularis and serratus anterior muscles Scapulotrapezial bursae: between superomedial scapula and the trapezius

Interposed Tissue

Minor/Adventitial Bursae Superomedial Angle of the Scapula Infraserratus bursae: between serratus anterior and chest wall Supraserratus bursae: between subscapularis and serratus anterior Inferior Angle of the Scapula Infraserratus bursae: between serratus anterior and chest wall Spine of Scapula Trapezoid bursae: between medial spine of scapula and trapezius From Kuhn JE, Hawkins RJ: Evaluation and treatment of scapular disorders. In Warner JJP, Iannotti JP, Gerber C (eds): Complex and Revision ­Problems in Shoulder Surgery. Philadelphia, Lippincott-Raven, 1997, pp 357-375.

syndrome,15 the rolling scapula,7 the grating scapula,16 and the scapulocostal syndrome.17 Boinet18 was the first to describe this disorder in 1867. Thirty-seven years later, Mauclaire19 classified scapulothoracic crepitus into three groups: “froissement” was described as a gentle friction sound and was thought to be physiologic, “frottement” was a louder sound with grating and was usually ­pathologic, and “craquement” was a loud snapping sound and was always pathologic. These scapular noises are thought to occur from two sources, either from anatomic changes in the tissue interposed between the scapula and the chest wall or by incongruence in the scapulothoracic articulation (Box 17E-3). Extrapolating from Milch,9 frottement may

Infraserratus bursa Supraserratus bursa

Infraserratus bursa Supraserratus bursa Trapezoid bursa Infraserratus bursa

Figure 17E-2   Bursae of the scapula. The locations of both anatomic and adventitial bursae are shown. (Redrawn from Kuhn JE, Hawkins RJ: Evaluation and treatment of scapular disorders. In Warner JJP, Iannotti JP, Gerber C [eds]: Complex and Revision Problems in Shoulder Surgery. Philadelphia, Lippincott-Raven, 1997, pp 357-375.)

Muscle Atrophy Fibrosis Anatomic variation Bone Rib osteochondroma Scapular osteochondroma Rib fracture Scapular fracture Hooked superomedial angle of scapula Luschka’s tubercle Reactive bone spurs from muscle avulsion Other Soft Tissue Bursitis Tuberculosis Syphilitic lues Abnormalities in Scapulothoracic Congruence Scoliosis Thoracic Kyphosis From Kuhn JE, Hawkins RJ: Evaluation and treatment of scapular disorders. In Warner JJP, Iannotti JP, Gerber C (eds): Complex and Revision ­Problems in Shoulder Surgery. Philadelphia, Lippincott-Raven, 1997, pp 357-375.

s­ uggest soft tissue pathology or bursitis, whereas craquement may suggest bony pathology as the source of symptomatic scapulothoracic crepitus. It is interesting to note that Codman was able to make his own scapula “sound about the room without the slightest pain”6 and was likely demonstrating froissement. In every instance, the air-filled thoracic cavity acts as a resonance chamber, much like a string instrument,20 and amplifies these noises. Pathologic conditions affecting muscle in the scapulothoracic articulation include atrophied muscle,9 fibrotic muscle,9,13,21 and anomalous muscle insertions.22 The most common bony pathology in the scapulothoracic space that may give rise to scapulothoracic crepitus is the osteochondroma, arising either from the ribs23 or the scapula (Fig. 17E-3).21,24,25 Malunited fractures of the ribs or scapula are also capable of creating painful crepitus.9,26,27 Abnormalities of the superomedial angle of the scapula, including a hooked superomedial angle24,28 or Luschka’s tubercle (which originally was described as an osteochondroma, but has subsequently come to mean any prominence of bone at the superomedial angle),9,27,29 have also been implicated as sources for scapulothoracic crepitus. Others11,20,30 implicate reactive spurs of bone that are created by the microtrauma of chronic, repeated small periscapular muscle avulsions. Certainly, any bony pathology that causes scapulothoracic crepitus is capable of forming a reactive bursa around the area of pathology.31,32 In fact, at the time of resection of bony pathology, a bursa is frequently seen. Bursae can become inflamed and painful in the absence of bony

888

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

who participate in sports that require repetitive overhead activity are commonly affected.12 There may be a familial tendency toward developing symptoms.7 Patients may relate a history of mild trauma that precipitates ­symptoms,41 and scapulothoracic crepitus may be bilateral in some patients.5 A space-occupying lesion, such as an osteochondroma, should be suspected if fullness or winging is identified on inspection of the scapula. Palpation or auscultation while the shoulder goes through a range of motion may help to identify the source and location of the periscapular crepitus.5,42 Supplemental radiographs, which include tangential views of the lateral scapula, computed tomography, or magnetic resonance imaging may be ­helpful in identifying anatomic pathology. Figure 17E-3   Osteochondroma of the scapula causing scapulothoracic crepitus. Note the increased signal in the bursa surrounding this osteochondroma of the scapula.

pathology and may, by themselves, become a source of crepitus.33 Other soft tissue pathologies that have been implicated in scapulothoracic crepitus include tuberculosis lesions in the scapulothoracic region and syphilitic lues,9 which are exceedingly rare in athletes. A common source of scapulothoracic crepitus in athletes involves abnormalities in congruence of the scapulothoracic articulation. Both scoliosis34,35 and thoracic kyphosis7 have been implicated as sources of scapulothoracic crepitus. In many athletes, thoracic kyphosis is common36,37 and may be the most likely source of scapulothoracic crepitus. Scapulothoracic dyskinesis38,39 may position the scapula in such a way as to promote contact between the superomedial angle and the ribs. This disorder is described later. Winging of the scapula has also been associated with scapulothoracic crepitus. Winging may be described as primary, secondary, or voluntary.40 Primary scapular winging results from identifiable anatomic disorders that directly affect the scapulothoracic articulation. Secondary scapular winging usually accompanies some form of glenohumeral joint pathology. This type of winging resolves once the glenohumeral pathology has been addressed. Voluntary winging, although rare, often has psychological overtones. Any form of winging may be associated with scapulothoracic crepitus. The more common causes of scapular winging in athletes are neurologic and include damage to the fifth cervical nerve root causing palsy of the rhomboid muscles; damage to the spinal accessory nerve causing trapezius palsy; and, as has been described in a number of athletic events, damage to the long thoracic nerve, causing serratus anterior palsy. Long thoracic nerve injury causing scapular winging is typically a neurapraxic or stretch injury to the nerve that occurs during play and typically resolves spontaneously within 1 year.40

Evaluation The patient with symptomatic scapulothoracic crepitus may be able to identify the location of the crepitus, pointing to the superomedial angle or the inferior angle. Athletes

Treatment As exemplified by Codman,6 it is important to recognize that scapulothoracic crepitus is not necessarily a pathologic condition. Scapular crepitus has been found in 35% of normal asymptomatic people.43 As a result of this, patients with hidden agendas or psychiatric conditions may not respond to treatment as well as other patients. However, if the athlete presents with pain, winging, or other disorders of the scapulothoracic articulation, the scapulothoracic crepitus is considered pathologic. Most athletes do not require surgical treatment of scapu­ lothoracic crepitus, particularly if the crepitus is related to soft tissue abnormalities, altered posture, or scapulothoracic dyskinesis.5,44 Treatment in these athletes should include postural exercises designed to prevent sloping of the shoulders.5,44,45 A figure-of-eight harness may be a useful tool to remind patients to maintain upright ­posture. Exercises to strengthen periscapular muscles are also thought to be important.5,12,44,45 Systemic nonsteroidal anti-­inflammatory drugs, as well as local modalities such as heat, massage, phonophoresis, and ultrasound, and the application of ethyl chloride to trigger points may also prove useful.5,12,44 Injections of local anesthetics and corticosteroids into the painful area have also been recommended.7,12,41,44-46 Caution must be used because there is a risk for creating a pneumothorax.41 Using these means, most athletes are expected to improve significantly12,46; however, for those who fail, a number of operations have been described. In addition, athletes with clearly defined bony pathology such as an osteochondroma are unlikely to improve with conservative treatment.47 Resection of the bony pathology is usually necessary to alleviate symptoms with a high likelihood of success.12,21,26 Historically, some authors have used muscle plasty operations to treat scapulothoracic crepitus, which include those described by Mauclaire, who reflected a flap of the rhomboids or trapezius and sutured it to the undersurface of the scapula.19 This is thought to be inadequate, however, because the muscle flap may atrophy with time, and symptoms could return.47 Rockwood has excised a rhomboid muscle avulsion flap with the elimination of snapping and pain.41 The most popular method for the surgical treatment of scapulothoracic crepitus involves a partial scapulectomy, which has been performed on the medial border of the scapula48 and, more commonly, on the superomedial angle.7,25,27,28,42,47,49

Shoulder

889

Suprascapular artery and nerve

Trapezius m. Spine of scapula

Area of scapula to be resected Rhomboideus minor m. Supraspinatus m. Infraspinatus m.

A

C

The surgical technique for the resection of the superomedial angle of the scapula begins with the patient in the prone position (Fig. 17E-4). An incision following Langer’s lines is made just lateral to the medial border of the scapula, from the superior angle to the scapular spine. The soft tissue is dissected down to the spine of the scapula. The periosteum over the spine is incised, and a plane is developed between the superficial trapezius and the underlying scapula. Next, a plane is developed between the supraspinatus and the rhomboids, and the levator scapulae muscles along the medial border of the scapula starting at the spine of the scapula. The supraspinatus is elevated in a subperiosteal plane from the supraspinatus fossa. The medial scapulothoracic muscles are dissected from the medial border of the scapula, and the dissection in this subperiosteal plane is carried around the medial border and to the subscapularis fossa, elevating the serratus and subscapularis with the rhomboids and levator. The superomedial angle of the scapula is resected with an oscillating saw. Caution is warranted, because the resection is carried laterally, to avoid injury to the dorsal scapular artery and the suprascapular nerve in the suprascapular notch. After resecting the bone, the reflected muscles fall back into place, and the medial border of the supraspinatus is repaired to the rhomboidlevator flap. Inferiorly, the periosteum is repaired back to the spine of the scapula using suture passed through drill holes. Postoperatively, the patient is placed in a sling and begins passive motion immediately. Active motion is

B

Figure 17E-4   Surgical approach for excision of the superomedial angle of the scapula. A, The trapezius is elevated from the spine of the scapula. B, The supraspinatus, rhomboids, and serratus are elevated in a subperiosteal plane from the medial border, and the superomedial scapula is resected while protecting the suprascapular nerve and artery. C, The supraspinatus is sutured back to the spine of the scapula. (Redrawn from Kuhn JE, Hawkins RJ: Evaluation and treatment of scapular disorders. In Warner JJP, Iannotti JP, Gerber C [eds]: Complex and Revision Problems in Shoulder Surgery. Philadelphia, Lippincott-Raven, 1997, pp 357-375.)

begun after 6 weeks, and resistance exercises follow after 8 to 12 weeks. Complications associated with partial scapulectomy include pneumothorax and postoperative hematoma; in younger patients, bone may try to re-form, but this rarely produces symptoms. The reported results for this procedure are generally good.7,25,28,47,49,50 However, it must be remembered that athletes typically do not require surgical intervention; as such, there are few data in the literature regarding the effect of superomedial angle resection of the scapula on athletic performance. It is also important to note that the bone resected is not pathologic and appears normal histologically, which has prompted some to perform bursectomies and avoid a partial scapulectomy.32,47

SCAPULOTHORACIC BURSITIS Symptomatic scapulothoracic crepitus is typically accom­ panied by an inflamed scapulothoracic bursa. It is important to realize that although these two conditions are frequently found together, an athlete may have crepitus without pain, and another may have scapulothoracic bursitis without crepitus. As described earlier, symptomatic ­ scapulothoracic bursitis appears to affect two areas of the scapula, the superomedial angle and the inferior angle. These bursae, when inflamed, are thought to be ­adventitious.3,7,12

890

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Evaluation Scapulothoracic bursitis may accompany painful scapular crepitus or may exist as a separate entity. Patients generally complain of pain with activity and may have audible and palpable crepitus of the scapulothoracic articulation. Usually the scapular crepitus associated with bursitis is of a much lesser quality and nature than that described with bony pathology. Periscapular fullness is frequently appreciated in thinner athletes. This may become significant enough to produce noticeable scapular winging. Scapular winging has been identified in 50% of patients with a snapping scapula and no bony abnormalities.12 Patients may describe minor trauma as a predisposing event.19,42 More commonly, repetitive overhead activities in work or athletics have been implicated.17,44,46 The repetitive motion may irritate soft tissues until a chronic bursitis and inflammation develops. The bursa then undergoes scarring and fibrosis, with crepitus and pain to follow. Like scapulothoracic crepitus, this scapulothoracic bursitis in athletes is related to postural abnormalities and scapulothoracic dyskinesis. O’Donoghue recommended injecting local anesthetic into the bursa as a diagnostic aid.12

Treatment The initial treatment of scapulothoracic bursitis regardless of its location is conservative, beginning with rest, analgesics, and nonsteroidal anti-inflammatory drugs. Physical therapy to improve posture, heat, and local steroid injections have also been recommended.10,17,44 Efforts to strengthen periscapular muscles and stretching are frequently added.10,18,44 For patients who continue to have symptoms despite conservative treatment, surgery may be beneficial. Sisto and Jobe10 described an open procedure for resecting a bursa at the inferior angle of the scapula in four major league baseball pitchers. All pitchers had pain during the early and late cocking phases, as well as during ­acceleration, and could no longer pitch (Fig. 17E-5). Only one of the four patients presented with scapulothoracic crepitus, but all had a palpable bursal sac ranging in size from 1 to 2 cm, best seen with the arm abducted to 60 degrees and elevated forward 30 degrees. All four pitchers failed conservative therapy and underwent a bursal excision through an oblique incision just distal to the inferior angle of the scapula. The trapezius muscle and then the latissimus dorsi muscle were split in line with their fibers, exposing the bursa. The bursa was sharply excised, and any bony prominence on the inferior pole of the scapula or ribs was removed. The wounds were closed routinely over a drain, and a compression dressing was applied. Physical therapy stressing motion was begun after 1 week and progressed to allow gentle throwing after 6 weeks. This progressed as symptoms permitted to full-speed throwing. After this procedure, all four pitchers were able to return to their former level of pitching. Similarly, McCluskey and Bigliani47 performed an open excision of a symptomatic superomedial scapulothoracic bursa in nine patients and noted a thickened, abnormal bursa between the serratus anterior and the chest wall at the time of surgery. Their surgical technique involved

Infraserratus bursa

Figure 17E-5   Bursa at the inferior angle of the scapula in throwers. This is an infraserratus bursa and has been described in baseball pitchers, in whom an excision of the bursa has allowed a return to throwing. (Redrawn from Kuhn JE, Hawkins RJ: Evaluation and treatment of scapular disorders. In Warner JJP, Iannotti JP, Gerber C [eds]: Complex and Revision Problems in Shoulder Surgery. Philadelphia, Lippincott-Raven, 1997, pp 357-375.)

making a vertical incision medial to the vertebral border of the scapula. The trapezius is dissected free, and a subperiosteal dissection is used to free the levator scapulae and rhomboids from the medial border of the scapula. A plane is developed between the serratus anterior and the chest wall. The thickened bursa is resected and any bony projections removed. The medial periscapular muscles and trapezius are reapproximated to the scapula. The skin is closed in a routine fashion. The patient uses a sling for comfort and begins passive motion and pendulum exercises immediately. After 3 weeks, active motion is allowed, with strengthening begun at 12 weeks. With this technique, 88% of patients with symptomatic scapulothoracic bursitis had good or excellent results. One patient with a fair result also required muscle transfers for trapezius winging.47 Similar results for bursectomy have been reported by others.51 Resection of the symptomatic scapulothoracic bursa has been performed endoscopically as well.3,5,46,52-56 Ciullo and Jones5 have one of the largest and earliest ­endoscopic series with 13 patients who underwent subscapular endoscopy after failing a conservative treatment program for symptomatic scapulothoracic bursitis. Débridement was performed for fibrous adhesions found in the bursa between the subscapularis and serratus muscles as well as the bursa between the serratus and chest wall. In addition, débridement or scapuloplasty of changes at the superomedial angle or inferior angle were performed. All 13 patients returned to their preinjury activity level, except for ­physician-imposed restrictions in a few patients, limiting the ­assembly line use of vibrating tools.5 Matthews and colleagues46,57 have described the technique for scapulothoracic endoscopy. Patients can be placed in the prone or lateral position; however, the lateral position is preferred because it allows for arthroscopic evaluation of the glenohumeral joint and the subacromial space. In addition, if the arm is extended and maximally internally rotated, the scapula will fall away from the thorax improving access to the bursae.

Shoulder

Figure 17E-6   Portals used for scapulothoracic bursoscopy. Portals are placed 3 cm medial to the medial scapular border with the most superior portal being placed below the level of the scapular spine.

Three portals are used that are placed at least 2 cm from the medial border of the scapula in the region between the scapular spine and the inferior angle (Fig. 17E-6). For the middle portal, a spinal needle is inserted into the bursa between the serratus anterior and the chest wall. This needle should be inserted midway between the scapular spine and the inferior angle, at least 3 fingerbreadths medial to the medial border of the scapula to avoid injury to the dorsal scapular artery and nerve. The bursa under the serratus anterior can be distended with fluid before a stab wound is made in the skin and the blunt obturator and endoscope are inserted. Deep penetration may traverse the serratus entering the axillary space and should be avoided. After this initial middle portal has been established, a superior portal placed 3 fingerbreadths medial to the ­vertebral border of the scapula just below the spine penetrates the interval between the rhomboideus major and rhomboideus minor. This portal allows access to the superomedial angle of the scapula. Portals placed superior to the scapular spine jeopardize the dorsal scapular nerve and artery, the spinal accessory nerve, and the transverse cervical artery and should be avoided. A third, inferior, ­portal can be made in a similar fashion at the inferior angle of the scapula. In the bursa between the serratus anterior and chest wall, landmarks are generally absent except the ribs. A motorized shaver and electrocautery are required to perform the bursectomy and obtain hemostasis (Fig. 17E-7). The

A

891

arthroscopic pump should be kept at low pressure throughout the procedure. After completing the ­bursectomy, the portals are closed in a standard fashion, and the patient is placed in a sling for comfort. Physical therapy, beginning with active range of motion, is initiated as tolerated by the patient. As presented at numerous meetings, the arthroscopic techniques for performing a scapulothoracic bursectomy appear to have early promising results, and to date no cases of injury to the long thoracic nerve, dorsal scapular artery, suprascapular nerve, axillary contents, or thoracic cavity have been reported. Despite this, few series of patients treated arthroscopically for scapulothoracic bursitis have been published in the peer-reviewed literature, and this technique remains investigational at this time.

SCAPULOTHORACIC DYSKINESIS As one might expect, abnormalities in scapulothoracic motion that change the position of the scapula relative to the chest wall are related to the development of scapulothoracic crepitus and bursitis. Burkhart and colleagues have described a condition known as the SICK scapula.58 The acronym SICK stands for scapula malposition, inferior medial border prominence, coracoid pain and malposition, and dyskinesis of scapular movement. The scapula assumes an abnormal position at rest characterized by a position that is inferior, protracted, and tilted anteriorly. Tenderness is typically found on the medial edge of the coracoid, and the pectoralis minor is thought to be in spasm. The authors recognized this pattern in throwing athletes with shoulder pain. This is a static finding at rest but is similar to scapular dyskinesis described by Kibler and McMullen, in which the scapula moves abnormally during arm ­elevation.59 Myers and colleagues studied scapulothoracic motion in a population of throwing athletes and compared this to a control population.60 They found that throwing athletes demonstrated significantly increased upward rotation, internal rotation, and retraction of the scapula during humeral elevation, implying that throwing athletes may develop these adaptations for more efficient performance of the throwing motion. Interestingly, all these positions may be related to the pectoralis minor muscle. Su and colleagues have demonstrated that scapular kinematics may be altered in symptomatic swimmers, an effect that is magnified with fatigue associated with a practice.61 Many

B

Figure 17E-7   Arthroscopic photographs of the scapulothoracic bursa. Landmarks are conspicuously absent. A, View through the superior portal. The shaver and cautery device are used to débride the bursa. The subscapularis fascia is seen superiorly. B, A similar view looking at the inferior aspect of the scapulothoracic space.

892

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

authors62-66 have evaluated muscle firing patterns in a number of athletes with shoulder pain and have noted that dysfunction of the serratus anterior is an early finding in many of these athletes. It is conceivable that weakness of the serratus anterior may predispose the overhead athlete to compensatory pectoralis muscle firing, which alters the dynamic position of the scapula during sport activities. This may lead to other problems as the normal function of the glenohumeral joint is altered. The altered position of the scapula would lead to increased strain on the posterior capsule structures during the deceleration of throwing, which could lead to tightness in the capsule and muscle leading to the glenohumeral internal rotation deficit. Static scapular position abnormalities then develop. With the altered position of the scapula, the acromion is in a position to produce impingement of the rotator cuff. In addition, the tight posterior capsule would lead to superior humeral head migration during elevation and external rotation of the arm, further jeopardizing the rotator cuff. Our understanding of this constellation of findings in the shoulder of overhead athletes is in its infancy, yet an interplay between a weak serratus and overfunctioning pectoralis minor appears to be important.

Evaluation Patients with scapulothoracic dyskinesis may not direct the physician toward the scapula and complain of pain in the glenohumeral joint. Standing behind the athlete and inspecting the scapulae from the back demonstrates asymmetry at rest, with the affected shoulder frequently depressed and the scapula protracted and tilted forward. Mild scapular winging may be present, with the inferior angle and the medial border of the scapula prominent. Patients frequently have pain to palpation just medial to the coracoid at the insertion of the pectoralis minor. Asking the patient to elevate the arm in the frontal plane and in the scapula plane reveals asymmetry in scapulothoracic motion, called scapulothoracic dyskinesis. In the presence of rotator cuff pathology, this may be related to decreased firing of the middle and lower trapezius.67 It appears to be also related to a shortened pectoralis minor tendon.68,69 Patients with this scapular dyskinesis often present with rotator cuff pain. These patients have positive impingement signs and pain when testing supraspinatus strength in a position of scaption. This pain will be reduced and measured strength will improve if the scapula is reduced to the chest wall, a test known as the scapular retraction test.70,71 It has been noted that core stability is typically poor in these patients.59 Checking the athlete for a Trendelenburg sign and observing the single-leg knee bend can be helpful to look for weakness and problems with balance. In baseball pitchers, asymmetry is often seen such that the stride leg has more weakness than the lead leg.

Treatment Treatment of scapulothoracic dyskinesis is through exercise and modalities of physical therapy. Kinetic chain– based rehabilitation programs have been recommended39,60 because many of the patients with scapulothoracic kinematic abnormalities have weakness in the core stabilizers of

the trunk. Stretching tight structures is important. A major contributor to scapular dyskinesis is a tight ­pectoralis minor tendon. Borstad and Ludewig demonstrated the door jamb stretch to be the best method to stretch a tight pectoralis minor.72 The posterior capsule, which is commonly found to be tight, can effectively be stretched.73,74 Although a variety of methods exist to stretch the posterior capsule, stretching across the body appears particularly effective.75 Strengthening of key muscle groups—the serratus anterior, low trapezius, rhomboids, and rotator cuff—is usually included in the rehabilitation.59 Many of these methods of rehabilitation have not been evaluated prospectively or with randomized trials, and as such, the work is currently in its infancy. Clearly much more work is needed to define pathologic scapulothoracic kinematics and their effect on other shoulder pathologies.

SUMMARY A variety of scapulothoracic conditions can affect the athlete’s shoulder. These include winging in a variety of forms, crepitus and bursitis, and dyskinesis of the scapulothoracic articulation. Scapular winging in athletes most commonly results from a long thoracic nerve neuropraxic injury, and these athletes recover spontaneously. Scapulothoracic crepitus and scapulothoracic bursitis are two related conditions but may be found independently in athletes with periscapular pain. In general, treatment for athletes is nonoperative and requires postural exercises designed to prevent sloping of the shoulders5,45 and periscapular muscle strengthening.5,12,45,46,59 A figure-of-eight harness may be a useful tool to remind patients to maintain upright posture. Local modalities, nonsteroidal anti-inflammatory drugs, and local injections have also been recommended. In athletes with refractory symptoms, surgical correction may be considered; however, there are only a few reports in the literature for this select population, so it is difficult to predict outcomes with regard to returning to sport. Scapular dyskinesis is only now under study as a source of shoulder pathology, and early results suggest the effects of scapular dyskinesis may be of critical importance. General considerations for treating scapular dyskinesis include exercises to improve core stability, stretch the pectoralis minor, stretch the posterior capsule, and strengthen the serratus and low trapezius.59 More work is needed to gain a complete understanding of scapulothoracic problems in the athlete.

C l Scapular

r i t i c a l

P

o i n t s

bursitis may exist with or without scapulothoracic crepitus, and scapulothoracic crepitus may exist with or without bursitis. l Scapular winging is most commonly neurologic and in athletes is the result of a neurapraxic injury to the long thoracic nerve. These injuries usually recover spontaneously but may take a year or longer. l Scapulothoracic bursitis is often seen in athletes. Initial treatment is nonoperative. When nonoperative measures fail, open or arthroscopic bursectomy may help.

Shoulder l Scapulothoracic

dyskinesis is common in athletes with shoulder pain and is characterized by an altered position of the scapula at rest (sick scapula) or during motion (scapula dyskinesis). l The esssence of treatment for scapulothoracic dyskinesis involves improving core stability, stretching pectoralis minor and posterior glenohumeral joint capsules, and strengthening the serratus and low trapezius.

893

Kuhn JE, Plancher KP, Hawkins RJ: Scapular winging. J Am Acad Orthop Surg 6:319-325, 1995. Kuhn JE, Plancher KP, Hawkins RJ: Symptomatic scapulothoracic crepitus and scapulothoracic bursitis. J Am Acad Orthop Surg 6(5):267-273, 1998. Manske RC, Reiman MP, Stovak ML: Nonoperative and operative management of snapping scapula. Am J Sports Med 32(6):1554-1565, 2004. Williams GR Jr, Shakil M, Klimkiewicz J, Ianotti JP: The anatomy of the scapulothoracic articulation. Clin Orthop 357:237-246, 1999.

R eferences Please see www.expertconsult.com

S U G G E S T E D

R E A D I N G S

Burkhart SS, Morgan CD, Kibler WB: The disabled throwing shoulder: Spectrum of pathology. Part III. The SICK scapula, scapular dyskinesis, the kinetic chain, and rehabilitation. Arthroscopy 19(6):641-661, 2003. Kibler WB: Role of the scapula in the overhead throwing motion. Contemp Orthop 22:525-532, 1991. Kibler WB, McMullen J: Scapular dyskinesis and its relation to shoulder pain. J Am Acad Orthop Surg 11(2):142-151, 2003.

S E C T I O N

F

Sternum and Rib Fractures in Adults and Children Matthew T. Provencher, Augustus D. Mazzocca, and Anthony A. Romeo

Rib and sternum injuries are relatively rare in athletes and more commonly result from motor vehicle crashes.1,2 In children, this type of fracture can be a marker of either severe trauma or abuse.3-6 This section defines the classification of rib and sternal fractures, the mechanism of injury, the physical assessment, and the treatment for both pediatric and adult athletes.

RIB FRACTURES Rib fractures can be divided into two main categories: stress fractures (indirect) and traumatic fractures (direct). Traumatic fractures are caused by either a blow from a blunt object, which fractures the ribs in direct contact, or compression of the entire thorax, which results in fractures of multiple ribs.7 Stress fractures can occur in ribs that are subjected to repetitive mechanical loading during a particular activity. The complete history generally provides insight into which of these categories applies (Table 17F-1).

Traumatic Rib Fractures The mechanism of traumatic rib fracture or injury is a direct blow from a blunt object. In athletics, this blow generally comes from an anterior direction, causing a more lateral

rib fracture or an injury to the costochondral junction.8,9 Athletes will recall the injury and will sometimes report that the “wind” was knocked out of them. It is important to identify which and how many ribs are injured because fractures of the first four or last two ribs, multiple fractures, and flail segments may result in injury to surrounding structures.8 The definition of flail chest has historically been involvement of three or more ribs that are fractured in two or more places. Clinically, this term has changed to include any chest segment that exhibits paradoxical motion during respiration. Injuries to the kidney, the spleen, or the liver may not be readily clinically apparent. Splenic trauma has been reported in up to 20% of left lower rib fractures, and liver trauma has been reported in 10% of right lower rib fractures.9 Pneumothorax and hemothorax must be ruled out by auscultating the chest and palpating for subcutaneous air. Progressive shortness of breath is also an ominous indicator in this situation. Rib fractures that presented to a level I trauma center9 (7147 total fractures over 5 years) demonstrated a high incidence of associated injuries (94%), hemothorax or pneumothorax (32%), or lung contusion (26%), with as many as one third developing pulmonary complications. Although not as prevalent in pure athletic trauma, one should be vigilant to evaluate for concomitant injuries, especially in the case of multiple rib fractures.

894

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  Table 17F-1  Typical Findings in Rib and Sternum Fractures

Acute Rib Fractures

Rib Stress Fractures

Sternum Fractures

Recall injury— More chronic onset “wind” knocked out

Direct motor vehicle trauma (automobile racing) Assess for shortness Suspect in most upper Direct injury: of breath, cardiac segment (1st rib) in relatively flexible symptoms weightlifting and inferior sternum is throwing flexed posteriorly and fractured, or sternomanubrial dislocation can occur Exclude Middle rib cage (ribs Indirect injury: pneumothorax 5-9) in rowers flexion-compression (auscultate) and of cervical spine— hemothorax posteriorly displaces (palpation and manubrium and inspection) anterior displacement of the sternum Pain during deep Evaluate personal Rib and spine inspiration dietary and fractures may be hormonal history associated Assess costochondral Lateral radiograph junction for step-off of sternum (sternal view) Chest radiograph— Ultrasound may aid assess for in diagnosis and pneumothorax identify retrosternal fluid Rib series—oblique Evaluate for views of involved mediastinal segment widening, which usually indicates more severe visceral or substernal pathology Ultrasound may Pediatric population: improve accuracy, sequential fusion of especially at sternum, which is costochondral not complete until junction about age 21 yr

In the case of infant, toddler, or child traumatic rib fracture, one should maintain a high level of suspicion for nonaccidental trauma (NAT).3-5 These rib fractures may be difficult to identify in NAT with a routine skeletal survey. The combination of conventional and high-detail skeletal radiography and possibly ultrasound may improve the diagnosis. The risk for complications and mortality increased with the number of ribs fractures and was usually a harbinger of additional pathology.3 Regardless, it has been argued that in children younger than 3 years, the positive predictive value of a rib fracture to identify NAT is greater than 90%.3

Examination Physical examination usually elicits pain during palpation as well as during deep inspiration. A hematoma is an indicator of a displaced rib fracture produced by injury to the intercostal vessel. In palpating the costochondral junction, one may also palpate a step-off, which aids in the diagnosis of costochondral fracture-dislocation. Two radiographic examinations (chest series and rib series) are important in diagnosing rib fractures. A chest

  Table 17F-2  Treatment Options in Rib and Sternum Fractures

Acute Rib Fractures

Rib Stress Fractures

Sternum Fractures

Symptomatic treatment Rest, ice, oral analgesics Intercostal rib block (injection) if especially painful on inspiration Rib binder may improve comfort

Symptomatic treatment Rest, ice, oral analgesics Investigate causes— sport mechanics (rowing, lifting, throwing) Physical therapy— critical to strengthen serratus anterior, and sports mechanics evaluation

Symptomatic treatment Rest, ice, oral analgesics Avoid contact sports until completely pain free

Bone mineral density may be indicated— dietary and hormonal evaluation

Sternomanubrial dislocation may warrant reduction and is usually stable once reduced. Treatment is similar to above.

series (posterior-anterior and lateral) is needed to rule out pneumothorax. A rib series includes oblique views of the clinically involved segment. Injuries to the costochondral junction are not well seen on radiographs. Magnetic resonance imaging (MRI) or computed tomography (CT) may be needed to confirm the suspected diagnosis of a costochondral injury. Ultrasound has also been advocated to diagnose acute rib fractures, with the potential for improved diagnostic accuracy over conventional radiography.10

Treatment Treatment of traumatic rib injuries is primarily symptomatic (Table 17F-2). Most of these fractures are stable. Stable rib fractures are nondisplaced or minimally displaced and do not involve more than two consecutive segments. The first mode of treatment is oral analgesic medication to relieve pain, which mainly occurs during inspiration. If this does not relieve the pain, an intercostal “rib block” may offer some temporary relief of spasm. A long-acting medication such as bupivacaine is used. It is infiltrated along the intercostal nerve of the fractured rib and the ribs above and below it. Some success has also been reported with the use of “rib binders,” which may splint the fracture, keeping it more stable so that activities of daily living are more ­comfortable.11

Return to Athletics The criteria for return to play with rib fractures are stability of the injury with no soft tissue complications (e.g., pneumothorax, hemothorax, or liver, spleen, or kidney contusion) and an improvement in pain (Table 17F-3). This is usually seen about 2 to 3 weeks after injury. The athlete can return to play at this time using a rib belt; he or she is further protected for about 6 to 8 weeks by a flak jacket (Fig. 17F-1) so that the rib can heal completely. It has been reported that nonunion is not a complication of

Shoulder

  Table 17F-3  Return to Play in Rib and Sternum Fractures

895

these injuries if there is persistent pain lasting longer than 6 to 8 weeks.

Acute Rib Fractures

Rib Stress Fractures

Sternum Fractures

Stress Fractures

Stability of the rib segments (usually 2-3 wk) Any soft tissue or visceral injuries healed (usually 2-3 wk)

Pain free for 2-4 wk

Pain free (usually 6-8 wk), pain >12 wk is rare If sternomanubrial reduction, may take 6-12 wk for return to play

Stress fractures or indirect injury to the ribs usually involves muscle contractions that result in subthreshold bending, ­causing microfracture. This microfracture is additive and eventually causes pain. A stress fracture is caused by the inability of the bone to withstand subfracture threshold force in a repetitive fashion. There is an imbalance between bone formation and bone resorption. It is also important to realize that a stress fracture is a normal response to an abnormal stress, whereas an insufficiency fracture is an abnormal response to a normal stress. Stress fractures have been reported to occur in the first rib in baseball players12-15 (Fig. 17F-2) and also as a more acute phenomenon in ­basketball players. Fractures of the lower ribs have been reported in golfers and rowers.16-20 Karlson19 reviewed rib stress fractures in rowers and found that those occurring in the anterolateral to posterolateral aspects of ribs five through nine are most often associated with long-distance training and heavy load per stroke. It was also noted that the similarity between these stress fractures and fractures caused by chronic coughing suggests a similar mechanism of injury. The actions of both the serratus anterior and the external oblique abdominal musculature on the rib may cause these stress fractures because of the repeated bending forces of both rowing and coughing. Mintz21 used an MRI analysis technique, and ­McKenzie22 analyzed rowing mechanics to implicate the serratus anterior in the causation of stress fractures. Avulsion fractures of the floating ribs may also result from the opposing pulls of the latissimus dorsi, the internal obliques, and the serratus posterior inferior muscles (Fig. 17F-3).15 It has also been suggested17,23 that exercise-induced rib stress

May return if no visceral pathology and ribs stable at 2-3 wk. Wear rib belt and flak jacket for up to 8 wk May return earlier (1-2 wk) if few (<2) ribs injured, stable segments, and no visceral issues Costochondral injuries take longer to heal (4-8 wk), and return to play is restricted longer.

Satisfactory completion of serratus anterior strengthening program and sportspecific training If return to sport aggravates ribs, stop and reassess.

this type of traumatic fracture, and malunion is of little importance.11 It is also important for the team physician to note that training can be started as early as 1 week after injury if the pain has significantly decreased. Costochondral injuries have a history of healing much more slowly. CT and MRI may be helpful in evaluating

A

B

Figure 17F-1   A and B, Flak jacket used to protect the athlete with rib fractures.

896

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Scalenes Break in groove for subclavian a.

Intercostals Serratus anterior Rectus abdominus

Figure 17F-2   Activities such as pitching may result in   a stress fracture of the first rib. (From Tullos HS, Erwin WD, Woods GW, et al: Unusual lesions of the pitching arm. Clin Orthop 88:169-182, 1972.)

fractures are associated with a decrease in bone mineral density and that treatment should focus on dietary and hormonal evaluation. Specific radiographs can be directed to the posterolateral portion of the rib cage because most stress fractures associated with rowers occur in this area. Athletes who engage in sports such as soccer24 and weightlifting25 or in sports that require throwing may be predisposed to rib stress fractures.26,27 Rib stress fractures have been reported to occur in golf as well. The leading side of the trunk in the posterolateral part of the rib is the most common site. Poor technique, frequent strokes, and large divots place the beginner or the poorly skilled golfer at greatest risk. These symptoms can also be confused with back strain, which is also quite common in this athlete. Electromyographic examination has shown that the serratus anterior muscle is the main force on the rib during these motions.

Examination The diagnosis of rib stress fractures is confirmed if deep inspiration causes significant pain. This motion causes the accessory muscles of inspiration to contract, pulling the periosteum of the rib under tension and causing an increase in pain. This test can differentiate a stress fracture from back strain, which is common in this population. Bone scanning and oblique radiographs of the ribs can also aid in diagnosis.

Figure 17F-3   Avulsion fractures of the floating ribs may result from the opposing pulls of the latissimus dorsi, internal obliques, and serratus posterior inferior. (From Tullos HS, Erwin WD, Woods GW, et al: Unusual lesions of the pitching arm. Clin Orthop 88:169-182, 1972.)

Treatment Rib stress fractures are treated in a symptomatic manner with analgesics and rest (see Table 17F-2). The key difference in this type of injury is that the mechanics must be addressed. Strengthening and stretching of the serratus anterior muscle with physical therapy is critical, as is kinesiologic evaluation of the basic mechanics of motion in the particular sport in question, followed by correction of those mechanics. There are few long-term sequelae associated with this injury.

STERNUM FRACTURE The sternum is a flat bone that forms a portion of the anterior wall of the thorax (Fig. 17F-4). It is comprised of three parts: the manubrium, the body, and the xiphoid process. The upper portion of the manubrium forms the jugular notch. It also articulates with the clavicle, forming the sternoclavicular joint. The first and second costal cartilages articulate with the upper and lower borders of the manubrium. The manubrium then articulates with the body of the sternum. This articulation forms the sternal angle. This level is generally associated with the fourth and fifth thoracic vertebrae and can be palpated. The seventh costal cartilage generally fits into the notch by the body in

Shoulder

Figure 17F-4   Structure of the sternum.

897

the xiphoid process. The xiphisternal joint is at the apex of the infrasternal angle. It lies at about the level of the 10th or 11th thoracic vertebra.9,11 Injuries to the sternum in athletes are rare (see Table 17F-1). Most sternal injuries occur in automobile racing.28 Fowler28 reported three mechanisms: direct, indirect, and muscular (Fig. 17F-5). Direct injuries occur when either a helmet or a steering wheel strikes the sternum directly. The relatively flexible lower portion of the sternum is ­displaced inward in a fracture, or a sternomanubrial dislocation can occur (Fig. 17F-6). This direct mechanism can also be associated with posterolateral rib fractures as well as cardiac and lung contusions. The indirect injury results from a flexion-compression injury of the cervical thoracic spine (Fig. 17F-7). This type of injury leads to fracture­dislocation of the sternomanubrial joint with posterior displacement of the manubrium and anterior displacement of the body of the sternum. The chin’s striking the chest may aid this mechanism. Increased intrathoracic pressure along with the mobility of the lower ribs may actually drive the body of the sternum forward. A wedge fracture of the thoracic vertebrae can also be associated with this condition.11 The last mechanism proposed by DeTarnowski,29 Fox,30 and Fowler28 is muscular action. This supposedly tears the sternum into proximal and distal fragments through the action of the opposing muscle groups. The most common cause of sternal fractures remains motor vehicle trauma.31-33 Most cases can be managed with

Figure 17F-5   Mechanistic classification of sternal fractures. (From Fowler AW: Flexion-compression injury of the sternum. J Bone Joint Surg Br 39:487-497, 1957.)

898

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Flexion

Impact

B

A

C

creatine kinase assays in the assessment of isolated sternal fractures is usually not indicated.37 Although this is rare in sports, Johnson and colleagues38 reported an indirect sternomanubrial fracture-dislocation in a football player as a result of a flexion-compression mechanism. Woo39 reported traumatic sternomanubrial joint subluxation in two basketball players from violent blows from the elbow to the sternum. Stress fractures of the sternum have been reported in a wrestler40 and a golfer.41

Examination

Figure 17F-6   Sternomanubrial dislocation caused by hyperflexion of the cervical spine combined with axial load. (From Johnson CD, MacKenzie JW, Zawadsky JP: Manubriosternal dislocation in a football athlete. Surg Rounds Orthop 2:45-50, 1989.)

observation, pain medications, and cardiovascular monitoring during a short inpatient stay.31-33 Interestingly, it has been shown34 that there is less overall concomitant injury if the sternum is fractured than without sternal fracture. It is possible that the sternal fracture in a seat-belted individual absorbs a substantial part of the energy, thus preventing greater visceral damage.32 There is an association of spinal fractures with sternal fractures34,35; however, cardiac or visceral chest or pulmonary damage is infrequent.31-34,36 Additionally, the routine use of electrocardiography and

A

Patients presenting with a sternal fracture have a history of a direct blow with pain and tenderness over the sternum. Sternal and rib fractures may be associated. These patients may also complain of shortness of breath that is not prolonged.42 Historically, sternal fractures have been thought to be associated with myocardial contusion and therefore a potential for arrhythmia. This has not been borne out by more recent studies.32-34,37,43,44 Isolated sternal fractures are not a significant risk factor for myocardial contusion.45 Peek and Firmin45 found no cases of arrhythmia in 162 consecutive isolated sternal fractures. Similarly, others have been unable to link sternal fractures to myocardial injury.46-48 It is believed that the energy is absorbed by the sternum and the fracture as opposed to being transmitted through the sternum to the cardiac muscle and then causing a contusion. Obviously, trauma protocol dictates that sternal injuries causing ecchymosis to the chest require 24-hour monitoring for arrhythmia.32-34 The diagnosis is confirmed if there is pain on palpation in the area of fractures as well as palpable crepitation at the level of the injury with respiration and any type of movement. Adequate radiography is the study of choice when

B

Figure 17F-7   A fall or blow to the upper part of the spine (A) or to the back of the head and neck (B) can result in sternal fracture. Posterior displacement of the manubrium can be exaggerated by a blow from the chin as the neck is hyperflexed (B). (From Fowler AW: Flexion-compression injury of the sternum. J Bone Joint Surg Br 39:487-497, 1957.)

Shoulder

899

Figure 17F-8   Lateral radiograph of the chest revealing a sternomanubrial dislocation.

a sternal fracture is suspected, particularly a lateral view of the sternum.42 There should be an increase in penetration over that normally used for chest radiography so that the sternum can be better penetrated and fractures can be appreciated (Fig. 17F-8). Huggett and Roszler49 reviewed data from a small series of sternal fractures, comparing computed tomographic scans with plain radiographs. CT identified only six of nine sternal fractures, whereas radiography identified eight of nine. A retrosternal hematoma, which is a specific finding for sternal fractures, was seen in only three of nine computed tomographic scans. Hendrich and associates50 used ultrasonography to correctly identify 16 of 16 sternal fractures, whereas radiography identified 15 of 16. Ultrasound has also been advocated by Fenkl and coworkers51 and Bitschnau and associates,52 who found both accurate interpretation of sternal fractures and also detection of substernal fluid (local hematoma and pleural effusion). One should also evaluate for mediastinal widening, which is fairly reliable for substernal visceral ­pathology.44 In children, it is important to recognize that the time of fusion commonly varies for the four segments of the body of the sternum.7 At 7 years of age, the third and fourth segments fuse. At 14 years of age, the second and third fuse, and at age 21, the first and second fuse. Fusion of these segments may be delayed at any stage, possibly leading to an incorrect diagnosis of fracture.43

Figure 17F-9   Method of reduction of fractured sternum. If necessary, an assistant may apply traction by grasping the patient’s arm at the axilla and pulling cephalad. Inset shows detail of the forces applied in reduction. (From DePalma AF: The Management of Fractures and Dislocations—An Atlas, 3rd ed. Philadelphia, WB Saunders, 1981.)

­ arrant a reduction attempt. Once pain has diminished, w the patient may return to athletics in a general conditioning and training program and is limited only by pain. Reduction of a dislocated sternomanubrial joint can be performed with the patient supine in Trendelenburg’s position with a sandbag beneath the scapula (Fig. 17F-9). Manual pressure on the manubrium and gentle traction on the arms may aid in reduction of the sternomanubrial ­dislocation.53

Return to Athletics Sternal contusion and most fractures are stable injuries that can be treated conservatively. Conservative measures consist of rest, ice, and various oral analgesics. Contact sports should be avoided until pain subsides (see Table 17F-3). It has been reported that pain persisting beyond 6 to 12 weeks is rare.11 Sternomanubrial dislocations are generally considered stable after reduction. Treatment is similar to that for sternal fracture.11

Treatment Most sternal fractures do not require more than symptomatic treatment (see Table 17F-2).53 Pain around the fracture site will last for an average of 11 weeks but persists longer in older patients.54 Rarely, displacement may

C l Rib

r i t i c a l

P

o i n t s

and sternum fractures are usually due to high-speed trauma. In children, there is the potential suspicion for child abuse.

900

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

l A

careful assessment of cardiovascular, pulmonary, and abdominal status is critical, especially with multiple rib fractures (pneumothorax, lung contusion, mediastinal injuries, and great vessel injuries). l Stress fractures of the rib may result from repetitive chest muscle sporting activities (rowing, kayaking, throwing, and baseball). l Intercostal rib blocks and anti-inflammatory medications may decrease pain associated with respiration due to rib fracture. l Rib binders may offer improved comfort in rib fractures, although there is no evidence for routine use. l Sternum injuries in athletes are rare, mostly due to automobile racing, and most require only symptomatic treatment after more serious debilitating injuries are addressed. l Sternum injuries need to be evaluated with adequate radiographs (lateral or computed tomographic scan) to assess for sternomanubrial instability or dislocation; this can be reduced with reduction maneuvers.

S U G G E S T E D

R E A D I N G S

Christiansen E: Rib stress fractures in elite rowers: A case series and proposed mechanism. Am J Sports Med 28(3):435–436, 2000. Foley N, Mattox K: Fractures of the sternum. Curr Concepts Trauma Care 8(3): 9–11, 1985. Garcia VF, Gotschall CS, Eichelberger MR, Bowman LM: Rib fractures in children: A marker of severe trauma. J Trauma 30(6):695–700, 1990. Gouldman JW, Miller RS: Sternal fracture: A benign entity? Am Surg 63(1):17–19, 1997. Griffith JF, Rainer TH, Ching AS, et al: Sonography compared with radiography in revealing acute rib fracture. AJR Am J Roentgenol 173(6):1603–1609, 1999. Hills MW, Delprado AM, Deane SA: Sternal fractures: Associated injuries and management. J Trauma 35(1):55–60, 1993. Vioreanu MH, Quinlan JF, Robertson I, O’Byrne JM: Vertebral fractures and concomitant fractures of the sternum. Int Orthop 29(6):339–342, 2005.

R eferences Please see www.expertconsult.com

S E C T I O N

G

Muscle Ruptures other than the Rotator Cuff Hussein Elkousy

The anatomy and function of the shoulder is complex. The shoulder girdle comprises four joints and several muscle and tendon units that contribute to its complex function. Several of these muscle-tendon units have been identified as common sites of pathology, such as the rotator cuff and the biceps tendon complexes. Pathology involving these groups is addressed in other chapters. This chapter addresses muscle ruptures associated with the shoulder joint other than the rotator cuff. In particular, two muscles are prone to injury: the pectoralis major and the latissimus dorsi. The pectoralis major has become a more recognized injury during the past 180 years. It was first described by Patissier in 1822.1 As of 2004, 180 cases had been reported in the literature. The largest series at that time was 33 cases collected over a 21-year period.2 Certainly many more cases are managed without being formally reported. The latissimus dorsi muscle has also been reported to tear traumatically. However, only a few case reports exist in the literature. Surgical intervention has been described with good results. In addition, a teres minor rupture has been reported. Both of these injuries will be addressed.

PECTORALIS MAJOR Relevant Anatomy and Biomechanics The pectoralis major is a broad muscle that lies along the anterior chest wall. It is composed of two heads: the clavicular head and the sternal head (Fig. 17G-1). The clavicular head originates from the clavicle and upper sternum. Its fibers run parallel and insert on the lateral aspect of the bicipital groove. The sternal head has a larger manubrial segment that arises from the manubrium and the first five costal cartilages. The sternal head also has a smaller segment that originates inferiorly from the external oblique fascia, transversalis fascia, and the fifth and sixth ribs. The sternal head also inserts on the humerus lateral to bicipital groove. However, the fibers of the sternal head twist posterior to the clavicular head so that the most inferior fibers insert on the most superior aspect of the groove and the most superior fibers insert inferiorly. This constitutes a 180-degree twist, which may affect pathologic rupture.

Shoulder Coracoacromial ligament

Deltoid (origin)

901

Clavicle

Supraspinatus tendon

Pectoralis major Sternum

Subscapularis tendon Deltoid Long head of biceps

Musculocutaneous nerve

Pectoralis major Coracobrachialis Axillary artery Biceps

Triceps

Teres major

Shaft of humerus

Latissimus dorsi

Figure 17G-1   Anatomy of the pectoralis major tendon. The sternal and clavicular head converge to insert lateral to the bicipital groove. The sternal head fibers make a 180-degree twist before insertion.

The insertion site itself is 4 to 6 cm in length and 5 mm in width.2-4 Despite this detailed description, the tendon has been described as either bilaminar or trilaminar. In a bilaminar description as given earlier, the anterior tendon is 1 cm in length, and the posterior tendon is up to 2.5 cm in length.3 The muscle itself is invested in a fascial layer. This fascia is continuous with the medial antebrachial and brachial fascia of the arm. It is important to recognize this anatomy intraoperatively.3 The pectoralis major is supplied by the pectoral nerves. The lateral pectoral nerve arises from the lateral cord of the brachial plexus that is supplied by the C5 through C7 nerve roots. The lateral pectoral nerve supplies the clavicular head and medial aspect of ������������������ the sternal �������������� head.3 The medial pectoral nerve arises from the medial cord, which is supplied by the C8 through T1 nerve roots. It supplies the lateral portion of the sternal head and the pectoralis minor.3 Both nerves enter the medial and inferior aspects of the muscle bellies of the pectoralis major.3 The blood supply to the pectoralis major arises from the pectoral branch of the thoracoacromial artery.3

The pectoralis major adducts, internally rotates, and forward-flexes the humerus.2,4 Because of the twisting of the fibers at the humeral insertion, the muscle fibers are shortest superiorly and inferiorly and longest centrally.4 With terminal humeral extension, the inferior sternal fibers are stretched disproportionately, which predisposes to rupture.4

Classification A classification system for pectoralis major tears has been proposed by Tietjen.5 However, most tears can be classified by a more simple description of the location and extent of the tear. The four sites of tears described from lateral to medial are a bony avulsion from the proximal humerus, rupture at the bone-tendon interface, rupture at the musculotendinous junction, and rupture in the muscle belly itself. The extent of the tear is more difficult to quantify but is described as either partial or complete. The Tietjen classification describes three types of rupture (Table 17G-1). A type I is a contusion or sprain of the muscle. A type II is a partial tear. A type III is a complete

902

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  Table 17G-1  Classification System for Pectoralis Major Tears

Tietjen Classification

Description

Type 1 Type II Type III   A   B   C   D

Muscle contusion Partial tear Complete tear   Muscle origin   Muscle belly   Myotendinous junction   Tendon

tear. Type III injuries are subclassified as A through D depending on the location of the tear. Type IIIA is a tear at the muscle origin. A type IIIB is a muscle belly tear. A type IIIC is a tear at the myotendinous junction. A type IIID is a tear of the tendon.5 Most tears are either at the muscle-tendon junction or tendon avulsions.2 In one meta-analysis of 81 tears, 56 were tendon avulsions, 21 occurred at the myotendinous junction, 4 were bony avulsions, 3 were intratendinous tears, and 2 were muscle ruptures.6 Tears may also involve one or both heads of the pectoralis major.4 The sternal head is more frequently involved in a partial or complete tear.4

Evaluation Clinical Presentation and History Virtually all pectoralis major tears occur in males.1 None have been reported in females from a sports injury.2 Several reports from one group have documented pectoralis major ruptures in female nursing home residents.7,8 However, these diagnoses were made clinically without magnetic resonance or surgical confirmation, and most were thought to be muscle tears and not traumatic tendon ­avulsions. It is safe to say that no documented traumatic tendon avulsion has occurred in a female to date. The average age of patients who suffer from a traumatic pectoralis major tear is between 20 and 40 years, but injuries have been reported in patients between 16 and 91 years of age.3,6 Most pectoralis major tears occur in weightlifters from bench pressing.4,6,9 Tears have also been reported to occur from football, water skiing, rugby, wrestling, and sailboarding.3,4 The common mechanism is an eccentric contraction of the pectoralis major, often due to a forced abduction against resistance.1,3 Nonavulsion injuries may occur from a direct blow.3 Patients will often report feeling or hearing a tearing sensation.1,6,10 They may notice bruising of the chest wall, axilla, or arm.1 Often, they will notice a defect or webbing of the armpit as well as weakness of adduction, forward flexion, and internal rotation of the arm.1,3 Some patients may give a history of steroid use. One study found that 4 of 14 patients admitted using anabolic steroids in the past.4 A second study found 12 of 33 ­ ruptures occurred in patients on steroids.2 It is not ­completely clear how steroid use may play a role in pectoralis rupture, but, based on rat studies, the pectoralis

Box 17G-1  Typical Findings of ­Pectoralis ­Major Tears Ecchymosis of chest wall, axilla, or proximal arm Webbed appearance of axilla Weakness of internal rotation, adduction, or forward flexion Pain with passive abduction

t­ endon may become stiffer, which predisposes to rupture, or the muscle bulk created from steroid use may overwhelm the capacity of the tendon to adapt.2 Interestingly, however, steroid use may also facilitate healing after ­surgical repair.2 Rarely, pectoralis major injuries may occur in association with a shoulder dislocation. A case report is documented of such an occurrence. The proposed mechanism of injury is failure of pectoralis tendon in a position of abduction and external rotation followed by dislocation.11

Physical Examination and Testing The diagnosis of pectoralis major tear may be difficult at first. In particular, a partial tear may confuse the issue. As a result, many tears are missed initially owing to misdiagnosing a partial tear if only the sternal head is torn.3 Some authors recommend a repeat examination at 4 weeks if the diagnosis is in question. This allows a more definitive assessment of asymmetry.4 Several findings are suggestive of a pectoralis major tear (Box 17G-1). Patients often present with weakness with adduction, asymmetry of the chest wall, ecchymosis, and pain with passive abduction. The asymmetry of the chest gives a webbed appearance to the axilla (Fig. 17G-2B). Active adduction accentuates the deformity (see Fig. 17G-2C). Ecchymosis can appear on the chest wall, in the axilla, or on the proximal medial arm (Fig. 17G-3). It is thought that more lateral bruising (i.e., on the proximal arm) is more consistent with a repairable tear of the tendon or the ­myotendinous junction. The differential diagnosis is small. If the patient pre­ sents with an isolated deformity and no other findings, one should certainly consider Poland’s syndrome.3 This is the congenital absence of the pectoralis major. In addition, a pop in the proximal arm followed by ecchymosis may occur from a biceps tendon tear, but the subsequent deformities should differentiate the two diagnoses.

Imaging Plain radiographs are generally not useful unless there is a bony avulsion of the insertion of the tendon.3 Computed tomography (CT) is also of limited use.3 Magnetic resonance imaging (MRI), however, is useful.1,3 It may show edema at site of the tear after an acute injury, particularly on T2-weighted images (Fig. 17G-4). MRI is generally the study of choice to confirm the clinical diagnosis and determine the site of the injury.

Shoulder

A

C

Treatment Options Nonoperative Pectoralis major tears may be managed nonoperatively (Table 17G-2). However, it is generally accepted that operative management achieves a better functional result. Certain situations are best managed nonoperatively. A tear in the muscle belly itself or even at the myotendinous junction is difficult to repair and is best managed ­conservatively.

Figure 17G-3   Ecchymosis of axilla and proximal arm from acute pectoralis major tear.

903

B

Figure 17G-2   Chronic pectoralis major tear. A, At rest, the deformity is only minimally noticeable. B, The deformity is more apparent with abduction revealing the webbed appearance of the left axilla. C, Active firing or adduction of the pectoralis major accentuates the deformity. This patient’s injury occurred 9 months previously. He elected to choose nonoperative management.

Partial tears should also be managed conservatively. This involves a short period of immobilization followed by a gradual increase in range of motion. Range of motion can be restored over a 6-week period. Strengthening may then be initiated, but bench-pressing should be avoided ­initially

Figure 17G-4   Magnetic resonance image of pectoralis major tear. Axial T2-weighted image demonstrates significant edema of the lateral aspect of the pectoralis major muscle belly with a fluid track filling the void of the tendon attachment on the proximal humerus.

904

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Operative

  Table 17G-2  Treatment Options for Pectoralis Major Tears

Indications

Management

Nonoperative

Elderly low-demand patient Partial tear Tear at or medial to myotendinous junction Patient choice (e.g., nondominant arm)

Operative

Active or high-demand patient Complete tear Tear at tendon-bone interface

Immobilize 3 wk or less Motion restored by 6 wk Strengthening except bench press by 12 wk Initiate limited/ protected bench press at 12 wk Trough technique Suture anchor technique

and gradually added, if the patient desires, at 10 to 12 weeks. Of course, the technique should be modified to avoid terminal extension and lower weights should be used.3 Often, it is prudent to maintain a permanent weight restriction. Even tears at the tendon to bone interface may be treated nonoperatively. Generally, the pain will subside, but patients will be left with residual weakness and a deformity. This management may be acceptable for older patients with lower demands or in a nondominant extremity. Certainly the patient should be counseled on the risk for residual weakness of internal rotation, adduction, and forward flexion as well as the cosmetic ­deformity.2

A

It is generally accepted that operative repair better restores strength and is therefore beneficial for young active athletes (see Table 17G-2).4,12,13 Disagreement still exists concerning the superiority of results of acute versus chronic repair, but most studies agree that acute repairs generally are technically easier and yield improved results.1,6 Several techniques have been reported in the literature. These include using screws and washers, staples, suture anchors, and bone tunnels.6 Generally, two methods have been implemented with documented success. In one technique, a strong nonabsorbable suture is passed in a locking or grasping pattern through the tendon. Usually the tendon will accept two or three of these depending on the quality of tissue and the breadth of the tendon. The suture is passed through a bone trough created at the insertion of the tendon lateral to the bicipital groove and out drill holes placed more laterally (Fig. 17G-5). The sutures are then tied over the bone bridge.4,14 A second technique places suture anchors at the humeral insertion of the pectoralis tendon (Fig. 17G-6). The tendon is then tied directly to these suture anchors.3 Chronic tears may be treated in a similar fashion if the tendon has not retracted significantly. A chronic tear is variably defined as a tear older than 6 to 8 weeks. There are case reports of good result with chronic repair as late as 13 years later. This can be achieved if the intact clavicular head tethers the sternal head, thereby preventing it from retracting. In the case report of the repair done 13 years

B

Figure 17G-5   Trough repair of pectoralis major tear. A, Illustration depicting sutures grasping the tendon, passed through the bone trough, and tied over a bone bridge. B, Operative view of bone trough.

Shoulder

A

905

B

Figure 17G-6   Suture anchor repair of pectoralis major tear. A, Illustration depicting suture anchors at site of humeral insertion. B, Operative view of suture anchors. Three 5.0-mm corkscrew anchors are buried in the bone. Each anchor has two sutures. The tendon can be viewed to the left with traction sutures in place.

after the injury, the same patient had an acute repair on the contralateral side and noted good results on both sides.15 Chronic tears may be retracted, and primary reconstruction may not be possible. In these cases, a bridging graft may be necessary. Two techniques reported in the literature with success involved use of an Achilles tendon allograft and a hamstring tendon autograft. Both patients achieved acceptable results according to the authors.1,16

Weighing the Evidence Surgical management generally results in improved patient satisfaction, strength, cosmesis, and return to sport as compared with nonsurgical management (Fig. 17G-7).3,12 Bak and colleagues reported in a meta-analysis that surgical repair resulted in 88% good or excellent results, compared with 27% for nonsurgical management.6 Zeman and associates treated four patients surgically and five patients nonsurgically.13 All four treated surgically had excellent results, and all five treated nonsurgically had strength deficits and difficulty returning to sports.13 Park and Espiniella reviewed 29 cases; they noted that 9 of 10 patients treated surgically had good or excellent results, and only 7 of 12 treated nonsurgically had good results.12 The largest single series of repair of pectoralis major tears included 33 patients reported on by Aarimaa and ­colleagues.2 They repaired the tears with a bone trough and drill holes in 10 patients, with suture anchors in 12 patients, and with direct tendon to tendon repair in 11

patients.2 They reported 13 excellent results, 17 good results, and 3 fair results. The patient outcomes were typically worse if a tear occurred in or near the muscle and if surgery was delayed. There was no statistically significant difference for outcomes for acute versus delayed repair due to a small sample size. The only factor that correlated with a positive outcome was use of anabolic steroids. In the same study, Aarimaa and colleagues also did a meta-analysis of outcomes of pectoralis repair in 66 studies with a minimal 6-month follow-up (73 cases).2 They concluded that acute repair yielded better outcomes than delayed repair or conservative management, with statistical significance.2 Even chronic repairs appear to yield better results than conservative management. Schepsis and associates ­compared acute repair, chronic repair, and conservative management.14 The study has some shortcomings, including a small number of patients and a lack of ­ complete physical examination follow-up for all patients. However, as a general trend, the patients repaired acutely yielded better subjective and objective results than the chronic repair group, although there was no statistical difference. Both groups did significantly better than the conservatively managed group.14 One case report of a chronic repair with Achilles tendon indicated that the patient was satisfied cosmetically, but there was no formal strength evaluation and the patient did not resume the bench press.16 Biomechanical testing confirms improved function and strength with surgical repair. In the initial study used to define the mechanism of pectoralis major injury, Wolfe

906

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

Figure 17G-7   Postoperative pectoralis major repair. A, At rest. B, Abduction showing diminution in size of the pectoralis major with symmetry. C, Active adduction showing symmetry.

and colleagues compared peak torque and work-repetition for surgically managed and nonsurgically managed injuries. They noted better peak torque and work-­repetition for patients who underwent surgical repair as compared with those treated nonoperatively.4 Both groups had 6 patients with peak torque 20% better and work-­repetition 15% to 40% better in the surgical group.4 Hanna and associates also noted improved peak torque results in a surgical group.17 Twenty-two injuries were treated surgically in 21 patients with evaluation by peak torque. Patients achieved 99% peak torque strength compared to the contralateral side for surgical management and only 56% for nonsurgical management.17

A u t h o r ’ s P r e f e r r e d M e t h o d I use both the bone trough and the suture anchor repair technique. Both techniques have yielded good results. The suture anchor technique is less technically demanding. I tend to use this for acute repairs. Often a cuff of tendon remains attached to the bone, which may facilitate healing. The site of the repair is first prepared with a bur to minimally decorticate the bony surface to promote healing. Generally three 5.0-mm suture anchors are sufficient. The three anchors are spaced about 1.5 cm apart. I use a corkscrew anchor, which is placed in a predrilled hole.

C

Each anchor has two sutures. The sutures are passed in a cruciform fashion through the tendon. This is achieved by passing one limb of one suture in simple fashion, followed by passage of both limbs of the second suture in a horizontal mattress fashion. The simple suture is tied down first to pull the tendon over to the anchor. The mattress suture reinforces the first suture. Other suture techniques have been described, including locking patterns, which may certainly work better for poor tissue. I use the bone trough technique for chronic repairs. One gets a sense that the tendon to bone contact is improved with the bone trough technique, but it is more time consuming, and the results are not better than the suture anchor technique in my hands. However, I have used both techniques interchangeably with good results and do not use any specific algorithm to determine which technique to use. In this technique, I create a 3- to 4-cm trough about 4- to 5-mm wide through the proximal humerus insertion. Three drill holes are spaced evenly 1 cm lateral to the edge of the trough. The holes are made with a 1.7- or 2.0-mm drill bit, depending on the size of the suture used for repair. I use a No. 5 FiberWire suture so that the 2.0 drill bit works better. Two No. 5 FiberWire sutures are passed in a locking fashion through the tendon. This leaves four suture ends, which are passed through the trough and out the lateral holes using a suture passer. Each suture is then tied over the bone bridge. The trough must be wide enough to accept a portion of the tendon end.

Shoulder

Postoperative Prescription, Outcomes Measurement, and Potential Complications Postoperatively, pectoralis major repairs are immobilized. After a period of immobilization, active range of motion is restored followed by strength training. Several protocols have been reported. Aarimaa and colleagues reported immobilizing the arm for 2 to 3 weeks followed by achieving the goal of full active range of motion by 6 weeks. Resistance training begins at 6 weeks, with unlimited function at 3 months.2 Most other authors report a postoperative regimen that includes immobilization in a sling for 3 to 6 weeks followed by active range of motion. Weightlifting begins at 3 months with limits on bench press. The patient is released to full activity between 4 and 6 months. Most studies report the outcome of pectoralis major repair clinically by the patient’s ability to return to sport. Biomechanical testing can quantify the outcome by measuring adduction and internal rotation strength. A final criterion used by many patients is cosmetic results. Many patients are bodybuilders or are aware of their appearance. Cosmetic results may not be reliably restored in the case of a partial repair or repair of a chronic tear. A partial repair may be necessary if a tear is not a pure tendon tear. Few complications have been reported with pectoralis major repair. No nerve or vascular injuries have been reported despite the relative proximity to the brachial plexus and axillary vascular bundle. Failed repairs have been reported and may need to be revised with repeat repair or an allograft reconstruction. Additionally, a suspected tendon avulsion may be explored and found to be a myotendinous or intramuscular rupture not amenable to repair.

Criteria for Return to Play If patients are treated nonoperatively, they generally do not wish to return to a high-level sport. However, if they have a desire to return to sports, they can generally do so when they have full range of motion and are free of pain. This also must take into account that they may not be capable of functioning in the given sport with their deficit. Patients treated operatively generally can return to contact sports by 4 to 6 months. Certain goals must be met, including full range of motion, painless resistance exercises, and strength approximating the contralateral side.

907

Box 17G-2 Typical Findings of ­Latissimus Dorsi Tears Tender mass posterior to axilla Ecchymosis Decreased adduction power

Classification Only four cases of traumatic latissimus rupture have been reported in the literature with one involving teres major also.18-21 There are no classification systems for latissimus tears. Certainly, partial tears may occur, but are not reported.

Evaluation Clinical Presentation and History Of the four reported cases of latissimus dorsi rupture, two occurred while water skiing, one from rock climbing, and the fourth from golfing.18–21 The mechanism appears to involve a powerful traction force. For example, the injuries sustained while water skiing occurred when the patients were either pulled up forcefully by the tow rope or when they did not release their grasp of the rope after falling.18,19 Complaints on presentation include pain, swelling, and weakness with adduction.18

Physical Examination and Testing On physical examination, the patient may have ecchymosis, a tender mass posterior to the axilla, and decrease in adduction power (Box 17G-2).18,19 With the arms abducted, asymmetry may be discernible.

Imaging MRI is the most useful imaging modality.18,19

Treatment Options Nonoperative

LATISSIMUS DORSI

There are only four reports in the literature of latissimus dorsi tears. Of those four, one was treated nonoperatively (Box 17G-3). That case had a good result according to the authors, but that case was a bony avulsion of the latissimus insertion. It also included an avulsion of the teres major insertion.21

Relevant Anatomy and Biomechanics

Operative

The latissimus dorsi originates from thoracic spine, thoracolumbar fascia, and iliac crest.18 It inserts on the proximal humerus from the inferior aspect of the lesser tuberosity to the medial aspect of the biceps groove.18 It mainly acts to internally rotate, adduct, and extend the shoulder.18 In addition, it also acts to lift the trunk forward and upward while climbing.18

The two acute traumatic ruptures reported in the literature were treated surgically (see Box 17G-3). Both acute repairs returned to full sport.18,19 One was repaired with a two-incision technique using an anterior cruciate ligament guide to create bone tunnels. Suture passed through the tendon is passed through bone tunnels and tied over the proximal humeral cortex. This patient had only an 11% to

908

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 17G-3 Treatment Options for Latissimus Dorsi Tears

• Nonoperative • Operative Suture anchors Bone tunnels One-or two-incision technique 14% adduction strength deficit at final follow-up.18 The second case was managed with a one-incision technique using suture anchors. The patient was placed in the lateral decubitus position. The tendon was identified and repaired anatomically using suture anchors placed at the site of the humeral avulsion. This patient recovered with improved strength as compared with the ­contralateral side.19 The third surgically treated case was for a chronic rupture.20 This patient presented 2.5 years after injury. He underwent surgical repair through a two-incision technique. The tendon was attached to the insertion through drill holes. The patient’s strength improved, although he was not able to return to his prior level of elite rock climbing at the time of final follow-up.20

C

r i t i c a l

P

o i n t s

l The pectoralis major has two major heads: clavicular and sternal. l The sternal head is most frequently involved. l The most common mechanism of injury is the bench press. l Tears may occur at many sites, but the site amenable to repair is at the tendon-bone interface. l Patients present with ecchymosis of the proximal arm or axilla, a webbed axilla, and weakness in adduction and internal rotation. l Operative management yields better results than nonoperative management. l Repair can be done with either a bone trough technique or suture anchors. l Chronic tears may be repaired with good success either directly or with an interposition graft. l The mechanism of latissimus dorsi tear is forceful ­traction. l The patient complains of weakness in adduction or a painful lump behind the axilla. l Latissimus dorsi tear is a rare injury; only four cases have been reported. l Because of its rarity, there is no proven benefit to operative intervention, but results have been good surgically.

Weighing the Evidence Because tears of the latissimus dorsi are so rare, it is difficult to conclude whether operative intervention offers any true advantage. Of the four cases reported in the literature, one was not an isolated injury, and it was treated nonoperatively with good success. The other injuries were treated surgically, also with a good result. Certainly the latissimus dorsi is not an expendable muscle, but it is transferred by orthopaedic and plastic surgeons surgically for other reasons. Multiple studies done on free latissimus flap transfers and postsurgical strength deficits demonstrate measurable deficits, but they are not always significant.18 Consequently, it is not clear that repair offers a significant advantage with the available data.

Postoperative Prescription, Outcomes Measurement, and Potential Complications After repair of the latissimus dorsi, most authors limited motion for 3 to 6 weeks before permitting full active range of motion. Patients were allowed to lift weights at 3 months and return to sport by 5 to 6 months.18-20 No complications have been reported in the four documented cases.

S U G G E S T E D

R E A D I N G S

Aarimaa V, Rantanen J, Heikkila J, et al: Rupture of the pectoralis major muscle. Am J Sports Med 32:1256-1262, 2004. Henry JC, Scerpella TA: Acute traumatic tear of the latissimus dorsi tendon from its insertion: A case report. Am J Sports Med 28:577-579, 2000. Lim JK, Tilford ME, Hamersly SF, Sallay PI: Surgical repair of an acute latissimus dorsi tendon avulsion using suture anchors through a single incision. Am J Sports Med 34:1351-1355, 2006. Livesey JP, Brownson P, Wallace WA: Traumatic latissimus dorsi tendon rupture. J Shoulder Elbow Surg 11:642-644, 2002. Petilon J, Carr DR, Sekiya JK, Unger DV: Pectoralis major muscle injuries: Evaluation and management. J Am Acad Orthop Surg 13:59-68, 2005. Schepsis AA, Grafe MW, Jones HP, Lemos MJ: Rupture of the pectoralis major muscle: Outcome after repair of acute and chronic injuries. Am J Sports Med 28:9-15, 2000. Spinner RJ, Speer KP, Mallon WJ: Avulsion injury to the conjoined tendons of the latissimus dorsi and teres major muscles. Am J Sports Med 26:847-849, 1998. Wolfe SW, Wickiewicz TL, Cavanaugh JT: Ruptures of the pectoralis major muscle: An anatomic and clinical analysis. Am J Sports Med 20:587-593, 1992.

R eferences Please see www.expertconsult.com

Shoulder

S E C T I O N

909

H

Glenohumeral Instabilities 1. Glenohumeral Instability in Adults Seth C. Gamradt and Russell F. Warren

Instability of the shoulder joint encompasses a broad range of pathology from subtle increased laxity to recurrent dislocation. Historical descriptions of the treatment of shoulder instability consisted only of relocation maneuvers for anterior instability. Later, surgical treatment of recurrent anterior shoulder dislocation evolved, and posterior instability and multidirectional instability (MDI) became recognized as separate entities in shoulder pathology requiring different treatments. With increasing knowledge of the anatomy and biomechanics of the shoulder and improved imaging studies (magnetic resonance imaging [MRI]), recognition, diagnosis, and nonoperative management of shoulder instability have improved considerably. In addition, with shoulder-specific implants and extensive evolution of open and arthroscopic surgery, operative indications have widened, and results are improved. No single entity or lesion is responsible for instability in the shoulder, and the joint should be looked on as a circle wherein increased translation in one direction demands laxity of the soft tissue on the side opposite the instability to allow the increased motion.1,2 Treatment of shoulder instability has become more focused on the pathoanatomy of traumatic or atraumatic instability. The goal of nonoperative and surgical treatment is a pain-free, stable shoulder with full range of motion of the glenohumeral joint. Most glenohumeral dislocations and subluxations are initially treated with rehabilitation and physical therapy. However, results of nonoperative treatment in young athletic patients suggest that early surgical repair might be warranted in certain cases. Understanding the pathology and surgical anatomy of the unstable glenohumeral joint is dependent on a thorough familiarity with the anatomy and the biomechanics of the normal, stable joint. This chapter reviews the relevant anatomy and biomechanics of the normal and injured glenohumeral joint. The indications, techniques, and results of operative and nonoperative treatment of glenohumeral instability are presented.

ANATOMY AND BIOMECHANICS Normal Anatomy of the Shoulder The shoulder joint has the greatest range of motion of all the joints in the body, and the bony articulation between the glenoid and humerus has little inherent stability. There are few bony restraints to movement and a relatively small

area of articular contact. Thus, the glenohumeral joint is dependent on soft tissue restraint, including capsule, labrum, ligament, and surrounding muscles.

Glenohumeral Joint The glenohumeral joint is the joint between the glenoid of the scapula and the humeral head. The glenoid, which is oval, is longest in its inferosuperior diameter. It is connected to the body of the scapula by the glenoid neck with an angle of about 7 degrees posterior and slightly superior version.3 The bony articular surface of the glenoid is almost flat; the center of the glenoid is only slightly deeper than the periphery. In the center of the glenoid, called the “bare area,” the articular surface is thin. Compared with the hinged joint of the knee or the ball-and-socket joint of the hip, the shape of the glenoid contributes little to stability of the joint. The body of the scapula lies on the posterior thoracic wall facing the glenoid articular surface anterior and superior toward the humeral head. The position of the scapula on the thoracic wall constitutes the scapular plane of motion, with the glenoid articular surface directed in 35 degrees of anterior and 5 degrees of superior version to the sagittal plane. This version of the glenoid increases the posterior and inferior stability of the joint. The shoulder joint superiorly is covered by the acromion of the scapula. The acromion arises posteromedially from the spine of the scapula, has a free posterior and lateral edge, and is connected to the coracoid of the scapula by the coracoacromial ligament anteriorly and to the clavicle through the acromioclavicular joint medially. The roof of the shoulder joint is important for superior and anterior stability. The humeral head is retroverted 30 to 40 degrees to the distal humeral epicondyles to meet the anteversion of the scapula. About half of the humeral head sphere is covered in articular cartilage. The cartilage ends about at the level of the surgical neck of the humerus and extends to the lesser tuberosity anteriorly and to the bicipital groove and greater tuberosity superiorly and posteriorly.

Labrum The labrum is firmly attached to the rim of the glenoid inferiorly and extends to a more loosely attached biceps anchor superiorly, the origin of the tendon of the long head of the biceps. The labrum contributes to the stability of the joint by adding depth and increasing the surface of

910

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the glenoid and mediates attachment of the capsule, ligaments, and biceps tendon to the glenoid. The anterosuperior portion of the labrum is variable, and normal variants include loose attachment and a sublabral foramen.4

Capsule

B S GH

The capsule of the glenohumeral joint is normally redundant enough to allow the extensive range of motion of the joint. The capsule also has normal thickenings within its substance (glenohumeral ligaments) that are critical static restraints to dislocation. The capsule is confluent with the labrum at the rim of the glenoid and attaches laterally on the humeral head, blending with the insertion of the rotator cuff tendons. The capsule between the superior edge of the subscapularis tendon and the anterior border of the supraspinatus tendon forms the rotator interval. In this interval, the capsule, the coracohumeral ligament, superior glenohumeral ligaments, and long head of the biceps are the restraints to anterosuperior migration.5,6 The rotator interval is an important secondary determinant of anterior stability of the joint; increased laxity can increase instability of the joint, whereas contracture of the rotator interval limits external rotation.5,7

P

L

M G H L

PC

PB

A

A B AP IGH L C

Ligaments The glenohumeral ligaments are thickened bands within the capsule that stabilize the glenohumeral joint (Fig. 17H1-1). The ligaments act as static restraints to glenohumeral translation at different arm positions (Table 17H1-1).2,8,9 The superior glenohumeral ligament (SGHL) consistently runs from the superoanterior rim of the glenoid across the rotator interval, above the biceps tendon, to its insertion with the capsule on the humeral head above the lesser ­tuberosity near the bicipital groove. The SGHL is an important restraint to inferior translation of the adducted arm. The middle glenohumeral ligament (MGHL) normally arises from the anterior upper third of the glenoid rim and the labrum and runs beneath and across the subscapular tendon to its humeral attachment. This ligament becomes taught with the arm in 45 degrees of abduction and is highly variable. Recognizing the variability of the capsulolabral complex in this region is important to avoid overtreatment arthroscopically. The Buford complex is a normal anatomic variant with a cord-like middle glenohumeral ligament and absent anterosuperior labrum. The inferior glenohumeral ligament (IGHL) complex consists of an anterior band, a posterior band, and an axillary pouch (Fig. 17H1-2). The IGHL stabilizes the shoulder in abduction. The anterior band of the inferior glenohumeral ligament runs from the anteroinferior rim of the glenoid to the undersurface of the humeral head. The posterior band runs in the posterior capsule from the posteroinferior rim of the glenoid to the posterior part of the humeral head. Normal redundancy in the axillary pouch is exaggerated in recurrent dislocators and in MDI. The coracohumeral ligament in the anterior part of the rotator interval runs between the base of the coracoid and the humeral head and inserts in the transverse humeral ligament, which acts as a roof of the bicipital groove between the lesser and the greater tuberosities. Both of these ligaments have minor roles in the stability of the joint.

Figure 17H1-1    Schematic drawing of the shoulder capsule showing the glenohumeral ligaments, highlighting the inferior glenohumeral ligament. A, anterior; AB anterior band; AP, axillary pouch; B, biceps tendon; P, posterior; PB, posterior band; and PC, posterior capsule. (From O’Brien SJ, Neves MC, Arnoczky SP, et al: The anatomy and histology of the inferior glenohumeral ligament complex of the shoulder. Am J Sports Med 18[5]:449-456, 1990.)

Muscles The muscles surrounding the shoulder joint act as dynamic stabilizers during shoulder motion. Particularly important are the muscles of the rotator cuff, whose tendons run anterior, superior, and posterior to the joint, and act as stabilizers by compressing the humeral head into the glenoid cavity during motion.10-13 The subscapularis muscle originates from the undersurface of the body of the scapula and runs anterior to the shoulder joint. The muscle inserts on the lesser tuberosity in a cord-like tendon at its upper part and a more muscular attachment at its lower part along the lesser ­tuberosity. The upper edge of the subscapular tendon is the lower border of the rotator interval. The subscapularis muscle is innervated by the subscapular nerve from the posterior cord of the brachial plexus. The muscle applies an anterior compression load to the joint and an internal rotation force to the humeral head; with increased abduction of the arm, it functions as a humeral head depressor. The supraspinatus muscle originates from the anterosuperior part of the scapula above the scapular spine and runs

Shoulder

  TABLE 17H1-1  Static Stabilizers of the Shoulder Position of Arm

Ligament

Abduction, external rotation Abduction, posterior translation 45 degrees of abduction Adduction

Anterior IGHL Posterior IGHL MGHL SGHL

IGHL, inferior glenohumeral ligament; MGHL, middle glenohumeral ligament; SGHL, superior glenohumeral ligament.

along the upper surface of the glenoid above the shoulder joint, where it becomes a tendon that inserts on the greater tuberosity lateral on the humeral head. Its anterior edge is the superior border of the rotator interval. The supraspinatus is innervated by the suprascapular nerve. The nerve arises in the posterior cord of the brachial plexus and runs underneath the suprascapular ligament in the suprascapular notch on the superior edge of the scapula, into and underneath the supraspinatus muscle in the supraspinous fossa. The supraspinatus functions as an important humeral head depressor and humeral joint compressor and stabilizer with increased abduction of the arm. The tendon of the long head of the biceps travels beneath the pectoralis major insertion under the transverse humeral ligament between the insertion of the subscapularis and the supraspinatus tendons on the humeral head. The tendon then enters the capsule on top of the humeral head and courses intra-articularly to its origin on the superior glenoid labrum. The long head of the biceps was thought to play a minor role in humeral head depression at

911

the beginning of abduction of the shoulder joint, although this is doubtful; a recent well-done dynamic electromyogram (EMG) study showed no significant contribution of the long head of the biceps during shoulder motion with the elbow locked in a brace.14 The infraspinatus muscle is posterior and inferior to the supraspinatus. It originates from the inferior surface of the scapula below the spine and runs lateral to its insertion on the posterior part of the greater tuberosity and posterior part of the humeral head, crossing the shoulder joint on its most superoposterior aspect. It is innervated by the suprascapular nerve as well. When the nerve leaves the supraspinatus muscle through the supraspinous fossa, it runs distally laterally around the base of the spine of the scapula through the spinoglenoid notch to the infraspinous fossa, where it innervates the infraspinatus muscle. The muscle functions as an external rotator of the shoulder and, with increasing abduction in the shoulder, as a humeral head depressor.15 The trapezius, rhomboids, latissimus dorsi, serratus anterior, and levator scapulae are important scapular stabilizers in that they position the glenoid in an anteverted and superior position to articulate with the retroverted humeral head. The position of the glenoid in relation to the humeral head increases the posterior and inferior stability of the glenohumeral joint under motion.

Vessels and Nerves The anterior and posterior circumflex arteries to the humeral head are branches of the axillary artery. The posterior circumflex artery runs with the vein and the axillary

SGHL MGHL PB

PB

SGHL MGHL AB AB

A

B

Figure 17H1-2    Function of the glenohumeral ligaments. The glenohumeral capsule is enhanced by ligamentous thickenings that provide static restraint at different functional positions. A, With the shoulder in adduction, the superior glenohumeral ligament (SGHL) and middle glenohumeral ligament (MGHL) are tight, and the inferior glenohumeral ligament (IGHL) is lax. B, With abduction and external rotation, the IGHL anterior band (AB) and posterior band (PB) tighten. (From Warner JP; Boardman ND III: Anatomy, biomechanics, and pathophysiology of glenohumeral instability. In: Warren RF, Craig EV, Altcheck DW (eds): The Unstable Shoulder. Philadelphia, Lippincott-Raven, 1999, pp 51-76.)

912

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

nerve underneath the capsule in the axilla and underneath the teres minor through the quadrangular space posteriorly. The anterior circumflex artery runs with its two veins along the lower border of the subscapularis tendon to its insertion on the lesser tuberosity. The axillary nerve is at risk during inferior capsular dissection during open surgery because it can lie as close as 2 mm from the inferior capsule. The musculocutaneous nerve arises medial to the conjoined tendon and passes about 3 to 5 cm below the coracoid, where it can be damaged by retractors placed under tendon.

BIOMECHANICS OF SHOULDER STABILITY As mentioned earlier, the articulation between the shallow glenoid and the round humeral head has little inherent stability. The shoulder is dependent on the surrounding soft tissue for its stability in various positions of motion. At rest, the joint is held in position partly by the vacuum effect of the negative pressure inside the capsule.16 With arm movement, however, a complex interaction between dynamic ­stabilization from the muscles and static ­ stabilization from the ligaments around the shoulder takes place. At the extremes of motion, rotation of the humeral head is ­coupled with translation on the glenoid cavity,11,12,16,17 but translation of the humeral head on the glenoid is normally confined to only a few millimeters during motion.12,18,19 The glenohumeral ligaments serve as important static stabilizers, particularly at the extremes of motion. With the arm adducted, the main stabilizers against inferior translation of the humeral head are the superior glenohumeral ligament, the coracohumeral ligament, and the rotator interval. These structures are under tension in both flexion and extension with the arm adducted and externally rotated. They also resist posterior translation but have no major role against anterior translation. With increasing abduction of the shoulder to 45 degrees, the middle glenohumeral ligament becomes the major restraint against anterior translation of the humeral head; with the arm in external rotation, the ligament comes under maximal tension.2,9,20 When the arm is abducted farther toward 90 degrees, the inferior glenohumeral ligament comes under increasing tension, resisting inferior translation of the humeral head; and at 90 degrees of abduction, the inferior glenohumeral ligament is the primary restraint to inferior translation. The anterior and posterior bands of the inferior glenohumeral ligament have important roles in the stability of the joint. With the arm in abduction and external rotation, the anterior band moves from inferior to more anterior, restraining anterior translation of the humeral head, and the posterior band moves from posterior to more inferior, restraining inferior translation. When the arm is internally rotated, the anterior band moves inferior and the posterior band moves posterior, protecting against translation of the humeral head in these directions. When the shoulder is flexed, the posterior band of the inferior glenohumeral ligament is under tension and is the primary restraint to anterior and posterior translation of the humeral head; during extension, the tension shifts to the anterior band as the primary restraint to translation.2,20,21

These reciprocal movements of the different portions of the inferior glenohumeral ligament are thought to contribute to the “rollback” of the head on the glenoid during motion. Dynamic stabilization by the muscles of the shoulder joint, like static stabilization by the ligaments, is related to the position of the arm. The muscles function in a complex balance to keep the joint stabilized during the full range of motion. The rotator cuff muscles are the main stabilizers in that they compress the humeral head into the glenoid cavity during motion, centering the humeral head. The scapular stabilizing muscles play a secondary role, maintaining the position of the glenoid relative to the humeral head during motion.

PATHOANATOMY Initial surgical treatments of shoulder instability focused on preventing further dislocations. Often these reconstructions were aimed at limiting external rotation (PuttiPlatt, Magnusson-Stack) or providing a bony block to anterior instability (Bristow). These procedures, although often effective in preventing further dislocations, decrease shoulder mobility, especially in external rotation, and are known to cause early osteoarthritis. Improved understanding of the pathoanatomy of shoulder instability has enabled surgeons to focus on arthroscopic and open restoration of anatomy, reserving nonanatomic stabilizations for complex or revision situations. History, mechanism of injury, physical examination findings, and imaging studies all play a role in determining the direction, severity, and chronicity of instability. Traumatic anterior instability most often produces a classic Bankart lesion involving a detachment of the anteroinferior capsulolabral complex below the equator of the glenoid. The Bankart lesion therefore disrupts the main static anterior stabilizer of the glenohumeral joint in abduction and external rotation. However, some degree of capsular injury or laxity must also accompany a Bankart lesion for a shoulder dislocation to occur. Speer and associates showed in a cadaver model that creation of a Bankart lesion alone was insufficient to allow dislocation.22 Although the Bankart lesion is the most common pattern of pathoanatomy in traumatic anterior instability, it is important to understand and recognize less common causes of anterior instability. In addition to labral detachment, capsular laxity is a common finding in anterior instability, especially in chronic cases or recurrent instability. Rotator interval capsular laxity or deficiency can also contribute to anterior instability. Addressing both labral detachment and capsular laxity is critical in successful surgical treatment of anterior shoulder instability. Occasionally, in less than 10% of cases of traumatic anterior instability, the capsule can avulse from the humeral side, producing a humeral avulsion of the glenohumeral ligament (HAGL) lesion.23-26 Bone injury or deficiency is also part of the spectrum of disease in anterior instability. Acute fracture of the anteroinferior glenoid (bony Bankart) or bony erosion of the glenoid due to recurrent instability affects surgical decision making and can require grafting. Lastly, the size of Hill-Sachs defects (posterosuperior compression fractures of the humeral head) must be noted during surgical planning for instability repairs. Failure to recognize and properly address a loss

Shoulder

of bony restraint to instability or an engaging Hill-Sachs lesion can lead to failure of instability surgery, especially in arthroscopic reconstructions. Multidirectional instability is recognized as a separate entity and is not associated with a classic Bankart lesion. Instability in this syndrome is not limited to the anteroinferior instability in abduction and external rotation. Congenital or pathologic hyperlaxity of the capsule and ligaments can result in repetitive and painful subluxation. Repetitive microtrauma in a susceptible shoulder is thought to play a role in the pathophysiology of this condition. Frank posterior dislocation is uncommon in the absence of high-energy trauma, seizure, or dislocations in the setting of electrical shock. Isolated or repetitive posterior subluxations are more common in athletics. Posterior instability pathoanatomy manifests as labral detachment and capsular laxity allowing increased posterior translation of the humeral head.

CLASSIFICATION OF SHOULDER INSTABILITY Glenohumeral instability encompasses a spectrum of disorders ranging from the dislocated shoulder to more subtle instability and covers a wide range of disabilities due to increased translation in the shoulder joint (Box 17H1-1). Although comprehensive classification systems have been devised,27 classification of shoulder instability is mostly described by four traits: degree, direction, frequency, and etiology.

CLINICAL PRESENTATION OF SHOULDER INSTABILITY Clinical Presentation and History A careful history and physical examination is the cornerstone of evaluating a complaint of shoulder instability. At a minimum, information collected during a history on a patient with acute or chronic dislocation should include the information contained in Box 17H1-2. The age of a patient at the time of dislocation is one of the most important prognostic factors for recurrent dislocations (Table 17H1-2). A greatly increased rate of dislocation is observed with younger patients with a first-time dislocation28; however, the redislocation rate in older individuals should not be regarded as insignificant. Second, rotator cuff tear is much more common in the older patient (40 years old or older) sustaining a shoulder dislocation. Bilateral shoulder instability is a hallmark of MDI but can be observed in unidirectional instability as well. By obtaining a careful history including the mechanism of injury (including position of the arm during dislocation), it is often possible to determine the direction of instability. Anterior instability often occurs with the arm in a vulnerable position of abduction and external rotation, whereas posterior subluxations are more ­common in adduction and internal rotation. Determining the chron­ icity and the number of dislocations or subluxations is

913

Box 17H1-1  Classification of  Glenohumeral Instability Degree Dislocation Subluxation Microinstability Direction Unidirectional Bidirectional Multidirectional Frequency Acute Recurrent Locked Etiology Traumatic Atraumatic Acquired

also important and can guide treatment. The treatment of a first-time dislocator can clearly be different from a recurrent dislocator with chronic instability. Patients may describe a history of a frank dislocation after acute trauma, repetitive microtrauma linked to the overhead actions of specific sports, or generalized joint laxity. The number of emergency room reductions required can often differentiate a patient who has recurrent dislocations from a patient with recurrent subluxations. The pain pattern in shoulder disability is also helpful in establishing a diagnosis. Pain in the neck and radiating arm pain may indicate a cervical spine or brachial plexus injury. Pain accompanied by weakness, pain that radiates to the deltoid, and night pain can be associated with a rotator cuff tear. Inquiring about pain with specific movements or in Box 17H1-2  History Taking for Shoulder ­Dislocation Age Handedness Unilateral or bilateral complaint Family history of shoulder dislocations Initial traumatic event Position of arm during subluxation or dislocation Number of previous dislocations or subluxations Degree of trauma required for recurrence Dislocation during sleep? Voluntary dislocation? Was the dislocation reduced in emergency room? Presence of and location of pain Sensory disturbance? Presence or absence of mechanical symptoms Sport or recreational activity Previous shoulder surgeries

914

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 17H1-2  Redislocation Rate after Anterior Dislocation Stratified by Age at Initial Trauma* Age (yr)

Recurrence Rate % (No.)

0-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 Total

16.7 (2/12) 61.3 (19/31) 34.6 (9/26) 10.5 (2/19) 27.0 (10/37) 13.6 (3/22) 22.2 (4/18) 26.7 (4/15) 29.4 (53/180)

*Note that the redislocation rate is highest in the 21- to 30-year-old group. The dislocation rate in older patients is still very significant, however. From Kralinger FS, Golser K, Wischatta R, et al: Predicting recurrence after primary anterior shoulder dislocation. Am J Sports Med 30(1):116–120, 2002.

specific positions can also aid in establishing the position and direction of dislocation. The sport or recreational activity pursued by the patient is critical for treatment decision making. A contact athlete is obviously at higher risk for redislocation when he or she returns to sport. In addition, athletes in certain sports often require full overhead range of motion and ­ hyperexternal rotation to be effective (e.g., baseball pitchers and volleyball players); this range of motion is important to preserve when treating these types of athletes. Lastly, a complete surgical history of each shoulder should be documented, and operative reports and arthroscopic photographs should be obtained. Patients with shoulder instability routinely present with a complex surgical history, especially in the tertiary care setting. The type of surgery should be delineated (arthroscopy alone versus arthroscopic stabilization versus open stabilization). The type (metal versus bioabsorbable) and location (glenoid versus humerus) of suture anchors should be documented and any surgical treatment of the capsule (plication versus shift) should also be noted. A complete understanding of the previous operations performed will greatly aid in nonoperative and operative management of recurrent dislocators.

Physical Examination The physical examination of a patient with an acute dislocation begins with inspection of the position and contour of the shoulder and palpation of shoulder and muscles around the joint. Range of motion of the shoulder in a dislocation is severely limited and is not tested before reduction. In an acute anterior dislocation, anteroinferior fullness can be palpated, and the normal contour of the deltoid is lost. In addition, the coracoid is prominent, and a void can be palpated lateral to the coracoid. The arm is usually held in slight abduction and external rotation. Posterior dislocation is more difficult to appreciate on physical examination, but a posterior prominence may be noted. With a posterior dislocation, the arm is often held in internal rotation and adduction. A neurovascular examination of the upper extremity should also be performed to rule out brachial plexus and axillary nerve injury.

Physical examination findings in patients who pre­sent with a chronic complaint of recurrent dislocation or subluxation are more subtle. The goal of the examination is to determine the direction and degree of instability for each shoulder and detect concomitant muscle and nerve injuries. We start by observing neck range of motion and perform a Spurling’s maneuver (external rotation and extension of the neck in each direction) to screen for cervical spine pathology. Next, active forward elevation and abduction mobility of the shoulder joint are observed in the shirtless patient from behind, paying close attention to scapular symmetry, mobility, and muscle atrophy compared with the opposite shoulder. Next, with the patient facing the examiner, a neurologic evaluation with strength testing is performed. Deltoid (abduction), biceps, triceps, and wrist/hand strength are tested. The supraspinatus is tested in 90 degrees of forward elevation in the plane of the scapula with the shoulder in internal rotation (“empty can” test). Passive and active external rotation with the arm at the side and in 90 degrees of abduction is then tested. A side-to-side increase in external rotation can indicate subscapularis insufficiency. A voluntary decrease in passive shoulder mobility or apprehension in certain positions can be detected in the chronic dislocator. Patients with hypermobility of the shoulders should be examined for generalized joint laxity by evaluating thumb hyperabduction. External rotation and internal rotation strength with the arm at the side is tested, and the subscapularis is tested with both a lift-off test and the belly press test. The strength of all the muscles examined is compared with the opposite site and graded in standard 0-to-5 fashion. After documenting bilateral shoulder range of motion as well as muscular symmetry and strength, specific tests for shoulder instability are then assessed. Specific tests for shoulder instability enable the clinician to classify the instability pattern present. It is often advisable to evaluate the opposite shoulder first to gain the patient’s confidence during the examination and for comparison. The sulcus sign (Fig. 17H1-3) is produced by pulling the arm inferiorly with the arm in adduction and should decrease with external rotation. The gap between the undersurface of the acromion and the upper surface of the humeral head is measured. The sulcus sign is graded 1+ when the gap is less than 1 cm, 2+ when the gap is between 1 and 2 cm, and 3+ when the gap is more than 2 cm. The patient is best evaluated for anterior and posterior instability in the supine position, with the arm and shoulder on the border of the table. The anterior apprehension test is performed with the arm in 90 degrees of abduction. The arm and shoulder are externally rotated, and the patient’s apprehension (i.e., the anxiety and muscle defense against the motion) is observed. A positive response is a feeling that the shoulder is about to dislocate. However, occasionally, this maneuver elicits only pain, especially in a patient with chronic, painful subluxations. The relocation test (Jobe’s test) (Fig. 17H1-4) is performed by placing a posteriorly directed force on the humeral head while abducting and externally rotating the arm. A positive test occurs if the pain and the apprehension produced with the apprehension maneuver are decreased, and the patient usually then can increase the external rotation of the arm. The axial load test, or load

Shoulder

Figure 17H1-3   Sulcus sign. Longitudinal traction on the adducted arm creates a space between the humeral head and acromion. Grading is 1+ when the gap is less than 1 cm, 2+ when the gap is between 1 and 2 cm, and 3+ when the gap is more than 2 cm. (From Altchek DW, Warren RF, Skyhar MJ, Ortiz G: T-plasty modification of the Bankart procedure for multidirectional instability of the anterior and inferior types. J Bone Joint Surg Am 73(1):105-112, 1991.)

915

Figure 17H1-4    Relocation test. Glenohumeral apprehension improves with posteriorly directed force placed on the humeral head by the examiner, stabilizing the humeral head and often allowing increased external rotation.

and can indicate posterior instability, acromioclavicular joint ­pathology, or SLAP tears.

Imaging Tests Radiography

and shift test, is conducted with the patient supine and the arm in 90 degrees of abduction and neutral rotation (Fig. 17H1-5). While one hand applies an axial load, crepitation and humeral head translation are noted when the other hand applies anterior and posterior stress to the joint. Grading reflects the degree of humeral head translation anterior and posterior to the glenoid rim. The grade is 1+ if the translation of the humeral head is to the edge of the glenoid, 2+ if the humeral head can be subluxated over the glenoid rim but reduces spontaneously, and 3+ if a frank dislocation of the humeral head over the glenoid rim does not reduce spontaneously.29 The axial load test must be performed carefully because there is a risk for ­dislocating the patient’s shoulder. Often the maneuver produces pain and guarding when the force is applied toward the direction of instability. The grading system is more applicable during examinations under anesthesia because muscle guarding can limit its usefulness in the office setting. We routinely perform an O’Brien active compression test to evaluate for a painful superior labrum.30 This test was designed to detect superior labral anteroposterior (SLAP) lesions and is performed in two parts. The arm is internally rotated, forward flexed, and adducted 15 degrees. A downward force is then applied by the examiner with the patient’s hand pronated. A positive test results in deep shoulder pain that improves when the downward force is applied with the hand in supination. Impingement tests are performed (Hawkins and Neer) as well as cross-arm adduction. Pain with cross-arm adduction is nonspecific

Radiographs are the first-line imaging study to obtain in an acute dislocation or in a patient who complains of chronic instability. Standard anteroposterior (AP), scapular Y, and axillary lateral views are obtained in most cases. The AP view is deviated 30 to 45 degrees from the sagittal plane of the body to parallel the plane of the glenohumeral joint.

Figure 17H1-5   Load and shift test. One hand applies an axial load while the second hand directs the humerus anteriorly or posteriorly.

916

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A trans-scapular (Y) lateral view and an axillary view are critical to evaluate the position of the humeral head with respect to the glenoid. A shoulder cannot be deemed relocated until it is documented on an axillary radiograph. The AP radiograph is not reliable for detecting shoulder dislocations, especially posterior dislocations. In the assessment of more chronic shoulder instability, additional views are helpful in determining bony anatomy and pathologic changes.31 The West Point view is useful for evaluating the glenoid rim.32 A bony Bankart fracture or ectopic bone production at the anterior glenoid rim can be detected. The West Point view is obtained by placing the patient prone with the arm in 90 degrees of abduction and neutral rotation. The cassette is placed at the superior aspect of the shoulder. The beam is directed from the inferior and is projected cephalad at an angle of 25 degrees from the horizontal and medially at an angle of 25 degrees. The Stryker notch view is useful in demonstrating a Hill-Sachs lesion. In the Stryker notch view, the patient is supine, and the palm of the hand is placed on top of the head. The shoulder is in 90 degrees of forward flexion and neutral abduction to expose the humeral head. The cassette is placed posterior to the shoulder.33 The Stryker notch view is also useful in revealing the Bennett lesion of the shoulder,34,35 which is a crescent-shaped region of mineralization at the posteroinferior rim of the glenoid at the insertion of the posterior capsule.

Three-Dimensional Imaging MRI and computed tomography (CT) have become indispensable tools in the evaluation of patients with shoulder instability. Although a three-dimensional imaging study is not indicated in every case, MRI and CT are often used for presurgical planning. Because image quality has increased dramatically, it is now possible to critically evaluate the

glenoid labrum and ligament-capsule complex with greater than 90% accuracy using MRI.36,37 In addition, MRI provides critical information about concomitant injuries to cartilage, bone, and rotator cuff. CT is used to evaluate the skeletal complications of instability in cases in which a bony Bankart lesion, glenoid erosion, or an engaging Hill-Sachs lesion is suspected. CT can provide critical preoperative information if the surgeon suspects that bone grafting of a large humeral or glenoid bone deficiency may be required.38

TREATMENT Treatment algorithms for acute and chronic dislocations are presented in Figures 17H1-6 and 17H1-7.

Reduction Acute dislocations of the shoulder are reduced as gently and expeditiously as possible to minimize damage to the surrounding structures. Trauma series radiographs of the shoulder are taken before and after the reduction to rule out fractures of the glenoid or the humeral head. Early relocation of the joint can sometimes be performed on the field without sedation if the reduction can be accomplished before muscle spasm occurs. Recurrent dislocations are often more easily reduced and may require less sedation. At the other extreme, posterior dislocations may need general anesthesia to secure a gentle and safe closed reduction. Although more than 20 techniques have been described, there are no studies demonstrating superiority of one technique of reduction over another.39 In the emergency room, we reduce shoulder dislocations using conscious ­intravenous sedation or after an intra-articular injection of lidocaine, depending on patient preference. After adequate anesthesia and a routine series of radiographs to rule out

Figure 17H1-6   Treatment algorithm for acute shoulder dislocation.

Acute Dislocation

Reduction and Radiographs +/– MRI

Traumatic

Young Patient

High Risk

Early Arthroscopic Stabilization

Low Risk

Immobilization/ Rehabilitation

Atraumatic

Older Patient

Immobilization Rehabilitation

Rotator Cuff Tear?

Yes

No

Early Surgery

Immobilization/ Rehabilitation

Shoulder

917

Chronic Dislocation/Subluxation

Radiographs/MRI Rehabilitation Physical Therapy If Rehabilitation Fails:

Diagnosis

Surgery

Anterior Bankart +/–Capsular laxity

Arthroscopic stabilization

Posterior Bankart +/–Capsular laxity

Arthroscopic stabilization

MDI 2+

Arthroscopic stabilization

MDI 3+

Consider open stabilization

Anterior or Posterior Instability With: History of failed stabilization HAGL

Consider open stabilization

Figure 17H1-8   Traction-countertraction method of reduction for acute anterior shoulder dislocation. (From Matsen FA, Titelman RM, Lippitt SB, et al: Glenohumeral instability. In Rockwood CA, Matsen FA, Wirth MA, Lippitt SB: The Shoulder. Philadelphia, Saunders, 2004, p 708.)

Open stabilization

Bone loss

Open stabilization +/– bone gratting or coracoid transfer

Capsular deficiency

Open stabilization +/– augmentation of capsule

Subscapularis rupture

Open stabilization, repair of subscapularis v. Pectoralis transfer

Figure 17H1-7   Treatment algorithm for recurrent instability chronic dislocation/subluxation. HAGL, humeral avulsion of the glenohumeral ligament.

fracture, the patient is placed in the supine position with a sheet wrapped around the ipsilateral side of the chest for an assistant to use for countertraction (Fig. 17H1-8). The elbow is flexed, and a second sheet is placed on the forearm to use for traction on the arm. Gentle traction is applied in the inferior direction on a slightly abducted arm while countertraction is applied by the assistant. This maneuver pulls the humeral head laterally, allowing it to disengage from the inferior edge of the glenoid. When the humeral head is free of the inferior glenoid edge, short arcs of gentle internal and external traction can facilitate the palpable reduction. It is important to note that this is not a leverage maneuver. Shoulder reduction is best accomplished with a slowly applied constant force, allowing muscle spasm to

relax and minimizing the risk for fracture. We also have had very good results using scapular manipulation in the prone position in thin patients. Two recent studies have demonstrated that intra­articular glenohumeral lidocaine is an effective method of anesthesia for the reduction of acute shoulder dislocations. In a randomized, controlled trial of 30 acute shoulder dislocations, both time in the emergency room (75 versus 185 minutes) and cost ($0.52 versus $97.64) were reduced when the intra-articular lidocaine was used instead of intravenous sedation.40 Another randomized trial demonstrated that intra-articular lidocaine was inferior to intravenous sedation for relieving prereduction pain in acute shoulder dislocations, but it did provide equivalent overall pain relief.41 Intra-articular lidocaine appears to be a safe and effective method of anesthesia for reduction of shoulder dislocations in the willing patient or if sedation is contraindicated medically. If a gentle attempt at closed reduction fails or if the patient presents with a chronic dislocation, open reduction is the safest method of relocating the joint and avoiding fracture.

Nonoperative Treatment and Rehabilitation The optimal nonoperative treatment for shoulder instability is not known. Despite a plethora of studies evaluating the results of surgical treatment of anterior instability and many studies comparing operative to nonoperative intervention, the literature is practically devoid of solid research

918

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

critically evaluating such important topics as position and duration of immobilization after a shoulder dislocation. Rowe published a classic study in 1956 evaluating the effect of length of immobilization on recurrence rate of shoulder dislocation. Six groups of patients were immobilized for 1 to 6 weeks at 1-week increments between groups.42 There was no difference in recurrence rates between groups. Kiviluoto and colleagues evaluated the effect of duration of immobilization on 226 shoulder dislocations followed for 1 year prospectively.43 They concluded that immobilization of 1 week was sufficient in patients older than 50 years and that younger patients should be immobilized for 3 weeks. Hovelius and colleagues have published extensively on the natural history of anterior shoulder dislocation in young patients.44-47 They divided patients into three groups: those with immobilization for 3 to 4 weeks in a sling or Velpeau dressing with the arm in internal rotation, those who wore a sling for comfort (removed when tolerated), and a mixed-treatment group who were noncompliant with immobilization. Although this was not a randomized trial, no difference in recurrence rate between the groups was observed at 5 years.46 In a total of 257 patients, the overall recurrence rate was 55% in patients who were younger than 22 years, 37% of patients who were 23 to 29 years, and 12% of patients who were between 30 and 40 years old. Hovelius concluded that the initial treatment did not affect the rate of recurrence. A systematic review of conservative management following acute traumatic shoulder dislocations by the Cochrane Database concluded that there was really no evidence from randomized controlled trials to guide nonoperative management after a shoulder dislocation.48 Because there is no consensus on a standard of care for treatment after shoulder dislocation, the duration of immobilization and rehabilitation protocols used vary widely and are not well documented.49 Three recent studies have renewed interest in the nonoperative management of shoulder dislocations; specifically, sling immobilization in internal rotation after dislocation has been called into question. First, Miller and associates

A

created Bankart lesions in cadavers and evaluated the contact forces between the glenoid labrum and the glenoid in 60 degrees of internal rotation, neutral, and 45 degrees of external rotation.50 They found no detectable contact force between the labrum and glenoid with the arm in internal rotation, but 83.5 g of contact force with the arm in external rotation. Second, Itoi and coworkers evaluated the coaptation of a Bankart lesion on the anterior glenoid using MRI in internal rotation and in external rotation.51 They found that external rotation of the arm tightened the anterior structures (subscapularis and anterior capsule) and prevented medial displacement of the labrum on the glenoid on MRI (Fig. 17H1-9). Third, Itoi and coworkers conducted a study comparing the recurrence rates of 40 patients who were immobilized in external rotation (10 degrees) versus those who were immobilized in internal rotation after an acute dislocation.52 The study was partially randomized and included first-time dislocators only. All patients were immobilized for 3 weeks. Average age was about 40 years. Recurrent dislocation rate in the internal rotation group was 6 of 20 for the internal rotation group and 0 of 20 in the external rotation group. Compliance was only 75% to 80%. Early evidence suggests that recurrence rate of shoulder dislocation could possibly be decreased with immobilization. Clearly, this topic deserves further study, and a larger randomized controlled study could validate these results. However, two main problems exist when conducting research into acute shoulder dislocations. First, rarely is the patient seen initially by an orthopaedic surgeon to perform the reduction and advise the patient to wear an external rotation sling. Second, external rotation after acute dislocation is poorly tolerated. Although the maximal contact of labrum on glenoid is observed with external rotation at 45 degrees, in Itoi and coworkers’ clinical study, external rotation beyond 10 degrees was poorly tolerated, and noncompliance with the external rotation sling was 20%. At our institution, it is more common for a patient to be referred several weeks after dislocation. Therefore, we

B

Figure 17H1-9   Axial shoulder magnetic resonance imaging after acute dislocation with the shoulder immobilized in internal   (A) and external (B) rotation. Notice apposition of labrum (arrow) on glenoid in external rotation in B. (From Itoi E, Sashi R, Minagawa H, et al: Position of immobilization after dislocation of the glenohumeral joint: A study with use of magnetic resonance imaging. J Bone Joint Surg Am 83(5):661-667, 2001.)

Shoulder

rarely are able to recommend external rotation bracing. When a patient presents with an acute first-time dislocation, certain patients are candidates for early reconstruction.53 However, most patients are prescribed a period of immobilization followed by extensive rehabilitation. We continue immobilization for 3 weeks, with the patient removing the sling only for pendulum exercises and passive forward elevation in the plane of the scapula. Muscle rehabilitation can be started early with isometric contractions. After 3 weeks, the sling is discontinued, and progressive active assisted range of motion begins. Abduction and external rotation are avoided for a full 6 weeks after ­dislocation.54 The emphasis of the rehabilitation program is on regaining safe and protected full-range motion, starting with passive range of motion and pendulum exercises. Full range of motion can be expected by week 8 and strengthening begins between 9 to 12 weeks. Strengthening of the rotator cuff is essential to restore and increase the dynamic stabilization of the joint, as is strengthening of the scapular stabilizing muscles to position the glenoid toward the humeral head. The goal of rehabilitation is to restore function in the shoulder to a level that allows the patient to return safely to previous activity. Each rehabilitation program should include a sport- or activity-specific test that proves the patient’s ability to return to the sport or activity. If the patient’s rehabilitation fails to progress or the patient is unable to return to previous activities, an MRI of the shoulder should be considered.

Operative Treatment of the Unstable Shoulder Although there are relative indications for operative treatment of the first-time dislocator, the main indication for operative intervention in anterior shoulder instability is recurrent dislocation or subluxation despite an adequate course of rehabilitation. The choice between open and arthroscopic reconstruction remains the choice of the surgeon. Although we have evolved to perform mostly arthroscopic reconstructions for routine anterior stabilizations, we still consider a well-done open anterior stabilization and capsular shift to be the gold standard to which we compare our arthroscopic results. In the following portion of this section, we present a literature review and the authors’ preferred technique of open and arthroscopic anterior and posterior stabilizations. In addition, we review the evidence directly comparing open versus arthroscopic stabilization. The indications for and results of operative stabilization of first-time dislocators is reviewed. Treatment of MDI is also reviewed.

Arthroscopic Anterior Stabilization Many arthroscopic shoulder specialists agree that arthro­ scopic anterior stabilization has become an accepted treatment for recurrent glenohumeral instability. This was not always the case because early techniques did not yield the reliable results usually noted with an open repair. Early arthroscopic repairs reported higher failure rates than an open repair. Initial strategies for treating anterior instability by arthroscopy used staples or transglenoid sutures.55-61 The initial studies using arthroscopic staple repair had a

919

high incidence of recurrence and hardware migration. Transglenoid suturing of the Bankart lesion showed some success, but fell out of favor because of variable recurrence rate, technical difficulty, and the risk for injuring the suprascapular nerve by tying sutures over the posterior fascia.62 Bioabsorbable tacks were also used with some success (recurrence rate, 10% to 20%),63 especially in patients with isolated Bankart lesions with no capsular laxity.64 Un­fortunately, a synovitis could develop if this device loosened with time,65,66 and capsular redundancy often associated with recurrent dislocation was difficult to treat with this device.63,67 Since first described in the mid 1990s,68,69 suture anchor repair has become the most common method of arthroscopic repair of a Bankart lesion in anterior instability. Published recurrence rates using suture anchors for arthroscopic Bankart repair vary from 0% to 33%.62 Hoffman and Reif reported good results using Mitek suture anchors with recurrence in 2 of 26 dislocators at 2 year follow-up.70 Gartsman and associates reported good to excellent outcome scores in 53 patients treated arthroscopically with suture anchors for anterior instability at a mean ­follow-up of 33 months. Four of 53 patients had recurrent instability. The authors attributed this low recurrence rate to repair of concomitant superior and inferior labral tears and closure of the rotator interval in 14 patients.71 Bacilla and coworkers reported on the results of arthroscopic stabilization using suture anchors in 40 high-demand athletes at an average of 30 months’ follow-up.72 Thirty-seven of 40 patients remained stable and returned to sports; three patients required revision stabilization. Tan and associates conducted a randomized controlled trial showing no difference between stabilizations conducted arthroscopically using metal suture anchors versus absorbable suture anchors in 124 patients.73 They reported an overall redislocation rate of 6% and an 85% return to previous level of sporting activity. Several authors have reported good results using arthro­ scopic shoulder stabilization in contact athletes.74 Cole and Romeo reported no recurrent dislocations in a series of 45 patients with 96% good to excellent shoulder scores.75 Mazzocca and associates reported on the results of 18 collision and contact athletes younger than 20 years treated with arthroscopic repair of a Bankart lesion using suture anchors.76 All patients returned to sport, and 2 patients had recurrent dislocations, leading the authors to conclude that participation in a collision sport is not a contraindication for arthroscopic anterior stabilization. The authors concluded that arthroscopic repair was equivalent to open repair in their hands. Kim and colleagues reported 95% good to excellent Rowe scores after arthroscopic ­ stabilization in 167 patients with traumatic recurrent instability of the shoulder.77 Overall recurrence rate was 4% and was related to glenoid bone deficiency. A prospective study by Carreira and coworkers evaluated the results of a consecutive series of 85 patients with Bankart lesions treated arthroscopically with suture anchors loaded with nonabsorbable sutures at a mean of 46 months.78 The authors reported an overall recurrence rate of 10% after repair and 90% good to excellent results on subjective outcome measures. The authors reported recurrent dislocation in 2 of 18 collision athletes at 22 and 60 months postoperatively after return to sport. The authors noted several key points as critical to their ­

920

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

success at arthroscopic Bankart repair: (1) use of a low anterior portal (5-o’clock), (2) repair of tear extension into the superior labrum, and (3) placement of the suture anchors 2 mm on the glenoid face articular cartilage. Most current literature suggests that arthroscopic anterior ­stabilization

A u t h o r s ’ P r e f e r r e d M e t h o d A n t e r i o r S t a b i l i z a t i o n

f o r A r t h r o s c o p i c

Based on recent results at our institution and recent results published using suture anchors for arthroscopic anterior stabilizations, we routinely use an arthroscopic approach for Bankart repair. The goal of an arthroscopic anterior stabilization or Bankart repair is to restore the glenoid labrum and attached capsule/ligaments back to the anatomic position on the face of the anterior inferior glenoid and reduce capsular redundancy associated with recurrent dislocation. A preoperative MRI, or MR arthrogram, although not a necessity, is commonly obtained to confirm the expected pathology and avoid unexpected pathology in the operating room (Fig. 17H1-10). Regional anesthesia with an interscalene block and sedation is administered. The patient is placed in the beach chair position. Arthroscopic landmarks are palpated and marked on the skin of the shoulder: acromion, the acromioclavicular joint, clavicle, and coracoid. Supplementary local anesthesia in the area of the posterior portal is injected. An examination under anesthesia is performed to determine the pattern of instability that needs to be addressed with surgery. We find that the examination in the operating room is more specific in determining both the degree and direction of instability. The axial load test, or load-and-shift test, is conducted, and translation is noted in the anterior, inferior, and posterior ­ directions. Grading reflects the degree of humeral head translation anterior and posterior to the glenoid rim. The grade is 1+ if the translation of the humeral head is to the edge of the glenoid, 2+ if the humeral head can be subluxated over the glenoid rim but reduces spontaneously, and 3+ if a frank dislocation of the humeral head over the glenoid rim does not reduce

A

can be performed with a 10% recurrence rate in most patients, including contact athletes. The controversies regarding the efficacy of arthroscopic versus open stabilization and the indications for surgical stabilization in the first-time dislocator are covered later in this chapter.

spontaneously.29 The arm is immobilized in the ­McConnell arm holder (McConnell Orthopedic Manufacturing Company, Greenville, Tex). The arthroscope is introduced through the posterior portal 2 cm below and 1 cm medial to the posterolateral corner of the acromion. The anterior working portal 1 cm lateral to the coracoid is localized using a spinal needle and established with a 5.5-mm cannula. Diagnostic arthroscopy then confirms the presence of a Bankart lesion (Fig. 17H1-11). It is important to rule out associated pathology, including rotator cuff tears, anterosuperior labral tears, and posterior labral tears. After confirming that there is no bone loss on the ­glenoid or an engaging Hill-Sachs lesion to prevent successful arthroscopic repair, a second anteroinferior portal is placed under direct vision using a spinal needle. The angle of ­ approach of this portal is critical: before inserting an 8.25-mm clear, twist-in cannula, we use the spinal needle to be certain that the direction afforded by the portal is both inferior enough (to reach the 5:30 position on the glenoid with an anchor) and lateral enough (to place anchors on the face of the glenoid rather than the glenoid neck). In addition, ­ensuring spread of the two anterior portal skin incisions (2 to 3 cm) will facilitate instrument and suture management. ­After ­establishing this second anterior portal, the labrum and capsule are ­mobilized from the glenoid using a rasp. In an acute dislocation, the tissue often exhibits minimal retraction, but in a chronic dislocator, the capsulolabral complex can be scarred down medially and inferiorly on the glenoid neck. In this situation, we use an electrofrequency probe to aid in mobilization

B

Figure 17H1-10   Preoperative magnetic resonance imaging in a 21-year-old contact athlete with an acute shoulder dislocation reveals detachment of the anterior labrum (Bankart lesion) (A) and Hill-Sachs defect (B, arrow).

Shoulder

921

A u t h o r s ’ P r e f e r r e d M e t h o d f o r A r t h r o s c o p i c A n t e r i o r S t a b i l i z a t i o n — cont ’ d

A

B

Figure 17H1-11   Arthroscopic probe under torn anterior labrum of a Bankart lesion in a 20-year-old with recurrent anterior instability. Note the medialized position of the labrum (A) before mobilization. B, The labrum rests in a reduced anatomic position after thorough mobilization with electrocautery and an arthroscopic elevator.

A

C

B

Figure 17H1-12   Insertion of suture anchor in glenoid. A, Placement of drill guide slightly (1-2 mm) on the glenoid face avoids a medial reconstruction. B, Seating the drill bit to the predetermined level on the glenoid face. C, Insertion of bioabsorbable suture anchor preloaded with No. 2 nonabsorbable suture through the anterior inferior portal. Continued

922

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A u t h o r s ’ P r e f e r r e d M e t h o d f o r A r t h r o s c o p i c A n t e r i o r S t a b i l i z a t i o n — cont ’ d ­ efore repair. This step is critical for arthroscopic stabilib zation; the glenoid labrum and capsule complex should be elevated to the 6-o’clock position to ensure restoration of the tension of these structures and closure of the axillary pouch. A probe is used to assess the mobility of tissue before placement of any anchors. Another indicator that the capsule and labrum have been mobilized sufficiently is visualization of the subscapularis muscle under the ­ mobilized capsule and labrum. After satisfactory mobilization of the labral and capsular tissue (see Fig. 17H1-11), the labrum should float up to its nearly anatomic position. Next, a motorized shaver is used to gently decorticate the glenoid to prepare a bleeding bed to aid in healing. This maneuver should be performed with the shaver suction off to avoid damaging the soft ­tissues. Three to four suture anchors are then placed and used to ­secure the glenoid and labrum in the following fashion. The first anchor, ideally at the 5:30 position on the glenoid, is drilled, inserted, and tied first. The other two or three anchors are evenly spaced more superiorly to the 2-o’clock or 3-o’clock position on the glenoid. Care is taken to ­ensure that the suture anchors are placed on the glenoid face, ­centered

A

C

about 2 mm from the edge of the glenoid cartilage (Fig. 17H1-12). Anchors placed more medial than this risk restoration of the glenoid labrum in a nonanatomic medialized position that will not restore the normal “bumper” effect of the anterior labrum. We use bioabsorbable suture anchors that are singly loaded with nonabsorbable No. 2 suture. Figures 17H1-13 and 17H1-14 demonstrate placement of anchors, passing of suture, and knot tying in arthroscopic anterior stabilization. It is critical for the success of an arthro­ scopic repair to pull the inferior capsule superiorly with the stitches. Occasionally, we will use a traction stitch placed in the inferior pouch to pull superior traction while passing the curved suture hook through the capsule and labrum. After completing an arthroscopic Bankart repair, we ­assess anterior stability and occasionally add capsular plication or rotator interval closure if necessary. Key portions of a successful arthroscopic repair include the following: (1) thorough diagnostic arthroscopy to identify associated pathology that needs concomitant treatment; (2) mobilization of the capsular and labral tissue using a rasp or an electrofrequency probe, or both, and performing a trial reduction of the tissue

B

Figure 17H1-13   A, A curved, sharp suture passing device is used to pierce the capsule and labrum lower than the anchor position, ensuring that the capsule and labrum will be shifted up to an anatomic position. The amount of capsule secured in each stitch with the labrum is based on the degree of laxity, the capacity of the inferior pouch, and the severity of the drive-through sign. B, Passing of a PDS stitch into the joint to use as a suture shuttle. C, Position of nonabsorbable suture after shuttling through capsulolabral tissue, ready to tie. The limb on the right should be used as the post.

Shoulder

923

A u t h o r s ’ P r e f e r r e d M e t h o d f o r A r t h r o s c o p i c A n t e r i o r S t a b i l i z a t i o n — cont ’ d to the glenoid to ensure that the mobilization is adequate; (3) a well-placed anteroinferior portal that is low and ­lateral enough to obtain proper angle on the 5:30 anchor (this can be ensured by assessing the angle of entry using a spinal needle before inserting a large cannula; anchors should be placed on the edge of the glenoid face, not on the glenoid

A

rim or neck); (4) grasping capsule and labral tissue lower than each anchor with the suture passing device to ensure an adequate capsular shift and restoration of labral anatomy; and (5) tying knots on the capsular side, which prevents mechanical abrasion of the knots on the glenohumeral cartilage and restores the bumper effect of the labrum.

B

Figure 17H1-14   A, A sliding arthroscopic knot is used to reduce the tissue to the anchor, taking care to place the knot as far from the articular surface as possible. We believe that knot position (capsular), quality of tissue contained in the “bite,” and loop security are far more critical than the type of sliding knot used. B, A completed repair with four anchors at 2:30, 3:30, 4:30, and 5:30 (the most inferior anchor is commonly not visualized at the end of the procedure because of tightening of the inferior pouch and elimination of the drive-through sign). Note the position of the knots on the capsular side of the repair and the restoration of an anterior bumper effect by the labrum.

Open Surgery for Anterior Stabilization Modern open anterior stabilization usually consists of a combination of a Bankart repair79 to restore the labral anatomy and a capsular shift to address capsular laxity. Addressing the capsular laxity is critical for this procedure because Speer and colleagues demonstrated that a simulated Bankart lesion in a cadaver model was not enough to create an anterior dislocation.22 Open anterior stabilization is often described as the gold standard operation for anterior instability. Rowe and associates described a long-term 2% redislocation rate in a series of 162 patients after open anterior stabilization between 1946 and 1976; however, only 69% of patients regained full range of motion.80 Recent studies have also shown good success rates with less range of motion loss. Gill and colleagues reviewed the 12-year results of 60 shoulders treated with open anterior stabilization and found only three recurrent dislocations.81 External rotation loss averaged 12 degrees. Pagnani and Dome reviewed their results when treating 58 American football players (average age, 18.2 years) with open anterior stabilization and found no recurrent dislocations in this high-risk group at a minimum of 2 years’ follow-up. There were two patients who continued to have anterior subluxations after stabilization. Range of motion loss was observed in 9 of 58 patients.82

Indications in our clinic for a primary open anterior stabilization include humeral avulsion of the capsule and humeral or glenoid bone loss that precludes an arthroscopic repair. We become concerned with the success rate of an arthroscopic stabilization when glenoid bone loss is greater than 25% on a preoperative computed tomographic scan. Second, certain athletes (e.g., wrestlers) anecdotally are at high risk for recurrence and are offered an open stabilization. Revision anterior stabilizations are often successful arthroscopically; however, revision of failed stabilizations are one of the most common indications for open surgery. If anatomy is relatively normal at the time of diagnostic arthroscopy in a revision situation, we perform a revision arthroscopic stabilization. However, if there is an engaging Hill-Sachs defect or glenoid bone loss83 or any question of tissue quality, capsular deficiency, or subscapularis rupture due to previous surgery (e.g., thermal capsulorrhaphy), the procedure is converted to open. In addition, if a patient has atraumatically failed a well-done arthroscopic anterior stabilization on one shoulder, a primary open stabilization is often recommended to stabilize a contralateral shoulder. An excellent recent review of indications for open anterior stabilization was published by Millett and colleagues in 2005.84

924

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A u t h o r s ’ P r e f e r r e d M e t h o d

for

O p e n A n t e r i o r S t a b i l i z a t i o n

Open anterior stabilization can be performed with either general anesthesia or regional interscalene block anesthesia and supplementary local anesthesia in the axillary ­region. The patient is placed in the beach chair position with 40 degrees of elevation. Examination under anesthesia is performed as described previously. A McConnell arm holder is useful for positioning the arm during the procedure. The skin incision is marked from just lateral to the coracoid process following the deltopectoral fold distally lateral to the axilla near the deltoid insertion. Administration of dilute epinephrine solution into the skin and subcutaneous tissue before incision decreases bleeding. Sharp dissection is carried down to the deltopectoral fascia. The location of the cephalic vein and deltopectoral interval is usually identified by a fat stripe. The cephalic vein is preserved and taken medially or laterally, depending on the preference of the surgeon. The deltopectoral interval is bluntly dissected, and the clavipectoral fascia is incised just lateral to the conjoined tendon coming off the coracoid. A self-retaining shoulder retractor (Tiemann) is then inserted under the deltoid laterally and under either the pectoralis major or conjoined tendon medially. Care should be taken not to place too much tension on the blades underneath the conjoined tendon because of the risk for damage to the musculocutaneous nerve that runs 5 cm below the coracoid medially and under the conjoined tendons. With external rotation, the subscapularis tendon is exposed and the anterior humeral circumflex vessels are identified, which help define the lower border of the tendon. The subscapularis tendon blends laterally with the capsule before it inserts on the lesser tuberosity. The subscapularis tendon is then released 1 cm from its lateral insertion and tagged with No. 2 nonabsorbable suture, leaving a stump of tendon for later repair. A subscapularis-splitting approach can also be used, but the visualization and amount of capsular shift can be limited with this approach. Often a patient who can be treated with a subscapularis split open stabilization could also be treated with an arthroscopic repair. After dividing

Posterior Instability When compared with anterior instability, posterior instability is a rare condition (making up 2% to 4% of dislocations) and is less well studied.85 Acute posterior shoulder dislocations are usually caused by high-energy injuries and can be associated with electrical shock and seizure.86 Another subset of patients who can present with posterior instability are athletes who undergo repeated load to a flexed, adducted, and internally rotated arm (e.g., football linemen).87,88 Athletes often present with recurrent subluxation and pain rather than acute posterior dislocation. Mair and associates described posterior labral detachment in eight American football linemen and one lacrosse player and hypothesized that the posterior loading of the shoulder joint creates a shear force at the posterior labrum.89 Acute posterior dislocations are notoriously missed on plain AP radiographs, necessitating an axillary view to document the diagnosis or confirm reduction. Acute reduction is accomplished with traction, external rotation, and gentle ­abduction.

the tendon insertion, the tendon is then separated from the underlying capsule with blunt or sharp dissection; the plane between capsule and subscapularis becomes more defined medially. It is often necessary to cauterize or ligate the three circumflex vessels below the subscapularis tendon. The capsular incision used depends on the amount of shift deemed necessary. In a lateral capsular shift, the capsule is incised along its lateral insertion on the humeral head. If a smaller shift is necessary, a horizontal capsular incision is made for later imbrication of one leaf over the other. Placement of the arm in abduction and external rotation while the capsule is being incised distally allows the most inferior part of the posterior capsule to be exposed. The axillary nerve should be protected during the inferior and posterior capsular incision. After incising the capsule, the Bankart lesion can be visualized and repaired with suture anchors along the glenoid edge. The rim of the glenoid is débrided of scar tissue, and the capsule and labrum are released to be brought up to the edge. It is important to obtain fresh bleeding bone along the edge of the glenoid to enhance healing between bone and capsule. Two to four suture anchors, depending on the size of the lesion, are inserted along the glenoid rim at the margin of the articular surface in a 45-degree angle. The sutures are brought through the capsulolabral complex in a mattress fashion and tied. Three or four suture anchors are then needed to shift the capsule back to its insertion along the humeral head, starting inferior to secure the most inferior limb of the capsule. The most superior anchor is placed just below the rotator interval and the biceps tendon. The inferior limb of the capsule is pulled superiorly and laterally to tighten it, and sutures are brought through the capsule tissue to create the desired tension. The superior limb of the capsule is then sutured on top of the inferior limb to close the defect. To ensure that the shift is not too tight, the arm is held in a 45-degree abducted, 45-degree externally rotated position while the sutures for the capsular shift are tied. ­Finally, if the rotator interval is lax, it is closed with the arm in adduction and external rotation.

Initial treatment of posterior instability is with physical therapy.90 As with anterior instability, physical therapy is more likely to be successful in the absence of a large reverse Bankart lesion or bony deficits. Fronek and ­coworkers described a 63% success rate of an exercise program in patients with posterior instability and moderately disabling symptoms.87 The success rate was higher in those with posterior instability and a nontraumatic cause. In that study, 10 of 16 patients in the exercise group were able to avoid surgery. In patients who failed physical therapy or had severe symptoms, the success rate of open posterior stabilization was 91%. When physical therapy fails to control recurrent posterior dislocation or subluxation, surgical treatment has been successful in controlling symptoms and preventing recurrence in most patients. Arthroscopic stabilization is often preferred for uncomplicated posterior Bankart lesions because the open posterior approach to the shoulder is associated with some morbidity. However, the role for arthroscopic stabilization (compared with open

Shoulder

stabilization) in shoulders in which the primary problem is posterior capsular laxity is less clear.91 The results of arthroscopic stabilization of posterior instability has been reported by several groups. Williams and colleagues evaluated the results of arthro­ scopic posterior surgical stabilization using a bioabsorbable tack in 27 shoulders at a mean of 5.1 years’ follow-up.92 There was no range of motion loss observed, and 92% had a good to excellent result. There were two ­reoperations. Kim and colleagues also reported on a retrospective series of 27 arthroscopic posterior stabilizations with suture anchors and a capsular shift at an average follow-up of about 3 years.93 There was one episode of recurrent instability, and 26 patients returned to previous athletic activities. Provencher and associates treated 33 consecutive patients with posterior instability using arthroscopic placement of suture anchors or capsular plication.94 Four patients failed because of recurrent instability (12.1%), and an additional 3 patients had persistent pain. These authors concluded that previous surgery was a risk factor for failure, especially thermal capsulorrhaphy. In addition, a voluntary component to the instability was a poor prognostic sign. Antoniou and coworkers treated 41 patients with arthroscopic posterior capsular shift, including nine revision procedures.95 They reported that 35 patients had stable shoulders postoperatively and that the average simple shoulder test score improved significantly. Workers’ compensation was a risk factor for failure. Arthroscopic posterior stabilization has been successful for returning athletes to sport in most cases. Bradley and colleagues recently reported a prospective evaluation of 91 athletes with

A u t h o r s ’ P r e f e r r e d M e t h o d P o s t e r i o r ­S t a b i l i z a t i o n

­ nidirectional posterior instability who were treated with u arthroscopic capsulolabral reconstruction.96 Eighty-nine percent of athletes returned to sport (albeit 22% at a limited level). This study had excellent documentation of the diverse ­ pathology that can be encountered in posterior instability: 43% of patients had only a patulous posterior capsule without labral tear, which often manifests as increased visualization of the posterior glenoid neck upon initial diagnostic arthroscopy through the posterior portal. Fifty-seven percent of patients had partial or complete posterior labral tears, and 23% of patients had both labral tear and capsular laxity. The arthroscopic findings in posterior instability are therefore less consistent than anterior Bankart lesions. The authors point out that the arthroscopic stabilization of posterior shoulder instability can consist of labral repair with suture anchors, capsular plication, or both, depending on the pathology viewed at arthroscopy. Our institution has previously reported an 81% success rate of open posterior stabilizations,91 which is comparable to the results of other published reports.97,98 With continued improvement in arthroscopic techniques, we prefer to reserve open stabilization for shoulders with significant capsular laxity not amenable to arthroscopic plication (e.g., 3+ posterior and inferior instability on examination under anesthesia or MDI picture), capsular and bony deficiency, or revision situations. Although Fuchs and colleagues have reported good results in surgical treatment of voluntary instability,99 we still consider voluntary posterior instability a relative contraindication to surgery unless a psychiatric component has been completely ruled out.

f o r A r t h r o s c o p i c

We conduct repair of a posterior Bankart lesion in a similar manner to anterior stabilization. Preoperative MRI confirms diagnosis of a posterior Bankart and reverse Hill-Sachs

A

925

(Fig. 17H1-15). Standard anterior and posterior portals are created, and the diagnostic arthroscopy is carried out with the arthroscope in the posterior portal. Posterior capsular

B

Figure 17H1-15   Magnetic resonance arthrogram demonstrating a preoperative posterior Bankart lesion. Continued

926

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A u t h o r s ’ P r e f e r r e d M e t h o d f o r A r t h r o s c o p i c P o s t e r i o r ­S t a b i l i z a t i o n — cont ’ d

Left Shoulder Right Shoulder 12 12 P

A 3

9

6

A 3

9

6

Figure 17H1-16   Position of posterior labral tear on left and right shoulders on the clock face model of the glenoid. Inset, Arthroscopy photos show posterior labral tear before and after arthroscopic repair. (From Williams RJ 3rd, Strickland S, Cohen M, et al: Arthroscopic repair for traumatic posterior shoulder instability. Am J Sports Med 31(2): 203-209, 2003.)

A

Figure 17H1-17   Accessory portal (clear cannula) established inferior and lateral to standard portal (black cannula) for placement of low (6 o’clock) anchors in the posterior glenoid. Note the arthroscope viewing from the anterior portal.

l­axity can manifest as increased visualization of the ­posterior capsule, labrum, and glenoid neck. Upon completion of a satisfactory diagnostic arthroscopy, including inspection for a reverse Hill-Sachs lesion noted at the anterior humeral head, the arthroscope is switched to the anterior portal using a switching stick, and the posterior labrum and capsule are evaluated. Compared with a standard clock face, the posterior capsulolabral complex is usually detached from the glenoid rim at the 6- to 10-o’clock position in affected right shoulders and at the 2- to 6-o’clock position in affected left shoulders (Fig. 17H1-16).92 Upon confirmation that a posterior Bankart repair is necessary, we have found that an accessory portal more ­lateral and inferior is helpful for obtaining the proper ­angle to the

B

Figure 17H1-18   Arthroscopic posterior Bankart repair. A, Posterior Bankart lesion with glenoid impaction viewed from the anterior portal. B, Preparation of glenoid and mobilization of Bankart lesion with a rasp.

Shoulder

927

A u t h o r s ’ P r e f e r r e d M e t h o d f o r A r t h r o s c o p i c P o s t e r i o r ­S t a b i l i z a t i o n — cont ’ d

C

D

Figure 17H1-18, cont’d   C: Adequate mobilization of the posterior labrum. D, Completed posterior Bankart repair.

low, posterior glenoid. This portal is often more ­lateral than is intuitive, and the angle should be confirmed with a spinal needle before insertion of an 8.25-mm screw-in cannula in the proper orientation for placement of suture anchors (Fig. 17H1-17). The preparation of the labrum and capsule, as well as suture passage, is similar to that for ­anterior ­stabilization

Open Posterior Capsular Shift The skin incision extends from 1 cm medial to the posterior corner of the acromion distally toward the axillary fold. The deltoid is split vertically between its middle and posterior thirds. The approach to the joint begins through a horizontal split in the infraspinatus muscle and tendon that will expose the capsule. The capsular shift is performed by opening the capsule along the glenoid rim and shifting the capsule medially and superiorly, using suture anchors as described before. The inferior dissection is critical in this operation, and capsular dissection is sometimes carried anterior to 6 o’clock to allow a sufficient shift. Anchors are also used to repair the labral tear. In the posterior approach to the shoulder, care should be taken to avoid the quadrangular space containing the axillary nerve and circumflex vessels. The space is just below the teres minor muscle and can be jeopardized if the incision and the split in the deltoid are too low.

WEIGHING THE EVIDENCE Treatment of the First-Time Dislocator Because of the high recurrence of anterior instability in the young population (<25 years old), some authorities have recommended operative stabilization for first-time ­dislocators who would like to return to high-risk activities. Several recent studies support operative intervention for

(Fig. 17H1-18). After bony débridement to a bleeding bed of bone on the posterior glenoid rim, we reattach the capsule and labrum to the ­glenoid margin using bioabsorbable suture anchors with No. 2 ­nonabsorbable sutures. Three suture anchors are usually satisfactory for repair. Arthroscopic capsular plication can be added if necessary.

the first-time dislocator in populations at high risk for ­recurrent dislocation. DeBerardino and associates reported the results of surgery on 48 cadets at the U.S. Military Academy who underwent acute arthroscopic stabilization with repair of the Bankart lesion with a bioabsorbable tack after anterior dislocation.100 At about 3 years’ follow-up, the average Rowe score was 92%, and there was a 12% redislocation rate. Strengthening this evidence, Bottoni and colleagues conducted a randomized controlled trial in 21 active-duty military personnel with acute, first-time anterior shoulder dislocations.101 They randomized these patients to receive arthroscopic stabilization with a bioabsorbable tack or nonoperative treatment. At 3 years’ followup, there was an 11% redislocation rate in the surgically treated group and a 75% redislocation rate in the nonoperative group. In addition, six of nine patients with recurrent dislocation in the nonoperative group required open anterior stabilization surgery. Kirkley and colleagues conducted a similar randomized controlled trial of 40 patients in the nonmilitary population and reported the short-term102 and long-term103 (75 months) results. Their conclusions were that there was a significant decrease in the redislocation rate after acute surgical stabilization of the first-time dislocator and a small but clinically significant improvement in this group in one of three shoulder scores when compared with patients treated with rehabilitation alone. In our practice, a young patient with a first-time anterior dislocation with a documented Bankart lesion on MRI is counseled that the recurrence rate with nonoperative

928

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

treatment will likely be higher than 50%, especially if the patient returns to high-risk athletic activities. A detailed discussion is conducted with the patient and family to delineate the patient’s athletic activities and goals as well as the optimal time for operative intervention based on the timing of the patient’s season. We do not discourage a trial of nonoperative therapy. However, we do recommend surgery if there is evidence of recurrent subluxations or a repeat dislocation because further damage can occur to the joint, and capsular tissue can become attenuated. The in-season athlete with an anterior shoulder dislocation can occasionally be managed with bracing. Buss and coworkers reported their results of physical therapy and bracing for 30 in-season athletes with acute or recurrent anterior dislocation.104 They reported an average of 1.4 instability episodes per athlete per season. About half of these patients underwent anterior stabilization after the season was over. The importance of competing must be weighed against the risk for further damage to the shoulder in these cases.

Open versus Arthroscopic Stabilization Although open anterior stabilization has a proven track record, the trend in recent years among many shoulder specialists is to repair uncomplicated Bankart lesions arthroscopically. Although initial studies of arthroscopic stabilization had a high recurrence rate and problems with hardware failure, more recent studies show that arthroscopic techniques have improved. The two main advantages purported by arthroscopic aficionados are decreased surgical morbidity, including avoiding subscapularis takedown, and improved range of motion, especially terminal external rotation. Freedman and associates conducted a meta­analysis of six studies comparing the results of arthroscopic anterior stabilization with transglenoid sutures or bioabsorbable tacks (172 patients) and open repair (156 patients).105 The recurrent dislocation rate was 12.6% in the arthroscopic group and 3.4% in the open group; in addition, the Rowe and Constant scores were significantly higher in the open group. In 1996, Guanche and colleagues compared open (12 patients) and arthroscopic (15 patients) stabilizations in 27 men with anterior dislocations. After a follow-up ­ average of 26 months, the arthroscopic group demonstrated 5 patients with recurrent dislocation or subluxation, and the open group had only 1 patient with recurrent subluxation. The study was not randomized, and decision for open versus arthroscopic stabilization was made by patient preference. Also, 10 of 15 arthroscopic repairs were conducted with transglenoid sutures, and mobilization of the Bankart lesion and capsular plication were not described.106 These results represent a summary of early arthroscopic results versus open repair and led to open repair being described as the gold standard for arthroscopic stabilization. However, the results of current arthroscopic stabilization techniques using suture anchors have improved these results dramatically. Karlsson and colleagues compared the results of 60 arthroscopic anterior stabilizations with 48 open stabilizations in a nonrandomized trial.107 The Rowe and Constant scores were similar at 28 months’ follow-up. Five of 48 (10%) shoulders had recurrent instability in the open group, and 9 of 60 (15%) shoulders developed instability in

the arthroscopic group. The open group had 10 degrees less external rotation in abduction, on average. Sperber and associates conducted a randomized trial comparing open stabilization and arthroscopic stabilization with a ­bioabsorbable tack.108 Seven of 30 patients in the arthroscopic group developed a recurrent dislocation, compared with 3 of 26 patients in the open stabilization group. There was a trend toward increased recurrence in the arthroscopic group, but the average follow-up in this study was short (about 1 year) and included patients with as short as 2 months’ follow-up. Fabbriciani and associates randomized 60 patients with traumatic anterior shoulder instability to open or arthroscopic repair using metal suture anchors and found no difference in recurrence rates or outcome scores at 2 years’ average follow-up except slight decreased range of motion in the open group.109 Rhee and ­coworkers conducted a retrospective cohort study evaluating the results of 16 arthroscopic stabilizations and 32 open stabilizations in collision athletes.110 They reported four redislocations in each group, resulting in a redislocation rate of 25% in the arthroscopic group and 12.5% in the open group at a mean follow-up of 72 months. Cole and colleagues also compared the results of open capsular shift and arthroscopic shoulder stabilization using a bioabsorbable tack.111 They compared the results of 39 patients with anterior instability only (demonstrated on examination under anesthesia) with the results of open Bankart repair and capsular shift in 24 patients with anterior and inferior instability. They found no significant difference in return to sport, shoulder scores, and recurrence rates between groups. There was a slight decrease in average forward elevation in the open group. Recurrence rate was high in this study: 24% in the arthroscopic group and 18% in the open group. Bottoni and associates randomized 64 young patients to receive arthroscopic or open stabilization by a single surgeon. After an average of 32 months, subjective shoulder scores were equivalent. The average operating room time was 59 minutes in the arthroscopic group and was 149 minutes in the open group. Average range of motion loss compared with the contralateral shoulder was higher in the open group. Three clinical failures occurred (two in the open group).112 Attempting to summarize these data is difficult. There is a relative lack of adequately powered randomized controlled trials comparing arthroscopic stabilization and open stabilization. Therefore, the following opinion is based more on a critical review of multiple case series and personal experience rather than level-one evidence. The most recent evidence suggests that arthroscopic anterior stabilization can lead to failure rates of less than 10%. Arthroscopic reconstruction may have a slightly higher recurrence rate than open repair, and open repair probably increases the incidence of range of motion loss. However, in current practice, many patients with recurrent anterior instability present well informed and expect an arthroscopic stabilization. More important, therefore, is deciding which candidates are not candidates for arthroscopic stabilization. Burkhart evaluated 194 patients who had arthroscopic Bankart repair and found a recurrence rate of only 4% in 173 patients without bony deficiency, but reported recurrence in 14 of 21 patients with bone defects (e.g., the “inverted pear” glenoid).83 Boileau and colleagues evaluated their

Shoulder

results in 91 consecutive arthroscopic stabilizations using suture anchors.113 At an average of 3 years’ follow-up, the recurrence rate was 15.3%. The risk for recurrent dislocation was increased significantly in the presence of significant glenoid compression fracture or Hill-Sachs defects. Also, significant inferior laxity diagnosed at the time of surgery also was a risk factor for recurrence. In summary, we perform primary arthroscopic anterior stabilization routinely for recurrent anterior instability, uncomplicated revision surgery, and high-risk first-time dislocators. We believe that the overhead athlete is not a good candidate for an open repair because the range of motion loss observed with open repair can preclude a return to the previous level. We do not consider participation in contact athletics a contraindication to arthroscopic stabilization. We do counsel collision athletes that recurrence after stabilization is certainly dependent not only on the repair type but also on the activity level chosen after surgery. We use a combination of history, physical examination, and preoperative imaging to diagnose glenoid bone loss, Hill-Sachs defects, HAGL lesions, and MDI, all of which are clear risk factors for recurrence after a standard arthroscopic Bankart repair. Lastly, it is important to be prepared to convert to open surgery if examination under anesthesia and diagnostic arthroscopy reveal any of the aforementioned diagnoses not amenable to successful arthroscopic reconstruction.

POSTOPERATIVE MANAGEMENT AND RETURN TO PLAY Postoperative management after arthroscopic or open stabilization surgery is critical and must be individualized based on the type of surgery performed and quality of the repaired tissue. We immobilize the shoulder in a sling for 4 weeks after a labral repair and up to 6 weeks for a capsular plication. If a subscapularis takedown was performed for an open repair, this is generally protected as well for 4 weeks. Pendulum exercises and passive forward elevation in the plane of the scapula to 90 degrees are allowed immediately. Passive external rotation is permitted to 0 degrees. Distal range of motion is also started immediately. Active range of motion is allowed after the sling immobilization is discontinued. Terminal passive stretching exercises are delayed until 8 to 10 weeks after surgery. After range of motion is nearly full, strengthening is started. Return to sport is generally delayed until 4 months after surgery when the patient has gained full range of motion and nearly full strength.114 McCarty and colleagues reviewed rehabilitation and return-to-play guidelines after shoulder surgery.115 These authors divided the rehabilitation after shoulder surgery into four phases. Phase one consists of primary healing of capsulolabral tissue consisting of sling immobilization, active-assisted or passive range of motion exercise, distal range of motion, and isometrics. Phase two spans from weeks 6 to 12 and involves gaining full range of motion. Phase three12-20 involves strength­ening of the shoulder, including dynamic strengthening. Four to 8 months after surgery, the patient returns to sport-specific rehabilitation; he or she returns to sport when range of motion and strength are nearly ­normal.

929

COMPLICATIONS Complications of shoulder instability surgery include recurrence of instability, stiffness, subscapularis rupture, arthrosis, neurovascular injury, hardware failure, and capsular necrosis.84,116 The risk factors for recurrence after stabilization include MDI, large Hill-Sachs or glenoid rim fractures, and voluntary instability.83,113,117 The keys to avoiding recurrence are proper diagnosis and proper surgical technique in addressing the pathology observed on imaging and ­arthroscopy. Arthrosis of the glenohumeral joint is a known sequela of recurrent shoulder dislocation in some shoulder joints with or without surgery. There is evidence that dislocation of the shoulder alone is associated with the development of osteoarthritis. Hovelius45 followed 247 anterior shoulder dislocations for 10 years and showed that 20% had at least mild osteoarthritis and 9% had severe osteoarthritis. Marx and coworkers conducted a case-control study and determined that the risk for developing severe arthrosis of the shoulder is between 10 and 20 times greater for individuals who have had a dislocation of the shoulder than for those with no dislocation.118 There is limited long-term evidence that shoulder surgery for instability prevents osteoarthritis. Arthritis after shoulder stabilization was very common after nonanatomic repairs such as coracoid transfers (Bristow and Latarjet), subscapularis advancement (MagnusonStack), and subscapularis shortening (Putti-Platt).119-121 The incidence of osteoarthritis after modern open Bankart repair is thought to be lower than in nonanatomic repair. Rosenberg and colleagues showed moderate to severe osteoarthrosis in only 4 of 33 shoulders 15 years after Bankart repair.122 However, one study by Pelet and associates (albeit with older surgical technique) showed a 40% osteoarthritis rate in 30 patients 29 years after open Bankart repair for anterior instability.123 Shoulder surgeons should evaluate the long-term results of modern arthroscopic and open stabilizations to determine whether osteoarthritis of the glenohumeral joint can be prevented with anatomic repairs. Although our success rates are often judged in the short term by dislocation rates, perhaps a better measure of surgical success is the long-term function of the joint and prevention of future arthrosis. Stiffness can complicate shoulder stabilization surgery, especially with overtensioning of the soft tissues and prolonged immobilization. To avoid overtightening of the soft tissues and losing external rotation, it is important that capsulolabral reconstructions be tensioned in 30 to 45 degrees of external rotation and 30 degrees of abduction.84 The incidence of neurovascular complications in open stabilization surgery has been reported as 1% to 8%.124 ­Neurovascular complications and subscapularis rupture are rare but devastating complications of open stabilization surgery and are best prevented by meticulous ­surgical technique and deliberate exposure of surgical planes ­during revision surgery. One of the major developments in shoulder surgery in the past 5 years has been a shift away from thermal capsulorrhaphy. Although many surgeons had initial success with this technique (especially in anterior instability), its failure rate has been high in some recent series. Recurrence, stiffness, and neurologic complications have all been

930

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

reported.125 This had led some authorities to recommend avoidance of thermal capsulorrhaphy, especially in MDI and posterior instability.125,126 Further basic science and clinical studies are needed to optimize thermal capsulorrhaphy for shoulder instability.

SPECIAL POPULATIONS Multidirectional Instability MDI describes a subset of shoulder instability patients who often have congenital ligamentous laxity or repetitive microtrauma in a susceptible shoulder (e.g., swimmers).127

A u t h o r s ’ P r e f e r r e d M e t h o d

for

Despite excellent results of open capsular shift for MDI, our institution has moved progressively toward arthroscopic capsular plication to stabilize shoulders with MDI (Fig. 17H1-19). The operation begins with an examination under anesthesia, which can guide the amount and direction of capsular plication necessary. A typical examination

A

C

This condition differs from traumatic anterior instability in that it is often atraumatic and bilateral.128 Diagnosis is made by history and physical examination, with common findings including 2+ instability in more than one direction and a sulcus sign. Treatment initially relies on physical therapy, which is successful in most cases but needs to be continued for 6 to 12 months. When rehabilitation fails, arthroscopic129 or open130 capsular plication can result in improved stability of the shoulder and decreased pain. Recent studies have shown that arthroscopic treatment success is approaching that of open surgery, with most studies demonstrating a 10% recurrence rate with few complications.131 Again, thermal capsulorrhaphy has fallen out of favor in this condition because of a high complication and failure rate.

M DI S t a b i l i z a t i o n in a patient requiring stabilization is 2+ or 3+ anteriorly, 2+ posteriorly, and a sulcus sign. Diagnostic arthroscopy shows a drive-through sign and patulous capsule. Capsular plication begins anteroinferiorly, taking a “bite” of capsule and a second pass near the capsule-labrum transition with a suture passing hook. The goal of each stitch is to draw the

B

Figure 17H1-19   Arthroscopic stabilization of multidirectional instability (MDI). A, Drive through sign and redundant anterior inferior capsule. B, Suture hook at 5:30 drawing inferior pouch of capsule superiorly and medially. C: Elimination of drive-through sign after placement of four anterior sutures. A single capsular plication suture was placed posteriorly.

Shoulder

A u t h o r s ’ P r e f e r r e d M e t h o d

for

capsule superiorly and medially. Four to five stitches are placed anteriorly, closing the anteroinferior pouch, centering the humeral head, and eliminating the drive-through sign. The arthroscope is then introduced anteriorly, and

C

r i t i c a l

P

o i n t s

l Improved understanding of the anatomy, pathoanatomy, and biomechanics of the shoulder has allowed surgeons to improve surgical techniques for stabilization. l A thorough history and physical examination will most often diagnose the degree, direction, frequency, and cause of shoulder instability. l The cornerstone of treatment for instability remains immobilization followed by rehabilitation, but the optimal nonoperative treatment of a shoulder dislocation is still unknown and deserves further study. l Early arthroscopic stabilization can decrease recurrence rates and improve functional outcome in young (<25 years), high-risk individuals with first-time anterior dislocations and a Bankart lesion. l Proper mobilization of capsulolabral tissue, placement of an accessory anterior portal low and laterally, and addressing capsular laxity are critical in arthroscopic stabilization surgery. l Open surgery for anterior and posterior stabilization remains an excellent option for primary stabilization surgery. Open surgery should also be strongly considered in cases of severe capsular laxity or deficiency, glenoid or humeral bone loss, avulsion of the capsule from the humeral side, and revision stabilizations. l Results of arthroscopic surgery using modern suture anchor techniques and shoulder-specific instrumentation have yielded results comparable to open surgery in some series. Most authorities would agree that open surgery takes longer and may have a slightly lower recurrence rate. However, this lower recurrence rate comes with the cost of some lost range of motion. l Athletes may return to play after a shoulder dislocation or after stabilization surgery when strength and mobility have normalized; generally this averages 4 months.

931

M DI S t a b i l i z a t i o n � — cont ’ d the posterior capsule is plicated as necessary. For patients with pain and a mild MDI picture with subluxation only, we perform an isolated capsular shift through the rotator interval.

l Multidirectional instability often presents as bilateral shoulder subluxations in a susceptible patient and should be treated with an extended course of physical therapy. Arthroscopic and open capsular shifts have been successful in restoring stability and decreasing pain, but thermal capsulorrhaphy has recently shown a high complication rate.

S U G G E S T E D

R E A D I N G S

Boileau P, Villalba M, Hery JY, et al: Risk factors for recurrence of shoulder instability after arthroscopic Bankart repair. J Bone Joint Surg Am 88(8):1755-1763, 2006. Burkhart SS, De Beer JF: Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: Significance of the inverted-pear glenoid and the humeral engaging Hill-Sachs lesion. Arthroscopy 16(7):677-694, 2000. Clavert PH, Millett PJ, Warner JJP: Traumatic anterior instability: open solutions. In Warner JJP, Iannotti JP, Flatow EL (eds): Complex and Revision Problems in Shoulder Surgery. Philadelphia, JB Lippincott, 1995, pp 23-52. Cole BJ, Millett PJ, Romeo AA, et al: Arthroscopic treatment of anterior glenohumeral instability: indications and techniques. Instr Course Lect 53:545-558, 2004. Cooper DE, Arnoczky SP, O’Brien SJ, et al: Anatomy, histology, and vascularity of the glenoid labrum: An anatomical study. J Bone Joint Surg Am 74(1):46-52, 1992. Gill TJ, Zarins B: Open repairs for the treatment of anterior shoulder instability. Am J Sports Med 31(1):142-153, 2003. Millett PJ, Clavert P, Warner JJ: Open operative treatment for anterior shoulder instability: When and why? J Bone Joint Surg Am 87(2):419-432, 2005. Miniaci A, Codsi MJ: Thermal capsulorrhaphy for the treatment of shoulder instability. Am J Sports Med 34(8):1356-1363, 2006. O’Brien SJ, Neves MC, Arnoczky SP, et al: The anatomy and histology of the inferior glenohumeral ligament complex of the shoulder. Am J Sports Med 18(5):449456, 1990. Wahl CJ, Warren RF, Altchek DW: Shoulder arthroscopy. In Rockwood CA, Matsen FA, Wirth MA, Lippitt SB (eds): The Shoulder. Philadelphia, Saunders, 2004, pp 283-353.

R eferences Please see www.expertconsult.com

932

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S E C T I O N

H

Glenohumeral Instabilities 2. Glenohumeral Instability in the Child Ralph J. Curtis Jr.

The glenohumeral joint of the shoulder is susceptible to dislocation and subluxation in high-demand sports activities. The anatomy of the glenohumeral joint provides a high degree of functional mobility with the sacrifice of stability. Although glenohumeral joint instability is common in adolescents and adults, it is less common in children with open physes. Traumatic forces applied to the upper extremity in the skeletally immature child are more likely to result in fracture through the proximal humeral physis as opposed to damage to the soft tissues of the shoulder joint that lead to instability.1-3 Therefore, traumatic dislocation of the shoulder in a child is uncommon. In the adolescent with open physes, however, patterns of instability begin to resemble those found in the adult. Varying degrees of atraumatic instability based on multidirectional laxity of the shoulder are more commonly symptomatic in younger patients and must be recognized when treating athletes in these age groups.

RELEVANT ANATOMY AND BIOMECHANICS Anatomy The glenohumeral joint is formed by the articulation between the proximal humerus and glenoid portion of the scapula. There are several unique aspects of the glenohumeral joint that anatomically allow for range of motion but sacrifice stability. The humeral head articular surface is about 3 times that of the glenoid so that only 25% to 30% of the humeral head contacts the glenoid at any given time. This would be similar to placing a golf ball on a tee in a horizontal position. The glenoid is relatively flat with a radius of curvature greater than that of the humeral head. However, a circumferential rim of dense collagenous fibrous tissue called the glenoid labrum deepens the socket and decreases the mismatch in radius of curvature. Therefore, the soft tissues, including ligaments, muscles, and tendons, are extremely important in stability of the shoulder. The glenohumeral joint is a true synovial joint with a capsule reinforced by thickened areas known as the glenohumeral ligaments. The humeral attachment of the capsular ligaments occurs along the region of the anatomic neck except medially, where the attachment extends distally along the shaft. The proximal humeral physis lies in an extracapsular position except medially, where it is an

intra-articular structure. The capsular ligaments have been more precisely defined in the anterior and posterior capsule as the superior, middle, and inferior glenohumeral ligaments. The inferior glenohumeral ligament complex along with its labral attachment is important in stability of the shoulder joint. The rotator cuff muscles, including subscapularis, ­sup­raspinatus, infraspinatus, and teres minor, originate on the scapula and attach to the greater and lesser ­tuberosities as a thickened sleeve of tendon that covers the joint ­anteriorly, posteriorly and superiorly. The insertion of the rotator cuff is contiguous with the attachment of the glenohumeral ligaments.

Biomechanics The range of motion at the shoulder is greater than at any other major joint in the body. It can be carried through an arc of elevation from 0 to 180 degrees. In addition, there is an internal rotation to external rotation range of about 150 degrees and a horizontal plane range of motion in adduction and abduction of about 170 degrees. To accomplish this range of motion, there is contribution from both the glenohumeral joint and the scapulothoracic articulation. The average ratio of glenohumeral to scapulothoracic motion is about 2:1. The instant center of rotation within the humeral head has a small amount of upward translation in the first 30 degrees of elevation equal to about 3 mm. There is also a small 1- to 2-mm translation in the anterior to posterior plane with normal shoulder motion. Static glenohumeral stability is accomplished through several factors. First, the congruency of the humeral head and the glenoid-labrum complex provide an element of stability. Most important, the glenohumeral ligaments function as the primary soft tissue stabilizer at the extremes of range of motion. In addition, the intra-articular long head biceps tendon and rotator cuff tendons provide some passive static stability along with their contribution to dynamic stability. The normal negative intra-articular pressure that is present within the closed glenohumeral joint capsule is another factor adding to static stability. Dynamic stability is a product of several factors that combine to enhance the static stability of the joint, particularly in the midrange of motion. Active contraction of the shoulder musculature compresses the humeral head into the glenoid, increasing joint reactive forces that resist translation, the so-called concavity-compression effect.

Shoulder

The combined rotator cuff tendons have an important dynamic stabilizing effect by forming a force-couple to counterbalance the shear forces resultant to the muscular power of the deltoid, pectoralis major, and latissimus dorsi. In addition, specialized nerve endings in the rotator cuff and glenohumeral joint capsule provide proprioception that allows for effective sequential muscular action during joint motion.

INCIDENCE Incidence data for glenohumeral instability in skeletally immature patients are difficult to extrapolate from the literature.4-6 A review of 500 dislocated shoulders by Rowe listed only 8 patients younger than 10 years old, whereas 99 patients were between 10 and 20 years of age.7,8 Wagner and Lyne presented a series of 9 children with open epiphyses of 212 patients with traumatic glenohumeral dislocations, representing a 4.7% incidence rate.9 Heck10 and Foster and colleagues11 individually reported cases of true glenohumeral dislocation due to trauma in children younger than 10 years. Endo and coworkers presented two cases of traumatic dislocation of the shoulder in children younger than 10 years old.12 Many reports in the literature include patients between 11 and 20 years of age, but data for skeletal maturity are not included. Most of these studies involve traumatic instability treated by surgical reconstruction and do not address instability treated by nonoperative means.13-19 Other studies describing multidirectional laxity with subluxation of the shoulder demonstrate this problem to be more common in younger age groups.20-22 Certainly, true traumatic glenohumeral dislocation in children younger than 10 years is rare. True traumatic instability of the shoulder in skeletally immature adolescents is more common, probably approaching the incidence in adults (Box 17H2-1). Box 17H2-1 Classification Degree of Instability 1. Dislocation—complete, requiring reduction 2. Subluxation—incomplete, not requiring reduction Direction of Instability 1. Anterior 2. Posterior 3. Multidirectional (MDI) 4. Inferior (luxatio erecta)80 Etiology of Instability 1. Traumatic instability a. Macrotraumatic—single event b. Microtraumatic—multiple subthreshold events 2. Atraumatic instability a. Voluntary—volitional b. Involuntary—not controllable Frequency of Instability 1. Acute instability—single episode 2. Recurrent instability—multiple episodes 3. Chronic instability—locked dislocation

933

TRAUMATIC ANTERIOR INSTABILITY Clinical Presentation and History Anterior instability of the glenohumeral joint is the most common type of instability associated with trauma. This type represents more than 90% of traumatic dislocations in all age groups and is frequently seen in collision and contact sports. In a true traumatic dislocation, there is a history of significant injury and an appropriate mechanism. Anterior dislocation is associated with an eccentric load applied to the outstretched hand, which forces the shoulder into an abducted and externally rotated position. The humeral head is translated anteriorly, damaging the anterior-inferior glenohumeral ligaments and labrum, eventually dislocating. The posterior-superior aspect of the humeral head is impacted against the anterior rim and neck of the glenoid, resulting in an impaction fracture called a Hill-Sachs lesion. The capsular injury that results in stripping of the labrum and capsule from its insertion on the rim of the glenoid is termed a Bankart or Perthes’ lesion (Fig. 17H2-1).

Physical Examination and Testing Patients with traumatic glenohumeral anterior dislocation present with obvious deformity, often with swelling and always with pain. The acromion is prominent with a void below it. The humeral head is located anteriorly, and it can sometimes be visualized or palpated in the axilla. The affected arm is usually supported by the opposite hand and held in a slightly abducted and externally rotated position (Fig. 17H2-2). There is pain with any attempted motion and occasionally crepitus. Many patients with traumatic anterior instability present after a subluxation episode or after spontaneous reduction of acute dislocation. In this situation, deformity is absent, and the examination demonstrates tenderness with guarding in the apprehension position of abduction and external rotation. Careful examination of the neurologic and vascular status is mandatory. The axillary nerve is the most commonly injured neurologic structure, but there may be additional brachial plexus involvement. Axillary nerve injury has been reported in between 5% and 35% of first-time anterior shoulder dislocations.23 Careful examination of the axillary nerve can be accomplished easily. The sensory distribution of the axillary nerve is along the upper lateral arm and can be tested by light touch or pinprick. The motor innervation of the axillary nerve includes the deltoid and teres minor muscles. This motor innervation can be tested by asking the patient to abduct the arm as the examiner supports the affected elbow with one hand while palpating the deltoid for contraction with the opposite hand. This can be accomplished without causing the patient undue pain and confirms function of the axillary nerve (Fig. 17H2-3). The presence of both the radial and ulnar pulses should be noted. Absence of the pulses in association with massive swelling or rapidly expanding hematoma suggests a rare vascular injury.24

934

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17H2-1   Diagram in the coronal plane of the relationship between the humeral head (HH) and the glenoid in anterior traumatic instability. Note the Perthes’/Bankart lesion, which includes stripping of the anterior-inferior labrum and ligaments off the glenoid rim. Also note the Hill-Sachs compression fracture on the posterior-superior aspect of the humeral head. A, anterior; P, posterior.

P

HH

Perthes’/Bankart lesion

Imaging Acute traumatic instability of the shoulder is best evaluated initially with routine radiography. Particularly in children with suspected injury to the shoulder, physeal fracture should always be ruled out with routine radiographs.2,25 The trauma series for the shoulder includes three views, including the anteroposterior view, an axillary lateral view, and a transscapular Y view.26,27 It is important to image the shoulder with at least two views oriented at right angles to the other to confirm the actual position of the humeral head in relation to the glenoid. In cases of anterior instability, the humeral head can be seen situated anterior and inferior to the normal position in the glenoid (Fig. 17H2-4). There is overlapping of the humeral head with the glenoid on the anteroposterior view. The lateral radiographs confirm the anterior position of the head. Postreduction films in both planes are important to confirm reduction and assess for fractures that are difficult to visualize in the dislocated shoulder. The postreduction films often reveal the common posterolateral humeral head impaction fracture (Hill-Sachs lesion) but can also reveal fractures involving

Hill Sachs lesion

A

the anterior glenoid rim. The glenoid rim is best evaluated for fracture or deficiency on an axillary lateral or modified axillary lateral view (West Point view).28 Magnetic resonance imaging (MRI) has improved our ability to assess soft tissue injury in the shoulder. Anterior capsular and labral injury can be successfully visualized in a high percentage of cases.29

A

B

Figure 17H2-2   A patient with acute traumatic anterior dislocation of the shoulder. Note the prominent acromion with characteristic deformity of the shoulder.

Figure 17H2-3   Clinical examination of the axillary nerve should be carried out in every patient with anterior dislocation. A, Sensory distribution noted in the upper lateral arm. B, Motor testing for deltoid function.

Shoulder

A

935

B

Figure 17H2-4   Anterior dislocation of the shoulder. A, Prereduction radiograph demonstrating an anteroinferior dislocation of the left shoulder. B, Postreduction radiograph demonstrating a prominent posterolateral compression fracture known as a Hill-Sachs lesion.

Nonoperative Treatment After completion of a thorough examination, initial treatment is accomplished by closed reduction. If the dislocation is treated promptly, reduction can often be carried out without anesthesia. If significant pain and muscle spasm have ensued, appropriate intra-articular anesthesia or intravenous analgesia and sedation are used.30,31 Several safe and effective methods for closed reduction of an anterior dislocation have been described, including the traction-­countertraction technique, Stimson’s maneuver, and the abduction maneuver. On-the-field reduction of a suspected dislocation in a child before radiographic confirmation is certainly more hazardous than in an adult because of the higher frequency of proximal humeral physeal fracture in this age group. The traction-countertraction technique (Fig. 17H2-5) is accomplished with the patient in the supine position. Longitudinal traction is applied to the arm on a continuous basis while countertraction is applied to the thorax by means of a sheet passed around the patient through the axilla. The humeral head is disimpacted from the anterior glenoid rim by overcoming the muscle spasm with traction, leading to reduction. In Stimson’s maneuver (Fig. 17H2-6), the patient is placed prone on an examination table. The dislocated arm is allowed to hang off the edge of the table while up to 10 to 15 pounds is suspended from the patient’s wrist. Spontaneous reduction occurs as the shoulder musculature is relaxed by the gravity-assisted traction. Closed reduction by the abduction maneuver is performed with the patient in the supine position.32 The arm is supported by the examiner with the elbow flexed. The shoulder is gently abducted and externally rotated into the overhead position, reproducing the mechanism of injury. Reduction occurs as the arm is then adducted and gently internally rotated.

It is important to repeat examination of the neurologic and vascular status after reduction. Postreduction radiographs are also obtained to confirm the position of the humeral head and to rule out the possibility of associated fracture. After reduction, the arm is immobilized in a sling for protection.

Operative Treatment Rarely is acute surgical treatment for open reduction of a traumatic anterior dislocation necessary. Closed reduction is successful when adequate muscular relaxation is obtained. After initial reduction, surgical stabilization as the definitive treatment for initial traumatic anterior dislocation in the young, at-risk athlete has been reported.33

Figure 17H2-5   The traction-countertraction maneuver for reduction of anterior dislocation of the shoulder.

936

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Postoperative Prescription

Figure 17H2-6   Stimson’s maneuver for reduction   of anterior dislocation of the shoulder.

Arthroscopic findings have documented the presence of a Bankart lesion with hemarthrosis in a large number of cases, providing a favorable milieu for arthroscopic fixation. Studies from the U.S. Military Academy using either an arthroscopic transglenoid suture technique or an arthroscopic bioabsorbable fixation device have resulted in an 88% success rate.34-36 Bottoni and associates reported on a prospective study in treating acute dislocations in which a 75% recurrence rate was observed after immobilization and rehabilitation compared with an 11% recurrence rate after arthroscopic stabilization.37 These results are appealing because the recurrence rate after anterior dislocation in patients younger than 20 years has been as low as 48% and as high as 100%. The more common arthroscopic techniques used today include repair of the Bankart lesion using suture anchors and capsular plication.

A u t h o r ’ s P r e f e r r e d M e t h o d Closed reduction is the initial treatment of choice after acute traumatic anterior dislocation. Physical examination with careful attention to the neurologic status is performed before and after closed reduction. In children younger than 14 years, I prefer radiographic evaluation with a trauma series to confirm the diagnosis and to rule out associated fracture before reduction. In adolescents 14 years and ­older, on-the-field gentle closed reduction can be attempted ­before radiographic evaluation. For stable dislocations, a sling is adequate for postreduction immobilization. Despite the high incidence of recurrence in young ­patients, nonoperative treatment is still our treatment of choice. Occasionally adolescent athletes with documented traumatic lesions who participate in high-risk collision sports are considered for primary surgery. Successful ­results in treating young athletes with primary traumatic dislocations have been obtained using arthroscopic suture anchor repair techniques.

Immobilization in a sling is followed by a period of supervised rehabilitation. Aggressive rehabilitation focuses on strengthening the rotator cuff, deltoid, and periscapular muscles. Plyometric exercises are added in the later rehabilitation phase in an attempt to improve proprioceptive function for return to sports-specific activities.38 The length of immobilization and the type and length of rehabilitation are somewhat controversial. Recurrent dislocation is the single biggest problem in shoulder instability after an acute traumatic dislocation. At least two studies have demonstrated lower recurrence rates in young, at-risk patients treated with 4 to 6 weeks of immobilization and a 3- to 6-month delay in return to athletic activities.39,40 Other reports suggest that neither the length of immobilization nor the type or duration of rehabilitation alters the natural history of recurrence.41 Hovelius and coworkers have reported a 10-year prospective study that confirmed that neither the type nor the duration of the initial treatment had any effect on the rate of recurrence.18

Criteria for Return to Play After closed reduction of acute traumatic anterior shoulder instability, the athlete must regain full range of motion and protective strength before return to participation. The athlete should demonstrate good stability in the overhead position without apprehension before return to play. If the athlete continues to demonstrate apprehension in the provocative position, the decision about return to play must be made on an individual basis.

TRAUMATIC POSTERIOR INSTABILITY Clinical Presentation and History Posterior instability of the shoulder is less common than anterior instability. When all age groups are included, posterior dislocation represents about 4% of traumatic dislocations of the shoulder. However, many recent studies of posterior instability of the shoulder reported over the past decade were likely based on improved imaging and greater awareness of the problem.29,42-45 Many of these series of posterior instability include adolescent patients, but no data are available regarding whether these patients have completed skeletal maturity. Foster and associates described a case of an isolated traumatic posterior shoulder dislocation in a 10-year-old child.11 Most cases of posterior instability in young patients are associated with multidirectional atraumatic instability rather than a result of trauma. Both direct and indirect mechanisms can result in posterior dislocation. Posterior dislocation results in stripping of the posterior-inferior labrum and capsular ligaments from their attachment on the glenoid, which is termed a reverse Bankart lesion. When the anterior aspect of the humeral head impacts the posterior rim of the glenoid, the result is often an impaction fracture called a reverse HillSachs lesion.

Shoulder

937

Figure 17H2-7   Posterior dislocation of the shoulder can occur in contact sports both by a direct blow to the front of the shoulder or indirectly during a fall on an outstretched arm with the shoulder adducted, internally rotated, and flexed.

A force directed on an outstretched arm with the shoulder adducted, internally rotated, and flexed can commonly drive the shoulder out posteriorly (Fig. 17H2-7). This would appear to be a common mechanism in football offensive linemen who develop posterior subluxation more commonly than dislocation. A direct mechanism of injury results from a blow to the anterior aspect of the shoulder that can result in posterior shoulder instability. Either of these mechanisms can occur in the course of contact sports. In addition, posterior instability can occur indirectly as a result of a seizure or electrical shock. During seizure activity, violent contraction of the strong shoulder internal rotators forces the humeral head out posteriorly. Any complaint of shoulder pain in an athlete who has suffered a convulsion should be taken seriously and a posterior dislocation suspected.23 Clinically patients with posterior instability of the shoulder complain of pain and limited ability to move the arm. In cases with complete dislocation, patients are usually aware that the shoulder is out of place but may not be able to describe the direction of the dislocation. When subluxation occurs, the prominent complaint is pain with a transient sensation of instability. One must always remember that in the skeletally immature athlete, fracture through the proximal humeral physis is more common than dislocation and may mimic instability clinically.3

Physical Examination and Testing Posterior dislocation is less apparent on clinical examination when compared with the more common anterior dislocation and can easily be overlooked. The arm is usually held across the abdomen with the shoulder internally rotated and adducted. The shoulder is painful, and the patient avoids motion. Only on close inspection is deformity noted. There is flattening of the anterior aspect of the shoulder with prominence of the coracoid. A fullness posteriorly created by the dislocated humeral head can sometimes be appreciated (Fig. 17H2-8). These features are best appreciated when they are visualized by the examiner from above. The hallmark of the diagnosis is a lack of shoulder external rotation and inability to supinate at the forearm. These examination findings are subtle and sometimes difficult to elicit in the acute situation because of swelling and pain. The lack of clinical findings can lead the examiner to a delay in diagnosis. Examination of the athlete after a posterior subluxation episode is usually relatively normal on clinical inspection.

Figure 17H2-8   A patient with a locked posterior dislocation of the shoulder.

The shoulder may demonstrate tenderness at the joint line, mild swelling, and decreased range of motion. Pain is usually elicited with cross-arm adduction and the posterior apprehension maneuver. The jerk test may be positive for crepitus and a catch when the arm is taken from the adducted to the abducted position.

Imaging The radiographic evaluation of posterior instability of the shoulder includes the three-view trauma series: the anteroposterior view, the axillary lateral view, and the transscapular Y view. Even when the shoulder is posteriorly dislocated, findings on the anteroposterior view are subtle, with overlapping of the humeral head on the glenoid. The diagnosis is confirmed with a lateral radiograph, either axillary or transscapular Y view, demonstrating the posterior dislocation of the humeral head. On the axillary view, the empty glenoid is apparent with the humeral head lodged on the posterior aspect of the glenoid (Fig. 17H2-9). Postreduction films are important in both the anteroposterior and lateral planes to confirm the position of reduction and to evaluate for associated fracture. Common radiographic findings include a reversed Hill-Sachs impaction fracture on the anterior surface of the humeral head and marginal fractures of the posterior glenoid rim.

Nonoperative Treatment Initial treatment of posterior dislocation of the shoulder includes adequate diagnosis followed by closed reduction. Appropriate analgesia and sedation should be used to

938

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

In cases of suspected posterior shoulder subluxation, work-up with magnetic resonance arthrography is obtained to confirm the diagnosis. Most cases are treated with a short period of immobilization followed by an intensive rehabilitation program. Occasionally, primary surgery is considered for a high-performance athlete in whom a significant reverse Bankart lesion is identified. Arthroscopic capsulolabral reconstruction has been successful in allowing athletes to return to their preinjury level of participation.

Postoperative Prescription

Figure 17H2-9   Magnetic resonance image depicting a posterior Bankart lesion in recurrent posterior instability of the shoulder in a college football player.

obtain adequate muscular relaxation. With the patient in the supine position, reduction is carried out by applying traction to the adducted arm in line with the deformity. Countertraction is applied to the chest while lateral pressure is placed on the upper arm, lifting the humeral head back into the glenoid fossa. The arm is usually stable at the side and can be immobilized in a sling. If instability is present after reduction, an external rotation splint or spica cast can be applied. Treatment of posterior shoulder subluxation is supportive with use of a sling, analgesics, and ice. Further evaluation with radiographs and MRI is often required to confirm the diagnosis.

Operative Treatment Surgical treatment is rarely required for open reduction of a locked posterior dislocation. Surgery may become necessary after an acute traumatic posterior dislocation associated with fracture of the lesser tuberosity or with a major fracture of the posterior glenoid rim resulting in uncontrollable instability despite external rotation bracing. Open reduction and internal fixation of the fracture are usually sufficient to provide stability.

A u t h o r ’ s P r e f e r r e d M e t h o d For the pediatric and adolescent athlete with acute posterior dislocation of the shoulder, radiographic confirmation is required before attempted reduction to exclude proximal humerus fracture. Closed reduction is accomplished by applying longitudinal traction to the arm with countertraction to the torso. Stability is assessed, and immobilization is tailored to the most stable position. Radiographs after reduction should confirm position and exclude fracture.

The goal of postreduction or postsurgical treatment is to regain full range of motion and strength without apprehension. After the prescribed immobilization for 3 to 6 weeks, an aggressive rehabilitation program is begun. Active and assisted range of motion avoiding cross-body adduction is started when appropriate. This is followed by focused strengthening of the rotator cuff and scapular rotators. As strength progresses, additional chest, shoulder, and back exercises are started in the weight room. Complete rehabilitation for return to overhead and contact sports usually requires 4 to 6 months.

Criteria for Return to Play As for most injuries to the shoulder, return to play after posterior instability treated with nonoperative or operative techniques requires relatively normal function before return. Full, painless motion in flexion and rotation is needed. Protective strength in the injured arm must be present without apprehension in the provocative position.

RECURRENT TRAUMATIC ANTERIOR INSTABILITY Clinical Presentation and History Recurrent instability of the shoulder is the most common complication associated with nonoperative treatment after an acute traumatic anterior dislocation. Predictably, the age of the athlete at the time of a first dislocation is the most important factor in assessing the risk for recurrent instability after an acute anterior dislocation. The literature is filled with studies confirming that the risk for recurrence is inversely proportional to the patient’s age.6,17,41,46 In patients younger than 20 years who want to continue to be athletically active, the risk for recurrent dislocation ranges between 48% and 100%. Rowe, in 1963, reported a 100% incidence of recurrence in children younger than 10 years and a 94% incidence of recurrence in adolescent and young adult patients 11 to 20 years old.7 Unfortunately, he did not distinguish in this group of patients whether or not they had open proximal humeral physes. Rockwood and colleagues reported a recurrence rate of 50% in a series of adolescent patients between 13.8 and 15.8 years of age.23 This was a mixed group of patients with traumatic and atraumatic instability. Postacchini and

Shoulder

associates recently documented a recurrence rate of 86% overall and 92% in traumatic dislocations for a group of adolescent patients aged 12 to 17 years.47 Wagner and Lyne reported an 80% recurrence rate in 10 patients with clearly open proximal humeral physes.9 Marans and associates reported on the natural history after anterior dislocation in 21 children between the ages of 4 and 15 years with open physes at the time of initial dislocation.48 They found a 100% recurrence rate no matter what postreduction treatment program was used. In a 10-year prospective study by Hovelius and colleagues, a 48% recurrence rate after primary anterior dislocation was documented in young patients.18 Evaluation of the patient with recurrent anterior instability complaints begins with a thorough history. Many of these patients have required multiple trips to the emergency department for reduction of anterior dislocation. Careful consideration of the initial trauma should be given to rule out the possibility of atraumatic instability. Patients often present with pain only with specific activities in the provocative position. Feelings of instability are often associated with a “dead arm” sensation when the shoulder is positioned in abduction and external rotation. This may represent recurrent traumatic subluxation, which can be confirmed by examination and appropriate imaging such as MRI.

Physical Examination and Testing Clinical examination usually reveals a full range of motion with apprehension or guarding in the abducted, externally rotated position. Strength is generally equal to the opposite side except for occasional mild weakness in external rotation against resistance. Provocative tests for anterior instability include the anterior apprehension test and the relocation test.23,49 The anterior apprehension test (Fig. 17H2-10) is performed with the patient upright or seated with the examiner ­ positioned at the side. The examiner stabilizes the

A Figure 17H2-11   A and B, The relocation test.

939

Figure 17H2-10   The anterior apprehension test.

scapula with one palm, placing the thumb on the posterior aspect of the shoulder joint while the opposite hand is used to bring the patient’s arm into abduction and external rotation. The result is positive when the test elicits a feeling of apprehension with or without pain in this position. The relocation test (Fig. 17H2-11) is performed with the patient supine. The examiner performs the apprehension test in this position and then repeats the test with one hand applying a posteriorly directed force to the anterior shoulder. A resultant decrease in apprehension with this maneuver represents a positive test.

B

940

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Imaging Radiographic work-up consists of the anteroposterior view, axillary lateral view, and transscapular Y view. Special views, such as the West Point axillary lateral view for imaging the anterior glenoid and the Stryker notch view for evaluating the posterior humeral head, may also be helpful.27-29 Common radiographic findings include the Hill-Sachs compression fracture on the posterolateral aspect of the humeral head and anterior glenoid rim lesions associated with a bony Bankart defect. The size and position of these bony lesions have been reported to affect the success of surgical treatment for recurrent anterior instability. MRI has become the standard for evaluation in shoulder instability. MRI is commonly combined with arthrography to assess labral, capsuloligamentous, articular, and tendinous injury around the shoulder (Fig. 17H2-12).29

Nonoperative Treatment Nonoperative treatment for recurrent anterior instability of the shoulder has not been successful. Aggressive rehabilitation programs and bracing have been described but are usually only marginally effective in eliminating recurrent dislocation or subluxation. These conservative methods may be useful in reducing symptoms in an attempt by an athlete to complete a season before later definitive treatment.

Operative Treatment When confirmed by appropriate history, physical examination, and radiographic studies, recurrent traumatic anterior instability of the shoulder should be treated surgically. Historically, many types of procedures have been used successfully in the treatment of recurrent dislocation. The most common types of anterior reconstruction of the shoulder are bone block-type procedures (Bristow, Latarjet),50-52 subscapularis-shortening procedures (PuttiPlatt, Magnuson-­Stack),19,53,54 and capsular procedures (Bankart, du Toit).55,56 All these procedures are reported to have good success (85% to 95%) with regard to eliminating recurrent dislocation. The capsular procedures are directed toward restoring normal anatomy with direct repair of the capsulolabral structures. These procedures have been reported to have a high success rate combined with a low complication rate and are the most commonly used today.15,16,57,58 For patients with open physes, it would seem prudent to consider anatomic procedures around the shoulder that do not employ the use of metallic implants to decrease risks to the physis.

Figure 17H2-12   Magnetic resonance image of the shoulder demonstrating an anterior Bankart lesion in recurrent anterior instability.

Results for surgical treatment of recurrent anterior instability in skeletally immature patients have been poorly documented. Many studies include patients who are skeletally immature, but results are not specifically categorized by age, so conclusions about treatment must be extrapolated from the adult literature. In a report by Wagner and Lyne, 10 shoulders in pediatric patients were treated by surgical reconstruction using the Magnuson-Stack and Bristow procedures.9 Marans and associates described 13 of 21 recurrent dislocations of the shoulder treated surgically with soft tissue procedures.48 In both studies, shortterm results were good, but no long-term follow-up results were reported. Postacchini and colleagues reported a 100% success rate in a group of adolescent patients treated with surgical reconstruction.47 Most authors today use the open Bankart repair or some variation of the capsular shift procedure with good results. Arthroscopic capsulolabral repairs have gained popularity as results approach those reported for open repairs.59,60 In separate reports by Cole and colleagues,61 Freedman and associates,62 and Karlsson and coworkers,63 results of both open and arthroscopic treatment for recurrent anterior dislocation were good; however, the open techniques still demonstrated 5% to 10% better results on direct comparison.

A u t h o r ’ s P r e f e r r e d M e t h o d In the skeletally immature patient who has documented r­ ecurrent traumatic anterior shoulder instability, surgical reconstruction is indicated. I prefer the open capsular imbrication procedure described by Wirth and associates combining a direct repair of the Bankart lesion using bioresorbable

­suture anchors with a capsular shift (Fig. 17H2-13).58 A sling is used postoperatively for 3 to 4 weeks, followed by a vigorous 6- to 9-month rehabilitation program before return to competitive sports.

Shoulder

941

A u t h o r ’ s P r e f e r r e d M e t h o d — cont ’ d

A

B Repair of Bankart lesion

C Lateral capsular shift

D Medial capsular shift

Completed reconstruction

Figure 17H2-13   The capsular shift reconstruction as described by Rockwood. A, The Bankart lesion is repaired directly with suture anchors or through drill holes. B, The medial capsular flap is passed beneath the lateral capsular flap, and the rotator interval is closed. C, The lateral flap is then closed over the medial flap. D, The end result with repair of Bankart lesion and tightening of capsular ligaments.

Postoperative Prescription The shoulder is protected in an immobilization device for 1 month while gentle range of motion is allowed. During the second month, range of motion is stressed, but the extremes of abduction and external rotation are avoided. At 6 weeks, rotator cuff and scapular strengthening exercises are begun. The athlete can begin core strengthening and running at the 6-week stage. At 12 weeks, the athlete is started on general weightlifting activities and proprioception retraining. Contact is avoided for 6 months.

the forward flexed position such as football linemen or in the follow-through phase of an overhead throw or serve. ­Occasionally the only symptom is pain with a dead-arm feeling in a specific sports position.

Physical Examination and Testing

RECURRENT TRAUMATIC POSTERIOR INSTABILITY

The examination is characterized by normal appearance and range of motion. The athlete usually demonstrates normal strength on manual muscle testing. Positive provocative tests include the posterior apprehension test and jerk test.49 The O’Brien test classically described for SLAP (superior labrum, anterior to posterior) tear lesions is often positive. The posterior apprehension test is performed with the patient upright standing or seated with the examiner positioned on the symptomatic side. The arm is taken into horizontal adduction with the shoulder flexed 90 degrees while the examiner applies a posteriorly directed force across the flexed elbow. A feeling of apprehension or instability is a positive test. The jerk test is similar to the posterior apprehension test in terms of position. The arm is taken into the horizontally adducted position, attempting to subluxate the shoulder posteriorly. When the arm is then moved rapidly into a horizontally abducted position the shoulder is reduced with a palpable and often visible jerk.

Clinical Presentation and History

Imaging

For athletes involved in contact sports, recurrent posterior instability after an acute episode is not uncommon. These episodes usually represent recurrent subluxation rather than dislocation. These patients are more likely to complain of pain associated with specific activities rather than instability. About half of patients with recurrent posterior subluxation have a sensation of instability but can rarely identify the direction. This problem is particularly disabling for athletes involved with contact in

Standard radiography of the shoulder should be completed with views in at least two planes oriented at 90 degrees to the other. The trauma series of radiographs include the anteroposterior view, the axillary lateral view, and the true scapular lateral view. Plain films are often unrevealing but may demonstrate a reverse Hill-Sachs lesion on the anterior aspect of the humeral head or a reverse bony Bankart lesion associated with a posterior soft tissue capsulolabral injury. Magnetic resonance arthrography is the

Criteria for Return to Play The goal of surgical reconstruction of the shoulder is to return the athlete to all activities without symptoms of instability. These goals usually require 6 to 9 months after surgical reconstruction to achieve. If the athlete has regained adequate range of motion and protective strength without signs of apprehension in the provocative position then the patient can return to contact and overhead sports.

942

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

standard for evaluating posterior instability of the shoulder. The reverse Bankart lesion and reverse Hill-Sachs injury can be well demonstrated. In addition, other soft tissue and bony injuries around the shoulder can be ­visualized, including chondral injuries, rotator cuff tears, and biceps lesions.

Nonoperative Treatment Rehabilitation and bracing are rarely effective in eliminating symptoms in patients with posterior recurrent instability of the shoulder. By improving protective strength and limiting range of motion in the provocative position, nonsurgical treatment may ameliorate the symptoms of instability to allow an athlete an opportunity to complete a season before obtaining definitive treatment.

Operative Treatment Surgical repair has been successful in the treatment of recurrent posterior shoulder instability. Both open and arthroscopic techniques that address the posterior capsulolabral injury (reverse Bankart lesion) and combine with capsular repair have been described.43,44,64,65 Results of both techniques have been similarly successful. Recent studies by Kim and colleagues,42 Bradley and colleagues,66 and Mair and associates67 have described greater than 90% success with arthroscopic posterior capsulolabral reconstructions in athletic populations. Wolf and colleagues45 and Misamore and Facibene68 have documented 81% to 92% good results with traditional open capsular ­techniques.

Postoperative Prescription The operative shoulder is immobilized for 1 month after ­surgery in an external rotation brace. At the 4- to 6-week stage, Thera-Band cuff strengthening and scapular strengthening is started along with range of motion exercises. Avoidance of provocative cross-arm adduction maneuvers and pressing activities is continued for 12 weeks. At 12 weeks, a full weight room strength program and proprioception retraining are begun.

Criteria for Return to Play The athlete is usually able to return to sports activities after 6 months of rehabilitation. Full range of motion, full protective strength, and no posterior apprehension are required.

ATRAUMATIC INSTABILITY Clinical Presentation and History Atraumatic instability of the glenohumeral joint in children and adolescents represents a common type. The actual incidence of this subset of instability is not available in the literature. As opposed to traumatic instability in which damage to the capsulolabral structures is macroscopic, atraumatic instability is characterized by redundancy and hyperelasticity of the capsule with increased intra-articular volume. An underlying multidirectional laxity of the shoulder is a prerequisite for pathologic atraumatic instability.22 Multidirectional laxity may be associated with a true ­syndrome

A u t h o r ’ s P r e f e r r e d M e t h o d In athletes with well-documented recurrent posterior ­instability associated with a posterior capsulolabral injury, I prefer arthroscopic repair. Bioresorbable suture anchors are placed along the posterior rim of the glenoid to repair

A

the reverse Bankart lesion (Fig. 17H2-14). Removal of loose bodies and capsular repair or plication can be performed as necessary. Closure of the rotator interval and bone grafting of the reverse Hill-Sachs lesion are unnecessary.

B

Figure 17H2-14   A, Arthroscopic view of posterior capsulolabral injury (reverse Bankart lesion) visualized from the anterior portal. B, Arthroscopic view following repair of posterior capsulolabral injury in recurrent posterior instability.

Shoulder

943

Figure 17H2-15   A and B,   A 10-year-old boy with voluntary instability of the shoulder.

A

B

of collagen deficiency, such as Marfan or Ehlers-Danlos syndrome. For most of these patients, however, the excessive joint laxity of the shoulder is just an extreme variant of normal. Emery and Mullaji have reported signs of instability in 57% of shoulders in boys and 48% in girls in a study of normal school-aged children.69 Atraumatic instability can be categorized as voluntary or involuntary. All these patients have multidirectional shoulder laxity with increased translation in at least three directions: anterior, posterior, and inferior. A shoulder “dislocation” in a child or adolescent without a clear-cut, significant history of trauma suggests that this may be an instance of atraumatic instability. These patients have inherent joint laxity, and the glenohumeral joint can be dislocated voluntarily or involuntarily as a result of minimal trauma. Episodes of instability may occur with activities such as throwing, hitting an overhead serve in tennis and volleyball, or swimming. These episodes do not constitute significant trauma, and a high index of suspicion for atraumatic instability should be maintained in these cases. Atraumatic instability associated with secondary impingement symptoms is a common cause of shoulder pain in children and adolescents involved in sports that require repetitive overhand motion such as swimming, baseball, and volleyball. These patients rarely complain of instability but rather complain of pain exacerbated by high-demand sports activity. Most of these individuals do not recognize their own inherent multidirectional laxity; therefore, it is difficult from the patient’s history to determine that instability is truly the primary underlying pathology. Voluntary instability is accomplished by patients with multidirectional laxity through conscious firing of certain muscle groups and inhibition of their antagonists while combining these muscle manipulations with certain arm positions that lead to subluxation of the glenohumeral joint (Fig. 17H2-15). A most notable finding in cases of voluntary

instability is the lack of pain associated with the subluxation or dislocation. Pathologic voluntary instability can be associated with psychological or emotional instability.21,70-72

Physical Examination and Testing On examination, signs of multidirectional laxity of the glenohumeral joint are present. The sulcus sign (Fig. 17H2-16), a dimpling of the skin below the acromion when manual longitudinal traction is applied to the arm, is due to inferior subluxation of the humeral head within the glenohumeral joint. Significant humeral head translation is often present on the anterior and posterior drawer test as described by Gerber and Ganz.73 The drawer test (Fig. 17H2-17) is performed with the examiner seated at the

Figure 17H2-16   The sulcus sign.

944

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17H2-17   A and B, The anterior-posterior drawer test.

side of the patient. The scapula is stabilized with one hand while the opposite hand manually translates the humeral head anteriorly and posteriorly. During the acutely painful phase, the affected shoulder may not demonstrate these signs because of guarding. If multidirectional laxity is suspected, the opposite noninvolved shoulder should be

A

examined to confirm the diagnosis. Many of these patients demonstrate evidence of ligamentous hyperlaxity in multiple other joints (Fig. 17H2-18). Examples include hyperextension at the elbows, knees, and metacarpophalangeal joints. Skin hyperelasticity and striae may also be present and may be suggestions of underlying collagen abnormality.

B

Figure 17H2-18   Note the hyperextension at the elbow (A) and the knees (B) often associated with multiple joint laxity.

Shoulder

Imaging The findings on radiographic examination in patients with atraumatic instability are usually normal. Stress radiographs can be used to supplement the clinical examination to demonstrate instability in the anterior, posterior, and inferior directions. The inferior component is easily demonstrated by applying weights to the arm during an anteroposterior radiographic film. Traumatic lesions, such as the Hill-Sachs lesion of the humeral head and anterior glenoid fracture, are not characteristic of atraumatic instability. Magnetic resonance arthrography commonly demonstrates an abnormally patulous, redundant capsule with a large intra-articular volume.

Nonoperative Treatment Treatment of patients with atraumatic instability begins with a thorough history and physical examination to confirm the diagnosis. A nonoperative approach is indicated as the initial treatment in every case of atraumatic instability. This nonoperative treatment emphasizes a vigorous rehabilitation program involving strengthening of the dynamic stabilizers, improvement of proprioception, and avoidance of provocative activities. Most patients who do not have significant emotional or psychiatric problems improve their symptomatic shoulder instability with this program. Burkhead and Rockwood have reported an 80% success rate in the treatment of atraumatic instability with a vigorous rehabilitation program alone (Fig. 17H2-19).40 Neer and Foster, in the classic description of multidirectional laxity, restricted surgical

945

intervention to patients who failed a 12-month rehabilitation program.74 Huber and Gerber have described 25 children who presented with voluntary subluxation of the shoulder at an average 12-year follow-up; 18 children were managed by “skillful neglect,” whereas 7 children had undergone stabilizing operations.75 At long-term follow-up, the nonoperative treatment group had 16 of 18 who were considered to have successful outcomes; in the surgical group, only 3 had good results. Their conclusion was that voluntary subluxation of the shoulder in children has a favorable prognosis, and there is no indication for surgical intervention during childhood. This conclusion has been confirmed by Postacchini and associates, who demonstrated 100% failure in adolescent patients with atraumatic instability treated ­surgically.47

Operative Treatment For patients with multidirectional atraumatic ­ instability who fail to improve after a thorough rehabilitation program, the inferior capsular shift procedure originally described by Neer and Foster has been successful.74 This procedure addresses the inferior capsular redundancy directly by diminishing the overall capsular volume. Mizuno and associates reported two cases of surgical treatment in young children with disabling multidirectional instability who were treated successfully with the inferior capsular shift.76 Many adolescent patients have been included in other large studies on surgical treatment of this difficult problem and appear to have successful results proportional to those reported for adults.20 Arthroscopic techniques have recently been described to accomplish capsulorrhaphy in cases of atraumatic instability with capsular redundancy.77 Thermal capsular shrinkage using electrothermal or laser probes as a means for reducing capsular redundancy have been used with somewhat unsatisfactory results.78,79 As previously discussed, care should be taken to exclude patients with true voluntary instability from surgical treatment.

A u t h o r ’ s P r e f e r r e d M e t h o d

Figure 17H2-19   Rehabilitation of the rotator cuff is important to provide dynamic stability of the shoulder.

In all cases of atraumatic instability, initial treatment should be conservative. Nonsteroidal anti-inflammatory medications can be used for reducing symptoms of secondary impingement along with avoidance of provocative activities. A diligent rehabilitation program should focus on improving dynamic stability through rotator cuff strengthening, scapular control, and proprioceptive retraining. A strict rehabilitation program is undertaken for 9 to 12 months before any more aggressive intervention is considered. In the patient with multidirectional instability who has failed to improve with the rehabilitation program, an inferior capsular shift reconstruction is performed to ­accomplish global tightening of the redundant capsule. Care must always be taken to exclude the voluntary dislocator, who will have a greater risk for poor outcome after surgical treatment.

946

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Postoperative Prescription

l Surgical

Patients who undergo a capsular shift reconstruction are treated with a very slow and controlled postoperative program. Immobilization in a sling continues for 6 weeks after surgery. At that time, a gentle range of motion program is begun along with a rotator cuff and scapular muscle strengthening program using Thera-Band. Ballistic training and heavier weightlifting activities are deferred for at least 4 months. Return to full activities requires 6 to 9 months.

Criteria for Return to Play Patients treated for atraumatic instability must have little or no pain with rehabilitation activities before consideration for return to athletics. They should also demonstrate full range of motion and adequate dynamic strength to provide stability. This level of improvement may be difficult to confirm on examination; therefore, a progressive trial of sports activity may be necessary in the decision for return to play.

C l Proximal

r i t i c a l

P

o i n t s

humeral physeal fractures in young patients with open physes are more common than traumatic instability of the shoulder in this age group. Therefore, in young athletes with suspected instability, fracture should always be ruled out. l Recurrent instability after a traumatic anterior shoulder dislocation is common, with recurrence rates of 48% to 100% well documented irrespective of type or duration of nonoperative treatment. Surgery after initial dislocation may be indicated for the high-level athlete. l Recurrent traumatic instability of the shoulder cannot be treated successfully with nonoperative methods. Open or arthroscopic anterior reconstruction is successful in a high percentage of cases. l Although less common than anterior instability, we now recognize the importance of traumatic posterior ­instability in young athletes. Initial treatment includes protection and rehabilitation.

treatment for recurrent traumatic posterior instability of the shoulder has a high success rate. Arthroscopic posterior capsulolabral reconstruction with suture anchors is the most commonly performed reconstruction. l Careful clinical and radiographic evaluation is needed to distinguish between traumatic and atraumatic instability in the child and adolescent. Atraumatic instability should always be treated with an extended nonoperative course of rehabilitation. l Magnetic resonance arthrography is useful in identifying specific pathology in athletes with shoulder instability. Treatment can then be directed based on specific f­indings.

S U G G E S T E D

R E A D I N G S

Bottoni CR, Wilckens JH, DeBerardino TM, et al: A prospective randomized evaluation of arthroscopic stabilization versus non-operative treatment in patients with acute, traumatic, first-time shoulder dislocations. Am J Sports Med 30:576-580, 2002. Bradley JP, Baker CL, Kline AJ, et al: Arthroscopic capsulolabral reconstruction for posterior instability of the shoulder: A prospective study of 100 shoulders. Am J Sports Med 34:1061-1071, 2006. Freedman KB, Smith AP, Romeo AA, et al: Open Bankart repair versus arthroscopic repair with transglenoid sutures or bioabsorbable tacks for recurrent anterior instability of the shoulder: A Meta-analysis. Am J Sports Med 32:1520-1527, 2004. Hovelius L, Augustini GB, Fredin OH, et al: Primary anterior dislocation of the shoulder in young patients: A ten-year prospective study. J Bone Joint Surg Am 78:1677-1684, 1996. Huber H, Gerber C: Voluntary subluxation of the shoulder in children: A long term followup study of 36 shoulders. J Bone Joint Surg Br 76:188-122, 1994. Karlsson J, Magnusson L, Ejerhed L, et al: Comparison of open and arthroscopic stabilization for recurrent shoulder dislocation in patients with a Bankart lesion. Am J Sports Med 29:538-542, 2001. Kraplinger FS, Gosler K, Wischatta R, et al: Predicting recurrence after primary anterior traumatic shoulder dislocation. Am J Sports Med 30:116-120, 2002. Marans HJ, Angel KR, Schemitsch EH, et al: The fate of traumatic anterior dislocation of the shoulder in children. J Bone Joint Surg Am 74:1242-1244, 1992. Postacchini F, Gumina S, Cinotti G: Anterior shoulder dislocation in adolescents. J Shoulder Elbow Surg 9:470-474, 2000. Wagner KT, Lyne ED: Adolescent traumatic dislocations of the shoulder with open epiphysis. J Pediatr Orthop 3:61-62, 1983.

R eferences Please see www.expertconsult.com

Shoulder

S E C T I O N

947

H

Glenohumeral Instabilities 3. Imaging of the Glenohumeral Joint Timothy G. Sanders

CONVENTIONAL IMAGING OF THE SHOULDER Conventional Radiography Conventional radiography is often the initial imaging examination performed for a patient presenting with shoulder pain, and although radiographs provide only a limited evaluation of the rotator cuff and glenoid labrum, they occasionally offer important information about the source of the patient’s symptoms. Radiographs depict an assortment of osseous abnormalities, including fracture, arthritis, and tumor, and they are frequently complementary to the more advanced imaging modalities, such as CT and MR imaging. A wide variety of radiographic views has evolved to aid in the evaluation of the glenohumeral joint, and knowledge about the advantages and disadvantages of each will assist in optimizing imaging protocols, depending on the clinical presentation (Box 17H3-1).1,2

Anteroposterior View The anteroposterior (AP) view (Fig. 17H3-1A) is typically obtained with the patient in the upright or supine position, and the beam is directed in a true AP direction relative to the body. The glenoid rim is normally tilted anteriorly about 40 degrees, which results in overlap of the humeral head and glenoid rim in the AP view. This view can be obtained with the humeral head in neutral position, internal or external rotation. When compared with other radiographic views of the shoulder, the AP view provides the best overview of the osseous structures of the shoulder girdle as the projection allows for relatively uniform

Box 17H3-1  Common Radiographic Views  of the Glenohumeral Joint Anteroposterior view Glenohumeral “true” anteroposterior (Grashey) view Axillary lateral view Scapular Y view Stryker notch view Acromioclavicular anteroposterior and posteroanterior views

­ istribution of soft tissue density across the entire should der. As a result, one or more of the AP views is nearly always included in the standard radiographic evaluation of the shoulder. These projections allow for adequate evaluation of the humeral head, glenoid, and body of the scapula as well as the acromioclavicular (AC) joint and coracoid process. The AP projection is very helpful in the evaluation following acute trauma for evidence of fracture or dislocation and is also of value in determining the cause of chronic pain from arthritis, impingement, calcific bursitis, tumor, or infection.

Glenohumeral “True” Anteroposterior (Grashey) View The true AP view, or Grashey view (see Fig. 17H3-1B), differs from the standard AP view in that the patient is rotated posteriorly 35 to 40 degrees, thus providing a tangential view of the glenohumeral joint. The advantage of the Grashey view is that it provides a superior evaluation of the glenohumeral joint. This view can demonstrate subtle subluxation in the superior or inferior direction and also will show subtle joint space narrowing associated with arthritis of the glenohumeral joint. The disadvantage is that there is a rapid change of soft tissue density from medial to lateral, and as a result, the lateral aspect of the shoulder, including the acromion and AC joint, is difficult to evaluate because of a rapid change in density on the radiograph and loss of anatomic detail laterally.

Axillary Lateral View Numerous variations of this projection exist, but the projection is most commonly obtained with the patient supine and with the arm abducted 90 degrees (see Fig 17H3-1C). The beam is then directed from distal to ­proximal while tilted 15 to 30 degrees toward the spine. This projection is best suited for evaluating for subtle anterior or posterior subluxation or dislocation and can also help detect subtle osseous Bankart fractures of the anterior glenoid rim. The osseous detail on this projection, however, is often quite poor, and its main benefit is in identifying anterior or posterior dislocation. Numerous variations of this projection have been developed, some with the goal of decreasing movement of the arm in the setting of acute trauma, and others with the intention of accentuating certain anatomic features. The West Point view is a variation of the axillary view that was developed

948

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Clavicle Clavicle

Acromioclavicular joint Acromion process

Acromion process

Coracoid process Coracoid process

Greater tuberosity Lesser tuberosity

Glenohumeral joint

Glenoid rim

A

B Clavicle Coracoid

Clavicle

Acromioclavicular joint Glenoid rim

Acromion process

Acromion

Coracoid process Humeral head

C Acromion process

Scapular body

Clavicle

Coracoid process

D Humeral head

Glenoid rim

E Figure 17H3-1   A, Anteroposterior view of the shoulder in external rotation. B, Anteroposterior view of the glenohumeral joint (Grashey view). C, Axillary view of the shoulder. D, Scapular Y view of the shoulder. E, Stryker notch view of the shoulder.

to optimize visualization of an osseous Bankart lesion of the anterior glenoid rim. It is obtained by placing the patient in the prone position on the x-ray table with the arm abducted 90 degrees from the long axis of the body and with the forearm draped over the edge of the top. The beam is directed 15 to 20 degrees in an inferiorto-superior direction and tilted 25 degrees toward the spine. This projection improves the detection of osseous Bankart lesions but is very difficult to obtain in the setting of acute trauma and is best reserved for the patient in the setting of subacute or chronic instability.

Scapular Y View The scapular Y view is easily obtained in the setting of acute trauma because it can be obtained with the arm immobilized by the side with very little or no movement required of the arm (see Fig. 17H3-1D). It can be very helpful in the setting of acute trauma to evaluate for anterior or posterior dislocation. The projection is obtained with the patient upright or prone and rotated approximately 30 to 45 degrees toward the cassette. The beam is directed down the body of the scapula and results in a projection in which the body of

Shoulder

949

the scapula is seen in tangent and the glenoid fossa en face as a Y-shaped intersection of the scapular body, coracoid process, and acromion process. This projection can replace the axillary lateral view as a lateral projection in the setting of acute trauma to evaluate for possible dislocation. It is also the projection commonly used when “typing” the undersurface of the acromion.

Stryker Notch View This view is best suited for viewing the posterolateral aspect of the humeral head and is an excellent radiographic view for detecting flattening of the posterior humeral head or a Hill-Sachs deformity (see Fig. 17H3-1E). It is very limited, however, in its evaluation of the glenoid rim for osseous Bankart lesions. The Stryker notch view is obtained with the patient in either the upright or supine position. The arm is positioned vertically overhead with the elbow flexed and the hand supported on the back of the head. The beam is directed toward the midaxilla and is tilted about 10 degrees in a cephalic direction.

Acromioclavicular Articulation Anteroposterior and Posteroanterior Views The AC joints are best evaluated with the patient either sitting or standing with the back flat against the cassette. The arms are usually freely hanging by the sides and are often holding weights or sandbags, which are helpful in accentuating AC joint separation. The beam is directed toward the midline of the body and centered over the AC joints. This view is very useful in evaluating for AC joint pathology, including fracture, AC joint separation, arthritis, and osteolysis of the distal clavicle. Comparison with the contralateral AC joint can be helpful in detecting subtle abnormalities.

Figure 17H3-2   Normal shoulder arthrogram. Contrast is seen within the glenohumeral joint. Ax, axillary recess; Bi, bicipital tendon sleeve; Su, subscapularis recess.

labrum and capsular structures, osseous outlet, acromion, and articular surfaces. Conventional shoulder arthrography (see Fig. 17H3-2) may be performed as either a single-contrast5 or doublecontrast6 procedure. The double-contrast technique is generally considered preferable,7 providing accuracy similar to that of the single-contrast method in the diagnosis of full-thickness tears while improving the detection rate of partial-thickness undersurface tears and cartilage defects. Although many variations in technique exist, a standard method for performing double-contrast arthrography

Conventional Shoulder Arthrography Conventional arthrography of the shoulder (Fig. 17H3-2) has long been considered the gold standard for preoperative diagnosis of full-thickness tears of the rotator cuff (Fig. 17H3-3); reported accuracies range as high as 98% to 99%.3 Partial-thickness rotator cuff tears involving the undersurface as well as adhesive capsulitis are also accurately depicted with use of this method. Partial-thickness rotator cuff tears involving the bursal surface and intra­ substance tears, on the other hand, are not demonstrated by conventional shoulder arthrography, and its usefulness in the evaluation of glenohumeral instability is limited.4 Conventional arthrography remains a cost-effective and highly accurate method of identifying full-thickness tears of the rotator cuff. During the past decade, however, magnetic resonance imaging (MRI) has largely replaced it as the primary imaging modality for evaluation of the rotator cuff, and in many practices, magnetic resonance (MR) arthrography has evolved into the modality of choice for the evaluation of glenohumeral instability. The primary factors contributing to the shift from conventional arthrography to MRI are the superb soft tissue contrast and multiplanar capabilities provided by MRI. This allows a global assessment of the painful shoulder, including the rotator cuff,

Bu

Figure 17H3-3   Rotator cuff tear, shoulder arthrogram. Frontal external rotation radiograph demonstrates air and radiographic contrast, with the subacromial-subdeltoid bursa (Bu) indicating a complete tear with extension of contrast through the cuff defect.

950

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

t­ ypically begins with the appropriate scout films of the affected shoulder. This ensures proper radiographic technique and identifies abnormalities, such as soft tissue calcifications, that may be obscured by the radiopaque contrast agent. Standard scout films include AP projections of the shoulder in internal and external rotation, an axillary lateral projection, and a bicipital groove view. The patient is then placed supine on the fluoroscopic table with the arm positioned next to the body in slight external rotation. The anterior skin is prepared and draped in a sterile manner, and the site of joint puncture is determined fluoroscopically. A point is chosen overlying the lower third of the humeral head about 0.5 cm lateral to the medial cortex of the humeral head. The skin is anesthetized with 1% lidocaine (Xylocaine) with a 25-gauge ¾-inch needle. Deeper anesthesia is achieved with a 22-gauge 1½-inch needle. A 22-gauge 3½-inch spinal needle is then used to enter the joint. Correct position is verified fluoroscopically by injecting a small amount of radiopaque contrast material. If the needle is correctly positioned within the joint, the contrast material will outline the medial surface of the humeral head and spill into the subscapularis recess. A total of 2 to 5 mL of radiopaque contrast material is injected, followed by 10 mL of air. A total volume of 12 to 15 mL of air and contrast material provides adequate distention of the shoulder joint without undue discomfort. Injection of a larger volume often leads to decompression of the joint through a weak point of the capsular insertion along the medial aspect of the subscapularis recess, resulting in leakage of contrast material and air into the adjacent soft tissues. This degrades the quality of the examination and should be avoided if possible. Rotator cuff tears are identified fluoroscopically as the contrast agent leaks through the defect. The diagnosis of adhesive capsulitis is made by identifying increasing resistance to the contrast material with a small injected volume as well as by noting an abnormally small axillary pouch and a small or absent subscapularis recess. If increased pressure is encountered during injection of the contrast agent, fluoroscopic observation should be performed to ensure that the needle tip remains intra-­articular. If the contrast agent is intra-articular, the diagnosis of adhesive capsulitis is made, injection of contrast material is ceased, and the needle is withdrawn. After injection of the contrast agent, spot radiographs are obtained of any abnormality noted fluoroscopically. Radiographs are then obtained, repeating the AP projections of the shoulder in internal and external rotation, an axillary lateral projection, and a bicipital groove view. If no abnormality of the rotator cuff is noted on these images, the patient exercises the arm for a total of 5 minutes, and the same views are repeated. The exercise portion of the examination is important because an inadequately stressed articulation is the most common reason for a false-negative study result. Spot radiographs can be obtained of any abnormality to localize and quantify the rotator cuff tears more precisely. Although conventional shoulder arthrography was in the past considered the gold standard for identifying fullthickness tears of the rotator cuff, it has many limitations, and MRI has largely replaced conventional arthrography in the evaluation of the rotator cuff. It is less invasive and provides a more thorough evaluation of the shoulder. When shoulder arthrography is performed, it is now frequently

combined with a more advanced imaging modality, such as computed tomography (CT) or MRI, in an attempt to evaluate the labrum and capsular structures better.

Computed Tomography CT of the shoulder is performed primarily as a means of evaluating the osseous structures following trauma.8,9 Multidetector computed tomographic examinations with sagittal and coronal reconstructions can accurately detect the extent of humeral head and neck fractures. The precise number of fracture fragments as well as the extent of displacement and angulation of fracture fragments can accurately be depicted. The scapula is a complex anatomic structure composed of the body, coracoid and acromion processes, and the glenohumeral articular surface. As a result, the full extent of scapular fractures is difficult to describe fully using conventional radiography, whereas the multiplanar capabilities of CT make it ideal for evaluation of complex scapular fractures. Following glenohumeral dislocation, computed tomographic examination is the study of choice to depict the size and position of a glenoid rim fracture fragment, which can be an important part of presurgical planning (Fig. 17H3-4).

Computed Arthrotomography Computed arthrotomography of the shoulder was widely used in the past and for years was considered the gold standard in the imaging of labral abnormalities.10-13 The combination of CT with intra-articular injection of contrast material provides a highly sensitive method of evaluating the glenoid labra and articular cartilage. It has been shown to be as sensitive as conventional shoulder arthrography in the detection of full-thickness rotator cuff defects and provides improved accuracy in detecting abnormalities of the glenoid labrum.14,15 Over the past decade, MRI and MR arthrography largely replaced CT arthrography in the evaluation of rotator cuff abnormalities and glenohumeral instability.16,17 However, with the development of multidetector CT and the capability to reconstruct images in any imaging plane, we have seen a resurgence of computed arthrotomography of the shoulder. It is especially useful in the evaluation of the glenoid labrum in a patient with a contraindication to MRI and for quantifying the extent of an osseous Bankart lesion.18 The technique of joint injection for computed arthrotomography14,19-21 is similar to that for conventional ­doublecontrast shoulder arthrography. A smaller volume of contrast material (usually 2 to 3 mL) and a smaller total volume (10 to 12 mL) are used to prevent spontaneous decompression of the joint through the subscapularis recess, which results in a significant degradation of the diagnostic accuracy of the study. The iodinated contrast agent is mixed with 0.1 to 0.3 mL of 1:1000 epinephrine to prolong retention of the contrast agent within the joint to allow adequate time for transportation of the patient to the CT scanner and imaging. CT should be performed as soon as possible after injection of the contrast agent to minimize resorption of the air and contrast material. Two sets of contiguous 3-mm axial images are obtained through the shoulder with the arm in internal and external rotation.

Shoulder

A

C

Sagittal and oblique coronal reformatted images may also be obtained to provide a better evaluation of the rotator cuff and superior labrum.11,22,23 Although CT arthrography clearly depicts rotator cuff abnormalities and many of the lesions associated with glenohumeral instability, MRI and MR arthrography improve sensitivity in the detection of many labral abnormalities, and they use no ionizing radiation. Relative to CT arthrography, MRI and MR arthrography provide superior soft tissue contrast and multiplanar imaging capabilities, resulting in a more accurate depiction of nondisplaced labral tears as

951

B

Figure 17H3-4   Osseous Bankart lesion demonstrated on CT of the shoulder. A, Axial computed tomographic scan shows a large osseous Bankart lesion with a slightly displaced and comminuted fracture fragment (arrow) involving the inferior glenoid rim. B, Sagittal reconstruction shows the size of the osseous defect (arrows) of the inferior glenoid rim. C, Three-dimensional reconstruction image shows the relationship of the fracture fragment (short arrows) with the glenoid rim osseous defect (long arrows).

well as tendinopathy and partial-thickness tears of the rotator cuff. Another limitation of CT arthrography is that air is rapidly absorbed into the adjacent soft tissues, and any delay in imaging will significantly diminish the diagnostic accuracy of the examination.

Ultrasonography Sonography of the rotator cuff (Fig. 17H3-5) was first popularized in the 1980s as a simple and noninvasive method of evaluating the rotator24-28 and in recent years has been

952

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

repopularized as an accurate and cost-effective method for evaluation of the rotator cuff.29-32 Ultrasound examination of the shoulder requires a high-resolution transducer (7.5 to 10 MHz). Evaluation of the shoulder is performed with the patient in the sitting position. The examination begins with the arm in neutral (thumb up) position. In this position, the bicipital tendon (see Fig. 17H3-5C) is clearly visualized at the superolateral margin of the shoulder, and the presence or absence of fluid can be noted in the ­ subacromial-subdeltoid bursa. Before evaluation of the rotator cuff, the shoulder is rotated internally, and the arm is placed behind the back. This maneuver results in retraction of the critical portion of the supraspinatus tendon from beneath the acromion, allowing maximal visualization of this portion of the rotator cuff.33 The rotator

A

C

cuff should be evaluated in both the sagittal and coronal planes. The normal rotator cuff (see Fig. 17H3-5A and B) is sharply defined, uniform in thickness, and homogeneous in echo texture; it measures 4 to 6 mm in thickness anteriorly, normally being somewhat thinner posteriorly.25 A thin echogenic band paralleling the upper surface of the cuff characterizes the subacromial-subdeltoid bursa.34 The overlying deltoid muscle is characterized by a speckled appearance that is distinct from the normal overlying cuff (Table 17H3-1). Although sonography of the rotator cuff has been shown to be nearly as accurate as MRI in the evaluation of the rotator cuff, it is clear that sonographic evaluation of the shoulder has a steep learning curve and is very operator

B

Figure 17H3-5   Normal shoulder sonography. Images from a normal shoulder demonstrate the appearance of the supraspinatus tendon in axial   (A) and sagittal (B) internal rotation projections.   C, The bicipital groove (arrows) is demonstrated in   an axial sonographic section. BT, bicipital tendon; GT, greater tuberosity; H, humeral head; SST, supraspinatus tendon.

Shoulder

  TABLE 17H3-1  Sonographic Imaging Signs of Rotator Cuff Disease Tendinopathy Partial-thickness tear Full-thickness tear Subacromial subdeltoid bursitis

Thickened and heterogeneous tendon Focal change in echogenicity Focal thinning of tendon Nonvisualization of the cuff Focal defect or absence of the cuff Discontinuity of the cuff Band of decreased signal superficial to cuff

dependent. This, coupled with the fact that MRI provides a more global evaluation of the shoulder, including the labrum and osseous structures, has led to the replacement of shoulder sonography by MRI in most practices.

Magnetic Resonance Imaging During the past decade, MRI has evolved into the imaging modality of choice for the evaluation of shoulder impingement.35-45 It offers soft tissue contrast that is superior to any other imaging modality and provides an excellent overall assessment of the osseous outlet and acromion, tendons, muscles, capsular structures, and labrum. The addition of intra-articular contrast material has also been shown to increase sensitivity in the detection of undersurface partialthickness tears of the rotator cuff  36,38,46 and abnormalities of the labrum.36,47 The type of magnet and various pulse sequences that are used to evaluate the shoulder differ according to equipment availability and the preference of the imager. Numerous commercial coils are suitable for MRI of the shoulder. Surface coils come in a variety of configurations and are generally adequate for most clinical applications. Recent advances in coil design include the quadrature and phased-array coils, which offer the highest signal-to-noise ratio and best overall image quality. The arm is typically positioned next to the side; for specialized views, it may be placed over the head. The arm should not be positioned on the abdomen because this transmits respiratory motion to the shoulder, with subsequent image degradation. The arm should be placed in the neutral position with the thumb up to provide the most comfort for the patient, thus limiting motion artifact. External rotation tends to be uncomfortable for the patient and tightens the anterior capsule, resulting in poor visualization of the anterior labrum.48,49 Internal rotation may obscure the posterior labrum. The abduction and external rotation (ABER) position reportedly improves visualization of the undersurface of the rotator cuff as well as of the anterior band of the inferior glenohumeral ligament and the anterior labroligamentous complex.49-51 The shoulder is usually imaged in three planes: the axial, oblique sagittal, and oblique coronal planes. A typical imaging protocol includes a T1-weighted sequence in the oblique coronal plane (Table 17H3-2). The T1-weighted images provide the best overall signal-to-noise ratio and

953

  TABLE 17H3-2  Magnetic Resonance Imaging Pulse ­Sequence Characteristics Sequence

Image

T1-weighted image

Anatomy image High signal-to-noise ratio

T2-weighted image

Characteristics

Fat—bright Muscle—intermediate Water—intermediate Calcium/fibrous tissue—dark Pathology image Fat—bright Lower signal-to-noise Muscle—intermediate ratio Water—bright Calcium/fibrous tissue—dark

are thus excellent for depicting anatomy. T1-weighted images also nicely demonstrate the osseous anatomy of the shoulder, and because fat is bright, this sequence will demonstrate fatty atrophy of the rotator cuff. Fast spin-echo T2-weighted images with frequency-selective fat saturation are commonly obtained in the oblique coronal and oblique sagittal planes. Fluid is bright on T2-weighted images, thus allowing the most accurate detection of shoulder disease, including rotator cuff abnormalities, labral tears, bone marrow edema, and paralabral cysts. Axial imaging may be performed with use of either fast spin-echo T2-weighted or gradient-echo sequences and typically provides the best opportunity for evaluation of the anterior and posterior labra as well as of the glenohumeral ligaments and subscapularis tendon. The use of low field strength magnets has been steadily gaining acceptance in the orthopaedic community as a cost-effective means of performing MRI of the extremities and in particular the shoulder. The advantages of a low field strength magnet include a large-bore opening of the magnet resulting in better patient acceptance and less difficulty with claustrophobia. As a result, the low field strength systems are less constrained by patient size and body part. The lower strength of the magnet, however, results in decreased signal-to-noise ratio and longer scan times. The images obtained on a low field strength system are generally considered “less pretty” than those images obtained on a high field strength system, but several studies have demonstrated comparable accuracy for depiction of rotator cuff pathology (see Table 17H3-3). High field strength systems, however, are still considered superior and provide higher diagnostic accuracy for the detection of cartilage and labral lesions in the shoulder.52

Magnetic Resonance Arthrography Although conventional MRI has been established as the imaging modality of choice in the evaluation of shoulder impingement syndrome, MR arthrography is steadily gaining acceptance as the method of choice for the evaluation of glenohumeral instability.36,47 Many of the lesions associated with instability are subtle and may undergo partial healing. This, coupled with the fact that the normal anatomic structures of the shoulder lie in proximity, can make the diagnosis of these lesions difficult even with high-quality conventional MRI.53-56 MR arthrography, on

954

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the other hand, is performed by distending the joint with fluid (saline or dilute gadolinium), which more ­accurately depicts subtle labral tears, cartilage and ligamentous abnormalities, and partial-thickness tears of the undersurface of the rotator cuff. The standard injection technique for MR arthrography of the shoulder is similar to that for conventional doublecontrast shoulder arthrography or computed arthrotomography described in previous sections. The injection is typically performed through an anterior approach under fluoroscopic guidance with a 22-gauge 3½-inch spinal needle. A small amount (1 to 2 mL) of an iodinated contrast agent is first injected to confirm intra-articular placement of the needle tip. About 12 to 15 mL of gadolinium diluted 1:200 with normal saline is then injected. The patient is taken directly to the MRI unit, and imaging is initiated within 30 minutes to avoid excessive resorption of the intra-articular gadolinium. Unlike postinjection exercise in conventional shoulder arthrography, which increases the sensitivity of detection of subtle tears, exercising the shoulder after injection of gadolinium is neither beneficial nor detrimental.57 Imaging protocols vary, but a standard set of imaging sequences typically includes T1-weighted images with frequency-selective fat saturation in the axial, oblique sagittal, and oblique coronal planes. The T1-weighted images have a high signal-to-noise ratio, resulting in the exquisite anatomic detail that is critical in detecting the subtle lesions associated with glenohumeral instability. A T2-weighted sequence performed in the oblique coronal plane is the most important sequence for depicting ­rotator cuff abnormalities. A T2-weighted sequence may also be helpful in detecting other pathologic processes, such as a paralabral cyst (see Figs. 17H3-21B to 17H3-29) or bone marrow edema (see Fig. 17H3-21B to 17H3-40C). The ABER view may be added to the standard imaging protocol for any patient thought to have an anterior labral pathologic process.50

IMAGING OF SPECIFIC SHOULDER ABNORMALITIES The shoulder is a complex joint, and the interpretation of a shoulder MRI examination is complicated, requiring the evaluation of numerous images obtained in several imaging planes and with various pulse sequences. A comprehensive MRI examination of the shoulder requires a search pattern that encompasses all of the pertinent anatomic structures. The following sections provide a systematic approach for MRI evaluation of the shoulder that includes a thorough review of the basic anatomy as well as of the imaging basis for recognizing common pathologic processes; when appropriate, complementary imaging modalities are discussed. The important structures that must be thoroughly evaluated on each MRI examination include: • Osseous outlet and acromion • Rotator cuff • Labrum and capsular structures • Biceps tendon • Osseous structures and articular surfaces

Osseous Outlet and Acromion The clinical syndrome of shoulder impingement refers to a painful compression of the soft tissues of the anterior shoulder (rotator cuff, subacromial bursa, and bicipital tendon) between the humeral head and the coracoacromial arch (coracoid process, acromion process, coracoacromial ligament, and AC joint).58-60 Pain occurs when the arm is elevated forward and internally rotated or placed in the position of abduction and external rotation.59 In the normal shoulder, the powerful upward pull of the deltoid on the proximal humerus is resisted by an intact rotator cuff so that the humeral head remains centered on the glenoid in all arm positions. If this stabilizing mechanism becomes weakened secondary to repeated trauma, overuse, or age, the humeral head is pulled upward under the structures of the coracoacromial arch. Impingement is initially followed by subacromial bursitis and rotator cuff tendinopathy. Over time, irreversible cuff trauma occurs with fibrosis and degeneration. In the latter stages, a bony excrescence (subacromial enthesophyte) tends to form at the anteroinferior margin of the acromion where the coracoacromial ligament inserts. Tears of the rotator cuff are frequent in this stage, undoubtedly in part owing to direct cuff trauma from the spur. After massive tears of the rotator cuff, bone-to-bone contact may result between the humeral head and the undersurface of the anterior third of the acromion, causing sclerosis and proliferative changes in this area. Neer60 first introduced the term impingement syndrome and described three stages in the rotator cuff disorder. Reversible hemorrhage and edema characterize stage 1, typically seen in individuals younger than 25 years. Stage 2 consists of cuff fibrosis and tendonitis and occurs in individuals between 25 and 40 years of age. In stage 3, osteophyte formation occurs along the anteroinferior margin of the acromion, and rotator cuff tears are common; this group is typically older than 40 years. Impingement is common in young athletes who participate in sports requiring repetitive overhead activities, such as tennis, baseball, and swimming. Neer postulated that chronic impingement is the most common cause of rotator cuff tear,60 although other authors have suggested that additional causes, including degeneration of the cuff secondary to aging, acute trauma, and inflammatory diseases, may also result in rotator cuff tear.61,62 The diagnosis of impingement syndrome is usually a clinical one based on appropriate historical and physical examination findings. A thorough history and physical examination by an experienced physician have an 84% to 90% sensitivity and a 75% to 95% specificity for diagnosis of a tear of the rotator cuff.45,63,64 Many imaging modalities are available to assist in the evaluation of the progressively painful shoulder, and their role is both to assess the extent of abnormality of the rotator cuff and to identify configurations of the osseous outlet that may predispose to rotator cuff impingement. Conventional radiography has met with limited success in the evaluation of the clinical syndrome of impingement. Many of the osseous changes that occur with impingement are seen late in the process and thus offer little in establishing an early diagnosis and preventing progression of the associated soft tissue injuries. Osseous abnormalities that

Shoulder

may be associated with the clinical syndrome of impingement include the following59,65-67:

955

MRI provides an excellent evaluation of the entire osseous outlet. This is due to its multiplanar capabilities and its ability to demonstrate the relationship of the entire osseous outlet to the underlying rotator cuff. Bigliani and colleagues65

described three different acromial shapes (Fig. 17H3-8) and related the configuration of the undersurface of the acromion to the presence of rotator cuff tears. A type I acromion (see Fig. 17H3-8A) has a flat undersurface, a type II acromion (see Fig. 17H3-8B) has a curved undersurface, and a type III acromion (see Fig. 17H3-8C) has an anterior hook. The acromial types II and III have an increased association with rotator cuff tear.65,66 A type IV acromion with a convex undersurface has subsequently been described, but no definite correlation has been shown to exist between the type IV acromion and impingement.73 On MRI, the shape of the acromion (see Fig. 17H3-8) is best assessed on the oblique sagittal view just lateral to the AC joint. One study, however, suggests poor correlation of acromial arch shape between conventional radiography and MRI.74 Anterior and lateral down-sloping of the anterior acromion (Fig. 17H3-9) can also narrow the supraspinatus outlet and potentially result in impingement.75-77 Anterior down-sloping (see Fig. 17H3-9B) is demonstrated on oblique sagittal MRI; lateral down-sloping (see Fig. 17H3-9C) is best seen on the oblique coronal images. An enthesophyte (Fig. 17H3-10) extending off the anteroinferior aspect of the acromion can also be clearly demonstrated on MRI. It typically appears as a marrow containing osseous excrescence (see Fig. 17H3-10), which should have MRI signal characteristics similar to the adjacent acromion marrow (bright on T1-weighted images). Potential pitfalls include the attachment of the coracoacromial ligament and the deltoid tendon insertion (see Fig. 17H3-9A and C) on the anterior acromion. These structures may mimic an osseous excrescence, but they can be differentiated from enthesophyte because they lack marrow signal and appear dark on all pulse sequences. The acromion should also be evaluated for os acromiale (Fig. 17H3-11). This is an accessory ossification center along the outer edge of the anterior acromion. It is normally fused by 25 years of age. There is an

Figure 17H3-6   Subacromial spur, conventional radiography. A frontal radiograph of the shoulder with a 30-degree caudal tilt demonstrates a large hook-like bony excrescence (s) arising from the anteroinferior margin of the acromion process in this patient with shoulder impingement syndrome.

Figure 17H3-7   Chronic rotator cuff tear, conventional radiography. A frontal radiograph of the shoulder demonstrates a high-riding humeral head (arrows) resulting in remodeling of the undersurface of the acromion and sclerosis of the greater tuberosity. An osteophyte (arrowhead) extends off the inferior aspect of the humeral head, resulting from osteoarthritis of the glenohumeral joint.

• Enthesophyte formation on the anteroinferior aspect of the acromion • Long anterior portion of the acromion with anterior or lateral down-sloping of the acromion • Unfused os acromiale • Hypertrophy of the AC joint Optimal conventional radiographic views have been described for identifying these variations of the osseous outlet.68-71 The anteroposterior radiograph at a 30-degree caudal angle (Fig. 17H3-6) is helpful in visualizing the anterior aspect of the acromion and in detecting inferiorly directed enthesophytes.69 A modified transcapular lateral view obtained with 10 to 15 degrees of caudal angulation (supraspinatus outlet view) helps further identify the anteroinferior aspect of the acromion.72 Fluoroscopy has also been shown to aid in the detection of subacromial enthesophytes.70 A high-riding humeral head with remodeling of the undersurface of the acromion and sclerosis of the greater tuberosity are conventional radiographic findings (Fig. 17H3-7) that are pathognomonic of a chronic rotator cuff tear. Although conventional arthrography, ultrasonography, and CT offer improved visualization of the rotator cuff relative to conventional radiography, they add little in the direct evaluation of the osseous outlet.

Magnetic Resonance Imaging of the Osseous Outlet and Acromion

956

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

association between persistent os acromiale and impingement of the rotator cuff.78-82 The deltoid muscle attaches to the inferior aspect of the accessory ossicle, and contraction of the deltoid results in a downward motion of the unstable segment, potentially leading to impingement of the underlying rotator cuff.80 Os acromiale is demonstrated best on axial images (see Fig. 17H3-11A), but it can also be seen on oblique sagittal or oblique coronal images (see Fig 17H3-11B),81 on which it should not be confused with the adjacent AC joint. MRI signs of instability of the os acromiale include fluid signal within the synchondrosis or sclerosis, cystic change, or marrow edema on either side of the synchondrosis (Box 17H3-2). Hypertrophic changes of the capsule and ­ inferiorly directed osteophyte formation of the AC joint (Fig. 17H3-12) can also be associated with impingement.59 MRI clearly demonstrates these changes as well as associated mass effect on the underlying rotator cuff. The ­coracoacromial ligament (Fig. 17H3-13) is a soft tissue structure that forms part of the coracoacromial arch. It

B

Figure 17H3-8   Acromial types, magnetic resonance imaging, oblique sagittal T1-weighted images of the acromion one section lateral to the acromioclavicular joint. A, Type I acromion demonstrates a flat undersurface (arrows). B, Type II acromion demonstrates a gentle curvature to the undersurface of the acromion (arrows). C, Type III acromion demonstrates a hook (arrow) extending off the anterior aspect of the acromion.

extends from the coracoid to the acromion and is well seen on oblique sagittal MRI. It normally measures less than 2 mm in thickness (see Fig. 17H3-13A) and extends across the rotator interval and anterior aspect of the supraspinatus tendon. The role of the coracoacromial ligament in impingement remains controversial; some believe that thickening or ossification (see Fig. 17H3-13B) of the ligament may be a potential cause of impingement, whereas others believe that thickening results from impingement.83,84 Coracohumeral impingement is an uncommon cause of extrinsic impingement that results from a narrowed distance between the coracoid process and the underlying humeral head. The normal coracohumeral distance should be greater than 10 mm as seen on axial MRI. A narrowed coracohumeral distance results in entrapment of the ­subscapularis tendon between the coracoid process and the humeral head and can lead to isolated tendinosis and disruption of the subscapularis tendon. Abnormalities isolated to the subscapularis tendon should prompt an investigation of the coracohumeral distance.85

Shoulder

A

C

Rotator Cuff The rotator cuff is composed of four tendons: the supraspinatus superiorly, the subscapularis anteriorly, and the infraspinatus and teres minor posteriorly. These tendons are important dynamic stabilizers of the glenohumeral joint, and any review of rotator cuff disease will rapidly expose the supraspinatus tendon as the weak link of the rotator cuff. Most cuff failures originate in the ­ supraspinatus ­tendon at or near its insertion onto the greater tuberosity of the humeral head.86 Many investigators attribute this propensity for cuff failure within the supraspinatus tendon to its blood supply. The supraspinatus tendon receives its arterial supply from the anterior humeral circumflex, subscapular, ­suprascapular, and posterior humeral ­circumflex arteries.87,88 A zone of relative avascularity has been described in the tendon proximal to its attachment site and may represent a “critical zone” for cuff failure.80,87 Other authors have found this zone to be richly vascularized by anastomosing vessels from the tendon and humeral tuberosity.89 Arterial filling of the cuff vessels in the critical zone depends greatly on the position of the arm; poor filling is

957

B

Figure 17H3-9   Acromial down-sloping, magnetic resonance imaging. A, Oblique sagittal T1-weighted image demonstrates no evidence of anterior down-sloping (arrows). The deltoid tendon (arrowhead) is noted as it inserts onto the anterior aspect of the acromion. B, Oblique sagittal T2-weighted image demonstrates marked anterior down-sloping of the acromion (arrows), with resulting mass effect on the underlying supraspinatus muscle. C, Oblique coronal image demonstrates lateral down sloping of the acromion (white arrow). The black arrow demonstrates the insertion site of the deltoid tendon onto the anterolateral acromion.

present when the arm is adducted.90 A high correlation has also been shown to exist between rotator cuff tears and subacromial impingement.60 It is probably a combination of avascularity and subacromial impingement that leads to most rotator cuff abnormalities originating in the critical zone of the supraspinatus tendon. Conventional radiography plays only a limited role in the direct evaluation of the rotator cuff, although it is frequently the initial imaging study performed for patients with the clinical syndrome of impingement. Radiographs allow identification of associated pathologic change, especially of the osseous outlet and acromion. Conventional radiography findings associated with cuff pathology include the following2,91,92: • Calcific tendinitis • Calcific bursitis • High-riding humeral head (<7 mm between the humeral head and the undersurface of the acromion) associated with chronic rotator cuff tear (see Fig. 17H3-7) • Scalloping and loss of the normal convexity of the undersurface of the acromion (see Fig. 17H3-7)

958

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

outlet, labrum and capsular structures, and articular surfaces.94 In addition, it is noninvasive, readily available, and capable of detecting the full spectrum of rotator cuff abnormalities from tendinosis to partial-thickness or full-thickness tear.35,42,45,94-96 MRI is also superior to ultrasonography and computed arthrotomography in the evaluation of the rotator cuff.97 This is primarily a result of the multiplanar capabilities of MRI and its superior soft tissue contrast.

Magnetic Resonance Evaluation of the Rotator Cuff

Figure 17H3-10   Acromial enthesophyte, magnetic resonance imaging. Oblique sagittal image demonstrates marrow containing osseous excrescence (arrowheads) extending off the anterior aspect of the acromion (arrow) Note that an enthesophyte contains marrow signal that is bright on T1-weighted images. This differs from the deltoid tendon attachment, which is of low signal intensity on all pulse sequences (see Fig. 17H3-9A and 17H3-9C).

Conventional shoulder arthrography, once considered the imaging gold standard for detecting a full-thickness tear (see Fig. 17H3-3) of the rotator cuff,93 has now been largely replaced by MRI. MRI offers a more complete evaluation of the shoulder, including the coracoacromial

A

The normal anatomy of the rotator cuff is accurately depicted with the multiplanar capabilities of MRI. The supraspinatus muscle originates along the posterosuperior portion of the scapula above the level of the scapular spine. A single tendon arises out of the muscle and extends superiorly above the humeral head to insert onto the greater tuberosity of the humeral head. The supraspinatus tendon is best evaluated on oblique coronal and oblique sagittal MRI. The subscapularis is located anteriorly and has multiple tendon slips. It has a broad origin along the anterior aspect of the scapula and attaches to the lesser tuberosity on the anterior aspect of the humeral head. An extension of the subscapularis tendon known as the transverse ligament extends across the intertubercular groove and helps to stabilize the long head of the biceps tendon within the intertubercular groove. The subscapularis muscle and tendon are best evaluated on axial MRI. The infraspinatus is located posterosuperiorly and has a broad origin along the posterior aspect of the scapula inferior to the scapular spine. The teres minor is located posteroinferiorly, below

B

Figure 17H3-11   Os acromiale, magnetic resonance imaging. A, Axial T2-weighted section at the level of the acromioclavicular joint demonstrates an accessory ossification center, the os acromiale (long arrow), which remains unfused to the remainder of the acromion (short arrows). AC, acromioclavicular joint. B, T2-weighted oblique coronal image of the same patient demonstrates an unfused os acromiale (arrow). The presence of marrow edema on either side of the os acromiale suggests an unstable synchondrosis.

Shoulder

959

Box 17H3-2  Magnetic Resonance Imaging  Findings of the Acromion  Associated with Extrinsic  Impingement Acromion type—assessed on the sagittal imaging plane on the image immediately lateral to the acromioclavicular joint Type I: flat undersurface Type II: curved undersurface Type III: anterior hook Type IV: convex undersurface Lateral down sloping—assessed on the same image as used to type the acromion Anterior down sloping—assessed on the coronal images Acromial spur—best assessed on T1 images (bright fatty marrow signal within the bony excrescence ­differentiates a spur from the deltoid tendon slip) Unfused os acromiale—best assessed on axial ­images ­(fluid signal within the synchondrosis suggests ­instability)

the level of the infraspinatus, originating along the axillary surface of the scapula and inserting on the most inferior aspect of the greater tuberosity of the humeral head. The infraspinatus and teres minor are best evaluated on oblique coronal and oblique sagittal images. Evaluation of the rotator cuff typically begins with T1-weighted images. These images possess a high signalto-noise ratio and are best suited for demonstrating cuff anatomy. The normal tendons demonstrate uniformly dark signal intensity on all pulse sequences, and increased signal on T1-weighted images is a nonspecific finding that may represent a wide array of conditions ranging from artifact

A

Figure 17H3-12   Osteoarthritis of the acromioclavicular joint. Oblique coronal T1-weighted image shows capsular hypertrophy (short arrow). There is an inferiorly directed osteophyte (long arrow) that results in mass effect on the underlying supraspinatus tendon.

to complete tear. The T2-weighted images have a lower signal-to-noise ratio and provide less anatomic detail, but they are both sensitive and specific for depicting the full range of rotator cuff abnormalities. There are many causes of increased signal within the rotator cuff tendons on T1-weighted images. When increased signal is noted on T1-weighted images, the morphologic features of the tendon as well as the T2-weighted

B

Figure 17H3-13   Coracoacromial ligament, magnetic resonance imaging. A, Oblique sagittal T1-weighted image demonstrates a normal coracoacromial ligament (arrows) extending from the coracoid process to the anterior acromion. The normal ligament measures less than 2 mm in thickness. B, Oblique sagittal T1-weighted image demonstrates a thickened nodular-appearing coracoacromial ligament (arrows).

960

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 17H3-3  Magnetic Resonance Imaging Appearance of Rotator Cuff Pathology Cuff Pathology

Magnetic Resonance Imaging ­Appearance

Normal tendon Tendinopathy

Low signal on T1 and T2 Intermediate signal on T1 and T2 Possible thickening of tendon Globular decreased signal on T1 and T2 within tendon Surrounding edema common Tendon intermediate signal on T1 and T2 “Blooming” artifact gradient echo imaging Fluid signal and gadolinium extending partially through thickness of tendon Bursal, articular, interstitial, intramuscular cyst No retraction of tendon Fluid extending completely through tendon superior to inferior Retraction of tendon Gap or discontinuity in tendon Measured as the length of the medial-to-lateral tendon gap Graded as mild, moderate, or severe Streaks of high signal on T1 (fatty streaking): irreversible Loss of muscle bulk: reversible

Calcific tendinitis

Partial-thickness tear

Full-thickness tear

Musculotendinous retraction Fatty atrophy

images should be evaluated. Normal tendon appearance and normal signal intensity on T2-weighted images ­suggest that the tendon is probably normal. Potential causes of increased signal within a normal tendon on the T1-weighted images include partial volume averaging and “magic angle” phenomenon. This MRI artifact results

A

in increased signal in normal tendons that are angled 55 degrees relative to the direction of the main magnetic field. It occurs primarily on T1-weighted images, and as echo time lengthens, as occurs with T2 weighting, the signal intensity again decreases. This MRI artifact is particularly likely to occur in the critical zone of the supraspinatus tendon and therefore occasionally causes confusion about the integrity of the cuff.64 There is a wide spectrum of rotator cuff pathology that can be accurately depicted on MRI ranging from tendinosis to a full-thickness tear of the rotator cuff (Table 17H3-3). Tendinosis (Fig. 17H3-14) is noted as intermediate signal on T1-weighted MRI that persists on T2-weighted MRI. The signal abnormality does not reach the same brightness as fluid on T2-weighted images (see Fig. 17H3-14B). The tendon may also demonstrate diffuse or focal thickening, but there will be no evidence of tendon ­disruption. Tendinosis represents a degenerative process of the tendon, which histologically represents a combination of inflammation and mucoid degeneration.98 On arthroscopic examination, the tendon may appear to be edematous and hyperemic, with occasional fraying, roughening, or degeneration of the surface of the tendon, and it may be difficult at times to differentiate tendinosis from early partial-thickness tearing both at arthroscopy and on MRI.99 Partial-thickness tears (Fig. 17H3-15) can occur on either the articular surface or the bursal surface or within the substance of the tendon. The articular surface tears are the most common type to occur and are the only type of partial-thickness tear demonstrated on conventional shoulder arthrography. A partial-thickness tear is seen on MRI as an area of increased signal on T1-weighted images (see Fig. 17H3-15A) that increases to fluid signal intensity on T2-weighted images (Fig. 17H3-15B). The fluid ­signal intensity extends only partially through the thickness of the tendon from superior to inferior. These tears may partially heal with granulation tissue, making them difficult to distinguish from tendinopathy on MRI. Direct MR

B

Figure 17H3-14   Supraspinatus tendinopathy, magnetic resonance imaging. A, Oblique coronal T1-weighted image reveals mild thickening of the supraspinatus tendon as well as a focal area of increased signal intensity (arrows) within the tendon, adjacent to its attachment site on the greater tuberosity. B, Persistence of the intermediate signal (arrows) within the tendon on the T2-weighted oblique coronal image confirms the diagnosis of tendinopathy.

Shoulder

A

961

B

Figure 17H3-15   Rotator cuff, partial-thickness articular-sided tear, magnetic resonance imaging. A, Oblique coronal T1-weighted image demonstrates thinning of the supraspinatus tendon adjacent to its attachment on the greater tuberosity. In addition, there is abnormal intermediate signal (arrow) involving the undersurface of the tendon. B, On the T2-weighted oblique coronal image, there is fluid-bright signal (arrows) involving the undersurface of the tendon, confirming the presence of an articular surface partialthickness tear.

arthrography is more sensitive than conventional MRI in detecting articular surface tears (see Fig. 17H3-40A); however, it does not increase sensitivity for detecting ­partial­thickness tears of the bursal surface. It has been suggested that the sensitivity for detecting partial-thickness undersurface tears can be further improved by adding the ABER view to the MR arthrography protocol.100 Interstitial tears of the rotator cuff represent tears that occur within the substance of the tendon but do not involve the bursal or the articular surface (Fig. 17H3-16). These tears are ­demonstrated on MRI as a focal linear area of fluid signal that is contained within the substance of the tendon, and they typically occur at the footprint of the supraspinatus tendon as it attaches to the greater tuberosity of the humeral head. A partial articular-sided supraspinatus tendon avulsion (PASTA) lesion is a subset of partial thickness tears that has been recently described in the literature.101-103 The tear represents a partial-thickness articular-sided avulsion of the supraspinatus tendon at its most anterior attachment site. This type of tear deserves special attention and should be accurately described on MRI as the recommended treatment for this subset of tendon tears differs from the standard partial-thickness tears described previously. A transtendon suture technique is performed to preserve the intact portion of the tendon while firmly reattaching the torn portion of the tendon to the humeral footprint. On MRI, a small articular-sided avulsion is seen as fluid signal extending into the articular surface of the supraspinatus tendon at its anterior attachment site with partial avulsion of the tendon at this level, and this lesion represents a subset of the articular surface partial-thickness tears (Fig. 17H3-17). A full-thickness tear of the rotator cuff tendon (Fig. 17H3-18) is defined as a tear that extends through the complete thickness of the tendon from superior to inferior. This allows communication between the joint space and the subacromial-subdeltoid bursa. MRI criteria for

e­ stablishing the diagnosis of a full-thickness tear include high (fluid) signal completely traversing the tendon from superior to inferior on T2-weighted images, a gap or absence of the tendon, and retraction of the musculotendinous junction.

Figure 17H3-16   Rotator cuff, partial-thickness   interstitial tear, magnetic resonance imaging. Oblique coronal T2-weighted image demonstrates a focal fluid collection within the substance of the supraspinatus tendon (long arrow) representing an interstitial tear. Note that the bursal surface (short arrows) and the articular surface (arrowhead) are both intact.

962

A

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

Figure 17H3-17   Rotator cuff, partial articular-sided supraspinatus tendon avulsion or “PASTA” lesion, magnetic   resonance imaging. Oblique coronal (A) and oblique sagittal (B) T2-weighted images demonstrate a focal fluid collection   (long arrows) that extends to the articular surface of the supraspinatus tendon at the level of attachment to the humeral head.   Note that the bursal surface (short arrows) of the tendon remains intact. This is the typical magnetic resonance appearance of   a PASTA lesion.

A

B

Figure 17H3-18   Rotator cuff, full-thickness tear with musculotendinous retraction, magnetic resonance imaging. Oblique coronal T2-weighted image with fat saturation (A) demonstrates a full-thickness tear of the supraspinatus tendon, with fluid in the expected location of the tendon (short arrow), with marked retraction of the musculoskeletal junction (long arrow). Oblique sagittal T2-weighted image (B) shows fluid (short arrows) in the expected location of the supraspinatus tendon, representing a full-thickness tear. The infraspinatus tendon (long arrow) remains intact more posteriorly.

Shoulder

Some of the terminology with regard to atrophy of the rotator cuff musculature is currently in a state of flux. The term “fatty infiltration” is typically used to describe actual fatty infiltration or replacement of the muscle while “muscle atrophy” is used to describe a loss of muscle bulk. Muscle atrophy is seen on MRI as loss of muscle bulk (best depicted on sagittal images through the cuff musculature), while fatty infiltration is seen as streaks of high T1-weighted signal within the substance of the muscle (Fig. 17H3-19). When normal, the four muscles of the rotator cuff demonstrate symmetrical bulk and signal characteristics. Asymmetric loss of bulk indicates muscle atrophy, whereas streaks of high T1-weighted signal within the substance of the muscle indicate fatty infiltration. Goutallier and colleagues reported a system for grading fatty infiltration of the cuff musculature on the basis of CT imaging, and although this system has been widely adapted for use with MRI, a recent study by Fuchs and associates demonstrated poor correlation between CT and MRI with regard to grading fatty infiltration. Although there remains some controversy regarding the accuracy of MRI, fatty infiltration is typically graded as mild, moderate, or severe based on the extent of fatty infiltration (high T1-weighted signal) with the belly of the muscle. Muscle atrophy can be graded separately as mild, moderate, or severe on the basis of muscle bulk depicted on sagittal images at the level of the supraspinatus fossa. The extent of retraction can be determined by measuring the gap between the normal footprint of the rotator cuff tendon on the humeral head and the torn tendon end, using the oblique coronal images (see Fig. 17H3-18). Calcific tendinitis (hydroxyapatite crystal deposition ­disease) can also be diagnosed with MRI (Fig. 17H3-20). The crystalline deposits are typically within the critical zone of the rotator cuff and appear as areas of low signal intensity on both T1-weighted and T2-weighted images. Calcific deposits within the rotator cuff may be difficult to identify on MRI because both the tendon and the calcific deposit appear dark on all pulse sequences. Ancillary

Figure 17H3-19   Fatty atrophy of the supraspinatus muscle, magnetic resonance imaging. Oblique coronal T1-weighted image reveals streaks of high-signal fat (arrows) replacing the normal intermediate signal intensity of muscle.

963

MRI findings that may aid in the identification of calcific tendinitis include globular thickening of the involved tendon and high signal within the tendon and surrounding tissues secondary to associated inflammation. Gradient-echo imaging may improve conspicuity of the deposits secondary to the “blooming” artifact associated with local magnetic field heterogeneity. Use of gradient-echo imaging and comparison with conventional radiographs improve the likelihood of detecting calcific tendinitis. Most cuff tears originate in the supraspinatus tendon (see Fig. 17H3-18); however, large tears may extend into either the infraspinatus or subscapularis tendon. Isolated tears, although less common, occasionally occur in either the infraspinatus (see Fig. 17H3-40A) or subscapularis (Fig. 17H3-21) tendon. An isolated tear of the infraspinatus tendon is usually associated with the internal impingement syndrome (discussed further in the section on glenohumeral instability).104 An isolated tear of the subscapularis tendon may result from shoulder dislocation or in ­association with coracohumeral impingement105 and is best demonstrated on axial MRI as high signal traversing the tendon with retraction of the tendon away from its normal attachment site on the lesser tuberosity (see Fig. 17H3-21). An extension of the subscapularis tendon known as the transverse ligament holds the long head of the biceps tendon in the intertubercular groove, and a tear of the subscapularis tendon may result in disruption of the transverse ligament, leading to medial subluxation or dislocation of the long head of the biceps tendon.106,107 Axial MRI is well suited not only for evaluating the integrity of the subscapularis tendon but also

Figure 17H3-20   Calcific tendinitis and bursitis, magnetic resonance imaging. The calcification is visible as ovoid areas of low signal intensity within the supraspinatus tendon (long arrow) and within the subacromial-subdeltoid bursa (short arrow). Note edema within the adjacent soft tissues and the adjacent humeral head, representing adjacent inflammatory changes that often accompany calcific bursitis or calcific tendinosis.

964

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17H3-21   Subscapularis tendon disruption with biceps tendon dislocation, magnetic resonance imaging. Axial T2-weighted image with fat saturation through the glenohumeral joint demonstrates complete disruption and retraction of the subscapularis tendon (arrow). Complete disruption of the subscapularis tendon has resulted in dislocation of the biceps tendon (arrowhead), which is now located intra-articularly.

A

C

for demonstrating medial subluxation of the biceps tendon out of the intertubercular groove (see Fig. 17H3-21). An intramuscular cyst within the rotator cuff (Fig. 17H3-22) has been described as a finding associated with small full-thickness tears or partial-thickness articularsided tears of the rotator cuff.108 Intramuscular cysts are similar to paralabral cysts of the shoulder or meniscal cysts of the knee. Fluid leaks through a defect in the cuff and tracks in a delaminating fashion along the fibers of the tendon, resulting in a fluid collection contained within either the muscle or fascia of the rotator cuff. These cysts have been reported in the supraspinatus, infraspinatus, and subscapularis muscles and appear as oval lobulated collections of low signal intensity on T1-weighted images and high signal intensity on T2-weighted images. Identification of an intramuscular cyst of the rotator cuff should prompt a thorough search for a small associated cuff tear.108 Denervation of a rotator cuff muscle can result from either a compressive neuropathy or an acute traumatic injury of a nerve. Compressive neuropathies most commonly result from a paralabral cyst associated with a labral tear, but they can also be caused by fractures or other masses about the shoulder. Paralabral cysts (Fig. 17H3-23)

B

Figure 17H3-22   Intramuscular cyst of the rotator   cuff, shoulder magnetic resonance arthrography. A, Oblique coronal T1-weighted image demonstrates a partialthickness tear (large arrow) of the articular surface of the supraspinatus tendon. Contrast material extends in a laminar fashion (small arrows) through the fibers of the tendon and into the muscle. B, A second oblique coronal T1-weighted image slightly posterior to A demonstrates an oblong collection of contrast material (arrows) within the supraspinatus muscle. C, Oblique sagittal T1-weighted image demonstrates contrast material contained within the intramuscular cyst (arrow).

Shoulder

A

965

B

Figure 17H3-23   Paralabral cyst, magnetic resonance imaging. A, Coronal T2-weighted image with fat saturation demonstrates a lobulated cystic structure (long arrow) containing high T2-signal intensity extending into the suprascapular notch. It is arising from a superior labral tear (short arrow). B, Axial T2-weighted image with fat saturation of the same patient reveals the paralabral cyst (long arrow) extending posteriorly into the spinoglenoid notch.

most commonly arise in association with a superior labrum anterior-to-posterior (SLAP) tear or a posterior labral tear. These cysts may extend into either the suprascapular notch or the spinoglenoid notch, resulting in entrapment of the suprascapular nerve, which innervates the supraspinatus and infraspinatus muscles.109,110 Paralabral cysts arising from an anteroinferior labral tear are less common, but they may compress the axillary nerve as it traverses the quadrilateral space.111 Compression of the axillary nerve can also result from adhesive bands in the quadrilateral space in athletes, such as pitchers, who participate in repetitive overhead activities.112 The axillary nerve innervates both the teres minor and deltoid muscles. Anterior dislocation can result in a stretching injury of the axillary nerve and give rise to a temporary or permanent denervation of the teres minor and deltoid muscles and can occasionally mimic a rotator cuff tear on clinical examination in a person with previous anterior dislocation (Fig. 17H3-24). Denervation atrophy initially results in edema of the affected muscles and over time will progress to an irreversible fatty replacement. On MRI, acute denervation edema appears as high T2-weighted signal within the affected muscle and is associated with reversible muscle atrophy (see Fig. 17H3-24). The more chronic and irreversible form of fatty atrophy appears as decreased muscle bulk and bright streaks (representing the fat) within the muscles on T1-weighted images (see Fig. 17H3-19).

Rotator Interval The rotator interval is an anatomically complex area of the glenohumeral joint that is best described as a triangular gap in the anterosuperior aspect of the rotator cuff that is

bordered by the anterior margin of the supraspinatus and the superior margin of the subscapularis muscle. The function of the interval is to allow passage of the long head of the biceps tendon from an intra-articular to an extraarticular location. The rotator interval plays an important role in glenohumeral stability and ensures normal function

Figure 17H3-24   Denervation edema, magnetic resonance imaging. Oblique coronal T2-weighted image with fat saturation demonstrates diffuse high signal intensity (arrows) throughout the deltoid muscle. This represents denervation edema within the deltoid in this patient who suffered an anterior dislocation of the glenohumeral joint and stretching of the axillary nerve.

966

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

of the long head of the biceps tendon. Injury or pathology of the interval can be associated with altered range of motion of the glenohumeral joint, adhesive capsulitis, and progressive signs of impingement or degenerative arthropathy. Isolated injuries of the interval can occur, but they are often associated with tears of the most anterior aspect of the supraspinatus tendon or the superior leading edge of the subscapularis tendon.

Rotator Interval Anatomy The rotator interval is a small triangular gap in the rotator cuff formed by the interposition of the coracoid process between the anterior margin of the supraspinatus tendon and the superior leading edge of the subscapularis tendon. The roof of the rotator interval is formed by the coracohumeral ligament along the bursal surface and the superior glenohumeral ligament along the articular surface. Together, the coracohumeral and superior glenohumeral ligaments form the biceps tendon “sling” and are primarily responsible for stability of the long head of the biceps tendon as it transitions from its intra-articular position within the rotator interval to its extra-articular position within the intertubercular groove.113-116 The coracohumeral ligament arises from the base of the coracoid process extending across the anterior aspect of the glenohumeral joint to insert onto the lesser and greater tuberosities of the humeral head, forming the bursal lining of the rotator interval roof. The coracohumeral ligament is composed of the medial limb, which courses inferiorly from its origin on the coracoid process to insert onto the superior margin of the subscapularis tendon and the lesser tuberosity; and the lateral limb, which courses in a more horizontal fashion to its insertion on the anterior edge of the supraspinatus tendon and the adjacent greater ­tuberosity. The medial limb is primarily responsible for preventing medial subluxation of the long head of the biceps tendon. The superior glenohumeral ligament arises from the superior glenoid tubercle adjacent to the origin of the long head of the biceps tendon and courses deep to the coracoid process and the coracohumeral ligament to insert onto the lesser tuberosity of the humeral head and comprises the articular lining of the roof of the rotator interval. Together the coracohumeral and superior glenohumeral ligaments surround the intra-articular portion of the long head of the biceps tendon functioning as the primary stabilizer of the tendon. The long head of the biceps tendon originates from the superior glenoid tubercle and then crosses the joint in an oblique fashion sitting within the rotator interval and is covered by the coracohumeral and superior glenohumeral ligaments. The long head of the biceps tendon exits the joint at the level of the intertubercular groove where it is completely covered and stabilized by these two ligaments. At this point, the combination of the medial limb of the coracohumeral ligament and superior glenohumeral ligament acts as a sling to prevent medial subluxation of the long head of the biceps tendon as it exits the joint. The rotator interval is best visualized and evaluated for ­pathology in the oblique sagittal plane, especially when the joint is distended either by a native joint effusion or by intra-articular contrast.

The rotator interval plays an important role with regard to glenohumeral stability. The coracohumeral ligament and the superior glenohumeral ligament combine to limit external rotation of the humeral head, whereas the superior glenohumeral ligament contributes to glenohumeral stability by limiting inferior subluxation of the humeral head while the arm is in 0 degrees of abduction. Disease processes that affect the rotator interval can limit external rotation of the humeral head, whereas injury to the rotator interval can contribute to anterior and inferior instability of the glenohumeral joint and can also allow superior migration of the humeral head, thus leading to progressive degenerative arthropathy. The rotator interval also plays an important role with regard to the stability of the long head of the biceps tendon (Table 17H3-4). A tear of the rotator interval can result from an acute traumatic event or from repetitive microtrauma, and although isolated tears of the rotator interval have been reported, they most commonly occur in conjunction with a tear of the anterior portion of the supraspinatus tendon. Disruption of the rotator interval will result in shoulder pain and allow superior migration of the humeral head ­(microinstability). MRI findings of an isolated rotator interval tear include high T2-weighted signal isolated to the rotator interval or disruption of the capsule of the rotator interval. Direct MR arthrography will improve detection of the disrupted capsule at the level of the interval and may actually demonstrate leakage of contrast through a defect into the subacromial-subdeltoid bursa (Fig. 17H3-25). The rotator interval is an unsupported part of the glenohumeral joint capsule and as such is prone to disorders that affect the synovium, including inflammatory arthritides and adhesive capsulitis. Evaluation of the rotator interval in these disorders is best performed with either direct or indirect MR arthrography using the oblique sagittal images.

  TABLE 17H3-4  Rotator Interval: Anatomy, Function, and Pathologic Entities Normal anatomy

Function

Pathologic conditions

Gap within rotator cuff formed by interposition of the coracoid process Superior border—anterior margin of supraspinatus tendon Inferior border—superior leading edge of subscapularis tendon Roof—coracohumeral ligament (bursal surface); superior glenohumeral ligament (articular surface) Contains long head of biceps tendon Limits excessive external rotation of humeral head Prevents superior migration of humeral head Traumatic disruption Inflammatory arthritis Adhesive capsulitis

Shoulder

967

  TABLE 17H3-5  Adhesive Capsulitis: Clinical and Imaging Findings Risk factors

Clinical presentation Arthrography Magnetic resonance imaging

Figure 17H3-25   Rotator interval injury, magnetic resonance arthrography. Oblique sagittal image demonstrates disruption of the coracohumeral ligament (long arrow), allowing contrast to leak through the disrupted capsule into the overlying subacromial-subdeltoid bursa (short arrow). Note the long head of the biceps tendon (LHBT) located within the rotator interval between the supraspinatus (SST) and subscapularis (Subscap) muscles. The infraspinatus (IST) is seen more posteriorly.

Adhesive Capsulitis (Frozen Shoulder) Adhesive capsulitis is a pathologic entity unique to the shoulder that is characterized by inflammation and thickening of the synovium. The inflammatory changes are most prominent in those areas of the capsule that lack reinforcement by the rotator cuff tendons, most notably the rotator interval and axillary recess. Women 40 to 70 years of age are most often affected, presenting clinically with insidious onset of shoulder pain followed by progressive stiffness and weakness of the glenohumeral joint. The onset of symptoms is often related to a previous episode of minor trauma; however, it may be associated with preexisting rheumatologic disorders, diabetes, or idiopathic in nature. At the time of clinical presentation, patients are often misdiagnosed with impingement syndrome or rotator cuff pathology as the signs and symptoms broadly overlap with those of adhesive capsulitis. Physical examination demonstrates a painful restriction of motion of the shoulder in all directions, but the most pronounced restriction of motion usually involves external rotation of the humeral head (Table 17H3-5). The clinical presentation and age distribution of patients with adhesive capsulitis overlap with those of rotator cuff pathology, and as a result, imaging can play an important diagnostic role. There are no conventional radiographic signs diagnostic of adhesive capsulitis, although radiographs may be helpful in excluding other sources of pain and decreased range of motion. Conventional arthrography demonstrates a small contracted joint capsule with a

Female 40 to 70 years old Minor trauma Rheumatologic disorders Diabetes Insidious onset pain, stiffness Decreased range of motion Misdiagnosed as impingement Decreased joint volume (<10 mL) Contracted axillary pouch Thickened capsule in axillary recess (>4 mm) Pericapsular edema and enhancement Synovitis in rotator interval Thickened coracohumeral ligament

decreased joint volume of less than 10 mL. The axillary pouch appears small and contracted, and there is a lack of contrast filling the long head of the biceps tendon sheath (see Table 17H3-5). MRI findings include abnormal soft tissue in the rotator interval, obliteration of the subcoracoid fat triangle, and thickening of the coracohumeral ligament. There may also be thickening of the joint capsule (normal < 4 mm) in the region of the axillary pouch. Inflammatory changes of the capsule may also be seen as increased T2-weighted signal adjacent to the capsule of the axillary pouch. The use of direct MR arthrography may be helpful in establishing the correct diagnosis in two ways. First, there will be a decreased joint volume of less than 10 mL, similar to conventional arthrography. Second, the contrast will help to outline and define the abnormal soft tissue in the rotator interval. Indirect MR arthrography, however, appears to be most specific for establishing the diagnosis of adhesive capsulitis on the basis of imaging. Findings include enhancement of the abnormal soft tissues within the rotator interval combined with enhancement of the capsule and adjacent soft tissues in the region of the axillary pouch (Fig. 17H3-26) (see Table 17H3-5).117,118

Labrum and Capsular Structures Glenohumeral instability is a complex topic, and in terms of diagnosis and treatment of shoulder injuries, it is the area that has undergone the most significant change during the past decade. The anatomic configuration of the shoulder allows extensive range of motion, but it also predisposes the shoulder to instability and frequent dislocation. The clinical definition of shoulder instability is the symptomatic displacement of the humeral head out of the glenoid fossa, and various schemes have been developed to classify shoulder instability. These include the temporal relationship of the instability to previous trauma (first-time versus recurrent), the degree of instability (subluxation versus dislocation), and the direction of instability (anterior, posterior, inferior, or multidirectional). Thomas and Matsen119 devised a classification system that places patients with recurrent glenohumeral instability into one of two broad categories. The first category includes patients with antecedent trauma and unidirectional instability. These

968

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17H3-26   Adhesive capsulitis, indirect magnetic resonance arthrography. Oblique coronal (A) and oblique sagittal   (B) images reveal marked thickening of the capsule (long arrow on A) in the region of the axillary pouch with adjacent enhancement. There is also enhancement in the region of the rotator interval (short arrows on B). The intraarticular portion of the long head of the biceps tendon (LHBT) is seen within the rotator interval just deep to a markedly thickened coracohumeral ligament (CHL).

patients usually have an associated Bankart lesion and require surgical repair of the anterior labroligamentous complex to regain stability. This group is referred to by the acronym TUBS (traumatic, unidirectional, Bankart, requiring surgery). The second group of patients has no history of antecedent trauma, presents with bilateral multidirectional instability resulting from capsular and ligamentous laxity, and is less likely to benefit from surgical repair. The acronym describing this group of patients is AMBRI (atraumatic, multidirectional, bilateral, rehabilitation, and occasionally requiring inferior capsular shift). Although the initial imaging study after suspected shoulder dislocation or recurrent glenohumeral instability is usually conventional radiography, it provides only a limited evaluation of the shoulder, detecting such abnormalities as fractures, persistent dislocation, and abnormal soft tissue calcifications. Several advanced imaging modalities are capable of providing a more comprehensive evaluation of the capsule and labroligamentous structures of the shoulder. These include conventional arthrography, CT, computed arthrotomography, conventional MRI, and MR arthrography. In the past few years, conventional MRI has become the primary modality for detecting abnormalities of the glenohumeral labroligamentous complex, and MR arthrography is gaining acceptance as a minimally invasive method that further improves detection of even the most subtle lesions associated with glenohumeral instability. Conventional radiography, although limited in its evaluation of the glenoid labrum and capsular structures, is often complementary to the more sophisticated imaging techniques in the evaluation of glenohumeral instability. The imaging protocol for conventional radiography

varies, depending on personal preference, but it should include the basic views, which will enhance the detection of osseous abnormalities. For patients with a history of antecedent trauma, views should be selected that optimize visualization of the Hill-Sachs deformity, anterior trough defect, and Bankart lesion.120 The Garth view, the anteroposterior view with the shoulder in internal rotation, the axillary view, and the Stryker notch view improve detection of the posterior Hill-Sachs deformity as well as of the anterior trough defect. The anterior and posterior glenoid rim is best imaged on the axillary view or one of its many variations. An osseous Bankart lesion involving the anteroinferior glenoid rim is best visualized on the Garth view or the West Point view. The patient’s condition should also be taken into consideration in selecting radiographic views because the standard axillary view may be difficult to obtain in the setting of acute trauma. In these situations, one of the many axillary variants may be easier to obtain. In the setting of recurrent subluxation with no antecedent trauma, it is recommended that two tangential views be obtained to identify subtle subluxation of the humeral head. A true anteroposterior glenohumeral (Grashey) view will aid in the detection of inferior subluxation; one of the axillary view variants will demonstrate anterior or posterior subluxation. The radiographic findings associated with glenohumeral instability include the Hill-Sachs and Bankart osseous abnormalities, which occur when the posterior aspect of the humeral head impacts the anteroinferior glenoid rim at the time of anterior dislocation. The Hill-Sachs lesion is seen as flattening or a wedge-shaped defect involving the posterosuperior aspect of the humeral head; the Bankart lesion

Shoulder

Box 17H3-3  Posterior Shoulder Dislocation: Radiographic Findings Positive rim sign (widening of glenohumeral joint >6 mm) Humeral head appears in same position on internal and external rotation anteroposterior views “Trough line,” or reverse Hill-Sachs defect Reverse Bankart lesion Fracture of lesser tuberosity

is seen as a small fracture or area of cortical irregularity involving the anteroinferior glenoid rim. Although these radiographic findings lack sensitivity, they are specific for previous anterior dislocation and indicate that injury has occurred to the labroligamentous complex, which may predispose to recurrent glenohumeral instability. At the time of posterior dislocation, the anterior aspect of the humeral head impacts the posterior glenoid rim, creating an anterior trough defect in the anterior aspect of the humeral head and a “reverse Bankart” lesion involving the posterior glenoid rim. Subluxation can be a subtle radiographic finding that may indicate glenohumeral instability, and it is best detected on tangential views of the glenohumeral joint. The center of the humeral head should be centered within the glenoid fossa, and any asymmetry may indicate subluxation or persistent dislocation (Box 17H3-3). Conventional arthrography of the glenohumeral joint can be useful in detecting abnormalities of the rotator cuff (see Fig. 17H3-3) but offers little in the evaluation of the labrum and capsular structures.93 A combination of arthrography with conventional tomography of the shoulder was briefly used in the evaluation of glenohumeral instability during the late 1970s and early 1980s with limited success.121 In the early 1980s, however, CT arthrography replaced it as the modality of choice in the evaluation of glenohumeral instability. More recently, conventional MRI

A

969

and MR arthrography have become the imaging modalities of choice in the evaluation of a patient with glenohumeral instability.47,36 CT imaging with sagittal and coronal reconstructions remains the study of choice in detecting and depicting the size and location of a fracture fragment of the anteroinferior glenoid rim (see Fig. 17H3-4), which can be useful in presurgical planning. Conventional MRI is widely accepted as the standard noninvasive method for evaluating glenohumeral instability. However, MR arthrography is gradually replacing it in many practices because MR arthrography offers improved accuracy in detecting the subtle soft tissue lesions often associated with instability.93,121 The normal anatomic structures of the glenoid labrum and capsule lie in close proximity to one another, and the lesions associated with instability often undergo partial healing. If there is a paucity of joint fluid, these subtle lesions can be difficult to observe on conventional MRI. With MR arthrography, the joint is distended with fluid, which outlines and separates the normal anatomic structures, allowing more accurate depiction of subtle lesions (Fig. 17H3-27). The addition of certain stress views, such as the ABER view, will further improve visualization of a nondisplaced tear (Fig. 17H3-28), especially in the region of the anteroinferior labrum.50 Although CT arthrography and conventional MRI remain acceptable, MR arthrography is evolving into the imaging method of choice in the evaluation of the young patient with recurrent glenohumeral instability.

Magnetic Resonance Arthrography of Glenohumeral Instability Articulation of the large surface area of the humeral head with the small surface area of the glenoid fossa allows extensive range of motion of the shoulder but also predisposes it to instability. The various soft tissue structures about the shoulder act synergistically to improve stabilization of the glenohumeral joint. The glenoid labrum and capsular

B

Figure 17H3-27   Bankart lesion, magnetic resonance arthrography. Axial (A) and oblique coronal (B) T1-weighted images with fat saturation after intra-articular administration of gadolinium demonstrate disruption of the anteroinferior labrum (arrows). The gadolinium distends the joint capsule and outlines the torn and irregular-appearing labrum.

970

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17H3-28   Anteroinferior labral tear (Perthes’ lesion), magnetic resonance arthrography. A, Axial T1-weighted image with fat saturation after intra-articular administration of gadolinium demonstrates a nondisplaced tear of the anteroinferior labrum. A small gap filled with contrast material (arrow) is present between the anterior labrum and the adjacent articular cartilage. B, T1-weighted image in the abduction and external rotation (ABER) position places tension on the anterior band of the inferior glenohumeral ligament (arrowheads), displacing the torn anterior labrum away from the glenoid (arrow), increasing conspicuity of the subtle lesion.

s­ tructures act as static stabilizers, while the rotator cuff provides dynamic stabilization of the glenohumeral joint. The glenoid labrum is a thickening of the joint capsule that inserts into the periphery of the osseous glenoid and functions to deepen the shallow glenoid fossa, providing extended coverage of the humeral head.122-124 The superior

A

labrum (Fig. 17H3-29A) is best visualized on coronal MR images and appears as a dark triangular structure extending off of the superior glenoid rim. The anterior and posterior labra (see Fig. 17H3-29B) are best seen on axial images and also typically appear as dark triangular structures; on occasion, however, the normal labrum may appear rounded,

B

Figure 17H3-29   Normal labrum, magnetic resonance arthrography. A, Coronal T1-weighted image with fat saturation demonstrates a normal-appearing superior labrum (arrowhead), which is best seen in the coronal plane. It is triangular and demonstrates low signal intensity throughout the substance of the labrum. A small collection of contrast material (arrow) is present between the superior labrum and the glenoid. This collection of contrast material is filling the superior labral recess. It can be differentiated from a superior labral tear because it is smooth and tapering and does not extend completely beneath the superior labrum. A labral tear, in contrast, is irregular in appearance, and the high signal extends into the substance of the labrum rather than between the labrum and the glenoid. B, Axial T1-weighted image with fat saturation demonstrates the normal appearance of the anterior (black arrow) and posterior (curved white arrow) labrum. The labra are triangular and demonstrate low signal intensity throughout. Notice the normal cartilage undermining (straight white arrow) of the anterior labrum. Cartilage undermining differs from a tear because it is smooth and tapering and does not extend completely beneath the labrum. The signal intensity of hyaline articular cartilage is intermediate, whereas a tear is irregular in appearance and contains high signal contrast.

Shoulder

blunted or flattened.123,125 Three glenohumeral ligaments, which represent thickenings of the capsule, also serve to improve stability of the shoulder. The inferior glenohumeral ligament (Fig. 17H3-30) is the most important of the three thickenings and functions primarily to improve glenohumeral stability with the arm in abduction and external rotation.126 It is composed of three separate components: the anterior band, posterior band, and axillary pouch.127,128 The origin of the inferior glenohumeral ligament is somewhat variable; it arises from either the inferior glenoid labrum or adjacent osseous glenoid. It then inserts in a collar-like fashion along the medial aspect of the humeral neck. It is lax when the arm is in neutral position and appears redundant when the arm is imaged in the standard planes on MR arthrography (see Fig. 17H3-30A and B). Tirman described a new imaging position during MR arthrography of the shoulder in which the arm is placed in the ABER position, demonstrating a taut anterior band (see Fig. 17H3-30C).50 This results in better visualization of the anterior band of the inferior glenohumeral ligament and improves the detection of nondisplaced tears of the anteroinferior labrum, an important stabilizing structure of the glenohumeral joint.50,51

A

Figure 17H3-30   Normal anterior band of the inferior glenohumeral ligament, magnetic resonance arthrography. A, Coronal T1-weighted image with fat saturation reveals a normal-appearing anterior band of the glenohumeral ligament (white arrows). It is redundant in appearance but nicely outlined by the intra-articular contrast material. The black arrow denotes a tear of the anteroinferior labrum. B, Axial T1-weighted image with fat saturation through the inferior aspect of the glenohumeral joint demonstrates the normal appearance of a lax anterior band of the inferior glenohumeral ligament (arrow). C, T1-weighted image of the shoulder in the abduction and external rotation (ABER) position demonstrates a taut anterior band of the inferior glenohumeral ligament (arrowheads). In this position, tension is placed on the anterior labrum (arrow), thereby improving visualization of the anteroinferior labroligamentous complex.

971

The middle glenohumeral ligament (Fig. 17H3-31) is less important as a stabilizer, but it is responsible for preventing external rotation of the arm during abduction between 60 and 90 degrees.129 In terms of size, it is the most variable of the three ligaments, ranging from a thick cord-like ligament to complete absence.130 It arises from the superior glenoid tubercle adjacent to the origin of the biceps tendon and superior glenohumeral ligament and then courses obliquely in an inferolateral direction to merge with the deep fibers of the subscapularis tendon just before attaching to the humeral neck. It is typically seen on axial MR images deep to the subscapularis muscle and superficial to the anterior labrum, occasionally mimicking an avulsed fragment of the anterior labrum. The superior glenohumeral ligament (Fig. 17H3-32) contributes the least to shoulder stability, although a report suggests that it provides some degree of restraint, preventing inferior subluxation of the humeral head when the arm is in 0 degrees of abduction.131,132 It arises from the superior glenoid tubercle adjacent to the attachment of the biceps tendon and then courses obliquely and anteriorly to merge with the coracohumeral ligament before its attachment on the humeral head. The superior glenohumeral ligament

B

C

972

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C Figure 17H3-31   Middle glenohumeral ligament, magnetic resonance arthrography. A, Axial T1-weighted image through the mid-glenohumeral joint reveals a normal middle glenohumeral ligament (black arrowheads), which is interposed between a normal anterior labrum (arrow) and the deep fibers of the subscapularis muscle (white arrowheads). B, Axial T1-weighted image through the upper aspect of the glenohumeral joint demonstrates a normal middle glenohumeral ligament (black arrow) adjacent to the anterior labrum (arrowhead). The middle glenohumeral ligament can mimic an anterior labral tear. It can be differentiated from a tear by tracing the ligament throughout its entire course on multiple sequential axial images. The white arrow demonstrates a normal sublabral foramen in the anterosuperior quadrant. This is a normal variant in which the anterior labrum is completely detached from the adjacent glenoid. This variation occurs only in the anterosuperior quadrant. C, Axial T1-weighted image through the mid-glenohumeral joint demonstrates a thick, cord-like middle glenohumeral ligament (arrow) and an absent anterior labrum (arrowhead), representing a normal anatomic variant known as the Buford complex.

forms the articular surface of the roof of the rotator interval and also plays a role in stability of the long head of the biceps tendon. It is consistently visualized on axial MR arthrography images as a thick band-like structure arising from the glenoid tubercle and paralleling the coracoid process.128 Several normal anatomic variations have been described that on MR arthrography can mimic a labral tear.133 The majority of these variations occur in the anterosuperior quadrant of the labrum and include: Cartilage undermining—smooth tapering intermediate signal hyaline cartilage that partially undermines the superior labrum (see Fig. 17H3-29B).120,122,134,135 Sublabral recess—potential space between the labrum and underlying glenoid, appears as a smooth tapering fluid

collection that only partially undermines the superior labrum with no labral displacement (see Fig 17H3-29A).136 A tear differs in that it will show an irregular surface of the labrum with possible labral detachment or displacement (see Fig. 17H3-37). Sublabral foramen—complete detachment of the superior labrum that occurs only in the anterosuperior quadrant.130,137 Appears on MRI as a fluid filled gap between the labrum and the osseous glenoid (see Fig. 17H3-31B). Buford complex—absent or diminutive anterosuperior labrum associated with a thick middle glenohumeral ligament (see Fig. 17H3-31C).122,125,128,130,137,138 Can mimic a displaced tear of the anterior labrum on MRI.

Shoulder

Figure 17H3-32   Normal superior glenohumeral ligament, magnetic resonance arthrography. Axial T1-weighted image with fat saturation through the superior aspect of the glenohumeral joint demonstrates the normal superior glenohumeral ligament (black arrow). It is invariably seen on axial images as a thick, band-like structure arising from the superior glenoid tubercle and paralleling the coracoid process (C). It then blends with the fibers of the coracohumeral ligament anteriorly. White arrow, superior labrum.

Magnetic Resonance Arthrography of Multidirectional Glenohumeral Instability As described before, patients with recurrent glenohumeral instability can be divided into two broad categories. The first group presents with recurrent unidirectional instability after a traumatic injury to the glenohumeral joint (TUBS). The second group presents with bilateral multidirectional instability of the glenohumeral joint but gives no history of antecedent trauma (AMBRI). The diagnosis of multidirectional instability is usually established solely on the basis of history and physical examination findings; MRI adds little to the evaluation. There are no specific MR findings to confirm the diagnosis of multidirectional glenohumeral instability, but in the atypical case, it may help to exclude a concurrent Bankart or other labral lesion.138

Magnetic Resonance Arthrography in Post-traumatic Anterior Glenohumeral Instability A wide variety of glenolabral lesions exist, and in a given patient, the specific lesion depends on the mechanism of injury that led to the labral abnormality. Unidirectional instability may be further classified as anterior or posterior in direction. Anterior glenohumeral instability usually results from a fall on the outstretched arm. In the young patient (younger than 35 years), the typical lesion resulting from an anterior dislocation is either a Bankart lesion or a Bankart variant. In the patient older than 35 years with a first-time anterior shoulder dislocation, the resulting lesion is usually either a tear of the supraspinatus tendon, a tear

973

Figure 17H3-33   Hill-Sachs defect, magnetic resonance arthrography. Axial T1-weighted image through the superior aspect of the glenohumeral joint demonstrates a concavity (arrow) in the posterosuperior aspect of the humeral head representing a Hill-Sachs deformity, secondary to previous anterior dislocation.

of the subscapularis tendon, or an avulsion injury of the greater tuberosity of the humeral head. Posterior instability occurs less commonly than anterior instability and can result either from a single traumatic event resulting in posterior dislocation of the humeral head or from repetitive microtrauma as occurs in weightlifters, football linemen, and swimmers. Finally, a miscellaneous category of lesions includes the SLAP tear, internal impingement syndrome, and the paralabral cyst. Acute anterior dislocation of the humeral head in the young patient typically results in disruption of the anterior labroligamentous complex, and Bankart is generally credited with the first description of this lesion in 1923. The classic Bankart lesion (see Fig. 17H3-27) is composed of an avulsion of the anterior labroligamentous complex from the glenoid with disruption of the medial scapular periosteum. The MRI appearance of a Bankart lesion varies, depending on the age of the lesion, but the anteroinferior labrum typically loses its normal triangular configuration and becomes irregular or amorphous in appearance. A discrete tear may be seen as an irregular collection of contrast material extending into the labrum. In the acute lesion, the anterior labrum may be displaced away from the adjacent glenoid; whereas in the more chronic injury, the labrum may scar back down to the glenoid and demonstrate minimal displacement, making it more difficult to identify at the time of imaging or at arthroscopy. On occasion, an associated fracture of the inferior glenoid rim (osseous Bankart lesion) or of the posterosuperior humeral head (Hill-Sachs deformity) (Fig. 17H3-33) may result from an impaction injury at the time of dislocation. Although both of these osseous abnormalities are accurately depicted on MRI, computed tomographic examination with sagittal reconstructions is often considered the imaging modality of choice for detecting and depicting the size and appearance of the osseous Bankart lesion (see Fig. 17H3-4).

974

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 17H3-6  Bankart Variants Lesion

Description

Bankart

Anteroinferior labral tear with disrupted medial scapular periosteum Anteroinferior labral tear with adjacent glenoid rim fracture Nondisplaced anterior inferior labral tear with an intact medial scapular periosteum Medial anteroinferior labral tear; torn labrum displaced medially by an intact medial scapular periosteum Posteroinferior labral tear

Osseous Bankart Perthes (nondisplaced Bankart) Anterior labroligamentous periosteal sleeve avulsion (ALPSA; medialized Bankart) Reverse Bankart

Two Bankart variants that have potential therapeutic and diagnostic ramifications have been described (Table 17H3-6). The classic Bankart lesion is composed of a tear of the anterior labrum coupled with disruption of the medial scapular periosteum, resulting in displacement of the anterior labroligamentous complex. Both Bankart ­variants differ from the classic Bankart lesion in that the medial scapular periosteum remains intact, preventing complete separation of the labrum from the adjacent osseous glenoid. The first of these variations was initially described by Perthes in 1906 and has subsequently been referred to as Perthes’ lesion.56,138 This Bankart variant consists of a nondisplaced tear of the anteroinferior labrum held in nearly anatomic position by an intact medial scapular periosteum (see Fig. 17H3-28). The tear may eventually resynovialize, making identification of the labral tear difficult at arthroscopy or on MRI. The clinical importance of Perthes’ lesion is that glenohumeral instability may persist in the face of a resynovialized nondisplaced tear, and at arthroscopy, the surgeon must probe the anterior labrum to determine whether it is incompetent. Identification of a nondisplaced anterior labral tear may be difficult at MR arthrography, but it may be aided by applying stress to the anterior labroligamentous complex. This is accomplished by positioning the arm in the ABER position (see Fig. 17H3-30C), which places traction on the anterior band of the inferior glenohumeral ligament, thereby pulling the partially detached labrum away from the glenoid, allowing the tear to fill with contrast material. When a nondisplaced tear of the anterior labrum is identified at MR arthrography, the surgeon should be alerted so that the anterior labrum is adequately probed to avoid inadvertently missing the nondisplaced tear. Neviaser54 described a second variation of the Bankart lesion, which he referred to as the ALPSA (anterior labroligamentous periosteal sleeve avulsion) lesion. The ALPSA lesion is similar to Perthes’ lesion in that the medial scapular periosteum remains intact. In the ALPSA lesion, however, the labrum is displaced medially and rolls up on itself like a sleeve is rolled up the arm. The labrum then heals in this abnormal medialized position. The ALPSA lesion is sometimes referred to as the medialized Bankart (Fig. 17H3-34). The acute lesion is easily identified at arthroscopy or on MRI, but the chronic lesion may be more difficult to identify at arthroscopy because scar tissue may mound up on

Figure 17H3-34   Anterior labroligamentous periosteal sleeve avulsion (ALPSA) lesion, magnetic resonance arthrography. Axial T1-weighted image with fat saturation demonstrates a variation of the Bankart lesion, in which the torn anterior labrum (arrow) is displaced medially. It may scar down in this position and can be difficult to identify at arthroscopy and during imaging.

top of the medialized labrum and subsequently resynovialize. MR arthrography accurately depicts the position of the labrum and demonstrates its medial displacement.138 The ALPSA lesion is treated arthroscopically by débridement of the mounded-up scar tissue followed by surgical completion of the Bankart lesion and finally reattachment of the labrum in its normal anatomic position.54 First-time dislocation in the older patient (older than 35 to 40 years) leads to a set of lesions different from those seen in the younger patient. As a patient ages, the rotator cuff tendons gradually degenerate and over time become the weak link among the various anatomic structures responsible for anterior shoulder stability.54,105,126 Firsttime dislocation after age 35 years therefore uncommonly results in a Bankart lesion but usually results in one of the following three lesions: • Tear of the supraspinatus tendon (Fig. 17H3-35A) • Avulsion fracture of the greater tuberosity of the humeral head (see Fig. 17H3-35B) • Avulsion or tear of the subscapularis tendon and anterior capsular structures (see Fig. 17H3-21) MR arthrography can play a pivotal role in directing the appropriate therapy in the older patient following a first-time anterior dislocation of the shoulder. The patient with rotator cuff rupture will have weakness on abduction and may be incorrectly diagnosed with an axillary nerve injury.105,139 This group of patients may benefit from surgical repair, depending on the extent of rotator cuff tear. The patient with a greater tuberosity avulsion fracture is usually treated conservatively, whereas the patient with an avulsion of the subscapularis and anterior capsule is likely to have recurrent instability, and persistent pain and will require surgical repair to regain stability.119 MR ­arthrography can

Shoulder

A

975

B

Figure 17H3-35   First-time dislocation in an older patient, magnetic resonance imaging. Oblique coronal T1-weighted image (A) with fat saturation after intra-articular administration of gadolinium reveals complete disruption and retraction (arrow) of the supraspinatus tendon in this 45-year-old man following anterior dislocation of the shoulder. Oblique coronal T1-weighted (B) image demonstrates a nondisplaced avulsion fracture of the greater tuberosity after a first-time dislocation in this 63-year-old man. The fracture line (arrow) appears as a dark line.

accurately differentiate these lesions, thereby directing clinical ­management by correctly identifying those patients who would benefit most from surgical intervention. MR arthrography clearly depicts a rotator cuff rupture as high signal contrast extending through the rotator cuff defect and retraction of the musculotendinous junction. Axillary nerve palsy, which initially results in denervation edema, is seen on MRI as high signal within the muscle on T2-weighted images (see Fig. 17H3-24). Chronic denervation of the rotator cuff eventually leads to fatty atrophy, which is characterized by increased signal on T1-weighted images.111,112,140-142 An avulsion fracture of the greater tuberosity is depicted as a dark fracture line extending through the greater tuberosity on all MR pulse sequences. Surrounding bone marrow edema in the acute fracture is depicted as increased signal on T2-weighted images or decreased signal intensity on T1-weighted images. Finally, an avulsion injury of the subscapularis and anterior capsular structures is best depicted on axial images using MR arthrography. Extension of contrast material will be seen extending between the subscapularis tendon and the lesser tuberosity. Retraction of the musculotendinous junction may also be noted.130,143

lesions are seen in anterior and posterior ­ glenohumeral instability, but the term reverse is typically applied to the lesions associated with posterior instability (reverse Bankart or reverse Hill-Sachs lesion). MR arthrography accurately depicts a posterior labral tear as high signal contrast extending into the tear with displacement and abnormal morphologic features of the labrum (Fig. 17H3-36). Osseous injuries, including a reverse Hill-Sachs lesion or a reverse osseous Bankart abnormality, are also accurately depicted.47,146,147

Magnetic Resonance Arthrography of Posterior Glenohumeral Instability Posterior shoulder instability is less common than anterior instability; it typically results from either a single traumatic event or repetitive microtrauma as in weightlifters or swimmers. Posterior glenohumeral instability can be a difficult entity to diagnose and treat.144,145 Traumatic posterior dislocation usually results from a fall on the outstretched arm with the arm positioned in adduction and internal rotation. In this position, the posterior capsule becomes taut, and any force directing the humeral head posteriorly can result in disruption of the posterior capsule or labrum. Similar

Figure 17H3-36   Posterior labral tear, magnetic resonance arthrography. T1-weighted axial image through the midglenohumeral joint demonstrates contrast material extending between the posterior labrum and the glenoid (arrow), representing a nondisplaced posterior labral tear.

976

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Magnetic Resonance Arthrography of Miscellaneous Lesions of the Glenoid Labrum The SLAP Lesion The superior labrum anterior-to-posterior (SLAP) lesion was named by Snyder and colleagues148 in 1990 to describe a tear of the superior labrum. Two different mechanisms have been proposed as potentially leading to the SLAP lesion. The first is a fall on the outstretched arm that results in compressive forces on the superior labrum; the second is repetitive microtrauma resulting from recurrent overhead activity, such as throwing or swimming. Patients with a SLAP injury typically present with shoulder pain, aggravated by overhead activities, and may also complain of a catching or popping sensation. Physical examination is often unreliable in establishing the diagnosis of a SLAP tear, although stress maneuvers may indicate that an abnormality involves the biceps anchor. Snyder and colleagues148 originally described four types of SLAP lesions: Type I lesion (Fig. 17H3-37A): marked degeneration and fraying of the superior labrum, but the labrum remains firmly attached to the glenoid Type II lesion (see Fig. 17H3-37B): an avulsion of the labral-bicipital complex from the glenoid, resulting in an unstable biceps anchor Type III lesion (see Fig. 17H3-37C): displaced buckethandle tear of the superior labrum with an intact biceps anchor Type IV lesion (see Fig. 17H3-37D): displaced buckethandle tear of the superior labrum that extends to involve the biceps anchor Maffet and coworkers149 expanded the original classification put forth by Snyder and colleagues to include three additional types of SLAP lesions: Type V lesion: Bankart lesion of the anteroinferior labrum, which extends up to involve the biceps anchor Type VI lesion: unstable flap or radial tear that involves the biceps anchor Type VII lesion: SLAP tear that extends anteriorly and inferiorly beneath the middle glenohumeral ligament Treatment varies, depending on the type of SLAP tear, but it generally involves arthroscopic débridement of the type I and type III lesions and arthroscopic repair of the type II and type IV lesions, which include an avulsion of the biceps anchor.150 Since the original report of the SLAP lesion by Snyder and colleagues, numerous articles have described the conventional MRI and MR arthrographic findings.94,151-155 MR arthrography is both sensitive and accurate as a technique for diagnosis of SLAP tears, and in particular, it provides pertinent information about the extent and type of the SLAP tear as well as defines the integrity of the biceps anchor.151 The SLAP lesion is usually best seen on oblique coronal MRI (see Fig. 17H3-37). Contrast material extends into the normally dark triangle of the superior labrum, representing either marked fraying or detachment from the glenoid. The SLAP lesion can easily be confused with several normal

a­ natomic variants that occur in the anterosuperior quadrant of the labrum. These are the superior labral recess (see Fig. 17H3-29A), the sublabral foramen (see Fig. 17H3-31B), and potentially the Buford complex (see Fig. 17H3-31C). A tear can be differentiated from these normal variations because abnormal signal extends into the tear, which is usually irregular in appearance, and the labrum may actually pull away from the glenoid. In type III and type IV SLAP lesions, a labral fragment is often seen dangling from the superior labrum and is outlined by intra-articular contrast material. Contrast material seen extending into the substance of the superior labrum or biceps anchor is abnormal and is ­diagnostic of a SLAP tear (Box 17H3-4). In addition to identifying the abnormality of the superior labrum, it is important to detect any extension of the SLAP tear to adjacent structures because involvement of adjacent structures can have an impact on proper surgical treatment. SLAP lesions can extend to involve the inferior labrum in either the anterior or posterior quadrants. SLAP lesions can also extend to involve the superior or middle glenohumeral ligaments and can also be associated with rotator cuff pathology.

The GLAD Lesion Neviaser55 first described the glenolabral articular disruption (GLAD) lesion as a superficial tear of the anteroinferior labrum associated with an injury of the adjacent articular cartilage. It is caused by a forced adduction injury to the shoulder with the arm positioned in abduction and external rotation. The injury results from an impaction of the humeral head against the glenoid fossa without anterior subluxation or dislocation. The patient presents clinically with persistent shoulder pain but demonstrates no evidence of instability on physical examination. This lesion is probably the first along the spectrum of instability lesions secondary to a fall on an outstretched arm. The labral tear is superficial and nondisplaced; as a result, it does not lead to clinically identifiable instability. The pain is most likely due to catching of the cartilage flap as it contacts the moving humeral head.64 The abnormalities associated with the GLAD lesion are subtle and may be undetectable on routine MRI of the shoulder. MR arthrography, however, has been shown to increase the sensitivity of detecting subtle labral lesions.36,47 The findings on MR arthrography in a patient with the GLAD lesion include the following (Fig. 17H3-38): • Flap tear of the articular cartilage, usually in the anteroinferior quadrant • Superficial, nondisplaced tear of the anteroinferior labrum The nondisplaced labral tear is best demonstrated in the ABER view and is seen as a tiny collection of contrast material filling a gap between the labrum and the osseous glenoid.156 The anterior band of the inferior glenohumeral ligament and the medial scapular periosteum remain intact. These abnormalities noted on MR arthrography could mimic findings seen in the chronic Bankart lesion with an adjacent chondral injury resulting from recurrent glenohumeral instability. The GLAD lesion, however, can be differentiated from the Bankart lesion on the basis of the mechanism of injury as well as by the absence of instability on physical examination.156

Shoulder

A

B

C

D

977

Figure 17H3-37   Superior labrum, anterior-to-posterior (SLAP) tears, magnetic resonance arthrography. Oblique coronal   T1-weighted images reveal various types of SLAP lesions. A, Type I SLAP tear. Abnormal signal is noted along the inferior margin of the superior labrum (arrow), indicating a degenerative pattern tear with no displaced or unstable fragment noted. B, Type II SLAP tear.   An abnormal collection of contrast material (arrow) extends into the substance of the superior labrum, indicating a partial avulsion.   C, Type III SLAP tear. A displaced bucket-handle fragment (long arrow) is seen extending off the inferior aspect of the superior labrum. Contrast (short arrow) completely surrounds the avulsed bucket-handle fragment. D, Type IV SLAP tear. A bucket-handle fragment (long arrow) is seen extending from the inferior aspect of the superior labrum, involving the biceps anchor (short arrows).

The HAGL Lesion The Bankart lesion and its many variants result from a tear or avulsion of the anterior labroligamentous complex at its insertion site on the osseous glenoid. A lesser known entity is the humeral avulsion of the glenohumeral ligament (HAGL). Cadaveric studies have demonstrated failure of the anterior labroligamentous complex under tension at the

glenoid attachment in 40%, in the midsubstance in 35%, and at the humeral attachment in 25% of the specimens.126 Although disruption of the anterior labroligamentous complex clinically occurs most commonly at the glenoid attachment site, reports have also shown that the inferior glenohumeral ligament may fail in the midsubstance or at the humeral attachment site.157,158 In a single series, the

978

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 17h3-4  Magnetic Resonance Imaging Signs Indicating a SLAP Lesion Abnormal morphology, fraying, and irregularity of the superior labrum High T2-signal extending into the substance of the ­superior labrum Avulsion or displacement of the superior labrum away from the glenoid Abnormal signal involving the biceps anchor

HAGL lesion was found as an isolated abnormality in 35% of shoulders with recurrent anterior instability. It was suggested that many cases of recurrent shoulder instability that in the past were thought to result from capsular laxity may actually be secondary to a missed HAGL lesion.158 The findings on MR arthrography of the HAGL lesion include disruption of the anterior band of the inferior glenohumeral ligament either in its midsubstance or near its humeral attachment (Fig. 17H3-39). Extravasation of contrast material is seen through the defect in the ligament; associated disruption of the adjacent subscapularis may or may not be present. Identification of this lesion at MR arthrography may be helpful to the arthroscopist because this lesion may be difficult to detect using the standard arthroscopic portals, and a miniarthrotomy may be required to identify the lesion and adequately repair it. Surgical repair is considered crucial in reestablishing shoulder stability after disruption of the inferior ­glenohumeral ­ligament.158,159 More recently, humeral avulsion of the posterior band of the inferior glenohumeral ligament has also been described with similar MRI findings involving the posterior band of the inferior glenohumeral ligament.160

A

Posterosuperior Glenoid Impingement Posterosuperior glenoid (or internal) impingement is a condition that occurs in athletes who participate in overhead activities, such as throwing. The injury occurs during the late cocking phase of throwing and is characterized by repeated impingement of the undersurface of the posterior aspect of the rotator cuff on the posterosuperior labrum and glenoid. It has been suggested that anterior instability is associated with internal impingement and may be contributory.161-164 This form of internal impingement may lead to undersurface tearing of the posterior aspect of the rotator cuff as well as fraying and degeneration of the posterosuperior labrum. Associated injuries may include bony changes in the greater tuberosity of the humeral head secondary to recurrent impaction of the humeral head against the posterior glenoid. If anterior instability is also present, a lesion of the anterior labroligamentous complex may also be noted. Treatment of internal impingement has two primary goals. The first is to rest the injured structures, and the second is to restore normal biomechanics. Because the initial recommended treatment is conservative, accurate diagnosis with MRI may eliminate the need for diagnostic shoulder arthroscopy and the associated recuperation time. The constellation of MRI findings in a patient with posterosuperior glenoid impingement includes the following (Fig. 17H3-40): • Partial-thickness articular-sided tear or fraying of the posterior aspect of the rotator cuff (infraspinatus tendon or the posterior aspect of the supraspinatus tendon) • Degenerative fraying or a discrete tear of the posterior aspect of the superior labrum • Subcortical cystic changes in the greater tuberosity of the humeral head

B

Figure 17H3-38   Glenolabral articular disruption (GLAD) lesion, magnetic resonance arthrography. A, Axial T1-weighted image through the inferior aspect of the glenohumeral joint demonstrates a flap tear (arrow) of the articular cartilage with an underlying osseous injury (arrowhead). B, Abduction and external rotation (ABER) view demonstrates a nondisplaced anteroinferior labral tear (arrow).

Shoulder

979

• Impingement of the posterior rotator cuff between the humeral head and the posterior glenoid seen on ABER imaging • Possible tear, stretching, or laxity of the anteroinferior labroligamentous complex104

Glenohumeral Internal Rotation Deficit Disorder

Figure 17H3-39   Humeral avulsion of the anterior glenohumeral ligament (HAGL), magnetic resonance arthrography. Oblique coronal T1-weighted image with fat saturation demonstrates disruption of the anterior glenohumeral ligament (long arrow) near its humeral attachment site. There is extravasation of contrast material (short arrows) from the joint into the adjacent soft tissues.

A

For many years, it has been believed that the primary pathophysiology of internal impingement is “microtrauma” to the anterior capsule resulting in anterior capsular laxity and allowing excessive external rotation of the humeral head. This leads to subtle posterior subluxation of the humeral head during maximal abduction and external rotation, which results in impingement of the posterior cuff. Recently, Burkhart and associates have questioned the long-standing theory that microinstability is the universal cause of internal impingement in the disabled throwing athlete. Burkhart has proposed that, in many cases, the primary lesion in throwing athletes is posterior capsular contraction or scarring and posterior muscular tightness, leading to an acquired loss of internal rotation with secondary increased external rotation and tertiary anterior capsular stretching.165 This lesion has been referred to as the glenohumeral internal rotation deficit (GIRD) disorder, indicating that the internal impingement results from posteroinferior capsular contracture leading to the subtle posterior subluxation of the humeral head during maximal abduction and external rotation and thus to superior instability and SLAP lesions. The ramifications of this theory

B

Figure 17H3-40   Posterosuperior glenoid impingement, magnetic resonance arthrography. A, Oblique coronal T1-weighted image reveals a partial-thickness articular-sided tear (long arrow) of the infraspinatus tendon with contrast extending into the substance of the tendon. Subcortical cystic change (short arrows) is also noted in the posterior aspect of the greater tuberosity.   B, Axial view demonstrates a posterior labral tear (long arrow). Subcortical cystic change (short white arrow) is noted in the posterior aspect of the humeral head, and contrast extends into the undersurface of the infraspinatus tendon, indicating a partial-thickness articular-sided tear (short black arrow).

980

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17H3-41   Glenohumeral internal rotation deficit lesion, magnetic resonance arthrography. Axial T1-weighted image reveals marked thickening of the posterior capsule   (long arrow) in this patient with symptoms of internal impingement. Note the normal-appearing posterior labrum   (short arrow).

are important because appropriate treatment will differ depending on whether the primary lesion is anterior capsular stretching or posterior capsular contracture and scarring. MRI is supportive of the GIRD theory in many cases of internal impingement with posterior capsular thickening and scarring accompanying the previously described lesions of internal impingement (Fig. 17H3-41).

Labral Cysts The glenoid labral cyst is the final lesion of the glenoid labrum to be discussed that is associated with glenohumeral instability. These cysts arise adjacent to the glenoid labrum secondary to either a tear or degeneration of the labrum and are analogous to cysts that arise adjacent to other joints, such as the meniscal cysts of the knee or labral cysts of the hip. They have a high association with labral tears and glenohumeral instability, and it has been hypothesized that fluid extrudes through a labral defect and collects in the adjacent soft tissues, thus forming a labral cyst.110 They most commonly occur superiorly in ­association with SLAP

lesions or posteriorly in association with posterior labral tears, but they have also been reported inferiorly in association with anteroinferior labral tears.110 Glenoid labral cysts have been implicated as the cause of compression neuropathies of both the suprascapular and axillary nerves.111,166 A glenoid labral cyst is depicted on MRI as a fluid-filled structure adjacent to the labrum (see Figs. 17H3-21B to 17H3-29). Glenoid labral cysts are variable in size, ranging from a few millimeters to several centimeters. They are frequently lobulated and may contain internal septations. They are bright on T2-weighted images and intermediate on T1-weighted images. They may communicate with the joint space or may be isolated from the joint space secondary to partial resynovialization of the adjacent labral tear. Thus, after the intra-articular injection of a contrast agent during MR arthrography, the cyst may continue to demonstrate fluid signal or may demonstrate high signal on T1-weighted images secondary to the presence of contrast material. Because labral cysts have a high association with labral tears, identification of a cyst on MRI should prompt a thorough search for an associated labral tear. A cyst that extends into the suprascapular or spinoglenoid notch may result in a compressive neuropathy of the suprascapular nerve; a cyst dissecting inferiorly into the quadrilateral space may produce a compressive neuropathy of the axillary nerve (Table 17H3-7).109,111,112,140-142,167-169 Compressive neuropathy initially results in diffuse edema and eventually leads to fatty replacement of the involved muscles in the chronic irreversible stages of the disease. On MRI, denervation edema appears as diffuse increased signal intensity within the affected muscle on T2-weighted images (see Fig. 17H3-24). The more chronic form of fatty atrophy demonstrates streaks of increased signal intensity within the involved muscle on T1-weighted images coupled with loss of the normal muscle bulk (see Fig. 17H3-19).

Biceps Tendon The tendon of the long head of the biceps is an important stabilizing structure of the glenohumeral joint contributing to both superior and anterior stability.170 Like any other tendon, the long head of the biceps tendon can be involved in a broad array of injuries. These can range from tenosynovitis and tendinosis to partial or complete tears. In addition, because of the unique anatomic configuration of the biceps tendon, subluxation or dislocation can occur in conjunction with large tears of the rotator cuff or isolated injuries of the subscapularis tendon.106 SLAP tears, which can also be associated with injuries of the proximal biceps tendon, are discussed in a preceding section and are not addressed here. The course and morphologic appearance of the long

  TABLE 17H3-7  Compressive Neuropathies Caused by Paralabral Cysts Location of Entrapment

Associated Labral Tear

Nerve Entrapped

Affected Muscles

Suprascapular notch

Superior

Suprascapular

Spinoglenoid notch

Superior Posterior Anteroinferior

Suprascapular

Supraspinatus Infraspinatus Infraspinatus

Quadrilateral space

Axillary

Teres minor Deltoid

Shoulder

­ icipital tendon can be evaluated with sonography, CT, b MRI, or MR arthrography; however, the superior soft tissue contrast and multiplanar capabilities afforded by MRI make it the imaging modality of choice. A thorough evaluation by any shoulder imaging study should include an assessment of the course and appearance of the long bicipital tendon. The normal course of the tendon begins as the tendon arises from the long head of the biceps muscle in the upper arm. The tendon then courses through the intertubercular sulcus, located between the lesser and greater tuberosities. As it passes through the intertubercular sulcus, it is held in place inferiorly by the tendon of the pectoralis major muscle as it attaches to the proximal humeral shaft. In a more cranial location, it is held in place by the transverse ligament, which is an extension of the subscapularis tendon. As the tendon moves even farther cranially, it becomes intraarticular, being located in the anatomic space referred to as the rotator interval. In this position, the tendon is covered by the coracohumeral ligament and finally by the superior glenohumeral ligament. These two ligaments form the “sling” or pulley mechanism that is primarily responsible for the stability of the biceps tendon as it transitions from its intra-articular to its extra-articular location. It then has a broad triangular attachment to the superior labrum at or near the superior glenoid tubercle. Conventional MRI accurately depicts both the location and the morphologic features of the extra-articular portion of the tendon, whereas MR arthrography improves detection of subtle intra-articular abnormalities of the tendon at or near its proximal attachment site. Tenosynovitis can result from repetitive trauma or from an inflammatory process such as rheumatoid arthritis, and it is usually depicted on MRI as increased fluid, sometimes containing complex debris or bodies within the bicipital tendon sheath. The tendon sheath normally communicates with the shoulder joint; therefore, only significant amounts of fluid isolated to the bicipital tendon sheath or fluid quantities out of proportion to associated joint effusions are suggestive of this diagnosis. Tendinosis is a little farther along the spectrum

A

981

of injury and includes degeneration of the bicipital tendon. This is depicted on MRI as thickening and increased signal within the substance of the tendon (Fig. 17H3-42).171,172 The “hourglass” biceps tendon describes a condition in which there is marked tendinosis and hypertrophy isolated to the intra-articular portion of the long bicipital tendon, which prevents the tendon from sliding into the bicipital groove during elevation of the arm.173 Entrapment of the long bicipital tendon causes a mechanical block and pain. This condition occurs most often in association with a full-thickness tear of the rotator cuff, although there are case reports of entrapment occurring in association with partial-thickness rotator cuff tears as well. Patients present with anterior arm pain and loss of passive elevation of the arm averaging about 10 to 20 degrees. The condition is treated with resection of the abnormal segment and tenodesis of the biceps tendon followed by appropriate treatment of any concomitant injury of the rotator cuff. Patients experience immediate recovery of their full range of motion. MRI will show marked tendinosis and thickening limited to the intra-articular portion of the biceps tendon (Fig. 17H3-43) and is often associated with either a partial- or full-thickness rotator cuff tear. Partial or complete tear of the long bicipital tendon may result from repetitive microtrauma or from an isolated traumatic injury. It most commonly occurs in the middleaged man as a result of impingement, with rupture usually occurring near the proximal aspect of the intertubercular sulcus. Findings on MRI may include absence of the tendon within the intertubercular groove secondary to retraction of the tendon. A less dramatic finding may be fluid signal (bright on T2-weighted images) extending partially through the tendon with a partial tear or completely traversing the tendon with a full-thickness tear. Subluxation or dislocation of the tendon out of the intertubercular sulcus can result from injury to the overlying stabilizing structures.106,107 A tear of the coracohumeral ligament will result in medial subluxation of the long bicipital tendon, and three patterns of subluxation

B

Figure 17H3-42   Biceps tendinopathy, magnetic resonance arthrography. Oblique coronal (A) and axial (B) T1-weighted images demonstrate thickening and intrasubstance signal (arrows) involving both the intra-articular portion of the tendon (coronal image) and the intertubercular groove portion of the tendon (axial image).

982

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

have been described that are related to the status of the subscapularis tendon and the coracohumeral ligament (Table 17H3-8). First, disruption of the coracohumeral ligament in conjunction with a complete tear or avulsion of the subscapularis tendon off of the lesser tuberosity will result in an intra-articular subluxation or dislocation. The long bicipital tendon will sublux medially into the anterior aspect of the glenohumeral joint, medial to the intertubercular groove. Second, a tear of the coracohumeral ligament in conjunction with a tear of the transverse humeral ligament (superficial fibers of the subscapularis tendon) will result in extra-articular subluxation of the

B

Figure 17H3-43   Hourglass biceps tendon, magnetic resonance imaging. Oblique coronal (A) and axial (B) T2-weighted images show marked thickening and tendinosis of the intraarticular portion of the long bicipital tendon (arrows). Axial   (C) T2-weighted image shows a normal-appearing biceps tendon (arrow) at the level of the intertubercular groove. This pattern of tendinosis with marked thickening isolated to the intra-articular portion of the tendon can result in a mechanical block with entrapment of the tendon, resulting in limited range of motion and pain.

  TABLE 17H3-8  Patterns of Medial Subluxation: Long Bicipital Tendon Torn Structures Coracohumeral ligament Subscapularis tendon Coracohumeral ligament Transverse humeral ligament Isolated coracohumeral ligament tear

Location of Medially Subluxed Tendon Intra-articular Extra-articular (superficial to subscapularis) Within substance of subscapularis tendon or muscle (“hidden lesion”)

Shoulder

long bicipital tendon, and the tendon will be located medial to the intertubercular groove but superficial to the intact fibers of the subscapularis tendon. Finally, an isolated tear of the coracohumeral ligament with an intact subscapularis tendon and transverse humeral ligament will allow medial migration of the long bicipital tendon into the substance of the subscapularis tendon, resulting in interstitial tearing of the tendon. This is the so-called hidden lesion because the abnormal position of the tendon may be hidden to the arthroscopist. Preoperative MRI can alert the surgeon to the abnormal position of the long bicipital tendon as axial MRI nicely depicts the location of the long bicipital tendon medial to the intertubercular groove sitting within the substance of the subscapularis tendon (see Fig. 17H3-21).

Osseous Structures and Articular Surfaces Osseous structures of the glenohumeral joint are usually adequately imaged by conventional radiography, although more detailed evaluation is required in some instances. In addition, osseous abnormalities are often incidental findings on a variety of imaging modalities. CT is generally considered the modality of choice for detailed evaluation of fractures about the shoulder (see Fig. 17H3-4). CT and MRI, however, can adequately characterize anterior and posterior glenoid rim fractures as to the size and displacement of fracture fragments as well as identify humeral head compression fractures (i.e., Hill-Sachs fracture). MRI has the added benefit of increased accuracy in the depiction of associated soft tissue abnormalities. CT is usually reserved for the evaluation of complex fractures involving the glenohumeral joint. CT can be used to characterize the size and number of intra-articular fragments as well as to identify rotated or dislocated fragments. Although CT is considered the imaging modality of choice for the evaluation of complex fractures about the shoulder, many other osseous abnormalities are evaluated best with MRI, which is more sensitive in detecting

A

983

­ arrow abnormalities. Osseous abnormalities that may m occur around the shoulder include primary or secondary tumors, infiltrative marrow processes, infection, and avascular necrosis. The humeral head normally appears bright on T1-weighted images and dark on T2-weighted images with fat saturation. This is due to the abundance of fat cells normally present in the epiphysis. The proximal humeral metaphysis and diaphysis, on the other hand, contain hematopoietic marrow cells and fewer fat cells and therefore demonstrate intermediate signal intensity on T1-weighted images and intermediate to bright signal intensity on T2-weighted images with fat saturation. The scapula and glenoid can contain either fat or hematopoietic cells and thus demonstrate a mixed MR appearance. Skeletal muscle can act as an internal standard in evaluating the appearance of the adjacent osseous structures. Normal hematopoietic marrow should always be brighter than adjacent muscle on T1-weighted images, and any marrow signal that appears darker than the adjacent muscle on T1-weighted images should be considered abnormal.174 This is a nonspecific finding and may represent a number of pathologic conditions including infection, trauma, and neoplasm. A variety of primary and secondary tumors can involve the osseous structures about the shoulder. The differential diagnosis is typically established on the basis of the conventional radiograph, but MRI may contribute additional information. The primary role of MRI is to ­demonstrate the extent of osseous and soft tissue involvement. Most osseous tumors have a nonspecific MR appearance demonstrating decreased signal intensity on T1-weighted images and increased signal intensity on T2-weighted images (Fig. 17H3-44). A few tumors have specific MRI characteristics that aid in the differential diagnosis. A simple cyst of the proximal humeral shaft displays homogeneous signal intensity similar to water on both T1-weighted and T2-weighted images (low signal on T1-weighted images and bright signal on T2-weighted images). An aneurysmal bone cyst also has a fairly specific

B

Figure 17H3-44   Metastatic adenocarcinoma involving the proximal humerus, magnetic resonance imaging. Oblique coronal T1-weighted image (A) and T2-weighted image with fat saturation (B) demonstrate a large destructive mass involving the proximal humeral shaft (arrows). The mass is isointense to muscle on the T1-weighted image, and it is heterogeneous with areas of bright increased signal on the T2-weighted image. Destruction of the cortex (arrowhead) of the humeral shaft is also noted.

984

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

MR appearance demonstrating expansion of cortex with multiple fluid-fluid levels and variable MR signal characteristics based on the presence of various blood breakdown products.175 MRI plays a critical role in preoperative planning by defining the extent of osseous and soft tissue involvement of malignant osseous neoplasms. MRI is more sensitive than conventional radiography in depicting the extent of osseous involvement, clearly depicting intramedullary extent as well as cortical disruption or breakthrough (see Fig. 17H3-44). Involvement of adjacent soft tissue structures, such as neurovascular bundles, is also clearly depicted with MRI. Infection of the glenohumeral joint or adjacent osseous structures is an unusual condition but can occur secondary to direct penetrating trauma, as a result of prior surgery, or by hematogenous spread. Osteomyelitis is identified much earlier on MRI than on CT or conventional radiography. In the past, nuclear scintigraphy was considered the imaging modality of choice, but it lacks the spatial resolution provided by MRI. In the setting of clinically suspected osteomyelitis or septic arthritis, MRI provides a high level of sensitivity as well as significantly improved spatial ­resolution relative to nuclear scintigraphy. Osteomyelitis appears as decreased signal intensity on T1-weighted images and increased signal intensity on T2-weighted images and will enhance after the intravenous administration of a contrast agent. The presence of abscess or septic arthritis is also depicted with MRI. Glenohumeral chondrolysis is a rare but devastating complication that has been reported following shoulder arthroscopy and is most commonly seen in young patients, typically between the ages of 20 and 30 years, following shoulder reconstruction for glenohumeral instability. The exact cause is unclear. Proposed causes include use a thermal probe at the time of capsulorrhaphy, use of a morphine pump, an unknown infectious agent, or possibly an event during arthroscopy that triggers an immune response and subsequent migration of inflammatory cells into the glenohumeral joint.176,177 Acute onset of chondrolysis is manifested by rapid destruction of the glenohumeral articular cartilage and leads to progressive shoulder pain, disability, and rapidly developing glenohumeral osteoarthritis. MRI reveals complete loss of the glenohumeral articular cartilage on both sides of the joint, joint space narrowing, and subchondral cyst formation. The MRI appearance is similar to that of advanced osteoarthritis of the glenohumeral joint, except there is a lack of osteophyte formation (Fig. 17H3-45). The MRI appearance differs from infection in that there is a paucity of joint fluid and no evidence of synovitis. Treatment is supportive, and most patients eventually require arthroplasty of the involved joint. The surgical history is very important in accurately establishing this diagnosis because most patients present within the first 2 years after shoulder arthroscopy with progressive shoulder pain and extensive loss of the normal range of motion. Invariably, at the time of shoulder reconstruction, the glenohumeral articular cartilage was reported as normal.

Avascular necrosis commonly occurs in the humeral head, and although there are many potential causes of avascular necrosis, the most common are trauma and steroid use. Conventional radiography is insensitive for the early detection of avascular necrosis. Nuclear scintigraphy and MRI are sensitive in the early detection of avascular necrosis, but MRI has a much higher specificity and commonly identifies other abnormalities about the shoulder.178 Because of its high sensitivity and specificity, MRI is considered the imaging modality of choice in the evaluation of avascular necrosis.179,180 Early MR findings in avascular necrosis of the humeral head (Fig. 17H3-46) include subchondral marrow edema (bright on T2-weighted images and dark on T1-weighted images). The subchondral edema is typically geographic in appearance. As the process evolves, a reactive interface develops between the avascular and living bone. This appears as the double-line sign on T2-weighted images (a bright line adjacent to a more proximal dark line that demarcates the interface between living and dead bone). This sign is nearly pathognomonic for avascular necrosis and is thought to represent an interface of granulation tissue.181 As the process progresses even further, the humeral head may collapse, leading to irregularity of the humeral head and eventually to secondary osteoarthritis of the glenohumeral joint.

Figure 17H3-45   Chondrolysis of the glenohumeral   joint, magnetic resonance imaging. Oblique coronal   T2-weighted image with fat saturation shows diffuse chondral loss (long arrow) and joint space narrowing. Numerous areas of subchondral marrow edema and early subchondral cyst formation (short arrows) are noted on both sides of the joint. This patient had undergone arthroscopic shoulder reconstruction for glenohumeral instability about 6 months previously and had normal articular cartilage at the time of arthroscopy.

Shoulder

985

l Sonography has been shown to be an accurate imaging method for detecting and characterizing lesions of the rotator cuff. Sonography, however, has a steep learning curve and is very operator dependent. This, coupled with the fact that MRI provides a more global evaluation of the entire shoulder including the labrum and osseous structures, continues to be a factor limiting the use of sonography in the evaluation of the glenohumeral joint. l The multiplanar capabilities and superb soft tissue contrast of MRI makes it the imaging modality of choice for the global evaluation of the glenohumeral joint, including the evaluation of the osseous outlet and acromion, muscles, tendon, capsular structures, and labrum. l Direct MR arthrography provides the most accurate assessment of partial-thickness articular-sided tears of the rotator cuff and of labral abnormalities. l Computed arthrotomography provides an acceptable alternative for the evaluation of labral abnormalities in patients that have a contraindication to MRI.

Figure 17H3-46   Avascular necrosis of the humeral head, magnetic resonance imaging. Coronal oblique T2-weighted image demonstrates the typical magnetic resonance imaging characteristics of avascular necrosis. There is a well-marginated geographic-appearing lesion in the subchondral region of the humeral head. The presence of the double-line sign—the bright signal intensity line paralleling an area of low signal (arrows)—is specific for avascular necrosis. The integrity of the overlying subchondral bone plate is maintained.

C

r i t i c a l

P

o i n t s

l Conventional radiography plays an important role in the initial evaluation of most suspected abnormalities of the glenohumeral joint. Knowledge of the numerous radiographic projections including the value and limitations of each view is crucial in tailoring an appropriate imaging examination. l CT is the examination of choice for detecting and describing complex fractures of the glenohumeral joint and in particular in detecting and describing the extent of an osseous Bankart lesion.

S U G G E S T E D

R E A D I N G S

Beltran J, Kim DH: MR imaging of shoulder instability in the athlete. Magn Reson Imaging Clin N Am 11:221-238, 2003. Bencardino JT, Rosenberg ZS: Entrapment neuropathies of the shoulder and elbow in athletes. Clin Sports Med 25:465-487, 2006. Chaipat L, Palmer WE: Shoulder magnetic resonance imaging. Clin Sports Med 25:371-386, 2006. Jung JY, Jee WH, Chun HJ, et al: Adhesive capsulitis of the shoulder: Evaluation with MR arthrography. Eur Radiol 16:791-796, 2006. Kassarjian A, Bencardino JT, Palmer WE: MR imaging of the rotator cuff. Magn Reson Imaging Clin N Am 12:39-60, 2004. Mohana-Borges AVR, Chung CB, Resnick D: Superior labral anteroposterior tear: Classification and diagnosis with MRI and MR arthrography AJR Am J Roentgenol 181:1449-1462, 2003. Moosikasuwan JB, Miller TT, Dines DM: Imaging of the painful shoulder in throwing athletes. Clin Sports Med 25:433-443, 2006. Morag Y, Jacobson JA, Shields G, et al: MR arthrography of rotator interval, long head of the biceps brachii, and biceps pulley of the shoulder. Radiology 235: 21-30, 2005. Sanders TG, Miller MD: Systematic approach to magnetic resonance imaging interpretation of sports medicine injuries of the shoulder. Am J Sports Med 33:1088-1105, 2005. Sanders TG, Morrison WB, Miller MD: Imaging of glenohumeral instability: Current concepts. Am J Sports Med 28:414-434, 2000.

R eferences Please see www.expertconsult.com

986

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S E C T I O N

I

Rotator Cuff 1. Impingement Lesions in Adult and Adolescent Athletes Kenneth C. Lin, Sumant G. Krishnan, and Wayne Z. Burkhead

HISTORICAL REVIEW Although anatomic descriptions of the rotator cuff date back to antiquity, appreciation for its function and therefore dysfunction is a relatively recent development. Monro in 1788 illustrated tearing of the supraspinatus and infraspinatus, but it was not until almost a half century later in 1834 that John Smith published the first detailed series of rotator cuff tears and not until another 50 years had passed that Muller published the earliest report of a rotator cuff repair in 1889. Although these early developments certainly laid the groundwork, it was not until 1934 when E. A. Codman published his landmark work, The Shoulder, that our modern understanding of the rotator cuff, rotator cuff injury, and its evaluation and management began (Fig. 17I1-1).1-3 In this one comprehensive text, Codman described the fundamental pathology and pathophysiology of the rotator cuff and detailed the associated clinical findings and treatment options. Our understanding of the etiology of cuff pathology has continued to evolve over the years. Codman believed in a purely traumatic cause for rotator cuff injury. In 1937, Meyer argued that repetitive “minor” trauma was in fact the underlying cause. He was the first to describe the concept of “outlet” impingement, in which the cuff is caught between the greater tuberosity and the acromion.4 Neer expanded on this concept and described the technique of anterior acromioplasty directed at relieving this outlet impingement.5 Neer considered impingement and rotator cuff injury as a spectrum of pathology that could

Fibrosynovial edge Erosion of articular cartilage Stub of tendon Recession of tuberosity Fluid axillary pouch

be divided into three separate stages. Initially, in stage I, there is inflammation and edema within the cuff. This is followed by the fibrosis and tendonitis seen in stage II. Finally, there is partial or complete tearing of the rotator cuff in stage III.6,7 Although this progression of pathology may be an accurate description of what occurs in the general population with degenerative tearing of the rotator cuff, caution must be exercised when extending this thinking to the athlete because it may be an oversimplification.8-15 The shoulder in the athlete can experience extremes of loading and motion that make it susceptible to a unique set of pathologic processes that deserve special consideration. This is especially true in the case of overhead athletes. Peak shoulder distraction forces have been reported to be 80% to 100% of body weight in baseball pitchers. Shoulder distraction forces as high as 108% of body weight were reported in a study of professional pitchers.16 Therefore, in considering rotator cuff injuries in athletes, it is important to take these unique conditions into account because the underlying pathology and therefore treatment may be distinctly different. In the older athlete, a primary tendinopathy may occur within the substance of the cuff, with secondary acromial changes and compression of the cuff under the coracoacromial arch.12,17 However, in younger overhead athletes, classical outlet impingement is most often caused by subtle glenohumeral laxity or instability leading to muscle imbalance, scapulothoracic dyskinesia, and subsequent subacromial space impingement.8,15,18 Another consideration in overhead athletes is the possibility of “internal impingement.” This phenomenon has been described in overhead athletes in whom physiologic contact of the articular surface of the rotator cuff with the posterosuperior glenoid rim may become pathologic owing to repetitive supraphysiologic activity leading to cuff pathology (Fig. 17I1-2). These less common circumstances, leading ultimately to impingement or cuff pathology, underscore the differences in the types and causes of rotator cuff pathology and impingement lesions in the athlete that should encourage vigilance for these variations because successful treatment necessitates their recognition.19-21

EPIDEMIOLOGY Figure 17I1-1   Diagram of a rotator cuff tear from E. A. Codman’s The Shoulder published in 1934.

Tears of the rotator cuff have been reported in up to 60% of cadaveric specimens. Using magnetic resonance imaging (MRI), Sher and colleagues found the prevalence of rotator

Shoulder

987

A

B

C

Figure 17I1-2   A-C, Contact of the articular surface of the rotator cuff with the posterosuperior glenoid rim.

cuff tearing to be 34% overall in asymptomatic individuals.22 More importantly, he showed a direct correlation between age and asymptomatic rotator cuff tears detected by MRI. In this study group, 4% of those between the ages of 19 and 34 years had tearing of the rotator cuff detectable on MRI scanning. In contrast, 54% of those over the age of 60 years had tearing of the rotator cuff, partial or complete, detectable on MRI. This reflects the likely degenerative tearing of the rotator cuff over time consistent with Neer’s three-stage description. Because rotator cuff tearing can exist in the absence of symptoms and because of the high incidence of rotator cuff tears in the older population, the question becomes which rotator cuff tears require treatment and which do not. Yamaguchi and associates have recently shed some light on this question in a study of asymptomatic rotator cuff tears over time.23 Their findings suggest that pain associated with a rotator cuff tear is likely indicative of progression of that tear. With this knowledge, the physician can counsel

the patient and recommend treatment, understanding that larger tears and chronic tears are more difficult to treat. Because degenerative changes may not be the singular cause of rotator cuff tearing in the athlete, other factors such as the demands of the athlete’s specific sport must be taken into consideration because the incidence of impingement and rotator cuff tearing may be quite different in each of these sport-specific subpopulations. For example, shoulder injuries in football players are usually the result of traumatic dislocations or acromioclavicular separations, with a much smaller incidence of rotator cuff or bicipital tendinitis.24 Conversely, 40% to 70% of swimmers have been reported to experience shoulder pain that is often the result of impingement. And in a series of world-class tennis players, Nirschl found that more than 50% suffered from shoulder problems involving the rotator cuff and biceps tendon.25 Although the original descriptions of rotator cuff tears were in laborers, rotator cuff lesions have been demonstrated as a considerable concern in athletes, and in

988

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

overhead athletes in particular. Because the shoulder is primarily involved in many overhead sporting activities (such as baseball, golf, and volleyball) the true incidence of both external and internal impingement lesions of the rotator cuff may be much higher than in the general populace.

Current Concepts and Controversies Since Codman’s pioneering work, our knowledge and understanding of the rotator cuff and rotator cuff pathology have continued to expand. Exactly how much more remains to be learned is illustrated by the fact that information as fundamental as the natural history of rotator cuff tearing is only recently becoming clearer. The importance of understanding the natural history of rotator cuff tears becomes obvious when one remembers that Sher and colleagues found that 54% of asymptomatic individuals older than 60 years had MRI findings consistent with rotator cuff tearing.22 Milgram had similar findings showing that 50% of patients older than 70 years had rotator cuff disease on ultrasound.25a Yamaguchi and associates retrospectively reviewed 588 symptomatic patients who had been evaluated by ultrasonography for unilateral shoulder pain.26 Like Sher and Milgram, they also found that rotator cuff tearing was highly correlated with age. They also found that patients older than 66 years had a 50% likelihood of rotator cuff tearing in the contralateral, asymptomatic shoulder. Because rotator cuff tears can be present in both symptomatic and asymptomatic shoulders, the presence of tearing is not prognostic and cannot be the sole guide to treatment. To clarify the natural history of rotator cuff tears, Yamaguchi and associates looked at asymptomatic rotator cuff tears over time.23 They reassessed 45 patients with asymptomatic tearing of the rotator cuff an average of 5.5 years after their initial assessment. About half of patients had become symptomatic at follow-up, with symptoms starting at an average of 2.8 years after initial evaluation. Twentythree of these patients were also available for re-evaluation by ultrasound at follow-up. Although this was not a randomized, prospective study and the number of patients was small, the study did suggest two interesting trends. First, none of the tears healed. Even when a tear continued to be asymptomatic, the tear itself did not decrease in size, and some tears even increased in size. Second, whereas 22% of those who remained asymptomatic had ultrasound documented progression, 50% of those who had become symptomatic had ultrasound evidence of tear progression. Given the increasing availability and image quality of MRI, our ability to detect rotator cuff pathology continues to increase. The presence of this pathology is therefore not as primarily important as the significance of this pathology. These preliminary studies suggest that whereas a lack of symptoms may indicate a stable rotator cuff tear, symptomatic pain may correlate with tear progression and a need to consider intervention. When surgical intervention is indicated, anatomic repair of the rotator cuff is certainly the treatment of choice. Our ability to repair rotator cuff tears through an arthroscopic approach has advanced, and results equivalent to those of open repair have been shown to be possible. The results of large and massive rotator cuff tears repaired either arthroscopically or with open techniques have been

reported to include a high incidence of re-tear on followup with ultrasound.27 Galatz and coworkers reported ultrasound evidence of tear recurrence in 17 of 18 rotator cuff tears that had undergone complete arthroscopic repair.27 Despite this structural failure, the clinical results can be satisfactory. Jost and colleagues followed 20 open rotator cuff repairs found to have structural failure of their repair over time.28 At an average of 7.6 years’ follow-up, patients continued to experience improvements in pain relief, function, strength, and satisfaction. Jost and colleagues also found that re-ruptures had not progressed when compared with prior MRI performed at 3.2-year follow-up and that tears smaller than 400 mm2 had the potential to heal ­consistently.28 Attempts to improve arthroscopic repair have focused on the footprint of the rotator cuff and a more anatomic re-creation of this footprint through improvements in repair technique involving dual rows of fixation. The footprint of the rotator cuff was described by Minagawa and associates in 199829 and Tierney and Scheller in 1999.30 Since then, a number of double-row rotator cuff repair techniques have been described in an effort to better re-create this area of contact and (in theory) result in higher contact pressures, better tendon healing, and therefore better clinical results (Fig. 17I1-3). Biomechanical in vitro and arthroscopic in vivo measurements have confirmed the strength and the increased contact with the footprint with double-row fixation when compared with single-row fixation.28,31-34 Intuitively, increased contact area would provide a greater opportunity for tendon healing. However, clinical benefits to such a construct remain to be proved. It is important to remember that the principles of successful rotator cuff repair are the same regardless of whether it is performed through an open incision or through an arthroscopic portal. Re-creation of the anatomic footprint, if it proves to be important to successful arthroscopic rotator cuff repair, should likewise be important to open repair.

Figure 17I1-3   Arthroscopic view of a double row repair viewed from a lateral portal. Two lateral row suture anchors are seen in the lower half of the screen. A single medial row suture anchor is seen in the upper left hand side of the picture.

Shoulder

989

human dermis show promise.37 Prospective, long-term, randomized controlled studies remain to be done to demonstrate the success of such grafts. The effectiveness of these materials is one issue, but the indications, technique, and mode of placement are also still unclear.

PERTINENT ANATOMY The shoulder is a ball-and-socket joint linking the axial trunk and appendicular upper extremity. The glenohumeral joint is the most mobile in the body, allowing for precise positioning of the hand in space. The shoulder is also the fulcrum for the long lever arm of the upper limb and consequently absorbs the majority of forces in sports that require propulsive action of the upper extremity. The rotator cuff is vitally linked to these motions in terms of both precision and propulsion. The cuff is ­composed of the confluent tendons of the supraspinatus, infraspinatus, subscapularis, and teres minor muscles (Fig. 17I1-5).38 The tendon of the long head of the biceps is intimately associated with the cuff and has been called the “fifth” tendon of the rotator cuff. The cuff envelops and blends with the glenohumeral capsule on all sides except at the redundant inferior pouch. The biceps tendon originates at the supraglenoid tubercle and traverses the glenohumeral joint as an intra-articular but extrasynovial structure because it is lined by a synovial sheath. The biceps passes deep to the interval between the supraspinatus and subscapularis (the “rotator interval”) and exits the joint in the intertubercular sulcus, which is bounded by the coracohumeral ligament superiorly and the superior glenohumeral ligament inferiorly. These ligaments form a “pulley” for the biceps tendon as it enters the intertubercular groove. The groove has a variable shape and depth, and the bony anatomy of the supratubercular region has been implicated in degenerative lesions of the biceps tendon.39-41 Distal to its articular portion, the biceps is held in the intertubercular groove by the transverse humeral ligament. The vascular supply of the biceps and rotator cuff has been extensively studied.42-45 Anatomic studies have demonstrated that the vascular supply of the rotator cuff comes

Figure 17I1-4   Sagittal magnetic resonance image of the shoulder showing fatty infiltration of the supraspinatus, infraspinatus, and subscapularis muscles.

Success of rotator cuff repair has also been shown to be dependent on a biologic environment conducive to rotator cuff tendon healing. With long-standing, chronic, massive rotator cuff tearing, fatty infiltration of the rotator cuff tendons can occur. Goutalier and coworkers correlated the importance of preoperative fatty infiltration as a predictor of postoperative repair success.35,36 Fatty infiltration has been shown to be irreversible (Fig. 17I1-4). To address the problems of repairing rotator cuff tears in the face of poor or inadequate tissue, some have turned to augmentation with tendon transfers or grafts. Tendon transfers can provide viable, healthy tissue for autologous augmentation. Depending on the specific transfer, such a procedure can potentially improve function not simply by improving the likelihood of healing but through the rerouted muscles activation with motion as well. Various biologic materials have been used with mixed results. ­Xenograft porcine intestinal submucosa has been shown not to be effective, but other materials such as allograft

Supraspinatus

Supraspinatus

Infraspinatus

Transverse ligament Biceps (long head)

Teres minor

A

B

Figure 17I1-5   Posterior and anterior views of the rotator cuff musculature.

Subscapularis

990

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

from six branches of the axillary artery with the largest contributions arising from the suprascapular and the anterior and posterior humeral circumflex arteries.45 The pattern of arterial supply appears to result in an area of relatively poor vascularity known as the “critical zone.” This area of the cuff corresponds with the area commonly associated with degenerative changes and where most degenerative rotator cuff tearing begins.1,2,46 This area lies within the supraspinatus tendon immediately proximal to its insertion onto the greater tuberosity.47 Despite this correlation, the theory of vascular insufficiency as a predisposing factor leading to rotator cuff injury remains controversial. The biceps tendon also demonstrates an area of hypovascularity in its intra-articular portion related to tension or pressure from the humeral head when the tendon is in the anatomic position. With arm abduction, these areas demonstrate complete vascular filling.47 Superficial to the rotator cuff is the deltoid and coracoacromial arch. The acromion is an extension of the spine of the scapula and has a variable shape and slope that forms the posterolateral bony roof of the arch.48 The acromion provides bony protection to the glenohumeral joint but also creates a finite space between its undersurface and the humeral head. The coracoacromial ligament extends from the outer edge of the coracoid and widens to insert on the anteromedial aspect and undersurface of the acromion. The coracoacromial ligament encompasses the anterior extent of the coracoacromial arch and, with the anteroinferior edge of the acromion and the coracoid process, is implicated in classic extrinsic impingement of the rotator cuff (Fig. 17I1-6).5,49 Matsen’s concept of the “humeroscapular articulation” is useful for understanding the coracoacromial arch with regard to normal shoulder motion and abnormal shoulder impingement. In this model, the shoulder is seen as two ­concentric spheres. In the inner sphere, the humeral head articulates with the glenoid. The outer sphere is represented by the proximal humerus and the coracoacromial

Acromion

Coracoacromial ligament

arch. With normal shoulder function, these two spheres are concentric with the substance of the rotator cuff acting secondarily as a passive spacer between the proximal humerus and the coracoacromial arch. Under normal circumstances, because of the concentric nature of the two articulations, there is very little demand placed on the spacer function of the cuff. When there is a loss of concentricity or an alteration in morphology of the coracoacromial arch, impingement can result. It remains a matter of controversy whether acromial arch morphology and potential extrinsic compression are clearly related to primary rotator cuff degeneration or rather are a secondary phenomenon.50 Deep to the coracoacromial arch lies the filmy synoviumlined sac known as the subacromial bursa, which attaches at its base to the greater tuberosity with its roof fixed to the undersurface of the acromion and coracoacromial arch (Fig. 17I1-7).51 The remainder of the superior and inferior surfaces of the bursa loosely articulate with the deltoid and rotator cuff, respectively. Although the roof and base of the bursa are in intimate contact, the two layers are separated by a thin interface of synovial fluid that allows relatively frictionless motion between the cuff and the overlying deltoid and coracoacromial arch. Inferior to the subacromial bursa, the rotator cuff and biceps tendon reside within the glenohumeral joint. The four tendons insert as a composite into the greater and lesser tuberosities. The “rotator interval” is an anatomic space defined by the inferior edge of the supraspinatus tendon and the superior edge of the subscapularis tendon.52-54 The superficial roof of the rotator interval is the coracohumeral ligament, and the floor of the interval is the superior glenohumeral ligament (SGHL). This interval is occupied by the biceps tendon as it enters the shoulder joint, with the coracohumeral ligament and SGHL forming a “pulley” for the biceps tendon (Fig. 17I1-8). The rotator interval has been demonstrated biomechanically to function as a suspensory structure for the humeral head, and lesions of the rotator interval have been recognized as an important pathology in the genesis of shoulder pain.38,55 As mentioned previously, some authors have suggested that the shape and slope of the acromion may be related to extrinsic rotator cuff pathology.5,49 However, it remains controversial whether the variability in acromial shape is the result or the cause of the underlying cuff degeneration (Fig. 17I1-9).50

RELEVANT BIOMECHANICS Coracoid

Because the glenohumeral joint lacks inherent bony stability, it relies heavily on both static and dynamic soft tissue stabilizers for its stability and function.56,57 The muscles of the rotator cuff do contribute to glenohumeral motion. But more importantly, they help maintain a stable fulcrum at the glenohumeral joint around which the other muscles of the shoulder girdle can effectively act.

Cuff Function Figure 17I1-6   Diagram of the coracoacromial arch consisting of the acromion, coracoacromial ligament, and coracoid.

Although previously thought to initiate abduction, the supraspinatus is presently considered to function ­primarily as a stabilizer of the glenohumeral joint. Its orientation 70 degrees from the plane of the glenoid means it provides

Shoulder Coracoacromial arch Rotator interval

Greater tuberosity

Supraspinatus Infraspinatus Teres minor

991

Figure 17I1-7   Diagram of the coracoacromial arch with underlying subacromial bursa and cuff tendons.

Subacromial bursa Subscapularis (lesser tuberosity)

Biceps (long head)

a compressive force driving together the humeral head and the glenoid cavity (Fig. 17I1-10).58 By maintaining this articular congruity through concavity-compression, a stable fulcrum is created for the more powerful muscles of the shoulder girdle. The powerful deltoid, for example, requires this stability at the glenohumeral joint to function effectively. Without the stabilizing, synergistic action of the supraspinatus, the humeral head would displace superiorly as the deltoid contracted and would result in impingement of the rotator cuff between the humeral head and the undersurface of the acromion.58,59 By virtue of their orientation, the action of the infraspinatus and teres minor muscles is external rotation of the

arm and depression of the humeral head.60 The infraspinatus is in fact the primary depressor of the humeral head. Along with the infraspinatus and teres minor muscles the subscapularis muscle likewise depresses the humeral head. (It also acts as an internal rotator of the arm.) The infraspinatus and subscapularis are two of the most important ­stabilizing muscles of the shoulder especially during eccentric contraction and overhead activity.61 The concept of the infraspinatus and subscapularis acting as a “force couple” is a useful one. While the rotator cuff provides the stability and the “fine tuning” of motion, the large superficial muscles around the glenohumeral joint (such as the deltoid, trapezius,

Supraspinatus Coracohumeral ligament

Biceps (long head)

A

Subscapularis

B

Figure 17I1-8   A, Anterior cuff structures with biceps tendon entering the rotator interval. B, Arthroscopic view from within the glenohumeral joint. The biceps tendon is seen entering the joint.

992

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17I1-9   Variability in acromial morphology. Lateral view of a normal acromion (A) and lateral view of a more hooked acromion associated with impingement (B).

l­atissimus dorsi, and pectoralis major) provide the power for movements of the shoulder. These muscles provide their propulsive force through their powerful concentric contraction. It is important to recognize that at the same time as this is occurring, the muscles of the rotator cuff are providing a crucial stabilizing force at the glenohumeral joint through their eccentric contraction.

Biceps The long head of the biceps, although implicated as a humeral head depressor, is most likely a passive player during most shoulder motions (Fig. 17I1-11).57,62 Yamaguchi and associates used electromyography to assess the activity of the long head of the biceps with shoulder-related activity.63 They controlled elbow function with the use of a brace

that locked the elbow at 100 degrees of flexion and neutral forearm rotation. With elbow motion thus eliminated, they demonstrated that the long head of the biceps is essentially inactive during shoulder-related activities in normal shoulders. Furthermore, they showed that the presence of a rotator cuff tear results in the same lack of activity. Although the long head of the biceps tendon may not actively participate in shoulder stability, it can be abnormally loaded. In the presence of a rotator cuff tear, for example, the biceps may become inflamed or torn through pathologic loading. This may especially be the case if the tear involves both the subscapularis and the anterosuperior cuff. In the absence of a rotator cuff tear, the biceps tendon may be abnormally loaded in the overhead athlete. Rodosky and colleagues demonstrated in an in vitro cadaveric model that the long head of the biceps may passively contribute to anterior stability of the glenohumeral joint in the abducted and externally rotated position by increasing the resistance of the shoulder to torsional forces.63a Hence, the biceps may be important in combined shoulder and elbow function in the overhand or throwing motion during athletic activity.

Static Stabilizers The static structures of the shoulder such as the glenohumeral ligaments are important for stability but also may be implicated in the impingement phenomenon.53 For ­example, tight posterior structures cause greater anterior translation of the humeral head with forward elevation and thus may contribute to secondary impingement.64 Similarly, anterior laxity and subluxation may result in compromise of the available subacromial space, leading to classic outlet impingement, or may result in increased hyperangulation of the humeral head in the abduction and external

70˚

Figure 17I1-10   Angle of pull of supraspinatus with direct line of force (solid arrow) and compressive component of force (broken arrow).

Figure 17I1-11   By virtue of its location, the biceps tendon can resist superior translation of the humeral head in situations where normal restraints have failed.

Shoulder

993

Figure 17I1-12   Schematic of the stages of overhead throwing.

rotation position, leading to posterosuperior glenoid or “internal” impingement.65,66

anterior, and triceps provide additional force through concentric activity.

The Throwing Motion

Deceleration or Follow-Through Phase

Overhead motion, as exemplified by throwing, is the most common motion affecting the shoulder in sports (Fig. 17I1-12).8,67 This action can be divided into three phases.11,58,68-70

The final phase, deceleration or follow-through, involves the entire rotator cuff. The subscapularis continues to fire, producing internal rotation of the arm. The posterior cuff is working to decelerate the arm and maintain the humeral head within the glenoid. The supraspinatus is also continuing to be active. In fact, the supraspinatus is active throughout all stages of throwing, but the intensity of its activity is only moderate. This likely reflects its role as a stabilizer by compressing together the humeral and glenoid articulating surfaces. The long head of the biceps tendon, apart from being active during the cocking phase in which the elbow is actively bent, also comes into play during deceleration. It is at peak activity following ball release when it acts eccentrically to decelerate terminal elbow extension. This constitutes the active or dynamic description of the throwing motion. Passively, it has been demonstrated that there are obligatory translations of the glenohumeral joint during this motion. The humeral head moves posteriorly with extension or external rotation and anteriorly with flexion or cross-body motion. The humeral head also hyperangulates during late cocking in abduction and external rotation. However, a clear understanding of the interplay between the dynamic and passive motions in the shoulder remains elusive. Recognition of these components implies that an even greater complexity of these functions exists. The actions comprising the overhead throw, the tennis serve, the javelin throw, and the various swimming strokes are all made up of relatively similar mechanisms. Differences exist in the equipment involved, the associated body motions, and the position of the shoulder in each action. Of importance is the degree, repetitiveness, and nature of the forces involved and whether any impact occurs such as in spiking a volleyball. Because of the biomechanical action of the rotator cuff, dysfunction due to injury or disease can easily lead to significant problems, particularly in the athlete’s shoulder in which the stresses are so great.

Wind-Up or Cocking Phase The wind-up or cocking phase involves abduction, extension, and external rotation. Some authors divide the windup and cocking phases and further subdivide the cocking phase into early cocking and late cocking. However, from the standpoint of muscle activation, they can be considered as one. During this phase, the primary muscle involved is the deltoid, and the result is an anterior levering of the humerus. This may also cause superior migration of the humeral head, especially during late cocking or with abduction and external rotation of the arm. A natural consequence of this is tension on the anterior joint structures and potentially secondary impingement. To counteract this, the muscles of the cuff (through their action) provide a stabilizing force. The supraspinatus contributes to stability by compressing the humeral head into the glenoid. The infraspinatus and teres minor also help to maintain the humeral head within the glenoid fossa by virtue of their inferiorly directed force vector. (Beyond their contribution as dynamic stabilizers the infraspinatus and teres minor also provide a concentric external rotation contraction force.) The subscapularis, aided by the ­pectoralis major, restricts the terminal external rotation of this first phase through eccentric control and also stabilizes the humeral head within the glenoid fossa but through a tethering effect.

Acceleration Phase The acceleration phase follows with a sudden and complete reversal of motion. The internal rotators, subscapularis, and sternal head of the pectoralis major all are active to provide the propulsive force. Opposing this are the posterior cuff muscles. It may be that no synergistic relaxation of the posterior cuff muscles actually occurs but rather eccentric contraction of these muscles contributes to a stabilizing balance in the joint. The latissimus dorsi, serratus

Concentric versus Eccentric Muscular Contraction Unlike joints with significant bony stability, the shoulder relies heavily on dynamic muscle balance for stability. With concentric muscular contraction producing motion

994

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

in one direction, there is a concomitant eccentric muscular contraction on the opposite side of the joint producing stability. These eccentric contractions are particularly evident with the muscles of the rotator cuff and point to their essential role in shoulder stability. For example, with external rotation of the arm, the infraspinatus contracts concentrically; at the same time, the subscapularis shows significant electromyographic activity as it contracts eccentrically.61,71-77 In this case, the infraspinatus is producing the propulsive power on one side of the joint while the subscapularis is producing a counteracting, stabilizing force on the opposite side of the joint. This balance is important biomechanically in fine-tuning the movements in the athlete’s shoulder. It has been suggested that an imbalance in these opposing muscle forces can predispose the overhead athlete to injury.

Scapular Lag Biomechanically, the scapula plays an intimate role in shoulder function. It is important to remember that shoulder function is composed of not only glenohumeral motion but also scapulothoracic motion. Many pathologic situations such as impingement and various instabilities result in subtle winging through dysfunction of the scapula as it moves on the chest wall. Fatigue of the scapular rotators on the chest wall may lead to inability of the scapula to rotate properly and may prevent the acromion from clearing out of the way when the arm is elevated. This situation, termed scapular lag, may result in secondary impingement.11,12,78

Eccentric and Intrinsic Failure In many shoulder problems, early musculotendinous fiber failure is likely, with resultant secondary changes.44,79-83 With time and superior migration of the humeral head, degenerative changes develop, causing wearing and thinning with eventual tearing of the rotator cuff. Many years are usually required to develop a rotator cuff tear; however, in the younger throwing athlete, partial tears may result because of the severe stresses placed on the cuff structures. The incidence of partial articular surface tears of the rotator cuff in the athletic population is unknown but may approach 30% to 40%.39 In addition, other mechanisms may come into play, such as instability with secondary impingement causing further stresses on the already weakened musculotendinous cuff. Classic full-thickness tears are not usually seen until the athlete is older, usually older than 50 years. These tears are related to chronic degeneration and attrition occurring over an extended period. This process leads to changes in the biomechanical function of the shoulder with abnormal instant centers of rotation and resultant compensation by the biceps, deltoid, and other shoulder girdle muscles.

CLINICAL EVALUATION The importance of the history and physical examination in evaluating the athlete with shoulder complaints cannot be overemphasized. This is because shoulder complaints (especially in athletes) can be more nuanced and the underlying pathology more complex, necessitating a certain level

of sophistication of the examiner. History can be vague, symptoms poorly defined, and physical findings subtle. Also, multiple lesions can often coexist in the shoulder of the athlete, with only one of them causing the patient’s symptoms. It is therefore not enough to simply identify a problem in the shoulder. It is necessary to identify the problem that is at the root of the patient’s symptoms. To assist in the diagnosis in light of these difficulties, an alternative perspective to the traditional differential diagnosis is recommended. A differential diagnosis–directed approach to both the history and physical examination is driven by the chief complaint, and a differential diagnosis is formulated at the beginning of the interview rather than at the end.84 The essential components of the history and physical examination (e.g., chief complaint, history of present illness, past medical history, review of systems) remain the same. However, the differential diagnosis–directed approach allows the examiner to test the initial diagnosis throughout the history and physical examination and allows for a more focused approach to the shoulder as it pertains to the athlete. Traditionally, physicians are taught to form their differential at the end rather than at the beginning of the encounter. As such, the differential diagnosis–directed approach may seem to run counter to established teaching. In actuality, such an approach may more closely approximate what a practicing clinician does on a day-to-day basis anyway. To take an extreme example, a 50-year-old who presents complaining of “left arm pain” immediately triggers concerns that the pain is of cardiac origin. Subsequent questioning and tests may prove or disprove this, but the subsequent history and physical examination is certainly geared initially toward proving or disproving this first impression. Conversely, a 14-year-old left-handed pitcher with the same complaint of left arm pain elicits a very different set of concerns. It is important to emphasize that with the differential ­diagnosis–directed approach, although a tentative diagnosis is made early, this diagnosis is tested by the subsequent history and physical examination and is thereby confirmed or refuted.

The Chief Complaint An athlete with a rotator cuff or biceps tendon problem in the shoulder may present with any of a variety of chief complaints. The most common complaint is that of pain. Other complaints include fatigue, functional catching in the shoulder, stiffness, weakness, and symptoms of instability. On occasion, an athlete may present simply with deterioration in athletic performance, for example, a pitcher who loses velocity on his fastball. Knowing the chief complaint as well as the age of the patient provides a high index of suspicion for the most probable diagnoses and establishes the framework for the differential diagnosis–directed approach that will guide the rest of the encounter.19,85-87 Each subsequent step in the examination will serve to help confirm or refute the diagnosis, providing confirmatory evidence when the initial diagnosis is correct and a series of red flags when the initial diagnosis is incorrect. With the differential diagnosis–directed approach, the chief complaint must be matched with various constitutional factors such as age and sex. The older athlete whose symptoms began with an

Shoulder

overhead smash in tennis or experienced a fall on an outstretched arm could have sustained a rotator cuff tear. This same history in someone younger than 40 years would be unlikely to result in a rotator cuff tear. Female athletes are more likely to demonstrate the generalized hyperlaxity that has been implicated in chronic painful shoulder ­conditions. A complete understanding of the chief complaint is critical because multiple pathologies may coexist in the shoulder of the athlete, only one of which may be symptomatic. Miniaci and coworkers imaged professional baseball pitchers who were completely asymptomatic and found labral abnormalities on MRI in 79% of the shoulders imaged.88 Concurrent pathologies present the so-called athlete’s dilemma, which makes evaluating the athlete’s shoulder so challenging. An operative “shotgun” approach without regard for whether or not the lesions addressed are the cause of the athlete’s symptoms can end the athlete’s career because the more surgery undertaken, the less likely the athlete is to return to high levels of participation. It is therefore critical to determine which pathologies are symptomatic and might be improved with surgery and which should be left alone or treated with rehabilitation. This determination must begin with an accurate understanding of the athlete’s chief complaint.

History of Present Illness and Injury The purpose of the history of present illness is to reconstruct the story of the chief complaint (from onset to present) so that the examiner has a clear understanding of how the symptoms began, what has been attempted, and the current state of the problem. The athlete may be unclear as to how or why the symptoms started and may describe an insidious onset. When a single traumatic event is responsible for the injury, information about the mechanism, degree of initial symptoms, and events surrounding the event will provide valuable information. Once the circumstances surrounding the onset are established, the clinical course of the complaint is determined from its inception to the present. The effects and timing of various treatments are carefully considered. Any response to treatment, even if temporary, is important. For example, if a lidocaine and steroid injection was administered to the subacromial space for shoulder pain, it is important to note whether this was effective, even if only temporarily, because this will yield diagnostic as well as therapeutic information. The effectiveness of other interventions such as anti-inflammatories, modalities, and physical therapy should also be noted. This information should give the examiner an understanding of what has already been done and the progression of the treatment already instituted. It is then important to note the current status of the complaint given the prior treatments. This current status should seen in the context of the athlete’s current level of activity, where he or she is in relation to the season and how long he or she has until the shoulder has to be in “playing condition.” A college football quarterback who dislocates his shoulder for the first time early in his senior year might pursue a different treatment course than the same player who dislocates in the first week of the off-season after his junior year. Such an understanding requires thorough

995

communication with the athlete and an understanding of his or her goals and will guide the patient and the physician to the best treatment option for their desired outcome. Likewise the degree of disability incurred by the athlete from the injury is important. This should be considered in the broader context of the sport and the intensity of the athlete’s participation in that sport (recreational versus professional). Athletes and patients in general present with complaints that range on a spectrum from minimal annoyance with certain high-level sports-related activities to complete disability with activities of daily living. Understanding where the patient is on this spectrum will greatly aid in guiding how aggressive the diagnostic workup and treatment plan should be. Accurate assessment of the degree of disability may require communication with the athletic trainer and physical therapist because some athletes may attempt to “play through” injuries that render them ineffective and put themselves in danger of further injury. An accurate assessment may be difficult for an athlete to make objectively, and often a trainer’s input is invaluable.

Past Medical History and Review of Systems Although a solid differential diagnosis should be in place at this point, and although athletes are among the healthiest patients in our population, questions about past medical history should not be neglected. These include questions about medications, allergies, and congenital or other medical problems. Finding out that a swimmer with shoulder pain has Ehlers-Danlos syndrome might not only point to multidirectional instability (MDI) as a diagnosis but also might influence treatment. Although often negative, a review of systems and questions regarding past medical history can avoid missing key aspects affecting the diagnosis and eventual treatment of the athlete.

Physical Examination Because shoulder pathology in the athlete can be quite complex, the specific physical examination tests that may elicit these symptoms can be quite specific and, even when positive, quite subtle. As such, it is not useful to apply the shotgun approach to diagnosis, whereby every test described for the shoulder is performed on every shoulder. The differential diagnosis–directed approach, as in the history taking part of the examination, helps to direct the physical examination to those tests that will either confirm or refute our tentative diagnosis. The physical examination should nonetheless be organized and thorough. At the very beginning of the physical examination, an initial impression is taken accounting for the ­athlete’s age, overall health, and level of specific ­ distress related to the shoulder problem. Inspection, ­ palpation, range of motion, strength testing, and neurologic and vascular stability assessment constitute an orderly sequence.12-14,17,19,71,72,78,89,91,110-115 Inspection considers symmetry (one must appreciate that pitchers or tennis players may have unilateral drooping of their dominant shoulder) or deformities such as old acromioclavicular injuries and muscle wasting, which is

996

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17I1-13   Clinical photograph of a patient with prominent infraspinatus wasting in the right shoulder.

most often located in the infraspinatus fossa with a rotator cuff tear (Fig. 17I1-13). A proximally ruptured biceps tendon shows the characteristic bulging distally with muscle contraction (Fig. 17I1-14). The location and degree of tenderness found on palpation often provide a reliable physical sign leading to an accurate diagnosis. Tenderness in the bicipital groove (2 to 5 cm distal to the anterior acromion and midway between the axilla and the lateral deltoid with the arm in the anatomic position) is a reliable sign of bicipital tendonitis (Fig. 17I1-15). Tenderness in this region with palpation and passive external rotation of the arm (“rolling” the bicipital groove under the examiner’s fingers) is another reliable sign of bicipital pathology.89 The supraspinatus insertion (Codman’s point) is palpated through the deltoid just distal to the anterolateral border of the acromion with the shoulder extended and internally rotated (Fig. 17I1-16).2 Maximal tenderness over the acromioclavicular joint (Fig. 17I1-17) may also indicate specific pathology. Range of

Figure 17I1-14   Clinical photograph of a patient with a ruptured long head of the biceps tendon in the right arm with ecchymosis and a “Popeye” deformity.

motion should be documented. A true discrepancy between active and passive ranges of motion is suggestive of a rotator cuff tear. However, in the athlete, superior strength and flexibility can mask these findings. Many athletes such as swimmers and gymnasts may have developed what appears to be a supranormal range of motion when compared with the general population but for them is normal. In contrast, the overhead throwing athlete may have sideto-side range of motion differences with increased external rotation and decreased internal rotation in the dominant, throwing arm when compared with the contralateral nondominant side. However, the total arc of motion is typically the same on both sides, albeit shifted.19 It is believed that this decreased internal rotation is an adaptive mechanism in athletes engaging in overhead throwing, possibly owing to hyperangulation of the humerus with relative stretching of the anterior capsular structures and relative tightening of the posterior capsular structures. This may be a contributing factor in rotator cuff overuse syndromes through excessive obligatory anterior translation and secondary impingement. When evaluating shoulder range of motion, ­stressing the shoulder at the extremes of motion can also provide clues to pathology in the shoulder. Active and passive ranges of motion should be documented in all planes. This includes elevation in the scapular plane and external rotation with the arm at the side, which can be recorded in degrees. It is also important, especially in athletes, to document external rotation (particularly passively) in the 90-degree abducted position in the coronal plane. This position represents a more functional measure of external rotation.89 Internal rotation can be recorded as the most cephalad vertebral level obtainable by the hitchhiking thumb or index finger) (Fig. 17I1-18). Strength is considered along with range of motion. Although it is part of the neurologic examination, assessment of strength is particularly important in athletes with rotator cuff pathology. Objective weakness beyond that which can be attributed to pain or a neurologic deficit is a very specific sign of rotator cuff deficiency. The remainder of the neurologic examination will help to rule out pathology such as a cervical root, brachial plexus, or peripheral nerve lesion. The assessment of shoulder stability is very important because rotator cuff signs and symptoms are often a secondary manifestation of an underlying problem of stability. It has been suggested in high-profile throwing athletes that shoulder pain is due to instability related to anterior subluxation with secondary impingement until proved otherwise. Stability is assessed by translating the humeral head in the glenoid fossa anteriorly, posteriorly, and inferiorly (the load-and-shift and sulcus tests, respectively), with the arm in varying degrees of abduction and rotation (Fig. 17I1-19).89 The presence of an anterior apprehension sign is also important and can represent either anterior instability or pain from internal impingement. This is performed by passively placing the arm in increasing degrees of abduction and external rotation (Fig. 17I1-20).39,90,91 The relocation test (Fowler’s sign or Jobe’s relocation test) is a variation of the apprehension sign (Fig. 17I1-21). The arm is placed in the abducted and externally rotated position until pain or apprehension is elicited. The same maneuver is then repeated, but with a posteriorly directed force

Shoulder

A

997

B

Figure 17I1-15   A, Patient with underlying acromion, acromioclavicular joint, coracoid, and coracoacromial ligaments drawn on the skin. B, Examination of the biceps tendon by palpation.

on the arm. Relief of pain or apprehension with improved external rotation is indicative of internal impingement or anterior subluxation, respectively. Examination of the regional vascular supply is necessary as a baseline and also to evaluate for conditions such as thoracic outlet syndrome. Finally, there are a number of special tests that should be considered. The signs of impingement are characteristic of rotator cuff tendinitis and tears. These include a painful arc of abduction between 60 and 120 degrees, pain on forced forward flexion in which the greater tuberosity is forced against the anterior acromion (Neer’s sign), and pain on forcible internal rotation of the 90-degree forward flexed arm (Hawkins’ sign or the impingement reinforcement test) (Figs. 17I1-22 and 17I1-23).6,92 The latter maneuver causes impingement against the coracoacromial ligament. Biceps tendon involvement is demonstrated by Speed’s test, in which pain is reproduced on resisted forward elevation of the humerus against an extended elbow (Fig. 17I1-24). Yergason’s test is performed with the elbow flexed to

90 degrees and the forearm pronated.93 The examiner grasps the wrist and resists active supination by the patient. Pain in the area of the bicipital groove is suggestive of pathology in the long head of the biceps (Fig. 17I1-25). The active compression test (O’Brien’s test, with resisted elevation and the arm at 90 degrees of forward flexion and 10 to 15 degrees of adduction) may also be positive with pathology of the long head of the biceps without a superior labrum, anterior to posterior (SLAP) lesion.94 Biceps tendon instability (medial subluxation or dislocation) can be determined by passively abducting the shoulder to 80 to 90 degrees and eliciting a palpable snap in the region of the bicipital groove with internal and external rotation.41,95-98 This is a rare presentation as an isolated entity and usually indicates a lesion to the superior fibers of the subscapularis tendon or the SGHL. The history and physical examination will lead to an appropriate diagnosis. As previously stated, in athletes, the main differential diagnosis is between instability and primary rotator cuff pathology. Shoulder pain in the athlete

Figure 17I1-16   Palpation of Codman’s point.

Figure 17I1-17   Palpation of the acromioclavicular joint.

998

DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

Figure 17I1-18   A-D, Active range of motion is tested for absolute range as well as symmetry.

can create a vicious cycle, with an overlap between shoulder instability and/or laxity and rotator cuff and biceps tendinitis and/or impingement. Whether or not instability causes tendinitis or tendinitis causes instability remains unclear.

Diagnostic Studies Diagnostic studies used in the evaluation of the athlete with a rotator cuff or proximal biceps tendon problem serve to support the clinical findings and help confirm the diagnosis. These studies are also important in ruling out other pathologic entities.

demonstrates impingement to be at least a component of the patient’s underlying problem. Nevertheless, it should be emphasized that this is a nonspecific test and can be misleading because it may be positive in both patients with primary impingement as well as patients with secondary impingement due to instability.99,100 Although not strictly an impingement test, injection of local anesthetic into the acromioclavicular joint or into the bicipital groove can supply additional information about the source of the pain. Subacromial anesthetic can mask or minimize the symptoms from these two areas. It is the clinical examination that is critical in guiding the selections and order of the injection sites.

The Impingement Test This test, as described by Neer, involves injection of local anesthetic into the subacromial region after a positive Neer’s sign.5-7 The injection is performed under sterile conditions with insertion of the needle from anterior, lateral, or posterior into the subacromial space. Subsequent to the injection, impingement signs should be sought as previously described. Subjective relief or significant ­diminution of the previously present pain with impingement testing

Radiographic Studies Plain Radiographs Standard plain radiographs should include an anteroposterior film at right angles to the scapular plane, a lateral film in the scapular plane with the beam tilted 10 degrees to evaluate acromial shape and slope, and an axillary view. Plain radiographs of the typical athlete with a rotator cuff

Shoulder

A

C

Figure 17I1-20   Apprehension test performed with the patient supine. This can elicit apprehension and pain.

999

B

Figure 17I1-19   A, Shoulder stability can be assessed with the load-and-shift test performed with the patient sitting. B, The load-and-shift test performed with the patient supine. C, Sulcus test.

Figure 17I1-21   Relocation test with a posteriorly directed force. This can relieve symptoms from the apprehension test.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1000 DeLee & Drez’s� O

Figure 17I1-22   Neer’s sign suggests classic impingement when this maneuver elicits pain.

Figure 17I1-24   Speed’s test.

complaint are most often normal.101 The characteristic changes of advanced rotator cuff disease include sclerosis and cystic changes in the greater tuberosity and osteophyte formation on the acromion. This has been described as acetabularization of the acromion and femoralization of the proximal humerus. There can be a more pronounced notch between the greater tuberosity and the articular surface, changes in the shape of the acromion and, in the presence of a massive rotator cuff tear, a narrowed acromiohumeral distance of less than 6 mm can be seen. There may be osteophyte formation on the inferior surface of the acromioclavicular joint as part of chronic rotator cuff disease. Although plain radiographs are often normal, they are nonetheless invaluable because they help to rule out other conditions that may present with shoulder pain such as glenohumeral arthritis, calcific tendinitis, or even neoplasm, conditions that would not typically be considered first in the athlete. Plain radiographs can also be used to evaluate the bicipital groove. A shallow groove may indicate a rare biceps

tendon instability problem. Osteophytes around the groove may be implicated in pathologic degenerative conditions of the biceps.

Figure 17I1-23   Hawkins’ sign.

Figure 17I1-25   Yergason’s test.

Arthrography Single- or double-contrast arthrography was once considered the gold standard for determining the presence of a full-thickness rotator cuff tear before the advent of MRI.102,103 Arthrography is now most commonly used in combination with other imaging modalities (such as computed tomography or MRI) to make subtle lesions of the cuff and glenohumeral joint more conspicuous.

Ultrasonography Diagnostic ultrasound is a noninvasive form of examination of the rotator cuff.104-106 It allows comparison with the other side and can provide a great amount of anatomic detail. It has reported 91% sensitivity and specificity rates,

Shoulder 1001

with a 100% positive predictive value when it shows nonvisualization or focal thinning. It has also been reported to be very useful in diagnosing bicipital pathology and is helpful in patients who have previously undergone a rotator cuff repair. However, the results are related to the operator’s experience, and the technique has inherent limitations because of the surrounding bony anatomy.

In most athletes, treatment of a rotator cuff or biceps tendon problem is nonoperative. Three important items to be considered are (1) the etiology of the condition, (2) the sport and level of performance, and (3) the severity of the problem.

Magnetic Resonance Imaging

Etiology

MRI has become the gold standard for the investigation of rotator cuff pathology, with sensitivities and specificities exceeding 90% in most current series (Fig. 17I1-26).103,105,107,108 MRI can demonstrate the size, location, and characteristics of the cuff pathology—whether full thickness, partial thickness, or intratendinous. There are drawbacks to MRI. Occasionally, patients are unable to tolerate the examination. This can be because of claustrophobia or an inability to remain still for the time necessary to obtain a useful scan. The presence of metallic implants may interfere with image acquisition and other implants such as pacemakers, or recently placed vascular clips may preclude MRI evaluation. The addition of intra-articular contrast may help augment MRI and may be of particular benefit in the identification of partial cuff tears and labral lesions in the athletic population. A good history and physical examination remain the most important components in establishing the diagnosis in an athlete with shoulder complaints. Because these complaints are often complex, MRI, examination under anesthesia (EUA), and arthroscopy can sometimes be used to help clarify the clinical picture. Such studies, however, are a supplement to, not a substitute for, a well-performed history and physical examination.

Rotator cuff pathology can result from a multitude of factors. Although the end result may be simplified to pain or instability or weakness, the underlying cause can be quite varied. An appreciation for the unique factors that resulted in the development of the pathology in the individual is necessary for focused and effective treatment. Classic, primary outlet or extrinsic impingement was described by Neer in 1972. The mechanism of primary outlet impingement is that the rotator cuff is impinged on by the coracoacromial arch. Neer classified the resulting pathologic changes of this impingement into three stages: stage 1, edema and hemorrhage (at any age); stage 2, fibrosis and tendinitis (usually in patients older than 25 years); and stage 3, degeneration, bony changes, and tendon ruptures (usually in patients older than 40 years).5-7,109 Secondary impingement is possible, with overuse and fatigue of the scapular stabilizers leading to a scapular lag. In this scenario, the anatomy of the coracoacromial arch itself is unchanged. However, because of abnormal mechanics of the scapula, the coracoacromial arch is brought into a position in which it impinges on the rotator cuff. In the absence of impingement, primary or secondary, other sources of pain include eccentric overload of the cuff tendons, which can lead to fatigue and pain. Long-term repetitive overuse, combined with the inherent poor blood supply of the tendon, can also lead to degeneration and tearing, particularly in older individuals.12-14,17,78,89,91,110-115 Instability may also cause rotator cuff pathology. Instability itself comes in two distinct flavors: multidirectional and unidirectional. True multidirectional instability can lead to secondary rotator cuff–related pain and pathology. Anterior subluxation can lead to secondary impingement with pain. In the overhead athlete, anterior subluxation is a common cause of secondary impingement, especially in patients with tight posterior structures and limited internal rotation.8,69 Recent authors have described posterosuperior glenoid impingement or “internal” impingement, particularly in overhead throwing athletes.19,21,66 This is also a secondary impingement but one in which the articular surface of the cuff comes into contact with the posterosuperior glenoid rim (Fig. 17I1-27). Contact between the cuff and the posterosuperior glenoid is physiologic when the arm is maximally abducted and externally rotated. However, repetitive contact in this area (as can be seen in throwers) leads to partial tearing of the supraspinatus and often the infraspinatus tendons on their articular surface. This is typically seen 1 cm posterior to the biceps tendon. Others have illustrated that subtle anterior glenohumeral laxity may exacerbate this internal impingement by allowing the humerus to hyperangulate in the late cocking (­maximal abduction and external rotation) phase of throwing. Furthermore,

Figure 17I1-26   Coronal magnetic resonance image of a supraspinatus tear.

TREATMENT OPTIONS

�rthopaedic ����������� S �ports ������ � Medicine ������� 1002 DeLee & Drez’s� O

A

C

­ osterior capsular tightness and contracture have been p demonstrated to shift the contact point of the humeral head on the glenoid in a posterosuperior position and may thereby exacerbate internal impingement. Acute trauma to the cuff secondary to either a dislocation of the glenohumeral joint or direct or indirect damage can occur.12-14,17,78,89,91,110-115 Although most athletes with these problems can be treated nonoperatively, the surgical management varies dramatically between an athlete with primary impingement and one with primary instability and secondary impingement.

Sport and Performance An appreciation for the patient’s aspirations and their chosen sport can be helpful in understanding the patient’s disability in the context of that sport and possibly aid in minimizing disability. For example, a soccer player with a rotator cuff problem may have only minimal disability.

B

Figure 17I1-27   A, Arthroscopic view of the glenohumeral joint from a posterior portal. The humeral head and cuff insertion is visualized along with the posterosuperior aspect of the glenoid rim. B, The arm is progressively brought into external rotation and abduction bringing the rotator cuff insertion closer to the glenoid rim and labrum. C, Contact is seen between the rotator cuff insertion and the glenoid rim as the arm is brought further into abduction and external rotation. When the shoulder is subjected to repetitive, supraphysiologic stresses (as in throwers), this normal contact can develop into internal impingement.

A baseball pitcher, conversely, may be significantly disabled and may be unable to throw at all. Understanding an athlete’s sport may allow the physician to make suggestions on activity modifications that are acceptable to the athlete. A gymnast may have her symptoms resolve simply by avoiding the rings and concentrating on vaulting. Or a ­swimmer may be able to change from freestyle to breaststroke events. The same consideration should be given to the athlete’s level of performance as well as participation. Enforcement of therapeutic rest has dramatically different implications to a weekend tennis player than to a professional quarterback in midseason.

Severity Finally, the severity of the problem should be considered. This can be considered in the context of the level of the clinical signs, symptoms, and disability as well as in terms

Shoulder 1003

of the underlying pathology. Both have obvious implications with respect to the treatment.

Types of Treatment With the principles discussed above in mind, we can divide the treatment options into three general categories: (1) preventive, (2) nonoperative, and (3) operative.

Preventive Treatment When dealing with athletes, a primary concern that should be paramount to the sports medicine physician, surgeon, trainer, therapist, or coach should be prevention of injury.25,43,83,116-130 Prevention is of particular importance in relation to the shoulder, in which most injuries are related to overuse. Intense training and practice may actually be detrimental to the athlete when poor technique or improper training methods are employed. The underlying principle of prevention is applied common sense. A musculotendinous unit is capable of resisting only as much as it has been prepared to resist. A 50-year-old who plays 2 hours of doubles tennis once per week cannot expect his shoulder to suddenly tolerate the 12 hours required in a 2-day weekend tournament. The same logic can be applied to the college or professional level pitcher who arrives in training camp with little or no off-season training. The basis of prevention is preparation.43 This involves overall body conditioning, flexibility, strengthening, and careful attention to technique while recognizing the stresses that both training and competition present. As an example of the need for preparation to prevent injury, it has been recommended that the little league pitcher develop the fastball as his primary pitch.131 From there, the emphasis should be on improving velocity while maintaining consistent mechanics and control before developing the full repertoire of pitches. The off-season is the key to an athlete’s development, with a general fitness and weight-­training and flexibility program being essential. It is during this time that the specific adaptation to imposed demand (SAID) principle is applied to training. For a pitcher, this involves alternating between long-toss and short-toss throwing at half speed with enforced rest at least 2 days per week. During the season, the same principle applies with a proportionate increase in the frequency and duration of training. The warm-up is also of critical importance.43 A satisfactory warm-up leads to increases in tissue temperature with improved oxygen uptake, nerve impulse transmission, and increased activity of metabolic enzymes. The influence of shoulder flexibility has been demonstrated in swimmers.132-134 Published reports have demonstrated a clear correlation between anterior shoulder inflexibility and shoulder pain. It follows that stretching is an important preventive measure. The particular goal in stretching is to try to maintain internal rotation and adduction (Fig. 17I1-28). It is not necessarily the intention to normalize internal rotation compared with the opposite side. The intention is to avoid posterior capsular contractures and maladaptive loss of internal rotation and adduction. Stretching is equally important in older athletes in whom the potential for stiffness is greater.

Figure 17I1-28   “Sleeper” stretch for stretching of the posterior capsule.

A strong, well-balanced shoulder is critically important in the prevention of overuse injuries.12,17,43,78,89,110-112 This balance begins with the scapular platform and scapular stabilizing muscles. Both the scapular stabilizers and the rotator cuff can be strengthened with the use of many aids, such as free weights, Thera-Band, or surgical tubing. Isokinetic machines are also recommended. It is evident that any strengthening program must be well controlled, particularly in the young athlete. In addition, the use of some isokinetic equipment does not allow eccentric muscle contraction. This may be of considerable importance given our understanding of the pathophysiology of rotator cuff lesions, which is that of eccentric overload. A preventive strengthening program for shoulder problems in athletes is critical, and the emphasis should be on eccentric exercises for the cuff and scapular stabilizers.12-14, 17,78,89,91,110-115 This can be easily performed by the athlete on a daily basis with some form of resisted Thera-Band or rubber tubing. The exercises consist of the following steps: 1. Resisted external rotation exercises with the arm at the side and also in the 90-degree abducted position (Fig. 17I1-29A to D) 2. Resisted internal rotation exercises (see Fig. 17I1-29E and F) 3. Push-ups with the arm adducted for scapular rotator control (see Fig. 17I1-29G and H) 4. Sitting rows for serratus and rhomboid scapular control (see Fig. 17I1-29I and J) 5. Shrugs for trapezius strengthening (see Fig. 17I1-29K and L) 6. Latissimus pull-downs for latissimus control, important in the deceleration phase of overhead motion and change (see Fig. 17I1-29M and N) 7. Supraspinatus strengthening exercises: resisted abduction in the scapular plane with internal rotation (see Fig. 17I1-29O and P) Finally, it is incumbent on the coaches to teach the correct technique, avoid overtraining, allow for rest periods and

�rthopaedic ����������� S �ports ������ � Medicine ������� 1004 DeLee & Drez’s� O

A

B

C

D

E

F

Figure 17I1-29   A-P, Eccentric, external rotation strengthening exercise.

Shoulder 1005

G

H

I

J

K

L

Figure 17I1-29, cont’d    Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1006 DeLee & Drez’s� O

M

N

O

P

Figure 17I1-29, cont’d   

(ideally) recognize the fatigued athlete who is heading for a painful shoulder problem.

Nonoperative Treatment Nonoperative treatment is in essence an extension of preventive management with the addition of specific measures dealing with the injury.12,17 It can be divided into four components: (1) modification of activity, (2) local or systemic measures to reduce and relieve the symptoms, (3) stretching and strengthening exercises, and (4) re-evaluation and maintenance treatment.

Modification of Activity In the athlete with mild symptoms, modification of activity means reducing the frequency and duration of the specific activity. It also involves activity substitution or what is sometimes called active rest. In a tennis player, avoiding the service action but still hitting ground strokes may be all that is required. In a baseball pitcher, it would be helpful

to cut back on the daily number of pitches and to decrease velocity. A swimmer should decrease yardage or use the kickboard. Other methods of treatment involve specific changes in technique such as throwing side-arm for a pitcher or using a higher arm entry for a freestyle swimmer. Changing equipment may also be of benefit. With time, these patients usually improve, but return to sport is a longer term goal, and more involved treatment may be necessary.

Local and Systemic Methods to Relieve Symptoms Although the athlete experiencing pain with activity that is not disabling may be able to manage with various methods of modified activity, most patients have a more involved problem and require more complicated therapies. The use of nonsteroidal anti-inflammatory drugs (NSAIDs) is ubiquitous. There is no question that they provide symptomatic relief. However, the essential problem (whether acute or due to overuse) is tissue injury, and the ­ inflammatory response to this injury is a normal part of the healing

Shoulder 1007

mechanism. There is also evidence that in chronic overuse injuries, the surgical pathologic tissue is noninflammatory. The value of these medications lies in the initial treatment given to decrease pain and allow rehabilitation. Ice ­treatment is also a recognized local treatment modality with its action of reducing vascularity, numbing the pain, and possibly reducing swelling and inflammation. Ice is particularly useful after an acute episode or injury. It is commonly advocated after overuse activity to minimize the pain and lessen the immediate postactivity inflammatory response. Therapeutic ultrasound may be helpful in the treatment of rotator cuff and biceps tendinopathies. It is believed to increase the local vascular response to the injured tissue. This in turn allows the release of the products of injury and the influx of the raw materials needed for repair. Other modalities include high-voltage electrical stimulation, transcutaneous nerve stimulation, electromagnetic field therapy, and the use of lasers. The specific benefits of some of these modalities are not clear, yet many patients appear to gain relief with few side effects. The use of local corticosteroid injections is a more invasive form of therapy. The deleterious effects of steroid injections have been documented. Critical analysis of the literature cannot lead to the conclusion that they are of any long-term benefit. Their use is still advocated and may be of value in a patient with an acutely painful lesion to halt the vicious circle of pain related to overuse. This would then allow earlier institution of corrective rehabilitation.

Stretching Stretching exercises are not only therapeutic but also quite clearly preventive and provide the basis for maintenance treatment.17,43 The warm-up, including stretching, helps the athlete improve timing and control and allows the muscles to function efficiently. Stretching increases muscle blood supply and improves contractility. Lack of flexibility has been associated with a higher incidence of shoulder problems in swimmers. An increased range of external rotation compared with internal rotation is also associated, especially in throwers. These factors in the athlete with a painful shoulder require specific attention to the treatment regimen. Stretching should be generalized but should focus on internal rotation and adduction across the chest and internal rotation and extension behind the back. These exercises should be performed before activity in addition to the usual routine.

Strengthening Strengthening is the mainstay of treatment for most athletes with rotator cuff or biceps tendon problems.810,17,74,135 Whether the problem is due to an acute direct injury or eccentric overload, the athlete is left with a compromised, weakened musculotendinous unit. This unit is usually contracted either primarily as a result of muscular imbalance or secondarily due to the injury mechanism. A complete tear or disruption of the tendons will obviously need to heal or be approximated through surgery before active use, but this is an uncommon circumstance in the young athletic population.

Strengthening of the cuff should emphasize the external rotators.17 The use of rubber tubing is simple and effective for both prophylaxis and treatment. Initially the exercises need to be performed with the arm at the side until the pain has been relieved. When pain is present, it is important to avoid the 90-degree abducted position with resisted rotational strengthening. The arm can be brought into the abducted position gradually, initially to 45 degrees, until the pain has completely disappeared and then to the functional range. Most recreational and nonthrowing athletes do not require strengthening at 90 degrees of abduction. As described previously, the associated musculature, particularly the scapular stabilizers, must be considered immediately. Kibler and Burkhart have described the effect of scapular dyskinesis on shoulder dysfunction.136-139 Shoulder motion involves not only the glenohumeral articulation but also the scapulothoracic articulation and depends on the coordinated interplay between the two. Scapular dyskinesis may arise primarily, as from a nerve palsy, or alternatively may arise secondarily, in which case it may exacerbate the symptoms and increase dysfunction. For example, a rotator cuff tear may be the inciting injury, but subsequent pain from this may lead to pain inhibition of the scapular stabilizers of the shoulder girdle. The resulting scapular dyskinesia may in turn exacerbate the pain originating from the cuff pathology by perhaps subtly altering the position of the scapula and leading secondarily to cuff impingement. A useful maneuver in evaluating scapular dyskinesis is for the examiner to mechanically stabilize the scapula and then have the patient elevate his arm to see if symptoms are improved or relieved. If scapular dyskinesia is a primary cause of shoulder dysfunction, nonoperative physical therapy aimed at retraining and restrengthening the scapular stabilizers can be curative. When scapular dyskinesia is secondary to other forms of shoulder pathology physical therapy may be necessary even after the primary lesion is addressed to return the shoulder to normal. In addition to the specific components of shoulder strengthening, it is important to understand that the shoulder cannot be viewed in isolation. Kibler has also described the importance of “core stability” to athletic performance.140 Equal concern should be paid to the associated joints and muscles to ensure that appropriate body mechanics are used in the rehabilitation of the athlete. Closed chain scapula stabilizer strengthening exercises include the following: 1. Ball rolling (Fig. 17I1-30A and B) 2. Protraction and retraction (see Fig. 17I1-30C and D)

Re-evaluation and Maintenance It is during re-evaluation that the treatment phase blends into that of prevention of further injury. Throughout the process of treatment, the athlete, coach, trainer, and (if applicable) parent should be part of an educational program. This involves education about the clinical problem, its course, and its ultimate prognosis. Most athletes are willing to perform a daily routine of exercises if it means participating in the sports they love. Unfortunately, compliance with this routine is more likely following rather than preceding an injury.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1008 DeLee & Drez’s� O

A

B

C

D

Figure 17I1-30   A-D, Closed-chain scapula stabilizer strengthening exercises.

Operative Treatment It cannot be restated enough that the surgical management of athletes with rotator cuff or biceps tendon pathology should be emphasized less than nonoperative treatment. Most athletes recover, modify their activities, or even give up their sport before undergoing surgery. The maturing athlete may have not only acute and overuse injuries but also the added concern of degenerative tendon pathology, which involves a far greater incidence of rotator cuff and biceps tears. These athletes have the same symptoms as any individual of a similar age with rotator cuff pathology, but usually they have greater demands and expectations. Greater understanding of the pathophysiology of shoulder pain has led to the realization that sometimes stabilization procedures rather than surgery are required to deal with primary cuff or biceps disorders. Surgical management involves open or arthroscopic procedures.8,15,37,51,59,78,107,119,141-155 Historically, the procedures designed to correct these problems involved various forms of cuff repair and biceps tenodesis and used a variety of approaches, including acromionectomy. Even

a paraglenoid osteotomy was devised to deal with subacromial bursitis and supraspinatus tendinitis. Today, three main procedures for subacromial-based pathology are used in the athletic population: subacromial decompression (and its variants such as débridement), rotator cuff repair, and biceps tenodesis.

Open Techniques Neer was the first to describe anterior acromioplasty based on the fact that extrinsic impingement occurred under the anterior third of the acromion.156 His initial patient population was nonathletic and included a heterogeneous group of patients with and without cuff tears, with satisfactory results in 80%. Neer’s method of anterior acromioplasty has been modified by others: some avoid detachment of the deltoid, whereas others include distal clavicle excision. Still others release or resect the coracoacromial ligament without bony decompression.17 The other common open procedure is that of rotator cuff repair. Usually this involves an anterior acromioplasty (although this remains controversial) with both direct

Shoulder 1009

side-to-side (McLaughlin or margin convergence technique) and tendon-to-bone repair.17,78,157 Many other techniques have been advocated in dealing with an athletic population, including arthroscopic procedures. However, restoration of normal anatomy is the obvious goal. Many factors relating to sport, level of activity, and arm dominance influence results.123 The demands are different for athletes versus nonathletes and for throwers versus nonthrowers. Surgery for cuff tendonitis produces different results in a soccer player than in a pitcher. The use of coracoacromial ligament resection has been successful in 95% of a series of patients who had failed to benefit from nonoperative management for impingement syndrome secondary to coracoacromial ligament entrapment.158 The use of this procedure, which has a 70% success rate in athletes, has been advocated as an alternative to the more surgically invasive anterior acromioplasty.77 Comparison of these two procedures in a group of athletes showed similar results.18 Of the patients with coracoacromial ligament resections, 13 of 17 returned to full activity with no further symptoms at an average follow-up of 3.5 years. These series did not clearly define the level of activity or throwing capability postoperatively. In a report of open anterior acromioplasty in athletes, 89% reported improvement subjectively, but only 43% showed good functional results. Of the athletes classified as primarily pitchers and throwers, the results are even less satisfactory, with only 22% achieving a good functional result.107 It is very difficult to compare these results. Patient selection is critical to the outcome in this type of operation, and there are no adequate controlled series. For athletes with rotator cuff tears, the decision to operate is easier in those with a complete cuff lesion but is not so clear with partial-thickness tears. In a series of 45 patients who underwent rotator cuff repair for both full-thickness and partial-thickness tears, 39 (87%) were satisfied and experienced subjective pain relief.107 Analysis based on sports participation showed that among pitching and throwing athletes performing at the college or professional level, only 32% had a good result. The results of surgery in similar athletes with complete cuff tears were slightly better, with 5 of 9 experiencing good results. Most authors recommend early surgical repair for fullthickness cuff tears that are symptomatic.3,17,65,80,159,160 Yamaguchi and associates’ recent work on the natural history of rotator cuff tears would likewise support this recommendation.23 In an athlete with higher demands, a prediction of full return to premorbid levels of play is not possible based on a current review of the literature. This caution applies both to decompression of the cuff regardless of the method used and to rotator cuff repair. A symptomatic full-thickness tear without surgery yields predictably poor results, and a large tear (>3 cm) has a poor functional recovery even with surgery.37,120,146,160,161 Surgery related to biceps tendon pathology is even less clearly delineated in the literature.6,162-166 It is usually assumed that the biceps is involved along with the ­rotator cuff based on similar pathophysiology, mechanisms of injury, and repair. It therefore follows that the treatment involves decompression of the coracoacromial arch. Again, it is difficult to ascertain the results of surgical ­management,

but success similar to that achieved with rotator cuff repair has been reported. With respect to subluxation or dislocation of the biceps tendon from the bicipital groove, the treatment is relatively straightforward, and tenodesis is the recommended approach in the athletic populace.41,167 One series reported that 77% of athletes resumed their sport and could throw satisfactorily after tenodesis.167 In another report, excellent results were cited. However, no specifics were documented about either the patient population or the actual results. It should be remembered that primary biceps instability in athletes is an extremely rare entity.

Arthroscopic Techniques Arthroscopic evaluation is now the standard of care for athletes with rotator cuff and impingement lesions. Diagnostic arthroscopy allows confirmation of the diagnosis. The intra-articular structures are visualized, and evidence of articular damage or instability is documented. SLAP lesions with undersurface cuff degeneration may be a manifestation of overuse associated with instability. The undersurface of the cuff is carefully examined. Partial-thickness tears can be examined and evaluated.145,151 Snyder and Ellman have both proposed classifications for the arthroscopic evaluation of rotator cuff tears. A spinal needle, marker suture, or both, can be placed through the articular surface partial tear for later bursal surface identification and evaluation. Most authors recommend débridement and subacromial decompression if the partial tear involves less than 50% of the cuff.143,147,168,169 If the tear involves greater than 50% of the cuff, repair and decompression are indicated. Recent work has demonstrated that partial tears may significantly predispose to the development of a full-thickness tear.39 Hence, some have advocated repair for all partial-thickness tears, although this clearly remains controversial, especially in the athletic population. Posterosuperior glenoid impingement, or internal impingement, can only be evaluated arthroscopically.19,21,66 Results have been mixed with débridement alone for this lesion. Some have reported improved functional results in the athletic population with the addition of anterior thermal capsulorrhaphy in this population of athletes, with a presumed diagnosis of subtle anterior instability allowing excessive hyperangulation of the humerus and causing the internal impingement. However, these results have not been reported in the peer-reviewed literature, and such treatment remains controversial until further studies clarify the issue. The results of subacromial decompression in 24 patients active in sports who had a diagnosis of primary impingement revealed that 87.5% returned to active participation.170 The overall success rate in another heterogeneous series (including both sport- and non–sport-related etiologies) was 88%.144 The complications of arthroscopic procedures are relatively few in all reported series.171 The arthroscope has also allowed the identification and treatment of lesions in the intra-articular portion of the biceps tendon.172 Biceps tenosynovitis is often located in the intertubercular groove portion and can be visualized and débrided arthroscopically. Partial tears of the biceps can

�rthopaedic ����������� S �ports ������ � Medicine ������� 1010 DeLee & Drez’s� O

be débrided, and if more than 25% to 50% of the tendon is disrupted, a tenodesis can be performed. Arthroscopic biceps tenodesis has been described in a mixed population of patients with good results.173 Superior labral lesions, which are discussed elsewhere, can be appropriately treated arthroscopically.

Authors’ Preferred Method

of

Evaluation of the anteroinferior labrum can identify labral pathology responsible for anterior subluxation and secondary cuff pathology and impingement. Anterior stabilization procedures can be performed arthroscopically or open to prevent the secondary impingement occurring in this situation.

Treatment

First and most important in the management of any athlete with a rotator cuff, impingement, or biceps problem is establishing the correct diagnosis. We place strong emphasis on a careful history, thorough physical examination, plain radiographs, and the judicious use of diagnostic injections. Further investigation (usually in the form of an MRI) is reserved for athletes with an atypical presentation, those who are older, those with a significant traumatic episode, and those in whom a lesion requiring surgery is suspected. The diagnosis and treatment of the rotator cuff or biceps injury is based on the following causes: 1. Acute trauma 2. Primary impingement 3. Instability and secondary impingement 4. Internal impingement 5. Overuse: (a) scapular lag with secondary impingement, or (b) eccentric overload 6. Combined cause The type of management used follows from this etiologic classification. It should be emphasized that the focus of treatment is nonoperative in most individuals. Acute Trauma

Patients with acute traumatic episodes resulting in a strain of these musculotendinous units require rest until the symptoms have subsided and then a rehabilitation program involving a gradually increasing regimen of stretching and strengthening. Immobilization should be avoided, and anti-inflammatory agents are sometimes helpful in this situation. The prognosis is good, and an early return to sport is possible depending on the severity of the injury. In an older individual (older than 40 years), acute trauma is more likely to result in a disruption of the rotator cuff. This should be initially treated with rest to allow sufficient healing to take place followed by range of motion and strengthening exercises. Persistent pain or weakness requires further investigation. Surgical repair should be considered early (within 2 months) to minimize the chronic effects of such an injury. Primary Impingement

Patients with primary impingement more commonly present at an older age. This diagnosis implies an anatomic narrowing of the subacromial space. Nonoperative management is attempted, but with failure of nonoperative treatment, we advocate surgical decompression earlier than in someone with secondary impingement.

Instability and Secondary Impingement 

Patients with associated instability are treated nonoperatively, especially if multidirectional instability is diagnosed. The emphasis is on strengthening the cuff after symptoms have subsided. Only after nonoperative management has been attempted for a prolonged period and has failed is surgical management pursued. This usually is in the form of stabilization (arthroscopic or open). Anterior subluxation causing secondary impingement and pain can produce both a diagnostic and a therapeutic challenge. In such cases, prolonged nonoperative measures are appropriate. If surgery is considered, the choice between anterior stabilization (arthroscopic or open), subacromial decompression, or a combined procedure remains unclear. We use our physical examination, EUA, and arthroscopic evaluation to determine the most appropriate course of action and usually perform all technical procedures (both stabilization and decompression) arthroscopically. Internal Impingement

Patients with internal impingement undergo similar nonoperative treatment. Extremes of abduction and external rotation are restricted until symptoms resolve. A program emphasizing deltoid, rotator cuff, and scapular stabilizer exercises is begun, with a gradual return to throwing. Intra-articular cortisone injections are employed on a selected basis to reduce pain in the early stages. Caution should be exercised in the management of these athletes because this can be a very difficult entity to treat. Surgical management is indicated with refractory symptoms and an adequate trial of rehabilitation. Again, the physical examination, EUA, and arthroscopic evaluation will aid in identifying and treating any subtle but pathologic associated anterior laxity. Overuse Problems

Overuse problems should be treated extensively with nonoperative methods. Surgery is recommended in chronic situations in which nonoperative management has failed or when the problem has progressed to the point of a cuff tear. Combined Causes

It must be appreciated that, in athletes, multiple underlying pathologies can often coexist. It is not always clear which of the lesions is primarily responsible for the patient’s symptoms, and there is no guarantee that addressing one lesion or even all the lesions will be curative. Such cases can be very difficult to treat. A nonoperative approach is prudent until the specific components can be determined, and the treatment is modified accordingly.

Shoulder 1011

NONOPERATIVE MANAGEMENT Nonoperative treatment, regardless of the level of performance, follows the same generic plan: 1. Activity modification 2. Medications 3. Stretching 4. Strengthening 5. Ice 6. Physiotherapy Usually, a period of rest is advised, specifically avoiding the overhead sporting activity involved. It should be emphasized that this is an active form of rest. This means that, although the specific activity causing the symptoms is avoided, substitute activities are used, for example, practicing ground strokes and avoiding serving and overhead swings in tennis, or changing from a butterfly to a breaststroke. This “active rest” is particularly important in the high-profile athlete who will not accept the prescription of total rest. We then prescribe NSAIDs, initially for a 2- to 4-week period, with the understanding that they can be used on an as-needed basis thereafter. The use of a steroid is reserved for patients not responding to this regimen of anti­inflammatory medications in addition to the other aspects of the protocol. Subacromial or intra-articular injections of 40 mg of water-soluble steroid combined with 6 to 8 mL of lidocaine and bupivacaine are used as a one-time measure. A peritendinous injection at the level of the transverse humeral ligament is used for bicipital tendinitis, with the same amount of steroid and 2 to 4 mL of lidocaine and bupivacaine. The injection is rarely repeated. Occasionally, for severely refractory cases, we use a tapered oral steroid protocol, but this is a rare circumstance. Stretching is used to maintain range of motion and to correct any obvious discrepancies or contractures, in particular posterior capsular tightness. Cross-arm adduction and overhead adduction stretching are important in the athlete, particularly one with tight posterior structures. Strengthening is the hallmark of nonoperative management of the adult or adolescent athlete with a biceps or rotator cuff lesion. The use of rubber tubing or simple free weights is the most practical method of strengthening the rotator cuff. The scapular muscles cannot be ignored, however, especially when they are implicated in the etiology. Rotator cuff strengthening exercises are performed initially at the side using the rubber tubing. The individual is then progressed to 45 degrees of abduction and then to a more functional level above 90 degrees in selected athletes (especially throwers). The supraspinatus muscle is isolated by abducting the arm in the plane of the scapula with the forearm and shoulder internally rotated. The infraspinatus and teres minor are exercised in external rotation and the subscapularis in internal rotation. Biceps function is improved through elbow flexion and forearm supination. Scapular stabilizers are strengthened by resisted scapular elevation, retraction, and protraction. In addition, the inside push-up with hands placed inside the parasagittal plane of the shoulder has been helpful. Sitting rows for the serratus and rhomboids, shrugs for

the trapezius, and latissimus pull-downs for the latissimus dorsi muscle help control, strengthen, and stabilize the scapula. Local ice application after workouts and competition can prove beneficial and is frequently employed depending on the athlete’s level of participation and response. Physiotherapy is instituted with specific modalities such as ultrasound, transcutaneous nerve stimulation, muscle stimulation, and laser therapy, depending on the individual athlete’s response to these modalities. We employ these modalities only occasionally in resistant cases. Nonoperative management of the athlete with a rotator cuff or biceps tendon problem initially involves regular visits to the therapist. However, it should be emphasized that long-term benefit will be gained through a regular and almost obsessive home exercise program rather than relying on specific physical therapy modalities or medications. Failure of this regimen over a prolonged period of 6 to 12 months constitutes an indication for surgery. The treatment of the high-level athlete may differ in subtle ways from the generic program previously outlined. Included in this group are individuals who may not necessarily be competing at a professional, national, or international level. Nevertheless, they take their sport very seriously and have the same motivation to compete and perform to capacity within their own level. The main difference between their treatment and the more general treatment program is the intensity and volume of the exercises prescribed. Because these individuals place great physical demands on their shoulders, management should reflect the anticipated return to a high level of function. Isokinetic machines can be of value, especially those allowing eccentric training of the musculature. Strengthening is the key to long-term success. Depending on whether the sport involves primarily aerobic or anaerobic shoulder function, strengthening exercises should be low-intensity and high-volume or highintensity, respectively. The other main difference encountered in treatment of high-level athletes is that communication with the coach, team trainer, and parent (if applicable) is essential. Communication creates the best possible environment, with all relevant people involved and working in unison. Specific techniques may need to be changed, equipment modified, and short-term as well as long-term competition goals established. This approach helps the athlete focus and maintain a positive attitude during the rehabilitation process.

SURGICAL MANAGEMENT The primary indication for surgery is the failure of an adequate, nonoperative management program. An adequate, nonoperative program should be carefully coordinated with good patient compliance, and the patient should be followed for at least 3 to 6 months (depending on the athlete and the sport) before surgery is considered. Inherent in establishing failure of this nonoperative program is significant pain and disability warranting intervention. With an acute traumatic injury, especially in an older athlete, the possibility of a full-thickness rotator cuff tear must be entertained. MRI is the gold standard in providing a diagnosis in this situation. If a full-thickness tear of

�rthopaedic ����������� S �ports ������ � Medicine ������� 1012 DeLee & Drez’s� O

the rotator cuff is present, surgical repair is performed. Another situation in which surgery may be considered early is the presence of very obvious primary impingement due to bony overgrowth of the acromion or an abnormally angled acromion, usually in an older athlete. In both these situations, the obvious anatomic abnormality causing the patient’s symptoms necessitates a surgical correction of this abnormality. Overall, the choice of surgical procedure depends on a number of factors, including the underlying cause and the extent of the abnormality. If a full-thickness tear of the rotator cuff is present, open or arthroscopic repair can be performed. The basic principles for successful repair are the same in both open and arthroscopic rotator cuff repair and should not be compromised regardless of the technique selected. The decision on how these basic principles are fulfilled (whether with an open or arthroscopic technique) is then individualized to the surgeon.

Partial-Thickness Rotator Cuff Tears and Internal Impingement Lesions With regard to partial-thickness cuff tears, we currently perform only a débridement of the cuff if the tear involves less than 50% of the tendon. If the tear involves greater than 50% of the tendon, we consider excision of the damaged tissue and formal repair to bone. In either case, a subacromial decompression can be performed arthroscopically as deemed necessary. Postoperatively, passive assisted motion and stretching are started immediately. The patient progresses rapidly to active and resisted motion as tolerated. The nonoperative routine is then instituted to build strength and regain ­function. For those athletes with internal impingement, we débride and/or repair both the posterosuperior labral lesion and the corresponding cuff lesion. We use the physical examination and EUA to determine whether subtle but pathologic differential anterior laxity and posterior contracture are present, and perform and manage appropriately. It should be emphasized again that this is a very difficult population of patients to diagnose and treat, and the surgeon should be certain of the appropriate treatment before proceeding with any surgical management that might result in ­ postoperative failure to regain external rotation in the overhead athlete.

Full-Thickness Rotator Cuff Tears Arthroscopic Surgical Technique Setup The authors’ preferred treatment technique is arthroscopic repair performed in a modified beach chair position. The patient is positioned on a regular bed with a beanbag, and standard positioning precautions are observed as with any operative procedure, taking care to pad and protect areas of potential pressure. The position is a slight modification of the usual beach chair position in that the patient is more upright and the acromion is brought to a position parallel to the floor. We call this the “dinner chair” position.

The patient is then prepared and draped, and a sterile articulated arm holder is used to aid in arm positioning throughout the procedure. The importance of arm positioning is that, as the arm is manipulated, the various compartments of the shoulder are differentially opened up or closed down. Appropriate arm positioning can therefore make an already technically demanding surgery easier. Conversely, inappropriate arm positioning can make it much more difficult.

Arm Positioning Although subtle adjustments to obtain the optimal view and access must be performed intraoperatively, some general principles of arm positioning should be kept in mind. These principles may seem obvious and beneath mentioning, but they are discussed here briefly because proper arm positioning is often overlooked. When working within the glenohumeral joint, the arm is best placed in slight abduction, slight flexion, and about 30 degrees of external rotation in line with the scapular plane. In this position, tension is removed from the supraspinatus, and the glenohumeral joint is easily distracted. As the arm is abducted and internally rotated, the cuff attachments are also brought medially and into the field of view. When working in the subacromial space, position of the arm depends on what structures are to be visualized. Bringing the arm into a more adducted position allows for traction to be applied to the arm to distract the humeral head inferiorly and away from the acromion. This improves access to the inferior aspect of the acromion, which may be useful during arthroscopic subacromial decompression. At the same time, however, adducting the arm collapses the subdeltoid space, making it more difficult to visualize the rotator cuff. To visualize the cuff, then, it is necessary to relax the deltoid by flexing and abducting the arm.

Portals Ultimately, portals are intended to safely provide optimal visualization and access. The arthroscopic portals described here do differ slightly from the classically described portals, but it is the authors’ belief that these subtle adjustments will improve the surgeon’s ability to work arthroscopically (Fig. 17I1-31). To provide a more global “bird’s-eye view” of the shoulder, a slight adjustment is made to the classic posterior portal. Our “high posterior” portal is slightly more superior and slightly more lateral than the standard posterior portal. Its position is established by placing it within the palpable sulcus just inferior to the posterolateral acromion and superior to the humeral head as it slopes inferiorly and medially to the glenoid. This results in an entry point on the skin that is roughly 1 cm inferior and 1 cm medial to the posterolateral corner of the acromion. The anterior portal is located at the midpoint between the coracoid and the anterolateral corner of the acromion. This puts it roughly in line with the acromioclavicular joint. Its trajectory should result in the anterior portal entering the glenohumeral joint through the rotator interval. To help establish this trajectory, a spinal needle may be placed first as a rough guide. Two portals, an anterolateral and a posterolateral portal, are established laterally in place of

Shoulder 1013

Figure 17I1-31   Right shoulder marked with the authors’ preferred portals for rotator cuff repair.

the standard, single “50 yard line” portal. This provides a greater access both anteriorly and posteriorly from the lateral aspect of the shoulder and provides a working and a viewing portal laterally during cuff repair (Fig. 17I1-32).

Arthroscopy The glenohumeral joint is entered through the posterior portal, and the arthroscope is introduced. The arm is immediately positioned in slight abduction, slight flexion, and external rotation to allow for glenohumeral work. The anterior portal is then established, and a 5.5-mm cannula

is introduced in its place. A routine diagnostic arthroscopy is then performed. The purpose of the diagnostic arthroscopy is to identify alternative or associated pathology. In particular, instability can be diagnosed by evidence of increased translation of the humeral head, associated labral pathology, or even a Bankart lesion. Internal impingement is diagnosed based on the pathological changes of the ­posterosuperior glenoid labrum and the articular surface of the cuff. The biceps tendon and its intertubercular portion are drawn into the joint and evaluated. If the biceps tendon is implicated as a contributing factor, synovitis and partial tearing can be débrided. If significant, an arthroscopic biceps tenodesis or biceps tenotomy can be performed, depending on the demands of the patient. The rotator cuff attachment can be easily visualized by internal and external rotation of the arm. If desired, a marker suture introduced through a spinal needle may be placed to mark the location of the rotator cuff tear if there are concerns it will not be easily visible from the subacromial space. Instruments can be introduced through the anterior cannula as needed. After the intra-articular work is completed, the subacromial space is then entered again through the posterior portal. Through the previously established skin incision, an arthroscopic trocar is used to feel the posterior border of the acromion. The trocar is then slid just underneath it and into the subacromial space. The trocar is advanced across the entire subacromial space and exited anteriorly just lateral to the coracoacromial ligament and through the anterior portal skin incision. The 5.5-mm cannula is advanced over the cannula from anteriorly, thus establishing a second portal to the subacromial space anteriorly. A third anterolateral portal is established immediately using a spinal needle. The placement of this portal may need to be adjusted to allow for entry parallel to the acromion and centered between the acromion above and the humeral head below.

Subacromial Decompression A subacromial decompression can now be performed with the arthroscope in the posterior portal and an ablation device followed by a burr in the anterolateral portal. The viewing and the working portal can then be switched to allow for a 90-degree change in perspective and biplanar confirmation that an adequate decompression has been performed (Fig. 17I1-33).

Bursectomy A bursectomy is then performed using a shaver through the posterior portal. Care is taken to resect any adhesions of the posterior cuff to the deltoid as well as any adhesions of the anterior cuff to the coracoid. The rotator cuff tear is identified at this point from the bursal side.

Arthroscopic Rotator Cuff Repair Figure 17I1-32   For repair of the cuff, the arthroscope is in the posterolateral portal. The anterolateral portal is used as a working portal. Sutures are managed through the anterior and posterior portals.

The posterolateral portal is now established between the middle and posterior thirds of the rotator cuff tear. The arthroscope is then placed in this portal, and the 5.5-mm cannula is placed in the anterolateral portal. For the duration of the rotator cuff repair, the posterolateral portal

�rthopaedic ����������� S �ports ������ � Medicine ������� 1014 DeLee & Drez’s� O

Figure 17I1-33   A burr in the subacromial space. A subacromial decompression has been performed.

will be the viewing portal, the anterolateral portal will be the working portal, and the anterior and posterior portals will be used as suture management portals through which sutures are retrieved and organized. The conceptual goal of the rotator cuff repair is re­creation of the disrupted, rotator cuff anatomy. To this end, various techniques and suture configurations have been proposed. The authors’ preferred technique is a double-row, triangle repair (Fig. 17I1-34). In this technique, a “triangle” of anchors is placed for every 8 to 10 mm of torn cuff. This triangle consists of a single medial anchor placed at the articular margin and two lateral anchors placed 5 to 8 mm inferior to the tip of the greater tuberosity on the lateral cortex of the greater tuberosity. This triangle repair has demonstrated anatomic re-creation of the footprint of the native rotator cuff and biomechanical restoration

Figure 17I1-34   Triangle repair.

of cuff tendon strength, theoretically leading to improved healing and therefore improved outcomes.174 The tendon edges are débrided using a shaver or biter. The bony bed of the repair is prepared using a burr to expose bleeding bone. The cuff is mobilized by releasing the cuff off of the superior glenoid (juxtaglenoid capsulotomy), releasing any adhesions to the scapular spine, and releasing any adhesions to the coracoid, including the coracohumeral ligament. Starting medially, the anchors are placed, and all sutures are passed through the tendon before the next anchor is placed. The sutures from the medial anchor are placed in a horizontal mattress as close to the musculotendinous junction as possible and retrieved out the posterior portal. This is repeated with all sutures of the anchor. It is important to note that by passing multiple sutures through the same portal, there is a risk for sutures becoming twisted or tangled with one another. However, by maintaining constant gentle traction on the sutures and keeping them taut, it is possible to pull them to one side of the portal while continuing to work within the remaining portal. Once all sutures have been placed and retrieved out the posterior portal, they are clamped together, and the next anchor is placed. Lateral row sutures are placed in a simple fashion. All sutures are tied at the end through the anterolateral 5.5-mm cannula. The lateral row is tied first because these sutures provide traction and pull the cuff laterally into anatomic position. The medial row of sutures is tied last and serves to compress the cuff down onto its footprint (Fig. 17I1-35).

Open Surgical Technique After diagnostic arthroscopy, if open rotator cuff repair is opted for, the shoulder is re-prepared with povidone-iodine (Betadine) to decrease the risk for infection before the open incision. The incision is then made starting from the posterior aspect of the acromioclavicular joint and extending to the anterolateral corner of the acromion and extending down the deltoid a few centimeters. The subcutaneous

Shoulder 1015

passage of the same sutures to achieve the biomechanically strongest repair of the cuff attachment.31 Heavy absorbable suture is then used to repair the subperiosteally raised flaps and the deltoid raphe, and the overlying skin is closed.

Postoperative Care and Rehabilitation Postoperatively, a shoulder SmartSling (Ossur/Innovation Sports, Pauling, Calif) with a small pillow is used to keep the arm in about 20 to 30 degrees of abduction and to take tension off the repair. Passive assisted motion is started immediately.110 The patient progresses to active motion after about 6 weeks, depending on the extent of the tear.17,114 Active motion is combined with terminal stretching, and resistive motion is added according to each individual’s progress, usually at the 10-week mark.113 The remaining postoperative regimen includes the components outlined in the nonoperative section. Figure 17I1-35   Double row rotator cuff repair viewed arthroscopically. The lateral row has been tied. More medially, the placed but as yet untied medial row can be seen.

plane is developed with a bovie and the acromioclavicular joint, acromion, deltoid, and deltoid raphe between the anterior and middle deltoid are identified. A No. 15 blade is used to first incise obliquely across the acromioclavicular joint, taking advantage of the thick tissue here. This is extended as a single incision to the anterolateral corner of the acromion and down the deltoid raphe. Care is taken to avoid distal extension beyond 5 cm where the axillary nerve could be in jeopardy.110 Subperiosteal flaps are elevated using the No. 15 blade exposing the anterior acromion. In the process of completely exposing the anterior acromion, the coracoacromial ligament is released from its attachment to the undersurface of the acromion. The acromioclavicular joint is likewise exposed. It is important to maintain this as a thick, single flap, taking the tissue right off of the bone so that a solid repair can be performed at the end of the procedure. With adequate exposure, a distal clavicle excision and subacromial decompression can easily be performed through this exposure. Attention is then turned to the rotator cuff. Bursectomy can be performed using Mayo scissors and releases performed including release of coracohumeral ligament. The cuff should now be mobilized. Just as with the arthroscopic procedure, the footprint is prepared using a burr, and a triangle repair is performed using the same configuration of anchors and sutures described previously. Alternatively, the medial row of anchors can be combined with a lateral transosseous

C

r i t i c a l

P

o i n t s

l Accurate diagnosis is essential and should be made through a differential-directed approach. l Judicious use of diagnostic injections and/or further imaging can be used to supplement the history and physical examination. l Both nonoperative and operative treatments can be utilized to restore the normal biomechanics of the shoulder. l Operative repair, when necessary, has as its goal restoration of normal anatomy regardless of the technique employed.

S uggested

R eadings

Burkhart SS, Morgan CD, Kibler WB: The disabled throwing shoulder: Spectrum of Pathology. Part I: Pathoanatomy and biomechanics. Arthroscopy 19:404-420, 2003. Burkhead WZ (ed): Rotator Cuff Disorders. Baltimors, Williams & Wilkins, 1996. Codman EA: The shoulder. Boston, Thomas Todd, 1934. Krishnar SG, Hawkins RJ, Warren RF (eds): The shoulder and the Overhead ­Athlete. Philadelphia, Lippincott Williams & Wilkins, 2004. Yamaguchi K, Ditsios K, Middleton WD, et al: The demographic and morphological features of rotator cuff disease: A comparison of asymptomatic and symptomatic shoulders. J Bone Joint Surg 88A:1699-1704, 2006.

R eferences Please see www.expertconsult.com

1016 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ection

I

Rotator Cuff 2. Superior Labral Injuries Mark W. Maffet and Walter R. Lowe

The recognition of the importance of the superior labrum– biceps tendon complex in normal shoulder function as well as in shoulder pathophysiology has been well demonstrated. This section summarizes normal and abnormal anatomy, biomechanics, diagnosis, treatment, treatment outcomes, and potential complications of injuries to the superior labrum. The importance of the superior labrum and the biceps tendon in shoulder pathology was emphasized in the 1990s. Early thought regarding impingement syndrome and rotator cuff pathophysiology dates back to Neer’s writings in 1972.1 The literature has evolved subsequently to include secondary impingement2 and internal impingement associated with overhead throwing athletes.3,4 Literature regarding shoulder instability treatment dates back to the “hot poker” days centuries ago. Although more modern treatment has evolved in the form of anatomic reconstructions and recognition of subtle instabilities associated with athletes, the acceptance of shoulder looseness as a source of morbidity is longstanding. The superior labrum–biceps tendon complex as a possible source of shoulder pain and dysfunction first was recognized by Andrews and colleagues in 1985.5 In an electromyography study, these investigators successfully showed that biceps contraction could lift the biceps tendon off its glenoid insertion. Andrews and colleagues postulated that tensile overload during eccentric biceps contraction in the follow-through phase of throwing in athletes could lead to injuries to this area. The widespread use of shoulder arthroscopy in the 1990s expanded knowledge of superior labrum injuries. Superior labral pathology has now taken its place alongside impingement and instability as a distinctly separate cause of shoulder dysfunction (Fig. 17I2-1). Its recognition has revolutionized the treatment of overhead throwing athletes Glenohumeral Joint Pathology SLAP

Impingement

Instability

Figure 17I2-1   SLAP (superior labrum, anterior to posterior) lesions are common causes of shoulder dysfunction. Just like instability can be related to impingement symptoms (secondary impingement), SLAP lesions can be related to both instability and impingement-type symptoms.

as well as the approach taken in treating other shoulder pathology. This chapter outlines modern thought in relation to underlying superior labrum and proximal biceps insertional anatomy, describes injury patterns and their relationship to suspected mechanisms of injury, describes new diagnostic methods, and discusses the treatment of these lesions with early outcome measures.

RELEVANT ANATOMY AND BIOMECHANICS Normal Anatomy Understanding of the anatomic relationships of the rotator cuff and glenohumeral ligaments of the shoulder is based on decades of research and study. Only in the 1980s and 1990s, however, has work attempting to understand the anatomy of the superior labrum been published.6-9 Many surgeons may not have a thorough understanding of the anatomy of the superior labrum–biceps tendon complex and its normal variants. The importance of this understanding cannot be overemphasized; otherwise, the surgeon at arthroscopy will not know what needs repair and what should be left alone. The glenoid labrum peripherally surrounds the glenoid and deepens the socket. Perry10 showed that humeral head contact area increased 75% vertically and 67% horizontally when the labrum was intact. The glenoid often is thought of as being flat because of its radiographic appearance. Differences in the glenoid articular cartilage depth (thicker peripherally, narrowing to a central clear area), along with the presence of the glenoid labrum, cause the glenoid to be more cup-shaped, however. This deepening effect appears to help stabilize the glenohumeral joint. When the labrum is intact, some authors11 have suggested a bumper effect stabilizing the humeral head within the glenoid. This effect appears to be more important in the superoinferior direction than in the anteroposterior direction.12 Others13,14 have studied the suction effect of an intact labrum and capsule and its shoulder-stabilizing properties. In addition to these biomechanical effects, the labrum is the point of attachment of the supporting shoulder ligaments. Specifically, the superior labrum is the seat of attachment of the superior glenohumeral ligament, the middle glenohumeral ligament, and the posterosuperior capsule.6,15 Recent studies show that the superior labrum anterior to the biceps tendon anchor often has significant contribution from the inferior glenohumeral ligament

Shoulder 1017 Figure 17I2-2   Histologic stain of a normal inferior labral attachment to the glenoid. Note the firm attachment to the glenoid rim. I, inferior glenohumeral ligament complex; L, labrum. (From Cooper DE, Arnoczky SP, O’Brien SJ, et al: Anatomy, histology, and vascularity of the glenoid labrum: An anatomical study. J Bone Joint Surg Am 74:46-52, 1992.)

(IGHL) in some patients as well.16 It is readily apparent based on these anatomic facts that the glenoid labrum serves an important function in the shoulder. The normal appearance of the labrum can vary significantly depending on the position around the glenoid. The inferior labrum anatomically appears to be an elevated extension of the articular cartilage blending into the capsule (Fig. 17I2-2), whereas the superior labrum above the mid-glenoid notch typically carries a more meniscoid appearance. This normal appearance can vary significantly. Some shoulders show a minimal labral fold all the way around the glenoid; other shoulders have an abundant meniscal appearance circumferentially (Fig. 17I2-3). The

Figure 17I2-3   Anterosuperior sublabral hole. Cadaveric specimen demonstrating a normal sublabral hole under the anterosuperior labrum. (From Cooper DE, Arnoczky SP, O’Brien SJ, et al: Anatomy, histology and vascularity of the glenoid labrum: An anatomical study. J Bone Joint Surg Am 74:46-52, 1992.)

histologic appearance of the glenoid labrum, once thought to be cartilaginous,17 appears to be more fibrous in nature, consistent with its appearance as an extension of the glenohumeral ligaments (see Fig. 17I2-2).18 There is a fibrocartilaginous transition zone from the articular cartilage to the fibrous labral tissue. The labrum above the mid-glenoid notch can have even more variability. The anterosuperior labrum can be attached firmly to the glenoid rim or can be detached with a sublabral foramen present. This normal variant can be seen in 12% of shoulders19 and should not be confused with a pathologic labral detachment. This sublabral hole communicates anteriorly with the subscapularis recess, both of which can vary in size. The middle glenohumeral ligament (MGHL) attaches onto the anterosuperior portion of glenoid labrum forming a continuous structure. Occasionally in shoulders with a prominent sublabral foramen, the MGHL has a cord-like appearance that merges with the anterosuperior labrum to insert at the anterior base of the biceps tendon (Fig. 17I2-4).7 A small percentage of

Figure 17I2-4   Arthroscopic photograph of a cord-like middle glenohumeral ligament (arrow) as visualized from the posterior portal. This normal variant should not be confused with a capsular tear.

1018 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

II

III

IV

I

A

B

C

D

Figure 17I2-5   Subtypes of superior labral attachment. Type I superior labral attachment is entirely posterior (A). In type II, it is mostly posterior (B). Type III is equally distributed anteriorly and posteriorly (C). In type IV, most of the attachment is to the anterior labrum (D). (Redrawn from Vangsness CT, Jorgenson SS, Watson T, Johnson DL: The origin of the long head of the biceps from the scapula and glenoid labrum. J Bone Joint Surg Br 76:951-954, 1994.)

these cord-like variants can have a prominent recess below the MGHL that has the overt appearance of a capsular rent. Williams and associates19 called this normal variant a Buford complex. The clinician must be aware of the appearance of this variation to avoid unnecessary repairs. Superiorly the biceps tendon has a close relationship with the glenoid labrum. The biceps tendon inserts onto the supraglenoid tubercle after its entry into the joint through the rotator interval. This tendon insertion begins about 5 mm medially from the superior edge of the glenoid. The articular cartilage often extends superiorly over the top of the glenoid 5 mm to the biceps. The fibers of the biceps tendon blend into the fibers of the labrum and capsule surrounding this insertion.6 Normal variations can exist as well in this biceps insertion onto the labrum. In the sagittal plane, as one faces the glenoid, four types of attachments have been described (Fig. 17I2-5).8 The tendon’s labral attachment can be (1) entirely to the posterior labrum, (2) entirely to the anterior labrum, (3) equally to both, or (4) mostly posterior with a small anterior contribution. Differences in the biceps insertion have implications in the type of superior pathology that develops during an injury. Cooper and colleagues6 carefully evaluated the cross­sectional anatomy of the biceps tendon–superior labral complex. Evaluating normal biceps insertional anatomy in the frontal plane shows an intimate relationship between the proximal biceps insertion and the superior glenoid labrum. Some authors have described the labrum as inserting onto the tendon,6 whereas other authors describe the tendon as inserting onto the labrum.18 The point is probably moot because the superior labrum itself appears typically to have only a loose connection to the glenoid periphery (Fig. 17I2-6). The supraglenoid tubercle onto which the biceps tendon attaches is about 5 mm medial to the glenoid rim. Because the superior labrum does not always attach firmly to the superior glenoid over this distance, a small synovial recess may exist beneath a meniscoid-appearing superior labrum. This particular cross-sectional anatomy is important to understand because it may explain why certain patterns of superior labral tears are seen. The vascularity of the glenoid labrum originates from the suprascapular artery, the posterior circumflex artery, and the circumflex scapular branch of the subscapular

artery.6 Apparently there is no vascular contribution to the labrum from the underlying bone. The penetration of vascular branches from the periphery into the substance of the labrum varies. Similar to the knee meniscus, most of the labral vascularity in the meniscoid variety is limited to its periphery. The superior and anterosuperior labrum is less vascular than the inferior and posterior portions. This diminished vascularity may contribute to delayed or

Figure 17I2-6   Histologic picture of a cross section of the superior labrum (L), the glenoid, and the biceps tendon (B). This photomicrograph demonstrates primary attachment of   the superior labrum to the biceps tendon before it inserts onto the supraglenoid tubercle. Note the lack of firm attachment of the superior labrum to the glenoid itself. (From Cooper DE, Arnoczky SP, O’Brien SJ, et al: Anatomy, histology, and vascularity of the glenoid labrum: An anatomical study. J Bone Joint Surg Am 74: 46-52, 1992.)

Shoulder 1019

Figure 17I2-7   Baseball pitcher in the cocking phase of the throw. Note the pitching arm in 90 degrees of abduction and external rotation with posterior rotation on the biceps in this position.

incomplete healing after injury and higher rates of failure after repair. The vascularity of the glenoid labrum often appears to decrease with increased age of the individual.

Biomechanics The structural importance of the biceps tendon–superior labrum complex to the glenohumeral joint became increasingly clear in the 1990s. Good biomechanical studies showed that this complex is important in normal shoulder function and that disruption of it has detrimental effects. That the biceps tendon functions as a humeral head depressor has been established previously.20,21 That function and its relationship to rotator cuff pathology have been well studied. Warner and McMahon22 showed the importance of the biceps tendon in superior stability of the glenohumeral joint as the arm is abducted in the scapular plane. Increased superior translation was noted varying

Figure 17I2-8   A chart of electromyographic activity of the biceps as a percentage of maximal manual muscle test for each phase of the overhand throw. Comparison is made between biceps activity in normal shoulders and in those with instability. Note increased biceps activity in the unstable shoulder. (Redrawn from Glousman R, Jobe F, Tibone J, et al: Dynamic electromyographic analysis of the throwing shoulder with glenohumeral instability. J Bone Joint Surg Am 70:220-226, 1988.)

Biceps

100 Instability

90 % Maximal Manual Muscle Test

2 to 6 mm when the biceps tendon was ruptured in vivo compared with the contralateral control shoulder. More recently, the importance of the biceps tendon in supporting the anterior shoulder has been studied.23,24 Pagnani and associates23 looked at translations of the glenohumeral joint with and without detachment of the superior labrum–biceps tendon complex. Significant increases of superoinferior and anteroposterior translations were found in shoulders with created SLAP (superior labrum, anterior to posterior) lesions when the arm was in the abducted position. Pagnani and associates23 created SLAP lesions not involving detachment of the biceps insertion as well and found no increase in translation. Rodosky and coworkers,24 using a dynamic cadaver shoulder model, showed that the biceps contributed to torsional rigidity of the shoulder in abduction and external rotation. With the shoulder in this position, the biceps seemed to support the function of the inferior glenohumeral ligament. Next, the authors created SLAP lesions with biceps tendon detachment and found decreased torsional rigidity of the joint and increased strain on the IGHL. The results of this work are complemented by electromyography studies performed at the Kerlan-Jobe Clinic.25,26 During the overhead throwing motion, high electromyography activity was found in the biceps during the late cocking phase when the shoulder was at an extreme of abduction and external rotation (Fig. 17I2-7).25 In a comparative study done on pitchers with known instability of the shoulder, there was higher biceps electromyography activity in these subjects compared with normal controls (Fig. 17I2-8). Others studies have also shown increased strain on the superior labrum27 and failure of the biceps tendon–superior labrum complex during this critical late cocking phase.28 It has been postulated that the shoulder-­stabilizing effects (Box 17I2-1) of the biceps occur through (1) compression of the joint, (2) tension on periligamentous fibers, (3) anterior and superior barrier function of the tendon, and (4) placement of the joint into a position that secondarily tightens the ligamentous structures.29,30 It is likely that a combination of these effects comes into play.

Normals

80 70 60 50 40

35

30 21

20 10

8

26 17

32

12

8

18

13

0 Wind Up

Early Cocking

Late Cocking

Acceleration p < .05

FollowThrough

1020 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 17I2-1 S  houlder-stabilizing ­Effects of the Biceps 1. Compression of the joint 2. Tension on periligamentous fibers 3. Anterior and superior barrier function of the tendon 4. Placement of the joint into a position that secondarily tightens ligamentous structure

Most recently, Burkhart and others have presented a unified concept of the disabled throwing shoulder with the SLAP lesion at the center of the pathoanatomy.31 Before this, most sports medicine surgeons have treated throwing athletes with a “dead arm” as having problems originating from microinstability. Burkhart instead postulates that a posteroinferior capsular contracture leading to a SLAP lesion is the culprit. This contracture first causes a glenohumeral internal rotation deficit (GIRD), which can worsen over time. In their reciprocal cable model (Fig. 17I2-9) of shoulder motion, a tight posteroinferior glenohumeral ligament causes a posterosuperior directed force on the humeral head during the cocking phase of the throw. This in turn allows hyperexternal rotation due to (1) a posterosuperior shift of the center of rotation of the humeral head (Fig. 17I2-10) and (2) the loss of the normal buttress or cam effect of the inferior humeral head across the IGHL. This functionally “loosened” IGHL allows even more external rotation (Fig. 17I2-11). This torsional posterosuperior centered force during the cocking phase of the throw, along with the altered direction of pull of the biceps tendon at 90 degrees of abduction and hyperexternal rotation, can cause a peel-back mechanism failure of the superior labrum–biceps tendon complex leading to a SLAP lesion. Once the superior labrum

Lesser tuberosity

AIGHL

Contracted PIGHL

Figure 17I2-10   Schematic of shift of center of rotation posterosuperiorly. Note that when the posterior band of the reciprocal cable model is contracted, the glenohumeral contact point shifts posterosuperiorly. The allowable arc of external rotation increases. AIGHL, anterior inferior glenohumeral ligament; PIGHL, posterior inferior glenohumeral ligament. (Redrawn from Burkhart SS, Morgan CD, Kibler WB: Current concepts. The disabled throwing shoulder: Spectrum of pathology. Part I: Pathoanatomy and biomechanics. Arthroscopy 19:404-420, 2003.)

is unstable, further external rotation and posterosuperior shift can occur. In throwing athletes, the SLAP lesion due to this mechanism is typically more posterior than anterior. Additionally, the hyperexternal rotation of the humeral head can lead to (1) hyper-twisting of the posterosuperior rotator cuff fibers and eventually articular side failure and

Greater tuberosity AIGHL

Lesser tuberosity PIGHL

Figure 17I2-9   Schematic of reciprocal cable model. The inferior glenohumeral ligament (IGHL) is represented as its anterior and posterior bands acting as interdependent cables. AIGHL, anterior inferior glenohumeral ligament; PIGHL, posterior inferior glenohumeral ligament. (Modified from Burkhart SS, Morgan CD, Kibler WB: Current concepts. The disabled throwing shoulder: Spectrum of pathology. Part I: Pathoanatomy and biomechanics. Arthroscopy 19:404-420, 2003.)

Figure 17I2-11   Schematic of functionally loosened inferior glenohumeral ligament (IGHL). Again using the reciprocal cable model, the contracted posterior cable causes a posterosuperior glenohumeral shift. This, in turn, reduces the space-occupying effect of the proximal humerus on the anteroinferior capsule. The relative capsular redundancy can look like anterior microinstability. (Redrawn from Burkhart SS, Morgan CD, Kibler WB: Current Concepts. The disabled throwing shoulder: Spectrum of pathology. Part I: Pathoanatomy and biomechanics. Arthroscopy 19:404-420, 2003.)

Shoulder 1021

Biceps tendon

A

B

Figure 17I2-12   The torsional peel-back mechanism of injury to the superior labrum and the biceps tendon. Note the difference between the resting position (A) and the position at 90 degrees of abduction and external rotation (B). The biceps tendon rotates posteriorly, peeling the posterosuperior labrum off the glenoid. (Redrawn from Burkhart SS, Morgan CD: The peel-back mechanism: Its role in producing and extending posterior type II SLAP lesions and its effect on SLAP repair and rehabilitation. Arthroscopy 14:637-640, 1998.)

(2) anterior ligament fiber failure in conjunction with the pseudolaxity already present (Fig. 17I2-12).

CLASSIFICATION Snyder and colleagues in 199032 originally classified the pathoanatomy they found in the superior labrum. They described four main patterns of injury (Fig. 17I2-13). Type I lesions involved fraying of the superior labrum but no detachment of the biceps tendon insertion. Type II lesions were characterized by detachment of the biceps tendon from the supraglenoid tubercle. In type III lesions, a bucket handle tear of a meniscoid-type superior labrum occurred. Type IV lesions involved a superior labrum detachment that extends into the substance of the biceps tendon (Fig. 17I2-14). Others7 later found pathologic patterns that did not fit within Snyder’s classification (Fig. 17I2-15): (1) a biceps insertion detachment extending anterior to include a Bankart lesion in a continuous labral detachment, (2) a cord-like MGHL with detachment of the biceps tendon (Fig. 17I2-16), and (3) a flap tear of the superior labrum with biceps separation (Fig. 17I2-17). There has been some confusion regarding Snyder’s original classification. Of Snyder’s four types, only the type II lesion was originally described as involving an actual detachment of the biceps insertion. The biceps attachment remains intact in types I, III, and IV lesions. Other authors have used the term SLAP lesion only when the biceps detachment is found. The authors believe that the basic classification of the SLAP lesion should not be made complicated. Although we described additional patterns of pathology,7 we now feel these types can be put within the original Snyder classification. Type V is a variant of type III, with a flap of labrum instead of a bucket handle pattern. Type VI and VII are variants of type II but with a coexisting Buford complex and Bankart lesion, respectively. There probably could be simply two categories of superior labral lesion: (1) lesions with the biceps detached and (2) lesions with

the biceps attachment intact. These two categories have significant bearing on the mechanism of injury that causes the lesion, associated pathology found, and the surgeon’s treatment plan. Morgan and coworkers8 divided type II lesions (biceps detached) based on whether the detachment was (1) ­ primarily posterior, (2) primarily anterior, or (3) primarily anterior and posterior. These subtypes probably are important to know because they may be related to certain injury mechanisms, affect the type of repair attempted, and affect postoperative care. When evaluating a suspected type II SLAP lesion, one thing not always clear is how much biceps tendon must be detached to consider the lesion pathologic. Complete detachments are not a problem diagnostically (Fig. 17I2-18). One often sees a tendon, however, that appears partially separated. The surgeon has to incorporate the superior labrum–biceps tendon complex appearance with other arthroscopic clues (see later) and with the patient’s history and physical complaints in deciding whether the anatomy seen is pathologic and in need of repair. The biceps tendon typically begins attachment to the supraglenoid tubercle 5 mm from the edge of the glenoid.6 A 3-mm probe used as a guide can help one determine whether the biceps is attached or not. Mihata and others33 have recently devised a scoring system based on the depth of the superior labral sulcus. This sulcus score is an attempt to quantitatively describe the size and severity of a type II SLAP lesion. One thing, though, is clear: the surgeon must be familiar with normal superior labral anatomy and its many variants before diagnosing and classifying a SLAP lesion.

EVALUATION Clinical Presentation and History Determining whether a patient has a superior labral lesion as the source of shoulder pain can be a challenge. The most important part of the work-up is to maintain a high level

1022 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

D

B

C

Figure 17I2-13   The original Snyder classification of SLAP (superior labrum, anterior to posterior) lesions. A, Type I has degenerative superior labrum tearing but attached biceps. B, Type II has detachment of the superior labrum–biceps tendon complex from the superior glenoid. C, Type III has a bucket handle tear of a meniscoid superior labrum but attached biceps. D, Type IV has tearing of the superior labrum up into the biceps tendon. Variable amounts of the biceps are left attached. (Redrawn from Snyder SJ, Karzel RP, Del Pizzo W, et al: SLAP lesions of the shoulder. Arthroscopy 6:274-279, 1990.)

Figure 17I2-14   Arthroscopic photograph of a type IV SLAP lesion. One can see the extension of the tear into the substance of the biceps tendon (arrow).

of suspicion for these problems. Symptomatic superior labrum–biceps tendon complex lesions can act like almost anything. The patient can present with complaints similar to an impingement syndrome, chronic acromioclavicular (AC) joint pain, refractory bicipital tendinitis, or a case of symptomatic instability. The most common presenting symptoms are pain,34 often with overhead activities, and mechanical catching, popping, or grinding. Depending on what secondary pathology accompanies the SLAP lesion, a combination of these symptom complexes may be ­present. What types of injury can cause damage to the superior labrum–biceps tendon complex? Opinion varies greatly, and several types of injury have been implicated (Table 17I2-1). Different mechanisms may cause different types of superior labral pathology. These differing mechanisms of injury need to be appreciated when eliciting a history from the patient with shoulder pain.

Shoulder 1023

A

B

C Figure 17I2-15   Additional patterns of superior labrum–biceps tendon complex injury. A, An anterior Bankart lesion continues superiorly to include separation of the biceps tendon. B, An unstable flap tear of the labrum is present in addition to biceps tendon separation. C, The superior labrum–biceps tendon avulsion extends anteriorly beneath the middle glenohumeral ligament. (Redrawn from Maffet MW, Gartsman GM, Moseley JB: Superior labrum/biceps tendon complex lesions of the shoulder. Am J Sports Med 23:93-98, 1995.)

Figure 17I2-16   Photograph of a cord-like middle glenohumeral ligament (MGHL) (arrow on right) with a concomitant SLAP (superior labrum, anterior to posterior) lesion (arrow on left).

Figure 17I2-17   Flap-type SLAP (superior labrum, anterior to posterior) lesion. As viewed from the posterior portal, the posterior-based unstable superior labral flap (arrow) can be noted. This variety is similar in origin and treatment to the type III SLAP lesion.

1024 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17I2-18   Photograph of a type II SLAP (superior labrum, anterior to posterior) lesion. Note the separation of the superior labrum from the superior glenoid rim (arrow).

In their original study, Snyder and colleagues32 found that many of their patients described a fall on the outstretched arm as the inciting trauma to the shoulder. Snyder extrapolated a superior translation and compression of the humeral head damaging the superior labrum. In a case report, Lee and Harryman35 noted the occurrence of superior labral damage in a paralytic patient. The mechanism of persistent superior compression force in a patient using the upper extremity as a weight-bearing joint is assumed. Andrews and coworkers5 noted that firing of the biceps muscle caused elevation of the superior labrum off the ­glenoid. Because their electromyography studies showed the greatest biceps firing during the deceleration phase of the overhead throw, Andrews and coworkers speculated that injuries to the superior labrum–biceps insertion occurred during this portion of the throw. Studies at the Kerlan-Jobe clinic25,26 support this finding of high levels of biceps firing during the overhead throw. ­Others36 evaluating the overhead throw noted that 1090 N of shoulder compression force was produced shortly after ball release and believed this could help explain the high incidence of SLAP lesions in overhead athletes. It is possible that biceps firing versus compression mechanisms during the throw may create different types of SLAP lesions. Traction injuries have been suggested to cause SLAP lesions. Maffet and associates7 polled 84 patients and found many describing a typically sudden traction injury as the onset of shoulder pain. A fall onto the point of the

  TABLE 17I2-1  Mechanisms of Injury and Possible Resultant Superior Labral Pathology Overhead Thrower Fall on outstretched arm Traction

Torsional Peel-Back Superior compression Avulsion

shoulder, pulling superiorly on a heavy object, waters­ kiing injuries, and traumatic dislocations were described. Bey and coworkers37 in a cadaver study found that stress on the superior labrum–biceps tendon complex was enhanced significantly when inferior instability was introduced. These investigators noted that the direction of pull of the biceps tendon changed when inferior subluxation was produced. Bey and coworkers37 speculated this change caused the increased stress on the superior labrum found in their study. Burkhart and colleagues38 described a torsional peeling-back mechanism in throwing athletes. This mechanism occurs with the humerus abducted and external rotation occurring with glenohumeral joint compression as in the overhead throw (see Fig. 17I2-12). These authors speculated that this mechanism may cause either a posterior subtype SLAP lesion or extension of a type II lesion posteriorly. This mechanism could cause injury in the late cocking phase of the overhead throw. They discussed the contribution of posterior capsular tightness in the evolution of SLAP lesions in pitchers. These authors speculated that tightness posteroinferiorly adjusts the resting position of the humeral head superiorly, making torsional peel-back of the biceps and superior labrum more likely to occur during the cocking phase of the throw. All these mechanisms likely contribute to injury to the superior labrum and biceps tendon. It is likely that different types of SLAP lesions occur with different injury mechanisms. This fact has not been documented yet and is under study. The more one understands the relationship between a specific mechanism and the pathology it typically creates, the more accurate the clinical diagnosis in the office will be.

Physical Examination and Testing Persistently positive biceps tendinitis signs may be found, but not always. A positive Speed’s or Yergason’s test makes sense given the involvement of the proximal biceps insertion but is not a reproducible finding.39 One always must consider superior labral and biceps tendon pathology when a biceps tendinitis fails to improve despite adequate conservative management. Because of the common association of rotator cuff damage with SLAP lesions, typically in the posterosuperior portion,31 a clinical examination entirely consistent with impingement or rotator cuff tendinitis or tear often is seen.40 This association is one reason why we believe all rotator cuff patients should have diagnostic arthroscopy before even a planned open repair. A SLAP lesion probably would be missed from a purely open approach and may lead to persistent shoulder pain after rotator cuff repair. The tear may recur if an unrecognized SLAP lesion and its resultant posterosuperior laxity are missed. Berg and Ciullo41 studied a group of patients who had persistent shoulder pain after an AC joint resection. These patients were diagnosed clinically with AC joint pain and had undergone a distal clavicular resection. Berg and Ciullo41 found that 75% of these patients with failed AC resections had SLAP lesions that had not been diagnosed at or before the first surgery.

Shoulder 1025

  TABLE 17I2-2  Provocative Tests for SLAP (Superior Labrum, Anterior to Posterior) Lesions SLAP Lesion Tests

Method

Compressionrotation test51

Supine patient Catching Shoulder 90 degrees Snapping abducted Pain Compression to joint Humerus rotated Upright patient Catching Shoulder 160 degrees Pain elevated in scapular Click plane Compression to joint Humerus rotated Upright patient Pain more in pronated Shoulder 90 degrees than supinated elevated and adducted Resisted upward force fully pronated Resisted upward force fully supinated Upright patient Pain more in pronated Shoulder 90 degrees position abducted Shoulder externally rotated with forearm pronated then supinated Supine patient Pain more with resisted Shoulder 120 degrees elbow flexion of flexion, maximal external rotation Forearm supinated Resisted elbow flexion

Crank test48

O’Brien’s test46

Pain provocative test45

Biceps load test44

Positive

Attempts have been made to improve the preoperative clinical diagnosis of SLAP lesions. Several clinical tests have been described with varying degrees of accuracy (Table 17I2-2).42-48 Most of these tests appear to be variations of maneuvers that either try to pinch a torn labrum between the humeral head and glenoid, causing pain, clicking, or popping, similar to a McMurray’s test in the knee, or place traction on the biceps tendon. Mimori and colleagues45 noted a difference in pain provoked when alternately pronating and supinating the forearm with the humerus at 90 degrees of abduction and external rotation. Increased pain in the pronated position relative to supination was considered positive for superior labral pathology. The pilot study by Mimori and colleagues45 showed a 100% sensitivity rate with 90% specificity. Similarly, O’Brien and associates46 and Berg and Ciullo42 described a test in which the patient has the arm extended in front with the elbow extended. The arm is adducted across the body with the forearm pronated, then with it supinated. Increased pain in the pronated position versus supination was consistent with a SLAP lesion. Pain in both positions more likely would indicate AC joint pathology. It is believed that increased biceps tension in the pronated position provokes pain. Other authors have described more specificity in the preoperative clinical examination. Morgan and colleagues8 believed they could differentiate between

­ osterior and anterior type II SLAP lesions. Postep rior lesions would show a positive Jobe’s relocation test, whereas anterior SLAP lesions would have positive Speed’s and O’Brien’s test. Kim and coworkers44,49 devised a test they thought could differentiate between anterior shoulder instability alone and anterior instability with a SLAP lesion. A standard apprehension test with the arm in the abducted and externally rotated position was performed with a positive result present in the anteriorly unstable patient. Active elbow flexion against resistance in the stressed position would improve symptoms unless a SLAP lesion also was present. No change or worsening pain with active elbow flexion was considered diagnostic of a SLAP lesion accompanying anterior instability in Kim’s study. In our experience, SLAP lesions are sometimes difficult to diagnose accurately by clinical examination. The ­previously mentioned tests are useful when employed in the context of the entire clinical history and examination along with a high level of suspicion. The clinical picture may be entirely consistent with rotator cuff pathology or entirely consistent with instability, depending on associated shoulder pathology.

Imaging Imaging studies can be helpful diagnostically. Magnetic resonance imaging (MRI) of the shoulder can show the anatomy of the superior labrum and the proximal biceps tendon insertion well. Limitations exist in the sensitivity of nonenhanced MRI to detect superior labrum detachments. Although some authors50 have shown success in detecting SLAP pathology on plain MRI, the presence of a highly qualified musculoskeletal radiologist directing scan parameters and interpretation in their study makes a huge difference. Not all orthopaedists in the community have access to this skill level or interest from an MRI radiologist, so some series show a high level of missed ­pathology.51,52 The use of gadolinium-enhanced MRI in the shoulder has increased greatly accurate preoperative diagnosis of superior labral pathology.53,54 If appropriate slices are obtained, dye often can be seen leaking between a detached superior labrum–biceps tendon complex and the glenoid (Fig. 17I2-19). This appearance can be diagnostic of a SLAP lesion. The orthopaedist must be cognizant of and familiar with normal superior labral variants, however, when interpreting these scans. The normal anterosuperior labral detachment and the normal sublabral recess under a meniscoid-type superior labrum can lead to falsepositive readings of MRIs. The orthopaedist must become familiar with interpreting superior labral anatomy on gadolinium-enhanced MRI. All younger patients with refractory shoulder pain being evaluated by MRI should have a gadolinium-enhanced MRI scan to increase diagnostic accuracy. Arthroscopy is still the gold standard in the diagnosis of superior labral lesions. The diagnosis more often than not is made or modified at arthroscopy. The surgeon should be aware of repair techniques and have necessary equipment available to handle SLAP repairs when performing shoulder arthroscopy.

1026 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

Figure 17I2-19   Gadolinium-enhanced magnetic resonance images of SLAP (superior labrum, anterior to posterior) lesions. A-C, Note the leakage of the dye under the biceps tendon insertion in three separate cases (arrowheads). These images are clearly diagnostic of a type II SLAP lesion. D, A type IV SLAP lesion is suspected owing to the imaging appearance.

TREATMENT OPTIONS Nonoperative Difficulty in evaluating any nonoperative treatment of SLAP lesions lies in the fact that diagnostic confirmation of superior labral tears may not occur until the time of operative arthroscopy. Very few studies document SLAP lesions, then do nothing to treat them operatively. Thus, conservative treatment has no proven role in the treatment of superior labral injuries. Although judicious use of nonsteroidal anti-­inflammatory medications and isolated corticosteroid injections may decrease symptoms, they do not cure the mechanical problems created by a superior labral lesion, often leading to symptom recurrence when the medication is stopped. A shoulder rehabilitation program centered around rotator cuff exercises, scapular stabilization exercises, posteroinferior capsular stretching, and regaining a full normal range of motion no doubt would optimize any surgical treatment

that ensues and help improve symptoms of associated instability or rotator cuff disease. This program may alleviate symptoms of SLAP lesions in which the biceps anchor is stable. Also, familiarity with this rehabilitation program benefits the patient recovering from any surgical procedures required to correct the superior labral pathology. In overhead throwers, there may be a prophylactic benefit to a posteroinferior capsular stretching program to halt the pathologic cascade leading to SLAP lesions. This benefit has not been proved yet with good prospective studies, but if Burkhart’s model of the disabled throwing shoulder is correct, the pathologic cascade should be interrupted. We believe that the problem of internal rotation contracture in the overhead throwing athlete should be recognized early and treated aggressively.

Operative Arthroscopic surgical techniques are invaluable in the evaluation and treatment of superior labral lesions. Variations in the normal anatomy of the superior labrum, superior

Shoulder 1027

glenohumeral ligament, and MGHL make interpretation of pathologic lesions and the extent of these lesions difficult. A thorough understanding of normal superior labral anatomy and normal superior labral anatomic variants as discussed earlier is essential. A cord-like MGHL (Buford complex) may confuse the surgeon and result in surgical treatments that may worsen the patient’s overall shoulder condition.19 Recognition of the pathologic aspects of the superior labrum is equally important in dictating the surgical treatment. Classifying superior labral pathology is the first step in the surgical management of SLAP lesions. Type I through IV lesions have been well described. More complex superior labral lesions also must be appreciated, however.7,8,55 Recognition of these lesions, as well as the associated pathology, such as concomitant anterior capsulolabral injury, allows arthroscopic and open surgery to be planned. Any information available preoperatively, such as the appearance of the anterior, posterior, and superior labrum on gadolinium MRI, may be helpful in planning not only the surgical repair but also the surgical positioning for that repair.

Diagnostic Arthroscopy Diagnostic arthroscopy of the shoulder joint may be performed in either the beach chair or the lateral decubitus position. The advantages and disadvantages of these two positions are debated among shoulder surgeons. Surgeon familiarity with either position is paramount to a successful surgical procedure. Snyder56 described a 15-point arthroscopic examination visualizing all shoulder structures from posterior and anterior arthroscopic portals. Familiarity with diagnostic arthroscopy of the shoulder through both of these portals greatly aids the arthroscopic surgeon in repair of superior labral lesions. Probing of the superior labrum allows the surgeon to assess the stability of the biceps anchor on the supraglenoid tubercle. When superior labral pathology has been visualized and biceps anchor stability assessed, surgical treatment can proceed.

Débridement of Superior Labral Lesions Most SLAP lesions have associated degenerative edges, flap tears, or bucket handle tears of the central edge of the superior labrum. Débridement of this pathologic tissue is simple. Care must be taken to ensure that the labral change is pathologic because normal variants and age-related fraying are much more common than true superior labral pathology. Débridement is performed with an arthroscopic shaver, arthroscopic baskets, or thermal devices to ablate and smooth the lesion. Care must be taken to avoid injuring the articular cartilage surface of the glenoid and preserving all normal functional labral tissue. Any of the available devices may cause injury if not used with care.

Preparation of the Superior Glenoid The superior labrum is known to be an area of questionable vascularity. Cooper and associates6 and Rouse and coworkers57 have shown abundant vascularity of the entire labrum with the exception of the superior portion. This area where SLAP lesions occur is vascularized only at the periphery with a relatively avascular central edge. Anything the arthroscopic surgeon can do to promote healing in this area is beneficial. Although there are no basic science studies comparing biceps anchor repair with and without abrasion of the superior glenoid, common sense dictates that preparation of the superior glenoid bony surface is necessary to maximize the healing of labral tissue to the bone (Fig. 17I2-20). Preparation of the bony glenoid surface can be done with an arthroscopic shaver, bur, or arthroscopic rasps; this part of the procedure is performed easily through a properly placed anterosuperior or anteroinferior portal.

Repair of the Biceps Anchor There are many different techniques to repair or stabilize the biceps anchor. Each of these techniques has its proponents. Surgeons who are not acquainted with

Operative Arthroscopy The surgeon should be familiar with the use of suture anchors, knot tying, and suture passing techniques. Proper portal placement is essential to success in repairing lesions that extend anterior and posterior to the biceps anchor. This may include, in addition to the standard anterior and posterior portals, a posterolateral portal typically located 1 cm lateral and 1 cm anterior to the posterolateral corner of the acromion. The principles of treatment of these lesions are simple after proper portal placement and include the following: 1. Débridement of degenerative labral tissue 2. Preparation of the superior glenoid for biceps anchor repair 3. Biceps anchor repair 4. Repair of other pathology Each of these steps is addressed separately.

Figure 17I2-20   Photograph of arthroscopic preparation of superior glenoid for repair. Preference is for débridement and decortication with a shaver or a bur on reverse.

1028 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

arthroscopic knot-tying techniques can choose tacktype repairs. Surgeons who possess knot-tying skills may prefer suture anchor-type repairs. Becoming acquainted with both of these techniques significantly increases the likelihood that an anatomic repair can be obtained. Studies show good results with the use of arthroscopic tacks34,58-60; however, complications have been demonstrated with these types of anchors not found after suture anchor repairs.61,62 Proper placement of the arthroscopic portals significantly aids placement of the anchors and secure knot fixation. An appropriate angle against the glenoid rim is necessary to avoid breaking into the articular surface during drilling or anchor insertion. A posterolateral portal is usually made through the rotator cuff to appropriately place anchors in the posterosuperior glenoid rim. This portal typically ends up being near the musculotendinous junction of the rotator cuff and not in the critical zone where rotator cuff tears typically originate. Newer anchors and passage instruments with smaller diameters allow smaller cannulas to be placed through the cuff. Several types of suture anchors are available for use in SLAP repairs. They may be metallic or bioabsorbable. Nonabsorbable suture should be used. Likewise, many types of suture passage instruments are available on the market. Personal preference often dictates what works best in a given surgeon’s hands. The angle of passage through the labrum needs to work. Minimal damage to the labrum should occur during passage. A single suture loop works best. Either sliding knots or nonsliding knots can be used, again based on familiarity. Sliding knots should be backed up by nonsliding knots.

WEIGHING THE EVIDENCE It is clear based on the evidence that SLAP lesions are a significant cause of shoulder pain. In reviewing the salient orthopaedic literature regarding the treatment of SLAP lesions, it is apparent that the orthopaedic surgeon’s op­ tions continue to evolve. Simple débridement alone has failed to achieve satisfactory long-term results, especially in athletes. In 1993, Cordasco and associates40 reported on 27 patients with unstable SLAP lesions treated with débridement alone. Although 78% of patients showed excellent pain relief at 1 year, these results deteriorated to 63% at 2 years, and only 45% of the patients were able to return to their previous level of athletic performance. The authors attributed these unsatisfactory and diminishing results to instability associated with the superior labral lesion. No patients in this study had a history of dislocation or instability on clinical examination, but 70% had instability on examination under anesthesia. This finding suggests that even subclinical instability associated with unstable biceps anchor may be the reason unrepaired SLAP lesions cause symptoms. Failure to regain the stability of the biceps anchor predictably leads to continued symptoms. Other authors have reported similarly poor results with simple débridement of unstable superior labral lesions.62,63,51 Because of the uncertain results after simple débridement regarding pain relief and return to athletic activity, the evolution of treatment of superior labral injuries has focused more recently on repair of the biceps anchor to the bony glenoid.

Many techniques have been described for repair of the unstable biceps anchor, including metal staples, metal screws, bioabsorbable tacks,64 transosseous suture techniques, and suture anchor techniques. DiRaimondo and associates65 biomechanically evaluated pull-out strength of initial fixation strength of tissue tacks compared with suture anchors and found no significant difference. In an early outcomes study, Snyder and colleagues66 reported their results with 140 patients treated with a variety of methods. Second-look arthroscopies were performed on 18 of the 140 patients. Of these 18 patients, five type II lesions were treated with abrasion and débridement alone. Three of the five had healed at second-look arthroscopy. Of five type II lesions treated with tack fixation, four of the five had healed. These five second-look procedures were to remove loose bioabsorbable tack material. Three type III lesions and a type IV lesion that were treated with débridement alone had a normal appearance on inspection. The biceps anchor at initial arthroscopy in these types was believed to be stable. Two unstable type IV lesions and a complex type II/III lesion were treated with suture anchor repair, and all were found to have healed at second-look arthroscopy. Snyder’s work laid the foundation for surgical treatment of unstable superior labral lesions. Suture anchor repair appeared to yield the highest percentage of surgical success and was not associated with the need to remove surgical implants. For these reasons, attachment of the unstable biceps anchor using suture anchor techniques is preferred. As described earlier, Burkhart and others have made a compelling case for an entirely new pathomechanism for the creation of SLAP lesions in overhead throwers. Their theory describes a pathologic cascade of the disabled thrower’s shoulder, originating with a posteroinferior capsular contracture. It also pays credence to the era of the microinstability theory by explaining how anterior capsular laxity can occur within their model. This may explain the earlier successes in treating throwers’ shoulders based on the assumption of subtle anterior instability as the originating event. Their theory does not, however, explain all cases of superior labral tears. Many cases we encounter are not found in the overhead throwing athlete. It may well be that different mechanisms of injury may lead to different types of superior labral pathology. A fall on the outstretched arm has often been described as an inciting event.7,67 This sudden superior compression with or without rotation can lead to avulsion of the superior labrum–biceps tendon complex and may better describe the origin of type III and IV lesions. A traction injury such as a sudden downward or outward force has also been found in the clinical history. This mechanism may be another cause of type II lesions in the ­nonoverhead ­throwing athlete. Additionally, certain types of anatomic variants predispose the patient to SLAP lesion development. Two studies have found the association of the Buford complex with SLAP lesions 56% and 83% of the time, respectively.68,69 Normal anatomy versus pathologic anatomy versus associated anatomy should be carefully studied and understood. It is not always absolutely clear that a SLAP lesion is present.70 In other words, there appears to be a continuum

Shoulder 1029

between an absolutely normal superior labrum–biceps tendon insertion and a clear separation. The surgeon needs to consider normal variations and look for evidence of trauma.71 Fraying and tearing of the labrum, erythema and synovitis (Fig. 17I2-21) surrounding the biceps insertion, and chondral changes on the superior glenoid rim can all illustrate to the surgeon that there is a pathologic attachment of the superior labrum–biceps ­ tendon complex.

Figure 17I2-21   Arthroscopic photograph of the superior labrum with a type II SLAP (superior labrum, anterior to posterior) lesion. Note the area of synovitis posterosuperiorly, indicating evidence of trauma (arrow).

Authors’ Preferred Method Understanding the pathology when planning a repair of the superior labrum is paramount to achieving a successful result. As stated earlier, a thorough diagnostic arthroscopy performed through posterior and anterior portals and a methodical examination under anesthesia for shoulder instability greatly aid the surgeon in planning surgical repairs. When operating on a shoulder with superior labral pathology, débridement and repair of multiple structures frequently is necessary. A planned approach to addressing these multiple pathologies simplifies the surgery. Débridement of superior glenoid soft tissue, and cartilage, generally is performed through an anterior portal using a 4.0 mechanical full radius shaver blade (see Fig. 17I2-20). Limit the extent of labral débridement to preserve as much tissue as possible for repair. If further bony débridement is necessary, an arthroscopic rasp is preferred instead of an arthroscopic bur. Bony débridement facilitates labral healing to the superior glenoid, where little potential for spontaneous healing exists. A thermal instrument, such as the ArthroCare 90-degree suction wand (Sunnyvale, Calif) can be used to seal the margins of the injured tissues. The surgeon should avoid using thermal energy around a planned suture repair site to prevent potential weakening of the repair tissue. For superior labral repair, two arthroscopic portals are generally required. The first portal is created anteriorly through the rotator interval. A spinal needle can confirm the appropriate angle of approach to the anterosuperior glenoid rim before insertion of the arthroscopic cannula. A second portal is created just lateral to the acromial border to ­ facilitate repair of the posterosuperior labrum. The 18-gauge spinal needle is also good for determining the appropriate anteroposterior position and the appropriate angle of approach to the glenoid rim for the portal (Fig. 17I2-22). Generally, this portal is 1 cm anterior to the posterior lateral corner of the acromion. Most superior labral injuries in throwers are posterosuperior and require this more

Figure 17I2-22   Arthroscopic photograph of the superior labrum. The 18-gauge spinal needle is used to localize the posterolateral portal and angle to the glenoid rim. It is preferable to establish both portals before repair is begun.

­posterior portal placed through the rotator cuff. After establishing the correct position, a No. 11 blade is passed into the subacromial space in line with the fibers of the rotator cuff, staying as medial as possible within the tendon. The path of the blade should follow that of the 18-gauge needle. A 6-mm transparent cannula is directed into the shoulder joint. This method provides access to the superoposterior margin with minimal trauma to the rotator cuff tendon. The articular opening of the portal is through the muscle-tendon junction of the supraspinatus and should heal without the need for closure. The authors believe that the posterolateral trans–rotator cuff portal is important to achieve the correct angle of insertion of the posterior suture anchors against the glenoid rim. We agree with others71,72 that this portal does Continued

1030 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

Figure 17I2-23   Arthroscopic photograph of drilling and placement of the suture anchor through a single arthroscopic guide. Note the appropriate location up on the ridge of the glenoid and angle of approach.

not compromise the function of the rotator cuff. If desired, however, the ­ rotator cuff incision from the portal may be closed easily with a side-to-side repair using arthroscopic suture techniques during evaluation of the subacromial space. Next, repair of the superior labrum is performed from anterior to posterior. The number of suture anchors and their position needed for anatomic repair of the superior labrum are determined next. Current preference is for the 3-mm BioSuture Tak with a single No. 2 FiberWire suture (Arthrex, Naples, Fla). This anchor is inserted through a cannulated guide after predrilling (Fig. 17I2-23). The handle for anchor insertion has a laser mark indicating the alignment of this suture eyelet. Proper alignment of this suture anchor eyelet has been shown to be advantageous for tying tight suture loops. The eyelet should be positioned perpendicular to the glenoid rim to allow vertical mattress sutures to be placed across the circumferential fibers of the labrum. This positioning also facilitates repairing the labrum up onto the glenoid rim in its normal anatomic position. Two suture anchors are used routinely to repair the unstable biceps anchor. More anchors may be required, however, should the lesion extend further in the posterior direction. The suture anchors placed posterior to the biceps are introduced through the posterior lateral portal, and the suture anchor positioned anterior to the biceps is placed through the high anterior portal. The order in which anchors are placed probably is not important. We typically place anchors from anterior to posterior. The next step in the repair of the SLAP lesion is passing sutures through the detached labral tissue. Many devices can be helpful during this process. Our current favorite is the ­Arthro-Pierce 35-degree Upbiter (Smith & Nephew, Andover, Mass). It is introduced adjacent to the selected anchor, penetrating the labrum as medial as possible. The labrum is lifted off of the superior glenoid, and the instrument is passed laterally between the labrum and superior glenoid. A probe through the alternate portal facilitates capture of the suture in the suture passer (Fig. 17I2-24). The suture

Figure 17I2-24   Arthroscopic photograph of passage of the anterior anchor’s suture through the anterosuperior labrum just in front of the biceps root. Note use of the “bird beak” suture passer to grasp the suture. A probe through the other portal eases capture of the suture. The suture passer is removed and pulls one limb of the suture anchor through the labrum.

can then be tied using arthroscopic knot-tying techniques (Fig. 17I2-25). The process is repeated for each successive suture anchor. It is preferable to tie knots as soon as suture is passed for each anchor (Fig. 17I2-26). Several principles are important regarding suture handling and knot tying during superior labral repair. The sutures to be tied should exit the cannula through which their attached anchor was placed; this allows the suture to slide easily through the labrum and anchor eyelet. The anchor eyelet must be aligned with the direction of the suture ­passing through the labrum. The suture must pass through the labrum, through the eyelet, and back through the labrum without twisting or crossing. A knot pusher must be used to tighten the loop before locking the knot.

Figure 17I2-25   Arthroscopic photograph of a knot pusher securing the anterior knot. The same sequence of steps is then performed for the posterior anchors.

Shoulder 1031

Authors’ Preferred Method—cont’d One type of arthroscopic knot will not work in all situations. There are two types of knots, sliding and nonsliding. Sliding knots can be divided into groups that lock themselves and those that must be locked. Nonsliding knots (square knot, Revo knot) are the most difficult to produce tight loops and suture knots. When the suture does not slide well, however, these techniques are essential. Sliding knots (e.g., Tennessee slider, Duncan loop, and Roeder knot) and sliding-locking knots (Westin knot, giant knot) produce tighter loops but must be secured with three reversed half-hitch throws on alternating posts. Current preference is for the Westin knot, which is easy to tie and reliably produces tight loops and secure knots with a low profile. The ability to tie secure arthroscopic knots is important to the successful completion of these superior labral repairs. The surgeon performing arthroscopic repairs must be aware of the differences in knot strength and security of different suture materials and knot patterns. Familiarity with one sliding knot and one nonsliding knot is essential to handle any clinical situation.

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Postoperative Prescription The postoperative program after SLAP repair is delineated in Table 17I2-3.73 This program may need to be modified based on any comorbid pathology repaired along with the SLAP lesion. Those patients who undergo a thermal capsulorrhaphy along with the superior labral repair typically have their motion restricted for the first 3 to 4 months. Throwers with posterior capsular contracture severe enough to require an arthroscopic release along with the SLAP repair require immediate postoperative posteroinferior capsular stretching (Figs. 17I2-27 and 17I2-28).

Figure 17I2-26   Arthroscopic photograph of a completed SLAP (superior labrum, anterior to posterior) repair. Note in this case that two anchors were placed posterior to the root of the biceps tendon to secure the superior labrum–biceps tendon complex to the superior glenoid.

Outcomes Measurement The outcome of nonoperative management of SLAP lesions is not well known. The definitive diagnosis of the superior labral–biceps tendon disruption is often not made by examination and imaging criteria alone. In cases in which the SLAP lesion is identified arthroscopically, few studies have documented the pathologic anatomy and not attempted surgical repair. Additionally, significant associated pathology commonly seen with superior labral lesions makes the diagnosis of a truly isolated SLAP lesion ­difficult.74,75

  TABLE 17I2-3  Postoperative Prescription Postoperative Time

Prescription

1 wk

Sling immobilization: elbow range of motion Passive range of motion, circumductions, external/internal rotation at 0 degrees of abduction, sling otherwise Progressive passive range of motion, start posteroinferior capsular strengthening Start rotator cuff and scapular rotator strengthening, no biceps work until 8-12 wk Start interval throwing program

2-3 wk 3-6 wk 6-16 wk 4 mo

Figure 17I2-27   The patient is positioned supine with the shoulder abducted 90 degrees and elbow flexed 90 degrees. The scapula is held against the table to concentrate the stretch on the posterior capsule and posterior cuff.

1032 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

CRITERIA FOR RETURN TO PLAY

Figure 17I2-28   The patient is positioned supine with the shoulder abducted 90 degrees and the elbow extended. The patient stretches into horizontal adduction. This exercise stretches the posterior cuff and scapulothoracic musculature.

Most authors recommend operative treatment of superior labrum–biceps tendon pathology.70,73,75,76 Studies have shown good results of operative repair, both using bioabsorbable tacks58,59,60 and suture anchors.31,75,77 These studies typically report the results of repair of type II SLAP lesions because the less common types are not reported with enough patient numbers. Burkhart and associates73 reported an 87% return to preinjury status in 53 baseball players (44 pitchers) with type II SLAP lesions. However, others found that only 16 of 31 patients returned to their preinjury level of sports after repair.71 Kim and colleagues75 evaluated their repair of type II SLAP lesions and found 94% good to excellent results with suture anchor repair. Again, results were less satisfactory with a lower return to preinjury sports levels in overhead throwing athletes. Results of repair of SLAP lesions in biomechanical studies have also been studied. Panossian and coworkers78 reported that the increased glenohumeral translocations and rotation found on biomechanical testing after creation of SLAP lesions were normalized after arthroscopic repair. Park and others,79 however, found that repair of type II SLAP lesions did not restore anterior and inferior translation values to control values in the vented joint.

Potential Complications Appropriate repair of pathologic anatomy is important, but care must be taken not to repair normal labral variations. If this is done, loss of shoulder motion and function may occur. Most shoulder surgeons now use suture anchor techniques to repair superior labral tears instead of bioabsorbable tacks. Although studies show good outcomes with their use, the level of complications due to breakage or dislodgment of the tack is too high.61 Some types of tacks can lead to a significant foreign body reaction within the shoulder joint as well.62

Return to play after the diagnosis of a SLAP lesion is not well delineated in the literature. With conservative management of rest, rehabilitation, and anti-inflammatory drugs, a graduated return to sport can occur as tolerated. Patients should be monitored for return of symptoms or compensatory mechanics. After surgical repair, a reasonable amount of time should be given for healing and rehabilitation of rotator cuff and scapular rotator muscles. Typically, 6 weeks of limitations of daily activities followed by 6 weeks of concentrated rehabilitation are recommended before an initial return to sports. This early return depends on any associated pathology found at the time of SLAP repair, especially injury to the rotator cuff. More severe cuff injuries require more prolonged rehabilitation. In the absence of severe cuff pathology, a throwing program can begin after 3 to 4 months, with a projected return to full throwing after 9 to 12 months.79 C

r i t i c a l

P

o i n t s

l SLAP lesions typically cause pain but can mimic other shoulder diagnoses. l Torsional peel-back is the probable mechanism of injury in the throwing athlete. l Gadilinium MRI is the best diagnostic study. l SLAP lesions are often found with associated pasthology. l Surgical repair is recommended for those patients in whom conservative management fails. l Repair of all associated pathology should be performed along with the SLAP repair.

S U G G E S T E D

R E A D I N G S

Burkhart SS, Morgan CD, Kibler WB: Current concepts. The disabled throwing shoulder: Spectrum of pathology part I: Pathoanatomy and biomechanics. ­Arthroscopy 19:404-420, 2003. Burkhart SS, Morgan CD, Kibler WB: Current concepts. The disabled throwing shoulder: Spectrum of pathology. Part II: Evaluation and treatment of SLAP ­lesions in throwers. Arthroscopy 19:531-539, 2003. Kim SH, Ha KI, Kim SH, et al: Results of arthroscopic treatment of superior labral lesions. J Bone Joint Surg Am 84:981-985, 2002. Maffet MW, Gartsman GM, Moseley B: Superior labrum-biceps tendon complex lesions of the shoulder. Am J Sports Med 23:93-98, 1995. Morgan CD, Burkhart SS, Palmeri M, Gillespie M: Type II SLAP lesions: Three subtypes and their relationships to superior instability and rotator cuff tears. ­Arthroscopy 14:553-565, 1998. Pagnani MJ, Deng X-H, Warren RF, et al: Effect of lesions of the superior portion of the glenoid labrum on glenohumeral translation. J Bone Joint Surg Am 77:1003-1010, 1995. Snyder SJ, Karzel RP, Del Pizzo W, et al: SLAP lesions of the shoulder. Arthroscopy 6:274-279, 1990. Tennent TD, Beach WR, Meyers JF: A review of the special tests associated with shoulder examination. Part II: Laxity, instability, and superior labral anterior and posterior (SLAP) lesions. Clin Sports Med 31:301-307, 2003. Tuoheti Y, Itoi E, Minagawa H, et al: Attachment types of the long head of the biceps tendon to the glenoid labrum and their relationships with the glenohumeral ligaments. Arthroscopy 21:1242-1249, 2005. Wilk KE, Reinold MM, Dugas JR: Current concepts in the recognition and treatment of superior labral (SLAP) lesions. J Ortho Sports Phys Ther 35:273-291, 2005.

R eferences Please see www.expertconsult.com

Shoulder 1033

S E C T I O N

J

Injuries of the Proximal Humerus 1. Injuries of the Proximal Humerus in Adults W. Anthony Frisella, Dan Guttman, Chang-Hyuk Choi, and Frances Cuomo

PROXIMAL HUMERUS FRACTURES Fractures of the proximal humerus are an uncommon sports injury,1-3 representing 4% to 5% of all fractures.4,5 They are most common in the early adolescent patient who has open physes and in the older, osteoporotic patient. In the sporting young and middle-aged patient, the bony structures of the proximal humerus are less vulnerable to injury than the soft tissue structures around the shoulder. When proximal humeral fractures do occur in the athlete, they are usually the result of high energy or some insidious pathology.2,6,7 Most proximal humerus fractures are minimally displaced and can be treated nonoperatively. In athletes, however, to allow rapid return to competitive sport, rigid fixation with anatomic reconstruction is often the goal of treatment. In the significantly displaced fracture, a reproducible classification system is essential to make the correct diagnosis and to choose appropriate treatment options. Minimal internal fixation techniques have proved in recent years to be of great benefit in efforts to maximize results and minimize complications. For severely comminuted fractures and fracture-dislocations in the elderly patient, the best surgical option is often primary humeral head replacement. The use of arthroplasty is not indicated in young athletic patients. The surgical techniques and designs for prostheses have evolved since first reported by Neer in 1955. Despite the impressive developments in surgical techniques and implants, controversies still exist when choosing treatment methods of severely displaced fractures. Also as important is an early and aggressive rehabilitation program designed to prevent residual stiffness and dysfunction, which can compromise treatment of these fractures.8,9

tuberosity, the pull of the supraspinatus, infraspinatus, and teres minor can displace the greater tuberosity fragment both posteriorly and superiorly.11 Similarly, fractures of the lesser tuberosity are displaced by the attached subscapularis muscle anteriorly and medially. The long head of the biceps sends its tendon through the intertubercular groove, which lies between the greater and lesser tuberosities, to enter the glenohumeral joint and attach to the superior pole of the glenoid.10 The long head of the biceps tendon can act as a tether in closed reduction, although intraoperatively it is a useful landmark in identifying the rotator interval and relative position of the tuberosity fragments.

Bony Anatomy

Blood Supply

The proximal humerus consists of four major bony components: the humeral head, the lesser tuberosity, the greater tuberosity, and the humeral shaft (Fig. 17J1-1). In fractures, the displacement of tuberosity fragments and shaft is dependent on the direction of pull of muscle fibers attached to the tuberosities and the proximal shaft (Table 17J1-1). The subscapularis muscle attaches to the lesser tuberosity and functions as a strong internal rotator of the humerus. The supraspinatus, infraspinatus, and teres minor tendons attach to the greater tuberosity and function in abduction and external rotation.10 In fractures involving the greater

The major blood supply to the humeral head is through the intraosseous vessels that cross the anatomic neck from the metaphysis.12 Proximal humerus fractures depend on the presence of an adequate blood supply to ensure union. Laing12 and Gerber and colleagues13 have reported that the anterolateral branch of the anterior humeral circumflex artery provides the main blood supply to the humeral head. At the superolateral aspect of the bicipital groove, it becomes the arcuate artery within the humeral head. The arcuate artery anastomoses with posteromedial branches of the posterior humeral circumflex artery and is ­responsible

Humeral head Greater tuberosity Lesser tuberosity

Shaft

Figure 17J1-1   The bony structure of the proximal humerus and its relationship to the scapula. Note the four major portions of the proximal humerus: (1) head, (2) greater tuberosity,   (3) lesser tuberosity, and (4) shaft.

1034 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 17J1-1  Proximal Humerus Bony Anatomy Description

Muscle Attachments

Relevant Fracture Anatomy

Humeral head

Articular portion of proximal humerus, retroverted 30 to 35 degrees from shaft

None

Humeral shaft

Portion of humerus inferior to tuberosities

Greater tuberosity

Superior and lateral bony fragment, separated from lesser tuberosity by bicipital groove Medial and inferior bone fragment separated from greater tuberosity by bicipital groove Between tuberosities. Contains long head of the biceps tendon Between articular cartilage and attachment of rotator cuff Inferior to attachment of rotator cuff

Proximally: deltoid, pectoralis major, latissimus dorsi, teres major Supraspinatus, infraspinatus, and teres minor

Fractures that involve the articular surface (head-splitting) have a poor prognosis and high chance of avascular necrosis. In fracture, this fragment is displaced medially by pull of pectoralis major.

Lesser tuberosity Bicipital groove Anatomic neck Surgical neck Acromion Clavicle

Extension of the scapular spine; forms part of the coracoacromial arch Articulates with acromion at acromionclavicular joint

Subscapularis None

None Trapezius superiorly, deltoid inferiorly

Muscle Anatomy The muscular anatomy of the shoulder is summarized in Table 17J1-2. There are two layers of muscle overlying the proximal humerus and glenoid. The outer layer consists of the deltoid, pectoralis, teres major, and latissimus muscles, and the deep layer is composed of the four muscles of the rotator cuff. The direction of pull of various muscles can influence the displacement of proximal humerus fractures (Fig. 17J1-3).

Nerve Supply The brachial plexus lies anterior to the scapula, passing below the coracoid to enter the upper arm (Fig. 17J1-4). The axillary nerve arises from the posterior cord and courses first inferiorly along the anterior surface of the subscapularis muscle belly and then below the glenohumeral joint capsule to innervate the deltoid and teres minor (Box 17J1-2). The musculocutaneous nerve arises from the lateral cord, penetrating the coracobrachialis muscle as near as 2 cm distal to the coracoid process.14 The brachial plexus therefore is tethered in its position anteromedial to the proximal humerus and is vulnerable to injury in

Box 17J1-1 Vascular Supply to the Humeral Head Vascular supply is from the anterior humeral circumflex artery, anterior branch, which becomes the arcuate artery and runs in the bicipital groove. It is often disrupted in fracture patterns that disrupt the medial calcar region of the humeral neck (see Fig. 17J1-2).

The biceps tendon and groove is the landmark between the greater and lesser tuberosities. Rarely fractured in isolation

None

for supplying the greater tuberosity (Box 17J1-1; Fig. 17J1-2). Injury to this artery with significant displacement of fracture fragments may result in avascular necrosis.

Fragment is pulled superiorly and posteriorly by supraspinatus, infraspinatus, and teres minor. Fragment is pulled medially by subscapularis.

Commonly fractured, with or without associated tuberosity fractures

displaced fractures. Electrophysiologic evidence of nerve injury is found in up to 45% of humeral neck fractures and primary dislocations, most commonly involving the axillary nerve, followed by the suprascapular, radial, and musculocutaneous nerves. Older patients and those with hematomas have more neurologic injuries. Most patients with low-energy injuries recover partially or completely in less than 4 months. In patients with permanent motor loss from brachial plexus injuries, treatment is focused on preservation of hand function.15

Biomechanics The shoulder is the most mobile major joint in the body. The articular surface of the humeral head is 2 to 3 times larger than the surface area of the glenoid. The range of motion of the glenohumeral joint is 2 times that of scapulothoracic motion, and their combined motion ­approximates

Ascending arcuate branch of anterior humeral circumflex a. Anterior humeral circumflex a. Posterior humeral circumflex a. Axillary a. Subscapular a. Figure 17J1-2   Vascular supply to the humeral head. The anterior humeral circumflex artery enters the humerus, via the ascending arcuate branch, at the intertubercular groove, providing the major source of proximal humeral vascularity.

Shoulder 1035

  TABLE 17J1-2  Muscle Anatomy Attachment

Insertion

Rotator cuff Subscapularis

Scapula Subscapularis fossa

Humeral tuberosities Lesser tuberosity

Supraspinatus

Supraspinatus fossa

Greater tuberosity

Infraspinatus Teres minor Pectoralis major

Infraspinatus fossa Inferior scapula fossa Clavicle, sternum, upper ribs

Deltoid

Lateral one third of clavicle, acromion, spine of scapula Spine, T7 and below

Greater tuberosity Greater tuberosity Humeral shaft, lateral to bicipital groove Humeral shaft deltoid tuberosity Medial humerus

Long thoracic nerve

Medial humerus

Lower subscapular nerve

Latissimus dorsi Teres major

Medial, inferior border of scapula

180 degrees of elevation. The humeral neck-shaft angle is about 45 degrees, and the head is 30 to 40 degrees retroverted relative to the epicondyles at the elbow.16 The synergy between the stabilizing effect of the rotator cuff combined with the power of the deltoid provides normal dynamic shoulder function. The deltoid is the primary motor source for the shoulder but also creates shear stress across the joint. The rotator cuff and biceps provide stability by counterbalancing the humeral head against the deltoid shear. As internal rotation and external rotation components are added, the rotator cuff muscles

Rotator cuff mm.

Subscapularis m. Deltoid m.

Pectoralis major m.

Nerve Supply Upper and lower subscapular nerve Suprascapular nerve Suprascapular nerve Axillary nerve Medial and lateral pectoral nerves Axillary nerve

Function Stabilize the glenohumeral joint Humeral internal rotation Humeral abduction, stabilization of humeral head against deltoid External rotation External rotation Adduction, internal rotation Flexion and abduction of humerus Adduction and extension of humerus Adduction and extension of humerus

­ rovide not only humeral head depression but also stability p against excessive anterior and posterior translation within the joint. Fractures of the proximal humerus that disrupt this fine anatomic balance can alter the biomechanics of the shoulder and therefore lead to limitation of motion and function. Restoration of this anatomy, along with appropriate muscle strength and coordination, is necessary to return the shoulder to normal function.16,17

Incidence and Causes of Proximal Humeral Fractures Proximal humeral fractures are more common in females than in males by a ratio of 2:1.18 This difference is largely attributable to osteoporosis of the humerus in elderly patients and does not necessarily apply to athletes. The injury usually is sustained after a simple fall onto the outstretched limb or onto the shoulder itself. In males, highenergy trauma is more common, with a higher incidence of associated dislocations. Sports-related fractures tend to be the result of high-energy impact or avulsion-type injuries, often associated with dislocations.

Associated Injuries Soft tissue pathology is often associated with fractures of the proximal humerus. Schai and coworkers performed arthroscopic examination in a series of 80 fractures around the shoulder.19 Significant numbers of labral, capsuloligamentous, and rotator cuff lesions, as well as cartilage damage, were reported. Arthroscopic assessment in shoulder fractures has recently been found to be a useful tool in better understanding the extent of the injury.

Classification

Figure 17J1-3   The muscular attachments and the direction of their pull (arrows) can influence displacement in proximal humeral fractures. These forces must be taken into account when one is attempting to reduce a fracture.

The most commonly adopted classification is the four­segment classification described by Neer in 1970.20 The AO/ASIF system, described in 1990 by Müller and based on the work of Jakob, is much less frequently employed.21,22 The Neer classification (Fig. 17J1-5) is based on the concept of four fragments of the proximal humerus as first proposed by Codman in 1934.23 When any of the four major

1036 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 17J1-4   The relationship of the proximal humerus to the brachial plexus and the axillary nerve.

Axillary n. and posterior humeral circumflex a.

Thoracoacromial a. Axillary a.

Anterior humeral circumflex a.

Lateral thoracic a.

Ulnar n. Median n.

fragments is displaced greater than 1 cm, or angulated more than 45 degrees, the fracture is considered displaced. Sidor and associates reported fair interobserver and intra­ observer agreement with the Neer classification.24 The Neer classification is the most widely used classification system because of its simplicity and rationale for surgical management (Table 17J1-3).

Clinical Evaluation History and Mechanism of Injury The patient with a proximal humeral fracture typically describes specific trauma and can often outline the mechanism of injury (Box 17J1-3).1,2,25-27 Pain, swelling, and inability to use the shoulder are seen immediately. The most common mechanism of fracture overall is a fall on an outstretched arm, but it can occur in both contact and noncontact sports, especially in association with a dislocation. In the older patient with osteoporotic bone, fracture of the proximal humerus may occur without significant trauma. In the younger patient, however, significant high-energy trauma is necessary, and the resultant fracture is often more serious. These younger patients commonly have displaced fractures or fracture-dislocations with substantial

BOX 17J1-2 Nerve Supply The axillary nerve courses along the subscapularis and dives posteriorly underneath the glenohumeral joint capsule. It then courses laterally on the deep surface of the deltoid. It is vulnerable to injury in fracture and trauma and is the most commonly injured nerve in proximal humerus ­fractures. A detailed examination of deltoid muscle function and sensation to the lateral upper arm is essential in the proximal humerus fracture to test the axillary nerve function (see Fig. 17J1-6).

soft tissue disruption. Neurologic or vascular injury can also occur and is related to the gravity of the soft tissue component of the injury. An additional mechanism of fracture of the proximal humerus occurs with excessive external rotation of the arm, especially in an abducted position in association with a dislocation. This mechanism is a much more common cause of fracture in the younger, athletic population. Fractures associated with primary dislocation of the shoulder are usually seen with forced abduction and external rotation (for anterior fracture-dislocations) and forced adduction with posterior displacement (for posterior fracture-­dislocations).16 A direct blow to the lateral arm can also cause injury to the proximal humerus but is less common. This mechanism usually results in a fracture of the greater tuberosity, a minimally displaced fracture, or a fracture involving the articular surface.

Physical Examination Physical examination reveals in most cases swelling and tenderness to palpation around the shoulder.28-30 Disruption of the normal bony contour of the shoulder is seen in displaced fractures and fracture-dislocations. Crepitus may be present if there is significant displacement of the fracture fragments. Ecchymosis occurs over 2 or 3 days, and discoloration may occur in the arm extending past the elbow and along the chest wall and upper back. These patients often hold the arm adducted. It is difficult for the patient to initiate active motion, which usually results in painful muscle spasm. A detailed neurovascular examination is essential. Brachial plexus and axillary artery injuries have been associated with proximal humerus fractures.31-34 The axillary nerve is the most commonly injured, but the entire plexus can be involved. The sensory distribution of the axillary nerve over the lateral upper arm should be examined for light touch and pin-prick. The sensory loss in the distribution of the axillary nerve may be the only clear evidence of injury to the axillary nerve because deltoid function will be difficult to assess in the setting of fracture. If possible,

Shoulder 1037 Displaced Fractures 2-part

3-part

4-part

Articular Surface

Anatomic Neck

Surgical Neck

A

Figure 17J1-5   Neer four-segment classification of displaced fractures. A fragment is considered displaced when greater than 1 cm of displacement or 45 degrees of angulation is present. A fracture-dislocation is present if the articular segment is no longer in contact with the glenoid. (From Neer CS: Displaced proximal humeral fractures. Part I. Classification and evaluation. J Bone Joint Surg Am 52:1077-1089, 1970.)

C B

Greater Tuberosity

Lesser Tuberosity

Anterior FractureDislocation Posterior

HeadSplitting

deltoid motor activity can be evaluated by palpating the muscle belly with one hand while supporting the elbow to resist abduction or extension with the other (Fig. 17J1-6). It is difficult to test the suprascapular nerve in isolation. The suprascapular nerve innervates the supraspinatus and infraspinatus muscles. The supraspinatus and the deltoid are both primary shoulder abductors, whereas the infraspinatus and teres minor are involved primarily in external rotation. While testing the relative strength in abduction and external rotation, there will subsequently be some degree of overlap. Also, the presence of fracture will make these tests difficult.

Imaging Accurate radiographic evaluation is essential for diagnosis and treatment of proximal humerus fractures. The mandatory trauma series consists of anteroposterior and lateral views in the scapular plane and an axillary view. A true anteroposterior view in the scapular plane is performed with the affected shoulder placed against the x-ray cassette and the body rotated toward that side about 40 degrees (Fig. 17J1-7). The true lateral view in the scapular plane is also termed the tangential view, Y view, or scapular lateral view (Fig. 17J1-8). This can best be accomplished by

1038 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 17J1-3  Treatment Options by Neer Classification Fracture

Box 17J1-3 Typical Findings in Proximal ­Humerus Fracture

Treatment Options

Two-Part Fracture

Surgical neck

Greater tuberosity

Lesser tuberosity Three-Part Fracture

Greater tuberosity and neck Lesser tuberosity and neck Four-Part Fracture

Young patient, good bone quality Older patient (>60 yr), osteoporosis

ORIF with locking plate,* ORIF with tension band suture with or without Ender nail, percutaneous pinning, ORIF with intramedullary nail ORIF with tension band suture fixation incorporating rotator cuff,* percutaneous screw fixation if acceptable bone quality ORIF with suture fixation incorporating rotator cuff,* percutaneous screw fixation ORIF with locking plate, ORIF with tension band suture with or without Ender nails,* percutaneous pinning ORIF with locking plate, ORIF with tension band suture with or without Ender nails* ORIF with locking plate, ORIF with tension band suture,* percutaneous reduction and fixation Humeral head replacement

Special Cases

Valgus-impacted, four-part ORIF with locking plate, ORIF with fracture tension band suture, percutaneous reduction and pinning* Head-splitting fracture For physiologically elderly patient: humeral head replacement For physiologically younger patient: ORIF with preferred technique, accept high risk for avascular necrosis *Authors’ preferred technique. ORIF, open reduction with internal fixation.

A

History: High-energy injury to the shoulder in the younger patient, often associated with dislocation. In an older patient, lower energy trauma is the rule (fall). Physical examination: Shoulder pain and ecchymoses. Rule out neurovascular injury, especially to the axillary nerve (most common). Radiographs: Anteroposterior, lateral, and axillary radiographs are absolutely required for accurate diagnosis. Classify the fracture by number of displaced fragments.

­ rawing a line along the scapular spine to help align the d x-ray tube with the spine. The axillary view allows evaluation in the axial plane (Fig. 17J1-9). The supine position is preferable but not mandatory. Alternatively, a Velpeau axillary lateral view may be taken in the patient with an injury that prevents abduction of the arm, allowing the patient to keep the arm immobilized in a sling. On this view, the humerus appears foreshortened, and the glenohumeral joint is magnified but still demonstrates the relationship of the humeral head to the glenoid (Fig. 17J1-10).25 The axillary view often provides the best information about the relationship of the humeral head and glenoid fossa (anterior or posterior dislocation versus rotary subluxation of the head). In addition, posterior displacement of the greater tuberosity and medial displacement of the lesser tuberosity are best delineated in this projection.35 Computed tomography (CT) is extremely helpful in evaluating the amount of articular involvement with headsplitting fractures, impression fractures, chronic fracturedislocations, and glenoid rim fractures. CT may also aid in judging the amount of displacement of tuberosity fragments.36,37 Sjoden and colleagues investigated whether the addition of three-dimensional reconstruction would increase the reproducibility of the Neer and AO fracture

B

Figure 17J1-6   A, Clinical examination of axillary nerve motor distribution. The deltoid muscle is palpated while an attempt to abduct the arm is made. B, This patient had an injury to the axillary nerve and deltoid atrophy. The lined area on the upper lateral arm depicts the region of decreased cutaneous sensation.

Shoulder 1039

classifications.38 Seven observers independently classified 24 fractures of the proximal humerus using both plain radiographs and three-dimensional CT reconstructions. Moderate interobserver agreement was noted with Neer’s four-segment classification, but only fair agreement with the AO classification. Intraobserver reproducibility was fair for both the Neer and AO classifications. The authors concluded that the addition of CT and three-dimensional reconstructions did not improve reproducibility when classifying these injuries.38

Treatment Historically, many different types of treatment have been described for proximal humerus fractures, including a variety of external immobilization devices such as spica and hanging arm casts, traction, and splints as well as internal fixation devices such as percutaneous pinning, plates and screws, rods, staples, and the humeral head prosthesis. In the competitive athlete, a rigorous effort should be made to restore anatomy and obtain rigid fixation. Even in the

True AP (45˚ lateral) Patient can be sitting, standing, or lying down. 45˚

A

B

ROUTINE

A-P

SHOULDER

Ant. Glenoid rim

Post. Glenoid rim

TRUE

A-P

SHOULDER

Ant. & Post. Glenoid rims superimposed

A

45˚

B

C Figure 17J1-7   A, The technique for taking a true anteroposterior radiograph of the shoulder is demonstrated. The affected shoulder is rotated about 45 degrees toward the cassette. This eliminates the bony overlap between the humeral head and the glenoid. B, The technique for a true anteroposterior radiograph in the scapular plane. Note the angle of the beam to the body is 45 degrees.   C, Here, the difference between an anteroposterior (A) and true anteroposterior in the scapular plane (B) is demonstrated.

1040 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

D

E

Figure 17J1-7����������� , cont’d���   D and E, The resulting films. Note the clear view of the joint space in E. (A and B, From Rockwood CA, Matsen DD [eds]: The Shoulder. Philadelphia, WB Saunders, 1990; C-E, from Rockwood CA, Green DP [eds]: Fractures, 2nd ed. Philadelphia, JB Lippincott, 1984.)

significant four-part, displaced, or head-splitting fracture, arthroplasty should be avoided in the young, active patient. Even with a significant risk for avascular necrosis, it is better to attempt reconstruction then to resort to arthroplasty in this patient population. In the older patient, however, arthroplasty remains a viable, and often better, option. Many factors contribute to the decision-making process of fracture treatment (Box 17J1-4). There is an increasing trend toward locked plated fixation or percutaneous reduction and fixation techniques. Conventional plates are being replaced by locking plates and applied with less aggressive soft tissue stripping. Some controversy remains regarding the management of four-part fractures and fracture­dislocations.39,40 Correlation of treatment modality and functional outcome remains difficult, as seen in a prospective multicenter study by Weber and Matter, in which confusion regarding classification and numerous techniques of fixation made statistical analysis impossible.41

Nonoperative Treatment Most proximal humerus fractures are minimally displaced and can be treated with sling and early passive range of motion exercises. Koval and colleagues reported outcomes at an average of 41 months in 104 patients treated for onepart fractures.42 Results were excellent in 77% of patients, fair in 13%, and poor in 10%. The authors reported that age greater than 70 years and a delay of passive range of motion exercises (≥14 days) had detrimental effects on range of motion outcome. Bjorkenheim and associates reported on 27 patients with minimally displaced proximal humerus fractures treated nonoperatively.43 Good results were reported, with an average Constant score of 84. Gaebler and coworkers showed that, for minimally displaced fractures, good or excellent results were obtained in 88% of patients in a very large series of 507 patients.44 Age was the primary determinant of outcome. Zyto reported the

results of nonoperative treatment of three- and four-part fractures.45 Seventeen shoulders with a minimum followup of 10 years were included. The mean Constant score in the three-part fracture group was 59, and 47 in the four-part fractures. Despite low functional scoring, poor fracture reduction, and mean flexion and abduction of 90 degrees, the author suggested that nonoperative treatment should be considered.

Operative Management Locking Plates Recently, locking plate technology has become widely applied to fracture care, especially for periarticular fractures (Fig. 17J1-11). The use of locking plates for fractures of the proximal humerus has greatly expanded in recent years. Locked plates have been demonstrated to have superior biomechanical characteristics compared with other methods of fixation, with greater load to failure when compared with traditional plates,46 greater resistance to torsional loading compared with blade plates,47,48 and greater resistance to bending and torsional loading when compared with locked nail constructs.49 Thus, the biomechanical characteristics of the locking plate make it an attractive choice for rigid anatomic fixation of proximal humerus fractures. Theoretical disadvantages of locking plates are similar to those for traditional plates: the need for ­dissection to place the plate, potentially compromising vascular supply to the humeral head; the placement of hardware in the subdeltoid space with the potential for adhesions; and the possibility of hardware impingement. As early as 1998, Plecko and Kraus described their results with a locking plate implant in two-, three-, and four-part fractures as well as surgical neck nonunions.50 Results were satisfactory in 75% in this diverse group of patients at 12-month minimal follow-up. Fankhauser and colleagues reported a series

Shoulder 1041

A

C B Figure 17J1-8   A, The technique for true scapular lateral radiography of the shoulder is demonstrated. The cassette is placed lateral to the affected shoulder, and the x-ray beam is directed along the scapular spine. B, The anatomy of the scapular lateral view.   C, The resulting image. Note that the body of the scapula is in profile. (From Rockwood CA, Green DP [eds]: Fractures, 2nd ed. Philadelphia, JB Lippincott, 1984.)

of 29 proximal humerus fractures treated with a locking plate specifically designed for the proximal humerus.51 The series included AO types A, B, and C fractures, and good to excellent outcomes were achieved in most patients, with an average 1-year postoperative Constant score of 75. There were some complications, the most common being loss of reduction, but there were no nonunions. The study by Fankhauser and colleagues showed improved outcome for locking plates for three-part fractures compared with a variety of other methods; for other fracture types (besides three-part), the locking plate was at least equivalent to other methods of fixation.52 Other authors have also demonstrated acceptable outcomes using this implant type, with Constant scores in the range of 70 to 80 and acceptable to excellent results in most patients, with low

rates of nonunion. Results deteriorated with more severe injury and the presence of avascular necrosis.43,53-55 A more recent study by Rose and colleagues showed a relatively high nonunion rate of 25%, which they attributed to comminution and cigarette smoking.56

Percutaneous Reduction and Pinning Percutaneous pinning may be indicated after closed reduction if the reduction is deemed to be unstable. It is technically demanding but offers the advantage of less disruption of soft tissue and minimal hardware, theoretically reducing the prevalence of avascular necrosis. Williams recommended that the most suitable indication for closed reduction and percutaneous pinning was in select two-part

1042 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Axillary Lateral arm abduction

90˚

A Figure 17J1-9   The technique for taking an axillary lateral radiograph of the shoulder is demonstrated. The beam is angled about 20 degrees off the horizontal and vertical axes toward the cassette, which is positioned above the shoulder. (From Rockwood CA, Matsen DD [eds]: The Shoulder. Philadelphia, WB Saunders, 1990.)

surgical neck fractures and valgus-impacted fractures with good bone quality and minimal comminution. This technique is ideally suited to a young, athletic population. It minimizes soft tissue disruption and thus scarring. Once fracture healing has begun, the pins are removed, leaving

no hardware to irritate the soft tissue. This technique has also been reported with good results in three- and fourpart fractures of younger patients without osteoporosis (Fig. 17J1-12).57 Biomechanically, multiplanar pins are required to augment torsional stiffness, and it has been shown that the addition of two bicortical tuberosity pins enhances bending rigidity.58 Jaberg reported satisfactory results in 70% of patients (34 of 48) who underwent percutaneous pinning, including 32 two-part fractures.59 In another study, the results of 31 percutaneous pinnings in patients averaging 68 years of age, including 7 two-part, 20 three-part, and 4 four-part fractures, were studied.60 The mean Constant score was 80, with avascular necrosis seen in 5 cases. The authors concluded that transitory percutaneous pinning was a reasonable technique for the management of displaced three-part fractures even in an older population but was not satisfactory for the management of four-part fractures. Some modification of percutaneous pinning was attempted by several authors.61 Resch and associates also reported satisfactory results with percutaneous pin fixation of three- and four-part fractures.62 Avascular necrosis was seen in only 11% of four-part fractures. Bungaro and coworkers discussed preliminary results of osteosynthesis with percutaneous wiring. They found the use of a traction device applied to the surgical table simplified reduction maneuvers and osteosynthesis.63

Traditional Plate Fixation Traditional plating has been used for all types of proximal humeral fracture, although concern exists regarding poor fixation in osteoporotic bone. Hessmann and colleagues evaluated the use of indirect reduction and internal fixation using a buttress plate in 99 cases of two-, three-, or four-part fractures of the proximal humerus.64 The mean

Figure 17J1-10   A, Velpeau axillary view may be taken; this allows the patient to keep the arm immobilized in the sling. The patient leans back and over the plate while the beam is directed from superior to inferior. B, Velpeau axillary radiograph showing proximal humerus fracture. Note that the head is located and the greater tuberosity is displaced posteriorly. (From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 143.)

A

B

Shoulder 1043

Box 17J1-4 T  reatment Options for Proximal Humerus Fracture

traditional plating has largely been replaced by fixed-angle locking plates.

Nonoperative: Generally indicated for nondisplaced fractures of any type. Range of motion is begun once the fracture moves as a unit with the head during the clinical examination. Earlier physical therapy achieves better outcome. Operative: There are many operative techniques, and no one technique has been demonstrated to be superior. In general, anatomic reduction with rigid fixation is the goal. Rigid fixation allows early motion, maximizing functional outcome. Percutaneous reduction and pinning: Minimally invasive. Indicated for displaced surgical neck fractures and selected three-part fractures in younger patients without osteoporosis. Valgus-impacted four-part fractures are also amenable to this technique. Technically demanding. Locking plate fixation: Indicated for two-, three-, and some four-part fractures. Locked plates have better fixation in osteoporotic bone. The addition of suture constructs incorporating the rotator cuff allows for ­better fixation and incorporation of the lesser tuberosity into the construct. Suture and wire tension band constructs: These have been reported with good outcomes for all fracture types. Their use does not rely on bone purchase, but rather the tendinous cuff; therefore, they may be more useful in the elderly patient with poor bone quality. The addition of intramedullary Ender nails through the cuff adds longitudinal stability (see Authors’ Preferred Method for three-part fractures).

Wire or Suture Tension Band

age was 63 years, and the average follow-up period was 30 months. Good to excellent results were reported in 69% of cases. Complete and partial humeral head necrosis developed in 3% and 1% of cases, respectively. Warner employed double-plate stabilization to 71 displaced two-, three-, and four-part fractures of the proximal humerus using one-third tubular plates.65 At an average of 17 months of ­ follow-up, outcomes were graded as good or excellent in 63% of patients and fair in 25%. T-plate osteosynthesis used for fracture dislocations demonstrated a rate of humeral head osteonecrosis of 39%, with poor results in almost 50%.66 Gerber and associates reported on their experience with open reduction with internal fixation (ORIF) of complex proximal humerus fractures with a rate of avascular ­ necrosis of 35%.67 If avascular necrosis did not develop, the result was good with an average Constant score of 78. The use of a buttress plate technique for three-part fractures has also been reported.68-71 However, several authors have reported a substantial complication rate using this technique, including avascular necrosis, plate impingement requiring removal, and failure of fixation in osteoporotic bone.68,72,73 High complication rates have also been reported using blade plates, the most frequent of which was unrecognized joint penetration in 22%.74 Given the inconsistent results using traditional plating methods, the use of

Darder and colleagues presented 64% excellent or satisfactory results in 33 patients (average age, 59 years) with four-part fractures employing modified K-wires through the tuberosities in conjunction with tension band techniques.75 Zyto found no benefit from tension band fixation in a more elderly group with a mean age of 74 years compared with nonoperative management.45 These results illustrate the need to consider the patient’s age, lifestyle, and bone quality as well as the fracture pattern in making a management decision. Screw and tension band technique has also been reported with 80% good results.76 Panagopoulos and coworkers reported the result of fixation of 16 patients with valgus-impacted four-part proximal humerus fractures with transosseous suturing with impressive results: only 1 patient developed avascular necrosis, with an average Constant score of 87.77 Several other studies have shown acceptable outcomes with suture fixation alone, with Park and colleagues reporting an average ASES score of 87 in 28 shoulders with displaced proximal humerus fractures; however, most of these were two-part greater tuberosity and surgical neck fractures.78 Hockings and Haines also showed good results at more than 6-years of follow-up of valgus-impacted fractures treated with reduction and suture-only fixation, with 10 of 11 patients having a Constant score greater than 70, and 8 having a score greater than 80.79 Others have reported good results with antegrade modified Ender nails in conjunction with suture or wire tension band fixation in an attempt to achieve better rotational and longitudinal control (Fig. 17J1-13).35,57

Retrograde Ender Nails Originally, a retrograde approach was recommended to avoid rotator cuff injury. Ogiwara and associates described the effectiveness of stabilization of displaced surgical neck fractures with retrograde Ender nailing.80 Thirtyfour patients were evaluated at the average follow-up of 9.9 months. Bone union occurred by 5.9 weeks in all but 1 patient.

Intramedullary Nailing Intramedullary nailing is still used for some proximal humerus fractures (Fig. 17J1-14). Lin and coworkers reported 86% excellent or satisfactory results with locked intramedullary nailing in 21 proximal humerus fractures.81 Union occurred at an average of 14.8 weeks with one case of postsurgical impingement. A later study by Lin demonstrated similar results in three-part fractures, with 78% excellent or satisfactory at an average of 24 months’ follow-up.82 In another study, 21 two- and three-part proximal humerus fractures treated by the Kapandji nailing technique were also reviewed. 83 Pin migration was frequent, but the technique was thought to be technically easy, noninvasive, and inexpensive. Agel and associates reported on a series of fractures treated with the Polarus nail and reported that 3 of 20 patients had migration of the proximal screws.84

1044 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B

D

Figure 17J1-11   A, Displaced two-part fracture of the surgical neck of the humerus. Note complete displacement of the fracture fragments. This fracture is at high risk for nonunion without intervention and would require an extensive period of immobilization and rehabilitation. B, Lateral radiograph of the injury. C, Anteroposterior view after treatment with a locked humeral plate. D, Four-year follow-up radiograph demonstrates healing in anatomic alignment. The patient had excellent function.

Shoulder 1045

A

B

C Figure 17J1-12   A, Anteroposterior view of a two-part angulated surgical neck fracture in a skeletally immature patient. This technique would also be used in a young adult. B, Anteroposterior view after placement of percutaneous pins. C, Axillary view after pin placement. The most proximal pin is first placed into the greater tuberosity and head fragment so that it can be used as a joystick to aid in reduction. It is then driven distally into the cortex medially in conjunction with lateral pins placed from distal to proximal into the head.

They recommended against this implant when there was associated lateral comminution, which was thought to contribute to screw migration. There has been some interest in the technique of retrograde intramedullary fixation to avoid injury to the greater tuberosity and rotator cuff. Retrograde intramedullary fixation was used in 74 unstable proximal humerus fractures, resulting in 60% good or excellent, 30% satisfactory, and 10% unsatisfactory or poor outcomes. The main complications associated with the procedure, especially in marked

osteoporosis, were secondary loss of reduction (16%) and pin migration (21%).85 Loitz and colleagues reported satisfactory results of retrograde unreamed nailing in 120 fractures of displaced two-part and occasionally three- and four-part fractures.86 This study included 110 cases of unreamed humeral nails with deployable proximal locking and 10 cases of solid interlocking nails. Functional results averaged 87% of the opposite side, with complications including nail migration (8.3%), instability (3.8%), nonunion (5.8%), and iatrogenic fractures (5.8%). The results

1046 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 17J1-13   A, Anteroposterior view of a three-part proximal humerus fracture treated with Ender nails and tension band fixation. B, Anteroposterior view after treatment.

A

again revealed that patients with high-grade osteoporosis, small proximal fragments, and poor compliance were poor candidates for this procedure.

Humeral Head Replacement Shoulder hemiarthroplasty is a well-accepted surgical ­procedure for the treatment of select proximal humerus fractures, including four-part fractures and fracture­dislocations, three-part fractures associated with severe osteoporosis, head-splitting fractures, and severe head impression fractures. The role of prosthetic replacement in treating acute proximal humeral fractures requires special consideration, especially in an athletic population. Prosthetic replacement is not an option for the young, athletic patient and should be reserved for older patients with osteoporotic bone and significant four-part and headsplitting fractures. Nevertheless, a discussion of humeral head replacement is appropriate with any discussion of proximal humerus fractures. Surgical reconstruction requires ­restoration of humeral length, center of rotation, and anatomic retroversion. Adequate positioning of the tuberosities and osteosynthesis are most critical to the successful outcome of prosthetic replacement surgery.45,87-89 Hartsock and associates reported satisfactory results in about 80% of cases in young patients, with acute fractures faring better than chronic cases.90 Ballmer and Hertel pointed out the value of proper evaluation of the patient and ­accurate fracture assessment.91 In young individuals with goodquality bone, they recommended careful techniques of reduction and fixation, even in the case of possible impairment of the vascular supply to the humeral head. In elderly patients with osteoporotic bone and limited compliance throughout aftercare, humeral head replacement is a superior indication with a more predictable expectation of a pain-free shoulder (Fig. 17J1-15). Robinson and coworkers reported a series of 138 patients who demonstrated Constant scores

B

of 64, with predictable relief of pain but variable function.92 Displacement of the tuberosities and age were both associated with poor outcome. Mighell and colleagues reported a series of 71 patients treated with hemiarthroplasty, with an average ASES score of 77.89 Tuberosity malunion was again correlated with a worse functional result. Most investigators have recommended arthroplasty as a primary rather than secondary procedure. Generally, secondary intervention requires a more demanding surgical approach as a result of soft tissue contracture, scarring, and altered anatomy with poor functional results. Gobel and associates evaluated the results of shoulder hemiarthroplasty in patients with 9 acute and 13 old fractures of the proximal humerus.93 The mean Constant score improved 28 points (27 to 55), especially distinct in the group with acute fractures. Bosch and colleagues compared outcome results after primary (>4 weeks) and secondary (≥4 weeks) hemiarthroplasty in elderly patients.94 The results of late reconstruction have been found to be inferior to those after reconstruction of acute fracture and to be more technically demanding with increased complications. Beredjiklian and colleagues95 and Norris96 evaluated 39 cases of malunion of the proximal aspect of the humerus. Sixty-seven percent of patients were deemed to have satisfactory results at an average of 44 months’ follow-up. The authors suggest that a delay in the operative treatment of the malunion of the proximal aspect of the humerus has a negative impact on outcome as a result of many factors, including disuse atrophy and more mature soft tissue scarring associated with prolonged malunion.

Open Reduction and Internal Fixation versus Humeral Head Replacement for Four-Part Fractures Some authors have questioned the effectiveness of prosthetic replacement, especially in terms of functional outcome. In 1993, Hawkins and Switlyk evaluated humeral

Shoulder 1047

A

C

B

Figure 17J1-14   A, Anteroposterior view of a three-part proximal humerus fracture treated with intramedullary rodding.   B, Axillary lateral view before treatment. C, Anteroposterior view after treatment.

head replacement in 20 three- and four-part fractures in patients with an average age of 64 years.97 The results were variable. Eighteen patients (90%) had mild or no pain; however, active forward elevation averaged only 72 degrees. Movin and coworkers reported poor functional results after shoulder replacement in 29 proximal humerus fracture patients followed for 2 to 12 years.98 They ­concluded that treatment of severe proximal humerus fractures with a prosthesis did not give complete pain relief and resulted in impaired shoulder function. Zyto and colleagues also reviewed the outcome of 27 patients after hemiarthroplasty for three- and four-part fractures of the proximal humerus.99 Results were disappointing, and they recommended open reduction and fixation. Bigliani and colleagues evaluated 29 cases of failed primary prosthetic replacement for displaced proximal humerus fractures.100 Most of the failure factors were technical errors of ­surgery. Detachment of either the greater tuberosity or both tuberosities occurred in 15 shoulders (52%). Malposition of the prosthesis occurred in 7 cases (24%). Loosening of the humeral component occurred in 12 shoulders (41%), and all but 1 of these were uncemented. Inadequate rehabilitation postoperatively or patient noncompliance with restrictions contributed to failure in 9 patients (31%). In 1991, Jakob reported a much lower rate of avascular necrosis in four-part valgus-impacted fractures than with true four-part fractures. In this series of 14 fractures, treated either with closed reduction and percutaneous K-wires or with minimal internal fixation, the authors obtained 74% excellent or satisfactory results at an average 4-year follow-up.101 Subsequently, Darder and associates presented 64% good to excellent results with tension band wiring and K-wires in the treatment of four-part fractures,

although the mean age of their patients was 59 years.75 They suggested that two groups, those with significant comminution plus fracture-dislocation, and patients older than 75 years, should be treated with hemiarthroplasty. Esser also treated 16 three-part and 10 four-part fractures employing the use of a modified cloverleaf plate in a young population.102 At an average follow-up of 6 years, results were excellent or good in 92%. In 1997, Resch and associates published the results of a 4-year study of 9 three-part fractures and 18 four-part fractures treated with percutaneous reduction and screw fixation.103 The average age was 54 years, with follow-up of 2 years. Average ­Constant­Murley scores of 91 for three-part fractures and 87 for four-part fractures were obtained. Thirteen of 18 fourpart fractures were of the valgus-impacted type. Hoellen and coworkers compared minimal osteosynthesis with primary prosthetic replacement in 30 four-part fractures of the proximal humerus.104 After 1 year, the results were similar in both groups, although there were two revision surgeries and four cases of implant removal in the minimal osteosynthesis group. In older patients, the authors recommended prosthetic replacement. Robinson and associates recommended using the vascularity of the head as a guide to prosthetic ­replacement in complex fracturedislocations, and advocated primary prosthetic replacement in the absence of arterial backbleeding in the older (>60 years) patient.105

Summary On the basis of recently published reports, some controversy remains regarding optimal treatment for fourpart fractures. It should be kept in mind that precise ­identification of fracture configuration and bone quality

1048 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 17J1-15   A, Anteroposterior view of a four-part proximal humerus fracture. B, Intraoperative radiographs of a trial prosthesis can be used to assess humeral height and the relationship of the humeral head to the greater tuberosity and the glenoid. C, Anteroposterior view after treatment with a humeral head replacement.

Shoulder 1049

are most important when choosing treatment methods. When ORIF is planned in the valgus-impacted fourpart fracture without osteoporosis, a soft tissue–sparing approach with minimal but stable fixation is recommended. However, it is suggested that older patients and true four-part fractures are best treated with primary replacement. In lesser fractures, such as three-part fractures, in which the tendinous attachments or bone is too frail to accept rigid fixation, arthroplasty remains the better alternative.

Complications Avascular Necrosis The most important factors in the development of avascular necrosis are fracture type, degree of displacement, and the treatment method selected. In four-part fractures, the reported incidence varies between 13% and 34%, whereas the incidence in three-part fractures is reported to range from 3% to 14%.75 Brooks and colleagues studied the effect of simulated four-part fractures on the vascularity of the humeral head.106 In most cases, perfusion was prevented, but if the head fracture extended distally below the articular surface medially, some perfusion of the head persisted by the posteromedial vessels. This finding suggested the importance of identifying the size of the fracture fragments and configuration of the fracture. Lower rates of avascular necrosis with valgus-impacted four-part fractures have been noted. It is postulated to be the result of the intact medial soft tissue hinge. This has been confirmed by Hertel and coworkers, who demonstrated that the most important factors predictive of loss of blood supply were loss of the medial calcar bony hinge and a fracture pattern with a short metaphyseal extension.107 When the medial hinge is intact, preservation of its integrity is of utmost importance in an attempt to prevent interruption of blood supply.100,103 Gerber and colleagues reported on 25 patients with a partial or complete collapse of the humeral head caused by post-traumatic avascular necrosis.97,108,109 Initial treatment consisted of ORIF in 19 cases, including 9 cases of minimal internal fixation. Five cases were treated with closed reduction and percutaneous pinning. The comparative outcome results were gained from two groups of patients based on radiographic analysis. Group 1 consisted of 13 patients with avascular necrosis and collapse in the absence of malunion, and group 2 included 12 patients with additional malunion. The clinical outcome was significantly related to the anatomic alignment of the fragments of the humerus at the time of healing. The exact pathogenesis of the poor outcome in patients with coexisting avascular necrosis and malunion is not well known, but it is believed that in addition to bony impingement caused by tuberosity malunion, malposition of the head segment prevents optimal function of the cuff muscles. The authors therefore recommended that a proximal humeral fracture that is at risk for avascular necrosis should be reduced anatomically if joint­preserving treatment is selected. If anatomic reduction cannot be obtained, other treatment options such as arthroplasty should be considered.

Neurovascular Injury Axillary nerve injury remains the most common peripheral nerve injury affecting the shoulder after glenohumeral joint dislocation and displaced proximal humeral fractures. During the acute phase of injury, the shoulder should be rested, and when clinically indicated, an extensive rehabilitation program should be initiated, emphasizing range of motion and strengthening of the shoulder girdle muscles. If no recovery is observed by 3 to 6 months after injury, surgical exploration may be indicated.110 Visser and associates compared electromyographic findings in shoulder dislocations and fractures of the proximal humerus with clinical neurologic examination in 215 patients.111 Electromyographic disorders were noted in 133 patients (62%), with testing of sensibility and clinical reflexes proving to be unreliable indicators for electromyographic abnormalities. The findings of this study implied that by clinical examination alone, a large number of axonal lesions remain undetected.

Nonunion Nonunions are most frequently encountered at the surgical neck. Nonunions are more frequent after ORIF.96 Surprisingly, up to 23% of patients undergoing these procedures may develop nonunion. Patients may have virtually no functional use of their shoulder and experience pain. Successful treatment is reliable in relief of pain and potentially can restore function. Usually, the diagnosis can be made with simple plain radiographs; however, CT may be helpful in the identification of tuberosity union and ­position. Surgical reconstruction for nonunion of the surgical neck often results in significant improvement in pain but much more modest improvement in active motion and function.112,113 Jupiter and Mullaji reported on the treatment of nonunion of proximal humerus fractures in the elderly patient with weak bone, resorption at the fracture site, contracture of the glenohumeral joint, and associated medical conditions.114 A blade plate, designed specifically for the treatment of these nonunions, was used in nine patients (mean age, 66 years) who had painful nonunion with a mean duration of 22.5 months. Autologous bone grafting was used in all cases. Eight nonunions followed displaced two-part proximal humeral fractures: five had been initially treated nonoperatively and four with medullary nails. After a mean of 6.5 months (range, 4 to 28 months), union had been achieved in all but one patient. Functional evaluation revealed good results in five patients, fair results in three, and one poor result. Use of the blade plate offered a successful method of stable internal fixation in these complex cases, although locking plates are gaining favor for this indication.

Malunion Tuberosity malunion can lead to significant shoulder dysfunction if it is associated with arthritic changes or fixed contracture. Conditions that contribute to the malunion or nonunion include osteoporotic bone, premature or aggressive rehabilitation, high-energy multitrauma, long

1050 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

head of biceps tendon interposition, and inadequate stability of operative stabilization. A relatively painless malunion may not require surgical management. However, symptomatic malunion may be treated with excision of the bony prominence with soft tissue release, osteotomy, and realignment, or prosthetic replacement if the articular surfaces are involved. Plain radiographs can be inadequate and even misleading in determining tuberosity displacement. The addition of CT in the evaluation of ­tuberosity ­malunions can improve assessment of displacement and assist in determining treatment options.37,115

and 4 three-part fractures.117 The results revealed that age and poor general condition of the patient, as well as the difficulty of the surgical technique, more than the rehabilitation, were related to the poor results observed after shoulder replacement. The discrepancy between active and passive elevation suggests that limited motion is not caused by shoulder stiffness and glenohumeral scarring but instead by weakness of the deltoid or external rotators, especially in the presence of greater tuberosity migration.

Loss of Motion

Myositis ossificans and heterotopic ossification can occur after fracture of the proximal humerus but most often pre­ sent little if any clinical significance. Because the degree of myositis can usually be correlated with the severity of the soft tissue injury, fracture-dislocations have the highest incidence.2 The heterotopic ossification rarely forms a block to motion, although many patients have decreased motion secondary to soft tissue contracture.

Early passive range of motion is the goal of both ORIF and replacement surgery. It is imperative to continue exercises because patients usually do not attain their maximal result until 12 to 18 months after surgery.116 Boileau and associates reported prognostic factors during rehabilitation after shoulder replacement for 42 four-part fractures

Authors’ Preferred Method

of

Myositis Ossificans

Treatment

Two-Part Greater Tuberosity Fractures

Lesser Tuberosity Fractures

Once the deltoid-splitting approach has been performed and adhesions lysed, debris and hematoma are removed to clean the fracture bed and allow for anatomic reduction of the tuberosity. The tuberosity fragment is identified and mobilized with heavy nonabsorbable sutures such as No. 5 Tevdek or Ethibond, placed at the bone-tendon interface to incorporate the strong rotator cuff tendon (Fig. 17J1-16). This area is often stronger and holds suture more securely than when the sutures are placed through osteoporotic tuberosity bone, where they may easily cut out. To completely reduce the tuberosity, it is important to place sutures at the level of the superior, middle, and inferior facets to overcome the displacement forces of the supraspinatus and teres minor. Based on these rotator cuff insertions, the fragments may be pulled superiorly, posteriorly, or in both directions at the time of injury, depending on the fracture location within the tuberosity. Sutures therefore must be placed in a manner that not only brings the superior facet downward but also brings the inferior facet forward. With this configuration of suture placement, the fracture can then be reduced. If the fragment is large, it may require removal of a small amount of cancellous bone to allow reduction. At this point, drill holes are placed around the periphery of the bed in the shaft and humeral head. By definition, there is an associated rotator cuff tear either in the rotator interval between the supraspinatus and subscapularis or more posteriorly between the supraspinatus and infraspinatus. It is helpful to repair this tear after reducing but before securing the fragment to the head and shaft to take tension off of the tuberosity fixation. It is essential to repair the tear in the cuff to restore optimal function. The tuberosity is now securely fixed to the shaft through the drill holes, using the heavy nonabsorbable sutures at the bonetendon interface in a figure-of-eight fashion (Fig. 17J1-17). The deltoid is meticulously closed, also with nonabsorbable suture.

The displaced isolated lesser tuberosity fracture is a rare lesion with a minimum of clinical experience reported. Although it is difficult to be dogmatic about treatment, most authors would agree that if a portion of the articular surface is involved with the tuberosity fracture or a significant block to internal rotation exists, then ORIF is warranted. Fixation of these fractures is performed in a manner similar to that described for greater tuberosity fractures, although exposure for the anteromedial displacement seen is better afforded by a deltopectoral or anterior axillary approach. The anterior axillary approach differs from the standard deltopectoral only in that the skin incision is vertical and made in the anterior axillary crease for better cosmesis. The skin is undermined superiorly and medially to gain the necessary exposure. The remainder of this approach is as described for the deltopectoral approach. The subscapularis and lesser tuberosity fragment are then identified with external rotation and forward elevation of the arm. The subscapularis and lesser tuberosity are then mobilized with heavy nonabsorbable suture and reduced into the freshened bed. Drill holes are placed in the head and shaft around the bed’s periphery, and the fragment is secured as previously described for the greater tuberosity. Some authors have used a cannulated screw in place of suture fixation, although this is not our preference (Fig. 17J1-18). Excision of the fragment by shelling it out of the subscapularis and anterior capsule and with subsequent reattachment of the subscapularis has also been described with some success. Inasmuch as the deltoid origin has been left intact, the interval between the deltoid and pectoralis is easily closed. Two-Part Surgical Neck Fractures

ORIF of displaced surgical neck fractures is most easily performed through the deltopectoral approach. Once the fracture is adequately exposed, the shaft and head are

Shoulder 1051

Authors’ Preferred Method

of

Treatment—cont’d

A

B

Figure 17J1-16   A, Superior skin incision (dotted line) is begun just lateral to the coracoid and extends 7 to 8 cm in Langer’s lines over the lateral margin of the anterolateral corner of the acromion. B, Mobilization of the greater tuberosity fragment with heavy nonabsorbable suture placed at the bone-tendon interface at the superior, middle, and inferior facets to correct both superior and posterior displacement. (A, From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 143.)

­ obilized, and the fracture site is curetted. The humeral m shaft is ­mobilized gently, and control of the humeral head is gained either by placing heavy suture at the tuberosity bonetendon interfaces or with skin hooks to reduce the head on top of the shaft. Reduction is obtained by forward elevation of the shaft while gently pulling on the sutures controlling the head. The fracture is then impacted and the arm lowered while keeping tension on the sutures through the cuff to prevent loss of reduction. Once adequate reduction is achieved, fixation is accomplished with a locking proximal humerus plate. It should be emphasized that sutures placed into the rotator cuff will allow better mobilization of the head fragment and will also be incorporated into the plate, increasing the stability of the construct. Heavy nonabsorbable suture incorporating the tuberosities and cuff to the shaft in addition to the locked plate construct will provide excellent, rigid fixation and allow for very early and aggressive rehabilitation, important in a patient considering a return to sport. This augmented fixation is of even greater importance when there is associated comminution of the fracture site with less inherent reduction stability. While reduction is maintained, the locked plate is placed over the lateral aspect of the humeral shaft. A key

c­ onsideration is the height of the plate. A plate placed too high along the greater tuberosity will cause impingement in the subacromial space and reduce the ability to forward-flex and abduct the arm. A plate placed too low will compromise fixation in the humeral head (Fig. 17J1-19). Once the correct height of the plate is determined, we fix the plate to the shaft with a nonlocking cortical screw. The proximal portion of the plate is then fixed provisionally into the head with K-wires. At this point, a fluoroscopic image is taken to check the position of the plate. If it is malpositioned, the provisional K-wires are removed and the position adjusted until it is correct: parallel to the shaft of the humerus and more than 5 mm distal to the tip of the greater tuberosity. Once the position of the plate is confirmed and appropriate, locking screws are placed through the appropriate guide into the humeral head. It is important to ensure that no screw has penetrated into the joint space, which can be accomplished only by a careful fluoroscopic survey encompassing an arc of at least 90 degrees. Unrecognized screw penetration causing chondrolysis is a disastrous complication that can only be avoided by fastidious checking of screw lengths. If a screw has been found to penetrate the joint, it can be exchanged for a screw that is 10 mm shorter without significantly compromising fixation. Nonlocking (to secure Continued

1052 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method

of

Treatment—cont’d

B

A

C

Figure 17J1-17   A, Displaced greater tuberosity fracture seen on an anteroposterior radiograph. B, On an axillary radiograph, the posterior displacement can be appreciated. C, Tuberosity fragment is secured to its bed with No. 5 Tevdek sutures in figure-of-eight fashion, incorporating the strong cuff tendon after closure of the rotator cuff tear. (A, From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 146.)

the plate to the shaft) followed by locking screws are then placed into the humeral shaft. At this point, an assessment of stability and range of motion is undertaken. Finally, previously placed sutures may be incorporated from the rotator cuff tendons into the plate. This allows fixation into the tendinous cuff to be combined with plate fixation into the strong cortical bone of the humeral shaft. Three-Part Fractures

ORIF of these fractures merely combines the techniques described for two-part tuberosity and surgical neck fractures. However, it is here that the anatomy becomes difficult. Strict attention to detail and anatomic landmarks can ease the task at hand. Again, the deltopectoral approach is employed. The

biceps tendon now becomes the lighthouse for identifying the somewhat complicated anatomy. It may be found underneath the pectoralis major insertion and should be protected if the pectoralis is released. Following the biceps superiorly will lead to the rotator interval, which is incised to the glenoid to gain further exposure and facilitate mobilization of fracture fragments. The displaced medial lesser or lateral greater tuberosity is then mobilized, as previously described, with heavy nonabsorbable suture. Once identified, the head and its attached tuberosity may be mobilized in a similar manner, gaining control of the possibly dislocated head by gentle traction on sutures placed at the tuberosity bone­tendon interface, assisted with a blunt Darrach retractor for leverage back into the joint.

Shoulder 1053

Authors’ Preferred Method

of

Treatment—cont’d

Figure 17J1-18   Treatment of a lesser tuberosity fracture with open reduction and minimal internal fixation with a cannulated screw. (From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 146.)

Once control of the fragments is obtained, the displaced tuberosity is first reduced and fixed to the head and intact tuberosity with heavy suture through drill holes (Fig. 17J1-20). The fracture is now converted to a two-part surgical neck fracture. The head-tuberosity fragment can be fixed to the shaft using a locking plate as for a two-part fracture. Alternatively, the head-tuberosity fracture can be repaired using a tension band construct augmented with intramedullary Ender rods. We prefer this technique for three-part fractures, described later. Once adequate reduction of the tuberosity to the head is achieved with heavy nonabsorbable suture, the head fragment is reduced onto the shaft. Again, heavy suture or wire incorporating the tuberosities and cuff are placed, which will be supplemented with intramedullary nails in a tension band configuration. Eighteen-gauge wire may be substituted for suture, inasmuch as it will provide greater immediate stability, but the risk for breakage, migration, and irritation in the subacromial space must be weighed. Therefore, we prefer suture whenever possible. Ender nails (3.5 mm) are superior to straight rods or pins such as Rush rods, in that they afford three-point fixation and therefore enhance rotational stability. The addition of the tension band configuration with intramedullary nails has been found to add even greater longitudinal and rotational stability over that of either tension banding or intramedullary nailing alone (Fig 17J1-21). This augmented fixation is of even greater importance when there is associated comminution of the fracture site with less inherent reduction stability. While reduction is maintained, small longitudinal incisions are made in the direction of the rotator cuff fibers over

the lesser tuberosity and just outside the articular surface for nail insertion. Ender nails are preferred here, not only for stability, as previously discussed, but also for the ability to place figure-of-eight suture or wire through the eyelet. The slot of the Ender nail is long, however, and an appreciable amount of metal may still protrude proximally. Therefore, the nail can be modified with an additional hole above this slot for suture or wire incorporation. This allows for deeper insertion of the nail into the humeral head, placing the tip well below the cuff tendons (Fig. 17J1-22). The site for nail insertion is dependent on the lack of associated fractures within the tuberosity chosen to achieve maximal rigidity. When no other fractures are identified, the greater tuberosity is the better choice, inasmuch as two nails may be used here as opposed to only one in the lesser tuberosity. Therefore, anterior and posterior longitudinal incisions are made in the supraspinatus tendon over the greater tuberosity, and an awl is used to penetrate the bone. The posterior nail is best placed initially because levering on this partially inserted nail will aid in holding the reduction and preventing the humeral head from falling posteriorly. The second nail is then inserted more anteriorly, about 1.0 to 1.5 cm from the first. It is advantageous to use nails of different lengths to prevent the possibility of a stress riser distally. Nails between 22 and 27 cm in length are generally adequate. Before the nails are completely buried, two drill holes are made in the shaft lateral to the biceps tendon for tension band suturing or wiring. Figure-of-eight suture or wire (No. 5 Tevdek or 18-gauge wire) is passed through the eyelets of the nails, passing it deep to the rotator cuff tendon between the nails to prevent proximal migration, and then through the predrilled holes in the shaft. Before the suture is tied or both limbs of wire are twisted, the nails are impacted well below the cuff, and fracture reduction is evaluated. Once the figure-of-eight is secured, range of motion and fixation stability are assessed carefully to guide the postoperative rehabilitation without stressing the repair. The rotator cuff incisions and deltopectoral interval are closed, usually over suction drainage. Four-Part Fractures, Head-Splitting Injuries

In four-part fractures and head-splitting injuries, humeral head replacement is the treatment of choice, especially in elderly patients. It should be emphasized that four-part fractures are exceedingly rare in young patients. Valgus-impacted fractures have been reported to have lower rates of osteonecrosis (9% to 11%), and ORIF is therefore possible. ORIF is performed by elevating the impacted head and relocating the tuberosities. It is also imperative to fill the void with cancellous bone chips.103 Minimal osteosynthesis can be performed with 1.8-mm K-wires to secure the head to the shaft, and numerous sutures are used to anchor the tuberosities to each other and the shaft. A locking plate may be added and the tuberosities secured to the plate, with additional fixation into the head. This is a procedure best performed in young, healthy bone. The technique of humeral head replacement uses a deltopectoral approach and leaves the origins and insertions intact. It is important to identify and then mobilize the ­fracture Continued

1054 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method

of

A

Treatment—cont’d

B

Figure 17J1-19   A, Placement of the locking plate at least 5 mm below the lesser tuberosity will reduce the chance of subacromial impingement when the arm is abducted. B, This plate is placed too proximal along the shaft, increasing the risk for plate impingement.

Figure 17J1-20   This three-part fracture is reduced to two parts by securing the involved tuberosity to the head and the lesser tuberosity with multiple sutures. The exact construct depends on the anatomy of the fracture. (From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 151.)

fragments. Skin hooks can aid in retrieving retracted fragments. No. 5 Tevdek/Ethibond suture is placed carefully at the ­ tuberosity bone-tendon interface. By avoiding placing sutures through bone, fragmentation of comminuted or ­osteoporotic bone is prevented. Humeral positioning is critical in determining the correct height, version, and sizing of the component and head. Secure prosthesis fixation always requires cement. Drill holes are placed in the shaft medial and lateral to the bicipital groove for tuberosity fixation, and then No. 5 nonabsorbable sutures are passed through these drill holes before cementing. The goal in tuberosity reconstruction is to obtain healing of the tuberosities to the shaft and to each other. Tuberosity stay sutures, which were previously placed at superior, middle, and inferior tendinous insertions, are now used to secure the tuberosities to the prosthetic fin, to each other, and to the shaft, respectively. It is important to secure the tuberosities below the head. The inferior sutures are first secured to the shaft, then to the fin, and finally to the other tuberosity. Wire fixation should be avoided because of the potential for breakage. Tuberosities are held together with a towel clip before tying the final sutures. The rotator interval is then repaired with nonabsorbable No. 1 Tevdek/Ethibond suture. Range of motion is then tested, and the tuberosity repair is inspected.

Shoulder 1055

Authors’ Preferred Method

of

Treatment—cont’d

A B

D

C

Figure 17J1-21   A and B, Fixed tuberosity and head unit are reduced and secured to the shaft with a nail inserted into the intact lesser tuberosity. C, Figure-of-eight sutures or wires are used to secure the greater tuberosity and the humeral head segments to the shaft through four predrilled holes. D, Clinical example of similar injury 1 year after open reduction and internal fixation of three-part anterior fracture-dislocation. (A, From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 152.) Continued

1056 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method

of

Treatment—cont’d

Figure 17J1-22   Modified 3.5-mm Ender nail with additional hole above the eyelet to allow deeper insertion below the rotator cuff. (From Cuomo F, Zuckerman JD: Proximal humerus fracture. In Browner BD [ed]: Techniques in Orthopaedics, vol 9. New York, Raven Press, 1994, p 152.)

REHABILITATION Intensive rehabilitation after proximal humeral fractures is essential to maximize the functional result after treatment. In the athletic patient, it is especially important to avoid the common complications of stiffness and decreased function. The limits of postoperative rehabilitation are based on the limits of the repair. Physician-supervised postoperative rehabilitation plays a vital and integral part in the management of displaced proximal humerus fractures and is paramount for optimal results. The primary goal of surgical reconstruction is to obtain fixation secure enough to allow early passive motion and active-assisted exercises. Early motion is necessary to prevent intra-articular and extra-articular adhesions, which may be extremely difficult to eliminate nonoperatively. In most cases, passive range of motion is begun on the first postoperative day in the form of pendulums and passive forward elevation in the plane of the scapula. Dependent on the fixation rigidity noted at surgery, external rotation with a stick and elevation are also added in the immediate postoperative setting, usually within 1 to 2 days. These passive exercises are performed 4 times daily until healing has occurred, at which point more advanced stretching is begun. Internal rotation behind the back is held for 6 weeks when the greater tuberosity has been fixed to avoid stressing the repair. In the case of lesser tuberosity fractures, external rotation is limited to a point, determined at surgery, that will not place tension on the tuberosity ­fixation. Active range of motion is not initiated until adequate bone and tendon healing has occurred, usually at 6 weeks, with resistive and strengthening exercises added at about 3 months in the form of Thera-Band or surgical tubing. ­Progressive improvement in range of motion and strength

will continue over a full year’s duration; therefore, a ­physician-supervised rehabilitation program is essential throughout this time to maximize results. A four-phase program is used for rehabilitation. In the initial phase, passive and assisted range of motion exercises are performed within the limits of the repair or fracture stability until the fracture is deemed to be healed. In the second phase, active range of motion and advanced stretching exercises are instituted. Early resistance exercises, both isometric and isotonic, are begun. The third phase moves toward advanced stretching exercises as well as isotonic and isokinetic strengthening. The final fourth phase is activityspecific exercise, which attempts to return the athlete to his or her sport. In the initial phase of rehabilitation, exercises are performed more frequently for a shorter duration. A common protocol consists of exercises performed 4 to 5 times daily for 10 to 15 minutes. The use of moist heat before exercise helps to relieve discomfort and make the soft tissues more supple. Analgesics are used for pain control as necessary. The first phase consists of pendulum exercises along with supine assisted elevation exercises. These can be accomplished with the use of either an overhead pulley mounted to a frame, a 3-foot stick, or the opposite hand. Hand and elbow range of motion exercises are begun in this initial phase (Fig. 17J1-23A and B). A stick is used to perform supine external rotation stretching. In the second phase, active range of motion is initiated with cane-assisted active forward elevation, activities of daily living, isometrics, and further stretching, that is, internal rotation behind the back (see Fig. 17J1-23C). When the fracture is deemed to be clinically and radiographically healed, the third phase is begun. The third phase of exercises includes more advanced ­ stretching

Shoulder 1057

exercises such as wall stretch for both elevation and external rotation. Isotonic exercises needed to strengthen the ­components of the rotator cuff and three components of the deltoid are performed using a pulley and weight system. The trapezius and rhomboids are strengthened using a hand-held weight for shoulder shrugs and scapular retraction exercises. As healing progresses and strengthening advances, isokinetic strengthening can be used in both internal and external rotation as well as flexion-extension planes. Attention should be given to the fact that a weak rotator cuff can allow superior shear of the humeral head and therefore can propagate impingement. Care is taken to keep the strengthening exercises out of the impingement planes until adequate cuff strength has been regained (Fig. 17J1-24). The final phase of rehabilitation is sport specific. Ballistic overhead activities in the throwing plane can be performed using surgical tubing or isokinetic equipment (Fig. 17J1-25). Subcomponents of the sport-specific activity can be performed until the athlete completes the entire sport-specific motion in a pain-free fashion. Return to play is variable, depending on the severity of the fracture (Box 17J1-5).16

Muscle Ruptures Involving the Proximal Humerus Region Excluding the Rotator Cuff Injuries to the musculotendinous unit are quite common in sport-related activities. Most cases result only in partial injury and rarely cause long-term disability for the athlete. Partial injuries or muscle strains may cause pain, loss of strength, and decrease in function leading to significant interference with athletic performance. Muscle strains and tendinitis probably represent the most frequent causes of missed playing time in competitive athletes.16 Complete avulsion injuries to major tendons can lead to substantial functional disability. With complete disruption of a musculotendinous unit or the muscle belly, significant alteration of joint biomechanics results. Because most complete tendinous avulsion injuries result in retraction of the tendon by muscle contraction and shortening, permanent dysfunction may occur unless appropriate anatomic repair is carried out.

A

C

B

Figure 17J1-23   Rehabilitation phase I and II. A, Pendulum exercise. B, Forward elevation using overhead pulley. C, Internal rotation stretching using opposite hand (a phase II exercise).

1058 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 17J1-24   Rehabilitation phase III (strengthening).   A, External rotation with Thera-Band. B, Internal rotation.   C, Scapular retraction.

A

B

Figure 17J1-25   Rehabilitation phase IV (sport-specific exercise). Ballistics using Thera-Band in throwing position.

Shoulder 1059

Box 17J1-5 R  eturn to Play in Proximal Humerus Fracture After painless range of motion and bridging callus on ­radiography are apparent, a conservative estimate should require 12 weeks of additional rehabilitation. Early range of motion will minimize stiffness and allow earlier return to play. Strengthening exercises are held until radiographic evidence of healing is apparent.

Most significant muscle ruptures occur when an actively contracting muscle group is overloaded by the application of a load or extrinsic force that exceeds tissue tolerance.118 In the athlete, this overload can occur acutely with the application of a single macrotraumatic force, or it can occur chronically with the application of multiple submaximal forces at a rate that exceeds the body’s ability to respond. Muscles and tendons clearly do respond to stress and applied demand by varying their strength and ­dimensions.119 Tendons become thicker with repeated stresses, and their collagen fibers become more appropriately oriented for function. Repetitive stress can eventually damage tendon tissue if the rate of application of the stress is more rapid than the body’s ability to respond. Healing of microscopic fiber failure occurs by the formation of scar and granulation tissue, which leads to an area within the tendon of altered mechanical properties. It is in this area that macrofailure will eventually occur.16 In the normal state, it has been shown that tendon appears to be stronger than the muscle belly and the tendon-bone interface.118 However, in the case of repetitive microtrauma to tendons, as often occurs in sports, this “normal” situation is modified and may lead more commonly to tendinous rupture. Both the rate of force application and the mechanism of injury affect the site of rupture.119 An additional factor that can affect the normal state of muscle-tendon unit physiology and biomechanics is the use of anabolic steroids. In response to reports in recent years of frequent abuse of steroids by certain groups of athletes, much research has begun to understand ­ better the changes that develop in the human system.120,121 Abuse of anabolic steroids may increase the risk for injury to the muscle-tendon unit both by direct effects on structure and physiology and indirectly through changes in biomechanics.120

Rupture of the Pectoralis Major Rupture of the pectoralis major is an uncommon injury that most commonly occurs in skeletally mature men. A recent review of the literature reports a meta-analysis of 112 cases with enough data to evaluate cause, rupture site, injury mechanism, and treatment outcomes.122 All patients were men. This injury occurred most commonly in sports during weight training, weightlifting, or wrestling when the arm was externally rotated and abducted. Most reported ruptures are complete and are located at the insertion to the humerus. Work-related injuries occur more often at the musculotendinous junction. The prognosis was related neither to the age of the patient

nor to the location of the rupture. Surgical treatment, preferably within the first 8 weeks after the injury, had a significantly better outcome than nonoperative treatment or delayed repair.122 McEntire and colleagues completed a review, revealing that only 45 cases had been reported before the 11 cases noted in their series.123 Pectoralis major rupture, therefore, is a relatively rare injury, with about 150 reported cases in the literature. Significantly, most of the patients sustained rupture of the pectoralis major while involved in sports. Weightlifting was by far the most common activity, followed by rugby, wrestling, other contact sports, and even windsurfing.124 This injury has been reported in all age groups, but patients in the third and fourth decades are affected predominantly. There has never been a case reported in a female.125-132 This most likely reflects the fact that men in the early middle-age groups are historically more commonly involved in contact sports and have traditionally had greater involvement in weightlifting.

Anatomy The pectoralis major muscle arises as a broad sheet of muscle from the mid-clavicle, sternum, ribs, and external oblique fascia (Fig. 17J1-26). The muscle is often described as having two parts or “heads,” an upper clavicular head, and a lower sternocostal head. Muscle fibers, composed of a flat tendon about 5 cm broad, converge toward their insertion distal to the crest of the greater tuberosity of the humerus. The tendon consists of two laminae placed one in front of the other that commonly blend together inferiorly. The fibers from the clavicular head run in line to form the anterior lamina of the tendon. The more distal and deep fibers of the sternocostal head run upward and laterally to form the posterior lamina, those with the lowest origin having the highest insertion, giving a twisted appearance to the muscle.10 Dissections performed by Kretzler and Richardson showed the tendon to be about 1 cm long on the anterior surface and perhaps 2.5 cm long on the deeper or posterior surface.128 The tendon is quite thin and appears to be a coalescence of the anterior and posterior vesting fasciae rather than a true tendon or musculotendinous junction.

Laminated tendon

Pectoralis major m.

Figure 17J1-26   The two “heads” of the pectoralis major muscle, which arise from a broad area on the chest and converge into a flat laminated tendon that attaches to the proximal humerus.

1060 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Grossly, the pectoralis major forms the smooth, rounded appearance of the anterior ­ axillary fold. The ­ muscle is innervated by the medial and lateral pectoral nerves, which branch directly from the medial and lateral cords. The primary function of the pectoralis major muscle is adduction and medial rotation of the humerus. The muscle can also function to flex the humerus if it is extended behind the plane of the body. With the arm at the side, the upper or clavicular portion of the muscle is most effective, but as the shoulder is abducted, the lower portion of the muscle provides the bulk of the power.10

Classification Pectoralis major ruptures can be classified by the degree and location of the rupture. The degree of rupture can be classified as either complete or incomplete. Most cases are undoubtedly partial or incomplete and represent strains of the muscle belly or musculotendinous junction. Most cases in the literature that have come to surgery involved complete ruptures. Avulsion-type injuries involving the tendon at or near the humeral insertion tend to predominate.122,123 Incomplete and complete ruptures at the musculotendinous junction or in the substance of the muscle belly have also been reported in association with a direct blow.125,126,132,134,135

Clinical Evaluation History and Mechanism of Injury Patients with an acute rupture of the pectoralis major usually present with a definite history of injury. The typical injury occurs when the patient is lifting weights or receives a direct blow. The patient experiences an immediate, severe pain and a tearing sensation with burning at the site of injury. Some patients describe an audible pop or snap associated with a complete rupture of the tendon. The most common mechanism of injury is indirect and is related to excessive tension on a maximally contracted muscle. Weightlifting, specifically the bench press, is an example of this mechanism.123,128,132 Ruptures have also been reported to occur while windsurfing, being dragged behind a moving vehicle, and attempting to break a fall. These injuries can be associated with a fracture or dislocation of the proximal humerus. Direct trauma to the muscle as in wrestling, rugby, football, or other contact sports has also been reported.123,124,135-138 The most common activity associated with complete rupture was the bench press weight lift. This indirect mechanism is associated with the development of excessive muscle tension and usually results in an avulsion type of injury at or near the tendinous insertion into the humerus. An attempt to break a fall on an outstretched hand can also apply a severe indirect force to the muscle-tendon unit that can result in rupture. Wolfe and colleagues reported on nine weightlifters with pectoralis ruptures and designed an anatomic cadaver study to determine the mechanism of injury.139 They concluded that the short, inferior fibers have a mechanical disadvantage in the final portion of the eccentric phase of the lift. Continued application of high loads to these ­maximally

stretched fibers of the sternal head produces rupture of the pectoralis major. Direct blows are also associated with injury to the pectoralis major and may occur within the substance of the muscle belly. In wrestling, there seems to be a propensity to disrupt the muscle at the sternoclavicular head by direct contact. McEntire and colleagues, in their review of the literature, proposed that some of the previously reported cases of pectoralis major rupture may actually represent congenital absence of the muscle.123 Complete avulsion of both the sternal and clavicular heads may also occur in older individuals.125

Physical Examination The patient with acute rupture of the pectoralis major pre­ sents with the affected extremity splinted across the chest and often supported by the opposite hand. Early swelling and ecchymosis occur across the chest, axilla, and upper arm region. Pain with shoulder motion is present. Distal rupture is associated with asymmetry of the anterior axillary fold as the muscle retracts medially and superiorly. There is commonly a prominent bulge involving the retracted muscle belly. A palpable defect is present at the site of injury but may be obscured initially by swelling. Weakness is present with attempted adduction and internal rotation of the arm. Clinical diagnosis immediately after injury can be difficult because ecchymosis, swelling, and severe pain can obscure the exact location and extent of injury. Consequently, it may be very difficult in the acute setting to distinguish between complete and partial ruptures. Some authors advocate repeated physical examinations over a 4week period (Fig. 17J1-27).132 In chronic cases, the retracted muscle belly may be very prominent, and asymmetry from side to side is noticeable. A defect in the anterior axillary fold gives a webbed appearance to the anterior axilla. The most reliable clinical examination finding is weakness in adduction and internal rotation. In chronic cases, the patient often complains of a dull, aching pain with activities that require heavy muscle power.

Imaging Plain radiographs of the shoulder, chest, and scapula are usually normal unless a bony avulsion has occurred.140 A loss of the normal pectoralis major shadow has been described but is very subtle.128,141 Liu and colleagues reported on Cybex isokinetic muscle testing, CT, and ultrasonography to help in making the diagnosis.142 Magnetic resonance imaging (MRI) is the optimal imaging technique to study the normal and abnormal conditions of the pectoralis major muscle and tendon unit. The normal pectoralis major myotendinous unit has low signal intensity on both T1- and T2-weighted images. Lee and associates described anatomic landmarks for recognition of injuries to the muscle and myotendinous unit.143 These include the quadrilateral space, or the origin of the lateral head of the triceps, as the superior boundary and the deltoid tuberosity as the inferior boundary of the intact tendon of insertion. Failure to visualize a normal insertion within these boundaries should prompt a search for

Shoulder 1061

A

B

Figure 17J1-27   A, Clinical presentation of an acute rupture of the pectoralis major tendon at rest. Note the mildly webbed appearance of the axilla. B, Webbing and deformity on the right are accentuated by stressing the pectoralis with resisted contraction.

r­ upture and retraction of the tendon medially. MRI can be valuable in making or confirming an early diagnosis, which can avoid surgical delay. Early diagnosis has the advantage of avoiding adhesions, muscle retraction, and atrophy and thus preventing the delayed return of the athlete to competition (Fig. 17J1-28).129

Treatment Many authors agree that partial ruptures of the pectoralis major tendon and incomplete lesions of the muscle belly itself respond to conservative treatment.123,141,144-146 These partial injuries are characterized by less swelling, ecchymosis, and pain than complete injuries. When adduction is

A

resisted, no defect in the tendon is palpable. Initial treatment begins with ice, rest, and control of the hematoma followed by a program of progressive range of motion and strengthening exercises. Slow return to strength is the rule. These injuries usually heal without major deformity or significant strength deficit. They often require a 6- to 8-week course of recovery before return to stressful lifting ­activities. Both surgical and nonsurgical treatment methods for complete rupture of the pectoralis major have been described.127,128,132,145 Three major parameters can be assessed when describing results of treatment after this injury: pain, strength, and cosmetic deformity. Nonoperative treatment is similar to that described for incomplete injuries.

B

Figure 17J1-28   A, T1-weighted magnetic resonance image of an acute rupture of the pectoralis major tendon. B, T2-weighted image. Note the detachment of the tendon from its insertion on the humerus as well as medial retraction.

1062 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

­ perative management for complete tears within the tenO don can be carried out by reattaching the tendon through drill holes or suture anchors in the humeral cortex at its anatomic insertion site.122,129,138,139,147,148 If the tear is at or near the musculotendinous junction, repair can be accomplished by an end-to-end technique. In a review of 29 cases by Park and Espinella, only 58% of patients with rupture of the pectoralis major treated by nonoperative means showed good results.149 In the same series, the authors report 90% good to excellent results in those patients treated surgically. Zeman and colleagues reported four of four cases treated surgically had excellent results.132 In five patients treated conservatively, significant strength deficits existed that limited return to athletic competition. Late repair has been shown to be effective in improving pain, strength, and function in a high percentage of cases.128,132,150 Delayed repair for up to 4 to 6 weeks does not appear to affect the result. Several reports of late repairs from 3 months to 5 years discuss successful results. Anbari and associates recently reported a delayed repair successfully performed 13 years after the initial injury, although the authors caution this was possible only because the ruptured sternal portion of the muscle was scarred to the intact clavicular portion and had not retracted.151 The authors described all

Authors’ Preferred Method

of

late repairs with some residual deficiency of strength but good overall results.128,147,150,151 Because many of these injuries occur in athletes, most authors have recently advocated surgical repair to obtain complete recovery and restoration of the full strength of the muscle.128,129,132,133,135 Pavlik and colleagues surgically repaired an acute rupture in one wrestler who 3 months later went on to win an Olympic gold medal.138 Wolfe and associates found that surgically treated patients showed comparable torque and work measurements, whereas conservatively treated patients demonstrated a marked deficit in both peak torque and work/repetition.139 Schepsis and colleagues retrospectively studied 17 cases of distal ­pectoralis major muscle rupture to compare the results of repair in acute and chronic injuries and to compare operative and nonoperative treatment.148 Isokinetic testing revealed that acute injuries treated surgically demonstrated the highest adduction strength (102% of the opposite side) compared with chronic injuries (94%) or nonoperative treatment (71%). There were no statistically significant subjective or objective differences in outcome between the patients treated operatively for acute or chronic injuries, but these patients fared significantly better than patients treated nonoperatively.148

Treatment

For partial ruptures involving predominantly the muscle belly, we prefer nonoperative treatment. Initially, the patient’s arm should be immobilized in a sling, and ice should be applied. Pendulum exercises are begun on the second or third day, followed by slow, progressive shoulder range of motion exercises. Strengthening exercises are begun at about the fourth week, and slow progression to isotonic and isokinetic exercises is allowed. Typically, recovery takes 6 to 8 weeks. Early diagnosis of complete rupture is helpful in advising the patient about treatment options. Diagnosis can usually be based on clinical data, but if there is a question,

MRI is a helpful adjunct to obtain additional information. Complete ruptures within the substance of the tendon or avulsion from the humerus should be repaired to bone (Fig. 17J1-29). An anterior axillary incision is used, and the tendon is isolated. The tendon is usually thin and small and is sometimes difficult to identify owing to rotation into the muscle belly and large hematoma. Heavy No. 5 nonabsorbable sutures are woven in a Bunnell-type fashion from the musculotendinous junction out through the end of the tendon. Even if some lateral tendon remnant is found at the humerus, we believe the musculotendinous unit should be

Pectoralis major m.

Distal portion of deltopectoral incision

A

Distal stump

Figure 17J1-29   Surgical repair of a ruptured pectoralis major tendon. A, The distal portion of a deltopectoral incision is used.

Shoulder 1063

Authors’ Preferred Method

of

Treatment —

cont’d

Bony trough

B

Repaired

Figure 17J1-29, cont’d   B, The tendon is repaired with multiple No. 5 nonabsorbable sutures.

anchored back to bone. A bony trough is developed in the cortex of the humerus, and multiple drill holes are made in the adjacent cortex. The tendon is advanced into the trough and anchored by passing the sutures through the adjacent

Postoperative Management and Rehabilitation A sling is worn postoperatively for the first month. Gentle shoulder range of motion exercises are begun early in the first week. Further stretching and range of motion exercises are initiated to obtain a full range of motion by the sixth or eighth week. Light isotonic exercises are begun at about 6 weeks, and slow progression to advanced strengthening and functional activities is made during the first 3 months. Return to full unrestricted weightlifting is often delayed until 6 months after the repair.

Criteria for Return to Sports Participation The patient with a partial or complete injury who is treated conservatively should regain full shoulder motion and protective strength before return to competition is allowed. The patient treated by surgical repair often requires 6 months to return to full strength. Before return to heavy athletic competition, full range of motion should be achieved.16

drill holes. If there is residual tendon remaining on the humerus, this can be oversewn for double reinforcement of the repair.

three major portions of the deltoid that converge on a common insertion point midway down the humerus at the deltoid tuberosity. The axillary nerve is the primary innervation for the deltoid muscle. It passes beneath the glenohumeral joint to exit posteriorly at the quadrangular space, then branches into anterior and posterior divisions that course on the deep surface of the deltoid about 6 to 8 cm distal to its origin. The axillary nerve innervates both the deltoid and the teres minor and provides sensation for the upper lateral arm. Functionally, the deltoid provides power for flexion, extension, and abduction of the glenohumeral joint. It also forms the bulk that covers and protects the underlying rotator cuff and the shoulder joint itself.10

Supraspinatus m. Posterior third of deltoid m.

Rupture of the Deltoid Complete rupture of the deltoid muscle is a rare clinical entity, although contusions and strains are not uncommon in both throwing and contact sports. Little has been written about injury to this muscle, yet even minor injuries can seriously affect athletic performance.141,152-154

Anatomy and Biomechanics The deltoid is the primary motor for the shoulder girdle. It arises from the anterior clavicle, the acromion, and the spine of the scapula (Fig. 17J1-30). There are

Middle third of deltoid m. Clavicular head of pectoralis major m.

Subscapularis m.

Figure 17J1-30   The three major portions of the deltoid, originating on the clavicle, the acromion, and the spine of the scapula. Insertion is midway down the humerus at the deltoid tuberosity.

1064 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Clinical Evaluation History and Mechanism of Injury Minor injuries to the deltoid such as strains and contusions are common in athletic activities. Usually a direct blow to the upper arm while it is in abduction or forward elevation is the cause. Deltoid muscle strains have also been described in throwing sports. The anterior deltoid can be injured during acceleration, whereas the posterior deltoid is subject to injury during deceleration. Injury to the origin of the deltoid can occur with grade V acromioclavicular dislocations when the distal clavicle ruptures through the deltotrapezius fascia. Complete disruption of the deltoid is quite rare and appears to be associated with crushing injuries or severe direct blows received during major trauma from the outside. A few rare examples have been reported in the literature, usually associated with cases of severe trauma such as those described by Gilcreest and Albi in 1939155 and McEntire and colleagues in 1972.123 Recently, Blazar and Morisawa both reported on the spontaneous detachment of the deltoid origin in patients with chronic, massive rotator cuff tears. These detachments were associated with an acute and sudden onset of shoulder weakness and minimal pain.152,153 One series described more than 1000 musculotendinous injuries that contained no cases of deltoid rupture.156 It is important to note that rupture of the deltoid is perhaps most commonly associated with surgical intervention.141,154,157 The deltoid muscle is commonly detached from its origin on the clavicle and acromion in a number of shoulder procedures. Inadequate reattachment of the deltoid origin followed by early institution of resistance exercises can result in retraction of the deltoid distally. This is an extremely difficult problem to reconstruct later, and therefore early diagnosis following detachment and retraction of the deltoid is important. Permanent loss of deltoid function owing to detachment can result in a severe functional deficit that may preclude return to ­ athletic ­competition.

Treatment In most strains and contusions involving the deltoid muscle, local conservative treatment is all that is necessary. Ice is applied in the acute phase, with immobilization in a sling as symptoms warrant. After several days, heat, gentle mobilization, stretching, and slowly progressive strengthening exercises are usually all that is required to return to full function. Management of postoperative dehiscence requires early diagnosis followed by surgical reattachment. Deltoid disruption after shoulder surgery is associated with poor function. Sher reported on 24 patients who underwent direct repair or rotational deltoidplasty reconstruction for detached muscle origin after shoulder surgery.154 Two patients required a shoulder fusion for intractable pain. Overall, 1 (4%) excellent, 7 (29%) good, and 16 (67%) unsatisfactory results were observed. A poor outcome was associated with a prior lateral acromionectomy, involvement of the middle deltoid, a massive rotator cuff tear with weakness in external rotation, and a residual postoperative defect larger than 2 cm. Only in select cases was repair or deltoidplasty thought to improve function and pain.

Authors’ Preferred Method of Treatment Partial injuries and contusions should be treated with ice and rest in the acute phase, followed by progressive range of motion exercises, heat, and strengthening. Complete healing and return to sport can be expected. Complete disruption and dehiscence following previous shoulder surgery should be treated surgically. Every attempt should be made to anchor these injuries back to bone. Protection in an abduction splint is almost always necessary to decrease any tension on the repair. Passive motion should be instituted early, but active exercise is delayed for at least 6 weeks. Prognosis after this injury is guarded.

Physical Examination

Rupture of the Subscapularis

In the acute deltoid strain without rupture, local tenderness and mild swelling may be the only clinical signs. Ecchymosis may be present in the case of a contusion caused by a direct blow. Shoulder motion is often limited, and weakness secondary to pain may be present. Examination after acute complete rupture will demonstrate massive injury with swelling, deformity, and ecchymosis. With complete avulsion, there is loss of the normal shoulder contour, direct tenderness, and often a palpable defect. Weakness and limited abduction are present. Because this injury usually occurs after massive multiple trauma, masking by the additional injuries may make immediate recognition difficult. In postsurgical dehiscence of the deltoid, the defect is usually palpable in the region of the previous attachment. This late detachment can also be difficult to diagnose because of local ­swelling.

As an isolated tear, this is an extremely rare injury.155,158,159 Recently, Gerber and Krushell presented a series of 16 cases of this injury that required surgical treatment.160 Partial ruptures have been documented by many authors in association with anterior dislocation of the glenohumeral joint.58,128 In the traumatic situation, subscapularis tendon avulsions are often associated with avulsion fractures of the lesser tuberosity.161,162

Anatomy The subscapularis muscle arises from the subscapular fossa on the deep surface of the scapula and inserts by a broad tendinous attachment into the lesser tuberosity of the humerus. It is the anterior part of the musculotendinous rotator cuff. The subscapularis forms the upper border of the quadrangular and triangular spaces with the axillary

Shoulder 1065

nerve, posterior humeral circumflex vessels, and scapular circumflex vessels passing beneath it. It has an important function in internal rotation of the humerus and acts as a dynamic humeral head stabilizer.10

Clinical Presentation History and Mechanism of Injury Although poorly described in the literature, the mechanism of injury is similar to that for an anterior dislocation. A fall on the outstretched arm in abduction as the patient attempts to bring the arm into an adducted position is often the mechanism of injury. The cause of injury as forceful hyperextension or traumatic external rotation of the adducted arm has also been described.160 The patient presents with pain, anterior swelling, and decreased mobility around the joint. If the injury is ­associated with anterior instability, the apprehension test is positive. With an isolated injury, weakness of internal rotation is present along with increased passive external rotation. Gerber and Krushell described a clinical test for evaluation of the integrity of the subscapularis that they termed the lift-off test.160 The patient places the arm in internal rotation with the dorsum of the hand on the back. If the patient is unable to lift the hand off the back, incompetence of the subscapularis can be suspected. Additionally, one can hold the patient’s hand in internal rotation away from the back and then release. Inability of the patient to hold the hand away from the back is consistent with subscapularis incompetence.

Recommended Treatment In cases associated with anterior dislocation, the subscapularis rupture is repaired during anterior reconstruction. Surgical repair or reconstruction of isolated ruptures has been reported with good results.159,160 In cases associated with lesser tuberosity avulsion, treatment by direct reattachment with either sutures or screw fixation has been described.161,162

Anatomy This large muscle arises in the back from the lower six thoracic spinous processes, the thoracolumbar fascia, the posterior part of the iliac crest, the lower three or four ribs, and the inferior angle of the scapula. It attaches to the upper medial humerus, forming the posterior axillary fold. It is a powerful adductor of the humerus and acts as an extensor of the humerus when the shoulder is flexed. The thoracodorsal nerve arising from the posterior cord supplies the latissimus.10

Clinical Evaluation History and Mechanism of Injury In most cases of muscle strain and contusion of the latissimus dorsi, swelling, ecchymosis, and deformity are notably absent. Tenderness to direct palpation of the muscle belly and pain when adduction of the shoulder is resisted are the major clinical signs. The latissimus is stressed during the throwing motion when the arm is rapidly decelerated during follow-through. Kawashima and colleagues164 and Spinner and associates166 have described the only cases of complete rupture of the latissimus dorsi tendon. In one case, this was ­associated with a complete rupture of the pectoralis major in a patient who suffered a severe industrial crush injury. Pain, ecchymosis, and swelling of the posterior axillary fold and tenderness to palpation were present. The other case was associated with an avulsion of the conjoined tendons of the latissimus and teres major muscles.164,166 There is also one report of an isolated teres major tendon injury in a baseball pitcher diagnosed by MRI.167 Lazio and coworkers reported on two cases of ruptures of the latissimus dorsi as a result of the surgical approach in transthoracic and thoracoabdominal approaches to the spine.165 In each case, the patient presented 3 to 6 months postoperatively with a large painful mass along the posterior axillary line adjacent to the surgical incision.

Recommended Treatment Authors’ Preferred Method of Treatment We recommend primary repair of injuries to the subscapularis tendon with or without avulsion fracture. If the bone fragment is large, it may be repaired with screw fixation. If it is a tendinous avulsion or a small portion of bone, repair with heavy sutures through drill holes is preferred.

Latissimus Dorsi Injuries Injuries to the latissimus dorsi are extremely rare. Complete rupture has been described only in conjunction with rupture of the pectoralis major and the teres major in severe trauma and after anterior spine surgery.163-166 Pain in the region of the latissimus dorsi can occur in throwing athletes as a result of muscular strain and in contact sports as a result of contusion.

Surgical repair of complete rupture by direct suture has been reported with complete return of function.164 Partial injuries should be treated like other acute strains of the musculotendinous unit. In the acute phase, rest, ice, and stretching are the mainstays, followed by a program of progressive motion, heat, ultrasound, and strengthening exercises.

C l The

r i t i c a l

P

o i n t s

surgeon undertaking fixation using a locking plate must use meticulous care to avoid penetration of screws into the joint. This involves a careful examination of the screw position in a minimum of 90 degrees of shoulder rotation, using either manipulation of the patient’s arm or arcing of the fluoroscope. Two isolated views are not adequate.

1066 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� l Fixation of three-part fractures using tension band and Ender nails requires that the tuberosity fragment first be reduced to the head fragment, converting the fracture to a two-part fracture. l We describe and prefer a technique of suture fixation through the cuff combined with intramedullary Ender nails to fix the head/tuberosity fragment to the shaft; however, a locking plate is also a good option (as for two-part fracture). l Proximal humerus fractures in younger sporting patients are usually high-energy injuries. l Fractures often occur in association with a dislocation, especially in a younger patient. l A neurovascular examination is essential to document function, especially of the axillary nerve. l An adequate axillary radiograph is an absolute requirement for classification. l Treatment is determined by fracture classification and patient function. l Anatomic, rigid stabilization is the goal of operative treatment, regardless of technique. l There are multiple methods of fracture fixation, none of which has been definitively proved superior to ­others.

S U G G E S T E D

R E A D I N G S

Egol KA, Kubiak EN, Fulkerson E, et al: Biomechanics of locked plates and screws. J Orthop Trauma 18:488-493, 2004. Fankhauser F, Boldin C, Schippinger G, et al: A new locking plate for unstable fractures of the proximal humerus. Clin Orthop 430:176-181, 2005. Gerber C, Schneeberger AG, Vinh TS: The arterial vascularization of the humeral head: An anatomical study. J Bone Joint Surg Am 72:1486-1494, 1990. Hertel R, Hempfing A, Stiehler M, et al: Predictors of humeral head ischemia after intracapsular fracture of the proximal humerus. J Shoulder Elbow Surg 13:427433, 2004. Keener JD Parsons BO, Flatow EL, et al: Outcomes after percutaneous reduction and fixation of proximal humeral fractures. J Shoulder Elbow Surg 16:330-338, 2007. Neer CS II: Displaced proximal humerus fractures: Parts I and II. J. Bone Joint Surg Am 52:1077-1103, 1970. Nho SJ, Brophy RH, Barker JU, et al: Innovations in the management of displaced proximal humerus fractures. J Am Acad Orthop Surg 15:12-26, 2007. Park MC, Murthi AM, Roth NS, et al: Two-part and three-part fractures of the proximal humerus treated with suture fixation. J Orthop Trauma 17:319-325, 2003. Rose PS, Adams CR, Torchia ME, et al: Locking plate fixation for proximal humeral fractures: Initial results with a new implant. J Shoulder Elbow Surg 16: 202-207, 2007. Sperling JW, Cuomo F, Hill JD, et al: The difficult proximal humerus fracture: Tips and techniques to avoid complications and improve results. Instr Course Lect 56:45-57, 2007.

R eferences Please see www.expertconsult.com

S E C T I O N

J

Injuries of the Proximal Humerus 2. Injuries of the Proximal Humerus in the Skeletally Immature Athlete Kaye E. Wilkins

PHILOSOPHY OF TREATING FRACTURES OF THE PROXIMAL HUMERUS IN CHILDREN The High-Performance Athlete It will become obvious that there are many methods of treating the various fractures that occur in the proximal humerus in the immature athlete. In many instances reference will be made for singling out a specific treatment for the so-called high-performance athlete. This usually entails achieving an anatomic reduction by surgical intervention. This is in contradistinction to waiting for some misalignment to naturally remodel over the ensuing months or

even years. Depending on remodeling alone to resolve the misalignment may result in the individual having less than total motion and possibly total function of the involved shoulder. For almost all individuals this may not result in any problem of performing the usual ordinary tasks of life. For an athlete who, for example, is a very talented baseball pitcher, any minor reduction in motion or shoulder function can inhibit the ability to achieve the ultimate pitching potential. The athlete may have to change to playing in the outfield rather than continuing pitching. Certainly for those individuals who are presently participating at the high-performance level, there is no argument but to provide the treatment that will have the best chance of restoring both the anatomic alignment and function of the shoulder.

Shoulder 1067

The Non–High-Performance Athlete What about those athletes who are not participating at high-performance levels? This can present a dilemma to the treating physician and parents. The real question is what the future holds for this injured athlete. At the point of the injury, this may not have yet been determined. Although this athlete may not become engaged in a life of high-performance sporting participation, he or she may be involved in an occupation that would involve the need to perform heavy overhead work such as would be required of a painter or carpenter. Should not this lower level athlete be offered the same level of treatment? That is a question that can be answered only by a full explanation to the parents (and possibly to the patient as well) of the advantages and risks of all the modalities of treatment available for the injury sustained.

Offer Only the Best In summary, all pediatric athletes, regardless of their level of performance, should be offered all the modalities of treatment for their specific proximal humeral injury. The advantages, risks, and rehabilitation needs for each treatment available for that injury pattern should be outlined in detail so that the athlete and family can make the appropriate decision. In addition, should the initial treating surgeon not be able to perform all of the modalities available, it may appropriate to refer the athlete to an individual who can. It is also important that the treating surgeon be objective and not allow his or her prejudices to color the description of the modalities available.

EPIDEMIOLOGY Incidence of Shoulder Injuries Uncommon Shoulder injuries are relatively uncommon in the overall picture of injuries to the pediatric musculoskeletal system. Although fractures to the upper extremities per se are the most common injuries seen in the pediatric age group, most are distal rather than proximal. In his study of 8682 fractures in children, Landin1 found that 22.7% involved the distal forearm, 8.1% involved the clavicle, and only 2.2% involved the proximal end of the humerus. Most clavicular fractures after the age of 10 years occurred in boys and occurred during falls or contact sports. In a more recent review of 6493 children’s fractures by Cheng and coworkers,2 fractures about the shoulder accounted for less than 5% of all fracture types.

Almost No Sequelae In looking at the long-term sequelae of pediatric sportsrelated injuries, Marchi and coworkers3 found that none of those injuries occurring in the shoulder region had any long-term sequelae in young athletes. The highest

i­ncidence of injuries with long-term sequelae was seen in the ankle and elbow regions.

Organized Sporting Events Increasing Incidence In Landin’s overall global review of all pediatric fractures, only 21% occurred in organized sporting events.1 In fractures associated with sporting events, there have been increases not only in participation but also in the number of injuries. Landin1 found that in the three decades from 1950 to 1980, there was a fivefold increase in the incidence of injuries to children from sporting activities.

Male Predominance in Sports In another study that looked only at fractures of the proximal humerus, Kohler and Trillaud4 found that 22% of 136 fractures occurred during sporting events. In their series, 60% of the patients were male. The peak age incidence was 10 to 14 years of age. Two thirds of the fractures involved the proximal metaphysis, and the other third involved the physeal plate.

Nonorganized Sports Higher Risks In the pediatric age group, most sporting activity occurs outside the organized educational setting. This has been a growing trend over the past 20 years. In the pediatric setting, many studies1,5,6 have found that far more injuries occur in nonorganized play activities. It was estimated in 1980 that nearly 30 million young people aged 6 to 21 years were involved in nonscholastic athletic programs.7 In contrast, in organized interscholastic sports in 1981,8 only 5.35 million young people were active participants. In the same study, the most popular sport reported for boys was football. For girls, basketball was the leader.

Female Preponderance In Landin’s study1 fractures of the proximal humerus accounted for only 2.2% of all fractures in children. More of these fractures occurred in girls. In his series of children in Sweden between the ages of 9 and 10 years, fully 50% of the fractures resulted from falls that occurred with horseback riding. Almost all of these patients were girls. Even when horseback riding injuries were removed from the overall group, there was still a preponderance of girls. In this same study, Landin found that the incidence of females sustaining fractures of the proximal humerus had increased dramatically since 1970. This large preponderance of girls sustaining fractures of the proximal humerus was also found in another study from Sweden.9

Horseback Riding In Landin’s series, the sporting event that produced the highest percentage of shoulder injuries was horseback riding (Fig. 17J2-1).1 Twenty-eight percent of the injuries

1068 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Summary In summary, the overall risk for injury to the pediatric athlete is related to the nature of the sporting event, the age of the participant, and the method by which the players are grouped. There appears to be a greater risk for injury in the skeletally immature individual among participants in nonorganized recreational or play activities than in players who are under some type of adult supervision.

Sport-Specific Trauma Sports injuries can be divided into two categories: macrotrauma and microtrauma. Macrotrauma usually results in an acute failure of the osseous structures. Microtrauma, conversely, is usually the result of repetitive stresses so that the failure occurs over a prolonged period.

Macrotrauma

Figure 17J2-1   A proximal humeral metaphyseal greenstick fracture in an 11-year-old girl who fell from a horse. Note that the fracture is incomplete with minimal displacement.

sustained from horseback riding involved the proximal humerus, and another 9% involved the clavicle.

Age Fear of Crippling Injuries Age is a factor in the overall incidence of athletic injuries in the pediatric age group. In 1956, the American Medical Association (AMA) published a statement warning against the participation of skeletally immature individuals in organized contact athletic events.10 The AMA stated categorically that such participation was unsafe because of the large number of physeal or growth injuries that could occur. Subsequent follow-up studies11 showed that this fear of so-called crippling injuries was unfounded. In fact, injuries to the physes accounted for less than 5% of all sports injuries.

Age-Related Injuries Injuries are age related.12 Among grade school athletes, the injury rate is very low. The injury rates increase steadily with age so that the maximal rates are seen in the high school age group.13

In 1978, Garrick and Requa14 surveyed the overall incidence of injuries over a 2-year period in high school sports. They mainly looked at macrotrauma injuries. Macrotrauma injuries occur as a sudden failure of bone. Predictably, football and wrestling produced the greatest number of injuries, whereas swimming and tennis had the lowest injury rates (Fig. 17J2-2). In reviewing the incidence of macrotrauma shoulder injuries in football,15-20 bicycling,20-23 skiing,24 snowboarding,25 and wrestling,8,26 the overall incidence of true acute failure of the osseous structures is very low. In these sporting events the failure usually occurs in the soft tissues securing the glenohumeral and acromioclavicular joints. There are three major sporting events in which there can be a high incidence of acute failure of the osseous structure of the proximal humerus while engaged in the sport: baseball, football, and horseback riding. Baseball

Fractures of the humeral shaft in pediatric baseball players are relatively rare but do occur.27,28 Usually, they are related to some inherent defect in the bony structures (Fig. 17J2-3). Ireland and Andrews27 described a case in a young pitcher with an acute avulsion of the coracoid epiphysis. Football

Most injuries to the shoulder in football result in macrotrauma (i.e., fracture of the clavicle or glenohumeral dislocation). The overall injury rate increases with age. In the little league age group, the rate of football injuries overall was only 7.8%, compared with 17% in high school players.17 When specific body areas are examined, the percentage of football injuries involving the shoulder is fairly consistent, ranging from 8% to 12%.16,17,26,29 The shoulder ranks second after the knee in overall injuries sustained in football. There appears to be an increased incidence of shoulder injuries in more recent years. Culpepper and Niemann16 theorized that this was due to the outlawing of spearing, which brought a return of shoulder-body contact to tackling. In one study of recurrent anterior shoulder dislocations requiring surgical correction, 49% of the patients sustained their initial injury in football.23 In another study

Shoulder 1069 Figure 17J2-2   Injury rates for macrotrauma for various athletic events in high school sports. (From Garrick JG, Requa RK: Injuries in high school sports. Pediatrics 61:465, 1978.)

90 FOOT BALL

80

WRESTLING 70 2-yr rate Male Female

Injuries / 100 participants

60

1st yr 2nd yr 50 SOFTBALL 40

CROSS COUNTRY

20

BASKET BALL

SOCCER

GYM

30

CROSS COUNTRY VOLLEYBALL

10 BADMINTON

TRACK & FIELD

GYM

TRACK & FIELD BASKET BALL

BASEBALL

SWIMMING

SWIMMING

TENNIS TENNIS

0

of acromioclavicular injuries, 41% of patients sustained their initial injury in football.20 Horseback Riding

Trauma sustained by young horseback riders most frequently involves head and neck injuries. Next to head ­injuries, however, is skeletal trauma, with two thirds of fractures occurring in the upper extremity.30 In Sweden, the major cause of fracture of the proximal humerus in girls (see Fig. 17J-1) is falling off a horse.1

Microtrauma Injuries due to microtrauma are especially prevalent in the shoulder region. The discussion of microtrauma injuries involving the physis will be presented in detail in the final section of this chapter dealing with stress injuries of the proximal humeral physis.

STRUCTURE OF THE PROXIMAL HUMERUS The proximal humerus is composed of the epiphysis, the physeal plate, and the metaphysis, all of which have unique biomechanical properties. These properties contribute to the specific injury patterns seen in this area. A review of the unique structural anatomy of this area will give the reader a better understanding of both the mechanism and the treatment of injuries that occur in the proximal humerus.

It also must be remembered that the soft tissues, such as the muscles, tendons, capsule, and periosteum, also play a prominent role in the susceptibilities of the osseous tissues of the proximal humerus to failure.

Epiphyseal Development Two Ossification Centers The proximal humeral epiphysis is a hemispherical structure containing a cone-shaped physeal plate. The physis is more proximally directed posteriorly, which provides some intrinsic stability (Fig. 17J-4). The hemispherical epiphysis contains the articular surface and the greater and lesser tuberosities. The epiphyseal mass is formed by two separate ossification centers, which fuse to form a single center by about 7 years of age and are completely fused to the proximal humerus by 17 or 18 years of age.31

Ossification Process This serial development of the proximal humeral epiphysis has been studied in great detail by Ogden and his coworkers.32 The initial contour of the physis is transversely oriented. At birth, there is usually no radiographically discernible ossification center. The first center develops as the capital or articular center at about 2 months of age (Fig. 17J2-5). At 7 months of age, a second center develops in the area of the greater tuberosity. By 3 years, these centers have enlarged and matured, with the physis assuming

1070 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

its conical shape. When the individual reaches 7 years of age, there is a complete fusion of these two centers. The physis becomes more conical in shape. From 10 to 13 years of age, the greater tuberosity ossification center expands until it completely fills the cartilaginous space on the lateral portion of the epiphysis.

Physeal Closure By 14 years of age, the physis begins to close, starting in the center and extending peripherally.

Metaphysis Development When the epiphyseal centers fuse, the lateral metaphyseal cortex becomes thicker. It is composed almost entirely of cortical bone up to the physis, whereas the cortex on the medial metaphysis remains thin with a trabecular structure. This is thought to be one of the reasons that the Thurston-­Holland metaphyseal fragment in the Salter-Harris II fracture occurs medially. As the physis closes, the medial metaphyseal cortex increases in density. Even though the cortex on the medial side is thicker, it has developed some trabecular pattern next to the physis. This produces an inherent weakness on the medial side of the proximal humerus.

Muscle Attachments Figure 17J2-3   This 12-year-old sustained an acute fracture while simply throwing a baseball during practice. This fracture occurred through a cystic osseous defect in the proximal humeral diaphysis (arrows).

Anterior view

The proximal humeral epiphysis has four muscles attached to it. In the greater tuberosity posterolaterally are the teres minor, infraspinatus, and supraspinatus. The ­ subscapularis

Posterior view

B

A Figure 17J2-4   A, Line drawings of the proximal epiphysis and physis demonstrating the conical configuration. B, This SalterHarris type I physeal injury demonstrates the conical configuration as well. (A, From Grant JCB: An Atlas of Anatomy, 5th ed. Baltimore, Williams & Wilkins, 1962.)

Shoulder 1071

2 Mo.

3 Mo.

7 Yr.

9 Yr.

7 Mo.

10 Yr.

2 Yr.

13 Yr.

3 Yr.

14 Yr.

Figure 17J2-5   The schematic development of the proximal humeral epiphysis and metaphysis from age 2 months to 14 years. (From Ogden JA, Conlogue GJ, Jensen P: Radiology of postnatal skeletal development: The proximal humerus. Skeletal Radiol 2:153-160, 1978.)

inserts into the lesser tuberosity anteriorly. The pectoralis major attaches distally to the anterior metaphysis. In addition, the deltoid muscle inserts more distally at the diaphyseal-metaphyseal junction. When these two parts are separated by a fracture, there can be considerable ­displacement of the fragments. The dynamic effects of these muscles’ insertions on the epiphysis and metaphysis of the proximal humerus are demonstrated in Figure 17J2-6.

Capsular Attachments The glenohumeral joint capsule on the medial side attaches distally past the edge of the articular surface to the metaphysis. This means that a portion of the proximal physis lies intra-articularly within the shoulder joint (Fig. 17J2-7).

­ osterior circumflex humeral arteries, which arise from the p third part of the axillary artery.

Anterior Source Most of the blood supply to the osseous humeral head comes from the anterior ascending branch of the anterior circumflex artery.33 This artery ascends proximally along the upper end of the bicipital groove and then enters the head by branches that pierce the greater and lesser tuberosities. Once inside the humeral head, this artery assumes

Attachment of capsule

Blood Supply With the exception of the suprascapular artery, the blood supply to the shoulder arises from the second and third parts of the axillary artery (Fig. 17J2-8). The major ­supply to the proximal humerus comes from the anterior and

Physeal plate

Supraspinatus Infraspinatus Teres minor

Subscapularis

Pectoralis major

Figure 17J2-6   Effect of muscle forces on the proximal humeral epiphysis and metaphysis. (From Dameron TB: Fractures and dislocations of the shoulder. In Rockwood CA Jr, Wilkins KE, King RE [eds]: Fractures in Children, vol 3. Philadelphia, JB Lippincott, 1984.)

Figure 17J2-7   The relationship of the physeal plate and the glenohumeral capsular attachment to the proximal humerus is shown. The medial end of the physeal plate extends across an area covered by articular cartilage in the area noted by the dashed line. This area of metaphysis is intra-articular. (From Dameron TB: Fractures and dislocations of the shoulder. In Rockwood CA Jr, Wilkins KE, King RE [eds]: Fractures in Children, vol 3. Philadelphia, JB Lippincott, 1984.)

1072 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Suprascapular a. Anterior circumflex humeral a.

Transverse cervical a. Inferior thyroid a. Thyrocervical trunk Subclavian a. Highest thoracic a. Thoracoacromial a.

Posterior circumflex humeral a.

Box 17J2-1 O  sseous Failure Patterns of the Proximal Humerus

I.    Fractures of the proximal humeral physis II.    Pure metaphyseal fractures III. Avulsion of the lesser tuberosity IV. Stress fracture of the proximal physis

Lateral thoracic a.

Profunda brachii a.

Subscapular a. Superior ulnar collateral a. Brachial a.

Figure 17J2-8   The arteries of the shoulder region. (From O’Rahilly R: Gardner-Gray-O’Rahilly: Anatomy, 5th ed. Philadelphia, WB Saunders, 1986.)

a form like that of the lateral epiphyseal artery of the femoral head. That is, it forms an arcuate system from which branches radiate at right angles to the periphery of the epiphysis.

Posterior Source A small amount of blood supply originates from the posterior humeral circumflex artery. This artery enters the epiphysis through a small portion on the posteromedial surface of the humeral head, which is similar to the medial epiphyseal arteries of the femoral head.

Muscular Sources A part of the blood supply also arises anterolaterally through the rotator cuff into the greater tuberosity. Because much of the blood supply enters through the muscular attachments to the proximal humeral epiphysis, avascular necrosis of the humeral head is extremely rare after a fracture through the proximal humeral physis (see “Avascular Necrosis.”).

INCIDENCE OF FRACTURES OF THE PROXIMAL HUMERUS It is difficult to determine the true incidence of pure physeal versus metaphyseal fractures of the proximal humerus because many of the reported series combine both into the generic category of fractures of the proximal humerus.3,34-37 The specific incidence of each of the various fracture ­subtypes is discussed in each of the sections dealing with the specific fracture types.

SPECIFIC FRACTURE PATTERNS Failure of the bony structure of the proximal humerus can occur in one of four patterns. These are listed in Box 17J2-1. The first three occur as a result of acute macrotrauma. The

last pattern is the result of repetitive microtrauma. Each of the failure patterns is discussed in detail in the following section. When there is failure of the proximal osseous structures, the position of the fracture fragments is determined by the muscles that remain attached to the separate fragments and the periosteal coverings. Thus, the nature of the fracture pattern is determined by whether the fracture occurs through the physis or metaphysis Because the proximal humerus is the area where most of the growth and remodeling of the humerus occurs, the incidence and patterns of fractures in this area are much different from those in the shaft. There is more leeway in accepting less than an anatomic alignment of the fracture fragments because of the increased remodeling potential in this area.

FRACTURES OF THE PROXIMAL HUMERAL PHYSIS Structural Aspects Variable Strength The presence of the physeal plates about the shoulder provides bony matrices of lesser strength than those provided by the adjacent capsules, ligaments, or even in some cases periosteum. The physes have an age-related variability in strength. The physis and its perichondrial ring weaken just before maturity.38 This fact is borne out in a clinical setting in the classic study by Peterson and Peterson.39 They found that the greatest incidence of physeal injuries occurred between the ages of 11 and 12 years in girls and 13 and 14 years in boys.

Role of the Periosteum The periosteum serves to provide some stability between the physis and the metaphysis. It can also help to direct the remodeling process. The thickness is not uniform around the proximal humerus. The varying degrees of thickness can be a factor in the failure pattern.

Thinnest Anteriorly In laboratory studies on humeri obtained from stillborn infants, Dameron and Reibel36 found that posterior displacement of the metaphysis was extremely difficult to accomplish. They attributed this to the fact that the periosteum of the metaphysis was considerably thicker posteromedially. Anterior displacement of the metaphysis was relatively

Shoulder 1073

easy to achieve because of the thinner periosteum on the anterolateral surface.

Provides Stability If it remains intact, the periosteal sleeve can also stabilize an undisplaced fracture. Once the metaphyseal fragment has torn the periosteum, all of its intrinsic stability is lost.

Guides Callus Formation The intact periosteal sleeve serves as a guide to the callous formation when the fracture fragments have become markedly displaced. This is an important factor in the remodeling process (see Fig. 17J2-12C).

Incidence Pure Physeal Injuries The most recent large review of physeal injuries has been that by Peterson and coworkers in 1994.40 Their series looked at 951 physeal injuries in Olmsted County, Minnesota, from 1979 to 1988. In this group, only 1.9% involved the proximal humerus, which was less than the range of 2% to 6.7% in four other previous series totaling 3326 cases.39,41-43

Salter-Harris I or II Almost all physeal fractures in this area are either SalterHarris type I or II.4,36,44 Type I fractures are less common and occur usually in younger children (i.e., younger than 10 years). Almost all the physeal fractures among individuals older than 10 years are Salter-Harris type II lesions.4,36 Because of the flexibility of the shoulder, forces acting directly against the articular surface or perpendicular to the physis are rarely applied to the proximal humerus.

Figure 17J2-9   Radiograph showing muscle force displacement. The epiphyseal fragment is rotated into flexion, abduction, and external rotation (white arrow). The metaphyseal fragment is forced cephalad by the pectoralis and deltoid muscles (black arrow).

Fragment Displacement Loss of Adduction Force When the failure occurs through the proximal physis, the adducting force of the pectoralis major muscle has been completely lost. Thus, the epiphyseal segment has become totally under the control of the muscles that rotate the proximal humerus to produce a distinct fracture pattern.

Epiphyseal Rotation When the proximal epiphysis becomes disrupted from the metaphysis, the unopposed action of the muscles on the epiphysis tends to pull this fragment into flexion, abduction, and external rotation (Fig. 17J2-9).48

Other Salter-Harris Injuries

Signs and Symptoms

As a result, Salter-Harris type III and IV fracture patterns are extremely rare.44-47 Lee and coworkers46 recently described a case of a Salter-Harris type III fracture pattern in which the humeral head epiphysis was dislocated and displaced posterior to the glenoid labrum. They pointed out that this was unusual because if there is complete separation of the epiphyseal fragment, it usually lies anterior to the glenoid. The type V lesions, which often result in the development of a varus deformity of the humerus, are usually the result of nontraumatic injuries that occur during early infancy or childhood and thus are usually not seen with athletic events.

Displacement Dictates Swelling

Anatomic Characteristics Intracapsular Fracture The capsule attaches to both the epiphyseal and metaphyseal cortices. Thus, a physeal fracture passes through the medial physeal plate, creating a line that is intra-articular in location (see Fig. 17J2-7).48

The differentiation between fractures of the proximal physis and metaphysis of the proximal humerus may be difficult clinically. In undisplaced metaphyseal fractures, especially those in which cortical integrity is maintained, there may be only minimal swelling. The tenderness is usually localized over the proximal humerus. In physeal fractures and metaphyseal fractures with displacement, there is usually considerable bleeding into the soft tissues of the deltoid area, which produces marked swelling. The athlete with these types of fractures is usually uncomfortable and holds the extremity adducted to the chest. The weight of the extremity is often supported at the elbow and forearm with the opposite hand.

Clinical Signs Not Specific The only major differential clinical finding occurs with complete fractures of the proximal physis. The maximal point of tenderness is usually more proximal. With the

1074 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

physeal injuries, the distal metaphyseal fragment usually is laterally displaced. In this case it may be possible to define the prominence of the proximal metaphyseal fragment by palpating it under the anterolateral deltoid. However, with the severe swelling and pain that occur with these displaced physeal injuries, the ability to palpate these structures is severely limited. The final differentiation is usually dependent on the radiographic findings.

Evaluate for Ipsilateral Injuries It is important to assess all the nerves of the upper extremity to rule out a concomitant injury to any of the peripheral nerves of the brachial plexus. The vascular status of the upper extremity needs to be evaluated. Because the force was transmitted from the hand longitudinally, the entire extremity must be checked for the occurrence of less obvious ipsilateral fractures, especially in the distal radial metaphysis.

Mechanism of Injury Older Age Group The exact mechanism of proximal humeral physeal injuries is not completely clear. Physeal fractures are more common among athletes nearing skeletal maturity.

Multiple Mechanisms One of the first to truly analyze the mechanisms of proximal humeral physeal injuries was Williams in 1981.49 He did this by examining different displacement patterns of these fractures in his patients. It was determined that four forces could be applied to the proximal physis either singly or in combination. Only six variations of this, however, are likely to occur in the clinical situation: (1) pure extension; (2) pure flexion; (3) forced extension with lateral rotation; (4) forced extension with medial rotation; (5) forced flexion with lateral rotation; and (6) forced flexion with medial rotation. It is hoped that a better reduction can be obtained when the mechanism is determined according to the fracture patterns and the position of the patient’s upper extremity.

Displacement Dictated by Local Architecture The periosteum is weaker on the anterolateral aspect of the proximal humerus. In fractures involving the humeral physis, the proximal metaphyseal fragment is usually forced anteriorly and laterally through this weakened area. Neer and Horowitz42 reasoned that this force resulted from a direct blow to the shoulder by a posterolateral shearing force that adducted the humeral shaft and forced it anteriorly. Dameron,27 in contrast, believed that the force was directed longitudinally up the upper extremity as it was used to break a fall in a backward direction. The force originating as the hand hits the ground is transmitted proximally through the humeral shaft with the shoulder extended and adducted. This forces the metaphysis anteriorly, laterally, and cephalad. The horizontal alignment of the physis in the anterolateral portion of the proximal

humerus facilitates this displacement in an anterolateral direction. The combination of a stronger posteromedial periosteum and compressive posteromedial forces results in the triangular metaphyseal fracture fragment occurring in this area.

Metaphysis Slips Anterolateral When the periosteum disrupts, the proximal portion of the metaphysis tends to displace anterolaterally to the intertubercular groove under the long head of the biceps (Fig. 17J2-10).47

Classification Two Types Proximal humeral physeal fractures can be either acute or chronic. In the acute injuries, there is immediate partial or complete displacement of the physis from the adjacent metaphysis. The chronic type represents a stress injury, which is discussed at the end of this chapter.

Acute Injuries Fractures of the proximal humeral physis can be classified by location, degree of displacement, and stability. The degree of stability usually depends on the degree of initial displacement and the magnitude of the injury. The most commonly accepted classification of displacement is that proposed by Neer and Horowitz,42 who separated the displacement into four grades (Box 17J2-2). It should be noted that Neer and Horowitz in their original article42 used the term width of shaft instead of width of physis to denote the degree of routine displacement. In their illustrative cases, however, they demonstrated displacement of the physis. Thus, I have taken the license to correct this anatomic inaccuracy because the actual displacement is measured by the amount of physeal displacement, not by the amount of displacement of the proximal shaft, which is actually the metaphysis.

Radiographic Studies Routine Imaging Usually, routine roentgenograms are enough to demonstrate the presence of the fracture. The proximal and lateral displacement of the metaphyseal portion is usually obvious on routine anteroposterior views of the shoulder. It may be necessary to use transthoracic lateral or oblique scapular views to determine the degree of apex anterior angular displacement between the separate fracture fragments (see Fig. 17J2-10B).

Computed Tomography In some cases in which there may be a complex fragment pattern, a computed tomographic scan with horizontal cuts or three-dimensional reconstruction may be helpful in determining the fracture patterns and degree of ­displacement.

Shoulder 1075

A

C

Radiographic Comparison There is a distinct difference in the displacement of the fracture fragments between the proximal physeal and proximal metaphyseal fractures (Fig. 17J2-11). With the completely displaced fractures of the proximal physis, the distal fragment lies lateral to the epiphyseal fragment. Because the proximal fragment is rotated, the long axes lie at an angle to each other. With the metaphyseal fractures, the Box 17J2-2 Grades of Displacement Grade I—less than 5 mm Grade II—up to one third the width of the physis Grade III—up to two thirds the width of the physis Grade IV—greater than two thirds the width of the ­physis, including total displacement From Neer CS, Horowitz BS: Fractures of the proximal humeral epiphyseal plate. Clin Orthop 41:24-31, 1965.

B

Figure 17J2-10   A Salter-Harris type II fracture of the proximal humerus physis. Posteroanterior (A) and transthoracic (B) views show that the metaphyseal fragment (solid line) is anterolateral to the epiphyseal fragment (dashed line). C, Axillary lateral view confirms that the proximal metaphyseal fragment is anterior to the head.

pectoralis tendon insertion on the distal fragment forces it to lie medial to the proximal metaphyseal segment. Because there is some attachment of the pectoralis tendon on the proximal fragment, this proximal ­ fragment lies with its long axis relatively parallel to the distal ­fragment.

Treatment In choosing the appropriate treatment plan, two major factors must be evaluated. The first is the degree of rotation and displacement of the proximal fragment. The second is the remodeling capacity of the patient.

Proximal Fragment Rotated In displaced fractures of the proximal humeral physis, rotation of the proximal fragment often occurs because there is an absence of adductor forces on this fragment. The adductor forces act entirely on the distal metaphyseal ­fragment

1076 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 17J2-11   Left, With the completely displaced fractures of the proximal physis, the distal fragment lies lateral to the epiphyseal fragment. Because the proximal fragment is rotated, the long axes (lines) are at an angle to each other. Right, With the metaphyseal fractures, the pectoral tendon displaces the distal fragment to lie medial to the proximal metaphyseal segment. Because there is some attachment of the pectoral tendon on the proximal fragment, this proximal fragment is not rotated and thus lies with its long axis relatively parallel to that of the distal fragment (lines).

(see Fig. 17J2-9). In addition, the distal insertion of the deltoid muscle on the distal fragment provides a force that tends to retract the metaphyseal fragment proximally.

At no other physis in the body is there a larger proportion of contribution to longitudinal growth than at the proximal humeral physis. About 80% of the longitudinal growth occurs in this area.3,44,50 As a result, the remodeling potential in this area is tremendous. The younger the athlete, the greater is the potential for remodeling. Because of the wide range of motion of the glenohumeral joint, the residual varus that may remain usually does not result in any functional limitation.44 Shortening of these fractures is also well tolerated because of the independent function and non–weight-bearing status of the upper extremity. It must be emphasized, however, that there needs to be at least 12 to 18 months of growth remaining in the proximal humeral physis for adequate remodeling to be achieved.

f­ ragments into a more anatomic position. In some cases, a primary closed reduction is performed, either with sedation or under general anesthesia, and then the reduction is maintained by external support. The reduction is usually achieved by bringing the distal shaft fragment into flexion and some abduction and external rotation to align it with the flexed, abducted, and externally rotated proximal fragment.48 The real question concerns whether an anatomic reduction is necessary. Many series34,37,42,44,45,52-54 have shown that for most individuals, even those who have considerable initial displacement, simple immobilization often produces satisfactory results for most athletic activities not requiring high-performance upper extremity skills (Fig. 17J2-12). Baxter and Wiley44 found in their retrospective review that the manipulative process improved the arm’s position in only one third of the patients in whom it was attempted. When there was an improvement in position, the final result was no better than that seen in patients in whom an equal displacement had been accepted. These authors questioned whether active manipulation had had any effect on the final outcome.

Specific Treatment Methods

Maintaining the Reduction

As with most fractures, there are both closed and surgically invasive techniques that can be used in the management of fractures of the proximal humeral physis.

Once obtained, the reduction must be maintained nonoperatively by either a cast or some type of traction.

Remodeling Capacity

Closed Methods Obtaining a Reduction The argument for reduction of the fragments is that it decreases the degree of shortening that develops if the malalignment is allowed to remain.48,51 Various closed methods have been advocated to realign the fracture

Cast Methods

If the fracture is stable after the reduction, it can be immobilized alongside the chest. If it is not stable, this position of flexed abduction and external rotation must be maintained with either a shoulder spica cast in the salute position or a commercial splint. Neither immobilization technique is well tolerated by the athlete or the parents. The “Statue of Liberty” position with a spica cast should be avoided because of its potential to cause injury to the brachial

Shoulder 1077

A

D

C

B

E

F

Figure 17J2-12   No initial reduction. A, Anteroposterior image of this 12-year-old soccer player who sustained a Neer type III displaced fracture of the proximal humeral physis. B, Lateral image taken along the axis of the scapula. The metaphysis is anterior (black dotted line), and the epiphyseal segment is angulated posterior (white dotted line). C, Images obtained 6 weeks after the fracture demonstrate new bone (black arrows) forming along the intact inferior periosteal sleeve. The prominence of the fracture surface of the metaphyseal fragment is being resorbed. D, By 7 months after injury, there is considerable remodeling of the proximal humerus. Clinically, this patient still lacks full abduction (E) and external rotation (F).

plexus or to lead to development of vascular ­compromise to the upper extremity.22,47 Although this method of treatment was recommended in the past,55 it has now been universally abandoned. With the minimally invasive surgical techniques now available, combined with the unreliability of a cast to adequately maintain the reduction achieved, there is almost no indication for the use of casts in the treatment of these fractures. Traction Methods

Some37 have used the traction produced by the hanging arm cast to achieve a reduction or to improve the final fracture position. This method may not be effective if the cast is applied late after the fracture clot has congealed. The hanging arm cast usually requires that the patient sleep in the upright position. Having to sleep in this position can produce considerable discomfort in the early phases of the fracture healing process. Overhead skeletal olecranon traction can be used for patients in whom the usual external immobilization techniques cannot be used. It is a good method of achieving and maintaining a reduction, but it requires extensive hospitalization and has all the problems associated with the management and care of the skeletal pin or screw. The major indication for this method in an athlete would be a severely comminuted fracture or concomitant injuries that require a period of recumbent positioning.

Maintenance Difficult Nonoperatively Often after the reduction is obtained, it can be lost unless something is added surgically to the treatment process to stabilize the fragments (Fig. 17J2-13).

Operative Intervention The High-Performance Athlete It must be remembered, however, that these aforementioned series using nonoperative methods were treating individuals who were engaged mainly in normal activity. What about the high-performance athlete? Is there a greater need to achieve a more anatomic reduction in athletes? This question must be answered when such an athlete sustains this type of fracture.

Percutaneous Stabilization The major indication for operative stabilization is to maintain as near anatomic reduction as possible in an athlete who needs to regain a full range of shoulder motion. A ­semiclosed method of operative management involves reducing the fracture by closed methods first, and then stabilizing it with pins placed across the fracture site ­ percutaneously under image intensifier guidance. There are two methods of stabilizing the fragment using a ­percutaneous technique.

1078 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B

D

One involves inserting the pins obliquely across the fracture site. The second involves passing the pins proximally up the intramedullary canal. Direct Insertion

Figure 17J2-13   Loss of reduction. A, Injury image of a   12-year-old soccer player who sustained this Neer stage III displaced fracture of his left proximal humerus. B, An anatomic reduction was obtained by inserting a threaded pin into the proximal shaft to facilitate the manipulation. However, nothing was done to stabilize the reduction.   C, After reduction, the upper extremity was simply placed aside the chest. As a result, the prereduction position of the fracture fragments recurred. By 2 weeks, some early callus developed along the inferior periosteal sleeve (arrows).   D, Fortunately, after 6 months, there had been considerable remodeling of the fracture site.

The stabilizing pins or nails can be inserted directly in an oblique manner across the fracture site. It is best to direct the pins cephalad from the metaphysis into the epiphysis rather than antegrade from the epiphysis in to the metaphysis (Fig. 17J2-14A and B). These percutaneous techniques have the advantage of maintaining fracture alignment, with the arm in the normal position supported only with a splint or collar and cuff. Because of rapid healing, the pins can be removed within 2 to 3 weeks. There are many ­disadvantages of this technique. The first is that the pins may be ­difficult to insert because the shoulder must be abducted to maintain the reduction. Because of the oblique angle insertion that must be used, it is often difficult to gain entrance into the lateral cortex of the proximal humeral metaphyseal segment. The tips of the pins often keep slipping off the cortex. The other disadvantage is that the penetration of the pins through the substance of the deltoid muscle often inhibits the initiation of early motion. The antegrade ­insertion of the pins from the epiphysis to the metaphysis should be avoided because it will eliminate any chance of starting early motion as the pins bind up the rotator cuff (see Fig. 17J2-14C). In inserting the pins in this manner, there can be production of scar tissue in the cuff, which can result in a permanent loss of shoulder motion. The decision about whether to leave the ends of the pins outside the skin is usually left to the treating surgeon. More recently, cannulated screws have become a popular method of stabilizing the fragments (Fig. 17J-15).The use

of these cross screws to stabilize the fragments eliminates some of the disadvantages of using the oblique pins. However, these screws are not appropriate in those immature athletes who have more than 2 years of growth remaining because the threads in the screws apply compressive forces across the fracture site, which can produce a growth arrest. Intramedullary Nails

Retrograde intramedullary flexible nails also can be used in those fractures that require an anatomic reduction be maintained (Fig. 17J-16).56 The nails are inserted distally from the metaphysis in the supracondylar area. The entrance points can be either both through the lateral supracondylar column or one each through the medial and lateral supracondylar columns. Passage of the nails proximally using the medial and lateral approaches is easier but requires care to avoid injuring the ulnar nerve when inserting the nail in the medial column. The advantage of the ­ intramedullary technique is that the deltoid muscle is not violated and bound by the pins when they are passed percutaneously. Thus, circumduction motion can be initiated earlier when the nails are passed intramedullary. The disadvantage of the intramedullary technique is that the prominence of the nail ends at their entrance sites may inhibit motion at the elbow if they are left protruding too much.

Open Reduction On rare occasions, a satisfactory reduction may not be able to be obtained by a simple manipulation. In these situations, the surgeon will need to resort to performing an open reduction.

Shoulder 1079

A

B

C Figure 17J2-14   Percutaneous pins. A, An image of a Neer type IV displaced proximal humeral physeal fracture in the dominant extremity of a 13-year-old high-performance quarterback. B, Anatomic reduction was achieved by closed reduction and was secured by percutaneous pin fixation placed in a retrograde direction.

Rarely Indicated

Primary open operative reduction just to improve the position has almost no role in treatment of this fracture. The major absolute indications are the rare open fractures and fractures with vascular injuries. Other relative indications include comminuted intra-articular Salter-Harris type III and IV injuries and cases in which large amounts of periosteum or the biceps tendon have become interposed in the fracture site (Fig. 17J2-17). In the high-performance athlete, the risks of an open surgical intervention must be weighed against the need to obtain a nearly anatomic reduction. Poor Results

series of 39 patients who were treated nonoperatively. In the 9 patients who were treated operatively, there was a high rate of complications. The deficiency in this review is that they did not evaluate specifically the patients who used their upper extremities for high-performance activities. Also, it can be said that those undergoing surgical intervention had more severe injuries.

Subjective and Objective Findings Unfortunately, closed reduction alone is undependable in ensuring that the reduction obtained will be maintained. There are many other series35,36,44,45,55,57 that reported

In general, the open reduction of these fractures has produced poor results, which often are worse than the results of comparable fractures managed closed.54 Nilsson and Svartholm37 believed that the poor outcomes seen after an open reduction were the result of the surgical intervention and not the fracture. This conclusion was echoed 20 years later by Baxter and Wiley,44 who stated that open reduction improved the displacement in only three of seven patients, inflicting a cosmetically unattractive scar for no obvious advantage.

Weighing the Evidence No Controlled Studies Unfortunately, there are no double-blind controlled or comparative studies to compare the results of operative (which includes closed reduction and percutaneous fixation) with pure nonoperative management. The most recent work to confirm the good results of nonoperative management is the article by Beringer and coworkers.34 They demonstrated excellent results in their

A

B

Figure 17J2-15   Percutaneous screw fixation. A, Injury image of a 15-year-old male basketball player who sustained the Neer stage II fracture of his dominant shoulder. B, Following the reduction, the fracture was stabilized with a cannulated screw placed percutaneously. (Courtesy of Dr. John Edeen, San Antonio, Texas.)

1080 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

3-4 cm

1-2 cm

1 cm

A

B

D

E

C

F

Figure 17J2-16   A, Skin incision and entrance sites. The skin is incised for 3 to 4 cm over the lateral aspect of the distal humerus starting about 1 cm proximal to the prominence of the lateral condyle. The two entrance sites (dotted circles) are placed 1 to 2 cm   apart on the anterior lateral surface of the lateral supracondylar column. If medial-lateral entrance sites are used, a separate anteromedial incision (bold circle) is also placed over the medial supracondylar column to make the medical entrance site.   B, First entrance site. The first site is usually the most proximal site. The awl is started first at 90° until it engages the cortex. It is then angled cranially gradually 45° as it drills through the cortex. C,The first nail is inserted through its entrance site and advanced proximally. The second entrance site is then made distal and anterior with the awl. The handle of the awl is leaned against the first nail to guide it until it has penetrated the cortex. (Note: This is a lateral projection.) D,The second nail is inserted and both nails are advanced proximally to the fracture site. E, Reduction and fixation. The fracture fragments are reduced by abducting and slightly externally rotating the distal fragment. Each nail is then separately advanced into the proximal fragment. F, Final position. Once both nail tips have been secured in the head, the pins are cut distally. Notice the tips have the desired divergence. (From Dietz HG, Schmittenbecher PP, Slongo T, Wilkins KE: AO Manual of Fracture Management, Elastic Stable Intramedullary Nailing (ESIN) in Children. Davos Platz, Switzerland, AO Publishing, 2006, pp 25-30.)

good results with nonoperative management alone. In some of these series, there were reports of both subjective complaints and objective findings of minor decreases in shoulder motion in the older patients. The major question is how these minor deficiencies affect the function of the upper extremity in the high-performance athlete. Are the athletes who are near skeletal maturity (the commonly accepted age for males is 15 years58) able to return to their preinjury level of performance?

Results after Operative Intervention Dobbs and coworkers58 recently studied a group of patients 15 years and older in whom a Neer stage I or II reduction was obtained by open or closed method followed by pin or screw stabilization. In all the patients who were examined 4 years after surgery, there was near-normal glenohumeral motion and excellent strength. The important finding in this series was that all of these patients were able

to regain preinjury functional use of the involved upper extremity.

Specific Poor Nonoperative Results The first citation in the literature that relates to poor results in the high-performance athlete has been described by Dameron and Rockwood.48 This involved a 14-year-old track star who sustained a displaced Salter-Harris II fracture of the proximal humerus while pole vaulting. At the time of healing, there was a small proximal lateral spur on the anterolateral aspect of the metaphysis due to proximal migration of the distal fragment. When the patient had fully recovered, he was able to play football but was unable to participate in throwing sports because of a restriction of about 20 degrees in flexion and abduction. Sherk and Probst55 also reported loss of shoulder motion along with high-performance function in the older athletes who had less than an anatomic reduction.

Shoulder 1081

A

B

Figure 17J2-17   Open reduction. A, Image taken after an attempted closed reduction of an almost skeletally mature male demonstrated a persistent gap between the proximal metaphyseal and epiphyseal fragments. (The anteroposterior and lateral radiographs are seen in Fig. 17J2-10A and B.) The decision was made that the sporting activities of this athlete required as close to an anatomic reduction as possible. B, At open surgery, the periosteum and biceps tendon were found to be interposed between the fragments. Once this tissue was removed, the fragments were reduced anatomically and secured with cannulated screws.

The Author’s Experience

of shoulder motion 18 months after fracture to require an osteotomy to regain her motion to a functional level (Fig. 17J2-19). Thus, there is both direct and indirect evidence to support the concept that with the high-performance throwing athlete nearing maturity, an aggressive surgical approach may be necessary to obtain and maintain an anatomic reduction.

I had experience with a high-performance pitcher who was treated nonoperatively and had a minor residual deformity (Fig. 17J2-18). After the fracture, he had enough residual restriction of motion that he could no longer pitch ­ effectively and had to change to playing the outfield. Another young pubertal girl whose fracture was initially inadequately reduced had sufficient restriction

A

B

Figure 17J2-18   High-performance athlete. Anteroposterior (A) and lateral (B) injury radiographs of a high-performance 14-yearold baseball pitcher who sustained a displaced proximal humeral physeal fracture.

1082 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

D C Figure 17J2-18, Cont’d   Anteroposterior �(C) and lateral (D) radiographs taken 6 months after the fracture show some incongruity of the proximal humerus. Although mild, this injury was severe enough to interfere with his ability to pitch at a high level.

A

C

B

Figure 17J2-19   Malunion. A, Injury film of an adolescent female basketball player who sustained a significantly displaced proximal humeral physeal injury. The proximal spike of the metaphyseal fragment (dotted line) was very prominent (arrow). B, Eighteen months later, she still could not abduct past 120 degrees, which interfered with her basketball skills. The prominence of the metaphyseal spike remained (arrow). C, A valgus osteotomy was performed, which corrected the varus of the head and allowed her to assume full abduction. (Photo courtesy of Dr. Earl A. Stanley, Jr, University of Texas Health Science Center at San Antonio.)

Shoulder 1083

Author’s Preferred Method

of

Treatment

Minimally Displaced

The Neer I or II fractures can usually be treated with a simple sling and allowing the initiation of early active motion when the pain and swelling have subsided. Because this involves the physis, the healing process is very rapid, and there is the laying down of callus very early. This early callus ­stabilizes the fracture fragments to allow the initiation of early motion. It must be emphasized that this must be active motion at all times. Passive motion should never be used in the early recovery process. Significantly Displaced

In markedly displaced physeal fractures, sufficient reduction must exist to ensure uninhibited motion of the ­glenohumeral

A

D

joint. In the athlete close to maturity, this may be as ­ little as a Neer I displacement pattern. This is especially true for those athletes who need full glenohumeral motion for throwing, gymnastics, or swimming activities. In such cases, an anatomic reduction is achieved first by manipulation. In the past, I had used percutaneous pins or screws to stabilize these fractures. In inserting these devices, the deltoid muscle is violated, which can delay the resumption of shoulder motion. Recently, I have found intramedullary flexible nails to be more satisfactory for fixation (Fig. 17J2-20). They allow the patient to start motion almost immediately and do not have risk for infection associated with pins protruding from the skin. In inserting the pins in the distal humerus, I prefer to use the separate medial and lateral entrance sites rather than the two ­lateral entrance sites. See Figure 17J2-16.

C

B

F

E

G Figure 17J2-20   Author’s retrograde nailing. A, The injury image of a mature 14-year-old female swimmer who sustained this Neer grade IV fracture of her left proximal humeral physis. B, The fracture was able to be reduced to a nearly anatomic position by abducting the extremity. C, When the extremity was allowed to return to the side of the body, the original fracture malalignment recurred. Because it was determined that a nearly anatomic alignment of the proximal humerus would need to be achieved for her to regain her ability to perform her swimming activities on a high level, the decision was made to secure the fragments in a nearly anatomic position with retrograde nails. ������������������������������������������������������������������������������������� The AO technique �������������������������������������������������������������������� (as demonstrated in Figure 17J2-16) was used; both nails are passed retrograde up to the fracture site. D, With this athlete, the nails were advanced retrograde from separate medial lateral entrance sites. E, Once the nails had been advanced retrograde to the fracture site, the fracture was reduced by abducting the extremity. F and G, After the reduction, the nails were then advanced into the proximal fragment a sufficient distance to provide adequate stabilization. Stabilization is enhanced by providing separation of the tips. Continued

1084 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Author’s Preferred Method

of

Treatment—cont’d

I have found that external immobilization methods are cumbersome and uncomfortable and are not well accepted by the patient, in addition to being unreliable in maintaining the position of the fragments. Open Reduction

In my experience, I have found few indications for an open reduction. In the very rare open fractures or fractures with vascular compromise, I would not hesitate to perform an open reduction. In an athlete in whom there is interposed tendon or other tissue, one would have to weigh the risks of surgical damage to the shoulder ­muscles against the ben-

efits of achieving an anatomic reduction. In my limited experience, the most common tissue interposed has been the ­local periosteum. The diagnosis of interposed tissue is made by having a persistent gap between the fragments when attempting a closed reduction. This interposed tissue can limit shoulder motion by either the persistent incongruity of the proximal humerus or tethering the long head of the biceps. Once the decision has been made to achieve as anatomic a reduction as possible, an open reduction with internal fixation using a deltopectoral approach (see Fig.17J-17A) usually achieves an adequate reduction.

Postfracture Management

Criteria for Return to Athletic Activity

Nonoperative Cases

Patients can return to their athletic activities in a noncompetitive setting as soon as they have regained nearly full range of motion of the shoulder. They should not be allowed to return to their athletic activities in a competitive setting until the upper extremity has regained full strength. A good parameter for determining the return of upper extremity strength is when the athlete can perform 10 to 20 push-ups.

Initial Rest In those cases managed totally nonoperatively, the patient is supported in a sling and allowed to wait until the pain and swelling have subsided enough to start early shoulder motion. Initially the motion is in the form of circumduction exercises. This type of motion can usually be initiated after the first week. Once early callus begins to appear, the athlete can begin to actively abduct the shoulder. This usually consists of the patient starting to walk the fingers up a wall. It must be emphasized that all this activity is in the form of active motion. In these early stages, passive motion should never be employed.

Physical Therapy Once the athlete is able to actively abduct the shoulder to at least 90 degrees, formal physical therapy can be initiated. The emphasis needs to be on strengthening the muscles of the shoulder girdle as well as the other muscles of the upper extremity that have become weakened because of disuse.

Operative Cases In those cases in which there has been operative intervention, postoperative management is essentially the same. The initial rest period may be slightly prolonged in patients who have undergone an open reduction to allow the incision to heal before starting active motion. Usually, there is sufficient healing after a week to allow the amount of motion that the patient will use in the early stages. In patients in whom the fragments were stabilized with pins obliquely passed through the skin and deltoid muscle, the pins may inhibit the initiation of early motion. Usually, there is sufficient callus and internal stability by 3 weeks to remove the pins so that uninhibited motion can begin. This inability to initiate early shoulder motion is one of the major disadvantages of using the pins, which are directly inserted through the skin and muscle.

Complications The complications associated with proximal humeral fractures can be divided into nonskeletal and skeletal ­complications.

Nonskeletal Nonskeletal complications are rare with injuries in this area. Baxter and Wiley44 described one patient with complete disruption of the brachial artery at the lateral border of the axilla. In some patients, they also found severe tenting of the skin, which required operative intervention to prevent skin necrosis. Dameron and Reibel36 reported one case of brachial plexus paresis in a patient treated in a Statue of Liberty cast. Transient axillary nerve paralysis has also been described.47 Usually, however, these problems resolve by the time the athlete is ready to start the recovery or rehabilitation phase.

Skeletal The major skeletal complications include growth arrest, avascular necrosis, late malunion, and dislocation of the humeral head.

Growth Arrest This complication has been described in only one case in the recent English literature.46 Although there was a varus deformity of the humeral head and neck, the function of the upper extremity was the same as the uninjured ­extremity.

Shoulder 1085

Avascular Necrosis Although avascular necrosis of the humeral head is not uncommon in comminuted fractures of the proximal humerus in adults, it is extremely rare in the skeletally immature athlete. An article by Martin and Parsons59 described a case of a 14-year-old who had a Neer II physeal injury. The patient was not symptomatic until 7 months after the fracture. One year later, the lesion had healed with no significant deformity. In the two other cases described in the recent literature, the patients were asymptomatic, and there was no effect on the function of the shoulder.46,60 Lipscomb61 in 1975 described a case of localized avascular necrosis producing a condition similar to osteochondritis dissecans of the knee that developed in an athlete from a chronic stress injury involving the proximal humeral epiphysis. This resulted in a loose body that needed to be removed surgically.

Late Malunion Residual varus angulation and shortening, although they do occur, are rarely a problem among athletic individuals.44 In some adolescents in whom a fracture is not reduced anatomically just before the onset of skeletal maturity, there may be sufficient deformity to produce a disabling loss of motion. In these patients, the surgeon must weigh the benefits of surgical intervention against the risks of the procedure (see Fig. 17J2-19).

Figure 17J2-21   Many proximal humeral metaphyseal fractures present as a simple, nondisplaced incomplete fracture (arrow) with minimal greenstick angulation.

fractures involving the physis and the metaphysis. These differences are discussed in detail in the following ­section.

Anatomic Characteristics

Dislocation

Area of Remodeling

Almost all the physeal fracture patterns described in the major series of fractures of the proximal humeral physis are Salter-Harris I or II fracture patterns.4,34,36,44,53,55,58 In the recent literature, six cases involving fractures of the proximal humeral physis were associated with a dislocation of the shoulder.46,60,62-65 These cases all had Salter-Harris III or IV fracture patterns. Salter-Harris III and IV fractures are rarely seen in the proximal humerus. The reason given is that the shoulder has such a free range of motion. The Salter-Harris III and IV patterns are more commonly seen in those joints in which the motion is limited to only one plane.46 Often it is difficult to determine the exact position of the epiphyseal fragment on the routine radiographic images. Usually either CT or magnetic resonance imaging (MRI) is needed to determine the exact location of the proximal and distal fragments. With the exception of one case,46 the epiphyseal fragment was situated anterior in the joint. Of the six cases reported, four required an open reduction

When bone is laid down at the physis, it is produced as cancellous or quantity bone. This is why the cross-sectional area of this region is greater. With time, this quantity bone is remodeled into the stronger cortical bone of the ­diaphysis. Thus, we have in the diaphysis quality bone. Because the bone in the metaphyseal area is undergoing remodeling, its cortex is inherently thinner and also weaker. Thus, it is less resistant to compressive and tension forces. As a result, we see in the pediatric age group failure patterns such as minimally displaced greenstick-type fractures (Fig. 17J2-21).

METAPHYSEAL FRACTURES OF THE PROXIMAL HUMERUS In most of the articles and textbooks dealing with fractures of the proximal humerus, there is little differentiation between those fractures involving the physis and the proximal metaphysis. Unfortunately, the structure, incidence, and treatment principles differ considerably between those

Pathologic Fractures The metaphysis is also the most common location for a unicameral bone cyst. As a result, the incidence of pathologic fracture is frequent in this area (Fig. 17J2-22). In Kohler and Trillaud’s review covering 20 years of proximal humeral fractures,4 one fourth of the metaphyseal fractures occurred through unicameral bone cysts

Muscle Forces Because the location of the fracture line is more distal in the metaphysis, it is at the insertion of the pectoralis major tendon on the proximal humerus. Thus, this tendon is situated on both fragments. There is usually enough of the insertion on the proximal fragment to counteract the rotation of the proximal epiphysis such as is seen with the pure physeal fractures. The major portion of the tendon

1086 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

is situated on the proximal portion of the distal fragment. If the facture is complete, the force of this tendon on this distal fragment tends to force it anterior and medial. On the anteroposterior projection, the two fracture fragments usually are in a somewhat parallel alignment to lie in bayonet apposition (Fig. 17J2-23).

Extra-articular Fracture The distal location in the metaphysis places the fracture line well outside the joint capsule. As a result, any malunion of the fracture fragments will not impinge on the motion of the joint. Limited abduction of the shoulder is rarely seen following these fractures. This makes the metaphyseal fractures less likely to result in any loss of shoulder function in the high-performance athlete. This fact also has a bearing on the indications for surgical intervention to achieve an anatomic reduction.

Incidence Younger Age Group Proximal metaphyseal fractures are characteristically predominant in children younger than 10 years of age. The mechanism of the fracture probably is the same as that causing proximal humeral physeal fractures. The combination of a weaker metaphysis and a stronger perichondrial physeal ring probably accounts for the occurrence of the fracture in the metaphysis rather than the physis in this younger age group.

Figure 17J2-22   This fracture through a large unicameral bone cyst occurred after minimal trauma. The developmental defect greatly weakens the bone, making it susceptible to pathologic fracture. Coracobrachialis m. Supraspinatus m.

Subscapularis m.

E

Deltoid m.

M Pectoralis major m.

A

B

Figure 17J2-23   Metaphyseal muscle forces. A, The pectoralis major is the main deforming force on the distal metaphyseal fragment. The deltoid muscle tends to decrease the rotation on the proximal fragment. B, Image of a complete fracture of the proximal humeral metaphysis in which the proximal and distal fragments are aligned in parallel bayonet apposition.

Shoulder 1087

Less Concern in the Athlete Because metaphyseal fractures are usually seen in the first decade, they are only rarely a factor of concern in the older high-performance athlete.

Signs and Symptoms As stated previously, the degree of swelling is dictated by the amount of displacement of the fracture fragments. The simple undisplaced metaphyseal fractures may have minimal swelling with no deformity. In those fractures that are completely displaced, there will be considerable swelling and discomfort. Because of the swelling, it may be difficult to palpate the ends of the fracture fragments. In those that are completely displaced and the fragments are in bayonet apposition, it is theoretically possible to palpate the fracture surfaces as being more distal than with the pure physeal fractures. From a practical standpoint, the young athlete is so uncomfortable that he or she will not allow much in the way of palpation at the fracture site. Thus, the differentiation between the two types of fracture patterns is dependent on the radiographic image (see Fig. 17J2-11).

Classification of Metaphyseal Fractures Ogden47 classifies metaphyseal fractures into two types. In the first type, the cortex remains intact. These are usually torus or greenstick fractures (see Fig. 17J2-21). In the ­second type, there is loss of cortical integrity with either angular or translocation displacement (see Fig. 17J2-23).

Radiographic Studies These fractures are usually straightforward and obvious on routine radiographic projections. The standard lateral projection, which is usually obtained by abducting the shoulder, may be difficult to obtain in the acute situation because of pain and swelling. In this situation, the transthoracic view may be used. The transthoracic projections are sometimes difficult to evaluate. There is almost never an occasion to obtain special studies such as CT for these fractures.

Treatment of Proximal Humeral Metaphyseal Fractures With few exceptions, metaphyseal fractures of the proximal humerus can be managed conservatively. There are very few operative indications.

Nonoperative Management Undisplaced Fractures Simple metaphyseal fractures that are undisplaced or only minimally angulated usually can be treated quite adequately with a collar and cuff. In the acute stage, added comfort may be achieved by binding the arm to the chest with a circular elastic bandage. Most of these fractures are intrinsically stable; thus, shoulder motion can be initiated early. In the pediatric athlete, it is extremely important to regain

A

B

Figure 17J2-24   Bayonet remodeling. A, Injury film of an 8year-old baseball player who sustained a proximal metaphyseal fracture. The bayonet apposition and shortening were accepted. Initially, the boy was treated with a simple collar and cuff, supplemented with an Ace bandage. B, Radiograph obtained 4 months after injury shows remarkable remodeling. The range of motion and strength of the shoulder have since returned to the preinjury level.

shoulder motion as soon as possible to achieve maximal rehabilitation.

Displaced Fractures In the completely displaced metaphyseal fracture, little abduction of the proximal fragment is usually present because some adduction force is maintained by both the pectoralis major and the latissimus dorsi on the proximal metaphysis (see Fig. 17J2-23). As a result, these fractures often develop bayonet apposition. Although shortening occurs because of the longitudinal pull of the triceps, biceps, and deltoid muscles, it is usually not sufficient to cause any functional or cosmetic residua. These fractures usually can be treated quite well using a collar and cuff plus a thoracic elastic bandage. Healing with the bayonet apposition (although it may initially concern the parents) usually results in an acceptable cosmetic and functional result, even in the adolescent patient (Fig. 17J2-24).

Operative Indications Olecranon Traction About the only time treatment with overhead olecranon traction would be warranted would be with the multipleinjury trauma patient who is confined to a recumbent ­position. This situation would be extremely unlikely to occur with the immature athlete. The only other possible indication would be severe comminution of the metaphyseal fragments. The traction maintains the reduction until there is sufficient callus and intrinsic stability at the fracture site

1088 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

to allow the traction to be discontinued. Once the traction is removed, the upper extremity can be supported by no more than a collar and cuff.

Percutaneous Stabilization Because the nonoperative methods produce such satisfactory results, about the only indication for pins would be in the athlete who is at or near maturity. As with the physeal fractures, the reduction is usually first obtained by closed methods and then stabilized with pins. These can be placed directly across the fracture fragments or retrograde with intramedullary pins.66 One special indication for surgical stabilization would be a fracture in either the ipsilateral or contralateral upper extremity, in which ­surgical stabilization would facilitate early mobilization (Fig. 17J2-25).

Open Reduction

A

Other than the rare open fracture, there is probably no indication to perform an open reduction of a proximal humeral metaphyseal fracture. The only possible ­exception is ­treatment of an athlete who is approaching skeletal maturity in whom an anatomic reduction is deemed necessary for the athlete to achieve full functional recovery.

B

Weighing the Evidence

Figure 17J2-25   Metaphyseal intramedullary pins. A, Radiograph of an 8-year-old girl who sustained a contralateral displaced lateral condyle in addition to this displaced proximal humeral metaphyseal fracture. B, In an effort to facilitate early recovery of both injuries, the proximal humeral metaphyseal fracture was initially stabilized with retrograde intramedullary pins.

Authors’ Preferred Method

of

The absence of articles devoted exclusively to the treatment of metaphyseal fractures of the proximal humerus makes the evaluation of any method of treatment based on the principles of evidence-based medicine impossible. Fortunately, in the citations dealing with the overall fractures of the proximal humerus, metaphyseal fractures heal well with any method of treatment.

Treatment

For undisplaced or minimally displaced fractures of the proximal humeral metaphysis, I use a collar and cuff supplemented with an elastic bandage, and I strap the extremity to the chest wall. Within a few days, the elastic bandage strap is discontinued, and the cuff is gradually lowered until the elbow is at 90 degrees. At this time, circumduction exercises of the shoulder are begun. The remainder of the postoperative course is carried out the same as for the physeal fractures. In those complete metaphyseal fractures that are in bayonet apposition, I usually accept this position and treat them similarly to the minimally or undisplaced fractures. Patients with these fractures may be slower to recover their range of motion. It usually requires a major degree of explaining to convince the parents that these will completely remodel. I usually keep pictures or radiographs of previous cases handy to convince them (Fig. 17J2-26). In patients who are very near skeletal maturity, I usually do a closed reduction and then secure the fragments with percutaneous intramedullary pins (Fig. 17J2-27). I have only one instance in which I treated a proximal humeral metaphyseal fracture with an open reduction. This patient had an associated closed head injury which made the

A

B

Figure 17J2-26   Full bayonet remodeling. A, Image of a young child who had sustained a complete fracture of the proximal humeral metaphysis 6 weeks previously. The bayonet apposition was accepted. B, Two years later, the proximal humerus had completely remodeled. This is a good example of the remodeling capacity of the proximal humerus.

Shoulder 1089

Authors’ Preferred Method

A

D

of

Treatment—cont’d

C

B

Figure 17J2-27   Author’s intramedullary metaphyseal nailing technique. Anteroposterior (A) and lateral (B) injury images taken of a high-performance tennis player who was involved in an automobile accident. Fortunately, this was his only major musculoskeletal injury. Because of his need to have an anatomic reduction to ensure his continued high-performance upper extremity function, it was elected to attain as anatomic a reduction as possible and stabilize it with retrograde intramedullary flexible nails. Anteroposterior (C) and lateral (D) images demonstrating the position of the ends of the nails separated in the humeral head.

usual closed treatment impossible. It was a difficult procedure. Had there not been an associated head injury, I would have treated this patient by the usual closed methods. It was difficult to determine whether the reduction obtained by an open procedure produced any better functional results in the long term over accepting the inevitable malreduction that was developing in this patient. The rehabilitation protocol and criteria for returning to competitive athletic activities are essentially the same as outlined following physeal fractures of the proximal humerus.

The complications associated with fractures of the proximal humeral metaphysis are much less than those associated with the fractures of the proximal humeral physis. There is much less loss of shoulder motion. The problems associated with growth disturbance are essentially nonexistent. The shortening that is seen with the fractures that heal in bayonet apposition is usually not of any functional or cosmetic significance. (Table 17J2-1)

  TABLE 17J2-1  Typical Findings: Differences between Proximal Humeral Physeal and Metaphyseal Fractures Age Fracture location Radiographic findings Treatment options

Proximal Humeral Physis

Proximal Humeral ­Metaphysis

More common after 10 years of age Intra-articular, thus more malunion may alter motion Fragments angulated to each other Metaphyseal fragment usually lies lateral to the epiphyseal fragment May require surgical reduction and stabilization in those patients near skeletal maturity

Usually seen before the age of 10 Extra-articular, thus malunion less likely to affect shoulder motion Fragments lie parallel in bayonet apposition Metaphyseal fragment usually lies medial to the epiphyseal fragment Extremely rare to require any surgical intervention

1090 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

AVULSION OF THE LESSER TUBERCLE Avulsion of the lesser tubercle by the subscapularis tendon is uncommon in the skeletally immature patient.67-70 Levine and coworkers provide the most extensive analysis of this type of fracture in their case report and review.57

Mechanism The most commonly reported mechanism is one of a forced external rotation against a resisting force. This can occur in wrestling.57

BOX 17J2-3 Typical Findings: Avulsion of the Lesser Tubercle

• This injury is rare in the skeletally immature. • Often the diagnosis in unappreciated on the

initial evaluation. • Radiographic findings may be subtle. • Most require the establishment of an anatomic ­reduction. • If left unreduced, it may result in significant loss of shoulder motion. • Because there is the need to achieve soft tissue healing, the recovery period may be slower and prolonged.

Diagnosis Late Diagnosis These fractures may be difficult to diagnose acutely. Levine and coworkers,57 in their review of the published reports of this injury, found that 50% of the cases reported in the skeletally immature and adolescents were diagnosed late. These were patients who presented with chronic shoulder pain following an injury.

Clinical Appearance There may be point tenderness over the anterior aspect of the shoulder. The patient’s apprehension is accentuated by abducting the shoulder to 90 degrees and attempting to externally rotate the shoulder. This apprehension can be lessened with shoulder stabilization using downward pressure on the upper arm in the supine position.57

Imaging Studies The findings may be subtle on the routine radiographs. The axillary view provides the best image of the fragment.71 MRI or CT usually is necessary to adequately assess the displacement of the tubercular fragment.57

Treatment Nonoperative Management If the fragment is essentially nondisplaced, it can be obtained with nonoperative management. This requires close follow-up to be sure that the fragment does not displace late.72

Operative Management Because of the fear of subsequent nonunion, malunion, or anterior impingement, it has been recommend by some that these fragments be replaced surgically.57 If untreated, these avulsions by the subscapularis and shoulder capsule can result in loss of shoulder motion. This can be disabling in a high-performance athlete who uses that upper extremity for throwing. Once the exostosis has developed, it may require surgical removal to permit full recovery of the shoulder motion. For this reason, some authors57,67,70 have

recommended primary reattachment of the tubercle in all young athletes if recognized acutely. The tuberosity is reattached using a variety of drill holes, suture anchor fixation, or small fragment screw fixation. Care must be taken in the skeletally immature athlete to avoid procedures that can create a premature growth arrest. In general, the results with acute fixation of the fragment have enabled a full return to athletic function in 3 to 6 months.57,71,73,74

Weighing the Evidence Because of the rarity of this injury, there have been no double-blind comparison studies. In the long-term review of their few cases, Ogawa and Takashi72 reported superior results in those treated operatively.

Postoperative Management The reports in the literature are simple case reports, so there are few guidelines to the postoperative management. Because this involves a soft tissue and bone repair, there is a longer healing time than with just fracture healing. Levine and coworkers57 placed the patient initially in a sling. Active shoulder internal and external rotation was restricted to 45 degrees for the first 6 weeks. The patient’s range of motion and strength were then gradually increased with aggressive physical therapy. By 4 months, when the motion and strength had returned to normal, the athlete was allowed to return to full competitive activity (Box 17J2-3).

STRESS FRACTURES OF THE PROXIMAL HUMERAL PHYSIS One area of skeletal weakness that is susceptible to failure with repeated microtrauma is the proximal humeral physeal plate. The failure usually occurs as a stress fracture of the proximal humeral physis.

Incidence Because rotational forces applied to the shoulder are especially prevalent in the pitching activity associated with baseball, the proximal humeral stress fractures involving

Shoulder 1091

the physis are most commonly seen in the immature baseball player. This injury has also been seen in gymnasts,75 badminton players,76 and cricketers.77

In baseball, however, elbow problems predominate in immature players. Shoulder pain and chronic problems do not develop until adolescents are in their late teens.78-81 Some authors have speculated that the late incidence of shoulder problems is related to abnormal pitching patterns because of other chronic elbow conditions that have developed during the immature athlete’s earlier years.78,81

behind. Next, the forearm and hand are whipped forward, owing in large part to forces generated by the pectoralis major and latissimus dorsi. The third, or follow-through, phase involves coordination of the forearm muscles to release the ball at the proper time and with the proper spin. The deceleration forces generated in this phase are unique to baseball and tennis. This phase puts stretch on the posterior capsule and ­ external rotators that can be a source of the posterior shoulder pain syndrome. Richardson29 found that during this phase there might also be stress on the rhomboids and levator scapulae insertions, which produces pain along the medial scapular border.

Little Leaguer’s Shoulder

Both Rotational and Compressive Forces

One area of skeletal weakness that is susceptible to failure with repeated microtrauma is the proximal humeral physeal plate. The failure usually occurs as a stress fracture of the proximal humeral physis. Dotter50 first described this entity as little leaguer’s shoulder in 1953. Since then, numerous cases have been described in the literature.60,61,66,82-85 All these cases occurred in high-performance male ­pitchers who were 11 to 13 years old. In the cases presented, all except one responded to rest for the remainder of the season plus a vigorous preseason conditioning program the following year.86 In only one case was operative intervention necessary. This was an individual described by ­Lipscomb61 in whom a localized avascular necrosis of the epiphysis developed, producing a loose body that had to be removed surgically.

Tullos and King88 found that the pitching patterns among adolescents and adults were remarkably similar. The forces generated during pitching are very large, especially when rotation is considered. Gainor and coworkers89 pointed out that pitching involves both rotational forces from the internal and external rotators of the shoulder and compressive forces from the flexors and extensors of the elbow. They calculated that internal rotational torque is 14,000 inchpounds just before release of the ball. The kinetic energy produced is 27,000 inch-pounds during the throw. These forces are four times greater than those generated in the lower extremity when an athlete is kicking a ball. In addition, they are greater than the forces required to fracture an isolated cadaver humerus. These are usually repeated rotational forces applied to the physis.

Shoulder Problems among Adolescents

Causative Factors Repetitive Rotational Forces This is usually the result of the repeated rotational forces that develop at the proximal humerus in the immature athlete. The opposing proximal muscular attachments (rotator cuff) and deltoid, pectoralis major, and triceps attachments distally render the extracapsular proximal humeral physis particularly vulnerable to repetitive ­rotational ­microtrauma.87

Forces around the Baseball Pitcher’s Shoulder Pitching Phases Tullos and King88 divided pitching activity into three phases (Fig. 17J2-28): cocking, acceleration, and follow-through. First is the cocking phase, in which the shoulder is markedly externally rotated. This tightens the triceps and biceps as well as both the internal and external rotators across the shoulder. This part of the throwing act in the adolescent pitcher results in increased external and decreased internal rotation arcs in the shoulder. Richardson29 pointed out that during this phase, the internal rotators and adductors are at maximal stretch. If the body or shoulder moves forward too soon, the arm has to catch up by putting an excessive load on these structures, thereby creating an inflammatory tendonitis that is the most common cause of anterior shoulder pain among adolescent pitchers. The second, or acceleration, phase consists of two parts. First, the shoulder is brought forward with the forearm

Pitching Techniques In addition to chronic repetitive rotational and compressive forces across the shoulder, there appear to be other factors that may create microtrauma in young, skeletally immature pitchers. Albright and associates78 observed in their extensive study of little league pitchers that the incidence of symptoms reflected the form of pitching rather than the age of the pitcher. Those who had poor pitching skills were more likely to become symptomatic. For this reason, Slager81 advised that the first emphasis of immature pitchers should be on the development of skills and control; as they mature, emphasis can then be placed on increasing the speed of pitching.

Social Factors Social pressures can also be a factor. Torg and associates85 found that in comparable age groups, those who performed in a less competitive environment were less likely to develop symptoms in the throwing arm than those who were subjected to high competitive pressures.

Signs and Symptoms Clinical Presentation and History Patients complain of gradually increasing lateral shoulder pain with aggravation by throwing and later at rest. There is usually no history of trauma or neurologic involvement.

1092 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

Cocking phase

B

Acceleration phase 1st stage

D

Follow-through phase

Acceleration phase 2nd stage

Figure 17J2-28   Phases of pitching. A, Cocking phase. B and C, Acceleration phase. D, Follow-through phase. (From Woods GW, Tullos HS, King JW: The throwing arm: Elbow joint injuries. J Sports Med 1[Suppl 4]:45, 1973.)

Physical Examination and Testing Physical examination shows localized lateral tenderness to palpation over the proximal humerus with painful external rotation. Loss of range of motion and swelling are uncommon findings.

Imaging The common radiographic finding is a widening of the proximal humeral physeal plate (Fig. 17J2-29). Associated findings in about half of patients may include demineralization, sclerosis, cystic changes, and lateral fragmentation of the proximal humeral metaphysis.90 MRI and bone scan studies do not add information but may be more sensitive in the early stage of physeal injury when plain radiographs remain normal. Oblique coronal T1-weighted images reveal widening of the proximal lateral humeral physis and

increased signal intensity indicative of periosteal and bone marrow edema on T2-weighted images. Follow-up MRI or bone scan studies in an uncomplicated course are not indicated.

Treatment These fractures usually respond well to rest for the remainder of the season. There must also be a vigorous muscle conditioning program. The pitching technique may need to be critically evaluated and any errors corrected.

Complications If the player continues to participate by ignoring the pain, a slippage of the proximal physis can occur75 (Box 17J2-4).

Shoulder 1093

A

B

Figure 17J2-29   Baseball shoulder. A, This 13-year-old high-performance little league pitcher experienced pain while throwing toward the end of the season. This radiograph demonstrates widening of the physis (arrows), which is indicative of a stress fracture through the physis, or baseball shoulder. B, Radiograph of normal left side for comparison.

BOX 17J2-4 Typical Findings: Stress Fractures of the Proximal Humeral Physis

• Commonly referred to as “baseball shoulder” • Due to repetitive rotational forces applied

to the ­ roximal humeral physis p • Manifest by local tenderness with pain accented by ­rotation of the shoulder • Radiographically demonstrated as widening and ­fragmentation of the physis • Treated by prolonged rest and correction of pitching techniques • If not treated early, the proximal humeral physis can slip, producing joint incongruity.

C l Most

r i t i c a l

P

o i n t s

injuries from sporting events in the pediatric age group occur in males in nonorganized sporting events. l The weakest periosteum of the proximal humerus is over the anterolateral aspect of the metaphysis. This is where the initial failure occurs with fractures involving the proximal humeral physis. l Most fractures of the proximal humeral physis are SalterHarris I or II failure patterns. l Fractures of the proximal humeral physis occur more commonly in the first decade of life. Fractures involving the proximal humeral metaphysis occur more commonly in the second decade of life. l Fractures involving the proximal humeral physis can usually be managed conservatively up until about 3 years before skeletal maturity. Once the athlete approaches skeletal maturity, efforts should be directed toward obtaining and maintaining as near anatomic reduction as possible. Failure to accomplish this in the high-performance athlete can result in loss of function of the upper extremity. l Because fractures of the proximal humeral metaphysis are extra-articular and occur in the younger age group, they rarely require operative intervention.

l The proximal humeral metaphysis often contains pathologic bone such as unicameral bone cysts. This predisposes this area to fractures occurring with minimal trauma. l Avulsions of the lesser tubercle of the proximal humerus often present late as the initial radiographic findings may be subtle and overlooked. l Most avulsion fractures of the lesser tubercle require an open reduction to obtain an anatomic alignment. l The management of stress fractures (little leaguer’s shoulder) of the proximal humeral physis requires a prolonged period of rest as well as an examination of the pitching techniques of the pediatric athlete.

S U G G E S T E D

R E A D I N G S

Adams JE: Little league shoulder: Osteochondrosis of the proximal humeral epiphysis in boy baseball pitchers. Calif Med 105:22-25, 1966. Beringer DC, Weiner DS, Noble JS, Bell RH: Severely displaced proximal humeral epiphyseal fractures: A follow-up study. J Pediatr Orthop 18:31-37, 1998. Dobbs MB, Luhmann SL, Gordon J, et al: Severely displaced proximal humeral epiphyseal fractures. J Pediatr Orthop 29:208-215, 2003. Levine B, Pereira D, Rosen J: Avulsion fractures of the lesser tuberosity of the humerus in adolescents: Review of the literature and case report. J Orthop Trauma 19:349-352, 2005. Neer CS, Horowitz BS: Fractures of the proximal humeral epiphyseal plate. Clin Orthop 41:24-31, 1965. Sanders JO, Cermack MB: Fractures of the proximal humerus. In Rockwood CA, Matsen FA, Wirth MA, Lippitt SB (eds): The Shoulder, 3rd ed. Philadelphia, WB Saunders, 2004, pp 1307-1325. Sarwark JF, King EC, Luhmann SJ: Fractures of the proximal humerus. In Beaty JH, Kasser JR (eds): Rockwood and Wilkins’ Fractures in Children, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2006, pp 104-715. Webb LX, Mooney JF: Proximal humeral fractures. In Green NE, Swintowski MF (eds): Skeletal Trauma in Children, 3rd ed. Philadelphia, WB Saunders, 2003, pp 334-337.

R eferences Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 1094 DeLee & Drez’s� O

S e c t i o n

K

Adhesive Capsulitis Gary M. Gartsman and Matthew D. Williams

The stiff shoulder presents a persistent diagnostic and therapeutic challenge. “Peri-arthritis scapulo-humerale” was the first description of the stiff shoulder by Dulay1 in 1872, and manipulation under anesthesia was reported as the treatment of choice. Frozen shoulder, a term coined by Codman,2 provided a clear description of the presentation and examination of patients with stiff shoulder regardless of etiology. In a natural history study, Reeves reported on three phases of the frozen shoulder that persists over a course of 1 to 3 years: freezing, frozen, and thawing.3 Thickened capsular tissues led Nevaiser to describe adhesive capsulitis as a term indicative of pathologic alterations in the capsule of stiff shoulders.4 There are four basic conditions that produce shoulder stiffness: idiopathic adhesive capsulitis,3,5 endocrine disorders,6,7 post-traumatic stiffness,8 and postoperative stiffness (Box 17K-1).9 Idiopathic adhesive capsulitis is a painful and disabling condition believed to be self-limited that resolves after 1 to 2 years.5 The primary cause of this condition is unknown and remains controversial. Restriction on active and passive motion is due to capsular thickening and contracture.4 Although patients do improve over time, comparison to the uninvolved contralateral shoulder demonstrates residual limitation of movement and function.10 In addition, many patients suffering from pain and disabling loss of motion are not willing to undergo a protracted conservative course and request early operative intervention. Although patients with diabetes are at increased risk for development of shoulder stiffness, the pathophysiology of the contracture and its relation to diabetes is not understood.11 The stiffness in diabetic patients is reported to be more painful and more recalcitrant to conservative therapy than idiopathic capsulitis in the nondiabetic population.6,7 Effective treatment resulting in improvements in motion

Box 17k-1 conditions causing a stiff shoulder Idiopathic adhesive capsulitis Diabetes Thyroid disorder Degenerative arthritis Post-traumatic stiffness Rotator cuff tear Labral tear Fracture malunion Postoperative stiffness

and pain is attainable in diabetic adhesive capsulitis with arthroscopic release.6,12 The degree of pain and impairment in post-traumatic stiffness is related to the severity of the trauma. Postoperative stiffness may result from scarring in the area of surgical treatment (subacromial adhesions after rotator cuff repair, anterior glenohumeral capsular contracture after a Bankart procedure),9 and profound glenohumeral joint contractures can occur after surgical treatment that does not violate the capsule.12,13

RELEVANT ANATOMY AND BIOMECHANICS Glenohumeral joint stability is preserved through concavity-compression by balanced muscular forces of the rotator cuff.14 Passive restraints such as capsular tissues and the glenohumeral ligaments serve as checkreins that become taught at extremes of motion to resist abnormal glenohumeral rotation and translation and maintain a congruent joint.14,15 The stiff shoulder may be characterized by the direction of motion loss. For instance, global loss of mobility in idiopathic adhesive capsulitis is characterized by circumferentially rigid capsular tissues and glenohumeral ligaments. Throwing athletes with restricted internal rotation have contractures isolated to the posterior capsule. Loss of external rotation with the arm adducted points to rotator interval stiffening.16,17 Decreased external rotation in abduction exposes the anteroinferior structures as the cause for restricted movement.18 Understanding the function of soft tissue restraints of the glenohumeral joint under normal conditions aids in understanding patterns of motion restriction in stiff shoulders with capsular contracture. Releasing a tight capsule eliminates the impediment to shoulder motion and allows the primary stabilizing mechanisms, the rotator cuff, to keep the joint congruent. Capsular release results in increased translation of the humeral head in all planes of motion, but not an increased incidence of dislocation.19 The captured shoulder is characterized by extra­articular adhesions secondary to prior surgery.20 Severe subacromial bursal and subdeltoid adhesions clinically simulate the global loss of shoulder mobility found in the contracted capsule of idiopathic capsulitis patients. However, arthroscopic examination proves the intra-articular capsule benign and compliant. Circumferential inspection of the bursal surface is possible arthroscopically and allows thorough adhesion release.

Shoulder 1095

EVALUATION History and Clinical Presentation The classic presentation of patients with adhesive capsulitis, regardless of the underlying cause, is that of painful and restricted shoulder motion.13,21 Pain is constant and often unremitting, interfering with sleep and activities of daily living. Overhead activities and reaching behind the back is painful or impossible secondary to contracture. Rapid shoulder movement causes severe pain. Patients report decreased motion, both active and passive, that is often less than half the mobility of the opposite side (Box 17K-2 ).

Physical Examination Measurements of shoulder mobility using reproducible tests should be recorded for both active and passive motion. Active and passive motion recordings that are discordant point toward a diagnosis other than adhesive capsulitis. Evaluation of motion should always be compared with assessments made of the uninvolved ­extremity. Motion planes recorded are elevation, abduction, and external rotation with the arm in adduction and 90 degrees or maximal abduction. The highest vertebral level the patient can reach behind their back with the thumb extended measures internal rotation. The examiner must remember that scapulothoracic motion may increase the effective range of motion and should be held in check. Stabilizing the scapula with a hand on the inferior border will protect against scapulothoracic motion and provide more accurate measurement of glenohumeral motion. In addition to motion, a complete neurovascular examination, including muscle strength about the shoulder, should be included in the physical examination (see Box 17K-2).

Imaging Radiographs including anteroposterior, axillary, and supraspinatus outlet views should be obtained and are often normal. Plain films document any glenohumeral abnormality, including arthritis; post-traumatic, postsurgical changes; or disuse

Box 17k-2 history and presentation History No memorable inciting event Minimal traumatic event Recent surgery Concomitant medical issues Diabetes Thyroid dysfunction Presentation Pain and limited motion Night pain Pain with activities of daily living (particularly overhead or reaching behind the back) Severe pain with rapid shoulder movement

osteopenia.21 Arthrography is described as an adjunctive test for adhesive capsulitis (a contracted glenohumeral joint has a limited joint volume),22,23 but has no relationship to loss of motion.22 Magnetic resonance imaging allows the evaluation of associated pathology, including rotator cuff and labral tears. Additionally, reports in radiology literature correlate the thickness of rotator interval and inferior capsular tissues (3 to 4 mm) with the diagnosis of adhesive capsulitis.24,25

TREATMENT OPTIONS The treatment options for adhesive capsulitis range from nonoperative physical therapy to operative intervention, including manipulation under anesthesia and either arthroscopic or open capsular release. The efficacy of nonoperative treatment depends on the cause of the stiffness; post-traumatic and postsurgical stiffness seem to respond poorly to physical therapy.8,21 Regardless of the cause, a trial of conservative therapy is warranted for a period (we use 6 months as a baseline) before operative intervention (Box 17K-3).

Nonoperative Modalities In a prospective functional outcome study of conservative treatment of idiopathic adhesive capsulitis, Griggs and colleagues5 reported on 71 patients over a mean follow-up of 22 months. Passive exercises, including pendulums, forward elevation, external rotation, horizontal adduction, and internal rotation, were completed 5 times per day (91% participated in organized physical therapy). Pain scores at rest and with activity were significantly improved, as were the increases in range of motion attained. Ninety percent were satisfied with their outcome despite residual differences in motion between the affected and unaffected shoulders. Five patients were unhappy with the nonoperative results and underwent operative intervention. Other reports support the conclusion that nonoperative therapy is effective at reducing pain and increasing motion in patients with idiopathic adhesive capsulitis.10

Closed Manipulation Closed manipulation under anesthesia is generally regarded as the second-line treatment of the stiff shoulder after ­failure of nonoperative therapy. Nevaiser and Nevaiser22 and ­Harryman8 have described methods of closed ­manipulation. Manipulation is reported as effective in recovering motion in patients with adhesive capsulitis.26,27 The procedure is not without risk, with possible complications including humeral

Box 17k-3 treatment options Nonoperative Physical therapy Nonsteroidal anti-inflammatory drugs Corticosteroid injection (intra-articular) Operative Manipulation under anesthesia Arthroscopic capsular release Open capsular release

�rthopaedic ����������� S �ports ������ � Medicine ������� 1096 DeLee & Drez’s� O

fracture, dislocation, and soft tissue injury such as rotator cuff and labral tears and possible neurovascular injury. Manipulation alone without direct arthroscopic visualization does not allow for documentation of concomitant lesions or visualization of the effectiveness of the manipulation.

Arthroscopic Release Arthroscopic treatment of adhesive capsulitis is advantageous in that it enables visualization and release of both intra-articular and extra-articular adhesions without the need for open surgery that requires dividing the subscapularis tendon. Active range of motion can be initiated immediately after surgery, reducing the incidence of new scar tissue formation.28 Arthroscopic capsular release safely and reliably improves shoulder mobility in adhesive capsulitis.6,12,13,27,29-33 Postoperative time to full motion and activity is 3 to 6 months,30 and in contrast to those treated with nonoperative modalities, most patients regain motion similar to the contralateral extremity.6,8,13 In a direct comparison of manipulation and arthroscopic release, OgilvieHarris and colleagues27 reported similar gains in range of motion between the two groups but significantly improved pain relief and ultimate restoration of function in the arthroscopically treated group.

Open Release Advances in arthroscopic techniques have decreased the indications for open capsular release. Familiarity with intra­articular and extra-articular arthroscopic shoulder anatomy allows the surgeon the freedom to address all the components of the stiff shoulder without open surgery. However, knowledge of open capsular release techniques should also be familiar. Open capsular release is indicated when anatomy is distorted precluding safe arthroscopic release, when arthroscopic treatment fails, or in patients with postoperative or post-traumatic stiffness associated with malunited fractures or hardware requiring removal. The surgeon should be prepared to abort an arthroscopic release and proceed with open surgery in the event of a complication, inability to adequately discern tissue planes, or difficulty identifying landmarks for release. Several reports in the literature ­discuss the success of open release for capsular contractures.34-36 Key points in

open release for stiff shoulders include lysis of deltoid adhesions, rotator interval release,17,37 adequate subscapularis release or lengthening of the subscapularis if necessary, and inferior capsular release.

Weighing the Evidence Literature discussing the treatment of adhesive capsulitis supports the efficacy of all modalities: nonoperative, manipulation under anesthesia, and arthroscopic and open capsular releases. Treatment recommendations have evolved over time with improvements in surgical techniques and outcomes. Conservative therapy for adhesive capsulitis is effective owing to the self-limited nature of the condition; stiff shoulders will regain motion over time.5,10 However, the prolonged course of adhesive capsulitis pushes the patient and physician toward more aggressive treatment. Manipulation under anesthesia effectively loosens the stiff shoulder and helps regain motion without surgery.26,27 However, manipulation does not allow visualization of the joint, resulting in the possibility of incomplete release. Surgical intervention with open capsular release provides improved mobility after surgery, but at the expense of subscapularis transection and its associated risks.34-36 Arthroscopic capsular resection and adhesion release provide visualization of intra-articular and extra-articular pathology and allow the surgeon to address both. Patients proceed with active motion immediately after arthroscopic release and may regain nearly complete return of function.6,12,13,27,29,30-33 There are few direct comparison studies of treatment options in the literature, making it impossible to define the most efficacious modality. Therefore, nonoperative methods should be exhausted before operative intervention, and the surgeon should choose the most appropriate operative procedure, either open or arthroscopic, consistent with his or her comfort and skill level.

POSTOPERATIVE PRESCRIPTION Steroids Regardless of the cause of the stiffness, all patients are started on a methylprednisolone dose pack the day of surgery. Cortisone is not injected into the joint after capsular

Author’s Preferred Method Indications and Contraindications

Surgical intervention is discussed with the patient if after 6 months of appropriate nonoperative treatment there ­remains persistent pain and restricted motion. Severe stiffness is ­ defined as 0 degrees of external rotation and less than 30 degrees of abduction, whereas moderate stiffness is a decrease of 30 degrees in either plane as compared with the opposite shoulder. Restriction of internal rotation is not considered an indication for arthroscopic release despite the fact that it may be significant to the patient. The throwing athlete poses an exception to this rule in that posterior ­capsular contracture and restricted internal rotation may be

an ­ isolated problem readily addressed arthroscopically. If after 6 months of therapy, the patient reports decreased pain but continued stiffness, nonoperative treatment is prolonged for at least 2 more months. Our rationale is that the decrease in pain may herald the “thawing” phase and spontaneous resolution without surgery. Operative release is indicated if the contracture has not improved by 2 additional months. We proceed more aggressively with surgery if motion has not improved or worsens by 4 to 6 months after the initiation of therapy. In our practice, there are few contraindications to ­arthroscopic release: intervention during the inflammatory

Shoulder 1097

A u t h o r ’ s P r e f e r r e d M e t h o d — c ont ’ d or freezing phase of adhesive capsulitis, and significant fracture malunion requiring osteotomy or hardware requiring removal. Any surgery during the inflammatory phase could increase capsular irritation and accelerate or worsen contractures and should be avoided during this time. Treatment of traumatic stiffness from mildly malunited greater tuberosity fractures or proximal humerus fractures may be attempted arthroscopically. Previous open procedures or trauma often cause scar formation extra-articularly, anterior to the subscapularis. Although we routinely address these contractures with the arthroscope, release in these situations is difficult and requires in-depth knowledge of arthroscopic anatomy and skill (Box 17K-4).

in abduction and adduction are applied to the shoulder only if it responds to elevation and abduction. If external rotation improves, proceed with internal rotation in maximal abduction, followed by cross-body adduction, then behindthe-back internal rotation. Ensuring that the contractures respond to abduction and elevation before proceeding with rotational torque will decrease the incidence of humeral fracture. Regardless of the amount of mobility achieved with manipulation, we proceed with arthroscopy. If full motion was obtained, the capsular release can be documented for completeness or surgically released if found to be insufficient. Manipulation techniques are effective for extracapsular contractures but often leave capsular contractures intact that must be addressed for optimal results.28,38

Technique for Arthroscopic Release

Examination proceeds under anesthesia. Range of motion of both shoulders is examined and recorded in elevation, ­abduction, and external rotation with the arm adducted at the side. Internal and external rotation are also recorded with the arm in maximal abduction. Before soft tissue release, hydrocortisone is administered intravenously. A trial of gentle closed manipulation is performed on all patients before proceeding with arthroscopic release. Gentle is a relative term but in this case applies to a small amount of force used to move the shoulder in abduction and elevation. Contracted shoulders amenable to closed manipulation will move with minimally applied force. External rotation forces

Box 17k-4 indications and contraindications Surgical Indications Failure of nonoperative treatment Persistent pain and stiffness after 6 months of care No improvement or worsening of external rotation after 4 to 6 months Quantitative Evaluation of Stiffness Severe stiffness 0 degrees of external rotation ≤30 degrees of abduction Moderate stiffness Loss of motion of 30 degrees in external rotation or abduction versus contralateral side Arthroscopic Contraindications Hardware requiring removal Significant fracture malunion requiring osteotomy Factors increasing difficulty of arthroscopic release Prior instability procedure requiring subscapularis takedown* Post-traumatic stiffness* Fracture malunion coupled with stiffness* Surgical Contraindications Release during the freezing phase of idiopathic adhesive capsulitis *These conditions may require open release.

Arthroscopic Technique—Joint Entry

Entering the stiff shoulder is difficult secondary to the ­reduced joint volume and thick capsule. Articular cartilage damage may be caused by forcefully pushing a trocar into the glenohumeral joint. Generally, a standard metal cannula and blunt trocar are effective owing to their stiffness and the surgeon’s ability to palpate the posterior bony structures. The entry point for the trocar is located superiorly ­because the glenohumeral joint is widest at this point and the potential for articular damage is reduced (Fig. 17K-1). The skin incision should be located at the posterior joint line to allow adequate maneuverability inside the joint. The trocar is inserted until bone is palpated. Internal and external rotation of the arm allows the surgeon to feel whether the trocar is positioned over the humeral head (will feel the rotation) or the posterior glenoid. The objective is to palpate the superior glenoid rim and advance the trocar at that point (Fig. 17K-2).

Superior

Superior entry

Inferior entry

Inferior

Figure 17K-1   Location of joint entry. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 146.) Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1098 DeLee & Drez’s� O

A u t h o r ’ s P r e f e r r e d M e t h o d — c ont ’ d

A

Figure 17K-3   Contracted rotator interval. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 146.)

B Figure 17K-2   Palpate bone to determine the entry point. A, Glenoid palpation—trocar is too medial. B, Palpate the humeral head—trocar is too lateral; aim medially and enter the joint. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 146.)

will often be found contracted (Fig. 17K-3). The soft ­tissues within these borders are excised using a motorized soft ­tissue resector. We use the Dyonics Electroblade (Smith & Nephew Endoscopy, Andover, Mass) that is a combination motorized soft tissue resector and electrocautery. The anterior cannula pierces the rotator interval. The resector is passed into the joint through the cannula, which is then pulled back out of the joint leaving the tissues accessible to the shaver. After tissue excision, the cannula is advanced back into the joint to maintain joint access, and the resector is removed. The arthroscope is ­removed from the posterior cannula at this point, leaving it within the joint. Gentle closed manipulation is completed as previously described. Should full motion be attained, we inspect the capsule with the arthroscope to ensure complete capsular release and humeral head location. We proceed to the next step and release the anterior capsule if full range of motion is not obtained with ­manipulation.28,38 Anterior Capsular Release

Inside the joint, attention is directed to viewing the rotator interval. A spinal needle is used to localize anterior portal placement superior to the subscapularis tendon. A 5-mm cannula and trocar are inserted after skin incision. After successful placement of the arthroscope and anterior cannula, we begin our capsular releases, proceeding from the rotator interval to the anterior capsule and finally to the inferior and posterior capsular tissues. The subacromial space is ­inspected for extra-articular adhesions following intra-articular releases. Rotator Interval Release

The boundaries of the rotator interval are the biceps tendon medially, the superior border of the subscapularis tendon inferiorly, and the humeral head laterally. The rotator ­interval

The key landmark for beginning the anterior release is the point at which the middle glenohumeral ligament crosses the subscapularis tendon (Fig. 17K-4). The plane between these two structures should be exploited using a resector or a blunt dissector. Initially the shaver is used to transect the middle glenohumeral ligament until the rolled edge of the subscapularis is clearly visualized. At this point, the blunt dissector is used to complete the tissue separation. Using a Harryman soft tissue punch (Smith & Nephew Endoscopy, Andover, Mass), a 5- to 10-mm strip of anterior capsule is removed. The resection includes the middle glenohumeral ligament and superior portion of the anteroinferior glenohumeral ligament (Figs. 17K-5 to 17K-11). The humeral head may be translated laterally after this release to improve visualization and access to the inferior capsule.

Shoulder 1099

A u t h o r ’ s P r e f e r r e d M e t h o d — c ont ’ d

Figure 17K-4   Contracture of the anterior capsule. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 147.)

Inferior Capsular Release

A lateral translation force to the proximal humerus is applied by an assistant who allows the arthroscope to enter anteriorly and inferiorly and improves visualization of the anteroinferior glenohumeral ligament and inferior capsule. Using the soft tissue punch, the bottom blunt jaw is used to transect the anteroinferior glenohumeral ligament and inferior capsule from anterior to posterior (Figs. 17K-12 and 17K-13). The transection should be competed as far from the glenoid rim as possible. Severe contracture of the axillary pouch will limit ability to reach the inferior capsule. Attention must be turned to the release of the posterior and posteroinferior capsule first to gain adequate access to the inferior axillary pouch.

Figure 17K-5   Identify the superior margin of the middle glenohumeral ligament. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 147.)

Figure 17K-6   Divide the superior portion of the middle glenohumeral ligament. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 147.)

Under arthroscopic visualization, the anterior cannula is removed, and a metal cannula and trocar are inserted in its place. The arthroscope is then moved from the posterior cannula to the anterior cannula. The posterior metal cannula is exchanged for a small plastic cannula under visualization. Using the motorized tissue resector, 5 to 10 mm of posterior capsule is excised (Fig. 17K-14). The small cannula is removed, and a cannula large enough to accommodate the soft tissue punch is then placed through the posterior incision (Figs. 17K-15 and 17K-16). The posteroinferior capsule is resected using the punch for a distance of about 10 mm. Similar to the anterior ­release, we keep the punch 5 to 10 mm lateral to the glenoid rim to avoid injury to the labrum. A third trial of ­ manipulation

Figure 17K-7   Cauterize or resect the middle glenohumeral ligament covering the subscapularis tendon. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 147.) Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1100 DeLee & Drez’s� O

A u t h o r ’ s P r e f e r r e d M e t h o d — c ont ’ d

Figure 17K-8   Cautery to the middle glenohumeral ligament covering the subscapularis. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 148.)

Figure 17K-10   Blunt dissection anterior to the subscapularis tendon. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 148.)

is often successful in releasing the remaining capsule in the axillary pouch. The inferior capsule is inspected from the posterior portal; if it is still intact after manipulation, it is transected with the punch through the anterior portal (Figs. 17K-17 to 17K-19). The blunt tissue dissector may be used to release adhesions anterior to the subscapularis tendon.

lateral subacromial portal, providing visualization anteromedially and anteroinferiorly. An anterior subacromial portal can be made, or a small cannula is directed into the subacromial space through the anterior incision. The shaver can be directed anteromedially and anteroinferiorly from here to release adhesions anterior to the subscapularis tendon. Far medial or inferior adhesions can be reached more easily with the shaver through an accessory portal located just distal to the anterolateral corner of the acromion. Acromioplasty should not be performed as an adjunctive procedure after capsular release because of increased risk for forming recurrent subacromial adhesions.

Subacromial Inspection and Bursectomy

The subacromial space is inspected, and the motorized ­tissue resector is used to excise the bursal tissue and any extra-­articular adhesions (Figs. 17K-20 and 17K-21). Extra-articular adhesions between the subscapularis and the conjoined tendon are evaluated with the arthroscope in the

Figure 17K-9   Cautery to middle glenohumeral ligament covering the subscapularis. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 148.)

Figure 17K-11   Blunt dissection posterior to subscapularis tendon. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 148.)

A u t h o r ’ s P r e f e r r e d M e t h o d — c ont ’ d

Figure 17K-12   Contracture of the inferior capsule. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 148.)

Figure 17K-15   Insert a large cannula posteriorly. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 149.)

Figure 17K-13   Soft tissue punch release of anteroinferior capsule. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 149.)

Figure 17K-16   View of completed posterior capsular resection. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 149.)

Figure 17K-14   Shaver resection of the posterior capsule. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 149.)

Figure 17K-17   Return arthroscope to the posterior portal and complete inferior capsular resection. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 150.) Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1102 DeLee & Drez’s� O

A u t h o r ’ s P r e f e r r e d M e t h o d — c ont ’ d

Figure 17K-20   Resection of subacromial adhesions. Figure 17K-18   Punch resection of inferior capsule. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 150.)

Figure 17K-21   Complete subacromial adhesion release. Figure 17K-19   Complete inferior capsular resection. (From Gartsman GM: Shoulder Arthroscopy. Philadelphia, WB Saunders, 2003, p 150.)

release because it will extravasate through the incompetent capsule and lose effectiveness. A subacromial injection of 1 mg of hydrocortisone is used after release of post­traumatic and postsurgical stiffness. Diabetes is a contraindication to steroid treatment (Box 17K-5).

Physical Therapy after Arthroscopic Release Following arthroscopic release, a pillow is used to keep the arm in slight abduction and to avoid internal rotation. During the initial postoperative examination in the ­hospital, the patient’s arm should be taken through a range of motion (not painful secondary to the interscalene block). The rationale for this important step is to ­demonstrate to

the patient that motion has been improved and to impress upon the patient that success will depend on compliance with postoperative therapy protocols. Continuous ­passive motion (CPM) is started the day of surgery and continued for 2 weeks at home. The CPM chair is used for 1-hour sessions 4 times each day. At 2 weeks’ follow-up in the clinic, the CPM is discontinued if shoulder range of motion is ­improving. Specific range of motion exercises prescribed are passive elevation and external rotation in a supine position using a dowel or pulley. Full range of motion and use of the shoulder is encouraged. Additional follow-up visits are at 6 weeks, 3 months, and 6 months. Patients are allowed immediate full use of the shoulder for all activities as pain allows (see Box 17K-5).

Shoulder 1103

Box 17k-5 Postoperative ­prescription Medications Hydrocortisone given intravenously during surgery Methylprednisolone dose pack, after surgery No steroids in diabetic patients Continuous Passive Motion Chair Begin the day of surgery Continue for 2 weeks: 1-hour session, four sessions per day Two-week follow-up: stop if motion is satisfactory Directed Home Exercises Supine passive elevation and external rotation Full active use of the extremity as tolerated

s­ tatistically ­significant improvements in pain, external rotation, abduction, and function in a group of diabetic patients with adhesive capsulitis released arthroscopically. Diabetic patients generally achieve slightly less improvement than those without diabetes and have a higher incidence of recurrence. A review of the complications of arthroscopic and open release is given in Box 17K-6. Open release of the stiff shoulder is also a successful operation for patients recalcitrant to conservative ­ management.34-36,39 Lusardi and associates34 and ­MacDonald and colleagues36 reported increases in motion and decreased pain in patients treated with open capsular release secondary to postoperative capsular contracture. Ozaki and colleagues17 and Omari and Bunker39 documented nearly complete return of normal motion and improved pain scores following open release of adhesive capsulitis.

Repeat Contracture Release Repeat contracture release is scheduled if full range of motion is not obtained by 3 months after the initial procedure. Manipulation under anesthesia is generally ­effective for achieving full range of motion in these repeat ­situations.

RESULTS The results of arthroscopic release for adhesive capsulitis is well documented. Harryman and colleagues30 reported on 30 patients whose affected shoulder motion improved from 41% of the unaffected shoulder to 93%. They also reported pain relief and significant functional improvements. ­Warner and colleagues13 documented average Constant score improvement of 48 points and motion improvements of 49 degrees flexion, 45 degrees external rotation, and eight spinous processes of internal rotation. Final motion regained was within a mean of 7 degrees of the unaffected shoulder. Ogilvie-Harris and Myerhall6 reported

Box 17K-6 Complications Manipulation Humerus fracture Glenohumeral dislocation Neurovascular injury Arthroscopic Release Axillary nerve injury Subscapularis tendon transection Labral injury Inadequate release Recurrence Open Release Neurovascular injury Subscapularis dehiscence Inadequate release Infection Recurrence

C

r i t i c a l

P

o i n t s

l Complete

a thorough history and physical examination with reproducible measurements of active and passive motion. l Take plain radiographs to document bony pathology. l Proceed with conservative therapy for 6 months. l Do not operate during the freezing stage of disease. l Follow a routine procedure for releases: rotator interval, anterior capsule, anteroinferior capsule, posterior capsule, and then inferior release. l Treat with preoperative methylprednisolone and postoperative oral methylprednisolone (Medrol) dose pack (except in diabetic patients). l Examine the subacromial bursa and release all extracapsular adhesions. l Encourage active and active-assisted range of motion immediately after surgery.

S U G G E S T E D

R E A D I N G S

Farrell CM, Sperling JW, Cofield RH: Manipulation for frozen shoulder: Longterm results. J Shoulder Elbow Surg 14:480-484, 2005. Griggs SM, Ahn A, Green A: Idiopathic adhesive capsulitis: A prospective functional outcome study of nonoperative treatment. J Bone Joint Surg Am 82:1398-1407, 2000. Harryman DT II: Shoulders: frozen and stiff. Instr Course Lect 42:247-257, 1993. Harryman DT, Matsen FA, Sidles JA: Arthroscopic management of refractory shoulder stiffness. Arthroscopy 13:133-147, 1997. Neviaser RJ, Neviaser TJ: The frozen shoulder: Diagnosis and management. Clin Orthop 223:59-64, 1987. Ogilvie-Harris DJ, Myerhall S: The diabetic frozen shoulder: Arthroscopic release. Arthroscopy 13:1-8, 1997. Shaffer B, Tibone JE, Kerlan RK: Frozen shoulder: A long-term follow-up. J Bone Joint Surg Am 74:738-746, 1992. Warner JJP: Frozen shoulder: Diagnosis and management. J Am Acad Orthop Surg 5:130-140, 1997. Warner JJP, Answorth A, Marks PH, et al: Arthroscopic release for chronic refractory adhesive capsulitis of the shoulder. J Bone Joint Surg Am 78:1808-1816, 1996.

R E F E R E N C E S Please see www.expertconsult.com

1104 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

L

Glenohumeral Arthritis in the Athlete Matthew D. Williams and T. Bradley Edwards

Glenohumeral joint degeneration is a disabling condition. Pain and loss of shoulder mobility reduce a patient’s functional capacity, even to the point of affecting basic activities of daily living. These disabilities are especially difficult hurdles for patients desiring active lifestyles. Primary arthritis of the shoulder is atypical, but especially rare in young people. Usually, glenohumeral degeneration in these patients is due to an underlying diagnosis or secondary cause. Treatment is challenging because it not only must reduce pain and improve function in the short term but also must be durable for the life of the patient. Modalities range from conservative care aimed at pain modulation to surgical intervention with joint-sparing orreplacing ­techniques. Proper treatment selection and outcomes depend on understanding the etiology of the joint degeneration, the age of the patient and the patient’s functional demands, and the efficacy and durability of treatment options. The ultimate goal in treating glenohumeral arthritis in the active population is pain relief and restoration of optimal function.

CLASSIFICATION Glenohumeral arthritis has multiple etiologies (Box 17L-1). Primary osteoarthritis may affect the shoulder joint but is uncommon in patients younger than their sixth decade. The secondary causes of arthritis are subdivided into atraumatic and traumatic conditions. Atraumatic conditions include inflammatory arthritis such as rheumatoid disease. Box 17l-1 Classification of ­Glenohumeral Arthritis Primary osteoarthritis Secondary osteoarthritis   Atraumatic conditions   Inflammatory arthropathy (rheumatoid arthritis)    Osteonecrosis—avascular necrosis   Post-traumatic conditions    Postfracture    Dislocation arthropathy    Post-traumatic osteonecrosis   Postsurgical conditions    Capsulorrhaphy arthropathy    Chondrolysis Implant complications Rotator cuff tear arthropathy

Osteonecrosis and its subsequent articular collapse is also an atraumatic cause of glenohumeral arthritis. Traumatic causes of shoulder arthritis include arthritis following fracture and instability (dislocation arthropathy). Capsulorrhaphy arthropathy (arthritis following anterior or posterior stabilization procedures) and chondrolysis are postsurgical complications. Massive rotator cuff tears do occur in the active and athletic population and left untreated may theoretically cause secondary rotator cuff arthropathy; however, the average age at presentation of these patients is 77 years, and most rotator cuff tears resulting in arthropathy are degenerative.1

Primary Osteoarthritis Primary osteoarthritis of the glenohumeral joint is rare, and most patients who present with primary glenohumeral arthritis are older than 60 years of age.2-4 The etiology of primary joint degeneration is unknown. Stiffness of the glenohumeral joint, joint space narrowing, humeral head osteophytes, and an intact rotator cuff characterize primary osteoarthritis of the shoulder as classically described by Neer.5 An inferior osteophyte present on the glenoid and humeral head is a classic finding (Fig. 17L-1),6,7 and chronic posterior subluxation, posterior glenoid erosion, and tight anterior structures limiting external rotation are also features.7,8 Recently, Walch and associates9 described a group of patients with static posterior subluxation of the humeral head preceding the development of osteoarthritis. Posterior subluxation may be the first radiographic indication of emerging arthritis (Table 17L-1).

Dislocation Arthropathy Traumatic dislocation of the shoulder produces articular cartilage injury secondary to shear and impaction, with an incidence of chondral and osteochondral lesions of 47% and 46% respectively.10,11 Neer described glenohumeral arthritis after anterior shoulder instability in 1982.7 The term dislocation arthropathy was coined and classified by Samilson and Prieto in their 1983 report on 74 patients with glenohumeral instability and arthritis.12 The development of arthritis was associated with increased age at the initial event; the direction of dislocation, with posterior dislocations inducing more degenerative change than anterior dislocations; and associated glenoid fractures. The number of dislocation events and previous stabilization procedures were not associated with the development of arthritis.12 Several reports have agreed that arthritis after instability

Shoulder 1105

is associated with older age of the patient at the time of injury and the length of time since the injury occurred.13-15 The impact of previous stabilization surgery and of numerous dislocation episodes has been a point of debate.15-18 Recently, in a multicenter study of patients undergoing arthroplasty for arthritis secondary to ­ instability, Matsoukis and colleagues19 reported no differences between patients treated operatively and those treated nonoperatively for their instability in regard to arthritis severity. The incidence rate of arthritis after dislocation treated nonoperatively averages between 10% and 20% (see Table 17L-1).13-15

Capsulorrhaphy Arthropathy

Figure 17L-1   Anteroposterior radiograph of primary osteoarthritis. Inferior osteophyte is visible at the inferior glenoid and humeral head.

Glenohumeral arthritis in patients with prior instability procedures is well documented and discussed in the literature. The term capsulorrhaphy arthropathy, coined by Matsen and colleagues,20 describes overtightening of the capsule either anteriorly or posteriorly resulting in abnormal translation of the humeral head opposite the capsulorrhaphy. Nonanatomic glenohumeral translation results in atypical biomechanics, asymmetric cartilage wear, and ultimately arthritis (Fig. 17L-2). Biomechanical models have demonstrated that selective capsular plication causes alterations in humeral head translation.21-24 Capsulorrhaphy arthropathy affects predominantly young male patients with a mean age of 45 years.25 Factors implicated in the development of arthritis include length of followup, external rotation contracture of the operative shoulder, and age at the time of initial trauma, with older patients

  TABLE 17L-1  Typical Findings Diagnosis

Presentation and History

Physical Examination

Imaging

Primary osteoarthritis

Pain and decreased motion No identifiable cause

Restricted range of motion External rotation contracture Palpable crepitus

Rheumatoid arthritis

Pain and decreased motion Joint swelling Diagnosis of rheumatoid arthritis

Restricted range of motion Shoulder effusion Painful crepitance

Avascular necrosis

Shoulder pain Suggestive history with positive risk factors

Near-normal motion Crepitus—often palpable Pain at midrange positions

Post-traumatic arthritis

Pain and decreased motion Subjective complaints of instability Antecedent fracture or dislocation

Restricted range of motion Positive or negative instability examination

Capsulorrhaphy arthropathy

Pain and decreased motion History of glenohumeral stabilization procedure

Restricted range of motion Painful at midrange positions

Chondrolysis

Pain with activity Difficulty achieving motion goals in therapy Recent shoulder surgery (arthroscopic)

Pain out of proportion to expected examination Restricted range of motion Palpable crepitus

Joint space narrowing Osteophytes Sclerosis Subchondral cysts Posterior humeral subluxation Regional osteopenia Joint space narrowing Periarticular erosions Medial glenoid erosion Superior humeral head migration if rotator cuff involved Sclerosis commonly Subchondral crescent sign Subchondral collapse Images dependent on stage of disease Joint space narrowing Radiographic signs of fracture Osteophytes Sclerosis Hill-Sachs lesions Anterior or posterior glenoid bone loss Joint space narrowing Sclerosis Osteophytes Signs of previous operative intervention Glenohumeral space narrowing

1106 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

cuff repair protocol.44 The remaining five cases occurred after shoulder arthroscopy in young patients, all younger than 20 years.42,43 Radiofrequency ablation (RFA) was used in four of the cases and thought to be a contributor to the onset of chondrolysis because RFA has been shown to cause cartilage destruction and chondrocyte death.45,46 However, the direct causal relationship of thermal devices and cartilage death was questioned in a review of more than 14,000 arthroscopic shoulder procedures using thermal devices with no incidence of chondrolysis.47 The final case involved use of a bupivacaine pump intra-articularly for pain control. Gomoll and colleagues recently reported on the chondrotoxic properties of bupivacaine, questioning the routine postoperative use of intra-articular bupivacaine infusion pumps.48 The pathophysiology of cartilage death and degeneration in the glenohumeral joint following arthroscopy is not understood. There is no described treatment protocol for glenohumeral chondrolysis in the literature (see Table 17L-1).

Rheumatoid Arthritis

Figure 17L-2   Anteroposterior radiograph of capsulorrhaphy arthropathy that developed in a patient after an anterior stabilization procedure.

being more susceptible.15 Dislocation arthropathy is associated with many of the same predisposing factors as ­capsulorrhaphy arthropathy with the exception of previous surgery. Reports have found no difference in patients treated operatively for instability and patients treated nonoperatively, calling into question whether dislocation ­arthropathy and capsulorrhaphy arthropathy are actually different entities (see Table 17L-1).19,26,27 Surgical procedures for anterior glenohumeral instability have evolved over time and may be divided into anatomic and nonanatomic repairs. Anatomy-preserving procedures are the Bankart28 and capsular shift.29 Nonanatomic procedures include the Putti-Platt,30 MagnusonStack,31 Bristow,32 and Latarjet.33 Arthropathy following anterior reconstructive procedures for instability occurs after both anatomic and nonanatomic repairs, and severity correlates to the length of time since surgery. Degenerative changes have been reported in 30% to 61% of shoulders treated with the Putti-Platt procedure at 9- to 26-year follow-up.17,34,35 The Latarjet-type coracoid transfer is associated with a 49% incidence of arthritis at an average 14 year’s follow-up.15,36,37 Anatomic Bankart repairs and later joint degeneration are also time dependent. The increasing incidence of arthritis is reported from 0% to 63% over a 6- to 19-year follow-up.18,32,38-41

Chondrolysis Chondrolysis of the glenohumeral joint is a rare and devastating condition described in only a handful of cases, all of whom were young patients.42-44 Two reported cases were caused by a chondrotoxic stain used during a rotator

Rheumatoid arthritis is an inflammatory condition of the synovium that results in synovial hyperplasia and a disabling secondary erosive arthritis. Overall prevalence is 1% with a female predominance and shoulder involvement in more than 90% of patients with chronic rheumatoid disease.49 Early complaints are pain, swelling, and decreasing shoulder motion. As the disease progresses, extra-articular structures become involved and painful: the subacromial bursa, acromioclavicular joint, and even rotator cuff.50 Rheumatoid shoulders undergo medial migration of the humeral head into the glenoid. The encroachment is typically into the central glenoid with joint space narrowing. The bone quality is generally osteopenic with periarticular erosions (Fig. 17L-3).51 Periarticular bony invasion of rheumatoid disease tends to affect the humeral head superior and medial to the greater tuberosity. Bony lesions (cysts) at the insertion of the rotator cuff tendons, coupled with the tendon degeneration associated with rheumatoid disease, may lead to cuff compromise. Rotator cuff–­deficient shoulders will have superior humeral head migration and uneven joint erosion. The late result of rheumatoid disease affecting the shoulder is painful joint destruction, loss of bone stock, rotator cuff compromise, and poor function.50,52 Evaluating patients with rheumatoid arthritis is difficult secondary to the concomitant pain generators that may be present. Rheumatoid patients often suffer with cervical involvement and radiculopathy or myelopathy that can confound the shoulder examination with secondary pain and weakness. A thorough history and physical examination in association with appropriate imaging—magnetic resonance imaging (MRI) and computed tomography (CT)53,54—and diagnostic injections may be necessary to delineate the causes of pain in order to properly direct treatment (see Table 17L-1).

Osteonecrosis Vascular compromise and death of bone are the end result of osteonecrosis, otherwise known as avascular necrosis (AVN) or aseptic necrosis.55 The pathogenesis

Shoulder 1107

Figure 17L-3   Rheumatoid arthritis—concentric medial migration of the humeral head.

of ­osteonecrosis of the humeral head parallels that of the femoral head. Loss of microcirculation in the epiphyseal bone leads to marrow necrosis. Resorption of necrotic bone proceeds more rapidly than trabecular bone replacement and subsequent subchondral weakness results. Collapse of the articular surface secondary to fractures in weak subchondral bone results in articular incongruency and degeneration. The vascular network supplying the humeral head stems from the anterolateral branch of the anterior humeral circumflex artery.56 Injury to this principle vessel threatens the blood supply to the entire humeral head. Osteonecrosis of the proximal humerus results from numerous traumatic and atraumatic causes, most frequently systemic corticosteroids and trauma (see Table 17L-1).57-64 Pain is the most common complaint of patients with osteonecrosis and includes night pain with difficulty sleeping and pain interfering with simple activities of daily living.64,65 These patients are generally younger than those presenting with primary osteoarthritis.66,67 Although near-normal range of motion may be present, mechanical symptoms are not uncommon secondary to articular cartilage lesions or joint incongruity. Pain is reliably reproduced with the shoulder flexed or abducted near 90 degrees owing to the involvement of the superior central portion of the humeral head that contacts the glenoid in this position.65 Osteonecrosis of the proximal humeral head is classified according to the system reported by Cruess.68 The six-stage system is based on the classification of femoral head osteonecrosis described by Ficat and Arlet.69,70 Similar to osteonecrosis in the femoral head,

Figure 17L-4   Collapse of the subchondral bone—Arlet and Ficat stage 4 osteonecrosis.

c­ ollapse of the ­ subchondral bone in the humerus is heralded by the crescent sign, and final stage of disease is illustrated by radiographic degeneration on both sides of the joint (Fig. 17L-4).

PATIENT EVALUATION Presentation and History Although there is not a classic or pathognomonic history for glenohumeral arthritis, most patients complain of increasing pain coupled with a progressive loss of shoulder mobility. Common complaints also include “noise” and crepitus, feelings of “catching,” and instability complaints. Instability complaints are subjective in nature, generally without history of frank dislocation, and are usually related to the mechanical catching of incongruent articular surfaces. Morning stiffness usually improves through the day, and sleep is affected by night pain. Primary glenohumeral arthritis is uncommon in the young population and an exhaustive history and physical examination should be completed to identify underlying primary diagnoses. The history should contain episodes of trauma, therapy or surgical procedures, medications including steroid use, family history, and recreational and social factors. Questions directed to participation in sports, organized or recreational, including questions about training, conditioning, and position should be answered.

1108 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The physical examination should begin with the cervical spine to rule out pathology that could confound ­examination and treatment outcomes in the shoulder. Should the cervical spine examination demonstrate positive findings, it is incumbent on the evaluating physician to treat or refer this patient for proper care of their associated conditions. The shoulder examination begins with visual examination with the patient exposed to evaluate muscular atrophy. Palpation for tenderness around the shoulder, including the sternoclavicular and acromioclavicular joints, is important. Disregarding a painful acromioclavicular joint can result in misdiagnosis and poor outcomes after treatment of other pathology. Mobility is paramount to successful outcomes in shoulder surgery and active and passive motion should be documented. Motion is evaluated by elevation in the plane of the scapula, abduction, external rotation with the arm at the side and abducted 90 degrees, and internal rotation by vertebral level reached with the thumb outstretched behind the back. Accurate delineation of painful motion is important in accurate diagnosis and treatment of shoulder arthritis. Typically, end range of motion pain is secondary to impingement, osteophytes, and capsular contracture. Midrange motion pain is indicative of articular surface damage or inflammation or synovial inflammation and is of prognostic value in directing surgical treatment.71 Glenohumeral crepitus should be noted, as should discrepancy between active and passive motionpain. Impingement signs should be documented as well as pain along the long head of the biceps. Strength testing of the rotator cuff is improved by isolating each tendon. Jobe’s test is used for the supraspinatus.72 The infraspinatus is tested using the external rotation lag sign and external rotation strength with the arm adducted at the side.73 The horn blower’s sign (external rotation with the shoulder abducted 90 degrees) tests teres minor

A

B

integrity.74 Subscapularis strength is tested using the belly press test and the lift-off test.75

Imaging Combining a thorough history and physical examination with proper imaging gives the best chance of proper diagnosis and successful treatment. Radiographic evaluation of glenohumeral arthritis provides an illustration of the extent of disease, and standard views should be taken of each patient. Standard radiographs for evaluation of the shoulder in our practice include an anteroposterior view in neutral rotation, a scapular outlet view, and an axillary view (Fig. 17L-5). Additional radiographic studies are obtained if necessary, for example, a glenoid profile or Bernageau view to assess instability.51,76,77 These views document the position of the humeral head in relation to the glenoid, the presence of osteophytes, bone quality, relative glenohumeral joint space, and visualization of glenoid bone loss. CT arthrograms are obtained to evaluate glenoid bone stock, morphology and version, and rotator cuff muscle and tendon quality if surgery is a consideration. The morphology, bone quality, and bone loss are important to document. Glenoid version is assessed using the ­technique described by Friedman and colleagues (Fig. 17L-6).78 The classification system of Walch and associates is used to assess glenoid wear in the anteroposterior plane for biconcavity (Fig. 17L-7).79,80 Rotator cuff muscle quality is documented because fatty infiltration has been shown to affect the results of unconstrained shoulder arthroplasty in patients with primary osteoarthritis.81 The classification system of Goutallier is used to determine prognosis.82 Routine MRI of patients with radiographically documented glenohumeral arthritis is not part of our diagnostic protocol.

C

Figure 17L-5   Standard radiographic views used in our clinical practice: anteroposterior (A), scapular outlet (B), and axillary (C) views.

Shoulder 1109 Figure 17L-6   Assessment of glenoid version as described by Friedman and colleagues. (From Friedman RJ, Hawthorne KB, Genez BM: The use of computerized tomography in the measurement of glenoid version. J Bone Joint Surg Am 74:1032-1037, 1992.)

α

α = Amount of Retrovision

A

Axis of Scapula

B

TREATMENT

Nonoperative

The aims of treating glenohumeral arthritis are pain reduction and restoration of mobility and function. In older patients, as in the knee and hip, a total shoulder arthroplasty effectively replaces the painful joint surfaces and allows return of function. In young patients, however, the durability of a treatment option must be weighed against the age of the patient and the patient’s functional goals. Nonoperative treatment should be aggressively pursued and exhausted before surgical intervention. Surgical options include joint-sparing techniques such as arthroscopic débridement and glenoidplasty that are temporizing treatments to delay replacement. Some diagnoses are not adequately treated with conservative or minimally invasive surgery and require arthroplasty at a young age to regain function. The mainstay of a successful outcome, whether nonsurgical or surgical, is patient education: explanation of the natural history of the disease process and functional prognosis, management of acute and chronic pain, and clear delineation of a treatment plan.

Nonoperative treatment modalities include physical therapy, nonsteroidal anti-inflammatory drugs (NSAIDs), injectable corticosteroids, and viscosupplementation. Physical therapy is the primary defense against stiffness and muscle atrophy caused by disuse secondary to pain.83,84 Therapists are excellent patient educators, helping patients modify tasks and activities of daily living to make them easier and less painful. Range of motion is maintained using both active and active-assisted exercises. Advancement of motion and resistance training should be tailored to the individual patient based on pain. Painful activities serve to decrease compliance and exacerbate the disease process. Medical interventions include NSAIDs and corticosteroid injections among others. NSAIDs are effective for the treatment of arthritis pain.85 Intra-articular corticosteroid injections also provide symptomatic relief, yet there remains much discussion and no definitive recommendations for their use despite being a fundamental part of arthritis care. Viscosupplementation using hyaluronic acid has been used for arthritis therapy in the knee with

A1

B1

A2 B2

C

Figure 17L-7   The classification system of Walch and associates is used to assess biconcavity and glenoid wear in the anterior-posterior plane. (From Walch G, Badet R, Boulahia A, Hhoury A: Morphologic study of the glenoid in primary osteoarthritis. J Arthroplasty 14:756-760, 1999.)

1110 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

c­ ontroversial results.86 Viscosupplementation agents are not yet approved for treatment of glenohumeral arthritis. The efficacy of oral glucosamine and chondroitin sulfate in the treatment of arthritis is also controversial, but this supplement remains popular with the patient population.87,88

Joint-Sparing Techniques Arthroscopic Débridement Diagnosis of glenohumeral arthritis and cartilage lesions in young patients is often made incidentally unless radiographic changes or a predisposing underlying condition is present (Fig. 17L-8).89-91 Arthroscopy provides a minimally invasive approach to directly address chondral lesions in the shoulder and allows early rehabilitation and return to activities for patients unresponsive to conservative modalities.92 Although arthroscopic débridement and associated procedures are temporizing treatments that do not change the course of the underlying process, they can provide pain relief and functional improvement.93-96 Weinstein and colleagues94 reported on 25 patients who underwent arthroscopic débridement for isolated chondral lesions. Results correlated positively with the extent of cartilage damage. Seventy-six percent of the patients had sustained pain relief at an average 34-month follow-up. In another report, grade IV osteochondral lesions were treated with arthroscopic débridement. Pain relief was at a maximum within 5 weeks of surgery, and 88% of patients had sustained pain relief at 2 years. Lesions larger than 2 cm2 were associated with earlier return of pain.93 In addition to débridement, full-thickness chondral defects have been managed with microfracture, similar to that performed in the knee for full-thickness cartilage defects.97,98 Shoulder pain in patients with glenohumeral arthritis may be due to concomitant pain generators as well as the cartilage lesions. Subacromial bursitis and impingement,99 acromioclavicular joint pain,100 partial rotator cuff

tears,101,102 labral tears, long head of the biceps tendon lesions,103 and capsular contractures104 should be addressed at the time of arthroscopic surgery. Failure to address these associated lesions will worsen postoperative results.

Glenoidplasty Glenoidplasty and osteocapsular arthroplasty are arthros­ copic alternatives to arthroplasty in the young patient with glenohumeral arthritis. Posterior subluxation of the proximal humerus causes nonconcentric glenoid wear and glenoidplasty is the restoration of glenoid morphology from a biconcavity to a single concavity. Recreating the concavity of the glenoid reduces posterior humeral subluxation, increases the effective joint surface, which decreases point loading, and increases joint stability. Using an arthroscopic bur, the anterior glenoid is removed until flush with the posterior glenoid. In addition to stabilizing the joint, glenoidplasty serves to reduce tension in anterior soft tissue restraints, thereby improving external rotation that is often restricted in these patients.105 The primary recommendation for glenoidplasty and other arthroscopic joint-sparing procedures is pain relief. Results suggest predictable relief of rest pain and mechanical glenohumeral impingement at extremes, but pain relief in mid-arc range of motion pain is not predictable. Postoperative functional recovery is related directly to pain relief. Motion gained intraoperatively is often lost in the postoperative period, resulting in unpredictable functional outcomes. Clinically, factors associated with good results include pain at extremes of motion, rest pain, and painless crepitus. Radiographically, large humeral osteophytes, glenoid biconcavity secondary to posterior humeral subluxation, and loose bodies are also positive prognosticators. Important negative prognostic signs for success are pain in the mid-arc of motion, painful crepitus, small osteophytes, and no glenoid biconcavity. Severe pain in the mid-arc of motion is a contraindication to glenoidplasty because of its association with severe arthritis as well as glenoids without biconcavity.105 Osteocapsular arthroplasty involves removing humeral osteophytes and releasing soft tissue capsular contractures. Capsular releases should be performed from the rotator interval to the posterior-inferior axillary recess to gain maximal mobility. Osteophytes can be a source of pain and may limit mobility—both of which are improved by this procedure. Accessory portals or enlarged portals may be required to remove large osteophytes or loose bodies. Kelly and colleagues106 reported on a group of patients with an average age of 50 years with glenohumeral arthritis and restricted range of motion treated using glenoidplasty and osteocapsular arthroplasty. More than 85% of the patients reported improvement in both pain and range of motion at average 3-year follow-up.

Arthroscopic Treatment of Rheumatoid Arthritis and Osteonecrosis Figure 17L-8   Isolated humeral head osteochondral lesion in a young patient found at arthroscopy for a suspected superior labral lesion.

Arthroscopic procedures are also used in the early treatment of rheumatoid disease. Hypertrophic synovial tissue incites the bony and articular destruction; therefore, synovectomy helps to slow disease progression. ­ Synovectomy

Shoulder 1111

may be effected medically, surgically, or with a combination of modalities.107-109 Arthroscopic synovectomy results in increased motion and decreased pain in 80% of patients; however, it is only indicated early in the disease process. When radiographic signs associated with joint destruction are visible, synovectomy is no longer a viable option.110 Associated arthroscopic procedures may be performed, including subacromial bursectomy and rotator cuff repair or débridement if irreparable. Regeneration of the hypertrophic synovium may occur, requiring repeat ­synovectomy. Early treatment of proximal humeral osteonecrosis includes arthroscopy and core decompression. Results have been shown to be related to the severity of humeral head involvement.111 Mont and colleagues112 first described core decompression for humeral osteonecrosis in 1993. Ficat and Arlet stages I and II osteonecrosis have successful treatment rates of 94% and 88%, respectively, after decompression.113 Results decreased to 70% for stage III and only 14% success for stage IV. Core decompression may increase the time until arthroplasty in stage III disease, but its indications are controversial. Core decompression is not indicated with stage IV or V disease.

Joint-Resurfacing Techniques Osteochondral Allograft Patients with full-thickness cartilage defects of the humeral head that fail treatment with arthroscopic débridement and microfracture techniques are candidates for open treatment. Humeral lesions are replaced with matched osteochondral allograft implantation (Fig. 17L-9). Championed in the knee and used in the shoulder for large defects secondary to instability,114-117 osteochondral allografts are a viable and effective means to fill localized and diffuse chondral defects in the humeral head. Two years after osteoarticular allografting in 18 patients for instability defects, Miniaci and colleagues117 reported no repeat instability and no allograft complications. Gerber and associates115 treated

A

B

four patients with humeral osteoarticular allografts. At 60 months, three patients had little or no pain and minimal functional disability; one failure occurred secondary to necrosis of the humeral head at 6-year follow-up. The non–weight-bearing glenohumeral joint provides a forgiving environment for allograft incorporation and sympto­ matic relief of grafted full-thickness defects.

Humeral Head Resurfacing In addition to traditional hemiarthroplasty and total shoulder arthroplasty, humeral head resurfacing implants are available as treatment options for the younger patient. A benefit of these implants is that implantation requires minimal bone resection, and no stem is inserted into the canal, alleviating the complication of periprosthetic humeral shaft fracture.118-121 Humeral head resurfacing implants provide for complete surface replacement or partial surface replacement. An indication for resurfacing humeral arthroplasty is AVN of the humeral head with an area of necrosis less than 25% of the humeral head. Larger defects compromise the fixation of a resurfacing design, and a stemmed implant is necessary. Another indication is a very young patient requiring arthroplasty, usually secondary to chondrolysis. In our practice, complete resurfacing implants are not used along with prosthetic glenoid reconstruction or in a humeral head without adequate proximal bone, that is, in proximal humeral fracture patients or large AVN defects. Without resection of the humeral head, as done for traditional humeral implant placement, glenoid exposure is difficult and may result in improper glenoid component positioning, leading to poor results. Partial humeral resurfacing implants (Hemicap, Arthrosurface Inc., Franklin, Mass) are used for focal chondral lesions that fail arthroscopic management (Fig. 17L-10). Contraindications for partial humeral head arthroplasty are lesions larger than 35 mm (represents the largest implant available), nonlocalized disease, and insufficient bone to support the implant. Although these implant designs are smaller and require less bony

C

Figure 17L-9   A, Large osteochondral lesion in a young patient. B, Intraoperative view of matched osteochondral allograft being impacted into position. C, Resurfaced lesion with matched osteochondral allograft.

1112 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17L-10   A, Focal full-thickness osteochondral lesion. B, Complete coverage of the lesion using a metallic resurfacing implant.

resection, they are no less invasive to implant than a conventional prosthesis.

Biologic Resurfacing Biologic resurfacing of focal glenoid chondral lesions or glenoid arthritis is a viable alternative to total shoulder arthroplasty in the young patient. Focal glenoid lesions that fail arthroscopic techniques (i.e., débridement and microfracture) can be “covered” with a soft tissue graft. Tissues used to resurface the glenoid are autologous fascia lata, anterior shoulder capsule, Achilles allograft, meniscal allograft, and porcine tissue.122-125 Morphologic abnormalities of the glenoid should be realized and addressed concomitantly with biologic resurfacing. Preparation of the glenoid for the graft includes reaming to correct any anterior or posterior wear and removing any remaining cartilage to bleeding bone. Hemiarthroplasty of the proximal humerus or a humeral resurfacing implant coupled with a biologically resurfaced glenoid alleviates many of the complications associated with glenoid component wear and loosening in younger patients in short-term follow-up (Fig. 17L-11). Nowinski and associates evaluated 26 shoulders treated with glenoid resurfacing using a variety of techniques over a 5- to 13-year follow-up.123 All were treated with cementless hemiarthroplasty for the humeral disease. An overall 81% good to excellent result was reported for the treatment group and improved results were documented with fascia lata or Achilles grafts. Twenty-one of the patients returned to predisease activities, including heavy manual labor and sports; however, biologic resurfacing did not protect the glenoid from erosion.123 Yamaguchi and colleagues126 reported on results of seven patients

Figure 17L-11   Biologic glenoid resurfacing using a fascia lata autograft. Graft being positioned for final suturing intraoperatively.

Shoulder 1113

treated using meniscal allograft over a 5-year follow-up. All reported improvements in pain and had restoration of the joint space on radiographs.

Arthroplasty Neer introduced shoulder arthroplasty in the 1950s.127 Advances in prosthesis design and surgical techniques have improved function in patients undergoing shoulder arthroplasty. Shoulder arthroplasty is an effective treatment option for degenerative conditions of the glenohumeral joint, including primary osteoarthritis, post-traumatic arthritis, inflammatory arthritis, osteonecrosis, and capsulorrhaphy arthropathy. The results of shoulder arthroplasty are dependent on the condition of bone and soft tissue structures and the underlying etiology of disease. Pain recalcitrant to other treatment modalities is the most common indication for replacement; however, patients also improve statistically in regard to motion, strength, and function after arthroplasty. Primary osteoarthritis of the glenohumeral joint that is unresponsive to nonoperative therapy or that fails joint-preserving surgical treatments such as arthroscopic débridement is an indication for shoulder arthroplasty (Fig. 17L-12). Results of primary osteoarthritis treated

Figure 17L-12   Anteroposterior radiograph of a total shoulder arthroplasty for primary osteoarthritis using an anatomic humeral component and an all-polyethylene glenoid component.

with shoulder arthroplasty are found in focused studies2,128-132 and in series of nonhomogenous diagnoses including primary osteoarthritis.7,133,134 In a series of more than 200 patients, Godenèche and colleagues128 reported 77% good or excellent objective results based on the Constant score.135,136 Mean forward elevation of their cohort improved 50 degrees from 94 degrees preoperatively to 145 degrees postoperatively, and ageadjusted Constant scores improved from an average of 38 preoperatively to 97 postoperatively. Subjective results were equally impressive; 94% of patients were satisfied or very satisfied.128 Edwards and associates2 showed statistically significant improvement in postoperative values over preoperative scores for all variables, including Constant scores, pain, ­ mobility, activity, and strength, in patients with osteoarthritis treated with arthroplasty. These results compare favorably with the results of the other series on arthroplasty for primary osteoarthritis. Preoperative factors affecting results of shoulder arthroplasty in osteoarthritis include rotator cuff tears, muscle degeneration, and glenoid morphology (in total shoulder arthroplasty). Small supraspinatus tears have little effect on outcomes after arthroplasty,81 but larger tears have a negative impact on active forward elevation and strength.128,133 Rotator cuff tears have little effect on postoperative pain relief.128 Fatty degeneration of the rotator cuff muscles affected postoperative results, and decreased forward elevation and strength were noted in patients with Goutallier82 grade 2 or higher fatty infiltration in the infraspinatus and grade 3 or higher in the subscapularis.128 Historically, glenoid resurfacing has been a point of debate because the indications for glenoid resurfacing were poorly defined and the decision to resurface was largely based on the skills and comfort level of the operating surgeon. Glenoid radiolucencies are a concern with total shoulder arthroplasty and have an incidence of 29% to 90%.7,128,134,137-139 Advancement in components, glenoid preparation, and implantation and cementing ­ techniques in response to glenoid radiolucencies have lowered their overall incidence (Fig. 17L-13).140-144 Indications for glenoid resurfacing and arthritis include an intact rotator cuff, small reparable rotator cuff tear, glenoid incongruity, and loss of glenoid cartilage.145,146 Arthritis in conjunction with massive rotator cuff tear or inadequate glenoid bone stock is an indication for hemiarthroplasty or reverse shoulder arthroplasty.145,146 The reverse shoulder prosthesis should be reserved for elderly patients and as an implant of last resort, not routinely considered in young and active patients with glenohumeral arthritis. The literature has indicated that total shoulder arthroplasty provides improved results over hemiarthroplasty,134,147-149 but early series were too small to prove statistical significance. However, a meta-analysis by ­ Kirkley and coworkers150 showed superiority of total shoulder arthroplasty compared with hemiarthroplasty. Additionally, Edwards and associates2 compared hemiarthroplasty to total shoulder replacement in a large multicenter study of 600 patients, which was powerful enough to show statistical significance. Total shoulder arthroplasty out­performed hemiarthroplasty, with statistical significance in nearly all measured parameters, including Constant scores, pain, motion, strength, activity scores, and active anterior

1114 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17L-13   A and B, Intraoperative views of a resurfaced glenoid using an all-polyethylene glenoid component.

elevation and external rotation. Of important note, there was no difference in complication rate or reoperation rate. Both hemiarthroplasty and total shoulder arthroplasty provide pain relief, but total shoulder replacement is statistically superior to hemiarthroplasty. Humeral head replacement alone is reported to have poorer results and an increased incidence of revision when compared with total shoulder arthroplasty.149

Arthroplasty for Inflammatory Arthritis Shoulder arthroplasty and its results for patients with rheumatoid disease and inflammatory arthritis have been reported.147,151-156 Although preoperative symptoms are improved or relieved with both hemiarthroplasty and total shoulder arthroplasty for this condition, choice of implant is a source of controversy. Basmania and colleagues156 reported improvement in pain and motion over preoperative values for 45 patients with rheumatoid arthritis treated with arthroplasty. Hemiarthroplasty achieved improved range of motion in this group as well as improved subjective satisfaction over total shoulder patients. Other studies recommend total shoulder arthroplasty in the face of inflammatory arthritis.147,155,157 In contrast, other studies show no significant differences in outcome between the two modalities in rheumatoid disease.153,158 Rheumatoid arthritis associated with a large rotator cuff tear is an indication for reverse shoulder arthroplasty, but only in older patients without another reasonable surgical ­alternative.159,160

Arthroplasty for Instability Arthropathy Neer and associates7 first reported on the results of shoulder arthroplasty for osteoarthritis following instability surgery, with 16 of 17 patients reporting ­ satisfactory to excellent results. Total shoulder arthroplasty and hemiarthroplasty for arthritis after instability surgery resulted in 77% excellent or satisfactory outcomes and 23% unsatisfactory results at 3-year follow-up in a study by Bigliani and colleagues.161 No distinction in postoperative outcome was made between hemiarthroplasty and total shoulder arthroplasty patients. Multiple procedures before the arthroplasty were cited as the reason for inferior outcomes compared with arthroplasty for primary arthritis. Other patient series report marked improvements in pain and mobility, including external rotation and abduction, after total shoulder arthroplasty for arthritis after instability repair.162,163 Revision rates are higher in these patients than in primary osteoarthritis.163 Matsoukis and colleagues19 reported on 55 patients with glenohumeral arthritis and a history of anterior dislocation treated with arthroplasty, 27 of whom had a previous stabilization procedure. Glenoid resurfacing was done on 39 patients. Younger patients scored higher on outcome studies than older patients but not significantly. Total shoulder arthroplasties demonstrated improved postoperative function compared with hemiarthroplasties. In contrast to other studies, patients with history of instability surgery and those treated nonoperatively before the arthroplasty demonstrated no differences in postoperative outcomes.19

Shoulder 1115

Author’s preferred method Patients who present and are diagnosed with glenohumeral articular cartilage degeneration in our practice are treated according to the severity of their disease and according to their functional goals. The major component in determining a treatment protocol is severity of disease. All patients are given a trial of conservative management consisting of NSAIDs, selective rest, and activity modification for 6 to 12 weeks regardless of their condition. After this initial trial, we either continue on a conservative course using corticosteroid injections as necessary—not more than one injection every 6 months—or discuss operative intervention. For focal articular humeral head defects less than 35 mm in diameter, arthroscopic débridement of the lesion, microfracture, and contracture release as indicated is the initial surgical option. A period of 6 months is allowed for improvement after débridement. If patients remain symptomatic after 6 months, we fill the defect with a matched osteoarticular allograft if the patient is younger than 30 years, or with a metallic device (Hemicap, Arthrosurface Inc., ­Franklin, Mass) if they are older (Fig. 17L-14). Focal chondral lesions of the glenoid are managed similarly to those found on the humeral head. Arthroscopic

techniques are used initially. Should symptoms not improve or worsen, biologic glenoid resurfacing using fascia lata autograft is completed. Isolated biologic glenoid resurfacing is used only in the presence of a normal humeral head articular surface. Nonfocal articular lesions or larger osteochondral defects caused by osteonecrosis or chondrolysis are treated using a variety of methods. In the absence of glenoid involvement and defects less than 25% of the humeral head, complete resurfacing implants are used without glenoid resurfacing. Large osteonecrotic lesions with bone loss contraindicating the use of a resurfacing implant are treated with a stemmed anatomic hemiarthroplasty.164 Involvement of the glenoid in patients younger than 40 years is addressed with biologic glenoid resurfacing using a fascia lata autograft. In older patients, we proceed to stemmed humeral components and glenoid resurfacing using all-polyethylene glenoid implants (total shoulder arthroplasty) (Fig. 17L-15). Glenohumeral arthritic conditions secondary to ­primary osteoarthritis, rheumatoid arthritis, extensive traumatic arthropathy, or capsulorrhaphy arthropathy are managed surgically after failure of conservative measures. Diffuse

Focal full-thickness chondral lesions of the humeral head Initial treatment: Arthroscopic débridement and microfracture

Treatment failure?

Isolated humeral head lesion

Isolated glenoid lesion

<30 years old: Matched osteoarticular allograft >30 years old: Metallic resurfacing humeral implant

Fascia lata autograft resurfacing

Humeral head lesion with glenoid involvement

Diffuse involvement (humeral head + glenoid)

<30 years old: Matched osteoarticular humeral allograft and fascia lata autograft glenoid resurfacing

<40 years old: Standard hemiarthroplasty or resurfacing humeral implant and biologic glenoid resurfacing >40 years old: Total shoulder arthroplasty

30–40 years old: Metallic resurfacing humeral implant and fascia lata autograft >40 years old: Total shoulder arthroplasty Figure 17L-14   Treatment algorithm for decision making in the treatment of focal osteochondral lesions. Continued

1116 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A u t h o r ’ s p r e f e r r e d m e t h o d — c ont ’ d i­nvolvement of both sides of the glenohumeral joint in these conditions precludes a lasting functional outcome from arthroscopic débridement in our opinion. For these patients, we proceed with total shoulder arthroplasty. Total

shoulder arthroplasty is used over hemiarthroplasty even in younger patients (>40 years), secondary to the improved functional results and durability of modern total shoulder prostheses.

Nonfocal chondral lesions

Primary osteoarthritis Dislocation arthropathy Capsulorrhaphy arthropathy Rheumatoid arthritis Chondrolysis

If <40 years old: Complete metallic humeral resurfacing implant or stemmed hemiarthroplasty — fascia lata autograft glenoid resurfacing If >40 years old: Total shoulder arthroplasty using stemmed humeral component

Osteonecrosis

Normal glenoid

Complete metallic humeral resurfacing implant or stemmed hemiarthroplasty— no glenoid resurfacing

Glenoid involvement

If <40 years old: Complete metallic humeral resurfacing implant or stemmed hemiarthroplasty — fascia lata autograft glenoid resurfacing If >40 years old: Total shoulder arthroplasty using stemmed humeral component Figure 17L-15   Treatment algorithm for decision making in the treatment of nonfocal osteochondral lesions.

REHABILITATION

Arthroplasty Rehabilitation Protocol

Rehabilitation after arthroscopic treatment of arthritic lesions is focused on regaining and maintaining functional range of motion. Range of motion activities, including active and active-assisted range of motion, are begun within 1 week of surgery. Strengthening is started after acceptable motion is achieved in a pain-free arc. Hydrotherapy, discussed later in our protocol following arthroplasty, is also used after arthroscopic treatments for glenohumeral joint lesions (Table 17L-2).165

The surgical procedure, arthroscopic or open, the type of prosthesis used, and associated procedures performed determine the type of postoperative orthosis and the ­duration it is required. Table 17L-3 presents postoperative orthosis protocols. All patients begin hand, wrist, and elbow mobility exercises on postoperative day 1; patients undergoing unconstrained shoulder arthroplasty or resurfacing procedures without performance of an ­associated posterior   TABLE 17L-3  Type of Postoperative Orthosis Used

  TABLE 17L-2  Time for Initiation of Hydrotherapy Procedure Arthroscopic procedures Unconstrained arthroplasty Resurfacing arthroplasty Osteochondral grafting Unconstrained arthroplasty and posterior capsulorrhaphy Reverse shoulder arthroplasty

Hydrotherapy Initiated ­(Postoperative Week) 1 1 1 1 1 3 to 4

and Duration of Usage Procedure Performed

Type

Duration (wk)

Arthroscopic procedures Unconstrained hemi or total shoulder arthroplasty Resurfacing arthroplasty Osteochondral grafting Unconstrained arthroplasty and posterior capsulorrhaphy Reverse shoulder arthroplasty

Simple sling Simple sling

1-2 as needed 2-4

Simple sling Simple sling Neutral rotation sling Neutral rotation sling

2-4 2-4 4 4

Shoulder 1117

  TABLE 17L-4 Limitation of Motion in Physical Therapy to Protect Repairs Procedure Performed

Motion Limited

Unconstrained total shoulder arthroplasty* Posterior capsulorrhaphy* (associated procedure)

External rotation past neutral Internal rotation, horizontal adduction

Duration of Limitation (wk) 4 4

*With the exception of these procedures, no limitations on range of motion are imposed.

c­ apsulorrhaphy are also instructed in pendulum exercises. The exercises are performed 3 to 5 times per day for about 15 minutes and are continued throughout the rehabilitative program. Hydrotherapy is used for all arthroplasty patients unless contraindicated because of deep venous thrombosis, fear of water, or chlorine allergy. Hydrotherapy is performed using a warm (35° C) rehabilitation pool to improve comfort without increasing body temperature and risk for inflammatory response. Supports accommodate straps and harnesses used in the rehabilitation process. The surgical wound is covered with a waterproof, air-permeable, hypoallergenic adhesive dressing, and patients are equipped with a mask and snorkel. The

A

C

inability to swim is not a contraindication to hydrotherapy rehabilitation, but anxiety and fearfulness of aquatic rehabilitation are contraindications. In this instance, we use a land-based program incorporating the same exercises used in the rehabilitation pool. Land-based therapy is effective in obtaining similar results to hydrotherapy; however, it generally takes longer to reach mobility goals, and patients tend to complain of more pain. Daily (5 to 7 days per week depending on availability) sessions in the pool last 30 to 45 minutes. Exercises designed to gain elevation, extension, horizontal ­adduction, internal rotation, and external rotation are performed in sets of 10 repetitions. The unaffected extremity is used for passive and active-assisted movement. An active gentle breaststroke motion with the palms horizontal to decrease water resistance is allowed with the patient supported by a harness and the shoulders submerged. Table 17L-4 provides motion limitations in physical therapy to protect soft tissue repairs. Short land-based verification sessions follow pool sessions to affirm improvements in mobility achieved in the pool. Modalities, such as cryotherapy, are used as required at the discretion of the therapist. Analgesic ­medication is provided for 6 weeks following surgery for postoperative discomfort.165 Shoulder motion exercises with total-body immersion are proposed to all patients but are not compulsory. Patients are fitted with weighted belts and instructed to

B

 Figure 17L-16  Capsular stretching exercises completed underwater in hydrotherapy or as part of a land-based therapy protocol. A, The siesta position with fingers interlocked behind the head. B, Anterior capsular stretching by external rotation.   C, Posterior capsular stretching with internal shoulder rotation.�

1118 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

hold their breath as they kneel or recline supine on the bottom of the pool. Mobility gains are accelerated when exercising in the insulated underwater environment.165 Once 140 degrees of elevation is obtained, the “siesta” position is reached with the hands clasped behind the head. Internal rotation of the shoulder in this position stretches the posterior capsule, and external rotation stretches the anterior (Fig. 17L-16). From the siesta position, the arms may be raised, and the triple-locking position is obtained, which stretches the inferior capsule (Fig. 17L-17). The siesta and ­triple-locking positions form a patient-directed program implemented after discharge from formal therapy sessions.165 Patients are re-evaluated after 5 weeks of hydrotherapy. Water therapy is discontinued and a land-based ­ self-rehabilitation regimen begins once acceptable motion is achieved. The siesta and triple-locking stretches are ­performed several times per day indefinitely. Additional sessions of hydrotherapy are ordered at 6-week intervals as needed. Most patients graduate to the self-directed program by 3 months after surgery. Patients without access to a physical therapist with a rehabilitation pool may learn the program from an experienced therapist and complete the program independently in any public or private pool. Additionally, patients adept at performing the exercises in the hydrotherapy program are allowed to work independently with only periodic monitoring by a trained therapist.

Return to Play After arthroscopic joint-sparing procedures, open procedures, and arthroplasty, activities of daily living generally resume by 6 weeks after surgery. Patients are asked to gradually resume their normal activities; this serves to increase strength and stamina with minimal risks.

  TABLE 17L-5  Timetable to Resume Golf and Tennis Six Weeks Three Months after Surgery after Surgery Golf

Putting

Tennis

Four or Five Months after Surgery

Six months after Surgery

Half-swing Full swing with Unrestricted with 7-iron all clubs from play from a tee a tee Gentle ground Slowly increase Unrestricted strokes intensity (no play overheads)

All patients are encouraged to pursue their preoperative activities and sporting events; however, we limit patients to noncontact sports. Most of our patients desire to continue or begin playing golf and tennis after arthroplasty. Restriction on immediate return to play protects the subscapularis. A timetable for return to golf and tennis is given in Table 17L-5. We do not specifically recommend strengthening exercises as part of the rehabilitation program; however, some of our younger patients ­ participate in weightlifting as part of their fitness regimen. Upper extremity weightlifting is allowed 6 months after surgery. Weightlifting for maintenance of muscle tone is encouraged, but powerlifting exercises are restricted. Our patients remain active after shoulder reconstruction, participating in trap shooting, water-skiing, snow-skiing, and mountain climbing 6 months after undergoing shoulder arthroplasty.

COMPLICATIONS Complications associated with arthroscopic procedures for treatment of the degenerative glenohumeral joint are extremely rare. Open shoulder reconstruction procedures are more prone to complications, with an incidence of nearly 20%. Tables 17L-6 and 17L-7 contain a list of common complications encountered during and after shoulder reconstruction and their treatment. The complications are discussed in relation to specific anatomic structures: humerus, glenoid, and soft tissues.   TABLE 17L-6  Intraoperative Complications and Treatment Location

Complication

Treatment

Humerus

Iatrogenic diaphyseal fracture

Humerus

Iatrogenic tuberosity fracture Iatrogenic glenoid fracture

Reduction and fixation— long stem prosthesis, allograft struts and cables as necessary Suture fixation (adjust postoperative rehabilitation) Rim fractures—no treatment required Glenoid body fractures— bone grafting and staged reconstruction Avoid with proper exposure—large defect may require reverse prosthesis Avoid overzealous retraction—follow with observation Avoid medial dissection; emergent intraoperative vascular surgery consult

Glenoid

Rotator cuff

Figure 17L-17   The inferior capsule is stretched using the triple-locking position by raising the arms from the siesta position.

Tendon disruption

Neurovascular Axillary and injury musculocutaneous nerve injury Neurovascular Large vessel injury injury

Shoulder 1119

  TABLE 17L-7  Postoperative Complications and Treatment Location/Diagnosis

Complication

Treatment

Wound

Hematoma

Humerus

Dehiscence Component loosening

Symptomatic; nonoperative protocol Irrigation and débridement if drainage persists > 7 days Symptomatic; local wound care Rule out infection Revision to a larger or cemented stem Nonoperative fracture care Operative indications: complete displacement, >30 degrees angulation, loosening, nonunion Revision surgery Revision to total shoulder arthroplasty Revision surgery Revision surgery with posterior capsulorrhaphy. or revision to reverse prosthesis Revision to reverse prosthesis Attempt at repair if acute. or revision to reverse prosthesis Revision to smaller component Arthroscopic release if failed 6 mo of nonoperative therapy Irrigation and débridement with component retention Six weeks of intravenous antibiotics Removal of components and intravenous antibiotics Revision surgery or resection arthroplasty, case specific

Periprosthetic fracture Glenoid

Component loosening Glenoid erosion (following hemiarthroplasty) Poor prosthetic alignment Capsule related (posterior instability most common) Rotator cuff related Failed subscapularis repair Prosthesis related Capsule related Early (within 6 wk)

Instability

Stiffness Infection

Late (after 6 wk)

C l Primary

r i t i c a l

P

o i n t s

glenohumeral arthritis in athletes is rare; therefore, an underlying diagnosis or secondary cause should be excluded. l Treatment should not be focused on short-term gains only, but directed toward durable results for the life of the patient. l Successful outcomes hinge on appropriate diagnosis and assessment of the functional demands of the patient and understanding the efficacy of treatment options. l A primary aim of the physical examination is to rule out confounding issues before focusing on the glenohumeral joint. l Pain is the most common indication for arthroplasty, yet motion, strength, and function also improve statistically after glenohumeral replacement. l Radiographs and CT should be used to define bony morphology and soft tissue quality; these influence treatment selection and results. l Treatment algorithms should be followed from conservative measures through surgical intervention in a stepwise manner in the treatment of glenohumeral arthritis. l A defined rehabilitation plan and structured goals should accompany treatment planning and patient education perioperatively.

S U G G E S T E D

R E A D I N G S

Edwards TB, Boulahia A, Kempf JF, et al: The influence of the rotator cuff on the results of shoulder arthroplasty for primary osteoarthritis: Results of a multicenter study. J Bone Joint Surg Am 84:2240-2248, 2002. Edwards TB, Kadakia NR, Boulahia A, et al: A comparison of hemiarthroplasty and total shoulder arthroplasty in the treatment of primary osteoarthritis: Results of a multicenter study. J Shoulder Elbow Surg 12:207-213, 2003. Friedman RJ, Hawthorne KB, Genez BM: The use of computerized tomography in the measurement of glenoid version. J Bone Joint Surg Am 74:1032-1037, 1992. Gartsman GM, Roddey TS, Hamerman SM: Shoulder arthroplasty with or without resurfacing the glenoid in patients who have osteoarthritis. J Bone Joint Surg Am 82:26-34, 2000. Liotard JP, Edwards TB, Padey A, et al: Hydrotherapy rehabilitation after shoulder surgery. Tech Shoulder Elbow Surg 4:44-49, 2003. Neer CS II: Replacement arthroplasty for glenohumeral osteoarthritis. J Bone Joint Surg Am 56:1-13, 1974. Neer CS, Watson KC, Stanton FJ: Recent experience in total shoulder replacement. J Bone Joint Surg Am 64:319-337, 1982. Samilson RL, Prieto V: Dislocation arthropathy of the shoulder. J Bone Joint Surg Am 65:456-460, 1983. Walch G, Boileau P: Prosthetic adaptability: A new concept for shoulder arthroplasty. J Shoulder Elbow Surg 8:443-451, 1999.

R E F E R E N C E S Please see www.expertconsult.com

1120 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

M

Nerve Lesions of the Shoulder Daniel C. Fitzpatrick and Kenneth P. Butters

SUPRASCAPULAR NERVE PALSY Suprascapular neuropathy has been reported in many types of athletes.1-3 Injury to the suprascapular nerve secondary to compression or traction most commonly occurs at either the suprascapular notch or the spinoglenoid notch. A direct blow or forceful scapular protraction may cause traction on the nerve at Erb’s point or kinking at either the suprascapular or spinoglenoid notch.4 When the injury occurs at the suprascapular notch, pain and motor weakness of both the supraspinatus and infraspinatus muscles may result. Compression of the suprascapular nerve by a ganglion at the spinoglenoid notch is a well-known clinical entity,5-9 resulting in isolated infraspinatus palsy, which often presents as painless posterior shoulder atrophy and weakness.

Anatomy and Biomechanics The suprascapular nerve arises from C5-C6 at the upper trunk of the brachial plexus, where it passes deep to the trapezius and the omohyoid. It enters the supraspinatus fossa through the suprascapular notch beneath the transverse scapular ligament (Fig. 17M-1). Together, the suprascapular notch and the overlying ligament form the suprascapular fossa. The suprascapular notch may assume various shapes, most commonly U-shaped (48% to 84%) and varies from flat to enclosed with bone.10,11 The width of the transverse scapular ligament parallels the size of the notch—that is, a larger bony notch results in a larger foramen.11 The nerve continues deep to the supraspinatus, innervating it with two motor branches, and sends sensory branches to the glenohumeral and acromioclavicular joints. There is no cutaneous sensory distribution from the suprascapular nerve. The nerve then reaches the lateral edge of the spine of the scapula and descends through the spinoglenoid notch, entering the infraspinatus fossa and innervating the infraspinatus. The spinoglenoid ligament passes from the spine of the scapula to the glenoid neck and posterior shoulder capsule. Its attachment into the posterior capsule results in tightening of the spinoglenoid ligament with cross-body adduction and internal rotation. The ligament is described in 14%12 to 100%13 of patients. Demirhan and associates14 found the spinoglenoid ligament present more commonly in men (64% to 36%), whereas Plancher and colleagues found the ligament to be present in equal proportions in men and women.13 Cummins and colleagues15 classified two types of spinoglenoid ligament: type I, a thin fibrous band (60%); and type II, a distinct ligament (20%), with an absent ligament in 20%. Bigliani and coworkers16 found the average

distance from the supraglenoid tubercle to the nerve at the suprascapular notch was 3 cm. The distance from glenoid rim to spinoglenoid notch is 1.8 to 2.1 cm.16,17 The suprascapular nerve is relatively fixed at its origin in the brachial plexus and at its terminal branches into the infraspinatus, resulting in several possible sites of injury.3 The two most commonly described locations of injury are the suprascapular fossa and the spinoglenoid notch. Although there is no translation of the nerve at the suprascapular fossa, the nerve forms an angle as it passes through the fossa. Nerve contact with the suprascapular ligament is accentuated with depression, retraction, or hyperabduction of the shoulder. This resulting “sling effect” may result in traction injury to the nerve.11 Cadaver studies18 also show that extremes of scapular motion can render the suprascapular nerve taut and clinically may result in suprascapular nerve injury.19,20 Ferretti and colleagues21 suggested an alternate mechanism of nerve compromise by hyperabduction of the shoulder with eccentric contraction of the infraspinatus resulting in compression of the ­suprascapular nerve at the spinoglenoid notch. Nerve compression against the

Suprascapular n.

Transverse scapular ligament

Spinoglenoid ligament

Infraspinatus m.

Figure 17M-1   Anatomy of the suprascapular nerve. (Redrawn from Black KP, Lombardo JA: Suprascapular nerve injuries with isolated paralysis of the infraspinatus. J Sports Med 18[3]: 225-228, 1990.)

Shoulder 1121

A

B

Figure 17M-2   Magnetic resonance image of the right shoulder shows a ganglion at the superior and posterior glenoid compressing the infraspinatus branch of the suprascapular nerve at the spinoglenoid notch.

lateral margin of the spine of the scapula by supraspinatus and infraspinatus tendons at their point of juncture is also thought to result in nerve injury.21 Ganglion cysts are a common cause of compressive injury to the suprascapular nerve. They result from superior labral tears, with the cyst expanding into the posterior scapular region, which is devoid of overlying muscle or tendon. Compression of the infraspinatus branch typically occurs as the nerve passes through the spinoglenoid notch (Fig. 17M-2).22 Suprascapular nerve palsy has also been reported following distal clavicle fractures and after resection of the distal clavicle.23,24 The nerve is located only 1.4 cm behind the clavicle and within 2 to 3 cm of the acromioclavicular joint (Fig. 17M-3).

Clinical Evaluation The athlete may have a history of trauma, but more often the complaint is vague posterior shoulder discomfort and weakness of insidious onset. Because there is no cutaneous distribution from the suprascapular nerve, pain is thought to arise from the articular branches to the acromioclavicular and glenohumeral joints. If the lesion is at the spinoglenoid notch, distal to the acromioclavicular and glenohumeral branches, the presentation may be one of painless atrophy of the infraspinatus and external rotation weakness. Diagnosis of suprascapular nerve compression by ­physical examination is difficult. It requires careful shoulder examination, including visual inspection of the posterior shoulder and testing of external rotation and supraspinatus strength. A complete neurologic evaluation of the neck and proximal extremity is also required. Pain is an inconsistent finding, but when present, it is usually located in the posterior shoulder and radiates to the arm; it may be worse with adduction of the shoulder.25 Posterior shoulder atrophy,

especially in the infraspinatus fossa, is an important finding (Fig. 17M-4). Because it is covered by the trapezius, supraspinatus atrophy may be difficult to observe. Weakness is a classic finding; however, weakness of the supraspinatus is not as easily exposed as that of the infraspinatus. Post and Mayer found suprascapular notch tenderness in seven of nine patients.25 Posterior ganglion cysts compressing the suprascapular nerve are thought to be associated with superior labral injuries. Compression typically occurs at the spinoglenoid notch, although cysts at the suprascapular notch have been described.26 Patients with clinical signs suggestive of a labral tear and wasting of the infraspinatus muscle warrant further diagnostic work-up, including magnetic resonance imaging (MRI) and electromyography (EMG) and nerve conduction velocity studies. Fractures of the scapula are also associated with associated nerve palsy. Edeland and Zachrisson described 18 scapular fractures, 7 with clinical involvement of the suprascapular nerve and only 1 with positive electromyographic findings.27 Treatment of suprascapular neuropathy after scapular fracture should probably include early exploration of the nerve with neurolysis and notch resection.

Diagnostic Studies A work-up for a patient with suprascapular nerve palsy should include shoulder views and, if necessary, a cervical spine series. A 30-degree cephalic tilt radiograph to visualize the suprascapular notch is helpful, especially in patients with fractures (Fig. 17M-5). MRI is useful in the evaluation of patients with suprascapular nerve palsy.28 Acute entrapment may be differentiated from chronic injury on T2-weighted images based on increased signal in the affected supraspinatus or

1122 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 17M-3   Superior view of the suprascapular nerve showing its proximity to the acromioclavicular joint and posterolateral clavicle.

Medial border of scapula

Infraspinatus

Spine of scapula

Superior angle

Acromion

Supraspinatus Suprascapular n.

Infraspinatus

Suprascapular notch and ligament

Greater tuberosity

1.4 cm 2-3 cm

Supraspinatus Bicipital groove

Coracoid process

Lessor tuberosity

Clavicle

Acromioclavicular joint

i­nfraspinatus muscles. Chronic compression appears as typical denervation changes, including decreased bulk and fatty infiltration of the muscles.29 Ganglion cysts in the supraspinatus fossa causing compression of the suprascapular nerve can be readily identified on MRI, as can associated pathology such as SLAP (superior labrum, anterior to posterior) tears and rotator cuff tears. Less common causes of suprascapular nerve palsy such as schwannoma30 and interneural ganglion31 have also been identified on MRI. Ultrasound is also reported as an effective diagnostic tool for the identification of paralabral cysts and ­associated rotator cuff tears.32 Ultrasound has the added benefit of

allowing aspiration of paralabral cysts, with ­ symptomatic improvement in 86% of patients in one series.33 The authors’ experience was not nearly as successful. A local anesthetic block in the suprascapular notch area is a useful part of a series of diagnostic injections. Electrical evaluation should include both EMG of the entire shoulder girdle, including the paraspinus muscles, and nerve conduction studies from Erb’s point to the supraspinatus. Normal latency values in nerve conduction studies are 1.7 to 3.7 msec to the supraspinatus and 2.4 to 4.2 msec to the infraspinatus. Nerve conduction studies should be abnormal to confirm the diagnosis of ­suprascapular nerve

Figure 17M-4   Rotator cuff atrophy.

Figure 17M-5   Radiograph of suprascapular notch fracture with 30-degree cephalic tilt.

Shoulder 1123

compression. Electromyographic abnormalities can also occur with brachial neuritis, cervical root compression, and incomplete brachial plexus stretch. Additionally, some think that electromyographic studies may be normal with an obvious clinical suprascapular nerve deficit,34,35 confirming the need for the nerve latency examination. Compression with ganglia may involve only one of the three or four suprascapular nerve branches to the infraspinatus, so electromyographic recordings should be done at more than one location within the muscle.

Treatment Treatment of a patient with closed, acute suprascapular nerve injury is initially conservative, with follow-up of the problem at frequent intervals, including electrical studies. A patient with a chronic condition (6 to 12 months) and well-established atrophy requires surgery, as does a patient with suprascapular nerve palsy associated with acute scapular fracture in the area of the suprascapular notch. The symptomatic patient with a ganglion cyst compressing the suprascapular nerve also benefits from surgical ­decompression.5-9,26

Spinoglenoid Notch Compression When a spinoglenoid ganglion is discovered as the cause of suprascapular nerve palsy, arthroscopy should be done to repair or débride associated labral lesions and decompress the labral cyst.36 Piatt and coworkers reported 97% patient satisfaction in a group treated with labral repair and open or arthroscopic decompression of the cyst ­versus 57% satisfaction in a nonoperative treatment group.8 Several authors have reported cyst resolution and return of nerve function following arthroscopic decompression of the cyst.5-9,26 Arthroscopic decompression of the cyst may be performed through a preexisting labral tear; however, if no labral tear is present, a capsulotomy is required. The preoperative MRI is useful for planning the exact location of the capsulotomy, which may be performed with an electrocautery device or a shaver. The cyst is visualized with the arthroscope in the posterior portal. A blunt probe is placed in the labral tear or capsulotomy until the characteristic amber-colored cyst fluid is visualized. Decompression is achieved by placing a shaver within the cyst and evacuating the fluid. The cyst wall may be removed using the shaver, but care must be taken to avoid iatrogenic injury to the suprascapular nerve. The shaver is pointed at the glenoid neck during removal, and dissection should not extend greater than 1 cm medial to the posterior labral attachment to the glenoid.9 Associated labral pathology is then addressed. Youm reported cyst resolution and return of nerve function following repair or débridement of an associated SLAP lesion without attempts at cyst débridement in 10 patients.37 If open decompression of the nerve at the spinoglenoid notch is necessary with excision of ganglia, a surgical approach to the posterior glenoid is performed. This approach is begun with a deltoid split over the glenohumeral joint with limited deltoid detachment laterally from the acromion. The superior edge of the infraspinatus is

identified, and at most, the upper one half of that tendon is detached, leaving a humeral side stump for repair.38 The size of the exposure needed is based on the MRI position of the ganglion and the size of the patient. Aspiration and steroid injection of posterior-superior ganglia is also good initial treatment, with one report of only one in five cases developing recurrent cyst. In painless infraspinatus muscle palsy without a cyst, function is usually good with nonoperative care. Asymptomatic ganglia without nerve findings may not require ­treatment.21,36,39,40

Suprascapular Notch Compression Compression of the suprascapular nerve at the suprascapular notch that fails nonoperative treatment may benefit from decompression. Most reports show good return of function in selected patients following open surgical decompression.41-43 Arthroscopic decompression has also been reported.44 Open surgical decompression of the suprascapular nerve at the suprascapular notch is performed with the patient in the lateral decubitus position. A skin incision is made parallel to the spine of the scapula, and subperiosteal removal of the trapezius attachment to the spine exposes the superior border of the supraspinatus. The upper border of the supraspinatus is carefully retracted inferiorly and posteriorly to expose the superior surface of the scapula and the suprascapular notch and ligament. The suprascapular artery crosses above the ligament and the nerve below. Ligament excision and appropriate bony resection should be performed with a laminectomy rongeur. Rask reported two cases in which repeat decompression of the nerve with bony resection gave good results; he recommends wide notch resection as primary treatment.45 Certainly, if there is any question about the nerve being free, notch resection is indicated. In thin individuals, a less extensile trapezius-splitting approach can be used through a strap incision across the spine of the scapula 2 cm medial to the acromioclavicular joint. The trapezius is split 5 cm in length centered over the skin incision.

Sports Suprascapular nerve injury may present after specific trauma, with chronic onset of pain or weakness, or with insidious painless muscle atrophy. Bateman stated that “athletic stress,” especially throwing, produces a backward and forward rotation of the scapula and suprascapular nerve compression at the notch.46 Jobe and colleagues have stated that in the athlete, the nerve is often injured as it passes around the lateral spine of the scapula, sparing the supraspinatus.47 In patients with spinoglenoid notch lesions, a program of therapy may allow the elite pitcher to return to high-level competition, assuming the infraspinatus is not completely denervated. Jobe and colleagues’ electromyographic studies showed that only 30% to 40% of the maximal strength of the infraspinatus is used during throwing. Therefore, in the case of a partial nerve injury, a return to pitching is possible.47 Ferretti and associates2 studied asymptomatic volleyball players and found that 12 of 96 had isolated partial infraspinatus paralysis mostly in

1124 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the dominant shoulder; some had electrical abnormalities, others had muscle atrophy, and there was a 15% to 30% loss of external rotation power.2 They suggested that the cause was nerve tension at the spinoglenoid notch when the arm is cocked in maximal stretch and during follow-through. In a long-term study, Ferretti had 35 such patients with isolated infraspinatus atrophy and re-­examined 16 patients at 5.5 years of average follow-up.21 All were still able to play volleyball at a high level with atrophy unchanged. He also found the incidence of subacromial impingement in this series to be no higher than in the general population of volleyball players. Suprascapular neuropathy has been reported with acute shoulder dislocation in a cyclist48 and with sudden onset after a hard throw from center field in a professional baseball player.49 The literature is confusing in that it refers to problems with the suprascapular nerve as both compression syndrome and nerve traction injury. The overall good response to conservative management suggests that traction injury rather than compression may be the cause. In the absence of a compressing lesion, rest from sports or other inciting causes may be helpful. Return to activity is permitted according to the judgment of the physician, based on factors in the course of follow-up, including the extent of the initial paralysis, electrical studies, symptoms, and improvements in the muscle examination with therapy. Surgical exploration of a well-localized lesion should be performed if conservative management of 3 to 6 months has failed.

LONG THORACIC NERVE PALSY Anatomy Long thoracic nerve palsy causing paralysis of the serratus anterior with winging of the scapula is a rather disabling lesion. This pure motor nerve is formed from roots of C5, C6, and C7, which branch shortly after they exit from the intervertebral foramina. Branches of C5 and C6 form the upper trunk of the nerve, which passes anteriorly through the middle scalene muscle. It then joins the lower trunk from C7 to form the long thoracic nerve. The nerve courses behind the brachial plexus to perforate the fascia of the proximal serratus anterior. It then passes medial to the coracoid on the frontal view and has an overall length of 30 cm (Fig. 17M-6).50 The nerve supplies a single muscle, the serratus anterior, which covers much of the lateral thorax and acts with the trapezius to position the scapula for elevation. It arises from the upper nine ribs and attaches at the deep surface of the scapula along the vertebral border. The muscles may be separated into upper, intermediate, and lower portions. The upper and intermediate portions are supplied by the upper division of the long thoracic nerve and produce shoulder protraction. The lower portion is primarily responsible for scapular stabilization.51 These portions typically work together to draw the scapula forward and rotate its inferior angle upward. The serratus anterior also acts as an accessory inspiratory muscle, as is seen in runners who fix their scapulae by holding their thighs to catch their breath after a race.

Long thoracic n.

Serratus anterior m.

Figure 17M-6   The brachial plexus. (Modified from Haymaker W, Woodhall B: Peripheral Nerve Injuries. Philadelphia, WB Saunders, 1956.)

Etiology of Disorders Isolated serratus anterior palsy may result from acute injury, chronic irritation, or brachial neuritis. Hester and colleagues described a tight fascial band between the inferior aspect of the brachial plexus and the region of the middle scalene insertion on the first rib.52 They noted the long thoracic nerve to “bow-string” over this band with shoulder abduction and external rotation. Medial and upward rotation of the scapula further compressed the nerve. Other authors have described internal traction on the nerve secondary to asynchronous motion between the arm and scapula as a cause of acute traction injury to the nerve.53 Long thoracic nerve palsy may also occur with prolonged recumbency or intraoperative stretch during thoracic surgery. Serratus anterior weakness following transaxillary first rib resection is not uncommon and has a good prognosis, although complete paralysis has a poor outlook.47 Other causes of nerve palsy include backpacking and shoveling. Proposed traumatic mechanisms include crushing of the nerve between the clavicle and the second rib, tetanic scalenus medius muscle contraction, and nerve stretch with head flexion or rotation and lateral tilt with ipsilateral arm elevation or backward arm extension.50 The outcome of acute traction injuries is good.11 Because the nerve is deeply located, a direct blow seems unlikely to cause isolated palsy. Serratus anterior rupture has been reported in patients with rheumatoid arthritis.4 The long thoracic nerve is often affected by the poorly understood syndrome of brachial neuritis. Brachial neuritis is a clinical syndrome of unknown cause and is the most common cause of serratus anterior palsy in our experience. Significant pain lasting a variable time—days to weeks— precedes loss of function in one or more shoulder girdle proximal extremity muscles. Sensory loss does not exclude

Shoulder 1125

Figure 17M-7   Scapular winging is often discovered during weight training as the scapula protrudes with resisted elevation or contacts the flat surface during bench press. If weightlifting is thought to be the cause, resumption of participation should await return of nerve function. Return to sports by patients with long thoracic nerve palsy depends on the demands placed on the upper extremity by the sport.

the syndrome. In the literature, there is a good prognosis for recovery, with 36% of patients recovered by the end of the first year and 75% by the end of the second year.54 Some improvement may occur after 2 years.48 Recurrent long thoracic nerve palsy is rare.50 Parsonage and Turner55 coined the term neuralgic amyotrophy (brachial neuritis) in 136 military personnel, 30 of whom had isolated serratus anterior paralysis. They also noted a right-sided ­predominance.

Clinical Evaluation Paralysis of the serratus anterior causes winging and a lack of scapular stabilization, limiting active shoulder elevation to 110 degrees in patients with complete lesions.50 Winging of the scapula is usually brought out with resisted active arm elevation or by doing a push-up while leaning against a wall (Fig. 17M-7). Scapular winging secondary to a long thoracic nerve palsy is characterized by elevation and retraction of the scapula such that the scapula moves toward the midline and slightly superior.53 Bertelli described the shoulder protraction test to identify upper trunk long thoracic nerve injuries.51 In this test, the patient is placed in the supine position and asked to elevate (protract) the shoulder. Ability to protract the shoulder indicates an intact upper trunk of the nerve. A patient with an early palsy may present with subtle changes in the ability to perform his or her sport, along with decreased active range of motion of the shoulder and altered scapulohumeral rhythm. The onset of long thoracic nerve palsy may be painful, as in brachial neuritis, or it may be more

subtle, involving problems with ­weightlifting or the feeling of pressure from a chair against the winging scapula while one is sitting. After an acute injury, several weeks may pass before marked scapular winging is evident. Gregg and colleagues believe that time is needed for the trapezius to stretch out and for scapular winging to become evident.50 Electromyographic studies confirm the diagnosis of long thoracic nerve palsy. Conduction studies should be performed from Erb’s point to the serratus anterior muscle on the anterolateral chest wall. Causes of winging other than serratus anterior palsy include trapezius palsy, painful shoulder conditions resulting in splinting of the glenohumeral joint, winging associated with multidirectional instability, and voluntary winging. The appearance of winging with arm elevation due to serratus anterior palsy differs from that of winging due to trapezius palsy. When the serratus anterior muscle does not function, the inferior tip of the scapula is pulled medially and posteriorly. With trapezius paralysis, the scapular body is held in position, and the medial border merely becomes more prominent, a more subtle deformity. In neither type of winging is the scapula rotated laterally to facilitate arm elevation.

Treatment Cessation of the suspected inciting activity is important. Shoulder braces cannot begin to normalize the force couple on the scapula between the serratus anterior and the trapezius. However, in the case of severe serratus winging, braces may prevent stretching out of the trapezius muscle. Surgically, pectoralis minor transfer56,57 to the lateral inferior scapula for dynamic support has been reported. Transfer of the pectoralis major (sternal head) with fascia lata extension to the inferior border of the scapula is the currently favored reconstruction.58 Fortunately, surgical treatment is seldom needed.

Sports Sports have been implicated as a cause of isolated serratus anterior palsy,48,50,59 with traction injury—single or ­repetitive—to the long thoracic nerve being the proposed mechanism. In one series, the repetitive trauma of ­tennis and archery was thought to be the cause of the lesion in 5 of 20 patients. Other sports implicated in this type of injury are basketball, football, golf, gymnastics, and ­wrestling.60

ACCESSORY NERVE PALSY Anatomy The spinal accessory nerve is a pure motor nerve innervating the trapezius and sternocleidomastoid muscles. The nerve leaves the jugular foramen at the base of the skull, goes through the upper third of the sternocleidomastoid muscle, and crosses the posterior triangle of the neck. It is here that it is superficial and vulnerable to injury. The nerve enters the trapezius and is the predominant motor nerve to that muscle. Root fibers from C3 and C4 also innervate the trapezius and may blend with the ­accessory

1126 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 17M-8   A, Drooping of the right shoulder when the patient is relaxed. B, There is no voluntary elevation of the shoulder on the right compared with the left.

nerve; some feel that this C3-C4 contribution is only proprioceptive.61 The accessory nerve is small—only 1 to 3 mm in diameter.62 Scapular stabilization and elevation result from the balance of the forces between the trapezius and serratus anterior. The upper trapezius elevates and tilts the ­scapula, raising the point of the shoulder and assisting in arm elevation, respectively. The lower trapezius works with the rhomboids to retract the scapula and balance the pull of the serratus anterior. The nerve may be damaged, as it is most commonly, during a posterior triangle node biopsy. Sports injuries are typically by a direct blow—for example, with a hockey stick or in a traction injury with a cross-face maneuver in wrestling.63 Stretch injury resulting from distal upper extremity distraction and contralateral head rotation has been reported.64

Clinical Evaluation The patient complains of a sagging shoulder and incomplete arm elevation with loss of strength (Fig. 17M-8). The symptoms may be quite severe owing to muscle spasm and brachial plexus traction neuritis. Examination shows a drooping of the shoulder and a deepening of the supraclavicular fossa after trapezius atrophy has occurred. Winging of the scapula occurs with resisted arm elevation and with active external rotation against resistance.65 Winging secondary to trapezius palsy differs from that seen with a serratus anterior palsy in that the scapula lowers and translates to the midline while the inferior angle is drawn up by the levator scapulae. The levator scapulae is palpable and is seen as a band of muscle in the neck; rhomboid contraction is also palpable on attempted scapular adduction. EMG is used to provide the definitive diagnosis, but its role in determining prognosis for recovery has been questioned.66

Treatment Most athletic injuries to the spinal accessory nerve are closed and initially treated with observation.53 Good functional results are generally expected with closed injuries.66 Teboul and colleagues thought that exploration was indicated if no clinical or electrical signs of improvement existed at 3 months.67 However, they also showed good results with neurolysis up to 20 months after injury if the trapezius responded to interoperative electrical stimulation of the nerve. If the nerve injury has lasted longer than 20 months, reconstruction is advised. Clinical indications for exploration or reconstruction include a symptomatic patient with upper extremity drooping, aching, numbness, and incomplete active arm elevation. Adjacent scapular muscles cannot substitute for a paralyzed trapezius with muscle strengthening alone. The current operation of choice was described by Bigliani and associates.68 The levator scapulae and rhomboids are moved to a more lateral insertion on the scapula to substitute for the upper, middle, and lower trapezius. Other operations described include a scapular suspension with fascial grafts from the vertebral spine to the medial scapula or from the ribs to the scapula, and scapulothoracic fusion.69

Sports Spinal accessory injuries in sports are rare. Cases have been reported of a wrestler and a hockey player63 with closed accessory nerve palsy, and the authors have seen a rugby player with palsy resulting from a direct blow; all recovered nerve function with observation. Winging is less obvious and often is less disabling with trapezius palsy than with serratus anterior palsy. However, shoulder function in an athlete with accessory nerve palsy is often inadequate for competition.

Shoulder 1127

c

r i t i c a l

P

S U G G E S T E D

o i n t s

R E A D I N G S

Bencardino JT, Rosenberg ZS: Entrapment neuropathies of the shoulder and elbow in the athlete. Clin Sports Med 25:465-487, 2006. Bertelli JA, Ghizoni MF: Long thoracic nerve: Anatomy and functional assessment. J Bone Joint Surg Am 87:993-998, 2005. Duralde XA: Neurologic injuries in the athlete’s shoulder. J Athl Train 35:316-328, 2000. Friedenberg SM, Zimprich T, Harper CM: The natural history of long thoracic and spinal accessory neuropathies. Muscle Nerve 25:535-539, 2002. Piatt BE, Hawkins RJ, Fritz RC, et al: Clinical evaluation and treatment of spinoglenoid notch ganglion cysts. J Shoulder Elbow Surg 11:600-604, 2002. Plancher KD, Peterson RK, Johnston JC, Luke TA: The spinoglenoid ligament: Anatomy, morphology, and histological findings. J Bone Joint Surg Am 87:361365, 2005. Teboul F, Bizot P, Kakkar R, Sedel L: Surgical management of trapezius palsy. J Bone Joint Surg Am 86:1884-1890, 2004. Westerheide KJ, Dopirak RM, Karzel RP, Snyder SJ: Suprascapular nerve palsy secondary to spinoglenoid cysts: Results of arthroscopic treatment. Arthroscopy 22:721-727, 2006. Youm T, Matthews PV, El Attrache NS: Treatment of patients with spinoglenoid cysts associated with superior labral tears without cyst aspiration, debridement, or excision. Arthroscopy 22:548-552, 2006.

l Suprascapular neuropathy occurring at the suprascapular notch presents with pain and motor weakness of both the supraspinatus and the infraspinatus. l Suprascapular neuropathy occurring at the spinoglenoid notch presents with isolated infraspinatus palsy occurring as painless atrophy. l Spinoglenoid notch cysts occurring as a result of superior labral tears are a common cause of suprascapular ­neuropathy. l MRI and EMG/NCV studies of the entire shoulder ­girdle are valuatble in the work-up of suprascapular nerve ­compression. l Scapular winging secondary to a long thoracic nerve palsy affecting the serratus anterior muscle is characterized by medial, posterior, and slight superior displacement. l Scapular winging secondary to a spinal accessory nerve palsy affecting the trapezius is characterized by prominence of the medial boarder of the scapula.

R E F E R E N C E S Please see www.expertconsult.com

S e c t i o n

N

Thoracic Outlet Syndrome Karim Abdollahi and Virchel E. Wood

HISTORY Thoracic outlet syndrome (TOS) was described by Paget as an effort thrombosis of the subclavian vein in 1875 and was discussed similarly by von Schroetter in 1884.1 In 1740, Hunauld2 first described compression of the thoracic outlet secondary to a cervical rib. In 1919, Stopford and Telford showed that the neurovascular structures could be compressed by the first thoracic rib and that surgical removal of this rib would alleviate symptoms of the compression.3 The first cervical rib removal was performed by Coote4 in St. Bartholomew’s Hospital in 1861. Twenty-nine years later, the second case of a cervical rib was removed. Murphy5 was the first to remove a normal first thoracic rib for TOS in 1910. TOS has been called by various names (Box 17N-1).

ANATOMY The thoracic outlet involves the area of the shoulder girdle and thorax in which the subclavian artery and vein exit the chest cavity and combine with the brachial plexus, passing

through the scalene triangle over the first rib and under the clavicle to enter the axillary region of the shoulder (Fig. 17N-1). The anatomic boundary of the thoracic outlet consists of the superior surface of the first rib and the anterior scalene muscle and the middle scalene muscle, both of which insert into the first rib. The clavicle overrides the neurovascular structures and applies pressure on the thoracic outlet if it is displaced or positioned posteriorly. Neurovascular compression within the thoracic outlet involves the subclavian artery, the subclavian vein, or the brachial plexus within this area. Compression may involve a cervical rib abnormality or the anterior or middle scalene muscle, or it may occur at the costoclavicular or subclavian tendon junction, at the level of the first rib, or as far laterally as the pectoralis minor insertion into the coracoid process.6 Anatomic factors include an inadequate intrascalene triangle (which may be due to anterior or middle scalene hypertrophy or spasm), a high first thoracic rib, or descent of the shoulder girdle with age, allowing a sagging effect and compression of the neurovascular structures. ­ Congenital

1128 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 17n-1 synonyms for thoracic outlet ­syndrome Shoulder-hand syndrome Paget-Schroetter syndrome Cervical rib syndrome First thoracic rib syndrome Scalenus anterior syndrome Brachiocephalic syndrome Scalenus minimus syndrome Scalenus medius band syndrome Costoclavicular syndrome Humeral head syndrome Hyperabduction syndrome Nocturnal paresthetic brachialgia Fractured clavicle syndrome Pneumatic hammer syndrome Cervicobrachial neurovascular compression syndrome Effort vein thrombosis Rucksack paralysis Pectoralis minor syndrome Cervicothoracic outlet syndrome Subcoracoid syndrome Syndrome of the scalenus medius band Naffziger’s syndrome Acroparesthesia

C1 C2 C3

Middle scalene m.

C4 Anterior scalene m.

C5

Upper branchial plexus A

Lower branchial plexus Coracoid process

B

C6 C7 T1

C Axillary

factors, such as a cervical rib, a rudimentary or anomalous first thoracic rib, variant scalene muscles, an elongated transverse process, or adventitial fibrotic bands, may be present.7 Wood and Marchinski8 described anomalous muscles such as the axillopectoral muscle (4% to 8%), chondroepitrochlear muscle, and subscapularis-teres-latissimus muscle (5.2%). Further considerations include traumatic factors, such as fracture of the clavicle, injuries to the cervical vertebrae, dislocation of the head of the humerus, and atherosclerosis of the major arteries at the isthmus of the neck of the humerus.9 Kofoed10 emphasized the necessity of ruling out cervical disk herniation in the evaluation of TOS. Differential diagnoses in TOS include any pathology creating pain in the neck, arms, or shoulders (Box 17N-2). Roos11 observed that 98% of his patients with TOS had anomalous fibrous muscular bands that probably irritated or compressed the brachial plexus. Nine different bands were described. The most frequent is type 3, which is a fibromuscular structure originating on the neck of the first rib and passing horizontally across the thoracic outlet to lie between the T1 root of the plexus and the subclavian artery (Fig. 17N-2). Type 7 is a fibrous cord attaching to the anterior surface of the anterior scalene passing under the subclavian vein to attach to the posterior surface of the sternum (Fig. 17N-3). The other seven types have been described in detail.11 TOS usually involves the lower plexus, but when the upper plexus is involved, other abnormalities may be present. Roos11 described five types of anomalies that primarily involve the relationship between scalene muscles and the upper brachial plexus. The patient presents clinically with symptoms of median nerve compression. The most frequent upper plexus anomaly is type 3, in which the anterior scalene muscle passes between the roots and trunks of the plexus (Fig. 17N-4). TOS may be caused by compression in the subcoracoid space, which is immediately posterior to the origin of pectoralis minor from the coracoid process.12 When performing Wright’s hyperabduction maneuver, the neu-

a. v.

Pectoralis minor m.

Figure 17N-1   Compression of neurovascular structures may occur at three points. The brachial plexus may be compressed between the anterior and the middle scalene muscles, causing upper thoracic outlet syndrome (A). Most commonly, compression occurs between the clavicle and the first rib (B). The pressure effect between the pectoralis minor and the rib cage is specifically assessed when performing Wright’s hyperabduction maneuver (C). a, Artery; m, muscle; v, vein.

Box 17n-2 Differential ­Diagnosis of Thoracic Outlet ­Syndrome Cervical radiculopathy (caused by disk protrusion, ­osteophytes) Cervical arthritis causing radiating pain to shoulder Shoulder bursitis (impingement syndrome, rotator cuff tendinitis) Glenohumeral joint instability (subluxation of humeral head irritating the brachial plexus) Brachial plexus neuritis (Parsonage-Turner syndrome) Peripheral nerve compression (cubital and carpal tunnel syndromes) Neoplasm of the spinal canal Rheumatologic conditions (fibromyalgia, myositis, ­fibrositis, Raynaud’s phenomenon) Thromboangiitis Neoplasm of the peripheral nerve Apical pulmonary neoplasm Mass effect (axillary tumor)

Shoulder 1129

Type 3 Type 7

Figure 17N-2   Type 3 is the most frequent type of anomalous band encountered in thoracic outlet syndrome.   (From Wood VE, Twito RS, Verska JM: Thoracic outlet syndrome: The results of first rib resection in 100 patients. Orthop Clin North Am 19:131-146, 1988.)

Figure 17N-3   A type 7 band found in thoracic outlet syndrome may cause venous thrombosis. (From Wood VE, Twito RS, Verska JM: Thoracic outlet syndrome: The results of first rib resection in 100 patients. Orthop Clin North Am 19:131-146, 1988.)

rovascular bundle gets stretched by the coracoid process and compressed by the pectoralis minor in the subcoracoid space and may lead to loss of radial pulse or to paresthesias, or both. Effort thrombosis of axillary or subclavian vein appears to be related to overstretching of the venous wall or occluding pressure against the vein by the first rib.

which the neck or upper torso was injured.13 In our cases, the problem with the ulnar nerve alone was the most common symptom,14 followed by problems of the artery and the median nerve alone in 12%. There were many combinations of symptoms. Two of our patients were admitted to the hospital for a heart attack, and two other patients had Raynaud’s phenomenon. Three women had unilateral breast swelling and severe breast pain. Shoulder pain in athletes usually is due to musculos­ keletal problems, such as impingement syndrome and glenohumeral instability. Less commonly, the shoulder pain may be caused by TOS. Repetitive throwing activities in the extended, abducted, externally rotated position of the arm aggravate the symptoms (Fig. 17N-5). Pressure on the brachial plexus and artery, especially during overhead exertions, may result in fatigue, aching, and inability to perform competitive activities such as swimming.15 The swimmer may present with complaints of inability to keep fingers together during the pull-through phase of the swimming strokes. The water polo athlete may have trouble grabbing, holding, and throwing the ball. These symptoms are due to weakness in the intrinsic muscles of the hand, which suggests compromise of C8 and T1 nerve roots. Effort thrombosis of the subclavian vein (PagetSchroetter syndrome) has been reported in a competitive swimmer.16 TOS of vascular origin occurs in only 2% of

SYMPTOMS Although pain, weakness, and neurovascular deficits are associated with TOS, often the symptoms are quite bizarre and may be intermittent. Awareness of the occurrence of TOS and the presence of clinical objective findings are necessary to diagnose this syndrome. Neurologic symptoms consist of weakness, fatigability, numbness, and tingling, particularly in the distribution of posterior and medial cords of the brachial plexus. A pain diagram with a questionnaire can be helpful. Vascular symptoms consist of ischemia, claudication, cold intolerance, swelling with venous congestion, and occasional thromboembolic phenomena with distal arterial occlusion. Symptoms are especially pronounced with arm elevation above the level of the shoulder, particularly during throwing, combing the hair, or sleeping with the arm above the head. Half of our patients indicated that a single traumatic event precipitated TOS, such as a motor vehicle crash in

1130 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Anterion scalene m.

Figure 17N-4   In a type 3 anomaly, the anterior scalene muscle passes between the roots and the trunks of the plexus. (From Wood VE, Ellison DW: Results of upper plexus thoracic outlet syndrome operation. Ann Thorac Surg 58:458-461, 1994.)

patients; 97% present with neurologic symptoms. Athletes may present with generalized aching, fullness, and swelling of the arm. If acute, swelling may be significant; superficial veins may not drain with arm elevation. The incidence is higher in young, physically active males 15 to 40 years old.15 Ten of our patients presented with a venous ­thrombosis, Figure 17N-5   Axillary artery compression by the pectoralis minor muscle at the coracoid process insertion in the throwing athlete.

which represented 3% of our patients with TOS. PagetSchroetter syndrome has been reported in several overhead athletes, including weightlifters (Fig. 17N-6). Because aquatic athletes are primarily overhead athletes, one may expect a higher incidence of TOS in this population. Swimmers require controlled, repetitive power strokes at the extremes of abduction and external rotation of the shoulder.12 If the athlete complains of tightness and pain about the shoulder at the point the hand enters the water, the physician should be alerted to the possibility of TOS. Water polo also subjects the shoulder to repetitive abduction and external rotation in throwing and in blocking a shot. In upper TOS, athletes may complain of pain in the lower face and ear; headaches; and radiation of the pain to shoulder, thumb, and index and middle fingers.12 This presentation should be distinguished from swimmer’s ear (otitis externa). Patients also may complain of weakness or fatigue in the upper extremity muscles. Priest and Nagel17 described tennis shoulder as a depression or drooping of the exercised shoulder that they attributed to stretching of the muscles that elevate the shoulder and hypertrophy of the extremity. Shoulder droop may induce TOS by increasing the pressure at the thoracic ­outlet. Symptoms may be reduced by strengthening the shoulder-elevating muscles (levator scapulae, rhomboids, and upper trapezius). Rayan18 reported on two young athletes with TOS resulting from a cervical rib. Their symptoms increased with sporting activities. They responded well to resection of the cervical rib. Four cases of TOS in athletes were reported by Strukel and Garrick.3 Their patients responded well to conservative treatment. TOS is seen not only in the throwing athlete3 but also in heavy, muscular athletes, such as weightlifters, football players, and athletes who may sustain traction injuries to

Shoulder 1131

A

B

Figure 17N-6   A, This weightlifter developed effort vein thrombosis with swelling and pain in the left arm. B, The venogram shows multiple clots. This was called Paget-Schroetter syndrome in the older literature. (From Wood VE, Twito RS, Verska JM: Thoracic outlet syndrome: The results of first rib resection in 100 patients. Orthop Clin North Am 19:131-146, 1988.)

the upper arm and chest. Direct trauma that results in rib fractures, transverse process fractures, clavicular fractures, and shoulder dislocations may precipitate thoracic outlet symptoms.19

PHYSICAL EXAMINATION The physical examination is the most important aspect of diagnosing TOS. Adequate time must be allowed to perform the initial evaluation. The female athlete should be in a gown with her hair up. The shoulders should be observed for slouching. The physician should note if the breasts are large. The physician should look for any shoulder asymmetry, especially unilateral drooping or hypertrophy, as may be seen in professional tennis players, shot putters, javelin throwers, and other overhead athletes. The physician should note any venous engorgement or arm swelling (vein thrombosis). The physician should palpate the clavicle for deformity and the supraclavicular area for a cervical rib. Muscles and bones around the shoulder should be palpated, and the physician should document areas of tenderness, especially areas where the patient complains of pain. Sensation may be tested by static and moving two-point discrimination. We prefer the 10 test, in which the same area in both upper extremities is touched and the patient is asked to rate the symptomatic side 0 to 10 with 10 being normal sensation (as felt on the asymptomatic side) and 0 being no sensation at all. By sensory testing, one can attempt to distinguish between lower TOS (ulnar nerve distribution) and upper TOS (median nerve distribution). If pathology at the nerve root level is suspected, the physician should test sensation by dermatomes and note any reflex asymmetry. Motor weakness, such as with intrinsic muscles of the hand, may be subtle. Comparison to the opposite side is essential.

The physician should test for distal sites of nerve compression, such as carpal or cubital tunnel syndromes, before doing provocative maneuvers for TOS.20 This practice helps minimize false-positive findings of peripheral nerve compression that are seen when the brachial plexus is irritated by provocative maneuvers first.

PROVOCATIVE SIGNS For a test to be positive, either the symptoms must be reproduced or the radial pulse must shut off. If the pulse shuts off, the opposite side also should be tested. Less value is placed on a positive test result if the pulse shuts off on the asymptomatic arm also. The physician should ask the patient if the baseline paresthesia gets worse with the maneuver and returns to baseline after the maneuver. Between tests, the physician should give the patient a few seconds to shake his or her hand and recover from paresthesias caused by the maneuver.

Adson’s Test The physician palpates the radial pulse and abducts the arm slightly (Fig. 17N-7). The physician asks the patient to hyperextend the neck and turn it to the affected side and inhale deeply. Diminution or obliteration of the pulse probably is due to compression of the axillary artery by the anterior scalene muscle. The patient should turn the head to the opposite side (reverse Adson’s test) to test compressive effect of the middle scalene.

Halstead’s Maneuver The physician has the patient retract the shoulders downward and backward to draw the clavicle closer to the first rib (Fig. 17N-8). The physician palpates the pulse and asks about worsening paresthesias.

1132 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Roos’ Test Both shoulders are abducted 90 degrees and externally rotated 90 degrees (Fig. 17N-10). The patient opens and closes both hands rapidly for 3 minutes. The physician asks the patient if the two hands feel different from each other and if there are any paresthesias or numbness in the involved hand. A cold sensation or rapid fatigue is suggestive of arterial compromise. A patient with TOS is unable to keep the arms and hands elevated because of an ischemic, vascular type of pain.

Retroclavicular Spurling’s Test

Figure 17N-7   Adson’s test: Hold patient’s arm in slight abduction while palpating the radial pulse. Ask the patient to extend the neck and rotate toward the affected side. Adson’s test is positive if the patient reports paresthesias or if the pulse fades away.

The physician places the thumb flat and deep into the retroclavicular space and attempts to compress the brachial plexus and vascular structures (Fig. 17N-11).21 The physician should ask the patient if he or she has increasing numbness or tingling in the hand. If the answer is yes, the test is performed on the opposite side as a control. Wright,22 in his series of 150 asymptomatic normal subjects, found that 92.6% had obliteration of the radial pulse in at least one upper extremity tested in the elevated position. In our experience, the best objective test to diagnose TOS is Roos’ test.7 The retroclavicular Spurling’s test is

Wright’s Hyperabduction Test The physician passively abducts the patient’s shoulder to 180 degrees and extends it to compress brachial plexus and vessels between the pectoralis minor and the rib cage (Fig. 17N-9). This maneuver also stretches the structures under the coracoid process. The physician should palpate the pulse and ask about worsening paresthesias.

Figure 17N-8   Halstead’s maneuver: Ask patient to pull shoulders backward and downward to depress the clavicle against the first rib. Halstead’s maneuver is positive if the patient reports paresthesias or if the pulse is diminished.

Figure 17N-9   Wright’s hyperabduction maneuver: Passively abduct the affected side to 180 degrees while palpating the radial pulse. Paresthesias or diminution of the pulse suggests a positive test.

Shoulder 1133

Figure 17N-10   Roos’ test: Ask the patient to abduct and externally rotate the shoulders and open and close both hands simultaneously for up to 3 minutes. Worsening paresthesias on the affected side indicate a positive Roos’ test.

our second most reliable objective test. We believe that at least three or four of the aforementioned five signs should be clearly positive to make the diagnosis of TOS.

DIAGNOSTIC STUDIES Diagnostic tests consist of routine chest, cervical spine, and shoulder radiographs. Radiographic evaluation for cervical ribs (Fig. 17N-12), anomalous first and second ribs, pathologic clavicular fractures, and space-occupying lesions such as tumor or aneurysm must be ruled out. Arteriography documents arterial compression and possible aneurysm formation about the first rib. Venography documents venous compression or occlusion (see Fig. 17N-6). Peripheral vascular studies, including pulses,

Figure 17N-11   Retroclavicular Spurling’s test: Press the thumb down into the space behind clavicle. Paresthesia into the ipsilateral hand indicates a positive test.

Figure 17N-12   Prominent cervical rib seen on the right can compress the neurovascular structures in the thoracic outlet.

blood pressure measurements, and Doppler studies, aid in the diagnosis of thoracic outlet compression and occlusion of the arterial supply to the arm. Electromyography and nerve conduction velocity are negative in most cases of TOS. This test still is recommended, however, because it can help diagnose cervical radiculopathy, carpal tunnel, and cubital tunnel syndromes, which can be seen commonly as part of a double-crush phenomenon.23 These other sites of compression can be treated to decrease ­symptoms. MRI has been used to evaluate TOS.24-26 Using special techniques in an open MRI scanner, patients were imaged at 0 and 90 degrees and compared with normal subjects. A significantly smaller distance between the rib and the clavicle was seen in the patients with TOS. On coronal views, the compression of the brachial plexus often could be visualized in abduction. Gadolinium-enhanced magnetic resonance angiography, in the neutral and the abducted position, is a good screening test for patients suspected of having TOS. More recently, lidocaine injection into the scalene muscles has been promoted to help confirm the diagnosis of TOS.27,28 Under fluoroscopic and electromyographic guidance the anterior and middle scalene muscles are anesthetized using lidocaine, which temporarily paralyzes the muscles. With relaxation of these two muscles, there will be less compression on the upper brachial plexus. Shortly after the injection the provocative tests for TOS are repeated and accurately documented. The athlete is also asked to estimate (in percentage) how much relief they obtained from the injection. If physical signs show improvement and the patient reports significant reduction of symptoms shortly after these injections, he or she would be a candidate for injection of botulinum toxin into the same scalene muscles or surgical decompression. Unfortunately, there is no single test that is diagnostic for TOS. The diagnosis is based on the history, physical findings, and supportive diagnostic testing.

1134 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

NONSURGICAL TREATMENT In most cases, nonoperative treatment is the initial form of management when the diagnosis of TOS is suspected.29 The treatment program includes patient education, behavior modification, and joint mobilization exercises.30 If successful, this program is followed continually and should not be terminated. Continued evaluation and monitoring of the patient are necessary. Behavior modification consists of altering sleep patterns, working patterns, and driving patterns and taking general precautions for activities that could compromise the thoracic outlet. Faulty posture must be corrected. Shoulder exercises with emphasis on gradual scapular retraction and shoulder range of motion increase joint motion and allow the cervical spine to achieve axial body and extremity extension and open the thoracic outlet space. Diaphragmatic breathing exercises often are indicated to aid in respiration when there is hypertrophy of the accessory muscles of the chest and neck. In particular, elevation of the rib cage by the pectoralis minor and scalene muscles should be eliminated because this tendency decreases the thoracic outlet space. Weak musculature about the neck and shoulder should be strengthened, and shoulder posture should be improved. The upper trunk muscles, such as the serratus anterior, middle and lower trapezius, latissimus dorsi, and rhomboids, must be strengthened. Joint mobilization techniques for the sternoclavicular, acromioclavicular, and scapulothoracic joints improve and increase the costoclavicular space. Likewise, mobilization of the occiput on the atlas facilitates axial extension body movements and improves the symptoms of TOS. Specific joint mobilization and therapy programs have been outlined by Smith.31 Conservative management should produce improvement in symptoms 1 to 3 months after the onset of symptoms. Physical therapy should not continue indefinitely and should be discontinued if it substantially exacerbates the symptoms.

SURGICAL TREATMENT When symptoms persist or become worse with conservative management, surgical intervention may be necessary.32 Patients with severe, intractable pain; disability; arterial or venous compromise, or neurologic compromise fall into this category.33 Surgery ranges from resection of the scalenus anterior, described by Adson in 1927,34 to removal of the first thoracic rib, described by Murphy5 in Australia and Brickner35 in the United States. Anterior and supraclavicular approaches have been used by these authors for first rib resection. Another approach used by Clagett36 was a limited posterior thoracotomy incision, and a transaxillary

Authors’ Preferred Method

of

approach was used by Roos.37 First rib resections now are done most commonly according to the techniques of Roos through a transaxillary approach. In addition to first rib resection, this approach allows the possibility of performing a thoracic sympathectomy at the same time for pain relief. Scalenectomy by itself does not provide predictable relief of TOS symptoms. Strict attention must be paid to the pathology, and when other causes appear to be operative, they must be corrected. Such correction may include cervical rib resection or excision of callus after a fractured clavicle. Vascular changes, particularly aneurysmal dilation or thromboembolism of the intima of the vessel, must be addressed and corrected surgically at the time of thoracic outlet decompression. In certain instances, a combination of supraclavicular and transaxillary approaches has been used for decompression of the thoracic outlet, first rib resection, and vascular reconstruction, including brachial plexus exploration as indicated. Edwards and colleagues38 performed 52 transaxillary first rib resections in 46 patients, and 42 patients (91%) had immediate improvement in symptoms after surgery, but symptoms recurred in 3 patients 6 to 8 months postoperatively. Donaghy and colleagues’39 surgical treatment of suspected neurogenic TOS relieved pain and sensory disturbance in 90% of patients but was less effective for relieving muscle weakness (50%). Thrombosis of the subclavian or axillary vein may require fibrinolytic therapy with intravenous streptokinase.16 When recannulation of the vein is confirmed by venogram, after an appropriate period of warfarin treatment (about 4 months), a first rib resection is recommended. The postoperative care is simple. The patient is allowed limited range of motion of the arm for 3 to 4 weeks for activities of daily living. Gentle active range of motion is encouraged. At 4 weeks, if the range of motion of the shoulder is restricted, we send the patient to physical therapy for range of motion, scar tissue management, and strengthening exercises. By 2 months, the patient should have full use of the operated extremity.

CRITERIA FOR RETURN TO SPORTS PARTICIPATION Resumption of sports depends on return of range of motion, strength, and endurance in the shoulder girdle and upper extremity. A recovery period of 6 months to 1 year usually is required to maximize functional return in competitive athletes. In the case of effort thrombosis of axillary or subclavian vein, retirement from competitive swimming is likely because of the long-lasting effects.15

Treatment

Conservative Treatment

Conservative treatment of TOS is aimed at reducing inflammation around the brachial plexus; improving the posture of neck, shoulders, and upper back; and treating muscle spasms.12,40 Nonsteroidal anti-inflammatory medications and oral steroids should help decrease nerve irritation.

Therapists experienced in dealing with TOS can work on improving flexibility of the shoulder to allow more space ­between the clavicle and the first rib. Correcting posture and improving muscle balance should decrease compression of the neurovascular structures (Fig. 17N-13).

Shoulder 1135

Authors’ Preferred Method

of

T r e a t m e n t — cont ’ d

ALGORITHM FOR TOS History & physical examination consistent with TOS

Equivocal

Work-up for other compression neuropathy (e.g., cervical radiculopathy, carpal tunnel syndrome)

Yes C-spine radiograph to R/O cervical rib

Physical therapy 4-8 wk

If better

If positive

Home exercise program

If not better Work-up for other compression neuropathy with EMG/NCS and/or neck MRI

If positive

Treat the diagnosed condition

If negative Lidocaine injection of scalene muscles

Consider Botox injection of scalenes or surgery

Good response

No good response Re-evaluate for other conditions: rheumatologic disorder fibromyalgia Figure 17N-13   Algorithm for the treatment of thoracic outlet syndrome.

Most upper extremity sports, such as tennis, baseball, and all water sports, involve repetitive overhead motions. A period of rest from the offending activity should help. The athlete should be advised to avoid sleeping with the arms overhead either prone or supine. Muscle spasm is thought to play a role in causing TOS. Stretching of scalenes, pectoralis major and minor, trapezius, and levator scapulae may help relieve symptoms or prevent symptoms in the future. Success of conservative treatment of TOS is reported to be 50% to 90%.12 Our experience has shown, however, that many patients cannot tolerate physical therapy and become much worse. A negative response to therapy may be helpful in the decision process as to when surgery should be done. Another nonoperative approach would be injection of the anterior and middle scalene muscles with botulinum toxin.27,28 This would be done under fluoroscopic and electromyographic guidance in athletes who responded to the lidocaine injection in the same muscles discussed previously under Diagnostic Studies. Chemodenervation of the scalene muscles in this manner can reduce symptoms for about

3 months on the average. These injections could be repeated every 3 to 6 months, but the subsequent injections can be less effective in some patients. Surgical Technique

Roos37,41,42 described the operative technique for the transaxillary approach in several articles and gave many valuable pointers for success. We briefly describe our current technique, which is modified from that of Roos. The patient is placed on a table (usually covered with an air cushion) with the hips in a straight lateral position and the thorax tilted 60 degrees, with sandbags supporting the back for easy manipulation of the arm. The buttocks are taped in a crisscross fashion for stability, a pillow is placed between the legs, and the legs are strapped to the table. The incision is made transversely in the axilla at the point where the hairline first breaks from the rib cage up to the axilla when the arm and shoulder are elevated properly toward the ceiling. If one gets too high in the axilla, the fat Continued

1136 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method

of

T r e a t m e n t — cont’d

and lymph nodes from axillary fat make dissection impossible. If one gets below the third rib, the hole becomes extremely deep, making dissection difficult. The transverse incision is curved slightly in the shape of a parabola so that it lies in the axillary skin lines, becoming almost imperceptible after 6 months. One of the first structures encountered is the intercostal brachial nerve in the midfield coming from the second intercostal space. Although this may be thought to be a blood vessel, it should not be ligated. The intercostal brachial nerve is protected best by dissecting it free along with a sleeve of adipose tissue. The arm position is extremely important. One assistant will be in charge of distracting the upper arm while the patient’s shoulder is held in 90 degrees of abduction. The patient’s elbow is flexed to 90 degrees and interlocks with the assistant’s elbow while the assistant’s other arm holds onto the patient’s wrist to get effective distraction and exposure in the axilla. The procedure is well described by Roos.37 The surgeon dissects immediately to the chest wall, at which point all of the structures fall away until the first rib becomes visible. Often the superior thoracic artery lies in the field near the first rib. The artery is ligated easily using vascular clamps. We use a Cobb elevator to remove the soft tissues from the anteroinferior surface of the first rib. When the soft tissues are dissected free, the Cobb elevator is directed posteriorly under the first rib to open up the surrounding field. Theoretically, removing the rib subperiosteally invites recurrence. We and others have not found this to be a problem, but resecting it extraperiosteally invites almost certain damage to the pleura. We next take a right-angle clamp and carefully tease all of the structures from the superior surface of the rib, including the scalenus anterior, subclavius, and a portion of the ligament between the first rib and the anterior clavicle (although this is cut more easily with a knife). All of the abnormal muscle structures inserting on the first rib are teased free, with the muscle fibers spread carefully and the pleura protected at all times. The scalenus medius muscle is teased from the first rib. The surgeon now can remove the first rib safely; a special rib cutter and a nerve root retractor (designed by Roos) are indispensable at this point. We have modified the rib cutter

C l The

r i t i c a l

P

o i n t s

main source of compression is between the first rib and the clavicle. l A physical therapist experienced in TOS is important to get good results. l During provocative maneuvers, shutting off of the radial pulse is more important than paresthesias alone. We do not recommend rib resection unless the pulse shuts off. l Rule out other sources of compression neuropathy before considering surgery for TOS. l Resection of the first rib is associated with more complications and longer recovery time than carpal tunnel release or ulnar nerve decompression at the elbow.

to one that is smaller and cuts at a 60-degree angle. The rib cutter is placed as far posterior as possible so that the T1 nerve root is visualized away from the tips of the rib cutter. The posterior part of the rib is cut, and using two Kocher clamps, the rib is pulled gently from the rib cage. Particularly in women, the rib can be avulsed by a gentle pull from the sternocostal junction. If it is impossible to remove the rib from the sternocostal junction, it can be cut anteriorly with the rib cutter. With these maneuvers, one can obtain, in most cases, the entire first rib except for the posterior stump. The stump should be cut and left short so that it lies posterior to the T1 nerve root; the remaining first rib should be less than 2 cm in length. A box rongeur is used to trim the first rib back to the level of the transverse process of the seventh cervical vertebra. If a cervical rib is 2 cm or less, resection usually is not necessary, but the muscles coming from its tip should be removed. At this point, the pleura is checked carefully for holes by putting saline into the wound and overinflating the lungs. If a pneumothorax is found, a chest tube is placed into the hole. The skin is reapproximated with a subcutaneous and subcuticular stitch, and the wound is left undrained because there have been few problems with infection. The chest tube usually is removed the following day. Roos11 suggested that a clinical presentation of upper or lower plexus symptoms is an appropriate criterion to use in selecting the surgical approach for relieving thoracic outlet compression. Because the upper plexus lies beneath the anterior scalene muscles, we recommend an anterior scalenotomy through the superior clavicular approach for upper plexus symptoms as well as a transaxillary resection of the first rib. A first rib resection is recommended for the relief of lower plexus symptoms; scalenectomy and first rib resection are performed easily through the transaxillary approach. If there is any indication that the thoracic outlet is not decompressed thoroughly, one should not hesitate to do a combined approach. Removal of the first rib is a surgical procedure that must be thought out carefully and executed meticulously because it is a procedure often associated with malpractice suits. The TOS operation is not a procedure that lends itself well to teaching, and it is not a procedure easily mastered.43 The procedure requires at least two assistants.

S U G G E S T E D

R E A D I N G S

Adson AW: Cervical ribs: Symptoms, differential diagnosis for section of the insertion of the scalenus anticus muscle. J Int Coll Surg 16:546, 1951. Brickner WM: Brachial plexus pressure by the normal first rib. Ann Surg 85:858872, 1927. Edwards DP, Mulkern E, Barker P: Trans-axillary first rib excision for thoracic outlet syndrome. J R Coll Surg Edinb 44:362-365, 1999. Hagspiel KD, Spinosa DJ, Angle JF, et al: Diagnosis of vascular compression at the thoracic outlet using gadolinium-enhanced high-resolution ultrafast MR ­angiography in abduction and adduction. Cardiovasc Intervent Radiol 23:152154, 2000. Nichols HM: Anatomic structures of the thoracic outlet. Clin Orthop 207:13-20, 1986. Roeder DK, Mills M, McHale JJ, et al: First rib resection in the treatment of thoracic outlet syndrome: Transaxillary and posterior thoracoplasty approaches. Ann Surg 178:49-52, 1973. Roos DB: Transaxillary approach for first rib resection to relieve thoracic outlet syndrome. Ann Surg 163:354-358, 1966.

Shoulder 1137 Roos DB: Experience with first rib resection for thoracic outlet syndrome. Ann Surg 173:429-442, 1971. Roos DB: Congenital anomalies associated with thoracic outlet syndrome: ­Anatomy, symptoms, diagnosis and treatment. Am J Surg 132:771, 1976. Wood VE, Biondi J: Double-crush nerve compression in thoracic outlet syndrome. J Bone Joint Surg Am 72:85-87, 1990. Wood VE, Marchinski LJ: Neurovascular abnormalities associated with congenital anomalies. In Rockwood CA (ed): The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998, pp 142-163.

R E F E R E N C E S Please see www.expertconsult.com

S e c t i o n

O

Vascular Problems of the Shoulder J. Michael Bennett

Vascular injuries of the shoulder are relatively uncommon and are often associated with direct trauma. Most acute sports-related vascular injuries occur with contact sports; however, repetitive motion and congenital malformation can lead to many chronic syndromes that can become painful and debilitating. Symptoms can be misinterpreted and attributed to one of the more common shoulder musculoskeletal abnormalities. Early recognition of vascular compromise is essential to avoid potentially catastrophic outcomes from misdiagnosis.

and pectoral branches. The third section is distal to the lateral border of the pectoralis minor and contributes the largest branch of the axillary artery, the subscapular artery. The subscapular artery further divides into the scapular circumflex and the thoracodorsal artery. The anterior and posterior humeral circumflex arteries are the last two remaining branches from the axillary artery before it becomes the brachial artery. The posterior humeral circumflex descends posteriorly into the ­ quadrilateral space

ANATOMY A thorough understanding of the vascular anatomy of the shoulder is necessary to fully understand the complete spectrum of vascular injury (Fig. 17O-1). Blood flow to the upper extremity begins with the heart. On the right, the subclavian artery branches from the innominate artery. On the left, the subclavian artery arises directly from the arch of the aorta. The subclavian artery then enters the thoracic outlet and extends to the lateral border of the first rib. The thoracic outlet is composed of the upper border of the first rib, inferior border of the clavicle, and anterior and middle scalene muscles. From the lateral border of the first rib to the inferior border of the latissimus dorsi, the subclavian becomes the axillary artery. The artery travels beneath the pectoralis minor and is divided into three sections, with the number of branches from each section corresponding with the number of the section. The first section is above the superior border of the pectoralis minor and gives off the superior thoracic artery inferiorly, which supplies vessels to the first, second, and third intercostal spaces. The second section travels deep to the pectoralis minor consisting of the lateral thoracic and the thoracoacromial arteries, which further arborize into clavicular, acromial, deltoid,

Anterior scalene muscle

Posterior humoral Suprascapular circumflex artery artery Axillary Anterior artery humoral circumflex artery

Subclavian artery

Pectoralis minor muscle

Subscapular artery

Figure 17O-1   Arterial anatomy of the shoulder demonstrates arterial relationships to surrounding musculoskeletal structures and potential sites of compression.

1138 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

with the axillary nerve. The anterior humeral circumflex artery is smaller than the posterior branch and travels laterally around the front of the surgical neck of the humerus, supplying most of the blood supply to the humeral head. The humeral head is perfused by the arcuate artery, which is an anterolateral ascending branch of the anterior circumflex artery that enters the bone in the area of the intertubercular groove and supplies branches to the lesser and greater tuberosities. The subclavian vein begins as a branch from the brachiocephalic vein medially and becomes the axillary vein at the lateral border of the first rib. At the inferior border of the latissimus dorsi, the axillary vein becomes the basilic vein, and it continues distally. The cephalic vein is a superficial vein that pierces the clavipectoral fascia and empties into the axillary vein. The axillary and cephalic veins are responsible for most of the venous drainage in the shoulder. The lymphatics end in the thoracic and right lymphatic ducts.

CLINICAL PRESENTATION The differential diagnosis of shoulder pain and swelling should include vascular injury. Vascular lesions have been described in baseball, volleyball, tennis, cycling, marksmanship, and kayaking athletes.1 Initial symptoms are vague and nonspecific; however, complaints of easy ­fatigability, venous congestion, pallor, coolness of the hand, ­paresthesias, diminished pulses, and cold intolerance should increase suspicion of a vascular lesion. The throwing athlete is at particular risk for developing a ­ vascular injury. The increased stresses across the shoulder can place vascular structures at risk. Other potential causes to consider include penetrating and blunt trauma, thrombosis, and compression by muscle, tendon, fascia, callus, or bone.

Figure 17O-2   Adson’s test indicates subclavian arterial compression between the scalene muscles when there is a diminished radial pulse with arm extension, external rotation, and the patient facing the involved extremity.

PHYSICAL EXAMINATION A thorough history emphasizing timing of symptoms, activities, and precipitating causes is necessary. Physical examination of both upper extremities should include inspection of the hands and fingers, looking for ulcerations, cold intolerance, capillary refill, color differences, and nailbed abnormalities.2 Standard range of motion, strength testing, blood pressure, shoulder swelling, auscultation of the axilla and brachial artery, Allen’s test, and pulse palpation from the wrist to the shoulder should be evaluated. The position of the arm is important and should be tested at the side and in abduction and external rotation to identify any clinically significant differences that would need further neurovascular evaluation. In addition to the extremities, the cervical spine and clavicle must be thoroughly examined. Vascular compression in the cervical region can be evaluated using Adson’s test, the costoclavicular maneuver, and the hyperabduction maneuver. Adson’s test (Fig. 17O-2) indicates subclavian arterial compression between the scalene muscles when there is a diminished radial pulse with arm extension, external rotation, and the patient facing toward the involved extremity. The costoclavicular maneuver (Fig. 17O-3) indicates compression among the structures between the clavicle and the first rib when there is a ­diminished pulse after thrusting the shoulders back in an erect posture. For the hyperabduction maneuver (Fig. 17O-4), the arm is extended, abducted, and externally rotated with the patient facing away from

Figure 17O-3   The costoclavicular maneuver indicates compression among the structures between the clavicle and the first rib when there is a diminished pulse after thrusting the shoulders back in an erect posture.

Shoulder 1139

Figure 17O-5   Arteriogram demonstrating an axillary artery aneurysm after a gunshot injury to the right shoulder. (Courtesy of Drs. Raphael Espada and James B. Bennett.) Figure 17O-4   For the hyperabduction maneuver, the arm is extended, abducted, and externally rotated with the patient facing away from the affected extremity, which compresses the axillary artery as it passes underneath the insertion of pectoralis minor and the coracoid process.

the affected extremity, which theoretically compresses the axillary artery as it passes underneath the insertion of the pectoralis minor and coracoid process.3

IMAGING Once a history and physical examination have been obtained, a systematic diagnostic work-up is indicated. Standard radiographs of the cervical spine and shoulder should be obtained to rule out bony abnormalities associated with vascular compromise such as a cervical rib, a mass occupying bone lesion, fracture, or dislocation. If a vascular lesion is suspected, a Doppler ultrasound of the extremities is the best initial screening test1,4; however, its use should be limited as a dynamic test. Arteriography remains the gold standard for the diagnosis of arterial injuries. Magnetic resonance angiography (MRA) has gained recent popularity for its detailed images of blood vessels and blood flow without having to insert a catheter into the area of interest, minimizing the risk for arterial damage. In all cases, a vascular consultation is recommended to aid in the initial evaluation and work-up.

VASCULAR TRAUMA Vascular injury to the shoulder can occur secondary to blunt or penetrating trauma. Most traumatic vascular injuries to the shoulder occur from penetrating trauma such as bone fragments or a foreign object (Fig. 17O-5). Axillary artery injury has been associated with scapular neck fractures, humeral neck fractures, and clavicle fractures. Vascular status and fracture pattern must be quickly assessed with plain films and a thorough physical examination followed by arteriography or MRA (if indicated). If surgery is indicated, the injury must be addressed within 6 hours to reduce risk for limb ischemia. Although blunt

trauma-induced injury is in the minority, the consequences from delay in diagnosis can be just as devastating. Shoulder mobility can create injury from traction and avulsion of underlying neurovascular structures without creating a bony injury. Low-energy shoulder dislocations or highenergy traction injuries such as scapulothoracic dissociation can lead to similarly poor outcomes if a misdiagnosis or delay in diagnosis is made.

Scapulothoracic Dissociation Scapulothoracic dissociation (SCD) describes a complete disruption of the scapulothoracic joint and its underlying neurovascular structures, which can be associated with acromioclavicular separation, displaced clavicle fracture, or sternoclavicular disruption. Vascular lesions have been reported in 88% of patients, and severe neurologic injuries occur in 94% of patients. One study reported a 10% mortality rate and nearly 100% of deaths were associated with vascular lesions.5 Overall, clinical outcome after SCD is uniformly poor. Outcomes such as flail extremity, early amputation, and death have been reported.6 The scapula, clavicle, acromioclavicular joint, and the surrounding ligamentous, tendinous, and capsular structures create a superior shoulder suspensory complex. Injuries to single components of the complex may be treated nonoperatively because the complex maintains a stable construct. However, if two or more components are compromised, the complex is unstable and requires at least partial repair to restore stability.6

Shoulder Dislocation Anterior shoulder dislocation is a common injury with potential complications associated with the initial dislocation as well as the reduction. Fortunately, vascular injury associated with anterior dislocation is rare. Because of anatomic location and primary restraints, the axillary artery remains at risk with this type of injury. There are a number of mechanisms that have been proposed to describe arterial injury. Some authors have proposed that the artery is fixed by the circumflex scapular artery, which reduces the

1140 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

artery’s mobility on impact and can lead to arterial disruption.7,8 Others have suggested9 that the pectoralis minor acts as a fulcrum against which the artery is angulated, contused, and ruptures as the humeral head displaces the artery anteriorly (Fig. 17O-6). Vascular injuries secondary to shoulder dislocation occur primarily in older patients with stiffer, calcified, more delicate vessels. Other injuries to the axillary artery include axillary artery occlusion, which has been documented with luxatio erecta, and pseudoaneurysm (Figs. 17O-7) and can occur after recurrent anterior dislocations.9 In addition to arterial injuries, venous injuries such as venous thrombosis (Fig. 17O-8) can have a delayed presentation with unilateral extremity swelling and pain. Noninvasive Doppler imaging or venography can be used for diagnosis.

Sternoclavicular Dislocation Posterior sternoclavicular joint dislocation is a rare entity, with only 120 cases documented in the medical literature since it was first described by Sir Astley Cooper in 1824.10 The sternoclavicular joint is the articulation between the medial clavicle and the manubrium of the sternum. The joint is considered to be gliding with an intra-articular

disk, allowing up to 60 degrees of angulation in extremes of shoulder girdle movement.10 Younger patients have a higher rate of dislocation due to increased joint laxity. Stability of the joint is maintained with anterior ligaments and a thicker, stronger posterior ligament. The proximity of the brachiocephalic vein and innominate artery on the right and the common carotid artery and subclavian vein on the left can make posterior dislocation of the ­sternoclavicular joint a potentially life-threatening injury. Compressive or violent force is usually required to cause a dislocation, although a few cases of atraumatic dislocation have been reported.11 Contact sports and motorcycle injuries are the most common causes worldwide.12,13 Symptoms include pain, inability to move the affected shoulder, and a palpable depression on the affected side. Rarely, patients may present with dyspnea or respiratory compromise and immediate reduction is indicated. Other life-threatening complications, which can occur in up to 25% of cases, include tracheal damage, hemopneumothorax, and damage to the larynx with vocal cord palsy.13 Standard radiographic views can be difficult to interpret; therefore, computed tomography (CT) evaluation in stable patients is the ideal method for confirming the suspected diagnosis. CT angiography can delineate related injuries, and intravenous contrast can be used to enhance computed tomographic interpretation of vascular injuries. Treatment of posterior dislocation is immediate reduction. Attempt at closed reduction is made and if unsuccessful or the joint is found to be unstable, open reduction is indicated. If open reduction is indicated, a cardiothoracic surgeon should be present or on standby. Arteriography can be used before and after reduction; however, because of its invasive nature, it is reserved for selected cases.10

VASCULAR INJURY IN THE ATHLETE Subclavian and Axillary Artery Occlusion

Figure 17O-6   During anterior dislocation, the pectoralis minor can act as a fulcrum against which the artery is angulated and contused, and it ruptures as the humeral head displaces the artery anteriorly.

The axillary artery is a continuation of the subclavian artery and is responsible for perfusing the entire upper extremity. There are a number of anatomic compression sites along this pathway that may lead to arterial occlusion. Compression can occur at the subclavian artery as it angulates over a cervical transverse process, cervical rib, or first rib or is compressed by the anterior scalene muscle. Symptoms may include intermittent blanching of the hand and fingers associated with cooler temperatures, fatigue, and exertional pain. Physical examination may reveal a diminished or absent radial pulse, supraclavicular bruits, and a positive Adson’s test. The diagnosis is confirmed with arteriography. Treatment involves a first rib resection. An acute arterial occlusion is an emergency, and immediate surgery is indicated with first rib resection, removal of the thrombus, and embolectomy.14 In 1945, Wright first demonstrated occlusion of the axillary artery from direct compression of the pectoralis minor as the arm is brought into a position of hyperabduction.3,15 Tullos and colleagues further expanded on this description to include a position of abduction, extension, and external rotation, which is consistent with the cocking phase of the throwing cycle. They ­concluded

Shoulder 1141 Figure 17O-7   Diagnosis and treatment of a large pseudoaneurysm in a patient after recurrent shoulder dislocations. (Courtesy of Drs. Raphael Espada, Baylor College of Medicine, Houston, TX, and James B. Bennett, the Fondren Orthopedic group, Houston, TX.)

that repetitive throwing can lead to repeated local trauma to the artery creating intimal damage and the development of subsequent thrombosis.3,16 Symptoms included claudication pain, rapid fatigue, poor control of the pitch, diminished or absent distal pulses, cyanosis, and decreased skin temperatures particularly in the position of hyperabduction and external rotation. Noninvasive Doppler studies can be diagnostic; however, definitive diagnosis is made with arteriography (Fig. 17O-9). Treatment of this condition is usually surgical, and options include thrombectomy,

sympathectomy, segmental excisions, bypass with vascular graft, anastomosis, and angioplasty.16-19

Thrombosis of the axillary venous system was first described by Sir James Paget in 1875 and by Von Schroetter in 1884. This condition has been termed effort-induced thrombosis because of its frequent association with repetitive vigorous activity or blunt trauma with direct or indirect injury to the

Figure 17O-8   Venogram demonstrating venous thrombosis, which can be a complication after dislocation. (Courtesy of Drs. Raphael Espada and James B. Bennett.)

Figure 17O-9   Arteriogram demonstrating axillary artery thrombosis. (Courtesy of Dr. Raphael Espada.)

Effort Thrombosis

1142 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

axillary vein. The basilic vein becomes the axillary vein at the lower border of the teres major muscle, which becomes the subclavian vein at the lateral margin of the first rib. There are several points along its anatomic course where venous compromise may occur. Compression occurs with hyperextension of the neck or hyperabduction of the arms and can occur between the first rib and the clavicle, the subclavian muscles, or the costocoracoid ligament.20,21 Risk factors for the development of a thrombus include a hypercoagulable state, dehydration, oral contraceptives, and vascular injury. The repetitive throwing motion involved with overhead athletes stretches the subclavian vein and can predispose to the development of tears within the intima of the vein. Symptoms occur within 24 hours of the inciting trauma or activity and consist of activity-related fatigue, dull and aching extremity pain, numbness, and swelling. Often, patients describe a “heaviness” of the upper arm and shoulder following activities. Physical examination may reveal superficial venous dilation, extremity cyanosis, swelling, and a painful axilla or deltopectoral groove. Pulses and neurologic examination can be normal. The physical findings may become more prominent with exercising testing. Diagnosis is made with history and physical examination and can be confirmed using venography. The venogram will show complete occlusion of the axillary or subclavian vein with extensive collateral venous return. Once the diagnosis is made, the first line of treatment for effort-induced thrombosis is conservative, emphasizing rest, heat, and elevation of the involved extremity. Pain and swelling will resolve within 3 to 4 days; however, many of these patients continue to suffer symptoms.17,22 In the acute phase, heparin, followed by warfarin, is often used to inhibit progression of the thrombus. Recently, thrombolytic agents such as streptokinase have been found to be effective in the lysis of acute clots less than 2 weeks old; however, these agents are ineffective with chronic clots. Early thrombectomy with simultaneous decompression of the thoracic outlet or first rib resection following fibrinolytic therapy has been associated with good long-term results.17,23

of the extremity for longer than 1 minute. Dampening of the radial pulse may also occur with the arm in this cocked position. In chronic cases, there may be atrophy of the deltoid. Neurologic examination and electromyographic studies are usually normal. Diagnosis may be confirmed using bilateral dynamic subclavian arteriograms. Cahill and Palmer2,24 found that with abduction and external rotation, the PHCA remained patent in the asymptomatic shoulder and became obstructed in the symptomatic shoulder. Mochizuki and associates25 found that asymptomatic volunteers demonstrated angiographic occlusion of the PHCA while in the cocked position. Angiography is nonspecific and should serve only as a supplement to a clinical diagnosis. The standard treatment is nonoperative for patients with quadrilateral space syndrome. Rest, modification of activities, and the initiation of a formal therapy program is the first line of treatment. Therapy should emphasize stretching the posterior capsule and teres minor. If conservative measures fail, surgical decompression of the quadrilateral space through a posterior approach is indicated. The surrounding muscles, tendons, and fibrous bands that constrain the space are released, removing all pressure on the neurovascular bundle when the shoulder is brought into abduction and external rotation. Cahill and Palmer reported 16 of 18 patients with good or excellent results and 2 of 18 with no change in symptoms after decompression.24 Further interpretation of the results associated with decompression is difficult owing to the small number of cases and the short follow-up.

Thoracic Outlet Syndrome Thoracic outlet syndrome involves compression of the neurovascular structures supplying the upper limb as they course from the neck to the axilla. The boundaries of compression include the clavicle, the scapula, and the first thoracic rib or cervical rib. Etiology, symptoms, and treatment are further discussed in Chapter 17N.

Quadrilateral Space Syndrome The quadrilateral space is defined as the area enclosed by the teres minor superiorly, the humeral shaft laterally, the teres major inferiorly, and the long head of the triceps medially. Within this space traverses the axillary nerve and posterior humeral circumflex artery (PHCA). In 1983, Cahill and Palmer first described the “quadrilateral space syndrome,” which involves compression of the PHCA or the axillary nerve within the quadrilateral space.1,24 Compression can occur from fibrous bands within this space, creating tension across the neurovascular bundle with abduction and external rotation of the affected extremity. Typical patients are between 25 and 50 years of age. Symptoms are unilateral and often involve the dominant extremity. The symptoms described are nonspecific and include pain and paresthesias not associated with a traumatic event. These symptoms progressively worsen with abduction and external rotation. Clinical findings include tenderness to palpation over the quadrilateral space in addition to reproduction of symptoms with abduction and external rotation

c

r i t i c a l

P

o i n t s

l Early recognition of vascular compromise is essential to avoid potentially catastrophic outcomes from ­misdiagnosis. l Initial symptoms are vague and nonspecific; however, complaints of easy fatigability, venous congestion, pallor, coolness of the hand, paresthesias, diminished pulses, and cold intolerance should increase suspicion of a vascular lesion. l The position of the arm is important and should be tested at the side and in abduction and external rotation to identify any clinically significant differences that would need further neurovasuclar evalution. l Standard radiographs of the cervical spine and shoulder should be obtained to rule out bony abnormalities associated with vascular compromise such as a cervical rib, a mass occupying bone lesion, fracture, or dislocation.

Shoulder 1143

l Arteriography remains the gold standard for the diagnosis of arterial injuries. l Vascular injuries secondary to shoulder dislocation occur primarily in older patients with stiffer, calcified, more delicate vessesls. Other injuries to the axillary artery include axillary artery occlusion, which has been documented with luxatio erecta, and pseudoaneurysm and can occur after recurrent anterior dislocations. l The proximity of the braciocephalic vein and innominate artery on the right and the common carotid artery and subclavian vein on the left can make posterior dislocation a potentially life-threatening injury. l Risk factors for the development of a thrombus include a hypercoagulable state, dehydration, oral contraceptives, and vascular injury. l The repetitive throwing motion involved with overhead athletes stretches the subclavian vein and can predispose to the development of tears within the intima of the vein.

S U G G E S T E D

R E A D I N G S

Baker CL, Liu SH: Neurovascular injuries to the shoulder. J Orthop Sports Phys Ther 18(1):360-364, 1993. Baker CL, Liu SH, Blackburn TA: Neurovascular compression syndromes of the shoulder. In Andrews JR (ed): The Athlete’s Shoulder. New York, Churchill Livingstone, 1994, pp 261-273. Baker CL, Thornberry R: Nerovascular syndromes. Injuries to the Throwing Arm. Philadelphia, WB Saunders, 1985, pp 176-188. Beeson MS: Complications of shoulder dislocation. Am J Emerg Med 17(3):288299, 1999. Durgas JR, Weiland AJ: Vascular pathology in the throwing athlete. Update on Management of sports Injuries 16(3):477-485, 2000. Marone PJ: vascular lesions about the shoulder girdle: Shoulder injuries in sports. Wellesley Hills, MA, Aspen Publishers, 1992, pp 129-132. Mirza AH, Alam K, Ali A: Posterior sternocalvicular dislocation in a rugby player as a cause of silent vascular compromise: A case report. Br J Sports Med 39(5):e28, 2005. Nuber GW, McCarthy WJ, Yao JS, et al: Arterial abnormalities of the shoulder in athletes. Am J Sports Med 18(5):514-519, 1990. Ryu RK, Dunbar WH, Kuhn JE, et al: Comprehensive evaluation and treatment of the shoulder in throwing athlete. Arthroscopy 18(9):70-89, 2002. Schulte KR, Warner JP: Uncommon causes of shoulder pain in the athlete. Orthop Clin N Am 26(3):505-528, 1995. Tullos HS, Erwin WD, Woods GW, et al: Unusual lesions of the pitching arm. Clin Orthop 88:169-182, 1972.

R E F E R E N C E S Please see www.expertconsult.com

S e c t i o n

P

Parsonage-Turner Syndrome Adam Nelson Whatley

Parsonage-Turner syndrome, also known as brachial ­neuritis or neuralgic amyotrophy, is a condition of unknown etiology. It affects the brachial plexus and causes pain followed by weakness of the shoulder and upper extremity. It has been described numerous times in the literature since it was first reported by Parsonage and Turner in 1948.1 The classic presentation begins with an acute onset of sharp pain in the shoulder region. As the pain subsides, weakness arises in the shoulder musculature. Diagnosis of this condition is primarily clinical in nature and is exceedingly difficult in the acute stage. There is a 3:2 male-to-female ratio in the idiopathic form of the disease, although a hereditary form of the ­disease has been reported widely in the literature.2 Age of onset is usually in the second or third decade, but cases have been reported ranging in age from neonates to patients in their eighth decade. The exact cause is unknown, but the current hypothesis is one of an immune-mediated response to a patient’s own peripheral nerves.2 This theory is supported by the fact that about half of attacks are ­preceded

by some event that can trigger the immune system, including infection, surgery, pregnancy and puerperium, mental and strenuous physical stress, immunizations, and immunomodulating treatment regimens (interleukin-2 or interferon-α2).2

RELEVANT ANATOMY AND BIOMECHANICS The brachial plexus comprises the ventral rami of spinal nerve roots from C5 to T1. These rami, or roots, form subsequent trunks, divisions, cords, and terminal branches that innervate the shoulder and upper extremity. The brachial plexus resides in the upper shoulder region between the anterior and middle scalene muscles. It encircles the subclavian artery and enters the upper arm through the axilla. The brachial plexus is the site of many traumatic and atraumatic conditions, which must be considered in the differential diagnosis of Parsonage-Turner syndrome.

1144 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

CLASSIFICATION There is no classification system for Parsonage-Turner syndrome at the time of the writing of this chapter.

EVALUATION Clinical Presentation and History Patients who present with Parsonage-Turner syndrome most commonly (96%) complain of severe pain in or around the shoulder girdle and upper arm.2 There are few if any disorders in the shoulder and arm region that are so extremely and persistently painful at onset. The pain usually (61%) wakes the patient from sleep in the middle of the night or early in the morning.2 Ninety-four percent of patients report an inability to return to sleep.2 The pain is most often described as burning, aching, or throbbing and quantified as severe in nature. Often the pain radiates down the arm and occasionally extends below the elbow. This distal radiation is characteristic of patients with diffuse lesions or lesions that involve the lower brachial plexus.3 Most patients present with right-sided symptoms, and two thirds of patients demonstrate scapular winging.2 These painful symptoms usually maintain their intensity for a period ranging from a few hours to 2 to 3 weeks.1 Pain is exacerbated in the acute stage by any movement around the shoulder. In contrast, Valsalva or neck motion does not affect the symptoms.1 Waxman described the flexionadduction sign as patients keeping the shoulder adducted and the elbow flexed to decrease discomfort.4 These painful symptoms precede generalized weakness, although in some cases pain and weakness coincide. Eightyfive percent of patients report some degree of subjective weakness during the first month of symptoms, and 70% report weakness during the first 2 weeks.5 This weakness involves lower motor neurons in the brachial plexus and is represented by flaccidity followed by generalized wasting of the affected musculature. The pattern of involvement most often follows an upper brachial plexus distribution (71%). Half of these upper brachial plexus cases demonstrate long thoracic involvement, and 21% do not.2 Females present with middle or lower plexus distribution in 23% of cases compared with 11% of males.2 Fifteen percent of cases demonstrate signs of distal autonomic nervous system dysfunction, such as hand edema and vasomotor instability.2 Seventeen percent of these idiopathic cases involve nerves outside of the brachial plexus, including the lumbosacral plexus, phrenic nerve, or recurrent laryngeal nerve. Very infrequently, the facial nerve or abdominal nerves are involved. These atypical cases are seen more frequently (56%) in the hereditary form.2 Often patients present demonstrating the “coat pocket sign,” which entails suspending the affected extremity by lodging the hand in the ipsilateral coat pocket to reduce painful symptoms. The differential diagnosis of Parsonage-Turner syndrome should include the following: rotator cuff pathology, subacromial impingement, adhesive capsulitis, calcific tendinosis, diskogenic cervical spine disorders, ­poliomyelitis, amyotrophic lateral sclerosis, herpes zoster, tumors of the

spinal cord and brachial plexus, traumatic compressive nerve injuries, and ganglion cysts.6 Accurate diagnosis is difficult to achieve if the disease is in the acute phase during presentation; however, it can prevent unwarranted testing or surgery that may be indicated by a misdiagnosis and can guide earlier appropriate therapies.7 It can mimic other conditions, but there is a characteristic rapid and spontaneous resolution of pain in Parsonage-Turner syndrome. It remains a diagnosis of exclusion in some patients.

PHYSICAL EXAMINATION AND TESTING Examination The characteristic weakness in Parsonage-Turner syndrome may present in one of several patterns: muscles innervated by one peripheral nerve, muscles innervated by multiple peripheral nerves, muscles innervated by one or more nerve trunks, or muscles innervated by a combination of peripheral nerves and trunks.8 The most commonly affected peripheral nerve is the axillary nerve, followed by the suprascapular, long thoracic, and the musculocutaneous nerve.9 However, cases involving the anterior interosseous, radial, and median nerves have been reported in the literature.5,10-12 Classically, the deltoid demonstrates some degree of weakness, but the supraspinatus, infraspinatus, serratus anterior, biceps brachii, triceps, and extensors of the wrist and fingers have also been known to be affected. Diaphragmatic paralysis can be caused by phrenic nerve involvement and presents with shortness of breath and tachypnea. Atrophic changes of the affected muscles usually occur to some degree during the course of the disease. Interestingly, one third of cases demonstrate other areas of subclinical involvement.2 It is imperative to pay close attention to other upper extremity musculature and cutaneous sensation that the patient does not directly complain about in order to avoid diagnostic errors.

Testing There is only a limited role in laboratory testing in the diagnosis of Parsonage-Turner syndrome because most standard laboratory values (complete blood count, erythrocyte sedimentation rate, electrolytes, liver function tests, and urinalysis) are within normal limits.12 Cerebrospinal fluid, although reported in few affected patients to be abnormally elevated in protein, is not an efficient marker because of the potential comorbidity of lumbar puncture.5,8 Immunoassays are largely noncontributory because immunoglobulin M and immunoglobulin G have been reported to be elevated in only one patient.13 Electromyography (EMG) has been used effectively in the diagnosis of Parsonage-Turner syndrome. Its uses range from aiding in accurate initial diagnosis to the prediction of eventual recovery. EMG is helpful in localizing the area affected by the disease process and confirming the presumptive diagnosis. It is also useful in differentiating between Parsonage-Turner syndrome and traumatic etiology. Although the reading is at times variable, an acute denervation indicating axonal neuropathy is often ­demonstrated.14 Nerve ­conduction

Shoulder 1145

velocity is often within normal limits but is sometimes delayed. Testing 3 to 4 weeks after onset often demonstrates fibrillation potentials, positive waves, delayed distal latencies, and decreased amplitude of action potentials.3,6,9,15-17 Interestingly, EMG in the face of Parsonage-Turner syndrome spares the paraspinal musculature. EMG is also useful in detection of subclinical pathology in the contralateral (otherwise silent) upper extremity.5 EMG demonstrates reinnervation and recovery of muscular function, which is important in helping to predict functional outcome.

Imaging Plain radiographic imaging of the cervical spine and shoulder usually is normal and noncontributory. One exception is the initial presentation of inferior glenohumeral subluxation occurring up to 2.5 weeks after onset in cases with deltoid and rotator cuff involvement.18 This finding usually resolves on overall symptomatic relief. Plain radiographs of the chest demonstrate hemidiaphragmatic elevation in cases involving one or both phrenic nerves.5,19 Standard magnetic resonance imaging (MRI) usually demonstrates only the secondary changes in affected muscles, including early high intensity in T2-weighted images in the subacute stage (1 to 3 months) and diffuse muscular atrophy and fatty replacement (mean of 5 weeks after onset) with T1-weighted images in the months following onset of symptoms.20 Only infrequently have pathologic changes been demonstrated in neural elements in the face of Parsonage-Turner syndrome with standard MRI (3 of 50 cases).21 Magnetic resonance neurography is more sensitive than standard MRI in peripheral neural element pathology22,23 and has been shown in studies to be sensitive in the acute stage (within 1 month of onset) in the diagnosis of Parsonage-Turner syndrome.20 It has been shown to confirm the diagnosis in the case of one 27-year-old man demonstrating thickened, hyperintense upper trunk changes consistent with plexitis.24 This methodology holds some promise in the advancement of a radiologic basis of diagnosis in patients with Parsonage-Turner syndrome.

TREATMENT OPTIONS Nonoperative The overall nature of Parsonage-Turner syndrome is one of a benign and self-limited disease. Treatment is largely supportive, and no one treatment regimen has been shown to alter the course of the disease process.12 Corticosteroids have been shown to reduce pain in a few patients during the very early stages.5 Analgesics are somewhat effective in the treatment of debilitating pain. Rest of the affected extremity is often advocated because movement often exacerbates pain,25 and immobilization has been used effectively to limit pain from stretching affected muscles. One method of immobilization is to employ a standard shoulder sling to ensure maintenance of full normal range of motion with dedicated active and passive range of motion exercises to the shoulder and elbow a few times per day. Physical therapy is widely recommended but has not been shown

to have any effect in reducing time to recovery compared with no formal therapy.5 Massage and electrical stimulation have been cited as beneficial, but there are no subjective data to support this claim.12 The neurology literature has reported good response in decreasing pain with a multimodal pharmaceutical approach. One such regimen includes the combination of a long-acting nonsteroidal anti-inflammatory drug and an opiate (sustained-release diclofenac, 100 mg, with sustained-release morphine, 10 to 30 mg twice daily).21 Co-analgesics such as gabapentin, carbamazepine, and amitriptyline are also recommended for control of second-phase pain, which is characterized by spontaneous or movement-induced shooting pains and tingling sensations due to aberrant impulse firing in damaged, hypersensitive portions of the brachial plexus.2 These medications are used in delayed-onset cases and are not indicated for the treatment of acute pain.

Operative Operative treatment of Parsonage-Turner syndrome is indicated only in cases in which there is no long-term improvement. Such cases involve muscle transfers or fusions to compensate for permanent muscular deficit. Reported procedures include scapulothoracic stabilization for a case with persistent serratus anterior and rhomboid deficit, and tendon transfers to thumb, fingers, and wrist for a case involving persistent radial nerve deficit.26

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Overall recovery is usually good, with complete restoration of strength and function. A longer duration of active painful symptoms and longer time to full recovery are ­typical in cases with multiple nerve involvement or bilateral involvement.1,5,8,9 Isolated upper trunk pathology, which is the most common distribution, usually resolves more quickly and without complications.5 In all cases of Parsonage-Turner syndrome, improvement in both strength and sensation occur as early as 1 month after onset but can take up to 3 years if recovery is reached at all. Tsairis and colleagues reported complete recovery in 36% by 1 year, 75% within 2 years, and 89% within 3 years.5 Sensory and motor recovery parallel each other. Recurrence of symptoms is seen occasionally (<25%) in patients with the idiopathic form of the disease.2 In these patients, the symptoms are less severe and shorter in duration (<2 weeks). The recurrence usually is seen within a few months to several years after full recovery is achieved.1,3,5,8,12,15 In contrast, patients in the families affected by the hereditary form of the disease have much more frequent and numerous reoccurrences (75%).21

CRITERIA FOR RETURN TO PLAY Patients with Parsonage-Turner syndrome should be considered as special cases and require much guidance and attention. It is recommended that a plateau in recovery

1146 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

of strength must be achieved to advocate return to play.14 Full strength may not be recovered, and in these few cases, limited participation in most sports and significant restriction or disallowance with regard to contact sports is advocated. These cases must be treated on an individual basis with regard to abilities and the particular sporting event in question.

SPECIAL POPULATIONS As mentioned earlier, Parsonage-Turner syndrome exists in both an idiopathic and hereditary form. The hereditary form is passed through an autosomal dominant pattern and demonstrates a high penetrance estimated at 80%.2 It is extremely rare, with only 200 families reported worldwide. In many of the affected families, the genetic locus for the hereditary form of the disease is found on chromosome 17q25. However, this locus is not consistently confirmed in other affected families, suggesting a heterogeneous nature of the hereditary form of the disease. C

r i t i c a l

P

o i n t s

l 

 arsonage-Turner syndrome is characterized by sudden, P sharp pain in the shoulder and upper arm followed by weakness in the affected shoulder musculature. l Parsonage-Turner syndrome is self-limited, and usually a full recovery occurs within a few months to 3 years. l The exact etiology of Parsonage-Turner syndrome is unknown but is thought to be related to an immune response to one of various stressors. l Accurate diagnosis of Parsonage-Turner syndrome is crucial to avoid potential surgical treatments and complications based on misdiagnosis. l Treatment of Parsonage-Turner syndrome is largely supportive, with an emphasis on pain control and maintenance of range of motion of the affected upper extremity.

l Diagnosis of Parsonage-Turner syndrome is primarily clinical. Electromyography is used to confirm the ­diagnosis. l Magnetic resonance neurography has been reported to show promise in the diagnosis of Parsonage-Turner syndrome in its acute stage. l Athletes with Parsonage-Turner syndrome should be allowed to return to play based on their particular ­sporting event (noncontact or contact) and achievement of a plateau if not full-strength recovery. Each case should be judged on an individual basis.

S U G G E S T E D

R E A D I N G S

Aymond JK, Goldner JL, Hardaker WT: Neuralgic amyotrophy. Orthop Rev 18:1275-1279, 1989. Favero K, Hawkins R, Jones M: Neuralgic amyotrophy. J Bone Joint Surg Am 69:195-198, 1987. Fibuch EE, Mertz J, Geller B: Postoperative onset of idiopathic brachial neuritis. Anesthesiology 84:455-458, 1996. Maravilla KR, Aagard BD, Kliot M: MR Neurography: MR imaging of peripheral nerves. Magn Reson Imaging Clin North Am 6:179-194, 1998. Misamore GW, Lehman DE: Parsonage-Turner syndrome (acute brachial neuritis). J Bone Joint Surg Am 78:1405-1408, 1996. Parsonage MJ, Turner JWA: The shoulder-girdle syndrome. Lancet 1:973-978, 1948. Tsairis P, Dyck PJ, Mulder DW: Natural history of brachial plexus neuropathy: Report on 99 patients. Arch Neurol 27:109-117, 1972. van Alfen N: The neuralgic amyotrophy consultation. J Neurol 254:695-704, 2007. van Alfen N: Engelen BGMV: The clinical spectrum of neuralgic amyotrophy in 246 cases. Brain 129:438-450, 2006. Walsh NE, Dumitru D, Kalantri A, et al: Brachial neuritis involving the bilateral phrenic nerves. Arch Phys Med Rehabil 68:46-48, 1987.

R E F E R E N C E S Please see www.expertconsult.com

S e c t i o n

Q

Development of Skills for Shoulder Surgery Hussein A. Elkousy and T. Bradley Edwards

The ability to perform successful shoulder surgery requires more than what can be obtained in reading a textbook. Motor skills must be developed and perfected to become adept at performing many of the procedures described in this chapter. For several years now, we have been involved in helping surgeons develop the skills necessary to become a successful shoulder surgeon mainly through our involvement in technique-intensive courses offered at the Joe W. King Orthopedic Institute in Houston, Texas.* This

chapter will relay our experience in teaching the technical aspects of shoulder surgery, both arthroscopic and open. This should provide the reader with a starting point in the development of the technical skills necessary to implement much of the knowledge provided in this section. *For information on courses offered by the Joe W. King Orthopedic ­Institute, contact Kayla Berndt by telephone at 713-794-3517 or by e-mail at [email protected].

Shoulder 1147

SHOULDER ARTHROSCOPY Shoulder arthroscopy has several applications in the management of shoulder pathology, and the indications continue to expand. Pathology amenable to management with shoulder arthroscopy has traditionally included supraspinatus and infraspinatus repair, labral repair, subacromial decompression, biceps tenotomy and tenodesis, distal clavicle excision, contracture release, and débridement of arthrosis. More recently, the indications have expanded to include subscapularis repair, biologic resurfacing, bone block transfer procedures, and suprascapular nerve ­decompression. Shoulder arthroscopy is a technically demanding procedure and is not an easily acquired skill. It is not something learned in a lecture hall. It requires hands-on learning and repetition to acquire motor skills, understand spatial relationships, and develop organizational skills that result in a successful procedure. Unfortunately, in the present-day litigious society, it is sometimes difficult to teach such a difficult skill without compromising patient outcomes in a clinical setting. Furthermore, most patients expect their surgeon to be an expert before performing a procedure. As a result, systems and models have been developed to facilitate such learning and acquisition of skills.

DEFINITIONS AND PRINCIPLES There are several principles critical to learning shoulder arthroscopy. These include developing and understanding spatial relationships, developing knot-tying skills, and being able to organize surgical steps both mentally and physically. The concept of mastering spatial relationships refers to the ability to move one’s hands in space to manipulate an instrument attached to the hand, being able to triangulate, establishing portals, and becoming oriented with shoulder structures. These concepts are self-explanatory except for triangulation. Triangulation refers to the ability to use an instrument in each hand and have the instruments meet at a point in space. This is useful, for example, in locating an instrument with a camera in an area of poor visualization. Knot tying is an essential skill for the shoulder arthroscopist. Certainly, several systems and implants have been developed that are knotless and obviate the need to tie a knot. However, these systems can be expensive and are not always available or applicable. The ability to tie an arthroscopic knot makes the arthroscopist more adaptable to manage shoulder pathology. However, becoming ­skillful at knot tying requires practice and a good understanding of the basic anatomy and function of a knot. An arthroscopic knot has two parts that confer different essential properties. The first is the pattern of the initial throw, which gives the knot its name. For example, different initial throw patterns are the Roeder knot, the Weston knot, the Tennessee slider, the SMC knot, the Duncan loop, and Nicky’s knot.1-5 These knot types can be further subdivided into sliding and nonsliding knots and locking and nonlocking knots (Fig. 17Q-1). These terms refer, respectively, to the knot’s property of sliding down a post to the target tissue and whether the initial throw can maintain its internal integrity without losing tissue tension

before additional knots are placed to back it up. This initial throw pattern serves to pull the two target tissues or structures together. The tightness of this initial throw pattern confers a property of knots known as loop security.6-11 This initial loop security defines how well apposed the target tissues are to each other. The second part of a knot is the additional half hitches placed on a knot that prevent the knot from losing its integrity in the face of external loads. This property is referred to as knot security.6-11 It has been demonstrated that knot security is best achieved by placing three half hitches with alternating directions of throw on alternating posts to the initial knot.4,6

METHODS OF LEARNING Shoulder arthroscopy may be taught on a model, a cadaver, or a live patient. Several models have been developed. These include simple tying blocks to help with knot-tying techniques first in the open and then in the arthroscopic setting. The spatial relationships and motor skills necessary for arthroscopic surgery can then be learned on shoulder models. These models can be uncovered, covered with a clear cover, or covered with an opaque cover with the use of an arthroscopic camera (Fig. 17Q-2). These models allow the novice shoulder arthroscopist to develop motor skills by simulating the clinical scenario and by allowing repetitions to hone skills. The stepwise progression from the uncovered to the opaque models allows for a logical and deliberate progression of skills and comfort. Models, however, cannot substitute for reality. They provide an excellent introduction to shoulder arthroscopy and allow for development of movement in space, triangulation skills, and organization of surgical steps and suture management. They do not allow the novice shoulder arthroscopist to establish portals or become familiar with anatomy and structures of the shoulder. Cadavers and live patients are necessary to master these aspects of shoulder arthroscopy. Cadavers provide an excellent opportunity to simulate a live patient. However, they are costly, and not everyone has access to them. Additionally, each cadaver has limited use, and the anatomy may be somewhat distorted depending on the condition of the cadaver. Live patients are certainly the gold standard to master shoulder arthroscopy, but as mentioned previously, the medicolegal consequences of “learning” on patients has become a difficult obstacle to ignore. This type of education can be implemented in a stepwise fashion. Most shoulder procedures are facilitated by a first assistant. The novice shoulder arthroscopist may act as a first assistant with little morbidity to the patient and still hone the motor skills and develop the organizational skills necessary to succeed. Additionally, it allows them to become more familiar with the variable anatomy not afforded with learning from a model. The novice shoulder arthroscopist can first learn how to place a blunt instrument in the joint and try to identify structures. Soon thereafter, they can learn the importance of portal placement and develop both insideout and outside-in techniques for establishing such portals. They can hold the camera while the primary surgeon ties suture or passes instruments.

1148 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E

F

G

H

I Figure 17Q-1   The Weston knot is an example of a sliding locking knot. A, The short post strand is in the left hand, and the longer nonpost strand is in the right hand. B-G, Successive steps in tying the knot. H, The final knot is slid down to the target tissue. I, After the knot is pulled down to the target tissue, tension is placed on the nonpost strand to “flip” the knot and lock it.

Shoulder 1149

A

B

C

Figure 17Q-2   Arthroscopic models. A, Arthroscopic shoulder model without a cover. B, Model with a clear cover.   C, Model with an opaque cover.

Once the skills of the assistant are mastered, the novice shoulder arthroscopist can gradually be introduced to the primary surgeon position. This entails holding the camera while performing a motor task with the other hand. Initially this can be a survey of the shoulder structures. It can then progress to more advanced tasks like passing suture. Once knot-tying skills are fully mastered on a model, this can be developed in vivo. Occasionally, an extra security anchor or suture can be passed that will enhance the repair but is not critical to the repair; this can be used to avoid compromising the clinical result if the novice surgeon is not successful.

ORGANIZATION Successful shoulder arthroscopy does not result simply from mastery of the skills previously mentioned. These motor skills are of relatively little utility if they cannot be applied in a logical and organized manner. This organization is predicated on assembling a series of steps based on the pathology and subsequently making a plan to manage portals, choose instruments, and manage suture. Certainly, there are several ways to approach any given problem, but it is useful to learn one approach first before venturing off in other directions. The next section demonstrates one way of approaching a rotator cuff tear and a labral tear on a model to illustrate the thought required to complete the procedure efficiently.

ROTATOR CUFF REPAIR The models that we use have several portals to allow for several approaches to learn a given procedure. A rotator cuff model has been developed that allows for placement of anchors in the tuberosity of a proximal humerus and ­ passage of suture through a mesh, which simulates the rotator cuff. The models are all left shoulders, but the ­procedure may be reversed to practice for a right shoulder. Most right-handed surgeons have more difficulty working on a left shoulder because the nondominant hand must perform most of the motor tasks. Repair of a simple crescent-shaped tear is demonstrated first. The camera is placed through a posterior portal. A lateral portal is used to place anchors and pass suture through the rotator cuff (Fig. 17Q-3A). An anterior portal is used to organize and hold suture from the anchor and suture that have been passed (see Fig. 17Q-3B). If a single-row repair is to be performed, anchors are placed from anterior to posterior about 1 cm lateral to the horizontal edge of the greater tuberosity. Each anchor contains two sutures, which are passed in simple fashion from anterior to posterior using a commercially available suture-passing device with a springloaded needle (see Fig. 17Q-3C to G). The sutures are then tied down from posterior to anterior (see Fig. 17Q-3H). If a double-row repair is to be done, an accessory portal is used more medially to allow for a vertical application of these anchors. The most anterior anchor is placed first at

1150 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E

F

G

H

Figure 17Q-3   Sample steps for a rotator cuff repair on a model. A, A suture anchor is placed into the lateral aspect of the greater tuberosity through a lateral portal. B, All four limbs of two sutures are passed out the anterior portal for suture management.  C, A spring-loaded suture-passing device is used to pass one limb of a suture through the anterior aspect of the rotator cuff tear. D, The passed suture is grasped and passed out the anterior portal. E, One limb of the second suture is passed through the more posterior aspect of the rotator cuff tear. F, One limb of each suture has been passed, and all four limbs are in the anterior portal. G, Both limbs of the posterior suture are passed out the lateral portal before being tied through the lateral portal.   H, Both sutures have been tied, and the completed repair is viewed through the lateral portal.

Shoulder 1151

the anterior margin of the tear just lateral to the articular surface. The remainder of the medial-row anchors (usually numbering between one and three) are placed from anterior to posterior. Only one suture from each anchor is kept if it is a double- or triple-loaded anchor. The sutures are then passed in a horizontal mattress fashion from anterior to posterior. The spatial placement of these suture limbs is an acquired skill based on the excursion of the rotator cuff and the pattern of the tear. After all sutures are passed, they are tied down from anterior to posterior. The sutures are not cut once they are tied down. These sutures are passed through the eyelet of a knotless anchor that will be placed laterally in the tuberosity to complete the lateral row. These anchors are placed from posterior to anterior. The number of anchors used laterally and the number of sutures in each anchor is dictated by the size and pattern of the tear. These procedures may seem straightforward, but several organizational principles must be followed. The anchors and sutures are placed in a deliberate order to avoid tangling. Cannulas are always used to pass suture and are often used to hold suture that has been passed. Hemostat clamps are always placed on sutures not being used to avoid inadvertent unloading of the tissue or of an anchor. Sharp instruments are never placed through a cannula with an existing suture. Finally, the pattern of the anchors to be used is defined before placing anchors. It can be modified, but a plan is made initially to allow for opening of implants and efficient use of time.

ANTERIOR LABRAL REPAIR The camera is placed in a posterior portal, and two anterior portals are used. One anterior portal is superolateral (anterosuperior), and the second one is inferomedial (anteroinferior). Anchors are placed through the anterosuperior portal by first predrilling the site of the anchor (Fig. 17Q-4A). Of significant importance, the anchor is placed on the glenoid face, not along the anterior neck. Once the site is predrilled, the anchor is placed. A cannulated tissue penetrator is placed through the anteroinferior portal and passed through the appropriate anterior capsule and labrum (see Fig. 17Q-4B). The two free ends of a 2-0 nylon passing suture are fed through this penetrator and taken out the anterosuperior portal (see Fig. 17Q-4C). This passing suture is reversed with a second suture by passing the looped end of the second suture so that it exits the anterosuperior portal (see Fig. 17Q-4D). This looped end will be used to pass one limb of an anchor suture through the anterior capsule and labrum (see Fig. 17Q-4E). Both ends of this suture are then passed out the anteroinferior cannula and then tied down (see Fig. 17Q4F). Subsequent anchors are placed in similar fashion from inferior to superior. Generally, up to three anchors can be used for a complete repair. Suture management is essential for this procedure. The same principles apply as for a rotator cuff repair. A sharp instrument is never passed through a cannula with suture, suture is always tied through a cannula, and hemostat clamps are always placed on suture limb ends. Also, we use a suture relay system with an extra step of passing a reversal suture, which can be cumbersome, but it allows use of a small-diameter tissue penetrator to protect the target

t­ issue. Care should always be taken when shuttling suture from one cannula to another so that the suture does not move through the anchor, which will result in unloading of the anchor.

OPEN SHOULDER SURGERY We teach open shoulder surgery using similar techniques to those proven effective in teaching arthroscopic shoulder surgery. Specifically, in teaching shoulder arthroplasty, we believe in repetition to aid in the acquisition of the skills necessary to perform the procedure. Ideally, fresh fullbody cadavers are used to teach shoulder arthroplasty. This allows practice of handling the soft tissue portions of the procedure not afforded by practice on synthetic bone models void of soft tissues. Unfortunately, the cost-prohibitive nature of using multiple full-body cadavers and limited usefulness of cadaveric forequarters (hard to stabilize in a clamp) provides a large obstacle to this teaching method. We were able to address these problems through the development of an arthroplasty model complete with synthetic soft tissues (Fig. 17Q-5). This model has proved successful at allowing surgeons to practice the multiple steps of shoulder arthroplasty in a cost-effective manner. At the Joe W. King Orthopedic Institute Advanced Shoulder Arthroplasty Course, participants are advanced to surgery on a full-body cadaver only after they have mastered all steps using the arthroplasty model.

SIMULATED SHOULDER ARTHROPLASTY USING A MODEL The key portions of shoulder arthroplasty are practiced on our shoulder arthroplasty model. This starts with retraction of the conjoined tendon medially to expose the subscapularis, axillary nerve, and anterior humeral circumflex vessels (Fig. 17Q-6A). The anterior humeral circumflex vessels are suture-ligated, and stay sutures are placed in the subscapularis tendon (see Fig. 17Q-6B). The rotator ­ interval is opened, and a subscapularis tenotomy is performed along the anatomic neck of the humerus (see Fig. 17Q-6C). A humeral head retractor is placed, and the capsuloligamentous structures are released from the subscapularis, including the superior, middle, and inferior glenohumeral ligaments, while noting the relationship of the inferior glenohumeral ligament to the axillary nerve (see Fig. 17Q-6D). A glenoid-based release of the inferior capsule is performed to obtain glenoid visualization (see Fig. 17Q-6E). The humeral head is dislocated, and an osteotomy is performed along the anatomic neck of the humerus, taking care to protect the posterosuperior rotator cuff (see Fig. 17Q-6F). The proximal humerus is prepared by reaming the diaphysis and broaching the metaphysis. The trial humeral stem is inserted, and the appropriate-sized humeral head trial is positioned on the stem to optimally cover the cut humeral surface (see Fig. 17Q-6G). The humeral trial is removed, and a cut protector is placed. The proximal humerus is retracted posteriorly, allowing glenoid visualization. A central hole is drilled in the glenoid vault (see Fig. 17Q-6H). The glenoid is reamed to a concentric surface (see Fig. 17Q-6I).

1152 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E

F

Figure 17Q-4   Sample steps for a labral repair on a model. A, A suture anchor is placed through an anterosuperior portal onto the glenoid face. B, A cannulated tissue penetrator is passed through the anterior labral tissue through an anteroinferior portal. C, Two limbs of a 2-0 nylon suture are passed through the cannulated tissue penetrator and are retrieved through the anterosuperior portal using a crochet hook. D, The loop of the 2-0 nylon is used to pull a 2-0 Prolene suture through the labral tissue in order to place the looped end of the suture on the glenoid side of the labrum. E, The looped end of the Prolene suture is used to pass one limb of the anchor suture through the labrum and out the anteroinferior portal. F, The knot is then tied to appose the labrum to the anchor in the glenoid.

Shoulder 1153

a slot with a bone compactor for a keeled component (see Fig. 17Q-6J). Glenoid component insertion is performed. Three transosseous sutures are placed through the lesser tuberosity for later subscapularis repair, followed by humeral component insertion (see Fig. 17Q-6K). Trialing for stability can then be performed. The subscapularis is repaired using the previously placed transosseous sutures (see Fig. 17Q-6L).

SUMMARY

Figure 17Q-5   Model used for teaching shoulder arthroplasty at the Joe W. King Orthopedic Institute Advanced Shoulder Arthroplasty Course.

Final preparation of the glenoid is accomplished by drilling an additional hole for a pegged component or ­creating

Teaching shoulder surgery can be a very difficult process. Indications have expanded, and the complexity of procedures has increased. Educators or mentors have several tools at their disposal. It is no longer possible or prudent to teach novice shoulder surgeons from start to finish in the clinical setting. Key motor and organizational skills are developed with observation and repetition. Models in particular provide an excellent avenue for the novice shoulder surgeon to hone his or her skills before venturing fully into a live setting. Once in the live setting, a period of gradual introduction into the first assistant and finally the primary surgeon position allows for a smooth transition to successful shoulder surgery with minimal patient morbidity.

A

B

C

D

Figure 17Q-6   Steps for performing shoulder arthroplasty using a model. A, Exposure of the subscapularis, axillary nerve, and anterior humeral circumflex vessels. B, Ligation of the anterior humeral circumflex vessels and placement of stay sutures in the subscapularis tendon. C, Tenotomy of the subscapularis tendon. D, Release of the inferior glenohumeral ligament, taking care to protect the axillary nerve.

1154 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

E

F

G

H

I

J

Figure 17Q-6, cont’d   E, Glenoid-based release of the inferior capsule. F, Humeral head osteotomy performed along the anatomic neck of the humerus. G, Placement of the trial humeral stem. H, A central hole is drilled into the glenoid vault. I, Reaming of the glenoid surface. J, Preparation of the glenoid for a pegged glenoid component. K, Reduction of the prosthetic glenohumeral joint following placement of sutures to be used in subscapularis repair. L, Completed subscapularis repair.

Shoulder 1155

K

L

Figure 17Q-6, cont’d   K, Reduction of the prosthetic glenohumeral joint following placement of sutures to be used in subscapularis repair. L, Completed subscapularis repair.

S U G G E S T E D

R E A D I N G S

Chan KC, Burkhart SS, Thiagarajan P, Goh JC: Optimization of stacked half-hitch knots for arthroscopic surgery. Arthroscopy 17(7):752-759, 2001. De Beer JF, van Rooyen K, Boezaart AP: Nicky’s knot—a new slip knot for arthroscopic surgery. Arthroscopy 14(1):109-110, 1998. Elkousy H, Hammerman SM, Edwards TB, et al: The arthroscopic square knot: A biomechanical comparison with open and arthroscopic knots. Arthroscopy 22(7):736-741, 2006. Elkousy HA, Sekiya JK, Stabile KJ, McMahon PJ: A biomechanical comparison of arthroscopic sliding and sliding-locking knots. Arthroscopy 21(2):204-210, 2005. Kim SH, Ha KI: The SMC knot—a new slip knot with locking mechanism. ­Arthroscopy 16(5):563-565, 2000. Kim SH, Ha KI, Kim JS: Significance of the internal locking mechanism for loop security enhancement in the arthroscopic knot. Arthroscopy 17(8):850-855, 2001.

Lee TQ, Matsuura PA, Fogolin RP, et al: Arthroscopic suture tying: A comparison of knot types and suture materials. Arthroscopy 17(4):348-352, 2001. Lo IK, Burkhart SS, Chan KC, Athanasiou K: Arthroscopic knots: Determining the optimal balance of loop security and knot security. Arthroscopy 20(5):489-502, 2004. Loutzenheiser TD, Harryman DT 2nd, Yung SW, et al: Optimizing arthroscopic knots. Arthroscopy 11(2):199-206, 1995. Nottage WM, Lieurance RK: Arthroscopic knot tying techniques. Arthroscopy 15(5):515-521, 1999.

R E F E R E N C E S Please see www.expertconsult.com

C H A P T E R  

  18

Arm Florian G. Huber

SOFT TISSUE INJURY AND FRACTURES OF THE ARM IN THE ADULT Anatomy The humerus is the skeletal support for the soft tissues that course distally from the glenohumeral joint and end in the articulation of the elbow. In addition to providing origins and insertions for the muscles that influence movement in the articulations of the shoulder and elbow, the brachium is a conduit for the nerves and vessels that innervate the muscles and provide sensation to the subcutaneous tissues of the upper extremity. The function of the arm is to position, support and move the hand. In addition to placing the hand in position to allow it to perform its functions, the arm also aids an athlete in the complex activity that enables us to balance our bodies despite carrying various loads in our upper extremities. The arm, unlike the leg, is held away from the midline of the body. The proximal portion of the leg, the hip joint, is planted in the pelvis’ acetabulum. The arm is abducted from the body by its skeletal and muscular attachments to the torso and to the scapula. In this position the arm, in an unconscious manner, helps us to maintain our balance as the hand, positioned by the arm, lifts and moves objects (Box 18-1). The origin of the functions of the upper extremity begins in the nuclei of the opposite cortex of the cerebrum. Any approach to the diagnosis of upper extremity dysfunction must proceed on this premise. A thorough understanding of the anatomy is the basis to our efforts in aiding ­athletes.

Box 18-1  Arm Function and Examination

• Provides attachments for muscles that position the hand

• Acts as a conduit for nerves and vessels traversing the arm from the thorax to the hand

• Aids in maintaining our balance • Nuclei in the opposite cerebral cortex control ­functions in the upper extremity

• Examination of the upper extremity begins with ­examination of the opposite cerebral cortex

• Lesions in the cranium, cervical spine, brachial ­plexus, and great vessels can cause symptoms in the upper ­extremity

The humeral shaft, proximally, is almost round, and as the shaft internally rotates 15 degrees, it is forced to take on a triangular appearance. Proximally, the crest of the greater tuberosity of the humerus provides an attachment for the pectoralis major tendon and serves as a landmark in the restoration of humeral length in reconstructive procedures.1 The crest of the lesser tuberosity provides attachment for the teres major muscle. About midshaft, the deltoid tuberosity provides attachment for the deltoid muscles, and at this level, the humeral shaft begins to rotate internally, and the medullary canal becomes more triangular than circular in configuration. The distal articular surface of the humerus is about 15 degrees internally rotated when compared with the proximal articular surface. The posterior aspect of the humerus is relatively flat from the deltoid insertion to the olecranon fossa and is ideal for plate placement in fracture fixation. The crosssectional anatomy of the humerus is suitable for nailing in selected fractures. Contouring the plates distally, laterally, and anteriorly allows fixation devices to be appropriate in these areas. The major veins of the arm superficially are the cephalic and basilic veins, which are found subcutaneously. The medial brachial and antebrachial cutaneous nerves arise from the medial cord of the brachial plexus and provide sensation to the anterior surface of the arm. The posterior surface of the arm is innervated by the cutaneous branches of the axillary and radial nerves. Coursing along the medial aspect of the arm is the medial antebrachial cutaneous nerve arising from the medial cord of the brachial plexus. The lateral antebrachial cutaneous nerve is the terminal branch of the musculocutaneous nerve. The significance of these cutaneous nerves lies in their propensity to develop painful neuromas in addition to a sensory deficit when injured. The cutaneous nerves traverse the subcutaneous fat and must be protected in surgical procedures. The anterior compartment of the arm is separated from the posterior compartment by the medial and lateral intermuscular septum. The lateral intermuscular septum arises from the lateral epicondyle and blends proximally with the fascia of the deltoid insertion. The medial septum traverses from the medial epicondyle to the fascia over the coracobrachialis muscle and ends with its insertion along the medial aspect of the intertubercular sulcus. The biceps and brachialis muscles are in the anterior compartment. The superficial muscle of the biceps has two proximal heads: the long head arising from the supraglenoid tubercle and the short head from the coracoid process of the scapula. The investing fascia of the belly of the biceps, when divided, allows easy separation from the underlying brachialis muscle. When displacing the biceps muscle, one can easily see the musculocutaneous nerve that supplies 1157

�rthopaedic ����������� S �ports ������ � Medicine ������� 1158 DeLee & Drez’s� O

the biceps and the medial portion of the brachialis muscle. A single muscular branch that patterns to both muscles is most common, but anatomic variations exist. The biceps brachii and the brachialis muscles are innervated at about 45% and 60% of the total acromion-lateral epicondyle distance, respectively.2 The brachialis arises near the deltoid insertion and from both intermuscular septa and inserts distal to the capsule of the elbow, into the tuberosity of the coronoid. The coracobrachialis arises from the lateral aspect of the tip of the coracoid process and inserts on the medial aspect of the humerus, blending with the origin of the brachialis and sharing its innervation with the musculocutaneous nerve (Fig. 18-1). Posteriorly the triceps covers the posterior aspect of the humerus. The three heads are innervated by branches of

the radial nerve. The medial or deep head is covered by the long and lateral heads. The long head arises from the infraglenoid rim of the scapula, the lateral head from the surgical neck of the humerus, and the medial or deep head inferomedial to the spiral groove with fibers coursing distally near the elbow joint capsule. All three heads insert distally into the triceps tendon, which begins as a two­layered structure near the mid-humerus and blends into one tendon distally as it inserts into the olecranon. The biceps muscle flexes the elbow and is the strongest supinator of the elbow. The brachialis muscle aids in flexion of the elbow. The coracobrachialis aids in adduction of the arm. The coracobrachialis flexes the arm forward and medially, and in abduction it acts with the anterior deltoid. The triceps is the major extensor of the elbow.

Figure 18-1   Muscles of the arm: anterior view, deep layer. (Reprinted from Netter Anatomy Illustration Collection, © Elsevier, Inc. All rights reserved.)

Biceps brachii tendons (cut) Short head Long head Coracobrachialis muscle Musculocutaneous nerve Branch to biceps brachii (cut) Deltoid muscle (cut)

Brachialis muscle

Lateral intermuscular septum

Medial intermuscular septum

Lateral epicondyle of humerus Lateral antebrachial cutaneous nerve Head of radius

Medial epicondyle of humerus

Biceps brachii tendon Radial tuberosity

Deep layer

Tuberosity of ulna

Arm 1159

Neurovascular Structures The brachial artery enters the arm at the lower border of the teres major, accompanied by both veins, the median nerve anteriorly, the ulnar nerve medially with both medial cutaneous nerves, and the radial nerve posteriorly. The median nerve gradually assumes a more medial position as the vessel enters the antecubital fossa. The brachial artery follows the medial border of the biceps in the anterior compartment, giving off its major branch, the profundus brachial artery just distal to the teres major. This artery is accompanied by the radial nerve. Other muscular branches supply the accompanying muscle fibers as well as the superior and inferior ulnar collateral arteries before the brachial artery divides distally into radial and ulnar branches. The major branch of several nutrient arteries to the humeral shaft arises medially near the junction of the middle and distal third of the humeral shaft. The distal humerus is supplied by multiple anastomoses with penetrating branches into each condyle and the osseous structures of the proximal ulna and radius3 (Fig. 18-2). The radial nerve arises from the posterior cord of the brachial plexus. It travels along the subscapularis to merge with the deep brachial artery at the triangular interval; both structures then separate the long and medial heads of the triceps. The nerve traverses the spiral groove of the humerus, where it is in direct contact with bone for an average span of 6.5 cm, to pierce the lateral intermuscular septum and enter the anterior compartment.4 The nerve emerges from the spiral groove 10.1 to 14.8 cm proximal to the lateral epicondyle4,5 and pierces the septum at an average of about 10.2 cm but at least 7.5 cm proximal to the articular surface of the elbow before contributing muscular branches to the lateral third of the brachialis muscle.4,6 Bono and colleagues7 quantified the mean percentage of distance from the nerve’s emergence at the lateral intermuscular septum to the distal humerus as 47% of the entire humeral length (range, 41% to 53%). Multiple proximal branches supply the lateral head of the triceps; distally, the radial nerve supplies the medial head of the triceps and branches out as the lower lateral brachial cutaneous nerve before continuation into the forearm. The deep head of the triceps can be split proximally during surgical exposure without risk for denervation. At the level of the lateral epicondyle, the main radial nerve trunk has an average diameter of 7 mm and splits into the posterior interosseous nerve (PIN) and the superficial branch8 (Fig. 18-3). The ulnar nerve arises from the medial cord and accompanies the brachial artery anterior to the medial intermuscular septum. About 8 cm proximal to the medial epicondyle, the nerve passes through the medial septum at the level of the internal brachial ligament,9 accompanied by the superior ulnar collateral artery to enter the cubital tunnel. The nerve gives off no branches in the arm. The musculocutaneous nerve arises from the lateral cord of the plexus, supplying the coracobrachialis, the biceps, and the major portion of the brachialis muscles. It exits the fascia along the lateral border of the biceps to terminate as the lateral antebrachial cutaneous nerve. The median nerve arises from the medial and lateral cords of the plexus and follows the brachial artery ­laterally,

gradually traversing medial to the artery as it enters the antecubital fossa. The median nerve supplies only the brachial artery with branches while traversing the upper arm. The anterior interosseous nerve (AIN) originates off the median nerve at a mean distance of 5.4 cm distal to the medial epicondyle.10 For internal fixation of fractures of the humeral shaft and exploration of nerves and vessels, the shaft can be approached from an anteromedial, anterolateral, and posterior skin incision. All surgical approaches are based on internervous intervals. The indications for the anterolateral approach are fractures of the humeral shaft and exposure of the radial nerve as it courses anteriorly in the arm. The patient is positioned supine with a bolster under the shoulder with the arm placed on a hand table. The entire upper extremity, including the shoulder and clavicle, is draped to the base of the neck. The anterior or the anterolateral approach can be extended proximally to the deltopectoral interval and distally into the forearm. Proximally, the incision between the deltoid and the pectoralis major is extended distally along the lateral border of the biceps, curving distally over the biceps-brachioradialis interval, into the antecubital fossa of the forearm. Superficially, the cephalic vein should be preserved for control of bleeding and to avoid postoperative swelling. It is advantageous to dissect medial to the vein and deltoid proximally, owing to the numerous muscular branches connecting the two structures. Distally, the cephalic vein is found 1 fingerbreadth medial to the lateral border of the biceps tendon in the subcutaneous tissue. The interval between the biceps and the brachialis is easily developed by incising the thinned veil of fascia over the biceps, and then the biceps muscle itself can be easily displaced either medially or laterally, exposing the brachialis muscle that is firmly fixed to the anterior aspect of the humerus, along with the musculocutaneous nerve that supplies both muscles. The brachialis muscle can be split along its lateral third to expose the distal humerus subperiosteally. The radial nerve is easily identified distally in the interval between the biceps and the brachioradialis or proximally by incising the lateral intermuscular septum at the insertion of the deltoid. The major disadvantage of this approach is the difficulty of placing a plate that abuts the articular surface distally. For the posterior approach, the patient can either be placed in a “sloppy lateral” position or fully prone with the arm supported by a hand table. It is my opinion that the sloppy lateral position is the one that most facilitates the exposure. The shoulder is draped posteriorly to the base of the neck and anteriorly to the sternoclavicular joint. For fractures, a sterile bolster can be placed under the arm to support the brachium. The advantage of this exposure is the ability to explore the radial nerve in its entirety and to achieve distal exposure of the articular surface of the distal humerus for placement of fixation devices both on the medial and lateral columns. A posterior incision is made down to the fascia of the triceps. The cutaneous branch of the radial nerve can be located in the superficial subcutaneous tissues, and this superficial branch can be traced back to the radial nerve. Once the radial nerve is identified, the entire muscle mass of the triceps can be elevated through this lateral approach, and portions of the triceps tendon

�rthopaedic ����������� S �ports ������ � Medicine ������� 1160 DeLee & Drez’s� O Coracoid process

Axillary artery

Deltoid muscle

Pectoralis minor muscle (cut)

Anterior circumflex humeral artery

Lateral cord, Medial cord of brachial plexus

Humerus

Musculocutaneous nerve

Pectoralis major muscle and tendon (cut)

Subscapularis muscle Biceps brachii muscle

Long head

Anterior and posterior circumflex humeral arteries

Short head

Teres major muscle

Coracobrachialis muscle

Latissimus dorsi muscle Brachial artery Profunda brachii (deep brachial) artery

Muscular branch

Medial brachial cutaneous nerve

Median nerve

Ulnar nerve Medial antebrachial cutaneous nerve

Muscular branch

Long head Biceps brachii muscle

Medial head

of Triceps brachii muscle

Superior ulnar collateral artery Brachialis muscle Radial recurrent artery

Biceps brachii tendon

Medial intermuscular septum Inferior ulnar collateral artery Medial epicondyle of humerus Bicipital aponeurosis

Radial artery

Pronator teres muscle Ulnar artery Flexor carpi radialis muscle Brachioradialis muscle

Figure 18-2   Muscles of the arm: anterior view, superficial layer with brachial artery. (Reprinted from Netter Anatomy Illustration Collection, © Elsevier, Inc. All rights reserved.)

Arm 1161 Capsule of shoulder joint Supraspinatus tendon Infraspinatus and Teres minor tendons (cut) Axillary nerve

Teres major muscle

Posterior circumflex humeral artery Superior lateral brachial cutaneous nerve Profunda brachii (deep brachial) artery Radial nerve Middle collateral artery Radial collateral artery Inferior lateral brachial cutaneous nerve

Long head of triceps brachii muscle Lateral head of triceps brachii muscle (cut) Medial head of triceps brachii muscle Medial epicondyle of humerus

Lateral intermuscular septum Nerve to anconeus and medial head of triceps brachii muscle Posterior antebrachial cutaneous nerve

Ulnar nerve Olecranon of ulna

Deep layer

Lateral epicondyle of humerus

Anconeus muscle

Figure 18-3   Muscles of the arm: posterior view, deep layer. (Reprinted from Netter Anatomy Illustration Collection, © Elsevier, Inc. All rights reserved.)

can be mobilized, exposing the humerus from distal to the proximal articular surface to the posterior aspect of both columns distally. The triceps-splitting approach does not allow as much distal access to the humerus. Increased intraoperative bleeding and working at significant depth in muscle tissue make exposure more difficult. The medial approach to the arm is useful for exposure of the brachial artery and the median, ulnar, and radial nerves, and possibly for nonunions that require bone grafting. The patient is positioned supine, with the arm extended on a hand table. Again, the entire anterior aspect of the shoulder to the sternoclavicular joint and to the base of the neck is draped out. The medial epicondyle, the medial bicipital groove, and the basilic vein are anatomic landmarks for placement of the skin incision. The skin and subcutaneous tissues are incised, and the basilic vein is located in the distal arm at the fascial interval between the brachialis and

triceps muscles. After identifying the medial intermuscular septum, the vascular bundle can be isolated and protected anteriorly. With the bundle retracted, the anterior and posterior aspects of the humerus can be exposed by subperiosteal dissection. The two major approaches used on the brachium are the anterior and anterolateral approaches, proximally incorporating the deltopectoral interval, the interval between the biceps and the brachialis muscles anteriorly, and the interval between the biceps and the brachioradialis muscles distally. The posterior approach described by Gerwin and associates4 is a utilitarian approach that allows the broadest exposure of the humerus, in particular over the medial and lateral columns for distal shaft and intra-articular fracture fixation. The radial nerve is the primary source of operative complications. It must be identified and protected at all times (Box 18-2).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1162 DeLee & Drez’s� O

Box 18-2  Surgical Pearls

• Always drape the extremity to the base of the neck. • Consider illumination and radiography when ­positioning the patient.

• Use extensile incisions. • Maintain meticulous hemostasis. • Keep tissues moist. • Avoid excessive retraction. • Identify and protect nerves and vessels.

TENDON AND MUSCLE RUPTURES Tendons and muscle-tendon units are exposed to dramatic tensile and compressive forces, particularly in the athlete. Often a tensile force of 5 times that of body weight is ­ delivered to the tendinous attachment site during high-stress activities. Predisposition to rupture includes conditions such as autoimmune disease, gout, hyperparathyroidism, steroid use (systemic or local injection), and advancing age.11-16 Long-standing tendinitis is also a contributing factor, although patients may not always be able to recall chronic symptoms in the face of acute injury. Degenerative attrition of the tendon is associated with calcification or exostosis at the attachment site. With advancing age, proteoglycan content is reduced, and collagen fibril diameter is decreased, reducing tensile strength. Subsequently, greater stress loading occurs at insertion sites. Tendons may fail either by direct transection, indirect rupture at the bone-tendon interface, or intrasubstance failure in either the tendon or myotendinous juncture. McMaster showed that normal tendon will not fail under tensile stress, even if 75% of its crosssectional area is sectioned.17 Failure occurs instead at the tendon insertion, myotendinous insertion, or muscle belly. In clinical practice, failure is largely at the tendon insertion. Nonlacerative rupture in the mid-substance of the muscle belly is extremely rare. In the upper extremity, the supraspinatus is most commonly ruptured, followed by the long head of the biceps and the finger extensors.18,19 In a large series of tendon injuries, 85% involved the upper extremity, with 54.7% involving the supraspinatus and 26.4% involving the biceps tendon (predominantly the long head).20 Because involvement of the long head of the biceps is generally part of shoulder soft tissue tendon disease, the inference is that a major portion of all tendon ruptures in orthopaedics occurs in the setting of shoulder pathology. With the exception of injury involving the rotator cuff, muscle and tendon ruptures of the arm are otherwise relatively infrequently observed. Nonetheless, these traumatic ruptures are increasing in prevalence, and recognition and treatment are critical for the physician who treats athletes requiring optimal function.

hematoma as a sequela.21 Pectoralis major rupture has been a fairly infrequently reported injury in the past, and Mc­Entire’s22 review of the literature in 1972 reported on only 45 cases before their series of 11. In recent years, multiple larger series have been added to the literature.23-26 Bak27 and colleagues performed a meta-analysis of 112 cases, and Aarimaa and coworkers28 reported on the largest series thus far at 33 patients, with a meta-analysis of 73 cases from the preceding English literature.

Relevant Anatomy and Biomechanics The origin of the pectoralis major is from the medial two thirds of the clavicle, the sternum, the first six ribs, and the aponeurosis of the external oblique muscle.29 It is commonly divided into a sternocostal head and the clavicular head. Its insertion is on the lateral aspect of the bicipital groove but has attachments to the greater tuberosity and deltoid insertion. The tendon can be separated into two laminae.25 The superficial lamina is formed by the clavicular head and the upper sternal fibers and inserts inferiorly. The deep lamina is formed by the lower sternal fibers and contributions from the abdominal origin and rotates 180 degrees before insertion. The muscle is innervated by the medial and lateral pectoral nerves. It functions as an arm flexor, adductor, and internal rotator.

Classification An anatomic classification system has been proposed by Tietjen30 as shown in Table 18-1.

Evaluation Clinical Presentation and History Most patients with pectoralis major rupture are males with an average age of 28 years (range, 16 to 67 years).27 The original publications on pectoralis major rupture focused on isolated case reports with sundry antecedent traumas. The common cause is forced extension and abduction against resistance. An increasingly athletic population, coupled with a heightened awareness of the injury, has led to association of pectoralis major rupture and sports-related activity. By far the strongest such association is with the bench press, whereby the rupture occurs in the terminal

  TABLE 18-1  Tietjen Classification Type

Injury Pattern

I II III III-A III-B III-C

Contusion, strain Partial tear Complete tear Complete tear of muscle origin Complete tear of muscle belly Complete tear of musculotendinous junction Complete tear at or near the tendinous insertion

Rupture of the Pectoralis Major Muscle

III-D

Patissier’s case report in 1822 is often cited as the first described pectoralis major rupture in the literature and was additionally notable for the complication of an infected

From Tietjen R: Closed injuries of the pectoralis major muscle. J Trauma 20:262-264, 1980.

Arm 1163

extension phase of the movement as the muscle contracts eccentrically in maximal abduction and extension. In one series, 47% of the patients were injured in this fashion.24 Other sports implicated include football, ice hockey, wrestling, and water sports.31-33 Associated injuries should be suspected with unusual mechanisms, and a concomitant shoulder dislocation has been reported.34 Two recent studies have noted pectoralis major ruptures in elderly nursing home patients sustained during transfers or forced repositioning.35,36 Generally, the patient notes a tearing or burning sensation in the anterior chest and axilla with sudden onset of severe weakness.

Physical Examination and Testing Passive abduction or external rotation is extremely painful, and the arm is generally held adducted and internally rotated. There is usually significant swelling with asymmetry but not always ecchymosis, and the defect may not be easily appreciated. It is best seen with abduction at 90 degrees with loss of the anterior axillary fold. Adduction and internal rotation against resistance allow for contraction of the muscle tendon unit away from the insertion and creates a “webbed” appearance to the anterior chest wall. In the case of an intact clavicular head, the defect may not be as noticeable. Palpation of the tender retracted tendon edge may be possible. Resisted adduction and internal rotation reveal significant weakness in acute cases but may be subtle in subacute or chronic presentations. Isokinetic strength testing has been advocated as a means of determining prognosis and outcome but is not routinely employed.37

Imaging Although the history and physical examination are largely sufficient for diagnosis, imaging can be helpful for complex presentations and treatment planning. Plain radiographs are warranted to rule out fracture or dislocation.38 In rare cases, a bony avulsion can be seen.39 Ultrasound is an inexpensive way to confirm the diagnosis and evaluate tear location. Magnetic resonance imaging (MRI) is the most typical imaging study used, and it has been shown to correlate well with pathology found at the time of surgery40 and allow for early diagnosis31 in the setting of extensive swelling. Partial ruptures and muscular or musculotendinous ruptures are best appreciated on MRI (Fig. 18-4A).

Treatment Options Nonoperative Most partial ruptures, muscular origin, intramuscular, or musculotendinous ruptures should be treated ­nonoperatively with emphasis on control of acute inflammation and edema. Early restoration of motion should be implemented when pain allows.41,42 Strength training begins 6 to 8 weeks from injury and functional return to sports after 3 to 4 months. Residual spasm at the injury site may persist in total ruptures but has minimal long-term ­consequences.

Operative For most athletes and patients with the desire to regain preinjury strength levels, operative treatment is indicated for both total distal tears and complete tears of either the ­sternocostal or clavicular head. Chronic and neglected ruptures in athletes can also be treated surgically with good results at even 5 to 13 years after injury26,43 (see Fig. 18-4B and C).

Weighing the Evidence Most studies comparing operative and nonoperative treatment of distal tendon rupture have been retrospective. In the first major retrospective review by McEntire and colleagues,22 75% of all patients did well irrespective of treatment method, but more excellent results were found in those surgically repaired. In a large meta-analysis study by Bak and associates,27 operative treatment was found to be more satisfactory than nonoperative treatment for strength and pain relief. Similarly, Aarimaa and coworkers’28 study found improved results in the surgical group regardless of time to repair, although the early repair faired better than delayed repair. Conversely, Hanna and associates23 and Schepsis and colleagues24 found no difference between delayed and early surgical treatment. A small study of four patients by Scott and coworkers37 suggests nonoperative treatment with monitoring by dynamometry for late weakness. The one patient with residual weakness had successful delayed repair. The evidence suggests that a measurable improvement is found in those treated with repair and earlier repair is preferable in those clearly indicated for surgery.

A u t h o r ’ s P r efe r r e d M e t h o d Patients are placed in the beach chair position. The exposure for acute repairs is through a limited anterior incision in the anterior axillary fold. In chronic cases, a typical deltopectoral incision is used for improved exposure. The distal tendon is exposed and mobilized from the underlying pectoralis minor and overlying subcutaneous tissue as needed. The bicipital groove is exposed laterally, and four or five bone tunnels are created in the lateral ridge. Heavy nonabsorbable suture is passed as a running grasping stitch (locking Krackow) and passed through the tunnels through a suture passer. The sutures are then tied over the bony bridge.

Postoperative Prescription Outcomes The two large retrospective series using meta-analyses have looked at outcomes of pectoralis major rupture in recent years and found better results with acute repair (<8 weeks) than with delayed repair (Table 18-2). Delayed repair was still better than nonoperative repair.27,28 Wolfe and associates25 used Cybex testing to find a 26% loss in peak torque and a 40% loss in work production compared with the uninjured limb in the nonoperative group, as opposed

�rthopaedic ����������� S �ports ������ � Medicine ������� 1164 DeLee & Drez’s� O

B

A

Figure 18-4   A, Axial T2-weighted magnetic resonance imaging of pectoralis major tendon rupture. Intraoperative images of pectoralis major tendon repair, including mobilized tendon with locking sutures (B) and completed repair to left humerus (C). (Courtesy of Glenn M. Garcia, MD, Department of Radiology, University of Texas Health Science Center, San Antonio.)

C

  TABLE 18-2  Postoperative Protocol for Pectoralis Major Rupture

Postoperative Period (weeks)

Activity

0-2

Postoperative shoulder immobilizer with elbow wrist and hand motion, no shoulder motion Maintain immobilizer between exercise sessions; pendulum, active-assisted, and gentle passive range of motion allowed Early resistance exercises and continued aggressive stretch; discontinue immobilizer Functional training, endurance and strengthening exercises Sport-specific training with return to noncontact sports at 5 mo, contact sports at 6-7 mo Restoration of 90% of strength of uninjured side

2-6

6-12 12-16 16-20 Return to play

to the operative group, which had 109% peak torque and 113% work production, respectively. Hanna and colleagues23 found that the surgical repair group had 97% of strength of the uninjured arm compared with 56% in the ­conservatively treated group. Such studies suggest repair to be successful in restoring near-normal function and satisfaction. Earlier repair also appears to yield improved results, although delayed repair is still warranted given its superiority to nonoperative treatment. Major complications are rare, but several case reports have noted infected hematoma as a potentially serious complication,44,45 including Patissier’s21 original paper.

Special Populations Two recent studies have examined pectoralis major rupture in elderly patients sustained in a nursing home setting.35,36 In this group of patients, repair is not indicated; however, the complication of hematoma formation must be monitored closely. Patients who have documented

Arm 1165

anabolic steroid use require counseling regarding the deleterious effects of steroids on tendon integrity, although in Aarima’s28 study, no negative effects on outcome after repair were identified in the steroid positive group.

Rupture of the Long Head of the Biceps Disease of the long head of the biceps is most frequently a component of the larger spectrum of rotator cuff pathology of the shoulder. In middle-aged patients, biceps tendinosis or frank rupture can occur concomitantly with rotator cuff disease. In the setting of a large rotator cuff tear, the biceps tendon frequently becomes inflamed and hypertrophic and can attritionally rupture, often causing relief of longstanding biceps-generated pain. Biceps instability leading to subluxation, dislocation, or rupture can occur in the setting of subscapularis tears and subcoracoid impingement. Isolated rupture of the long head of the biceps should thus be a diagnosis made only after other pathology around the shoulder has been clarified. Isolated rupture of the long head of the biceps in athletes is rare but can occur secondary to the same traumatic forces that may otherwise cause superior labrum, anterior to posterior (SLAP) tears. Two cases of concomitant long head of the biceps ruptures and SLAP tears have been reported.46 A series of ruptures in three bodybuilders at the labral attachment has also been published.47

Relevant Anatomy and Biomechanics The long head of the biceps is one of only two intra­articular tendons in the human body (along with the popliteus tendon). It originates from the supraglenoid tubercle and superior glenoid labrum. Three variations of origin have been described48: completely labral (50%), labral and supraglenoid tubercle (30%), and supraglenoid tubercle alone (20%). Its function in the shoulder has been debated, but in a younger athletic population, the healthy tendon appears to function as a shoulder depressor49,50 and dynamic stabilizer preventing anterior instability.51-55 In contrast, electromyographic studies with elbow motion controlled suggest no significant function of the long head of the biceps in shoulder function.56,57

Classification No commonly used classification exists exclusively for rupture of the long head of the biceps. Habermeyer and Walch’s48 system for biceps lesions in general is useful. Lesions that occur in the rotator interval are divided into the following categories: A, tendinitis; B, isolated ruptures; and C, subluxation. Lesions associated with rotator cuff tears are divided as follows: A, tendinitis; B, dislocation; C, subluxation; and D, rupture.

Evaluation Clinical Presentation and History Most patients with acute rupture of the long head of the biceps are older and may have had chronic shoulder pain before rupture. Although many patients may recall

a sudden tearing or “pop” in the shoulder, pain may not be extreme. In some cases, there may even be a notable improvement in pain after the acute inflammatory phase subsides. Patients with concomitant rotator cuff tears will complain more of overhead weakness, night pain, and pain with shoulder dependency.

Examination Examination will typically reveal the classic “Popeye” muscle deformity (Fig. 18-5A), but in obese patients with diminished muscle tone, the defect may not be dramatic. Ecchymosis is usually present and may be extensive, involving the entire anterior brachium. Elbow function is generally preserved, but shoulder function may be diminished, and a careful evaluation of rotator cuff integrity is required. Specialized tests for biceps pain include Speed’s, Yergason’s and Ludington’s tests58,59 (see Fig. 18-5A).

Imaging With isolated tendon rupture, plain films are negative but are suggested because presentation may be quite similar to that of a proximal humerus fracture in an elderly patient. Bicipital groove views may be obtained to characterize osteophytes and evaluate the slope of the medial wall of the intertubercular sulcus (medial wall angle).60 MRI is indicated whenever rotator cuff pathology is suspected and should be ordered liberally to avoid misdiagnosis of concomitant rotator cuff tears. The absence of the tendon in the proximal bicipital groove is typically found on MRI arthrogram.61,62

Treatment Options Nonoperative Traditionally, nonoperative treatment of isolated tears of the long head of the biceps ruptures has been recommended for most patients.63 It is important to counsel patients on the cosmetic sequelae of muscle deformity, but also to emphasize that only minimal loss of flexion and supination strength can be anticipated. Shoulder and elbow passive and active range of motion is initiated immediately, and strengthening can begin in 4 to 5 weeks or when pain resolution permits. Return to unrestricted activities is allowed after 2 to 3 months.

Operative Operative treatment of rupture of the long head is often coupled to treatment of concomitant pathology of the rotator cuff. In the setting of open or arthroscopic rotator cuff repair, the rupture of the biceps tendon can be stabilized with open or arthroscopic tenodesis. Biceps tenodesis has traditionally been performed through the deltopectoral interval to visualize the retracted tendon distally in the groove. Multiple techniques exist for tenodesis, including bone tunnels, keyhole technique, suture anchors, and interference screw fixation. Transfer of the long head to the coracoid has also been advocated and may prevent pain at the humeral tenodesis site.64,65

�rthopaedic ����������� S �ports ������ � Medicine ������� 1166 DeLee & Drez’s� O

B

A

Supraspinatus

Supraspinatus Acromion

Driver

Clavicle Coracoid

Mini-open deltoid split

Subscapularis Subscapularis

Arm in external rotator

Mini-open technique

Interference screw technique

C Figure 18-5   A, Clinical appearance of ruptured long head of the biceps muscle. B, Mini-open incision with retrieved long head of the biceps tendon. C, Mini-open technique with interference screw fixation.

Weighing the Evidence Most authors recommend conservative treatment of isolated rupture of the long head of the biceps. This is supported by studies that suggest no consequence to function with the untreated condition.66 One study even rated ­satisfaction with cosmesis of surgical tenotomy to tenodesis and found no significant difference.67 A number of studies have reported elbow flexion strength losses of 8%

to 16% and supination strength losses of 11% to 21%,68,69 with endurance loss of 25%.70 Most ruptures occur in middle-aged and elderly patients, so nonoperative treatment remains the standard because marginal differences in strength and endurance are outweighed by surgical morbidity. However, in the uncommon event of isolated traumatic rupture in an avid athlete or younger patient, consideration for tenodesis should be considered.

Arm 1167

A u t h o r ’ s P r efe r r e d T e c h n i q u e Disease of the long head of the biceps is generally addressed surgically when it is coupled with rotator cuff surgery in general. If the tendon has not ruptured completely, then it is tenodesed in the proximal bicipital groove after arthroscopic or open subacromial decompression and rotator cuff repair. The tendon can be easily approached in this setting with a small mini-open deltoid split. The arm is flexed and externally rotated to bring the bicipital groove under the deltoid split. The tendon can be brought out of the wound and prepared for tenodesis with either two suture anchors in the groove or an interference screw as described later. As an isolated procedure for complete rupture of the biceps with retraction, the patient is placed in the beach chair position. The arm is prepared and draped free. A deltopectoral approach is made, and the retracted stump of the tendon is identified within the bicipital groove. The tendon is mobilized circumferentially, and an attempt is made to liberate the muscle belly from the underlying brachialis and adjacent pectoralis tendon and overlying subcutaneous tissue. After maximal excursion is obtained, the tendon is mobilized proximally. A soft tissue interference screw technique is preferred for its improved fixation strength71 (Arthrex, Naples, Fla). A 2-0 permanent suture is woven into the tendon in a Krackow fashion for a distance of 23 mm. The tendon is sized for width. An equal-sized reamer is used to ream to a depth of 25 mm at the level of intended tenodesis. The free ends of the 2-0 suture are passed through the cannulated core of the screw using a wire suture passer. The screw is then driven into the previously reamed tunnel while maintaining countertraction on the sutures at the other end of the screw driver. After the screw is seated, the sutures are tied over the screw to reinforce the construct. In chronic cases, tendon stump may be more distal in the arm, and a second longitudinal incision may be required directly over the tendon stump in the mid-brachium. The tendon is mobilized in

Postoperative Prescription Outcomes Outcomes with biceps tenodesis alone have in the past had intermediate results.72,73 This is likely due to missed rotator cuff pathology and reinforces the need for attention to additional lesions beyond the biceps. Recent reports of tenodesis and associated shoulder surgery have been good.74-77 The interference screw technique has similarly had success, with restoration of 90% of strength.78 The results using any of the accepted techniques are generally excellent.

Complications Complications of rupture treated operatively or nonoperatively are rare. Heterotopic ossification79 and fracture after tenodesis80 have both been reported. A compartment syndrome of the anterior compartment of the brachium has also been noted.81

the distal wound and then passed under the pectoralis major tendon to be found in the proximal wound where the tenodesis is performed in the aforementioned fashion (see Fig. 18-5B and C). Wounds are closed in a standard fashion, and the arm is placed in a sling. Shoulder motion is ordered with respect to any shoulder pathology addressed and passive elbow motion is permitted with gravity-assisted active elbow extension and passive elbow flexion (Table 18-3).

  TABLE 18-3  Postoperative Protocol for Long Head of Biceps Rupture

Postoperative Period Activity Weeks 1-4 Wear sling Rehabilitation day 1 Passive and active-assistive range of motion to week 5 elevation in the scapular plane to 60 degrees to week 2; continue adding 15 to 20 degrees every 2 wk depending on repair site Add external and internal rotation in the scapular plane and active-assisted range of motion at week 3 or 4; 25-30 degrees of external rotation, internal rotation of 55-60 degrees No active range of motion; external rotation, extension abduction No biceps contraction Gentle isometric contractions of rotator cuff musculature Weeks 5-6 Initiate active range of motion of the shoulder, all planes Active range of motion of the elbow Weeks 7-9 Progress with passive and active range of motion Gentle isotonic rotator cuff and shoulder strengthening Week 10 Pain-free biceps isometrics Weeks 14-16 Gentle sotonic biceps strengthening Endurance training

Distal Biceps Rupture Rupture of the distal biceps tendon occurs almost exclusively in males and generally in the age range of 40 to 60 years.82,83 Although reported cases in the literature were relatively rare in the past, increasing awareness has led to an appreciation of the frequency of the lesion.

Anatomy and Biomechanics The biceps has a long and short head proximally, which form a bipennate muscle in the arm. It is innervated by the musculocutaneous nerve, which enters the ­coracobrachialis muscle 3.1 to 8.2 cm distal to the coracoid tip and terminates as the lateral antebrachial cutaneous nerve of the forearm, which exits from behind the musculotendinous junction of the biceps to supply sensation to the anterior lateral forearm.84 The biceps tendon attaches to the bicipital tuberosity of the radius, where the bicipitoradial bursa lies between the tendon and bone. The ­cubital

�rthopaedic ����������� S �ports ������ � Medicine ������� 1168 DeLee & Drez’s� O

Box 18-3  Ramsey Classification of Distal Biceps Rupture Partial rupture • Insertional • Intrasubstance Complete rupture • Acute (<4 wk) • Chronic (>4 wk): aponeurosis intact • Chronic (>4 wk): ruptured aponeurosis From Ramsey ML: Distal biceps tendon injuries: Diagnosis and management. J Am Acad Orthop Surg 7:199-207, 1999.

bursa lies above the ­ tendon. The tendon traverses the antecubital fossa, which is bordered by the pronator teres on the medial side and the brachioradialis on the lateral side. The brachial artery and median nerve lie medial to the biceps tendon. The bicipital aponeurosis or lacertus fibrosis blends with the forearm fascia and may or may not be torn in the setting of tendon rupture. The biceps functions primarily as a forearm supinator and secondarily as an elbow flexor.

Classification Biceps tendon injuries are categorized based on injury pattern and chronicity83 (Box 18-3).

Evaluation Clinical Presentation and History Most patients are males, frequently muscular, with a history of an eccentric force applied to an arm at 90 degrees of flexion. The patient reports a tearing sensation and sudden onset of pain and loss of flexion and supination strength. Flexion strength usually improves with late presentation, but weakness in supination remains profound. The dominant arm is affected in 80% of patients.85

Physical Examination The appearance of the antecubital fossa is marked by moderate swelling and ecchymosis with loss of the normal fullness in the region of the tendon as it crosses the flexion crease. A palpable defect can usually be felt, although the presence of an intact lacertus fibrosis may make the defect less notable. Active biceps flexion can accentuate the retraction of the muscle belly well. In the acute phases after injury, the patient may have significant loss of active flexion and supination. A partial rupture may have many of the same features as a complete rupture, but generally the tendon can still be palpated in continuity. A biceps squeeze test has been described whereby the elbow is held in 60 to 80 degrees of flexion and the muscle belly squeezed, eliciting supination of the slightly pronated forearm.86

Imaging Although the diagnosis is usually made on clinical grounds alone, atypical presentations may require further imaging studies such as ultrasound or MRI. MRI is the study of choice for imaging partial tears and evaluating cubital bursitis but cannot always reliably discriminate the severity of partial tears.87,88 Plain radiographs are warranted to rule out fracture and assess morphology of the tuberosity89 (Fig. 18-6A and B).

Treatment Options Nonoperative A trial of nonoperative treatment is advocated for patients with partial ruptures and elderly or sedentary patients with limited functional goals. Patients who opt for nonoperative treatment should be advised of a loss of 30% flexion strength and 40% supination strength85 and 86% decrease in supination endurance.90 Patients are allowed early active-assisted range of motion initiated in the first week after injury. As motion returns to normal, progressive strengthening is advanced as tolerated.

Operative As distal biceps rupture becomes a more commonly recognized and treated entity, so have the numerous techniques for operative repair multiplied. The initial reports on repair were performed using an anterior approach and had a high incidence of radial nerve injury.91 Boyd and Anderson92 described the double-incision technique with bone tunnels, which had become the standard procedure for many until the introduction of anterior single-incision repairs with improved techniques and implants. The ­ single-­incision repair can be performed with suture anchors, interference screws, or pull-through fixation with bone or soft tissue buttons.93-97 Techniques for chronic ruptures range from fixation to the ulna to improve flexion strength only,85 to recent descriptions of tendon grafting with autogenous semitendinosus,98,99 flexor carpi radialis,100 or allograft Achilles tendon.101 Partial ruptures that do not respond to conservative treatment are also indicated for surgery with a ­single-incision posterior approach allowing detachment and re-repair to the tuberosity.102

Weighing the Evidence Operative treatment for distal biceps ruptures is strongly supported by the improved flexion and supination strength and endurance found in repaired patients.85,90,103 Anatomic and early repair is further supported by a study showing that nonanatomic repair to the brachialis results in minimal improvement in supination strength despite restoration of flexion power.104 The merits of which technique to use for acute repair remains an issue of debate. The typical concern with the single-incision anterior approach is radial nerve injury, whereas the major concern with the double-incision technique is the formation of heterotopic bone and radioulnar synostosis.105 A clinical prospective

Arm 1169

A

B

Figure 18-6   T2-weighted magnetic resonance imaging of distal biceps tendon tear. A, Sagittal image with retracted tendon stump. B, Axial image with visible hemorrhage surrounding insertion site at bicipital tuberosity of proximal radius. (Courtesy of Glenn M. Garcia, MD, Department of Radiology, University of Texas Health Science Center, San Antonio.)

study found that the double-incision technique had better flexion and supination strength at 3 and 6 months than the single-incision suture anchors.106 Pereira and coworkers107 found bone tunnels to be stronger in younger bone, with no difference between anchors and bone tunnels in osteoporotic bone. In contrast, another in vitro study by Lemos and colleagues108 found improved strength with their suture anchor construct compared with bone tunnels. Traditionally, single-incision repair has been performed with suture anchors,93,94,109 but more recently, a comparison of absorbable soft tissue screws to suture anchors has suggested screws are superior to anchors for in vitro pullout strength.110 Ultimately, both the classical ­double-incision and the single-incision anterior approaches appear to have comparable profiles with regard to safety and efficacy, and technique is based on surgeon preference.

A u t h o r ’ s P r efe r r e d T e c h n i q u e The double-incision technique of Boyd and Anderson is used.92 The patient is placed supine with the arm on a hand table. A sterile tourniquet is applied. A limited anterior incision in line with a Henry approach to the anterior arm is made 2 cm above the flexion crease of the elbow. The lateral antebrachial cutaneous nerve is identified, and the retracted tendon is found in the sheath. Two running

Krackow sutures are passed into the tendon using 2-0 nonabsorbable suture. A curved clamp is passed distally to palpate the radial tuberosity and advanced just to the ulnar aspect of the radius to exit in the proximal forearm. A second incision is made over the clamp, and a muscle-splitting approach is employed to expose the radial tuberosity, taking care not to expose the ulna. The forearm is maintained in pronation to prevent injury to the posterior interosseous nerve. Retractors are placed around the radius, and the tuberosity is excavated with a small bur to create a 1.5-cm long trough. Three drill holes are prepared 7 mm apart and at least 5 mm from the edge of the trough.82 Sutures are pulled into the distal wound and passed through the tunnels with a suture passer. The arm is slightly supinated, and the tendon is seated in the trough before tying over the bone bridge. All bone dust is irrigated, and closure over a distal drain is performed. The elbow is splinted in 90 degrees of flexion and neutral rotation.

Postoperative Prescription Outcomes Morrey and colleagues85 showed 97% flexion strength and 95% supination strength of injured extremity in repaired patients; Klonz and associates104 similarly found 96.8% flexion and 91% supination strength after repair (Table 18-4).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1170 DeLee & Drez’s� O

  TABLE 18-4  Postoperative Protocol for Distal Biceps Rupture

Postoperative Period

Activity

Initial

Posterior splint at 90 degrees, neutral rotation Active-assisted extension and passive flexion allowed with an increase of 10 degrees of extension/wk to full extension by 6 wk Hinged elbow brace with extension limits Discontinue splint wear, unrestricted motion with emphasis on extension and pronation; progressive resistance allowed Advanced strengthening and functional restoration After 6-7 mo or when strength is 90% of uninvolved side

Weeks 1-8

Weeks 8-12 Week 12 to month 6 Return to play

deep and superficial laminae. Madsen and colleagues114 have recently described the medial head, forming a separate deep head that may rupture independently of the common tendon. The tendon has aponeurotic extensions to the anconeus beyond the olecranon to the proximal forearm fascia. The triceps functions as the only major elbow extensor, and its disruption essentially results in complete loss of the elbow extensor mechanism.

Classification Given its rarity, there is no commonly accepted classification of triceps tears.

Evaluation Clinical Presentation and History

Outcomes after chronic repairs are less satisfactory and have a higher incidence of complications.111 Complications include radial nerve injury, heterotopic ossification, radioulnar synostosis, pain, and loss of rotation, with a case report of median nerve entrapment. However, multiple allograft and autograft procedures have successfully restored supination and flexion strength, with only a 14% supination deficit measured in a patient treated with hamstring grafting.98,99,101 The incidence of heterotopic ossification is dramatically reduced by using the musclesplitting approach with the double-incision technique111 and meticulous hemostasis and bone debris removal. Heterotopic ossification has been reported with the anterior technique as well.105 Most radial nerve palsies resolve, but persistent nerve lesions were found in 3% to 4% of patients in Klonz’s104 meta-analysis of 277 patients. Rerupture is extremely rare but has been reported as well.82

Distal Triceps Rupture Distal triceps rupture may be the rarest of upper extremity tendon injuries, but, like distal biceps and pectoralis major ruptures, it has received increasing attention in the literature. In Anzel and associate’s20 large retrospective review of tendon ruptures at the Mayo Clinic, only eight cases of triceps disruption were noted, and four of these were lacerations. However, recent series have reported on larger numbers and association with modern preconditions, such as anabolic steroid usage, local steroid injections, weightlifting, and lineman injuries in American football players.112,113 Prompt recognition and treatment of this injury pattern are critical because the modern athlete may be more predisposed to this condition than his or her ­predecessors.

Relevant Anatomy and Biomechanics The triceps is a tripennate muscle with three origins: the long head originates from the infraglenoid tubercle, the lateral head arises from the posterior surface of the humerus distal to the teres minor above the radial groove, and the medial head arises from the posterior humerus distal to the radial groove. The muscle is innervated by the radial nerve. Its tendinous insertion is to the olecranon, where it forms

Triceps rupture is typically associated with the injury pattern of a fall on the outstretched hand with eccentric contraction of the triceps at time of impact. A second proposed mechanism is associated with a direct blow to the posterior arm. The male/female ratio is 2:1, with similar rates in dominant and nondominant extremities. The average age at time of presentation is 26 years, with a range of 7 to 72 years.115 Multiple predisposing factors have been reported, including a population of systemic conditions (hyperparathyroidism, chronic acidosis, Marfan syndrome),11,12,116 local and systemic steroid exposure,112,117 and chronic bursitis. Particular high-risk activities include weightlifting,112,118 football,113 baseball pitching,119 and the javelin throw. Bilateral ruptures have been reported in an otherwise healthy patient.120 Patients generally complain of pain and tearing and acute loss of extension strength.

Physical Examination The elbow region is usually somewhat swollen, and this may obscure any gross deformity of the tendon’s contour. A palpable defect may be felt but can be quite subtle. Extension strength should be tested by inspecting the ability of the patient to extend the arm against gravity. In partial ruptures, side-to-side differences in strength can be appreciated even in the face of preserved extension. The Thomson triceps test has been described by Viegas,121 whereby the arm is held in flexion against gravity by the examiner and the muscle of the triceps is compressed. Passive elbow extension is produced in a negative test and is deficient in a positive test.

Imaging Plain radiographs are important to evaluate not only for the “flake sign” (Fig. 18-7A) produced by avulsion of a small portion of the olecranon in many ruptures but also to rule out associated injuries. Levy and colleagues122 described six cases of radial head fracture found concomitantly with triceps rupture. The mechanism of an axial loading injury of the upper limb associated with a fall on the outstretched hand should raise concern for occult skeletal injury. ­Additionally, large avulsion fractures amenable to internal fixation or epiphyseal separations in adolescents may be found on radiographs. MRI can be invaluable in

Arm 1171

clarifying nebulous examination findings and assessing the extent of partial ruptures123 (see Fig. 18-7B).

Treatment Options Nonoperative Nonoperative treatment is generally reserved for partial ruptures and includes 4 to 6 weeks of splinting in 30 degrees of flexion followed by progressive range of motion and strengthening. Nonoperative treatment of complete triceps ruptures is retained for patients with major contraindications to surgery and low functional expectations. Consideration for repair should even be given to infirm patients of low functional status but who require upper extremity power for transfers.

Operative Acute tendon ruptures are generally repaired using a direct posterior approach (see Fig. 18-7C) and grasping sutures passed through bone tunnels. If a large fragment of bone is available, the bone-tendon unit can be reattached with screw fixation. Smaller fragments and comminuted ­fragments up to 50% of the olecranon can be excised and the tendon repaired directly to the articular margin of the

A

C

residual joint surface of the ulna. Suture anchors may be an alternative in select situations, but are not generally recommended. Multiple techniques exist for reconstruction of chronic ruptures, including turn-down flaps,124 posterior forearm fascial flaps,125 anconeus flaps,126 autograft hamstring graft,127 and Achilles allograft.128

Weighing the Evidence Loss of the elbow extensor mechanism is poorly compensated for. The anconeus provides a small measure of redundancy for triceps function but is insufficient to accomplish functional utility. As such, there are no comparison studies that evaluate nonoperative and operative treatment of ­ complete triceps rupture. The benefit of early repair (within 3 weeks) in allowing direct suture repair suggests that early operative repair is indicated whenever possible. Partial tears have been treated nonoperatively but require close observation for continued symptoms or progression to complete rupture. In a series of professional ­football players, 4 of 10 players with partial tears ultimately required surgery (1 for early rupture, 3 for refractory pain and ­ weakness). The other 6 had complete clinical recovery without surgery.113 One author recommends surgical repair if more than 50% of the tendon is torn on MRI and is coupled to significant triceps weakness.96

B

Figure 18-7   Distal triceps tendon rupture.   A, Radiographic “flake sign” associated with tendon avulsion. B, Sagittal T2-weighted magnetic resonance image of distal triceps tendon rupture C, Intraoperative image of complete distal triceps tendon avulsion (posterior approach to right elbow with patient in prone position).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1172 DeLee & Drez’s� O

A u t h o r ’ s P r efe r r e d T e c h n i q u e The patient is placed in the lateral decubitus position. The arm is prepared and draped free, and a sterile tourniquet is used. A posterior incision is made, curving just lateral to the tip of the olecranon. The ulnar nerve is identified and protected. Single large fragments are internally fixed with a screw and washer and tension band or with tension banding alone. Small bone fragments are excised, and the remaining tendon is prepared by passing a 5-0 nonabsorbable suture in a Bunnell crisscross pattern. The sutures are passed through crossed bone tunnels as described by ­Morrey129 and tied over the bone bridge distally. Local periosteum and fascia can be used to reinforce the repair as well. The tourniquet must be released before suturing the repair because it may prohibit tendon excursion. The patient is splinted in 30 degrees of flexion.

Postoperative Prescription Outcomes Generally, outcomes after triceps repair are successful at restoration of strength with minimal morbidity (Table 18-5).116,130,131 Isokinetic peak strength was found to be 92% of the uninjured side in primary repairs.130 Cybex testing in a bodybuilder confirms satisfying results even in high-demand patients.118

Complications Complications are rare, but olecranon bursitis from wire over the repair site has been reported.132 Three reruptures and one case of transient ulnar neuritis were noted in 22 patients in a retrospective review.130 Slight loss of range of motion with flexion contracture of about 5 to 20 degrees can be anticipated.131,133

  TABLE 18-5  Postoperative Protocol Distal Triceps Rupture

Postoperative Period

Activity

Weeks 1-6

Long arm splint; elbow flexed 30-45 degrees Can use hinged splint and block range of motion, yet allow range of motion during therapy and gradual elbow flexion Passive elbow extension Active elbow flexion Night-time extension splint if needed Full passive elbow extension Passive or gentle active elbow flexion to 30 degrees, increasing by 15-20 degrees/wk depending on repair Full active flexion Active extension after 6 wk Strengthening beginning with midrange isometrics, then isotonic concentric contractions, and finally, eccentric muscle contractions

Weeks 2-6

Week 6 Weeks 10-12 to month 4

Special Populations Patients with underlying systemic disease may present without a major antecedent trauma. Calcification at the tendon-bone interface and hypovascularity have been proposed as causes. Bilateral ruptures have been reported.11

FRACTURES Although the upper limb is the most commonly injured body part during sports participation, humerus fractures in adolescent and adult athletes are relatively rare compared with other anatomic locations.134,135 Characteristic injury patterns of the humerus have been well described for several sports and fracture locations. Most injuries are stress induced, caused by chronic overuse, poor technique, and deconditioning, and are associated with fatigue of surrounding musculature.136 They can be prevented with early recognition of injury patterns, activity modification, and appropriate training and conditioning. The diagnosis is generally made by obtaining a medical history, physical examination, and appropriate imaging studies. A high level of suspicion for stress fractures in upper extremity–dominated sports is indicated. Treatment is mostly nonoperative because low-energy fractures generally heal with a period of immobilization followed by a gradual and structured rehabilitation regimen. Only severely displaced, unstable, open, high-energy patterns or fractures associated with progressive neurologic impairment may warrant surgical intervention.137 Fractures in adolescent athletes are becoming more frequent because of increasing numbers of participants and higher levels of competitiveness in youth sports.

Proximal Humeral Epiphysiolysis (Little Leaguer’s Shoulder) Relevant Anatomy and Biomechanics Repetitive rotational shear injury or stress fracture of the proximal humeral epiphyseal growth plate in overheadthrowing athletes has been termed little leaguer’s shoulder.138-141 Three distinct ossification centers of the humeral head, greater tuberosity, and lesser tuberosity coalesce at about 7 years of age. The proximal humeral physis usually fuses between ages 14 and 20 years and contributes about 80% of the longitudinal growth of the humeral shaft. The injury most commonly occurs in adolescent, highperformance male pitchers between ages 11 and 16 years (average age, 14 years) but has also been reported in ­athletes participating in volleyball, gymnastics, badminton, tennis, and cricket.142-145 In baseball, the proximal humerus and shoulder are exposed to significant external rotation torque during the late cocking phase. Shear forces during early throwing motion and distraction during ball release in pitching have been well described. From an external rotation and abduction position, the shoulder is forcefully rotated internally and adducted. Opposing proximal muscular attachments (rotator cuff) and deltoid, pectoralis major, and triceps attachments distally render the extracapsular proximal humeral physis particularly vulnerable to repetitive rotational microtrauma.146 Imbalance among

Arm 1173

underdeveloped muscle mass, joint laxity, accelerated growth spurts in the growing skeleton, and high-frequency pitching with poor mechanics may predispose to injury.147

Box 18-4  Evaluation of Proximal Humeral Epiphysiolysis

Classification

• Lateral tenderness over proximal humerus • Painful external rotation

Physical Examination

Epiphyseal plate injuries of the proximal humerus have been classified by Neer and Horwitz according to severity of displacement148 (Table 18-6).

Evaluation Clinical Presentation and History Patients complain of gradually increasing lateral shoulder pain with aggravation by throwing and later at rest. There is usually no history of trauma or neurologic involvement.

Radiographs

• Lateral widening of proximal humeral physis • Lateral fragmentation of metaphysis • Demineralization • Sclerosis • Cystic changes

Treatment Options

Physical Examination and Testing

Nonoperative

Physical examination shows localized lateral tenderness to palpation over the proximal humerus with painful external rotation. Rotator cuff strength, shoulder stability, and labral integrity should be assessed. Loss of range of motion and swelling are uncommon findings.

Treatment usually consists of activity modification with rest from throwing for at least 6 weeks or until symptoms subside (but more often the remainder of the season).

Imaging

Operative interventions are indicated only for fractures in association with polytrauma, neurovascular injury, or ­tenting skin or in patients with severe head injury with spasticity. A small subset of adolescent athletes with severe displace­ment (see “Special Populations”) poses a relative indication.

Bilateral anteroposterior plain film radiographs in internal and external rotation show widening of the proximal humeral physis consistent with a Salter-Harris type Ⅰfracture.149 Associated findings in about half of patients may include demineralization, sclerosis, cystic changes, and lateral fragmentation of the proximal humeral metaphysis.142 The physis is usually stable because of its inherent pyramid shape and a strong band of posteromedial ­periosteum. MRI and bone scan studies do not add any information but may be more sensitive in the early stage of physeal injury when plain radiographs remain normal.150 Oblique coronal T1-weighted images will reveal widening of the proximal lateral humeral physis and increased signal intensity indicative of periosteal and bone marrow edema on T2-weighted images. Follow-up MRI or bone scan studies in an uncomplicated course are not indicated.142,149-152 Rotator cuff tear, glenohumeral instability, acromioclavicular joint injury, clavicle or proximal humerus fracture, and osteolysis of the distal clavicle should be considered as differential diagnoses153 (Boxes 18-4 and 18-5; Fig. 18-8A).

  TABLE 18-6  Neer and Horwitz Classification of Proximal Humeral Epiphysiolysis Grade

Displacement

I II III IV

Less than 5 mm Less than one third of shaft width Two thirds of shaft width More than two thirds of shaft width

From Neer CS 2nd, Horwitz BS: Fractures of the proximal humeral epiphysial plate. Clin Orthop 41:24-31, 1965.

Operative

Weighing the Evidence A sports-related physeal overuse injury in a little league pitcher was first described in 1953 by Dotter.138 Subsequent reports have since substantiated mechanism, ­ treatment, and expected outcomes in affected athletes.142,146,154 When comparing highly competitive, year-round pitching with 44 less competitive pitchers in recreational leagues, Torg and colleagues were unable to identify any patients with proximal humeral epiphysiolysis.155 In the largest series of 23 patients, Carson and Gasser142 reported on 23 competitive baseball players with an average age of 14 years (range, 11 to 16 years). Two thirds of patients played baseball year round, and 26% played on two teams at the same time. ­Twenty-one ­players reported gradual onset of pain with an average duration of 7.7 months (range, 1 week to 2 years). Although all

Box 18-5  Differential Diagnosis of Proximal Humeral Epiphysiolysis

• Rotator cuff tear • Glenohumeral instability • Acromioclavicular joint injury • Clavicle fracture • Proximal humerus fracture • Osteolysis of the distal clavicle (weightlifting!)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1174 DeLee & Drez’s� O

A

C

patients presented with lateral widening of the proximal humeral physis, and 50% showed signs of metaphyseal fragmentation on plain radiographs, Song and colleagues152 reported on one patient without ­ significant radiographic findings but MRI findings consistent with a physeal stress injury. With rest from throwing (average, 3 months; range, 1 month to 1 year), 21 patients (91%) were able to return to asymptomatic play.142

B

Figure 18-8   A, Little leaguer’s shoulder: proximal humeral epiphysiolysis (arrow) in an adolescent pitcher. B, Radiograph of Neer grade IV proximal humeral epiphyseal fracture. C, Neer grade IV proximal humeral epiphyseal fracture. (B, From Dobbs MB, Luhmann SL, Gordon JE, et al: Severely displaced proximal humeral epiphyseal fractures. J Pediatr Orthop 23:208-215, 2003. C, Redrawn from Dobbs MB, Luhmann SL, Gordon JE, et al: Severely displaced proximal humeral epiphyseal fractures. J Pediatr Orthop 23:208-215, 2003.)

Lyman and associates156 studied the association of elbow and shoulder pain with number, type, and mechanics of pitching in a prospective cohort of 476 pitchers aged 9 through 14 years during one season. Half of all pitchers experienced shoulder or elbow pain at some point throughout the season. Use of curveball pitches increased the relative risk for developing shoulder pain by 52%. There was a significant relationship between number of game pitches

Arm 1175

and the risk for developing shoulder pain. Joint pain was differentiated from postactivity arm pain and was regarded as an early sign of throwing injury.156-158

Rehabilitation and Prevention Complete cessation of throwing for 6 to 12 weeks with gradual range of motion exercises as tolerated is prescribed. Physical therapy is not necessary. Return to throwing activities is allowed after complete resolution of symptoms. Prevention and rehabilitation should consist of a specific preseason conditioning program with predetermined limits on pitch count and work-out frequency at practice and at home. A survey discussed by Andrews and Fleisig159 for USA Baseball News recommended specific, age-dependent pitch counts and a specific sequence of throwing techniques that pitchers should acquire through their adolescence.156,159,160 These studies recommended limitations on pitch counts to no more than 75 pitches during a single game and less than 600 competitive pitches per season for pitchers between 9 and 14 years of age. Exclusive use of fastball and changeup pitches should replace curveball and sliding pitches as breaking pitches in this age-group. Control skills should be emphasized over pitch velocity, and year-round pitching on several teams is discouraged.156,157,161,162

Potential Complications Premature physeal closure142 with possible limb-length discrepancy, osteonecrosis of the humeral head, and subluxation of the glenohumeral joint have all been described but are extremely rare occurrences.162 Development of humeral retrotorsion is a common, protective adaptation of humeral head version observed in adult pitchers to accommodate the repetitive external rotation stresses of overhead throwing.163,164

Criteria for Return to Play Return to activity is based on clinical, not radiographic, findings because healing may occur much sooner than radiographic remodeling and reestablishment of normal physeal width. A temporary shift to a different position until physeal closure should be considered on an individual basis.

Special Populations Dobbs and colleagues165 have reported on a very small subsegment of severely displaced, Neer grade III and IV proximal humeral epiphyseal fractures in mostly adolescent patients with limited remodeling potential ������ (see Fig. 18-8B and C)�������������������������������������������������� . Of 28 patients treated during a 15-year period, 8 patients sustained their injuries during participation in wrestling, hockey, football, and basketball. All patients underwent a closed reduction under anesthesia with or without fixation as indicated. Five of 28 patients required an open reduction due to interposed periosteum and biceps tendon. All fractures were converted to a Neer grade I or II displaced configuration and healed without residual impairment. On long-term follow-up (average, 4 years; range, 2 to 14 years) all patients had regained full preinjury function.165

Avulsion Fracture of the Lesser Tuberosity Relevant Anatomy and Biomechanics Isolated avulsion fractures of the lesser tuberosity are rare injuries and are usually reported in patients sustaining a posterior shoulder dislocation or violent falls and accidents.166 Few injuries have been reported in adolescent athletes participating in wrestling or football activities.167-171 A forced external rotation against active resistance by an intact subscapularis muscle is consistently described as the injury mechanism.

Classification The Neer four-segment classification of proximal humerus fractures is based on angulation and displacement of one or more of the consistent fragments.172-174

Evaluation Clinical Presentation and History The injury is difficult to diagnose and is frequently encountered as persistent, chronic anterior shoulder pain after a remote traumatic injury.168

Physical Examination and Testing Physical examination findings include ecchymosis, tenderness to palpation over the lesser tuberosity, increasing pain with external rotation, and resisted internal rotation. Subscapularis insufficiency manifests as a classic positive lift-off test175 and frequent pain with the anterior apprehension test.167-175 In chronic injuries, excessive external rotation is noted.

Imaging Radiographic findings are most apparent on axillary shoulder images or anteroposterior films in internal rotation. Computed tomography (CT) allows assessment of displacement, comminution, and fragment size; MRI further assists preoperative planning by detection of associated labral tears, biceps tendon displacement, or rotator cuff pathology.176

Treatment Options Nonoperative Nondisplaced and chronic injuries can be treated nonoperatively with a rotator cuff strengthening program and close observation.

Operative Acutely displaced fractures (greater than 5 mm of displacement or 45 degrees of angulation), mechanical block, and failed nonoperative management with persistent disability are operative treatment indications.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1176 DeLee & Drez’s� O

Weighing the Evidence Sports-related fracture avulsions of the lesser tuberosity are extremely rare and have been described for wrestling, football, and skateboarding injuries.167-169,177 Nondisplaced injuries can be treated nonoperatively but should be followed closely to avoid secondary displacement with medial subscapularis retraction.169,176 Severe disability with persistent pain and failure of nonoperative management are indications for secondary open reduction with internal fixation (ORIF). Fragment excision is not recommended.167 Ogawa and Takahashi167 reported 10 cases of isolated and displaced lesser tuberosity fractures in patients 11 to 68 years of age with various mechanisms of injury. Two patients sustained their fractures during wrestling matches. Of four chronic cases, two were treated operatively with excellent outcome after conservative treatment had failed. All chronic cases had excellent outcomes. Of six acute fractures, three patients refused surgery and had satisfactory outcomes with residual pain. Two of the remaining three patients had excellent outcomes after open screw fixation. For displaced and acute injuries, an ORIF through a deltopectoral approach with cannulated AO cancellous screws and washers166-168 or nonabsorbable suture fixation in adolescents169 has reliably yielded excellent outcomes when combined with repair of the anterior capsule and closure of the rotator interval. Operative indication for all but nondisplaced injuries is based on concern for subsequent nonunion, malunion, medial displacement of the long head of the biceps tendon into the joint, anteromedial impingement, and functional deficits.167,168 Paschal and associates168 reported on two operative patients with delayed diagnosis of 14 and 36 months with persistent pain. Both patients achieved full functional recovery with osteotomy, anatomic reconstruction, and screw fixation of the lesser tuberosity. In contrast, Ross and Love170 reported on two patients of similar age with acute and nondisplaced injuries from wrestling and skateboarding. Both were successfully treated nonoperatively, with asymptomatic nonunion occurring in one patient.

A u t h o r ’ s P r efe r r e d M e t h o d Operative repair is achieved through a standard deltopectoral approach. Both the subscapularis and the long head of the biceps tendons should be assessed for significant displacement. Depending on size and quality of the lesser tuberosity fragment, it is reduced into its humeral bed and repaired with either 4-mm cannulated screws with a washer or nonabsorbable, transosseous sutures. We prefer to use No. 2 FiberWire (Arthrex, Naples, Fla). A displaced or significantly damaged biceps tendon is treated with a metaphyseal in situ tenodesis after reduction of the tuberosity has been achieved.

Postoperative Management and Rehabilitation Postoperative treatment consists of sling support for 3 weeks with intermittent passive-assisted range of motion exercises followed by active range of motion in ­conjunction

with ­ isometric exercises. The repair is protected from active internal rotation against resistance and limited to 45 degrees of external rotation for a total of 6 weeks. Strengthening of internal and external rotators is then advanced with rubber bands.168,169

Potential Complications Possible complications include secondary fragment displacement, biceps tendon subluxation, malunion, nonunion, limited range of motion, external rotation stiffness, and chronic pain.167,168

Criteria for Return to Play Return to play is allowed after regaining full shoulder range of motion and ­preinjury-level internal rotation strength in neutral and abduction with radiographic signs of fracture healing.

Humeral Shaft Fractures Relevant Anatomy and Biomechanics Humeral shaft fractures in sports occur either through high- or low-energy mechanisms. Although high-energy injuries are treated according to trauma guidelines, several sport-specific low-energy fracture patterns have been identified and described. First accounts178,179 of spiral humeral shaft fractures sustained during the act of throwing objects can be found as early as 1805. “Ball thrower’s fractures” have been documented in athletes participating in overhead throwing activities, such as softball, baseball, shot put, football, and javelin, and in soldiers during grenade throwing.178,180-186 Caused by sudden opposing muscular forces during throwing motion, this injury occurs in healthy, young patients without prior history of trauma or pathologic bone lesion. Various mechanical explanations have been proposed with little scientific evidence. Lack of bony adaptation, muscle fatigue, excessive torque, sudden muscular contraction, repetitive stress in unconditioned athletes, and incorrect throwing technique have all been implied in the literature.178,183 None of these factors are proved, but their combination is likely. A similar spiral fracture pattern with a distinct mechanism has been described in participants of “arm-wrestling” matches against man or machine.187,188 Through a shift in body weight and the opposition’s counterattack, the shoulder internal rotator muscle group suddenly changes from peak concentric contraction to maximal eccentric contraction with subsequent spiral fracture of the humeral shaft due to extreme rotational torque178 (Fig. 18-9). Stress fractures of the upper limb are considered to be a common and frequently preexisting condition in the previously described throwing and wrestling activities but can also occur with other upper extremity–dominated sports, such as swimming, tennis, and weightlifting.178,180-185,189192 The injury occurs with a recent and significant increase of activity. In response to repetitive stress with traction forces through Sharpey’s fibers, the cortex adapts with remodeling bone resorption and osteoblastic activity. After

Arm 1177 AO CLASSIFICATION—SIMPLE FRACTURE A1 A2 A3 Simple fracture, Simple fracture, Simple fracture, Spiral Oblique (≥30°) Transverse (<30°) Posterior View of Humerus

A AO CLASSIFICATION—WEDGE FRACTURE B1 B2 B3 Wedge fracture, Wedge fracture, Wedge fracture, Spiral wedge Bending wedge Fragmented wedge

Figure 18-9   Muscular torsional and rotational forces producing a spiral fracture of the humerus.

Posterior View of Humerus

the rate of adaptive remodeling is overcome by external forces, stress fractures can occur.193

Classification For all clinical purposes, humeral shaft fractures are mainly classified in descriptive terms stating the location, energy, and mechanism pattern; articular extension; and associated soft tissue, vascular, or neurologic injuries. A more useful categorization for clinical trials and research purposes is the AO/ASIF Comprehensive Long Bone Classification system, modified by the Orthopaedic Trauma Association in 1996194 (Fig. 18-10).

B AO CLASSIFICATION—COMPLEX FRACTURE C1 C2 C3 Complex fracture Complex fracture, Complex fracture, Segmental Irregular

Evaluation Clinical Presentation and History

Posterior View of Humerus

Prodromal activity-related pain may be present in more than 50% of patients.184,195 Many authors suspect an underlying stress fracture, with patients retrospectively complaining of specific muscle fatigue and pain during and after play at the fracture site or in referred locations (shoulder and elbow). Symptoms of stress fractures are nonspecific, with insidious onset of localized deep pain associated with activity. The osseous injury is often overlooked owing to initially normal-appearing plain radiographs and is attributed to more common reactive soft tissue inflammation.

Physical Examination and Testing Most torsional shaft fractures occur in the distal third and present with varying degrees of angulation, swelling, and hematoma. A neurovascular examination with particular attention to dorsal wrist extension and radial nerve function should be diligently documented before and after

C Figure 18-10   AO classification of humeral shaft fractures. (Redrawn from DeFranco, Lawton J: Radial nerve injuries associated with humeral fractures. J Hand Surg 31[A]:655-663.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1178 DeLee & Drez’s� O

reduction maneuvers. Cases of nerve palsy, usually a neu­ rapraxia, should be monitored for progression with serial ­examinations.183,196

Imaging Plain radiographs are mandatory in all patients and commonly reveal a minimally displaced or angulated mid-third–to–distal third humerus fracture, with a characteristic medial butterfly fragment in 28% of patients (Fig. 18-11).178,183 A localized periosteal stress reaction, endosteal thickening, or a radiolucent cortical line may be apparent before completion of a fatigue fracture.184,191 In a patient with a classic presentation and no constitutional medical symptoms or structural abnormalities on plain radiographs, an extensive work-up to rule out metabolic or neoplastic bone disease is not indicated.197,198 A high level of suspicion should be maintained in evaluation of a throwing athlete with arm pain. An early MRI is the preferred initial imaging technique to rule out an impending fracture.197,199-201 Fredericson and colleagues202 and Arendt and Griffiths203 reported on MRI-based classification systems for assessment of stress reactions and fractures of the tibia in runners and correlated their grading with time to return to play. The classification had predictive value in estimating the duration of disability. A detailed gradual rehabilitation program guided by pain was established203 (Table 18-7). Three-phase bone scan studies are less specific but allow for early detection as well, with diffuse in­­creased cortical uptake before fracture occurrence189,204 (Fig. 18-12).

Treatment Options Nonoperative Treatment for most uncomplicated and isolated fractures with minimal displacement and axial alignment is conservative with an initial coaptation splint or hanging arm cast for 1 to 2 weeks (Fig. 18-13). Alignment should be followed closely with weekly radiographs (Box 18-6).

Operative Operative treatment is reserved for failed closed treatment, interposed soft tissue preventing reduction, open or segmental fractures, vascular injuries, “floating elbow” ­injuries, intra-articular extension, burns, and spinal cord injury. Patient-associated factors include polytrauma, traumatic brain injury, noncompliance, morbid obesity, and large breasts preventing maintenance of reduction. Internal fixation with compression plating according to AO principles remains the gold standard, intramedullary nailing devices are used with increasing frequency as well but are contraindicated in patients with preoperative radial nerve palsy.

Weighing the Evidence In a biomechanical evaluation of 25 major league pitchers with high-speed motion analysis, Sabick and colleagues163 found that the proximal humerus undergoes maximal external rotation torque (up to 92 ± 16 N-m) during late cocking

phase before release. Torque peaks support the likelihood of fracture at this point of the throwing cycle during abduction by the deltoid, supraspinatus, infraspinatus, and teres minor muscles. As the arm progresses into the acceleration phase it is exposed to a reversal internal rotation torque by the latissimus dorsi, pectoralis major, and teres major muscles.163,205 With uncoordinated throwing technique, these opposing forces may overlap and could cause a spiral shaft fracture. The weight of the object thrown appears to also contribute to the probability of sustaining this injury.183,206 Branch and colleagues195 identified the following specific risk factors for torsional fractures in baseball players: age over 30 years, prolonged layoff period from pitching, lack of regular exercise, and prodromal throwing arm pain. The two largest nontraumatic series to this day have been reported by Chao and coworkers183 following throwing of hand grenades in 129 soldiers and by Ogawa and Yoshida178 in 90 baseball players with an average age of 25 years (range, 12 to 43 years). All players in Ogawa’s178 series sustained their injuries during recreational play while throwing with maximal effort (64 of 90 patients threw a fastball, and 18 patients attempted a curveball). Eighty-five of 90 patients described their arm breaking in the late cocking or acceleration phase. Eighteen of 90 patients had experienced prodromal pain at the fracture site, raising the question of preexisting stress fractures at the later site of injury. According to the authors, no radiographic signs of stress fractures were identified on any of the injury films, and additional imaging studies such as MRI or bone scan were not obtained. Forty-six initial patients were treated with plate or screw fixation with good outcomes. The last 44 patients in the study were all treated nonoperatively with cast immobilization with or without functional bracing, with 38 of 44 having excellent results. These results are consistent with reports on high-energy fracture outcomes and union rates.207 Injuries in adolescents tended to occur in the proximal half of the humeral shaft.205 Ogawa and Ui208 also reported 30 patients with midthird to distal-third spiral humeral shaft fractures sustained during peak force exertion in an effort to decide an arm-wrestling match. Fracture patterns appeared similar to those seen in “ball thrower’s fractures,” with medial ­butterfly fragments present in 23% of cases and no comminuted fractures. There was no correlation between physical strength and fracture occurrence. The authors reported nonoperative treatment in 13 and surgical fixation in 17 patients with equally good results. Indication for surgery was mainly the patient’s desire for faster recovery. Both Moon and colleagues209 and Whitaker210 reported good results for nonoperative treatment in hanging arm casts. Heilbronner and associates211 reported two-screw fixation in three of seven patients because of interposed muscle tissue and inability to obtain adequate reduction. All patients had excellent or good outcomes on follow-up after 1 to 2 years. These treatment results are confirmed in a retrospective comparison of operative versus nonoperative treatment in 40 patients with extra-articular distal-third humeral diaphysis fractures.212 The authors conclude that both treatment options provide ­reliable, excellent results and advocate an individual, patient-oriented decision after discussion of all treatmentspecific risks and benefits.212

Arm 1179

A

B

C

E

D

F Figure 18-11   Spiral distal humerus fracture in a 20-year-old woman with characteristic medial butterfly fragment. A and   B, Anteroposterior and lateral injury radiographs. C and D, Anteroposterior and lateral follow-up radiographs 6 weeks after open reduction internal fixation with precontoured medial and posterior locking plates. E and F, Final follow-up radiographs demonstrating bony union with surgical treatment.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1180 DeLee & Drez’s� O

To compare compression plating versus intramedullary nailing techniques, Bhandari and associates213 conducted a meta-analysis of three randomized trials and found a lower relative risk for reoperation in the patient group treated with open reduction and internal plate fixation but concluded that a larger, randomized trial was needed. Chapman and colleagues214 found no statistically significant differences in the two treatment groups, besides increased elbow versus shoulder pain and stiffness for plating and intramedullary nailing, respectively. Preexisting stress fractures have been detected with bone scan, CT, and MRI studies before fracture completion.184-199 Polu and coworkers199 reported on a 21-year old Division I collegiate baseball pitcher with a 6-month history of arm pain related to throwing with normal plain

  TABLE 18-7  Fredricson Magnetic Resonance Imaging Classification of Stress Fractures Grade

Positive Sequence

Duration of Rest

1 2 3 4

STIR STIR, T2 STIR, T1, T2 T1 and T2 with visible fracture line

3 wk 3-6 wk 3-4 mo At least 4 mo

STIR, short T1 inversion recovery. From Fredericson M, Bergman AG, Hoffman KL, Dillingham MS: Tibial stress reaction in runners: Correlation of clinical symptoms and scintigraphy with a new magnetic resonance imaging grading system. Am J Sports Med 23: 472-481, 1995.

B

A

C

D

Figure 18-12   Magnetic resonance imaging findings in runners with tibial stress reaction according to the classification proposed by Fredericson and colleagues. A, Grade 1, T2-weighted image. B, Grade 2, T2-weighted image. C, Grade 3, T1-weighted image. D, Grade 4, T1-weighted image with visible fracture line. (From Fredericson M, Bergman AG, Hoffman KL, et al: Tibial stress reaction in runners. Am J Sports Med 23:472-481, 1995.)

Arm 1181

A

B

C

E

D

F Figure 18-13   Humeral shaft fracture sustained by a 22-year-old patient while throwing a baseball non-competitively. A and   B, Anterior-posterior and lateral injury radiographs. C and D, Anterior-posterior and lateral follow-up radiographs 2 weeks after closed reduction and coaptation splint immobilization. E and F, Final follow-up radiographs demonstrating bony union with non-operative treatment.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1182 DeLee & Drez’s� O

Box 18-6  Reduction Criteria for Humeral Shaft Fracture

• Shortening < 3 cm • Rotation < 30 degrees • Anteroposterior angulation < 20 degrees • Varus/valgus angulation < 30 degrees From Klenerman L: Fractures of the shaft of the humerus. J Bone Joint Surg Br 48:105–111, 1966.

radiographs but a nondisplaced distal humeral shaft stress fracture apparent on MRI and bone scan. The authors recommend the use of MRI for the early detection of stress fractures in order to prevent progression to acuteon-chronic humeral shaft fracture. Lee and associates200 and Hoy and colleagues215 reported similar MRI findings in 12 and 8 elite tennis players respectively, with activityrelated pain over the mid and distal humerus. Compared with healthy volunteers, all athletes showed increased marrow edema on short T1 inversion recovery (STIR) imaging sequences, which correlated in intensity with duration of prodromal pain. All patients entered rehabilitation and rest from activity, and none progressed to stress fracture. A MRI grading system may be predictive of recovery time.199-215 Rettig and Beltz191 reported a similar case with minimal periosteal elevation on plain radiographs, but with detection of a vertical stress fracture in the anterior humeral cortex on three-phase bone scan and CT.

A u t h o r ’ s P r efe r r e d M e t h o d Nonoperative therapy is initiated after closed reduction as indicated. The humerus is immobilized for 10 to 14 days in a well-molded coaptation splint. After acute pain and swelling are decreased, the patient is fitted with an adjustable and molded functional brace to facilitate early range of motion exercises of elbow and shoulder joints. Daily readjustment of the brace is emphasized to facilitate union and alignment through hydraulic compression.207 Operative fixation is approached through an anterolateral incision for proximal-third and mid-third shaft ­fractures; distal-third involvement with or without intraarticular ­ extension usually requires a posterior approach. We favor compression plate fixation through a posterior triceps-sparing or triceps-reflecting anconeus pedicle (TRAP) ­approach.216,217 The use of broad 4.5-mm locking compression plates (LCPs) allows for rigid fixation, preservation of periosteal blood supply, and a biomechanically favorable, staggered screw configuration. An interfragmentary lag screw and proximal and distal fixation with purchase of at least eight cortices is attempted whenever possible. Physically small patients may require application of a narrow LCP plate. Distal extension of the fracture may require use of anatomically contoured 3.5-mm locking plates, preferably in perpendicular planes,218 or the use of a combined 4.5/3.5-mm metaphyseal locking plate. Intramedullary fixation is reserved for pathologic fractures at our institution.

Postoperative Management and Rehabilitation Nonoperatively managed patients are fitted with a functional fracture brace 2 weeks after surgery and are started on a gradual range of motion and strengthening program for 6 to 8 weeks. Surgical patients are placed in a sling for 48 hours and are started on active range of motion exercises of shoulder and elbow joints. Full range of motion should be achieved by 6 weeks, at which point light weights are permitted.

Potential Complications The treatment of humeral shaft fractures in light of a radial nerve palsy remains controversial. Iatrogenic nerve deficits after closed reduction most often tend to resolve spontaneously,219-221 whereas nerve injuries associated with open fractures are more likely to involve a nerve laceration.222,223 Ogawa and colleagues178,208 reported a radial nerve palsy in 14 of 90 thrower’s fractures (16%) and in 7 of 30 arm-wrestling injuries (23%). Full recovery occurred in all patients within 2 to16 months with nonoperative treatment.178,197,208 Radial nerve laceration is extremely rare and has been reported in 3 of 129 patients by Chao and colleagues.183 All three patients required exploration and repair of complete transections after persistent palsy for 6 months after injury. Another complete radial nerve transection in a 28-year-old skier was treated with early primary nerve repair and ORIF by Takami and associates.224 The patient regained full function after 30 months. Possible causes for recurrent fractures are a combination of early return to throwing, occult nonunion, incorrect throwing mechanics, bone atrophy, and pathologic fracture. Because most recurrent fractures occurred either distal or proximal to the well-healed initial fracture site, nonunited extensions of the initial fracture could act as stress risers.182,225,226 Complications associated with nonoperative treatment are malalignment, nonunion, elbow stiffness, and skin breakdown. Operative complications include iatrogenic radial nerve palsy, infection, nonunion, and loss of fixation.212 Plate fixation is associated with increased incidence of elbow stiffness and pain, and antegrade intramedullary nail fixation has a higher rate of associated shoulder pain and ­stiffness.214

Criteria for Return to Play After 12 weeks, the patients should resume regular activities; return to manual labor and upper extremity–dominant sports can be resumed after 4 months if full strength, range of motion, and radiographic signs of osseous union are established. Preventive measures include emphasis on proper throwing mechanics with the elbow in full extension during stride foot contact.163 Well-conditioned professional athletes develop protective mechanisms to improve their torque threshold. Adaptations in the form of increased external

Arm 1183

rotation and shoulder range of motion, cortical hypertrophy of the humeral shaft, and increased axial ­ preload on the humeral shaft through well-coordinated biceps and triceps brachii contraction227 have been well described in the literature.163,228,229

Special Populations Because of the steadily increasing popularity of snowboarding, winter sports–related injuries have significantly increased during the past decade. Snowboarders are 6 times more likely to get injured, experience more than twice as many fractures, and sustain more severe injuries when compared with alpine skiers. Injuries usually result from a fall or a collision, and 77% of all snowboarding injuries involve the upper extremity. Humerus fractures are the second most common fracture (11%) in a large prospective study230 and most commonly involve the proximal and distal periarticular regions. In 90% of cases, injuries occur to novices and are more likely to occur without prior formal instruction and use of protective equipment.230-233

Box 18-7  Differential Diagnosis of Medial Epicondyle Avulsion

• Symptomatic

supracondylar process with or without fracture • Lateral or medial humeral epicondylitis • Osteochondral injury • Biceps or brachialis tendinopathy • Axillary artery aneurysm260

Imaging Plain radiographs in two planes are sufficient; computed tomographic scans are recommended to rule out intraarticular extension through the medial condyle or incarcerated fragments when associated with traumatic elbow dislocation.

Treatment Options

Medial Epicondyle Avulsion

Nonoperative

Relevant Anatomy and Biomechanics

Nonoperative treatment should be considered for nondominant extremities, nonathletes, and minimally ­displaced fractures.238

The term little leaguer’s elbow has been used for several different injury entities predominant not only in baseball but also in football, tennis, and gymnastics. Medial epicondyle and chondral involvement of the capitellum are most frequently observed. These injuries should be addressed separately with precise terminology. The medial epicondyle is the point of origin for the flexor-pronator muscle group and the medial collateral ligament. The anterior band of the medial collateral ligament serves as the primary stabilizer of the elbow joint. Of the distal humerus ossification centers, the medial epicondyle is the last to fuse after age 15 years. Fractures at this site occur during adolescence and are associated with posterolateral elbow dislocations (50%) and intra-­articular fragment incarceration (15%). Similar to the spiral humeral shaft fractures in adult wrestlers, adolescent wrestlers (aged 13 to 16 years) appear predisposed to sustain an avulsion injury through the medial epicondylar growth plate (Salter-Harris I) where the distal humerus is weakest just before physeal closure.209,234-237

Evaluation Clinical Presentation and History Patients usually present acutely with a classic injury mechanism or after a fall on the outstretched hand (Box 18-7).

Physical Examination and Testing Findings are severe tenderness to palpation and swelling over the medial epicondyle, with pain and weakness most noticeable on wrist flexion. Ulnar nerve function should be carefully assessed and documented.

Operative Relative operative indications include highly competitive athletes with displacement greater than 2 to 5 mm, valgus instability of the elbow, or progressive ulnar nerve palsy. An absolute indication is an intra-articular fragment entrapment after elbow dislocation. Rigid open reduction and internal fixation can be achieved with cannulated screws (3.5 or 4.5 mm) to allow for early mobilization. Washer augmentation is reserved for small or comminuted fragments. Kirschner wires are prone to failure with early postoperative range of motion exercises and should therefore be avoided. Fragment excision and ligamentous repair in severely comminuted fractures may be necessary but should be avoided when possible (Box 18-8; Figs. 18-14 and 18-15).

Box 18-8  Treatment Indications: Medial Epicondyle Avulsion Nonoperative Nondominant extremity Elbow stable Low-demand patient Operative Progressive ulnar nerve deficit Gravity valgus test positive High-demand athlete Intra-articular fragment

�rthopaedic ����������� S �ports ������ � Medicine ������� 1184 DeLee & Drez’s� O

A B

D

C

E

F

Figure 18-14   A 15-year-old patient after fall on outstretched hand with posterolateral elbow dislocation and medial epicondyle avulsion fracture. A, Lateral view with intra-articular medial epicondyle fragment. B and C, Lateral and anteroposterior views after reduction with incarcerated fragment. D, Sagittal computed tomographic image. E and F, Lateral and anteroposterior views 4 weeks after open reduction and internal fixation with single cannulated 3.5-mm screw and washer.

Arm 1185

A

B

C

Figure 18-15   A 17-year-old athlete sustained a displaced medial epicondyle avulsion to his left, nondominant, arm. A, Anteroposterior view at the time of injury. B and C, Anteroposterior and lateral views 2 months after injury with nonoperative treatment. The patient had regained full active and passive range of motion and was pain free.

Weighing the Evidence In a series of 10 wrestlers (aged 13 to 15 years) with medial epicondylar avulsion, Ogawa and Ui236 found that all injuries were sustained during an end-of-the-match power surge and the subsequent counterattack both as the attacking and defending opponent. The patients had positioned themselves with an unstable center of gravity, and the authors concluded that the combination of a mechanically weak epiphyseal plate with extreme traction force generated by a sudden shift from maximal concentric to eccentric contraction causes the injury. Similar mechanisms were postulated in other series.209,239,240 Both Moon and colleagues209 and Ogawa and Ui236 reported on arm-wrestling injuries that were treated operatively in 80% of cases and showed excellent short- and long-term outcomes for both groups. Operative fixation with screws and Kirschner wires yielded bony union in all cases without functional deficits. Ogawa observed one case of ulnar nerve palsy with residual mild discomfort but no distal muscle atrophy or sensory changes at the 8-year follow-up. Case and Hennrikus238 reported on 8 fractures (patients aged 9 to 15 years) after a fall on an outstretched hand; half occurred in combination with posterolateral elbow dislocations. All patients underwent open reduction with cannulated screw and washer fixation. The ulnar nerve was identified in all patients. All patients healed with osseous union after 4 to 8 weeks, and all regained full, pain-free elbow range of motion and returned to preinjury level athletic activity without valgus instability. Two studies239,240 reported on nonoperative treatment of 5 and 8 patients, respectively, with arm-wrestling ­injuries

(aged 13 to 39 years). All patients were immobilized for 1 to 2 weeks in a posterior splint or cuff-and-collar and were then gradually started on a mobilization regimen over the next 3 weeks. One year after injury, all but one patient demonstrated full functional recovery. Asymptomatic nonunion on radiographs was noticed in 11 of 13 cases. On long-term follow-up (average, 45 years) of 42 pediatric patients, Farsetti and colleagues241 found equally good functional results with stable elbow joints and full, pain-free range of motion with or without surgical fixation. Asymptomatic fibrous nonunion was a common finding (89%) in the nonoperative group, whereas bony union was achieved in all surgically treated patients. Excision of fragments led to long-term sequelae such as chronic pain, ulnar nerve paraesthesia, joint instability, decreased grip strength, and joint stiffness.

A u t h o r ’ s P r efe r r e d M e t h o d Most patients can be treated nonoperatively and can expect a good functional result even with development of a fibrous nonunion. Operative fixation should be considered in high-demand athletes with involvement of the dominant upper extremity (for throwing activities) because residual stiffness and instability can potentially jeopardize a career or ­ collegiate opportunity. We prefer the use of 3.5- or 4.5-mm cannulated screws according to patient and ­fragment size. Washers are used for comminuted and small fragments only because they tend to increase hardware prominence. Absolute stability must be achieved to allow early range of motion in the first postoperative week.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1186 DeLee & Drez’s� O

A

B

Figure 18-16   A and B, Supracondylar process (white arrow) and ligament of Struthers (black arrows) causing compression entrapment of the median nerve.

Postoperative Management and Rehabilitation Postoperative care consists of temporary immobilization in a posterior splint at 90 degrees of flexion over 3 days until the first dressing change. Full active range of motion is then resumed in a protective elbow brace for 4 weeks with avoidance of varus and valgus stress. Out-of-brace range of motion exercises for an additional 4 weeks are then ­initiated.238

Potential Complications Stiffness is the most common complication. Early institution of range of motion exercises is mandatory. Nonunion occurs frequently with nonoperative treatment but remains asymptomatic in most cases.239,241 Ulnar nerve palsy can occur at the time of injury or postoperatively and will usually resolve spontaneously. Symptomatic fragments causing localized tenderness or ulnar nerve compression may be excised. Refracture has been reported and should be avoided by gradual return to athletic activity.209

Criteria for Return to Play Return to noncontact and non–weight-bearing activity occurs after 8 weeks, and full return after 3 months.

Special Populations Repetitive overhead throwing in baseball places a high tension and valgus overload on the medial restraints during late cocking phase and can lead to medial epicondyle

fragmentation and avulsion. In gymnastics, the elbow becomes a weight-bearing joint exposed to large load forces, which can lead to medial epicondyle traction or avulsion injuries.

Supracondylar Process Fracture The supracondylar process of the humerus presents as an abnormal, downward-pointing exostosis over the anteromedial distal humeral cortex. It is present in 1% to 3% of the population and is located 5 to 7 cm proximal to the medial epicondyle, with a fibrous band frequently connecting the two structures. Abnormal insertions of the coracobrachialis and pronator teres muscles, along with median nerve and brachial artery compression syndromes, have been described.242 Detection may require oblique radiographs, and the process is usually easily palpated and very tender with deep palpation after acute injury. Doane243 reported an acute fracture of the process in a 17-year-old football player (Fig. 18-16). Kolb and Moore244 reported a fracture through a supracondylar process sustained during a wrestling match in a 12-year-old. Treatment is generally nonoperative. Rest, elevation, ice, and anti-inflammatory medications should ­ provide ­sufficient pain relief. Persistent pain upon return to ­athletics or neurovascular compromise is an indication for excision. The patient should be aware of the risk for myositis ossificans and spur recurrence.245

Arm 1187

Rib Fractures in Athletes

Criteria for Return to Play

Relevant Anatomy and Biomechanics

Return to play should be gradual and guided by the absence of pain. Discipline-specific conditioning and endurance exercises should be integrated into the rehabilitation program. Correct technique should be ­emphasized.

Rib fractures in athletes have been reported in numerous, mostly upper extremity–oriented, sports, most notably baseball, rowing, and golfing.246-252 Muscular forces exerted during upper extremity and trunk rotation and bending are responsible for these repetitive fatigue fractures. Injuries are most likely to occur in the training period when prolonged practice sessions and accelerated muscle development overcome the skeleton’s ability to remodel and adapt.253,254 The anterolateral 1st rib and the posterolateral 4th through 9th ribs are the most common locations for stress-induced fractures. First-rib fractures occur predominantly with repetitive overhead activities near the subclavian groove in between the insertions of the anterior and medius scalene muscles, which are counteracted by the downward pull of the serratus anterior and intercostal musculature.255,256 Mid and lower thorax rib fractures occur with rowing and swinging activities such as golf and tennis. Opposing forces of serratus anterior (lateral and superior pull) and external oblique muscles (medial and inferior pull) can cause progressive microfractures during long rowing strokes. Increased demands on the leading-arm side serratus anterior muscle are thought to be responsible for golfrelated stress fractures.251,257,258

Evaluation Clinical Presentation and History Patients will present with a gradual onset of dull pain localized over the clavicle, with occasional radiation toward the sternum or chest.

Physical Examination and Testing Localized tenderness to palpation, painful coughing, and activity-related pain are nonspecific findings. Neurovascular compromise has not been reported.

Imaging Plain films may be negative in recent injuries and should be followed-up with technetium bone scanning. Healing fractures may present with callus formation.

Treatment Options Treatment is generally nonoperative and aimed at ­symptomatic pain relief through temporary cessation and eventual reduction of inciting activities over 4 to 6 weeks.

Potential Complications Nonunion of a 1st rib fracture has been reported in baseball pitchers but tends to be asymptomatic.256,259

C

r i t i c a l

P

o i n t s

l Pectoralis major rupture most commonly occurs in male weightlifters during bench press; the diagnosis is primarily based on history and physical examination; MRI is useful to verify location of the tear; early surgical repair is indicated in active patients and outcomes of repair are significantly better than after nonoperative treatment. l Disease of the long head of the biceps tendon accompanies impingement syndrome pathology of the shoulder; rupture is often pain-relieving; there are minimal clinical consequences to untreated rupture in most patients; tenodesis should be considered in younger athletic patients and multiple new techniques are presently available for less invasive tenodesis. l Distal biceps rupture occurs in the dominant arm of males 40 to 60 years old during eccentric loading of the flexed arm; acute diagnosis is based on history and physical examination alone in most cases; MRI is a useful adjuvant; early repair yields improved results and decreased complications; single- and double-incision techniques both provide acceptable results; there is significant loss of supination strength without anatomic repair. l Distal triceps tendon rupture is rare but can easily be misdiagnosed; partial tears should be treated nonoperatively initially; complete tears are best repaired with early bone tunnel fixation; multiple late reconstruction techniques are available; the physician should look for associated fractures; good results with early surgical treatment can be anticipated. l Proximal humeral epiphysiolysis warrants a high level of suspicion in high-risk, overhead-activity athletes with increasing shoulder pain; early detection and diagnosis are accomplished with MRI; prevention is through preseason conditioning, with an emphasis on technique development with predetermined limits on pitch count and frequency. l Avulsion fracture of the lesser tuberosity is generally diagnosed with axillary lateral and anteroposterior radiographs in internal rotation; CT is used for questionable fractures; preoperative planning is done with MRI. l Humeral shaft fracture warrants a high level of suspicion in throwing athletes with activity-related arm pain; MRI is used for early detection of stress fractures; radial nerve function is evaluated with dorsal wrist extension; ­operative versus nonoperative therapy is decided on an individual basis. l For medial epicondyle avulsion injuries, early range of motion exercises with both operative and nonoperative treatment are used to prevent elbow stiffness.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1188 DeLee & Drez’s� O

S U G G E S T E D

R E A D I N G S

Blaine TA, Bigliani LU, Levine WN: In Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB (eds): The Shoulder. Philadelphia, WB Saunders, 2004, pp 355-412. Boyd HB, Anderson LD: A method for reinsertion of the distal biceps brachii tendon. J Bone Joint Surg Am 43:1041, 1961. Chen FS, Diaz VA, Loebenberg M, Rosen JE: Shoulder and elbow injuries in the skeletally immature athlete. J Am Acad Orthop Surg 13:172-185, 2005. Hoppenfeld S, DeBoer P: Surgical Exposures in Orthopaedics: The Anatomic ­Approach. Philadelphia, Lippincott Williams & Wilkins, 2003, pp 67-104. Kubiak EN, Koval KJ, Zuckerman JD: Chapter 2, Anatomy of the shoulder, pp 23-33; Leibmann MI, Zuckerman JD: Chapter 3, Proximal humeral fractures: Clinical evaluation and classification, pp. 34-45. Liporace FA, Koval KJ: Chapter 5, Two-part fractures and fracture dislocations, pp 58-85. In Zuckerman JD, Koval KJ: Shoulder Fractures: The Practical Guide to Management. New York, Thieme, 2005.

Sarwark JF, King EC, Luhmann SJ: In Beaty JH, Kasser JR (eds): Fractures in Children. Philadelphia, Lippincott Williams & Wilkins, 2006, pp 703-771. Templeman DC, Sems SA: In Stannard JP, Schmidt AH, Kregor PJ (eds): Surgical Treatment of Orthopaedic Trauma. New York, Thieme, 2006, pp 263-284. Warner JJP, Costouros JG, Gerber C: In Bucholz RW, Heckman JD, CourtBrown C: Fractures in Adults. Philadelphia, Lippincott Williams & Wilkins, 2006, pp 1161-1209. Wilkins KE, Morrey BF, Jobe FW, et al: The elbow. Instr Course Lect 40:1-87, 1991. Zlotolow DA, Catalano LW 3rd, Barron OA, Glickel SZ: Surgical exposures of the humerus. J Am Acad Orthop Surg 14:754-765, 2006.

R eferences Please see www.expertconsult.com

C H A P T E R  

 19

Elbow and Forearm S ect i o n

A

Biomechanics of the Elbow and Forearm Bernard F. Morrey

A basic understanding of joint biomechanics is critical to understanding both the nature of pathology and treatment rationale. This chapter discusses the biomechanics of the elbow according to joint motion, stability, and forcestrength. The biomechanical features of the forearm are discussed separately.

JOINT SURFACES The elbow is a trochoginglymoid joint; that is, it possesses 2 degrees of freedom: flexion-extension and forearm rotation.1 Certain special features of this articulation account for the motion and stability attributed to it. The joint surfaces of the humerus, radius, and ulna have a high degree of compliance and specific, discrete orientations with regard to the long axis of each. Recognition of these orientations is especially important for proper treatment of some fractures around the elbow.

Ulna In the anterior plane, the articular surface of the ulna makes about a 5- to 7- degree valgus angulation with regard to the long axis of the shaft. Laterally, the articular surface is oriented about 30 degrees posteriorly referable to the long axis, complementing the anterior rotation of the distal humeral articulation (Fig. 19A-1). This relationship allows the elbow to be stable when completely extended, an important function for some athletic activities.

Articular Surface The greater sigmoid fossa (often referred to as the olecranon) consists of the coronoid process distally and the olecranon process proximally. The arc of curvature of this

Humerus From the lateral projection, the articulation surface of the humerus is rotated anteriorly about 30 degrees in reference to the long axis of the humerus. Although there is considerable individual variation, a mean of 5 to 7 degrees of valgus tilt of the axis of rotation is thought to be normal. Viewed end-on, the distal humeral articulation is also externally rotated about 3 to 5 degrees in reference to the plane of the posterior surface of the medial and lateral columns.2

Articular Surface The articulate surface of the distal humerus is composed of a trochlea containing medial and lateral contours. Cartilage covers an arc of about 300 degrees. The capitellum is an almost perfect geometric hemisphere oriented anteriorly with an arc of curvature of about 180 degrees in both medial lateral and proximal distal meridians. A fixation plate can thus be contoured and applied to the posteroinferior aspect of the lateral column without violating the articular surface of the capitellum.3

Figure 19A-1   The 30-degree anterior rotation of the distal humerus is matched by the 30-degree opening of the greater sigmoid notch, providing stability of the elbow joint in flexion and in full extension.

1189

�rthopaedic ����������� S �ports ������ � Medicine ������� 1190 DeLee & Drez’s� O

articulation is about 185 degrees. The articular surface is thin, measuring 2 to 3 mm. An important anatomic feature is the lack of articular cartilage at the center of the greater sigmoid fossa.2 Hence, the contact area consists of the anterior coronoid and posterior olecranon surfaces. When investigating the ulnohumeral joint arthroscopically, one must take care not to interpret this normal variation as articular disease. Exposure by means of osteotomy should also be directed through this portion of the olecranon. A second articulation is present on the lateral side of the greater sigmoid fossa, the lesser sigmoid notch. With an arc of curvature of about 70 degrees, it accommodates the radial head.

Radial Head The radial neck makes an angle of approximately 15 degrees away from the long axis measured at the radial tuberosity (Fig. 19A-2). This angular relationship allows the forearm to undergo an arc of forearm rotation of about 180 degrees while maintaining a precise and constant orientation in regard to the capitellum. The slightest abnormality or alteration of this angle markedly alters forearm rotation.

Articular Surface The articular surface of the radial head is composed of a disk-shaped impression with an arc of curvature of 40 degrees.2 The margin of the radial head is covered by hyaline cartilage through an arc of about 240 degrees and articulates with the lesser sigmoid fossa. The remaining 120 degrees constitutes the nonarticular portion of the radial head. This portion is weaker because there is no subchondral bone, which may account for the ­predilection for shear fractures to occur through this region.

JOINT MOTION The three-dimensional motion of the elbow joint has been carefully studied.1,4 The carrying angle undergoes a linear change from valgus to varus from extension to flexion (Fig. 19A-3). Furthermore, there is a slight axial rotation of the ulna internally during the initiation of extension and then externally at the completion of extension as a result of the geometric characteristics of the ulnohumeral joint. The subtle axial rotation of the ulna may contribute to the osteophyte that develops on the radial tip of the olecranon in the throwing athlete.5

Axis of Rotation The locus of the instant center of rotation of the elbow is less than 3 mm in its widest dimension.6 Hence, elbow flexion may be considered primarily a spinning motion for all practical purposes. The hinge axis may thus be approximated by a line that pierces the lateral projection of the center of the capitellum and the center of the trochlea (Fig. 19A-4). The axis is located medially at the anteroinferior aspect of the medial epicondyle.

Axis of Forearm Rotation The axis of forearm rotation passes directly through the center of the radial head and thus through the center of the capitellum in both anteroposterior and lateral projections.7 This axis then traverses the interosseous membrane and emerges through the base of the styloid process of the ulna, at the center of curvature of the distal ulna.

Normal Elbow Motion The normal arc of elbow flexion is 0 to 145 degrees, with considerable individual variation.8 Lax-jointed individuals may hyperextend 10 or more degrees, and well-muscled individuals may flex only 130 degrees. Pronation and supination arcs are usually symmetrical. Pronation averages about 80 degrees, which is about 5 to 10 degrees less than supination, which averages about 85 degrees. Normal average forearm rotation thus is not typically a full arc of 180 degrees but about 165 to 170 degrees.8 Calculation of flexion-extension is one of the easiest joint measurements in the body. Hand-held goniometers are accurate to within 5 degrees.

Carrying Angle

Figure 19A-2   The radial neck makes a 15-degree angle referable to the proximal radial shaft in the direction away from the radial tuberosity.

The normal valgus orientation of the forearm varies as a function of both age and sex. It is less in children than in adults and averages 3 to 4 degrees more in females than in males.9,10 The carrying angle is formed by the valgus tilt of the axis of rotation (humeral articulation) and the valgus orientation of the ulnar shaft in reference to the olecranon.11 The “normal” angle varies greatly and averages about 10 degrees in the male and 13 degrees in the female. The proper definition of the carrying angle is the orientation of the forearm in reference to the humerus when the elbow is in full extension. The concept loses its

Elbow and Forearm 1191

Figure 19A-3   A, During elbow flexion, the carrying angle changes in a linear fashion from valgus to neutral or varus. B, The ulna undergoes a slight axial rotation during flexion and extension. (Used with permission of the Mayo Foundation.)

Figure 19A-4   The axis of rotation of the elbow is approximated by a line running through the middle of the lateral epicondyle and the center of the trochlea, emerging on the anterior inferior aspect of the medial epicondyle.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1192 DeLee & Drez’s� O Figure 19A-5   With serial removal of the olecranon, a linear decrease in ulnohumeral stability is observed. Note the 50% loss with 50% resected. (Used with permission of the Mayo Foundation.)

significance as the elbow flexes.12 With throwing sports, the carrying angle may increase 2 to 3 degrees. This occurs from osseous deformation rather than ligament laxity.5

Functional Motion As with many joints, the full available arc of motion is not required for routine daily activities. Our study has shown that activities involving personal hygiene and daily sustenance are accomplished with an arc of 30 to 130 degrees of flexion and with 50 degrees of pronation and 50 degrees of supination.13 For the athlete, functional motion requirements for various sports have not been addressed. It is known, however, that the thrower frequently develops a flexion contracture of up to 10 degrees that does not impair function.5 Greater motion might be anticipated during some athletic ­activities, such as gymnastics. Less motion, however, is associated with increased muscle mass, as might be seen with body builders.

ELBOW STABILITY Stability is achieved through static and dynamic components. Two elements contribute to static stability of any joint: the articular surface, and the capsule and ligamentous structures. Dynamic stability is achieved with the muscles that cross the joint. Unlike the shoulder, the dynamic ­contribution of the musculature to elbow ­stability under normal circumstances is minimal.14 Hence, the elbow relies on the contributions of the static constraints: the articular surface and the capsular ligamentous ­complex.15 Rehabilitation programs have little value in the unstable elbow.

Articular Contribution to Stability Olecranon Ignoring the dynamic contribution, resection of ­successive portions of the ulnar articulation decreases the stability of the joint in proportion to the amount that has been removed.16 Hence, a 50% resection of the olecranon causes a 50% reduction in ulnohumeral stability (Fig. 19A-5).

Coronoid Fracture of the tip of the coronoid does not compromise articular stability but indicates that the elbow has ­undergone dislocation or subluxation. The critical amount of coronoid needed for stability is 50%. This amount may be estimated by a line drawn from the tip of the intact olecranon through the remaining coronoid. If the line converges with the long axis of the ulna distally, the articular stability of the ulnohumeral joint has been lost. If the line converges proximally, the stabilizing effect of the coronoid is probably intact if the radial head is present (Fig. 19A-6). If the radial head is absent, the elbow is markedly more unstable in the absence of less than 50% of the coronoid (Fig. 19A-7). Hence, a fracture­dislocation involving the coronoid and radial head is a serious injury and typically ends sports participation for the athlete.

Radial Head Studies of the contribution of the radial head to elbow ­stability reveal that about 20% to 30% of an applied valgus load is transmitted through, and thus resisted by, the radiohumeral joint.17-20 Yet the role of the radial head in resisting valgus stress cannot be considered without accounting for the associated ligamentous structures.

Elbow and Forearm 1193 50% coronoid removed

Elbow stability

120

Figure 19A-6   Absence of 50% of the coronoid is estimated by drawing a line from the olecranon trip to the coronoid release. The line is parallel to the shaft at the 50% level.

With simulated flexion and extension, the radial head does not resist ­physiologic valgus stresses in the presence of an intact medial collateral ligament.21 If the medial collateral ligament has been removed or violated, however, the radial head plays a major role in resisting the valgus torque. If both the medial collateral ligament and the radial head are removed, the elbow is predictably dislocated (Fig. 19A-8). Interpretation of these data is easier when considering the contributions of the meniscus and the anterior cruciate ligament to knee stability. The radial head has a role similar to that of the meniscus, and the role of the medial collateral ligament is analogous to that of the anterior cruciate ligament. The radial head may be considered a secondary stabilizer in valgus elbow instability. Its contribution is seen only if the medial collateral ligament complex is deficient.

Capsuloligamentous Complex Medial Collateral Ligament Traditionally, the role of the medial collateral ligament in resisting valgus stability has been well described.22 Acute rupture of the medial complex may have been shown to

Radial Head as Secondary Stabilizer to Valgus Stress 1

2

Stability

1

RH MCL

2

MCL RH

Figure 19A-7   Left, Removing the radial head (RH) (1), but leaving the medial collateral ligament (MCL) intact (2), results in no alteration in valgus stability. Right, When the medial collateral ligament is removed first (1), the radial head is observed to provide some resistance to valgus stress. When both constraints are released (2), the elbow is grossly unstable. (Used with permission of the Mayo Foundation.)

90

60

R head present

R head absent

30

0 0

15

30

45 60 75 Elbow flexion

90

105

120

Figure 19A-8   With 50% of the coronoid resected, the elbow may be stable with, but not without, an intact radial head.

be effectively addressed by reconstructing or repairing the anterior bundle of this complex. The anterior element of the medial collateral ligament originates at the site of the axis of rotation for the elbow (Fig. 19A-9). This bundle is taut throughout the arc of motion; the anterior fibers are most taut in extension, and the posterior bundles become tightened in flexion. Hence, the role of the anterior portion of the medial collateral ligament is not unlike that of the anterior cruciate ligament at the knee. These experiments specifically show that the anterior bundle is the essential component of the medial complex. Given this fact, precise restoration of the humeral origin of the ligament must be attained with ligamentous reconstruction procedures.23 The posterior bundle originates off the axis. Hence, a cam effect is present, and the posterior bundle is taut only in flexion. This structure is now recognized to be contracted in those with contractures limiting flexion and may need to be surgically released.

Lateral Collateral Ligament Complex The lateral ligament complex is becoming better ­understood. A discrete portion of the lateral collateral ­ligament complex is termed the lateral ulnar collateral ligament (LUCL)22 and should be distinguished from the radial collateral ligament, which attaches to the annular ligament (Fig. 19A-10). The LUCL originates from the lateral tubercle at the center of rotation and inserts on a small tubercle in the ulna at the crest of the supinator. The existence of this particular structure explains the maintenance of varus stability after the radial head has been excised. Because the radial collateral ligament becomes lax as the annular ligament loses its tension, resistance to varus stress is provided by the LUCL. The most appropriate concept of collateral ligament stability of the elbow is that of a medial collateral ligament and LUCL complex that functions independently in the presence or the absence of a radial head (Fig. 19A-11). This view is consistent with the anatomic findings as well as with clinical experience. Deficiencies of the LUCL result in posterior lateral rotatory instability of the elbow.24 This condition most frequently occurs after elbow dislocation or from release and inadequate reconstitution after surgical procedures involving this structure.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1194 DeLee & Drez’s� O

A

B Figure 19A-9   A, Anatomic distribution of the components of the medial collateral ligament complex (MCL). B, The anterior bundle originates at the locus of the axis of rotation. However, the origin of the posterior bundle does not lie along the axis of rotation. Thus, some change in length of the posterior band is seen with changes in elbow flexion angle. (Used with permission of the Mayo Foundation.)

Relative Contribution of the Collateral Ligaments Experimental capsule and articulation data may be summarized by stating that the articular surfaces provide about 50% of elbow stability, and the collateral ligaments provide the remaining 50%. When the elbow is in full extension, the anterior capsule contributes about 15% of the resistance to varus-valgus stress.18

Forces across the Elbow Joint Accurate determination of forces across the joint is difficult owing to the numerous muscles and possible joint orientations (Fig. 19A-12). Nonetheless, a rather accurate estimation of the resultant forces across the joint has been calculated.

Muscle Contribution Calculation of muscle force is based on three factors: crosssectional area, orientation or line of action, and specific activity during a given function.3,25-27 This information allows one to estimate the resultant force at the elbow in certain positions. The greatest amount of force generated at the elbow occurs with the initiation of flexion. During flexion, the moment arm of the muscles increases so that less force of

contraction is required. The resultant vector with greatest magnitude is directed axially toward the humeral head. Greater strength of flexion is generated with the elbow in 90 degrees of flexion, however, because the mechanical advantage of the elbow flexors has improved from the fully extended position so that less force is actually applied to the joint. The actual force across the elbow is thus less in this position than in full extension, and the resultant force is deviated posteriorly and superiorly. Calculations suggest that about 3 times the body weight may be transmitted across the elbow joint when it is flexed at 90 degrees.28,29 It is important to note that the resultant vector undergoes a change in direction depending on whether the elbow is actively flexing or extending (Fig. 19A-13).30

Posteriorly Directed Forces One interesting feature of particular note at the elbow is that a posteriorly directed force component originates from the line of action of both agonist and antagonist muscles: biceps, brachialis, and triceps. Hence, for this specific joint, there is a tendency for posterior displacement as the elbow flexes and as it extends (Fig. 19A-14). These large forces are responsible for the frequently observed loss of fixation of distal humeral fractures and for the tendency toward posterior displacement of the joints after severe ligament and articular injury.

Elbow and Forearm 1195

A

B Figure 19A-10   A,The lateral collateral ligament complex consists not only of the radial collateral ligament (RCL) but also a lateral ulnar collateral ligament. B, With flexion, there is no change in length of the radial collateral ligament complex. This suggests that the origin is at the axis of rotation. (Used with permission of the Mayo Foundation.)

Ulnohumeral and Radiohumeral Forces Using a physiologic model of similarly active joint loading, the greatest amount of force across the radiohumeral joint was shown to occur in extension.30 The maximal amount of force across the radiohumeral articulation alone, however, approaches but does not exceed body weight.18 With the

elbow in full extension, about 40% of the applied forced is transmitted across the radial head.31 Release of the interosseous membrane did not affect the measurements (Fig. 19A-15).13 Finally, with varus-valgus stress, the load pivots medially and laterally at a point located in the center of the lateral face of the trochlea.29

Figure 19A-11   The ulnohumeral joint is stabilized by both medial and lateral constraints, independent of the presence of the radial head. LUCL, lateral ulnar collateral ligament; MUCL, medial ulnar collateral ligament. (Used with permission of the Mayo Foundation.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1196 DeLee & Drez’s� O

BR

BC

TR

Resultant force Figure 19A-12   Cross-sectional area showing the size and distribution and thus the moment arms of the muscles that cross the elbow joint. BIC, biceps; BRA, brachioradialis; PRO, pronator teres; FCR, flexor carpi radialis; FDS, flexor digitorum sublimis; FCU, flexor carpi ulnaris; TRI, triceps; ANC, anconeus; ECU, extensor carpi ulnaris; EDC, extensor digitorum communis; ECR, extensor carpi radialis; BRD, brachialis. (From Funk DA, An KN, Morrey BF, Daube JR: An EMG analysis of muscles controlling elbow motion. J Orthop Res 5:529-538, 1989.)

Figure 19A-13   With flexion and extension, the resultant force of the elbow undergoes a cyclic change in direction. The order of magnitude can be as much as three times the body weight. (Used with permission of the Mayo Foundation.)

Figure 19A-14   The posteriorly directed force component of the triceps (TR), the biceps (BC), and the brachialis (BR) all tend to displace the forearm posteriorly. (Used with permission of the Mayo Foundation.)

Figure 19A-15   The distributive forces with the elbow in full extension and an axial load placed at the wrist show that about 60% of the force goes across the radiohumeral joint and 40% goes across the ulnohumeral joint. This does not change the section of the interosseous membrane. (Used with permission of the Mayo Foundation.)

Elbow and Forearm 1197

C

r i t i c a l

P

S U G G E S T E D

o i n t s

l Understanding the relevance of biomechanics is essential to clinical practice. l The lateral collateral ligament has a role in stabilizing the elbow and in preventing recurrent instability. l The interaction of the stabilizing elements of the articular components is increasingly appreciated, specifically the role of the coronoid in transmitting force and stabilizing the joint. l The elbow tends to be unstable posteriorly because of the posterior-directed forces that occur at the elbow with both flexion and extension.

R E A D I N G S

An KN, Hui FC, Morrey BF, et al: Muscles across the elbow joint: A biomechanical analysis. J Biomech 14:659-669, 1981. Boone DC, Azen SP: Normal range of motion of joints in male subjects. J Bone Joint Surg 61A:756, 1979. Ishizuki M: Functional anatomy of the elbow joint and three-dimensional quantitative motion analysis of the elbow joint. J Jap Orthop Assn 53:989-996, 1979. London JT: Kinematics of the elbow. J Bone Joint Surg Am 63:529-535, 1981. Morrey BF, An KN: Functional anatomy of the ligaments of the elbow. Clin Orthop 201:84-90, 1985. Morrey BF, Chao EYS: Passive motion of the elbow joint. J Bone Joint Surg Am 59:501-508, 1976.

R eferences Please see www.expertconsult.com

S ect i o n

B

Tendinopathies around the Elbow William D. Regan, Philippe P. Grondin, and Bernard F. Morrey

LATERAL EPICONDYLITIS (TENNIS ELBOW) What we refer to today as tennis elbow was originally described in relationship to lawn tennis by Major in 1883.1,2 Today, however, the condition of lateral epicondylitis is well known to occur spontaneously or in association with other recreational and occupational pursuits.

Demographics Although 95% of reported cases occur in individuals other than tennis players,3,4 it is estimated that 10% to 50% of people who regularly play tennis experience the symptoms of tennis elbow in varying degrees some time during their tennis lives.5,6 Shiri and colleagues7 found that 1.3% of people aged 30 to 64 years suffered from lateral epicondylitis in a Finnish study involving 4783 subjects. Men and women were equally affected, and the peak incidence occurred in the 45- to 54-year age group. An analysis of 2500 patients at the Vic Braden Tennis Camp revealed a 50% incidence of tennis elbow.6 Another study involving 200 tennis players revealed an incidence of 50% in players older than 30 years experiencing symptoms characteristic of tennis elbow for less than 6 months; the remaining 50% had major symptoms of an average duration of 2.5 years. This malady can affect participants of any sport involving the use of the upper extremity. The incidence of tennis elbow is equal among men and women, but among

tennis players, it is more common in men.3 It is rare in blacks, as evidenced by a series of 1000 patients in the southern United States in an area of equal racial distribution in which the occurrence was limited to whites only.3 Although frequently seen in the age group spanning the fourth to sixth decades, it occurs 4 times more commonly in the fifth decade, with a peak incidence at age 42 years.3,4 It involves the lateral epicondyle about 7 times more frequently than the medial epicondyle. Lateral epicondylitis occurs most often, however, in the workforce when repetitive elbow and wrist motion is involved. In a 31-month study, Kurppa and associates8 found the annual incidence in a meat processing factory housing 377 employees to be 1% for those working in nonstrenuous jobs and 7% to 11% for those working in strenuous jobs.

Macroscopic Pathology Some variation exists in the precise gross pathologic anatomy of this condition. Local inflammation,9,10 muscle9,11 or ligament strain,9,12 synovial fringe inflammation,10,13-15 tendoperiosteal tears,2,5,12,16,17 disturbance in local metabolism,18 degenerative changes in the annular ligament,19,20 and cervical root irritations21 have been held responsible for the symptoms. The studies of Coonrad and Hooper16 confirmed earlier work by Goldie,10 who demonstrated the essential lesion in tennis elbow to be a degenerative tear of the common extensor or flexor origin at or near the

1198 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

respective lateral or medial epicondyle. It was believed that these tears were produced in a degenerating tendon fiber by mechanical overload in sports or at work. Tendon tears were found to range from microscopic to macroscopic. The origins of the extensor carpi radialis brevis (ECRB) and the superficial part of the supinator are blended and inseparable.22 Both originate from the lateral epicondyle, the elbow joint capsule, and the annular ligament. Furthermore, the extensor carpi radialis longus (ECRL) arises from the lateral epicondyle as well as more proximally along the lateral epicondylar ridge. Finally, the extensor digitorum communis (EDC) takes origin in part from the lateral epicondyle and does indeed contribute to this condition. Careful gross inspection of the ECRB tendon reveals a characteristic grayish, gelatinous, and friable immature tissue. In Nirschl and Pettrone’s series of patients with tennis elbow,5 97% of cases demonstrated various degrees of this pathologic tissue. In their experience, a macroscopic tear of the tendinous origin was found in 35% of cases. In Coonrad and Hooper’s work,16 a tear of the extensor or flexor tendon was demonstrated in 28 of 39 patients, or 72%. In the remaining 11 patients, there was no actual tear, but 9 patients demonstrated excessive scar tissue replacement of the tendinous origin. The macroscopic tears can be superficial or deep; when they are deep, the superficial tendon attachment to bone may completely obscure the pathologic process. Hence, a macroscopic tear of the ECRB tendon, with a possible contribution from the EDC, either superficial or deep, can be defined as the most common pathologic anatomy producing the pain of lateral epicondylitis.3,12,16

Whereas Nirschl described the pathologic lesion as vascular granulation termed angiofibroblastic hyperplasia, the lesion is recognized today as a degenerative, necrotic process. A single-blind, randomized study of specimens from control subjects and patients was conducted at the Mayo Clinic to determine this as definitively as possible.23 Two questions were addressed: (1) Are there any consistent microscopic pathologic changes? and (2) Is there an objective means of grading the changes? All 11 patients with an unequivocal clinical diagnosis of refractory lateral epicondylitis experienced relief after surgery. Macroscopic pathologic change at the origin of the ECRB was observed in all, and 40% were also found to have involvement of the common extensor origin.23 Twelve unembalmed and radiologically normal cadaveric specimens were employed as a control. Fifteen microscopic pathologic features reported as characteristic of lateral epicondylitis were identified from the literature. All specimens were reviewed, and each histologic feature was graded in a single, randomized, blinded manner by a single pathologist. The presence of vascular proliferation and hyaline degeneration was statistically higher in the surgical group than in the control subjects (P < .001) (Fig. 19B-1). Patients with calcific debris had preoperatively received 2, 3, and 12 steroid injections. The precise pathologic microscopic features characteristic of lateral epicondylitis in our material did include a significant vascular and fibrous proliferation (P < .001), but there was a definite absence of any inflammatory component. These findings are consistent with the extensor tendon disruption noted by Cyriax1 and later by Coonrad and Hooper,16 who described the process as resembling scar tissue.

Microscopic Pathology

Clinical Evaluation

Today, less controversy exists about the etiology and his­ topathology of the clinical entity of lateral epicondylitis. Most authors have placed the macroscopic pathologic ­process at the ECRB origin.3,5,10,16,17 The microscopic ­features now seem well defined as well.

History

A

The onset of symptoms can be sudden or gradual. On occasion, direct trauma is the clear cause of the process. More often, a history of repetitive flexion-extension or

B

Figure 19B-1   A, Vascular proliferation is a consistent feature of surgically excised tissue. B, Focal hyaline degeneration is also present but has infrequently been recognized in the literature.

Elbow and Forearm 1199

pronation-supination activity and overuse is obtained, but no predisposing event can be determined in many. Among tennis players and athletes performing other racquet sports, the backhand was the most common swing initiating symptoms.6 The incidence of lateral epicondylitis markedly decreases with a superior level of play, suggesting that faulty swing mechanics initiate the process. Nirschl and Pettrone5 found that the backhand was the most common stroke initiating symptoms (81%). The high-quality backhand is characteristically initiated with the forearm in mid-pronation, the front shoulder down, and the trunk leaning forward. In this position, the elbow, forearm, and wrist are in an overall position of anatomic strength, and this allows better stroke mechanics. In Nirschl’s series of world-class players,5 only 13% were affected by lateral ­epicondylitis.

Physical Examination Point tenderness precisely over the origin of the extensor carpi radialis brevis is the single most rewarding finding. The tenderness may be more diffuse in some instances and can even be migrating in others. The pain is often worsened when wrist extension is tested against resistance with the forearm in pronation. The chair test as described by Gardner14 may be helpful. The patient is asked to lift a chair with one hand, which is pronated. Severe pain in the region of the lateral epicondyle confirms the diagnosis. Coonrad3 has described the “coffee cup test,” which involves picking up a full cup of coffee with associated localized pain at the lateral epicondylar origin. He believes that pain after this test is almost pathognomonic of lateral epicondylitis. A useful finding in our hands is the increased pain with wrist extension when the elbow is fully extended.

Imaging Studies Routine anteroposterior and lateral radiographs are usually of little help in the diagnosis. Pomerance, reviewing 294 radiographs of patients diagnosed with lateral ­epicondylitis, found that only 16% had positive findings, the most common being faint calcific deposits along the ECRB tendon occurring in 7%.24 In only 2 cases (0.7%) did radiographs change management. Although ultrasonography and magnetic resonance imaging have reported sensitivities of 64% to 82% and 90% to 100%, respectively, and specificities of 67% to 100% and 83% to 100%, respectively, they are not usually necessary for this clinically based diagnosis.25

Electromyographic Studies Electromyographic studies are of little or no help in the diagnosis of lateral epicondylitis. Even when a concurrent posterior interosseous nerve irritation is present, the electromyogram remains normal.4

Differential Diagnosis The diagnosis is confused in some instances, such as ­distal neurologic entrapment, localized interarticular disease such as osteochondritis dissecans, radiocapitellar osteoarthrosis,

and bone tumors around the elbow. Subtle instability of the radiocapitellar joint may also be considered. Localized pathologic processes within the elbow joint or localized tumor processes are usually ruled out by a careful history and routine anteroposterior, lateral, and oblique radiographs of the elbow. A localized cervical root irritation is best diagnosed by a careful history and examination of the cervical spine for point tenderness and a distal neurologic examination for cervical radiculopathy. Subtle instability of the radiocapitellar joint can also be diagnosed by a history of a previous dislocation of the elbow, a severe varus stress, or previous surgery for release of the common extensor origin. Radial nerve compression may occur in 5% of those with lateral epicondylitis.4 In this condition, the pain is located directly over the point of nerve compression, which is generally 3 or 4 cm distal to the lateral epicondyle. Werner4 has shown that the most common site of entrapment of the posterior interosseous nerve occurs as the nerve courses through the arcade of Frohse. Other common causes of compression in this area include adhesions of or pressure on the nerve from the vessels at the leash of Henry overlying the nerve, constriction from the extensor carpi radialis muscle as it passes anteromedial to the nerve, and the ­supinator muscle itself.

Treatment Options Nonsurgical Treatment Two large prospective randomized controlled trials have shown that patient education and avoiding aggravating activities alone lead to resolution of lateral epicondylitis symptoms in most patients within 6 months to 1 year.26,27 Smidt and colleagues26 compared the results of a wait and see approach to corticosteroid injections (triamcinolone, 10 mg × 1 to 3 injections) to 6 weeks of physiotherapy in a population of 185 patients who had suffered from lateral epicondylitis for a median of 11 weeks (about 2.5 months) before the study. With success defined as symptoms much improved or completely resolved, Smidt and colleagues found that activity modification alone led to success in 32% of patients at 6 weeks, 52% at 3 months, 78% at 6 months, and 83% at 1 year (Fig. 19B-2). Symptom severity during the day (rated on a scale of 0 to 100, with 100 being most severe) decreased from an average of 70 at onset to 37 at 3 months, 23 at 6 months, and 17 at 1 year. Physiotherapy consisting of nine treatments of pulsed ultrasound, deep friction massage, and a home exercise program of stretching and strengthening over 6 weeks led to better results than simple activity modification and education but not by a statistically significant amount. The success rate with physiotherapy was at 91% at 1 year, and pain severity decreased from 70 to 11. Bisset and associates,27 in a population of 198 patients who had symptoms for a median of 22 weeks (about 5 months) before the study, found similar results, with 27% of patients having success defined as symptoms completely resolved or much improved 6 weeks after activity modification and education, 59% at 3 months, and 90% at 1 year. Pain in the preceding week rated on a scale of 0 to 100, with 100 being most severe, decreased from 61 at onset to 30 at 3 months, 20 at 6 months, and

1200 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� 100

Success rate (%)

80 60 40 Wait and see (n�59) Corticosteroid injection (n�62) Physiotherapy (n�64)

20 0

0 3

6

12

26 Weeks

52

Figure 19B-2   Success rates of three treatment regimens. (From Smidt N, van der Windt DA, Assendelft WJ, et al: Corticosteroid injections, physiotherapy, or a wait-and-see policy for lateral epicondylitis: A randomised controlled trial. Lancet 359[9307]:657-662, 2002.)

14 at 1 year. Physiotherapy consisting of eight sessions of elbow manipulations and exercises performed better than the wait and see arm but again not by a statistically significant amount. Success with physiotherapy was 94% at 1 year and pain in the preceding week decreased from 58 at onset to 7 at 1 year. Cortisone injections, acupuncture, and extracorporeal shock-wave therapy (ESWT) have fallen out of favor because multiple prospective randomized trials have shown no benefit over placebo. Corticosteroid injections outperformed physiotherapy and wait and see at 6 weeks, but the benefits were no longer seen at 3 months, and patients did worse than wait and see and physiotherapy at 6 months and 1 year. Smidt and associates found a 92% rate of success with injections at 6 weeks (versus 32% with wait and see), but this decreased to 69% at 1 year (versus 83% with wait and see), as shown in Figure 19B-2. Bisset and colleagues found a 78% rate of success with injections at 6 weeks (versus 27% with wait and see), but this decreased to 45% at 3 months (versus 59% with the wait and see approach) and 68% at 1 year (versus 90% in wait and see). Clearly, cortisone injections provide short-term relief only. Three prospective, randomized, blinded, controlled trials (Fink and associates,28 of 45 patients assessing relief using the disability of the arm, shoulder, and hand [DASH] form; Molsberger and Hille,29 of 48 patients; and Davidson and coworkers,30 of 16 patients, again using DASH to assess outcome) showed acupuncture relief to be superior to placebo at 2 weeks or less but no different to placebo beyond this point (1 and 2 months). Three randomized controlled studies,31-33 favored ESWT over placebo, whereas four trials8,34-36 showed no advantage to ESWT. When available data from the trials were pooled by Buchbinder and coworkers37 in a Cochrane review, however, most benefits observed in the positive trials were no longer statistically significant. For example, pooled analysis of three trials (446 participants31,34,35) showed that ESWT is no more effective than placebo with respect to pain at rest 4 to 6 weeks after the final treatment. Pooled analysis of three trials (455 ­participants32,34,38) showed that

ESWT is no more effective than placebo 12 weeks after the final treatment with respect to pain provoked by resisted wrist extension (Thomsen test) and grip strength. Two randomized controlled trials looking at oral nonsteroidal anti-inflammatory drugs (NSAIDs) versus placebo produced contradictory results. Labelle and Guibert39 found that diclofenac use plus cast immobilization decreased pain by 1.4 (of 10) compared with cast immobilization alone, but at the cost of increased abdominal pain and diarrhea, and thus concluded against using NSAIDs. Hay and coworkers40 found naproxen to be no better than vitamin C in improving pain at 1 month and 1 year. Botox injections are a recent addition to the nonoperative arsenal, but the four randomized controlled trials are contradictory, and weakness is a significant side effect of treatment. Wong and associates41 found that botulinum injection resulted in superior pain improvement than saline injection at 4 weeks and 3 months (2.4 [of 10] less pain and 1.9 [of 10] less pain, respectively, than placebo), but at the cost of finger paresis in 13% at 4 weeks and 3% at 3 months. In a larger study involving 130 patients, Placzek and colleagues42 found that Botox injection gave modest benefit over saline injection over 6 months. The average reduction in pain in the last 48 hours was 0.84 (of 10) more than placebo, but at the cost of increased weakness. Hayton and associates43 found no benefit to Botox over placebo at 3 months, noting significant weakness interfering with quality of life in 11% of patients receiving Botox. Topical NSAIDs have shown to be superior to placebo, but only in the short term. Green and colleagues44 meta-analyzed three randomized controlled trials with a combined population of 130 participants and found that patients using topical NSAIDs (diclofenac in two trials, Benzydamine in one trial) had 1.88 (of 10) less pain than those receiving placebo in weeks 1 to 3 after treatment. Further studies are needed to see whether this benefit persists beyond this period however. In the athlete, specifically the tennis player, one must first obtain relief of acute pain and then increase the forearm extensor power, flexibility, and endurance. In combination with these efforts, the athlete must attempt to decrease the moment of force placed against the elbow by altering sport mechanics or changing equipment. Proper stroke techniques are emphasized for the club tennis player, particularly for backhand strokes, so that the forearm is not placed in the fully pronated position with this stroke. Avoidance of impact without the proper forward body weight transference is also stressed. For selecting the proper racquet, Nirschl suggests that the distance from the mid-palmar crease to the ring finger is helpful in determining handle size (Fig. 19B-3). We have also successfully used this technique in treating patients with this condition. One should also consider reducing racquet string tension, enlarging the grip handle, and avoiding the use of heavy tennis balls.3 Isotonic eccentric hand exercises with use of graduated weights of usually no more than 5 pounds may also be helpful. Repetitions are increased on a daily basis as the patient is able to perform them. If pain recurs, return to a lower level of exercise is advised in combination with NSAIDs or rest. Both Froimson45 and Nirschl5,17 advocate the use of counterforce braces. These are about 5 to 6 cm wide and

Elbow and Forearm 1201

Figure 19B-3   Nirschl technique for determining proper handle size measured from proximal palmar crease to tip of ring finger. Place measuring rule between ring and long fingers for proper ruler placement on palmar crease. The measurement obtained is the proper handle size—that is, if this distance is 4.5 inches, the proper grip size is 4.5 inches. (Courtesy of the Mayo Foundation.)

consist of a band of heavy-duty, nonelastic fabric lined with foam rubber padding to prevent slipping (Fig. 19B-4). Velcro fasteners allow easy application of the band, which encircles the forearm just below the elbow. Tension is adjusted to a comfortable degree with the muscles relaxed. The patient is advised to use the support only during actual play, to avoid excessive tightness, and to remove it during periods of inactivity. It is believed that such bracing provides gentle compression of the muscle-tendon areas and partially decreases muscle expansion at the time of intrinsic muscle contraction. This was investigated by Groppel and Nirschl46 with electromyographic studies using counterforce bracing. A lateral counterforce brace demonstrated lower muscle activity in the two extensor muscles across all skill levels with the serve and one-handed backhand. Two meta-­analyses on bracing find insufficient data, however, to definitively support bracing as an effective measure in lateral epicondylitis.47,48 Conclusions are hard to draw because the handful of studies, although prospective and randomized, have small numbers and short follow-up and cannot be pooled because of heterogeneity of brace type, outcome measures, and comparison arm. In summary, we educate the patient on how to avoid aggravating activities, teach proper racquet handle choice and swing mechanics in the case of tennis players, and offer physiotherapy, bracing, and NSAIDs as potentially helpful treatment options but inform the patient that simple ­education and avoiding aggravating activities will bring symptom relief in 80% of patients at 6 months.

Figure 19B-4   Lateral elbow counterforce brace. Note that the wide, nonelastic support is curved to fit the conical forearm shape. This concept does not allow full muscle expansion, thereby diminishing intrinsic muscle force on the lateral epicondyle. (From Morrey BF: The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985.)

Surgical Treatment Surgery is indicated if patient symptoms persist for longer than 6 months to 1 year despite conservative measures. If the pain is disabling and recurrent despite faithful compliance with the nonoperative program, surgical intervention may be initiated sooner than 6 months. Surgical intervention falls into three broad categories, each with advantages and disadvantages: open release and resection of diseased ECRB and EDC tendons, percutaneous release of ECRB, and arthroscopic excision of diseased ECRB and EDC tendons along with treatment of concomitant intra-articular pathology.

Open Release and Resection Open lateral release and resection affords good exposure to the diseased tendons and can be done completely under local anesthetic in less than 15 minutes but provides poor access to intra-articular pathology. Nirschl and Pettrone5 described an open technique to excise diseased ECRB tendon in 1979, achieving good results. In this approach, an oblique 3- to 4-cm skin incision distal and anterior to the lateral epicondyle is made, and the interval between ECRL and EDC is identified and opened to expose the ECRB tendon (Fig. 19B-5). The involved ECRB tendon, which often looks grayish and edematous, is removed along with any diseased EDC tendon, and one or two holes are drilled in the epicondyle, causing bleeding to enhance healing. The joint is not routinely examined unless pathology is

1202 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Degeneration, extensor carpi radialis brevis

Extensor carpi radialis longus

Extensor aponeurosis interface

Lateral epicondyle

A2 Removal angiofibroblastic degeneration extensor brevis

Lateral epicondyle

D2

Lateral epicondyle

B2

Decortication by drilling anterior lateral condyle

Lateral epicondyle

E2

Extensor carpi radialis longus

Lateral epicondyle Extensor aponeurosis

C2

Closure, extensor longus to extensor Synovial aponeurosis opening

Extensor aponeurosis

Extensor carpi radialis longus

Lateral epicondyle

F2

Extensor aponeurosis

Figure 19B-5   Mini-open technique (Nirschl) for lateral tennis elbow surgery. A1 and A2, Incision is 2.5 cm long, passing just anteromedial to lateral epicondyle. B1 and B2, Interface between ECRL and extensor aponeurosis (extensor digiti communis [EDC]). C1 and C2, Incision is made 2 to 3 cm deep and ECRL is retracted as extensor carpi radialis brevis (ECRB) comes into view. D1 and D2, Removal of degenerated ECRB. E1 and E2, To enhance vascular supply, two to three holes are drilled through cortical bone of anterior lateral condyle to cancellous bone level. F1 and F2, ECRL is firmly repaired to anterior margin of extensor aponeurosis. Because the ECRB is still attached to the underside of the ECRL, it is unnecessary to suture the distal brevis. Filled chevron, lateral epicondyle; open chevron, tendinosis tissue; filled arrow, extensor carpi radialis longus (ECRL); open arrow, extensor aponeurosis. (From Nirschl RP: Elbow tendinosis/tennis elbow. Clin Sports Med 11:851-870, 1992.)

Elbow and Forearm 1203

suspected, ECRB is not reattached because it is still attached to the underside of ECRL, and the ECRL-EDC interval is closed with absorbable suture. With this technique, Nirschl obtained 75% excellent results and 10% good results in 88 elbows. The improvement rate was 97.7% and 85% returned to full activities. In a prospective study involving 63 patients with symptom duration averaging 50 weeks (about 11.5 months), Verhaar and colleagues50 reported no or slight pain in 76% of cases 1 year after surgery operatively. Thirty-two percent of patients had excellent results, 37% good results, 19% fair results, and 11% poor results at 1 year. After 5 years, there were 32 excellent results (56%), 19 good results (33%), 4 fair results (7%), and 2 poor results (4%). Rosenberg and Henderson51 reported on open lateral releases in 19 patients with a mean prior symptom duration of 21 months (minimum, 6 months) and noted that 95% achieved pain relief and strength gain on average 3 and 4 months after surgery, respectively.

Percutaneous Release Percutaneous release can be done in an office setting and requires less dissection than an open release, which may lead to quicker recovery but in which visualization of the diseased tendon is poor and no excision is made. In the percutaneous release described by Savoie,52 a No. 11 blade knife is introduced 2 to 3 mm anterior to the epicondyle with the elbow flexed at 90 degrees. The blade is inserted parallel to the long axis of the humerus and is then pivoted so that its tip reaches the more superior aspect of the ECRB proximally. Distally, the blade tip stops short of the articular margin of the humerus. The surgeon’s thumb is kept on the posterolateral aspect of the radiocapitellar joint to protect the radial collateral ligament. A small rasp is then used to excoriate the epicondyle in the area to stimulate a healing response. Using this technique, Savoie had an average increase of 32 in Andrews-Carson score in 21 percutaneous releases after surgery, obtaining an average score of 198 out of a possible 200. A high Andrews-Carson score indicates good function. One advantage of a percutaneous release over the open release is lower morbidity. Dunkow and colleagues53 con­ ducted a randomized trial comparing an open release with a percutaneous release in 47 elbows of patients having failed 1 year of conservative treatment and found that the percutaneous release group returned to work at an average of 2 weeks after surgery whereas the open group returned to work 5 weeks after surgery (P = .001). Hand dominance and involvement with a workman’s compensation board were not controlled. Light duties and full duties were also not distinguished in the designation of back to work, but the percentage of manual laborers in each group was similar (percutaneous release, 63% manual laborers, versus open release, 70% manual laborers). The decrease in DASH score after surgery showing improved function was significantly larger in the percutaneous release group. Additionally, 50% of patients were very pleased with procedure in the percutaneous group, compared with 25% in the open group. There were no significant complications noted in either group.

Arthroscopic Release In 1995, Grifka and colleagues54 described an arthroscopic method of performing a Hohmann lateral release for recalcitrant tennis elbow. Kuklo and associates55 conducted a cadaveric study using 10 specimens to study the safety of the lateral arthroscopic release. He found the radial nerve to be in close proximity to the proximal lateral portal (average distance separating the two was 5.4 mm) and failed to find instability in the elbow post intervention. Smith and associates56 confirmed the efficacy of the arthroscopic procedure using seven cadavers. They noted that the ECRB origin was completely excised in all cases and that, on average, 90% of the EDC origin was resected. Elbow stability was maintained when resection did not extend posteriorly to an intra-articular line bisecting the radial head. Arthroscopy also allows intra-articular pathology to be diagnosed and addressed concomitantly and may enable the patient to return to work more quickly than after open release. Arthroscopic release, however, requires general anesthesia in most patients (regional anesthesia may be used in selected patients) and, because of the set-up time, cannot be performed as quickly as an open or percutaneous release under local anesthesia. Baker and coworkers57 published their results in 42 arthroscopic releases. Patients had on average 14 months of symptoms and were followed for an average of 2.8 years. The average pain at rest after surgery was 0.9 (of 10), pain with daily activity was 1.4, and pain with sports was 1.9. Of the 39 elbows in the 37 patients who were available for follow-up, 37 (95%) were rated “better” or “much better.” Patients returned to work after an average of 2.2 weeks. Grip strength averaged 96% of the strength of the unaffected limb. Of note, there was a 69% rate of associated intra-articular pathologies found at the time of ­arthroscopy. Comparisons between open release and resection and arthroscopic release and resection have all been retrospective and have all shown that both procedures lead to similar results. Rubenthaler and associates58 retrospectively compared 10 open lateral Hohmann releases (symptom duration mean, 13.6 months) with 20 arthroscopic Hohmann lateral releases (symptom duration mean, 10.6 months) and found similar results in the two groups. Peart and associates33 retrospectively compared 54 open lateral releases (16 months of prior symptoms) with 33 arthroscopic ones (22 months of prior symptoms) and found no significant differences in outcomes, but patients treated with arthroscopic release returned to work earlier than patients treated with open release and required less postoperative therapy. Szabo and colleagues59 retrospectively compared 38 open lateral release cases with 41 arthroscopic cases and 23 percutaneous cases and found similar results in each arm 4 years after surgery. The analysis was confounded by two factors, however: first, the groups were not perfectly matched preoperatively; and second, concomitant surgery was often performed, and these secondary procedures were unequally distributed in each group. The percutaneous group had a better preoperative AndrewsCarson score (164.4 versus 160.2 for open and 158.9 for arthroscopic) and had a shorter duration of ­conservative care

1204 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

(6.6 months versus 15.3 for open and 12.9 for arthroscopic). Extra-articular concomitant surgery was done in 22% of the arthroscopic cases and in 21% of the open cases, whereas no surgery was performed in the percutaneous group. Szabo found that postsurgery pain visual analog scale, Andrews-Carson score, number of complications, failures, and recurrences were not significantly different for the three groups. The Andrews-Carson scores were 195.3, 195.4, and 193 for the open, arthroscopic, and percutaneous groups, ­respectively.

Authors’ Preferred Method Because of the ability to visualize and arthroscopically ­ anage the entire joint while performing minimal dissecm tion, we prefer an arthroscopic release and excision of the ECRB-EDC. A general anesthetic is preferred, but regional anesthesia is feasible in selected patients. A nonsterile inflatable tourniquet is used and placed as proximal as possible, and the ­patient is positioned in the lateral decubitus position using a beanbag positioner padding all the bony prominences. An arm board is used for the dependent arm, and a well-­padded bolster placed just distal to the arm board keeps the operative elbow bent at 90 degrees. The joint is inflated with 15 to 20 mL of saline using the direct lateral entry site (soft spot) after preparing and draping and examination ­under general anesthesia for crepitus, range of motion, and ­instability. Joint distention diminishes the risk for nerve ­injury with subsequent portal placement. A proximal medial portal is established by cutting the skin with a No. 11 blade 2 cm (1 fingerbreadth) proximal to the medial epicondyle just anterior (0 to 5 mm) to the intermuscular septum. Only the skin is cut with the blade, while blunt dissection spreading with a hemostat is used to reach the joint capsule. The 4-mm cannula (2.9 mm can also suffice) and blunt trocar are used to penetrate the joint, palpating the supracondy­ lar ridge and staying anterior to it and aiming 10 degrees posteriorly and 10 degrees distally toward the radial head. Hugging the anterior surface of the humerus without scraping the cartilage ensures entry through the capsule into the joint instead of sliding between the capsule and brachialis—a common mistake. Intra-articular entry is confirmed by fluid return. A 30-degree scope is used to inspect the anterior compartment. Often, there is a rent in the capsule already present in sufferers of lateral epicondylitis. A spinal needle under direct visualization is used to make the proximal anterolateral portal situated 0.5 cm proximal to the radiocapitellar joint as seen from inside the joint with the arthroscope. Usually this is situated externally 2 cm (1 fingerbreadth) proximal and 1 cm (1⁄2 fingerbreadth) anterior to the lateral epicondyle. The No. 11 blade, followed by the hemostat and blunt trocar, is used to establish the proximal anterolateral portal. A 3.5 shaver is introduced and used to make an 8- by 4-mm hole in the capsule 1 cm anterior to the epicondyle revealing the diseased ECRB. Care must be taken to stay above the equator of the radial head to avoid damaging the radial ulnar collateral ligament.

Extra-articular Lateral Epicondylar Release

At this point, our technique differs from that described by Baker and associates57 because the camera is introduced into this lateral-based portal using a switching stick introduced under direct visualization. The arthroscope is then slowly withdrawn to look at the ECRB and the common extensor origin extracapsularly. Under direct vision using a spinal needle, an accessory proximal lateral portal is created 1 cm proximal to the proximal anterolateral portal containing the arthroscope. A shaver is introduced through this high proximal anterolateral portal and used to resect the diseased ECRB and EDC tendon. The epicondyle is then decorticated using a bur (or a shaver if the bone is soft). The capsular rent created at the onset allows visualization of the radial head and provides the landmark through which we can avoid resecting the radial ulnar collateral ligament. We then inspect the posterior and posterolateral compartments in a standard fashion. We believe this extracapsular view gives a superior visualization to the pathology akin to an open procedure yet keeps the advantages of an arthroscopic technique. Postoperatively, we place a soft dressing on the elbow and allow active and passive range of motion as tolerated immediately. Anti-inflammatory measures (ice, elevation, compression, range of motion) are begun immediately. A smaller dressing is placed at 48 to 72 hours (three Elastoplast bandages), and stretching is continued, but resisted wrist and finger extension exercises are withheld until the 6-week mark. Tennis can usually be resumed gradually at about 12 weeks when adequate strength has returned and little or no pain remains. A counterforce support band may be worn indefinitely.

Using this technique, we have achieved 91% good to excellent results in 11 patients with an average follow-up of 12 months (range, 3 to 23 months). The average pain at rest decreased from 5 (of 10) preoperatively to 1 (of 10) postoperatively. The average pain with activity decreased from 7.1 (of 10) preoperatively to 2.2 (of 10) postoperatively. There was one recurrence at 1 year, which was successfully treated by revision with the same technique.

MEDIAL EPICONDYLITIS (GOLFER’S ELBOW) Medial epicondylitis develops from repetitive valgus and flexor forearm stress. Sporting activities producing this condition include squash, racquetball, tennis, and golf. Tennis and any racquet strokes most likely to initiate difficulty are the serve and forehand.

Symptoms and Signs Symptoms of aching pain in the flexor musculature at the medial epicondyle are most common. Weakness of grip strength is also common. Patients may have either an acute injury of the common flexor origin or rupture of the medial collateral ligament from a throwing injury. In an acute injury of the common flexor origin, forearm flexor

Elbow and Forearm 1205

muscle pain or medial tendinitis is enhanced by flexing and pronating the wrist against resistance (Fig. 19B-6). In ­rupture of the medial collateral ligament, valgus stress to the elbow produces pain medially (Fig. 19B-7). In either condition, there may be associated symptoms of mild ulnar neuropathy. Although this does not occur in all instances, forearm pain or radicular symptoms must be considered if surgical intervention is undertaken. Nirschl2 found ulnar nerve dysfunction in 60% of patients undergoing surgery for medial epicondylitis. Concomitant osteocartilaginous loose bodies and triceps tendinitis were also found in 1% of his cases. Symptoms are generally mild, intermittent, and primarily sensory, occurring after prolonged use and with heavy forearm activity. Tinel’s sign may be present. This is presumably secondary to local inflammation and edema, causing a compressive neuropathy of the ulnar nerve in the region of the cubital tunnel. In most cases, electromyographic conduction study findings are normal. ­Decompression by release of the flexor carpi ulnaris arcade generally resolves these symptoms.17 It is important to obtain radiographs before surgery to ensure that there is no evidence of degenerative changes in the posterior medial aspect of the olecranon, mimicking medial epicondylitis. Such changes can be visualized radiologically by simple radiographic assessment.

Treatment Nonsurgical Treatment The treatment of medial epicondylitis is similar to that of lateral epicondylitis; the principles center on the relief of the acute or chronic inflammatory process. This involves the use of NSAIDs in combination with ice and a decrease or modification in the activities that produce the tension overload. Finally, a therapy program aimed at gradually increasing flexibility, power, and endurance is initiated. A counterforce elbow splint has been devised that is similar to the lateral counterforce brace but provides an

additional support just distal to the medial epicondyle. Unfortunately, this has not proved to be of much value in our hands. Gradual resumption in play is usually recommended when symptoms have subsided, generally between 6 and 12 weeks after injury.

Operative Treatment The data regarding surgical management of medial epicondylitis are limited but slowly increasing.3,17 Precise localization of the maximal point of tenderness is necessary before surgery.60 A needle may be used for this purpose, and the medial epicondyle is exposed while the needle is in position. The tendon origin of the pronator teres and a portion of the flexor carpi radialis, often at the interval between these two muscles, are generally the sites of involvement. Torn or scarred tissue is excised, and repair is done in a manner similar to that described for the ­lateral epicondyle. All ­normal tissue is left attached to the medial epicondyle for fear that total excision of the common flexor origin may include a portion of the origin of the medial collateral ligament, which may then lead to subtle posteromedial instability. Coonrad3 identified the ulnar nerve in each case with this exposure, although he does not transpose the ulnar nerve or perform a neurolysis unless evidence of associated neuropathy is present. Management of the ulnar nerve usually involves decompression just distal to the medial epicondylar groove. Decompression by release of the flexor ulnaris arcade generally resolves these symptoms in our experience and in that of others. Indications for anterior ulnar nerve transposition include (1) nerve subluxation or dislocation from the epicondylar groove; (2) cubital valgus, with symptomatic tension neurapraxia, which is common in throwers; and (3) a hostile environment, such as scarring from previous surgery. It appears that there are few complications except a rare loss of extension of about 5 degrees. In our experience, the major problem has been the overall lack of predictability of the procedure when the ulnar nerve is involved. In fact, reports from the Mayo Clinic and other reports

Pain

Wrist flexion

Pain

Figure 19B-6   Medial epicondylitis may be diagnosed clinically from pain localized to the medial epicondyle during wrist flexion and pronation against resistance. Pain is often elicited after a tight fist is made, and grip strength is usually diminished on the affected side. (From Morrey BF: The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985.)

Valgus stress Pain in ulnar collateral ligament Figure 19B-7   Medial joint-line pain elicited by applying a valgus stress to the elbow identifies injury to the ulnar collateral ligament. (From Morrey BF: The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985.)

1206 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method The patient lies supine with the arm abducted and flexed to 90 degrees and resting on an elbow table. A gentle, curved incision is made, beginning just posterior to the medial epicondyle. After excision down to the fascia overlying the medial epicondyle, the ulnar nerve is identified and protected proximally. If there is any sign of compression of the nerve, or if symptoms have suggested that this is the case, the forearm fascia is split where the nerve enters the forearm between the two heads of origin of the flexor carpi ulnaris. We avoid a detailed dissection of the nerve unless intrinsic disease is found. Specifically, the cubital tunnel retinaculum is not violated unless constriction occurs at this point (Fig. 19B-8). The fibers between the pronator teres and the flexor carpi radialis are then excised longitudinally. The tendinous origin of these muscles is sharply excised from the medial epicondyle in the direction of the joint. The fascia and fleshy muscle fibers are sutured back to the cuff of tissue left on the medial epicondyle; the scarred and degenerative tendon is excised in an elliptical and longitudinal fashion. Little attempt is made to close the defect, but normal tissue, if detached, is repaired. The longitudinal incision made through the common flexor origin is then closed with 2-0 absorbable suture. When the ulnar nerve is extremely irritable or subluxed, or when there is objective clinical or electromyographic evidence of ulnar compromise, this is addressed by a subcutaneous transposition of the nerve. A soft dressing is applied to the elbow and the patient is allowed active and passive range of motion of the elbow, wrist, and hand immediately. Anti-inflammatory measures of ice, compression, and elevation are also instituted immediately. After 7 to 10 days, the bulky dressing is changed for a lighter dressing, and the incision checked. Active and passive range of motion and stretching are continued. At the end of 6 weeks, we encourage the restoration of grip strength

reveal that the 90% satisfactory rate is adversely affected by ulnar nerve involvement.61-63 Kurvers and Verhaar,64 in a ­retrospective review of 40 cases with average follow-up of 44 months, found that only 12.5% of patients (3 of 24) are symptom free after surgery if there are ulnar nerve symptoms preoperatively versus 69% (11/16) if there are not.

TRICEPS TENDINITIS AND RUPTURE Anatomy The tendon of the triceps brachii muscle consists of two aponeurotic lamellae that join together above the elbow and insert into the posterior portion of the dorsal surface of the olecranon. A lateral band extends over the anconeus muscle to attach to the dorsal fascia of the forearm. This has important reconstruction implications.

Triceps Flexor origin

Arcade zone 3

Zone 1 Zone 2

Figure 19B-8   Ulnar nerve zones at the cubital tunnel: zone 1, proximal to medial epicondyle; zone 2, at medial epicondyle; zone 3, distal to medial epicondyle. Zone 3 includes penetration of the nerve through the flexor ulnaris arcade and is the most common site of compression neurapraxia of the ulnar nerve. (From Morrey BF: The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985.)

by having the patient squeeze a rubber ball or by the use of ­exercise putty. Between 8 and 10 weeks after surgery, ­resisted elbow and wrist flexion exercises are initiated with the use of light weights. At the end of 4 to 6 months, patients have usually regained adequate strength to resume light duties or sporting activities. We recommend no return to their level of sport or work until we have evidence by objective testing that they have regained 80% or more of strength compared with the opposite extremity. In our experience, this usually takes about 6 months. We recommend the use of a forearm support band while working or playing any overhand sports for up to 6 months after surgery.17

Triceps Tendinitis Triceps tendinitis is thought most often to be an isolated entity associated with loose bodies in the posterior compartment of the elbow or with lateral tennis elbow.11,65 This condition occurs in baseball players, weightlifters, and those with occupations involving repetitive elbow extension, such as carpenters. The same conservative management program is used for triceps tendinitis as for lateral or medial tennis elbow. We have operated on only a single case of triceps tendinitis and doubt that operative treatment is indicated unless the pathologic process is actually a partial rupture. In cases that have been refractory to all conservative treatment for more than 1 year, Nirschl17 has elected surgical excision of a small portion of the triceps insertion. The site of resection coincides with the area of maximal point tenderness, previously eliminated by local lidocaine injection. The technique is essentially that of tennis elbow débridement.

Elbow and Forearm 1207

Triceps Rupture A rupture or avulsion of the triceps tendon is a rare injury. In 1868, Partridge66 reported the first case, which was managed by rest and graduated exercise. The site of disruption was at the tendo-osseous junction. Reported ages of those affected range from 7 to 72 years. Most patients are men, and right and left sides are equally affected. Injury in the skeletally immature is rare.67,68 There is no correlation of the side of rupture with the dominant or nondominant side. Anzel and associates69 reported a 9-year experience of consecutive tendon ruptures in 781 patients involving 1014 tendon ruptures. Eight triceps ruptures represented 1.9% of the tendon injuries. Our experience, discussed, is from a review of 22 patients treated for triceps tendon repair or reconstruction.70

Mechanism of Injury The mechanism of injury is either a fall onto an outstretched hand or direct laceration from a direct blow.71 Disruption of the triceps may occur spontaneously with minimal trauma in individuals who have been compromised by a systemic disease process, such as hyperparathyroidism,72 and in those receiving steroid treatments for lupus erythematosus.73 The most common site at the tendo-osseous attachment represents avulsion of the insertion, occasionally with a fleck of bone. Musculotendinous injuries have also been reported.74,75 Mair and associates retrospectively reviewed 10 partial and 11 full-thickness triceps tear in the National Football League over a time span of 6 years.76 Thirteen of 19 magnetic resonance images were reviewed (68%) and all tears were located at the olecranon insertion or within 2 cm proximal to it. Eight of the 10 partial tears were predominantly medial sided. The association of additional injuries cannot be overlooked. Levy and coworkers77,78 described 16 patients who had incurred triceps avulsions in association with radial head fractures. Clayton and Thirupathi79 found an association with chronic olecranon bursitis.

Diagnosis The most common mechanism of injury is a deceleration stress superimposed on a contracted triceps muscle, with or without a concomitant blow to the posterior aspect of the elbow. The diagnosis is usually evident in patients who present with a characteristic history and a palpable depression just proximal to the olecranon. On examination, pain, swelling, and a palpable depression just proximal to the olecranon and posterior ecchymosis may be present. Because of the expansion of the triceps laterally with the anconeus, however, active extension is still possible in some. In Mair’s study,76 7 of 21 patients (33%) complained of pain in the area before the rupture (5 of those had received cortisone injection in the past for presumed olecranon bursitis), eccentric contraction of the triceps was the mechanism in 80%, weakness and pain was found in all, and a palpable defect was found in 76% (16 of 21). Radiographs are particularly helpful, with flecks of osseous tissue visible on the lateral radiograph, but this

Figure 19B-9   Anteroposterior and lateral radiograph illustrating the small avulsion fracture off the tip of the olecranon, indicative of a triceps tendon rupture. (From Morrey BF: The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985.)

occurs in less than 20% of patients (Fig. 19B-9).70,75,80 Magnetic resonance imaging and even ultrasonography have more recently proved useful in the diagnosis of subtle cases.81,82

Treatment Acute repair is recommended, but a reconstruction procedure is necessary for those with a delayed intervention. At exploration, the tendinous central portion of the triceps may be found to be retracted, sometimes accompanied by a bone avulsion from the olecranon. Delamination due to a partial disruption of the triceps tendon is also seen.70,83 For acute rupture, a nonabsorbable suture placed in drill holes in the olecranon is effective (Fig. 19B-10).70,77,78 Stainless steel wire is not recommended because the wire fatigues and fragments.20 If treatment has been delayed or if the patient has a systemic disease that compromises the quality of tissue and hence the quality of repair, a reconstructive technique may be considered. Farrar and Lippert80 suggested using a periosteal flap from the olecranon to reinforce the repair. Bennett84 used a flap of fascia taken from the posterior aspect of the forearm with its base attached to the medial and lateral epicondyles to the humeral anterior olecranon process. Clayton and Thirupathi79 achieved successful results by splitting the triceps tendon into a partial-thickness flap. We have had excellent success with a rotation flap employing the anconeus (Fig. 19B-11); if there is marked tissue deficiency, an Achilles tendon allograft is employed.79

1208 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 19B-10   A crisscross suture is secured to the olecranon by a cruciate attachment through drill holes in the olecranon. (Courtesy of the Mayo Foundation.)

Figure 19B-11   The anconeus serves as a rotational flap providing continuity with the triceps mechanism. (Courtesy of the Mayo Foundation.)

Partial tears can be treated nonoperatively with success.76 Of the 10 partial tears Mair reviewed in the National Football League, 6 healed without any residual pain or weakness, causing the patients to miss on average 4.8 weeks of play (range, 0 to 9 weeks). One player completed the rupture upon returning to practice 5 days after injury and was repaired surgically, and three players finished the season but had surgery after the season for residual pain and weakness. It was not possible to correlate percentage of tear to those requiring surgery in this small series. All the players returned to play at least one season of professional football after their injury.

­ rogressive active and active-assisted exercises at 2 months; p by 4 months after the procedure, he had resumed weight training.74 In our experience, recovery is slow and may even require a full year for complete functional return.70

Postoperative Rehabilitation Postoperative immobilization varies from 10 days to 6 weeks, but most patients are generally immobilized for 2 to 3 weeks in a posterior splint at 35 to 40 degrees of elbow flexion. Active range of motion exercises are then begun. In a single report, an Olympic weightlifter began

Results In most instances of acute or delayed repair, nearly normal strength and full motion are restored with no pain. Levy and associates77,78 noted a 10% to 20% limitation of range of motion in two patients. Anderson and LeCoco83 and Bennett84 have noted 5-degree flexion-contraction in two patients. Sherman and colleagues75 noted that a professional body builder was able to bench-press 370 pounds without difficulty 3 months after repair. Most of these reports are without the benefit of precise objective strength measurements. In our experience with 22 procedures, only 5 occurred from athletics; the 3 with acute repair had slightly better functional results than the 9 with reconstructions. Overall, 21 of the 22 considered the result satisfactory.

Authors’ Preferred Method The treatment of choice of acute rupture is immediate ­repair by use of a posteriorly based incision just lateral to the ­midline. No. 5 nonabsorbable suture or No. 2 FiberWire (Arthrex, Naples, Fla) is preferred. The first is placed in a Bunnell

fashion through the proximal torn triceps tendon and then through crossed holes drilled through the olecranon (see Fig. 19B-11). A second transverse suture secures the precise site of attachment. The sutures are tied with the elbow at 60 degrees

Elbow and Forearm 1209

Authors’ Preferred Method—cont’d of flexion. This is the same technique that is used to reattach the triceps after elective reconstructive procedures. If repair is not possible or requires additional support, we prefer to mobilize the anconeus and lateral triceps ­extension and relocate this over the olecranon rather than create free tissue flaps (see Fig. 19B-11). In those with extensive ­deficiencies, the Achilles tendon allograft is used (Fig. 19B-12). This is attached to the olecranon first, either by screw fixation of the calcaneal portion or by advancement of the ­Achilles tendon into the bone of the olecranon. The fascial portion is attached to the triceps muscle with the elbow flexed about 45 degrees. We advocate protected immobilization for 2 to 3 weeks with the arm held in 30 to 40 degrees of flexion. Active­assisted motion is begun at 3 to 4 weeks. Training for avid athletes is not recommended until Cybex testing indicates that 80% strength has returned; this may take 6 months. We allow full participation in all active sporting events and expect 90% to 100% return of strength if the injury has occurred at the site of attachment. We are careful to advise that it takes a full year for complete recovery. In Mair’s review,76 in which all repairs were primary and unaugmented, one player returned to play at 7 weeks, but the rest returned to play the next season only. One player had a re-rupture dur­ ing rehabilitation at 6 weeks post operatively and was revised successfully. All had no discernable weakness, did not complain of pain, and had normal range of motion. One player retired from play, and 10 returned to play at least one more season of professional football. Figure 19B-12   When the anconeus is unavailable and for large defects, an Achilles tendon allograft is employed to reconstruct the extensor mechanism. (Courtesy of the Mayo Foundation.)

OLECRANON BURSITIS Athletic trauma causing bursitis most frequently occurs at the patellar bursa of the knee and at the olecranon bursa of the elbow. Football and ice hockey have been the chief athletic events implicated in the development of olecranon bursitis at the elbow. We are careful to advise that it takes a full year for complete recovery from this problem and that chronic conditions may need resection.

Anatomy The bursae in the olecranon region exist in three locations: (1) the subcutaneous bursa, so commonly seen clinically; (2) an intratendinous bursa, in the substance of the triceps tendon near its insertion; and (3) the subtendinous bursa between the tendon and capsule (Fig. 19B-13), as described by Morrey.85 Although there have been no recognizable clinically documented presentations concerning the two deep bursae about the olecranon, the intratendinous bursa may indeed be involved with tears of the triceps tendon as described earlier in this chapter. In the literature, the only ­subtendinous

Subtendinous bursa Intratendinous bursa

Olecranon bursa Figure 19B-13   Lateral illustration of the elbow demonstrating the superficial olecranon bursa, the intratendinous bursa found in the substance of the tendon, and the subtendinous bursa lying between the tip of the olecranon and the triceps tendon. (Courtesy of the Mayo Foundation.)

1210 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Classification Olecranon bursitis may be classified as acute, chronic, and suppurative. The acute entity results either from a direct blow or from repeated acute insults to the superficial elbow. Chronic bursitis develops as a sequel to recurrent acute episodes when the trauma is relatively mild. It occurs when the resorptive phase of an acute bursitis is repeatedly interrupted by further trauma. In this situation, the lining of the bursa is replaced by fibrous tissue, which becomes the predominant characteristic. Suppurative bursitis may also develop from an infective process in an acute or chronic bursitis after contamination through a skin wound or ­dermatitis.

Acute Bursitis Clinical Presentation

Figure 19B-14   Lateral radiograph demonstrating a spur off the tip of the olecranon, indicative of olecranon bursitis.

bursa was described by Vizkelety and Aszodi86 and was idiopathic in nature. We have not encountered this bursa in our practice. Inflammation of the superficial olecranon bursa is the bursitis of the elbow. The bursitis may be acute or chronic, septic or nonseptic; it is associated most commonly with occupational or sports trauma. It has been called a ­miner’s elbow or student’s elbow. In sports injuries, Larson and Osternig87 have documented that this is a common football injury, almost exclusively associated with artificial turf. In one season, 14 of the 16 cases (87.5%) were sustained on artificial turf. We have encountered this lesion in hockey as often as in football. The relation to artificial turf centers on the construction features of the artificial turf. This turf is composed of an upper layer of durable, synthetic grass, which is applied over a layer of padding of varying thickness and resiliency. These two surfaces are usually applied over a hard surface, such as asphalt. Consequently, artificial turf gives an even and consistent upper playing surface but has a relatively unyielding quality at the base compared with actual grass. Repeated falls on the partially flexed elbow result in trauma to the region, and sometimes an olecranon spur develops (Fig. 19B-14). Such traumatic episodes may produce an acute inflammation and subsequent bursitis. On occasion, with severe trauma, vascular disruption with hemorrhagic distention of the bursal sac may occur. An inflammatory response then ensues. The initial episode is usually one of hemibursitis. After resolution, recurrent episodes develop with less trauma and are not associated with intrabursal hemorrhage.

A painless distention of the bursa after a direct blow is the most common presentation. The differential diagnosis includes acute arthritis, ligament injury, and tendinitis if the swelling is diffuse. Otherwise, the diagnosis is easily made; the most important distinction is whether this represents a septic process. Joint motion is usually not limited in patients with olecranon bursitis except when flexion produces skin tension and increased pressure over the tender, distended bursa. This is an important point because septic bursae are painful. Fluctuation of the distended bursa in the absence of true joint findings further localizes the injury to the bursa. Canoso88,89 characterized the clinical features of 30 patients with acute traumatic olecranon bursitis. Repetitive trauma was cited as the most important factor in 14 cases (47%). When symptoms were present more than 14 days, the bursa was discretely swollen; but if symptoms were observed in less than 2 weeks, parabursal edema was present, and swelling was also observed in the arm and forearm in 50% of cases. We have not observed this associated swelling except with septic bursitis. Aspiration of the acute entity shows evidence of recent hemorrhage. The bursal fluid is characterized by a low white blood cell count (<2.0 white cells × 109/L), with a high percentage (80%) of monocytes.88,90

Treatment In the athletic population, olecranon bursitis is most commonly a post-traumatic disorder; but in the older patient, systemic inflammatory processes such as calcium pyrophosphate dihydrate crystal deposition disease91 must be considered. Once a bursitis has occurred, recurrence becomes more frequent with less trauma, resulting in the need to provide additional protective covering to lessen the forces of impact. In the acute situation, if the bursa is distended and uncomfortable or interferes with use of the joint, aspiration done under sterile conditions is recommended, followed by the application of continuous compressive bandages for 48 hours. If the distention is not too severe, there is no need to aspirate the bursa. Compression and cold packs at the first sign of swelling help to minimize bleeding into the sac,

Elbow and Forearm 1211

Figure 19B-15   Trabeculae and villi fill the bursal sac to produce a scarred, thickened bursa. This is excised through an incision centered lateral to the midline.

Figure 19B-16   The olecranon bursa is resected without opening the bursal sac.

Operative Intervention which usually reaches a maximum 24 hours after injury. After this, warm packs can be used to hasten absorption of bursal fluid. The bursal fluid often resorbs if it is not subjected to recurrent trauma. If continued trauma occurs, however, accumulation of fluid recurs. This can result in fibrous tissue formation, which becomes the predominant characteristic and then leads to subacute and chronic bursitis, which may require surgery.

Chronic Bursitis Presentation Chronic bursitis follows recurrent traumatic episodes, and in this situation, the bursal walls have usually become much thickened. Trabeculae and villi form, increasing in number and density and filling the bursal space.92 The villi that arise represent granulation in the floor of the bursa, consisting of central blood vessels surrounded by fibrous tissue cells. In a mild subacute form, this may be clinically evident as a slight, palpable thickening of the bursa; in the chronic form, a large rubbery mass containing numerous hard mobile bodies occurs in the subcutaneous tissues (Fig. 19B-15 and Fig. 19B-16). The causal factor in the development of a chronic bursitis is frequently not sports but the patient’s occupation. Occurrences have been so common in certain occupations that they have been given names such as miner’s elbow for chronic olecranon bursitis. By far the most cases of chronic olecranon bursitis develop from an acute traumatic episode, such as a football or hockey injury. In such cases, the acute injury is superimposed on a chronic or recurrent bursitis. Definitive measures are usually required in this situation.

Surgery is indicated if the process is refractory to nonoperative intervention and bothers the athlete to the point that he or she cannot participate in the sport of choice or if a septic episode has been imposed on the chronic process. Surgical technique falls under two broad categories: open and arthroscopic resection. In the open technique, a longitudinal incision just medial to the midline93 or centered directly over the olecranon bursa94 has been recommended. The bursa is carefully dissected without opening it if at all possible. On occasion, a bursa must be removed piecemeal; if this is necessary, all pieces of the bursal sac should be carefully removed. Freeing the bursa from the skin can devitalize the skin over the olecranon process or cause problems with healing. We recommend a compressive dressing with the elbow held in less than 45 degrees of flexion.85,95 Breck and Higinbotham94 have suggested placing mattress sutures on either side of the incision centered over the appropriate half of the dead space under the skin flaps. These sutures are passed down through the skin and then into the underlying deep tissues of the muscle and fascia of the triceps. The suture is then brought back up to the skin and snugly tied over a button. This brings the skin into firm contact with the underlying structures and tends to prevent the formation of a hematoma and recurrence of a bursa or skin breakdown. Quayle and Robinson96 have avoided the problem of wound healing secondary to the subdermal dissection of the bursa by simply reflecting the skin with the bursal tissue from a medial to a lateral direction. The tip of the olecranon is obliquely osteotomized, leaving the bursal tissue intact. The subcutaneous tissue plus bursa is then reflected over the olecranon, and the wound is closed with a drain. In the arthroscopic technique as described by ­OgilvieHarris and Gilbart,97 the patient is positioned supine

1212 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

and given general or regional anesthesia. A tourniquet is used, and the arm is placed across the chest with a pillow between the arm and chest to keep the elbow flexed about 45 degrees. The bursa is inflated with saline using a smallcaliber needle, and a lateral portal 1 cm distal and lateral to the bursa is created using the No. 11 blade to cut skin, followed by spreading with a hemostat to the bursal edge and puncturing the bursa with the camera sheath and sharp trocar. The portal is placed away from the bursa so that a subcutaneous tract can be used to diminish the risk for decompressing the bursa with arthroscope entry. A standard 4-mm, 30-degree arthroscope is used, and a pressure pump set at 35 mm Hg maintains the bursa open. A medial skin incision 1 cm medial and distal to the bursa is then made with the No. 11 blade, and the subcutaneous tissues are spread with the hemostat down to the bursa. The shaver (4.5 mm curved or 4.2 mm straight) is used to puncture the bursa medially and to resect first the subcutaneous part of the bursa, being careful not to penetrate the skin. The areas free of bursa should transilluminate. The deep surface of the bursa is then resected until fibers of the triceps are seen. Loose bodies can be resected using the shaver, and olecranon spurs should also be resected if seen. The tourniquet is released, the shaver tip is used to aspirate the bursa, and the portals are sutured closed. Long-acting local anesthetic is injected in the bursa, and a compressive dressing is kept for 10 days.

Results Stewart and associates,95 reviewing 16 open resections in nonrheumatoid patients at the Mayo Clinic, had success defined as complete relief without further need of surgery in 94% of cases (15 of 16 patients) at an average of 5 years. Rheumatoid patients fared less well, with only two of five patients (40%) obtaining a successful outcome. Postoperatively in the nonrheumatoid patients, there was one hematoma (6% of cases) requiring drainage and one stitch abscess that required no intervention. Quayle and Robinson,96 resecting only the olecranon spur and leaving the bursa, experienced no recurrence in 11 open resection cases. Some, like Degreef and associates98 from Belgium, have reported higher complications rates noting wound healing problems in 10 of 38 (27%) cases. Ogilvie-Harris and Gilbart,97 reviewing 31 cases of endoscopic olecranon bursal resection with an average of 2.7 years of follow-up, found one recurrence (3%) at 6 months in a rheumatoid patient, two cases of delayed healing at portal sites that required longer than 10 days of compression dressing, and one locally irritated portal site that resolved with topical antibiotics. Postoperatively, 86% had no pain at the elbow (compared with 17% preoperatively), 7% (two cases) had pain only on deep pressure (compared with 45% preoperatively), and 7% (two cases) had pain such that they could not rest their elbows on a table comfortably (compared with 28% preoperatively) (P = .01). Schulze and associates99 retrospectively compared nine endoscopically resected bursae with nine openly resected cases and found similar postoperative Morrey scores but found that the endoscopically treated patients returned to work significantly earlier (10 days versus 18 days; P = .04).

Authors’ Preferred Method We prefer a skin incision over the lateral aspect of the olec­ ranon.95 The skin and subcutaneous tissue are carefully incised, exposing the bursal sac. If possible, the bursal sac is enucleated in total, without violating its contents. The minimal amount of subdermal dissection required to accomplish this is performed equidistant on the medial and lateral aspects of the olecranon. Any olecranon spur or prominence is removed. The tourniquet is deflated to assess the viability of the skin, and meticulous hemostasis is attained. Closure is accomplished with absorbable sutures placed in the medial and lateral skin flaps and sewn down to the deep fascia. In this way, any dead space is obliterated, and the chance of subdermal hematoma is minimized. The wound is then closed in the usual fashion, and the elbow is immobilized in 0 to 45 degrees of flexion to prevent further recurrence. A compressive dressing is maintained for 2 to 3 weeks. Normal elbow motion is easily attained.

SEPTIC BURSITIS A large number of patients with a septic bursa give a history of previous idiopathic or traumatic bursitis. The presentation varies widely and includes the acute onset of localized cellulitis, a generalized cellulitis involving the forearm, a low-grade subacute process of 10 to 14 days in duration, or a fulminating process with systemic symptoms. In general, the patient with septic bursitis is not likely to be febrile, but unlike in the patient with an aseptic process, pain is present over the olecranon bursa and with motion. The infected bursa is often tender to palpation, and there is sometimes an abrasion or a skin lesion. These findings are not diagnostic of infection, however; some nonseptic cases also show these features. In addition, their absence does not rule out an infected bursa. In all cases of a painful bursa, the diagnosis is made by needle aspiration of the bursal contents. With aspiration, a cell count, crystal determination, and Gram stain should be done in all cases. Crystals are determined because gout may coexist with or even predispose the patient to olecranon bursitis.91 The drawn septic fluid generally appears purulent as either frank or bloody pus but may only appear serosanguineous or sanguineous.88,90,94 Fluid white blood cell count is more sensitive and specific than macroscopic appearance. Greater than 2.0 × 109/L is 79% sensitive and 95% specific for septic bursitis and has a positive predictive value of 80%.90 In addition, bursal fluid analysis in septic cases shows a predominance of polymorphonuclear cells (79% in septic versus 36% in aseptic cases90), whereas nonseptic fluids have a predominance of mononuclear cells. Gram stain smears of bursal fluid demonstrated organisms in all 10 septic cases reported by Ho and Tice.100 The most common organism was Staphylococcus aureus in 94% of cases in their series. In a prospective series of 15 consecutive septic olecranon bursitis seen in a London emergency department, Stell found liquid media cultures to be positive for S. aureus in 12 of 14 cases (93%).101

Elbow and Forearm 1213

Treatment of septic olecranon bursitis consists of adequate drainage coupled with antibiotic therapy. Adequate drainage varies from one or more aspirations, to aspiration with irrigation with a wide-bore cannula to formal open incision, irrigation and débridement, and drainage. Antibiotic regimen varies from admission and intravenous antibiotic treatment, to outpatient intravenous antibiotics followed by oral antibiotics, to oral outpatient antibiotic with close follow-up at 48 to 72 hours. Treatment duration varies from a 10-day oral antibiotic dose, to a single intravenous dose of antibiotic at presentation followed by a 10-day course of oral antibiotics, to parenteral antibiotics for 1 to 3 weeks followed by oral antibiotics for an additional 2 weeks.101,102 Stell101 treated 15 consecutive septic olecranon bursitis cases arriving at an emergency department in London in an outpatient manner, with one aspiration and oral flucloxacillin (13 cases) or erythromycin (2 cases) for 10 days and follow-up at 48 to 72 hours. Two cases not improving at 48 hours were reaspirated, and one case only slightly better at 48 hours was reaspirated. Only 1 case did not respond to three aspirations and erythromycin and needed an open incision, irrigation, and débridement, after which the patient did well. Three patients developed discharging sinuses (in the center of the bursa and not distally at the site of needle drainage) which appeared 5, 18, and 21 days after presentation, respectively, and closed 4 weeks, 6 weeks, and 1 week, after appearing, respectively. Most cases had significant symptom improvement by 48 hours (average pain level, 4.8 [of 10] at onset, improving to 1.7 at 48 hours), and symptom resolution took an average of 6.6 weeks (range, 1 to 13 weeks). Laupland and Davies in Calgary reviewed the outpatient treatment of 118 cases of septic olecranon bursitis and found that only 1 case needed admission for systemic symptoms.102 S. aureus constituted 88% of culture-proved cases, and the most common antibiotic regimen was cefazolin for a median of 4 days, followed by clindamycin for a median 8 days. Sixty patients (51%) were drained in an outpatient fashion.

Authors’ Preferred Method All painful bursae (even mildly painful) are aspirated, and the liquid is sent for analysis and culture. Bursae ­containing thick purulent material are irrigated with saline using a largebore (14-gauge) cannula until clear effluent is seen. The febrile patient or patient showing systemic signs is provided with parenteral antibiotics on a home intravenous program or hospital admission until symptoms improve. After significant improvement on intravenous therapy (usually 1 to 3 days), the patients are switched to a 10-day course of oral antibiotics with follow-up at 48 to 72 hours to confirm continued response to oral treatment. The afebrile patient not showing systemic signs but having fluid analysis (>2 white cells × 109/L or gram-positive cocci) or clinical signs suggestive of septic bursitis is given one dose of parenteral antibiotic active against S. aureus, such as cefazolin, and

sent home on a 10-day course of oral antibiotic such as cephalexin (other options include clindamycin, flucloxacillin, and erythromycin) with follow-up at 48 to 72 hours to confirm response to treatment. The patient with little clinical and laboratory evidence of septic bursitis is discharged without antibiotic but is seen at 48 to 72 hours to confirm improvement and culture-negative results. At the follow-up appointment, culture results are reviewed, and sensitivity to the antibiotic is checked if an organism is found. Nonresponders are reaspirated (septic and nonseptic cases) and given another dose of intravenous antibiotic if the bursitis is septic or occasionally methylprednisolone, 40 mg, if clearly aseptic and symptomatic and followed up in 48 to 72 hours. Responders are seen at the end of treatment and on an as-needed basis after. Failure to respond to three or more aspirations or a loculated abscess is treated with open débridement and irrigation.

C

r i t i c a l

P

o i n t s

l Tennis elbow is an expression of apoptosis—planned cell death. l The natural history of both medial and lateral epicondylitis is that it will heal with time. l Most documented surgical releases are effective but are probably not as reliable as the literature would suggest. l Triceps tendon rupture may be missed because continuity of the anconeus still allows extension. l The rehabilitation after triceps repair or reconstruction is markedly more prolonged than that of biceps tendon rupture and may take up to a year. l Septic bursitis is characterized by a painful process and may be associated with marked swelling of the extremity. This may lead to septicemia. Urgent treatment is required.

S U G G E S T E D

R E A D I N G S

Basker CL Jr, Murphy KP, Gottlob CA, Curd DT: Arthroscopic classification and treatment of lateral epicondylites: Two year clinical results. J Shoulder Elbow Surg 9:475-482, 2000. Dunkow PD, Jatti M, Muddu BN: A comparison of open and percutaneous techniques in the surgical treatment of tennis elbow. J Bone Joint Surg Br 86:701-704, 2004. Gable GT, Morrey BF: Operative treatment of medial epicondylitis: Influence of concomitant ulnar neuropathy at the elbow. J Bone Joint Surg Am 77:1065-1069, 1995. Laupland KB, Davies HD, for the Calgary Home Parenteral Therapy Program Study Group: Olecranon septic bursitis managed in an ambulatory setting. The Calgary Home Parenteral Therapy Program Study Group. Clin Invest Med 24:171-178, 2001. Nirsch RP, Pettrone FA: Tennis elbow: The surgical treatment of lateral ­epicondylitis. J Bone Joint Surg Am 61:832-839, 1979. Regan W, Wold L, Coonrad R, Morrey BF: Microscopic pathology of lateral ­epicondylitis. Am J Sports Med 20:746, 1992. Stell IM, Gransden WR: Simple tests for septic bursitis: Comparative study. BMJ 316:1877, 1998. Vangness T, Jobe F: The surgical treatment of medial epicondylitis. J bone Joint Surg Br 73:409-411, 1991. van Riet RP, Morrey BF, Ho E, O’Driscoll SW. Surgical treatment of distal triceps ruptures. J Bone Joint Surg Am 85A:1961-1967, 2003. Werner CO: Lateral elbow pain and posterior interosseous nerve entrapment. Acta Orthop Scand Suppl 174:1-62, 1979.

R eferences Please see www.expertconsult.com

1214 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S ect i o n

C

Throwing Injuries 1. Throwing Injuries in the Adult Jeffrey R. Dugas, Scott B. Reynolds, James Lebolt, and James R. Andrews

BASIC SCIENCE The throwing motion is a dynamic activity that requires extremes of glenohumeral motion that place extraordinary stresses on the athlete. Forces generated during the throwing motion are astonishing, and the muscles of the rotator cuff must offset these high-energy forces on the shoulder and stabilize the humeral head within the ­glenoid. Although the most widely studied sport involving overhead throwing is baseball, several other sporting activities require the same type of motion. Some such sports include football, volleyball, handball, javelin, softball, racquetball, squash, and tennis (the serve).1-7 The overhead throwing motion can be broken down into several discrete steps, which include the wind-up, early cocking, late cocking, acceleration, deceleration, and follow-through phases. The process of overhead throwing, regardless of specific sport, involves the generation of potential energy and the subsequent transfer of that potential energy to kinetic energy, which is imparted to the object being thrown. In total, the overhead throwing motion takes about 2 seconds to complete, with nearly 75% of that time taken up by the preacceleration phases.8-11 The first phase of throwing, the wind-up, is when the body readies itself by raising the center of gravity, and the shoulder is placed in slight abduction and internal rotation. During this phase, virtually no stress is placed on the upper extremity.12-15 In the early cocking phase, the arm is placed into the abducted, externally rotated position. In addition, the arm rotates behind the body axis about 15 degrees. This phase ends at the “top” of the motion just before the beginning of forward arm and body motion. Early in this phase, the deltoid is active as it abducts the arm, followed by activity in the rotator cuff musculature to cock the arm into a more externally rotated position. The third, late cocking, phase begins as the lead leg contacts the ground and ends when the arm reaches maximal external rotation of nearly 180 degrees (Fig. 19C1-1). During this phase, the scapula retracts in order to provide a stable glenoid surface for the humeral head to compress against. The upper arm is maintained in 90 to 100 degrees of abduction, and the elbow moves even with the plane of the torso. As the humerus progresses into external rotation, the humeral head translates posteriorly on the glenoid owing to increasing tightness in the anterior structures. The external rotators (infraspinatus and teres minor) are active early in this phase, as are

the supraspinatus and deltoid. The subscapularis is active toward the end of this phase as the internal rotation of the arm begins. During this phase, the rotator cuff musculature generates a compression force of 650 N.9 The acceleration phase begins as the arm initiates its internal rotation and ends at ball release. During this phase, the arm rotates at an angular velocity greater than 7000 degrees/second.8,9 Despite this tremendous movement, little stress is noted in the shoulder musculature during this phase.14 The arm is maintained in the same abduction as in the late cocking phase. Other important muscles that are active during this phase are the triceps early on, followed by the pectoralis major and latissimus dorsi later.11,14 Some have postulated that throwing curveballs would exert a greater amount of joint load and force on the shoulder than other pitches; however, Escamilla and coworkers found that not to be the case.12 Resultant joint loads were similar between the fastball, slider, and curveball. The greatest differences in peak shoulder angular velocities occurred between the changeup and fastball pitches, with a mean change-up value of 5800 degrees/second versus a mean of 6500 degrees/second for the fastball, curveball, and slider. The change-up produced the slowest and lowest kinetics in the shoulder; therefore, the authors implied that this is likely the safest pitch to throw.16 The deceleration phase begins just after the ball is released and ends when humeral internal rotation ceases. This phase is heralded by tremendous loads generated by the rotator cuff muscles as the rapidly rotating arm is slowed to a halt.8,9 The scapula protracts while maintaining a stable glenoid surface for the humeral head. In essence, the deceleration phase is when the energy not imparted to the ball is dissipated. Eccentric loads are seen in the posterior cuff musculature as compressive joint loads exceed 1000 N.11,14 The distraction force acting on the glenohumeral joint has been shown to range from 90% to 108% of body weight 1,9,17,18 Finally, the follow-through phase concludes the throwing motion as the body regains balance and stability. During this phase, muscle firing ceases, and joint compression loads drop to 400 N.8,9 Shear loads also diminish during this phase. Stresses across the elbow joint have been measured during the overhead throwing motion. Maximal elbow velocity reaches more than 2300 degrees/second during the acceleration phase.19 Just before reaching maximal external humeral rotation in the late cocking phase, valgus torque at

Elbow and Forearm 1215

A

B

Figure 19C1-1   Adult male pitcher at the beginning (A) and end (B) of the late cocking phase of the throwing motion. This phase begins as the foot contacts the ground and ends as the arm reaches maximal external rotation.

the elbow has been measured at 64 Nm.9 Cadaveric studies have demonstrated the tensile strength of the ulnar collateral ligament (UCL) to be about 32 Nm.20 The UCL provides the most static stability against a valgus stress, taking up nearly 55% of the valgus stress at 90 degrees of elbow flexion.21,22 Because 55% of 64 Nm is greater than the 32 Nm tensile strength of the intact UCL, contributions from the bony architecture and surrounding soft tissues are needed to assist the UCL in providing medial elbow stability during the overhead throw. These contributions come mainly from the flexor carpi ulnaris, flexor carpi radialis, and the pronator teres.13,23,24 When the medial soft tissues fatigue, more stress is placed on the lateral radiocapitellar articulation as well as on the UCL. With increased compression through the radiocapitellar articulation, avascular necrosis, osteochondritis dissecans, and loose body formation may occur.25,26 Regardless, the UCL remains the primary stabilizer to the medial side of the elbow during throwing. Secondary adaptive changes can occur in the structure of the tendon in response to the bending and twisting that tendons endure. These changes are considered protective in nature and consist of increased synthesis and accumulation of the large proteoglycan aggrecan.27 It is thought that accumulation of aggrecan can protect the tendon by providing compressive stiffness, allowing collagen fascicles to slide relative to one another, and by protecting vascular elements. In the older population, adaptation is altered. Both tensile strength and stiffness decrease, although it remains unclear whether these compositional changes are adaptive or pathologic, or both.28-30 Knowledge of these “normal” adaptations is necessary when evaluating the throwing athlete. In comparison with nonthrowers, the humeral head and glenoid of the thrower are in a more retroverted position, allowing more external rotation.31 Throwers have been shown to have no significant difference in total range of shoulder rotation, but the

arc of motion is “spun back” with increased external rotation and decreased internal rotation.31,32 The soft tissues around the shoulder are affected similarly, with laxity of the anterior structures to allow for the increased external rotation and contracture of the posterior capsule preventing normal internal rotation. The thrower’s elbow may have an increased carrying angle (more valgus) as well as a loss of extension and hypertrophy of the flexor pronator muscle group.32 Up to 50% of throwers without symptoms have some degree of extension loss.32 In most cases, the motion loss is due to contracture of the soft tissues anteriorly, but in some, the cause is posterior bony apposition caused by osseous overgrowth. Increased laxity is present in the UCL in the throwing elbow versus the nonthrowing elbow in pitchers.33

EVALUATION The most important aspect of the evaluation of the throwing athlete is the history and physical examination. The history should include a precise chronologic description of the symptoms, with particular attention to the phase of the throwing motion that causes the symptoms. The individual’s pitching mechanics should be reviewed. Activity should be monitored; we believe that year-round participation in athletics contributes to the injury mechanism. Albright demonstrated that there is an increased risk for medial elbow injury in pitchers who throw with a sidearm delivery.34 Also, any changes in the routine training regimen to which the thrower had been accustomed may provide useful information to the clinician. Along with clinical symptoms, throwing symptoms such as decreased velocity, loss of control, and early fatigue are important parts of the history. Any previous treatment that has been rendered, including rest, anti-inflammatory medications, injections, physical therapy, and surgical procedures, should be documented. Any changes in the sensory function in the hand

1216 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 19C1-2   A and B, Test for internal impingement. The arm is placed in the plane of the scapula and in maximal external rotation. Impingement of the undersurface of the rotator cuff between the humeral head and the posterosuperior glenoid and the labrum occurs in this position. The patient experiences pain in the posterior aspect of the shoulder at the level of the glenohumeral joint when internal impingement is present.

and arm should be carefully noted. Tingling or discoloration in the fingers may be the first and only sign of a vascular or neurologic abnormality related to throwing. Reviewing the entire physical examination of the shoulder and elbow is beyond the scope of this chapter. Several examination tests are highlighted as part of the routine evaluation in the throwing athlete. In the shoulder, a precise range of motion should be documented for both shoulders. A careful examination for normal laxity versus instability should be performed with specific attention to discomfort experienced by the patient as the humeral head is translated anteriorly and posteriorly. The internal impingement test is done with the patient supine and the arm abducted. The arm is forced into maximal external rotation (Fig. 19C1-2).35,36 Pain in the region of the infraspinatus insertion is considered a positive test, indicating possible injury to the undersurface of the posterior rotator cuff. Internal and external rotation strength is of particular importance in the thrower.37 Impingement testing is a routine part of the thrower’s evaluation owing to the high incidence of rotator cuff disorders. Palpation and inspection of the posterior shoulder with particular attention to the infraspinatus fossa and the posterior deltoid are of particular importance in the thrower’s evaluation. Infraspinatus wasting may be the only clinical indication of a suprascapular nerve entrapment, whereas pain to palpation posteriorly in the quadrilateral space or posterior deltoid wasting may be indicative of axillary nerve compression or impingement within the quadrilateral space.38 The active compression test (O’Brien’s test) or other provocative test is performed to determine whether there is any labral disorder (Fig. 19C1-3).39 O’Brien’s test, as with other tests for labral injury, is very sensitive but limited in its specificity. Owing to the location of the labrum and the difficulty in performing direct examination, O’Brien’s test appears to be the most helpful in our hands. Mimori’s testing with

the arm abducted and externally rotated has also been shown to have an increased sensitivity and accuracy for diagnosis of superior labrum, anterior to posterior (SLAP) lesions.40 At the elbow, special attention should again be paid to the range of motion of the throwing elbow compared

Figure 19C1-3   O’Brien’s active compression test (ACT). This maneuver attempts to entrap the anterosuperior labrum between the humeral head and the glenoid. The patient is asked to place the arm in 90 degrees of forward elevation and 20 degrees of cross-body abduction. The arm is maximally internally rotated with the elbow straight. The patient is then asked to resist a downward force placed on the forearm. The occurrence of pain in the anterosuperior shoulder that re-creates the symptoms is considered a positive test result.

Elbow and Forearm 1217

ultrasound.45 Gadopentetate contrast magnetic resonance arthrography with coronal oblique fat-suppressed images has been recently shown to have a sensitivity of 84%, a specificity of 96%, a positive predictive value of 93%, and an overall reported accuracy of 91%, which was found to be superior to conventional MRI. Contrast is particularly important at the elbow when looking for UCL injury; contrast seen leaking through the ligament indicates a fullthickness injury, and contrast within the ligament indicates a partial-thickness injury. If stress fracture of the olecranon or osteochondral injury is suspected, computed tomography or bone scan may also be obtained to assist in the diagnosis and subsequent management.

SHOULDER DISORDERS IN THROWERS Figure 19C1-4   The test for valgus stability at the elbow. The arm is abducted and externally rotated as a valgus stress is placed across the elbow. Any appreciable increase in valgus opening or pain with valgus stress is considered a positive test result, indicative of possible ulnar collateral ligament rupture.

with the opposite side. Test for valgus stability should be ­performed with the patient supine and the arm abducted and externally rotated in order to limit scapular motion (Fig. 19C1-4). The elbow is flexed 20 degrees, and a valgus stress is applied. Pain in the region of the UCL or increased ­laxity with a soft end point may be indicative of UCL disorder. The test for valgus extension overload is done by repeatedly forcing the elbow into hyperextension while a valgus force is applied. This test attempts to re-­create the stress across the elbow with throwing. Pain in the posteromedial aspect of the elbow is considered a positive test and may indicate bony or soft tissue impingement in the area of the pain. After the history and physical examination are completed, additional tests for diagnostic purposes may be obtained. The first radiographic studies should include plain radiographs in the anteroposterior, lateral, oblique,41 and axial planes of the elbow. Comparative stress views, using 15 kilopascals of weight, are taken in the anteroposterior plane of both elbows to determine whether increased medial laxity exists when compared with the nonaffected side. Anteroposterior internal rotation, anteroposterior external rotation, Stryker notch, axillary, and supraspinatus outlet views are reviewed for the shoulder.42 At the elbow, the previous standard views will demonstrate bony osteophytes on the posteromedial olecranon tip as well as general joint space integrity and bony alignment. In some cases, plain radiographs may detect loose bodies within the elbow. After plain radiography, magnetic resonance imaging (MRI) is the study of choice for both the shoulder and elbow. However, newer techniques in shoulder ultrasonography have been found to provide very similar accuracy rates when compared with MRI.43,44 Wiener and Seitz reported a sensitivity of 94% and a specificity of 93% for the diagnosis of partial-thickness rotator cuff tears with

As a result of the extreme forces generated by the throwing motion, the soft tissues and bony architecture around the shoulder are susceptible to injury caused by repetitive overuse or acute trauma. The two most common types of pathologic process seen in throwers are injuries to the rotator cuff and the labrum. Other types of shoulder disease seen include osteochondral lesions of the humeral head or glenoid, osteoarthritic changes, loose body formation, gross instability, nerve entrapment syndromes, and neurovascular injuries. The following sections discuss the diagnosis and treatment of these conditions as well as the unique aspects of the throwing athlete that make the care of these injuries challenging.

Rotator Cuff Injuries As noted earlier, the rotator cuff musculature is active during various stages of the throwing motion. Owing to the high stresses experienced by the cuff, a spectrum of disorders is possible. Disorders within the rotator cuff tendons range from mild tendinitis to full-thickness tears of the cuff. Debate still exists as to the mechanism of cuff injuries. Anterior glenohumeral instability may lead to rotator cuff impingement, with repetitive throwing causing progressive injury or attenuation of the static stabilizers of the shoulder. The increased glenohumeral translation thus requires increased rotator cuff activity to constrain the shoulder.46 Certainly tension is a possible and even probable mode of failure as the cuff tendons decelerate the arm in the deceleration and follow-through phases of the throwing motion.15 Jobe and colleagues demonstrated eccentric contraction of the infraspinatus and supraspinatus during the follow-through phase of throwing.15 Compression has also been demonstrated when the arm is in maximal external rotation. In this position, the infraspinatus has been found to be compressed or pinched between the posterior superior glenoid rim and the humeral head. This situation was termed internal impingement by Walch and coworkers.47 External impingement occurs when the bursal side of the rotator cuff is pinched or abutted by some structure. Such offending structures as a thickened coracoacromial ligament, an acromial spur, or thickened bursa may cause significant impingement pain and ­dysfunction.

1218 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Tendinitis Tendinitis is, by definition, inflammation of a tendon. In many cases of tendinitis, it is the tendon sheath that is actually inflamed rather than the tendon itself. In the case of the rotator cuff, the term tendinitis is frequently used when a patient has shoulder pain with activity but no obvious detachment of the cuff. In these cases, tendon inflammation is likely not present, but rather inflammation in the subacromial bursa, which is termed bursitis. Tendinosis implies true intratendinous disease, such as intrasubstance degeneration or tearing. The careless use of these terms is not insignificant in the case of a thrower because of the potential for catastrophic failure of the cuff, which may disable the player completely. The hallmarks of tendinitis or tendinosis of the rotator cuff are pain with overhead activity and weakness secondary to the pain. In the thrower, the symptoms are pain at the top of the motion when the arm is in maximal external rotation or pain after ball release as the cuff is trying to slow the arm down. Weakness of the supraspinatus or infraspinatus, or both, is a frequent finding. In most cases, the athlete does not recall a specific throw when the shoulder became sore. Rather, many notice the pain as they try to warm up and then are unable to get rid of the pain as they progress to full speed. In almost all cases, tendinitis in the thrower represents an overuse injury rather than an acute trauma. For this reason, the first act of the treating physician is to rest the player. For mild tendinitis, short-term rest (3 to 5 days), along with oral anti-inflammatory medication and a rehabilitation program designed to strengthen the cuff muscles, is enough to eliminate the symptoms. With more severe or recurrent tendinitis, more prolonged rest from throwing along with extended cuff rehabilitation may be necessary. In refractory cases, subacromial injection of corticosteroid may be considered, although this should be the last conservative option.

Full-Thickness Tears Full-thickness rotator cuff injuries are rare in the throwing population. However, natural history of progression has been shown; therefore, surgical treatment is recommended.48 In some cases, the injury occurs as a result of a nonthrowing trauma to the throwing arm. Regardless of the mechanism, full-thickness rotator cuff tear is an absolute indication for surgery. The literature remains inconclusive on whether the best approach for cuff repairs in the throwing shoulder is through a mini open or arthroscopic repair. Our trend of late has been migration to arthroscopic repair. Mazoué and Andrews reported only an 8% return to play rate in professional baseball players with full-­thickness repairs undergoing mini open treatment. In a study by Tibone and coworkers, only 55% of overhead athletes requiring open rotator cuff repair were able to return-totheir previous level of competition.49 With the advent of and increasing experience with arthroscopic repair, as well as the decreased morbidity associated with arthroscopic techniques, fewer open procedures are being performed on these athletes. It is imperative at the time of repair to achieve as strong a repair as possible and to ensure that there is no impingement on the cuff by the acromion, coracoacromial ligament, or distal clavicle. Emphasis should be

given to restoring the anatomic footprint of the supraspinatus, which has been shown to insert along the anterior 2 cm of the greater tuberosity.50 The postoperative course for these patients is very much the same as with the partial-thickness injuries. The early portion of the rehabilitation is spent regaining motion, followed by strength gains, followed by return to throwing using an interval throwing program.

Partial-Thickness Tears Partial-thickness tears generally occur on the articular side of the rotator cuff, although bursal-sided tears have been seen. Biomechanical tests of the supraspinatus tendon reveal that the bursal-side layer has the greatest deformation and tensile strength and is more vulnerable to a tensile load than the bursal-side layer. The articular surface of the rotator cuff was shown to have an ultimate stress to failure of 6.3 N/mm2, about half that of the bursal surface at 2.8 N/mm2. The ultimate failing stress of the supraspinatus was shown to be about 9 N/mm2.51 Itoi and colleagues showed that the anterior strip of the supraspinatus had a greater modulus of elasticity and is mechanically stronger than other portions of the supraspinatus.52 Other studies have shown that tendon degeneration can weaken the supraspinatus and that incomplete tears at the articular surface act as stress risers.52-55 In the cases of internal impingement, the deepest fibers of the infraspinatus become entrapped between the glenoid rim and the humeral head as the arm moves into hyperexternal rotation. This impingement causes fraying of these deep fibers, leading to recurrent pain at the top of the throwing motion.47,56-59 In contrast, some attribute internal impingement to posterosuperior subluxation of the humerus from posterior capsule contracture.60 Players report pain in the back of their shoulder as opposed to pain on the top of their shoulder with this condition. Partial-thickness tears of the supraspinatus are also seen in throwing athletes, and these patients tend to complain of pain on top of the shoulder. In treating partial-thickness tears, it is important to determine, as accurately as possible, the extent of the tear. As noted earlier, MRI with or without contrast is the study of choice in these cases. Although there are no absolute guidelines as to the surgical indications for throwers with partial-thickness tears, it is always wise to begin with conservative management. If MRI fails to demonstrate a fullthickness tear, rest and rehabilitation, along with modalities and anti-inflammatory medications, should be initiated. If the player fails to improve or return to throwing after a 6- to 12-week period, arthroscopy should be entertained. At the time of arthroscopy, a better determination regarding the thickness of the rotator cuff injury should be made. Protocol has been suggested based on the extent of tear with an examination of the footprint of the supraspinatus for articular-sided tears and direct visualization for bursal-sided tears. The mean thickness of the supraspinatus tendon has been shown to be about 12 mm.61 We use an arthroscopic probe to assist in palpation of the cuff remaining, but most importantly, gauge the amount of exposed footprint from inside the joint. Currently, our treatment for partial-thickness cuff tears involving less than 25% of the cuff is débridement

Elbow and Forearm 1219

Figure 19C1-5   Anchor and suture placement during partial-thickness rotator cuff repair.

of the undersurface tendinosis tissue to stimulate bleeding without acromioplasty. Débridement likely removes inflammatory cells and inflammatory mediators present in the torn rotator cuff tissue.47,62,63 For tears involving about 25% to 75% of tendon, we arthroscopically repair to the footprint using suture anchors in a technique similar to one described by Snyder.64 Excellent results using this method, which restores the medial footprint of the cuff while avoiding a length-tendon mismatch, have been recently published.65 Tears of more than 75% of the tendon are completed and treated as a full-thickness tear with repair of the cuff laminae using arthroscopic or mini open technique. This technique is also appropriate for those tears that have failed attempts at débridement and decompressive procedures and for those patients with significant shoulder instability. A conservative acromioplasty is performed in tears involving more than 75% of the tendon. Care must be taken during the acromioplasty to avoid release of the coracoacromial ligament and detachment of the anterior deltoid (Fig. 19C1-5).66-68 Mazoué and Andrews reported 85% good and excellent results with arthroscopic débridement of partial-­thickness tears in 34 young athletes with a mean follow-up of 13 months.69 Conway reported an 89% return to the same or higher level of play in a group of 14 baseball players who underwent repair using an intratendinous technique for partial cuff tears.70 Payne and associates evaluated 43 athletes younger than 40 years with partial-thickness tears treated with arthroscopic débridement and subacromial decompression.71 A satisfactory outcome was reported 86% of the time in acute, traumatic injuries; however, a return to preinjury sports was only 64%. An insidious onset of pain was a poor predictor of results, with only a 66% satisfaction rate and a 45% return to preinjury sports. Unpublished data obtained at our institution reviewed preoperative and intraoperative findings on 82 professional pitchers who had undergone débridement of a partial-thickness rotator cuff tear. This review included return-to-play data on 67 players, finding that 76% were able to return to competitive pitching at the professional level after débridement. A total of 55% were able to return to the same or higher level (Major or Minor League) of

competition.72 Other data obtained on professional Major and Minor League throwers suggest that 59% with an arthroscopic partial-thickness rotator cuff repair were able to return to professional baseball; however, only 29% returned at the same level or higher. Position players in this review had a greater percentage of return to play and a greater percentage of return to play at the same or higher level.73 These data should not be interpreted on a whole but examined independently, noting that the consequence of injury requiring repair (greater tear) has a less favorable outcome than those injuries requiring débridement only. In cases of internal impingement and labral ­ fraying, débridement of the frayed tissue is usually enough to eliminate symptoms in nonthrowers, as noted earlier. Simple débridement, however, will not address the rotational instability that most throwers with shoulder disease possess. The results of shoulder arthroscopy with ­débridement alone in the throwing population have been met with lesser rates of success. In Andrews and associates’ initial report of 73 pitchers, 76% had an excellent result, but follow-up was limited to only 13 months.74 A later study by Payne with longer follow-up demonstrated less than 50% return to competition in a similar population with associated instability.71 Therefore, in most cases of internal impingement and labral disorders, arthroscopic stabilization using thermal capsular shrinkage or capsular plication is carried out in the anterior, posterior, and inferior aspects of the joint capsule as indicated. Treatment of instability in the throwing athlete is a delicate matter owing to the balance that must be maintained between motion and stability. Throwers have increased external rotation owing to both bony and soft tissue factors mentioned earlier. Some throwers, however, develop pathologic instability that leads to increased stress on the capsule and rotator cuff tissues without frank dislocation. This subtle increase in instability can be demonstrated by an increase in humeral translation on the glenoid in either the anteroposterior or the inferior direction. Regardless of the direction, the instability must be addressed in order to correct the underlying pathologic process. Although open stabilization procedures certainly accomplish the goal of obtaining stability, the subsequent rate of return to competitive throwing is significantly decreased when compared with arthroscopic means of regaining stability.75,76 At this institution, the amount of capsular shrinkage has decreased over the past few years without an apparent detriment to outcome. The postoperative treatment of throwing athletes is just as important as the procedure itself. In general, the first 4 to 6 weeks after surgery are spent regaining motion, after which strength in the rotator cuff is regained. By 3 to 4 months after the procedure, the athletes are ready to begin an interval throwing program under supervision by a trainer or therapist. Once the interval throwing program is completed, the player is ready to return to competition. The time for return to competition varies with pathologic process and procedure as well as from individual to individual.

Labrum Injuries The labrum is a fibrocartilaginous ring around the bony glenoid rim that not only deepens the socket but also serves as the attachment site for the glenohumeral ligaments and

1220 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

biceps tendon. Although compression of the labrum against the glenoid by the humeral head certainly occurs, it is more likely in throwers that the disorders seen in the labrum are due to failure in tension related to the biceps tendon and glenohumeral ligaments. Morgan and colleagues described the “peel-back” phenomenon, during which the biceps tendon rotates behind the axis of the humeral head, leading to a tensile stress on the biceps anchor and superior labrum.77 Posterior labral fraying is common in throwers, likely owing to compression as is seen in internal impingement of the infraspinatus tendon, discussed earlier. Complete disruption of the posterior labrum away from the glenoid rim, however, is less common than its anterior counterpart. As noted earlier, the physical examination tests for labral disorders are less specific than they are sensitive. We prefer the active compression test (O’Brien’s test) and Mimori’s testing for physical diagnostic purposes, along with confirmation by contrast-enhanced MRI. In the absence of MRI findings, conservative management is usually initiated. If symptoms persist after adequate rest and rehabilitation, shoulder arthroscopy may be indicated. At the time of arthroscopy, examination under anesthesia should also be performed to gauge the laxity in the soft tissues. The labrum should be inspected and probed thoroughly along its entire course around the glenoid rim. Several normal anatomic variants have been described, particularly involving the anterior labrum. Knowledge of these variants is mandatory to avoid overconstraining the shoulder by repairing a normal structure. Among the more common anatomic variants is the sublabral foramen. This hole beneath the labrum is most commonly seen at the anterior labrum just below the biceps anchor. No true labral detachment is present, with no evidence of trauma or labral detachment. No increased laxity is attributed to this condition, and the sublabral foramen should not be closed. If a true labral detachment is noted, arthroscopic repair is the procedure of choice, inasmuch as débridement alone has led to disappointing results.78-81 Obviously, anterior lesions can be treated through the standard anterior portal; however, superior lesions are more difficult to repair through the anterior cannula. This is due to the angle of insertion for the anchor or other fixation device that is used. For this reason, we routinely use a small permanent anchor placed through a 4-mm trocar using a direct lateral stab incision. By doing this, the smallest possible defect is made in the rotator cuff without compromising the arthroscopic repair and fixation. Regardless of which device or approach is used, arthroscopic fixation should be as firm as possible. Before placing any anchor or tack, débridement of the glenoid rim down to bleeding bone should be undertaken. The fixation should be at the articular margin, not recessed on the glenoid neck. Fixation on the glenoid neck, as opposed to the correct periarticular location, may lead to instability and subsequent reinjury. Careful inspection of the superior labrum should be carried out before repair to ensure that the attachment is in fact detached. The normal superior labrum may be recessed or “meniscoid,” with its fibers attaching medial to the articular surface. This condition should not be mistaken for a true labral tear. Needle localization is used to develop the anterior portal in the rotator interval. A 7-mm threaded cannula positioned in the anterior portal is used for bony débridement, the

passage of suture retrieving instruments, and knot tying. A “near-lateral” percutaneous portal is used directly off the lateral border of the acromion for passage of the anchor and drill guide. Needle localization is used to optimize placement of the bioabsorbable anchor. A No. 11 blade is passed through the skin and into the supraspinatus, making a small 5-mm defect in the rotator cuff. The drill guide or trocar and then anchor are passed through the cuff onto the glenoid. We currently use the Weston knot for most of our arthroscopic knot tying.82 The “far-lateral” portal can be used for subacromial work and cuff repair as indicated (Fig. 19C1-6). Generally, one double-loaded suture anchor is placed below the insertion of the biceps tendon. One loop of suture is tied posterior and one tied in front of the insertion. The strand not being tied at the time is left outside the near-lateral acromial portal. If another point of fixation is needed, a single-loaded suture anchor is used. Knots are tied using the most medial strand as the post; therefore, the knots will lie away from the glenoid. Fraying of the biceps tendon is also seen at the time of arthroscopy in the throwing population. The mechanism of injury to the biceps tendon is not entirely clear at this time. The frayed tissue should be débrided during arthroscopy with caution so as not to completely amputate the tendon. Synovitis in the anterior recess of the shoulder can also be débrided. True traumatic dislocation is rare in the throwing shoulder. In most cases of frank dislocation, the injury occurred as the result of a nonthrowing action. Regardless of mechanism, it is of paramount importance to regain the stability and strength necessary to throw. For the overhead thrower with a dislocation history, it is possible to regain

Figure 19C1-6   Lateral decubitus position using the accessory far-lateral portal for superior labrum from anterior to posterior (SLAP) anchor placement.

Elbow and Forearm 1221

full ­function using conservative treatment, as would be the case for many simple shoulder dislocations. Many, however, require operative fixation of the detached structures, usually the anterior labrum and capsule. With increasing frequency, these procedures are performed arthroscopically, although no clinical data are available in the throwing population. It is reasonable to expect less difficulty with postsurgical scarring and range of motion with arthroscopic techniques than with traditional open techniques in this patient population; however, this remains to be proved clinically.

Other Shoulder Disorders Suprascapular Nerve Entrapment In some cases, wasting of the infraspinatus may be noted when compared with the nonthrowing shoulder. In these cases, suprascapular nerve entrapment or injury must be suspected. Although this represents a pathologic condition, many throwers with infraspinatus wasting are completely asymptomatic and do not warrant intervention. If, in contrast, the patient has symptoms of nerve entrapment such as pain, weakness, or decreased ability to perform, it is reasonable to perform decompression of the suprascapular nerve. Spinoglenoid notch cysts responded significantly better to operative treatment, and the results for open surgery were the same as the results for arthroscopic decompression. In addition, compressive lesions attributable to suprascapular notch entrapment had the best improvement with surgical decompression.83 MRI should be obtained before surgery to determine whether there is any cyst or other structure of articular origin that may be causing the compression. In either case, shoulder arthroscopy should be carried out at the time of nerve decompression to be certain that no perilabral cyst exists. If a cyst is present, it should be removed. Otherwise, simple decompression of the suprascapular nerve using spinoglenoid notchplasty is generally sufficient to relieve symptoms.84

ELBOW DISORDERS IN THROWERS As with the shoulder, tremendous forces are generated at the elbow during the overhead throwing motion. The soft tissues around the elbow, along with the bony articulations, provide stability to this complex joint. The UCL, specifically the anterior band, provides the most static stability at the elbow joint to oppose a valgus stress.21,85 The discrepancy between the tensile strength of the UCL and the forces imparted to the ligament during the throwing motion demonstrates the importance of the dynamic muscular stabilizers as well as the bony contributions to the resistance to valgus stress. Not surprisingly, flexor tendinitis and partial- and full-thickness injuries to the UCL are common in throwers. Also, bony disease is seen, particularly in the radiocapitellar articulation and at the posteromedial aspect of the olecranon. Although some players will not recall any specific throw that initiated their symptoms, an equal number are able to

pinpoint exactly when symptoms began. Players may complain of numbness or tingling in the ulnar nerve distribution with medial elbow injuries, owing to the proximity of the nerve. Also, any increase in valgus opening at the elbow increases the stress placed on the nerve, which can cause paresthesias.

Tendinitis Tendon injuries around the elbow range from minor inflammation to complete rupture of the flexor musculature.86-89 Flexor tendinitis and tendinosis are common in throwers owing to the muscles’ dynamic contribution to elbow stability during the valgus torque produced while throwing.24 The flexor carpi ulnaris has been determined to be the primary dynamic stabilizer against valgus torque.24 In such cases of tendinitis, patients have pain to palpation of the flexor-pronator mass just distal to the common tendon origin from the medial epicondyle as well as pain with a valgus stress. Careful attention to the location of pain will likely distinguish tendinitis from UCL injury, which most often demonstrates tenderness posterior and distal to the pain of flexor tendon injury along the anterior band of the UCL. Resisted pronation of the forearm and wrist volar flexion will also elicit pain in the region with tendinitis, but rarely in cases of UCL damage. Triceps tendinitis is also seen in throwers owing to the rapid extension required during the acceleration phase. In these cases, the pain is easily traced to the distal triceps tendon and insertion onto the olecranon. In addition, resisted elbow extension should elicit symptoms. As with other locations of tendinitis, the treatment of tendinitis around the elbow begins with rest from throwing along with anti-inflammatory medication and a stretching and strengthening program. Therapeutic modalities and corticosteroid injection have also demonstrated some beneficial effect. In general, corticosteroid injection is reserved for recurrent or recalcitrant tendinitis cases. With aggressive stretching and strengthening, most players return to competition after 1 to 3 weeks of treatment. In more chronic or recurrent cases, the required period of rest and treatment may be prolonged. In the few cases that fail conservative measures, operative débridement of the medial flexor tendon origin or triceps insertion can be performed through an open approach. Similarly, in cases of chronic resistant tendinosis, we prefer open débridement and repair of the flexor-pronator mass. In these cases, we reattach the flexor tendon to the medial epicondyle using a suture anchor. Although the results of operative treatment are excellent, most cases do not require this type of management. In fact, if nonoperative methods fail and surgery is thought necessary, more serious injury, such as UCL insufficiency, must be considered.

Ulnar Collateral Ligament Injuries The UCL is the most commonly injured ligamentous structure in the thrower’s elbow for the reasons described earlier. The spectrum of UCL injury ranges from minimal fraying to complete ligament disruption and frank instability.34,90-93 The distinction between partial-thickness and full-thickness injuries is difficult to make on clinical

1222 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

e­ xamination in the awake patient. In cases of complete ligament disruption with frank elbow instability or dislocation, the diagnosis of UCL injury may be easier on clinical examination. Unfortunately, most throwers do not fall into this category. A complete clinical history, along with appropriate physical findings and supporting radiographic studies, is usually sufficient to make the correct diagnosis. It is also important to note that not all full-thickness injuries require reconstruction and that not all partial-­thickness injuries will be treated successfully without surgery. The symptoms and clinical history are the most important facets of the decision-making process for the clinician. Partial-thickness injuries to the undersurface of the UCL and intrasubstance degeneration occur in the throwing population.93 It is reasonable to suspect that these lesions are precursors to full-thickness injury. In the thrower who has no frank instability or increased medial opening with a valgus stress but has clinical symptoms of a UCL injury, MRI with contrast is the diagnostic study of choice. Partial-thickness injury or edema within ligament is easily detected with MRI.94 Dynamic ultrasonography has also been found to demonstrate UCL pathology; however, its use in the clinical setting is not yet common.95,96 In cases of partial-thickness injury, short-term active rest (3 to 6 weeks), followed by a supervised return to throwing in an interval throwing program, is the treatment of choice. Anti-inflammatory medications, as well as other modalities, may be used during this time. If the player cannot return to throwing without pain after an adequate rest period and throwing program, prolonged rest (3 to 6 months) may be indicated. In professional athletes and in many amateur, college, and even high school athletes, these periods of prolonged rest are not well tolerated because of the rigorous schedules imposed on such players. In players who have recurring bouts of difficulty with partial-thickess injury or fail to improve with adequate conservative ­ management, UCL reconstruction may be considered. If, conversely, the diagnosis of full-thickness injury is made, it is less likely that conservative management will lead to favorable results in the throwing athlete. UCL reconstruction may be considered in these players without the need for prolonged rest and rehabilitation in an attempt to heal the ligament injury. It is important to note at this point that the UCL is important for the purpose of throwing in athletics. Rarely is it necessary to reconstruct the nondominant UCL or to perform reconstruction in nonthrowing athletes or in those who wish to discontinue their throwing activities. Nonoperative treatment is generally successful in this lower-demand population.97 ­ However,

even for throwing athletes, functional expectations must be considered. At times, a football or position baseball player can perform at an acceptable level with more significant UCL pathology. Conservative treatment may be successful. Conversely, a high-demand thrower, such as a pitcher, is much more likely to require UCL reconstruction, even with a more minor partial injury. These athletes do not respond well to nonoperative management.98,99 After the decision is made to proceed with ligament reconstruction, the choice of graft site must be made. In general, we prefer the ipsilateral palmaris longus tendon if present. In the general population, 85% of people have a palmaris longus. It is imperative at the time of physical examination and preoperative counseling that the presence or absence of a palmaris longus tendon is documented on either the operative or the nonoperative side. Reconfirmation just before induction of anesthesia is a routine part of our practice in order to avoid any graft-related complications. Because 15% of the normal population has no palmaris longus tendon, other graft choices are necessary. In our practice, a gracilis tendon is harvested from the opposite leg. Other graft choices include the plantaris tendon, lesser toe extensor tendon, a portion of the Achilles tendon, or allograft tendon.

Techniques Jobe92 first described the technique for UCL reconstruction using a free tendon graft in figure-eight fashion through bone tunnels, suturing the graft to itself. The flexor­pronator mass was detached from the medial epicondyle, and submuscular ulnar nerve transposition was performed. Since that time, several modifications have been made to Jobe’s original method; however, the basic tenets of the procedure remain the same.90 Dr. Andrews, as detailed in a report by Azar and colleagues,100 modified the procedure by elevating the flexor-pronator mass without detaching it from the medial epicondyle and by performing a subcutaneous versus submuscular ulnar nerve transposition. Dr. Yocum, in a report by Thompson101 and associates, modified the technique by performing a muscle-splitting approach without ulnar nerve transposition. Rohrbough and coauthorsl83 described how Dr. Altchek modified his technique through a muscle-splitting approach with a single humeral tunnel and suture fixation called the docking procedure. Finally, Dr. ElAttrache’s technique, outlined by Ahmad and associates,102 involves a single tunnel in the medial epicondyle and sublime tubercle with graft fixation achieved with soft tissue interference screws.

Authors’ Preferred Method Our preferred method for UCL reconstruction requires the patient to be positioned supine with the arm on a hand table. A nonsterile tourniquet is inflated for the case. The incision is made extending from 4 cm above the medial epicondyle to 6 cm distal to it. The medial antebrachial cutaneous nerve is identified and protected during the case. The ulnar nerve is isolated above and below the medial epicondyle. The ­medial

intermuscular septum is released to prevent tethering of the nerve when transposed. Next, the flexor-pronator mass is elevated off the UCL until the UCL can be completely visualized. The flexor-pronator mass is, however, left attached to the medial epicondyle. The native ligament is split longitudinally in line with its fibers to expose the underlying ­ulnohumeral joint. The torn or degenerative ligament tissue

Elbow and Forearm 1223

Authors’ preferred method—cont’d can be débrided. A 9⁄64 drill bit is then used to drill two connecting holes in the sublime tubercle of the proximal ulna, one from medial to lateral, and one from anterior to posterior, leaving an adequate bone bridge. Curved curets can be used to connect the two holes if necessary. The same drill is used to drill two converging holes in the medial epicondyle, one from proximal to distal, and one from medial to distal. These two holes should converge to exit the epicondyle at the origin of the UCL. Straight curets can be used to connect these holes if necessary. Suture passers are then used to pull suture loops through these tunnels. The distal end of the native ligament is repaired side to side using nonabsorbable suture, leaving the proximal-most ligament unrepaired. By closing the native ligament, the joint surfaces are covered, protecting the graft from abrasion. The proximal ­ligament

The “docking procedure” for UCL reconstruction, described by Rohrbough and colleagues,83 involves passing the graft through the ulna in the same fashion as described earlier; however, the medial epicondyle is reamed from distal to proximal using a 3-mm round bur without violating the proximal aspect of the medial epicondyle. A small dental drill is then used to make two small holes in the epicondyle connecting to the larger bur hole. The sutures from the graft are brought through these two drill holes through the tunnel and tied on the outer surface of the epicondyle. This technique does not typically include ulnar nerve transposition and has been noted to involve less soft tissue dissection than the standard technique as first described by Jobe. Several recent studies have compared versions of these modified techniques for UCL reconstruction.103-107 Armstrong and associates103 reported the docking technique and an EndoButton technique to be stronger than the interference screw or figure-eight technique, with all four techniques demonstrating inferior peak load to failure compared with an intact ligament. A comprehensive literature review by Langer and colleagues104 revealed that decreased dissection of the flexor-pronator mass and decreased handling of the ulnar nerve lead to improved outcomes. Paletta and coworkers105 found that neither the docking nor figure-eight technique reproduced the biomechanical profile of the native UCL, but did suggest that the docking construct may offer an initial biomechanical advantage over the figure-eight construct. A study that compared the effects of cyclic valgus loading on the docking and interference screw procedures, by McAdams and associates,106 demonstrated increased valgus angle widening with the docking technique after 10 and 100 cycles, but no difference between the two techniques at 1000 cycles. Finally, Large and colleagues107 biomechanically compared the transosseous figure-eight and interference screw reconstructions. Their study showed that failure strength, as well as initial and overall stiffness of the transosseous figure-eight technique, was superior to the interference screw procedure. The UCL originates on the distal aspect of the medial epicondyle and inserts onto the sublime tubercle of the

is left without repair in order to visualize the entrance to the epicondylar tunnels. If the native ligament was torn away from either insertion, sutures can be placed in the leading edges of the ligament for the purpose of repair through the tunnels for the graft. The graft of choice is then harvested. The minimal graft length needed is 12 cm. Nonabsorbable braided sutures are placed in each end of the graft tissue in Bunnell or Krakow fashion. The graft is then passed through the tunnels using suture passers and sewn to itself. We routinely perform subcutaneous transposition of the ulnar nerve at the time of UCL reconstruction. The nerve is held in place with a single fascial sling from either the flexor­pronator muscle fascia or the released intermuscular septum left attached to the medial epicondyle. The sling is tensioned loosely to prevent any compression of the ulnar nerve.

medial proximal ulna. Many adult throwers have either a bony extension from the sublime tubercle directed proximally along the course of the ligament or one or more small ossicles of bone within the ligament (Fig. 19C1-7). In most cases, the bony ossicles represent old avulsion fractures, likely occurring during the childhood or adolescent years. The same may be true of the development of a bony extension from the sublime tubercle, although the exact cause of this finding is uncertain. These bony findings are critical in the treatment of elbow instability in throwers. Because

Figure 19C1-7   Magnetic resonance image with contrast of the elbow demonstrates bony osteophyte of the medial proximal ulna. In such cases, the ulnar collateral ligament attaches to the tip of the osteophyte. The bone is acting in a tension mode, which places it at risk for failure. If the ulnar collateral ligament is reconstructed in individuals with osteophytes, such as the one pictured here (arrow), it is important to obtain a graft that can replace the native ligament because the distal ligament is deficient after removal of the bony prominence.

1224 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

bone is stronger in compression than in tension, these bony abnormalities will create a weak link in the medial stabilizing tissues. Both the ossicles and the bony extension are actually within the substance of the UCL. In the case of the bony projection from the ulna, the terminal distal fibers of the UCL actually insert onto the prominence. The presence of these types of bony disease may make the decision to proceed with UCL reconstruction somewhat less difficult. If the decision is made to perform UCL reconstruction in these patients, a larger graft should be used if possible. This is necessary because of the soft tissue deficiency that will result from débridement of the bony disease from within the native ligament. In the absence of such a deficiency, the native ligament can be left in place or reattached at the site of rupture along with ligament reconstruction, creating significant tissue thickness at the UCL site. In fact, the native ligament is usually placed beneath the graft tissue to protect the graft from abrasion by the joint surfaces. If the native ligament cannot be retained or if it is insufficient, a larger thickness graft should be used to make up for the missing tissue. We prefer the gracilis tendon in such cases.

Results In Jobe’s original report, 10 of 16 (63%) throwing athletes were able to return to their previous level of competition.92 In Conway’s group of 70 patients, only 50% of the athletes who underwent repair of the ligament returned to competition, compared with 68% (45 of 56) of those who underwent UCL reconstruction.90 Later, Andrews reported on 72 professional baseball players with elbow disorders, 14 of whom underwent UCL reconstruction, with 12 of those (86%) returning to their previous level of competition.108 Azar and Andrews then reported on 67 patients, with 81% (48 of 59) of those undergoing UCL reconstruction returning to the same or higher level of play.100 Thompson and colleageues101 reported that 93% of 33 patients had an excellent result with UCL reconstruction. In a study by Paletta and Wright,109 92% of 25 professional and collegiate baseball players were able to return to their preinjury level of competition. Ninety percent of 100 patients evaluated by Dodson and coworkers110 were able to compete at the same or a higher level.

Valgus Extension Overload and Loose Bodies UCL injury is not the only cause of medial elbow pain in the throwing athlete. Valgus extension overload (VEO) is a condition most commonly seen in throwers in which the valgus stress across the elbow causes impingement of the posteromedial olecranon tip against the medial wall of the olecranon fossa. With repeated impingement, a bony osteophyte may grow on the olecranon at the site of impingement. Bony growth within the olecranon fossa has also been seen. In addition, soft tissue, including the synovium of the elbow joint, may become hypertrophied with repeated impingement at the same location, leading to additional symptoms and swelling. With VEO, throwers typically complain of posteromedial elbow pain at the initiation of the acceleration phase of throwing.26 The

­ istinction between this condition and UCL injury is difd ficult but can usually be made by carefully determining the exact location of the pain that the patient experiences. With VEO, the pain of greatest magnitude is typically more proximal with direct palpation of the posterior medial tip of the olecranon. The provocative test for VEO was described earlier and is an important aspect of this diagnosis. In the case of UCL injury, the VEO test may also elicit pain, which is generally is located more distal in the area of the UCL. After the diagnosis of VEO has been made, a conservative treatment protocol similar to those described earlier should be instituted. If conservative measures fail, arthroscopic evaluation and removal of the impinging bony structures can be undertaken. At arthroscopy, the patient is positioned supine with the arm suspended from a boom. The shoulder should be abducted 60 to 70 degrees, with the elbow flexed to 90 degrees. The elbow is insufflated through the lateral soft spot, and an anterolateral portal is established. When the anterior joint space is visualized, an anteromedial portal may be established if any débridement is necessary. In throwing athletes, the anteromedial portal is not established unless it is absolutely necessary because of complaints of postoperative pain and stiffness related to this portal. After the anterior work has been completed, a lateral “soft spot” portal may be established using a small 2.7-mm arthroscope. Through this portal, the lateral radiocapitellar joint can be visualized, and a second, more distal, lateral portal can be established if any débridement or loose body removal is necessary. With the scope in the lateral portal, the posterior joint space can be visualized, and a regular arthroscopic cannula can be inserted lateral to the triceps tendon, directed toward the olecranon fossa. When the arthroscope is switched to this portal, a posteromedial portal can be established, with careful attention to avoiding the ulnar nerve. Posteromedial osteophytes on the olecranon tip and within the olecranon fossa can then be removed using an osteotome and high-speed bur. Careful attention must be paid to avoid over-resection of the olecranon tip. Removing too much bone may lead to increased stress on the soft tissue stabilizers around the elbow owing to decreased bony stability. No more than 4 to 8 mm of bone should be resected.111 After the resection, an intraoperative lateral radiograph of the elbow is obtained to ensure that the level of resection is adequate and that all loose fragments have been removed. After elbow arthroscopy for the treatment of VEO or loose body removal, immediate range of motion exercises are initiated. When range of motion has returned, strengthening of the upper extremity musculature and functional exercises are initiated. Return to rehabilitation throwing following these procedures is usually possible by 6 to 8 weeks, with an accelerated throwing program prescribed. When the throwing program is completed, the athlete may return to competition, which typically takes 3 to 4 months. In 1998, Hepler reported on 28 patients undergoing arthroscopic débridement of the posteromedial osteophyte with greater than 90% success based on objective and subjective data.112 Andrews and Timmerman reported on 56 professional baseball players who underwent excision of the posteromedial olecranon osteophyte

Elbow and Forearm 1225

either as an isolated arthroscopic procedure or as a part of UCL reconstruction.108 In their report, 68% returned to play at least one season; however, 41% required reoperation. The authors concluded that arthroscopic débridement was superior to open techniques, but warned against over-resection, which may eventually lead to medial instability. Finally, Reddy and colleagues reported on 187 elbow arthroscopies, noting that posterior olecranon impingement was the most common diagnosis (51%).113 In their series, 47 of 55 (85%) professional athletes returned to their previous level of competition. They also noted that players with either loose bodies or VEO tended to have better results than did those with degenerative disease in the elbow.

Other Elbow Disorders Olecranon Stress Fracture Fractures around the elbow are less common in adult throwers than in adolescents and children. Olecranon stress fractures, however, are not uncommon in adults. The ulnar olecranon is the most common stress fracture site among baseball athletes.114 These stress fractures typically present with pain during the acceleration phase of throwing that is localized to the posterior and sometimes the lateral border of the ulna at the level of the olecranon articular surface. Physical examination in these cases may not demonstrate pain with a valgus stress. These throwers will have point tenderness to palpation over the affected site. In some cases, these stress fractures can be seen on plain radiographs (Fig. 19C1-8). Computed tomography or bone scan is the study of choice in these cases if there is any question about the diagnosis (Fig. 19C1-9). No large clinical series exists in throwing athletes with this condition. Several anecdotal reports have

Figure 19C1-8   Plain lateral radiograph of the elbow in a throwing athlete with posterior elbow pain. The arrow indicates lucency through the proximal olecranon consistent with a stress fracture. Typically, these types of injury cause pain with rapid extension of the elbow, as seen in throwers. Also, pain on palpation of the stress fracture site is typically present.

Figure 19C1-9   Computed tomographic scan of the same elbow depicted in Figure 19C1-8. This scan demonstrates the olecranon stress fracture in this thrower (arrow).

­ emonstrated some success with both conservative and d operative means.115-118 The treatment of stress fractures of the olecranon in our clinic begins with rest from throwing along with strictly enforced lifting restrictions. Rotator cuff exercises, along with plyometrics and light triceps and biceps strengthening, can be initiated once the point tenderness ceases. If the player can progress through these rehabilitation modes, the interval throwing program may be entered and progressed through in the standard fashion. If symptoms persist, surgical intervention may be considered. A single axial large cannulated screw across the fracture site inserted through the distal triceps tendon is the treatment of choice in these cases (Fig. 19C1-10). Early range of motion exercises are instituted, followed by strengthening and return to throwing when healing is apparent. Screw removal is entertained only if symptoms at the insertion site are present once the fracture is healed. Not all throwers will require screw removal.

Figure 19C1-10   After failure of conservative management, percutaneous cannulated screw placement was performed.

1226 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Snapping Triceps Tendon Pathologic bands of the distal triceps tendon have been described on the medial side of the elbow.119,120 These bands can snap over the medial epicondyle, creating a “pop” that is both palpable and sometimes audible. Ulnar nerve irritation caused by the motion of these bands is possible.119,120 Although conservative management may decrease the inflammatory component, in many cases surgical resection of the bands is necessary to completely alleviate symptoms. No permanent triceps strength loss or ulnar nerve disorders should be expected in these patients after excision of the pathologic bands.

Ulnar Neuritis and Nerve Subluxation Chronic irritation of the ulnar nerve related to repeated valgus stress at the elbow has been seen in throwing athletes with medial elbow instability.90 A significant amount of stress is placed on the ulnar nerve during the throwing motion. A study by Aoko and colleagues121 found the maximal strain on the ulnar nerve during the acceleration phase to be close to the elastic and circulatory limits of the nerve. Besides traction, other factors thought to contribute include friction, compression, or the presence of an anconeus epitrochlearis muscle over the cubital tunnel.122,123 In some of these cases, the ulnar nerve is unstable and can easily be dislocated from the cubital tunnel.124,125 In others, the nerve is stable but tender to palpation above and within the cubital tunnel. If short-term rest, anti-inflammatory medications, and physical therapy fail to improve symptoms, ulnar nerve transposition can be considered. At the time of surgery, it is important to resect the intermuscular septum from the distal humerus to prevent tethering of the nerve when transposed and to isolate the nerve above and below the medial epicondyle. The motor branches to the flexor carpi ulnaris must be protected. The cubital tunnel should be closed, and a fascial sling may be used to prevent return of the nerve to its native position. When the wound is healed, strengthening followed by return to throwing can be accomplished.

OTHER DISORDERS IN THROWERS The most common types of disorders seen in throwing athletes have already been discussed. However, several less common injuries warrant attention owing to the potential for loss of playing time and loss of function. Among these injuries, humeral shaft fractures are perhaps the most serious. In one reported case of a professional baseball pitcher, a malignant bone tumor was the underlying cause. Other humeral shaft fractures have been reported in the absence of malignancy. These fractures may represent completion of an incomplete fracture or stress fracture, but in many cases, no previous symptoms or disease exists. In few, arm pain may herald the presence of a humeral stress fracture. Interestingly, in our practice, we have anecdotally noted a higher incidence of these fractures in left-handed throwers than in right-handed throwers. With nonoperative management and prolonged rehabilitation, some of these

a­ thletes will return to throwing. The potential for refracture is present, however, owing to the tremendous loads imparted to the bone.

Neurovascular Injury Several neurologic and vascular pathologies have been described in throwing athletes.126-129 Among the most serious are vessel aneurysm and thrombosis, as well as distal embolization to the hand. Quadrilateral space syndrome, in which the posterior humeral circumflex artery or axillary nerve becomes entrapped by fibrous bands, has been described in throwing athletes.128,129 Clinicians who treat throwing athletes must have a high index of suspicion for these conditions inasmuch as the complaints and the physical examination findings are typically subtle. Early symptoms include coolness, numbness, or discoloration in the fingertips as a result of digital vessel embolization. Fingertip ulceration may occur owing to recurrent or chronic vessel occlusion. Also, posterior shoulder pain or weakness of the deltoid may indicate nerve or vessel obstruction. Any player complaining of the aforementioned symptoms or any other problem that is not easily explained by common throwing disorders should be evaluated for a vascular or neurologic injury. Besides those diagnoses mentioned earlier, other considerations for these type of symptoms include thoracic outlet syndrome, cervical rib, and peripheral nerve entrapment syndromes. Following a careful history and examination, diagnostic evaluation should include noninvasive modalities such as a duplex ultrasound and pulse volumetric recording to document any flow abnormalities. If any abnormality is documented on these studies, an arteriogram may be obtained. Electromyography with nerve conduction velocities can be obtained in cases of nerve entrapment.126 Conservative management options for players with vascular disorders include cessation of smoking and tobacco use and possibly anticoagulation or thrombolysis depending on symptoms. If these methods fail to alleviate symptoms, surgical resection of the lesion causing the distal symptoms or open thrombectomy may be performed. Excellent results with full return to competitive throwing have been documented in the literature with no permanent sequelae noted.126

SUMMARY The overhead throwing motion comprises several steps that impart extreme forces to various segments of the upper extremity. These forces are capable of causing numerous types of disorders with concomitant symptoms and loss of function and playing time. The thrower’s shoulder is a delicate balance between stability and instability that must be preserved to allow for efficient use of energy in the overhead throwing motion. The clinician must differentiate between laxity and instability because most good throwers possess some element of congenital laxity, especially in the shoulder. The clinician involved in the care of such athletes must make every attempt first to determine accurately the location and nature of the disorder and then to recommend and carry out appropriate treatment. In most cases, conservative management is the first option of ­treatment. With few exceptions,

Elbow and Forearm 1227

surgical management should be considered only after failure of conservative options. As technology and knowledge regarding the diagnosis and treatment of throwing athletes continue to evolve, the ability of clinicians and trainers to care for these athletes will also continue to grow. C

r i t i c a l

P

o i n t s

l The throwing motion is a dynamic activity that requires extremes of glenohumeral motion that place extraordinary stresses on the athlete. l The most important aspects of the evaluation of the ­throwing athlete are the history and physical examination. l The soft tissues and bony architecture around the shoulder are susceptible to injury caused by repetitive overuse or acute trauma. l Tremendous forces are also generated at the elbow during the overhead throwing motion.

S U G G E S T E D

R E A D I N G S

Hill JA: Epidemiologic perspectives on shoulder injuries. Clin Sports Med 2:241246, 1983. Meister K, Batts J, Gilmore M: The posterior impingement sign: Evaluation of internal impingement in the overhand athlete. Presented at AOSSM Annual Meeting, Vancouver, British Columbia, July 12-15, 1998. Morrey BF, An K: Articular and ligamentous contributions to the stability of the elbow joint. Am J Sports Med 11:315-319, 1983. Perry J: Anatomy and biomechanics of the shoulder in throwing, swimming, gymnastics, and tennis. Clin Sports Med 2:247-270, 1983. Werner SL, Fleisig GS, Dillman CJ, et al: Biomechanics of the elbow during baseball pitching. J Orthop Sports Phys Ther 17:274-278, 1993. Wilk KE, Andrews JR, Arrigo CA, et al: The strength characteristics of internal and external rotator muscles in professional baseball pitchers. Am J Sports Med 21:61-66, 1993.

R eferences Please see www.expertconsult.com

S ect i o n

C

Throwing Injuries 2. Elbow Injuries in Children and Adolescents James P. Bradley, Russell S. Petrie, and Samir G. Tejwani*

Elbow injuries in skeletally immature athletes are unique compared with those in adults. The biomechanical and anatomic properties inherent to the epiphyseal plate, musculotendinous units, and articular cartilage, as well as the sport-specific mechanism of injury, determine the site and pathologic response in the skeletally immature elbow. Most pathologic immature elbow conditions can be predicted based on the age and sport of the patient.1 In young athletes, knowledge of these unique injury patterns, combined with early modification of activity and appropriate treatment, can often prevent functional disability, permanent deformity, and the need for surgical intervention.

EPIPHYSEAL DEVELOPMENT Skeletal maturation of the elbow occurs from the primary ossification centers of the humerus, radius, and ulna and six distinct secondary ossification centers. The chronologic appearance and closure of these centers has been well studied and documented.2-5 The typical sequence of ossification is the lateral humeral condyle (capitellum), proximal radius, medial epicondyle, trochlea, olecranon, and lateral epicondyle. Usually, secondary ossification centers appear *No party involved in the writing of this manuscript has a financial interest in any company or product mentioned in this manuscript.

radiographically as single bony foci; however, variations do occur. Variations in size, density, position, or number of secondary ossification centers, compared with the uninvolved extremity, often signal potential injury. A thorough understanding of the normal developmental sequence of primary and secondary ossification centers of the elbow is critical in the evaluation of the young athlete.

Elbow Ossification Distal Humerus, Radius, and Ulna Ossification of the distal humerus extends distally to the condyles by birth, and proceeds at a predictable rate throughout childhood.4 The ossification rate in girls exceeds that in boys in most instances.3 During the first 6 months of life, the distal humeral metaphyseal ossification line is symmetrical, and differentiation of the medial from the lateral side is difficult.2 Beginning late in the first year or early in the second year of life, the ossific nucleus of the lateral condyle (capitellum) appears, and the distal humeral metaphysis becomes asymmetric. Initially, the lateral humeral metaphysis slants laterally and then straightens; it then becomes well-defined and sometimes concave to conform to the ossific nucleus of the lateral condyle.6 The capitellum has the most variable pattern of ossification and time of appearance. Initially, it ossifies as a sphere,

1228 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 19C2-1   A line drawn along the anterior shaft of the distal humerus, the anterior humeral line (AHL), normally intersects the anterior third of the ossific nucleus of the capitellum.

and later flattens with maturation into its final shape. Until about 8 years of age, the posterior portion of the physis is broader than the anterior portion.7 In the absence of supracondylar fracture, on a true lateral radiograph, the anterior humeral line (AHL) normally intersects the anterior third of the ossific nucleus of the capitellum (Fig. 19C2-1). Contralateral views are useful for comparison, particularly when any doubt exists about the diagnosis. Radiographic changes in the elbow remain quiescent until late in the third year of life, when the ossific nucleus of the proximal radius begins to ossify. Elgenmark noted the appearance of the proximal radius in 50% of girls at 3.8 years, whereas it was absent in the same percentage of boys until 4.5 years.3 Commonly, the proximal radial epiphysis begins as a sphere, but often develops one or more flat sclerotic centers. Notches or clefts are sometimes noted in the proximal radial metaphysis, and represent normal variations of maturation.8,9 Ossification of the medial epicondylar ossific nucleus begins between the fifth and sixth years of life, with the semblance of a small concavity on the medial aspect of the humeral metaphyseal ossification border. The medial epicondyle may arise from more than one ossific nucleus and is commonly the last epiphyseal center to fuse with the humeral shaft in the normal child, sometimes as late as age 15 or 16 years.10 During evaluation of acute or chronic elbow injury in the skeletally immature patient, it is important to identify the presence and position of the medial epicondyle in each case.11 Avulsion fractures of the medial epicondyle are commonly displaced into the normal position of the trochlear ossification center. Because the medial epicondylar nucleus appears chronologically before the trochlear center, any radiograph showing the presence of a trochlear center with no visualization of a medial epicondylar center should suggest that, in fact, a fracture or dislocation of the medial epicondylar center is present.10 At birth, the proximal ulnar metaphyseal ossification margin lies halfway between the coronoid process and the

tip of the olecranon. This margin usually progresses to enclose about two thirds to three fourths of the capitellar surface by 6 to 7 years of age.6 The secondary ossification nucleus of the olecranon appears between 7 and 9 years of age. Sometimes two secondary ossification centers are visible, one being articular, the other a traction type.12 The anterior center is typically smaller than the posterior one, and both may occasionally remain visible into late ­adulthood.5,10 The trochlear ossification center usually emerges between 9 and 10 years of age. The appearance of multiple irregular secondary trochlear ossification centers is not uncommon. The trochlear center often has an irregular outline, and this appearance should not be confused with an aberrant process.10 The last secondary center to ossify is that of the lateral epicondyle, which appears initially after 10 years of age. It is often small and rapidly fuses to the lateral condyle. The lateral epicondylar center first appears as a thin sliver of bone, rather than as the typical round or spherical ossific nucleus. Considering the relatively short time between the appearance and fusion of the center, it is sometimes uncertain whether ossification is delayed or fusion to the humerus has already occurred. Before cessation of growth, the capitellum, trochlea, and lateral epicondyle fuse to produce one epiphyseal center, while the metaphyseal bone divides the extra-articular medial epicondyle from the new humeral epiphyseal center. The humeral epiphyseal center then fuses with the distal humeral metaphysis between 14 and 16 years of age.13 At about this time, fusion of the proximal radial and ulnar epiphyseal centers with their appropriate metaphyses takes place. Last to fuse to the humeral metaphysis is the medial epicondylar center, at about 14 years of age in girls and 17 years of age in boys6 (Fig. 19C2-2). An easy mnemonic for remembering the sequence of progression of ossification of the elbow secondary centers of ossification is the word CRITOE. The letter C represents capitellum (1 to 2 years), R represents radial epiphysis (3 to 4 years), I represents inner epicondyle (medial epicondyle, 5 to 6 years), T represents trochlea (9 to 10 years), O represents outer epicondyle (lateral epicondyle, older than 10 years), and E represents common epiphysis (14 to 16 years) (Fig. 19C2-3). Meticulous examination of the injury and of contralateral normal radiographs is helpful when evaluating elbow injuries in the young athlete; each secondary epiphyseal center can be expected to appear as a single bony focus.13 Sometimes two or more foci of ossification are apparent early, only to fuse later into a single bony focus. The ossific nucleus is typically homogeneous; variations in size, density, or position on comparative radiographs are harbingers of abnormal development. Irregular islets of ossification and fragmentation are considered abnormal. These anomalies usually represent alterations of the normal vascular genesis and ossification patterns of the ­ secondary ossific centers. Repetitive throwing in young athletes may account for many of these aberrant ossific patterns.14 Clearly, a thorough knowledge of the normal sequential pattern of appearance of the secondary ossification centers and temporal fusion rates is needed to properly evaluate elbow injuries in young athletes. Failure to appreciate the

Elbow and Forearm 1229

Girls 12 – 14 yr Boys 13 – 16 yr

Girls 14+ yr Boys 17+ yr

­ ndermineralized focus. This appearance should not be misu interpreted as a destructive lesion of the bone.10 Likewise, the thin humeral olecranon fossa occasionally may appear to be totally lucent, representing the so-called perforated olecranon fossa.10 Occasionally, a separate bony ossicle may be found within the perforated olecranon fossa. An anatomic anomaly that appears sporadically on the anterior medial distal humerus is a bony projection called the supracondylar process, appearing in 0.4% to 2.7% of the population.15 It is thought to be an embryologic remnant of little significance, although it can be associated with the ligament of Struthers, a band-like structure attaching it to the medial humeral epicondyle or brachial fascia above the epicondyle. This ligament can cause entrapment of the median nerve and the brachial artery in athletes who perform repetitive vigorous elbow flexion.15-17

PHYSICAL EXAMINATION

Girls 10+ yr Boys 12+ yr

Girls 10+ yr Boys 12+ yr

Figure 19C2-2   The usual ages at which the ossific centers fuse to each other and to the distal humerus in males and females.

process of normal development may lead to incorrect diagnosis and treatment.

Normal Anatomic Bony Variants During elbow ossification, a few normal variations or unusual appearances can occur and should be noted. The developing radial tuberosity, which is the site of insertion of the biceps brachii muscle, may appear as an

When elbow injury is suspected, physical examination begins with inspection of both upper extremities. Loss of motion, muscle atrophy or hypertrophy, bony deformity, and elbow asymmetry are ascertained. There may be some degree of hypertrophy or flexion contracture, or an alteration in the carrying angle of the dominant extremity. At this time, concomitant examination of the neck, wrist, shoulders, and hand is completed. The range of motion, including flexion, extension, pronation, and supination, is performed in comparing both extremities. A complete neurologic and vascular examination of the extremity is performed routinely, with special attention given to the ulnar nerve. In assessing tenderness, the medial and lateral epicondyles, olecranon process, radial head, and collateral ligaments are palpated. Palpation of the ulnar nerve in both flexion and extension is done to asses for ulnar nerve subluxation. Slight flexion of the elbow is needed to examine the olecranon fossa; with gentle pressure, the examiner should be able to differentiate posteromedial from posterolateral pathology. The olecranon is unlocked with slight flexion (15 to 25 degrees), which permits evaluation of the ligamentous stability of the elbow. The lateral ligaments are ideally tested with a varus stress and internal rotation of the arm, and the ulnar collateral ligaments are tested with a valgus stress and external rotation of the arm.18 Often times, laxity testing must be performed with great detail in order to detect subtle differences indicative of elbow ­instability.

RADIOGRAPHIC EXAMINATION Girls 8 – 11 yr Boys 9 – 13 yr Girls 1 m o– 11 mo Boys 1 m o– 26 mo

Girls 5 – 8 yr Boys 7 – 9 yr

Girls 7 – 11 yr Boys 8 – 13 yr

Figure 19C2-3   The average ages of the appearance of the secondary ossification centers of the humerus for males and females.

Routine radiographs are essential in the diagnosis of elbow pathology. Standard anteroposterior, lateral, reverse axial,19 and comparison radiographs are needed. Forty-five degree flexion views can also be obtained and may demonstrate pathology not seen on an extension film. A thorough understanding of skeletal maturation and epiphyseal development is necessary for adequate evaluation of pathology in pediatric elbow injuries. The importance of comparison elbow views cannot be overemphasized in this population. Stress films, when positive, are helpful in evaluating ligamentous compromise, but a negative stress film, even

1230 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� X-ray Plate

X-ray Beam Figure 19C2-4   The gravity stress test of the medial elbow ligamentous complex. The arm is placed in full external rotation, permitting the weight of the forearm to deliver a valgus stress to the elbow.

under anesthesia, does not exclude ligamentous disruption (Fig. 19C2-4). Common medial findings in the immature elbow include enlargement, fragmentation, or breakage of the epicondyle, and occasionally avulsion of the medial epicondyle. Lateral lesions usually involve the subchondral bone, manifest as osteochondrosis or osteochondritis dissecans (OCD) of the capitellum or radial head, and may eventually result in loose bodies and terminally degenerative arthritis. The initial finding is a lucent area in the capitellum best seen on the oblique film. A loose body that resides in the lateral compartment or anteriorly may develop and should be sought. Posterior lesions commonly present with hypertrophy of the ulna that causes chronic impingement of the olecranon tip into the olecranon fossa. Frequent impingement of the olecranon results in osteophytic enlargement with resultant loose bodies in the olecranon fossa. Rarely, stress fractures of the ulna, olecranon apophysitis, or delayed union of the olecranon apophysis occurs. Three-phase bone scans may be helpful in evaluating subtle changes noted in overuse injuries of the elbow. These are usually correlated with tomograms or computed tomography of the area of increased activity. Tomograms are helpful in detailing the articular changes, loose bodies, spur formation, and trabecular changes. Although not commonly employed, ultrasound has been found to be effective in diagnosing medial epicondylar fragmentation and OCD of the capitellum in young baseball players.20 Magnetic resonance imaging (MRI) is a very useful modality in the evaluation of pediatric elbow injuries. MRI potentially has specific advantages in defining nonossified structures, such as developing epiphyses and apophyses, joint capsules, fractures through cartilaginous structures, early avascular changes, and ligaments and soft tissues not identified on plain radiographs. MRI is particularly useful in the diagnosis of osteochondral lesions, epicondylitis, and ulnar collateral ligament (UCL) injuries.21

ANATOMIC CONTRIBUTIONS TO ELBOW STABILITY The elbow articulation is one of the most congruous joints in the body and therefore is one of the most stable. This characteristic is the result of an almost equal contribution

from the soft tissue constraints and the articular surfaces.22 Static stability is provided by the articular surfaces, ligaments, and capsular structures, whereas dynamic stability is provided by the musculotendinous units crossing the elbow.

Osseous Stability Elbow stability can generally be considered 50% a function of the collateral ligaments and anterior capsule and 50% a function of the bony articulations, primarily from the ulnohumeral joint.23 Morrey and An demonstrated that serial excision of the olecranon (25%, 50%, 75%, and 100%) resulted in near linear decreases in elbow stability provided by the ulnohumeral joint in both 0 degrees and 90 degrees of flexion.23 The stabilizing effect of the radial head on the elbow has been examined as well, as a secondary restraint. The radial head provides between 15% and 30% of resistance to valgus stress, depending on the load conformation and orientation of the elbow joint.23 The resistance of the radial head to valgus stress may be greater during throwing, and much clinical study has been devoted to elucidating this complex relationship. The amount of force transmitted across the elbow joint varies with the loading configuration and angular orientation of the joint, and a magnitude of up to nearly 3 times body weight has been estimated.24,25 Activities of daily living necessitate a force of about half of body weight transmitted across the joint, with maximal loads noted at about 90 degrees of flexion.26,27 Halls and Travill noted that 60% of the axial load through the wrist is transmitted across the elbow at the radiohumeral joint and 40% at the ulnohumeral joint.24 These investigations primarily examined the elbow during activities of daily living and isometric lifting, and the results cannot be extrapolated to the tremendous demands imposed on the elbow during throwing. Considering this, it is not surprising that a small deficiency in the elaborate stability-controlling mechanisms of the elbow may have a significant and cumulative effect on elbow function.

Ulnar Collateral Ligament The medial (or ulnar) collateral ligamentous (UCL) ­complex of the elbow is broad and fan shaped and is ­composed of three essential parts: an anterior oblique

Elbow and Forearm 1231

Annular ligament Accessory collateral ligament Anterior oblique ligament

Radial collateral ligament

Posterior Oblique Ligament

Transverse Ligament Figure 19C2-5   The medial (ulnar) collateral ligamentous complex of the elbow.

­ undle, a posterior oblique bundle, and a transverse ligab ment (Fig. 19C2-5). The anterior oblique component of the medial collateral ligament is a thick, substantial structure originating on the medial epicondyle and inserting into the coronoid process. As the elbow flexes, the posterior fibers become tight while the anterior fibers become less tense.28 The anterior oblique component of the UCL is the major ligamentous support of the medial aspect of the elbow. This is especially true during throwing, when tremendous valgus tension is generated along the medial aspect of the elbow.

Radial Collateral Ligamentous Complex The lateral (or radial) collateral ligamentous complex offers varus stability and has been better understood since O’Driscoll and colleagues’ description of posterolateral rotatory instability in 1991.29 The complex is composed of three individual parts: (1) radial collateral ligament, (2) lateral ulnar collateral ligament (LUCL), and (3) accessory lateral collateral ligament. The radial collateral ligament originates from the lateral epicondyle and inserts onto the annular ligament. The LUCL, which originates from the posterior aspect of the lateral epicondyle, traverses the annular ligament and attaches to the ulna at the crista supinatoris.30 The accessory lateral collateral ligament originates from the inferior margin of the annular ligament and inserts onto the tubercle of the supinator (Fig. 19C2-6). The entire radial collateral complex accounts for the stability of the elbow after the radial head has been excised, but it is the LUCL that is primarily responsible for posterolateral stability of the elbow. Although repetitive microtrauma from throwing typically does not disrupt the radial collateral complex, injury to the LUCL acutely from elbow dislocation or iatrogenically during surgery can result in instability.29 The accessory lateral collateral ligament further stabilizes the annular ligament during varus stress. The anconeus muscle and the lateral collateral

Lateral ulnar collateral ligament Figure 19C2-6   The lateral collateral ligamentous complex of the elbow.

l­igaments form a complex that functions as both a static and dynamic stabilizer to the lateral elbow.28

BIOMECHANICS OF THE THROWING ELBOW The Kinetic Chain The throwing motion of the elbow is common to many sports, most notably the tennis serve, javelin throw, and football pass; however, the prototype in terms of abundance of biomechanical information is the baseball pitch. Of paramount importance in understanding the biomechanics of the throwing elbow is to appreciate that the elbow is not an isolated functioning structural entity, but is part of a complex kinetic chain involving core and proximal musculature. Anticipatory proximal muscle activation results in postural adjustments that allow for the body to balance the forces required for throwing.31,32 These muscle activations create force moments, which are a function of the motion and position of adjacent segments.33 These proximal moments occur in the central body segments and are the key to developing significant force at distal joints, such as the elbow. Analogous to significant force generated in the throwing elbow from the lower extremities, pelvis, trunk, and shoulder is the cracking of a whip, in which proximal core activation provides moments which allow efficient distal segment function.34 Although there are few differences between the mechanics of young pitchers and those of adult pitchers,35 young pitchers demonstrate increased trunk and leading hip velocity from cocking to acceleration compared with adult pitchers, probably owing to a decreased capability for lower core force production.36 This alteration in the proximal kinetic chain can manifest as elbow pain in pitchers.34 Similarly, the “dropped elbow” occurs in pitchers when the elbow is positioned below the level of the shoulder in the acceleration phase, owing to lack of elbow elevation and extension before maximal shoulder rotation. This leads to increased tensile loads at the medial elbow ligaments and has been termed the kiss of death for the young pitcher.37

1232 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 19C2-9   Throwing, phase III: late cocking. Figure 19C2-7   Throwing, phase I: wind-up.

To minimize valgus load at the elbow, maximal elbow extension must occur before maximal arm rotation. This long-axis rotation allows coupled shoulder internal rotation and elbow pronation around the long axis of the arm from the glenohumeral joint to the hand and prevents medial tensile injury.38

Pitching Numerous investigations have studied the elaborate pattern and synchrony of bony, ligamentous, and muscular interactions that occur in pitching.39,40 Particular ­interest lies in the regulation of breaking pitches in the young pitcher, given its observed association with the development of elbow pathology. Electromyographic observations have shed new light on the biomechanics and mechanisms of injury sustained by the throwing athlete.41 In pediatric athletes, repetitive valgus and distraction forces are seen

Figure 19C2-8   Throwing, phase II: early cocking.

in the elbow and are thought to be causative in injuries including medial epicondylitis, collateral ligament injuries, and epicondyle avulsions.42 The pitch is divided into five stages: phase 1, the windup or preparation phase, ending when the ball leaves the nonthrowing glove hand; phase 2, early cocking, is a period of shoulder abduction and external rotation that begins as the ball is released from the nondominant hand and ­terminates with contact of the forward foot on the ground; phase 3, the late cocking phase, continues until maximal external rotation at the shoulder is obtained; phase 4, acceleration, is the short propulsive phase that starts with internal ­rotation of the humerus and ends with ball release; and phase 5, the follow-through phase, starts with ball release and ends when all motion is complete (Figs. 19C2-7 to 19C2-11).39,40

ELBOW INJURY: THROWERS In both the young throwing athlete and the mature thrower, four distinct areas are vulnerable to throwing stress: (1) tension overload of the medial elbow restraints,

Figure 19C2-10   Throwing, phase IV: acceleration.

Elbow and Forearm 1233

Figure 19C2-11   Throwing, phase V: follow-through.

(2) ­compression overload on the lateral articular surfaces, (3) posteromedial shear forces on the posterior articular surfaces, and (4) extension overload on the lateral restraints.14,43,44 Pathologic conditions associated with the thrower’s elbow include UCL injuries, ulnar neuritis, valgus extension overload with osteophyte formation and posteromedial impingement, flexor pronator strain, medial epicondyle pathology, OCD of the capitellum, loose body formation, bony spur formation, and capsular ­contracture.45 Specific injury patterns can be discerned during each phase of pitching. During early cocking and especially during late cocking, a significant distraction force is applied to the medial aspect of the elbow.14,43 The resultant force presents as tension on the medial epicondylar attachments, including the flexor muscle origin and the ulnar collateral ligaments. Commonly, with overuse or altered mechanics, the weakest link in the medial complex can be injured. In young athletes, subsequent injury or avulsion of the medial epicondylar ossification center is often encountered (Fig. 19C2-12). The UCLs may become stretched, resulting in traction spurs on the coronoid process. Traction ­ injuries

Figure 19C2-12   Avulsion fracture through the physis of the medial epicondyle with an attached anterior oblique ligament and flexor muscle mass.

to the ulnar nerve and flexor muscle strains may also ensue.23,25,43,45,46 Compression of the lateral articulation, in which the radial head abuts the capitellum, occurs mainly during early and late cocking. Sequelae include growth disturbances, chondral or osteochondral fractures of the capitellum (with resultant loose bodies), and growth disturbances and deformation of the radial head (Fig. 19C2-13).14,43 Medial elbow injury can also be associated with glenohumeral internal rotation deficit (GIRD), which is defined as a side-to-side asymmetry of more than 25 degrees, an absolute value of less than 25 degrees, or a side-to-side loss of total arc of motion of more than 25 degrees.47 GIRD is thought to be due to acquired contracture of the posterior shoulder capsule and results in abnormal scapular mechanics.47,48 In addition to being associated with glenoid labral injury, it is associated with increased posteromedial shear

Ulnar nerve

Figure 19C2-13   Compression of the capitellum against the radial head and tension on the medial collateral ligament, flexor muscles, and ulnar nerve medially usually are the forces that cause injury during early and late cocking when throwing.

1234 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 19C2-14   The sites of injury during the follow-through phase of throwing (coronoid and olecranon spurs).

at the elbow due to abnormal force dissipation in the upper extremity.47,49 Initial treatment consists of early detection on physical examination and aggressive utilization of “sleeper stretches” for the posterior shoulder capsule.50 Posterior articular surface damage develops in two phases of throwing. During late cocking, a posteromedial shear force develops around the olecranon fossa. Throughout follow-through, hyperextension of the elbow is prominent, placing stress on the olecranon and anterior capsule. These stresses commonly produce pathology at three sites: (1) posteromedial spurs, (2) true posterior olecranon spurs (from triceps strain), and (3) traction spurs of the coronoid process (Fig. 19C2-14).14,43 Lateral extension overload occurs during acceleration when extreme pronation of the forearm results in a tension force applied to the lateral ligaments and lateral epicondyle. Consequently, lateral epicondylitis may develop (Fig. 19C2-15).14,43

LITTLE LEAGUER’S ELBOW The term little leaguer’s elbow is used to depict a group of pathologic entities in and around the elbow joint in young developing throwers. It is important to realize that the throwing motion is common to multiple sports, including the tennis serve, javelin throw, and football pass. Many authors have emphasized that these abnormalities are secondary to the biomechanical throwing stresses placed on the young developing elbow.44,51-66 The physical stresses associated with throwing produce exceptional forces in and around the elbow.14 These include traction, compression, and shear, which are localized to the medial, lateral, and posterior aspects of the elbow.14,43 Any or all of these

forces may contribute to the alteration of normal osteochondral development of the elbow.14 The constellation of injuries include (1) medial epicondylar fragmentation and avulsion, (2) delayed or accelerated apophyseal growth of the medial epicondyle, (3) delayed closure of the medial epicondylar growth plate, (4) osteochondrosis and OCD of the capitellum, (5) deformation and OCD of the radial head, (6) hypertrophy of the ulna, and (7) olecranon apophysitis with or without delayed closure of the olecranon ­apophysis.52,58,67-70

History Age, position, handedness, activity level, location of pain, duration of pain, radiation, trauma, mechanism of injury, nature of onset, and past medical history are all salient factors in the history. A plethora of symptom complexes, some with subtle variations, may be evident on evaluation of young athletes with elbow pain, and only with a careful history can the orthopaedist begin to localize the underlying pathology. The age of young throwers can be divided into three groups: (1) childhood, which terminates with the appearance of all secondary centers of ossification, (2) adolescence, which terminates with the fusion of all secondary centers of ossification to their respective long bones, and (3) young adulthood, which terminates with completion of all bone growth and the achievement of final muscular form.14 During childhood, the most frequent complaints are pain around the medial epicondyle, which is usually secondary to repetitive microinjury at the apophysis and ossification center. Throwing stresses impede the normal chondro-osseous transformation and result in an

Figure 19C2-15   The sites of injury during the acceleration phase of throwing.

Elbow and Forearm 1235

i­rregular ossification pattern of the secondary ossification center.14 When the athlete enters adolescence, muscle strength, muscle mass, and throwing force increase. The athlete increases the valgus stresses on the elbow, and the result can be an avulsion fracture of the entire medial epicondyle. Partial avulsion of the medial epicondyle becomes apparent as the thrower approaches the end of adolescence because the medial epicondyle begins to fuse at this time. Some adolescents develop enough chronic stresses to cause delayed union or possibly nonunion of the medial epicondyle.14 By young adulthood, the medial epicondyle is fused, and injuries of the muscular attachments and ligaments of the epicondyle become more prevalent. During this time the flexor muscles and ulnar collateral ligaments are at increased risk of injury.14 Therefore, the age of the thrower can provide the examiner with useful information regarding the possible etiology and location of the elbow pathology. The position played by the thrower provides insight into the magnitude of stresses placed on the elbow and the relative incidence of elbow complaints. Pitching inherently places more stress on the elbow during play, and pitchers most commonly complain of elbow injuries.14,43 The usual order of prevalence of elbow complaints among players is pitchers, infielders, catchers, and outfielders.14 Although the magnitude of injury has not been proved to depend on position, intuitively it seems that pitchers have the greatest risk for elbow injury. Handedness is very germane in the initial history. Most throwers present with elbow problems in the dominant extremity unless direct trauma is the cause of the problem. Pain is the most common complaint. Localization, duration, character, temporal sequence (night, day, during or after activity), activity level, and nature of onset are all clues to the underlying pathology. Although pain is the most frequent complaint, related but less frequent problems include decreased elbow motion, mild flexion contracture, swelling, decreased performance, and local sensitivity of the elbow.14 The pain is most often localized to the medial epicondyle; however, lateral and posterior pain may accompany or be the presenting complaint. The duration of pain is usually an indirect measure of the severity of the problem. Pain before, during, and after throwing is usually an ominous finding. The relationship of the pain to the specific activity must be delineated. Medial pain in a young adult that occurs with a specific phase of throwing (such as late cocking or acceleration) may be the sentinel sign of early instability, although medial pain in the same phases of throwing in childhood most commonly represents medial epicondylar injury. Nocturnal pain is very uncommon, and the possible presence of a neoplastic process must be addressed. Burning or a vague ache centered around the medial aspect of the elbow and associated with dysesthesias or paresthesias of the small and ring fingers signifies ulnar nerve involvement and is a significant finding. The duration of pain (acute versus chronic) is a very helpful sign. If a child presents with a single episode of injury, an acute traumatic condition such as avulsion of the medial epicondyle must be considered. If a young adult complains of similar symptoms, UCL injury is possible. In

other instances, the history may reveal an insidious onset of chronic pain connoting a form of overuse syndrome or possibly an osteochondrosis type of injury. Ancillary information such as activities that aggravate or relieve the pain, types of pitches thrown, innings pitched, typical pitching rotation, and changes in the training schedule should be elucidated. Attention to detail is paramount because a neglected sprain or strain or the slightest change in the training schedule can sometimes lead to the correct diagnosis and treatment. It is extremely helpful to remember that the elbow may be the site of referred pain, although this is uncommon in young throwers. Therefore, associated neck, shoulder, and wrist pain or restricted motion must be appraised. Knowledge of prior surgical intervention, even remote, is essential when sorting through possible causes of the present pain. For example, previous fixation of a displaced supracondylar fracture resulting in subtle malunion can alter the intricate biomechanical relationships necessary for effective throwing. Likewise, prior shoulder surgery may place the elbow at risk for overuse syndromes secondary to altered shoulder-girdle mechanics. The past medical history and any recent medical work-ups should be evaluated. Inquiries should be made about a family history of osteochondrosis, including OCD, Kohler’s disease, Legg-Calvé-Perthes disease, and Osgood-Schlatter’s disease. When such patients participate in activity that involves a high articular demand around the elbow, the likelihood of variations in epiphyseal osteochondral development is increased.14 A history of delayed skeletal maturation, combined with participation on an age-determined team, commonly requires the child to throw beyond his physiologic tolerance and often leads to elbow ­problems.14

Diagnosis A timely and accurate diagnosis is the keystone to successful treatment of the many conditions associated with little leaguer’s elbow. A meticulous history and physical examination are the primary tools in the orthopaedist’s arsenal in achieving this goal. Special tests such as arthrography, computed tomography, MRI, magnetic resonance arthrography (MRA), and bone scans are often necessary, but play a confirmatory role rather than a diagnostic one. An understanding of the normal geometry of the throwing elbow is needed to evaluate elbow injuries. At 90 degrees of flexion, the medial and lateral epicondyles form an equilateral triangle with the tip of the olecranon. As the elbow is extended, these landmarks fall into a straight line. By understanding the relationship of these landmarks, the examiner can appraise the elbow for alignment and rotation, especially when evaluating the young athlete. Young throwers often have unilateral hypertrophy of the muscles and bone of the dominant extremity.67 Twelve percent of male little league pitchers in the Houston study (595 pitchers) had a flexion contracture of the elbow, and 37% of these young pitchers had a valgus deformity of the elbow.67 Thus, the presence of hypertrophy, valgus deformity, and flexion contracture should not be considered uncommon in young throwers and should be sought on physical ­examination.

1236 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

ELBOW INJURY Tennis Players The most common anatomic locations for injury in the pediatric tennis player are the shoulder and elbow.71 Survey data from the United States Tennis Association (USTA) National Championships for 16- to 18-year old boys and girls revealed that about 25% reported previous or current elbow pain.72 Lateral epicondylitis (tennis elbow), medial epicondylitis, and injury to the medial epicondylar growth plate can all be seen in skeletally immature tennis players.42 Lateral epicondylitis has been associated with preventable equipment issues, including incorrect grip size; metal racquets; heavier, stiffer, more tightly strung racquets; and racquets with increased vibration.73 In addition, poor backhand mechanics have been found to correlate specifically with the development of lateral epicondylitis.74 Therefore, ensuring proper equipment selection and the development of correct biomechanics plays an important prophylactic role in elbow injury in the young tennis player. As in baseball, GIRD is a common maladaptation in pediatric tennis players that appears at an early age and progresses with age and years of play.75,76 It has been associated with an increased incidence of elbow injury and should be screened for and treated as in baseball pitchers, with posterior capsular stretching. Proximal kinetic chain mechanics are also important in the young tennis player. In a study of two groups of Olympic tennis players who generated the same ball speed, Elliot and associates found that the group that exhibited knee flexion of less than 10 degrees in the cocking phase of the tennis serve increased the normalized valgus load at the elbow by 21%, resulting in an overall value above the “safe” level of repetitive load.77,78 Thus, lack of proximal muscle activation can increase the distal loads for the same force output, thereby placing the upper extremity at risk for sustaining an overload injury.

Gymnasts The frequency of participation of young athletes in gymnastics during the past decade has increased significantly. Nearly one third of injuries reported in gymnastics

Figure 19C2-16   Medial tension injury with widening and fragmentation of the medial epicondylar ossification center.

occur in the upper extremity, with at least 7% involving the elbow.79 In essence, the elbow, which normally is a non–weight-bearing joint, becomes a weight-bearing joint during routines such as one-arm balancing, handstands, tumbling, and trunk stabilization on the bars.80 Goldberg noted the presence of elbow compression and traction injuries similar to those seen in little leaguer’s elbow, the most common of which presented as traction injuries of the medial aspect of the elbow.80 These injuries included partial tears of the flexor muscle mass, collateral ligament strains, and medial epicondylar traction injuries. The most prevalent problem afflicting the elbow in gymnasts is a posterior elbow injury.81 Biomechanically, to support the body weight, the gymnast must repetitively “lock out” the elbow in full extension, thereby forcing the olecranon into the olecranon fossa, which results in posterior fossa impaction and inflammation.81 Multiple studies have also reported subluxation-dislocation of the elbow, often associated with an avulsion fracture of the medial epicondyle.80,82 Although infrequent, chondral or osteochondral fractures of the capitellum may occur in gymnasts as well.11 Additionally, OCD lesions of the capitellum can develop and carry a guarded prognosis for return to play, given the high demands and forces placed on the elbow during gymnastics.83

MEDIAL ELBOW PATHOLOGY Medial Tension Injuries Most cases of little leaguer’s elbow present with medial elbow complaints. The triad of symptoms includes progressive medial pain, diminished throwing effectiveness, and decreased throwing distance. Repetitive valgus stresses and flexor forearm pull usually produce a subtle apophysitis or stress fracture through the medial epicondylar epiphyses. Physical manifestations include point tenderness, swelling over the medial epicondyle, and an elbow flexion contracture that is often greater than 15 degrees.28,43,84 ­Radiographs show fragmentation and widening of the epiphyseal lines compared with the contralateral elbow (Fig. 19C2-16).28,52,67,70

Elbow and Forearm 1237

Authors’ Preferred Method of Treatment: Medial ­E p i c o n d y l e A p o p h y s i t i s In most cases, a 4- to 6-week course of abstinence from throwing results in cessation of symptoms. Initially, ice and nonsteroidal anti-inflammatory medications help to alleviate the symptoms.28,43,84 After the symptoms resolve, a gradual return to throwing is advisable. The use of steroid injections is neither necessary nor generally beneficial. Rarely, when the symptoms are advanced, the use of a ­removable posterior splint may be helpful. After 6 weeks, when the patient has no symptoms and a pain-free range of motion, strengthening exercises are begun. A progressive throwing program is initiated at about 8 weeks. Occasionally, disability may continue for an extended period, and elbow pain may continue when throwing is resumed.85 In these cases, throwing should be disallowed until the next baseball season. Medial apophysitis is usually a benign entity, and a good outcome is expected. Medial tension injuries in the adolescent pitcher result, in part, to overuse. It has become clear that a key to the treatment of little leaguer’s elbow is prevention, by avoiding year-round participation in one sport. This responsibility lies not only with the evaluating or team physician but also with the coach, trainer, parents, and officials.86,87 Additionally, proper mechanics must be emphasized at an early age. Lyman and associates have reported on the biomechanics of the pediatric baseball pitcher, aged 9 to 14 years old.88,89 Although the authors found no correlation between the number of pitches thrown in a game and elbow pain, throwing more than 600 pitches per season was associated with increased elbow pain, suggesting a cumulative rather than acute effect. Further, younger pitchers were found to be 80% more likely to experience elbow pain when throwing a slider, which correlates with findings by Sciascia and Kibler that throwing a curveball requires an increased amount of force and torque at the elbow and shoulder.34 Thus, delaying the use of breaking pitches in the young pitcher’s arsenal will likely have a protective ­effect on the medial elbow structures.

Medial Ligament Ruptures Injuries to the UCL, although generally uncommon in adolescent throwers, are becoming increasing prevalent in young adults as muscle mass and force during throwing increase.1 UCL injuries occur mostly in adults and occasionally in young adults. Most patients have an insidious progression of discomfort and tenderness around the medial aspect of the elbow for months to years before the ligament is qualitatively injured. Commonly, a rupture occurs as a sudden catastrophic event, after which the elbow is so painful that further throwing is not possible.90 Clinically, with these injuries, subtle findings of medial elbow instability are present, demonstrated by flexing the elbow to 25 degrees to unlock the olecranon from its fossa and gently stressing the medial side of the elbow.90 Woods and Tullos noted that a radiographic gravity medial stress test of the elbow is sometimes useful in diagnosing UCL injuries.91

Treatment of complete tears of the UCL with resultant instability in young throwers who wish to return to repetitive throwing should consist of surgical intervention.1,28,90 Irelano and Andrews have recommended direct surgical repair of the UCL in select cases.1 When bony avulsion of the ulnar sublime tubercle occurs in the overhead throwing athlete, resulting in insufficiency of the UCL, Salvo and coworkers have reported successful treatment with either repair using bone anchors or UCL reconstruction.92 Likewise, Bennett and Tullos stated that UCL injuries associated with elbow instability may necessitate surgical reattachment.28 Jobe supports direct repair if elbow stability can be reconstituted, but if a tenuous repair is imminent, reconstruction of the UCL using a palmaris longus tendon graft should be performed in association with an anterior submuscular transposition of the ulnar nerve (Fig. 19C2-17).93 Multiple alternative surgical techniques have been proposed for UCL reconstruction, with generally good results reported.94-96 Rarely, patients give a history of multiple episodes of pain over the medial elbow during throwing that initially restricts the throwing motion. However, after a few days to weeks of self-instituted conservative care, the pain resolves, only to become symptomatic with resumption of throwing. Valgus stress elicits pain and snapping at the medial elbow. Instability testing and stress radiographs are negative. If these patients do not respond to 6 months of supervised conservative treatment, MRI or MRA and surgical intervention are generally warranted.

Authors’ Preferred Method of Treatment: Ulnar ­C o l l a t e r a l L i g am e n t I n j u r y If UCL injury is detected in the early stages, conservative treatment is appropriate. Rest, for a longer time than may ordinarily be recommended, is crucial. Injections of steroid in the region of the ligament are not currently used in our practice. If significant scarring and calcification are present and are accompanied by pain that does not respond to rest, the calcifications should be removed and the ligament repaired primarily, if possible. If stability of the joint is all that is required, surgery is not necessary; however, if the patient desires to continue participation in a throwing sport, reconstruction is usually required. Primary direct ligamentous repair is indicated when it is believed that a stable repair is attainable. Bony avulsion of the ulnar sublime tubercle can be treated with fixation of the fragment or UCL reconstruction.92 Surgical reconstruction of the UCL with a tendon graft is also indicated in the following circumstances: (1) acute rupture in throwers who lack enough remaining ligament for a primary ­repair, (2) reestablishment of valgus stability in the presence of symptomatic chronic laxity, (3) following débridement for calcific tendinitis in athletes if there is not sufficient viable tissue left to effect a primary repair, and (4) multiple episodes of recurring pain with throwing after periods of conservative care. When surgical reconstruction of the medial collateral ligament is necessary, the technique described by Jobe and colleagues is preferred.90,93

1238 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Ulnar nerve

Figure 19C2-17   Medial collateral ligament reconstruction using a tendon graft with an anterior submuscular transposition of the ulnar nerve.

Medial Epicondylar Fractures

Osteochondritis Dissecans

When more substantial acute valgus stress is applied through forceful muscle contraction during throwing, an avulsion fracture of the medial epicondyle may ensue. The consequence is a painful elbow with point tenderness over the medial epicondyle and an elbow flexion contracture that may exceed 15 degrees. Radiographs show variable size and displacement of the epicondylar fragment.28

OCD is viewed as a singular entity within the group of pathologic conditions termed little leaguer’s elbow.63,98-100 Tullos and King analyzed the throwing motion and concluded that OCD of the capitellum was secondary to compressive forces occurring between the radial head and the capitellum from valgus overload.66 Many other authors have noted the relationship between throwing and OCD.51-54 Additionally, Schenck and coworkers demonstrated a biomechanical mismatch between the radial head and the capitellum that may contribute to the genesis of OCD.101 Osteochondritis is a focal lesion of the capitellum occurring in the 13- to 16-year-old age group, usually characterized by elbow pain and a flexion contracture of 15 degrees or more.28,97 The onset is insidious, with a focal island of subchondral bone demarcated by a rarefied zone on radiographs. Infrequently, the radial head appears larger than that on the uninvolved side.97 Sequelae include loose bodies, residual deformity of the capitellum, and often residual elbow disability.28,100 OCD must be differentiated from Panner’s disease, which is a distinct entity.28,97,100,102 Older age, onset, loose body formation, radiographic findings, and deformity of the capitellum all aid in the differentiation. Panner’s disease usually affects a younger population (<10 years), and the onset is acute with fragmentation of the entire capitellar ossific nucleus. The absence of loose bodies, minimal residual deformity of the capitellum, and no late sequelae are also unlike the findings in OCD. The etiology of OCD of the elbow has not yet been determined. Three popular theories include ischemia, trauma, and genetic factors.103-105 One or all three may play a role in the development of OCD of the capitellum. Recent histopathologic assessment of the articular cartilage and bone from specimens retrieved from 25 young athletes undergoing osteochondral autograft surgery for OCD of the capitellum has been performed.106 Findings suggest that the primary pathologic changes in OCD are due to damage of articular cartilage induced by repetitive stress following a degenerative and reparative process of articular and subchondral fracturing. Separation of the fragment subsequently occurs beginning on the cartilage surface and progresses to the subchondral bone in advanced stages.106

LATERAL ELBOW PATHOLOGY Panner’s Disease (Osteochondrosis) Panner’s disease is a disorder of growth of the ossification centers in children that begins as degeneration or necrosis of the capitellum and is followed by regeneration and recalcification.97 The child, typically aged 7 to 10 years, presents with dull, aching elbow pain that is aggravated by activity, especially throwing a ball. The elbow is usually swollen, and inability to fully extend is common. An important distinction must be made between osteochondrosis and OCD. The difference is the age of the patient and degree of involvement of the capitellar secondary ossification center.97 In the child, the most common cause of chronic lateral elbow pain is Panner’s disease, but in the adolescent (aged 13 to 16 years), the most common cause of recurrent lateral elbow pain and limited motion is OCD of the capitellum.

Authors’ Preferred Method of Treatment: Panner’s Disease Initial treatment involves rest and avoidance of throwing. Sometimes pain and tenderness necessitate the application of a posterior splint until the acute symptoms resolve. Radiographic follow-up is necessary to ensure that an adequate healing response is present. Conservative treatment in most cases is satisfactory. Late deformity and collapse of the articular surface of the capitellum are uncommon.

Elbow and Forearm 1239

Authors’ Preferred Method of Treatment: Osteochondritis Dissecans Stable lesions identified early appear to have the best prognosis with conservative management. Indications for surgery include persistent or worsening symptoms despite prolonged conservative care, loose bodies, or evidence of instability. When possible, open reduction and internal fixation of large, intact osteochondral fragments of the capitellum are performed with a 3.0-mm cannulated minifragment screw. Screw removal is routinely performed 3 months after surgery. Whether to excise and débride, fix, or remove the unstable fragment and replace with autograft or allograft is a controversial topic.83,107 In the case of small loose bodies generated from an OCD lesion, arthroscopy removal is indicated. In these cases, microfracture of the capitellum can be performed at the site of residual defect.

POSTERIOR ELBOW PATHOLOGY Posterior injuries are uncommon in young throwers; however, as the thrower matures, the incidence of posterior problems increases. Posterior injuries can be divided by age: (1) osteochondrosis of the olecranon in childhood, (2) avulsion fragments and lack of apophyseal fusion in adolescents, and (3) partial avulsion of the olecranon and osteophyte formation at the tip and medial border of the olecranon in young adults. Injuries involving the posterior elbow are uncommon in young throwers. Pappas reported only 18 instances in 111 elbows that involved the posterior joint structures, 11 of which were young adults.14 The causes of the various injuries are associated with the final stages of followthrough. During this throwing phase, extension overload at the olecranon tip and valgus stress results in abutment of the medial aspect of the olecranon process against the olecranon fossa.108 Injury during childhood usually centers on

the secondary ossification center of the olecranon, which produces irregular patterns of ossification and secondary pain due to stress.14 During adolescence, the injury pattern progresses to avulsion fragments or lack of fusion between the ossified secondary center and the olecranon, causing pain due to either loose bodies or nonunion of the olecranon.14,109 Young adults may present with two additional problems: partial avulsion of the olecranon and secondary osteophyte formation at the tip and medial border of the olecranon, which limits elbow extension (Fig. 19C2-18).14,18,108 Posterior olecranon resection can be performed to remove the secondary osteophytes; however, care must be taken to remove bone judiciously because studies have implied that there is an increase in UCL strain with valgus load with increasing resection of bone.110,111 Treatment is tailored to the individual according to the age of the patient and the specific condition affecting the posterior elbow. Children with osteochondrosis usually respond to rest from pitching along with gentle range of motion exercises and a flexibility and strengthening program.14,18,108 In young adults, partial avulsion of the olecranon requires surgical reattachment.14 Adolescents with persistent pain associated with loose body formation or lack of fusion may require surgical removal of the loose body and possibly a bone graft to induce union to the underlying physis.14 Persistent olecranon physes have been successfully treated with open stabilization and internal fixation with autogenous iliac crest bone grafting in adolescent baseball players.112 Similarly, nonunion of olecranon stress fractures have been observed in adolescent baseball pitchers and treated successfully with open reduction and internal fixation using a 7.0-mm cancellous screw and washer with or without 18-gauge tension banding.113 The indications for bone grafting are controversial, and rest and immobilization should precede any attempt at grafting and fixation. Yocum,18 Andrews,108 and Pappas14 all recommend removal of symptomatic loose bodies by arthrotomy or arthroscopy. Posterior and posteromedial osteophytes, as well as loose bodies, should be removed in symptomatic patients, preferably through minimally traumatic arthroscopic approaches.1,14,18,108 Figure 19C2-18   Typical location of posterior and posteromedial osteophytes of the olecranon associated with extension overload.

1240 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method of Treatment: Posterior Elbow Pathology Osteochondrosis of the olecranon in childhood is best treated by rest from pitching and gentle range of motion exercises. When the acute symptoms abate, flexibility and strengthening programs are initiated. When normal strength and range of motion return, throwing is allowed. Olecranon avulsion fragments in adolescence may become chronically bothersome and usually require arthroscopic removal. Lack of olecranon apophyseal fusion is usually responsive to conservative measures, including rest and sometimes immobilization. In recalcitrant cases, the delayed union or nonunion is treated with internal fixation using a partially threaded cancellous screw, varying in diameter from 4.5 to 7.0 mm, depending on the proximal ulna diameter. The repair is routinely augmented with cancellous bone graft from the ipsilateral proximal tibia. The graft is obtained through a 1-cm incision at Gerdy’s tubercle using an osteochondral autograft transfer system (OATS) harvester (Arthrex, Naples, Fla). After the 1-cm cortico-cancellous plug is harvested, the cancellous bone is removed for use in the olecranon. The proximal tibial metaphyseal defect is then filled with demineralized bone matrix putty (DBX, Arthrex, Naples, Fla), and the cortical cap is replaced to its original position to close the defect. Partial avulsion of the olecranon in young adults requires open surgical reattachment of the olecranon and triceps. Symptomatic posterior and medial olecranon ­osteophytes should be removed to decrease pain and improve range of motion. Arthroscopic removal of osteophytes using a posterior and posterolateral portal is preferred, with the aid of an arthroscopic bur.

force at the distal segments; any slight imbalance proximal to the elbow thus alters the method of force transmission and dissipation distally during throwing, leading to decreased performance and elbow pathology. l Pathologic conditions associated with the thrower’s elbow include UCL injuries, ulnar neuritis, valgus extension overload with osteophyte formation and posteromedial impingement, flexor-pronator strain, medial epicondyle pathology, OCD of the capitellum, loose body formation, bony spur formation, and capsular contracture. l The constellation of injuries falling under “little ­leaguer’s elbow” include medial epicondylar fragmentation and avulsion, delayed or accelerated apophyseal growth of the medial epicondyle, delayed closure of the medial epicondylar growth plate, osteochondrosis and OCD of the capitellum, deformation and OCD of the radial head, hypertrophy of the ulna, and olecranon apophysitis with or without delayed closure of the olecranon apophysis. l Prevention is key to the development of elbow pathology in adolescent pitchers; it is the responsibility of the evaluating or team physician, coach, trainer, parents, and officials to limit the cumulative effect of participation in sport and thus decrease the risk for overuse injury. l Panner’s disease (osteochondrosis) is a distinct entity from OCD and can be differentiated from it by younger age at presentation, acute onset, fragmentation of the entire capitellar ossific nucleus, absence of loose bodies, minimal residual deformity, and minimal late sequelae. l Posterior elbow injuries can be divided by age: osteochondrosis of the olecranon in childhood, avulsion fragments and lack of apophyseal fusion in adolescents, and partial avulsion of the olecranon and osteophyte formation at the tip and medial border of the olecranon in young adults.

S U G G E S T E D C l A

r i t i c a l

P

o i n t s

mnemonic to remember the sequence of progression of ossification of the elbow secondary centers of ossification is the word CRITOE (capitellum, radial head, inner epicondyle, trochlea, outer epicondyle, common ­epiphysis). l A thorough knowledge of the normal sequential pattern of appearance of the secondary ossification centers and temporal fusion rates is needed to properly evaluate elbow injuries in young athletes; contralateral comparison radiographs are invaluable. l Elbow stability can generally be considered 50% a function of the medial and lateral collateral ligaments and anterior capsule and 50% a function of the bony articulations, primarily from the ulnohumeral joint. l The anterior oblique component of the UCL is the major ligamentous support of the medial aspect of the elbow and is subject to significant load during valgus force from throwing; complete rupture can recur, necessitating surgical reconstruction for return to pitching. l In the kinetic chain, proximal moments occur in the central body segments and are the key to developing adequate

R E A D I N G S

An KN, Morrey BF: Biomechanics of the elbow. In Morrey BF (ed): The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985. Andrews JR: Bony injuries about the elbow in the young throwing athlete. Instr Course Lect 34:323-331, 1985. Hoffmann AD: Radiography of the pediatric elbow. In Morrey BF (ed): The Elbow and Its Disorders. Philadelphia, WB Saunders, 1985. Hunter SC: Little League elbow. In Zarins B, Andrews JR, Carson WG (eds): Injuries to the Throwing Arm. Philadelphia, WB Saunders, 1985. Hutchinson MR, Ireland ML: Overuse and throwing injuries in the skeletally immature athlete. In Ferlic DC (ed): Instructional Course Lectures. Rosemont, Ill American Academy of Orthopaedic Surgeons, 2003, pp 25-36. Jobe FW, Stark H, Lombardo SL: Reconstruction of the ulnar collateral ligament in athletes. J Bone Joint Surg Am 68:1158-1163, 1986. Lyman S, Fleisig GS, Andrews JR, et al: Effect of pitch type, pitch count, and pitching mechanics on risk of elbow and shoulder pain in youth baseball pitchers. Am J Sports Med 30:463-468, 2002. Pappas AM: Elbow problems associated with baseball during childhood and adolescence. Clin Orthop 164:30-41, 1982. Yadao MA, Field LD, Savoie FH 3rd: Osteochondritis dissecans of the elbow. Instr Course Lect 53:599-606, 2004. Yocum LA: The diagnosis and non-operative treatment of elbow problems in the athlete. Clin Sports Med 8:439-451, 1989.

R eferences Please see www.expertconsult.com

Elbow and Forearm 1241

S ect i o n

D

Osteochondritis Dissecans of the Elbow Felix H. Savoie and Larry D. Field

Osteochondritis dissecans (OCD) is defined as an inflammatory condition of bone and cartilage resulting in localized necrosis and fragmentation of bone and cartilage.1 However, since the original description by Konig (1889) in the knee, no inflammatory cause has been elucidated.2 In the elbow, the most common affected area is the capitellum, although it has been reported in the olecranon and the trochlea. It has also been confused with Panner’s disease, or osteochondrosis, a disease of the ossification center of the capitellum.1-5

RELEVANT ANATOMY AND BIOMECHANICS Classification The initial attempt at classification was a radiologic one by Minami and colleagues, based on lateral radiographs. Grade I lesions demonstrated a shadow in the middle of the capitellum.6 Grade II lesions had a clear zone between the lesion and the adjacent subchondral bone. Grade III lesions were those with one or more loose bodies. Bradley and Petrie subdivided this classification to include magnetic resonance imaging (MRI) as a part of the classification.5 Baumgarten and associates classified the lesions according to the appearance on arthroscopy based on Ferkel’s classification of similar lesions of the talus.7 However, none of these schemes have consistently been able to predict the course of the disease process and thus are of limited value.

Evaluation Clinical Presentation and History OCD is primarily a disorder of the young athlete. The usual age of presentation is 12 to 14 years of age. Males are affected more often than females, but there is a high prevalence in young female gymnasts. The dominant arm is most often involved, with bilateral involvement in 5% to 20% of patients.8 There is usually a history of overuse, most commonly throwing or overhead sports.9 Early on, the symptoms may be obscure, with pain location difficult to determine. It most often begins with some mild aching after activity. Most patients will try self-medicating with anti-inflammatory medicine and ice, which may provide some temporary relief. Symptoms often worsen slowly and are associated with a gradual loss of the ability to fully extend the elbow.10-14

On presentation the usual complaints are pain that increases with activity, loss of motion, and swelling on the lateral side of the elbow. Additional complaints of popping, clicking, or sudden “giving way” may also be present.15-17

Physical Examination and Testing The classic physical findings include loss of terminal extension, swelling along the posterolateral joint line, and inflammation of the normal posterolateral plica.9 Valgus extension overload testing produces pain over the lateral aspect of the elbow and results in a measurable increase in the loss of terminal extension. This is one of the key differentiating factors in the physical examination of OCD. Most throwing or hyperextension overuse injuries, when tested in valgus, have the primary component of pain along the medial aspect of the elbow: the medial apophysis, the medial ulnar collateral ligament, or the flexorpronator muscle. These may coexist with OCD, but in these patients, this stress, especially with the moving valgus overload test, results in pain more on the lateral than medial side.

Radiology and Magnetic Resonance Imaging The initial testing involves standard posteroanterior and lateral radiographs. These usually show the classic findings of radiolucency in the central aspect of the capitellum. There may be a small area of increased opacity in the center of the radiolucency.10 In the later stages, loose bodies may be present (Fig. 19D-1). Additional testing may be warranted early in the course of treatment. Computed tomography, computed tomography arthrograms, and ultrasound have all been used in the evaluation of these lesions. MRI has become the standard modality for evaluation of these lesions. The key points to evaluate—extent of the lesion, integrity of the cartilage cap, and presence of loose bodies—are best assessed by MRI or magnetic resonance arthrography.18,19 Early lesions show change on T1-weighted images but no change on T2-weighted images. The cartilage covering is intact, indi­ cating an improved prognosis. Advanced lesions show changes on both T1 and T2 images and may demonstrate a loose in situ bone fragment (Fig. 19D-2). Assessment of the integrity of the cartilage cap remains paramount. As the condition advances, the cartilage cap is violated, and loose bodies may result. However, the simple presence of loose bodies does not definitively define a rupture of the cartilage, so MRI assessment remains necessary. In the most

1242 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 19D-1   Posteroanterior (A) and lateral (B) views of the elbow delineate the area of lucency in the mid-capitellum indicative of early osteochondritis dissecans.

advanced cases, three-dimensional imaging may be helpful. The lesion that advances to the lateral edge or “shoulder” of the capitellum is a much more severe injury that requires more extensive reconstructive surgery. Three-dimensional imaging can help to define the loss of this critical lateral cortex.20,21

Figure 19D-2   Magnetic resonance imaging may indicate either an intact or disrupted cartilage cap as well as the presence of loose bodies within the defect.

TREATMENT OPTIONS Nonoperative Treatment options for OCD remain controversial. Options vary from a total cessation of any irritating activities to immediate surgery.9 The current classification systems do not help in predicting the course of treatment or the prognosis. The critical determinant in our opinion is the presence or absence of an intact cartilage cap. Patients with intact articular cartilage should be managed nonoperatively. Patients with disruption of the articular cartilage cap may still be managed nonoperatively but with less chance for a full recovery. The hallmark of nonoperative treatment has been rest and cessation of aggravating activities. This restriction is continued until symptoms resolve (usually 6 to 12 weeks) and radiographs or MRI demonstrate complete healing of the lesion (6 to 12 months). This lengthy recovery may be too long for most patients and families to tolerate, resulting in a too early return to sports and recurrence and worsening of the problem (Fig. 19D-3).1,5,6,7,9-17 One alternative method of management that has been used in our clinic has been the protection of these lesions with an offloading double-hinged elbow brace. The straight hinges correct the normal valgus tilt of the elbow and offload the lateral side, protecting the capitellum from injury. Initially, the brace is set at the limits of pain-free range of motion. As the inflammation in the plica decreases and pain-free motion increases, the brace is loosened to allow full range of motion. Sports and normal activities are allowed with the brace in place as long as symptoms do not occur during these activities. This form of treatment allows the patient to continue normal activities without aggravating the lesion in the capitellum. In most cases, the patient is able to return to normal activities with the brace in place within 2 weeks of initiating treatment. The time of the normal healing process remains unchanged, and close monitoring with monthly radiographs and quarterly MRI testing is necessary to follow the healing process if this treatment method is chosen.

Elbow and Forearm 1243

A

B

Figure 19D-3   A late-stage osteochondritis dissecans lesion as seen on radiographs.

Operative Progression of the disease process, loose bodies, or disruption of the cartilage cap that continues despite rest, bracing, and cessation of activity may be considered indications for surgery. Arthroscopic evaluation of the lesion, along with removal of the loose fragments and débridement of the base, is currently the mainstay of operative management. Although much of the older literature focused on open management, most current studies delineate the efficacy of arthroscopic treatment.1,9 The procedure begins with a diagnostic arthroscopy of the anterior compartment of the elbow. In these cases, it is most useful to begin with a proximal anteromedial portal for the arthroscope and to visualize the radiocapitellar joint. In most cases, the anterior aspect of the capitellum is normal. In the case of a more anterior lesion, the 70-degree arthroscope may be used to visualize through

A

the ­lateral portal to manage the lesion, but the damage is usually more posterior and best treated by visualizing from posterior. The anterior compartment is then evaluated for loose bodies, which are removed through an anterolateral portal if necessary. The inflow is then left on the canula in the proximal anterior medial portal, and the posterior compartment of the elbow is entered. The olecranon fossa is also evaluated for loose bodies, which are removed through a posterior central or posterior lateral portal. Attention is then directed to the lateral gutter. The posterolateral plica is evaluated and removed if necessary to allow visualization of the capitellum. The OCD lesion is best visualized through a superior posterolateral portal with a 70-degree arthroscope. This leaves the soft spot and straight lateral portal free for instrumentation (Fig. 19D-4). The cartilage cap may be probed through these portals for softness and fissures. Increasing the flexion of the elbow may be necessary to visualize the entire lesion. A shaver is introduced,

B

Figure 19D-4   A, The osteochondritis dissecans (OCD) lesion of the capitellum visualized with a 30-degree arthroscope from the posterolateral portal with the patient in the prone position. B, The OCD lesion of the capitellum as visualized from the same portal with a 70-degree arthroscope.

1244 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

and the necrotic bone is débrided to a stable bed. All loose bodies are removed as well. A stable rim of cartilage is left in place. It is important to preserve the lateral aspect of the capitellum (the shoulder) because it provides both bone stability and the attachment of the lateral capsule and ligaments. Once the débridement has been completed, the base of the lesion is drilled or picked to stimulate increased blood flow. Occasionally one may encounter a large and viable fragment. In these cases, the fragment is “hinged” open and the base débrided. The fragment is then replaced into the defect and stabilized using a Kirschner wire. Many techniques for permanent stabilization have been described, including Herbert-Whipple screw fixation, dynamic stapling, cancellous screws, and bioabsorbable implants. In the most severe cases, osteochondral grafting from the knee, ribs, synthetics, or allograft has been reported. None of the synthetic implants have been approved for use in this area, and their use should be considered experimental at this time. In these cases, the procedure is as listed above for débridement, and then the distal soft spot posterior portal is used to place the plugs. The elbow must be hyperflexed to allow correct orientation of the implants, which are then contoured to match the capitellar surface (Fig. 19D-5).

WEIGHING THE EVIDENCE There is no level-one study available on OCD of the elbow. Most of the long-term studies are case series identified retrospectively. Three case series studies on the longterm natural history of OCD of the humeral capitellum are reviewed. Takahara followed 24 patients with OCD of the capitellum nonoperatively for an average of 5.2 years and found the results correlated directly with the severity of the lesion on presentation, with 6 of 11 “early” lesions healing or improving.12 In the same year, he presented a comparison study of 53 patients, with 14 managed nonoperatively and 39 managed by surgery. The end results correlated mostly with the size of the lesion rather than type of treatment. This was not a randomized study.13 Woodward retrospectively reviewed 42 patients and followed 24 patients for 2 to 34 years after initial treatment. He found that, although subjectively most thought their elbow was normal, there was a consistent loss of full extension.11 Surgery in athletes has been reported by several case series with mixed results. Baumgarten and colleagues14 and Ruch and associates15 have reported relatively good results (13 of 16 and 9 of 12, respectively) with short-term followup after arthroscopic débridement. In each series, the more extensive the lesion at the time of surgical treatment, the less successful was the result. In contrast to the above work, Byrd and Jones presented a retrospective cohort series of 10 baseball ­players, with only 4 returning to unrestricted play and 5 cases of arthrofibrosis.16 Jackson and colleagues ­ presented the ­classic article on elite female gymnasts with osteochondritis and found that only 1 of 10 was able to continue her career.17

Other series have echoed the findings of the aforementioned authors, with results related to the size of the lesion rather than treatment. Of the cases of grafting for the more severe defects, only anecdotal evidence exists. ElAttrache and Savoie have presented, but not published, cases of osteochondral grafts placed in the more severe cases for loss of the structural integrity of the capitellum, with satisfactory short-term results.

Authors’ Preferred Method The authors prefer nonoperative management in most cases of early lesions. If the radiographs and MRI reveal an intact cartilage cap or a relatively small (8 mm or less) lesion, then the patient is managed with full-time doublehinged elbow bracing and physical therapy. Once the elbow is pain free, the patient is allowed to return to sports with the brace in place and with improvement of core strength, posture, and any improper sporting techniques that may produce increased stress on the elbow. The patient is allowed to return to sports with the brace in place once the elbow has regained a full and painless arc of motion. All activities are allowed with the brace in place until ­radiographic evidence of healing occurs. In the case of female gymnasts, we usually recommend quarterly MRI if possible and also suggest radiographs and at least 1 MRI of the opposite, asymptomatic elbow. If the patient has loose bodies, extensive involvement of the capitellum, or progression of the disease, then surgery is recommended. The procedure is performed arthroscopically with removal of loose bodies and débridement and drilling of the base of the lesion. Large lesions are repaired with a cannulated Herbert-Whipple–type screw, but this type of lesion is extremely rare in our practice. In the more extensive lesions with loss of the lateral cortex, we use osteochondral grafting using the proximal olecranon as the source of the graft.

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS In most cases of débridement, the patient is started on a continuous passive motion machine on the day of surgery. The elbow is placed in a brace to offload the capitellum. Physical therapy for gentle stretching, compressive pumping out of edema, and hand-wrist exercises are initiated as early as possible, usually with 1 week of surgery. The patient is started on general conditioning within the first 3 weeks of surgery. The elbow is followed on serial radiographs as well as clinically. After pain-free range of motion and satisfactory strength are obtained, return to sports is allowed, usually between 6 and 16 weeks. The main outcome measurement used in most studies is the return to unrestricted sporting activity. This usually occurs between 10% and 90% of the time, with both nonoperative and operative management. This success or failure is determined by the extent of the lesion and the level of sport rather than by the type of treatment.

Elbow and Forearm 1245

A

B

C

D

E

Figure 19D-5   The sequence in the treatment of osteochondritis dissecans (OCD) lesions of the capitellum.   A, Simple débridement and removal of the loose fragment.   B, Drilling of the residual defect to stimulate cartilage formation. C, In rare late-stage OCD or when the lateral aspect of the capitellum has been lost, arthroscopic osteochondral grafting may be indicated to bolster the lateral joint and possibly prevent instability and progressive arthritis. D and E, The radiograph of the patient in C, 6 weeks after osteochondral grafting.

1246 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Complications include recurrence or advancement of the lesion and arthrofibrosis. In most studies, the more extensive lesions were associated with increased risk for arthritis, stiffness, and ongoing pain. Failure to return to sport at the preinjury level is quite common, occurring 10% to 20% of the time in the best case series.1,4-7,10-17

CRITERIA FOR RETURN TO PLAY A pain-free elbow with full range of motion is the criterion for return to play with a protective off loading brace. Resolution of the lesion by radiographs or by MRI is the criterion for return to play without the brace.

SPECIAL POPULATIONS The elite female gymnast represents a special population. Although osteochondrosis (Panner’s disease) is more common in this age range (<10 years), the lesions in these athletes behave more like a severe osteochondritis. Osteochondrosis may often occur bilaterally and should be managed in an aggressive but nonoperative manner. Surgery in these patients will often end their career, so early detection and management by bracing and protection are essential. Surgery should be undertaken with great reluctance and with full knowledge of the possibility of a good patient but poor athletic result.

C

r i t i c a l

P

o i n t s

l Lateral elbow pain in a repetitive athlete should signal a warning sign of OCD. l Presenting symptoms are loss of motion and lateral swelling. l Radiographs may be equivocal. l MRI is diagnostic. l Early detection and treatment give the best results. l Gymnasts should undergo bilateral examination. l Bracing assists nonoperative management. l Arthroscopy is the best surgical treatment but results in less than 90% return to play.

S U G G E S T E D

R E A D I N G S

Cain EL, Dugas JR, Wolf RS, Andrews JR: Elbow injuries in throwing athletes: A  current concepts review. Am J Sports Med 31(4):621-635, 2003. Schenk RC Jr, Goodnight JM: Osteochondritis dissecans. J Bone Joint Surg Am 78:439-456, 1996. Yadao MA, Field LD, Savoie FH 3rd: Osteochondritis dissecans of the elbow. Instr Course Lect 53:599-606, 2004.

R eferences Please see www.expertconsult.com

S ect i o n

E

Olecranon Bursitis Richard J. Thomas and Larry D. Field

Usually a benign anatomic structure, the olecranon bursa can be a source of significant morbidity when it becomes inflamed or infected. The olecranon bursa is a synovium-lined sac filled with highly viscous lubricating fluid to allow the skin to glide smoothly over the bony prominence of the olecranon.1 It is susceptible to trauma because of its superficial location. Olecranon bursitis is more likely to develop in patients who commonly endure trauma to the tip of their elbow such as swimmers, weightlifters, gymnasts, skiers, and those with occupations requiring frequent elbow ­flexion and extension or constant pressure on the olecranon. About 40% of patients who develop olecranon bursitis give a history of trauma to the elbow.2

ANATOMY AND PATHOLOGY The olecranon bursa is the only bursa of the elbow. It is a subcutaneous synovial pouch that reduces friction between the skin and the tip of the olecranon process during flexion and extension of the elbow.3,4 The bursa is a closed space that does not communicate with the elbow joint. It lies between the skin and the insertion of the triceps tendon. The olecranon bursa is rarely found in children younger than 7 years. It has also been found to be larger in the dominant elbow.1,5 The floor of the bursa is usually adherent to the olecranon. However, a chronically inflamed bursa may have an absent floor. A normal olecranon bursa typically has a smooth synovial lining, whereas an inflamed bursa tends to

Elbow and Forearm 1247

have thickened synovium with inflammatory and fibrinous tissue present.1 Traumatic or nonseptic olecranon bursitis is the most common type of olecranon bursitis.6 The patient often gives a history of repetitive microtrauma or a specific traumatic event to the tip of the elbow. The trauma initiates an acute inflammatory response that causes bursal wall thickening and increased fluid production. Occasionally the bursa will fill with blood because of vessel disruption from the trauma. A pedunculated fibrous mass within the bursa has been described in association with traumatic olecranon bursitis and may have a pathogenetic role.7 Gout and pseudogout have also been associated with a crystal-induced bursitis.3 Minor bursal trauma may lead to intrabursal crystal shedding, increased bursal fluid production, and inflammation. Septic olecranon bursitis usually develops secondary to local small breaks in the skin over the tip of the elbow. Steroid injections have been shown to increase the risk for developing septic bursitis.3,8 Hematogenous seeding is rare, probably owing to the limited vascularity of the area.9 Osteomyelitis of the olecranon can occur from a chronic septic bursitis.10 Staphylococcus aureus is responsible for 80% of all olecranon septic bursitis.11,12 Streptococcal species contribute to 5% to 30% of septic bursitis, and Staphylococcus epidermidis contributes to smaller percentages. S. aureus septic bursitis has a seasonal trend that peaks in the summer.12 Immunocompromised patients may present with more uncommon pathogens, have a refractory infection, and may take longer to resolve the bursitis.12 Risk factors for septic olecranon bursitis include frequent or sustained pressure on the bursa, recent trauma, alcoholism, pre-existing bursal disease, chronic obstructive pulmonary disease, diabetes, chronic renal failure, systemic corticosteroid therapy, and prior aspiration or steroid injection of the bursa.1,12

CLINICAL EVALUATION As with any medical problem, a good history and physical examination are crucial in the evaluation of olecranon bursitis. Because the initial treatments for septic and nonseptic bursitis differ, it is important to try to differentiate the pathology between the two possible diagnoses. Therefore, detailed questioning pertaining to a history of trauma, fevers, chills, malaise, or a history of injections to the elbow should be carried out. About half of patients with septic bursitis have an abrupt onset of pain in the elbow. The other patients have an insidious onset of pain.13 On examination, patients may have palpable fluid in the bursa ranging from a few drops to 40 mL (Fig. 19E-1). However, patients may exhibit only bursal tissue thickening.1,3 Typically, patients have a painless range of motion except at extreme flexion, where the bursal tissue is stretched. If the patient has a painless bursa with no redness or warmth, the bursitis is rarely septic. Patients with nonseptic bursitis have only mild tenderness about half of the time and have mild peribursal edema, warmth, and erythema 25% of the time.3 A painful, red, and warm bursa is concerning for infection and should prompt the clinician to aspirate the bursa. The needle should be introduced from a lateral approach to prevent a needle tract directly over the lines of tension of the elbow. This minimizes the chance for a chronic draining

Figure 19E-1   Photograph of an enlarged olecranon bursa consistent with nonseptic olecranon bursitis.

sinus.4,14 The fluid should be sent for culture, Gram stain, crystals, and cell count. The white blood cell count in septic bursitis is typically between 1500 and 30,000/μL, with a predominance of polymorphonuclear cells. The white blood cell count in nonseptic bursitis is usually less than 2000, ranging from 50 to 11000 with a predominance of mononuclear cells.3,4,6 Empirical antibiotics should be started following aspiration if there is suspicion for infection. Radiographs are rarely useful in the diagnosis of olecranon bursitis. They are usually normal, although patients may have olecranon spurs or calcium deposits visible on plain films (Fig. 19E-2).15 Magnetic resonance imaging (MRI) is rarely indicated in making a diagnosis of olecranon bursitis. However, it may be used to evaluate for olecranon osteomyelitis or abscess. There is considerable overlap in MRI findings in septic and nonseptic olecranon bursitis.16 However, marked lobulation, complexity, soft tissue edema, and thickening of the triceps tendon are twice as common in septic compared with nonseptic olecranon bursitis.16

Figure 19E-2   Radiograph showing the soft tissue shadow of olecranon bursitis.

1248 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

TREATMENT Nonseptic Olecranon Bursitis Nonseptic olecranon bursitis can usually be treated conservatively and rarely requires surgical intervention. When the bursitis is caused by an acute traumatic event, the blood in the bursa should be aspirated, and a compression dressing should be placed. Ice can be applied to the elbow and may help to prevent the bursitis from becoming chronic. Multiple aspirations may be required if blood or fluid ­reaccumulates.1 When a specific traumatic event has not occurred, and the bursa is minimally tender, the patient may be treated symptomatically with aspiration for large fluid collections, 7 to 10 days of nonsteroidal anti-inflammatory drugs (NSAIDs), and a compression dressing (Fig. 19E-3). A small amount of steroid may be injected into the bursa, although multiple injections should be avoided because of the increased risk for the development of septic bursitis.17 The patient should wear an elbow pad when the elbow is at risk for trauma. If the patient has a history of gout, and the aspiration has monosodium urate crystals, oral colchicine should be considered.17 Smith and associates18 performed a prospective randomized study that supported the use of a single injection of steroid for nonseptic olecranon bursitis. The study showed a rapid decrease in swelling and a decreased incidence of recurrence. McAfee and Smith3 showed that 70% of bursitis effusions resolved a month after aspiration and conservative management without steroid injection. Ninety percent of these patients had complete resolution by 6 months without injection. McAfee and Smith showed that 85% of effusions resolved within 1 week of aspiration and steroid injection. However, the injection patients had a higher incidence of complications. Twenty percent of these patients had skin atrophy over the olecranon, 30% of these patients had chronic pain over the olecranon, and 10% of these patients developed septic bursitis.

Hassel and associates19 described aspirating and injecting 250 mg of tetracycline into the bursa of chronic cases with good results. The tetracycline theoretically acts as a sclerosant. In patients who have failed conservative management for their nonseptic olecranon bursitis, surgery is indicated. Most authors recommend complete excision of the bursa.1 A transverse or longitudinal incision may be used over the olecranon bursa. Placement of the incision directly over the tip of the olecranon should be avoided to prevent postoperative scar sensitivity. Care should be taken to keep the skin flap as thick as possible to prevent wound problems or skin necrosis. After bursal excision, meticulous closure is crucial, and a compressive dressing is placed.7 Stewart and colleagues7 performed 21 complete olecranon bursectomies with 94% complete and long-term relief in patients without rheumatoid arthritis and 40% complete and long-term relief in patients with rheumatoid arthritis. Three of the five patients with rheumatoid arthritis had recurrence of their bursitis. Endoscopic bursectomy has also been described.20,21 Ogilvie-Harris and Gilbart21 performed 31 endoscopic bursectomies. Eighty-six percent of the patients had no pain after recovery from surgery. The study showed less morbidity, less skin trauma, faster recovery, and less stiffness than conventional bursectomies. There were no wound infections in this study. Ogilvie-Harris and Gilbart21 describe the technique of endoscopic bursectomy. A 5-mm arthroscope and a 4.5-mm curved shaver are used. The bursa is injected with fluid before portal placement, and the portals are then placed 1 cm distalmedial and distal-lateral to the bursal sac with the elbow in extension. The bursa is then resected from the inside to the outside. However, several studies have shown that patients with rheumatologic disease do not do well with endoscopic bursectomies.7,20-22 Goldwirth and colleagues23 describe a technique using sterile talcum powder as a sclerosant for recurrent olecranon bursitis. An open bursectomy through a posteromedial approach is performed, and talcum powder is then added to the dead space before closure over a drain. This technique was performed on 11 patients with no complications and 1 minor recurrence that did not require further surgery.

Septic Olecranon Bursitis

Figure 19E-3   Photograph of an aspiration of olecranon bursitis.

Treatment of septic olecranon bursitis requires a more aggressive approach. If there is suspicion for septic bursitis, the patient’s bursa should be aspirated and cultures sent. If the patient has septic olecranon bursitis with no systemic signs and minimal cellulitis, the bursa should be aspirated and kept decompressed with serial aspirations as needed, and the patient should be placed on oral broad-spectrum antibiotics. Because most cases of septic bursitis are caused by S. aureus, oxacillin, cephalexin, or erythromycin should be initiated empirically until culture results return. Antibiotics can then be adjusted according to culture and sensitivity results.1 A 14-day course of oral antibiotics will cure most cases of septic bursitis.22 A longer course of antibiotics may be required if the patient has been infected for more than 2 weeks before treatment, the patient has a past history of septic or nonseptic bursitis,

Elbow and Forearm 1249

rheumatoid nodules or gouty tophi are present, the patient is immunocompromised, systemic signs and symptoms are present initially, or unusual organisms are cultured.3 If the patient shows no clinical improvement with oral antibiotics and serial aspirations, irrigation and débridement with bursectomy and intravenous antibiotics are indicated.1 A drain should be left in place for 24 to 48 hours after the bursectomy, and the patient should be immobilized until the wound is healed. For unusual organisms, an infectious disease consult may be required. Another technique for treating septic bursitis is described by Knight and associates,24 in which a 3-mm polyethylene tube is placed in the bursa, an antibiotic solution of kanamycin and polymyxin is instilled for 3 hours, and then suction is placed on the tubing for 9 hours. Intravenous antibiotics are used until cultures from the tubing are negative for 3 consecutive days, at which point the patient is placed on oral antibiotics for 1 week, and the tubing is removed. No recurrences occurred in 10 patients using this technique. It has been shown that the length of time necessary to sterilize the bursal fluid is proportional to the length of time that the infection has been present.2

absorbable subcutaneous sutures to tack the skin down to the underlying triceps fascia to prevent space for hematoma formation. A Penrose drain is left in place for 24 to 48 hours, and the patient is placed in an anterior splint with the arm in 60 degrees of flexion for 2 weeks. A compressive wrap is placed on the elbow for the next 4 weeks.

POSTOPERATIVE CARE AND REHABILITATION After bursectomy, wound drainage for 24 to 48 hours is recommended with a Penrose or suction drain. The patient’s elbow should be splinted in 45 to 60 degrees of flexion with an anterior splint. Too much flexion places stress on the wound, whereas too much extension may compromise future motion. The clinician may consider allowing early controlled range of motion in patients with rheumatologic disease because of the risk for stiffness in these patients.7 Athletic participation should be allowed 4 weeks after surgery, with restriction from contact for a total of 6 weeks. Soft elbow pads should be used until no tenderness is ­present.

Authors’ Preferred Method In our practice, operative treatment is rarely required. In our initial evaluation, we determine whether the bursitis is fluctuant or not. If a significant amount of fluid is present, the bursa is aspirated under sterile conditions, and the fluid is sent for culture and crystal evaluation. If monosodium urate crystals are present, the patient is started on indomethacin or colchicine and referred to the primary care physician for preventive treatment of gout. If the fluid is purulent and the patient is afebrile, a course of oral antibiotics is initiated. If the patient is febrile, we admit the patient and start the patient on intravenous antibiotics. ­Finally, if the fluid is nonseptic, a compressive dressing is applied for 1 week, and the patient is started on an NSAID. Occasionally, a patient will present with a nonfluctuant bursitis in which no fluid can be aspirated, and the bursa is merely thickened and inflamed. In this case, the patient is treated with an NSAID or a Medrol dose pack. An ­elbow pad is also prescribed to protect the olecranon from continued trauma. All patients with olecranon bursitis are ­instructed to avoid inciting activities. We inject 40 mg of Depo-Medrol only in resistant nonseptic cases that have not responded to aspiration and compression. The patient is educated to the fact that this ­injection increases the risk for septic olecranon bursitis. Finally, surgical excision of the bursa is reserved for patients who have failed all the previously mentioned conservative treatments or have a significantly septic olecranon bursitis that is not responding to intravenous antibiotics. We prefer using a longitudinal incision curved around the lateral aspect of the tip of the olecranon and take care not to thin the skin over the olecranon too much. We use a 69 “beaver” blade to carefully shell out the olecranon bursa. Once removed, we send it to pathology for evaluation and culture. Closure should be meticulous, and we like to use

C

r i t i c a l

P

o i n t s

l Staphylococcus aureus is the most common cause of ­septic olecranon bursitis. l Painful olecranon bursitis is concerning for infection or gout and should prompt the clinician to aspirate the bursa and send the fluid for evaluation. l Nonseptic olecranon bursitis rarely requires surgical intervention and usually responds to aspiration, compression, and NSAIDs. l Surgical excision of the olecranon bursa can result in wound complications and skin necrosis if care is not taken to leave the flap as thick as possible and to close the dead space. l Septic olecranon bursitis can initially be treated with 14 days of oral antibiotics and serial aspirations. If this treatment is not effective, however, the patient may need IV antibiotics and ultimately a bursectomy.

S U G G E S T E D

R E A D I N G S

Kerr DR: Prepatellar and olecranon bursectomy. Clin Sports Med 12:137-142, 1993. Ogilvie-Harris DJ, Gilbart M: Endoscopic bursal resection: The olecranon bursa and prepatellar bursa. Arthroscopy 16:249-253, 2000. Raddatz DA, Hoffman GS, Franck WA: Septic bursitis: Presentation, treatment and prognosis. J Rheumatol 14:1160-1163, 1987. Stewart NJ, Manzaneres JB, Morrey BF: Surgical treatment of aseptic olecranon bursitis. J Shoulder Elbow Surg 6:49-54, 1997. Smith DL, McAfee JH, Lucas LM, et al: Treatment of nonseptic olecranon bursitis: A controlled, blinded prospective trial. Arch Intern Med 149:2527-2530, 1989.

R eferences Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 1250 DeLee & Drez’s� O

S e c t i o n

F

Forearm Fractures 1. Fractures of the Elbow in the Adult David Ring

Trauma to the adult elbow can be challenging to treat by virtue of the complex articular structure, complex capsuloligamentous and musculotendinous arrangements, and the proximity of neurovascular structures. An increasing understanding of elbow injuries has led to a rapid evolution in treatment concepts.1 Awareness of the patterns of injury and the pitfalls of each can lead to restoration of a functional elbow in most patients. Most complications can be addressed with secondary surgery, but certain key structures, such as the ulnotrochlear relationship, must be reconstructed and protected by the initial treatment, or salvage measures will become necessary. Some simple elbow injuries (e.g., simple fractures of the olecranon or radial head or simple dislocation of the elbow) are associated with a relatively rapid recovery and return to high-level sports activities; however, the more complex elbow fractures and fracture-dislocations can severely compromise future athletic activity. Fortunately, such complex elbow injuries are uncommon in athletes.

FRACTURES OF THE DISTAL HUMERUS

proximal radius and ulna are subtracted from the image.4 Combined with an awareness of the typical injury patterns of the distal humerus, I encounter fewer surprises upon operative exposure when three-dimensional CT has been used; however, operative exposure remains the definitive method for characterizing the fracture, and adequate exposure of complex fractures is important. Mehne and Matta’s description of bicolumnar fracture types reflects the increasing experience with operative treatment.5 They identified the fact that the columns can fracture very low on one or both sides, making internal fixation much more challenging. In addition, there are frequently completely separate articular fragments. Many fracture texts depict bicolumnar distal humerus fractures as creating a simple articular split in the distal humerus. This is unfortunately rarely the case, at least in my experience. There is usually a rotational component to the failure of the articular surface that creates a posterior or anterior butterfly fragment of articular surface (Fig. 19F1-1). There may also be a shearing component to the fracture that creates a coronal split in the articular surface (Fig. 19F1-2).6 In addition to metaphyseal comminution, there may also be metaphyseal impaction (Fig. 19F1-3).

Indications and Contraindications Minimally displaced fractures of the distal humerus are uncommon, and most distal humerus fractures require surgery. Total elbow arthroplasty is used increasingly for these fractures and the early results can be seductively good,2,3 but arthroplasty is not an option in the athlete because it requires a lifetime 5-kg lifting limit. Total elbow arthroplasty is appropriate only for infirm and inactive patients and patients with preexisting elbow destruction (e.g., from rheumatoid arthritis) and are therefore not described in this chapter.

Preoperative Evaluation Standard radiographs will identify that the distal humerus has been fractured, but they are notoriously poor at demonstrating the complexity of the fracture, particularly at the articular surface. Radiographs taken with axial traction applied to the limb can improve visualization of the fragments but are usually not available until the patient is anesthetized and are therefore not ideal for planning the surgery. Two-dimensional and three-dimensional ­computed tomography (CT) images are useful for characterizing articular fragmentation, particularly when the

Figure 19F1-1   Most bicolumnar fractures include some fragmentation of the articular surface, most commonly as an anterior or posterior wedge of articular surface. Better access to the anterior articular surface can be obtained by rotation of the columns. Small articular fragments can be secured with small threaded Kirschner wires placed into the subchondral bone and buried.

Elbow and Forearm 1251

column fractures, involve complex comminution of the articular surface.7 Capitellum fractures have long been classified as either a (1) thick fragment, (2) thin fragment, or (3) comminuted fragment.8 This classification system is no longer adequate because it has become clear that isolated fractures of the capitellum are uncommon.7,9 Most fractures of the capitellum involve part of the trochlea,9-19 and many are much more complex injuries.7,20 There appears to be a progression of injury, with any fracture involving more than the anterior surface of the capitellum and trochlea also creating a fracture of the lateral epicondyle, followed by the next stage of impaction of the lateral column of the distal humerus (see Fig. 19F1-3), then fracture of the posterior aspect of the trochlea, and finally fracture of the medial epicondyle (Fig. 19F1-4).7

Operative Treatment in the Athlete Operative Exposure

Figure 19F1-2   Some distal humerus fractures create entirely articular fracture fragments (no nonarticular areas for implant placement). The fragment in the lower left part of the wound is the capitellum and part of the trochlea, split off in the coronal plane and flipped 180 degrees.

This usually becomes apparent when the fragments cannot be realigned accurately. Many of these factors are not accounted for in current classification systems, but their identification and appropriate treatment are important to optimize outcome. Isolated fractures of the lateral column above the base of the olecranon fossa are uncommon and are usually simple fractures. Some lateral column fractures, and most medial

A

Exposure of the distal humerus poses several challenges and is an active area of debate. A study in cadavers has demonstrated that olecranon osteotomy provides the best exposure of the articular surface of the distal humerus,21 but concern regarding problems with healing and symptomatic implants used to repair the olecranon osteotomy has limited enthusiasm for the technique. Alternatives to olecranon osteotomy include elevating the triceps from the posterior humerus, but leaving it attached to the olecranon (credited to Alonso-Llamas22,23); splitting the triceps and partially elevating it from the olecranon (the traditional Campbell exposure recently repopularized by McKee and colleagues,24,25 particularly for open fractures of the distal humerus); elevation of the triceps from both the humerus and the proximal ulna either as an intact sleeve (Bryan-­Morrey exposure26) or as a long tongue that is reflected ­proximally and then reattached (the triceps reflecting, anconeus pedicle, or TRAP, exposure described by O’Driscoll27); or an extensile lateral exposure, most often used for complex

B

Figure 19F1-3   Apparent capitellar fractures are often far more complex. A, The fracture usually extends into the trochlea.   B, Even more important, if the fragments do not fit back onto the distal humerus as illustrated here, there is likely some impaction   of the posterior part of the lateral column or the posterior aspect of the trochlea.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1252 DeLee & Drez’s� O

1 5

2

4 3

For operative exposure of bicolumnar fractures and complex articular fractures of the distal humerus, I prefer olecranon osteotomy. For simple articular and extra-­articular fractures, I will often start by developing the Alonso-­Llamas exposure and then cut the olecranon only if the fracture is more complex than anticipated or difficult to reduce. For most complex articular fractures (variations of the capitellum fracture), I use an extensile lateral exposure.

Olecranon Osteotomy Figure 19F1-4   Apparent capitellar fractures are often more complex. Complex fractures involving the articular surface and not the columns are being recognized more frequently and appear to follow a progression of injury. The first stage is fracture of the capitellum and anterior trochlea (1), followed by fracture of the lateral epicondyle (2), impaction of the posterior aspect of the lateral column (3), fracture of the posterior trochlea (4), and finally fracture of the medial epicondyle.

articular fractures of the distal humerus7 and more complex variations of the capitellum fracture. One study documents comparable elbow extension strength after the Campbell and olecranon osteotomy exposures,24 but triceps-­elevating exposures sometimes cause elbow extension weakness, and these exposures are occasionally complicated by triceps avulsion.28,29 Access to articular fragments is one of the major advantages of the olecranon osteotomy, but these can be accessed through other exposures by rotating the fractured condyles posteriorly into the wound so that the olecranon does not impede access (see Fig. 19F1-1), or by using a fractured condyle or epicondyle to allow rotation or subluxation of the distal humerus away from the ulna.

A

B

The patient is placed in a lateral decubitus position with the arm supported over a bolster. Either the prior incision or a straight posterior incision is used. Medial and lateral skin flaps are elevated, with care taken to protect cutaneous nerve branches and keep them within the skin flaps. The ulnar nerve is identified along the medial border of the triceps, dissected at least 6 cm proximally and distally, and left in an anteriorly transposed position in the subcutaneous tissues. The insertion of the anconeus onto the proximal ulna is partially elevated to directly view the arc of the trochlear notch. Alternatively, if the ulnar nerve has been transposed, the medial elbow capsule can be incised in the bed where the ulnar nerve originally lay. If the fracture is simple enough that reduction of the metaphyseal fracture lines can be relied on to achieve a good articular reduction, the triceps can be elevated from the posterior aspect of the distal humerus on both the medial and lateral sides and the osteotomy made only if necessary. In most patients, however, an osteotomy is necessary for adequate reduction and fixation. An apex distal chevron-shaped osteotomy is planned so that it enters the joint at the depths of the trochlear notch. An oscillating saw is used to start the osteotomy (Fig. 19F1-5A). A small straight osteotome is then used to complete it by levering the osteotome proximally and

C

Figure 19F1-5   An olecranon osteotomy can be used for exposure of fractures of the distal humerus with limited complications if careful technique is used. A, A chevron-shaped osteotomy will be easier to reposition and more stable. The medial or lateral capsule can be opened to judge the position of the osteotomy. B, The osteotomy is started with a saw but finished with an osteotomy to create additional interdigitations to help with repositioning and the stability of fixation. C, The posterior capsule is then incised, and the triceps is mobilized proximally.

Elbow and Forearm 1253

Extensile Lateral Exposure

Figure 19F1-6   An extensile lateral exposure is useful to expose complex anterior fracture of the articular surface. The lateral epicondyle with the attached lateral collateral ligament and common extensor is mobilized distally, the radial wrist extensors are elevated anteriorly, the triceps is elevated posteriorly, and the elbow is subluxated slightly.

cracking the subchondral bone (see Fig. 19F1-5B). This maneuver creates an uneven surface that facilitates repositioning and may enhance stability. The posterior elbow capsule is then incised, and the olecranon fragment and triceps are ­elevated from the posterior aspect of the humerus (see Fig. 19F1-5C). Repair of the osteotomy proceeds as described later for the fixation of olecranon fractures with a tension band wire or screw.

A

A midline posterior or a lateral skin incision can be used. The patient is positioned supine with the arm supported on a hand table. A sterile pneumatic tourniquet is used. When there is a fracture of the lateral epicondyle, it can be mobilized and retracted distally along with the attached origins of the wrist and digital extensor muscles and the lateral collateral ligament complex; otherwise, the lateral collateral ligament may need to be released (either directly or by osteotomy) for exposure. The more proximal origins of the radial wrist extensor muscles are elevated from the lateral supracondylar ridge to improve the exposure to the anterior articular fragments. The exposure is completed by the elevation of the lateral aspect of the triceps from the distal humerus and the proximal olecranon, permitting the elbow joint to be hinged open and providing exposure to both the anterior and posterior articular surfaces of the distal humerus (Fig. 19F1-6).

Fixation Techniques: Columnar Fractures Fixation techniques continue to evolve, reflecting, in part, dissatisfaction with fixation of some of the more complex fractures. As a result of the support of several biomechanical studies30-32 and the teaching of numerous authoritative surgeons,33 it has become standard practice to repair bicolumnar fractures of the distal humerus with two plates placed orthogonal to one another (e.g., one plate posterolateral and the other directly medial, or one plate posteromedial and the other directly lateral) (Fig. 19F1-7). Conversely, for many of the fractures that occur very low on one or

B

Figure 19F1-7   Orthogonal plating. A, The long-standard and popular approach has been to apply two plates orthogonal to one another. B, Usually one is directly medial and one posterolateral.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1254 DeLee & Drez’s� O Figure 19F1-8   Triple plating. A, The fracture creates very small articular fragments and metaphyseal comminution fixation of the articular fragments may be limited to a single screw. B, Adding a third or fourth plate can enhance fixation in this circumstance.

A both columns—Mehne and Matta lambda or low-T fractures5—there may be room for only a single screw in the distal fracture fragment. One suggestion for dealing with this is to add a third, and even a fourth, plate, the idea being that two orthogonal plates for a single low columnar fracture provide an additional screw for fixation and that the fixation in orthogonal planes will add additional stability (Fig. 19F1-8).34-36 Another option represents the evolution from an earlier attempt to repair distal humerus fractures using a single, thick, precontoured lateral plate (the so-called Dupont plate), which has led to the concept of placing two plates parallel to one another directly on the medial and lateral surfaces of the distal humerus (Fig. 19F1-9). This concept also has good biomechanical support.32,37 Furthermore, O’Driscoll and colleagues offer several principals for secure internal fixation of the distal humerus,1 many of which are based around the use of parallel rather than perpendicular plates, that I have found applicable and appealing in the treatment of fractures of the distal humerus. Screws for fixation of the distal fragment should (1) pass through a plate, (2) be as long as possible, (3) be as numerous as possible; (4) engage as many articular fragments as possible, and (5) engage a fragment on the opposite column that is secured by a second, parallel plate. The fixation of the distal fragments to the shaft should be strong enough to allow immediate motion. One-third tubular plates are therefore inadequate. O’Driscoll believes that compression and load sharing of the distal and proximal fracture fragments at the supracondylar level is so important that comminution should be excised rather than bridged and a so-called supracondylar shortening osteotomy performed. He recommends anterior translation of the fragments and burring out a new olecranon fossa to enhance motion. It is popular to attempt to place a long screw from the distal fragments up the medial or lateral column of the ­distal humerus. Although it can help enhance fixation, caution is

B merited. In most cases, the security of the fixation would be better enhanced by ensuring that the distal screws are well fixed in the distal articular fragments (Fig. 19F1-10). The recent concept of angular stable, or locking screw plates (plates with screws that thread directly into and lock to the plate), can also been used in the distal humerus.38

Figure 19F1-9   The use of parallel plates (one placed directly medial and one directly lateral) provides for a greater number and length of screws in the distal fragments, and the smaller distal screws can help secure smaller articular fracture fragments. The strength of the construct is further enhanced by having every screw go through a plate hole.

Elbow and Forearm 1255

Figure 19F1-10   Although the so-called home run screw—a long screw aimed up the medial or lateral column—is favored by some surgeons, it is important to remember to obtain adequate hold on the distal fragments as this nonunion demonstrates.

One difficulty of using locking plates in the distal humerus is the need for precise placement of distal screws through the central portion of the trochlea in order to avoid violation of an articular surface. Finally, distal humerus fractures often create entirely articular fragments that can only be secured with implants that are countersunk beneath the articular surface.6,7 Small threaded Kirschner wires can be very useful in this regard because there is often limited bone opposite these fragments for internal fixation. The small threaded Kirschner wires can be placed through the edge of the fracture site connecting the subchondral bone of adjacent fragments (see Fig. 19F1-1). When the major articular fragments are reduced, the small wires are buried. For larger articular fragments with an opposite bony support, headless screws (the Herbert screw and variations, many of which are now cannulated) are useful. I would caution against the use of a countersunk threaded screw because these allow settling, which results in protrusion of the screw head into the joint—it is better if the head is threaded so that it cannot back out as easily. I favor the concept of parallel plating but will use whatever plate configuration gives the best fixation. I attempt to repair small articular fragments and rarely discard them. I believe these provide support and assist with healing, but my primary reason for repairing them is to monitor and ensure accurate reduction of the major articular fragments. Fragments of the thin bone between the olecranon and coronoid facets are occasionally discarded, and small articular fragments that appear to be doing more harm than good despite attempts to repair them can also be discarded.

Many small articular fragments can be secured with headless screws to major fragments of the distal humerus, but others are better suited to repair with small threaded Kirschner wires inserted through the fracture surface and connecting the subchondral bone of adjacent fragments. Nonthreaded Kirschner wires may be more risky because they have been known to migrate throughout the body. In my experience, most bicolumnar fractures have at least one small separate articular fragment or additional fragmentation of the articular surface; therefore, an interfragmentary compression screw between the major medial and lateral fragments—as is often described—can rarely be used. Metaphyseal comminution can be addressed by either bridging the fragments with stable internal fixation or resecting comminution and shortening the distal humerus. Bridging the fragments preserves the olecranon and coronoid facets, avoids the need for shortening of a relatively intact column given that metaphyseal comminution often involves only one column, and usually does not impede healing, provided that stable internal fixation is achieved (this is a metaphyseal region with an excellent blood supply). Conversely, resecting the comminution to obtain stable apposition of major fracture fragments increases the initial stability of the fixation. In practice, I balance these two rationales and make the decision intra-operatively. In my experience, the need to shorten the bone is uncommon. The traditional operative tactic for columnar fractures involves fixation of the articular fragments to one another with interfragmentary compression screws followed by fixation of the repaired distal humerus back to the shaft. The screw for fixation of the articular fragments can be initiated from inside the fracture site to facilitate central placement. Two 3.5-mm plates (usually reconstruction plates) are applied perpendicular to one another (usually posterolateral and direct medial). Fractures that are low on one side (Mehne and Matta lambda fractures) often benefit from modifications of the traditional technique, including fixation of the larger column fragment back to the shaft as a first step, followed by repositioning and fixation of the smaller fragments, and use of a third plate to enhance fixation of the distal fragments on the lower columnar fracture. A third plate also proves useful when the bone quality is poor. Fixation of the medial epicondyle can be enhanced by transposing the ulnar nerve and contouring the plate around the epicondyle to the medial surface of the trochlea. An alternative sequence is to repair all the fragments back to the humeral shaft provisionally with Kirschner wires, then to apply plates. Care must be taken to place the provisional wires in places that will not interfere with plate and screw application. In most cases, I prefer this approach. An alternative method of fixation is to use parallel plates. Use of precontoured plates with smaller (2.7-mm) distal screws greatly facilitates internal fixation, with the caveat that a very long 2.7-mm screw made of titanium can be very easy to strip, break, or get stuck in good quality bone. I follow O’Driscoll’s principles as outlined earlier. In particular, many of the provisional wires can be placed through the distal plate holes. If a 2-mm wire is used, the Kirschner wire serves as the drill, and the holes can be filled sequentially with 2.7-mm screws. The plates are placed over the

�rthopaedic ����������� S �ports ������ � Medicine ������� 1256 DeLee & Drez’s� O

periosteum and muscle origins in most cases to maintain blood supply to the fragments. I have observed avascular necrosis of the medial epicondyle after direct medial plate placement. In my experience, few fractures of the distal humerus create a simple sagittal split of the articular surface. There is usually an anterior or posterior wedge of articular surface, and occasionally there is a more substantial fracture. These are repaired with small threaded Kirschner wires placed in the subchondral bone of adjacent fragments or with countersunk, headless screws. The smaller 2.7-mm screws of newer implants are often useful for engaging these smaller fragments.

allowing immediate elbow mobilization. Operative treatment introduces the potential for several additional complications, particularly those related to handling of the ulnar nerve and the triceps mechanism. Furthermore, the combination of operative trauma and unstable internal fixation can do more harm than good. • Conversely, a healed, well-aligned fracture takes precedence, particularly with complex fractures, because capsular contracture can be addressed with subsequent surgery if needed, but early loss of fixation or articular damage may be irrecoverable.

Fixation Techniques: Complex Articular Fractures

Ulnar Neuropathy

Apparent capitellar fractures are often more complex fractures of the articular surface of the distal humerus that spare the columns.7 When the posterior trochlea or medial epicondyle is involved, I prefer to use an olecranon osteotomy. Three-dimensional CT with the radius and ulna subtracted can usually determine this preoperatively, but the surgeon should be prepared in case an osteotomy becomes necessary. If the fragments are primarily anterior, then an extended lateral exposure is used. A fracture involving the capitellum and trochlea only can usually be realigned and secured without taking down the lateral collateral ligament complex (Fig. 19F1-11A, B). The metaphyseal fracture lines are realigned, the fracture is held provisionally with Kirschner wires (see Fig. 19F1-11C), and then the fracture is secured with headless countersunk screws or screws entering from the posterior, nonarticular portion of the lateral column (see Fig. 19F1-11D, E). More complex fractures nearly always involve a fracture of the lateral epicondyle. Mobilization of the epicondyle with the attached common extensors and lateral collateral ligament origin provides excellent exposure of the anterior articular surface. If the fracture fragments do not appear to fit back on to the distal humerus, there is likely some impaction of the posterior aspect of the lateral column and possibly the posterior trochlea as well (see Fig. 19F1-3A, B). This impaction must be realigned to properly reduce the fragments. Very complex fractures can be treated with fixation of one fragment to the next with small threaded Kirschner wires or headless countersunk screws. I have found that plates with smaller distal screws are useful for this type of fracture. The lateral epicondyle is usually a small fragment; it can be secured with either a plate or small tension wires that engage the soft tissue attachments.

Optimizing Outcome • Maintenance of alignment—including the anterior tilt and translation of the distal articular surface, restoration of the medial and lateral lips of the trochlea, and avoidance of major articular incongruity—is necessary to maintain motion and limit the potential for arthrosis. • The elbow is prone to stiffness with prolonged immobilization, and the goal of treatment is secure fixation

Complications Management of the ulnar nerve is debated. Some surgeons claim to always transpose the nerve,35 some never,39 and most fall somewhere in between. In my opinion, where the nerve is placed at the end of the procedure is not as important as how it has been handled during the case. Acute postoperative ulnar neuropathy is a common but underreported problem that is likely due to a combination of devitalization and bruising from the original trauma, devitalizing dissection to move the nerve out of the way, pressure and traction from retraction, and handling during the fixation (e.g., abutment by drill guides). In any case, as yet unpublished retrospective data document a rate of about 20% ulnar nerve palsy in patients in whom the nerve is moved out of the cubital tunnel (whether it is transposed at the end of the case or not). It is highly likely that prospective evaluation will identify slight weakness and numbness more frequently.

Stiffness and Heterotopic Ossification Failure to restore the anterior translation of the trochlea with respect to the humeral shaft will lead to limitation of flexion that cannot be improved with capsular release. Articular incongruities and arthrosis can also result in unsalvageable stiffness. Heterotopic ossification that blocks motion usually occurs on the anterior aspect of the distal humerus after distal humerus fracture, causing limitation of flexion. Resection of this bone is straightforward, restores flexion, and does not require prophylaxis against recurrence because recurrence after resection of this anterior heterotopic bone is unusual.

Nonunion Nonunion is nearly always related to inadequate fixation40-42 but can also be related to infection, avascular necrosis, or excessively forceful use or manipulation of the arm before healing. The weakest point is fixation of the distal, articular fragments to the shaft, and nonunion tends to occur at the supracondylar region and only occasionally involves the articular fragments. A sufficient number of screws in the distal fragments, good contact between the shaft and the distal fragments, and strong plates will help limit the risk for nonunion. Nonunion can nearly always be salvaged with repeat open reduction, internal plate fixation, and autogenous

Elbow and Forearm 1257

A

B

C

D

E

Figure 19F1-11   Fracture of the capitellum and trochlea. A, The radiograph suggests a capitellum fracture. B, The fragment extends well into the trochlea. C, The lateral epicondyle was not fractured, and there was no metaphyseal impaction. Realignment of the metaphyseal fracture lines resulted in good articular reduction. Provisional 0.062-inch Kirschner wires were used. D, The Kirschner wires were exchanged for countersunk variable pitch (Hebert) screws. E, Healing with good alignment and good function was obtained.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1258 DeLee & Drez’s� O

cancellous bone grafting. The addition of capsulectomy of the elbow joint and neurolysis and transposition of the ulnar nerve appear to have improved the success of surgery for nonunion.43 Total elbow arthroplasty can also be used to salvage nonunion of the distal humerus but is most appropriate in older, low-demand patients.44,45 In addition, total elbow arthroplasty for post-traumatic reconstruction may have an increased rate of complications.28

Implant-Related Complications The frequent need for a second surgery to remove the wires used to repair an olecranon osteotomy has been viewed by some as representing a complication. This may be a biased view because it is also appropriate to say that the need for two surgeries (the second being relatively straightforward) to restore elbow function after such a complex injury is reasonable. It could be detrimental to the patient for the surgeon to compromise his or her exposure of and access to the fracture in the attempt to avoid a relatively simple second surgery for implant removal. Furthermore, with careful attention to the technique of creation and fixation of the osteotomy, the need for a second surgery to remove the screw or wires does not arise routinely.

RADIAL HEAD FRACTURES Indications and Contraindications Most radial head fractures are not associated with another fracture or ligament injury. They are isolated, stable, and do well with nonoperative treatment. Displaced but stable isolated fractures require surgery only in the unusual circumstance in which forearm rotation is restricted by the fracture fragment; all other indications for operative treatment of stable isolated fractures are relative, and the benefits of operative treatment are not established. There are no radiographic criteria that have been consistently associated with a poor result with nonoperative treatment.46-48 In one recent study from Malmo, Sweden, at least 74% of patients with a fracture involving greater than 30% of the head and displaced greater than 2 mm had good or excellent results decades after the fracture according to a strict rating system.46 The major problem patients encounter after an isolated fracture of the radial head is elbow ­stiffness. Displaced fractures are often associated with other fractures or ligament injuries. Radial head resection is rarely appropriate in a healthy, active athlete with an unstable displaced fracture. The decision is usually between open reduction with internal fixation and prosthetic replacement. The indications for operative treatment of a displaced fracture relate primarily to restoration of stability of the elbow and forearm. In many patients, this goal may be achieved more predictably with prosthetic replacement than with open reduction and internal fixation; however, high level arm-specific athletes may benefit from an attempt to salvage the radial head. Although partial head fractures are usually considered good candidates for open reduction and internal fixation, widely displaced fractures associated with complex injuries

can be challenging to treat because of fragmentation, the small size of the fragments, lost fragments, poor bone quality, limited subchondral bone on the fracture fragments, and metaphyseal comminution and bone loss. Open reduction with internal fixation is performed when stable, reliable fixation can be achieved. Discarding the unrepairable fragments (partial radial head resection) has been documented to have poor results in older series,49 but a recent report on the operative treatment of terrible triad injuries reported partial head resection in several patients with no apparent problems related to it.50 Because small fragments can make important contributions to elbow stability,51 for a very unstable elbow or forearm injury, it may be preferable to resect the remaining intact radial head and replace it with a metal prosthesis to enhance stability. When treating a fracture-dislocation of the forearm or elbow with an associated fracture involving the entire head of the radius, open reduction and internal fixation should be considered a viable option only if stable, reliable fixation can be achieved. There is a risk for early failure that can contribute to recurrent instability.52 Many patients with a chronic Essex-Lopresti lesion (longitudinal instability of the forearm) had failed internal fixation of a radial head fracture. Other factors such as loss of fragments, metaphyseal bone loss, impaction and deformity of fragments,53 and the size and quality of the fracture fragments may make open reduction with internal fixation a less predictable choice. In particular, if there are more than three articular fragments, the rates of early failure, nonunion, and poor forearm rotation may be unacceptable (Fig. 19F1-12).52 The silicone rubber prostheses popular during the past three decades of the 20th century did not provide much stability and often caused a destructive synovitis.54-56 Metal prostheses, used for years in some centers,57,58 are now more widely available. Some prostheses are intentionally smooth and lie somewhat loose in the radial neck, serving as spacers rather than fixed prostheses.57,58 Others are either press-fit or cemented. Some designs have a mobile head.59 The major problem with a metal radial head prosthesis is “overstuffing” the joint.60 A radial head prosthesis that is more than 1 mm proximal to the lateral edge of the coronoid process may hinge the elbow open on the lateral side and lead to capitellar wear, arthrosis, and synovitis (Fig. 19F1-13).61

Preoperative Evaluation Even relatively minor fractures of the radial head (e.g., radiographically occult fractures) can be quite painful because the elbow joint is usually distended with blood. There is a variable amount of swelling and ecchymosis, which may correspond with the degree of associated ligament injury. The distal radioulnar joint, interosseous space, and medial side of the elbow should be examined for signs of associated ligament injury. There can be crepitation of the radial head with forearm rotation, and rarely, a fracture fragment will block forearm rotation. One of the keys to successful management of a fracture of the radial head is to identify and address associated injuries. This is particularly important for displaced fractures

Elbow and Forearm 1259 Figure 19F1-12   Early failure and nonunion are common after open reduction and internal fixation of fractures that involve the entire radial head (Mason type 3), particularly those fractures that create greater than three articular fragments. A, This displaced fracture was part of a fracture-dislocation. B, Stable internal fixation was achieved. C, Six months later, the plate was broken and the radial neck remained unhealed.

A

B

and fractures that involve the entire head of the radius. In the study by Davidson and colleagues, all 11 patients with a displaced fracture involving the entire radial head had associated injury to the elbow or forearm.53 In my experience, complex fractures of the entire head do occur without associated ligament damage on occasion, particularly in older patients, but I agree with Davidson and colleagues that one should assume there is an associated injury until it has been proved otherwise. In fact, a markedly displaced partial head fracture should raise similar concerns. There are several patterns of complex injury that include a fracture of the radial head. Identification of these injury patterns can help guide treatment. These patterns include (1) fracture of the radial head associated with rupture of the medial collateral ligament, (2) concomitant fracture of the radial head and capitellum, (3) posterior dislocation of the elbow with fracture of the radial head, (4) posterior ­dislocation of the elbow with fracture of the radial head and the coronoid process (the so-called terrible triad of

C

the elbow), (5) posterior Monteggia fractures, including posterior olecranon fracture-dislocations, and (6) EssexLopresti lesions and variants. The radiographic evaluation alone may not disclose associated ligament injury. In particular, intraoperative examination after removal of the radial head is important to avoid missing injury to the interosseous ligament of the forearm. After removing the radial head fragments, the surgeon should push and pull on the radius. If the radial neck is mobile and collides with the capitellum, the surgeon should assume that the interosseous ligament of the forearm is injured.62 In combination with the evaluation of the signs and symptoms described previously, radiographs of the elbow and wrist disclose most associated injuries. For isolated partial fractures of the radial head, the ability of the patient to fully pronate and supinate the forearm will influence treatment. Pain can make this difficult to evaluate during the first few days after injury. If the radiographs reveal a

�rthopaedic ����������� S �ports ������ � Medicine ������� 1260 DeLee & Drez’s� O

Figure 19F1-14   These fracture fragments demonstrate impaction and deformation of the head in the largest fragment (left), metaphyseal fragmentation, and small unrepairable articular fragments (right).

Figure 19F1-13   The major pitfall with a metallic radial head prosthesis is implantation of a prosthesis that is too long. This radiograph shows gapping on the lateral side of the ulnohumeral joint and a large cyst in the capitellum.

fracture that may restrict forearm rotation and operative treatment is being considered, it may be useful to aspirate some of the blood from the elbow joint and instill some local anesthetic (usually lidocaine). This can be done at the anatomic soft spot (roughly at the center of a triangle formed by the dorsal point of the olecranon, the radial head, and the lateral epicondyle) on the lateral side of the elbow. Alternatively, if the patient returns to the office a few days to a week after injury, he or she is likely to feel much better and be capable of demonstrating forearm rotation. A true block to forearm rotation is uncommon, so injection or delayed serial examination, or both, are important in decision making. Operative treatment of widely displaced fractures of the radial head typically reveals a more complex fracture than was apparent on radiographs. Although two-dimensional and three-dimensional CT will depict these aspects in greater detail and thereby facilitate planning of the operation, it is not necessary to obtain these studies provided that the surgeon is prepared for all possible treatment options, including repair with plates or screws or excision of the fractured radial head with insertion of a metal prosthesis if there is an associated forearm or elbow injury. Mason classified fractures of the radial head at a time when fractures were either excised or treated nonoperatively.63 He distinguished nondisplaced fractures that did well with nonoperative treatment (type 1), comminuted fractures of the entire head of the radius (type 3) that were best treated by excision, and displaced fractures involving part of the radial head (type 2), which presented a treatment dilemma in that most of the head was intact, but some fractures had poor results. His classification did not include radial neck fractures, did not account for associated injuries, and did not quantify displacement.

Morrey64 modified Mason’s classification to (1) include fractures of the radial neck, (2) provide a quantitative definition of displacement (a fragment involving 30% or more of the articular surface that is displaced more than 2 mm), and (3) incorporate fracture-dislocations of the elbow as suggested by Johnston65 as a Mason type 4 fracture. There are few data to support the quantitative definition of displacement that is offered in this system. Other factors that may have an important influence on treatment, but are not well accounted for in current classification systems, include: (1) lost fragments—a very common occurrence with displaced fractures, (2) fragments that are too small to be repaired and must be discarded, (3) fragments with little or no subchondral bone, (4) fragments with osteoporotic bone, (5) impaction and deformation of the fracture fragments, and (6) metaphyseal bone loss (Fig. 19F1-14). Partial resection of the radial head has long been associated with inferior results49 and was one factor associated with problems in our study of operative treatment.7 Therefore, when fragments are lost, are too small to fix, or have inadequate or poor quality bone and must be discarded, the surgeon probably ought to err toward resection of the radial head with or without prosthetic replacement depending on the presence or absence of associated injuries. Impacted fractures may be less suitable for operative fixation because enlargement and deformation of the radial head have been observed in long-term follow-up and appear to hinder forearm rotation. Metaphyseal bone loss and impaction are observed even with partial radial head fractures, and a plate may be superior to screws alone in this circumstance.

Operative Approach in the Athlete Operative Exposures The most popular interval for the exposure of fractures of the radial head is between the anconeus and extensor carpi ulnaris (Kocher exposure).66,67 This interval is the most posterior interval and provides good access to fragments of the radial head that displace posteriorly. It also provides

Elbow and Forearm 1261

greater protection for the posterior interosseous nerve. Conversely, attention must be paid to protecting the lateral collateral ligament complex. The anconeus should not be elevated posteriorly, and the elbow capsule and annular ligament should be incised diagonally, in line with the posterior margin of the extensor carpi ulnaris.68 A more anterior interval protects the lateral collateral ligament complex, but places the posterior interosseous nerve at greater risk.69 Some authors recommend identifying the nerve if dissection onto the radial neck is required.69 Kaplan described an interval between the extensor carpi radialis brevis and the extensor digitorum communis,66 whereas Hotchkiss recommends going directly through the extensor digitorum communis muscle.69 I find these intervals difficult to define precisely based on intraoperative observations. A useful technique for choosing a good interval and protecting the lateral collateral ligament complex was described by Hotchkiss69: starting at the supracondylar ridge of the distal humerus, if one incises the origin of the extensor carpi radialis, elevates it and incises the underlying elbow capsule, it is then possible to see the capitellum and radial head. The interval for more distal dissection should be just anterior to a line bisecting the radial head in the anteroposterior plane. In my practice, most fractures of the radial head that merit operative treatment are associated with fracture­dislocations of the elbow. In this context, exposure is greatly facilitated by the associated capsuloligamentous and muscle injury.52,70,71 When the elbow has dislocated, the lateral collateral ligament has ruptured, and the injury always occurs (or nearly always according to some authors71) as an avulsion from the lateral epicondyle. Along with a variable amount of muscle avulsion from the lateral epicondyle,71-75 these injuries leave a relatively bare epicondyle. There is often a split in the common extensor muscle that can be developed more distally. In the setting of a posterior olecranon fracture­dislocation or posterior Monteggia fracture, the radial head often displaces posteriorly through capsule and muscle. Accentuation of the injury deformity usually brings the radial head up into the wound (Fig. 19F1-15). In some

cases, the surgeon will extend the posterior muscle injury in order to mobilize the olecranon fracture proximally and to expose and manipulate the coronoid fracture through the elbow articulation. Slight additional dissection between the radius and the ulna is acceptable given the usually extensive injury in this region, but extensive new dissection in this area has been suggested to increase the risk for proximal radioulnar synostosis.76,77 When treating a complex fracture of the radial head with the lateral collateral ligament complex intact (e.g., an Essex-Lopresti injury), it may be difficult to gain adequate exposure without releasing the lateral collateral ligament complex from the lateral epicondyle. This can be done either by directly incising the origin of the lateral collateral ligament complex from bone or by performing an osteotomy of the lateral epidcondyle.22,78-81 In either case, a secure repair and avoidance of varus stress in the early postoperative period are important.

Open Reduction and Internal Fixation Increased enthusiasm for open reduction and internal fixation of the radial head paralleled the development of small screws (2.7-, 2.0-, and 1.5-mm) and the techniques for using them.80 At the same time small, headless, ­ variablepitch compression screws (such as the Herbert screw) were developed, allowing for fixation of entirely articular fragments.82-84 Standard screws can be used in this way as well, countersinking the head below the articular surface, although they are prone to backing out into the joint with the slightest amount of settling at the fracture site. Some small fragments can be repaired only with small Kirschner wires. Threaded wires are usually used because of the tendency for smooth wires to migrate and potentially travel to various parts of the body.85 Absorbable pins and screws are being developed for similar uses86,87 but are still somewhat brittle and sometimes associated with an inflammatory response. Small plates are available for fractures that involve the entire head. Plate types include T- and L-shaped plates with standard screws, small (condylar) blade plates, and new plates designed specifically for the radial head (many of which incorporate angular stable screws—screws that thread directly into the plate). The use of plates that are placed within the radial head or countersunk into the articular surface has also been described.79

Prosthetic Replacement

Figure 19F1-15   When treating a posterior olecranon fracture-dislocation or posterior Monteggia injury, the radial head can often be addressed through the posterior traumatic interval.

When preparing for prosthetic replacement, if some of the radial head is still attached to the neck, it is separated at the point on the radial neck where the flare of the head begins. I prefer to use a prosthesis with a smooth neck that serves as a loose spacer. The laxity in the neck facilitates insertion and removal of the prosthesis and accommodates for some of the nonanatomic features of the prosthesis compared with the native radial head. I use a neck diameter one size smaller than the reamer that can be passed with slight effort. I use a head size of slightly smaller diameter than the native head. I almost never add more length through the head. It is important to realize that the prosthesis will sit on the most prominent ­portion

�rthopaedic ����������� S �ports ������ � Medicine ������� 1262 DeLee & Drez’s� O

of the radial neck; therefore, one should be careful to choose a head thickness based on the thinnest portion of the radial head. I smooth off, but do not evenly plane, the remaining radial neck. It is important to check the level of the radial head with respect to the lateral edge of the coronoid process; it should be no more than 1 mm more proximal. In the unusual patient in whom the comminution extends into the neck, a prosthetic head of greater thickness can be used. In cases of extreme neck comminution, the prosthesis can be cemented in the neck.

Optimizing Outcomes • Only repair simple fractures of the radial head. Repair of complex fracture may be attempted in high-level athletes (particularly those who use their arms in their sport), but the patient and surgeon should be prepared for early fixation failure, later nonunion, and poor forearm motion, which occurs in many cases. • Use prosthetic replacement of complex partial and whole head fractures to restore stability of the elbow and forearm. Avoid placing a prosthesis that is too large. • Protect the lateral collateral ligament complex and the posterior interosseous nerve.

Complications Laceration or permanent injury to the posterior interosseous nerve during open reduction and internal fixation of a radial head fracture is unusual. Most commonly this complication is experienced as a palsy related to retraction or exposure that resolves over weeks to months. To limit the potential for this complication, retractors should not be placed around the radial neck, the forearm should be pronated during exposure of the radial neck, and consideration should be given to identifying and protecting the nerve when more distal dissection and internal fixation are needed, particularly when a more anterior muscle interval is used for exposure. Early failure of fixation and later nonunion are not infrequent, particularly after open reduction with internal fixation of complex fractures involving the entire head. In a recent series, 3 of 14 fractures involving the entire radial head and creating greater than three articular fragments had failure of fixation within the first month.52 Because this situation can contribute to instability of the forearm or elbow, unstable or unpredictable fixation is undesirable, and such fractures should probably be treated with prosthetic replacement. Among fractures of the entire radial head, 6 of 11 in one series81 and 8 of 26 fractures in another series52 (including 2 of 12 fractures with three or fewer fragments and 6 of 14 fractures with greater than three articular fragments) had nonunion. Delayed resection of the radial head has usually been performed to improved forearm rotation, and not for painful arthrosis of the radiocapitellar joint.88,89 Incongruity of the proximal radioulnar joint presents as stiffness rather than pain or arthrosis, and incongruity of the radiocapitellar joint inconsistently and unpredictably leads to radiocapitellar arthrosis, which appears to be an uncommon problem.

A radial head prosthesis that is too large can cause malalignment of the elbow, capitellar wear, and synovitis and usually needs to be removed. Removal of a metal radial head prosthesis can be difficult. Given the alternative of releasing the origin of the lateral collateral ligament complex to subluxate the elbow, in most patients, I prefer to excise a portion of the radial neck so that I can pry the prosthesis out. The difficulty encountered in removing these prostheses is one reason I prefer to use a slightly loose prosthetic stem.

TRAUMATIC ELBOW INSTABILITY Traumatic elbow instability occurs in three basic forms: posterolateral rotatory instability (elbow dislocations with or without associated fractures), varus posteromedial rotational instability (anteromedial coronoid facet fractures), and olecranon fracture-dislocations. Identification of the specific pattern of traumatic elbow instability will indicate which structures are likely to be injured, the morphology of the injuries, and the prognosis, all of which will help to guide management. Posterolateral rotatory instability results in dislocation of the elbow with or without fractures of the radial head and coronoid. Posterolateral rotatory injuries occur during a fall onto the outstretched arm that creates a valgus, axial, and posterolateral rotatory force. The ulna and the forearm supinate away from the humerus and dislocate posteriorly. Sometimes this results in injury to the radial head or coronoid. The soft tissue injury proceeds from lateral to medial, with the anterior band of the medial collateral ligament being the last structure injured.90 It is possible to dislocate the elbow with the anterior band of the medial collateral ligament remaining intact. Varus, posteromedial rotational instability occurs with a fall onto the outstretched arm that creates a varus stress, axial load, and posteromedial rotational force to the elbow.1 This results in fracture of the anteromedial facet of the coronoid process and either (1) injury to the lateral collateral ligament, (2) fracture of the olecranon, or (3) an additional fracture of the coronoid at its base. Anterior olecranon fracture-dislocations are the result of a direct blow to the flexed elbow, but the mechanism of posterior olecranon fracture-dislocations is more speculative, with some authors suggesting they may result from the same mechanism that usually creates elbow dislocations, particularly in older osteopenic individuals.91,92 Olecranon fracture-dislocations are addressed along with olecranon fractures in the final section of this chapter.

Indications and Contraindications Elbow dislocations without associated fractures are rarely unstable after closed reduction in the athlete, and immediate active mobilization will optimize motion, function, and stability. Mobilization of the elbow within 2 weeks results in less stiffness and pain.93,94 Simple elbow dislocations that redislocate after manipulative reduction appear to be related to a greater degree of soft tissue avulsion from the distal humerus and are typically observed in frail older patients and younger patients with high-energy ­dislocations.95

Elbow and Forearm 1263

Any elbow dislocation or subluxation associated with fracture should be considered for operative treatment in the athlete. Loss of support between the radial head and capitellum by fracture of the radial head creates an additional measure of instability, but this injury pattern is not as problematic as when the coronoid process is fractured as well. Broberg and Morrey96 and Josefsson and colleagues97 documented good results treating these injuries with or without radial head resection and 1 month of cast immobilization with two caveats: (1) patients with associated coronoid fractures redislocated in the cast and (2) patients treated nonoperatively often needed secondary procedures to address restriction of forearm rotation owing to the fracture of the radial head. Although the radial head may not be necessary to keep the elbow concentrically reduced in this injury pattern, some authors have suggested that preservation of the stabilizing influence of radiocapitellar contact may delay the onset of ulnohumeral arthrosis.98 Therefore, I would not recommend radial head excision in the context of a fracture-dislocation in an athlete. The addition of a coronoid fracture, no matter how small, to a dislocation of the elbow and fracture of the radial head dramatically increases the instability and the potential for problems.50,97,99 As a result, the term terrible triad of the elbow has been coined by Hotchkiss100 to refer to this injury pattern. Not all terrible triad injuries are unstable, but it can be difficult to predict which injuries will be unstable. Recurrent subluxation or dislocation can destroy the elbow articulation, particularly if it is not recognized and if abnormal contact of the articular surface persists. Nonoperative treatment is risky because the elbow can dislocate in a cast, unknown to the patient. Radial head resection without prosthetic replacement is unwise because recurrent dislocation of the elbow is common.99 Good results have been documented with repair of the coronoid or anterior capsule, repair or replacement of the radial head, and lateral collateral ligament repair.50 This restores stability in most cases, but in some patients, medial collateral ligament repair or hinged external fixation is also necessary. The use of cross-pinning of the ulnohumeral joint has been presented at a national meeting (K. Cramer and colleagues, annual meeting of the Orthopaedic Trauma Association, 1999) with reasonably good results, but has not yet been published in a peer-reviewed journal. With specific treatment of each of the injury components, this step should rarely be necessary.50 Varus posteromedial rotational instability pattern injuries are unstable injuries that can lead to chronic subluxation and arthrosis of the elbow. It has only recently been recognized and is relatively uncommon, so our knowledge of the results of nonoperative treatment is limited. Early experience would suggest that these injuries benefit from operative treatment.1

Preoperative Evaluation Although elbow dislocation is usually evident, if there is any doubt about alignment, the point of the olecranon process and the medial and lateral epicondyles should form a triangle in the coronal plane with the elbow flexed 90 degrees. If the point of the olecranon is well posterior to the epicondyles, the elbow is likely dislocated.

Acute neurovascular injuries are uncommon, but the ulnar and median nerves are most commonly involved. The brachial artery may also be injured, particularly with an open dislocation, which is also unusual. Most elbow dislocations and fracture-dislocations result in injury to all of the capsuloligamentous stabilizers of the elbow joint.73,75,90,97 The exceptions include fracture­dislocations of the olecranon and other injuries with fractures of the coronoid involving nearly the entire coronoid process.1,70,101,102 The capsuloligamentous injury progresses from lateral to medial, and the elbow can completely dislocate, with the anterior band of the medial collateral ligament remaining intact.90 There is also a variable degree of injury to the common flexor and extensor musculature.73-75,95 One recent study notes that the lateral collateral ligament complex fails by avulsion from the lateral epicondyle in more than 75% of patients with elbow dislocations.71 In my personal observations treating more than 60 fracture-dislocations of the elbow, I have found that the lateral collateral ligament is always avulsed from the lateral epicondyle. In many patients, there are small pieces of the ligament or other long strands of musculotendinous tissue, which may lead the surgeon to misinterpret the situation (see Fig. 19F1-13). Defined practically, reattachment of the soft tissue sleeve to the lateral epicondyle is nearly always sufficient. O’Driscoll and associates described several stages of elbow instability.90 Stage 1 involves partial or complete disruption of the lateral collateral ligament, which may result in slight posterior subluxation of the radial head with respect to the capitellum. Stage 2 involves an incomplete posterior dislocation with disruption of the lateral ligamentous complex and further injury to the osseous or ligamentous supporting structures anteriorly or posteriorly. The medial edge of the ulna may be found to rest on the trochlea. This gives the appearance of the coronoid being perched on the trochlea on a lateral radiograph.90 Stage 3 is divided into three subgroups (A to C). Stage 3A involves injury to all the soft tissue support except the anterior band of the medial collateral ligament. The elbow dislocates in a posterolateral direction rotating around the intact anterior medial collateral ligament. Stage 3B involves injury to the entire medial ligamentous complex, resulting in varus, valgus, and rotatory instability. Stage 3C injuries are unstable because of complete soft tissue disruption from the distal humerus, with the elbow having the ability to dislocate even when immobilized in a cast.1 Elbow dislocations that are associated with one or more intra-articular fractures are at greater risk for recurrent or chronic instability.70,97,103 Fracture-dislocations of the elbow usually occur in one of several distinct, recognizable injury patterns: (1) posterior dislocation with fracture of the radial head; (2) posterior dislocation with fractures of the radial head and coronoid process—the so-called terrible triad injury (Fig. 19F1-16); (3) varus posteromedial rotational instability pattern injuries (Fig. 19F1-17); (4) anterior olecranon fracture-dislocations (Fig. 19F1-18); and (5) posterior olecranon fracture-dislocations (Fig. 19F1-19). Each of these patterns is associated with characteristic injury components and fracture morphologies, the knowledge of which can help guide effective management.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1264 DeLee & Drez’s� O

A

C

B

Figure 19F1-16   The so-called terrible triad of the elbow consists of dislocation of the elbow with fractures of the coronoid and radial head. A, The coronoid fragment is the triangular fragment anterior to the trochlea. After manipulative reduction, this elbow could not be kept reduced despite cast immobilization. B, The coronoid fracture is a transverse fracture of the tip as seen on this three-dimensional computed tomography (CT) reconstruction. C, Operative fixation of the coronoid, replacement of the radial head, and reattachment of the lateral collateral ligament complex to the lateral epicondyle restored good elbow function.

Varus posteromedial rotational instability pattern injuries and olecranon fracture-dislocations are not true dislocations in that apposition of the articular surfaces is not lost. Rather, they are usually fracture-subluxation injuries in which the major problem is disruption of the trochlear notch. Recent reports on elbow instability have emphasized the importance of the coronoid process.1,70,99 The injuries that give surgeons the most trouble are the terrible triad, varus posteromedial, and olecranon fracture-dislocations with associated coronoid fractures.1 In each case, the fracture of the coronoid is the most important and challenging part of the injury. Regan and Morrey104 classified coronoid fractures based on the size of the fragment: type I, avulsion of the tip of the coronoid process; type II, a single or comminuted fragment involving 50% of the process or less; and type III, a

single or comminuted fragment involving more than 50% of the process.104 They also included a modifier to indicate the presence (type B) or absence (type A) of an elbow dislocation. However, it has become clear that the pattern of the overall injury and morphology of the fracture may be equally or more important than the size of the fragment and the presence or absence of dislocation. O’Driscoll proposed a new classification system for coronoid fractures based on the anatomic location of the fracture. Fractures may involve the tip, the anteromedial facet, or the basal aspect of the coronoid. The three groups are further divided into subtypes based on the severity of coronoid involvement. His system considers the mechanism of injury along with the associated fractures and soft tissue injuries and helps to dictate treatment.1 The first group of coronoid fractures involves the tip but does not extend medially past the sublime tubercle or

Elbow and Forearm 1265

A

B

C

E

D

F

G

Figure 19F1-17   Small coronoid fractures are often problematic. A, This appears to be a small isolated fracture of the coronoid. B, On the anteroposterior view, it is clear that the anteromedial facet of the coronoid process is fractured. There is varus subluxation and opening of the joint on the lateral side betraying the associated lateral collateral ligament injury. C, Three-dimensional CT depicts external rotation of the distal humerus with respect to the forearm as the trochlea rotates forward into the coronoid defect. D, There are separate coronoid tip and anteromedial facet fracture fragments. E, Exposure is obtained by transposing the ulnar nerve anteriorly and elevating the anterior portion of the flexor-pronator muscles off of the medial collateral ligament and the coronoid process. F, The coronoid fractures were secured with a buttress plate, and the lateral collateral ligament origin was reattached to the lateral epicondyle with a suture anchor. G, A concentric reduction and good elbow function resulted.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1266 DeLee & Drez’s� O

into the body. Tip, subtype 1 fractures involve less than 2 mm of the coronoid and may be found in isolation or with a fracture dislocation. Tip, subtype 2 fractures involve greater than 2 mm and are largely associated with terrible triad injuries. In my experience, all these fracture fragments contain the insertion of the anterior capsule and the 2-mm distinction between subtypes 1 and 2 is arbitrary and does not influence treatment. The second group of coronoid fractures involves the anteromedial aspect of the coronoid. Anteromedial subtype 1 fractures extend from just medial to the tip of the coronoid to the anterior half of the sublime tubercle (insertion of the anterior band of the medial collateral ligament). Anteromedial subtype 2 fractures are subtype 1 injuries with extension of the fracture line into the tip. Anteromedial subtype 3 fractures involve the anteromedial rim and the entire sublime tubercle with or without involvement of the tip of the coronoid. The mechanism of injury is usually a varus, posteromedial rotation injury with axial ­ loading.

A

The lateral collateral ligament complex is ­ generally ­disrupted unless the olecranon is also fractured. Radial head fractures may be seen in higher energy, subtype 3 injuries. Anteromedial coronoid fractures cause incongruent articulation of the ulnohumeral joint, which may lead to an earlier onset of post-traumatic arthritis. Basal coronoid fractures make up the third category and involve at least 50% of the height of the coronoid. Basal subtype 1 fractures involve the coronoid alone, whereas subtype 2 fractures are associated with fractures of the olecranon. In general, these fractures have less soft tissue disruption than those that involve only the tip of the ­coronoid. The following observations may be useful in guiding treatment: (1) Terrible triad injuries nearly always have a small transverse fracture of the tip of the coronoid, including the anterior capsular attachment (see Fig. 19F1-16B). Much less commonly, the coronoid fracture is either very large or involves the anteromedial facet of the coronoid

B

D

C Figure 19F1-18   Anterior (or trans-olecranon) olecranon fracture-dislocations are relatively uncommon injuries that are characterized by anterior translation of the forearm, an intact radial head, and fracture of the proximal ulna. A, This very complex proximal ulna fracture involves the coronoid process. B, The coronoid is split in the sagittal plane and can be repaired with interfragmentary compression screws. C, The metaphyseal and diaphyseal fragmentation is bridged with a long plate, contoured to wrap around the dorsal surface of the ulna. Tension wires are used to enhance fixation of the small proximal fragments. D, Six months later, the fracture has healed, and good elbow function has been restored.

Elbow and Forearm 1267

preferentially. (2) Varus posteromedial rotational ­instability pattern injuries are defined by a fracture of the ­anteromedial facet of the coronoid process (see Fig. 19F1-17). (3) In the setting of an olecranon fracture-­dislocation, the coronoid fracture can be one simple large fragment; it can be fragmented into two or three large pieces (anteromedial facet, central, and lesser sigmoid notch) with or without a tip fragment as well; or it can be more ­comminuted.

Operative Techniques in the Athlete The treatment of fracture-dislocations of the elbow is intended to restore the inherent bony stability of the elbow that allows us to treat most simple elbow dislocations with immediate active motion with a high degree of success. Critical to achieving this is restoration of the trochlear notch of the ulna, particularly the coronoid ­ process. Anatomic

A

C

B

D

E

Figure 19F1-19   Posterior olecranon fracture-dislocations can be very complex injuries. A, In this patient, the coronoid and radial head are fractured. B, The coronoid is split into three fragments. C, A posterior skin incision discloses muscle injury.   D, If the muscle injury is opened up and extended somewhat, the olecranon fragment can be translated proximally like an olecranon osteotomy, exposing the elbow articulation. E, An additional medial exposure with transposition of the ulnar nerve helps with manipulation of the anteromedial fracture fragment.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1268 DeLee & Drez’s� O

F

G

H

I Figure 19F1-19, cont’d   F, The coronoid is split into three large fragments: anteromedial, central, and lesser sigmoid notch.   G, A long dorsal plate is applied, bridging the comminution and securing the coronoid. H, The radial head is replaced. I, The anteromedial portion of the coronoid is secured with screws. Healing and good elbow functions were restored.

alignment of anteromedial and basal coronoid fractures is necessary for elbow stability and function. Radiocapitellar contact is also important to the stability of the injured elbow. The lateral collateral ligament is far more important than the medial collateral ligament in the setting of most cases of traumatic elbow instability. The trochlear notch (coronoid and olecranon), radial head, and lateral collateral ligament are repaired or reconstructed, but the medial collateral ligament rarely needs to be repaired. Some surgeons are still becoming comfortable with the idea that medial collateral ligament repair is not necessary for most fracture-dislocations.105 If the elbow is stable, or can be made stable with surgery on the lateral side, the medial collateral ligament will heal properly with active motion, and its repair is not necessary for stability.50

Intraoperative Testing of Elbow Stability Intraoperative testing of elbow stability is important. Substantial subluxation or redislocation of the elbow is challenging to treat, and the elbow can be unstable despite cast immobilization. Therefore, it is important not to leave the operating room until adequate stability has been achieved. Morrey recommended that the elbow should not redislocate before reaching 45 degrees of flexion from a fully flexed position,106 and Jupiter and I recommended that the elbow should be able to go to 30 degrees before substantial subluxation or dislocation.70 These are relatively arbitrary numbers. My current practice is to test stability in gravity extension with the forearm in neutral rotation (Fig. 19F1-20).

Elbow and Forearm 1269

l­igament facilitates exposure of the radial head. If the radial head cannot be repaired (lost fragments, more than three fragments, poor bone quality), resection and replacement with a metal prosthesis enhance immediate and long-term stability but also expose the patient to potential complications. When the lateral collateral ligament is repaired, immediate active motion is usually possible (particularly if radiocapitellar contact has also been restored), but up to 10 days of immobilization is reasonable.50

Terrible Triad Fracture-Dislocations

Figure 19F1-20   A useful method for intraoperative testing of elbow stability is to place the arm in gravity extension with the forearm in neutral. After this maneuver is performed, the elbow is gently flexed and palpated for recurrence of any subluxation or dislocation. It is also useful to monitor this maneuver on an image intensifier.

In other words, I support the upper arm or humerus, straighten the elbow, and let gravity apply a stress to the extended elbow. I then gently flex the elbow and feel for any clunk or crepitation. I repeat the maneuver under image intensification. In most cases, the elbow is stable with this maneuver, and no more than slight subluxation is seen radiographically. If greater subluxation occurs, I consider additional treatment such as the use of hinged external fixation.

Unstable Simple Elbow Dislocations Patients with persistent subluxation have been treated with a cast brace with the forearm held in pronation.107 An alternative is to forgo the brace or cast and encourage confident active motion of the elbow.108 Patients with persistent subluxation of the elbow resemble patients with so-called pseudosubluxation of the shoulder related to pain-related inhibition of the shoulder muscles. Active elbow mobilization adds an additional dynamic muscular contribution to elbow stability. This should be attempted only in patients with slight subluxation or opening of the joint and not in patients with so-called perched dislocations in which the trochlea is resting on and scraping against the coronoid process. When the elbow cannot be held in a concentrically reduced position, redislocates before getting a postreduction radiograph, or dislocates later despite splint immobilization, the dislocation is deemed unstable, and operative treatment is required. There are three general approaches to this problem: (1) open relocation and repair of soft tissues back to the distal humerus, (2) hinged external fixation, and (3) cross-pinning of the joint.

Posterior Dislocation and Fracture of the Radial Head When feasible, repair of the radial head restores the native anatomy and contributes to the immediate and longterm stability of the elbow. The injured lateral collateral

It might be possible to restore stability with radial head repair or replacement and lateral collateral ligament repair only in many patients; however, if doing so results in inadequate stability, these repairs must be taken down to restore access to the coronoid. Respecting that these injuries are infamously problematic, I prefer to repair the coronoid fracture, the radial head, and the lateral collateral ligament, working from inside out. This nearly always restores good stability in gravity extension of the elbow without the need for medial collateral ligament repair, hinged external fixation, or cross-pinning of the joint.

Surgical Procedure: Internal Fixation of a Tip Fracture of the Coronoid Exposure and fixation of the small transverse fractures of the coronoid that are usually associated with a terrible triad injury pattern can be accomplished through a lateral exposure. The unusual anteromedial coronoid fracture will require a separate medial exposure. I prefer to use a single midline posterior longitudinal incision, elevating a lateral skin flap. This provides access to the dorsal ulna for passing the coronoid sutures and to the medial side of the elbow in case the ulnar nerve or medial collateral ligament needs to be addressed or a hinged external fixator will be applied. If there is a traumatic rent in the fascia, it can be developed proximally and distally (Fig. 19F1-21A), but usually the overlying fascia is intact. In such cases, I identify the supracondylar ridge of the humerus and incise and elevate the radial wrist extensors off of the ridge anteriorly. As one works distally, it will become apparent that the lateral collateral ligament and some or all of the common extensor musculature has been avulsed from the lateral epicondyle. An interval in the common extensors is selected based on what is injured, and it is developed distally. This interval should be anterior to the midline of the radial head and neck. The supinator is spread bluntly over the radial neck, but it is rarely necessary to dissect far distally where the posterior interosseous nerve would be placed at risk. Removal of the radial head fragments from the wound, elevation of the radial wrist extensors from the supracondylar ridge, slight splitting of the supinator distally, and subluxation of the elbow all help to expose the coronoid fracture in the depths of the wound. The coronoid is repaired with sutures passed through drill holes in the ulna. Creation of the drill holes can be facilitated with a drill guide such as that used to assist with repair of the anterior cruciate ligament of the knee, or they

�rthopaedic ����������� S �ports ������ � Medicine ������� 1270 DeLee & Drez’s� O

A

B

C

D

Figure 19F1-21   Operative treatment of a terrible triad elbow fracture-dislocation. A, A traumatic rent in the common extensors was developed distally. The strand of tissue still attached to the lateral epicondyle is musculotendinous, not ligament. B, Sutures passed through drill holes in the ulna are passed through drill holes in the large coronoid fragment. C, The coronoid is reduced, restoring an anterior buttress and the function of the anterior capsule. D, The lateral collateral ligament complex is reattached to the lateral epicondyle using a suture anchor. The radial head has been replaced with a prosthesis.

can be made freehand. Two drill holes, one medial and one lateral, are created in the bed of the coronoid fracture. A simple method for passing the sutures is to put a suture loop on a Keith needle, pass the needle and loop into the drill holes, and pull the suture loop from the coronoid base out into the wound. A suture can then be easily passed through the loop and pulled through the ulna. Suture passers can also be used, but they require passing the suture deep in the confined space of the wound. The suture is passed through drill holes in the coronoid fragment for larger fragments (see Fig. 19F1-21B) and around the fragment and through the anterior capsular attachment in small or comminuted coronoid fragments. It is then passed through the other hole in the ulna. Precise anatomic alignment is not crucial—the goal is to restore the anterior capsular insertion and an anterior bony buttress (see Fig. 19F1-21C). This suture is not tied until the radial head has been treated because subluxation of the joint facilitates treatment of the radial head. The radial head is either repaired or replaced as described earlier in the section on radial head fractures. The lateral collateral ligament is repaired as described earlier (see Fig. 19F1-21D). The coronoid suture is tensioned and tied, followed by tying of the lateral collateral ligament repair

sutures (see Fig. 19F1-21D). Stability in gravity extension is tested. If there is substantial subluxation or dislocation of the elbow in gravity extension, consideration is given to repairing the medial collateral ligament. Alternatively, a hinged external fixator is applied. If the elbow remains unstable after the surgeon has done everything within their skill and resources to stabilize the elbow, it is reasonable to cross pin the elbow.

Varus Posteromedial Rotational Instability Injuries This injury is treated with operative repair of all injured structures. The lateral collateral ligament or olecranon fracture is repaired, and the coronoid is repaired through a medial exposure using some combination of wires, plates, and screws. A buttress plate is particularly useful.

Surgical Procedure: Anteromedial Coronoid Facet Fracture A medial skin flap is elevated, with care taken to protect the medial antebrachial cutaneous nerve and the ulnar nerve. We prefer to mobilize the ulnar nerve from the

Elbow and Forearm 1271

cubital tunnel, allowing it to remain anteriorly transposed in the subcutaneous tissues at the end of the procedure (see Fig. 19F1-17E). From the medial side, the coronoid can be exposed directly medially—between the heads of the flexor carpi ulnaris where the ulnar nerve usually lies, or superiorly through the over-the-top approach described by Hotchkiss109 for contracture release. Taylor and Scham110 described elevation of the entire flexor-pronator mass, but this requires far more dissection than elevation from within the split in the flexor carpi ulnaris. It is useful for very large fragments. The dissection on the medial side should remain superficial to the anterior band of the medial collateral ligament. In addition, capsular attachments to the tip of the coronoid should be preserved. Through this exposure, a small T-plate or a plate designed specifically for internal fixation of the coronoid can be applied.

During the fixation of anteromedial facet coronoid fractures, extensive dissection, transposition, and retraction of the ulnar nerve are needed to perform a medial repair, and ulnar nerve palsies sometimes occur. Large tapes and tapes with hemostats should not be used because they may lead to excessive traction. A means for providing adequate retraction of the nerve without prolonged pressure needs to be devised. I have sutured the skin to the forearm ­fascia to hold the nerve anteriorly as I work on the coronoid. Subluxation of the elbow can occur with rotation of the humerus into an anteromedial coronoid defect, leading to arthrosis.1 Heterotopic bone formation is uncommon after this injury.

Optimizing Outcomes

The key element in the treatment of a fracture of the proximal ulna, no matter how complex, is to restore the contour and dimensions of the trochlear notch of the ulna. Small articular incongruities and comminution in the relatively nonarticular transverse groove will be of little consequence provided that stable realignment of the coronoid and olecranon facets is achieved.1,101,102 These fractures often occur in patients with poor bone quality.102 Careful technique and an understanding of how to achieve reliable internal fixation are important. Restoration of the trochlear notch is the key to restoring elbow stability. Also important are repair or replacement of the radial head and repair of the lateral collateral ligament complex when they are injured. Stable fixation will allow immediate mobilization, thereby diminishing the risk for stiffness and heterotopic ossification.­

• Understand the overall injury pattern. The pattern of injury tells you what is injured and what needs to be repaired. • Identify any coronoid fracture and its pattern. Tip fractures are associated with terrible triad injuries and can be repaired from lateral. Anteromedial facet fractures associate with varus posteromedial instability injuries and are usually treated with medial buttress plating. Basal fractures are usually associated with olecranon fracturedislocations and are treated through the fracture. • Protect unstable elbows treated late (after 2 weeks) or with comminuted coronoid fractures with hinged external fixation.

Complications Varus, valgus, and posterolateral rotatory instability are uncommon after elbow dislocation. Persistent subluxation and dislocation can destroy the ulnohumeral articulation if not promptly identified and treated as the joint surfaces scrape against one another. Slight subluxation with relative congruence of the articular surfaces can be treated with active exercises as described earlier. Wide dislocation is actually preferred over subluxation with abnormal contact of the articular surfaces because the articular surfaces are less likely to be damaged. We have had substantial success treating a chronic simple elbow dislocation with relocation and 6 weeks of hinged external fixation without ligament repair or reconstruction, likely because the articular surfaces have been relatively well preserved.111 Acute redislocation, recurrent dislocation, and chronic instability are uncommon with this injury pattern. Untreated or inadequately treated fractures of the radial head can contribute to loss of forearm motion and arthrosis and are treated with a second operation for excision of the radial head. An oversized radial head prosthesis can cause painful radiocapitellar wear and arthrosis. Instability, arthrosis, heterotopic ossification, and ulnar neuropathy are all relatively more frequent after terrible triad injuries than simple dislocations and dislocation with fracture of the radial head.50,99

OLECRANON AND PROXIMAL ULNA FRACTURES

Indications and Contraindications The rare stable, minimally displaced fracture of the olecranon can be treated with 3 to 4 weeks of immobilization followed by active exercises to regain motion. Most olecranon fractures are displaced and require surgery. All fracturedislocations require operative treatment. Excision of the olecranon and triceps advancement is rarely used, mostly of historical interest, and should not be used in the athlete.

Preoperative Evaluation The initial radiographs obtained after the injury are often of limited quality owing to the deformity and pain in the limb. Nevertheless, it is usually possible to discern the overall pattern of the injury, which, in turn, leads one to suspect other injury components that may not be immediately obvious. For example, a posterior olecranon fracturedislocation is often associated with fractures of the radial head and coronoid process as well as injury to the lateral collateral ligament complex,102 whereas an anterior fracture-dislocation rarely involves injury to the radial head or collateral ligaments.101 Radiographs obtained after manipulative reduction and splint immobilization of the limb (when appropriate) may provide better views of the elbow and additional

�rthopaedic ����������� S �ports ������ � Medicine ������� 1272 DeLee & Drez’s� O

i­nformation about the injury. When additional information about fractures of the radial head or coronoid may influence decision making, CT is useful. In particular, three-dimensional reconstructions with the distal humerus removed can provide an accurate characterization of the injury. Using such images, the preoperative planning will be more accurate. Additional information regarding the character of the injury can be obtained by viewing the elbow under the image intensifier once the patient is anesthetized. For some complex injuries, complete characterization of the injury pattern—and therefore, a final treatment plan— can only be made based on operative exposure. The surgeon must therefore be comfortable with ­ extensile

exposures providing adequate access to the injury components. The Mayo classification of olecranon fractures distinguishes three factors that have a direct influence on treatment: (1) fracture displacement, (2) comminution, and (3) ulnohumeral instability (Fig. 19F1-22).112 Type I fractures that are nondisplaced or minimally displaced are either noncomminuted (type IA) or comminuted (type IB) and are treated nonoperatively. Type II fractures feature displacement of the proximal fragment without elbow instability—these fractures require operative treatment. Type IIA fractures, which are noncomminuted, are well treated by tension band wire fixation. When the fracture is oblique, an ancillary interfragmentary compression screw

TYPE I Undisplaced

TYPE II Displaced– Stable

A – Noncomminuted

B – Comminuted

A – Noncomminuted

B – Comminuted

TYPE III Unstable

Figure 19F1-22   The Mayo classification of olecranon fractures divided fractures according to displacement, comminution, and subluxation-dislocation.

Elbow and Forearm 1273

can be added. Type IIB fractures are comminuted and require plate fixation. Type III fractures feature instability of the ulnohumeral joint and require surgical treatment. Fractures of the proximal ulna can appear extremely complex. The identification of basic injury patterns can facilitate management. Even a simple fracture pattern of the olecranon can have associated injuries, which the surgeon must be careful not to miss. Varus posteromedial rotational instability pattern injuries have only recently been recognized and described. The central element of this injury is a fracture of the anteromedial facet of the coronoid process, resulting in varus ­instability.72 There is an associated injury—either an avulsion of the lateral collateral ligament complex from the lateral epicondyle or a fracture of the olecranon, but rarely both. The radial head is rarely fractured. Most olecranon fracture-dislocations occur in either an anterior or a posterior direction.101,102 Anterior olecranon fracture-dislocations have been described as trans-olecranon fracture dislocations because the trochlea of the distal humerus implodes through the trochlear notch of the ulna as the forearm translates anteriorly (see Fig. 19F1-18).101,113 This pattern can be confused with posterior fracture-dislocations with a similar appearance, so the term anterior olecranon fracture-dislocation may be preferable. Anterior fracture-dislocations are injuries to the ulnohumeral articulation, with the radioulnar relationship being relatively preserved and the radial head rarely injured. The fracture of the proximal ulna can be a simple oblique fracture but is often complex, including fragmentation of the olecranon, fragmentation extending into the ulnar diaphysis, and fracture of the coronoid. Associated collateral ligament injury is unusual.101,113 It is useful to consider posterior fracture-dislocations of the olecranon as the most proximal type of posterior Monteggia injury.114 Common factors of posterior Monteggia injuries include an apex posterior fracture of the ulna, posterior translation of the radial head with respect to the capitellum, fracture of the radial head, and frequent injury to the lateral collateral ligament complex. With posterior olecranon fracture-dislocations (or type A posterior Monteggia fractures according to Jupiter and colleagues114), the fracture of the ulna occurs at the level of the olecranon and is nearly always associated with a fracture of the coronoid process. When a complex olecranon fracture-dislocation is identified as being posterior in direction, fractures of the radial head and coronoid, and injury to the lateral collateral ligament should be suspected (see Fig. 19F1-19). The fracture of the coronoid varies somewhat with each specific type of olecranon fracture-dislocation. Those associated with a varus posteromedial mechanism will involve the anteromedial facet and the tip and the radial head will not be fractured. The fracture of the coronoid that occurs with an anterior olecranon fracture-dislocation is usually a single, large fragment involving nearly the entire coronoid, but it is occasionally split once or twice in the sagittal plane. The fractures of the coronoid associated with posterior olecranon fracture-dislocations are more variable, including the occasional fracture of the tip, a single large fragment, comminution into three fragments (anteromedial, central, and sigmoid notch), and more extensive comminution (see Fig. 19F1-19).

Operative Techniques in the Athlete Skin Incision A midline posterior skin incision is used for all complex fractures of the proximal ulna. Traumatic wounds are incorporated. Some surgeons prefer that the incision not pass directly over the olecranon, and curve it slightly.66 A direct midline incision may cut fewer cutaneous nerves.115

Tension Band Wiring The fracture is opened and hematoma removed to be sure that comminution and articular involvement are limited. Periosteum and muscular attachments are elevated minimally, just enough to ensure accurate reduction of the fragments. A large tenaculum clamp can be used to maintain reduction of the olecranon. A drill hole made in the dorsal surface of the ulna can provide a good anchor point for the distal tine of the clamp.

Kirschner Wire Technique Two parallel Kirschner wires are drilled across the osteotomy site. Most surgeons use 0.062-inch wires, but we use 0.045-inch wires with few problems. The wires are often drilled parallel to the ulnar diaphysis, but we and others favor drilling the wires obliquely so that they pass through the anterior ulnar cortex, just distal to the coronoid process.116,117 This is intended to limit the potential for wire migration. After exiting the anterior cortex, the wires are retracted between 5 and 10 mm, anticipating subsequent impaction of the wires into the olecranon process ­proximally. The extensor carpi ulnaris and flexor carpi ulnaris muscles are partly elevated from the apex of the ulna distal to the osteotomy site to expose the cortex. The appropriate distance between the fracture and this drill hole has been commented on based on mechanical calculations, but the placement of these holes is determined more practically by the transition from the flat proximal ulna to the apex posterior triangular shape of the diaphysis. Likewise, the placement of the drill holes in the anteroposterior plane is not critical except that they should not be so dorsal as to risk breaking out of the ulna. Large drill holes (2.5-mm) facilitate wire passage. Many surgeons use a single 18-gauge stainless steel wire for the tension wire, but we prefer to use two 22-gauge stainless steel wires, each passed through its own drill hole distally (Fig. 19F1-23). The smaller wires are less ­prominent. The tension wires are placed in a figure-of-eight over the dorsal ulna. The proximal end of the wire is passed deep to the Kirschner wires, through the insertion of the triceps using a large-gauge needle. The tension wires are then tensioned on both the medial and lateral sides of the ulna until the wire rests flush with the ulna. Some surgeons prefer to twist the wires until they are very tight, but this cannot be done with smaller gauge wires because they will break. The wire does not need to be tight, it is only important to take up all of the slack in the wires. This is

�rthopaedic ����������� S �ports ������ � Medicine ������� 1274 DeLee & Drez’s� O

A

B

D

C

E

G

F

Figure 19F1-23   See legend on opposite page.

Elbow and Forearm 1275

H

I

Figure 19F1-23, cont’d   Tension band wiring is useful for simple isolated fractures of the olecranon. A, Lateral radiograph demonstrates a transverse fracture of the olecranon. B, The fracture is realigned and secured with two 0.045-inch Kirschner wires drilled obliquely so that they engage the anterior ulnar cortex distal to the coronoid. C, Two 22-gauge stainless steel wires are passed through drill holes in the ulna distal to the fracture. D, The wires are passed underneath the triceps insertion adjacent to the Kirschner wires. E and F, The wires are then tensioned both medially and laterally.�G, The Kirschner wires are then bent 180 degrees and impacted into the olecranon beneath the triceps insertion. H, The wires are bent into the adjacent soft tissues so that the entire construct has a very low profile. I, Postoperative lateral radiograph.

done by twisting the wire until it starts to bend over itself. The twisted ends are trimmed and bent into adjacent soft tissues to limit prominence. The proximal ends of the Kirschner wires are bent 180 degrees and trimmed. The triceps insertion is then incised, and these bent ends are impacted into the proximal olecranon with a bone tamp. The strength of the fixation can be tested by completely flexing the elbow; the fracture should not separate.

Screw Technique Some surgeons prefer to use screws instead of ­ Kirschner wires. Some recommend using a very long screw that engages the medullary canal of the ulnar diaphysis distally.118 Others recommend aiming the screw anteriorly to engage the anterior ulnar cortex. An oblique screw is particularly well suited to an oblique fracture. The remaining portion of the technique is as described for the Kirschner wire technique.

Plate and Screw Fixation When a plate is applied to the proximal ulna, it should be contoured to wrap around the proximal aspect of the ulna (see Figs. 19F1-18 and 19F1-19). A straight plate will only have two or three screws in metaphyseal bone proximal to the fracture. Many patients with complex proximal ulna fractures have osteopenic bone, which further compromises the strength of plate and screw fixation. Bending the plate around the proximal aspect of the olecranon provides additional screws in the proximal fragment. In addition, the most proximal screws are oriented orthogonal to the more distal screws. Finally, the most proximal screws can be very long, crossing the fracture line into the distal fragment. In some cases, these screws can be directed to engage one of the cortices of the distal fragment, such as the anterior ulnar cortex.

A plate applied to the dorsal surface of the proximal ulna also has several advantages over plates applied to the medial or lateral aspects of the ulna. Placing the plate along the flat dorsal surface can assist in obtaining and monitoring reduction. The dorsal surface is in the plane of the forces generated by active elbow motion so that the plate functions to a certain extent as a tension band. Finally, dorsal plate placement requires limited soft tissue stripping. Exposure of the ulna should preserve periosteal and muscle attachments. A plate contoured to wrap around the proximal ulna can be placed on top of the triceps insertion with few problems. This is particularly useful when the olecranon fragment is small or fragmented. Alternatively, the triceps insertion can be incised longitudinally and partially elevated medially and laterally sufficiently to allow direct plate contact with bone. Distally, a dorsal plate will lie directly on the apex of the ulnar diaphysis. This might appear unsettling to some surgeons but has not been a problem in our hands. One advantage of this situation is that the muscle need only be split sufficiently to gain access to this apex; there is no need to elevate the muscle or periosteum off either the medial or lateral flat aspect of the ulna. No attempt is made to precisely realign intervening fragmentation; once the relationship of the coronoid and olecranon facets and the overall alignment are restored, the remaining fragments are bridged, leaving their soft tissue attachments intact. Despite extensive fragmentation, bone grafts119 are rarely necessary if the soft tissue attachments are ­preserved.101,102

Operative Technique for Fracture-Dislocations Fractures of the radial head and coronoid process can be evaluated and often definitively treated through the exposure provided by the fracture of the olecranon process. With little additional dissection, the olecranon fragment can be mobilized proximally, providing exposure of the coronoid through the ulnohumeral joint. If the exposure of

�rthopaedic ����������� S �ports ������ � Medicine ������� 1276 DeLee & Drez’s� O

the radial head through the posterior injury is inadequate, a separate muscle interval (e.g. Kocher’s or Kaplan’s intervals66)—accessed by the elevation of a broad lateral skin flap—can be used. If the exposure of the coronoid is inadequate through the straight dorsal skin incision, a separate medial or lateral exposure can be developed. Posterior olecranon fracturedislocations often require a lateral exposure to address a fracture of the radial head or coronoid, or to repair the lateral collateral ligament. When the lateral collateral ligament is injured, it is usually avulsed from the lateral epicondyle. This facilitates both exposure and repair. The lateral collateral ligament origin and common extensor musculature can be included in an anterior or posterior flap, or mobilized distally. Improved exposure of the coronoid can be obtained by releasing the origins of the radial wrist extensors from the lateral supracondylar ridge and elevating the brachialis from the anterior humerus and by excising the fractured radial head.72,120 A medial exposure, between the two heads of the flexor carpi ulnaris, or by splitting the flexor-­pronator mass more anteriorly may be needed to address a complex fracture of the coronoid, particularly one that involves the anteromedial facet of the coronoid process. The fracture of the coronoid can often be reduced directly through the elbow joint using the limited access provided by the olecranon fracture.103,121,122 Provisional fixation can be obtained using Kirschner wires to attach the fragments either to the metaphyseal or diaphyseal fragments of the ulna, or to the trochlea of the distal humerus when there is extensive fragmentation of the proximal ulna.123,124 An alternative to keep in mind when there is extensive fragmentation of the proximal ulna is the use of a skeletal distractor (a temporary external fixator).101,123 External fixation applied between a wire driven through the olecranon fragment and up into the trochlea and a second wire in the distal ulnar diaphysis can often obtain reduction indirectly when distraction is applied between the pins. Definitive fixation can usually be obtained with screws applied under image intensifier guidance. The screws are placed through the plate when there is extensive fragmentation of the proximal ulna. If the coronoid fracture is comminuted and cannot be securely repaired, the ulnohumeral joint should be protected with temporary hinged or static external fixation, or temporary pin fixation of the ulnohumeral joint depending on the equipment and expertise available. A long plate is contoured to wrap around the proximal olecranon. A very long plate should be considered (between 12 and 16 holes), particularly when there is extensive fragmentation or the bone quality is poor. When the olecranon is fragmented or osteoporotic, a plate and screws alone may not provide reliable fixation. In this situation, it has proved useful to use ancillary tension wire fixation to control the olecranon fragments through the triceps insertion.

Optimizing Outcomes • The success of tension band wiring is predicated on selecting simple fractures and using meticulous technique with attention to all aspects that may affect wire prominence or migration.

• Plates should be reinforced with tension wires engaging the triceps insertion when the olecranon fragment is small, fragmented, or osteoporotic. • Complex proximal ulna fractures are often fracture­dislocations. Optimal management requires recognition of the injury pattern and treatment of each particular aspect of the injury.

Complications Tension band wire constructs can fail when used for complex fractures or fracture-dislocations but rarely fail when used for simple fractures unless the patient returns to forceful activity too soon. Plate loosening is most common in older patients with fracture-dislocations when a noncontoured plate has been placed on either the medial or lateral side of the proximal ulna. Failed internal fixation can be salvaged with realignment and repeat internal fixation using a dorsal contoured plate and screws. If there is a bone defect or delayed union, autogenous cancellous bone graft can be applied to the fracture site. Nonunion after simple olecranon fractures is very unusual.125 Proximal ulnar nonunion usually occurs after a fracture-dislocation of the proximal ulna. Union can usually be achieved with contoured dorsal plate fixation and autogenous bone grafting.125,126 Ulnohumeral instability is sometimes a surprise to the surgeon treating a complex proximal ulna fracture. It is usually the result of some combination of fixation of the proximal ulna with apex dorsal deformity, as well as inadequate treatment of the coronoid, radial head, and lateral collateral ligament complex. These can often be salvaged by secondary surgery, often including the use of hinged external fixation.95,127-129 Both the elbow and the forearm can become stiff with these injuries, particularly posterior olecranon fracturedislocations. Proximal radioulnar synostosis occurs fairly frequently with these injuries.

REHABILITATION The hand is typically swollen and ecchymotic after an elbow injury and can become permanently stiff if confident active exercises are not encouraged. Active exercises and gentle functional use of the arm for daily activities is usually initiated within a few days, most often the morning after surgery. The patient is encouraged to use their other hand, gravity, and pushing against other objects to help assist with elbow mobilization. Passive manipulation by a therapist or family member is discouraged. There is a long-standing belief that this will contribute to heterotopic bone formation. It may also be more likely to loosen implants or impede healing. Finally, the patient must learn how to mobilize his or her own arm if the exercise program is to be effective. In the treatment of traumatic elbow instability, when there is slight radiographic sagging or subluxation of an otherwise concentrically reduced joint, this can usually be addressed by encouraging confident active motion of the elbow.74,108 This adds a dynamic muscular component of stability, which overcomes what is likely a type of pseudosubluxation of the joint. More substantial subluxations

Elbow and Forearm 1277

should be treated operatively because they risk damage to the articular surface. Given the useful dynamic component of stability, combined with the fact that the elbow can dislocate in a cast when unstable, the value of casts or braces for enhancing elbow stability must be questioned; I do not use them. The idea of an extension block brace is also common, but probably not usually necessary because patients usually struggle to regain extension. For some elbow injuries—distal humerus fractures in particular—many surgeons prefer to immobilize the elbow in maximal extension overnight before initiating active exercises.36 The rationale for this is that flexion is easier to regain than extension. Considering the following counterpoints, I no longer immobilize the elbow in extension: (1) even an overnight immobilization will make the patient stiff in that position, and flexion is more important functionally than extension—an anxious patient who has trouble with exercises often fails to regain flexion; (2) if the ulnar nerve is transposed, immobilization in extension will bring it to a more posterior position where it may be more vulnerable at subsequent operations; and (3) the braces for elbow extension are much better than those for elbow flexion—I think it is easier to regain extension in a very stiff elbow. Elbow braces for stiffness can be either static-progressive (a position of static stretch is applied and an adjusted as the pain dissipates) or dynamic (a constant dynamic force is applied to the elbow). Some hinged external fixators incorporate a static-progressive stretch ­mechanism. When a lateral collateral ligament injury or an anteromedial facet coronoid fracture is repaired, it is useful to avoid shoulder abduction for a few weeks, so-called varus stress precautions.108 The use of continuous passive elbow motion (using various devices), with or without continuous anesthesia through a brachial plexus catheter, is gaining in popularity despite the lack of evidence that it improves elbow motion. It is not clear that the additional risk and cost are justified. I believe that active elbow motion is the key and that motivated patients will do better—I therefore shy away from passive treatments. Return to sports may be possible within a few months after relatively simple elbow fractures and dislocations but is risky earlier than 6 months after complex elbow trauma. Many of the complex elbow injuries discussed in this chapter result in substantial elbow impairment that may limit athletic effectiveness and could be career ending.

COMPLICATIONS Stiffness and Heterotopic Ossification Stiffness is a complication that is common to all elbow injuries.109 Some permanent loss of elbow motion is to be expected for all but the simplest injuries; usually, it is a loss of extension. Severe loss of motion is often associated with heterotopic ossification, joint incongruity, ulnar neuropathy, or arthrosis. In the absence of these associated problems, a program of active-assisted exercises supplemented by static-progressive or dynamic elbow splints can improve motion to a functional range in a large percentage

of patients.130,131 With complex or unresponsive stiffness, operative release of the contracted capsule, the constricted ulnar nerve, heterotopic bone, and osteophytes can often improve motion.109 Passive manipulation of the elbow in an attempt to regain motion is generally discouraged for fear of causing heterotopic bone, inhibiting healing, or fracturing the arm. The formation of heterotopic ossification can be reduced by the administration of a single local 700-Gy dose of radiation132 and, to some degree, by nonsteroidal anti-inflammatory medications.133 Both of these interventions may also inhibit fracture healing. Radiation also has a potential cancer risk that must be respected. It is still recommended that these prophylactic measures be used selectively. Patients with head injuries, high-energy fracture-dislocations, and repeat surgery during the first few weeks are at the highest risk for substantial heterotopic ossification.

Ulnar Neuropathy The ulnar nerve is vulnerable to constriction as it runs through the cubital tunnel fascia and Osborne’s fascia. Cubital tunnel syndrome—a gradually developing, chronic ulnar neuropathy—is quite common after elbow trauma,134 likely owing to the swelling, bleeding, tissue injury and distortion, scar formation and tissue contracture, heterotopic ossification, and arthrosis that develop after injury. Consideration has been given to routine in situ decompression of the ulnar nerve for some fractures that appear to be particularly at risk, such as fracture-dislocations of the elbow. There is an ongoing debate about whether to perform a complete anterior subcutaneous transposition of the ulnar nerve when treating a fracture of the distal humerus, or whether a limited mobilization is sufficient to protect the nerve. An ulnar neuropathy that develops after treatment of the injury can contribute to stiffness and pain in addition to affecting hand function.135 It is important to be aware of the importance of the ulnar nerve and always evaluate for symptoms and signs of ulnar nerve dysfunction, being particularly suspicious when stiffness or pain are greater than might otherwise be expected. Acute ulnar nerve dysfunction with an intact nerve (either due to the injury or handling of the nerve during surgery) usually, but not always, recovers. In can take a long time, in some cases more than a year or more, to recover. Serial clinical examination and nerve conduction studies and electromyography can help determine whether surgical intervention is worthwhile. Operative treatment is considered if serial neurophysiologic testing does not demonstrate improvement or if there is obvious worsening of the deficit on examination.

Instability Truly unstable elbows or flail elbows are usually related to bone loss or nonunion.38 Instability after a fracture­dislocation usually refers to residual malalignment and incongruity of the elbow.128,129 This will lead to arthrosis and must be addressed as soon as possible. Even a few weeks of elbow motion and use in a malaligned position can

�rthopaedic ����������� S �ports ������ � Medicine ������� 1278 DeLee & Drez’s� O

­ ermanently damage the articular surfaces. At that point, p only interpositional arthroplasty or total joint arthroplasty can be considered; neither are great options in young, active individuals. If the elbow articulation remains malaligned after operative treatment, additional surgery is necessary as soon as possible, even though early reoperation increases the risk for extensive heterotopic ossification. Operative treatment of persistent elbow malalignment consists of restoration of the stabilizing anatomy of the elbow (radial head, coronoid, collateral ligaments) and hinged external fixation to maintain concentric reduction during the initial treatment period.127-129 Although this is fortunately uncommon, it is my impression that patients in whom the trochlear notch is relatively spared do better than those with instability associated with large and complex coronoid fractures.

Nonunion The metaphyseal bone around the elbow is well vascularized, and nonunions are relatively uncommon overall; ­ however, several specific injuries are more prone to nonunions than others and require specific attention in ­treatment. Whereas nonunion of the olecranon is unusual, fractures of the proximal ulna with extensive metaphyseal comminution have been more problematic.125,126 Much of this appears to relate to inadequate fixation of the proximal, metaphyseal fragment, particularly in the setting of osteoporotic bone. In the treatment of both fresh fractures and nonunions, it seems better to apply a long plate on the dorsal surface of the ulna that wraps around the olecranon process, thereby providing additional screw fixation.102,126,127 It is also important not to remove the muscular and periosteal attachments to the comminuted fragments in the metaphysis, but rather to bridge this area with a long plate and use the fragments as vascularized bone graft, rather than depending on them for stability. In the setting of nonunion, débriding the fracture site of sclerotic, inflammatory, and devitalized tissues and applying a nonstructural cancellous bone graft has been successful in our experience.101,102 Nonunion of the radial head may be more common than previously recognized.52,81,136,137 Minimally displaced fractures involving the radial neck often fail to heal.136,137 We do not know the true incidence of nonunion because it rarely causes symptoms, we do not usually reevaluate the elbow radiographically, and un-united fractures of the radial head may eventually heal without additional intervention if followed for more than 2 years.52,136 Complex fractures of the radial head that are repaired with a plate and screws are also prone to early failure and nonunion.52,81 Usually the reconstructed radial head has served well as a stabilizer of the elbow, and with the ligaments now healed, it can safely be resected without ­replacing it. Nonunions of the distal humerus tend to occur at the supracondylar level. Osteochondral fracture fragments also occasionally fail to heal or develop avascular necrosis. Nonunion is usually related to inadequate fixation, overvigorous rehabilitation, or bone loss and devitalization of

the fragments. It can be salvaged in most cases with ulnar nerve release, elbow capsular release, stable internal fixation, and autogenous cancellous bone grafting.43

Infection Infections are fortunately uncommon after the operative treatment of elbow fractures. They are usually related to complex open injuries, devitalized fracture fragments, and immunocompromised patients. These infections are often treated with serial débridement, retention of implants, and parenteral antibiotics, particularly when the fracture is complex. Healing of the fractures can usually be achieved with this regimen (see Fig. 19F1-9). Eventually, complete eradication of the infection usually requires implant removal. Elbow mobility is typically allowed during treatment of infection. It can be assisted with external fixation or hinged external fixation when there is associated elbow instability or an unsupported fracture or nonunion.

Wound Problems Wound problems are also uncommon, owing to the excellent blood supply. Most patients with wound edge necrosis or slight wound separation can be treated with dressing changes, but patients with exposed implants or an underlying total elbow arthroplasty should be treated operatively to obtain better skin cover. This can often be accomplished with local rotational flaps, pedicled flaps (such as a radial forearm flap), and on occasion a free microvascular tissue transfer.138,139 I follow the so-called reconstructive stepladder, using the simplest procedure that will address the problem.

Arthrosis Despite the sometimes dramatic claims made by nutritional supplement and pharmaceutical companies, there is no cure for post-traumatic arthrosis. Patients must adapt to and live with the arthrosis or consider reconstructive procedures, none of which are perfect. Débridement of osteophytes and loose bodies and capsulectomy can be useful in the short term and may be best used in conjunction with ulnar nerve release because ulnar neuropathy is a commonly associated problem.140 Patients with severe articular damage or incongruity must consider interpositional arthroplasty141,142 or total elbow arthroplasty.143 Total elbow arthroplasty has a finite ­ life-span (with each revision becoming increasingly more ­difficult), is more prone to infection and major complications than knee or hip arthroplasty, and requires strict activity limitations (a 5-kg lifting limit). Total elbow arthroplasty is suitable only for older, low-demand patients. Fascial interpositional arthroplasty is better suited for younger, more active patients. The material used for interposition has traditionally been the cutis layer of skin or fascia lata, but more recently allograft Achilles tendon has been used. Interpositional arthroplasty does not eliminate elbow pain and leaves the elbow somewhat unstable; it is best suited for patients with severe stiffness related to arthrosis.141,142

Elbow and Forearm 1279

C

r i t i c a l

P

o i n t s

• Apparent capitellum fractures are often much more complex, involving the anterior trochlear, the lateral epicondyle, the posterior aspect of the lateral column, and even the posterior trochlea and medial epicondyle. • Long-term data suggest that most isolated radial head fractures, even those displaced 2 mm more, do not benefit from operative treatment, but operative treatment is straightforward and can be considered. • More displaced fractures (no contact between fracture fragments) of the radial head indicate important forearm or elbow ligament injuries and associated fractures. Even partial head fractures can be complex and difficult to repair, so the surgeon should be prepared for radial head replacement. • Most simple elbow dislocations are stable after reduction and are treated with immediate active motion. Although still vulnerable, some athletes have been able to return to their sports—accepting the attendant risks—within a few weeks. • Even small coronoid fractures can be troublesome. The terrible triad fracture-dislocation (dislocation with fractures of the radial head and coronoid) is an obvious severe injury, but beware of anteromedial coronoid fractures— there is often an associated lateral collateral ligament injury, and these can subluxate and cause problems.

• Most olecranon fractures are displaced and benefit from operative treatment. • If the repaired elbow will remain concentric with active exercises, these should be initiated as soon as possible. If there is concern about the fixation or stability, it is reasonable to immobilize the elbow to protect the repairs, knowing that it may be difficult to regain motion and might require surgery.

S U G G E S T E D

R E A D I N G S

Hotchkiss RN: Displaced fractures of the radial head: Internal fixation or excision. J Am Acad Orthop Surg 5:1-10, 1997. Morrey BF: Complex instability of the elbow. J Bone Joint Surg 79A:460-469, 1997. O’Driscoll SW, Jupiter JB, Cohen M, Ring D, McKee MD: Difficult elbow fractures: Pearls and pitfalls. Instructional Course Lectures 52:113-134, 2003. Pugh DM, Wild LM, Schemitsch EH, et al: Standard surgical protocol to treat elbow dislocations with radial head and coronoid fractures. J Bone Joint Surg 86A:1122-1130, 2004. Ring D, Gulotta L, Jupiter J: Articular fractures of the distal part of the humerus. J Bone Joint Surg 85A:232-238, 2003. Ring D, Jupiter JB: Fracture-dislocation of the elbow. J Bone Joint Surg 80A:566580, 1998. Ring D, Quintero J, Jupiter JB: Open reduction and internal fixation of fractures of the radial head. J Bone Joint Surg 84A:1811-1815, 2002.

R eferences Please see www.expertconsult.com

S e c t i o n

F

Forearm Fractures 2. Pediatric Elbow Fractures and Dislocations Anil S. Ranawat, Michael P. McClincy, Samuel P. Robinson, and James P. Bradley

The elbow is a common site of injury in children. It is ­estimated that nearly 80% of athletic-related fractures and 65% to 75% of all fractures in children occur within the upper extremity.1-5 Nearly 7% to 10% of these injuries involve the elbow region.3-6 Fractures within this region occur more frequently in younger populations compared with adults, and the estimated peak age of injury is between 5 and 10 years. The frequency of pediatric elbow fractures is multifactorial, but the most significant cause is the ­process of bone remodeling that occurs during childhood. Skeletal ossification of the pediatric elbow proceeds in a highly ordered fashion, often permitting accurate diagnoses based entirely on age, sex, and proper radiographs (Fig. 19F2-1). The distal humerus has four sites of secondary ossification: the capitellum at 1 year, the medial epicondyle at age 4 to 5 years, the trochlea at age 8 to 9 years, and finally the lateral condyle at age 10 years.7,8 In the forearm, the radial head appears at about age 4 to 5 years, whereas the olecranon ossification center appears at about 8 to

9 years.7,8 These patterns occur at an earlier age for females and are slightly delayed in males by 1 to 2 years.7 A systematic approach to evaluating pediatric elbow radiographs is essential to accurately diagnose pediatric elbow trauma. First, the proximal radius should align with the capitellum in all views. Second, the longitudinal axis of the ulna should align directly or slightly medial to the longitudinal axis of the humerus on true anteroposterior radiographs. Third, the anterior humeral line should bisect the capitellum on a true lateral radiograph. Fourth, the humeral-capitellar (Baumann’s) angle should be within 9 to 26 degrees of valgus.8,9 Finally, the fat pad shadow should be evaluated for an occult fracture because it is ­visible in 76% of elbow fractures. The most common injuries presenting with a fat pad shadow are supracondylar fractures (53%), proximal ulna fractures (26%), and lateral condyle fractures (9%).10-12 If it is still difficult to distinguish a secondary ossification from a fracture, additional imaging modalities may be needed such as contralateral

1280 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

F: 1–11 mo M: 1–26 mo F: 8–11 yr M: 9–13 yr

F: 4–5 yr M: 5–8 yr

F: 4–5 yr M: 5–8 yr

F: 1–11 mo M: 1–26 mo

F: 8–9 yr M: 9–10 yr

F: 4–5 yr M: 5–8 yr

F: 12–14 yr M: 13–16 yr

F: 8–9 yr M: 9–10 yr

F: 14+ yr M: 17+ yr

B

A

F: 10+ yr M: 12+ yr

F: 10+ yr M: 12+ yr

Figure 19F2-1   A, Secondary ossification centers of the pediatric elbow with expected age range of appearance for each gender. B, Expected age of fusion of secondary ossific centers in children, resulting in a skeletally mature elbow.

radiographs, arthrography, ultrasound, or magnetic resonance imaging (MRI). The pediatric elbow is composed of three main articulations. The first two articulations form a hinge joint to permit flexion and extension. In this hinge joint, the distal humerus articulates with the proximal aspects of both the ulna (olecranon) and the radius (radial head). The ­proximal

radius and ulna also articulate with each other in this region, and this interaction permits proper rotation of the elbow joint. Each anatomic region has unique anatomy and fracture patterns associated with it. This chapter reviews the common fractures encountered in the pediatric elbow and details treatment plans, complications, and patient outcomes after these injuries (Table 19F2-1).

  TABLE 19F2-1  Age, Classification, and Treatment of Common Pediatric Elbow Fractures Fracture

Age (yr)

Classification

Treatment

Supracondylar

5-8

Lateral condyle

6-10

Medial epicondyle

10-14

Gartland I Gartland II Gartland III Jakob I—no displacement Jakob II—minimal displacement (<4 mm) Jakob III—full displacement, malrotation General population Mild displacement (<5 mm) Moderate displacement (>5 mm) Incarcerated fragment Throwing athletes Mild displacement (<2 mm) Moderate displacement (>2 mm) Nondisplaced (<2 mm) Displaced (2-4 mm) Type I (0-30 degrees angulation) Type II (30-60 degrees angulation) Type III (60-90 degrees angulation)

Immobilization CRPP CRPP (rare ORIF) Immobilization CRPP CRPP or ORIF

Olecranon

<14

Radial neck

8-12

CRPP, closed reduction with percutaneous pinning; ORIF, open reduction with internal fixation.

Immobilization Immobilization or CRPP ORIF Immobilization ORIF Immobilization ORIF Immobilization CRPP CRPP (or ORIF)

Elbow and Forearm 1281

DISTAL HUMERUS FRACTURES The most commonly fractured site of the pediatric elbow is the distal humerus, accounting for nearly 75% of all fractures.13-15 This site is highly predisposed to fracture primarily because of the reduced anteroposterior diameter caused by local bone remodeling. Other factors associated with an increased incidence of distal humerus fractures in children include ligamentous laxity permitting excessive hyperextension and a relatively thicker anterior versus posterior capsule.16,17 The most common injuries found in this region are supracondylar, T-condylar, and transphyseal fractures.

Supracondylar Fractures Supracondylar fractures account for nearly 60% to 75% of all pediatric elbow fractures and 10% of pediatric fractures in general, making them the most common fracture ­pattern.13-15 Most of these fractures occur between the ages of 5 and 8 years.13-15 In general, older populations approaching skeletal maturity tend to experience a greater number of elbow dislocations and intercondylar distal humeral fractures.13,17 Supracondylar fractures are classified by their mechanism of injury and degree of displacement. Most supracondylar fractures are extension type, accounting for nearly 98% of cases.11,12 They occur by a fall on an outstretched, hyperextended arm.14 The downward force produces tension within the joint anteriorly and the olecranon process imparts compression posteriorly. These injuries typically present with the distal bone fragment lying posteriorly, with posteromedial displacement occurring more frequently than posterolateral displacement.18-21 The rarer form of supracondylar humerus fracture is a flexion-type variant, which accounts for only 2% of all supracondylar fractures. These fractures usually present with anterior displacement of the distal fragment and are believed to be the result of a direct fall on the posterior portion of the distal humerus.22 Both extension and flexion fractures are further classified by the Gartland classification.23 Type I fractures are nondisplaced and are usually diagnosed by a fat pad sign.10 Type II fractures are displaced but maintain some degree of cortical contact, and type III fractures are fully displaced with no cortical contact. Evaluation of these injuries requires a thorough history and physical to explore for associated injuries. A careful examination including proper documentation must include all motor, sensory, and vascular structures. In addition, soft tissue swelling and ecchymosis should also be carefully documented as indicators of more significant injury and possible compartment syndrome. Other associated injuries include ipsilateral upper extremity fractures in 5% of cases, and a 1% to 3% incidence of open fractures has been reported.24,25 Treatment methods for supracondylar fractures depend primarily on the degree of fracture displacement and fracture stability. Type I injuries are treated by ­immobilization in a full arm cast for 2 to 3 weeks in 90 degrees of flexion with no formal reduction followed by protected active range of motion. Type II fractures may be treated conservatively especially when the anterior humeral line passes

through the capitellar ossification center on the lateral radiograph. If there is more significant angulation, especially medial impaction and varus angulation, a formal reduction is required to adjust for any sagittal, coronal, or combined malalignments. Traditionally, type II injuries require extreme position of flexion to stabilize the fracture. This position has been associated with neurovascular compromise and compartment syndrome. In addition, closed reduction and casting have been associated with higher rates of both early and late complications when compared with closed reduction and pinning.26 This has led many to favor operative treatment for type II fractures with closed reduction and percutaneous pinning. With this approach, the extremity is immobilized in only 90 degrees of flexion, and the extreme position of flexion is avoided.27,28 All type III fractures require operative intervention, usually consisting of closed reduction and percutaneous pinning. In about 8% of cases, an open reduction may be necessary to obtain an acceptable reduction.29-31 Historically, it was thought that all displaced supracondylar humerus fractures should be treated emergently. A growing body of literature has shown that delaying surgery to the following day for closed fractures presenting in the late evening or at night with normal neurovascular examination has similar outcomes.32-34 For the most part, extension and flexion fractures are handled similarly. Extension fractures are usually immobilized in 90 degrees, whereas flexion fractures may be casted in a more extended position. Some authors report that displaced flexion-type fractures require open reduction more often than extension-type fractures.11,12,35,36 In general, most fractures require a general anesthetic during reduction. The reduction usually begins by longitudinal traction with the arm in hyperextension, followed by properly aligning the distal fragment for medial or lateral displacement. The elbow is then gradually flexed and either supinated (lateral) or pronated (medial), depending on the initial direction of displacement.36 Usually two lateral divergent pins are first placed with the elbow in hyperflexion.37,38 If the fracture requires more rotational stability, either a third lateral pin or a medial pin may be placed (Fig. 19F2-2). The medial pin should be inserted with the arm in extension to avoid iatrogenic injury to a subluxable ulnar nerve. Although this can be done percutaneously, most recommend a mini open approach during medial pin fixation.39 Specific techniques for reduction and pin configuration will vary on a case-wise basis. Pin configuration is ultimately dependent on the fracture’s stability, direction of displacement, and severity of localized swelling. Cross-pin formations are more biomechanically sound but increase the risk for ulnar nerve injury, especially when done percutaneously.40,41 Studies have shown that lateral pins do provide clinical outcomes similar to crosspins without the risk for neurologic injury.42-44 Divergent ­lateral pinning, as opposed to the traditional parallel pinning, has been shown to increase stability while avoiding the neurologic injury associated with cross-pins.15,39,45 Casts and pins are usually removed after 3 weeks. In most cases, early ranges of motion exercises are encouraged to prevent ­stiffness. The incidence of neurologic injuries for supracondylar fractures ranges from 5% to 19%.46,47 Some reports cite

1282 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 19F2-2   A and B, Injury radiographs of an extension type II supracondylar humerus fracture.   C and D, Postoperative radiographs after closed reduction and percutaneous pinning of supracondylar humerus fracture. (Courtesy of Dr. Jan Grudziak).

B

A

C

D incidence up to 40% for type III injuries.48,49 Injuries to the radial nerve are most commonly reported, but it is hypothesized that anterior interosseous nerve injury is more common but frequently undiagnosed.46,50,51 Median and ulnar nerve injuries are also possible from either the initial injury or operative treatment. Ulnar nerve injuries have been linked to flexion-type fractures as a result of displacement by the proximal fragment and also from medial pin placement. When evaluating supracondylar fractures and neurologic injury, the degree of displacement and the position of the proximal fragment suggest the injured nerve.52-54 Radial nerve injuries are more likely when the proximal fragment is displaced laterally, whereas median and anterior interosseous nerve injuries are more common when the proximal fragment is displaced medially.51,52 Most neurapraxias relating to the original injury recover over 3 to 6 months, whereas iatrogenic injuries may require more invasive interventions.55 Vascular injury resulting from supracondylar fractures occurs in less than 1% of cases, although an increased ability to document injury to the vessels has characterized ischemia in up to 19% of displaced injuries.49,56,57 Vascular injuries are most common with type III fractures.53 Posterolateral displacement of the distal fragment has been correlated with higher frequency of brachial artery injuries.49,58 Examination should consist of assessing peripheral pulses, capillary refill, and compartment checks.59-61 A Doppler device may be useful to confirm distal blood flow in both the emergency and operating room settings. Significantly diminished or absent pulse in the setting of

a displaced fracture is a possible indicator of vascular compromise or injury. In cases of possible vascular compromise, prompt gentle reduction and vascular reassessment should be performed immediately. If the extremity has good capillary refill and a pulse on Doppler, then pin fixation is appropriate without an open procedure because the vessels are likely in spasm owing to the original displacement.49 If, however, the extremity has poor capillary refill and is still pulseless, open exploration is indicated.62 Postoperatively, careful neurovascular examinations are necessary for about 24 hours. The patients should be assessed for appropriate immobilization, monitoring the patient’s discomfort level at rest and with passive range of motion of the fingers.63 If compartment syndrome is suspected, the splint or cast should be released. If symptoms continue, then compartment pressures should be checked, and fasciotomies may be necessary. Cosmetic or functional complications can arise during treatment. Malalignment can present in any direction. Cubitus varus is the most commonly cited deformity, occurring in up to 5% to 10% of all supracondylar fractures.55 It is usually linked to a poor reduction or loss of reduction or fixation.20 Besides the cosmetic effect, a cubitus varus deformity increases the risk for tardy ulnar nerve palsy and subsequent lateral condylar fracture in some patients.64-66 The added stability of percutaneous pinning has dramatically decreased the incidence of cubitus varus.19,21,42,43,51 Although little functional impairment is associated with cubitus varus, distal humeral osteotomies will correct it. Poor outcomes have also been observed in patients with

Elbow and Forearm 1283

undiagnosed neurovascular injuries, compartment syndrome, myositis ossificans, and avascular necrosis.67

T-Condylar Fractures Unlike supracondylar fractures, T-condylar fractures are most common in the adolescent age group, with the peak incidence of these injuries occurring between the age of 11 years and the middle teens.68-71 T-condylar fractures are essentially an intra-articular subset of supracondylar fractures in older pediatric patients. These fractures generally resemble adult-type fracture patterns as the ossification centers of the distal humerus begin to fuse.72 The distal free condylar components can be nondisplaced or significantly malrotated in both the coronal and sagittal planes. As with all pediatric elbow fractures, a careful history and physical examination must be performed and documented on presentation. The entire arm should also be closely examined for potential ipsilateral fractures that may go unnoticed by the patient or examiner. There is usually significant swelling and crepitation on attempted range of motion because these are relatively higher energy injuries.69 There are two common injury patterns for T-­condylar fractures. Extension patterns result from the coronoid ­process driving into the trochlea and cause posterior displacement. Flexion patterns arise from the olecranon acting as a wedge on the posterior portion of the condylar region and effectively splitting it.68,69 Either fracture pattern may be displaced or nondisplaced. Nondisplaced fractures can be treated with immobilization for 4 weeks at 90 degrees of flexion. Fractures with any degree of displacement should be treated with operative fixation.69,70,73 Internal fixation is accomplished in one of two ways, dependent on the degree of skeletal maturity. Younger, skeletally immature patients are treated with Kirschner wire fixation to stabilize each elbow column.74 It has recently been shown that closed reduction with percutaneous placement of partially threaded pins may also be a viable option for fixation.68,75 For older, skeletally mature patients, rigid fixation is used, consisting of lag screws to reconstruct the articular surface, followed by bicondylar plate fixation.70 Numerous approaches are available, including a triceps split, a Bryan-Morrey triceps elevation,69 or an olecranon osteotomy. In all cases, the ulnar nerve should be mobilized and may need to be transposed, especially in patients with preoperative ulnar symptoms.76 Early range of motion is crucial in preventing elbow stiffness, which is the most common complication. Good results have been documented in several series.70,73,74

Transphyseal Fractures Unlike other distal humeral fractures, transphyseal fractures are much less common. This fracture-separation of the distal humerus epiphysis is a Salter-Harris type II injury. Most of these injuries occur before the age of 4 years, with the peak incidence at early infancy.77-79 These injuries are commonly seen secondary to birth trauma or child abuse when the humeral physis fails owing to a shear mechanism.78-80 Transphyseal fractures occur just proximal to the epiphysis but more distal than supracondylar

fractures. The humeral physis is especially susceptible to shear forces before the fusion of the medial and lateral condyles, which occurs by the age of 10 (girls) or 12 (boys) years.35 DeLee and colleagues described three types based on age and the Thurston-Holland fragment.79 Most notable in the clinical presentation is young age (<4 years) and a less pronounced deformity than is commonly seen in supracondylar fractures. Severe swelling should be monitored because it may lead to neurovascular compromise. Posteromedial displacement of both the radius and ulna in a young patient is usually indicative of a transphyseal fracture. Radiographically, these injuries are indistinguishable from elbow dislocations until the capitellum is ossified. In transphyseal fractures, unlike elbow dislocations, the capitellum maintains its relationship with the proximal radius. Definitive diagnosis can be difficult, and arthrograms, ultrasound, and MRI may be needed.80-82 Treatment requires operative intervention, including closed reduction, percutaneous fixation, and immobilization. Complications include loss of reduction, varus angulation (cubitus varus), and osteonecrosis of the medial humeral condyle.83

Authors’ Preferred Method Type II and III supracondylar fractures are treated operatively with closed reduction and percutaneous pinning. With the patient supine on the operative table, we invert the fluoroscopy machine to use the wide base as a “hand table.” We then prepare and drape the patient, obtain a closed reduction under fluoroscopy, and place two lateral divergent pins with the elbow in hyperflexion to secure the fragments. If the stability of the construct is in question, a mini-open technique is used to provide additional ­fixation from the medial side. Both the reduction ­maneuver ­required for the reduction and the fixation technique required to provide stability vary with the specific fracture pattern. Displaced T-condylar fractures in skeletally immature ­ patients are treated with closed or open reduction ­followed by percutaneous fixation with Kirschner wires for each ­ elbow column. In older children, who are skeletally ­mature, open reduction and rigid internal fixation is used. The surgical approach varies depending on the specific fracture pattern, but the surgeon should be prepared to perform an olecranon osteotomy if adequate visualization is compromised. It is important to identify, mobilize, and consider transposition of the ulnar nerve. Once the fracture is exposed, we first reduce the articular surface and fix that segment using lag screws. We then apply a bicondylar plate for fixation of the articular segment to the shaft. Displaced transphyseal fractures require surgical intervention through a closed reduction and percutaneous fixation technique. We use an inverted fluoroscopy machine to help visualize the closed reduction; however, the adequacy of the reduction may still be difficult to determine. The surgeon should be prepared to obtain an intraoperative arthrogram to confirm the reduction before conclusion of the case. Fixation is primarily achieved by laterally based Kirschner wire placement; however, the required construct depends on the stability of the individual fracture pattern.

1284 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 19F2-3   Lateral condyle drawing—Milch classification. A, Milch type I fracture. The fracture line is   through the ossific center of the capitellum. B, Milch type II fracture. The fracture line is lateral to the ossific center of the capitellum.

LATERAL-SIDED FRACTURES Unlike the supracondylar region, fractures to the lateral elbow in the skeletally immature patient are subtle in presentation and require a high index of suspicion for accurate diagnosis. Most lateral elbow fractures involve the lateral condyle, whereas only a few rare cases involved the lateral epicondyle. They are most commonly seen in children between the ages of 6 and 10 years. Although the ossification center of the lateral condyle is one of the first to appear, the lateral epicondyle is the last to appear.84,85

Lateral Condyle Fractures Lateral condyle fractures represent most lateral-sided injuries in the pediatric elbow. These fractures represent about 12% to 20% of all pediatric elbow fractures, second only to supracondylar fractures.24,86 Fractures of the lateral condyle are most commonly seen in children between the ages of 6 and 10 years of age.14 The mechanism of injury is usually related to a varus force applied to a hyperextended elbow with the forearm in supination.87 These fractures are usually intra-articular and require aggressive treatment to attain good functional outcomes. There are two common classification systems. The Milch criteria differentiates by location,87-91 with type I fractures proceeding lateral to the trochlea and through the ossific

A

B

nucleus, whereas type II fractures course ­medially into the trochlear groove (Fig. 19F2-3). Type I fractures represent a true a Salter-Harris IV injury. These fractures preserve elbow stability because the trochlea is not involved. Type II fractures, in contrast, are considered similar to a SalterHarris II injury and may affect elbow stability owing to the involvement of the trochlea. The Milch classification does not guide treatment and is difficult to apply accurately.92 Unlike the Milch criteria, the Jakob criteria separate the injuries by degree of displacement (Fig. 19F2-4). Type I fractures are nondisplaced, type II fractures are moderately displaced (<4 mm), and type III fractures are completely displaced and malrotated with associated joint instability.20,87,93 The Jakob criteria tends to be more useful because they help guide treatment and may predict ­outcome.94 Lateral condyle fractures present with minimal observable physical deformity. Light swelling is only visible directly over the lateral condyle in most cases. Common clinical features that should be observed include crepitation during supination and pronation as well as pain during resisted wrist extension exercises. Radiographs are critical in diagnosing these fractures, and an oblique radiograph may be the best view for detecting this fracture pattern. MRI may also be helpful in assessing the stability of these subtle fractures.95 Treatment of lateral condyle fractures is dependent on the degree of displacement. All fractures with displacements of less than 2 mm are typically casted for 3 to 4 weeks with the forearm in a neutral position and the elbow flexed to 90 degrees.96,97 These fractures must be closely monitored with repeat radiographs to check for any displacement because late displacement and nonunion have been reported in up 23% of cases when the fracture line extends into the epiphysis.98 All fractures with greater than 2 mm of displacement should be treated with operative intervention.94,99 Closed reduction and percutaneous pinning should be attempted but must be monitored closely postoperatively owing to excessive rotational forces at the elbow.94 If an anatomic reduction is not obtained, open reduction with internal ­ fixation is suggested and is often necessary for most Jakob type III injuries. During this procedure, dissection must be done carefully to prevent avascular necrosis, especially avoiding posterior and distal soft tissue stripping. Thus, a lateral approach is usually favored over a posterolateral approach. Kirschner wires,

C

Figure 19F2-4   Lateral condyle drawing—Jakob classification. A, Type I fracture. The fracture line does not enter the articular surface, permitting it to remain stable. B, Type II fracture. Fracture extends into the articular surface but is minimally displaced. C, Type III fracture. Fracture extends into the articular surface and is highly displaced.

Elbow and Forearm 1285

which diverge within the metaphysis, are typically used to maintain reduction.94,100 Cancellous screws can be used as alternatives but have no clinical advantage.86,101 Complications are common, but early diagnosis and anatomic reductions are correlated with good outcomes. Any degree of incomplete or inadequate reduction can lead to a malunion or nonunion with a cubitus valgus deformity.102 Treatment is dictated by the cause of valgus deformity and the position of the nonunion. Malunions are treated with osteotomies only when necessary, whereas nonunions are best treated with open reduction and bone grafting when the displacement is less than 1 cm.103 For significantly ­displaced nonunions, the treatment is controversial.76 Other chronic symptoms include osteonecrosis, growth arrest and fishtail deformity, lateral spur formation, cubitus varus, and myositis ossificans.87,96,104,105 Ulnar nerve palsy can also develop, but it typically presents decades later and can be treated with a transposition.106,107 Cubitus varus is possible either from overgrowth of the lateral column or osteonecrosis of the trochlea. Osteonecrosis is more likely caused from extensive soft tissue stripping than from the actual injury and may also be related to the timing of surgery.87 Careful precautions taken during diagnosis, reduction, dissection, and immobilization can greatly decrease all these risks.

Lateral Epicondyle Fracture Fractures of the lateral epicondyle are extremely rare in children of all ages, and literature is only available in sparse case reports.89,108,109 These fractures are usually caused by direct traumatic forces to the lateral epicondylar apophysis and may involve an elbow dislocation.108 They may also occur in the overhead-throwing athlete from traction during follow-through.14 Most commonly, the fractures are immobilized for 2 to 3 weeks in a supinated position,110,111 but surgical excision is suggested if an incarcerated bone fragment is present.110 In either case, early motion is ­recommended to prevent elbow stiffness.

Authors’ Preferred Method Lateral condyle fractures that are displaced greater than 2 mm are ideally treated with closed reduction and percutaneous pinning using laterally based divergent Kirschner wires within the metaphysis. If an anatomic reduction is not possible, a careful open reduction is performed using a lateral approach with limited posterior and distal soft tissue stripping. Lateral epicondyle fractures rarely require surgical intervention, but surgical excision of incarcerated bone ­fragments is occasionally required. Care must be taken to preserve or repair soft tissue attachments as needed at the time of excision.

MEDIAL-SIDED FRACTURES Unlike lateral-sided fractures, medial-side injuries are likely to be extra-articular in nature. The medial epicondyle experiences a delayed fusion with the metaphysis, not occurring until between the ages of 14 and 17 years. Conversely, the trochlea, lateral condyle, and lateral epicondyle become fused at 10 or 12 years of age.35 As such, the

medial epicondyle is the most commonly fractured aspect of the medial elbow. Intra-articular injuries of the medial condyle and trochlea are extremely rare in all pediatric ­populations.

Medial Epicondyle Fracture Medial epicondyle fractures encompass about 11.5% of all pediatric elbow injuries, and these fractures occur most often in pediatric patients between the ages of 10 and 14 years.14,112 These are apophyseal injuries and do not involve the true physis. Fractures of the medial epicondyle are also highly correlated with posterolateral elbow dislocations, with reported incidences ranging from 30% to 55% in published reports.112-114 The most common mechanisms of injury are indirect and arise from avulsion forces and overexertion of the immature elbow joint with excessive valgus stress. Direct injuries are rare.112 Indirect injuries commonly result from a fall on an outstretched arm in valgus or through avulsive forces supplied by either the flexor muscles of the forearm or the medial collateral ligament during a throwing motion.76 Chronic cases are frequently seen in skeletally immature overhead athletes.113,115,116 There is no well-established classification system. Medial epicondyle fractures present with tenderness and mild swelling of the medial elbow. Pain and tenderness often become worse when the patient is asked to flex the wrist or with resisted wrist flexion. There is usually some degree of decreased range of motion, which is either a secondary effect of pain or a direct result of an incarcerated fracture fragment. Suspicion of an incarcerated fragment must be considered seriously because many times unrecognized incarcerations are present. These undiagnosed fragments may develop chronic fibrous union ­ scarring to the coronoid process, resulting in a severe decrease in elbow range of motion.72,112,114 Valgus instability is also a ­ common feature and is best identified with the arm in mild flexion (15 degrees) and pronation to isolate the soft tissue of the medial side by unlocking the olecranon from its fossa and locking the radius on the capitellum.117 Any excessive swelling or gross instability of the elbow would suggest a greater injury such as an elbow dislocation or an intra-articular medial condyle fracture. A careful neurologic examination is also critical because of the proximity of the ulnar nerve. Ulnar nerve dysfunction has been reported in about 10% to 16% of all cases, and nearly 50% of cases involve some degree of incarceration.22 Subtle ulnar symptoms usually resolve on their own.118-120 Chronic cases are most likely to present with ulnar nerve symptoms and may affect outcomes.121,122 Proper radiographic evaluation is critical. A high index of suspicion is necessary for identifying incarcerated fracture fragments in the setting of an elbow dislocation. Incarceration may occur in up to 20% of cases with a concomitant elbow dislocation.113,114 It is critical that this is identified because the incarcerated fragment may hinder the process of reduction.123 A posterior fat pad sign is possible and would also suggest a concomitant intra-articular injury. Historically, nonsurgical management was the mainstay of treatment. For nondisplaced or minimally displaced fractures with normal neurologic examinations, treatment consisted of cast immobilization.124 Presently, displacement

1286 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

i­ncarcerated fragment or ulnar nerve in the setting of an elbow dislocation.14

Medial Condyle Fractures

A

B

Figure 19F2-5   A, Injury radiograph of displaced medial epicondyle fracture in an overhead athlete. B, Postoperative radiograph after open reduction and internal fixation of medial epicondyle fracture.

is defined as less than 5 mm in the general population and more than 2 mm in competitive overhead athletes.116,125-129 In these cases, the elbow is casted in neutral to slight forearm pronation. For both populations, early motion is begun upon cast removal to prevent stiffness.20,118 If significant displacement exists, treatment is much more controversial. Historically, patients with more than 5 mm of displacement did well with conservative treatment.113,114,118,119,124,130-132 Conservative treatment usually yielded a fibrous union, which was well tolerated by most individuals. If these fibrous unions became symptomatic, then surgical excision was also possible. Although conservative treatment is a reasonable option in certain patients, recent literature has promoted more aggressive operative interventions to achieve osseous union and lower the risk for future complications.14,116,125-129 Fibrous unions, especially in the overhead athlete, have shown to be more symptomatic than originally described, with compromised flexor function, valgus instability, and loss of motion.14,116,125-129 At present, overhead throwing athletes are considered separately from the general population because of the excessive stress they place on the medial epicondyle. To ensure complete union, open reduction with internal fixation is now considered the treatment of choice in these patients even with ­ minimal ­ displacement (>2 mm). Within both populations, an ­ incarcerated ­ fragment warrants ­ surgical intervention. Common surgical technique for medial epicondyle fractures uses a small longitudinal posteromedial incision behind the medial epicondyle and rigid internal fixation (Fig. 19F2-5). Washers may also be used to protect the delicate apophysis bone from injury.112,116 After only a brief period of immobilization, early range of motion is stressed to prevent any loss of motion. The relative indications for open reduction with internal fixation of medial ­epicondylar fractures include more than 2 mm of ­ displacement or ­valgus instability in competitive athletes such as pitchers and gymnasts, whereas absolute ­ indications include

Although rare, medial condyle fractures generate 1% to 2% of all pediatric elbow injuries. These injuries most commonly occur in children between 8 and 12 years of age and are labeled as Salter-Harris IV fractures.133 In some rare cases, medial epicondyle fractures can have an intraarticular extension. These patients are slightly younger than patients with medial epicondylar fractures and have an unossified trochlear physis. They usually have a large hemarthrosis and a small metaphyseal fragment attached to the epicondyle fragment. Common treatment methods include immobilization of nondisplaced or minimally ­displaced fractures. For fractures with high degrees of displacement, open reduction is suggested, with care taken to avoid the ulnar nerve. Fixation by Kirschner wires in parallel position is the most common method to ensure that reduction remains permanent.133 The elbow should remain immobilized until radiographic evidence shows that the fracture has achieved union, at which point active range of motion therapy should begin.

Authors’ Preferred Method Our patient population has a high incidence of overhead athletes, so our indications for surgery for medial epicondyle fractures include patients with as little as 2 mm of displacement. We use a small longitudinal posteromedial incision behind the medial epicondyle, reduce the fragment as necessary, and achieve fixation using one or two small lag screws with or without a washer. Medial condyle fractures with a high degree of displacement are treated with open reduction to identify and ­protect the ulnar nerve. Medially based Kirschner wires or small cannulated screws are used for fixation in parallel or slightly divergent construct.

FRACTURES OF THE PROXIMAL FOREARM Olecranon Fractures Pediatric olecranon injuries are rare, accounting for about 4% to 6% of all pediatric elbow fractures.134,135 The fusion of the apophysis and metaphyseal elements of the ­olecranon occurs at about the age of 14 years, and most olecranon fractures occur before this time, but they can happen at almost any age.35 The most commonly described mechanisms of injury include falls on an outstretched hand, twisting injuries, and direct blows to the posterior elbow. Olecranon fractures have been associated with other elbow fractures in 20% of cases, including radial neck fractures, medial epicondylar fractures, coronoid fractures, and osteochondral injuries.4,136-138 Classification of olecranon fractures begins by distinguishing between apophyseal and metaphyseal fractures. Injuries to the apophyseal region can present as several

Elbow and Forearm 1287

injuries, including apophysitis, stress fracture, and complete fracture. Any suspicion of fracture to this region should be confirmed by a radiograph of the contralateral elbow or an MRI scan of the affected elbow. Avulsive forces causing a complete fracture in this region are rare. Metaphyseal fractures of the olecranon can be categorized as greenstick injuries or complete fractures. Fracture due to an avulsive force of the triceps is rare because the triceps insert in the ulna distal to the metaphysis.139,140 Most injuries here are from direct blows, and different fracture patterns have been identified for flexion, extension, and shear injury mechanisms.141 Metaphyseal fractures are more common than apophyseal fractures and must be carefully examined to rule out Monteggia’s injuries. Most apophyseal fractures are nondisplaced, and these are treated through immobilization (usually in 20 degrees of flexion to relax the triceps) and early range of motion therapy.135,136,142 Any fractures that are intra-articular with severe displacement (2 to 4 mm) should be treated with open reduction with internal fixation. Tension band wiring using heavy suture and removable Kirschner wires is increasingly performed because it obviates additional surgery to remove hardware.135,143-145 Metaphyseal fractures are typically nondisplaced or minimally displaced fractures treated with immobilization. Displaced metaphyseal fractures (>5 mm of metaphyseal displacement or 2 mm of articular displacement) require open reduction with internal fixation with either tension bands or screw fixation.12,20,136,146 Early range of motion therapy motion is recommended. Complications for olecranon fractures include transient ulnar neuropathy, decreased range of motion and fracture displacement being documented in a small number of cases.135,147

Monteggia’s Fractures Monteggia’s fractures are relatively rare but challenging injuries. These fractures consist of a proximal fracture of the ulna followed by dislocation of the radiocapitellar joint. They are low-energy injuries most frequently seen in children between the ages of 7 and 10 years.35 The most common causes of fracture are direct blows to the ­posterior aspect of the proximal forearm or by a fall on an outstretched arm. These fractures are usually classified by the adult Bado criteria by the position of the radial head after dislocation.148 Type I fractures are anteriorly dislocated, type II fractures are posteriorly dislocated, type III fractures are laterally dislocated with varus malalignment of the proximal ulna, and type IV fractures involve an anterior dislocation with a fracture of the proximal radius. Type I injuries are most commonly seen, accounting for more than 70% of all cases.22 Common pediatric ­ Monteggia’s fracture variants include radial head ­ dislocations with ­plastic deformation of the ulna and incomplete (greenstick) or complete ulnar shaft fractures. A high index of suspicion is paramount for any radial head dislocation. Careful inspection of the radial head and ulna is necessary for treatment. Unlike an acute dislocation, a chronic congenital dislocation has a hypoplastic capitellum and a small convex radial head. Assuming an acute Monteggia’s injury, the mainstay of treatment depends on the ulna injury.149 With incomplete or plastic deformity of the ulna, the radial head should be closed reduced and then

followed very closely postoperatively to show no loss of reduction. These injury patterns are commonly missed initially and then present as chronic cases.150 Late reconstruction of these are difficult and require ulnar osteotomies, open reductions, annular ligament reconstruction, and either plating or external fixator application.151 If the ulna injury is a complete fracture, then the radial head should be closed reduced, followed by internal fixation of the ulna with Kirschner wires, flexible nails, or plates.152,153 If the ulnar fracture or radiocapitellar dislocation is irreducible, open reduction with internal fixation is suggested.154 In all cases, immobilization for 4 to 6 weeks is suggested. During this time, frequent radiographs should be taken to ensure stability of the radial head. Complications involved with this fracture pattern include radial nerve injuries, fractures of the distal forearm, and missed or delayed diagnosis. Delayed treatment necessitates open reduction, and these cases have a higher risk for complication.149,154

Radial Head and Neck Fractures Fractures to the proximal radius account for between 5% and 9% of all pediatric elbow fractures.14,22 These injuries include fractures through the physis (Salter-Harris II) in younger children or metaphyseal fractures in ­ children between 8 and 12 years of age.8,155,156 Intra-­articular radial head fractures can also occur in adolescents. For the most part, radial head fractures occur in older patients, whereas on average, radial neck fractures occur in younger patients.157 Two common mechanisms of radial neck fracture are falls on a hyperextended arm and chronic compressive valgus forces seen in overhead athletes. Chronic throwing injuries of the proximal radius are much less common than those of the medial epicondyle mentioned previously.158 All fractures are classified by the O’Brien scale, which rates injuries by the degree of angular deformity of the radial neck.8,14,155,156 In an uninjured child, the normal angle between the head and neck is usually between 0 and 15 degrees of valgus. This angle is measured by drawing one line directly through the metaphysis of the radial neck and another perpendicular to the epiphysis of the radial head. In general, type I injuries are between 0 and 30 degrees, type II fractures are between 30 and 60 degrees, and type III fractures are greater than 60 degrees.159 Patients with radial neck injuries usually present with tenderness and swelling around the lateral elbow. They frequently report an increase in pain with forearm rotation. Treatment depends on degree of displacement and the patient’s age. Type I fractures are usually immobilized for only 7 to 10 days, whereas type II and III fractures need a formal reduction consisting of either (1) extension, varus stress, and direct pressure; or (2) flexion, forearm rotation, and direct manual pressure.149,160-165 If the fracture is still significantly angulated and the patient has limited forearm rotation, percutaneous Kirschner wires can be used as a joystick, or the intramedullary method can be performed to gain a satisfactory reduction.158,163,165,166 Occasionally, these fractures are still unstable and require placement of Kirschner wires. Controversy exists on how to define an acceptable reduction. By strict definition, a reduction is successful when there is less than 30 degrees of ­angulation

1288 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

and 2 mm of translation,76,167 but ultimately a painless full arc of motion, with complete rotation and joint alignment, is most significant.76,168 In rare circumstances, irreducible fractures may require an open reduction with minimal internal fixation. The posterolateral Kocher approach is recommended, with care taken to avoid the posterior interosseous nerve and minimal dissection of the annular ligament.158,164,169-171 Kirschner wires are inserted obliquely, whereas transcapitellar fixation is not recommended.158,160,170 In adolescents, radial head fractures may also occur and should be treated by adult guidelines with open reductions. Radial head excision should almost never be performed in skeletally immature patients. About 15% to 23% of all radial head or neck fractures will achieve poor results regardless of the quality of care. Common complications after injury include decreased motion (usually forearm rotation), radial head overgrowth (20% to 40%), avascular necrosis (10% to 20%), and myositis ossificans (>30%).158 Numerous studies have shown that closed reduction attains more desirable outcomes, whereas open reductions have the highest rate of complication.164,168,172,173 Adequate reduction, good surgical technique, and immediate treatment are predictors of a favorable outcome.158

Authors’ Preferred Method Displaced and intra-articular olecranon fractures are t­ reated with open reduction and internal fixation. The ­specific construct used depends on the skeletal maturity and the specific fracture pattern. In general, minimally comminuted fractures are regularly treated with a tension band construct using heavy suture and removable Kirsch­ner wires. However, in fractures in which there is ­ significant fragmentation or significant intra-articular involvement, we prefer rigid fixation with plate and screws. Monteggia’s fractures are carefully assessed for radial head injury, and open reduction of the radial head is ­occasionally required. Ulnar fractures are typically treated with open reduction and rigid fixation with plates and screws; however, the location of the fracture may require Kirschner wire or flexible nail fixation. Type II and III radial neck fractures are treated with ­percutaneous Kirschner wire fixation. We typically use a Kirschner wire as a joystick to obtain a satisfactory ­reduction, then additional Kirschner wires are inserted obliquely to achieve fixation as needed. If an open reduction is required, it is performed through a posterolateral ­Kocher approach. Care is taken to avoid the posterior interosseous nerve and minimal dissection of the annular ligament is desired.

PEDIATRIC ELBOW DISLOCATIONS Elbow dislocations are uncommon in children, ­appearing in only 1% to 3% of all elbow injuries.174 Most elbow dislocations are due to hyperextensions, and they most commonly present with a posterolateral dislocation.174 Associated fractures of the medial epicondyle are most common,

whereas fractures of the olecranon, radius, trochlea, and lateral epicondyle have also been noted.112-114,161,174-176 Other injuries seen with elbow dislocations include disruption of the anterior capsule, tearing of the brachialis muscle, incarcerated fractures of the medial ­ epicondyle, and tearing of the ulnar collateral ligament.174,175 Rare impairments of the brachial artery and median and ulnar nerves have been reported.177-179 Most dislocations are treated through closed reduction.180 A sling should be worn for 7 to 10 days, and passive motion encouraged. Upon removal of the sling, aggressive range of motion therapy should begin.180 C

r i t i c a l

P

o i n t s

l

 he pediatric elbow has a predictable course of physeal T development and closure, which is critical to understanding pediatric elbow trauma. l Radiographs have to be analyzed systematically. l Supracondylar humerus fractures are the most common injury pattern and can be associated with neurovascular injuries. l Lateral condyle fractures are subtle and need a high index of suspicion to diagnose. l Medial epicondyle fractures are associated with elbow dislocations in up to 50% of cases and may need operative fixation in the throwing athlete. l Proximal ulna fractures must be carefully assessed to properly diagnose Monteggia’s fractures and their variants. l Displaced radial neck fractures have good results with percutaneous operative techniques. l Pediatric elbow dislocations are rare except in the adolescent population; in younger patients, the radiographic appearance of an elbow dislocation is more likely to be a transphyseal fracture.

S U G G E S T E D

R E A D I N G S

Bhandari M, Tornetta P, Swiontkowksi MF: Displaced lateral condyle fractures of the distal humerus. J Orthop Trauma 17(4):306-308, 2003. Campbell CC, Waters PM, Emans JB, et al: Neurovascular injury and displacement in type III supracondylar humerus fractures. J Pediatr Orthop 15(1):47-52, 1995. Case SL, Hennrikus WL: Surgical treatment of displaced medial epicondyle fractures in adolescent athletes. Am J Sports Med 25(5):682-686, 1997. Gordon JE, Patton CM, Luhmann SJ, et al: Fracture stability after pinning of displaced supracondylar distal humerus fractures in children. J Pediatr Orthop 21(3):313-318, 2001. Gortzak Y, Mercado E, Atar D, Weisel Y: Pediatric olecranon fractures: Open reduction and internal fixation with removable Kirschner wires and absorbable sutures. J Pediatr Orthop 26(1):39-42, 2006. Lee HH, Shen H-C, Chang J-H, et al: Operative treatment of displaced medial epicondyle fractures in children and adolescents. J Shoulder Elbow Surg 14(2): 178-185, 2005. Radomisli TE, Rosen AL: Controversies regarding radial neck fractures in children. Clin Orthop 353:30-39, 1998. Re PR, Waters PM, Hresko T: T-condylar fractures of the distal humerus in children and adolescents. J Pediatr Orthop 19(3):313-318, 1999. Skaggs DL, Cluck MW, Mostoti A, et al: Lateral-entry pin fixation in the management of supracondylar fractures in children. J Bone Joint Surg Am 86(4):702-707, 2004. Sullivan JA: Fractures of the lateral condyle of the humerus. J Am Acad Orthop Surg 14(1):58-62, 2006.

R eferences Please

see www.expertconsult.com

Elbow and Forearm 1289

S e c t i o n

G

Heterotopic Bone Around the Elbow Mir Haroon Ali and Scott P. Steinmann

Heterotopic ossification can be defined as the formation of abnormal mature lamellar bone in soft tissues. This abnormal bone may or may not contain bone marrow. Heterotopic ossification was first identified in 1883 by Reidel, a German physician.1 It was also subsequently described during World War I by Dejerine and Ceillier, two French physicians.2 During the last century, this entity has been given many names, including paraosteoarthropathy, myositis ossificans, periarticular ectopic ossification, periarticular new bone formation, neurogenic osteoma, neurogenic ossifying fibromyopathy, and heterotopic calcification. The lack of consistent nomenclature has made understanding of this problem difficult and made comparison studies nearly impossible. Recently, most investigators have agreed on the classification of abnormal bone formation into three major categories. The first two conditions fall under the heading of ectopic ossification, referring to heterotopic ossification when abnormal bone forms in soft tissues and myositis ossificans when ectopic bone forms within muscles, where it seldom causes problems of clinical significance. The third category, termed periarticular calcification, refers to the presence of calcium phosphate crystal collections visualized on plain radiographs. These amorphous crystal depositions do not demonstrate trabecular organization and are found in specific locations about the elbow, such as within the collateral ligaments and the within the joint capsule. It is thought to result from alterations in pH and ion concentrations about a healing elbow joint. It rarely carries clinical significance. Heterotopic ossification is of clinical significance in the elbow, where it can be a source of pain and limited mobility. Heterotopic ossification can result from various local or systemic insults. Although the most common cause is direct elbow trauma, cerebrospinal trauma, burns, and genetic disorders also have been associated with heterotopic ossification about the elbow. Although most heterotopic ossification is clinically asymptomatic, it can cause severe elbow stiffness, even ankylosis. This stiffness may cause significant disability in the performance of one’s daily activities. Morrey and colleagues demonstrated that a 100-degree flexion arc is needed for to perform 90% of daily activities (usually from 30 degrees of flexion to 130 degrees flexion), and a 100-degree rotation arc is needed for most daily activities (usually 50 degrees of ­supination and 50 degrees of pronation).3 When these arcs cannot be achieved because of heterotopic ossification about the elbow joint, patients often present for intervention and treatment. Patients may also present with nerve

c­ ompression as a result of ectopic bone formation; the ulnar nerve is most commonly affected, but median and radial nerve palsies have been reported. Optimal treatment of these patients by an orthopaedic surgeon requires knowledge of the pathophysiology, natural history, presentation, and treatment of heterotopic ossification.

PATHOPHYSIOLOGY The exact mechanisms responsible for the formation of heterotopic bone are unclear. Heterotopic ossification is thought to result from inappropriate differentiation of pluripotent mesenchymal stem cells into osteoblastic stem cells. Urist and colleagues demonstrated that demineralized bone matrix induced ectopic bone formation when implanted in the muscles of animals; the authors hypothesized that “bone morphogenic proteins” present in muscle were responsible for this transformation of mesenchymal cells to osteoblastic cells.4 Chalmers and associates demonstrated that whereas implantation of demineralized bone matrix in muscle and fascia induced bone formation, implantation into other abdominal organs did not result in bone formation.5 These results suggested that, in addition to the presence of pluripotent stem cells, inducing agents and a permissive environment are required for formation of heterotopic bone. Local inducing agents implicated in the development of heterotopic ossification have been better characterized during the past 10 years. Bone morphogenic proteins (BMPs) have been implicated in fracture healing and endochondral osteogenesis and may be involved in heterotopic ossification. Members of the transforming growth factor-β family, BMPs have been shown to facilitate osteogenic chemotaxis, mitosis, and differentiation. These proteins have been shown to induce heterotopic ossification in multiple in vitro and in vivo studies. Moreover, BMP inhibition has been shown to decrease heterotopic ossification in a dose-dependent manner in established animal models. BMP-4 has been the most extensively studied in models of heterotopic ossification. The identity of systemic factors regulating bone formation has remained elusive during the past 15 years. Many patients with head injury have increased rates of fracture healing and commonly develop heterotopic ossification despite no history of direct trauma to the elbow joint. These observations imply a circulating factor that predisposes these patients to faster bone healing and ectopic bone

1290 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

formation. One proposed factor is prostaglandin E2, which has been supported by evidence that prostaglandin inhibitors such as indomethacin significantly reduce or prevent heterotopic ossification.

ANATOMY Most commonly, heterotopic ossification occurs in the posterolateral aspect of the elbow. Typically, ectopic bone extends from the lateral humeral condyle to the posterolateral olecranon, and ectopic bone fills the olecranon fossa, causing loss of terminal extension. The radial and ulnar collateral ligaments are also affected, with ectopic ossification within the ligaments rarely causing limitation of motion. However, ectopic bone in the vicinity of the medial collateral ligament may result in delayed ulnar nerve palsy. If the coronoid is enlarged with ectopic ossification, this is likely to restrict elbow flexion because of impingement at the coronoid fossa. This limitation of motion usually is accompanied by an anterior capsular contracture. Anterior heterotopic ossification may extend from the humerus to the radius and ulna to the level of the bicipital tuberosity, thus locking the elbow at 90 degrees of flexion and possibly causing median or radial nerve palsy. A concomitant radioulnar synostosis may be present, preventing pronation and supination. Extensive reports have documented a wide spectrum of anterior heterotopic ossification, ranging from involving the brachialis, biceps brachii and tendon, and anterior capsule. It is clear from these reports that the ectopic ossification can ignore anatomic fascial planes completely and present in a variety of ­locations.

HISTORY AND PRESENTATION In most cases, heterotopic ossification of the elbow is benign and does not cause clinical problems. However, many patients present to an orthopaedic surgeon because of heterotopic ossification causing major difficulty in their ability to perform activities of daily living. These difficulties appear amplified when the dominant extremity is affected. Although posttraumatic elbow ossification typically begins 2 weeks after trauma, surgery, burn, or head injury, the processes that initiate ectopic ossification likely start immediately after the injury. Patients present with pain, ­ swelling, induration of subcutaneous tissues, and occasional warmth and redness usually within 1 to 4 months after the insult. However, the most bothersome to the patient is the progressive loss of motion at the elbow. The patient will complain of difficulty with motion required to maintain personal hygiene (both flexion and extension) and inability to compensate for loss of forearm rotation. Because of the anatomic precision of the radiocapitellar joint required for fluid forearm rotation, many patients complain of limited forearm rotation before limitation in flexion-extension at the ulnohumeral joint. Occupational limitations may also uncover limitations in terminal extension or decreased forearm rotation. Although rare, patients have also presented with nerve compression symptoms as a result of ectopic ossification, so a history of sensory loss or motor weakness must be elicited.

RISK FACTORS AND ASSOCIATIONS Heterotopic ossification is associated with a number of disease states and risk factors. Studies have demonstrated genetic predisposition, local trauma, mechanical factors, neurologic abnormalities, and systemic conditions contributing to heterotopic ossification. In vivo, multifactorial causes most likely increase the risk for formation of ectopic bone. Evidence implicating a genetic role in the formation of heterotopic ossification has been provided by the study of fibrodysplasia ossificans progressiva and primary osteoma cutis. These autosomal dominant connective tissue disorders are characterized by diffuse heterotopic ossification. Shafritz and associates demonstrated that patients with fibrodysplasia ossificans progressiva overexpress BMP-4, suggesting that genetic overexpression of this protein may be causing extra bone formation.6 Local tissue trauma sustained in an injury can increase the likelihood of heterotopic ossification. In general, the severity of injury correlates positively with the likely degree of heterotopic ossification. Although severe acute trauma is most often associated with heterotopic ossification (Fig. 19G-1), case reports have shown that repetitive microtrauma can also lead to heterotopic ossification, as observed in riflers, fencers, equestrians, and dancers. Similarly, the degree of surgery and the presence of surgical complications correlate positively with extent of heterotopic ossification. Studies have demonstrated that surgical approach to the elbow, operative time, extensive dissection, dissemination of bone debris and dust, and hematoma formation all increase the likelihood of postoperative heterotopic ossification. However, these generalizations for both injury and surgery are not always true. Case reports have shown that even a relatively mild procedure such as elbow arthroscopy (Fig. 19G-2) or distal biceps repair (Fig. 19G-3) with minimal tissue injury can result in postoperative heterotopic ossification.

Figure 19G-1   Heterotopic ossification in a 35-year-old man 2 years after open reduction of proximal ulna fracture and radial head fracture (Bado IV Monteggia fracture-dislocation).

Elbow and Forearm 1291

Although the mechanisms are not clear, neurologic injury and thermal injury predispose patients to heterotopic ossification. Studies have demonstrated that muscle denervation alone does not lead to heterotopic ossification, but spinal cord or brain injury results in robust ectopic bone formation. Serum from these patients increases osteoblast stimulation in a dose-dependent manner, suggesting the release of a humoral mediator of bone formation in these patients. In contrast, this hypothesis is questioned owing to the observation that heterotopic ossification occurs only below the level of spinal cord injury or on the hemiplegic or spastic side in the case of brain injury. Thermal injury also increases the likelihood of heterotopic ossification and has been most commonly found to involve the elbow. The reasons for this are unclear; it may be due to immobilization, recumbency, passive stretching, or a humoral factor released by the thermal injury itself.

PHYSICAL EXAMINATION A thorough physical examination is required of patients undergoing evaluation for suspected complications from heterotopic ossification. Early in the course of ectopic elbow ossification, the elbow range of motion (ROM) may be preserved, but the patient may complain of redness, tenderness, swelling, and subjective stiffness of the joint, particularly in the area of the heterotopic bone.

­ uring this early period, static and dynamic splinting may D be helpful in preserving range of motion. However, some patients will lose ROM despite aggressive physical therapy and ­splinting. As the ROM decreases and firm end points to motion are established, the patient’s other complaints, such as tenderness and erythema, usually diminish. At this time, the ectopic bone may be palpable to the examiner. Once the heterotopic ossification has matured (usually 3 to 9 months after the injury), elbow ROM remains stable. When examining the patient throughout the course of the disease, it is important to document the active and passive range of motion for flexion, extension, pronation, and supination. The examiner should also note if the end point in each of these planes is compliant (suggesting soft tissue restriction) or rigid (suggesting osseus restriction). A thorough neurologic examination should also be performed on each patient undergoing evaluation for heterotopic ossification because delayed nerve compression is another recognized complication. Most commonly, the ulnar nerve is affected, but median and radial nerve palsies have been reported. These palsies may occur months or years after the inciting injury. Two-point sensation, Semmes-Weinstein sensation, and motor strength testing and grading should be performed on each patient on initial evaluation, and changes from the previous examination should be noted. Patients suffering neurologic injury with residual cognitive or physical impairments should be

B

A

C

Figure 19G-2   Heterotopic ossification after elbow arthroscopy in a 58-year-old man.   A, Two months after surgery. B, Six months after surgery. C, Ectopic bone formation in posterior elbow removed at surgery.

1292 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

t­ horoughly evaluated by a physiatrist and a neurologist before any surgical intervention to ensure compliance with postoperative rehabilitation. Garland and colleagues reported that successful results of ectopic ossification resection correlated with level of neurologic recovery after head injury.7

DIAGNOSTIC TESTING Laboratory evaluation—some authors have advocated monitoring serum alkaline phosphatase values to determine the maturation of heterotopic ossification for timing surgical intervention. However, recent data have demonstrated that this is not necessary nor reliable. Although the levels of serum alkaline phosphatase appear to increase during the first month following an inciting injury and peak about 12 weeks after injury, these levels are not different in patients with and patients without heterotopic ossification. Moreover, Garland and colleagues recently reported that alkaline phosphatase levels did not correlate with recurrent heterotopic ossification after resection.8 Most clinicians have had successful results treating heterotopic ossification

A

C

without using serum alkaline phosphatase, and thus these laboratory tests are not routinely indicated.

Radiologic Imaging Radiographs can show the general location of the ectop­ic bone and are useful is assessing joint congruity and fracture healing. Radiographs may show ectopic ossification as early as 2 weeks. In addition to plain radiographs of the elbow (anteroposterior and lateral), further advanced imaging is often helpful in determining the extent and location of heterotopic ossification. Computed tomography (CT) with two- and three-dimensional reconstructions is useful in determining if ectopic bone is likely responsible for the patient’s limitation in motion. The increased anatomic understanding of the ectopic bone afforded by these images also reveals ectopic bone around the medial and lateral margins of the ulnohumeral joint and allows better visualization of ectopic bone formation in the proximal radioulnar joint that may not be appreciated on radiographs. This additional information helps surgical planning by helping determine the optimal ­surgical approach

B

D

Figure 19G-3   Heterotopic ossification in a 38-year-old man after repair of distal biceps rupture. A, Lateral radiographs.   B, Computed tomographic scan with three-dimensional reconstruction. C, Radial tuberosity seen at surgery. D, Excised bone fragments.

Elbow and Forearm 1293

and techniques, thus decreasing surgical time and rate of complications.

Other Imaging Modalities Other imaging modalities may be used to visualize heterotopic ossification about the elbow but are not commonly indicated. Ultrasound imaging has been shown to be useful for early diagnosis of heterotopic ossification; studies have shown that ultrasound may allow visualization of ectopic bone before radiographic presentation. However, because this early diagnosis does not usually alter the treatment course, this modality is rarely used. Bone scans can also detect heterotopic bone formation earlier than plain radiographs. A three-phase bone scan, with vascular and blood-pooling images, may detect circulation abnormalities as early as 2 weeks after the inciting injury, several weeks before corresponding calcification is visualized on plain films. Normally, bone scan changes return to baseline activity levels in 7 to 12 months, indicating maturation of the circulatory milieu and stabilization of the heterotopic bone mass. In some cases of ectopic bone formation, the increased activity may be prolonged. Recent studies have also demonstrated a decreased risk for recurrence after resection even if the bone scan demonstrates increased activity. Bone scans thus are not used commonly in the evaluation and treatment of heterotopic bone formation about the elbow because they do not alter the surgical treatment or timing of intervention. Finally, magnetic resonance imaging (MRI) is seldom used in the evaluation of heterotopic ossification because of its inferior imaging of mineralized tissues. It may be useful in evaluating calcification within soft tissues, such as myositis ossificans. However, this ectopic bone is usually of little functional significance and rarely requires treatment. MRI may also be helpful when prior trauma or anomalous anatomy necessitates more understanding of the relationship between the ectopic ossification and the neurovascular structures in the elbow.

DIFFERENTIAL DIAGNOSIS The presence of heterotopic bone around the elbow does not always explain a patient’s loss of motion. Before offering surgical resection of heterotopic bone, other causes of elbow stiffness must be evaluated and excluded. These include infection, osteochondral lesions, thrombophlebitis, complex regional pain syndrome, malunion, synovitis, myositis, and rare fibrodysplastic syndromes such as fibrodysplasia ossificans progressiva. Although elbow trauma is the most common cause of elbow stiffness, it may result in limitation of motion in other ways besides heterotopic ossification. One of the most common causes of post­traumatic stiffness is capsular contracture secondary to capsular thickening. This is likely due to increased chronic synovial inflammation after an injury and subsequent rehabilitation. Second, early post-traumatic arthritis with the formation of marginal osteophytes may also limit elbow motion independent of heterotopic ossification. Usually, these osteophytes are most symptomatic in the posterior ulnohumeral joint and limit terminal extension. If present, both of these problems need to be addressed in addition to surgical treatment of heterotopic ossification.

Inherited musculoskeletal disorders, although rare, can also cause elbow stiffness and pain. The best-studied of these is fibrodysplasia ossificans progressiva. It is an autosomal dominant disorder caused by defective induction of endochondral osteogenesis. This results in the deposition of abnormal bone in periarticular soft tissues. More than 90% of patients have ectopic calcification in soft tissues by the age of 15 years. A family history and a history of exuberant soft tissue calcification after a previous injury suggest this diagnosis. Operative intervention in these patients is met with disastrous results, with massive amounts of heterotopic ossification after surgery. It is imperative to exclude this disorder before any operative resection of ectopic bone.

CLASSIFICATION Anatomic Several anatomic classification systems have been developed for heterotopic ossification in the upper extremity. Vince and Miller’s classification system focused on the forearm and divided ectopic bone formation into three types based on the anatomic location of the synostosis.9 Type I involved the distal third of the forearm, type II involved the middle third of the forearm, and type III contained the proximal third of the forearm. Ring and Jupiter further classified type III with respect to the proximal radioulnar joint.10 Type IIIA is located at or distal to the bicipital tuberosity, type IIIB involves the radial head or proximal radioulnar joint, and type IIIC is a proximal radioulnar synostosis with contiguous bone extending across the elbow joint and onto the distal humerus. Viola and Hastings have developed a more detailed classification system for heterotopic ossification about the elbow that allows for more precise surgical planning.1 By categorizing the location of the ectopic bone more systematically (proximal or distal, anterior or posterior), a surgical approach can be selected more easily. Type I refers to ectopic ossification of the proximal radioulnar joint, type II refers to ectopic ossification of the proximal radioulnar joint with distal extension to the level of the bicipital tuberosity, and type III refers to heterotopic ossification between the radius and ulna distal to the proximal radioulnar joint. A subclassification then denotes anterior or posterior heterotopic ossification: subtype A signifies anterior involvement, subtype B represents posterior involvement, and subtype C denotes intra-articular involvement. Although these types and subtypes are not mutually exclusive, the anatomic classification predicts which surgical approach or approaches will be needed to surgically efficiently resect the ectopic bone.

Functional Hastings and Graham have proposed a classification system for heterotopic ossification about the elbow based on functional range of motion.11 Class I patients present with radiologically evident heterotopic ossification without any functional limitation. Most cases of class I heterotopic ossification are clinically insignificant, and it is usually an incidental finding. It should, however, be recorded and noted

1294 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

because it demonstrates the patient’s tendency toward extra bone formation. This information should be documented to avoid complications with the patient’s future orthopaedic care. Class II patients have subtotal functional limitation of motion in flexion or extension, supination or pronation, or in both. Subcategories exist to denote which plane of motion is limited. Based on Morrey’s previous findings, the range of functional movement and limitation is determined by the examiner and is usually present if the patient cannot extend to 30 degrees short of full extension, flex to 130 degrees, pronate to 50 degrees, and supinate to 50 degrees. Patients with class IIA have limitation in flexionextension, patients with class IIB diagnosis have loss of pronation-supination, and patients with class IIC have limitation in both flexion-extension and pronation-supination. Class III patients have ankylosis that completely eliminates flexion-extension or pronation-supination, or both. This can be further subdivided into class IIIA when patients present with loss of flexion-extension, class IIIB when patients have complete loss of pronation-supination, and class IIIC when patients lack both planes of motion. This classification helps stratify the severity of the clinical problem and the extent of surgical treatment required, but does not ease surgical planning questions regarding which approach or approaches should be used to facilitate elbow motion.

PROPHYLAXIS Chemotherapeutics Patients who sustain an elbow injury and have significant risk factors for heterotopic ossification should receive one of the two available forms of prophylaxis against ectopic bone formation. These risk factors include neurologic injury, burns, diffuse skeletal hyperostosis, hypertrophic osteoarthritis, ankylosing spondylitis, Paget’s disease, and a history of ectopic ossification in other joints. The first treatment option is the use of chemotherapeutic agents to prevent heterotopic ossification. Diphosphonates have been shown to decrease the calcification of ectopic osteoid, but are not commonly used owing to the rebound ossification that occurs when the agent is discontinued. Several studies have shown no difference in the ultimate amount of heterotopic bone or in the range of elbow motion, pain, or function. Additionally, the risk for gastrointestinal complications and the rare but dramatic complication of ­maxillary-mandibular necrosis prevent their regular use in the prevention of formation of heterotopic bone. Nonsteroidal anti-inflammatory drugs (NSAIDs) are the standard chemotherapeutic agents used to prevent heterotopic ossification. Most of the studies supporting the use of nonsteroidal anti-inflammatory agents such as indomethacin have been focused on the hip and pelvis. Doubleblind placebo trials assessing the amount of heterotopic ossification about the hip after total hip arthroplasty have demonstrated the efficacy of NSAIDs in decreasing ec­topic bone formation. Indomethacin has been shown in multiple laboratory studies to prevent precursor cells from ­differentiating into osteoblastic cells and other studies have indicated that it may inhibit angiogenesis and ­ haversian

remodeling. Moreover, multiple studies have demonstrated decreased bone formation and fracture healing in laboratory animals. Similar results have been demonstrated in laboratory animals treated with cyclooxygenase-2 inhibitors. Both classes of agents are thought to act by inhibiting prostaglandin formation. The most commonly selected regimens are indomethacin, 25 mg by mouth 3 times daily, or indomethacin sustained-release, 75 mg by mouth once daily for 3 to 4 weeks. The optimal length of treatment is unclear, but most clinicians agree that treatment should be initiated within 5 days of injury or surgery, based on the findings of Schmidt.12 Failure of NSAID treatment is commonly due to two factors. The most common side effect of gastrointestinal discomfort is caused by gastritis or ulcer formation. This is due to the decreased prostaglandin synthesis and the subsequent decrease in the production of gastric alkaline mucus. The second likely reason for failure is the inability of the patient to comply with drug therapy, even with the advent of a sustained-release once-daily form of indomethacin. These limitations often force the clinician to consider alternative treatment options.

Radiation The second manner in which to prevent heterotopic ossification after elbow injury is to use low-dose external-beam radiation. Radiation is thought to prevent ectopic bone formation by preventing the differentiation and proliferation of precursor cells into osteoblastic cells by inhibiting DNA transcription in these precursor cells. Like studies supporting the use of NSAIDs, clinical studies in patients with total hip arthroplasty have demonstrated that this form of treatment inhibits ectopic bone formation postoperatively. Since Coventry and coworkers demonstrated that radiation therapy can be used to prevent heterotopic ossification in the hip in the early 1980s, multiple studies have been ­supportive of radiation treatment in the prevention of heterotopic ossification of the elbow.13 However, both of these studies did not have a matched control group, and thus the degree of absolute risk reduction with radiation treatment is unclear. Jupiter and Ring recently demonstrated that the baseline risk for recurrence after resection of proximal radioulnar synostosis is low because none of the patients in their series had recurrence despite no radiation or chemotherapeutic prophylaxis.14 Thus, because the baseline incidence of recurrent heterotopic ossification about the elbow is unclear, the effect of radiation treatment for prevention of recurrent heterotopic ossification about the elbow is also unclear. Most clinicians prescribe indomethacin for 3 to 4 weeks if the patient’s risk for postoperative heterotopic ossification is low or moderate. They resort to radiation treatment if the patient is deemed high risk. The single radiation dose of 700 cGy is ideally administered within 72 hours of elbow injury or surgery. The site is confirmed before radiation delivery by fluoroscopy simulation. These protocols are based on well-established models preventing the formation of ectopic bone in the hip following hip and pelvic surgery. The biggest drawback of using radiation therapy is the fear of secondary carcinogenesis. This fear persists despite several ­studies demonstrating that the doses used for the prevention of heterotopic ­ ossification have not been shown to cause tumors over the past

Elbow and Forearm 1295

50 years. Moreover, since the early 1980s when Coventry first reported the successful use of radiation to prevent ectopic bone formation, no case has been reported that a soft tissue or bone tumor resulted secondary to this treatment.13 Although radiation-induced sarcomas have not been shown to develop with radiation doses of less than 3000 cGy, pediatric vertebrae growth arrest has been demonstrated with 2000 cGy. These data indicate that although the standard 700-cGy dose may be safe for adults, the optimal safe dose for younger children may be less.

TREATMENT Nonoperative Management Before offering surgical intervention for a patient affected by stiffness due to heterotopic ossification about the elbow, a patient should be started on an aggressive motion program to prevent the progressive loss of elbow motion during the period of maturation of the ectopic bone. These programs consist of a combination of active exercises, passive motion exercises, continuous passive motion, dynamic splinting, and static splinting. The success of these programs is unclear. Splinting is commonly used to attempt to restore elbow range of motion. Spring-loaded dynamic splinting has been used to attempt to counteract flexion and extension contractures. Patients are instructed to wear these splints for approximately 6 hours each night. Alternatively, patients can be given a dynamic flexion splint and a dynamic extension splint and instructed to use them in an alternating fashion. Turnbuckle splints can be used to attempt to overcome rigid contractures by applying a static stretch to the soft tissues by tightening the turnbuckle. Once these modalities have been exhausted, if the patient is still having difficulty in daily activities owing to stiffness secondary to heterotopic ossification, surgical treatment can be considered.

in children who suffer a reversible neurologic injury; as their neurologic recovery continues, their ectopic ossification resolves in a similar manner. Surgical resection of heterotopic ossification can be challenging and requires technical expertise in order to successfully restore elbow motion. The most commonly applied treatment involves identification of the ectopic bone and removing it until normal elbow motion is restored. A capsulectomy is used for accompanying capsular contracture. A surgical release is considered adequate if the patient has flexion to 140 degrees and lacks less than 10 degrees of terminal extension, and has pronation to 80 degrees and supination to 80 degrees. Meticulous effort is made to preserve the collateral ligaments, and any calcification within these ligaments is ignored because this is rarely a factor in limiting motion. Surgical principles must be followed in order to optimize patient motion and satisfaction. These include preoperative and intraoperative attention to selection of the proper technique (arthroscopic versus open), proper approach (or portal placement in arthroscopy), to facilitate removal of all clinically significant ectopic bone transposition or decompression of any compressed nerves, ensuring smooth and clear joint surfaces at the coronoid and the olecranon fossae, and excision of joint capsule as needed. If these principles can be followed during the surgical procedure, a good outcome can be expected with minimal complications. The surgical approach to meeting the surgical goals needs to be individualized based on the location of the heterotopic ossification, need for nerve decompression, degree of elbow contracture, and patient’s previous incisions. For example, patients who require ulnar nerve decompression have traditionally required a medial approach, and patients with limitation of forearm rotation have required laterally based incisions (Fig. 19G-4). Recently, with the increased use of arthroscopy for removal of ectopic ossification, larger exposures and dissections have been increasingly

Operative Management If patients continue to have functional limitation in range of motion and significant pain with range of motion, surgical excision of heterotopic ossification can be considered. The timing of surgical intervention is controversial. A balance must be found between worsening soft tissue contracture, cartilage destruction, and maturation of ectopic bone. Conventional wisdom suggests waiting 1 to 2 years to allow for maturation of the ectopic bone and minimize risks for recurrent formation of heterotopic bone. Although this may be appropriate in special circumstances (e.g., brain injury or burns), recent data have demonstrated little risk for recurrence and good functional outcomes with earlier surgical resection. These early operative interventions are accompanied by adjunctive prophylaxis in the form of NSAIDs, radiation therapy, or an aggressive motion program. Most surgeons have now gone from avoiding surgical intervention for up to 2 years, to offering patients surgical resection of ectopic bone as early as 4 months after the initial inciting injury. In children, surgical intervention is still delayed because spontaneous resolution of heterotopic ossification has been reported. This is particularly evident

Figure 19G-4   Example of surgical approach for heterotopic ossification in lateral elbow and extensive heterotopic intraarticular ossification. The black line represents the lateral edge of the distal humerus, the green line represents Kocher’s interval between the extensor carpi ulnaris and the anconeus, and the yellow line represents Kaplan’s interval between the extensor digitorum communis, and the extensor carpi radialis longus. This approach can grant maximal access to the elbow laterally and into the joint with minimal risk for creating lateral elbow instability.

1296 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

avoided. The more minimal approaches have been able to meet the surgical goals with much less morbidity than the more extensile approaches in many cases.

Posterolateral Ectopic Ossification Arthroscopic excision of heterotopic ossification from the posterolateral elbow can be performed with proper surgical planning and portal placement. The ideal portal placement is about 1 cm distal and 1 cm anterior to the lateral epicondyle (over the radial head). A second lateral portal can be placed about 2 cm proximal to the lateral epicondyle along the humeral ridge. If the amount of ectopic bone is deemed too extensive for arthroscopic removal, an open approach may be used instead. A modified Kocher approach is ­usually used for this exposure. The skin incision is directly over the lateral epicondyle and extends to the lateral subcutaneous border of the ulna. The extensor digitorum communis is divided in line with its fibers to more directly access the radiocapitellar joint. Alternatively the extensor carpi ulnaris can be retracted anteriorly, and the triceps and anconeus retracted posteriorly. The surgeon must pay careful ­attention to preserving the lateral radial collateral ligament. One or two posterior portals can be established to remove ectopic bone from the posterior elbow. The most common portals are the posterior portal 2 to 3 cm proximal to the tip of the olecranon and a second portal 2 to 3 cm proximal to the medial epicondyle on the medial edge of the triceps. However, if the degree of heterotopic ossification warrants, an extensile excision may be necessary. The posterior elbow may be accessed through the same extensile Kocher approach by elevating the triceps tendon and its fat pad. With severe heterotopic posterior ossification, the triceps may need to be detached from the olecranon to access the ectopic bone and resect it. Once the excess bone is removed and the congruity of the distal humerus, olecranon fossa, and olecranon is established, range of motion should be checked to ensure that smooth flexion and extension are established. If there is no osseous obstruction and full motion cannot be obtained, an anterior capsular release may be indicated. Once the posterior resection is completed, the triceps is reattached to the olecranon using drill holes if it was detached. Upon completion of the procedure, the integrity of the lateral radial collateral ligament and the ulnar collateral ligament must be tested and repaired or reconstructed if necessary.

Anterolateral Ectopic Ossification Usually, anterolateral ectopic bone forms a bridge from the anterolateral humerus to the radial head or the coronoid. During resection of ectopic bone from this area, the surgeon must be careful to protect and decompress the radial and posterior interosseus nerves. Although the previously mentioned lateral arthroscopic portals can be used to remove ectopic bone from the anterior elbow, large amounts of bone may require an extensile approach. Although a modified Kocher approach can be used, the anterior portion of the common extensor group may need to be released to expose the radial nerve. The supinator may also need to be decompressed to expose the posterior interosseus nerve. The radial nerve is then released from

any scar proximally and distally. After all the ectopic bone is resected in this area, gains in flexion and extension are checked. Alternatively, some surgeons may use a traditional Henry approach to the proximal forearm and elbow. If full flexion and extension are not possible after all ectopic bone is removed from the anterior elbow, a posterior capsular release may need to be performed.

Intra-articular Ossification Heterotopic ossification limited to the intra-articular space should be addressed arthroscopically. Most often, this intra-articular bone forms after a fracture-dislocation of the elbow. Commonly, a clear plane is visible between the ectopic bone and the articular surfaces of the capitellum and radial head. If the resection cannot be performed with the lateral and medial arthroscopic approach, a Kocher approach can be used. Once the extensor origin is elevated from the distal humerus and the anconeus and triceps are elevated posteriorly, the ectopic bone can be removed, usually without violating the lateral ulnar collateral ­ligament and the orbicularis ligaments. After removing all the ec­topic bone and restoring full motion, the lateral collateral ligament and extensor origin are repaired with drill holes through the lateral epicondyle as needed.

Posteromedial Ectopic Ossification Medial ectopic ossification can be complicated owing to the proximity of the bone and the required incision to the ulnar nerve and medial collateral ligament. Axial CT or MRI for surgical planning is helpful in determining the extent of heterotopic ossification and the relative position of the ulnar nerve. Most cases of heterotopic ossification from the medial elbow require ulnar nerve transposition or arthroscopic decompression. Arthroscopically, two lateral portals or one lateral and one medial portal can be established to remove ectopic bone from the medial aspect of the elbow. These allow for bone resection and ulnar nerve decompression in most cases. However, extensive heterotopic ossification may not be amenable to arthroscopic intervention, and an extensile approach may be indicated. The skin incision extends from the proximal medial epicondyle to the proximal ulna between the heads of the flexor carpi ulnaris. Anterior and posterior flaps are raised, and the ulnar nerve is transposed subcutaneously through its course around the elbow from the distal third of the humerus to the level of the flexor digitorum superficialis. Ectopic bone resection should focus on finding the plane of the ulnar-trochlear joint, while protecting the ulnar, lateral antebrachial, and medial brachial cutaneous nerves. Any posterior bone may be accessed by raising the triceps as similarly described earlier for the posterolateral approach. A posterior capsulectomy or proximal olecranon excision, or both (up to 1 cm), may be done if necessary.

Anteromedial Resection of Heterotopic Ossification Anteromedial débridement can be safely performed after identifying the median nerve and brachial artery. Although this can be done arthroscopically with one lateral and one

Elbow and Forearm 1297

medial portal, open techniques are commonly employed. The ligament of Struthers and pronator teres fascia are excised, and the neurovascular structures are identified and retracted anterolaterally. The flexor pronator origin is elevated off the anteromedial capsule and anterior band of the medial collateral ligament. The brachialis muscle is elevated and retracted laterally as needed to complete a thorough débridement. After completing the excision, the integrity of the medial collateral ligament must be tested; if incompetent or lax, it must be either repaired to the medial epicondyle with drill holes or reconstructed with tendon graft.

Heterotopic Ossification at the Proximal Radioulnar Joint With ankylosis or synostosis of the proximal radioulnar joint, the approach can vary based on the most extensive location of the ectopic bone. Although arthroscopic excision and débridement can be successful, most cases entail an open approach. The type of approach is dictated by the location of the heterotopic bone. Anterior-based synostosis can be approached with a Henry or a modified Kocher approach. In the Henry approach, the interval between the brachialis and the brachioradialis is developed, the recurrent radial vessels are cauterized, and the radial and posterior interosseus nerves are identified. These nerves are retracted radially, and the biceps tuberosity is found. The biceps tendon must be preserved throughout the process of débridement of the radioulnar and radiocapitellar joints. Occasionally, fat interposition graft may be placed in between the bicipital tuberosity and the lateral ulna to prevent recurrent synostosis. Posterior ossification at the proximal radioulnar joint is usually addressed with a modification of the Kocher approach with triceps elevation. In this approach, the anconeus, extensor carpi ulnaris, and proximal supinator are elevated anteriorly together, providing access to the posterior radioulnar joint and interosseus membrane. Again, the biceps tendon must be preserved, and a fat interposition may be needed between the lateral ulna and the bicipital tuberosity. If excessive bone also continues proximally to involve the humerus, this approach can be extended by elevating the triceps off of the humerus. If extensive intra-articular involvement is present, radial head resection through a modified Kocher approach is considered to be the conventional treatment. Moreover, if the ectopic ossification is extensive, both a medial and a Kocher approach may be necessary to remove all of the ectopic bone.

POSTOPERATIVE MANAGEMENT Mobilization Patients are placed in a soft compressive dressing and scheduled to return for a wound check and mobilization exercises within 48 hours of surgery. If collateral ligaments were repaired or reconstructed, a posterior plaster splint is placed. Most patients are discharged home the day of surgery. If the patient has had extensive surgery and requires intravenous medications, the patient will be discharged

home the following day. When the patient returns to clinic 2 days after surgery, the dressings are removed, the wound is examined, and the patient is instructed to begin ROM exercises. Depending on the degree of motion, static and dynamic splinting may be prescribed to help obtain full flexion. Extension splints may be prescribed at night to help preserve terminal extension. Some surgeons advocate the use of a continuous passive motion machine and inpatient hospitalization to improve early motion. The authors do not employ this method and prefer early discharge with expeditious follow-up. The patient is seen on a weekly basis for the first several weeks to ensure proper wound healing, lack of hematoma formation, and no loss of recovered motion.

Pain Control Although some surgeons advocate regional anesthesia for postoperative pain control, many surgeons are reticent about not being able to examine nerve function immediately after surgery. Thus, the authors do not use regional anesthesia for postoperative pain control, fearing late detection of an intraoperative or postoperative nerve injury.

Prophylaxis As discussed earlier, NSAIDs or radiation therapy can be used to prevent recurrent heterotopic ossification. Lowand medium-risk patients are treated with a 3- to 4-week regimen of indomethacin sustained-release, 75 mg by mouth once daily. Patients are given a prescription for a proton pump inhibitor to prevent gastric irritation from the indomethacin during the course of treatment. Highrisk patients or patients who cannot tolerate NSAIDs are treated with radiation therapy, with a one-time dose of 700 cGy delivered to the elbow within 72 hours of surgery.

COMPLICATIONS In addition to the expected possible surgical complications common to all elective orthopaedic surgical procedures, such as bleeding, infection, and nerve injury, patients undergoing surgical resection of heterotopic ossification of the elbow are at increased risk for additional complications due to the complexity of the anatomy and the surgical treatment. There is an increased risk for triceps rupture, which can be minimized by a strong, secure repair and avoiding extensive flexion exercises postoperatively. There is increased risk for damage to the collateral ligaments, resulting in elbow instability. The patient may initially be immobilized if the collateral ligaments are repaired or reconstructed, leading to a higher likelihood of residual elbow stiffness. Despite this possible need for prolonged immobilization to ensure ligamentous healing, most patients maintain functional arcs of motion of at least 100 degrees of flexion-extension and 100 degrees of supination-pronation, with an aggressive therapy program started whenever the soft tissue healing permits. There is an increased risk for hematoma formation, most likely due to increased bleeding surfaces created from bone resection. Hematomas can be minimized by using suction drainage for the first 24 to 48 hours after surgery. A significant hematoma may limit motion, serve as a

1298 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

source of infection, and require subsequent operative drainage. Skin necrosis can be minimized by trying to preserve adequate skin bridges, using previous incisions whenever possible, and avoiding raising extensive soft tissue flaps. Edema control with compression and elevation also minimizes postoperative swelling and skin compromise. Aseptic resorption of the capitellum is uncommon, but can be avoided by limiting extensive soft tissue dissection around the lateral elbow. The risk for recurrent heterotopic ossification is relatively low with proper surgical treatment and postoperative ­prophylaxis.

activity. When nonoperative treatments such as splints and aggressive therapy are not able to restore motion, surgical resection can provide functional elbow motion in up to 90% of patients. During the past decade, successful surgical results have been demonstrated, with resection as early as 3 months following the inciting injury. Preoperative CT of heterotopic bone can be very helpful in determining the optimal surgical approach (Fig. 19G-5). Although much progress has been made during the past decade with arthroscopic techniques for removal of heterotopic ossification, knowledge of extensile approaches is required to excise extensive ectopic ossification. With careful surgical planning, technical expertise, and close postoperative ­follow-up, complications can be minimized. Prophylaxis with either NSAIDs for about 4 weeks or single-dose radiation therapy is strongly recommended to prevent recurrence of heterotopic ossification. Long-term results indicate that patients can be expected to recover functional range of motion with appropriate preoperative, intraoperative, and postoperative care.

SUMMARY Heterotopic ossification is an uncommon complication of trauma, neural injury, and burns. Post-traumatic elbow trauma should be managed with adequate débridement, wound drainage, and appropriate prophylaxis. However, despite adequate surgical care, heterotopic ossification occurs and can cause significant stiffness and limitation of

A

B

C

D

Figure 19G-5   Example of clinical heterotopic ossification and treatment in a 48-year-old male after revision open reduction with internal fixation for a proximal ulna fracture. A, Radiographs 1 week after surgery. B, Radiographs 3 months after surgery. C, Computed tomographic scan with three-dimensional reconstruction. D, Computed tomographic scan with two-dimensional reconstruction (sagittal).

Elbow and Forearm 1299

F

E Figure 19G-5, cont’d  E, Intraoperative elevation of ectopic bone in olecranon fossa. F, Postoperative radiographs.

C l

r i t i c a l

P

o i n t s

 eterotopic ossification can result from various local or H systemic insults. Although the most common cause is direct elbow trauma, cerebrospinal trauma, burns, and genetic disorders also have been associated with heterotopic ossification about the elbow. l The exact mechanisms responsible for the formation of heterotopic bone are unclear. Heterotopic ossification is thought to result from inappropriate differentiation of pluripotent mesenchymal stem cells into osteoblastic stem cells. Bone morphogenic proteins and prostaglandins have been implicated in the pathogenesis of heterotopic ossification. l Most commonly, heterotopic ossification occurs in the posterolateral aspect of the elbow. However, it can occur anywhere in the elbow, including within the collateral ligaments and surrounding the nerves about the elbow. Although most heterotopic ossification about the elbow is not clinically significant, some patients may present with restriction of motion about the ulnohumeral or radiocapitellar joints due to the ectopic bone. l A thorough physical examination is required of patients undergoing evaluation for suspected complications from heterotopic ossification. When examining the patient throughout the course of the disease, it is important to document the active and passive ROM for flexion, extension, pronation, and supination. The examiner should also note whether the end point in each of these planes is compliant (suggesting soft tissue restriction) or rigid (suggesting osseus restriction). l In addition to plain radiographs of the elbow (anteroposterior and lateral), further advanced imaging is often helpful in determining the extent and location of heterotopic ossification. CT with two- and three-dimensional reconstructions is useful in determining whether ectopic bone is likely ­ responsible for the patient’s limitation in motion. This additional information helps surgical planning by helping determine the optimal surgical approach

and techniques, thus decreasing surgical time and rate of complications. l Patients who sustain an elbow injury and have significant risk factors for heterotopic ossification should receive one of the two available forms of prophylaxis against ectopic bone formation. The first treatment option is the use of chemotherapeutic agents such as NSAIDs to prevent heterotopic ossification. The second manner in which to prevent heterotopic ossification after elbow injury is to use low-dose external-beam radiation. Radiation is thought to prevent ectopic bone formation by preventing the differentiation and proliferation of precursor cells into osteoblastic cells by inhibiting DNA transcription in these precursor cells. l Before offering surgical intervention for a patient affected by stiffness due to heterotopic ossification about the elbow, a patient should be started on an aggressive motion program with or without the use of dynamic splinting to prevent the progressive loss of elbow motion during the period of maturation of the ectopic bone. l Operative intervention may be considered when nonoperative measures have failed to restore elbow motion. Most surgeons have now gone from avoiding surgical intervention for up to 2 years to offering patients surgical resection of ectopic bone as early as 4 months after the initial inciting injury. l Surgical resection of heterotopic ossification can be challenging and requires technical expertise to successfully restore elbow motion. The most commonly applied treatment involves identifying the ectopic bone and removing it until normal elbow motion is restored. A capsulectomy is used for accompanying capsular contracture. l Recently, with the increased use of arthroscopy for removal of ectopic ossification, larger exposures and dissections have been increasingly avoided. These more minimal approaches have been able to meet the surgical goals with much less morbidity than the more extensile approaches in many cases.

1300 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S U G G E S T E D

R E A D I N G S

Adams JE, Steinmann SP: Nerve injuries about the elbow. J Hand Surg [Am] 31(2):303-313, 2006. Balboni TA, Gobezie R, Mamon HJ: Heterotopic ossification: Pathophysiology, clinical features, and the role of radiotherapy for prophylaxis. Int J Radiat Oncol Biol Phys 65(5):1289-1299, 2006. Davila SA, Johnston-Jones K: Managing the stiff elbow: Operative, nonoperative, and postoperative techniques. J Hand Ther 19(2):268-281, 2006. Jupiter JB, O’Driscoll SW, Cohen MS, et al: The assessment and management of the stiff elbow: Treatment of ectopic ossification about the elbow. Heterotopic ossification of the elbow. Instr Course Lect 52(370):93-111, 2003. Steinmann SP, King GJ, Savoie FH 3rd, et al: Arthroscopic treatment of the ­arthritic elbow: Anatomic relationship between elbow arthroscopy portals and neurovascular structures in different elbow and forearm positions. Instr Course Lect 55(4):109-117, 2006.

Steinmann SP, King GJ, Savoie FH 3rd: Arthroscopic treatment of the arthritic elbow. J Bone Joint Surg Am 87(9):2114-2121, 2005. Summerfield SL, DiGiovanni C, Weiss AP: Heterotopic ossification of the elbow. J Shoulder Elbow Surg 6(3):321-332, 1997. Viola RW, Hastings H 2nd: Treatment of ectopic ossification about the elbow. Clin Orthop 370:65-86, 2000.

R E F E R E N C E S Please see www.expertconsult.com

S e c t i o n

H

Elbow Dislocations in the Adult Athlete and Pediatric Patient Sami O. Khan and Larry D. Field

INCIDENCE The rate of elbow dislocations is 6 to 13 cases per 100,000 people and accounts for 11% to 28% of all injuries to the elbow.1 Elbow dislocations occur more frequently in males than in females. The highest incidence occurs in the 10to 20-year age group. About 60% of dislocations occur in the nondominant extremity.2 Ten percent to 50% of elbow dislocations are sports related, although not unique or common to any specific sport. Incidentally, the incidence of recurrent dislocation in the sports setting is exceedingly rare. A retrospective review by Kenter and colleagues demonstrated 91 elbow injuries in the National Football League over a 5-year period; of these, only 15% sustained a recurrent elbow dislocation-subluxation.3 Dislocations of the elbow in children, in contrast to dislocations of other joints in children, are common, constituting about 6% to 8% of elbow injuries. In general, because the attachments of muscles and ligaments are stronger than the adjacent growth plate, forces across the joint in the setting of an injury often result in physeal injury as opposed to dislocation.

MECHANISM Posterior elbow dislocations most commonly occur from a fall onto an outstretched hand or wrist. Mehlhoff and associates reported a fall as the mechanism of injury in 75% of patients.4 Posterolateral elbow dislocations make up more

than 90% of all elbow dislocations (Fig. 19H-1). The more uncommon anterior dislocation may be caused by impact on the posterior forearm in a slightly flexed position. Exactly how these forces contribute to an elbow dislocation is still a subject of debate. As the force from a fall is transmitted to the extended elbow, a resultant anterior force is generated that levers the ulna out of its trochlear articulation. With continued hyperextension, the anterior capsule and collateral ligaments are subjected to significant tensile forces and, ultimately, failure. Additional valgus or varus forces result in lateral or medial displacement of the posterior elbow dislocation. Cadaveric studies by O’Driscoll and coworkers have demonstrated that an extension and varus moment associated with elbow dislocation disrupts the lateral ulnar collateral ligament first (Fig. 19H-2).5 If this force is dissipated, a simple perched dislocation ensues; however, additional force further rotates the forearm and tears the remaining anterior capsule and, finally, the ulnar collateral ligament, resulting in a complete dislocation. The most commonly accepted mechanism in the pediatric population for posterior dislocations is a disruption of the ulnar collateral ligaments.6 This produces valgus instability. The force applied to the medial aspect of the elbow can produce an avulsion fracture of the medial epicondyle with its associated flexor muscle group. The proximal radius and ulna displace laterally, with the intact biceps tendon acting as the center of rotation for the displaced forearm. The far more common injury pattern seen in younger children is subluxation of the annular ligament (nursemaid’s elbow).

Elbow and Forearm 1301

A

B

Figure 19H-1   Anteroposterior (A) and lateral (B) radiographs demonstrating a posterolateral elbow dislocation. This pattern is by far the most common dislocation pattern.

Longitudinal traction on the extended elbow is the usual mechanism of injury. Cadaveric studies have shown that longitudinal traction on the extended elbow can ­produce a partial slippage of the annular ligament over the radial head and into the radiocapitellar joint. This ­displacement of the annular ligament most commonly occurs with forearm

pronation. Such an injury typically occurs when a young child is lifted or swung by the forearm or when the child suddenly steps down from a step or a curb while one of the parents is holding the hand or wrist (Fig. 19H-3). Reports of acute surgical exploration of this injury have confirmed this observation.7 Amir and colleagues performed a controlled study comparing 30 normal children with 100 who had a documented nursemaid’s elbow.8 They found an increased incidence of ligamentous laxity in the nursemaid’s elbow group. In addition, there was a higher incidence of laxity in one or both of the parents in the involved group as well. Based on their findings, the authors have postulated that increased laxity could be a factor predisposing children to this condition.

RELEVANT ANATOMY 2 1 3

LUCL

MUCL 2

Figure 19H-2   The ring of instability with elbow dislocations describes the progression of stresses from the ulnar part of the lateral ulnar collateral ligament (LUCL) to the anterior capsule and finally ending with injury to the ulnar part of the medial ulnar collateral ligament (MUCL).

Intrinsic stability of the elbow is provided by both ­osseous and ligamentous structures in the adult. The ulnohumeral articulation is the cornerstone of osseous stability and mobility in the flexion-extension plane. The coronoid ­process resists posterior subluxation in extension. The medial facet of the coronoid imparts an osseous stability to varus stress. The radial head also provides elbow stability as a secondary stabilizer to valgus loads.9 Most activities in athletes rely on a combination of ligamentous integrity and bony articulation to provide stability to the elbow. The medial ulnar collateral ligament is composed of three major bundles—anterior, posterior, and oblique bands (Fig. 19H-4). The anterior band provides the primary restraint to valgus stress at 30, 60, and 90 degrees of flexion (Fig. 19H-5).10 It originates from the anteroinferior aspect of the medial epicondyle and inserts onto the sublime tubercle of the coronoid. The posterior bundle is fan shaped and also originates from the medial epicondyle. It inserts onto the medial margin of the semilunar notch and provides a secondary restraint to valgus forces at greater than 90 degrees of flexion. The lateral ligament complex consists of four components: the lateral (radial) collateral ligament, the lateral ulnar collateral ligament (LUCL), the accessory lateral

1302 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 19H-3   Pulled elbow syndrome (nursemaid’s elbow) most commonly occurs with a longitudinal pull with the forearm pronated, allowing the annular ligament to sublux into the radiocapitellar articulation. (Redrawn from Beaty JH, Kasser JR [eds]: Rockwood and Wilkins’ Fractures in Children, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2006.)

collateral ligament (LCL), and the annular ligament (Fig. 19H-6). The LCL attaches to the lateral epicondyle and fans out to merge indistinguishably with the annular ligament. It functions as a varus restraint and stabilizes the annular ligament. Martin in 1958 described some details of the LCL complex.11 He reported that the lateral capsule of the elbow consists of three layers. The deep layer he described as the joint capsule. The intermediate layer was the true annular ligament. The superficial layer was derived from the lateral ligament and fanned out from the lateral epicondyle to attach to the anterior and posterior aspects of the proximal ulna. Morrey and An12 named the LUCL and recognized its clinical significance.

The LUCL is a thickening of the capsule that attaches proximally to the lateral humeral epicondyle and distally to the tubercle of the supinator crest of the ulna. The humeral attachment of the LUCL is at the isometric point on the lateral side of the elbow and is well defined. The distal attachment of the ligament is deep to the fascia surrounding the extensor carpi ulnaris and supinator muscles. Besides stabilizing the lateral aspect of the elbow, the LUCL also acts as a posterior buttress for the radial head to prevent its subluxation. Josefsson and colleagues demonstrated that all 15 patients who underwent surgical exploration after acute complete elbow dislocation demonstrated injury to both the medial and lateral ligament complex.13

W Anterior band

Posterior band Transverse band Figure 19H-4   The ulnar part of the ulnar collateral ligament comprises three bands—anterior, posterior, and transverse.

C L E 13% 67% 20% Figure 19H-5   The anterior band of the ulnar collateral ligament is the primary restraint to valgus stress at 30 to 90 degrees of flexion. It originates from the anteroinferior aspect of the medial epicondyle and inserts on the sublime tubercle. It is about 5 mm wide and 27 mm long.

Elbow and Forearm 1303

r­ ecurrent. The temporal definition of chronic elbow dislocation is not clearly defined, but if a joint is unreduced for more than 7 days, attempts at closed reduction are often unsuccessful. A descriptive classification based on the relationship of the radius and ulna to the distal humerus has also been proposed and is as follows: posterolateral (>90%), posteromedial, anterior, medial, and divergent (exceedingly rare). O’Driscoll and colleagues have proposed a classification system with clinical relevance (Fig. 19H-7).5 It is more a spectrum of instability, from subluxation to dislocation. The three stages correspond with the pathoanatomic stages of the capsuloligamentous disruption as mentioned earlier. A perched dislocation is one in which the elbow is subluxed, but the coronoid is impinged on the trochlea. In this setting, the ligaments should not be severely injured and often can be reduced in the emergency room with simple intra-articular analgesia and sedation. Furthermore, rehabilitation should begin earlier, and a more complete recovery should be expected. A complete elbow dislocation implies a more substantial ligamentous injury. A general anesthetic and muscle relaxant is often required for reduction. When a complete dislocation occurs, both medial and lateral ligamentous structures have been torn, as well as the injury to the anterior capsule and brachialis. The higher degree of soft tissue injury explains the long-term loss of motion seen with complete dislocations. The overwhelming majority of pediatric dislocations can be classified through the descriptive classification. However, a few distinct clinical entities should be noted as specific to the pediatric population. Congenital elbow dislocations must be considered in the setting of a new elbow injury. The key to differentiating a congenital from an acute traumatic elbow dislocation is examination of the radiographic architecture of the articulating surfaces. In a congenitally dislocated elbow, there is marked atrophy of the humeral condyles and the semilunar notch of the olecranon. Often in the setting of elbow injury in the pediatric population, an occult fracture is a subtle

Lateral (radial) collateral ligament Annular ligament

Accessory lateral collateral ligament

Articular capsule Lateral ulnar collateral ligament Figure 19H-6   The lateral ligament complex of the elbow.

It is important to emphasize some of the anatomic ­ ifferences that are unique to the pediatric elbow joint. d The elbow joint in children younger than 10 years is basically a ligamentous joint. Because of this, there is considerable flexibility in the elbow joint in children. It is not unusual for a child to hyperextend 10 to 15 degrees. It is this combination of hyperflexibility and lack of osseous stability that makes the elbow joint predisposed to dislocation in the setting of trauma. The major stabilizing ligaments on both the medial and lateral sides of the elbow originate from the distal humerus through apophyses. As discussed earlier, these are structurally weak regions prone to avulsion injuries and subsequent loss of joint integrity.

CLASSIFICATION Multiple classification systems have been proposed for elbow dislocations. Chronologically, elbow dislocations can be described as acute, chronic (unreduced), and

0 Reduced

1 PLRI

2 Perched

3 Dislocated

Supination Axial compression Valgus Figure 19H-7   The spectrum of elbow instability, from subluxation to dislocation. Stages 1 to 3 correspond with the stages of capsuloligamentous disruption (see text).

1304 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

­ nding on ­radiographic ­evaluation. Obtaining views of the fi ­contralateral extremity is usually helpful. Additionally, evidence of a posterior fat pad sign on the lateral radiograph may be the only indicator of injury. If clinical history and physical examination findings are consistent with an elbow injury, the extremity should be treated as a fracture until proved otherwise. Subluxation of the annular ligament, or pulled elbow syndrome, is a far more common elbow injury in young children. The term nursemaid’s elbow and other synonyms have been used to describe this condition. The mean age of injury is usually 2 to 3 years. It rarely occurs after 7 years of age. It is difficult to determine the actual incidence because many subluxations are treated in primary care physicians’ offices or resolve spontaneously before being seen by a physician.

EVALUATION It is of paramount importance that the neurovascular status of the dislocated elbow be assessed and documented before manipulation. Specifically, the status of the brachial artery and median and ulnar nerves should be determined because they are most vulnerable to entrapment during manipulation. Serial neurovascular examinations should be considered in cases in which massive antecubital fossa swelling exists or in which the patient is believed to be at risk for a compartment syndrome. Angiography may be necessary to evaluate for vascular compromise. If clinical symptoms are indicative of a compartment syndrome, emergent fasciotomies of the forearm and hand should be performed. Radiographic evaluation should include standard anteroposterior and lateral radiographs and should be obtained to determine the dislocation pattern and scrutinized for associated fractures about the elbow. If there is any concern of vascular compromise or potential compartment syndrome, the patient should be admitted for ­observation.

ASSOCIATED INJURIES Vascular Injury Many authors have reported injuries to the brachial artery associated with elbow dislocation. In 1913, Sherill described brachial artery transection following elbow dislocation and successfully treated it with primary repair.14 Since then, the history of treatment has included observation, ligation of the brachial artery, interpositional vein graft, and direct suture repair. The consensus today is that arterial repair with saphenous vein grafting should be the standard of care. Occasionally no graft is needed, but if the zone of injury is extensive and full mobility after repair is the ultimate goal, the reconstructed vessels must have adequate length to permit full extension. Loss of pulse does not preclude an attempted closed reduction; although, if immediate reperfusion of the extremity does not occur after closed reduction, surgical exploration with possible saphenous vein grafting should be planned immediately. Angiography should not delay surgery; if needed, it should be performed intraoperatively.

The treating physician also needs to be aware of a potential compartment syndrome that can accompany a vascular injury. If clinical symptoms warrant, fasciotomies of the forehand and arm should be performed.

Nerve Injury The median, ulnar, radial, and anterior interosseous nerves are all susceptible to injury at the time of elbow dislocation. Typically, the injury is a stretch injury that improves with time. Mehlhoff and colleagues reported on a series of 52 elbow dislocations, of which 9 had nerve injury treated nonoperatively.4 The median nerve can be injured at the time of dislocation or even during reduction, when it can become entrapped in the joint. It is therefore imperative that nerve function be determined before and after manipulation for the independent function of each nerve. If nerve deficit is not worse after closed reduction, the patient should be followed closely. Baseline electromyographic studies 6 weeks after injury may be helpful in demonstrating signs of denervation but should not substitute for careful clinical evaluation. Spontaneous recovery usually occurs, but if none is noted after 3 months, an operation should be considered.

Fracture Associated fractures have been reported in 25% to 50% of cases of elbow dislocation, the most common being radial head fractures. In general, radial head fractures occur in 10% of elbow dislocations.15 Other fractures associated with dislocation include the coronoid, olecranon, and medial and lateral epicondyles. Fracture-dislocations of the elbow represent a high-energy mechanism and should be treated as a separate entity.

TREATMENT OPTIONS Closed Reduction for Posterior Dislocation Closed reduction should be performed under conscious sedation in a monitored emergency department setting for complete elbow dislocations. This will allow for relaxation of muscle spasm, such that with slow, continuous, gentle, longitudinal traction, combined with gradual flexion with an anterior directed force over the olecranon, the elbow should reduce without difficulty. Some authors report hyperextension of the deformity first to unlock the olecranon, but the potential danger of median nerve entrapment remains. Repeated attempts at reduction should be avoided because soft tissue injury can worsen. After obtaining a reduction, the elbow should be taken through a range of motion to determine the arc of stability. Often the elbow remains reduced throughout the range of motion; on occasion, the elbow tends to redislocate in extension. Knowledge of the stable arc of motion enables the treating physician to allow early motion within this arc without compromising the stability of the elbow joint. Neurovascular status is again evaluated after reduction. Post-reduction radiographs should be obtained to confirm a congruent

Elbow and Forearm 1305

Figure 19H-8   With Lavine’s pusher method, the child is held in the parent’s lap while the affected extremity is draped over the edge of the chair’s armrest. The treating physician manually pushes on the olecranon distally with his or her thumb while pulling axial traction with the other arm. (Redrawn from Beaty JH, Kasser JR [eds]: Rockwood and Wilkins’ Fractures in Children, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2006.)

reduction. In a perched dislocation, an analgesic with intra-articular local anesthetic is often adequate to perform a reduction maneuver. Direct pressure is applied over the olecranon while the elbow is slightly extended and gentle axial distraction is performed. In the instance of simple pediatric elbow dislocations, most series demonstrate overwhelming success with closed reduction.16 Of note, there is a tendency toward rapid progressive swelling secondary to soft tissue injury with children that makes it imperative that the joint be promptly reduced. Royle17 found that dislocations reduced soon after the injury had better outcomes than those in which reduction was delayed. The Lavine pusher method has been described and found to be especially helpful in smaller children (Fig. 19H-8). The child is held by the parent, while the elbow is draped over the edge of the chair. The olecranon is pushed distally past the humerus by the thumb of the physician while the other arm pulls distally along the axis of the forearm.

Closed Reduction for Annular Ligament Subluxation Virtually all annular ligament subluxations are successfully treated by closed reduction (Fig. 19H-9). The child usually is seated on the parent’s lap. The patient’s forearm is grasped with the elbow semiflexed while the thumb of the surgeon’s opposite hand is placed over the lateral aspect of the elbow. The forearm is first maximally supinated. If this fails to produce the characteristic snap of reduction, then the elbow is flexed maximally until the snap occurs.

Figure 19H-9   The reduction maneuver for a nursemaid’s elbow entails placing the thumb over the radial head and maximally supinating the forearm. If no audible or palpable click is heard, the forearm is then flexed up while maintaining a supinated position. (Redrawn from Beaty JH, Kasser JR [eds]: Rockwood and Wilkins’ Fractures in Children, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2006.)

Open Reduction for Elbow Dislocation The need for open reduction after simple, acute dislocation is uncommon. Radial head entrapment has been rarely shown to become trapped in the soft tissues of the forearm or even to buttonhole through the forearm fascia. Medial epicondyle fractures can also become incarcerated within the joint and prevent closed reduction. Similar difficulties with closed reduction have been shown in other fracturedislocations seen with radial head, olecranon, and coronoid fractures. Such findings would indicate the need for surgery. The surgical approach of arthrotomy and removal of incarcerated tissue should be tailored to the specific injury pattern. For example, radial head entrapment causing a locked elbow dislocation should be approached through a lateral Kocher approach.

Surgical Treatment For the grossly unstable elbow (i.e., unstable from 0 to 90 degrees of flexion), surgical treatment should be considered. Recurrent dislocation in the acute setting usually occurs in the face of severe trauma in which the radial head and coronoid process are fractured (“terrible triad”). It is

1306 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Conversely, the lateral ligaments should be repaired or reconstructed. Care should be taken that the repair or graft reconstruction be performed through the points of isometry of the lateral ulnar collateral ligament.

WEIGHING THE EVIDENCE

A

Closed treatment of elbow dislocation in adults has overall demonstrated satisfactory results. Most authors report good results with nonoperative treatment for simple elbow dislocations. Van der Ley, in a series of 20 adults treated, found that 80% had good or excellent results.18 He reported no results of long-term instability after 5 years of follow-up, but reported poor results in those patients sustaining nerve injury at the time of dislocation. Primary repair of the elbow ligaments has demonstrated mixed results. Durig and colleagues demonstrated satisfactory results with acute ligament repair of gross instability of the elbow.19 A prospective study by Josefsson and coworkers compared operative and nonoperative treatment of elbow dislocations, in which they performed ligament repairs and found similar results in regard to pain and long-term stability.20 There is no justification for isolated ligament repair with uncomplicated dislocations, especially in the athlete, because this will only delay rehabilitation. The use of the hinged elbow external fixator has not been looked at with isolated elbow dislocations. However, there are multiple studies to validate its use in the setting of complex elbow fracture-dislocations. Mckee and colleagues evaluated the utility of hinged external fixators for complex recurrent instability of the elbow and found good to excellent results in 75% of its study population.21

Authors’ Preferred Method

B Figure 19H-10   A and B, The hinged external fixator is a useful adjunct in the setting of gross instability. Again, this is seen with high-energy mechanisms; often, major soft tissue trauma and fracture-dislocations are seen.

exceedingly rare to have acute instability in the setting of isolated elbow dislocation. In most cases, a high-energy mechanism (e.g., fall from height, motor vehicle crash) is seen. Isolated primary repair of the elbow ligaments after acute dislocation is becoming a procedure of historical ­significance. A hinged external fixator offers another surgical solution to the grossly unstable elbow in the setting of high-energy trauma, especially when compliance in a postoperative hinged, controlled motion brace is an issue (Fig. 19H-10). If despite fracture fixation and ligamentous repair, the elbow continues to be unstable, a hinged external fixator should be considered. Application of the hinged ­ external fixator will allow full unrestricted motion without the loss of a congruent reduction. Even if it is not repaired, the ulnar collateral ligament heals normally if protected.

The authors recommend that complete elbow dislocations be reduced under conscious sedation in the emergency room setting. Perched elbow dislocations should be attempted with a local intra-articular anesthetic and pain medication. After reduction is obtained, the stable arc of motion is determined. The patient is immobilized for a short time. In the extremely rare setting of an acutely unstable simple dislocation, we recommend isometric repair or reconstruction of the lateral ulnar collateral ligament with or without ulnar collateral ligament repair. In an exceedingly rare situation, a hinged external fixator may also be required to achieve stability.

POST-REDUCTION RECOMMENDATION For simple elbow dislocations, the elbow is immobilized for 3 to 5 days in slightly less than 90 degrees of flexion, depending on the degree of anterior soft tissue swelling, in a posterior splint. The trend in rehabilitation of the dislocated elbow has been toward less immobilization and earlier motion. For a stable elbow, full active elbow motion should be started about 1 week after injury. In the setting

Elbow and Forearm 1307

of the unstable elbow, protected range of motion can begin in a hinged brace within the stable arc of motion. An extension block should be placed at the point of instability. A progressive decrease of extension block by 15 degrees each week should allow the patient to obtain full motion by 4 to 6 weeks.

fractures, violates the lateral collateral ligament complex, and it is imperative that the complex be repaired at the completion of the procedure.23 PLRI of the elbow also can develop as a complication of the Boyd surgical approach, which involves release of the soft tissues from the lateral side of the proximal ulna.23

POTENTIAL COMPLICATIONS

CLINICAL PRESENTATION OF RECURRENT INSTABILITY

Recurrent Instability Although it is uncommon for the athlete, recurrent instability is probably the most devastating complication after elbow dislocations. LUCL insufficiency is the essential lesion leading to elbow instability. Although most studies have indicated that both the medial and lateral collateral ligaments are acutely disrupted with an elbow dislocation, the residual insufficiency most commonly involves the lateral collateral ligament complex for reasons that have not been fully elucidated. This insufficiency gives rise to recurrent posterolateral rotatory instability, a clinical entity first described by O’Driscoll, Morrey, and colleagues.5,22 Posterolateral rotatory instability (PLRI) is a known sequela of elbow dislocations regardless of whether they reduced spontaneously or were treated by closed reduction. One half of patients with PLRI have a history of documented dislocation requiring reduction.23 However, some may have only experienced either chronic elbow sprains or fractures of the radial head or the coronoid process. PLRI can also be seen as a complication of the lateral surgical approaches to the elbow. The Kocher approach, although providing a good exposure for management of radial head

Recurrent painful clicking, snapping, clunking, and locking of the elbow are the most common symptoms. These often occur in the extension half of the arc of motion, with the forearm in supination. Patients may report that the elbow feels loose or slides out of joint when they perform activities. Patients are often apprehensive about performing activities that precipitate the instability, for example, when pushing off to rise from a chair. On initial examination, the patient appears to have a normal elbow. It is not tender (unless there has been a recent injury), and there is usually a full pain-free range of motion. It is not uncommon to have hyperextension, particularly with the atraumatic type of instability. Applying a valgus or varus instability test does not cause pain and will not usually produce any symptoms.

Physical Examination The clinical examination is usually unremarkable except for the PLRI test. This is best performed with the patient supine and the extremity over the patient’s head (Fig. 19H-11). The shoulder is fully externally rotated, which

Valgus Axial compression

Supination

Subluxation

Figure 19H-11   The posterolateral rotatory instability test—with extension, valgus, axial load, and supination, the elbow subluxes. With flexion and pronation, the elbow reduces.

1308 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

stabilizes the humerus so that the elbow can be assessed independent of shoulder motion. The examiner then grasps the patient’s forearm, which is placed in full supination. In this position, the elbow looks like a knee, and the maneuver is analogous to the pivot shift test used to assess anterior cruciate ligament instability. Starting with supination and extension, the elbow is slowly flexed while the examiner applies a slight valgus force and axial load and maintains the supination. This produces a rotatory supination torque on the forearm that can produce a rotatory subluxation of the ulnohumeral articulation. The ulna tilts externally on the trochlea of the humerus, and this rotation dislocates the radial head posteriorly because it is coupled to the ulna by the annular ligament. As the elbow is flexed to about 40 degrees, the rotatory displacement is at a maximum. At this point, the subluxated radial head produces a posterior prominence associated with an obvious dimple in the skin proximal to the radial head.

Diagnostic Imaging Standard radiographs of the elbow are usually normal, although there can be slight widening of the radiohumeral joint. On lateral radiographs, the radial head may appear slightly posterior to the capitellum. Stress views, magnetic resonance imaging (MRI), and computed tomography have not been found to increase the accuracy of diagnosis.

Treatment Options Conservative management with protected range of motion and activity modification rarely eliminates symptoms of instability unless a patient is seen early after injury. The prospect of permanent use of an elbow brace is ­cumbersome for the patient. By definition, this is a recurrent problem, and surgical intervention is often required. Surgical options include open ligament repair, arthroscopically assisted posterolateral corner reconstruction, or open reconstruction with graft. Arthroscopic posterolateral ligament reconstruction is a relatively newly published technique that can be performed for symptomatic PLRI of the elbow, but certain criteria must be met.24 Specifically, adequate ligamentous and capsular tissue must be confirmed by MRI preoperatively and by arthroscopy intraoperatively. Also, standard contraindications to elbow arthroscopy include severe bony or fibrous ankylosis, and previous surgery that has distorted the native anatomy, such as a previous ulnar nerve transposition, should be ruled out. It is important to be prepared to perform an open procedure with tendon graft for LUCL reconstruction in case inadequate tissue is encountered at arthroscopy.

Surgical Technique We prefer the prone position for all elbow arthroscopy because it allows the elbow to be stabilized as well as giving improved access to the posterior compartment. We routinely use a pneumatic tourniquet and a prone arm holder. The elbow should be positioned and draped so that the arm is supported by the holder at the proximal upper arm, the elbow rests at 90 degrees of flexion, and the antecubital

Figure 19H-12   Examination under anesthesia. Often, the posterolateral rotatory instability test cannot be elicited in the awake patient. Thus, the elbow should be re-examined after induction of anesthesia for instability.

fossa is free from contact with the holder. An examination under anesthesia is performed first. Again, the PLRI test is performed as described earlier (Fig. 19H-12). A standard proximal anteromedial portal is created, and the arthroscope is placed inside the joint to begin the diagnostic arthroscopy. This portal will allow excellent visualization of the radiocapitellar joint. While pronating and supinating the wrist, the radiocapitellar articulation should be inspected; with supination, if the radial head will sublux in relation to the capitellum while rotating, then the posterolateral complex is injured. This assessment is known as the arthroscopic posterolateral rotatory load and shift test (Fig. 19H-13). Next, the posterolateral ligament and capsule structures are visualized to assess adequacy and quality of tissue. A spinal needle is then used to confirm the proper starting point and trajectory for the proximal anterolateral portal. A soft tissue shaver is introduced through this portal to abrade the capsule to obtain a bleeding surface to enable healing (Fig. 19H-14). Next, a spinal needle is inserted along the proximal radial border of the olecranon and directed toward the radiocapitellar articulation. A polydioxanone suture (PDS) is introduced into the joint through the spinal needle (Fig. 19H-15). Next, a suture retrieval device is placed percutaneously through the midlateral portal, just posterior to the lateral epicondyle, and is delivered into the joint (Fig. 19H-16). It is then used to retrieve the PDS and is drawn out. This sequence is repeated several times until multiple sutures have been passed. With each successive pass, the spinal needle pierces through the capsular tissue more proximally (Fig. 19H-17). Both limbs of each suture are retrieved through the midlateral portal and sliding knots are tied.

Surgical Reconstruction of the Lateral Collateral Ligament Again, surgery is indicated in patients with symptomatic instability. Open reconstruction with ipsilateral palmaris longus graft should be pursued if MRI demonstrates

Elbow and Forearm 1309

A

B

Figure 19H-13   A and B, Arthroscopic posterolateral rotatory instability test. With supination, the radial head is seen to sublux in relation to the capitellum, indicating instability. Also, note the presence of adequate capsular tissue to undergo reconstruction.

Figure 19H-14   The capsule is abraded with a soft tissue shaver in order to help healing after repair.

Figure 19H-15   A spinal needle is passed 5 to 6 cm distal from the superior subcutaneous border of the ulna and is directed toward the radial head inside the annular ligament.   A 1-0 polydioxanone suture is passed into the joint.

Figure 19H-16   A suture retrieval device is inserted just posterior to the lateral epicondyle and then pierces through posterior capsular tissue before being delivered into the joint to retrieve the suture.

Figure 19H-17   Multiple sutures have been passed for repair of the posterolateral capsule. When the sutures are tied, the arthroscope is often progressively driven out of the joint as each stitch is subsequently tied.

1310 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

1

2

Point of isometry

A

B

Figure 19H-18   A and B, Lateral ulnar collateral ligament reconstruction with palmaris graft. Care is taken to place the graft through holes in the ulna and brought into the isometric origin near the lateral epicondyle. Taking sutures through the osseous tunnel in the ulna, the isometric point is determined by the point where there is no laxity or change in tension on the suture as the elbow is taken through range of motion. (Redrawn from Morrey BF: Master Techniques in Orthopedic Surgery: The Elbow, 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2002.)

i­nadequate capsular tissue for capsular plication (Fig. 19H-18). A relative contraindication of this procedure is skeletal immaturity. Pediatric patients with open physeal plates should not undergo ligament reconstruction with tendon graft crossing the physis. Instead, the lateral collateral ligamentous tissue should be imbricated and reattached to bone isometrically. The absence of a radial head may adversely affect outcomes, but has not been found to be an absolute contraindication to surgery.

POSTOPERATIVE PRESCRIPTION After posterolateral capsular reconstruction, the patient is splinted in 60 degrees of flexion and near full pronation. Next, protected motion in the brace with a block beyond 30 degrees of extension is performed beginning 1 week after surgery and continued until 4 weeks. After 6 to 8 weeks, full motion is begun out of the brace.

CRITERIA FOR RETURN TO PLAY After undergoing an extensive postoperative ­rehabilitation course for 2 months, the patient is reassessed. If nearly full extension and equal strength compared with the uninjured extremity are obtained, the patient can begin an interval training program specific to his or her sport and eventually return to full activity without restrictions.

C

r i t i c a l

P

o i n t s

l

 valuation of elbow dislocation involves a prereduction E and post-reduction neurovascular examination. l Classification is based on distinguishing between perched and complete dislocation. l For the pediatric elbow injury, assess the radiograph for the posterior fat pad sign. After reduction, determine a stable arc of motion. l Plan for a short period of immobilization. l In patients undergoing arthroscopic posterolateral capsular reconstruction, confirm preoperatively with MRI and intraoperatively with arthroscopy the integrity of the capsular tissue. l An arthroscopic posterolateral capsular reconstruction is based on the arthroscopic posterolateral rotatory instability test.

S U G G E S T E D

R E A D I N G S

Bucholz R (ed): Rockwood and Green’s Fractures in Adults, 5th ed. New York, Lippincott Williams & Wilkins, 2002, pp 921-934. Josefsson PO, Gentz CF, Johnell O: Surgical versus non-surgical treatment of ligamentous injuries following dislocation of the elbow joint. J Bone Joint Surg Am 69:605-608, 1987. Josefsson PO, Johnell O, Gentz CF: Long-term sequelae of simple dislocations of the elbow. J Bone Joint Surg Am 66:927-930, 1984. Mehta J, Bain G: Posterolateral rotatory instability of the elbow. J Am Acad Orthop Surg 12:405-415, 2004. Morrey BF, An KN: Articular and ligamentous contributions to the stability of the elbow joint. Am J Sports Med 11:315-319, 1983. O’Driscoll S, Jupiter J, King G, et al: The unstable elbow. J Bone Joint Surg Am 82(5):724-738, 2000. O’Driscoll SW, Morrey BF, Korinek S, An KN: Elbow dislocation and subluxation: A spectrum on instability. Clin Orthop 280:186-197, 1992.

R eferences Please see www.expertconsult.com

Elbow and Forearm 1311

S e c t i o n

I

Entrapment Neuropathies around the Elbow Richard Y. Kim, Valerie M. Wolfe, and Melvin P. Rosenwasser

ELECTROPHYSIOLOGIC TESTING Electromyography and nerve conduction studies are important in confirming a clinical diagnosis of compression neuropathy. These studies objectify the severity of injury and clarify etiology, especially in multifocal entrapment pathology, the double-crush syndrome. Baseline electrical postoperative electrodiagnostic studies can be obtained for comparison and to monitor either spontaneous or postsurgical recovery of nerve conduction changes. The variability of clinical presentation requires physical examination and history to be as important as the electrodiagnostic studies in the management of elbow entrapments. It appears obvious that a positive nerve study with absent clinical signs should be questioned. The corollary of normal nerve studies and a focal and reproducible examination for a compression neuropathy still warrants the appropriate clinical response. Neurogenic pain is also not measurable on nerve tests. Finally, nerve studies are generally done in a static and neutral limb posture, which cannot emulate the provocative position of elbow flexion, and are also done at rest instead of following strenuous activity with muscle swelling and engorgement.1-5

MAGNETIC RESONANCE IMAGING Magnetic resonance imaging is a powerful and increasingly accurate tool for assessing the soft tissue anatomy around the elbow and can accurately identify anomalous muscles, fascial bands, or neoplasms, which may directly impact neural structures about the elbow.6-10

mass effect, as can elbow synovial ganglia.11-13 Any posttraumatic deformity like a lateral condyle malunion may result in a cubitus valgus deformity and a tardy ulnar nerve injury.14-16 Similarly, after other fractures and dislocations, the ulnar nerve may lose its smooth gliding path because of adhesions. The act of throwing is particularly stressful and creates tensile loads on the medial collateral ligament, which is adjacent to the ulnar nerve in the cubital tunnel.17 Elbow flexion in the cocking phase of throwing requires the ulnar nerve to elongate about 4.7 mm.18 Changes in the medial collateral ligament can generate tissue reaction, with a requisite reaction leading to impingement of the ulnar nerve (Fig. 19I-1).19 In the late cocking phase of the overhead throwing motion, the combination of elbow flexion and wrist extension can cause a sixfold increase in the pressure seen within the cubital tunnel.20 This is analogous to a localized compartment syndrome, as one can see in exertional lower extremity presentations. A recent cadaveric study demonstrated that the strain on the ulnar nerve was increased in all phases of the throwing motion and that in the acceleration phase, the maximal strain approached the elastic and circulatory limits of the nerve.21 In baseball pitchers, the chronic valgus extension overload forces applied to the elbow eventually lead to ulnar collateral ligament instability medially and compressive forces laterally.22,23 In fact, more than 50% of professional pitchers have some degree of acquired valgus elbow deformities.24

CUBITAL TUNNEL SYNDROME Etiology Ulnar neuritis is the most common entrapment neuropathy at the elbow. As the ulnar nerve descends from the brachial plexus, there are several choke points above, at, and below the elbow, including the arcade of Struthers, the cubital tunnel, and the flexor carpi ulnaris aponeurosis. The anatomic compression points are often normal fascial bands or muscle origins that have become hypertrophied with use, or whose neural passageways no longer allow nerve gliding and so provide both a traction and compression nerve injury. Of course, soft tissue tumors such as lipomas and anomalous muscles like the anconeus epitrochlearis can also cause compression of the ulnar nerve through their

Figure 19I-1   Attenuation of Osborne’s ligament can lead to ulnar nerve subluxation around the medial epicondyle, which may or may not be symptomatic.

1312 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

In addition to a slight loss of native elbow extension due to compensatory posterior humeroulnar osteophytes, these anatomic changes predispose pitchers to ulnar neuritis. Triangular arcuate ligament tears in association with ulnar nerve subluxation have also been documented in baseball pitchers.25 Any localizing pain to the medial elbow must be carefully differentiated between mechanical anterior oblique ligament pain closer to the coronoid or posterior cubital tunnel pain elicited by compression (Tinel) or provocative flexion tests.

Anatomy The ulnar nerve arises from the C8 and T1 roots, which coalesce into cords that traverse the supraclavicular and infraclavicular paths and then enter the arm. In the arm, it may have an anomalous motor branch to the medial triceps, but passes from the anterior to the posterior compartment. It travels deep to a thick aponeurotic band between the medial head of the triceps and the medial intermuscular septum called the arcade of Struthers, where the brachial artery branches into the superior ulnar collateral artery. The artery and nerve continue distally and medially on the anterior surface of the medial head of the triceps muscle. The nerve continues to descend posterior to the medial epicondyle, entering the cubital tunnel (Fig. 19I-2). The walls of the cubital tunnel are the medial trochlea, the medial epicondylar groove, and the posterior portion of the ulnar collateral ligament. The triangular arcuate ligament (Osborne’s ligament), extending from the olecranon to medial epicondyle, forms the roof of the tunnel and blends distally with the aponeurosis between the two heads of the flexor carpi ulnaris. The cubital tunnel is the most common area of compression.26 Distal to the cubital tunnel, the ulnar nerve travels between the ulnar and humeral heads of the flexor carpi ulnaris. Further distally, it passes between the flexor carpi ulnaris and the flexor digitorum profundis muscle bellies, providing motor branches to these muscles before continuing into the forearm and hand. Muscular hypertrophy of the flexor carpi ulnaris in athletes can be a source of ulnar nerve compression.27

Figure 19I-2   The ulnar nerve after in situ release. The roof of the tunnel has been removed.

History and Physical Examination In the general population, symptoms of ulnar neuritis typically start as paresthesias and hypesthesias involving half of the ring and small finger. Sensory changes often bring the patient to see a physician but, when not painful, are often tolerated for long periods, resulting in intrinsic muscle weakness and hand dysfunction. Ulnar-supplied extrinsic muscles like the flexor carpi ulnaris and flexor digitorum profundis 4 and 5 are usually unaffected with cubital tunnel syndrome because of the protected location of the extrinsic motor fibers in the nerve microanatomy and topography.28 Severe ulnar neuropathys result in muscle paralysis and a claw hand deformity, with weak pinch and grasp (Fig. 19I-3). In the athlete, ulnar neuropathy can present with pain along the medial arm both proximal and distal to the elbow, especially with throwing. In addition, ulnar nerve subluxation secondary to an attenuated or torn Osborne’s ligament with forceful elbow flexion and extension can create a painful “snapping” or “popping” sensation. Sensory findings are more common and occur earlier than motor ­weakness. Localizing signs such as Tinel’s sign at the cubital tunnel are confirmatory for ulnar compression but are not always present. The elbow flexion test is a useful provocative test much like Phalen’s sign at the wrist for carpal tunnel ­syndrome.18 To perform this test, the patient’s elbow is fully flexed and the wrist fully extended for 1 minute (Fig. 19I-4). Tingling or numbness in the ulnar nerve distribution helps to confirm compression neuropathy at the elbow. A clinical scale of ulnar neuropathy based on physical findings has been described by McGowan.29

Treatment Nonoperative Treatment Initial treatment consists of rest and activity modification. Nonsteroidal anti-inflammatory drugs and physical modalities like ice or heat are helpful, as is the use of an elbow sleeve to limit terminal flexion and to cushion the area over the nerve. If symptoms improve, a ­rehabilitation program

Figure 19I-3   Claw hand deformity in a patient with severe ulnar neuropathy at the cubital tunnel.

Elbow and Forearm 1313

Figure 19I-5   A medial curvilinear incision is extended approximately 8 cm proximal and 6 cm distal to the medial epicondyle. This allows exposure from the arcade of Struthers to the flexor carpi ulnaris.

Figure 19I-4   The elbow flexion test should be performed with the elbows in full flexion and the wrists in full extension. The position should be held a minimum of 1 minute; paresthesias in the distribution of the ulnar nerve are considered a positive result.

including reconditioning will hasten return to competitive sports.30 A study by Dellon and colleagues reported that 50% of patients with stage 1 compression will improve with conservative measures.31 Injection therapy at or around the cubital tunnel with corticosteroids is controversial and unpredictable. Patients with persistent symptoms that prevent return to activity or progress to motor weakness should be considered for surgical decompression.

Operative Treatment The optimal operative treatment for cubital tunnel syndrome is still uncertain. The options include in situ decompression alone, either open or arthroscopic; medial epicondylectomy; anterior subcutaneous transposition; and anterior submuscular transposition. A recent meta-analysis reported that in patients with mild ulnar neuritis, all these options had similar outcomes.32 However, global symptom relief was best after medial epicondylectomy and the least after subcutaneous transposition. For patients with moderate disease, submuscular transposition was the most effective. And for severe disease, most treatment options had similar results, except for medial epicondylectomy patients, who had poorer outcomes.32 Each procedure has advocates and can be effective for mild to moderate cases.

In situ Decompression The cubital tunnel is the most common site of compression, making an in situ release the simplest and least invasive procedure. This concept has led to endoscopic cubital tunnel release as a primary surgical option.33,34 Favorable

equivalent results to anterior transposition have been shown in prospective studies.35,36 However, athletes with ulnar neuritis associated with medial collateral ligament laxity and elbow instability may not respond to an in situ release alone. In situ decompression did not significantly reduce tensile strains on the ulnar nerve in a cadaveric study.37 A limited approach for in situ release does not permit inspection of the other potential sites of nerve compression or tethering. Caputo and Watson reported that at revision surgery for failed in situ decompression of cubital tunnel syndrome, the most common unrecognized sites of compression were the medial intermuscular septum and the flexor-pronator aponeurosis.38 Neither of these two areas are routinely visualized with in situ decompression.

Anterior Subcutaneous Transposition A skin incision is made posterior to the medial epicondyle and parallel to the course of the ulnar nerve along the medial intermuscular septum (Fig. 19I-5). The landmarks for the incision are the anterior edge of the medial triceps proximally and the two heads of the flexor carpi ulnaris distally. Two thirds of the incision should extend above the elbow and one third distal to it. The nerve is identified and released proximally at the arcade of Struthers. The medial intermuscular septum is excised at its distal insertion into the humerus and the nerve is dissected within the medial head of the triceps over the cubital tunnel (Fig. 19I-6). It is important to maintain the vascular mesentery with the nerve to minimize any ischemic injury. Osborne’s ligament, which composes the roof of the cubital tunnel, is carefully incised while protecting the nerve at all times. The nerve is traced distally until it passes underneath the fascia between the two heads of flexor carpi ulnaris (Fig. 19I-7). This fascia is divided and the nerve decompression is complete. The ulnar nerve is now free with its vascular pedicle and can be transposed anterior to the medial epicondyle, placing it atop the forearm fascia. After transposition, the nerve alignment is checked in both flexion and

1314 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

can begin an interval throwing program and advance to full activity when arm conditioning is completed.

Anterior Submuscular Transposition The incision and dissection for submuscular ulnar nerve transposition are similar for nerve exposure, but the flexor pronator origin is transected, leaving a robust stump for reattachment. The ulnar nerve is placed adjacent to the elbow capsule and beneath the flexor carpi ulnaris, pronator teres, flexor carpi radialis, palmaris longus, and a portion of the flexor digitorum sublimis muscle bellies. This places the ulnar nerve next to the median nerve. The common flexor origin is then reattached with sutures and sometimes supplemented with suture anchors. One advantage of this technique is the deep and wellprotected transposition, which is stable. A disadvantage is that the flexor-pronator muscle detachment requires prolonged protection and rehabilitation. This dissection may result in muscle scarring and loss of nerve glide.43 This type of procedure is not favored in throwing athletes.

Figure 19I-6   The medial antebrachial cutaneous nerves are identified and protected to minimize neuroma formation as well as an insensate posterior skin flap.

extension to ensure there are no secondary compressions or kinking on adjacent structures like the medial intermuscular septum stump. The nerve is maintained in its anterior location by a fascial sling created from the flexor-pronator origin (Fig. 19I-8). It can act as either a sling over the nerve or a trapdoor behind the nerve to prevent posterior resubluxation. At our institution, we prefer to wrap the nerve with a proximally pedicled fat flap from the anterior skin flap, thereby protecting the nerve with vascularized fat. One advantage of this technique is to diminish potential scarring around the fascial flap and exposed muscle. An inadequate sling may lead to recurrent subluxation and symptoms. Also, the nerve may be vulnerable to contusion in thin patients because of the lack of adequate padding in its new subcutaneous position.39 After surgery, immediate mobilization has been shown to reduce the time needed to return to work.40,41 For the athlete, rehabilitation should be specific to the sport. For throwing athletes, strengthening exercises are gradually added over the first month.42 After 3 months, the patient

A

Medial Epicondylectomy For medial epicondylectomy, a more limited skin incision is made just posterior to the epicondyle and over the cubital tunnel. The medial epicondyle is exposed, and the common flexor origin is reflected. The supracondylar ridge is removed, following the medial border of the trochlea. Excision of the medial epicondyle decompresses the cubital tunnel and allows the nerve to slide forward.44 The main advantage of this procedure is the minimal handling of the nerve itself. This procedure, like the in situ release, minimizes disruption to the perineural blood supply, which may enhance neurologic recovery. The main disadvantage, as with in situ decompression, is the inability to assess other proximal and distal sites of compression. This procedure must preserve the epicondylar origin of the medial collateral ligament anterior band, which is essential for stability, especially in the throwing athlete. A cadaveric study cautions that only 20% of the epicondyle can be excised without detaching a portion of the medial collateral ligament.45

B

Figure 19I-7   A fascial flap incision is outlined on the flexor muscle mass (A), and a fascial flap is elevated (B).

Elbow and Forearm 1315

A

B

Figure 19I-8   The ulnar nerve is transposed, and the subcutaneous transposition is completed.

This has led to a modification resulting in a minimal epicondylectomy. This minimal approach, compared with the historical partial medial epicondylectomy, has shown less valgus instability.46 A recent study reported no equivalent outcomes between minimal medial epicondylectomy and anterior subcutaneous transposition.47 It is still unclear which minimal procedure is best, but to elect this option, there should be no other more proximal or distal compression demonstrable by either physical examination, imaging studies, or electrophysiologic testing.

continues deep between the brachialis and brachioradialis ­muscles and exits the arm anterior to the lateral epicondyle, before dividing into the superficial and posterior interosseous nerve branches at the level of the radiocapitellar joint. The superficial radial nerve branch travels distally under the brachioradialis and emerges superficially between the brachioradialis and the extensor carpi radialis longus muscles. The posterior interosseous nerve branch enters the

RADIAL TUNNEL SYNDROME Etiology Many athletes perform repetitive forceful forearm rotations and may develop radial neuropathies around the elbow. Reported injuries occur in racquet sports, baseball, rowing, and track and field. Forceful pronation and supination in these athletes can compress the radial nerve on its path through the supinator and under the forearm extensor muscles.48 Lateral epicondylitis may overlap the symptoms of radial tunnel syndrome in 5% of cases, resulting from a coalescence of the supinator and extensor carpi radialis brevis origins to the lateral epicondyle.4 Radial neuropathy also results from extrinsic compression by tumors, vascular anomalies, and distal humerus fractures.49-52 Ganglion cysts in the proximal radioulnar joint may compress the posterior interosseous nerve and are amenable to arthroscopic evaluation.53

Anatomy In the distal third of the humerus, the radial nerve passes between the medial and lateral heads of the triceps, accompanied by the radial collateral artery. As it continues distally, it enters the anterior compartment by piercing through the lateral intermuscular septum about 10 cm above the lateral humeral epicondyle (Fig. 19I-9). The radial nerve

Figure 19I-9   Cadaveric dissection demonstrating the close proximity of the radial nerve to the intermuscular septum, where it enters the flexor compartment 10 cm above the lateral humeral epicondyle.

1316 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

proximal edge of the supinator at the arcade of Frohse, and then traverses the supinator until it emerges from its distal lateral margin. The posterior interosseous nerve continues distally under the extensor digitorum communis and on top of the abductor pollicis longus and extensor pollicis brevis muscle bellies, until it dives underneath the extensor pollicis longus to lie on top of the interosseous membrane. The radial tunnel is considered the course of the nerve from the point that it is located between the brachialis and brachioradialis muscles, to where it emerges at the distal end of the supinator.54 Proximally in the radial tunnel, fibrous bands overlying the radial head and capsule can be a compressive force on the nerve as it passes. The fibrous edge of the extensor carpi radialis brevis has been described as a potential compressive source as well.55 The most common site of radial nerve compression, however, is at the proximal entrance of the nerve into the supinator muscle, the arcade of Frohse. Spinner described the arcade of Frohse as a fibrous sling “arising in a semi-circular manner from the tip of the lateral epicondyle, its fibers arch downwards 1 cm and then gain attachment to the medial aspect of the lateral epicondyle just lateral to the articular surface of the capitellum.”56 The nerve can also be compressed as it travels through the supinator by the fascia that lines the superficial head of the muscle. The vascular leash of Henry, which is the broad leash of radial recurrent artery branches overlying the nerve in the radial tunnel, may also mechanically compress the nerve. Cadaveric dissections have demonstrated fibrous connections between the radial nerve and the radiohumeral capsule in 50% of cases, the superomedial margin of the extensor carpi radialis brevis in 72%, the arcade of Frohse in 87%, the distal border of the supinator in 65%, and the leash of Henry in 72%.57

History and Physical Examination The chief complaint associated with radial tunnel syndrome is generally pain at the dorsal aspect of the upper forearm, about 1.5 cm anterior and distal to the lateral epicondyle, an area distinctly distal to the typical location for lateral epicondylitis. In the athlete, activities involving forceful elbow extension or supination (discus and javelin, tennis, baseball) can initiate symptoms. Motor symptoms, when present, often demonstrate weakness of the extensor digitorum communis muscles and early fatigability. Sometimes pure posterior interosseous nerve injury will also be painful, and this is thought to be due to sensory fibers associated with nutrient arteries of the posterior interosseous nerve. On physical examination, the sentinel finding is tenderness to direct palpation over the supinator muscle dorsally. With the elbow extended, pain can be elicited with resisted supination. Pain can also sometimes be provoked by resisted middle finger extension with the elbow extended, but this test is not very sensitive. If the patient suffers from concomitant lateral epicondylitis, passive wrist and finger flexion with elbow extension will cause pain at the lateral epicondyle.58 Occasionally, the patient may report referred distal radiation of pain to the dorsal wrist, which is the termination of the sensory portion of the posterior interosseous nerve.

Treatment Nonoperative Treatment Conservative treatment includes anti-inflammatory medications, rest, and activity modification. A resting splint keeping the forearm in supination and the wrist slightly extended can be helpful as well. Following symptom resolution, a specific stretching regimen is employed.

Operative Treatment Dorsal Approach The incision begins at the lateral humeral epicondyle and continues 6 to 8 cm distally. The lateral cutaneous nerves of the forearm should be identified and protected. The extensor carpi radialis brevis and extensor digitorum communis muscle bellies are identified. The extensor carpi radialis brevis is retracted radially, and the extensor digitorum communis is retracted ulnarly. This will expose the underlying supinator muscle. The posterior interosseous nerve is identified at the arcade of Frohse. The supinator fascia is incised from its proximal to distal margins over the course of the posterior interosseous nerve. If the patient’s complaints are localized to the arcade of Frohse and the supinator, the dorsal approach offers distinct advantages. First, the approach to the supinator is direct and requires limited intermuscular dissection. Second, the radial recurrent vessels are not exposed or sacrificed. Third, this extensile approach allows access to the lateral epicondyle and extensor origin for concomitant treatment of epicondylitis.

Anterolateral Approach: More Proximal Lesion An incision is made starting just proximal to the radial head and lateral to the biceps tendon. The brachiora­ dialis muscle is retracted laterally, whereas the brachi­ alis, biceps, and pronator teres are retracted medially. The radial nerve is found proximally in the interval between the brachialis and the brachioradialis. The nerve is followed distally past the fibrous leading edge of the extensor carpi radialis brevis. At this point, the forearm is fully pronated and the wrist flexed. If the extensor carpi radialis brevis origin compresses the nerve in this position, it will need to be released. The radial nerve then travels underneath the radial recurrent artery vessels, which are ligated and cut. By this point, the radial nerve has bifurcated into the superficial sensory branch and the posterior interosseous nerve. The posterior interosseous nerve is followed to the arcade of Frohse. The supinator is incised, decompressing the posterior interosseous nerve. The anterolateral approach allows access to the entire length of the radial tunnel by inspection of all sites of possible compression. Therefore, this approach should be used when the localizing signs or symptoms are not restricted to the supinator arcade. After radial tunnel release, the wrist is splinted in neutral until the limb is comfortable, then exercise is begun usually after 7 to 10 days. Sport-specific conditioning can then be initiated.

Elbow and Forearm 1317

THE PRONATOR SYNDROME Etiology Median nerve compression around the elbow tends to pre­ sent after high-load repetitive activities that require welldeveloped musculature, causing compression of the median nerve or the anterior interosseous nerve branch.59 Athletes at risk compete in car racing and fast-pitch softball, which require repetitive forceful forearm flexion and pronation. Other extrinsic compression may result from tumors or anomalous muscles (Gantzer’s muscle, an accessory head of the flexor pollicis longus).60 Vascular anomalies, such as a persistent median artery, have also been reported to cause median nerve compression.61,62

Anatomy In the arm, the median nerve travels lateral to the brachial artery and medial to the short head of the biceps brachii muscle. As it reaches the elbow, the median nerve crosses anterior to the brachial artery and eventually lies medial to it before reaching the antecubital space. A small population of patients have an anomalous bony protuberance located about 5 cm proximal to the medial epicondyle called the supracondyloid process.63 Occasionally, a fibrous band, the ligament of Struthers, connects the supracondyloid ­process with the medial epicondyle, creating another possible compression focus for the median nerve.64,65 In the antecubital fossa, the median nerve is bordered laterally by the biceps tendon, anteriorly by the lacertus fibrosis, medially by the pronator teres, and posteriorly by the brachialis. The nerve continues between the two heads of the pronator teres and travels underneath the fibrous flexor digitorum sublimis arch. The median nerve can be compressed as it passes under the lacertus fibrosis, between the heads of the pronator teres, or underneath the flexor digitorum sublimis arch.66-69 The most common site of compression is the pronator teres, followed by the flexor digitorum sublimis arch, and finally the lacertus ­fibrosus.70 Motor branches rarely leave the median nerve proximal to the elbow.71 The first motor branch to arise from the median nerve is usually the branch to the pronator teres, and the second branch is typically the flexor carpi radialis branch. At the level of the pronator teres muscle, the anterior interosseous nerve branches off the median nerve before passing under the flexor digitorum sublimis arch, which is the most common site for anterior interosseous nerve compression.43

History and Physical Examination Athletes who present with proximal median nerve compression are heavily muscled and are involved in highforce repetition forearm pronation. Compression of the common median nerve at the pronator will result in a pain syndrome with minimal motor findings. If the area of compression is more distal around the anterior interosseous nerve, the symptoms may include a dull, deep ache in the volar forearm and complaints of hand weakness, especially

Figure 19I-10   The patient with anterior interosseous nerve syndrome will not be able to form an “O” between the tips of the thumb and index finger. Instead, the thumb interphalangeal joint and the index finger distal interphalangeal joint will extend.

with pinch, secondary to thumb and index finger paralysis of the long flexors. As with the other compressive neuropathies in athletes, symptoms generally worsen with activity and improve with rest. Physical examination can reveal clues about where the potential site of compression might be. Proximally, if the median nerve is compressed under the ligament of Struthers, pain can be elicited with resisted elbow flexion at 120 to 130 degrees.71 Compression under the lacertus fibrosus is likely to cause pain with resisted elbow flexion and with the forearm in pronation.68 Pain with resisted pronation would point to the pronator teres as the site of compression. This test should be done with the elbow extended and the wrist flexed to isolate the pronator teres. Finally, compression at the flexor digitorum sublimis arch can be tested by resisted middle finger flexion. In cases of anterior interosseous nerve syndrome, the patient will not be able to form an “O” between the tips of the thumb and index finger (Fig. 19I-10). Instead, the thumb interphalangeal joint and the index finger distal interphalangeal joint will assume an extended posture because of inability to exert force with the flexor pollicis longus and flexor digitorum profundis to the index ­finger.72

Treatment As with ulnar and radial nerve compression neuropathies, conservative treatment including anti-inflammatory medications, splinting, and activity modification should be employed initially and for up to 6 months. If there is an established paralysis, surgical exploration should be ­considered.

Operative Technique The skin incision begins in the distal arm about 5 cm above the elbow and along the medial aspect of the biceps muscle. The incision curves toward the lacertus fibrosus at the elbow crease and then continues distally over the

1318 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

flexor-pronator mass. If a supracondyloid process was seen by radiography, exploration should start proximally to identify the supracondyloid process, to check for the ligament of Struthers, and to release the ligament if present. The median nerve is traced distally to the lacertus fibrosus, the next potential compression site. The lacertus fibrosus is divided or partially excised, and the median nerve is inspected for compression effects like neuroma formation. Dissection proceeds through the two heads of the pronator teres. Antegrade dissection with magnification protects the specific muscle motor branches and the anterior interosseous emanating from the posterior and ulnar aspects of the nerve. The superficial humeral head of the pronator teres is elevated, and its insertion is separated from the deeper

ulnar head. The decompression is completed as the median nerve descends under the flexor digitorum sublimis arch. While doing this, the fibrous flexor digitorum sublimis arch is incised, freeing the nerve. Mobilization begins within 1 week. Sport-specific reconditioning progresses to patient tolerance and is prolonged, with the patient taking up to 6 months to regain full strength.

R eferences Please see www.expertconsult.com

C H A P T E R  

 20

Wrist and Hand S ect i o n

A

Wrist 1. The Adult Wrist Jack V. Ingari

The wrist in the adult athlete is amazingly complex. Understanding wrist anatomy and kinematics helps explain why this complex arrangement of bones, ligaments, tendons, and neurovascular structures is susceptible to athletic injury. This chapter section outlines the anatomy, kinematics, and physical examination of the wrist. A discussion of the various specific injuries, their diagnosis, treatment options, and author’s preferred method is presented with emphasis on the athlete as the patient and return to sports as the goal.

WRIST ANATOMY AND BIOMECHANICS The bony anatomy of the wrist, or carpus, is composed of eight carpal bones and their articulations with one another as well as with the radius, ulna, and metacarpals (Fig. 20A1-1),

A

B

The scaphoid, lunate, triquetrum, hamate, capitate, trapezoid, trapezium, and pisiform create a uniquely mobile, interconnected bony network that allows motion in a limitless number of planes, from flexion to extension, radial and ulnar deviation, and any combination of those. For example, the motion involved in throwing a dart exemplifies the combination of the wrist moving from relative dorsiflexion and radial deviation to a position of flexion and ulnar deviation in a fluid, almost effortless action.1,2 The easy fluid motion of the dart throw belies the complexity of muscle action, tendon excursions, and carpal bone movements that is occurring in the wrist, simultaneous with grip and release in the hand. The basic carpal anatomy and kinematics have been likened to an oval ring,3,4 a proximal and distal row5-7 and three columns, those being lateral, central, and medial columns,8 as well as combinations of these. Each of the carpal bones

C

Figure 20A1-1   A, Posteroanterior radiograph of the carpus, with each of the carpal bones identified. S, scaphoid; L, lunate; Tq, triquetrum; P, pisiform; Tz, trapezium; Td, trapezoid; C, capitate; H, hamate. B, Artist’s rendition of the dorsal view of the carpus. C, A three-dimensional computed tomogram depicting a volar view of the carpus. Notice how the hook of the hamate projects palmarly.

1319

�rthopaedic ����������� S �ports ������ � Medicine ������� 1320 DeLee & Drez’s� O

has a ligamentous attachment to the adjacent bone within its row. These are the intrinsic intercarpal ligaments, such as the scapholunate interosseous ligament (SLIOL) and lunotriquetral (LT) ligament joining the adjacent bones in the proximal row. These ligaments are stouter dorsally than palmarly. The extrinsic ligaments, conversely, provide connection between the carpals and the radius and ulna proximally as well as to the metacarpals distally. Extrinsic ligaments include the radiocarpal ligaments and the carpometacarpal ligaments. The radiocarpal ligaments are stouter palmarly than dorsally and are usually best seen from within the joint as in wrist arthroscopy.9 Viewed arthroscopically, the radial-most ligament is the radioscaphocapitate (RSC) ligament. Progressing ulnarly, next is the long radiolunate ligament (LRL) followed by the radioscapholunate (RSL). The RSL is directly in line with the articulation between the scaphoid and the lunate when viewed arthroscopically and is otherwise known as the ligament of Testut. It is primarily a vascular structure, contributing little structural support. Immediately ulnar to the RSL is the short radiolunate ligament (SRL). Continuing further ulnarly is the ulnolunate ligament followed by the ulnotriquetral ligament. All these ligaments constitute the palmar or volar radiocarpal ligaments. They are thickenings of the joint capsule that underscore the complexity of providing stability while allowing free, unhindered movement of the wrist. Pathology involving any of these ligaments, whether intrinsic or extrinsic, can lead to pain, limitation of motion, and eventually carpal arthritis and collapse.

VASCULAR ANATOMY The vascular supply to the wrist is from a rich network of anastomosing vessels that originate primarily from the radial artery, ulnar artery, and anterior interosseous arteries (Fig. 20A1-2). Gelberman’s classic cadaveric studies delineated three dorsal arches and three palmar arches that are longitudinally fed by the radial artery laterally and the

Figure 20A1-2   Vascular anatomy of the carpus. The proximal and distal palmar arches are part of a rich, anastomotic blood supply of the wrist. Bottom arrow, radiocarpal arch; middle arrow, deep palmar arch; top arrow, superficial palmar arch.

ulnar artery medially.10 These arches, along with several recurrent branches, form a rich network of vessels supplying the carpus and hand. On the palmar side, the most distal arch is the superficial palmar arch, formed by the main continuation of the ulnar artery with an anastomosis to the radial artery. The deep palmar arch is more proximal and is formed by the anastomosis of the radial artery and the deep branch of the ulnar artery. The superficial palmar arch can be visualized immediately distal to the distal edge of the transverse carpal ligament. Gelberman also described the arterial anatomy of the carpal bones themselves, and specifically noted that single vessels supply the scaphoid, capitate, and 87% of lunates examined, which explains why those three bones are at risk for vascular disruption with fractures or other trauma.10,11 The vascular anatomy of the wrist and the carpal bones most at risk for osteonecrosis was recently re-examined by Botte and colleagues, confirming Gelberman’s original work.12

PHYSICAL EXAMINATION OF THE WRIST Physical examination of the wrist should begin only after the history is complete. A detailed history of specific injury will direct the physical examination and aid in final diagnosis. Certain portions of the physical examination are important in every wrist evaluation and are discussed here. Other special wrist tests are linked to specific injuries and are discussed along with those injuries. Inspection of the wrist is the first task. Visible evidence of swelling, ecchymosis, or skin changes can be important clues in evaluation of wrist injury. Next, assessment of active and passive range of motion of the wrist should be recorded and compared with the contralateral wrist. Flexion, extension, radial deviation, ulnar deviation, pronation, and supination should all be measured with a goniometer. Limitation in the range of motion compared with the contralateral wrist adds significant information in ultimate diagnosis. Asking about pain at the extremes of motion is also important when assessing wrist motion. Once the range of motion is recorded, palpation of the wrist, using the index finger of the examiner, further hones diagnosis. Palpation should proceed from nonpainful areas to areas of maximal tenderness last, because once tenderness is elicited, the willingness of the patient to allow further palpation may become limited. Knowledge of specific anatomy can greatly aid in diagnosis. The scapholunate ligament, for example, lies about 1 cm distal to Lister’s tubercle, and by palpating Lister’s tubercle, then moving the examiner’s index finger 1 cm distally, pressure is placed directly over the scapholunate ligament area. If acutely injured, tenderness in that locale is readily elicited with the examiner’s finger. Similarly, the anatomic snuffbox and the tuberosity of the scaphoid are tender in the face of acute scaphoid fractures, which can accompany ligamentous injury. Precise palpation of the carpal bones and stressing their articulations is a great aid in diagnosis. Several of the physical examination maneuvers are described later with the individual injuries. In assessing vascular flow to the hand, Allen’s test can be used at the wrist or more distally at the base of each digit. In the classic Allen’s test at the wrist, the patient makes a

Wrist and Hand 1321

A

B

C

tight fist, and the examiner then occludes the radial and ulnar artery with digital pressure. The patient then opens the hand, and pressure is released from the radial artery to assess vascular fill, with an associated return of normal palmar color if the artery is patent. The test is repeated, with pressure released from the ulnar artery to assess its arterial flow into the hand (Fig. 20A1-3).

RADIOGRAPHIC EVALUATION OF THE WRIST The minimal radiographs that should be obtained are two orthogonal views, an anteroposterior (AP) or posteroanterior (PA) view and a lateral view of the wrist. A 45-degree semipronated view and an AP view in ulnar deviation to delineate the scaphoid are also valuable to detect subtle fractures not seen on the AP view. A clenched-fist AP view can also be extremely useful in assessing carpal instability and possible ligament injury. Whenever doubt exists, a radiograph of the contralateral wrist is also warranted because widening of the scapholunate interval, for example, may be a bilateral finding and not indicative of specific injury. Additional studies should be directed by history and physical examination findings and should be used to confirm or deny the presence of a suspected injury, rather than used as a “shotgun” to aid in diagnosis. Magnetic

Figure 20A1-3   A, Allen’s test with both radial and ulnar artery occluded. Notice the blanched, mottled appearance of the palm. B, In a negative, or normal, Allen’s test, when pressure is released from the radial artery, arterial flow is restored to the hand. C, Similarly, pressure released from the ulnar artery restores arterial flow to the hand in a normal, or negative, Allen’s test.

resonance imaging (MRI), with intra-articular administration of gadolinium, is an excellent study to assess the presence of intrinsic ligamentous injury as well as to delineate pathology of the triangular fibrocartilage.13,14 The gadolinium injected into the radiocarpal joint is seen to leak into the midcarpal joint or into the distal radioulnar joint if ligament injury or injury to the triangular fibrocartilage is present. This magnetic resonance arthrogram (MRA) is an extremely useful tool but, again, only when used in conjunction with history, physical examination, and plain radiographic findings that suggest specific injury. Nuclear medicine studies, like bone scintigraphy (bone scan), can be helpful as a sensitive indicator of acute injury, such as occult scaphoid fracture, but may lack the specificity to define exact injury.15,16 Bone scans should be ordered no less than 24 to 48 hours after injury to eliminate the possibility of false-negative results. Finally, wrist arthroscopy has emerged as an outstanding diagnostic tool, as well as offering therapeutic benefit. A thorough discussion of wrist arthroscopy is found in the wrist arthroscopy ­section of this chapter.

LIGAMENTOUS INJURIES In the early 1980s, Mayfield classified the perilunate instability patterns that are commonly seen in the wrist.17,18 Mayfield described a four-stage injury pattern, which

�rthopaedic ����������� S �ports ������ � Medicine ������� 1322 DeLee & Drez’s� O

Figure 20A1-4   Measuring the scapholunate angle. Step 1. Draw a line from the volar lip to the dorsal lip of the lunate (1). Step 2. Draw a second line perpendicular to line 1, essentially bisecting the lunate (2). Step 3. Draw a third line along the volar cortex of the scaphoid (3). The scapholunate angle is the angle measured between lines 2 and 3 (green arrow). In this example, the scapholunate angle is 100 degrees, which is abnormal. Normal is 30 to 60 degrees.

occurs when the wrist undergoes progressive ulnar deviation, carpal supination, and dorsiflexion, combined with an axial load. Simply stated, a forceful fall onto the palm of the hand, or forced extension of the wrist, can transmit forces through the carpus, causing perilunate injury patterns. The most common injury pattern is scapholunate ligament disruption, which is classified as a Mayfield I injury.18 With disruption of the scapholunate ligament, the lunate tends to fall into a dorsiflexed posture owing to its connection to the triquetrum, which tends to pull the lunate into extension. The scaphoid, now ligamentously detached from the lunate, assumes a flexed posture owing to the axial load transmitted to the distal scaphoid from the trapezium. Seen on a lateral radiograph, the scapholunate angle, normally between 30 to 60 degrees, is increased, often to greater than 70 degrees (Fig. 20A1-4). The lunate on the lateral view is seen to be dorsiflexed in the classic dorsal intercalary segmental instability (DISI) posture. When viewing the lateral radiograph, assessing the position of the lunate is key. The capitolunate angle is also assessed on the lateral radiograph and should be less than 15 degrees in a normal wrist (Fig. 20A1-5). A capitolunate angle greater than 15 degrees may be an indication of ligamentous laxity or of midcarpal instability. DISI is most often due to scapholunate disruption. The intercalary segment refers to the lunate, which is tethered between the scaphoid and triquetrum in the proximal carpal row. The DISI pattern refers to the position of the lunate, or intercalary segment, as seen on a lateral radiograph (Fig. 20A1-6). Additionally, on an AP or PA radiograph, the scapholunate interval, normally less than 3 mm, may be increased. The foreshortened, flexed scaphoid has a visible cortical ring sign, which is a manifestation of “looking down the barrel” of the scaphoid tuberosity. The lunate on the AP view looks elongated, almost

Figure 20A1-5   Measuring the capitolunate angle. Step 1. Draw a line (1) connecting the volar and dorsal lips of the lunate. Step 2. Draw a second line (2) perpendicular to line 1. Step 3. Draw a line (3) down the central long axis of the capitate. The capitolunate angle in measured between lines 2 and 3. Normal is 0 to 15 degrees. This example measures 20 degrees, which is abnormal.

assuming a triangular shape (Fig. 20A1-7). With progressive injury, the capitolunate and lunotriquetral intercarpal ligaments become disrupted, indicating a Mayfield II and Mayfield III perilunate injury pattern, respectively. Finally, in a Mayfield IV injury, the lunate is completely dislocated relative to the radius and rotates volarly into the space of Poirier, tethered by the stout radiocarpal ligaments, namely the long and short radiolunate ligaments and the ulnolunate ligament (Fig. 20A1-8). Rarely, the radiocarpal ligaments attached to the lunate can be disrupted, allowing further migration of the lunate in the volar distal forearm. The Mayfield classification is useful in terms of delineating the severity of injury as well as directing treatment options. In the case of complete lunotriquetral ligament disruption, the lunate as seen on the lateral radiograph assumes a palmarly flexed posture as the intact tether to the scaphoid tends to pull the lunate into a flexed position because the axial loads on the scaphoid tend to cause flexion. This palmarly flexed position of the lunate seen on lateral radiograph defines the volar intercalary segmental instability (VISI) pattern, which is most commonly due to lunotriquetral ligament disruption (Fig. 20A1-9).

Clinical Presentation and History A high index of suspicion is important in recognizing any but the most obvious of the perilunate injuries. A fall onto the palm of the hand with the wrist in a dorsiflexed position may, in combination with various carpal fractures, lead to perilunate instability patterns.19 History, along with physical examination and appropriate radiographic evaluation, should lead to the correct diagnosis. It is important to remember that ligament injuries can be

Wrist and Hand 1323 Figure 20A1-6   A, Dorsal intercalary segmental instability (DISI) posture of the lunate as seen on a sagittal plane magnetic resonance image in a patient with scapholunate dissociation. The arrow is pointing to the intercalary segment, the lunate, which is seen to tip dorsally. B, Artist’s rendition of DISI pattern.

A B Figure 20A1-7   A, Anteroposterior radiograph depicting a scapholunate ligament injury. Notice the foreshortened scaphoid, with a cortical ring sign (blue arrow) as well as the widened scapholunate interval (black line). B, A lateral radiograph of the wrist, demonstrating a scapholunate ligament disruption. Notice the dorsiflexed posture of the lunate and the increased scapholunate angle.

A

B

Figure 20A1-8   Posteroanterior and lateral radiographs of a Mayfield IV perilunate dislocation. Notice the difficulty of assessing the dislocation on the posteroanterior view alone. On the lateral radiograph, the lunate is completely dislocated, consistent with a Mayfield IV injury.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1324 DeLee & Drez’s� O Figure 20A1-9   A, Volar intercalary segmental instability (VISI) deformity as seen on the lateral radiograph. Notice how the lunate (arrow) appears to be tipping palmarly. This defines VISI. B, Artist’s rendition of VISI deformity. The lunate is the intercalary segment.

A

B

incurred even in the face of fracture, whether it is fracture of the distal radius or various carpal fractures. Geissler and associates demonstrated arthroscopically that 19 of 60 patients with distal radius fracture had associated scapholunate ligament disruption.20 The scapholunate injury pattern is the most common of the perilunate injuries described by Mayfield and is covered first.18 Perilunate injury has been described in children but is decidedly rare. It merits the same level of treatment in children as in adults because it is a high-energy injury with severe longterm consequences if untreated. These pediatric ligament injuries are discussed in the pediatric wrist section of this chapter.

Scapholunate Ligament Injury Classification No simple classification exists for all the parameters of scapholunate injury. As previously mentioned, a scapholunate injury defines a Mayfield I pattern. Equally important, however, is the timing of injury, whether it is a complete or partial injury, and the etiology of injury.

Time from Injury Time from injury affects the ligaments’ ability to heal and certainly affects surgical repair outcomes. Scapholunate injuries are considered acute if they present less than 3 weeks from time of injury. Subacute injury refers to scapholunate disruptions presenting between 3 weeks and 3 months after injury, and chronic injuries present more than 3 months after injury.21 Left untreated, the natural history of scapholunate injury leads to a predictable pattern of carpal collapse and arthrosis, the so-called scapholunate advanced collapse (SLAC) pattern.7,22-26 In addition, whether the ligament injury is partial or complete is important for both treatment options and prognosis after surgery.27

Clinical Presentation and History The clinical presentation often involves a swollen, painful wrist, with a history of a fall onto the palm of the hand with the wrist typically in dorsiflexion. Initial radiographs may show evidence of associated fractures of the distal radius or carpal bones, or there may be no fractures in a purely ligamentous injury. A fracture of the scaphoid does not preclude the possibility of scapholunate injury because the two can present simultaneously.28 In the absence of fractures, patients may be told they have a “wrist sprain,” which may be treated minimally with a splint or short cast. Although simple immobilization may be effective in some injuries, careful physical examination and radiographic examination are warranted to assess for more serious injury to the scapholunate ligament or other pathology.

Physical Examination The hallmark of physical examination is evaluation of static tenderness, coupled with directed evaluation of the wrist in various positions. History of a fall with the wrist in extension or forced dorsiflexion of the wrist from a sports injury is classic. Inspection of the wrist may be normal, or there may be swelling over the dorsoradial wrist in the area of injury. Range of motion is typically limited, especially in extension, with pain elicited at the extremes of extension and radial and ulnar deviation. The maximal point of tenderness lies over the scapholunate ligament, which is palpable about 1 cm distal to Lister’s tubercle on the dorsal aspect of the wrist. Watson described a classic maneuver, bearing his name, which should be assessed in the physical examination of suspected scapholunate injury.5-7 In Watson’s maneuver, the thumb of the examiner’s hand is placed directly over the tuberosity of the scaphoid, palmarly, and the wrist is passively moved to full ulnar deviation. With thumb pressure exerted on the scaphoid tuberosity, the wrist is then passively moved

Wrist and Hand 1325

A

C

B

Figure 20A1-10   A, Watson’s test depicting the initial wrist position in ulnar deviation with the examiner’s thumb exerting dorsally directed pressure on the tuberosity of the scaphoid. B, Watson’s test depicting the passive motion to radial deviation with continued pressure on the scaphoid tuberosity. In scapholunate dissociation, the scaphoid is subluxated dorsally onto the dorsal rim of the radius. C, Watson’s test completed. Removing thumb pressure from the scaphoid with the wrist held in radial deviation. In a positive test the scaphoid will “clunk” to its reduced position, with associated pain.

into radial deviation. In the face of scapholunate ligament disruption, the scaphoid subluxates dorsally on the dorsal rim of the radius. Finally, the thumb pressure of the examiner’s hand is released. A painful, palpable “clunk,” sometimes audible, accompanies the movement of the scaphoid into its reduced position in the scaphoid fossa of the radius. A positive Watson’s test, then, is a maneuver in which the reduction of the scaphoid from its dorsally subluxated position to its reduced position causes the painful clunk associated with scapholunate ligament injury (Fig. 20A1-10).

Radiographic Examination Radiographic examination for scapholunate ligament injury should include AP and lateral radiographs, a semipronated 40-degree oblique view, and a clenched-fist AP at a minimum. A PA view in ulnar deviation with the beam centered on the scaphoid is also quite helpful. On the AP view, several notable findings are seen. First, foreshortening of the scaphoid is evident. In complete scapholunate disruption, the scapholunate interval, normally 2 to 3 mm, is widened. The lunate appears triangular in shape owing to its extended posture. Lastly, an en face view of the scaphoid tuberosity, reminiscent of looking down a gun barrel, demonstrates the classic cortical ring sign. The lateral radiograph in complete scapholunate dissociation shows a DISI pattern with the lunate in an extended posture, “tipped” dorsally. It is this dorsally tipped lunate on the lateral radiograph that is the hallmark of the DISI pattern. In addition to plain films, MRI with intra-articular administration of gadolinium can be helpful in delineating ligamentous injury.21,29,30 The gadolinium is injected into the radiocarpal joint, and following its administration, an MRI is obtained. If the scapholunate ligament is torn, the gadolinium will be seen to leak into the ­midcarpal joint

traversing the disrupted scapholunate joint. This MRA is a helpful adjunct in the evaluation of wrist ligament injury.

Treatment Options The treatment of scapholunate ligament injury depends on the chronicity, severity, and etiology of the tear. In acute sports injuries to the scapholunate ligament, restoration of the normal anatomy is the goal. Closed reduction with casting is ineffective in this regard, and most authors have advocated operative intervention to restore anatomy and function in the face of acute scapholunate ligament injuries.21,31-35 Arthroscopic débridement of partial tears to the scapholunate ligament remains among the most advocated procedures.27 With complete scapholunate ligament tears, percutaneous pinning of the scaphoid to the lunate, with or without open reduction and ligament repair, are the mainstays of treatment. In subacute and chronic injury to the scapholunate ligament, the ability of the ligament to heal primarily is suboptimal, and many surgical treatment options for this difficult problem have been advocated, all with varying, but overall less than perfect, long-term outcomes.21,35-45 A dorsal capsulodesis, as described by Blatt, and some form of ligamentous tether to act as a checkrein for the scaphoid to prevent hyperflexion, using the dorsal intercarpal ligament as described by Linscheid and Dobyns, remain popular for this unsolved problem.21 Brunelli and Brunelli advocated using the flexor carpi radialis tendon weaved through the scaphoid as a means of restoring normal carpal alignment in chronic scapholunate dissociation.46 Rosenwasser and colleagues have described a procedure in which a Herbert screw is used to reduce the scaphoid to the lunate while soft tissue healing occurs.47 This reduction-association of the scapholunate (RASL) procedure is relatively simple to do and is gaining in popularity (Fig. 20A1-11).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1326 DeLee & Drez’s� O

A

B

Figure 20A1-11   A, Radiograph showing a compression type screw reducing and associating the scaphoid and lunate in the reduction-association of the scapholunate (RASL) procedure, 3 months after surgery. A small suture anchor is also visible in the scaphoid as part of the ligament reconstruction (arrow). B, The RASL as seen on lateral radiograph 3 months after surgery. The arrow shows the suture anchor embedded in the scaphoid. Notice the normal scapholunate angle of 48 degrees.

Author’s Preferred Method Acute Scapholunate Injury

For acute scapholunate ligament tears, my preferred treatment depends on whether the ligament injury is complete or partial. For incomplete symptomatic scapholunate ligament tears, I prefer arthroscopic débridement of the partially injured scapholunate ligament. (See Section C of this chapter for further detail.) In some incomplete tears, there may be significant attenuation or stretching of the ligament, and in these cases, I add percutaneous pinning of the scaphoid to the lunate to the arthroscopic débridement to reduce the scapholunate interval and to correct any rotational deformity of the scaphoid. In acute complete scapholunate ligament tears, I firmly believe that restoration of normal anatomy is the primary goal. For this reason, I begin with wrist arthroscopy to confirm the complete nature of the tear. If the tear is complete, I proceed with an open procedure to repair the torn ligaments. Specifically, through a 6-cm dorsal midline approach, I identify the torn ligament, which is usually evident as a stout ­dorsal ligament with rupture typically from its scaphoid attachment. Next, a single 0.062-inch K-wire is placed into the scaphoid, and a second 0.062-inch K-wire into the lunate. Using these “joysticks,” the scaphoid and lunate are reduced, such that the bones are in close approximation and in the correct rotational alignment. The rotation involves dorsiflexion of the scaphoid with palmar flexion of the lunate using the joysticks as moving levers. With the bones in this reduced position, I place multiple (three or four) 0.045-inch K-wires between the scaphoid and lunate to maintain their reduced positions and use a mini C-arm intraoperatively to verify adequate pin placement (Fig. 20A1-12A). Once the bones are in acceptable alignment, I proceed with ligament repair using small suture anchors into the scaphoid to reattach the dorsal (stoutest) portion of the scapholunate ligament. I then close the ­ dorsal wrist capsule with nonabsorbable suture

(2-0 Ethibond). Subcutaneous tissues are reapproximated with 4-0 absorbable suture, and the skin is closed with a running subcuticular suture, such as 4-0 Prolene. The pins are left outside the skin to facilitate later removal in the clinic (see Fig. 20A1-12B). Postoperative Care. A short arm thumb spica splint is placed in the operating room. Sutures are removed between 10 and 14 days after surgery, and a short arm thumb spica cast replaces the splint, leaving the pins in place. The exposed pins are padded with felt to prevent motion in the cast. Six weeks after surgery, the pins are removed, and radiographs are obtained. A commercially available, removable short arm thumb spica splint is applied and directed to be worn for an additional 6 weeks for any activity requiring hand or wrist motion. The splint is to be worn at night but can be removed for showering, eating, watching television, or other similar sedentary events. Gentle range of motion of the wrist is also started at 6 weeks. Exercises can be shown to the patient, or the patient can be sent to a therapist to begin gentle wrist range of motion. At 12 weeks, the splint is discontinued, and grip strengthening exercises are begun, along with continued range of motion. Criteria for Return to Sports. Three months after surgery, the patient’s range of motion and strength are evaluated, along with subjective complaints of pain. The primary determinant of return to sports activity is the patient’s ability to perform sports maneuvers of the hand and wrist relatively pain free. Some sports, such as golf, bowling, and racquet sports, may require a wrist splint or brace for added support. Protective splints are advised for skiers, snowboarders, skateboarders, and others where falling is common in the activity. Contact sports can be resumed when strength and pain allow. For some ­patients, that may be at 3 months, whereas for others, 6 months or longer may be required before resuming strenuous sports.

Wrist and Hand 1327

Author’s Preferred Method—cont’d

A

B

Figure 20A1-12   A, Multiple pins from the scaphoid into the lunate seen radiographically. B, All pins are left outside the skin to facilitate later removal.

Chronic Scapholunate Ligament Dissociation

As mentioned previously, no consensus on the best method of treatment of subacute and chronic scapholunate ligament injury exists. A recent outcome study by Moran and colleagues suggests that final wrist range of motion, strength, and Mayo wrist scores were similar for two groups of patients undergoing a Berger-type capsulodesis or Brunelli’s tenodesis procedure.48 Both groups averaged “fair” in the Mayo scoring system, and both groups had limitations in wrist flexion at long-term follow-up. The RASL procedure is attractive for its technical simplicity, but as yet, no long-term outcome studies are available. At the present time, I prefer a capsulodesis using the dorsal intercarpal ligament attached to the distal radius, with open reduction and percutaneous pin fixation, using multiple pins across the scapholunate interval to promote a pseudarthrosis between the scaphoid and lunate. The joystick method, as described in the acute scapholunate treatment, is used to reduce the scaphoid and lunate. A rongeur is used to débride the scapholunate interval and to partially decorticate the apposing surfaces of the scaphoid and lunate until bleeding bone is visible, to help promote a pseudarthrosis between the two bones. A true fusion between the scaphoid and lunate is difficult to achieve owing to the small abutting surface areas, but a stable pseud­ arthrosis is acceptable. Three or, preferably, four 0.045-inch K-wires are used from the scaphoid into the lunate, and if any dorsal ligament remains, it is repaired as well. The dorsal intercarpal ligament is elevated from its attachment to

Salvage Procedures Degenerative changes predictably occur after chronic scapholunate dissociation.6,7,24 These changes are referred to as scapholunate advanced collapse, or SLAC. Watson described three stages of SLAC wrist as seen on plain radiograph (Table 20A1-1). In stage I, arthrosis occurs

the triquetrum and is sutured to the proximal edge of the dorsal wrist capsule, or attached by suture anchor to the distal radius in line with the long axis of the scaphoid. The remaining capsule is closed with 2-0 Ethibond, and routine subcutaneous and skin closure follows. Postoperative Care. At the time of surgery, a short arm thumb spica cast is placed. The patient returns at day 10 to 14 for suture removal and placement of a short arm thumb spica cast, incorporating the pins, padded with felt. Six weeks after surgery, the pins are removed, and a short arm thumb spica cast is placed for an additional 6 weeks. This added immobilization optimizes the tissues ability to form a fibrous pseudarthrosis between the scaphoid and lunate. At 3 months, the cast is removed, and range of motion and grip strengthening exercises are begun. Criteria for Return to Sports. After 3 months in a cast, the patient will need an additional period of usually at least 6 weeks of therapy to include range of motion and grip strengthening, before returning to sports. In the best of situations, the patient will be pain free and have nearly complete return of range of motion. As stated previously, however, chronic scapholunate repair most often results in a fair outcome, and some persistent disability due to pain or loss of motion is expected. If the athlete can participate in his or her sport relatively pain free, then a full clearance is given 4 to 6 months after surgery, with recommendation for wrist support, especially for contact sports and sports with an inherent risk for falling, such as skiing, skateboarding, and snowboarding.

between the distal pole of the scaphoid and the radial styloid. With progression to stage II SLAC wrist, the entire scaphoid fossa and matching proximal surface of the scaphoid become arthritic and incongruous. In the final stage, stage III SLAC wrist, the capitate migrates proximally into the widened scapholunate interval, and the proximal pole

�rthopaedic ����������� S �ports ������ � Medicine ������� 1328 DeLee & Drez’s� O

  TABLE 20A1-1  Scapholunate Advanced Collapse (SLAC) Wrist

SLAC Type

Radiographic Features

Treatment

I

Osteophytes between radial styloid and distal pole of scaphoid Arthritic changes involving entire scaphoid fossa and proximal scaphoid Same as SLAC II with proximal migration of capitate with arthritic changes to proximal capitate

Radial styloidectomy

II III

Proximal row carpectomy or four-corner fusion Four-corner fusion or total wrist fusion

of the capitate is now involved in the arthritic process. ­Salvage options for stage I SLAC wrist include radial styloidectomy, although this does not address the radioscaphoid incongruity and is at best, a temporizing measure. In stage II SLAC wrist, options include four-corner fusion

and scaphoidectomy, the so-called SLAC wrist operation, as well as proximal row carpectomy, in which the scaphoid, lunate, and triquetrum are excised, allowing the capitate to articulate with the distal radius. Long-term results of each of these operations are remarkably similar with pain relief expected, but the patients lose nearly half of the flexion and extension of the wrist compared with the contralateral side.49,50 Wyrick and associates found patient satisfaction to be higher with proximal row carpectomy compared with four-corner fusion, with fewer complications in the proximal row carpectomy cohort.49 For stage III SLAC wrist, with the proximal pole of the capitate now involved with arthrosis, proximal row carpectomy is a less attractive option. Salomon and Eaton, however, have advocated doing a modified proximal row carpectomy, resecting the proximal capitate and interposing the dorsal wrist capsule between the partially excised capitate and the lunate fossa, with good results even in the face of proximal capitate articular cartilage loss.51,52 Other options include arthrodeses, including four-corner fusion with scaphoidectomy or total wrist arthrodesis.

Author’s Preferred Method Stage I SLAC

For stage I SLAC wrist, I prefer to perform a radial styloidectomy along with a capsulodesis and pinning as described previously for chronic scapholunate dissociation. The patient is warned that results are variable and further surgery may be necessary if pain and symptoms persist despite styloidectomy and capsulodesis. If the patient requires further surgery, I proceed with proximal row carpectomy, described later in the treatment of stage II SLAC. Stage II SLAC

The risk for complications, including nonunion and hardware failure, is higher with four-corner fusion than proximal row carpectomy. In addition, long-term outcomes of proximal row carpectomy have been shown to be at least equal to, if not better than, four-corner fusion with scaphoidectomy.49,50,53,54 For these reasons, as well as the relative technical ease of the proximal row carpectomy compared with four-corner fusion, I prefer proximal row carpectomy for stage II SLAC wrist (Fig. 20A1-13A and B). Using a dorsal longitudinal midline approach, I incise the extensor retinaculum over the third dorsal compartment and retract the extensor pollicis longus (EPL) tendon radially. Either a longitudinally or radially based Mayo-type capsulotomy is performed. The proximal carpal row is then removed, one carpal bone at a time. Threaded 0.062-inch K-wire ­joysticks into the carpals greatly facilitate their removal (see Fig. 20A1-13C). Occasionally, a rongeur is used to help with bone excision. One technical caveat is to take care to leave the volar radioscaphocapitate ligament intact to prevent ulnar migration of the carpus after surgery. After excision of the scaphoid, lunate, and triquetrum, the capsule is closed with nonabsorbable 2-0 suture. The EPL is left to “fly free”

rather than being returned to its third compartment, and clinically this has proved quite acceptable. The subcutaneous tissues are approximated with 2-0 or 4-0 absorbable suture, and the skin is closed with a running 4-0 Prolene subcuticular suture. Postoperative Care. A volar forearm splint is placed at the time of surgery, and this is removed 10 to 14 days after surgery, along with the running Prolene suture. The patient is placed in a removable wrist splint for protection when active. The splint is removed for the patient to begin range of motion and grip strengthening exercises. Six weeks after surgery, the splint can be discontinued based on the patient’s comfort. Criteria for Return to Sports. Six weeks after surgery, the patient is seen in clinic and cleared for sports participation if pain is minimal or absent and range of motion is about 50% of the contralateral wrist. Based on the athlete’s particular sport, a wrist brace may be recommended for protection for an additional 6 weeks. Stage III SLAC

Stage III SLAC wrist represents a chronic disease state, and the competitive athlete will rarely present with advanced stage III SLAC because previous symptoms will likely have caused him or her to seek care or stop competing. In the rare event of an athlete presenting with newly symptomatic stage III SLAC, the options are limited. If the proximal lunate and lunate fossa are spared from arthritic change, I prefer four-corner fusion to retain some wrist motion, while addressing the arthrosis of the proximal capitate by including it in the fusion mass (Fig. 20A1-14). For those with pancarpal changes, including carpal collapse and arthrosis, I prefer total wrist arthrodesis, but this may prove to be limiting for

Wrist and Hand 1329

Author’s Preferred Method—cont’d

B

A

C Figure 20A1-13   A, Scapholunate advanced collapse (SLAC) II wrist. The entire scaphoid fossa is involved with arthritic change. A styloidectomy is not enough here, and consideration of a proximal row carpectomy or four-corner fusion is needed.   I prefer proximal row carpectomy. B, A proximal row carpectomy for stage II SLAC wrist. Notice the radial styloidectomy as well. Although the pisiform appears to abut the ulnar styloid, it is well palmar to the styloid, so no impingement occurs ulnarly.   C, Threaded 0.062-inch K-wire inserted into scaphoid to facilitate removal as part of the procedure.

Figure 20A1-14   Four-corner fusion with scaphoidectomy for scapholunate advanced collapse (SLAC) III to retain some wrist motion. The capitate is contained in the fusion mass, which negates any arthritic changes involving the proximal capitate.   A combination of pins and headless compression screws is used here. This is early because pins are removed 6 weeks after surgery.

many athletes, and they will need to be counseled about the lack of wrist motion after surgery (Fig. 20A1-15). I prefer to place all my wrist fusion patients in a trial short arm cast for a period of no less than 4 weeks to allow them to assess their lack of wrist motion and potential pain relief. I have found this method to be an excellent screening tool. Those patients who are highly satisfied with the cast because of pain relief and are not disturbed by the lack of wrist motion are good candidates for wrist fusion. Conversely, the patients who dislike the cast are not ready for fusion by their own standards. I have no experience using Eaton’s technique of proximal capitate resection and dorsal capsular interposition, although it may represent a motion-sparing option for someone with stage III SLAC who is concerned about loss of motion associated with total wrist fusion. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1330 DeLee & Drez’s� O

Author’s Preferred Method—cont’d

A

B

Figure 20A1-15   A, Scapholunate advanced collapse (SLAC) III wrist with diffuse carpal involvement. Notice how the capitate has significant proximal arthrosis and proximal migration. B, Wrist fusion for SLAC III, using a Synthes precontoured plate.

Lunotriquetral Ligament Injury Injury to the lunotriquetral ligament can be the result of a sports injury, such as a fall onto the palm of the hand. Reagan and coworkers defined lunotriquetral sprains in 1984 and stated they were the result of hyperextension and twisting of the wrist.55 Biomechanical cadaveric studies by Viegas have demonstrated that the lunotriquetral ligament may be injured in an isolated fashion, without involvement of the scapholunate ligament.56,57 Any sport involving throwing, catching, or having a predilection for falling onto the hand or wrist can result in a lunotriquetral injury. Certainly contact sports must be considered a risk factor as well.

Classification Distinguishing between complete and incomplete ligament tears, as well as between acute and chronic ligament injury, is probably the simplest, most effective way of classifying lunotriquetral ligament injury, with implications for treatment options as well as prognosis. If associated with a scapholunate ligament dissociation, the combination of scapholunate progressing to the ulnar lunotriquetral ligament constitutes a Mayfield III perilunate injury.58 However, if a lunotriquetral ligament injury presents as an isolated entity, Viegas proposed an injury pattern based on cadaveric studies showing that the ulnar side of the carpus, specifically the lunotriquetral ligament, may be injured primarily.56,57

Clinical Presentation and History Falls onto the palm of the hand and forceful hyperextension of the wrist with an associated twisting moment are the most likely mechanisms resulting in lunotriquetral

ligament injury. History of an appropriate mechanism of injury, coupled with ulnar-sided wrist pain, should lead to consideration of lunotriquetral ligament injury. Occasionally, patients complain of painful snapping or clicking with wrist extension and ulnar deviation. The diagnosis remains difficult because ulnar-sided wrist pain offers myriad differential diagnoses, including triangular fibrocartilage complex tears, fracture of the pisiform, fracture of the hook or body of hamate, ulnocarpal impaction syndrome, snapping extensor carpi ulnaris, distal radioulnar joint arthrosis or instability, and dorsal sensory branch of the ulnar nerve neuritis, just to name some of the more common causes. Perhaps because of the extensive list of etiologies for ulnar-sided wrist pain, as well as trials of conservative splinting or casting, the diagnosis of acute lunotriquetral ligament injury remains elusive. I have never diagnosed an acute injury of the lunotriquetral ligament in my practice and have treated only subacute and chronic injuries.

Physical Examination Various authors have described the physical examination for lunotriquetral ligament injury. Reagan and coworkers described a ballottement test, in which the lunate is held dorsally and palmarly by the examiner’s index finger and thumb, with subsequent dorsal to palmar ballottement of the triquetrum eliciting crepitus and painful motion in the face of lunotriquetral ligament injury.55 Kleinman proposed a similar test in which the examiner exerts dorsal to palmar pressure on the lunate and simultaneously palmar to dorsal pressure on the pisiform, and thereby the triquetrum.59 In a positive Kleinman’s shear test, the patient experiences pain and crepitus. Although these tests certainly are provocative for pain associated with lunotriquetral ligament sprains, they may be nonspecific, eliciting pain with

Wrist and Hand 1331

­ isiform ­fracture, pisotriquetral arthritis, or other causes of p ulnar-sided wrist pain.

later in Section C of this chapter, is an excellent means of identifying lunotriquetral ligament tears as well.60-65

Radiographic Examination

Treatment Options

The notable finding on radiographic examination in static lunotriquetral dissociation is a volar intercalary segmental instability (VISI) pattern on the lateral radiograph, which connotes a palmar tilting of the lunate. The lunotriquetral interval may appear widened on an AP view as well. In partial lunotriquetral ligament injury or dynamic injuries that are manifested only with certain motions, the radiographs may appear entirely normal. I have used MRA of the wrist to help identify lunotriquetral ligament injury. Finally, wrist ­ arthroscopy, covered

Once recognized, lunotriquetral ligament injuries have multiple treatment options. For partial tears, simple immobilization in a cast or splint may be effective.62 Weiss and colleagues advocate arthroscopic débridement for partial tears with percutaneous pinning.65 For complete tears, arthroscopic débridement and percutaneous pinning, ligament repair, ligament reconstruction, ulnar shortening, and lunotriquetral arthrodesis have all been performed by various authors.27,62,64,65

Author’s Preferred Method As previously stated, most complete lunotriquetral ligament injuries present late, and as such, I have found lunotriquetral arthrodesis to be an effective procedure to alleviate pain and allow return to full activity. Through a curved dorsoulnar skin incision paralleling the course of the dorsal sensory branch of the ulnar nerve, the nerve is carefully identified but left undisturbed. I incise the interval between the fourth and fifth extensor compartments, elevating the tendon of the extensor digiti minimi, with its retinaculum intact, off the dorsal capsule. Next, I longitudinally incise the dorsal capsule directly over the lunotriquetral interval. A 22-gauge needle is sometimes valuable to identify the interval before capsulotomy. Once identified, I use sharp osteotomes or a motorized bur to decorticate the dorsal two thirds of the ­facing surfaces of the lunate and the triquetrum. I leave the volar lip of each bone undisturbed to prevent over-­reduction. Once the decortication is complete, three 0.045-inch K-wires are placed percutaneously from ulnar to radial, firmly fixing the triquetrum to the lunate, both in reduced position (Fig. 20A1-16). A headless compression-type screw (Herbert or Acutrak) can be used if preferred over pins, but care must be taken to avoid overcompression of the joint because this will potentially cause radiocarpal incongruity. An intraoperative mini C-arm has proved invaluable to assist in the reduction and pinning. Once pinned, the interval between the two carpal bones is packed with local autologous bone graft obtained from the distal radius. The capsule is closed with 2-0 or 4-0 nonabsorbable suture; the retinaculum is repaired with 4-0 absorbable suture taking care to avoid the dorsal sensory branch of the ulnar nerve. The 4-0 absorbable suture is used for reapproximation of the subcutaneous tissues, and running subcuticular 4-0 Prolene is my preferred skin suture. In patients with an ulnar-positive variance, I also perform a diaphyseal shortening of the ulna. I have been pleased with the TriMed (Valencia, Calif) ulnar shortening plate, and it has become my standard for ulnar shortenings. For symptomatic partial tears, typically identified at arthroscopy, I prefer arthroscopic débridement and percutaneous pinning using two or three 0.045-inch K-wires placed from the ulnar side of the wrist, through the lunate into the triquetrum. A high index of suspicion based on history, ­coupled with ulnar-sided wrist pain, must include evaluation

Figure 20A1-16   Radiograph showing pins placed for lunotriquetral arthrodesis. An ulnar shortening was previously performed in this patient.

of the lunotriquetral interval to ascertain the diagnosis and provide effective treatment. Postoperative Care. After lunotriquetral arthrodesis, the patient is placed into a short arm volar plaster splint at surgery, followed by suture removal and placement into a short arm cast at 2 weeks. Pins are left outside the skin and are protected with felt pads when the cast is placed. Pins are removed at 6 weeks, and a short arm splint or cast is worn for an additional 6 weeks. The cast or splint is removed at 3 months, and range of motion and strengthening exercises are begun. Criteria for Return to Sport. Once radiographic and clinic union are established, and no tenderness exists to palpation over the fusion mass, the patient may return to sports. Typically this is a 3-month period from surgery. Additional time to regain motion and strength, although optional, is highly recommended. If the sport allows cast wear, then the patient can potentially return to sports at 6 weeks wearing the cast.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1332 DeLee & Drez’s� O

Perilunate Dislocations and Fracture-Dislocations (Mayfield IV)

Clinical Presentation and History

Complete dislocation of the capitate relative to the lunate, in association with lunate dislocation from its lunate fossa on the distal radius, is defined as a Mayfield IV injury, or complete perilunate dislocation (see Fig. 20A1-8). A dorsal perilunate dislocation refers to the position of the capitate relative to the lunate. Dorsal perilunate dislocations are far more common than volar perilunate dislocations. Fractures may occur in association with the dislocations, or carpal fractures may occur rather than ligament injury. A lesser arc injury implies the injury travels through the ligaments, beginning radially, then moving distally, then ulnarly in clockwise fashion around the lunate. A greater arc injury, conversely, implies the same clockwise injury pattern, but fractures occur through the scaphoid, capitate, and triquetrum in conjunction with, or instead of, interosseous ligament disruption (Fig. 20A1-17).

Perilunate dislocations and fracture-dislocations represent high-energy injuries, and acute presentation is the rule.67 Motor vehicle crashes, falls from heights, and other high-energy trauma are the most common histories. As Mayfield showed in his classic work, wrist dorsiflexion, ulnar deviation, and carpal supination, coupled with an axial load, make up the mechanism of injury.18 Most athletes present remembering the forceful hyperextension of the wrist, or a significant fall onto the palm of the hand, but may not be able to describe the subtle twisting or ulnar deviation mechanisms characteristically involved. Wrist pain, swelling, and an inability to move the wrist are most common. Some patients experience paresthesias in the median nerve distribution as the nerve is stretched or contused by the injury. Despite their severity, some perilunate dislocations may be missed on initial presentation, and thorough radiographic evaluation is ­paramount.68

Classification

Physical Examination

As noted previously, a complete perilunate dislocation is classified as a Mayfield IV injury. More information, however, can be helpful in choosing treatment options. Associated fractures, for example, are also named as part of the injury. A perilunate dislocation associated with a scaphoid fracture is referred to as a trans-scaphoid perilunate dislocation (Fig. 20A1-18). The scaphoid fracture represents a component of a greater arc injury. A pure greater arc injury, involving fracture through the scaphoid, capitate, and triquetrum, is rare, and few case reports exist.66 Identifying the fractures is important because fracture fixation can greatly assist in the overall treatment of the perilunate fracture-dislocation.

Physical examination is limited because the patient has a painful, swollen wrist and will move it only minimally. Tenderness is diffuse. Attention to a thorough neurovascular examination is mandatory because the high-energy injury may involve vascular insufficiency or, more commonly, nerve involvement with paresthesias.

Radiographic Examination Initial PA, lateral, and 45-degree oblique views are usually sufficient to delineate a complete perilunate dislocation. The PA view alone can be subtle, and the injury can be missed.69 If a complete perilunate injury is identified on plain radiographs, reducing the injury takes precedence over obtaining more studies, such as computed tomography (CT) or MRI.70 After reduction, an MRI can be helpful in defining the extent of ligamentous disruption and help direct operative planning.

Treatment Options

Figure 20A1-17   Greater arc injuries involve fractures through the scaphoid, capitate, or triquetrum (yellow arc). Lesser arc injuries involve the perilunate ligaments (red arc).

Early reduction of the dislocation will help the patient immensely with pain relief and allow operative planning on a scheduled, rather than urgent or emergent, basis. This can be difficult, and the reduction may have to wait until the patient is brought to the operating room for general anesthesia, at which time definitive care can be rendered. Closed reduction and casting, although historically used, are rarely indicated in these injuries because longterm outcomes have been shown to be inferior to surgical treatment.69 Arthroscopic reduction with percutaneous pinning, open reduction using a dorsal approach, and combined dorsal and volar approaches have all been espoused.67,70-72 Perhaps the most important factor to remember is that each injury is unique and should be approached based on the specific injury pattern, including associated ­fractures.

Wrist and Hand 1333

A

B Figure 20A1-18   A, Trans-scaphoid dorsal perilunate dislocation. The arrow points to the scaphoid fracture. B, Open reduction and internal fixation of the scaphoid accompanied ligamentous repair of the torn lunotriquetral ligament for the trans-scaphoid perilunate dislocation seen in part A.

Author’s Preferred Method At presentation, an initial attempt at closed reduction in the emergency department is warranted. Under intravenous sedation and intra-articular administration of lidocaine, the lunate is reduced with the examiner’s thumb, with an assistant providing longitudinal traction. Next, with the lunate stabilized by the examiner’s thumb exerting palmar pressure on the volar lip of the lunate, the wrist is fully extended and then subsequently flexed, while maintaining longitudinal traction. If successful, the capitate will “snap” back into position distal to the lunate. This is the ideal situation, but significant swelling and imprecise ability to palpate the lunate can make closed reduction in the emergency department difficult, if not unachievable. If reduction is achieved, then surgery can be scheduled within the next week to 10 days. If unsuccessful with closed reduction, I prefer to proceed with urgent surgical care, within 24 hours, to minimize the risk to neurovascular structures with the carpus in an unreduced

position as well as to alleviate the pain associated with the carpal dislocation. My surgical preference is to perform both a dorsal and volar approach as initially described by Cooney and colleagues.72 The dorsal approach begins with a 6- to 8-cm longitudinal skin incision centered on Lister’s tubercle. The extensor retinaculum overlying the third dorsal compartment, containing the extensor pollicis longus, is incised over the tendon, and the tendon is retracted radially. Next, the fourth dorsal compartment and second dorsal compartment are elevated subperiosteally off the distal radius. The dorsal wrist capsule is typically ruptured with the injury, and no formal capsular incision is required. The carpus is easily visible through the capsular defect. At this point, the wrist is supinated, and an extensile carpal tunnel approach is performed through an 8-cm incision, taking care to create an apex or angle at the level of the wrist flexion crease. The Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1334 DeLee & Drez’s� O

Author’s Preferred Method—cont’d transverse carpal ligament and volar forearm fascia are incised in line with the skin incision. Retraction of the tendons shows the floor of the carpal tunnel, with a transverse rent invariably present in the volar wrist capsule, through which the lunate may partially or completely extrude. The lunate is manually reduced into the lunate fossa, and the surgeon’s thumb and index finger can hold the lunate in place through the dorsal and volar incisions. A K-wire to hold the lunate in its reduced position is optional at this point, but I rarely do this. Returning to the dorsal incision by pronating the wrist, the scaphoid is reduced to the lunate, still held between the index finger and thumb. The capitate is likewise reduced relative to the distal lunate and scaphoid. With the scaphoid reduced, two 0.045-inch K-wires are used to fix the scaphoid to the lunate. A third K-wire is placed from the scaphoid into the capitate. Once these three bones are reduced and pinned, attention is directed to the lunotriquetral ligament disruption. The triquetrum is reduced to the lunate, and two additional 0.045-inch K-wires are placed percutaneously, affixing the triquetrum to the lunate (Fig. 20A1-19). Any associated fractures are reduced and fixed in the manner

described later for bony injury. In greater arc injuries, fracture may take the place of ligament injury, but both ligament injury and fracture can coexist and should be sought. After pinning and fracture fixation, any ligament repair that can be done is done, although the ligaments, having been violently disrupted, are often not amenable to primary repair, other than simple sutures to close rents. The exception is the volar capsular rent, which is amenable to suture repair. I use nonabsorbable suture; either 2-0 or 3-0 works well. The subcutaneous tissues are reapproximated with 4-0 absorbable suture, and the skin is closed dorsally with a running subcuticular suture of 4-0 Prolene. Volarly, the skin is closed with 4-0 nylon in horizontal mattress fashion because the volar skin is less amenable to running subcuticular repair. Postoperative Care. The patient is placed into a short arm thumb spica cast at the time of surgery, leaving the pins trimmed and bent back on themselves, but outside the skin, to facilitate later removal. The patient is seen between 10 and 14 days after surgery, and sutures are removed. The pins are left in place, and a well-padded thumb spica cast is placed. Six weeks after surgery, the pins are removed in the clinic,

A

B Figure 20A1-19   A, My preferred pinning technique for Mayfield IV perilunate dislocations. Two additional 1.5-mm screws were used to address a large lunate volar lip fragment. B, Four months after surgery, the patient continues to have a well-reduced carpus. Notice the suture anchors in the scaphoid used to repair the scapholunate ligament.

Wrist and Hand 1335

Author’s Preferred Method—cont’d and the cast is replaced with a commercially available thumb spica splint or equivalent molded Thermoplast (Laval, Quebec, Canada) thumb spica splint. Supervised therapy to regain motion and strength is begun at 6 weeks as well. Criteria for Return to Sports. Patients must be ­thoroughly counseled about the severity of the injury and the likelihood of long-term loss of strength and motion in the involved wrist.73 This may preclude return to athletic participation for

BONY INJURIES Acute Scaphoid Fractures The scaphoid is the most commonly fractured carpal bone and remains the subject of much ongoing research.74,75 Many athletes have lost playing time from a fractured scaphoid, and alternatives in treatment for scaphoid fractures have seen numerous advances over the past decade.50,76-87 In recent years, more and more surgeons are taking a more aggressive approach to surgical fixation of the scaphoid to allow earlier return to sports or work.15,50,77,80,83,84,88 Percutaneous fixation of stable, nondisplaced scaphoid fractures has become an acceptable option in certain populations because it allows earlier return to athletic or military endeavors.50,75,77 Cast immobilization remains the gold standard for nondisplaced scaphoid fracture with union rates of more than 95% in nondisplaced scaphoid waist fractures, but increasing attention is being paid to early surgery to allow rapid return to activity, including sports.89,90 Displaced unstable fractures of the scaphoid commonly warrant surgical intervention early to optimize final outcome, but surgical treatment of the stable nondisplaced scaphoid fracture has evolved into an acceptable option in the past few years, especially in athletes who desire minimal time out of sports participation. The early percutaneous fixation of scaphoid fractures in young active athletes and military members is one of the most significant ongoing evolutions in the management of scaphoid fractures and will be discussed further.

some. For others, using a splint or brace long term may be a viable option, depending on the sport, and 3 months after surgery, patients are allowed to return to sports activity wearing a brace or splint. For those unable to wear a brace or splint, 6 months or longer may be required to allow sufficient return of strength and motion for sports. The individual nature of these devastating injuries should translate into individual counseling about appropriate time for return to sports.

Classification Several classification schemes have been presented for scaphoid fractures, but some common features are quite important. Specific location of the fracture can be an important determinant of healing. Distal pole or tuberosity fractures typically unite uneventfully; scaphoid waist fractures, if nondisplaced, heal readily with cast immobilization; and proximal pole fractures are at risk for avascularity and subsequent nonunion (Fig. 20A1-20). Fracture displacement is also a key point in addressing scaphoid fracture. Nondisplaced fractures are treatable, at least initially, with immobilization in a cast, whereas displaced, unstable fractures are more likely to be subject to delayed union, nonunion, and avascular nonunion.77,91 Comminution at the fracture site renders the fracture unstable and is treated similarly to displaced fractures. Finally, the chronicity of the fracture is also important. An acute, nondisplaced, scaphoid waist fracture is expected to heal uneventfully in a cast. Conversely, a fracture that has been present for months, or in some cases years, and is not healed is not likely to heal without surgical intervention and behaves far differently from the acute fracture (Table 20A1-2).

Clinical Presentation and History Any athlete who presents after a fall onto the palm of the hand or forceful dorsiflexion of the wrist, with or without a twisting mechanism, should be examined for possible

Figure 20A1-20   From left to right, a distal scaphoid tuberosity fracture, a scaphoid waist fracture, and a proximal pole fracture. The fracture location has implication on healing and treatment options.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1336 DeLee & Drez’s� O

  TABLE 20A1-2  Classification Scheme for Scaphoid Fractures* Location

Displacement

Chronicity

Distal pole Middle third (waist) Proximal pole

Nondisplaced Displaced (>1 mm)

Acute Chronic (nonunion)

*All three categories need to be assessed individually to decide on an optimal treatment plan. All displaced fractures should be addressed surgically.

scaphoid fracture. Skiers, snowboarders, skateboarders, and all those who participate in contact sports are at risk for scaphoid fracture.88,92-94 Painful swelling of the wrist, especially in the area of the anatomic snuffbox, completes the clinical presentation of acute scaphoid fracture. Careful evaluation for ligament injury must be considered as well because scaphoid fracture can occur with scapholunate ­ligament injury or other ligamentous wrist injury.54,94

Physical Examination The hallmark physical examination finding for scaphoid fracture is tenderness to palpation in the anatomic snuffbox, that area immediately distal to the radial styloid, bordered by the extensor pollicis longus ulnarly and the extensor pollicis brevis radially. Palpation over the tuberosity of the scaphoid will also elicit pain with most scaphoid fractures. Pain with range of motion, especially extension and radial deviation, is also common. Range of motion is also diminished, secondary to pain.

Radiographic Examination Plain radiographs including AP, lateral, 45-degree semipronated, and PA in ulnar deviation are often sufficient to identify scaphoid fracture. In the case of occult ­ scaphoid

A

fracture, however, initial radiographs appear normal despite clinical findings consistent with fracture. Adjunctive studies, including MRI and CT, are highly accurate in detecting occult scaphoid fractures in the setting of clinically suspected fracture but normal plain radiographs.77,95,96 MRI has been shown to be superior to CT in detecting trabecular abnormalities without cortical extension and has been our modality of choice in the evaluation of occult scaphoid fractures (Fig. 20A1-21).95,96 Alternatively, bone scintigraphy (bone scan) is also valuable in the detection of occult scaphoid fracture, but although sensitive, it is less specific than either MRI or CT.97,98 Finally, the time-tested method of repeat plain film after 7 to 10 days, following immobilization for clinically suspected scaphoid fracture, remains an acceptable diagnostic plan.

Treatment Options Nondisplaced Fractures The treatment of nondisplaced scaphoid fracture is undergoing an evolution toward offering early surgical intervention, specifically percutaneous screw fixation for certain populations.77,90,99 Those populations include active-duty military members as well as competitive athletes who desire the fastest return to sports.100 Surgery may allow earlier discontinuation of a cast or splint and subsequent return to work or sports. Bond and colleagues reported a return to work at 8 weeks for a group of patients treated with percutaneous screw fixation, compared with 15 weeks for return to work in group treated with cast immobilization.100 The risks of surgery, however, must be weighed against the greater than 95% expected union rate with casting. Casting of nondisplaced scaphoid fractures has long been a standard practice, and the decision between a short arm and long arm cast has also been debated.77,101-103 Thumb spica cast immobilization for 3 months should be

B

Figure 20A1-21   A, This film was read as normal. No obvious fracture line is visible. B, Because of clinical suspicion, magnetic resonance imaging (MRI) was performed, showing a scaphoid waist fracture. MRI has become our modality of choice in detecting occult scaphoid fracture.

Wrist and Hand 1337

expected to give entirely satisfactory results in the treatment of nondisplaced scaphoid fractures and should be considered as the first option, only offering surgery for a select population, after thorough discussion of nonoperative treatment with a cast.

Displaced Scaphoid Fractures Literature continues to support operative fixation of displaced, unstable, or comminuted scaphoid fractures. Open reduction and internal fixation of displaced scaphoid fractures allows anatomic reduction and bony stability to optimize healing, and other methods should be compared with that standard. Chen and colleagues reported successful union in 11 unstable displaced scaphoid fractures treated with closed reduction and percutaneous screw fixation.84 Arthroscopically assisted reduction and percutaneous screw fixation has also met with success for treatment of

displaced scaphoid fractures.94 But the most commonly recommended treatment for displaced, unstable scaphoid fracture is open reduction with internal fixation using a compression screw.50,77,91,94,99,104 This technique allows anatomic reduction and rigid fixation of the scaphoid, which is the goal of treatment. Displaced scaphoid fractures have a much higher nonunion rate if treated by casting alone, which also favors surgical treatment.77,91,104 Szabo and Manske reported nonunion rates as high as 55% and avascularity of the proximal pole occurring in 50% of scaphoid fractures with 1 mm or more displacement.104 Acute scaphoid fractures with displacement of more than 1 mm should therefore be strongly considered for open reduction and internal fixation. Multiple headless compression screws are available, and choice of screw fixation is less important than accurate reduction of the scaphoid, whether that be by arthroscopic or open means.

Author’s Preferred Method Acute Nondisplaced Scaphoid ­ racture F

Because of the predictable healing rates in a thumb spica cast, I always offer that as an acceptable treatment option. As Gellman showed in a classic study, scaphoid fractures treated initially in a long arm thumb spica heal, on average, 3 weeks earlier (9.5 weeks compared with 12.7 weeks) than those treated with a short arm thumb spica.102 I have used that information in recommending an initial 6 weeks of treatment in a long arm thumb spica cast, followed by 6 weeks in a short arm thumb spica cast. This has proved effective in the treatment of acute nondisplaced scaphoid waist fractures, and I remain a strong advocate of cast treatment as an excellent option. I have changed my practice, though, over the past 5 years, now offering, even recommending, surgery to a subset of patients who need the most rapid return to full function as possible. In young active military members, competitive athletes whose wrist motion is mandatory for sports participation, and those in whom a cast is an intolerable option, I have used percutaneous screw fixation in a retrograde fashion for nondisplaced fractures with highly satisfactory results, allowing patients to remove their splints 2 weeks after surgery to allow early motion and return to duty or sports. Using intraoperative mini C-arm fluoroscopy, the scaphoid is visualized, and a guide pin is placed from distal to proximal as close to the center axis of the scaphoid as possible. The triquetrum overlies the “perfect” starting spot on the distal pole, but an acceptable starting spot “hugging” or even traversing a small portion of the triquetrum is easily reached. The guidewire is placed retrograde and evaluated for optimal positioning through fluoroscopy. The optimal position is in the center axis of the scaphoid, allowing the longest screw to be placed. I next place a second pin, purposely not in the center axis, to act as a derotation pin. I use the Acutrak system, and the guidepin is overdrilled with the manufacturer’s custom drill bit. The length is ­measured,

typically 22 or 24 mm, and the screw is placed over the guidewire. The most challenging part of the case is to get the guidewire in the optimal center axis of the scaphoid, and I recommend multiple attempts, rather than accepting suboptimal placement (Fig. 20A1-22A and B). Postoperative Care. After percutaneous screw placement, I place the patient in a short arm thumb spica splint for 2 weeks. After suture removal at 2 weeks, a commercially available hand-­based thumb spica splint is worn, and range of motion and strengthening exercises are begun (see Fig. 20A1-22C). Criteria for Return to Sports. When motion and strength have returned and tenderness in the snuffbox has subsided, return to sports is acceptable, typically at about 6 weeks. Although continued bony healing is ongoing, I feel the addition of the internal screw fixation is adequate to allow return to sports by 6 weeks. To date, I have not had refracture occur, although with poor screw placement or avascularity of the proximal pole, the screw is destined to failure. Displaced or Unstable Scaphoid Fractures

Although closed reduction and pinning, as well as arthroscopically assisted reduction and pinning, have been described, I prefer open reduction with internal fixation with a headless compression screw. My screw of choice has been the Acutrak screw, although several others are on the market with similar headless compression design. Beginning with a standard Russe approach, I incise the skin, taking care to create at least a 40-degree angle across the wrist flexion crease to minimize late scar contracture. Next, the flexor carpi radialis (FCR) is identified and its sheath opened along its radial margin to protect the palmar cutaneous branch of the median nerve, which lies just ulnar to the flexor carpi ulnaris at the wrist. The tendon is retracted ulnarly, and the tuberosity of the scaphoid is palpated. An incision beginning Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1338 DeLee & Drez’s� O

Author’s Preferred Method—cont’d

A

B

C Figure 20A1-22   A, An acute nondisplaced scaphoid fracture in a young active-duty soldier. B, The patient opted for immediate percutaneous retrograde screw fixation. C, Two weeks after surgery, the patient is placed in a commercially available hand-based thumb spica splint. Motion and strengthening are begun, minimizing time out of work or sports.

at the palpable tuberosity is carried proximally in the floor of the FCR sheath along the length of the scaphoid. Distally, the incision typically is into the proximal-most portion of the thenar musculature. Careful dissection allows splitting between longitudinal muscle fibers, rather than transecting

them, to expose the distal pole of the scaphoid. The dissection involves opening and visualizing the scaphotrapezial joint because the starting spot for the guidewire is the articular surface of the distal pole of the scaphoid. The radioscaphocapitate ligament, lying perpendicular to the long axis

Wrist and Hand 1339

Author’s Preferred Method—cont’d of the scaphoid, is transected at the level of the scaphoid waist. This should be identified for later closure. The scaphoid, now exposed, is irrigated and curetted to remove hematoma. An anatomic reduction is performed under direct visualization, usually by axial loading of the carpus and slight rotation. A Freer elevator is sometimes helpful, as is a dental pick, to finely tune the reduction. Once ­reduced, a guidewire is placed from distal to proximal under direct visualization, assisted with fluoroscopic guidance. When the guidewire is in the center of the long axis of the reduced scaphoid, a second derotation is pin is added, taking care to account for the width of the drill bit to follow. The central guide pin is overdrilled with the cannulated drill bit. Although the protocol advocates hand drilling, I prefer to use a power drill in young males because the scaphoid cortex is quite hard to drill by hand. The length is measured, and the appropriate headless compression screw is then placed using a hand-held screwdriver to allow the surgeon to assess the “feel” of the screw as it is tightened. The screw’s position is checked fluoroscopically. The derotation pin is then removed. The radioscaphocapitate ligament is reapproximated using 2-0 ­ nonabsorbable ­suture. The remaining capsule is closed with 4-0 absorbable

suture. The subcutaneous tissues are reapproximated with 4-0 absorbable suture, and the skin is closed with 4-0 nylon suture in horizontal mattress fashion. Postoperative Care. A thumb spica splint, placed at the time of surgery, is removed at 2 weeks along with skin sutures. A short arm thumb spica cast is then placed for an additional 4 weeks. At 6 weeks, the cast is removed, and range of motion and grip strengthening exercises are begun. A splint can be worn for comfort at night and for heavy activity as needed for an additional 6 weeks. Criteria for Return to Sports. The soft tissue dissection required in open reduction requires more time than the simple percutaneous screw placement for nondisplaced fractures. The 6 weeks of casting is followed by a period of mobilization and strengthening. This typically takes an additional 6 weeks before the athlete is ready to return to sport. In some sports, if acceptable, playing in a cast or splint is allowed, with therapy progressing between periods of play. This would potentially allow a player to return to his or her sport, wearing a cast, as soon as comfort allows following surgery. I recommend waiting at least 2 weeks after surgery to allow soft tissue healing, wound evaluation, and suture removal.

Scaphoid Nonunions

Classification

Invariably, some patients will present following relatively innocuous trauma with a fractured scaphoid. Often a history of prior fracture, or “wrist sprain,” sometimes untreated, is given. Alternatively, even fractures treated appropriately with a cast become displaced and go on to nonunion. For displaced fractures, nonunion rates between 10% and more than 50% are reported.104,105 Szabo and Manske further reported 50% avascularity in the face of displaced scaphoid fractures.104 Clearly, the risk for nonunion is particularly high for displaced scaphoid fractures, and symptomatic scaphoid nonunions will present in athletes along with the acute fracture group. Dealing with scaphoid nonunions and avascularity has been troublesome at best for many decades, since Green reported zero out of five healed Russe nonvascular bone grafts in patients with an avascular proximal pole in his seminal article in 1985.106 Zaidemberg and colleagues, in their classic 1991 paper, reported healing in 11 of 11 avascular nonunions using a wedge of bone obtained from the distal radius fed by what he termed the ascending irrigating branch of the radial artery. The era of vascularized bone grafting of the avascular scaphoid nonunion was under way.107 Many variations of vascularized bone grafting have been described, but the arterial anatomy of the distal radius by Sheetz and associates in 1995 outlined what has become the most popular vascularized bone grafting techniques.108 In fact, Zaidemberg and colleagues’ ascending irrigating branch of the radial artery is the same vessel Sheetz and colleagues termed the 1,2 intracompartmental supraretinacular artery. That artery supplies a portion of distal radius bone, which when harvested provides a reliable source of vascularized bone graft for treating scaphoid nonunions.

The most important prognostic information regarding scaphoid nonunions is the vascular status of the proximal pole. Because the primary vascular supply enters the bone from the dorsoradial ridge and perfuses from distal to proximal, fractures of the scaphoid, especially when displaced, can lead to avascularity of the proximal pole. Knowing the vascular status of a scaphoid nonunion before attempting surgery is valuable in determining whether standard nonvascular techniques of bone grafting and fixation will suffice, or whether a vascularized bone graft procedure must be considered. MRI evaluation of scaphoid nonunions is now commonplace, and the addition of intravenous gadolinium further helps elucidate the vascular status of the scaphoid.77,109-111 The accuracy of the gadolinium MRI has been called into question, with poor correlation between MRI predictions of vascularity and the ability of the bone to heal after surgery.106,111,112 The gold standard for assessing the vascular status of the scaphoid remains intraoperative visualization of the bone.106 In addition to the vascular status of the scaphoid, associated arthritic changes also affect the surgical plan. Predictably the scaphoid and adjoining surface of the radius undergo arthritic changes.113 Mack and Ruby, in nearly simultaneous classic articles, elucidated the natural history of scaphoid nonunion over time.114,115 In Ruby’s study, 97% of 55 patients developed symptomatic arthritis after 5 years from fracture. Both authors concluded that unless treated surgically, scaphoid nonunions will go on to degenerative arthritis of the wrist in a predictable pattern, and that all patients with scaphoid nonunions should be advised to have surgery, even if asymptomatic, to minimize the risk for future arthritic change. Both Mack and Ruby’s

�rthopaedic ����������� S �ports ������ � Medicine ������� 1340 DeLee & Drez’s� O

studies have potential design flaws in that only patients with symptoms presenting for medical care were enrolled in the study. This selection bias leads to the possibility of a population of patients with long-­standing scaphoid nonunions who may remain asymptomatic. Despite the potential study design flaw, all patients with scaphoid nonunions should be advised of the greater than 90% likelihood of developing arthritic changes and should be offered surgical intervention to minimize the chance of future arthritis. The last factor to evaluate in determining the appropriate surgical plan is the architecture of the scaphoid nonunion. A humpback deformity of the scaphoid may develop in association with a nonunion.116 Correcting the humpback deformity, along with achieving bony union, should be the goal.116 Determining whether a humpback deformity is present is difficult to determine by plain radiograph. Computed tomographic scans and more recently MRI have been used to delineate scaphoid architecture, specifically humpback deformity.117-119

Treatment Options Vascular Scaphoid Nonunions without Humpback Deformity or Arthritis The vascular scaphoid is the easiest of the nonunions to treat. The goal is osteosynthesis to achieve fracture healing in anatomic alignment. Surgical options include open reduction with AO screw placement, corticocancellous bone grafts harvested from the iliac crest or distal radius, and cancellous chips with pinning or headless compression screw placement.116,120-122 Recently, Bullens and associates described percutaneous placement of corticocancellous bone grafting with a minimally invasive approach, reporting union with their method in 29 of 33 nonunions.121

Author’s Preferred Method For all scaphoid nonunions, I prefer a preoperative MRI with intravenous gadolinium enhancement for two reasons. First, it will help delineate the scaphoid architecture and assess for humpback deformity as described by Topper.119 Second, the intravenous gadolinium allows some, although to date not perfect, assessment of the vascular status of each of the scaphoid fragments. If the proximal pole is vascular by MRI and is confirmed intraoperatively by direct inspection, and if there is no humpback deformity, I perform a volar Russe approach and use a curette to freshen the fracture surfaces. If there is any incongruence at the fracture site, I add autologous cancellous bone graft to the fracture site and fix the fracture with a headless compression screw. I prefer the Acutrak screw for its technical ease and the lack of a required jig for screw insertion. This technique is essentially the same as for acute displaced scaphoid fracture fixation. The important exception is that the chance of avascularity of the proximal pole must be assessed intraoperatively and surgical planning must include a vascularized bone graft option if needed.

Scaphoid Nonunions with Avascularity Green’s classic 1985 paper reported healing in zero out of five patients treated with nonvascularized Russe graft technique.106 Zaidemberg and colleagues’ work with vascularized scaphoid bone graft technique paved the way for the now commonplace use of vascularized bone graft to treat this problem.107 Several review articles comment on the efficacy of using vascularized bone graft techniques to achieve satisfactory union rates in scaphoid nonunions with associated avascularity, or osteonecrosis.50,77,123 More recent literature by Chang and associates has have described some of the failures with vascularized bone grafting, reporting eight complications in 34 patients, including graft extrusion, superficial and deep infection, and hardware failures.81 The authors cite the need for proper patient selection and appropriate surgical technique to optimize outcomes. Tobacco use was also found to be a risk factor for failure. Humpback deformity in the scaphoid, alone, can lead to intercarpal incongruence and carpal collapse and arthrosis.81,124 Coupled with proximal pole avascularity, humpback deformity becomes even more problematic. Trumble and coworkers in 2003 called the combination of avascularity and a humpback deformity an unsolved problem in hand surgery because vascularized bone grafts were typically placed dorsally.116 Since that time, the placement of volar vascularized bone grafts has been described in an attempt to solve this problem as well.125,126

Hamate Fractures Racket sports, golf, lacrosse, and baseball have all been reported as the source of hamate fractures, especially those involving the hook or hamulus of the hamate.88,94,127-129 They can be difficult to diagnose because plain films often miss the fracture, especially if nondisplaced, but they can be the source of ongoing pain if left untreated.

Classification Hirano and Inoue classified hamate fractures into two main types.130 Type 1 fractures involve fracture of the hook or hamulus, and type 2 involve hamate body fractures. The authors further subdivided type 2 fractures into type 2a, or coronal fractures, and type 2b, or transverse fractures. In addition to Hirano’s classification of location of fracture, nondisplaced and displaced fractures also warrant differentiation because nondisplaced fractures may be amenable to immobilization, whereas displaced fractures may require surgical care for best results.

Clinical Presentation and History Any athlete involved in a racket sport or sports that involve holding onto a rigid piece of equipment, such as lacrosse, hockey, baseball, or golf, is at risk for hamate fractures. The typical scenario is a rapid deceleration of the stick, bat, golf club, or racket when tightly held in a clenched fist. The rapid deceleration of the implement impacting the palm, along with the active contraction of the deep flexor tendons, puts the hook of the hamate at risk for fracture. Hamate body fractures can also occur in this way or can Text continued on page 1347

Wrist and Hand 1341

Author’s Preferred Method Avascular Scaphoid Nonunions with or without Humpback Deformity

Although controversy remains regarding the significance of a healed scaphoid with a humpback deformity, I prefer to address all avascular scaphoid nonunions, with or without humpback deformity, with a vascularized graft placed volarly as a corticocancellous strut graft. Specifically, I now prefer the use of a 1,2 intercompartmental supraretinacular artery vascularized bone graft, placed volarly as a corticocancellous strut that simultaneously provides vascularity and the advantage of volar placement to achieve correction of humpback deformity, if needed. Preoperative MRI, with intravenous administration of gadolinium, is used to assess for vascularity and humpback deformity (Fig. 20A1-23). My approach is a modification of the original Russe approach to allow excellent visualization of the palmar surface of the scaphoid (Fig. 20A1-24A). About 2 cm proximal to the wrist flexion crease, the incision is curved radially and dorsally for an additional 3 cm. This portion of the incision is made after intraoperative evaluation of scaphoid blood flow because it may not be necessary to proceed to vascularized graft if the scaphoid fragments show good bleeding. If the scaphoid fragments show good bleeding intraoperatively, which remains the gold standard for assessing vascularity, the fracture can be fixed with a headless compression screw, with local nonvascular bone graft as needed. Attention is first focused on the approach to the scaphoid, leaving the proximal 3 cm of planned skin incision until intraoperative evaluation of the scaphoid is performed (see Fig. 20A1-24B). The radioscaphocapitate ligament is transected, after mobilization of the FCR tendon as previously described in the palmar (Russe) approach. Once the scaphoid is exposed, the nonunion site is evaluated, and intraoperative inspection of the scaphoid is used to evaluate blood supply (see Fig. 20A1-24C). If poor bleeding (punctate) or no bleeding (chalk-white) bone is encountered, I proceed with vascularized bone grafting. If excellent (red lush) bleeding is encountered from both proximal and distal fragments, I proceed with nonvascularized compression

A

screw placement as outlined previously, using cancellous bone graft as needed. If a humpback deformity was present on preoperative MRI, I prefer to proceed with volar vascularized strut grafting, through the modified Russe approach using the proximal extension of the incision to harvest the vascularized graft. Once lack of vascularity is established in the scaphoid, a trough is fashioned, using a 3-mm round bur and curettes, excavating as much necrotic bone as possible, while leaving the dorsal cortex intact. One to 2 mm of overhanging cortical bony rims are left as well to help hold the planned bone graft in place (see Fig. 20A1-24D). Upon completion of the scaphoid trough, the proximal 3 cm of the skin incision is made, taking care to identify and protect the superficial branch of the radial nerve, which is in the operative field. The interval between the first and second dorsal compartments is identified, and the 1,2 intercompartmental supraretinacular artery is visualized (see Fig. 20A124E and F). An appropriate-sized bone graft incorporating the artery is then marked and sharply harvested using a 5-mm wide osteotome (see Fig. 20A1-24G). My preferred graft size measures 5 mm wide and 12 mm long. A subperiosteal dissection of the vascular pedicle is performed sharply with a No. 15 scalpel, and a 5-mm swath of tissue is harvested with the pedicle to protect the vascular supply. This dissection continues to the radial artery. The vascular graft is then passed deep to the tendons of the first dorsal compartment and swung volarly into position into the trough created in the scaphoid (see Fig. 20A1-24H). The graft is carefully grasped with an Adson-Brown forceps and fashioned to fit snugly into the trough, using sharp rongeurs and careful burring of sharp corners. One technical tip is to fashion the ends in a sort of bullet shape to allow easy placement of the graft into the scaphoid trough. The graft is then placed into the trough, and a snap-fit of the graft deep to the cortical rims left on the scaphoid completes the process (see Fig. 20A1-24I). If a humpback deformity was present, the humpback is corrected using a small bone hook and distal traction on the

B

Figure 20A1-23   A, Radiograph showing a scaphoid nonunion. Nonunion was diagnosed based on injury 6 years before presentation (arrow points to the fracture). B, Magnetic resonance image depicting proximal pole avascularity (arrow) in the same scaphoid nonunion as seen in part A. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1342 DeLee & Drez’s� O

Author’s Preferred Method—cont’d

A

B

C

D

E

F

Figure 20A1-24   A, Incision for volar vascularized scaphoid graft. Distal portion of incision is made first, extended proximally if vascular graft is needed. B, Initial Russe portion of incision is used to evaluate the scaphoid. The flexor carpi radialis is mobilized, and the incision is carried deeply through the radioscaphocapitate ligament. C, The scaphoid is visible, showing a chalk-white, or avascular, proximal pole. The arrow indicates the proximal pole fragment with no visible bleeding, compared with the visibly lush bleeding in the distal pole seen on the left, distal to the fracture line. D. Hasty but effective intraoperative schematic of the snap-fit of the volar vascular scaphoid graft. Notice the proximal and distal overhangs to optimize graft security. E, The proximal portion of the incision is made, and the 1,2 supraretinacular, intercompartmental artery is identified. Planned bone graft is marked. This view is dorsoradial with the fingers pointing downward. F, Close-up view of the 1,2 supraretinacular, intercompartmental artery. The tendons of the first dorsal compartment (abductor pollicis longus and extensor pollicis brevis) are being held in the retractor. The arrow points to the artery. Distal is to the right in this picture.

Wrist and Hand 1343

Author’s Preferred Method—cont’d

G

H

J

I

K

L

Figure 20A1-24—cont’d   G, The graft is elevated with its pedicle. Notice that the width of the pedicle is larger than that of the artery to protect the vascular supply. The tendons of the first dorsal compartment are elevated to facilitate passing the graft from its dorsal harvest position “around the corner” to its volar insertion position into the scaphoid. H, Graft is readied for insertion by opening the fracture and trough. The thick arrow points to graft, and the thin arrow points to scaphoid trough and fracture. I, The graft is tightly snap-fit into the volar trough in the scaphoid. The dark blue background material is deep to the pedicle. The scissors are pointing to the pedicle. J, A single 0.045-inch K-wire is placed in retrograde fashion to secure the bone graft in place. Its position is checked with intraoperative fluoroscopy. K, A graft is seen embedded in the scaphoid at 6 weeks, at the time of pin removal (arrow). The bone graft harvest site is also visible on plain radiograph. L, After 3 months, the graft is incorporating well (thin arrow). The harvest site for the vascularized bone graft is also visible in this radiograph (thick arrow). Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1344 DeLee & Drez’s� O

Author’s Preferred Method—cont’d

M

N

O

Figure 20A1-24—cont’d   M, Healed vascularized volar strut graft at 6 months. The scapholunate angle measures 50 degrees. The arrow points to the cortical margin of graft along the volar scaphoid contour. The patient reported complete resolution of preoperative pain. N, Arrows point to the resultant scar 6 months after surgery. O, Motion 6 months after volar vascular scaphoid graft. Dorsiflexion is reduced on the operated side. Palmar flexion and ulnar deviation are symmetrical. The arrows indicate the operated side.

Wrist and Hand 1345

Author’s Preferred Method—cont’d index and middle fingers (see Fig. 20A1-24H). The graft is then placed with cortical bone facing palmarly, forming a vascularized volar corticocancellous strut to correct the deformity. Cancellous chips can be packed into any remaining gaps. Following placement of the graft, a single 0.045-inch K-wire is placed in retrograde fashion (see Fig. 20A1-24J). The combination of the snap-fit of the graft plus the K-wire has been adequate fixation in my hands. I worry that placement of a compression screw would at least partially obliterate the graft, and I have not used a compression screw for that reason. The capsule and the radioscaphocapitate ligament are closed with 2-0 or 4-0 nonabsorbable suture, taking care to protect the vascular pedicle. Skin is closed with 4-0 nylon suture. A thumb spica splint is placed at the time of surgery. Finally, in those cases with minimal arthritic changes involving the distal pole of the scaphoid and the radial styloid, a radial styloidectomy can be added following graft harvest. Graft passage from dorsoradial to volar is actually facilitated by radial styloidectomy. Postoperative Care. The pin is left protruding from the skin. Two weeks after surgery, the patient has suture removal and is placed into a short arm thumb spica cast, taking care to pad the pin with felt or similar padding. Six weeks after surgery, the pin is removed, and the patient is placed into a thumb spica cast for an additional 6 weeks, or a ­total of 3 months in a cast. Radiographs are taken at 6 weeks,

occur from falls with the wrist in a hyperextended position. The patient presents with pain in the hypothenar aspect of the palm, with exacerbation with grip and pressure in the palm, especially when directed over the hamate.

Physical Examination Given a history of acute onset of palmar pain in the hypothenar area, coupled with the appropriate history, hamate fracture, especially of the hamulus, should be suspected. By placing the interphalangeal flexion crease of the examiner’s thumb directly over the patient’s pisiform and directing the tip of the examiner’s thumb toward the base of the index finger, the tip of the examiner’s thumb, when flexed, will put pressure onto the hook of the hamate, eliciting tenderness in the face of hamate fracture (Fig. 20A1-25). An additional provocative maneuver is to have the patient actively flex the small finger against resistance. Because the flexor digitorum profundus of the small finger runs along the radial margin of the hook of the hamate, putting that tendon under tension may cause the fracture to move, and elicit pain directly over the hamate, in the hypothenar palm.

Radiographic Examination Routine radiographic examination may be inadequate to identify hamate fractures.88 Special views, including a carpal tunnel view and a semisupinated oblique view, may help detect the hook of the hamate fracture.131,132 If clinical suspicion remains high and plain radiographs fail to

3 months, and 6 months (see Fig. 20A1-24K to O). Graft incorporation can be seen on plain radiograph, but if there is any question of bony healing, I obtain an MRI to assess for graft incorporation and viability. At 3 months, the patient begins range of motion and grip strengthening therapy. Of the first 22 grafts I have placed using this procedure, 20 have incorporated with bony healing. One graft that did not heal had a persistent visible nonunion between the graft and the avascular proximal pole. The second graft was extruded from its position within the scaphoid. Both patients remained symptomatic and were eventually converted to a proximal row carpectomy. Two additional patients had ­persistent pain with progression of radiocarpal arthritis. Both had humpback deformity preoperatively. One pin tract infection was successfully treated with oral antibiotics. Four patients complained of transient anesthesia over the radial styloid, which resolved spontaneously by 3 months. Criteria for Return to Sports. I see the patients in the clinic after 6 weeks of motion and strengthening therapy. If pain is minimal and fracture healing has occurred, a return to sports is allowed. If the sport allows cast or splint wear, a custom orthosis or thumb spica cast is recommended for the remainder of the season. Contact sports should be avoided unless a cast or brace is worn, or until all pain has resolved and motion and strength are nearly equal to the contralateral hand and wrist.

show fracture, CT and MRI are useful diagnostic tests for hook or body fractures of the hamate (Fig. 20A1-26).131,132 MRI is particularly useful to delineate associated soft tissue injury, including visualization of tendons and the ulnar nerve, both of which may be injured along with hamate fracture.

Treatment Options Dividing hamate fractures into displaced and nondisplaced fractures helps direct treatment. Hook or hamulus fractures of the hamate that are nondisplaced can be treated effectively by cast immobilization.130,133 Displaced fractures of the hook of the hamate are treatable either by open reduction and internal fixation or by excision.88,94,129 For open reduction and internal fixation, typically a single AO type screw is used with direct visualization of the reduction, assisted by fluoroscopy. Hamate body fractures, if undisplaced, can also be treated with cast or splint immobilization. Displaced body fractures are best treated by open reduction and internal fixation.130

Trapezium Fracture Fractures of the trapezium are uncommon but can occur in athletes, particularly motorcyclists.134,135 The mechanism may be related to an axial load through the thumb in a ­vertical-type fracture or from avulsion of the transverse carpal ligament at the attachment to the trapezial ridge.136 From a series of 11 patients with displaced or comminuted

�rthopaedic ����������� S �ports ������ � Medicine ������� 1346 DeLee & Drez’s� O

A

C

B

Figure 20A1-25   A, The pisiform is located just distal to the wrist flexion crease on the ulnar margin of the carpus as shown. B, The intraphalangeal joint of the examiner’s thumb is placed directly over the pisiform as shown. C, By flexing the thumb toward the base of the patient’s index finger, the thumb pad comes to rest directly over the hook of the hamate. Pain is elicited if the hook of the hamate is fractured.

Figure 20A1-26   Plain films may not be adequate to diagnose hook of the hamate fracture. When clinically suspected, magnetic resonance imaging provides excellent visualization of the fracture through the base of the hook (arrows).

trapezial fractures, McGuigan and Culp recommend operative fixation for intra-articular fractures with greater than 2 mm of displacement, or carpometacarpal subluxation.134 In addition, trapezium fracture may occur in the setting of other carpal fractures, indicating a relative high-energy pattern of injury.94,137

Classification The avulsion fractures of the trapezial ridge should be differentiated from the trapezial body fractures because their mechanism of injury is different as well as their treatment. Finally, with the intra-articular propensity for vertical

Wrist and Hand 1347

Author’s Preferred Method For acute nondisplaced fractures of the hook of the hamate, a trial of 6 weeks of immobilization is warranted, but operative options are also discussed early with the patient. For special populations, including athletes who desire the quickest return to sports, excision of the fractured hamulus or hook is offered. Return to sports is usually within 2 to 3 weeks, once soft tissue healing allows. If the fracture is discovered late, even if nondisplaced, excision offers an excellent chance to return to pain-free sports participation. For displaced fractures of the hook of the hamate, my preference is excision, rather than open reduction and internal fixation. Because there is no risk for nonunion or hardware failure with excision, and the long-term outcomes are virtually identical, I have opted for the simpler option, which is excision. In addition, return to activity at 2 to 3 weeks is expected with excision, whereas a full 6 weeks to allow bony healing to occur is needed for open reduction and internal fixation. I begin with a zigzag incision over the hypothenar region overlying the hamate. I start the incision at the level of the wrist flexion crease and extend it in zigzag fashion to the level of the thumb-index web, in line with the ulnar border of the ring finger. This allows identification and protection of the ulnar nerve and artery. The deep branch of the ulnar nerve lies very close; in fact, it wraps around the base of the hook of the hamate and must be identified and protected when the hook is excised. A small portion of the transverse carpal ligament attaches to the hook of the hamate, and I have found no adverse consequence from detaching this from the excised hamate hook. For nondisplaced hamate body fractures, my preference is immobilization for 6 weeks in a short arm cast. For displaced hamate body fractures, I prefer open reduction and internal fixation, typically with small screws from the AO modular hand set (Table 20A1-3). Postoperative Care. After excision of the hook of the ham­ ate, the patient is placed into a short arm volar splint at the

fractures of the trapezial body, long-term prognosis includes a risk for arthritis, not seen with trapezial ridge fracture. Palmer subclassified the fractures of the trapezial ridge as type I, those involving the base; and type II, those involving only the tip of the trapezial ridge.138

Clinical Presentation and History A patient with a history of an axial load involving the thumb presenting with radial-sided pain at the base of the thumb or thenar area of the palm should be considered at risk for trapezial fracture. A fall onto the palm of the hand may also predispose to trapezial fracture, especially of the trapezial ridge. Because the fractured ridge involves the radial attachment of the transverse carpal ligament, this injury may be the result of an indirect trauma resulting in a ligament avulsion with a bony fragment.

time of surgery. At 2 weeks, sutures and splint are removed, and the patient may return to sports as soon as soft tissue healing and adequate motion and strength return, often within a few weeks after suture removal.   TABLE 20A1-3  Treatment of Fractures of the Hamate Fracture Type

Author’s Preferred Method

Nondisplaced hook or hamulus fracture Displaced or late hook or hamulus fracture Nondisplaced hamate body fracture Displaced hamate body fracture

Cast or excision (earlier return to sport with excision) Excision Cast for 6 weeks Open reduction with internal fixation

For open reduction and internal fixation of displaced ham­ate body fractures, a short arm volar splint is placed at the time of surgery, then replaced with a short arm cast at 2 weeks, when sutures are removed. Six weeks after surgery, the cast is removed, and the patient is allowed to return to sports, following an additional 6-week period of motion and strengthening. If brace or cast wear is allowed in the sport of interest, playing in a cast or brace is an acceptable ­alternative. Criteria for Return to Sports. As mentioned previously, following excision of the hook of the hamate, athletes are allowed to return to sports once soft tissue healing allows, typically 3 to 4 weeks after surgery. This compares favorably, even with nonoperative cast treatment, which requires 6 weeks to allow bony healing. For open reduction and internal fixation of the displaced hamate body fractures, 6 weeks in a cast followed by an additional 6 weeks of therapy has been adequate to allow return to sports.

Physical Examination The hallmark of physical examination for fracture of the trapezium is tenderness at the site of fracture. Because the trapezium is immediately distal to the scaphoid, the clinical examination is similar with tenderness to palpation inevitably when pressure is exerted over the scaphoid tuberosity. In addition, with an axial load on the thumb, patients with fractures of the trapezial body will experience pain.

Radiographic Examination Displaced fractures of the body of the trapezium may be readily seen on plain radiographs of the wrist, whereas subtle fractures of the trapezial ridge may be easily missed on plain film evaluation.139 If clinically suspected, three views of the thumb, to include a PA, lateral, and

�rthopaedic ����������� S �ports ������ � Medicine ������� 1348 DeLee & Drez’s� O

­ yperpronated (Robert’s) view, will show trapezial body h fractures and their displacement (Fig. 20A1-27A).140 Additionally, a carpal tunnel view may help identify fractures of the trapezium not seen on standard wrist films.141 If plain films do not reveal a fracture despite a high level of clinical suspicion a computed tomographic scan, or bone scintigraphy is recommended.94,142 We use CT to look for subtle fractures of the trapezium, which helps with surgical planning if needed, as well as diagnosis (see Fig. 20A1-27B).

Treatment Options For nondisplaced fractures of the body of the trapezium, a short arm thumb spica cast for 6 weeks works well. Conversely, open reduction and internal fixation is warranted in displaced fractures of the trapezium as advocated by McGuigan and Culp.134 For trapezial ridge fractures, Geissler advocates early surgical excision of the avulsed portion of the trapezium to allow early return to sports.94 Fracture of the trapezial ridge, especially if diagnosis is delayed, may not readily heal with cast treatment alone, leading to persistent pain associated with delayed or ­nonunion.139

Author’s Preferred Method Fracture of Body of the Trapezium and Return to Sports

For nondisplaced fractures, a short arm thumb spica cast is placed for 6 weeks. Upon cast removal, range of motion for the thumb and wrist, as well as grip strengthening exercises, are begun. Return to sports takes about 3 months, unless cast wear is allowable in the particular sport, which favors immediate return. For displaced fractures of the body of the trapezium, open reduction and internal fixation as per McGuigan and Culp is my preferred method134 (see Fig. 20A1-27C). Through a Wagner incision, the trapezium in reduced under direct visualization and provisionally pinned with a Kwire. Next, two AO screws are placed to hold the fracture fragments, and the K-wire can be removed. A thumb spica splint is placed, and the patient is seen 2 weeks after surgery for suture removal and placement of a thumb spica cast. The cast is removed after 6 weeks, then range of motion and grip strengthening exercises are begun. Return to sports is allowed at 3 months. Trapezial Ridge Fracture

For trapezial ridge fractures, I follow the advice of Geissler and perform early excision to allow return to sports in 3 to 4 weeks, after soft tissue healing.94 These fractures are difficult to diagnose, as noted previously, often presenting late with persistent palmar pain, and simple excision offers the best opportunity to alleviate pain and allow return to sports.

Capitate Fracture Isolated capitate fractures are rare.143,144 More commonly, capitate fractures occur in conjunction with other carpal fractures as part of a greater arc injury in a perilunate pattern.145-147 A well-recognized pattern of capitate fracture, although uncommon, is that associated with scaphoid fracture and termed the scaphocapitate syndrome.145,148 Along with the scaphoid waist fracture, a fracture of the neck of the capitate occurs. The proximal fragment of the capitate may be rotated 180 degrees, such that the proximal articular surface points distally.

Classification The most important aspect of the relatively rare fractures of the capitate are to recognize those that are isolated and those that are combined with other carpal injuries because the treatment is often dependent on the other carpal injury, rather than the capitate fracture alone.149 If diagnosed in isolation, determining whether the fracture is displaced or comminuted becomes important as well, in determining treatment options.150 Finally, the intraosseous blood supply of the capitate has been likened to that of the scaphoid, with retrograde blood flow from the distal to the proximal pole.144,151 For this reason, capitate fractures, especially displaced fractures or nonunions, can lead to osteonecrosis of the proximal fragment.144,151,152

Clinical Presentation and History The mechanism of injury for capitate fracture is similar to that for other carpal fractures, especially in association with a perilunate pattern, such as the scaphocapitate syndrome. Specifically, wrist hyperextension with a fall onto the palm of the hand is the most common mechanism for capitate fracture. Given such a history, capitate fracture must be considered in the differential diagnosis, and although rare, a high clinical suspicion can aid in making the diagnosis.

Physical Examination The physical examination should include palpation of the midcarpus as well as examining for evidence of other associated carpal fractures, especially of the scaphoid. The physical examination will elicit tenderness in the area of the capitate, but this may be overshadowed by positive findings for other involved carpal injuries. Diffuse swelling of the wrist may be evident, as may motion limited by pain. Because the capitate is rarely fractured in isolation, it should be considered part of a spectrum of carpal injury and remembered as a potentially fractured carpal bone along with scaphoid fracture or other ligamentous or bony carpal injury.

Radiographic Examination Plain radiographs may not readily demonstrate nondisplaced capitate fracture, and if clinically suspected, other modalities, such as bone scintigraphy and MRI, have been advocated.98,143,153 In the case of displaced fracture or scaphocapitate syndrome with the inversion of the ­proximal

Wrist and Hand 1349

A

B

C Figure 20A1-27   A, A vertical fracture of the trapezium with intra-articular involvement both proximally and distally. B, Computed tomography can help identify main fracture fragments for surgical plans. C, Open reduction and internal fixation of the fracture restores articular congruity. An additional thumb metacarpal base fracture was also fixed.

capitate fragment, plain films of the wrist, including PA, semipronated oblique, lateral, and clenched fist PA, should demonstrate the fracture.

Treatment Options For nondisplaced acute fracture of the capitate, immobilization in a cast for 6 weeks is espoused.154 For fractures with displacement, or in association with other fractures, operative fixation with pins, or more commonly, a headless compression screw is advocated.144,145,150 Operative fixation, including open reduction and internal fixation with pins or screws, is also warranted in cases of scaphocapitate syndrome because the proximal pole of the capitate will not heal in its pathologic inverted position.

Triquetrum Fractures Fractures of the triquetrum typically occur in two types. First is the dorsal chip or avulsion fracture, and second is the body fracture (Fig. 20A1-28B). Both fracture types

Author’s Preferred Method For nondisplaced capitate fractures discovered acutely, cast immobilization for 6 weeks is an effective treatment. For all other fractures of the capitate, those with displacement, or associated with scaphoid fracture, and certainly those presenting late, I prefer open reduction and internal fixation with a headless compression screw. A dorsal approach to the wrist is used, and the capitate is reduced under direct visualization. A guide pin is placed and overdrilled, similar to scaphoid fracture fixation, and a headless compression screw is placed. I have no personal experience with treatment of avascular proximal pole fractures of the capitate, but I would approach the problem much in the same manner as an avascular scaphoid nonunion, using a vascularized bone graft and internal fixation. Postoperative Care. After isurgical fixation of a capitate fracture, a short arm volar splint is applied at the time of surgery. Ten to 14 days after surgery, the sutures and splint Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1350 DeLee & Drez’s� O

Author’s Preferred Method—cont’d are removed, and a short arm cast is placed. Six weeks after surgery, the cast is removed, radiographs are obtained, and if tenderness is resolved and radiographic progression is satisfactory, range of motion and grip strengthening exercises are ­initiated. Criteria for Return to Sports. After operative fixation of a capitate fracture, an athlete may return to sports if cast wear is allowed in the sport in question. Otherwise, after cast removal at 6 weeks, an additional 6-week period of therapy to include range of motion and strengthening exercises is recommended before return to sports. In the case of nondisplaced fracture treated with casting, the cast is removed at 6 weeks, and an additional 6-week period of range of motion and strengthening exercises is advocated.

respond well to cast treatment and rarely require surgery. Triquetral fractures also present as part of a larger spectrum of carpal injury, such as perilunate dislocation with fracture, in which case operative intervention is warranted for the more global injury (see Fig. 20A1-28C).155 Suzuki and colleagues reported an osteochondral fracture of the triquetrum which caused post-traumatic arthrosis of the pisotriquetral joint.156 The patient was symptom free following excision of the pisiform.

Author’s Preferred Method Isolated fractures of the triquetrum, whether dorsal chip fractures or triquetral body fractures, respond well to nonoperative cast treatment. I follow Hocker’s outline for 3 weeks of cast immobilization for dorsal chip fractures.155 For triquetral body fractures, cast immobilization is extended to 6 weeks. Criteria for Return to Sports. Return to sports is allowed immediately for those sports allowing cast wear. Otherwise, return to sports is delayed until the cast is removed and motion and strength have returned, typically 3 months after injury. The dorsal chip fracture may not unite radiographically, but unless associated with a more serious ligamentous injury, typically is asymptomatic. Occasionally, avulsion fracture of the triquetrum represents a more serious ligament injury and further diagnostic workup to include MRI to assess for associated ligament injury is warranted. In this subset of patients, the clinical clue that more serious injury may be involved is the persistence of ulnar-sided pain following removal of cast or splint.33

Pisiform Fracture Fractures of the pisiform are uncommon, but because a fall onto the hand is common in many sports, and that is the usual mechanism of injury for pisiform fracture, those fractures warrant discussion here. Diagnosis of pisiform

fractures requires a high index of suspicion based on history and clinical examination findings.94 Tenderness over the pisiform to direct palpation leads to clinical suspicion of pisiform fracture. Plain films may not reveal the fracture unless special views are employed (Fig. 20A1-29A).157 A carpal tunnel view or 30-degree oblique supination view may demonstrate the fracture.158 CT, MRI, and bone scintigraphy have all been touted as potential adjuncts in discovering fracture of the pisiform (see Fig. 20A1-29B).94,142,159 Treatment with cast immobilization or excision if the fracture is comminuted, displaced, or develops sympto­ matic nonunion are the two mainstays of treatment.

Author’s Preferred Method For acute nondisplaced fracture of the pisiform, I prefer cast immobilization in a short arm cast for 4 to 6 weeks. If the athlete is pain free following immobilization, return to sports play is warranted. I prefer the nonoperative casting rather than immediate excision for nondisplaced fractures, in that it avoids all the potential pitfalls of ­surgical ­treatment, although the time to return to sports is actually longer in athletes treated with a cast (4 to 6 weeks for casted patients versus 2 weeks in patients undergoing surgical excision of the pisiform). If early return to sports is the overriding concern for the patient, early excision of even nondisplaced fractures can be undertaken, as long as the athlete is willing to accept the risks of a surgical procedure. If symptoms persist beyond 6 weeks, I offer elective excision of the pisiform (see Fig. 20A1-29C and D). If the initial fracture is displaced, or comminuted, initial excision of the bony fragments and suturing the defect in the flexor carpi ulnaris tendon following bony excision is my preferred method. Return to sports following excision of the pisiform requires soft tissue healing time only. Often 2 to 3 weeks is enough for a motivated athlete to safely return to sports activity. Similarly, if the diagnosis is delayed or if the fracture presents late, I prefer excision of the pisiform, followed by a 2-week period of splinting, allowing soft tissue healing before return to sports.

Fractures of the Lunate Fractures of the lunate are rare, unless associated with Kienböck’s disease (Fig. 20A1-30). The lunate may be fractured in association with other carpal fractures, and treatment is directed at the entire spectrum of injury rather than just the lunate.160,161 I have recently treated a volar lip fracture of the lunate in association with a perilunate dislocation, and the fracture required a volar approach, through an extensile carpal tunnel approach, and was fixed adequately with two small AO screws, along with reduction and fixation of the perilunate dislocation (see Fig. 20A1-19A). Kienböck’s disease is essentially a vascular pathology involving necrosis of the lunate, with fractures that can lead to fragmentation and collapse (Fig. 20A1-31). This is not truly a sports injury, and the reader is referred to more definitive hand surgery texts for further discussion.

Wrist and Hand 1351

B

A

C

Figure 20A1-28   A, Patient with a triquetral body fracture. Arrows indicate fracture, best seen on oblique view. B, With cast immobilization, the patient is symptom free at 6 weeks, although the fracture line remains evident (arrows). C, Triquetral fractures can occur in the setting of more global carpal injury. In this case, the triquetral fracture is associated with a perilunate dislocation and fracture of the radial styloid. This is a manifestation of a greater arc perilunate injury, discussed earlier.

TENDINITIS Tendon inflammation and injury are extremely common in athletics. Any sport that requires a repetitive motion, such as a tennis serve or volley, a basketball free-throw, or “turning the wrists over” as in the completion of a golf or baseball swing, puts an athlete at risk for tendon inflammation, instability, or even rupture.162,163 Even recreational rock climbing, well known for its association with flexor pulley rupture, has been implicated in tendinopathy of the wrist.164 Swimming, bowling, gymnastics, weightlifting, cycling, and skiing are among the many other sports associated with repetitive use tendinopathy.165 A recent study by Montalvan and colleagues categorized extensor carpi ulnaris (ECU) tendinopathy from 28 cases recorded in elite tennis players.166 The authors divided ECU tendon injuries into three categories: tendinopathy, instability, or rupture. Allende and Le Viet reported on 28 patients who underwent surgical treatment for ECU tendinitis between 1990 and 2002. Seventeen of the 28 patients reported onset of symptoms after sports activity. Of the 28, 15 had tenosynovitis or tendinitis, 5 had dislocation of the tendon, 4 had subluxation, and 4 had an ECU rupture.167 Twenty-two of the 28 were able to return to their previous level of activity at a mean of 23 months after surgery. The first dorsal compartment, containing the abductor pollicis longus (APL) and extensor pollicis brevis (EPB) tendons, is also subject to tendinopathy, specifically de

Quervain’s tenosynovitis, at the wrist, especially in those who participate in racket sports.168 Soejima and colleagues reported a case of FCR tendinitis in a professional baseball player associated with a malunited trapezial ridge fracture.169 He was successfully treated by excision of the trapezial ridge. Buterbaugh, in evaluating ulnar-sided wrist pain in seven professional athletes, reported calcific flexor carpi ulnaris (FCU) tendinitis as one of the contributing sources.170 Some sports, in particular, rowing and powder skiing, each requiring repetitive wrist dorsiflexion and radial deviation, have been found to be associated with intersection syndrome.165,168,171 For all the tendinopathies described in sports, early diagnosis and nonoperative treatment with activity modification, splint or brace wear, and nonsteroidal anti-inflammatory drugs (NSAIDs) has been a cornerstone.

Extensor Carpi Ulnaris Tendinopathy Situated in the sixth dorsal compartment on the ulnar aspect of the wrist, the ECU tendon is commonly involved in tendinopathy in sports. Golfers, tennis players, and other racket sport participants are at particular risk.93,162,166,167 The ECU tendon can be involved in inflammatory tendinitis, subluxation or dislocation, and rupture.167 One specific form of tendinopathy, acute calcific tendinitis, presents with intense pain in the area of the involved tendon and its etiology is poorly understood.172

�rthopaedic ����������� S �ports ������ � Medicine ������� 1352 DeLee & Drez’s� O

A

B

C

D Figure 20A1-29   A, Pisiform fracture seen on lateral radiograph. This subtle fracture is easily missed. B, This patient had tenderness over the pisiform. Plain films did not show a fracture. Magnetic resonance imaging, depicted here, shows the subtle fracture line and marrow edema in the pisiform. C, The pisiform, a sesamoid within the flexor carpi ulnaris, can be excised if fractured. D, A small bone hook can be used to facilitate excision of the pisiform, which is “shelled out” of the flexor carpi ulnaris by scalpel dissection.

Wrist and Hand 1353

Figure 20A1-30   Isolated lunate fractures are rare, as most are associated with other ligamentous injury or Kienböck’s disease.

A

B Figure 20A1-31   A, Lunate fracture, fragmentation, and collapse are often associated with osteonecrosis, or Kienböck’s disease. B, Kienböck’s disease affecting the lunate is dramatically seen with magnetic resonance imaging, whereas the plain film image of the same wrist, on the right, is more subtle.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1354 DeLee & Drez’s� O

The ECU is situated anatomically adjacent to the ulnar styloid, and dorsal to palmar dislocation or subluxation of the tendon can lead to a painful snapping sensation when the wrist undergoes dorsiflexion and rotation.166,167 Because the floor of the ECU sheath is inherently considered a part of the triangular fibrocartilage complex (TFCC), tendinopathy can and does coexist with TFCC injury in some cases.162

with active wrist extension and ulnar deviation against resistance further implicate ECU tendinitis. To detect instability, the examiner places an index finger lightly on the patient’s supinated wrist, with the wrist in full extension. The patient then ulnarly deviates the wrist and brings the wrist into flexion. A painful snap indicates ECU instability. Finally, an inability to extend and ulnarly deviate the wrist may be an indication of rupture.

Classification

Radiographic Examination

Differentiating among three types of tendon pathology for the ECU is useful in directing treatment. Montalvan and colleagues classified ECU tendon pathology into acute instability, intrinsic tendinopathy, or rupture.166 The authors indicate that treatment differs for each of the subtypes of tendon pathology, making the classification relevant and important.

Except in cases of acute calcific tendinitis, radiographs are unremarkable. A calcific density in the area of the involved tendon is suggestive of calcific tendinitis. MRI can be of significant help in the diagnosis of tendinitis because the T2-weighted images may show bright fluid signal both within and around the involved ­tendons.

Clinical Presentation and History

Treatment Options

For any athlete who presents with acute onset of ulnarsided wrist pain, the diagnosis of ECU tendinitis is in the differential. Because it may exist in the presence of other pathology, such as TFCC injury, the exact etiology may not be easily ascertained. Occasionally, swelling over the ECU sheath is present, and active extension with ulnar deviation of the wrist should elicit pain. A history of snapping, especially associated with twisting movements of the wrist, should be sought. Any sudden loss of motion, especially extension combined with ulnar deviation, is also important.

For acute tendinitis without instability, immobilization in a short arm cast, splint, or brace is warranted, along with the administration of oral NSAIDs. Selective use of corticosteroid injection for symptoms unresponsive to splinting and oral anti-inflammatory agents can be used, but patients need to be warned about the possibility of tendon rupture with the use of a steroid injection.166 Early splint treatment for instability involves the use of a long arm splint or cast with the wrist in pronation, as the supinated wrist, when brought from extension into ulnar deviation and flexion may elicit the painful snap.172 If splinting fails to alleviate the instability, then surgery to reconstruct the ECU sheath and reroute the tendon dorsally is warranted.173 For rupture, surgical intervention with primary repair, or tendon transfers may be needed.

Physical Examination and Testing The hallmark of physical examination in acute tendinitis is tenderness to palpation over the involved tendon. Pain

Author’s Preferred Method In all cases of acute ECU tendinitis without instability, I prefer immobilization in a short arm brace or cast for a period of 4 weeks, along with administration of an oral NSAID. In most cases, that is enough treatment to allow acute ­inflammation to resolve, including cases of acute calcific tendinitis. If splinting and NSAIDs are ineffective, I offer corticosteroid injection, followed by an additional period of splinting. My preference is 1 mL of 4 mg/mL of ­dexamethasone mixed with 1 mL of 1% lidocaine without epinephrine. Patients are counseled about the possibility of tendon rupture after steroid injection and are free to decline. Only in recalcitrant cases is surgery warranted, unless instability is present. Surgical release of stenosing tenosynovitis typically brings good relief.166 I use an incision that overlies the course of the dorsal sensory branch of the ulnar nerve, and that nerve should be identified and protected. The extensor retinaculum from proximal to distal is incised over the ECU tendon. No cases of late instability have been reported from surgical release of the ECU tendon from its sheath.172

In cases of instability, including ECU subluxation and dislocation, a period of immobilization in a long arm splint in pronation is warranted in acute cases. In cases ­refractory to initial splinting or in recurrent episodes of dislocation, surgical reconstruction of the ECU sheath, using the extensor retinaculum, is preferred (Fig. 20A1-32). The patient is immobilized in a short arm splint for 6 weeks after surgery, followed by a period of wrist motion and strengthening ­exercises. Criteria for Return to Sports. After nonoperative splinting and NSAIDs, athletes are allowed to return to sports based on symptom resolution, which may vary from 1 or 2 weeks to 3 months in some cases. After surgical release of the ECU if no tendon sheath reconstruction is performed, postoperative splinting can be discontinued after 2 weeks with therapy directed at restoring motion and strength. When the patient demonstrates full motion without pain and adequate strength similar to the contralateral side, return to sports is allowed. This is typically 6 weeks after surgery.

Wrist and Hand 1355

Author’s Preferred Method—cont’d In the case of ECU sheath reconstruction, the long arm splint is used for 6 weeks following surgery, followed by an additional 6 weeks of therapy before allowing return to sports. Resolution of pain and instability, coupled with near-normal

strength and motion, is the benchmark for allowing safe return to sports. Three months is a guideline, but individual recovery times may vary. If allowable and practical for the given sport, a short arm wrist brace is encouraged.

A

B Figure 20A1-32   A, The damaged extensor carpi ulnaris (ECU) sheath is opened along its volar margin, exposing the ECU tendon (arrow). B, ECU stabilization using a sling created from the dorsal retinaculum, placing the ECU tendon dorsal and radial   to the ulnar head (arrows). Blue marker is used to outline flap.

De Quervain’s Tenosynovitis Tendinitis of the first dorsal compartment involves the APL and EPB tendons. It is more common in women than in men because it is notably associated with pregnancy and the postpartum period but has also been noted with certain sports, especially rowing and racket sports.168,174-177 The area deep to the extensor retinaculum, which overlies the radial styloid, may be swollen and markedly tender. In Finkelstein’s test, the patient holds his or her thumb inside a closed fist, and then the examiner passively ulnarly deviates the wrist (Fig. 20A1-33). This maneuver produces pain in de Quervain’s tenosynovitis. Specific tenderness is elicited with digital pressure placed over the radial styloid area.

Treatment Options The treatment of de Quervain’s tenosynovitis begins with immobilization in a thumb spica splint or cast. NSAIDs are also useful. Injection of corticosteroid into the first dorsal compartment is another nonsurgical option. Avci and coworkers compared the efficacy of cortisone injection to splinting in pregnant or lactating women and found cortisone injection to be superior in alleviating symptoms.176 Finally, for those patients who fail nonoperative means, surgical release of the first dorsal compartment is an ­effective solution.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1356 DeLee & Drez’s� O

Author’s Preferred Method I prefer a stepwise treatment plan for de Quervain’s tenosynovitis. The diagnosis is made by history and physical examination findings as noted previously. I have not felt the need to use MRI as a diagnostic tool. Radiographs are a must because the thumb carpometacarpal joint may cause symptoms similar to de Quervain’s tenosynovitis. Physical exa­ mination findings are typically sufficient, however, along with plain radiographs to make the correct diagnosis. My initial treatment is oral administration of an NSAID, coupled with a thumb spica splint or cast and activity ­modification. I add a corticosteroid injection if a patient returns with continued pain after 4 to 6 weeks of splint wear. I use 1 mL of dexa­ methasone, 4 mg/mL mixed with 1 mL of 1% lidocaine without epinephrine. Patients who have persistent pain despite 3 months of nonoperative management of de Quervain’s tenosynovitis are offered elective release of the first dorsal compartment. I prefer a 2-cm transverse incision about 1 cm proximal to the radial ­styloid. Care is taken to identify and protect branches of the superficial branch of the radial nerve, which are inevitably encountered. The extensor retinaculum is incised from distal to proximal under direct visualization. I ensure that the tendons are fully released by using a small retractor to pull on the tendons, ensuring that both the APL and EPB tendons have been released. Based on passive extension of the thumb metacarpal and proximal phalanx when the appropriate tendon is pulled, full release of both tendons is verified. I think this is an important step because the APL may have multiple tendon slips, and a vertical septum may separate the smaller EPB tendon into a separate “chamber” of the first dorsal compartment. Postoperative Care and Criteria for Return to Sports. After surgical treatment for de Quervain’s tenosynovitis, a thumb spica splint is placed at the time of surgery, and the patient is seen 10 to 14 days after surgery for suture removal and thumb spica splint placement. Those athletes who require surgical release for refractory de Quervain’s tenosynovitis typically need 6 weeks of postoperative splinting to allow symptoms to resolve. Return to sports is allowed after thumb range of motion and strength returns. Athletes who do not require surgery are allowed to return to sports wearing a brace or cast, for those sports amenable to cast wear. Otherwise, athletes are allowed to return to sports based on symptom resolution and when return of motion and strength can be demonstrated.

Intersection Syndrome Intersection syndrome has been associated with sports that involve repetitive extension and radial deviation of the wrist, such as powder skiing, weightlifting, rowing, and racket sports.165,168,171,178 Radial-sided, dorsal wrist pain is the hallmark symptom, but it is typically more proximal than that associated with de Quervain’s tenosynovitis.179 Athletes may refer to intersection syndrome as “crossover tendinitis” because it occurs in the region of the distal forearm where the tendons of the first dorsal compartment intersect, or crossover, the tendons of the second dorsal compartment.165 Diagnosis depends on clinical examination with tenderness

Figure 20A1-33   Finkelstein’s test. The test causes pain along the tendons of the first dorsal compartment (arrows) owing to the passive stretch on those tendons in the maneuver. Note the thumb is held by the other four digits.

and, in more severe cases, crepitus with wrist extension in the anatomic region where the tendons of the first and second dorsal compartments intersect, about 4 cm proximal to the wrist joint. MRI is a useful adjunct in making the diagnosis, with increased signal evident on T2-weighted images in the region where the tendons of APL and EPB cross over the tendons of the extensor carpi radialis longus (ECRL) and extensor carpi radialis brevis (ECRB).165,180

Treatment Options Treatment is directed at resting the involved tendons in a thumb spica cast or brace, with oral administration of NSAIDs and cessation of the activity that caused the ­problem. Grundberg and Reagan demonstrated the pathologic anatomy to be a stenosing tenosynovitis of the investing sheath of ECRL and ECRB.181 Refractory cases may require release of the stenotic sheath of the second dorsal compartment and débridement of injured tendons and hypertrophic tenosynovium.181

Author’s Preferred Method Rest, ice, and a thumb spica splint are the cornerstones of therapy. NSAIDs are also prescribed. I have never had to surgically treat intersection syndrome, so I refer the reader to other texts, such as Green’s Operative Hand Surgery, for surgical treatment. Criteria for Return to Sports. Athletes are allowed to return to sports once the symptoms have dissipated, or alternatively while still in a thumb spica cast or splint if it is practical for the sport in question.

Flexor Carpi Radialis Tendinitis Tendinitis of the FCR is distinguished from the extensor tendinoses listed previously by its symptoms being on the volar radial aspect of the wrist, exacerbated by flexion and radial deviation of the wrist against resistance.182 The FCR

Wrist and Hand 1357

tendon is closely associated with the trapezium, scaphoid, and trapezoid, running in a fibro-osseus tunnel through the carpus, and is especially vulnerable to inflammatory changes as it courses adjacent to the ridge of the trapezium on its way to inserting on the volar aspect of the bases of the index and middle finger metarcarpals.183,184 Arthritic conditions of the scaphoid, trapezium, or trapezoid can lead to FCR tendinitis, even rupture.185 A report by Soejima and colleagues described FCR tendinitis in a professional baseball player associated with a malunited trapezial ridge fracture.169 Resolution of the tendinitis followed excision of the malunited trapezial ridge. Most commonly, ­ repetitive wrist motion, especially flexion with radial deviation can lead to a primary tendinitis of the FCR.182 Treatment varies according to the exact etiology, but initial splint treatment, with oral NSAIDs, is usually indicated. Conservative treatment with activity modification, splints, and NSAIDs is highly successful in the treatment of primary FCR tendinitis. In the unusual case that is unresponsive to conservative care, carefully planned surgical release of the FCR tunnel, along with any other pathology, such as trapezial ridge malunion, can be extremely beneficial.182

Flexor Carpi Ulnaris Tendinitis In an athlete presenting with ulnar-sided wrist pain, tendinitis of the flexor carpi ulnaris should be in the differential diagnosis. The clinical presentation with ulnar-sided wrist pain, exacerbated with active flexion and ulnar deviation, is classic. Other causes, such as triangular fibrocartilage tears, fractures of the hook of the hamate or pisiform, and even extensor carpi ulnaris tendinitis can lead to diagnostic difficulty. For that reason, more MRI evaluations are ordered for ulnar-sided wrist pathology than for radial-sided pathology and can be quite helpful when the diagnosis is unclear.131 With a history of repetitive or forceful wrist flexion and ulnar deviation, along with clinical examination findings of tenderness over the flexor carpi ulnaris tendon and exacerbation of pain with flexion and ulnar deviation against resistance, the diagnosis of FCU tendinitis is reasonable. A trial of NSAIDs, splint, or brace wear, coupled with activity modification, is warranted. If simple conservative measures are unhelpful, then adjunctive studies, including MRI, may be warranted to help differentiate FCU tendinitis from other sources of ulnar-sided wrist pain.170

VASCULAR INJURIES Hypothenar Hammer Syndrome Hypothenar hammer syndrome or ulnar hammer syndrome refers to the specific pathology of vascular thrombosis or aneurysmal dilation of the ulnar artery at the wrist that is associated with blunt trauma to the hypothenar area of the palm.186 Painful dysesthesia, typically involving the ring and small finger is common.

Clinical Presentation and History An athlete who presents with pain in the ring and small finger, worsened with cold intolerance, who also gives a history of blunt trauma to the hypothenar region of the palm

Figure 20A1-34   Left, Angiogram depicting occlusion of the ulnar artery at the level of the carpus. Right, A digital subtraction angiogram depicting the same ulnar artery occlusion.

should raise concern about possible hypothenar hammer syndrome. Although classically described as the result of repetitive trauma to the hypothenar area, it may arise following a single episode of significant trauma, such as a fall onto the hypothenar area. A history of smoking should be sought because the condition has been found to have an association with smoking. In their series of 29 patients treated for hypothenar hammer syndrome, Dethmers and Houpt reported that 18 were smokers.186 Ischemic pain in the ring, small, and sometimes middle fingers, with associated painful dysesthesias, is the hallmark of hypothenar hammer syndrome. Symptoms are often exacerbated with cold exposure.

Physical Examination Inspection of the hand may reveal a relative blanching to the ring and small fingers compared the thumb and index fingers, and in severe cases, ulceration and even gangrenous changes can occur at the tips of the involved ­ digits. The most impressive physical examination finding, ­however, is the positive Allen’s test. With occlusion of the radial artery, no flow into the hand is seen when pressure is released from the ulnar artery. A Doppler Allen’s test, in which a Doppler probe is placed over the superficial palmar arch in the palm, confirms the absence of flow through the ulnar artery, when the radial artery is manually compressed.

Radiographic Evaluation Plain radiographs are of little help in assessing hypothenar hammer syndrome but can be useful to look for associated bony pathology and should be taken. The most ­ helpful radiologic evaluation is angiography (Fig. 20A1-34). Newer CT angiograms can be diagnostic.131 A cessation or absence of flow in the ulnar artery at Guyon’s canal is diagnostic for ulnar artery occlusion. Aneurysmal dilations can also be detected by angiography. MRA and digital subtraction angiograms are also used in detecting vascular lesions, and consult with the radiologist may be helpful in choosing the best study in a given institution.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1358 DeLee & Drez’s� O

Author’s Preferred Method My preference for initial treatment is oral administration of a calcium channel blocker and a wrist splint for comfort. Elimination of the blunt trauma, especially if repetitive, is paramount. If after 6 weeks of conservative therapy, symptoms remain unaltered, I proceed with surgical treatment, as described by Troum and coworkers.189 Using a zigzag incision over the ulnar aspect of the wrist, the thrombosed

A

segment of ulnar artery is easily identified visually because it appears totally white in the region of thrombosis (Fig. 20A1-35A). The ulnar nerve is also identified and protected. The entire thrombosed section of artery is resected. Next, a suitable vein is harvested from the subcutaneous area of the volar forearm and placed as an interpositional reverse vein graft (see Fig. 20A1-35C to F). Although some authors

B

C

E

D

F

Figure 20A1-35   A, A thrombosed segment of the ulnar artery is visible as thickened and white (between thin arrows). Notice the anomalous muscle belly proximally (wide arrow). B, With the anomalous muscle belly retracted, normal-appearing ulnar artery is visible proximally (arrow). C, Suitable volar forearm veins are marked preoperatively, and an appropriate vein is harvested at surgery for interposition. D, Blue marker on the vein graft indicates the proximal end. E, The thrombosed segment of artery is excised, and the vein graft is anastomosed proximally and distally to the ulnar artery (arrows). F, The interpositional vein graft is checked with a pencil Doppler intraoperatively to verify flow is restored to the ulnar digits.

Wrist and Hand 1359

Author’s Preferred Method—cont’d have recommended simply resecting the thrombosed segment of ulnar artery, I prefer to use an interpositional graft to restore maximal blood flow to the hand. The graft can be seen to improve blood flow at the time of surgery, and intraoperative Doppler examination confirms it (see Fig. 20A1-35F). Postoperative Care. After surgical treatment of hypothenar hammer syndrome, a volar splint is placed at the time of surgery and removed 10 to 14 days after surgery, along with suture removal. I have patients take 1 aspirin, 325 mg, twice

Treatment Options Initial conservative treatment for suspected hypothenar hammer syndrome is well described, with calcium channel blockers, such as nifedipine, coupled with splinting and activity modification.186,187 Smoking cessation should be strongly encouraged. Intra-arterial administration of thrombolytic agents, like urokinase, have been used as well.188 In four patients with ulnar artery occlusion treated with urokinase, however, only one demonstrated angiographic improvement following treatment.188 Finally, for those patients who fail conservative measures, surgical treatment is indicated. Resection of the thrombosed segment of the artery, with or without interpositional grafts, has been described.186 Troum and coworkers, in a retrospective analysis of nine patients with hypothenar hammer syndrome treated with excision of the thrombosed segment and interpositional vein grafting, reported improvement in all nine patients, making this an attractive technique.189

Thenar Hammer Syndrome Thenar hammer syndrome is rare. Most of the literature discusses case reports or very small series.186,190,191 It involves similar pathology as ulnar hammer syndrome, but with thrombosis occurring in the radial artery at the wrist secondary to blunt trauma such as using the palm of the hand as a pounding tool.

Clinical Presentation and History A patient who presents with a painful index finger, cold intolerance, especially of the radial digits, and mottling or blanching of the index finger is the classic presentation of thenar hammer syndrome. A history of repetitive blunt trauma to the radial side of the palm, specifically the thenar eminence, is the most important clue in identifying the correct diagnosis. The differential diagnosis includes Raynaud’s disease and hand vibration syndrome, which can have similar vasculopathy symptoms, but missing the classic history of using the palm as a hammer.

Physical Examination and Imaging Allen’s test should indicate severely diminished or no flow through the radial artery at the wrist. Mottling, or a blanched appearance, typically most apparent in the index

daily for the first 6 weeks for the antithrombotic effect. The splint is removed at 2 weeks, and patients may begin using the wrist for activities of daily living at that point. Criteria for Return to Sports. Athletes are allowed to return to sports if, 6 weeks after surgery, symptoms are resolved, and soft tissue healing is complete. The aspirin is discontinued at 6 weeks. I have had good success using this regimen, with symptom improvement in all four patients that have required surgical care for this relatively uncommon pathology.

finger, should be sought. The most valuable imaging study is angiography, with specific options described earlier in the section on hypothenar hammer syndrome.

Treatment Options Treatment options are exactly the same as those described for hypothenar hammer syndrome, with conservative care including oral administration of a calcium channel blocker, coupled with avoiding blunt trauma to the involved area. Surgery has been described, with excision of the involved segment of the radial artery, with or without an interposition graft.

Author’s Preferred Method I have no personal experience treating thenar hammer syndrome but would approach treatment in the exact same fashion as ulnar hammer syndrome, beginning with oral calcium channel blocker administration and activity modification. If after 6 weeks, no symptom improvement or worsening has occurred, surgical consideration is warranted, with resection of the thrombosed segment and interposition grafting. Postoperative Care and Criteria for Return to Sports. The postoperative care and criteria for return to sports is exactly the same as in hypothenar hammer syndrome.

SPORTS-RELATED NEUROPATHY Median Neuropathy The median nerve is the most commonly affected nerve by sports participation and injury. Yet median nerve compression, or carpal tunnel syndrome, is rarely considered a sports injury.192 The scope of carpal tunnel syndrome greatly exceeds the limits of this chapter, but a few specific points are warranted. Some sports, notably wheelchair basketball and wheelchair racing, have been found to be associated with the development or exacerbation of carpal tunnel syndrome.193,194 Reports of carpal tunnel syndrome in rock climbers has also been noted.195 Cyclists, although more

�rthopaedic ����������� S �ports ������ � Medicine ������� 1360 DeLee & Drez’s� O

commonly compressing the ulnar nerve while gripping the handle bars, may also exacerbate carpal tunnel syndrome.196 Because carpal tunnel syndrome is the most common nerve entrapment in the upper extremity, it is safe to conclude that many individuals who participate in sports have preexisting carpal tunnel syndrome, and any provocative position, pressure, or movements may exacerbate symptoms.197,198 Finally, reports of median neuropathy in athletes who took growth hormones underscores the need for proper supervision and restrictions when taking “sports supplements.”199,200

Clinical Presentation and History The classic history of numbness, tingling, and pain, typically radiating from the palm of the hand to the radial three digits is most common. Any sports activity that causes pressure over the median nerve or continuous wrist flexion, such as rock climbing, should also be considered as a potential risk for median nerve compression ­ neuropathy. Wheelchair athletics, in particular, should raise the ­suspicion for potential nerve compression involving the median nerve.

Physical Examination Physical examination for carpal tunnel syndrome is well known to all who have completed an orthopaedic residency. Phalen’s test, a maximal flexion of the wrist, can reproduce symptoms associated with carpal tunnel syndrome. The test is correctly performed with the elbow in extension to avoid confusing median and ulnar nerve symptoms. A positive Tinel’s percussion test refers to the reproduction of shock-like tingling in the affected region when a nerve is tapped by an examiner’s finger. Direct compression, sometimes referred to as Durkan’s compression test, can also elicit symptoms in the median nerve.201 Pressure from the examiner’s thumb over the median nerve at the wrist, noting which fingers become symptomatic, and after how long, is the correct method. It may take as long as a full minute for direct compression to have an effect. Finally, static two-point discrimination in the digits should be recorded any time a physical examination for suspected carpal tunnel is performed.

Author’s Preferred Method When faced with an athlete presenting with carpal tunnel symptoms, the first challenge for me is to determine whether the sport is the problem or merely an exacerbating event in someone with preexisting carpal tunnel syndrome. Patients and athletes may be ready to blame the sporting activity as the cause of their problem, but perhaps incorrectly. Simply stated, there are more athletes who do not get carpal tunnel syndrome than athletes who do, so to blame a particular sport may be erroneous. Instead, the patient should understand that he or she may have a predilection for carpal tunnel syndrome that the sports activity simply made apparent. ­Although I think this is an important distinction, any athlete or patient with symptomatic carpal tunnel syndrome warrants treatment. I begin with a commercially available cockup wrist splint that holds the wrist in slight extension. In mild cases, night splinting may be all that is needed. If the history and physical examination clearly point to a diagnosis of carpal tunnel syndrome, I do not order electrodiagnostic studies initially. I do not routinely prescribe anti-­inflammatory medication because it has not been shown to be of significant benefit. If conservative measures fail, I then proceed to further testing. I have found carpal tunnel injection to be an invaluable diagnostic and prognostic tool. I inject 1 mL of 1% lidocaine mixed with 2 mL of 4 mg/mL dexamethasone into the carpal tunnel. I then tell the patient to return if, or when, symptoms recur. If the injection gives more than a few days of relief of symptoms, it supports the diagnosis of carpal tunnel syndrome and, as Green points out, correlates well with a good surgical outcome, if needed.205 I offer surgical release of the transverse carpal ligament if 3 months of conservative treatment has failed to alleviate symptoms, or if symptoms continue to worsen despite conservative measures. If the patient experienced a period of good relief from a carpal tunnel injection, I do not order

electrodiagnostic tests. If the patient experienced no relief from the carpal tunnel injection, I do order electrodiagnostic testing, ensuring that the median nerve at the wrist, ulnar nerve at the elbow, and cervical nerves are tested. Given the tests are consistent with carpal tunnel syndrome, I proceed with surgical release. I prefer a 4-cm longitudinal incision, beginning at the wrist flexion crease and extending distally in line with the ring finger ray. I try to follow any palmar creases in the area, preferring small arcs and zigzags to a straight linear skin incision to improve long-term cosmesis (Fig. 20A1-36). I use a No. 15 blade and transect the palmar aponeurosis in line with the skin incision. If a palmaris brevis muscle is encountered, I incise along its ulnar margin. Next, the transverse

Figure 20A1-36   Typical incision planned for carpal tunnel release, following the creases of the palm.

Wrist and Hand 1361

Author’s Preferred Method—cont’d carpal ligament is encountered and transected from proximal to distal under direct visualization. The distal-most fibers are transected using tenotomy scissors to protect the superficial palmar arch. I use my gloved small finger distally to ensure complete release. Next, using Metzenbaum scissors and a push-cut technique, I transect about 1.5 cm of the volar forearm fascia, in continuity with the transection of the transverse carpal ligament. The tourniquet is deflated, and hemostasis is obtained with monopolar electrocautery. The skin only is closed with 4-0 nylon suture in horizontal mattress fashion. Postoperative Care. A volar splint is placed at the time of surgery, and both sutures and splint are removed 10 to 14 days after surgery. Patients are then allowed to resume

activities of daily living as tolerated. I rarely send patients for supervised therapy after carpal tunnel release. Criteria for Return to Sports. After conservative splinting and activity modification, athletes are allowed to resume sports as symptoms allow, with no specific time frame mandated. After surgical release of the transverse carpal ligament, splinting is discontinued 2 weeks after surgery, and at 6 weeks, the patient is seen for re-evaluation. If at 6 weeks there is sufficient healing of the soft tissues, coupled with satisfactory symptom resolution and adequate motion, athletes are allowed to return to sports activities as tolerated. I warn patients that grip strength takes several months to improve fully, but that is not necessarily a reason to prolong sports avoidance.

Electrodiagnostic Studies

Treatment

Nerve conduction studies and electromyography are considered by some as the gold standard in diagnosing carpal tunnel syndrome.202 Yet others have noted that as many as 25% of patients with clinical evidence of carpal tunnel syndrome have negative or normal electrodiagnostic tests.203 Other studies point to a poor correlation with both nonsurgical and surgical outcomes when compared with the results of electrodiagnostic tests.203,204 Therefore, an electrodiagnostic study, although a useful adjunct, is helpful only when accompanied by a careful history and appropriate physical examination.

The goals of treatment are alleviation of the pain, numbness, and tingling that present with ulnar nerve entrapment at the wrist. Conservative measures, including splinting and activity modification, are usually successful. Surgery is limited to those with other pathology, such as hook of the hamate fractures or ulnar artery thrombosis, and those few patients who fail to respond to nonoperative care.

Treatment Options Initial treatment with splinting and activity modification is well accepted. Surgery is warranted in those cases that do not respond to 3 months of conservative care, or in cases with worsening neurologic examinations. The types of surgery for carpal tunnel release are myriad, and the reader is referred to other comprehensive hand surgery textbooks for more information.

Author’s Preferred Method I have never had to surgically treat isolated ulnar nerve symptoms at the wrist with anything but splinting and activity modification until symptoms resolve, usually within 6 weeks. I see far more cases of cubital tunnel syndrome that have many of the same ulnar nerve symptoms, but that is the subject for another chapter. I have treated patients with other pathology, specifically hook of the hamate fractures and ulnar artery thrombosis, surgically, as previously discussed in this chapter.

Ulnar Nerve Entrapment at the Wrist Although compressive neuropathy at the wrist is described as the nerve courses through Guyon’s canal, it is far less common than carpal tunnel syndrome. A particular entity, known as cyclist’s palsy, refers to an ulnar nerve compression at the wrist associated with tightly gripping handlebars in long-distance cycling sports.196,206 Montoya and Felice reported ulnar nerve compression at the wrist in a weightlifter, purportedly from direct effects of the bar compressing and contusing the nerve.207 The patient recovered with conservative measures. Other pathologies, including hook of the hamate fracture and ulnar artery thrombosis, have been shown to be associated with ulnar nerve pathology and should be considered as part of the differential in any athlete.

WRIST SPLINTS AND SPORTS Athletes participating in many sports, including snowboarding, skateboarding, rollerblading, hockey, weightlifting, and bowling, commonly wear protective braces or padded gloves that extend over the wrist.208-211 Although some studies question their efficacy, it is generally accepted that wearing wrist guards or padded braces in these particular sports can reduce injury severity.211,212 Conversely, some sports activities for children, such as scooter riding, monkey bars, and bicycle riding, have been identified as unsuitable for wrist brace wear because grip strength is actually degraded.213 Each sport must be approached individually to measure the potential benefit of wearing some

�rthopaedic ����������� S �ports ������ � Medicine ������� 1362 DeLee & Drez’s� O

sort of wrist protection. Padded gloves for cyclists, for example, do not hinder wrist flexibility because they are designed primarily to protect the palm of the hand from vibration and pressure.196 Inputs from trainers, coaches, experienced players, and therapists can be invaluable in ensuring proper wrist protection without impairing function in a given sport.

PLAYING CASTS Many sports now allow players to participate in sports while wearing a so-called playing cast. This can allow earlier return to sports following a wrist injury than having to wait until complete healing has occurred. Each sport has its own guidelines, and those guidelines may differ from region to region, school to school, league to league, and even within an individual sport, so close consultation with coaches, league officials, and athletic trainers is mandatory before advising the use of a playing cast.

TAPING Televised athletics in the United States are a veritable advertisement of the commonplace use of taping for ankle, wrist, finger, shoulder, and even rib support for sports participation. Pregame training rooms are filled with athletes getting “taped,” some prophylactically, others to allow playing with an injury. The high demands on professional athletes encourage “playing with pain.” The role of the sports medicine physician and orthopaedic surgeon is often to temper an athlete’s, or coach’s, overwhelming desire to participate in sports despite injury. Some sports, like rock climbing, use taping as a means of injury prevention.214 Skiers are known to tape their wrists and thumbs to help minimize the risk for developing intersection syndrome that can come from pole planting in deep powder.171 The practice of taping in sports is clearly widespread but should be done by those who have training and skill levels that allows safety to be a part of the process. Tape applied too tightly, or incorrectly, may not only lack benefit but also be the cause of harm.

REHABILITATION Returning the injured athlete to sports as expeditiously, yet safely, as possible is the result of teamwork. Orthopaedic surgeons, sports medicine physicians, trainers, coaches, therapists, and the players themselves all have a role. Ice, heat, whirlpool therapy, motion, strengthening, and sportspecific training all play key parts in the rehabilitation of an injured athlete.215 Pain relief, tissue healing, and return of normal preinjury function are the goals of modern rehabilitation. A good therapist is a welcome addition to the postoperative care for many athletes. Supervised therapy, whether conducted by a therapist or a trainer, can be of tremendous benefit in returning a player to his or her sport. A trainer’s or therapist’s in-depth knowledge of the sport, coupled with necessary medical knowledge, greatly facilitates return of motion and strength to the injured wrist in the athlete.

C

r i t i c a l

P

o i n t s

l The

anatomy of the wrist is complex, composed of bones, tendons, ligaments, and neurovascular structures all working in concert to produce seamless fluid motion in an infinite number of planes. The complex interaction of all component parts must be considered when evaluating and treating athletic wrist injuries. l Restoration of normal anatomy in acute injury is the goal to allow optimal long-term stability, strength, and return of function. This is true for bony injuries as well as for ligament injuries. l Tendinitis is common in athletics involving repetitive wrist motions. The mainstay of treatment should be nonoperative for most, with splinting, oral NSAIDs, and activity modification. Only those few who fail conservative measures should be considered for surgical intervention. l Vascular injury can, and does, occur in athletic wrist injury. Restoration of maximal flow to the hand and wrist through medical means or surgical means, including interpositional vein grafting, is recommended. l Athletes by nature desire rapid return to sports. The enthusiasm for rapid return must be balanced with appropriate time for healing and, more important, rehabilitation to optimize strength and motion before return to athletics. The use of playing casts or braces may allow earlier return in carefully selected sports.

S U G G E S T E D

R E A D I N G S

Berger RA, Landsmeer JM: The palmar radiocarpal ligaments: A study of adult and fetal human wrist joints. J Hand Surg [Am] 15:847-854, 1990. Bond CD, Shin AY, McBride MT, Dao KD: Percutaneous screw fixation or cast immobilization for nondisplaced scaphoid fractures. J Bone Joint Surg Am 83:483488, 2001. Buterbaugh GA, Brown TR, Horn PC: Ulnar-sided wrist pain in athletes. Clin Sports Med 17:567-583, 1998. Markiewitz AD, Stern PJ: Current perspectives in the management of scaphoid nonunions. Instr Course Lect 54:99-113, 2005. Mayfield JK: Mechanism of carpal injuries. Clin Orthop 149:45-54, 1980. Rettig AC: Athletic injuries of the wrist and hand. Part I: Traumatic injuries of the wrist. Am J Sports Med 31:1038-1048, 2003. Rettig AC: Athletic injuries of the wrist and hand. Part II: Overuse injuries of the wrist and traumatic injuries to the hand. Am J Sports Med 32:262-273, 2004. Rosenwasser MP, Miyasajsa KC, Strauch RJ: The RASL procedure: Reduction and association of the scaphoid and lunate using the Herbert screw. Tech Hand ­Up Extrem Surg 1:263-272, 1997. Walsh JJ, Berger RA, Cooney WP: Current status of scapholunate interosseous ligament injuries. J Am Acad Orthop Surg 10:32-42, 2002. Wyrick JD, Stern PJ, Kiefhaber TR: Motion-preserving procedures in the treatment of scapholunate advanced collapse wrist: Proximal row carpectomy versus four-corner arthrodesis. J Hand Surg [Am] 20:965-970, 1995.

R eferences Please see www.expertconsult.com

Wrist and Hand 1363

S ect i o n

A

Wrist 2. Wrist Injuries in the Child Eugene T. O’Brien

EPIDEMIOLOGY It has been estimated that 30 million children in the United States participate in an organized sports program. Dalton has noted that of the entrants in a Melbourne marathon in 1991, 190 of the 5807 runners were between 7 and 17 years of age.1 One 7-year-old completed the race in 3 hours and 31 minutes, and a 13-year-old ran it in 2 hours 55 minutes. Young gymnasts can train for 3 hours or more each day, and runners can run 100 km a week. As of 1992, 50% of males and 20% of females 8 to 16 years of age had participated in organized competitive sports, and in more than 20 years (1960 to 1980), childhood sports involvement doubled in the United States.1 More than one third of school-aged children will sustain an injury severe enough to be treated by a doctor or nurse.2 There were an estimated 2.7 million nonfatal sports and recreational injuries in children that were treated in emergency departments annually during the years 2001 to 2003.3 Traditionally, child and adolescent athletic injuries have been composed of acute injuries, including fractures, sprains, contusions, and lacerations. The types and severity of injuries in adolescence, however, are changing. Increased and more intensive sports participation at younger ages has led to more overuse injuries and changes in the patterns of acute injury within this population. Several excellent reviews have documented the increasing prevalence of overuse injuries in young athletes.4-8 Micheli noted that although overuse has traditionally been attributed to breakdown of aged tissues in the “weekend warrior,” it now is being seen increasingly in younger populations, and he suggested that the common denominator in overuse injury is too much stress in too short a time.7 He also identified training errors and muscle-tendon imbalance associated with the growth spurt as major factors contributing to this phenomenon. This is echoed by Bylak and Hutchinson,4 who reported that most tennis injuries are caused by overuse of the muscle-tendon unit. These injuries are more common among adolescents, whereas fractures predominate among more skeletally immature athletes. Others have documented changes in the pattern of acute injury.9-11 Deibert and associates detailed the changing pattern in skiing injuries in older compared with younger populations over 3 decades and noted that besides increases in upper extremity fractures overall across the age groups (pediatric, +402%; adolescent, +202%; adult, +35%),

pediatric gamekeeper’s injuries rose 36%.9 Stanciu and Dumont documented the changing distribution of scaphoid fractures; they reported 50% of scaphoid fractures as occurring at the waist rather than the distal pole, which were more commonly reported in previous studies.11

RISK FRACTURES FOR INJURY Adirim and Cheng2 stress these fractures as increasing the child’s vulnerability to injury: children have a larger surface area–to-mass ratio, children have larger heads proportionately, children may be too small for protective equipment, growing cartilage is vulnerable to stresses, and children often do not have the complex motor skills needed for certain sports until after puberty. Different levels of maturity in skeletal and muscle development are often seen in the same age groups. Strength-flexibility imbalance associated with the growth spurt in adolescence also contributes to the adolescent’s vulnerability to injury. The type of sport has some influence on the risk for injury. Logically, one might suspect that activities with more contact put the athlete at a greater risk for injury. Markiewitz and Andrish noted that more upper extremity injuries occurred among football players compared with other sports.12 The association between contact or impact sports and risk for injury has been supported by other ­studies as well. Bhende and associates reported that hand injuries among pediatric patients reporting to the emergency room were highest among football players, followed by basketball and gymnastics athletes.13 Chamber’s report on orthopaedic injuries among athletes 6 to 17 years of age participating in six organized sports followed the same risk pattern (football → basketball → gymnastics → soccer → baseball → swimming).14 Injuries of the immature wrist are more apt to occur in gymnastics, contact sports, racquet sports, skateboarding, soccer, and the martial arts. Gender differences have also been noted when types of injuries and their prevalence are compared. Bylak and Hutchinson examined a population of young tennis players and reported more wrist and hand injuries among females, whereas males presented with more shoulder and elbow problems.4 Weir and Watson reported their data on 266 Irish adolescents 12 to 15 years of age and noted a higher injury rate among males, although they also had higher participation rates.15 Micheli and others have raised awareness of overuse syndromes in young athletes, emphasizing training errors

1364 DeLee & Drez’s Orthopaedic Sports Medicine

as the number one culprit.1,5,7,8 Training errors include failure to limit exposure time (gymnastics and baseball), too rapid transition from practice to competition without proper training and conditioning, and inadequate supervision. Strength-flexibility imbalance associated with the growth spurt in adolescence also contributes to adolescent vulnerability. During the growth spurt, adolescents are particularly at risk for physeal injury. Relative weakness of the physis compared with surrounding ligamentous constraints appears to be responsible. Before and after a growth spurt, however, the physis resists injury better than ligament, muscle, and tendon; therefore, sprains and strains are more common. During the first 4% of tendon strain, collagen fibers become tight. In the 4% to 8% range, microtrauma occurs whereby molecular cross-links break, and collagen fibers slide past one another. With continued strain, complete tendon failure ensues.16 Evaluation of patients with injuries around physes should therefore be carried out with this age-dependent vulnerability in mind.

CATEGORIES OF INJURIES There are two types of injuries: those more chronic injuries that result from overuse, and acute injuries involving the bone, joints, cartilage, ligaments, or tendons. The immature carpus is infrequently injured, but when injury occurs, diagnosis may be difficult. The rarity of childhood carpal injuries results in a low level of awareness, and the diagnosis is often difficult in the incompletely ossified carpal bones. Fortunately, the diagnosis and treatment of carpal injuries in the child has been significantly improved by the advances in magnetic resonance imaging (MRI) and arthroscopy.

CARPAL DEVELOPMENT All carpal bones are completely preformed in cartilage and present at birth. Each carpal bone develops from an ossification center that expands centrifugally by the same process that occurs in the epiphyseal cartilage plate of all the bones. When ossification is complete, the last remaining surface cartilage becomes the articular cartilage. Carpal bones are ossified in a fairly orderly sequence. The pattern varies somewhat between the sexes, and slight asymmetry may be seen between the two sides.17 The capitate bone is the first bone to appear, and this usually occurs at about 2 to 4 months, followed closely by the appearance of the hamate at about the same time or slightly later. The triquetrum ossifies during the second year. The lunate begins its ossification at about the third year, and the scaphoid follows in the fifth year. The scaphoid begins its ossification in the distal part of the bone and expands centrifugally and proximally. The trapezium and trapezoid ossification centers appear at about 5 years. The pisiform, the last bone to ossify, makes its appearance between the 9th and 10th years (Table 20A2-1).

CARPAL FRACTURES Carpal fractures are rather unusual in childhood because of the resilient nature of the surrounding unossified cartilage. As a result, strong forces are necessary to injure the

  TABLE 20A2-1  Average Age of Appearance of Ossification Centers and Physeal Closure Around the Wrist* Bone

Appearance

Closure (yr)

Capitate Hamate Radius Triquetrum Lunate Scaphoid Trapezium Trapezoid Ulna

2 mo 3-4 mo 10-12 mo 2-3 yr 3 yr 4-5 yr 4-6 yr 4-6 yr 6 yr

13-15 13-15 16-17 13-15 13-16 13-16 13-16 13-16 15-17

*These values represent averages and do not specifically reflect gender differences. From Stuart HC, Pyle SI, Cornoni J, Reed RB: Onsets, completions and spans of ossification in the 29 bone-growth centers of the hand and wrist. Pediatrics 29:237-249, 1962.

incompletely ossified carpal bones. The force generated by a fall on the outstretched hand in the younger age group tends to be dissipated in a fracture involving the distal radial physis and metaphysis. Carpal fractures in younger children tend to result from more forceful trauma than a simple fall on the outstretched hand and are often associated with other injuries. These may include a fracture of the distal radius, other carpal injuries, and combinations of fractures and ligamentous injuries in the carpus. Fortunately, most carpal fractures in children, if diagnosed and treated promptly, heal without significant complication.

Scaphoid Fractures Epidemiology Fractures of the scaphoid account for between 0.34%18 and 0.5% of all upper extremity fractures and 3% of all hand and wrist fractures in children.19 The peak incidence is about 15 years of age. Fracture of the scaphoid is the most common carpal injury in a child, as it is in the adult. The diagnosis of an acute injury may be difficult in both age groups. The resilient cartilage surrounding the ossific nucleus in the immature scaphoid makes fracture less likely in the first decade.20-24 When a fracture does occur early in childhood and involves a chondral fracture, it may be diagnosed only in retrospect.20,25 Currently, diagnosis should be possible with computed tomography (CT) or MRI. In adolescence, the scaphoid becomes more completely ossified, and the bone takes on the more adult type of fracture pattern.22

Evaluation Because of the rarity of scaphoid fractures in children, the index of suspicion tends to be low. On clinical examination, there is snuffbox tenderness and swelling after a fall on the outstretched hand. Painful limitation of wrist motion is usually present. Graham has described a diagnostic test in which the scaphoid is trapped between the examiner’s two thumbs—one pressing on the snuffbox and one on the tubercle, which elicits pain in the patient with a scaphoid fracture.26

Wrist and Hand 1365

Standard radiographs include a posteroanterior view in ulnar deviation with the patient making a fist, a lateral view, and two oblique views. Radiographs of the acute injury may be falsely negative in up to 37% of patients.27 If the initial radiographs are negative, they should be repeated after 10 to 14 days of splint or cast immobilization. If the radiographs are still negative but clinical symptoms persist, consideration should be given to obtaining an MRI. An MRI has a 100% negative predictive value, picks up other injuries, and does not involve radiation or injection of radioactive material.28 Some of the positive MRIs show the plane of the fracture to be oblique or spiral, confirming the difficulty in recognizing the fracture with plain films.

Types of Scaphoid Fracture Distal one-third fractures are the most common scaphoid fracture in the preadolescent patient (Fig. 20A2-1). This is thought to be because the ossification nucleus expands from distal to proximal. The fracture may involve the distal pole but most often consists of an avulsion fracture of the dorsal radial aspect of the scaphoid tubercle. The avulsion occurs from the strong pull of the scaphotrapezial and volar capsular ligaments of this joint. Cockshott postulated that the fracture occurs here because it is the most recently ossified part of the bone.29 The fracture is best, and frequently only, seen in the pronated oblique view.30 Healing occurs rapidly after only 3 to 4 weeks of immobilization because of the abundant blood supply to the area. Wilson-MacDonald described an 11-year-old boy with a delayed union of a distal pole fracture, which ­ initially (8 months earlier) had appeared as a small avulsion fracture of the tubercle.31 Four weeks of immobilization was

carried out, and 2 years after injury, the fracture was healed with the remodeling of the distal scaphoid. As Stanciu and Dumont observed in the adolescent, the scaphoid fractures resemble the adult form, and the incidence of tubercle fractures tapers off.11 Middle-third scaphoid fractures are second in frequency to distal pole injuries. Displacement is uncommon unless the fracture is associated with an intercarpal dislocation. Occasionally an initially undisplaced fracture can become displaced in a cast. Chistodoulou and Colton encountered displacement in 5 of their 64 childhood scaphoid fractures, and 1 of these was initially undisplaced but subsequently became displaced; it was the only fracture in their series that failed to unite.19 Displacement of an undisplaced fracture can also occur during percutaneous screw fixation.32 Incomplete, single-cortex fractures are not uncommon in the middle third of the child’s scaphoid. Proximal-third fractures, thankfully, are very uncommon in the immature carpus. Vahvanen and Westerling identified one proximal-third fracture in their series of 108 children’s scaphoid fractures.24 Christodoulou and Colton recorded two proximal-third fractures in the chart documenting their series.19 An 8-year-boy with a proximalthird fracture that went on later to become a middle-third nonunion was reported by Pick and Segal.33 A refracture of a healed (by CT) proximal-third fracture in a 17-year-old football player was reported by Barrick and associates.34 The patient reinjured the scaphoid at the same site 10 months after his original injury after he had resumed playing football. Bone grafting and Herbert-Whipple screw fixation secured union. Toh and coworkers recorded one nonunion of a proximal pole fracture,35 and Waters and Stewart reported three nonunions of proximal pole fractures in patients with open epiphyses.36

Combined Injuries Distal radial physeal injuries are the most common associated injury with a fractured scaphoid.21,24,37,38 The associated scaphoid fracture is usually undisplaced and is liable to be missed unless it is looked for in all children with distal radius fractures. Kay and Kuschner reported a 13-year-old boy who sustained bilateral proximal radius and distal pole of scaphoid fractures, which were undisplaced. Healing of the injuries followed 4 weeks of immobilization.39 Scaphoid fractures associated with other carpal fractures are discussed separately.

Stress Fracture of the Scaphoid

Figure 20A2-1   A 15-year-old boy with an avulsion fracture of the dorsal radial end of the scaphoid tubercle. Healing followed 3 weeks of cast immobilization.

Six stress fractures in the scaphoid have been reported from ages 13 to 19 years. Five of these fractures were in gymnasts, and one was in a shot-putter.40-42 All had experienced the insidious onset of radial wrist pain without an acute precipitating injury. Initial radiographs were reported to be negative for four of the six fractures. A bone scan showed increased uptake and was helpful in making the diagnosis of a stress fracture in two of their patients. In one of Hanks’ patients, the initial radiographs and bone scan had been reported as negative.40 Because the specificity of a bone scan is limited, an MRI is a much more reliable test to rule out or in a stress fracture. All reported cases

1366 DeLee & Drez’s Orthopaedic Sports Medicine

involved the waist of the scaphoid, remained undisplaced, and healed with cast immobilization (6 weeks to 4 months) (Fig. 20A2-2). Baker postulated that muscle fatigue can alter the stress distribution in the bone because the normal protective action of the muscles in preventing stress concentration is lost.43 He also observed that fatigue macrodamage might be the normal stimulus for the production of properly orientated new haversian bone which is needed most. Matzkin and Singer suggested that the change in mechanical stress may overwhelm the remodeling osteoplastic and osteoblastic activity, resulting in a fatigue fracture.42

The Bipartite Controversy Louis and associates studied 196 human fetuses and 17,439 hand radiographs and were unable to find an example of a congenital bipartite scaphoid.44 Doman and Marcus described a case of bilateral congenital bipartite scaphoid, which was followed from early ossification at age 8 years and 6 months to age 17 years.45 At that time, an MRI was consistent with articular cartilage surrounding the circumference of the bipartite scaphoid. Bunnell listed five criteria to support the diagnosis of congenital bipartite scaphoid46: (1) bilaterally symmetrical bipartite, (2) absence of history or signs of previous trauma, (3) equal size and uniform density of each part, (4) absence of progressive degenerative changes between the two components or elsewhere in the wrist, and (5) smooth, gently rounded components of each part of the scaphoid. Larson and associates believed that it was possible that fracture of the scaphoid could occur before its ossification (<5 years) and that a nonunion might escape diagnosis until a later time when an incidental radiograph is misdiagnosed as a bipartite scaphoid.25

A

Treatment of Scaphoid Fractures Scaphoid fractures in children heal rather promptly after 6 to 8 weeks of immobilization if diagnosed acutely. Healing may be delayed if the diagnosis is initially missed. Open reduction with internal fixation is required if the displacement between the fragments is greater than 1 mm.47

Delayed Union The most common cause of delayed union is that either the initial fracture was missed, because of no follow-up radiograph or examination, or the child did not present for treatment. This can happen, especially in an athletic injury when the injury is passed off as a sprain by the coach or trainer. Fortunately, the missed scaphoid fracture, especially in a young child, has a fairly good chance of healing with prolonged cast immobilization (Fig. 20A2-3). DeBoeck and associates described an 8.5-year-old girl with a fracture that was 7 months old and that healed after cast immobilization for 14 weeks.48 They recommended that a cast be tried for at least 3 months in any child with an ununited fracture that has never been immobilized. Greene and associates presented a 6.5-year-old girl seen 4 months after a fall from her bike with a fracture through the ossific nucleus with sclerosis and cystic changes. Immobilization for 6 weeks resulted in healing.22 They presented three other children who had delayed union or nonunion. One was asymptomatic and received no treatment; two others demonstrated sclerosis and cystic changes at 3 and 8 months after injury, respectively. One fracture healed after plaster immobilization for 10 weeks, and the other fracture required a bone graft. Vahvanen and Westerland identified three neglected scaphoid fractures: a 6-year-old girl immobilized for 7 weeks, and one of the two 14-year-old patients immobilized for 12 weeks and the other for 16 weeks. All three healed.24

B

Figure 20A2-2   A, A 17-year-old elite gymnast had complained of chronic wrist pain for about 3 months. Two previous radiographs were normal. This radiograph demonstrates a transverse wrist fracture of the scaphoid consistent with a stress fracture. B, Five months later, after volar autogenous bone grafting without internal fixation, there is complete healing of the scaphoid fracture, and the patient has returned to competitive gymnastics.

Wrist and Hand 1367

Nonunion of the Scaphoid Because many scaphoid delayed unions heal with cast immobilization, it is difficult to precisely define a nonunion in a child. Theoretically, a nonunion is present when the fracture cannot heal without surgical intervention. Acute scaphoid fractures that are treated by prompt immobilization have a very low (0.8%) nonunion rate.49 About 150 childhood scaphoid nonunions have been reported in the literature.18,19,20,22,25,27,30,33,35,36,50-60 Most of these, which involve the middle third of the scaphoid, are found in the adolescent age group, and they were usually treated just as an adult nonunion is treated. In general, union is easier to achieve in nonunions in children than in adults, with healing rates in most series approaching 100%. Only six nonunions have been reported in children 9 years or younger.20,22,25,33,60,61 Two of these injuries involved the unossified portion of the scaphoid. In these two incidences, scaphoid nonunions were noted 7 years20 and 3 years25 after the original injury. These injuries most likely could have been diagnosed at the time of injury using CT or MRI had the techniques been available at that time. In 9 of 13 nonunions reported by Mintzer and Waters,58 healing followed iliac crest bone grafting and fixation with a Herbert screw. The other four healed following a MattiRusse procedure, but one of these had to be redone. There was a large discrepancy in immobilization times between the two groups. The graft and screw patients required immobilization for only 9 weeks, whereas the patients without fixation were immobilized for 6 months. Other successful techniques to treat childhood scaphoid nonunions have included AO screw fixation with autograft,59 bone grafting with K-wire fixation,56 and bone grafting

A

without osteosynthesis.53,54,58,60 Humpback deformity can occur in childhood scaphoid nonunions, and its correction can be achieved, as in adults, with a wedge graft and screw fixation.55 Toh and coworkers reported the largest series of scaphoid nonunions in children.35 In the 35 nonunions that were treated by volar grafting and screw fixation, primary healing occurred in all but two patients. Healing followed a second procedure in both cases. Percutaneous screw fixation was recommended for fractures having fibrous union in good position, but how the determination of fibrous union was made was not clear. Henderson and Letts reported the ­ second largest series of childhood scaphoid nonunions, some 20 patients aged 14 to 18 years.54 Thirteen patients were treated using a bone graft without fixation, and 1 patient had screw fixation alone. All the fractures healed, but the ones with fixation healed in an average of 8 weeks, compared with 3.7 months for those without fixation. In one patient, the screw backed out into the distal radius and had to be removed, but the fracture healed. Eleven patients who healed after interposition wedge graft and fixation with three K-wires were reported by Duteille and Dautel.52 These authors advised against a trial of cast immobilization, believing that the fracture might heal in a shortened position. Garcia-Mata achieved union in four childhood and scaphoid nonunions using a Matti-Russe graft from the proximal ulna without fixation.53 Chloros and associates achieved excellent results in 12 children, aged 9 to 15 years, using a volar iliac bone graft and Herbert screw fixation.51 All patients had preoperative computed tomographic scans to delineate any shortening or collapse of the fragments.

B

Figure 20A2-3   A, A 10-year-old boy was first seen after a fall on the outstretched hand. The radiograph shows an old fracture of the waist of the scaphoid. There was no definite history of an old injury. B, The fracture went on to solid healing after 36 months of immobilization. Motion and strength of the wrist were normal. Note that healing of an established nonunion is possible with prolonged immobilization. However, a trial of immobilization past 12 weeks is not advocated. (From Graham TJ, Waters PM: Fractures and dislocations of the hand and carpus, In Beaty JH, Kasser JR (eds): Rockwood and Wilkins’ Fractures in Children. Philadelphia, Lippincott Williams & Wilkins, 2001, p 352.)

1368 DeLee & Drez’s Orthopaedic Sports Medicine

Waters and Stewart treated three late adolescents, all athletic injuries, having proximal one-third nonunions with avascular necrosis.36 The use of a vascularized bone graft based on the 1,2 intercompartmental supraretinacular vessel with concept wires resulted in union and revascularization. All three were able to return to sports.

Malunion of the Scaphoid Fracture Grundy noted that four of the eight healed scaphoid fractures he collected had irregularity of the radial margin, and three of these patients complained of a painless clicking in the wrist.62 Suzuki and Herbert described children in whom a scaphoid malunion with dorsal intercalary ­ segmental instability (DISI) deformity spontaneously improved with continued growth over a 4-year period.61

possibility if the fracture is not healed.63 It should be noted, however, that a scapholunate advanced collapse (SLAC) wrist has not been reported in a child. A question occasionally comes up whether there is a place for percutaneous screw fixation of the undisplaced scaphoid fracture in the immature wrist. In the circumstance of a senior athlete who is a candidate for a college athletic scholarship and who presents with an undisplaced scaphoid fracture, percutaneous screw fixation may be indicated so that he or she can play his senior year. Consultation with the patient, the parents, and the coach is a necessary prerequisite.

Other Carpal Fractures Fractures of the other carpal bones are relatively unusual. In Nafie’s series of 82 carpal fractures in children, only 11 were fractures of carpal bones other than the scaphoid.27

Author’s Preferred Method Undisplaced scaphoid fractures, if diagnosed early, heal in a cast within 6 to 8 weeks depending somewhat on the age of the patient. If the diagnosis is delayed, union can still usually be anticipated, although immobilization will have to be continued longer. I prefer to immobilize the childhood scaphoid fracture in a long-arm thumb spica cast the first month, converting it to a short-arm thumb spica for another 2 to 4 weeks or until healing is complete. A cast is applied with the wrist in slight flexion, in radial deviation, and extends to just proximal to the interphalangeal joint of the thumb. Physical therapy is usually not required after the cast is removed, and return to athletics is allowed when a functional painless range of motion and good strength return. In the younger patient (≤12 years), if the diagnosis is delayed, I will try cast immobilization even if there are cystic changes and sclerosis at the fracture. CT should not show any shortening or resorption at the fracture site. If no signs of progress toward healing are noted after 3 months of immobilization, then operative intervention is indicated. An Acutrak screw fixation with a radial or iliac bone graft is used in a child with a scaphoid nonunion. If a preoperative computed tomographic scan in the plane of the scaphoid shows resorption with a humpback deformity, correction of the deformity with an interposition wedge graft and screw fixation should be carried out. A long arm thumb spica cast is worn for the first 6 weeks, and then a short arm thumb spica is worn until healing is verified on CT. If the child with a nonunion is very young, it may be advisable to let the fragments ossify further before grafting and fixation is attempted. In children younger than 10 to 12 years of age, fixation with K-wire may be preferable to a screw. At times an obvious nonunion may be completely asymptomatic. Even though we do not know the natural course of an untreated scaphoid nonunion in a child, I believe that union should be sought even if symptoms are not present. Union is easier to achieve in a child, and degenerative changes, as an adult, would appear to be a definite

Capitate Fractures The capitate is the second most commonly injured carpal bone next to the scaphoid. Thirteen capitate fractures in children were collected from the literature, and seven were associated with a scaphoid fracture.64-74 In general, capitate fractures are undisplaced or minimally displaced and heal with simple cast immobilization. Two of the reported capitate fractures were true scaphocapitate syndrome patients in which the proximal capitate fragment was rotated 180 degrees.70,72 Open reduction of the scaphoid and capitate with K-wire fixation was carried out in both cases, and neither patient developed avascular necrosis. This is rather surprising in view of the fact that the blood supply to the capitate is retrograde as it is in the scaphoid. None of the other authors reported avascular necrosis. Minami and associates reported a 13-year-old girl with a nonunion of the waist of the capitate in which the injury had been missed for 6 months.71 Iliac bone grafting resulted in healing. The combination of a fracture of the capitate with a fracture of the scaphoid can mean that there is some ­associated carpal instability even when appearing to be undisplaced in radiographs. A traction view is helpful in this circumstance. All displaced scaphoid and capitate fractures require open reduction with internal fixation. Early Acutrak screw ­fixation after reduction of the rotated proximal capitate fragment gives the best chance for rapid revascularization and healing.

Fractures of the Triquetrum Letts and Esser collected 15 triquetral fractures in children, 2 of which were complete fractures of the body and the remainder being flake fractures best seen on the 45-degree oblique roentogram.75 Triquetral fractures are more common in the 11- to 13-year-old group. The injury resulted, in their series, from impingement of the ulna on the triquetrum in hyperextension and ulnar deviation or by avulsion by the ulnotriquetral or radiocapitate triquetral ligament. Ten of the 15 patients in Letts and Esser’s study had

Wrist and Hand 1369

ulnar-plus ­variance. Healing was prompt in all patients after 3 weeks of immobilization, but 2 of the patients had some persistent discomfort at follow-up. Fractures of the triquetrum may be a more common cause of post-­traumatic wrist pain than is usually appreciated. Many fractures go undiagnosed as a wrist sprain.76 Larson and coworkers reported a nondisplaced ­triquetral fracture in a 5-year-old boy whose scaphoid was also fractured by a crushing mechanism.25 The fractured ­ triquetrum healed but the scaphoid went on to a nonunion. DeCoster and associates reported simultaneous fractures of the triquetrum, scaphoid, lunate, and radius in a 10-year-old.77 Wulff and Schmidt recorded a 12-year-old girl with a triquetral fracture after a fall from a monkey bar, which healed after 6 weeks of immobilization.73 Duteille and Dautel reported a 14 year-old boy with an ­intra-­articular fracture of the body of the triquetrum, which healed after 3 weeks of immobilization (Fig. 20A2-4).52

A

C

Hamate Fractures Ali reported a 16-year-old boy who sustained a fracture of the body of the hamate and a dislocation of the pisiform when a heavy cylinder fell on his wrist.78 The crush injury also caused a compartment syndrome of the wrist and distal forearm. Healing followed open reduction with internal fixation, but the patient had a partial paralysis of the deep branch of the ulnar nerve when last seen. Successful healing after 7 weeks of immobilization was achieved by Nafie in two childhood hamate fractures.27 An isolated hamate fracture in a 16-year-old boy that healed after 3 weeks of immobilization was reported by Wulff and Schmidt.73 Multiple unstable carpal bone fractures, including the scaphoid, capitate, and hamate, were percutaneously pinned by Kamano and associates in an 11-year-old boy who injured his wrist when a stone garden lantern fell on his hand.69 A good clinical result was noted 29 months after surgery. Duteille and Dautel reported a 14-year-old girl with an

B

D

Figure 20A2-4   A-D, A 12-year-old boy sustained this displaced fracture of the triquetrum in a fall from his bike. Open reduction with internal fixation was followed by prompt healing.

1370 DeLee & Drez’s Orthopaedic Sports Medicine

un-united fracture of the hamate that healed after a bone graft and internal fixation with a compression screw.52 Hook of the hamate fractures may be encountered in the older adolescent, usually the result of impact from a racquet handle or a golf club. Fractures of the hamate hook may present with tenderness and soft tissue swelling over the fracture site and may be accompanied by ulnar sensory deficits and intrinsic weakness. These can be difficult to diagnose, but carpal tunnel views, bone scans, and CT can be helpful imaging techniques. A trial of casting may be successful in promoting healing, but if the patient continues to be symptomatic, excision of the hook may be performed with little morbidity.

Lunate Fractures DeSmet and associates reported a displaced transverse fracture of the lunate in an 11-year-old boy treated by open reduction.79 Avascular changes were noted on the 12-month follow-up radiographs. A fracture-dislocation of the lunate in a 12-year-old boy that required open reduction was recorded by Blount.80 The lunate healed, but an associated fracture of the distal radial epiphysis resulted in a growth arrest. DeCoster and associates described a 10-year-old boy with a transcarpal fracture-dislocation that included a separated lunate fracture.77 Open reduction and

A

C

stabilization with K-wires was followed by some abnormal development of the proximal pole of the lunate seen on radiographs at 2 years (Fig. 20A2-5).

Trapezium and Trapezoid Fractures Nafie reported one child with a trapezial fracture that healed after 5 weeks of immobilization.27 Wulff and Schmidt reported a trapezial fracture in a 14-year-old girl after a motor vehicle crash, which healed after 4 weeks in a short arm thumb spica.73 Two children with trapezial fractures were reported by Duteille and Dautel.52 One 11-year-old boy was asymptomatic and received no treatment. The other patient, a 9 year-old boy, sustained a comminuted fracture of the trapezium, and healing followed K-wire fixation.

Pisiform Fractures Ashkan and associates recorded a dislocated pisiform in a 9-year-old girl who also had a widely displaced SalterHarris II fracture of the distal radius and ulna.81 Reduction of the dislocated pisiform followed closed reduction of the radius and ulna fractures. The two children (a 12-year-old boy and a 13-year-old boy) reported by ­ Mancini and ­ associates were both treated by closed

B

Figure 20A2-5   A-C, A 15-year-old boy sustained a fracture of the lunate in a fall off a skateboard. Open reduction and K-wire fixation resulted in good healing without avascular necrosis.

Wrist and Hand 1371

r­ eduction for a fracture-dislocation in the pisiform.82 They, too, had associated physeal fractures in the distal radius. Long-term follow-up in both patients disclosed no residual abnormality.

INTERCARPAL LIGAMENTOUS INJURIES Literature Review Intercarpal ligament injuries in the immature wrist are uncommon and thus liable to be missed. Difficulty is apt to be encountered in interpreting intercarpal spaces and measuring angles in the incompletely ossified carpal bones. Attention is likely to be focused on a more obvious associated fracture of the distal radius, causing the associated ligamentous injury to be missed.

Scapholunate Dissociation In a review of the literature, only six classic scapholunate dissociations verified at surgery were collected.83-86 Zimmerman and Weiland presented a 13-year-old boy who injured his wrist in a fall from a skateboard 6 months before.86 Reduction and fixation of a static scapholunate dissociation with Blatt’s tenodesis resulted in an asymptomatic wrist, return to sports, and maintenance of reduction 13 months after surgery. Dautel and Merle reported a 14-year-old girl with scapholunate instability diagnosed arthroscopically, which they treated by closed percutaneous pinning of the scapholunate joint.85 Six months later, she was free of pain, but a scapholunate gap persisted. Cook and associates described a 14-year-old boy who achieved normal carpal alignment and normal wrist motion 1 year after repair of scapholunate ligament supplemented with Blatt’s tenodesis.84 Alt and associates reported three patients, ages 9, 11, and 12 years, who had surgical repair of a scapholunate ligament tear.83 All the patients had an associated fracture of the distal radius. All three patients were asymptomatic and had normal radiographs after an average follow-up of 2.4 years. The authors stressed the need for early diagnosis as a prerequisite for a good result. Jansen and associates described a 10-year-old girl who had a successful repair of a lunotriquetral ligament tear.87 Hankin and associates reported four cases of a hypermobile scaphoid among 14 teenagers with wrist pain after injury.88 However, no cases were documented as having a true scapholunate dissociation.

Midcarpal Instability Gerard reported the first childhood case of post-traumatic wrist instability in a 7-year-old girl in 1980.89 The patient had increasing wrist deformity for several years, and there was a remote history of a fall. Radiographs were interpreted as showing palmar flexion instability with a widened scapholunate interval. After preliminary traction, the deformity was reduced open, and a reconstruction of the

scapholunate ligament with a tendon was accomplished. Eight months after surgery, the patient was pain free, but motion was significantly restricted. Radiographs showed no change in the restored carpal alignment. A 15-year-old girl with hemiatrophy and a chronic atraumatic palmar midcarpal dislocation was reported by Graham and Jacobson.90 Midcarpal fusion resulted in an increase of 40 degrees in wrist motion, and the patient returned to playing volleyball and bowling. A somewhat similar case, thought to be a chronic volar perilunate dislocation, was reported by Graham and O’Brien.26 Midcarpal fusion was followed by improved motion and function (Fig. 20A2-6).

Volar Dislocation of the Lunate A 10-year-old boy who sustained a volar dislocation of the lunate and a Salter-Harris type II fracture of the distal radius was reported by Giddins and Shaw.5 Delayed open reduction and pin fixation through a dorsal approach was followed by some persistent lunate subluxation at 6 months, and the patient had a sensation of instability. Sharma and associates reported an 8-year-old boy who sustained a volar dislocation of the lunate and a Salter-Harris type III fracture of the distal radius.91 The lunate was replaced through a volar approach, and at 12 months, there was no evidence of avascular necrosis, and motion was full.

Perilunate Dislocations Graham and O’Brien described a 6-year-old girl with a dorsal perilunate dislocation who was seen 6 months after injury.26 She had been treated for juvenile rheumatoid arthritis. Open reduction and pin fixation was followed several months later by recurrent deformity. Trans-scaphoid perilunate fracture-dislocation has been reported in five children. Closed reduction and cast immobilization was successful in a 10-year-old boy reported by Peiró and associates92 and in a 9½-year-old boy reported by Graham and O’Brien (Fig. 20A2-7).26 Aggarwal and associates reported a 12-year-old boy who sustained a trans-scaphoid perilunate dislocation when he fell 5 m.93 Open reduction with internal fixation was followed by transient avascular necrosis of the scaphoid, which resolved during the 5 months of immobilization. At follow-up, motion was excellent. Cases have also been mentioned by Christodoulou and Colton19 and by Light,94 but clinical details were lacking. An unusual carpal fracture-dislocation in a 10-year-old boy was reported by DeCoster and associates.77 The line of demarcation extended through the distal radial epiphysis, across the lunate and triquetrum, and through the ulnar collateral ligament. Two years after open reduction and ligament repair, function was quite good, but there was some abnormal osseous development of the lunate. It should be noted that there is a high incidence of associated injuries of the distal radial physis and metaphysis in these severe intercarpal injuries.5,80,85

Partial Scapholunate Ligament Injuries Reports of children with partial scapholunate ligament injuries having normal radiographs and lacking provocative signs for instability have appeared in the recent ­literature.

1372 DeLee & Drez’s Orthopaedic Sports Medicine

B A

Figure 20A2-6   A and B An 11-year-old girl sustained this injury 7 months previously. She had been x-rayed several times,   but the correct diagnosis of a volar trans-scaphoid perilunate dislocation was not made. C, Intercarpal fusion was performed, and at 1 year post-operative the range of motion was dorsiflexion and palmarflexion, each 30 degrees; radial deviation, 10 degrees; and ulnar deviation, 25 degrees. She has minimal discomfort with strenuous use. (From Graham TJ, Waters PM: Fractures and dislocations of the hand and carpus, In Beaty JH, Kasser JR (eds): Rockwood and Wilkins’ Fractures in Children. Philadelphia, Lippincott Williams & Wilkins, 2001, p 373.)

C A series of 32 children (ages 6.7 to 17.3 years), who had persistent dorsal radial wrist pain after an injury (21 were sports injuries) with normal radiographs and absent provocative signs, was reported by Erp and associates.95 After failed conservative treatment for at least 6 months, arthroscopy confirmed partial scapholunate interosseous ligament tears, which were débrided. Thirty children had Geissler96 II scapholunate tears, and two had Geissler III tears. At follow-up 2 years later, the average modified Mayo wrist score was 91.6, compared with 66.3 preoperatively; however, eight patients (including the two Geissler III repairs) required additional surgery. A large percentage of the patients were found at arthroscopy to have associated injuries, including 13 triangular fibrocartilage tears. The authors thought that most of these partial scapholunate ligament injuries involved the proximal aspect of the central membranous portion of the ligament. The preservation of the strong dorsal component of the scapholunate ligament prevented instability. Snider and associates reported three gymnasts who had successful débridement of a partial tear of the scapholunate ligament and associated chondromalacia changes.97 One patient was a 15-year-old boy, and the other two patients were 19 and 21 years old.

Evaluation Clinically, a patient with scapholunate dissociation has persistent dorsoradial wrist pain and tenderness after a fall on the outstretched hand. Wrist motion is limited by pain. A positive Watson’s sign may or may not be present. The diagnosis of a scapholunate dissociation in a younger child is difficult because of the pseudo-­widening of the scapholunate interval. Comparative radiographs and reference to a chart of ages with appropriate scapholunate intervals is useful.98,99 According to Kaawach and associates, scapholunate adult values of 2 mm are reached at 12 years in males and 11 years in females (Table 20A2-2).98 An increase in the scapholunate angle may be seen in the lateral view, but this may be difficult to measure in the younger child’s radiograph. An anteroposterior view in supination with the patient making a fist is helpful in diagnosing a dynamic scapholunate instability.

Treatment Options Treatment guidelines generally follow those for adult intercarpal ligamentous injuries. Trans-scaphoid ­perilunate ­dislocations need to be reduced by closed or open methods,

Wrist and Hand 1373

B

A

D

C

E

Figure 20A2-7   A and B, A 9.5-year-old boy sustained a trans-scaphoid perilunate dislocation of the wrist when he fell 60 feet off a viaduct. C and D, Closed reduction was performed 8 days after injury because the correct diagnosis had initially been unrecognized. Plaster immobilization was maintained for 4.5 months because of transient avascular necrosis of the proximal pole and delayed healing. E, Three years after injury, there was a slight irregularity of the proximal pole, but the patient had normal function. (From Graham TJ, Waters PM: Fractures and dislocations of the hand and carpus, In Beaty JH, Kasser JR (eds): Rockwood and Wilkins’ Fractures in Children. Philadelphia, Lippincott Williams & Wilkins, 2001, p 357.)

and the fractured scaphoid needs to be internally fixed. Complete scapholunate tears with scapholunate dissociation that are diagnosed early (≤6 weeks) can ­usually be reduced and repaired with or without an augmentation with a dorsal capsulodesis. Partial scapholunate and lunotriquetral tears without demonstrable instability may be symptomatically improved by débridement alone. Chronic scapholunate dissociations in which the ligament cannot be repaired but are reducible, can usually be improved with a reduction of the subluxated scaphoid and a dorsal capsulodesis.100-102

Author’s Preferred Method Scapholunate dissociation in the acute and subacute (≤6 weeks) stages is explored using a longitudinal dorsal incision, which allows for a dorsal tenodesis if necessary. Using “joysticks,” the scapholunate dissociation, including any associated DISI, is reduced, and the scapholunate and scaphocapitate joints are pinned for 6 weeks. The ligament, which is usually torn in substance or avulsed from Continued

1374 DeLee & Drez’s Orthopaedic Sports Medicine

Author’s Preferred Method—cont’d the scaphoid, is repaired using suture anchors or sutures, or both. If a repair is performed after the acute stage has passed, a dorsal tenodesis is usually added using suture anchors distally in the scaphoid. The pins are removed in 6 weeks, and part-time splinting is continued for another 4 weeks. Physical therapy is often required to regain motion and strength, particularly if the dorsal tenodesis was added. Return to competitive sports should be possible after 6 months. Scapholunate dissociation that is older than 6 weeks is treated in a similar manner if reducible. In the chronic stage, dorsal tenodesis is always necessary, even if the ligament is repairable. Patients with persistent dorsoradial wrist pain after an injury, despite conservative treatment for at least 6 months, need an arthroscopic examination even though there are no radiographic or clinical signs of scapholunate instability. If a partial tear of the scapholunate ligament is found, it should be débrided, along with any associated chondromalacia changes, and this gives a reasonable chance of relieving pain. Trans-scaphoid perilunate dislocations are usually reducible by longitudinal traction. Open reduction with internal fixation of the scaphoid is then carried out using a volar approach. Removable K-wires are used in the younger child, and a cannulated Acutrak screw is employed in the adolescent patient. After the fracture is healed (this should be checked by CT) and a serviceable range of motion and good strength are obtained, the patient may return to ­athletic participation.

  TABLE 20A2-2  Range of Scapholunate Distances in Children and Adolescents

Scapholunate Distances (mm) Males

Females

Age (yr)

Range (yr)

Mean (yr)

Range (yr)

Mean (yr)

 6  7  8  9 10 11 12 13 14-15

4.2-9.2 3.6-8.6 3.1-9.0 2.5-7.4 2-6.9 1.4-6.3 0.8-5.8 1.2-5.2 0-4.7

6.7 6.1 5.6 5 4.4 3.9 3.3 2.7 2.1

5.8-11.9 5.1-11.1 4.4-10.4 3.7-9.6 2.9-8.9 2.2-8.1 1.5-7.4 0.7-6.7 0-6

8.1 7.4 6.6 5.9 5.2 4.4 3.7 3 8.1

From Kaawach W, Ecklund K, Zurakowski D, Waters P: The 54th Annual Meeting of the American Society for Surgery of the Hand, Poster Session, September 1999, Boston, Mass.

reduced and immobilized in supination. Volar dislocation of the joint is reduced and immobilized in pronation.

Evaluation Patients with a distal radioulnar joint injury present with ulnar-sided wrist pain after a fall on the outstretched hand and often have a painful click on rotation of the forearm. The distal radioulnar joint may be clinically unstable when compared with the normal side. Direct dorsal pressure over the TFCC should elicit pain if there is a tear.

Radiographic Examination

Injuries of the Distal Radioulnar Joint Terry and Waters recorded 29 children and adolescents, 18 of whom were women, who had surgically documented triangular fibrocartilage complex (TFCC) tears.103 All except one of the injuries resulted from a fall, and 52% had concomitant fractures of the distal radius. Twentysix of the TFCC tears were repaired (23 Palmer type IB, 1 type C, and 1 type D), 4 arthroscopically and the rest open (Box 20A2-1). Three Palmer type IA tears were débrided. Coexisting pathology was present in 25 patients. Twelve ulnar styloid nonunions were excised, 12 ­associated distal ­radioulnar joint instabilities were stabilized, 7 ulnar shortenings were done for ulnocarpal impingement, and 3 corrective distal radial osteotomies were performed. There were 24 excellent and 2 good results at 2 years using a modified Mayo wrist score. Kuntz and associates reported an 11-year-old girl who was asymptomatic 1 year after having arthroscopic débridement of a full-thickness TFCC tear.104 Erp and associates recorded 10 children and adolescents with TFCC tears diagnosed arthroscopically.95 Seven had a repair of the torn TFCC, 5 had ulnar shortening, and 2 had excision of an ununited ulnar styloid fracture. All of the patients noted substantial improvement at 2 years’ follow-up. Dislocation of the distal radioulnar joint is a rare injury in children and is more likely to be seen in older adolescents. Acute dorsal dislocations of the distal radioulnar joint are

Radiographs may show a nonunion of the ulnar styloid. A zero position anteroposterior radiograph will show the state of ulnar variance. CT is useful for diagnosing subluxation or dislocation of the distal radioulnar joint. MRI may be helpful in delineating a tear of the TFCC. Arthrograms have been largely supplanted by arthroscopy.

Treatment Arthroscopic examination of the distal radioulnar joint is carried out, and if possible, repairs are accomplished through the scope. This is often achievable with type 1B tears. Type 1D tears require open repair, usually with suture anchors in the radius. A type 1A tear should be débrided. The large un-united ulnar styloid fragment can usually be internally fixed and will aid in reestablishing continuity of the TFCC. Small, more distal symptomatic Box 20A2-1 T  reatment of Traumatic ­Triangular Fibrocartilage Complex Tears 1A. Central tear—débridement 1B. U  lnar avulsion with or without an ulnar styloid ­fracture—repair 1C. Distal avulsion or tear of ulnocarpal ligament—repair 1D. Radial avulsion—repair

un-united fragments can be excised. Ulnocarpal impinge-

Wrist and Hand 1375

ment syndrome with positive ulnar variance requires ulnar shortening. This may be combined with débridement of the TFCC. Associated distal radioulnar joint instability105,106 and malunion of previous radius fractures need to be addressed separately.

Gymnast Wrist Epidemiology Children are entering the sport of gymnastics in increasing numbers and at an early age. Since the early 1970s, participation in this sport has grown 500% to include nearly 1.8 million gymnasts at schools, clubs, and collegiate levels.107 From the time of early entrance into this sport until maturity, the distal radial physis is vulnerable to injury. No other sport stresses the wrist of a child and adolescent more than does gymnastics. In gymnastics, the upper extremity is transformed to a weight-bearing limb in which the transmitted force across the hyperextended wrist can exceed 2 times body weight.108 The stresses induced on the pommel horse and balance bar, combining dorsiflexion and rotation, are especially prone to cause symptoms.

Evaluation According to DiFiori and associates, 45% of gymnasts report pain of at least 6 months’ duration.109 Symptoms consist of dorsal wrist pain accentuated by loading in hyperextension and rotation. Some dorsal swelling may be present, and there is usually some tenderness over the dorsum of the distal radius in patients with distal radial physeal injury. Soreness over the distal radioulnar joint with pain on hyperextension and ulnar deviation may indicate an ulnocarpal impaction syndrome. Gymnasts appear to be at particular risk for developing ulnocarpal impingement, especially in the presence of positive ulnar variance. Examination typically shows tenderness in the ulnar snuffbox and pain with passive ulnar deviation and extension. Ulnocarpal impaction is more apt to be seen in the mature gymnast than in the younger gymnast. Dorsal wrist pain in the immature gymnast can be caused by impaction of the dorsal scaphoid against the rim of the radius (scaphoid impaction syndrome). Radiographs may show a small ossicle or hypertrophied ridge on the dorsal scaphoid. If symptoms persist after conservative treatment, cheloidectomy of the radius or scaphoid, or both, may be required.110

Radiographic Diagnosis Read was the first investigator to report abnormal radiographic findings in the distal radial physis and metaphysis in two young female gymnasts with chronic wrist pain.111 He termed the injury a stress fracture to describe the widening and irregularity of the distal radial physis. Roy and associates reviewed 21 gymnasts with an average age of 12 years who had stress reaction involving the distal radial physis.112 They established the criteria for diagnosis of this entity to include one or more of the following: (1) widening of the distal radial physis, (2) cystic changes

and irregularity of the metaphyseal margin of the physis, (3) a beak effect of the distal aspect of the epiphysis, and (4) a haziness of the physis. Similar changes were noted by Carter and associates in eight adolescent patients—seven competitive gymnasts and one champion roller skater.113 They attributed the changes to a Salter-Harris I stress fracture of the physis and noted that immobilization of the wrist resulted in rapid healing. Shih and associates reviewed 93 MRI examinations in symptomatic adolescent gymnasts.114 Partial closure of the physis was noted in two patients. In 47 radii, there were abnormal findings consisting of horizontal metaphyseal fractures, widening of the physis, physeal cartilage extensions into the metaphysis, and bone bruise. These MRI findings were similar to the MRI findings Jaramillo and associates produced in rabbits by interfering with the metaphyseal blood supply.115 If the metaphyseal vessels are injured, the chondrocytes do not die and calcify but rather continue to proliferate resulting in a widening of the physis. In the immature athlete, the growth plate is especially vulnerable to both acute and chronic trauma during the adolescent growth spurt. During the growth spurt, there is an increase in the rate of growth, which is accompanied by a thicker and more fragile physeal plate. Bone mineralization may lag behind bone growth during the growth spurt, resulting in the bone being temporarily weaker and susceptible to injury.116 Fortunately, most physeal stress reactions heal without significant residual. Only nine gymnasts have developed a distal radial physeal arrest.109 There is, however, a disturbingly high incidence of positive ulnar variance in gymnasts compared with nongymnasts and normative values in skeletally mature subjects of both sexes. In a study of 38 collegiate gymnasts, Mandelbaum and associates found males averaged 2.82 ± 1.9 mm of positive ulnar variance compared to controls who averaged −0.52 of negative ulnar variance.107 To date, however, studies of skeletally immature and mature gymnasts have not linked positive ulnar variance either with chronic wrist pain or radiographic findings of distal radial physeal injury.109 Avascular necrosis of the capitate was identified by Murakmi and Nakajima in two collegiate gymnasts.117 Excision of the necrotic proximal capitate with drilling successfully returned both patients to athletic participation. Snider and associates describe three gymnasts (ages 21, 19, and 15 years) who had partial tears of the scapholunate ligament on arthroscopy and responded well to débridement.97 All three were able to return to training.

Treatment Patients who have wrist pain with radiologic findings of radial physeal injury should cease their gymnastic activities for at least 6 weeks. The wrist should be immobilized in a short arm cast for 4 to 6 weeks. Follow-up radiographs or MRI studies are necessary and should show a significant sign of healing before gymnastic activities are resumed in the asymptomatic patient. If the diagnosis is made early, the prognosis for healing and return to training and

1376 DeLee & Drez’s Orthopaedic Sports Medicine

c­ ompetition is good. When gymnastic activity is resumed, it should be on a gradual basis. Ulnocarpal impingement is fairly common among gymnasts, especially those participating in events that involve ulnar deviation and pronation such as the pommel horse.118 Under normal circumstances, the ulnocarpal joint carries about 15% of the load across the wrist joint, and with 2.5 mm of positive ulnar variance, this can increase to 40%.119,120 Gymnasts with positive ulnar variance are particularly susceptible to developing ulnocarpal impingement. Examination typically reveals tenderness in the ulnar snuffbox, and pain when the wrist is maximally ulnar deviated and pronation and supination stress is applied.121 Initial management includes nonsteroidal anti-inflammatory drugs and activity modification. If there is no improvement, wrist arthroscopy can be performed to débride or repair associated TFCC lesions. Ulnar shortening osteotomy may be the procedure of choice if the young athlete exhibits positive ulnar variance. This has the advantage of preserving radioulnar and ulnocarpal stability but requires an intact TFCC. An interesting case of presumed secondary ulnocarpal impingement was reported by Vender and Watson.122 Their patient was a 17-year-old gymnast who presented with a 3- to 5-year history of ulnar-sided activity-related wrist pain. Her workouts initially lasted 15 hours per week, but during the previous 5 years, she had increased her workouts to 25 hours per week. Synovitis was noted over the distal radioulnar joint, and pain was elicited at the extremes of pronation and supination, which was mildly limited, although there was no tenderness over the distal radioulnar joint. Radiographically, her physes were closed, and she was found to have bilateral Madelung’s deformity, which they thought was acquired in nature. Matched ulnar arthroplasty was performed on her symptomatic left wrist. Patients who experience a growth arrest of the distal radial physis require an ulnar shortening with a concomitant epiphysiodesis of the distal ulna if growth potential remains (Fig. 20A2-8).121

A

Prevention Training and skill development need to be individualized. Increases in training loads and skill progression should be more gradual and reduced during periods of rapid growth.123 Early detection and treatment of wrist pain in young gymnasts will prevent the development of distal radial physeal injury. Wrist braces may be used to prevent wrist pain and may protect against acute injury.124 The use of thicker mats may dampen hyperextension stresses on the wrist.

Kienböck’s Disease Review of Literature Kienböck’s disease is uncommon in the immature wrist. When it does occur in this age group, it has a greater capacity for healing with cast immobilization and a greater chance of remodeling after surgery than does Kienböck’s disease in the adult.125 Kahn and Sherry reported the youngest child with Kienböck’s disease—a 6.5-year-old girl with dermatomyositis on long-term corticosteroid therapy.126 Healing with extended cast immobilization was reported in three patients younger than 12 years by Kim and associates.127 Four other children in their series, all older than 12 years, required surgery. Rasmussen and Schantz presented an 8-year-old with Kienböck’s disease treated with 1 year of cast immobilization.128 Two years after onset, full range of motion was present, the child was asymptomatic, and radiographs showed significant revascularization. ­ Hosking noted clinical resolution of pain and a full range of motion in an 8-year-old boy who was immobilized for 6 months.129 Radiographs, however, showed persistent elongation and flattening of the lunate 1 year after onset. Greene described a 10-year-old girl with cerebral palsy who had Kienböck’s disease.130 Immobilization for 15 months was followed by complete resolution of the osteonecrosis and restoration of carpal height 8 months after onset. A 12-year-old boy who had 4 months of immobilization in a cast with temporary

B

Figure 20A2-8   A and B, A 13-year-old gymnast with chronic wrist pain and 4 mm of positive ulnar variance. Ulnar shortening and epiphysiodesis of the distal ulna resulted in an ulnar-neutral wrist, and eventually she returned to gymnastics.

Wrist and Hand 1377

K-wire fixation of the scaphotrapezial joint was reported by Yasuda and associates.131 In this case at 16 months’ ­­ follow-up, his pain had resolved, range of motion was ­normal, and MRI was interpreted as showing revascularization. Nakamura and associates reported seven patients ranging in age from 9 to 17 years who had Kienböck’s disease and who actively participated in sports—three played ­ tennis, one played volleyball, two participated in Asian martial arts, and one was a fencer.132 Five of the seven patients had surgery (one proximal row carpectomy, three radial shortenings, and one lateral closing radial osteotomy), and five were able to resume sports. Radial shortening has been the most popular surgical option for children and adolescents with Kienböck’s disease. Nakamura and associates reported on 23 patients with Kienböck’s disease, and 5 of these were 10 to 18 years of age.125 After radial shortening in the children, there were four excellent and one fair result. The children’s results were notably better than those in the 18 adults in the series. Iwasaki and associates performed radial shortening or a radial closing wedge osteotomy on 11 patients ranging in age from 11 to 19 years.133 Ten of the patients had excellent results, and 8 of the patients had radiographic and MRI evidence of revascularization of the lunate. Five of the six patients with stage IIIB Kienböck’s disease achieved excellent results. Three authors reported complications following radial osteotomy in children. Recurrence of the Kienböck’s disease after an initially successful radial shortening was reported in a 12-year-old girl by Edelson and associates.134 Failure of fixation of the radial osteotomy was recorded in an 8-year-old girl by Foster.135 Healing followed the second internal fixation, and some carpal height was regained at 36 months. The patient was asymptomatic. Herdem and associates encountered radial overgrowth in a 15-year-old boy who eventually ended up with 9 mm of negative ulnar variance.136 Later follow-up revealed that the patient was pain free and working as a carpenter. Vascularized bone grafting for Kienböck’s disease in a 15-year-old boy was reported by Herdem and associates.136 Although there were some signs of revascularization, persistent pain necessitated a radial shortening procedure. Although there was some healing and remodeling of the lunate, on follow-up radiography 80 months after surgery, there was marked loss of height on the radial aspect of the bone. Moran and associates reported 26 patients (ages 13 to 19 years) having a 4+/5 extensor compartment vascularized bone graft.137 Significant improvement in pain and

Box 20A2-2 R  adiographic Classification of Kienböck’s Disease Stage I: Normal roentgenograms except for a possible linear or compression fracture; positive magnetic ­resonance imaging or bone scan Stage II: Lunate sclerosis with normal size and shape Stage IIIA: Lunate sclerosis and collapse without carpal collapse Stage IIIB: Lunate sclerosis and collapse with rotary ­subluxation of the scaphoid

Stage IV: Carpal degenerative changes

grip strength, prevention of further collapse, and some evidence of revascularization in 71% of the patients were noted. Temporary unloading of the lunate with an external fixator or K-wire fixation of the capitatohamate joint was recommended by these authors.

Evaluation Clinically, there is decreased range of motion and loss of grip strength. Some swelling and dorsal tenderness is common because of the secondary synovitis. Radiographic or MRI evidence of avascular necrosis of the lunate makes the diagnosis of Kienböck’s disease. Staging of the disease using Lichtman’s classification should be performed. In stage I, there is a linear or compression fracture but otherwise normal configuration and density. In stage II, density is abnormal without collapse. Lunate collapse is present in stage III, separated into IIIA without carpal collapse and IIIB with carpal collapse. In stage IV, extensive osteoarthritis changes are present (Box 20A2-2).

Treatment Options Treatment depends somewhat on the stage of the disease. Operative treatment for Kienböck’s disease in stages I to IIIB includes joint-leveling procedures (radial shortening or ulnar lengthening) in those patients with a negative ulnar variance. Intercarpal fusions (triscaphe arthrodesis, capitatohamate fusion), capitate shortening, and a variety of vascularized bone grafts are available. For stage IV disease, a proximal row carpectomy or radial carpal ­arthrodesis is indicated.

Author’s Preferred Method Prolonged cast immobilization for at least 3 months should be tried in patients with Kienböck’s disease who are younger than 12 years. If there is no clinical or radiographic improvement in 3 months and symptoms warrant surgery, I proceed with a radial shortening osteotomy in the patient

with a ­negative ulnar variance. In the symptomatic patient with ulnar neutral or positive ulnar variance, I would consider a vascularized graft with temporary wire fixation of the scaphocapitate joint for 8 weeks to unload the lunate (Fig. 20A2-9). Continued

1378 DeLee & Drez’s Orthopaedic Sports Medicine

Author’s Preferred Method—cont’d

A

B

C

D

Figure 20A2-9   A, A 13-year-old girl was seen with a 8-month history of right wrist pain. X-rays demonstrate Lichtman stage IIIA KienbÖck’s disease with neutral ulnar variance. B, An MRI confirmed uniform avascular necrosis of the lunate. C, A 4-5 ECA vascularized bone graft was carried out, with temporary fixation of the scaphocapitate joint to unload the lunate. D, An MRI at 3 months postoperative shows early evidence of revascularization.

S U G G E S T E D C

r i t i c a l

P

o i n t s

l KienbÖck’s disease is a rare cause of adolescent wrist pain. l Immobalization of the wrist has a greater chance of revascularization in the child than in the adult. l Staging the disease is important in determining treatment. l The surgical treatment of KienbÖck’s disease in the child is the same as in the adult. l The prognosis for remodeling following surgery is greater in the child than in the adult.

R E A D I N G S

Herdem M, Özkan C, Bayram H: Overgrowth after radial shortening for Kienböck’s disease in a teenager: Case report. J Hand Surg [Am] 31:1322-1325, 2006. Iwasaki N, Miriami A, Ishikawa J, et al: Radial osteotomies for teenage patients with Kienböck’s disease. Clin Orthop 439:116-122, 2005. Moran SL, Cooney WP, Berger RA, et al: The use of the 4+5 extensor compartment vascularized bone graft for the treatment of Kienböck’s disease. J Hand Surg [Am] 30:50-57, 2005. Nakamura R, Imaeda T, Kiyoshi S, et al: Sport-related Kienböck’s disease. Am J Sports Med 19(1):88-91, 1991. Nakamura R, Imaeda T, Miura T: Radial shortening for Kienböck’s disease: Factors affecting its outcome. J Hand Surg [Br] 15:40-45, 1990.

R eferences Please see www.expertconsult.com

Wrist and Hand 1379

S E C T ION B

Hand 1. Athletic Injuries of the Adult Hand Robert Atkinson

Every physician who cares for patients will treat athletes with hand or wrist injuries. Although not every team or sports medicine physician needs to be a “hand surgeon,” the frequency of hand and wrist injuries in the athlete ensures that we will all be called on to diagnose and treat these injuries. Whether it is the little league catcher or the elite football player, athletes with hand injuries intuitively appreciate the interaction of upper extremity function and optimal athletic performance. Each individual sport has its distinguishing characteristics, with specific demands placed on the athletes’ hands. Most sports are hand intensive, and athletes recognize that a hand or wrist injury will have a severe impact on their performance.

The desire to participate and drive for optimal performance makes most athletes special patients. Caring for this unique patient population requires an understanding of their wishes for a fast and complete recovery. As treating physicians we need to understand not only the anatomy and pathomechanics of the athlete’s injury but also the financial, social, and psychological consequences. As the treating physician, surgical skill and intelligence needs to be coupled with an ability to make decisions that keep the athlete’s long-term optimal recovery paramount. The elite athlete’s special functional demands should inspire us to help by optimizing treatment and recovery.

EPIDEMIOLOGY Each sport has its specific hand injuries.1 Golfers, for example, are at risk for hook of hamate fractures.2 Berger has studied the epidemiology of hand sports injuries.3,4

SPORTS-RELATED BIOMECHANICS AND INJURY PATTERNS IN THE HAND The complexities of hand biomechanics are beyond the scope of this chapter, and an excellent review has been published by Dr. Paul Brand.4a Four mechanisms of injury and mechanical factors responsible for hand injuries have been proposed by Mirabello and colleagues: throwing, weightbearing, twisting, and impact.5 Frequently a combination of factors is responsible for the injury. Werner and Plancher categorized sports potential for injury.6 Mechanisms included were impact with ball or competitor; contact with racket, stick, or club; and external contact, seen in gymnastics, rock climbing and weightlifting. Although almost any injury can be seen in any athlete, certain patterns of

injury have been established. Interphalangeal (IP) fracture­ islocations are frequent in high-level ball sports (volleyball, d football), hamate fractures are seen in golfers and tennis players, gamekeeper’s thumb (ulnar collateral ligament [UCL] tear) is seen in skiers. Some injuries are notoriously easy to miss, such as a profundus flexor tendon avulsion in a football player. A high index of suspicion is the first step toward an accurate diagnosis in the injured athlete.

CARING FOR THE ATHLETE In general, caring for the athlete can be more difficult and charged with emotion, especially as the athlete reaches elite or professional status. Coaches, managers, agents, and parents may all have different agendas and levels of knowledge that can be quite disparate. Our goal is to optimize treatment to ensure the best clinical and functional result for the athletes. The patient management questions set forth by Green and Strickland remain a good guide to treatment7: 1. Is the method of treatment expected to provide the best long-term result? 2. Would we manage this injury in a similar manner in a nonathlete? 3. Are the potential complications of my treatment significantly greater than might be expected from a more conservative approach? 4. Will the treatment allow the athlete to return to competition with little risk for reinjury? 5. Would reinjury unfavorably influence the prognosis for a satisfactory recovery? These guidelines apply across the treatment spectrum of sports medicine.7 Careful communication, confidentiality, and a patient-centered attitude are all essential to gain the athlete’s trust and obtain the maximal outcome.

OSSEOUS AND SOFT TISSUE INJURIES OF THE FINGERS Ligamentous Injuries and Dislocations of the Fingers A dislocation of the metacarpophalangeal (MCP) or IP joint is a frequently encountered sports injury. Torsion, rotational and axial stresses, and hyperextension or ­flexion can result in dislocation. Most IP dislocations can be

1380 DeLee & Drez’s Orthopaedic Sports Medicine

e­ ffectively treated closed, whereas some MCP dislocations require operative treatment. Recognition of the complex MCP dislocation allows timely, appropriate intervention.

Metacarpophalangeal Joint of the Fingers The bony architecture of the MCP joint allows for significant motion, including hyperextension and flexion in the sagittal plane, adduction-abduction in the frontal plane, and rotatory motion of the base of the proximal phalanx (P-1) on the metacarpal head. The cartilaginous surface of the metacarpal head has a trapezoid shape, being broader on the palmar surface. MCP joint stability is a function of surrounding collateral and accessory collateral ligaments, volar plate, capsule, and extrinsic flexor and extensor tendons.8 The collateral and accessory collateral ligaments provide lateral stability. The collateral ligaments originate dorsal to the metacarpal head axis of motion and insert into tubercles on the sides of the P-1. Accessory collateral ligaments have their origin palmar to the proper collateral ligaments and insert into the palmar base of P-1 and the volar plate.9 Because of the dorsal metacarpal origin of the collateral ligaments and the cam shape of the metacarpal head in the sagittal plane, the ligaments are lax in extension, but taught in flexion. This is the basis for the recommendation that MCP joint injuries be splinted in full flexion, the socalled safe position. The safe position for the IP joints is in extension, as a result of the anatomy in that area. Accessory collateral ligaments, along with interosseous and lumbrical tendons, provide additional lateral (adduction-­abduction) stability. The palmar (volar) plate provides a block to hyperextension and is the third side of the anatomic box that provides MCP stability. The palmar plate has a broad firm distal attachment to P-1 and a membranous, loose origin from the metacarpal neck. The laxity of the collateral ligaments in extension puts the palmar plate at risk for rupture (usually proximally) with MCP hyperextension. The dorsal capsule is loose, and the extrinsic extensor tendons extend the MCP joint through the sagittal bands’ attachment to the base of P-1 and the volar plate.

Lateral (Coronal Plane) Injuries with Collateral Ligament Injury Collateral ligament injuries in the fingers are much less common than similar thumb MCP ligament ruptures. Most collateral ligament injuries in the fingers are partial

Figure 20B1-1   Note the sesamoids and the attached volar plate perched at the dorsal metacarpal head.

(grade I or II) and can be treated with early active range of motion and buddy splinting to an adjacent finger. Collateral ligament injuries usually occur in an ulnar direction when the finger is flexed and the collateral ligament is taut. Middle, ring, and little fingers are at increased risk.10 All suspected collateral ligament injuries should be radiographed. In a complete (third-degree) tear, the joint is unstable, and avulsion fractures can occur. Testing of the affected collateral ligament is done in flexion.11 If the joint is unstable, or if an avulsion fragment is displaced, I prefer to perform an open repair. If the ligament is avulsed from bone, then a mini Mitek suture anchor works well for fixation. If a fracture fragment is avulsed, my preference is to tension-band the fragment or fix it with a minifragment screw, depending on its size. Immobilization of the joint in flexion prevents an extension contracture from developing. In chronic tears, pinning of the MCP joint in flexion may be indicated for 2 to 3 weeks, with active motion begun thereafter.

Dorsal Dislocations of the Metacarpophalangeal Joint Classification Although MCP dislocations can occur in any digit, the index and small fingers are most commonly affected because of their unprotected location. Dorsal MCP dislocations are classified as simple (easily reduced) or complex. Simple dislocations present with the MCP joint cocked up in hyperextension of 70 to 90 degrees (Fig. 20B1-1). These are frequently easily reduced with gentle flexion and slight traction (Fig. 20B1-2). Complex dislocations of the MCP joint are defined by interposition of the volar plate between the base of P-1 and the metacarpal head (Fig. 20B1-3). The volar plate, trapped between the metacarpal head and the base of the proximal phalanx, prevents reduction with simple traction. The lumbrical muscle is wrapped around the radial side of the metacarpal head, and the flexor tendons loop around its ulnar side. Simple traction will only tighten this noose and prevent reduction. The key on physical examination is to note the palmar prominence of the metacarpal head and often dimpling of the palmar skin. The MCP joint is only in slight extension. Posteroanterior radiographs show widening of the joint space, and sesamoid interposition may be noted in the index and small fingers.12,13 A Brewerton view (with the forearm supinated and the MCP joints against the x-ray plate) may show a metacarpal head fracture.14

Figure 20B1-2   Reduction of the metacarpophalangeal joint with the volar plate in its normal position.

Wrist and Hand 1381

Volar Metacarpophalangeal Dislocations Volar dislocations of the MCP joints are rare. Closed reduction is possible and should be attempted soon after injury, under adequate anesthesia.17 Entrapped volar plate or MCP joint capsule can prevent reduction. Open reduction is frequently necessary through a dorsal approach.18 Again, early active motion is encouraged after reduction, and with buddy taping, early return to sports is possible.

Author’s Preferred Method

Figure 20B1-3   Note parallel alignment of P-1 and the metacarpal (complex dislocation).

Always radiograph the hand before attempting a closed r­ eduction. Every athlete deserves an attempt at closed reduction for a dislocated MCP joint under adequate anesthesia. If closed reduction is not successful, an open reduction through a dorsal capsulotomy is my preferred method. Return to sports is guided by return of range of motion and comfort level because chronic instability is not a problem.

Treatment Options Closed treatment is indicated for all simple MCP dorsal dislocations, with early active range of motion. Extension past neutral should be avoided. Buddy taping is appropriate, and the athlete can return to contact activities in 2 to 3 weeks, with protective taping. Complex dislocations usually require open surgical treatment, and palmar and dorsal approaches have been described. If a closed reduction of a complex dorsal MCP dislocation is attempted, a successful outcome is made more likely by the following: (1) Do a wrist block (e.g., median nerve for the index finger), and inject 1% lidocaine (Xylocaine) into the MCP joint itself; (2) flex the wrist to relax the flexor tendons that are wrapped around the metacarpal head; (3) gently guide the volar base of P-1 to contact the dorsal metacarpal head and push the proximal phalanx over the metacarpal articular surface, pushing the volar plate into a reduced position. Check Fluoroscan images for a concentric reduction. If this is the case, and active flexion and extension of the joint is possible, then a successful closed reduction has been achieved.

Author’s Preferred Method Irreducible Metacarpophalangeal Dislocation

If closed reduction is not possible, my preference is to perform a dorsal approach as described by Becton and colleagues in 1975.15 This approach opens the MCP joint capsule and then splits the volar plate in the midline from proximal toward the base of P-1. The joint is easily reduced, and osteochondral fragments can be fixed if needed. Recently, a percutaneous approach to reducing complex dislocations has been reported.16 Early return to sports (1 to 2 weeks) is possible with buddy taping and a protective splint for contact sports.

C M D

r i t i c a l

P

o i n t s

o n

e t a c a r p o p h a l a n g e a l i s l o c a t i o n s

• Simple dorsal dislocations present with a hyperextension posture, whereas complex dorsal dislocations present with the P-1 parallel to the metacarpal. • In simple dorsal dislocations, the base of P-1 is in contact with the dorsal surface of the metacarpal head and is easily reduced. • In complex dorsal dislocations, one attempt at closed reduction is indicated. If unsuccessful, an open reduction is performed through a dorsal approach. • MCP joints should be splinted in flexion to avoid shortening of collateral ligaments and an extension contracture. • If an index finger MCP dislocation is reduced through a volar approach, be careful of the radial proper digital nerve draped over the prominent metacarpal head.

Subluxations, Dislocations, and Ligament Injuries at the Proximal Interphalangeal Joint Anatomy The proximal interphalangeal (PIP) joint is a congruous hinge or ginglymus joint, in which stability is provided by the matching articular surfaces and the combination of a thick volar plate and stout collateral ligaments. The tight fit of the opposing articular contours increases stability when the PIP is under axial load.19 The collateral ligaments are thick and are composed of proper and accessory parts. The proper ligaments insert into the base of the middle phalanx (P-2) and the volar plate, whereas the accessory collateral ligaments insert only into the volar

V.I .P.

1382 DeLee & Drez’s Orthopaedic Sports Medicine

Figure 20B1-5   Note dorsal subluxation with impacted fracture at the base of P-2.

Figure 20B1-4   Eaton’s three sides of a box at the proximal interphalangeal joint. VIP, volar plate.

plate. The volar plate is thick distally where it inserts into the volar lip of P-2, whereas proximally it thins out and has two ­proximal ­projections that attach to P-1, the checkrein ligaments. This arrangement allows the PIP joint to flex 110 degrees. The condyles of the head of P-1 are not cam shaped, and the PIP joint does not stiffen in extension, when immobilized at 0 degrees. On the contrary, the PIP joint has a propensity to develop flexion contractures, with shortening of the volar plate, when immobilized is a flexed position. This propensity to PIP flexion contracture after trauma is most pronounced in the ring and little fingers. This fact and the low tolerance of this joint for any stepoff in the articular contours make the PIP joint a small joint that causes big clinical problems. Eaton described the soft tissue constraints of the PIP joint as three sides of a box (Fig. 20B1-4); instability occurs when at least two sides of this box are disrupted.20 The central slip of the dorsal apparatus attaches on the dorsum of P-2 just past the joint surface and is frequently avulsed in palmar PIP ­dislocations.

Classification and Physical Examination PIP dislocations are classified as follows: 1. The direction of P-2 relative to P-1 (dorsal, volar, ­lateral) 2. The presence or absence of associated fractures (central slip dorsal avulsion fracture in a volar dislocation, or a pilon volar articular comminuted fracture with a dorsal dislocation (Fig. 20B1-5) 3. Pure collateral ligament injuries, with or without ­avulsion fractures

History and Clinical Presentation Taking a history from the athlete is often helpful if the athlete can relate “which way the finger went.” If the patient cannot remember and the mechanism of injury is also unknown, the physical examination takes on added importance. Most PIP dislocations are closed injuries,

but open dislocations are not uncommon and require appropriate respect and treatment. A central slip injury or ­avulsion should be suspected in all PIP injuries with dorsal ­tenderness. Three tests can be done to ensure the central slip is still attached to the dorsal base of P-2: 1. Check “tenodesis extension” of the PIP joint with the MCP joint in full flexion. The PIP should extend to within 15 degrees of full extension. 2. Passively extend the PIP joint and ask the patient to flex the distal interphalangeal (DIP) joint. Inability to do this implies retraction of a ruptured central slip, with extension forces concentrated at the DIP joint through the lateral bands. 3. Flex the PIP to 90 degrees. If the patient can actively resist flexion of the DIP joint, then the central slip is ruptured. (See the section “Closed Boutonnière” for a description of Elson’s test to document central slip ­rupture.) Radiographs of the injured finger are mandatory, and anteroposterior, lateral, and oblique views are routine.

Treatment Options and Author’s Preferred Method Pure Hyperextension Injury (Type I)

These injuries imply a partial injury to the volar plate and are stable on examination to active and passive extension. Treatment involves buddy taping for 3 to 4 weeks, with Coban elastic wrap used at night to decrease edema. Simple Dorsal Dislocation

By definition, the volar plate is injured, but the articular surfaces of the joint are still in contact. This is treated with closed reduction and buddy taping. If the joint has a tendency to hyperextend, then a splint can be fashioned to sit dorsally and keep the joint flexed 15 degrees. Even simple dislocations can be quite painful, and the athlete will need to work with an occupational therapist to regain full flexion within the first 3 weeks after injury. Active and passive flexion is performed during the first 3 weeks. At 3 weeks, active extension is emphasized, with a resting splint in full extension. If a flexion contracture remains 5 weeks after injury, dynamic splinting with an LMB splint or Capener splint should be prescribed. Return to sport can happen within

Wrist and Hand 1383

Fracture-Dislocation (Type III)

Figure 20B1-6   Note the bayonet position of phalanges.

7 to 14 days, depending on pain, swelling, and improvement in digital motion. Complex Dorsal Dislocation (Type II)

This injury is defined by displaced articular surfaces and bayonet position of the phalanges (Fig. 20B1-6). ­Rupture of the palmar skin can occur, and residual stiffness is common.21 When no major fracture fragments are present, a closed reduction under a Xylocaine digital block is attempted by hyperextending the PIP joint and gently pushing P-2 over the articular surface of P-1. Occasionally, collateral ligaments or volar plate may be entrapped, necessitating an open reduction through a dorsal approach. The dorsal surgical approach splits the interval between the central slip and the lateral band on one side. The volar plate is cut in the midline and the joint reduced.15 Stability is checked after reduction, and a dorsal splint, blocking terminal extension, is fabricated for use in the first 2 to 3 weeks, with the PIP joint in 20 to 30 degrees of flexion. Again, residual flexion contracture is a concern, and active extension is emphasized at 3 weeks. Return to sports, if contact is anticipated, is delayed for 3 to 4 weeks. Buddy taping during sport activity is recommended until 6 weeks after injury or until motion is nearly full.

A

Dislocations associated with significant volar fracture fragments are difficult to treat and are classified as stable or unstable. Various authors have classified these injuries, and there is a variety of fracture configurations, from simple large volar fragments to complex comminuted pilon injuries involving the entire articular surface of P-2.22 I find it helpful to categorize these injuries by percentage of articular surface involved. Fractures with 0% to 30% involvement are usually stable after reduction and can be treated with dorsal block splinting or a dorsal K-wire inserted into the dorsal head of the proximal phalanx, under Fluoroscan control, to block the last 30 degrees of extension. The pin is removed at 3 weeks, but active flexion is encouraged during this time. If 30% to 50% of the joint surface is involved, these injuries are usually unstable and require operative fixation. Fractures with involvement of more than 50% of the joint surface at the base of P-2 are almost always unstable and frequently require surgical care. If a stable closed reduction cannot be obtained, operative fixation is indicated. If the volar fragments are large, operative fixation is performed with miniscrews (Fig. 20B1-7) or K-wires (Fig. 20B1-8) through a volar approach.23 A Bruner or curvilinear ­incision in made based at the level of the A-3 pulley, and flexor tendons are retracted after the flexor sheath between the A-2 and A-4 pulleys has been opened. The volar fragments are reduced and held with 1- or 1.5-mm screws (Figs. 20B1-7 and 20B1-9). This procedure is demanding technically, and residual stiffness may result. Athletes must be warned that their range of motion may not return to normal and that residual swelling at the PIP joint is the norm. If the volar fragments are comminuted, open reduction needs to be combined with pin fixation, either statically fixed or with dynamic traction.24 Badia and associates described a modification of a traction technique, originally reported by Gaul and Rosenberg (Fig. 20B1-10).25 Once the fixator is applied, a limited open reduction can be effected through a midaxial approach to elevate any articular fragments not reduced by the traction device. This is my preferred method for these complex and challenging fractures. Chronic Fracture-Subluxation

It is surprising how many athletes, and patients in general, ignore an injured, stiff PIP joint. When a PIP ­ fracturesubluxation or fracture-dislocation is chronic (older than 4 weeks), often only a reconstructive procedure will improve

B

Figure 20B1-7   A, Note the impacted proximal interphalangeal (PIP) joint surface and PIP joint subluxation. B, Note reduction of joint alignment and fixation of volar fragments with 1-mm screws.

1384 DeLee & Drez’s Orthopaedic Sports Medicine

Figure 20B1-8   Proximal interphalangeal joint fracture-subluxation (left) treated with static pinning, with restoration of joint surface (right).

the PIP function. Although a detailed description of the treatment of this entity is beyond the scope of this chapter, both volar plate arthroplasty26,27 and the newer technique of hemihamate arthroplasty for reconstructing the base of P-2 are effective options.28 I prefer the hemihamate reconstruction in young active patients who have a neglected PIP fracture-dislocation because it comes closest to providing a nearly normal joint (Fig. 20B1-11). The reader is directed to the previous two references if a detailed description of these procedures is desired. Volar or Palmar Subluxation or Dislocation

Volar dislocations of the PIP joint are rare and must not be missed.29,30 Their treatment as a chronic problem results in suboptimal function. An avulsion of the central slip often occurs with volar dislocations. Most acute volar dislocations are successfully reduced closed, and the central slip avulsion is allowed to heal with the PIP joint splinted in extension for 5 to 6 weeks. A static Bunnell, Stack, or Orthoplast splint with the DIP joint free works best, and the patient is encouraged to flex the DIP joint in the splint. At 6 weeks, a dynamic extension splint is used for an additional 6 weeks. Volar PIP dislocations with avulsion fractures of the central slip attachment or central slip avulsion are repaired

open, with a tension band technique using a 26-gauge wire or a minifragment (1.1- or 1.3-mm) screw, or suture anchor, through a dorsal approach (Fig. 20B1-12). Rehabilitation for these injuries is similar, protecting the dorsal central slip insertion for 12 weeks. Return to sports can occur 3 months after fixation.

Distal Interphalangeal Joint Dislocation of the Fingers and Interphalangeal Dislocation of the Thumb Dislocations of the DIP joints of the fingers and IP joint of the thumb are less common than PIP dislocations for the following reasons: 1. The short lever arm of the distal bony segment 2. The tight fit of skin, flexor and extensor insertions, and congruous articular surfaces 3. Snug collaterals, inserting into lateral tubercles of the distal phalanx Dislocations of the terminal phalanx are dorsal or lateral and can often be associated with open wounds. Concomitant fractures and injuries to the flexor and extensor insertions need to be assessed.

Treatment Options and Author’s Preferred Method

Figure 20B1-9   Volar approach for fixation of P-2 volar lip and pilon fracture. Arrow points to the volar fragment.

A closed injury is reduced, under digital block anesthetic, after radiographs rule out a significant fracture. I prefer 1% or 2 % plain Xylocaine, given as a digital block at the level of the metacarpal heads. Reduction is performed with traction and pressure from the dorsal aspect of the distal phalanx. Stability is checked after reduction. Radiographs are obtained, and the digit is splinted is slight flexion. Active motion can begin after 5 to 7 days, with a block to terminal extension at 20 degrees.31 Dorsal splinting can be removed at 3 weeks. Open injuries need irrigation and débridement in a sterile setting. Skin closure is done with 5-0 nylon or Prolene suture. Irreducible dislocations have been reported but are rare.32,33 Also rarely reported are simultaneous dislocations of both IP joints in the same digit.34 Always check flexor and extensor tendon function after reduction.

Wrist and Hand 1385 Figure 20B1-10   A, Proximal interphalangeal joint impacted pilon fracture with comminution. B, Reduction of joint surface with K-wires and traction device (Rosenburg’s technique).

A

B

Figure 20B1-11   A, Neglected proximal interphalangeal joint fracture-subluxation with degenerative changes. B, Note the chronic dorsal fracture-subluxation, treated with hemihamate arthroplasty on the right.

A

B

1386 DeLee & Drez’s Orthopaedic Sports Medicine

A

C

B

Figure 20B1-12   A, Rare volar proximal interphalangeal joint dislocation. B, Volar dislocation reduced. C, Central slip avulsion fixed with suture anchor.

Postoperative Care and Return to Sports Most of these DIP and IP dislocations are stable after reduction. Treat open injuries with an operative washout, and document flexor and extensor function after reduction.

 r C I n D i F r l Open l

l

l

l

i t i c a l

P

o i n t s

o n

t e r p h a l a n g e a l s l o c a t i o n a c t u r e

- D

a n d

i s l o c a t i o n s

dislocations necessitate an appropriate washout and antibiotic (cephalosporin) coverage. Most dorsal PIP dislocations can be reduced closed and radiographed to rule out concomitant fractures. Dorsal PIP fracture-dislocations demand respect as a significant injury. Pilon injuries and those with residual dorsal subluxation are best treated operatively with restoration of joint surfaces and congruity. Comminution at the base of P-2 may require traction to maintain ­reduction. Volar lip P-2 fractures associated with PIP dorsal subluxation are best approached volarly, with reduction and fixation of the fragment with miniscrews. Volar PIP dislocations frequently tear the central slip attachment to P-2, and the central slip tear must be treated to prevent a boutonnière deformity.

Low-profile splinting and buddy taping can allow early return to sports.

Carpometacarpal Dislocations of the Fingers Carpometacarpal (CMC) injuries are less frequent than the more distal dislocations of the digits. Usually, significant force is necessary to cause a CMC dislocation, and sports mechanisms include the following: 1. Clenched-fist injuries, in which the more mobile fourth and fifth metacarpal-hamate articulations are at risk. 2. Crush injuries from whatever cause can disrupt even the stabile columns of the index-trapezoid and long finger– capitate joints.

Anatomy The ring and small finger metacarpal bases are more mobile and less stable than the index and long metacarpal bases. The ring and small metacarpal bases articulate with two separate facets of the hamate. The base of the small metacarpal is convex from dorsal to palmar.35 This and the insertion of the extensor carpi ulnaris tendon on the dorsal base of the metacarpal make the fifth CMC more unstable in a dorsal direction. The index base articulates with the trapezoid and also has facets that contact the trapezium, capitate, and long finger metacarpal. The long ­ finger

Wrist and Hand 1387

metacarpal base articulates primarily with the capitate and less with the index and ring metacarpal bases. The index and long metacarpals have strong interosseous ligaments and are inherently more stable. The extensor carpi radialis longus and brevis tendons insert into the base of the index and long metacarpals, respectively.

and pinned if the ­articular surfaces of the metacarpal bases are preserved. If a concomitant dorsal coronal fracture of the hamate is present, ORIF of the hamate fracture is performed using a 2- or 2.4-mm screw, and the ring and small metacarpal bases are reduced and pinned to the volar part of the hamate. Not uncommonly, the ring and small metacarpal bases are fractured and comminuted. In this situation, I prefer a reduction with finger trap traction (10 lb) and then percutaneous pinning if reduction is adequate.41,42 Open reduction is reserved for neglected cases or when a closed anatomic reduction is not possible. If articular surfaces are destroyed in chronic cases, a CMC fusion is recommended.

Clinical Presentation and History A history of ulnar-sided hand pain and deformity is typical, with crushing or axial force common mechanisms of injury. Although volar dislocations have been reported,36 dorsal dislocations and fracture-dislocations make up the majority at the CMC level and usually involve axial and shear forces across the articular surfaces.37-39 Radiographic evaluation is mandatory, and anteroposterior, lateral, and 30-degree pronated lateral radiographs of the hand help to define the injury (Fig. 20B1-13). Concomitant fractures at the bases of the metacarpals and the dorsal aspect of the hamate are common. If the extent of fracture associated with the CMC dislocation is unclear, I prefer computed tomography (CT) to help define the injury.

Author’s Preferred Method Return to Sports

and

I prefer closed reduction and percutaneous pinning for acute cases of CMC dislocation. Pins enter the ulnar base of the ring and small finger metacarpals and cross into the hamate. They can be removed in 4 to 5 weeks. MCP motion begins early with this pinning technique, and pins can be buried just under the skin. Pin removal is performed in the office under local anesthesia. If ORIF is necessary, I use a dorsal zigzag incision with an ulnarly based skin flap for ring and small CMC stabilization. Care is taken to preserve ulnar nerve dorsal sensory branches that cross the field. After pin removal, a low-profile hand-based splint is made, and emphasis is placed on MCP joint flexion arc of motion. Return to sports can occur shortly after pin removal, with precautions for contact sports.

Treatment Options Closed reduction of acute dislocations at the CMC level can usually be achieved with adequate anesthesia (auxiliary or intravenous Bier block), although irreducible dislocations have been reported.40 Stability of the reduction is checked, and if unstable, percutaneous K-wire fixation is recommended. If the common fourth and fifth CMC dorsal dislocation is encountered, a K-wire is placed from the proximal ulnar side of the fifth metacarpal into the hamate under fluoroscopic control. One 0.045-inch K-wire for each metacarpal is sufficient. Uncommon central CMC dislocations (index and long) may require open reduction with internal fixation (ORIF) and pinning. Neglected dislocations seen late require ORIF, through a dorsal approach. In these situations, the CMC joint space must be cleared of fibrous tissue before reduction. Dislocations over 1 month from injury can be openly reduced

A

Tendon Injuries Because most sports are hand intensive, open and closed injuries to the flexor and extensor tendons are not ­uncommon. For open injuries, examination, exploration, and repair of injured structures are done acutely. Closed injuries to the extensor and flexor systems in the hand can

B

Figure 20B1-13   A, Ring and small carpometacarpal (CMC) dorsal dislocation. B, Reduction (open) of chronic CMC dislocation with pinning.

1388 DeLee & Drez’s Orthopaedic Sports Medicine

be subtle and neglected and can cause sufficient morbidity that they deserve special mention. On the dorsal side of the digits, mallet finger at the DIP joint, boutonnière deformity at the PIP joint, and sagittal band rupture, or “boxer’s knuckle,” at the MCP joint represent injuries to the dorsal apparatus that can interfere with integrated digital extension and flexion. On the flexor side, flexor digitorum profundus avulsion, or “jersey finger,” is a serious and often neglected injury that can ruin finger composite flexion.

Mallet Finger The anatomy of the dorsal apparatus of the fingers is complex and has generated detailed descriptions.43-45 Its delicate balance allows the integration of intrinsic and extrinsic muscle function to coordinate fine digital motion. Disruption of the terminal extensor tendon’s attachment into the dorsal base of the distal phalanx is common in sports. Axial load and forced flexion of the DIP joint can stretch the terminal tendon, avulse the tendon attachment, or cause an avulsion of a variable amount of bone from the dorsal ridge of the distal phalanx. Synonyms for this injury are baseball finger and drop finger, and jamming injuries in ball sports are common. Warren and associates described an area of terminal tendon hypovascularity as an area of frequent injury.46 Often players who present with an inability to extend the DIP joint after an injury also have laxity at the PIP joints, with hyperextension possible. This results in the DIP joint having a slightly flexed posture and may make it prone to mallet injury. The physical examination demonstrates the drooped posture of the DIP joint with an inability to completely extend the joint. Swelling and tenderness are variable. Radiographs are obtained to define any bony injury, especially an avulsion fracture associated with subluxation of the joint.

Treatment Options Splinting is the treatment of choice for almost all mallet finger injuries.47-51 Open treatment can result in frequent complications,52 and the surgeon should resist the temptation to “make good better” by operating on these injuries. This is one of the few clinical entities in which the overwhelming evidence favors nonoperative treatment, even if slight subluxation of the DIP joint is seen on the initial lateral radiograph.

Author’s Preferred Method Return to Sports

and

I treat most mallet finger injuries, acute and chronic, nonoperatively. I use a commercially available plastic STACK splint, or a dorsal aluminum foam-padded splint, with a slightly hyperextended posture. I use ½-inch adhesive tape to apply the splint and instruct the patient to wear the splint continuously, and not to remove it. I see the patient every 2 weeks in the office to ensure compliance and to clean the finger. Palmar and dorsal splinting can be performed in contact sport athletes who want to play during the period of immobilization. Using a plastic glove or finger cot makes showering manageable. An unstable fracture-dislocation of the DIP joint (rare) is the only scenario in which operative treatment warrants consideration. Splinting is done for 6 weeks routinely. In neglected cases, splinting can be carried out for a longer period.

Closed Boutonnière The communis extensor tendon divides into three parts over the PIP joint of the digit. The central slip inserts into the dorsal base of the middle phalanx just distal to the joint surface. Two lateral slips of the extrinsic extensor tendon separate from the central slip just proximal to the PIP joint and accept a tendinous contribution from the intrinsic muscles to become the conjoined lateral bands at the level of the middle phalanx. These conjoined lateral bands are joined by the oblique retinacular ligament, which facilitates conjugate extension of the PIP and DIP joints. In a closed injury, with forced flexion of the PIP joint, the central slip can rupture from its insertion onto the middle phalanx base. In addition, frequently the lateral bands’ connection to the central slip will be torn along with the flimsy ­triangular ligament that connects the lateral bands at the level of the middle phalanx. This combination allows volar subluxation of the lateral bands. They migrate volar to the PIP joint axis of motion. Thus, the lateral bands become flexors of the PIP joint and extensors at the DIP joint, and this defines the boutonnière posture (Fig. 20B1-14). Avulsion fractures at the central slip attachment site can also cause this deformity.53 Boutonnière deformities can be acute or chronic.

Figure 20B1-14   Boutonnière with central slip rupture. Note distal interphalangeal hyperextension.

Wrist and Hand 1389

Figure 20B1-15   Stack boutonnière splint, allowing distal interphalangeal flexion for boutonnière treatment.

Physical Examination and Testing Elson’s test has been shown to accurately diagnose central slip ruptures.54,55 In this test, the affected finger is placed over a table top, and the PIP is flexed 90 degrees. Any active extension of the middle phalanx is due to an intact central slip. In addition, if active extension of the distal phalanx is noted, the central slip is ruptured, and the DIP is extended by tight lateral bands. Elson’s test will not diagnose a partial injury to the central tendon and can be impeded by pain or lack of patient cooperation.35 A confirmatory test is to hold the PIP joint passively extended and ask the athlete to flex the DIP joint. An inability to flex the DIP joint indicates a tear of the central slip.

strict attention and must be uninterrupted. If compliance is an issue, percutaneous pinning of the PIP joint is an option, with pin removal at 3 weeks and protected mobilization with a Capener splint for 2 to 4 additional weeks. For the athlete who presents with severe DIP hyperextension (indicating retraction of the central slip), O’Dwyer and Quinton describe operative repair of the central slip with a suture anchor (Fig. 20B1-16) and pinning for 2 to 3 weeks.60 In neglected or chronic boutonnière deformity, the PIP contracture can be supple or fixed. If passive PIP extension is full, splinting can be tried, unless the athlete refuses to devote the 6 to 8 weeks necessary for successful closed treatment. If the PIP contracture is fixed, aggressive active splinting or serial casting is done to effect full PIP extension. Once full extension is attained, static PIP splinting is maintained for 6 weeks. Rarely, a palmar PIP contracture release is necessary, cutting the checkrein ligaments, proximal to the volar plate. This is sometimes combined with open reconstruction of the central slip and mobilization of the lateral bands.

Author’s Preferred Method Most closed boutonnière injuries can be treated with static and then dynamic splinting. It is important to emphasize to the athlete the necessity of following the prescribed treatment protocol. Those who cannot comply are treated with pinning of the PIP joint for 3 weeks and gradual protected mobilization with a Capener splint. Return to sports in the nonprofessional athlete at 6 weeks is possible with buddy taping of the finger. In the elite athlete, many factors need to be weighed to arrive at the best treatment for an individual patient.

Treatment Options Acute open lacerations of the central slip are repaired openly, and the repair is protected with a dynamic extension splint, such as the Capener or Bunnell splint, for 6 weeks. Although operative repair of closed boutonnière deformity has been reported, most can be treated with PIP extension splinting, leaving the DIP joint free (Fig. 20B1-15).56,57 DIP flexion is encouraged because it pulls the lateral bands distally and helps the central slip to heal to the middle phalanx.58,59 Extension splinting requires

A

Pseudo Boutonnière Deformity A pseudo boutonnière represents a volar plate injury by a hyperextension mechanism, with resultant scarring and contracture of the volar plate. This postures the PIP joint in flexion, mimicking a boutonnière position.61 The

B

Figure 20B1-16   A, Dorsal repair of central slip and proximal interphalangeal (PIP) pinning for boutonnière deformity. B, Central slip repair with anchor and PIP pinning.

1390 DeLee & Drez’s Orthopaedic Sports Medicine

­ yperextension mechanism of injury, however, and subh sequent examination can readily distinguish a true boutonnière from a pseudo boutonnière. In the latter, a chip avulsion fracture fragment may often be seen on the volar lip of the middle phalanx where the injury occurred.62 In addition, with a pseudo boutonnière, the tenderness is on the palmar side of the PIP, and the DIP motion is almost always normal. In the true boutonnière deformity, active and passive flexion of the DIP joint is impaired because of excessive tension on the terminal tendon due to proximally migrated lateral bands. Treatment of pseudo boutonnière is directed at progressive stretching of the PIP volar plate with dynamic splints.

Boxer’s Knuckle (Sagittal Band Rupture with Extensor Tendon Subluxation) Although any boxer’s hand can be subjected to a multitude of injuries, boxer’s knuckle specifically refers to rupture of the sagittal band, over the dorsum of the MCP joint. The extensor tendons are maintained over the dorsal apex of the MCP joints by a dorsal sling of transverse fibers, the sagittal band.35 The sagittal band acts as a tether to prevent radial or ulnar subluxation of the extensor tendon at the level of the MCP joint. The sagittal band invests the extensor tendon, crossing both volar and dorsal to it.63 The sagittal band inserts into the palmar plate of the MCP joint. It also acts to extend the proximal phalanx, by lifting P-1, when the extensor retracts proximally. Radial sagittal band rupture can occur from forceful finger extension, a direct blow, or an ulnarly directed force.64 Pain and snapping may occur at the long finger MCP joint, and a careful examination will document the tendon subluxation or dislocation. The radial sagittal band of the long finger usually ruptures, although ulnar sagittal band rupture has been reported.65 Clinically, the patient presents with an extensor lag at the involved digit and often an ulnarly deviated MCP joint. Frequently, there is snapping of the extensor tendon over the metacarpal head with flexion and extension of the MCP joint.

Treatment Options Acute sagittal band ruptures can be treated in an extension splint, called the sagittal band bridge, or a static extension splint.66 Splinting continues for 4 to 5 weeks, until there is no evidence of extensor subluxation when making a fist. In chronic cases, Wheeldon described a surgical technique using the ulnar junctura, which is flipped over and sutured to the radial side.67 Postrepair immobilization is brief (2 to 3 weeks), and the MCP joint is flexed 60 degrees.

Author’s Preferred Method Return to Sports

and

For injuries diagnosed acutely, and with minimal tendon subluxation, closed splinting is effective.68 Chronic cases and those with large dislocation of the extensor tendon are repaired surgically through a curvilinear dorsal incision, usually with reinforcement using a junctura tendinum.

I prefer to use a nonabsorbable suture such as 4-0 nylon or 3-0 Supramid for the repair, with the knots buried. In chronic or neglected cases, or when mechanical symptoms of tendon dislocation are prominent, surgical repair is my preferred treatment, with centralization of the tendon at the MCP joint and repair of the rent in the radial sagittal band. Frequently, I use a junctura tendinum to reinforce the repair. Contact activities are avoided until a pain-free arc of motion is possible and the tendon is stable at all MCP joint positions (12 to 16 weeks).

Jersey Finger Jersey finger is an avulsion of the flexor digitorum profundus (FDP) from its insertion on the distal phalanx. This injury classically occurs when an athlete grabs an opposing player’s jersey with flexed fingers. Usually, this injury occurs in the ring finger, and various theories have been proposed to explain why the ring finger and not the long finger is most commonly affected.69 Anatomic reasons for ring finger involvement include the following: 1. Strength of the FDP insertion of the ring finger is significantly less than that of the adjacent digits.70 2. During grip, the distal segment of the ring finger pro­ jects farther and becomes more prominent owing to the increased mobility (palmar flexion) at the ring finger CMC joint.71

Classification Leddy and Packer described three types of injury, based on the following72: 1. Presence or absence of a bony fragment on radio­graphy 2. The level to which the tendon retracted 3. The status of the blood supply of the avulsed tendon The tendon can avulse with or without a bony fragment. The level of the bony fragment on lateral radiograph does not reliably predict the level of tendon retraction because the tendon end and bone fragment may separate.73 In type I, the tendon has retracted into the palm, without a bony fragment. This injury needs early repair because retraction of the tendon into the palm and loss of blood supply lead to contraction. Type II avulsions often have a small bone fragment and retract only to the level of the PIP joint, and the vinculum longum may still be intact. This can be successfully repaired weeks after the injury because tendon length has been maintained. Type III injuries have a large bony fragment and are held distally by the A-4 pulley (Fig. 20B1-17A) at the DIP joint. These can be successfully repaired weeks after the injury. Because it is not possible to know the exact position of the avulsed tendon end, it is best to repair all these injuries acutely.60 After Leddy and Packer’s original three part description in 1977, a fourth type was added to the classification scheme.74 This fourth type involves an intra-articular fracture of the distal phalanx and separation of the tendon and fracture fragment.

Wrist and Hand 1391

A

B

Figure 20B1-17   A, Profundus flexor avulsion with fracture of distal phalanx (type III). B, Fracture fixation and pinning of distal interphalangeal subluxation.

Physical Examination and Testing On physical examination, it is important to palpate the flexor sheath. There may be a prominence where the tendon end is located and frequently tenderness at the A-1 pulley area if the tendon has retracted into the palm. Loss of isolated DIP joint flexion is the sine qua non of this injury and must be tested for by immobilizing the PIP joint in an extended position while profundus function is checked. Ultrasonography and magnetic resonance imaging (MRI) are not necessary to diagnose this entity.

Treatment Options Surgical repair should be carried out as soon as possible after the injury. The distal FDP insertion site is exposed with a Bruner incision. If the tendon has migrated proximally into the palm, the pulley system and fibro-osseous canal must be preserved. The tendon can be passed distally with a No. 8 pediatric feeding tube as a leader, or by milking the tendon distally with smooth forceps from the palm just proximal to the level of the A-1 pulley. Distal insertion and repair are done with a pullout suture or wire, although suture anchor fixation with mini anchors has been advocated recently.75

Author’s Preferred Method, Postoperative Care, and Return to Sports I prefer to do the distal repair with a pullout suture, using 3-0 Supramid. The suture is passed through the avulsed distal tendon end with a Kessler stitch and tied over the

nail plate after being passed around the narrow waist of the distal phalanx. Advancement of the profundus tendon should be limited to 1 cm or less, to avoid flexion contractures. Before tying the suture, the periosteum of the distal phalanx is curetted. If additional fracture fragments are present, they are fixed with fragment miniscrews (see Fig. 20B1-17B) before tendon repair. After repair, a place-andhold flexor tendon protocol is initiated under the supervision of an occupational therapist for the first 4 weeks. The pullout suture is removed at 6 weeks, and graduated active flexion is encouraged. Resistive grasp and return to sports are delayed until 10 to 12 weeks after repair.

Neglected Flexor Digitorum Profundus Tendon Avulsion Numerous reasons exist why players may not seek immediate treatment for an FDP avulsion injury. They may think it is a minor injury that can be treated after the season is over, or they may not want to miss playing for a large part of their season. It is imperative to communicate to the player the long-term consequences of neglecting this injury, and documentation of your conversations with the player is always a good thing. Treating a neglected FDP avulsion is never as satisfying as doing a primary repair, and the clinical results are not nearly as functional. The treatment options for a neglected avulsion are as follows: Late reinsertion: This is unlikely to be possible if the tendon has retracted a significant amount and the time from injury is long.

1392 DeLee & Drez’s Orthopaedic Sports Medicine

Acute single-stage reconstruction with a tendon graft: This option is fraught with potential complications, including loss of additional range of motion, scarring, and worsening of FDP function. Two-stage flexor tendon reconstruction: This approach requires a large amount of effort by the patient, surgeon, and therapist and will probably be considered too aggressive a course to regain some DIP joint flexion. Stabilization of the DIP joint by fusion or tenodesis: This is often the most acceptable choice for the patient when risks and benefits are carefully considered. Nonsurgical treatment: Some players will choose this option when weighing the pros and cons of reconstruction. If a retracted tendon is a tender mass in the palm, excision of the scarred tendon end may relieve symptoms.

Author’s Preferred Method Early repair is the best option, and my preferred technique is listed above. In the situation of a neglected avulsion, fusion of the DIP joint is the option that I explain is most reliable, if it is important for the athlete to flex his finger tip pulp to distal palmar crease. If the athlete has concerns about any surgical reconstruction, then foregoing surgery and allowing the patient to move has been the best course in my practice.

Treatment Options Immobilization, rest, and anti-inflammatory medication are used early. Frequently, return of complete range of motion is delayed. In the nonoperative treatment of acute pulley rupture, a pulley ring, fashioned by an occupational therapist, will relieve stress on the pulleys. Buddy strapping and gentle range of motion can begin, as edema decreases. Be mindful of preventing PIP joint contractures because their development can complicate the postinjury course. If there is rupture of the entire pulley system, loss of digital flexion will be obvious, and bowstringing of the tendons is evident. This situation requires surgical reconstruction of the A-2 and A-4 pulleys. The principles of reconstruction are to maintain the flexor tendons near the PIP and DIP centers of rotation. Reconstructed pulleys should be of stable material and sufficiently strong to allow early mobilization. Tendon graft (palmaris or plantaris) or dorsal wrist retinaculum is used to reconstruct the pulleys. At the A-2 pulley level, the graft is placed under the extensor and around the flexors and sutured to itself. At the A-4 level, the graft is placed over the extensor apparatus and around the flexors, and is sutured to itself. Postoperatively, tendon gliding protocols are used, with ring splints helpful in minimizing the stresses to the newly reconstructed pulleys.

Author’s Preferred Method Treatment

of

Disruption of the Flexor Pulley System The fibro-osseous canal and system of pulleys in the digits were elegantly described by Doyle and Blythe in 1977.76 Biomechanically, the A-2 (at the “proximal aspect of the proximal” phalanx) and A-4 (at the “middle of the middle” phalanx) pulleys are critical for normal flexor tendon ­function.77,78

Clinical Presentation and History Attenuation or frank rupture of a digital pulley can occur in rock climbers and pitchers and during a fall from a height. This occurs as a result of acute or chronic exposure to forcible contraction of the FDP tendon against a greater than physiologic load. In free rock climbers, the flexed DIP joint repetitively supports body weight. The rock climbing hand posture of crimping puts a large strain on the distal part of the A-2 pulley and can result in pulley rupture.79,80

Physical Examination and Testing On clinical examination, the patient will complain of pain over the flexor sheath. There may be swelling over the pulley and loss of DIP motion. The DIP can still be actively flexed, with discomfort. The athlete may perceive weakness, and pitchers will lose velocity on their fast balls. There is a spectrum of injury from partial injury of a pulley to complete rupture of multiple pulleys with overt bowstringing of the flexor tendons. Rarely, both the A-2 and A-4 pulleys will fail. MRI can aid in confirming the diagnosis and may show edema in the flexor sheath and palmar translation of the flexor tendons when the scan is taken with applied resistance.81

Closed conservative treatment is best unless complete traumatic rupture of the mechanically critical A-2 and A-4 pulleys has occurred. Pulley reconstruction requires a great deal of work on the patient’s part with a knowledgeable therapist and realistic expectations about ultimate ­function.

C

r i t i c a l

o n i n

T T

P

e n d o n h e

F

o i n t s

I

n j u r i e s

i n g e r s

l Most mallet finger injuries at the DIP joint and most acute boutonnière deformities can be treated with splinting and the supervision of an occupational therapist. l Use the physical examination to differentiate a boutonnière deformity from a pseudo boutonnière because the treatment is markedly different (see earlier). l If a boutonnière deformity with rupture of the central slip has severe stiffness of the DIP joint, central slip repair is recommended. l Sagittal band rupture treated with splinting may take longer than with operative repair, and this may be a factor in the elite athlete. l Primary repair of a jersey finger FDP rupture is dramatically better than delayed repair. l It is important to test for FDP function in any player who presents with DIP pain or stiffness. (Passively extend the PIP joint and observe for active FDP pull through at the DIP joint.)

Wrist and Hand 1393

FRACTURES OF THE METACARPALS AND PHALANGES This section highlights some common athletic injuries involving the metacarpals and phalanges. Although most phalangeal and metacarpal closed fractures can be treated with closed means in the general population, athletes, with their special requirements, often require a more aggressive (operative) approach. Because athletes want to be “better, faster, stronger,” they want their fractures to heal faster and not interfere with their return to play. Although improved surgical technique and implant technology allow aggressive fracture management in elite or professional athletes, it is important to inform the athlete of the treatment options, possible common complications, and time frames for rehabilitation and return to their sport. Certain principles of injury treatment and fracture fixation are universal. Indications for fracture fixation in the athlete include the following: 1. Irreducible and malrotated fractures 2. Displaced intra-articular fractures 3. Subcapital (unstable) fractures 4. Open fractures and the hand with multiple fractures Uncomplicated nondisplaced metacarpal and phalangeal fractures can best be treated with a short (2-week) period of protection in a brace, or splint, followed by an active range of motion program with buddy taping and removable dorsal protective splint. The hand suffers immobilization poorly, and extended immobilization can lead to stiffness, adhesion, and scar formation around the gliding tendon surfaces and shortening of collateral ligaments (especially at the MCP joint). Complex injuries such as PIP fracture-dislocations, CMC fracture-dislocations, and collateral ligament avulsions are covered elsewhere in this chapter. In this section, oblique fractures of the metacarpals and phalanges, unstable P-1 metaphyseal fractures, boxer’s fractures, and fractures at the base of the thumb metacarpal are featured.

A

Metacarpal Fractures Boxer’s Fracture The boxer’s fracture of the metacarpal neck is common, is often neglected in the nonathlete, and can be unstable. The usual mechanism is axial load to the metacarpal head and often but not always involves a clenched fist injury. Knowledge of the anatomy of the metacarpal head and neck is crucial for optimal treatment of these injuries. The metacarpal heads are cam shaped in the sagittal plane, and their collateral ligaments are taut, with the MCP joints in 60 to 70 degrees of flexion. This fact necessitates splinting the MCP joint in 60 degrees of flexion after injury or surgery that takes place around the MCP joint. To do otherwise invites the development of an extension contracture of the MCP joint, which is difficult to treat. MCP extension is a nonfunctional position for grasp and is poorly tolerated, especially in the ring and small rays, the most common sites of metacarpal neck fractures. The lumbrical and interosseous muscles exert a flexion force at the metacarpal level.82-85 These muscle actions, combined with the axial load mechanism of injury, result in an apex dorsal ­angulation deformity for most metacarpal neck and transverse shaft fractures. Anteroposterior, lateral, and oblique radiographs are routinely obtained.

Treatment Options and Author’s Preferred Method Nondisplaced or minimally displaced neck fractures are treated with a dorsal protective splint or cast with the MCP joints flexed 60 degrees and the wrist in slight extension. Braakman and coworkers reported that functional taping of fifth metacarpal neck fractures produced good results.86 Rarely is malrotation a problem in metacarpal neck fractures. In displaced metacarpal neck fractures, a reduction should be attempted under hematoma or wrist block. I prefer to reduce the angulation (Figs. 20B1-18A and 20B-19) with direct pressure, and then use the proximal phalanx as a plunger, with the MCP joint flexed, to lock the reduction.

B

Figure 20B1-18   A, Small finger metacarpal neck angulation. B, Reduction and dorsal splinting.

1394 DeLee & Drez’s Orthopaedic Sports Medicine

A

C

B

Figure 20B1-19   A, Angulated metacarpal neck fractures of the long and ring fingers. B and C, Reduced metacarpal neck fractures of the long and ring fingers.

A dorsal functional brace or cast is applied (see Fig. 20B118B), allowing digital motion, and is removed at 3 weeks. A recent study has shown functional casting with immediate range of motion to produce results superior to complete immobilization.87 For unstable metacarpal neck fractures with large apex dorsal angulation, reduction and pinning to an adjacent metacarpal works well. A great deal has been written recently about intramedullary pinning for metacarpal shaft and neck fractures.88 In a well-designed study, intramedullary��������������������������������������� pinning and transverse pinning worked equally well for unstable neck fractures.89 It is my preference to use transverse pinning when necessary and to place the pins below the skin for removal in the office at 3 weeks. Composite digital motion must be emphasized in the early postoperative period. Intra-articular extension of a neck fracture is treated with ORIF, using miniscrews and K-wires as needed.

Oblique and Transverse Metacarpal Shaft Fractures Closed nondisplaced or minimally displaced shaft fractures are treated with functional bracing. Always check the hand for malrotation clinically because sometimes radiographs can be misleading. Any malrotation or significant angulation is usually treated operatively in the high-demand athlete. Long oblique metacarpal fractures can be well fixed with multiple lag screws (Fig. 20B1-20). In transverse shaft

fractures, 2- or 2.4-mm plates in larger individuals work well. Respect for soft tissues, especially extensor tendons, is important during the dorsal approach to metacarpal fixation.90 Miniscrew (1.5 or 2 mm) lag fixation is my preferred fixation for displaced oblique fractures. Transverse fractures may be plated (Fig. 20B1-21). Reduction should be anatomic, and rotation of the digit is checked before fixation. The reader is referred to a classic text on the use of Association for the Study of Internal Fixation (ASIF) techniques for further reading.91 Crush injuries with multiple metacarpal fractures can be complicated by intrinsic muscle contracture and “intrinsic tightness.”92,93

Phalangeal Fractures Open fractures at the phalanges are much more common than open injuries at the metacarpal level. Associated injuries to neurovascular bundles and tendons, as well as the condition of the soft tissue envelope, have a great deal of influence on the patient’s ultimate recovery after phalangeal fractures.90,94 In one study, the worst functional results were seen in phalangeal fractures associated with tendon injuries.95 This information can help the treating physician counsel the athlete with an open phalangeal fracture, or one associated with a tendon laceration. Malrotation of P-1 fractures is common and usually requires ORIF. Comminuted fractures are effectively treated with percutaneous pinning (Fig. 20B1-22).

Wrist and Hand 1395

Figure 20B1-20   Lag screw fixation of the metacarpals.

Anatomy The extensor apparatus is closely applied to the dorsal and lateral surfaces of P-1 and P-2. Interference with this gliding structure or shortening of the dorsal apparatus can cause loss of motion at the IP joints. Shortening of proximal phalangeal fractures relatively lengthens the dorsal apparatus and causes an extension lag at the PIP joint.96 The PIP joint is spherical in the sagittal plane, and collateral ligaments on each side of the PIP joint do not shorten with PIP motion. The safe position of immobilization for the PIP joint is in full extension. The volar plate is sensitive to injury, immobilization, and post-trauma edema. Contracture of the volar plate leads to PIP joint stiffness and a flexion contracture. Because soft tissue injuries frequently accompany phalangeal fractures, an accurate assessment on physical examination is important. Injured structures are repaired

at the time of fixation, and the postoperative regimen adjusted accordingly. Always check rotation of the digits by looking for parallel nail plates in extension and flexion. Anteroposterior, lateral, and oblique radiographs are obtained. The treatment of spiral oblique phalangeal fractures and transverse, unstable P-1 metaphyseal fractures is discussed.

A

B Figure 20B1-21   Plating of transverse metacarpophalangeal fracture.

Figure 20B1-22   A, Comminuted P-2 fracture. B, Crossed pinning of P-2 fracture with the proximal interphalangeal joint free.

1396 DeLee & Drez’s Orthopaedic Sports Medicine

B

A

Figure 20B1-23   A, Oblique P-1 fracture. B, Open reduction and internal fixation of P-1 fracture with lag screws.

Oblique Phalangeal Fractures Because torsional injuries are common in sports, oblique fractures of P-1 are frequent. Displaced, shortened, and malrotated oblique fractures are all indications for ­operative fixation. ORIF with lag screws is recommended when the fracture length is twice as long as the bone diameter (Fig. 20B1-23). Minimal operative disruption of the extensor apparatus is best. Sometimes one “wing” of the lateral bands needs to be taken down to allow reduction and fixation. Lag screws of 1.3, 1.5 or 2 mm diameter are used, based on fracture length and bone size (see Fig. 20B1-23B). If bone loss or comminution is present, a mini condylar plate interferes with PIP motion least.97,98 Operative fluoroscopy is necessary to document reduction and screw lengths.

A

Transverse Metaphyseal Proximal Phalanx Fractures These fractures are notoriously unstable, especially if significant displacement was seen on the injury films (Fig. 20B1-24A). Comminution and complex injuries may require miniplate (see Fig. 20B1-24B) fixation.98,99 Placing the plate (1.5 or 2 mm) away from the extensor tendon toward the lateral aspect of P-1 is recommended. Angulated fractures (Fig. 20B1-25A and B) can also be treated with percutaneous fixation, after reduction by placing two pins (0.045 inch) across the fracture site from each lateral tubercle of the proximal phalanx. The PIP joint is not violated by the pins (see Fig. 20B1-25C). Pins are removed at 3 weeks, and limited MCP and PIP motion can begin before pin removal.

B

Figure 20B1-24   A, Thumb P-1 angulated fracture. B, Open reduction and internal fixation of P-1 thumb fracture with 1.5-mm ladder plate.

Wrist and Hand 1397

A

B

C Figure 20B1-25   A, Angulated P-1 metaphyseal fractures of the ring and small fingers. B, Apex palmar angulation of the ring and small finger P-1 fractures. C, Reduction and percutaneous pinning.

1398 DeLee & Drez’s Orthopaedic Sports Medicine

Author’s Preferred Method and Return to Sports Lag screw fixation after reduction is my preference for ­ blique fractures. Motion is instituted within 5 days of o ­surgery, with a protective splint worn between therapies. If a stable reduction of a transverse fracture is obtained, I prefer percutaneous K-wires as mentioned previously. Splinting the PIP joint in full extension helps to avoid ­flexion ­ contractures. Edema reduction techniques are ­necessary in all these injuries, and formal occupational therapy is highly recommended for optimal and timely ­recovery. Return to sports in 3 to 5 weeks is possible with protective taping and splinting in contact situations.

C r M e P h

i t i c a l

P

o i n t s

t a c a r p a l a l a n g e a l

o n

a n d

­ F

or chronically lax ligaments at either level lead to weakness, degenerative joint disease, and painful dysfunction.

Carpometacarpal Dislocation or Subluxation Pelligrini has noted that as the thumb has evolved, it has sacrificed stability for mobility at the trapeziometacarpal joint.101 The base of the thumb metacarpal articulates with the trapezium in a biconcave saddle joint. This joint has some inherent instability, which allows circumduction of the thumb and opposition of the thumb with the fingers. Although the saddle shape of the thumb metacarpal joint inherently allows instability, a group of ligaments at this level provides stability. One anatomic study found the dorsoradial and deep anterior oblique ligaments are most important for basilar joint stability.102 Although dorsal dislocations of the CMC joint are uncommon, involvement of this basilar joint complex is common in Rolando and Bennett fracture types.

r a c t u r e s

l Most closed metacarpal and phalangeal fractures can be treated with splinting and early active flexion, with the MCP joints in a position of flexion (60 degrees) in a brace. l Oblique fracture configurations lend themselves to lag screw fixation. Remember that multiple small screws are a stronger construct than fewer (large) screws. l Phalangeal fractures with soft tissue and tendon injuries will have a more complicated postfixation course and less optimal outcome. l PIP joints are prone to flexion contractures, and MCP joints are prone to extension contractures. l Crush injuries with multiple fractures are frequently ­associated with intrinsic muscle contracture.

INJURIES OF THE THUMB Athletes and physicians alike are familiar with the thumb’s crucially important role in hand function and sports activities. Whether a ball sport activity, pole vaulting, stick handling, or checking the reins, the opposable thumb’s function is indispensable in the interaction of the athlete and his or her competitive environment. Essential functions of large-object (cylindrical) grasp, key, and tip pinch are dependent on normal thumb stability, mobility, sensibility, and length. The thumb is frequently injured because of its location out of the plane of the palm and its involvement in the most demanding of tasks. The thumb’s unique osseous structure, with a specialized basilar joint, and its at-risk position makes injury patterns in the thumb unique. The MCP joint is exposed, and collateral ligaments are frequently torn. In this section, special injuries of the thumb are reviewed.

Treatment Options Radiographs are obtained to document the dislocation and rule out fractures. Three views are obtained, and the overpronated view gives the most information regarding the CMC joint. Reconstruction of the basilar joint ligaments acutely has been advocated,103 but my preference is to save ligament reconstruction for those patients who fail reduction and pinning and demonstrate chronic instability. Once a closed reduction is performed, stability is checked under anesthesia. Almost all are unstable, and percutaneous fixation is performed under Fluoroscan guidance. The thumb metacarpal is reduced and held in abduction and metacarpal extension. Pins are buried under the skin and removed in 5 to 6 weeks. A short arm thumb spica cast is worn. In cases of chronic subtle symptomatic instability of the CMC joint, frequently ligament reconstruction is necessary, using a portion of flexor carpi radialis tendon.104-106 These references give an excellent description of the procedure, which is done through a Wagner incision along the glabrous border of the thenar eminence.

Author’s Preferred Method and Return to Sports When a dorsal dislocation occurs, closed reduction and percutaneous fixation is my treatment of choice. For frank dislocations, I prefer acute reduction and percutaneous pin fixation for 6 weeks. If a patient fails with persistent instability, then ligament reconstruction is performed as outlined by Eaton.104 Return to contact sports, or sports in which hand use is intense, is possible at 12 to 16 weeks, with taping or hand-based splinting.

Ligament Injuries of the Thumb

Metacarpophalangeal Dislocation

Thumb stability at the CMC and MCP joints is crucial for opposing the thumb in pinch and grasp. Stout dorsal and palmar ligaments at the CMC level and collateral ligaments at the MCP level maintain normal stability.100 Torn

There is a great deal of variability in the anatomy of the thumb metacarpal head and range of motion of the MCP joint. Some metacarpal heads are round, and others are flat. Range of motion may vary from 30 to 90 degrees of

Wrist and Hand 1399

flexion. The metacarpal head is surrounded on three sides by strong stabilizers. The ulnar and radial collateral ligaments provide lateral support, and the volar plate prevents hyperextension. The most common MCP dislocation is dorsal. A hyperextension force produces the disruption of the volar plate. The collateral ligaments may fail if a torsional force is dissipated at the MCP level. MCP dislocations can be simple (reducible) or complex (irreducible). In a simple dislocation, there is still some contact of the base of P-1 with the metacarpal head (see Fig. 20B1-1). Often a hyperextension posture of the thumb is seen on lateral radiograph in this instance. Frequently in irreducible dislocations of the MCP joint, the sesamoids are seen interposed between the base of P-1 and the metacarpal head. The proximal phalanx is usually parallel to the metacarpal in complex dislocations. Rupture of the volar plate occurs in MCP dislocations, and it usually occurs proximally at the metacarpal neck area. However, reports of volar plate rupture distally, or through the sesamoids, have been made. In the more common simple dorsal dislocation of the thumb MCP joint, the hyperextension force leaves the thumb with apex palmar angulation (see Fig. 20B1-1). After a proper neurovascular examination, radiographs document that the P-1 is in contact with the metacarpal head (see Fig. 20B1-1). Reduction of a simple dislocation can be done under wrist block, augmented with a local hematoma block in the MCP joint area. The key to the reduction maneuver is to avoid traction and push the P-1 over the metacarpal head. A gentle push on the palmar surface of the metacarpal neck may help. Once reduced, MCP dislocations are usually stable. After a closed reduction, I prefer to cast the thumb for 2 to 3 weeks, with 20 degrees of flexion at the MCP joint. Complex dislocations present with the P-1 parallel to the metacarpal. Interposition of sesamoids between the base of P-1 and the metacarpal indicates a complex dislocation. These injuries usually require open reduction, through either a dorsal or radiopalmar approach.107 If an attempt at closed reduction is planned, this should be done in the operating room suite under adequate anesthesia. If this fails, proceeding to an open reduction is easily done. Multiple attempts at a closed reduction of a complex dislocation is discouraged. Flexor pollicis longus and volar plate entrapment are cited as reasons necessitating open reduction.107,108 A dorsal approach to reduction has been found safe and effective.109 A palmar approach may be done with special attention to digital nerves draped over the prominent metacarpal head. After open reduction of a complex dislocation, I immobilize the MCP joint for 2 to 3 weeks, in gentle flexion. Active range of motion is then begun with a resting hand-based splint. Rarely, a volar MCP dislocation of the thumb can occur. This is usually the result of direct dorsal trauma. This results in a torn dorsal capsule and frequently one or both collateral ligaments. In addition, the extensor pollicis brevis insertion may be torn from the base of the proximal phalanx. These injuries frequently require open reduction and stress testing of collateral ligaments, with collateral ligament repair as needed.12,110 MCP fusion for neglected dislocations has been reported.111

Author’s Preferred Method and Return to Sports Closed reduction is effective for most dorsal MCP dislocations. In the complex dislocations, I prefer a dorsal approach to reduce the joint, and remove incarcerated flexor tendon and volar plate. If the volar plate has been ruptured distally and hyperextension instability persists, I use a suture anchor to reattach the volar plate to the base of the proximal phalanx. Timing of return to sports for these injuries depends on treatment (closed versus open) and return of range of motion and strength. Taping and static splinting are recommended on return to contact or hand-intense sports. Elite and professional athletes may return to sports earlier, especially after a closed reduction.

Collateral Ligament Injuries of the Metacarpophalangeal Joint The thumb is exposed out of the plane of the palm and is called on frequently to participate in cylindrical grasp. Radial and ulnar deviation stresses on the MCP joint are resisted primarily by the collateral ligaments and volar plate. Ulnar collateral ligament injuries are more common than radial collateral ligament (RCL) tears. The importance of the ulnar ligament for tip and three-jaw pinch is unquestioned.112 A complete tear of the RCL, although less common, can cause thumb subluxation at the MCP joint with resultant pain, instability, and weakness. Ligament tears are classified as partial or complete. Complete tears should be repaired surgically.113,114 The anatomy of the MCP joint is different on its ulnar and radial sides, owing to the adductor aponeurosis, which passes directly over the UCL.

Ulnar Collateral Ligament Injuries The MCP joint of the thumb is an ellipsoid joint with the elliptical articular surface of the proximal phalanx moving on a convex metacarpal head. The shape of the thumb metacarpal head is different from that of the fingers. The thumb metacarpal is wider dorsally, and in the frontal plane, it is flatter than the finger metacarpals.35 Range of motion in the thumb MCP joint varies owing to variation in the curvature of the metacarpal head. Spherical metacarpal heads are associated with more motion.115 The anatomy of the UCL has been described by Bean and associates and is important in operative fixation of complete tears and when reconstruction of the UCL is performed.116 It has a predictable course from dorsal on the metacarpal head’s ulnar surface to palmar on the base of the proximal phalanx. The accessory collateral ligament is palmar and parallel to the proper collateral ligament and attaches to the palmar plate and the ligament. Most complete UCL tears occur at the distal insertion. The adductor aponeurosis, a thin sheet of tissue, lies superficial to the UCL. This thin expansion inserts onto the extensor tendon. Stener described the interposition of this adductor expansion between the torn distal end of the UCL and its attachment on P-1.117 This precludes any healing of the torn ligament to its insertion and is the indication for operative repair when Stener’s lesion is diagnosed.

1400 DeLee & Drez’s Orthopaedic Sports Medicine

Figure 20B1-26   Positive ulnar collateral ligament stress test. Figure 20B1-27   Ulnar collateral ligament rupture with positive stress radiograph.

Clinical Presentation and Imaging UCL rupture, “skier’s thumb,” and “gamekeeper’s thumb” all imply loss of continuity of the UCL at the MCP joint. The mechanism of injury is radial deviation of P-1 on the metacarpal head due to a radially directed force. On physical examination, it is important to detect a mass, if present, at the metacarpal neck. This implies Stener’s lesion and was the most reliable sign of complete UCL rupture in a study by Abrahamsson and colleagues.118 The radial deviation force on testing should be done with the MCP joint flexed to lessen the volar plate’s contribution to MCP stability. Lack of a firm end point indicates a complete tear (Fig. 20B1-26). If less than 10 degrees difference in side-to-side stability is noted, a partial tear has occurred, and closed treatment of these lesions yields good results.119 However, it is important not to miss a complete rupture and Stener’s lesion. When the stress testing is equivocal, MRI, ultrasound, and arthrography have all been found helpful in confirming a complete rupture.120,121 A stress anteroposterior radiograph (Fig. 20B1-27) may help in the measurement of the side-toside difference and aid in surgical decision making. Usually a complete tear examined acutely will have a floppy feel on stress testing, with no end point. This type of examination, a palpable Stener’s lesion, and more than 15 degrees of sideto-side difference are all indications for surgical repair. If an osteoarticular fragment has been pulled off, any displacement should be treated with surgical fixation of the fragment. If

Author’s Preferred Method

and

a mass is not palpable, stress testing should be performed under local anesthesia. Plain radiographs are obtained before stressing the MCP joint to rule out an avulsion fracture.

Treatment Options After an accurate diagnosis of partial injury or attenuation of the UCL has been made, the athlete can be treated with a splint, which protects against radial deviation. In one study of splinting versus casting, there was no difference in functional outcomes or pinch strength.119 A forearm-based opponens splint is fashioned, and the athlete is allowed out of the splint for gentle flexion and extension at the MCP and IP joints and for washing. If the patient is ­unreliable, an opponens cast is applied for 4 weeks, followed by ­supervised range of motion with a therapist. Use of a short (hand-based) opponens splint can protect the thumb for an additional 4 weeks. Return to sports take place at 12 weeks with the thumb protected. As with most injuries, professionals may choose to return to their sport earlier. Complete tears of the UCL are usually treated surgically, as are ulnar injuries with avulsion fracture fragments.114,122,123 Functional results are good with slight loss of MCP range of motion after operative repair. Pinch strength should equal that on the uninjured side.113 The operative approach causes little postoperative morbidity.

Return

If the athlete is reliable and has a partial injury, I prefer to treat with functional bracing. A long opponens splint is worn for 3 weeks and then changed to a short opponens splint for an additional 3 weeks. For acute complete tears, I prefer to use a suture anchor rather than a pullout tunnel in P-1. I prefer the use of a suture anchor (mini Mitek) to reapply the torn ligament to the P-1 insertion site.124 A small curvilinear incision is made at the MCP joint’s ulnar aspect with the distal limb slightly palmar to the midaxis of P-1. Dorsal sensory nerves are protected, and the adductor aponeurosis is incised. The torn collateral ligament is teased out, and the insertion site is roughened

to

Sports

with a curette. There is no need to make a trough for the ligament. The joint is inspected and irrigated. The joint surface is rarely injured. The suture anchor is drilled and placed just distal to the ligament’s insertion site, aiming away from the articular surface of P-1. Pinning of the MCP joint with a 0.045-inch K-wire with the joint in slight ulnar deviation is ­usually done, especially if some time has elapsed since the injury. Volar subluxation of P-1 on the metacarpal is rare with UCL injuries, but if present, it would be an indication for pinning at the time of repair. The ligament is sutured and tied down. The adductor ­ aponeurosis is sutured, and a subcuticular skin closure is used. If an avulsion fragment

Wrist and Hand 1401

Authors’ Preferred Method

and

Return

of bone is attached to the UCL, it is reduced and pinned with a 0.028-inch K-wire. If the fragment is large enough, a small screw is placed. An additional tension-band wire (26-gauge stainless steel) or suture is passed through the ligament. The K-wire comes out in the office at 7 to 10 days, and the patient is then splinted or casted, depending on the reliability of the athlete. Range of motion exercises at 2 weeks may be done for the motivated athlete, with protective splinting as mentioned previously. I am more likely to cast postoperatively for 2 weeks, if patient reliability is an issue, to protect the repair. Subsequently, active range of motion is emphasized. For the reliable patient, active motion can begin at 7 days, with functional splinting for protection. Pinching activities should be avoided for 4 to 6 weeks, to avoid stressing the repaired ligament. Postoperative rehabilitation emphasizes early range of motion. Pinch activities are delayed for 5 weeks ­ because this increases UCL strain.125 The athlete must be ­protected from reinjury for 4 to 6 months after repair, and sports ­participation will depend on the ability to protect the thumb during play. Chronic UCL laxity is well recognized, and ­ treatment with ligament reconstruction is effective.126 Glickel and ­others have described the technique for UCL reconstruction, usually using a palmaris longus tendon graft.127 The graft is

to

Sports—cont’d

placed through a bony tunnel in the ­proximal ­phalanx (Fig. 20B1-28). An apex proximal triangular ­ configuration has been found to be biomechanically strongest.128 The tendon graft is sutured to the collateral ligament stump proximally, or an anchor in the metacarpal head may aid in fixation of the graft. With a good reconstructive option for chronic UCL laxity at the MCP joint, fusion for instability is not indicated, unless degenerative joint changes are present.

Figure 20B1-28   Reconstruction with palmaris longus tendon graft.

Radial Collateral Ligament Injury

Extensor Tendon Injury: Mallet Thumb

RCL injuries are less frequent than UCL tears, so a high index of suspicion is necessary in the athlete with radialsided MCP joint pain and tenderness. There is no radial counterpart to the adductor expansion overlying the UCL. The diagnostic examination and criteria for instability are similar to those used for the UCL tears. Greater that 20 to 30 degrees difference in side-to-side comparison stress testing indicates a complete (type III) rupture. Radiographs are mandatory, and special attention is paid to the alignment of the MCP joint on the lateral as well as the anteroposterior film. Any palmar subluxation indicates a complete rupture, and surgical repair is indicated in this situation. Partial ruptures can be treated with casting or functional splinting until healed. Most hand surgeons would surgically repair an acute complete tear of the RCL.129 Some surgeons have reported good results with RCL soft tissue sleeve advancement only in chronic RCL instability.130 Others prefer a reconstructive procedure with a tendon graft using palmaris longus, in the chronically unstable RCL injury.129

A tear of the extensor pollicis longus (EPL) insertion into the distal phalanx is an uncommon injury and is much less common than the garden-variety mallet finger injury. The thumb distal phalanx is longer, thicker, and wider than the distal phalanges of the fingers.35 The terminal insertion of the EPL tendon into the dorsal base of the thumb’s distal phalanx is wider and thicker than the terminal tendon insertion in the fingers. The IP joint of the thumb has variability in its arc of motion. Full extension may range from 0 to 60 degrees of hyperextension.

Author’s Preferred Method I prefer to treat partial RCL injuries with casting or functional splinting, based on the patient’s reliability. For grade III ruptures, acute surgical repair is preferred, using a suture anchor. If the ligament is of poor quality, reinforcement with the abductor pollicis brevis tendon is performed. In the chronic situation, a reconstruction with palmaris longus tendon through proximal phalangeal drill holes is my preference.

Clinical Presentation and History If the athlete is not seen soon after the injury, it is important to determine the date of injury. As in mallet finger injuries, extension of the terminal phalanx is markedly less than the uninjured side. In jamming or contact injuries, it is important to rule out fractures, IP joint collateral ligament injury, and proximal injuries. Tenderness at the dorsal prominence of the distal phalanx is present and helps rule out a proximal EPL lesion. Radiographs are obtained.

Treatment Options Because this is an uncommon injury, the literature is mixed on treatment recommendations. Most published reports are small case series. Reports of closed treatment, with the IP joint splinted in extension for 8 weeks, have documented good results.131-134 Operative treatment has also been recommended.132,135

1402 DeLee & Drez’s Orthopaedic Sports Medicine

Figure 20B1-29   Bennett’s fracture-subluxation of thumb carpometacarpal joint. Figure 20B1-30   Reduction and pinning of Bennett’s fracture.

Author’s Preferred Method

Clinical Presentation and History

If the athlete’s uninjured thumb extends to neutral or 10 degrees of hyperextension, then I prefer conservative care with the IP joint splinted for 6 to 8 weeks. If a great deal of hyperextension is present in the normal thumb, operative treatment is considered. Some athletes may opt for operative repair to decrease the amount of immobilization time. Fixation of terminal phalanx avulsion fractures are fraught with all the possible complications seen with ­operative treatment of bony mallet fingers.136 The athlete must remain out of play for at least 6 weeks regardless of the treatment regimen, especially in contact sports. It is impossible to protect the thumb’s IP joint and have a functioning thumb ray simultaneously.

With axial load and some flexion of the thumb metacarpal, a fracture line forms at the base of the thumb (Fig. 20B1-29). The anterior oblique ligament attaches to the palmar-ulnar fracture fragment, the “beak” portion of the thumb metacarpal. The shaft of the thumb displaces in a dorsal radial direction. The articular surface is usually subluxated (see Fig. 20B1-29). Bennett’s description of this entity in pathologic specimens led to more than 25 proposed ­treatments.139 Patients present with pain, swelling, and subluxation of the metacarpal base. Radiographic evaluation is important, and a true lateral view of the base of the thumb is obtained by pronating the hand 30 degrees to the x-ray cassette.140

Treatment Options

Fractures of the Thumb Bennett’s Fracture Bennett’s fracture is an intra-articular fracture-subluxation of the base of the thumb metacarpal. The trapeziometacarpal articulation fits two saddle-shaped surfaces together and allows 6 degrees of freedom (flexion-extension, ­adduction-abduction, and pronation-supination) for metacarpal motion. The biomechanics of this unique articular arrangement allows motion in three planes.137 This joint, with reciprocal saddle surfaces, is stabilized by the anterior oblique, dorsal radial, and ulnar ligaments.138,139

Author’s Preferred Method

and

Once the diagnosis is made, treatment is directed to reducing the metacarpal shaft to the Bennett’s fragment and making sure the articular surface is anatomically reduced. Multiple studies have shown that accurate articular reduction leads to the best functional results.138,141,142 In most acute cases, a closed reduction can result in anatomic alignment of the joint surface. Percutaneous pinning can then be performed under fluoroscopic guidance (Fig. 20B1-30). In delayed cases, an open reduction may be necessary if articular alignment is not normal. This is done through a Wagner approach on the radial side of the CMC joint, reflecting the thenar muscles.143 Pinning or screw fixation of the volar fragment is dictated by the size of the fragment.

Return

I prefer to do a closed reduction on all but the very delayed Bennett’s fractures. The reduction maneuver is important. It involves traction, radial abduction, and then opposition ­(pronation) of the thumb metacarpal. The pronation at the end locks the fragment together. One 0.045-inch pin is driven from the metacarpal shaft into the trapezium (Fig. 20B1-31). If radiographs confirm an anatomic reduction, the fragment may also be pinned. Pins are placed under the skin and left for 5 weeks. Open reduction is reserved for those in

to

Sports

whom an anatomic reduction of the joint surface is not possible. A dental pick can help with fragment reduction. Provisional pinning is followed by screw fixation. After closed or open reduction and fixation, a thumb spica splint is worn for 7 to 10 days. This is replaced with a long opponens splint that is worn for 4 to 6 weeks. Return to play is dictated by ­position (a lineman can return sooner than a quarterback). If secure screw fixation is obtained in the elite athlete, return can ­occur 4 weeks after fixation.

Wrist and Hand 1403

Author’s Preferred Method

and

Return

A

to

Sports—cont’d

B

Figure 20B1-31   A, Bennett’s fracture with volar fragment (arrow). B, Carpometacarpal joint pinned with capture of volar fragment.

Rolando Fractures The Rolando eponym describes a fracture of the base of the thumb with three or more fragments. Frequently, the articular surface is split in the dorsal palmar plane of the metacarpal. Occasionally, there can be severe comminution associated with this fracture pattern. Treatment principles are similar to those of Bennett’s fracture. Restoration of articular surface geometry and anatomic shaft alignment are the treatment goals. If fragments at the base of the thumb are of a reasonable size, ORIF through a dorsal and radial approach can produce good results.144,145 If fragments are small and comminuted, external fixation is an option. Even closed reduction and pinning has been championed for these injuries.146

Author’s Preferred Method and Return to Sports I prefer ORIF for the Rolando variant, using the Synthes modular hand set plates. Provisional fixation with pins can be helpful, and lag screw fixation of the articular fragments at the base of the thumb is recommended. The postoperative regimen is similar to the Bennett’s variety. If fixation is secure, early active motion can be started. Again, return to play should be delayed 4 to 6 weeks.

S U G G E S T E D

R E A D I N G S

Bowers WH: Sprains and joint injuries in the hand. Hand Clin 2:93-98, 1986. Eaton RG, Malerich MM: Volar plate arthroplasty of the proximal interphalangeal joint: A review of ten years’ experience. J Hand Surg [Am] 5:260-268, 1980. Freiberg A, Pollard BA, Macdonald MR, Duncan MJ: Management of proximal interphalangeal joint injuries. Hand Clin 22:235-242, 2006. Glickel SZ: Thumb metacarpophalangeal joint ulnar collateral ligament reconstruction using a tendon graft. Tech Hand Up Extrem Surg 6:133-139, 2002. Green DP: In Strickland JW (eds): The Hand. Philadelphia, WB Saunders, 1994, pp 945-1017. Hamilton SC, Stern PJ, Fassler PR, Kiefhaber TR: Mini-screw fixation for the treatment of proximal interphalangeal joint dorsal fracture-dislocations. J Hand Surg [Am] 31:1349-1354, 2006. Kalainov DM, Hoepfner PE, Hartigan BJ, et al: Nonsurgical treatment of closed mallet finger fractures. J Hand Surg 30:580-586, 2005. Kozin SH, Thoder JJ, Lieberman G: Operative treatment of metacarpal and phalangeal shaft fractures. J Am Acad Orthop Surg [Am] 8:111-121, 2000. Sarris I, Goitz RJ, Sotereanos DG: Dynamic traction and minimal internal fixation for thumb and digital pilon fractures. J Hand Surg [Am] 29:39-43, 2004. Zemel NP: Metacarpophalangeal joint injuries in fingers. Hand Clin 8:745-754, 1992.

R eferences Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 1404 DeLee & Drez’s� O

S E C T ION B

Hand 2. The Pediatric Hand Thomas Shepler Several characteristics differentiate children’s hands from adults’ hands. The presence of growth plates, ossification centers, and a tendency to fracture without as much compressive force compared with adult bone are key examples.1 Depending on age, children may be less verbally capable than adults. They contain more subcutaneous fat as a percentage of body mass. Children may also exhibit congenital abnormalities that may require attention, yet do not have the degenerative problems associated with aging that plague adults. Orthopaedic problems in children are commonly related to trauma, either directly or indirectly.

ANATOMY The hand in the child is more than just a miniaturized version of the adult hand. In the child, the collateral ligaments insert predominantly into the epiphysis at the metacarpophalangeal (MCP) joint. At the proximal interphalangeal (PIP) and distal interphalangeal (DIP) joints, the collaterals insert more broadly, including areas of the middle and distal phalanges, distal to the physeal line. The nonspherical shape of the metacarpal head causes the collateral ligaments to tighten progressively as the joint is flexed (Fig. 20B2-1).2 The central slip of the extensor tendon is that portion of the extensor tendon that continues from the MCP joint to its insertion on the dorsal epiphysis of the middle phalanx. The distal extensor mechanism, or terminal tendon, is a confluence of both lateral bands, which are motored by the

interosseus and lumbrical muscles.3 The volar plate exists as a restraining mechanism to hyperextension at the MCP, PIP, and DIP joints. At the MCP joint, the volar plate is more lax, allowing more hyperextension than at the PIP and DIP joints, where the volar plates are strong, thick, and white. The sublimis tendon attaches over a significant longitudinal distance in two lateral grooves in the middle phalanx. The flexor profundus tendon attaches more discreetly on the proximal aspect of the distal phalanx.

EVALUATION Clinical Presentation and History The most important clinical question to ask the child with a hand injury is, “What brings you to me today?” This is followed by one instruction: “Point to the area with one fingertip that bothers you the most.” Much of the history may come from a parent in the very young child, and asking about the child’s past medical history, as well as ­immediate history of trauma, is important.

Physical Examination and Testing After the history, an initial physical examination of the child’s hand is key. The examination can be more thorough after an initial set of radiographs. The examination

Proximal origin is eccentric to axis of rotation (• = center) Collateral ligament is longer in flexion Further lengthened by condylar flare when in flexion

Figure 20B2-1   Anatomy of metacarpophalangeal joint collateral ligament.

Wrist and Hand 1405

Figure 20B2-2   Measurement of a person with a protractor.

should include range of motion of the affected parts combined with range of motion of the comparative parts on the opposite extremity. Examining range of motion personally is recommended (Fig. 20B2-2). By ranging the joints of the digits and thumb, additional physical findings may surface that can help in making an accurate diagnosis.

Imaging Radiographs should be obtained with most hand injuries because bony pathology may accompany soft tissue trauma. The basic initial radiographic views should be 90 degrees, or orthogonal, from one another. Another helpful hint is to obtain anteroposterior and lateral views with the beam of the radiograph centered on the specific bone area involved. There are many special views that can add important information that radiology technicians and radiologists may not know. Parents may need to assist in the radiographic positioning for very small children.

A

FRACTURES OF THE HAND Many fractures of the hand in children can be treated closed with casting or splinting without surgery.4-6 Bone often fractures through a physis (Salter-Harris I to III fractures).7 Even the injuries involving the physis are amenable to manipulation and casting.4-6 When one sees a “chip” radiographically, it is imperative to understand that many times the small fragment of bone may represent a fracture that travels through a cartilaginous growth plate and emerges in the metaphyseal region of the bone, as in Salter-Harris II fractures. Alternatively, some anatomic structure is often attached to that chip, such as tendon or ligament, representing an avulsion with a more significant soft tissue injury. In children, fractures occurring in the metaphyseal region of a bone in the hand heal rapidly. Three weeks of casting when fractures occur in the metaphyseal region of the bone is often sufficient in children. This is helpful knowledge because some bones may not look ­radiographically healed

B

Figure 20B2-3   Mallet avulsion fractures. A, Mallet avulsion fracture in a thumb. This occurs from a quick downward (volarward) force on the tip of a digit. Rarely does a major arthritic process result, even though this is an intra-articular fracture. B, Hyperextension produces a reasonable, although not perfect, reduction.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1406 DeLee & Drez’s� O

and the subungual bed.8 Fractures from crush often do not often involve the DIP joint; however, Salter-Harris II fractures occur frequently.4,8,9 If the terminal tendon is involved, a mallet avulsion fracture or a pseudomallet with fracture through the physis may result mimicking the ­classic mallet (Fig. 20B2-3).9 Rupture of the flexor digitorum profundus, or “jersey finger,” is rare in the child (Fig. 20B2-4).10,11 With associated pediatric nail bed injuries, efforts should be made to reattach an avulsed nail plate using absorbable suture because it (1) is biologically compatible with the patient, (2) provides excellent coverage over raw disrupted nail bed, and (3) often offers excellent reduction and ­stabilization of an underlying fracture.

Treatment Options

Figure 20B2-4   Jersey finger injury.

at 3 weeks but clinically are nontender and can be allowed to move. Periosteal new bone does not form as prolifically as one might imagine in many digital fractures, and the radiograph typically lags behind the clinical healing.

Phalangeal Fractures Distal Phalanx Clinical Presentation and History A common injury, especially in the small child, involves a crush of the tip of the digit. This may be open or closed and often involves not only the tuft of the distal phalanx but also some disruption of the soft tissue, the nail plate,

Most pediatric fingertip injuries are treated nonoperatively, as noted earlier, requiring only cleansing, splinting, and bandaging. Occasionally, operative intervention is indicated for open injuries of the distal phalanx. The two most common circumstances in which open reduction with internal fixation (ORIF) may be required are (1) an unstable near amputation injury of the fingertip that contains a small fracture fragment within the avulsed soft ­tissue ­element and (2) an acutely flexed Salter-Harris II or III fracture of the distal phalanx (P-3) that may protrude from the nail fold and have the nail bed interposed in the fracture site, or lie superficial to the nail fold.12 In this type of open physeal injury, wherein the distal portion of P-3 has been acutely flexed with concomitant disruption of the base of the nail plate, its accompanying subungual bed can emerge from underneath the eponychial edge of skin and lie superficial to the nail fold, preventing reduction.13

Author’s Preferred Method In unstable near-fingertip amputations, my preference is to stabilize the soft tissue avulsion by using a technique of placing a K-wire in a retrograde manner through the small element of bone (distal tip of P-3) contained in the avulsed

tip (Fig. 20B2-5). The avulsed tip is then reduced over the proximal bone, and the K-wire is driven back through the proximal end of the distal phalanx, through the DIP joint, and into the shaft of the middle phalanx. The distal end of Figure 20B2-5   K-wire stabilization of P-3 fracture with soft tissue injury. A, Fracture of the nail plate with an underlying P-3 fracture can be treated by stabilization with a K-wire. B, Comminuted P-3 fracture stabilized by K-wire awaiting soft tissue coverage. C, Flap.

B

X-finger flap

A

C

Wrist and Hand 1407

Author’s Preferred Method—cont’d the K-wire is bent over on itself, cut off, and left extracutaneous for 2 to 3 weeks until removal. The soft tissues are sutured when possible. This works well with a fractured nail plate and underlying P-3 fracture. The plate does not require removal, nor does the nail bed requiring suturing. Sterilizable, battery-powered K-wire drivers are available for this purpose (Fig. 20B2-6).

Figure 20B2-6   Sterilizable battery-powered K-wire driver.

Mallet Avulsion Fracture (Bony Mallet Injury) An avulsion of the distal extensor mechanism from its insertion to the proximal portion (epiphysis) of the dorsal distal phalanx is called a mallet injury. In young children, a pure tendinous avulsion is extremely rare. Usually some sort of Salter-Harris I or II fracture of the distal phalanx (P-3) occurs instead, resulting in the so-called pseudomallet.9 If

A

the bone is reduced, the results are excellent. Immobilization for 3 weeks is adequate in the child (Fig. 20B2-7). Occasionally in adolescents, an avulsion of the extensor mechanism with a piece of epiphyseal bone (Salter-­Harris III) develops. An extensor lag may follow bony healing. Cosmetically, this may be frustrating to the patient but rarely is of functional concern.

B

Figure 20B2-7   Mallet avulsion fractures. A, Acute mallet fracture. B, Healed mallet fracture.

Author’s Preferred Method The largest studies consistently show that closed treatment provides satisfactory results.8,9,14 The rare exception requiring surgical treatment occurs when the remaining distal phalanx is dislocated volarly at the DIP joint (Fig. 20B2-8). Otherwise, these are splinted with the DIP joint in hyperextension for 3 to 4 weeks if some reasonable fragment of bone is still attached to the distal extensor mechanism.

­ adiographic union is not universal, but a persistent fracture R line is of no consequence. As Wehbe and Schneider point out in their classic work, many have a resultant harmless dorsal “bump” and minimal extensor lag, but an excellent functional result.14 Sporting activity can be resumed after splint removal, although warning the patient and the parents of potential extensor lag is important. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1408 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d Figure 20B2-8   A-D, Mallet avulsion fractures with volar dislocation.

A

B

C

D

Middle Phalanx In children, the middle phalanx (P-2) is injured the least of the three phalanges. Most fractures can be treated closed.5,7,8 The most common injury in the middle phalanx is the volar plate avulsion fracture, and the avulsion component is usually small and represents the volar aspect of the epiphysis of the middle phalanx. A bony avulsion of the central slip of the extensor tendon slip, like the collateral ligament avulsion fracture, involves the articular surface of the distal side of the PIP joint. The central tendon slip attaches to the epiphysis of P-2. To avoid a boutonnière deformity, this fracture pattern must be recognized and splinted in extension. Rarely, pinning of the joint may be needed. Operative intervention may be needed in phalangeal neck fractures that are rotated, intra-articular condylar

fractures, oblique unstable shaft fractures, or transverse midshaft fractures that are totally displaced. The most common portion of the bone that may require an open reduction or preferably percutaneous pinning occurs distally at the neck of P-2. In this P-2 neck fracture, distal condyles (if intact) typically rotate dorsally (collapse apex volarly) (Fig. 20B2-9). This injury can cause loss of DIP joint flexion. The volar recess is lost, and flexion is diminished. In the young child, remodeling often corrects this with time.6,15 In the adolescent, permanent loss of full flexion at the DIP joint may result. Displaced condylar fractures of P-2 are rare but should be reduced accurately and stabilized. If the condyles are split longitudinally, the collateral ligaments pull on the fragments, causing them to open and splay proximally from one another. Figure 20B2-9   A, Distal neck of condyle fracture rotated 90 degrees.   B, Anteroposterior radiograph of same injury looks innocuous. C, Same injury in adult. D, The injury may be amenable with reduction to K-wire fixation.

A

B

C

D

Wrist and Hand 1409

Author’s Preferred Method medullary canal. If image intensification is used, after one side is pinned openly, the opposite side may be approached percutaneously and pinned successfully. If unsuccessful, the opposite side can be opened in a similar midaxial way. Pins are buried, and the wound is closed. In the adolescent with a long oblique shaft fracture of P-2, the fracture is usually unstable, and I prefer to percutaneously pin these with two or three K-wires perpendicular to the fracture line to prevent foreshortening and malrotation. Intraoperative imaging is mandatory. A helpful hint is to start the pinning in metaphyseal bone (usually distally) heading toward diaphyseal bone. If one starts in the midshaft (diaphysis), the pin may just skid off the hard cortical bone instead of engaging it. During casting, some active and passive motion of the distal joint is permitted. I begin range of motion exercises 3 to 4 weeks after surgery (Fig. 20B2-10). Transverse, displaced, midshaft, middle phalangeal fractures may be difficult to percutaneously pin. However, every attempt to do this closed is made before open reduction is attempted. Typical K-wire fixation is the easiest form of fixation to use. Sometimes the operative technique of temporarily inserting a longitudinal K-wire from the distal tip of P-3 through the DIP joint into the shaft of the middle phalanx to “shish kebab” the fracture will keep it stable enough to allow the insertion of several cross K-wires. Otherwise, direct visualization through a midaxial incision may be necessary. For fractures involving the insertion of the central slip, the fracture involves the articular surface, and accurate reduction optimizes the result, avoiding late boutonnière ­deformity.5,16 If I am unable to obtain an anatomic alignment through closed PIP joint extension and splinting,

Phalangeal Fractures of P-2

For phalangeal neck fractures, closed reduction and percutaneous pinning accomplished in an extra-articular fashion is my preference if the distal fracture fragments are large enough. The transverse neck fracture is manipulated into position and held in a reduced position while a small temporary 0.028-inch K-wire is inserted longitudinally from the tip of the distal phalanx (P-3), through the DIP, and into the proximal shaft of the middle phalanx. If the reduction is acceptable on a portable C-arm, I proceed with retrograde crossed K-wire fixation of the distal fragment to the proximal shaft fragment. If both crossed K-wires have good purchase in bone, the longitudinal K-wire is withdrawn. If a satisfactory closed reduction is unobtainable, I also prefer an open technique. My preference is to make a midaxial incision over the DIP joint. The lateral band is released along its volar edge and is retracted dorsally. The proximal side of the P-2 fracture is exposed, and its dorsal periosteum is reflected dorsally so that the transverse fracture line is seen. The DIP joint is passively flexed to envision where the joint line is located. Then the distal portion of P-2 is reduced to the proximal portion of P-2 by rotation volarly, and small crossed K-wires are inserted obliquely from the distal (condylar) end of P-2 in a proximal direction and through the proximal side of the fracture. It is easier to start the K-wire in soft metaphyseal bone where the proximal portion of the collateral ligament attaches because the collateral ligament prevents drift of the K-wire as the oblique drilling process begins. The proximal direction of the drill also engages and drills the proximal (shaft) side of the fracture, especially when it approaches the hard cortical shaft from within its

A

B

Preop

5.1 wk

C

5.1 wk

D

Figure 20B2-10   A-D, Middle phalangeal oblique shaft fracture.

Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1410 DeLee & Drez’s� O

Author’s Preferred Method—cont’d

A

Preop

1 wk postop

B

1 wks postop

C

Volar Plate Avulsion Fx

D

Figure 20B2-11   Boutonnière avulsion fracture with dislocation: preoperative (A) and postoperative (B) radiographs. Volar plate avulsion fracture with dislocation: preoperative (C) and postoperative (D) radiographs.

I prefer closed reduction and percutaneous pinning, and if needed, open reduction with pinning. If the fracture fragment is large, closed reduction and pinning can be accomplished by one or two vertical K-wires that may be splayed in such a manner to grab a lateral cortex of P-2 and not spear the volar flexor tendons. If the avulsion fracture fragment is small, I prefer closed pinning of the PIP joint in full extension for 3 to 5 weeks. At 5 weeks, the K-wire is removed, and the digit is immobilized again for an additional week in full PIP joint extension (Fig. 20B2-11).

Postoperatively, at the end of 1 week, I place the hand in a boxer’s cast (Fig. 20B2-12) until 3 weeks. The patient is allowed active and passive range of motion of the DIP joint. This will minimize adhesions of the lateral bands and keep the profundus tendon from becoming adherent. It is important to check the stability of the rest of the PIP joint. It is imperative to know that most of the middle phalanx is located and not dislocated volarly so that the boutonnière avulsion fracture fragment can be reduced against the intact articular surface of the distal end of the proximal phalanx. An image intensifier is critical to this procedure

B

A

C

Figure 20B2-12   Boutonnière avulsion fractures: boxer’s cast (A) with distal joint motion (B and C).

Wrist and Hand 1411

Proximal Phalanx The most common fracture in the proximal phalanx is a Salter-Harris II fracture, involving the proximal physis.5 The distal portion of the collateral ligaments, unlike those in the middle and distal phalanges, attaches only to the epiphysis of the proximal phalanx, making disruption through the physis more likely to occur. Extra-articular proximal one-third fractures of the proximal phalanx are amenable to cast or splint treatment.4,5 The treatment of the proximal phalanx virtually mimics that of the middle phalanx with regard to fracture patterns and treatment options as described earlier for fractures of P-2.

Metacarpal Fractures Metacarpal fractures can be divided into two groups: the first metacarpal of the thumb and the other four (index, middle, ring, and little) metacarpals. Intra-articular displaced fractures are often opened, but the shaft fractures can consistently be treated by closed means.

Thumb Metacarpal Fractures Classification Thumb metacarpal base fractures are classified as extraarticular or intra-articular. The intra-articular fractures are further subdivided into noncomminuted Bennett’s fractures and comminuted Rolando’s fractures.17 As in all

childhood fractures, fracture through the physis, which is proximal in the thumb metacarpal, is common. In children, most of the fractures of the thumb metacarpal are extra-articular.18 However, an intra-articular fracture should be reduced to optimize functional outcomes.19

Treatment Options Extra-articular Fractures The thumb ray tends to collapse into an apex dorsal angulated fracture. Usually, these are amenable to closed treatment.17,20 Although it is possible to manipulate this fracture into a more normal position, it is hard to maintain because the thumb ray tends to settle back into its original position. The consequence of this is some loss of extension of the thumb ray at the metacarpal-trapezial joint, but function is rarely compromised.

Bennett’s Fracture Some authors prefer to manipulate Bennett’s fracture into a good position and percutaneously pin it through the dorsal base of the thumb metacarpal into the trapezium in a position of palmar abduction.17,21 Others prefer to open this in such a way that one can see the articular surface of the base of the metacarpal from its radial side and yet have access to the dorsal aspect of the metacarpal to install a screw or apply some small T-plate once the articular surface is reduced.21,22

Author’s Preferred Method Extra-articular Fracture

Closed treatment usually produces a reasonable result in ­extra-articular fractures of the thumb metacarpal. Under ­local anesthesia, a thumb spica cast or splint can be applied after the apex dorsally angulated thumb ray undergoes a closed reduction. A combination of a longitudinal pull and

volarly directed pressure on the dorsal apex of the fracture at the base of the thumb metacarpal reduces this fracture. Some of the reduction may be lost as swelling diminishes. The thumb is positioned into palmar abduction opposite the index finger ray.

B

A

C

Figure 20B2-13   A-C, Bennett’s fracture is an intra-articular fracture of the base of the first metacarpal that will displace and must be fixed. Otherwise, metacarpal-trapezial joint osteoarthritis may result. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1412 DeLee & Drez’s� O

Author’s Preferred Method—cont’d Bennett’s Fracture

My preference is open reduction through a dorsal radial approach, being wary of the dorsal radial branch of the superficial radial nerve. The opponens and the dorsal edge of the abductor pollicis brevis are released from the base of the thumb metacarpal. The volar portion of the abductor pollicis longus is released to see the displaced articular surface from the radial aspect of the thumb metacarpal. Next, an incision is made dorsally between the abductor pollicis longus and the extensor pollicis brevis tendon. The periosteum and the extensor pollicis brevis tendon are

reflected ulnarly, allowing access to the dorsal aspect of the metacarpal for plate or screw placement. A small diameter (0.028-inch) K-wire used to transiently fix the shaft to the displaced volar piece of the base of this metacarpal while the screw or plate is put into place is helpful. This is casted or splinted for 3 to 4 weeks after surgery before motion is begun (Fig. 20B2-13). Rolando’s Fracture

Because this fracture is rare in the child, the reader is referred to the adult treatment section for Rolando’s ­fracture.

The Other Four (Index, Middle, Ring, and Small) Metacarpals

Distal Metacarpal Extra-articular Neck (Boxer’s) Fracture

The index through small metacarpals are connected to one another by the deep transverse intermetacarpal ligaments, which provide support to one another and can prevent some degree of foreshortening when one is fractured. They also form a dorsally convex arch, which can make them difficult to pin to one another through a lateral approach (Fig. 20B2-14). The motion of the index ­metacarpal-­trapezoidal joint and the middle ­ metacarpal­capitate (carpometacarpal [CMC]) joint is negligible, whereas the motion between the base of the ring and small fingers with the bifaceted hamate joint is considerable. With flexion of the ring and little fingers, the metacarpal head descends volarly because of motion of the CMC joints. All these anatomic points are important when assessing children’s metacarpal ­fractures.

The pattern of skeletal injury is different in fractures that occur in the distal portions of the index and middle metacarpals from those of the ring and small metacarpals. In the ring and small metacarpals, the fracture often is extraarticular and occurs at the “neck” of the metacarpal. With fracture of the index and middle metacarpals, it is often intra-articular and comminuted. The articular surface is often “mushroomed” over the distal shaft of the bone in fractures of the second and third (Fig. 20B2-15) metacarpals. Historically, teaching dictated that metacarpal neck fractures should not have too much apex dorsal angulation because the metacarpal head may become too prominent and cause trouble in the palm. However, many have seen angulations of 45, 60, and 75 degrees or more without an apparent deficit,23,24 possibly because the motion available

Figure 20B2-14   Percutaneous pinning of adjacent metacarpals.

Arc of metacarpals

 

Wrist and Hand 1413

Figure 20B2-15   Mushroomed second metacarpal head fracture. The mushroomed head is splayed out over the second metacarpal. This fracture results from a straight punch with axial load along the ends of the second or third metacarpal.

at the proximal CMC (metacarpal-hamate) joint allows for dorsal adjustment of a distal metacarpal head that might be too prominent because of an apex dorsal angulation that can form after such a fracture. It is difficult to know how much angulation of the distal metacarpal neck fracture is acceptable (Fig. 20B2-16).

Author’s Preferred Method Boxer’s Fracture

If seen early, casting alone for 3 to 4 weeks (depending on whether the fracture is predominantly metaphyseal or diaphyseal) yields very good results. I prefer to immobilize the MCP joint in some flexion to allow elongation of the collateral ligaments. My preferred cast allows for (1) DIP joint passive and active flexion and extension to prevent tenodesis of the flexor tendons and (2) full passive range of motion of all three joints of the adjacent digits.

A

Extra-articular Metacarpal Shaft Fracture As a general rule, it is rarely necessary to surgically treat a metacarpal shaft fracture.25,26 The deep transverse intermetacarpal ligaments link the volar plates of the index through small finger metacarpals with each other, helping to hold the fractured metacarpal out to length. In my series of 313 consecutive single metacarpal fractures treated closed, the most foreshortening was 8 mm. This is quite acceptable and does not create any major disability. The one exception to the dictum of closed treatment for metacarpal shaft fractures might be when all four metacarpals are fractured because there is nothing remaining to hold them out to length. Closed treatment works very well with long oblique or spiral fractures of the shaft of the metacarpal as well.25

Author’s Preferred Method A short arm cast is used. The affected digit is placed in the intrinsic positive position (i.e., flexion of the MCP joint and extension of the two distal joints). We have termed this the boxer’s cast (Fig. 20B2-17). Malrotation can ­easily

B

Figure 20B2-16   Severe boxer’s fracture. A, Anteroposterior view of an apex dorsally angulated fifth metacarpal fracture. B, Oblique view reveals the degree of angulation. Despite the severity of angulation, palmar prominence of the fifth metacarpal head is not symptomatic.

Figure 20B2-17   Boxer’s cast. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1414 DeLee & Drez’s� O

Author’s Preferred Method—cont’d be controlled in a cast. It is often taught that the adjacent digit should be put into a cast to assist in finding the correct rotation. We have found this unnecessary. A clinical tip is to allow active DIP joint flexion while in the cast, which ­prevents the flexor tendons from becoming adherent during the immobilization period. In closed treatment, the closed shaft fractures can consistently be moved at 4 weeks out of the cast or splint. If all four metacarpals are fractured, the border metacarpals (second and fifth) can be surgically stabilized with pins or plates. This holds the adjacent middle and ring metacarpal fractures sufficiently out to length so that they can be treated closed. Fortunately, fractures of all four metacarpals, especially in children, are extremely rare.

Intra-articular Fracture Base of Second through Fifth Metacarpals This is rare in the pediatric and early teenager populations, and the reader is directed to the adult hand injury section for further discussion of these fractures.

LIGAMENTOUS INJURIES OF THE HAND In pediatrics, pure ligamentous injuries occur mostly at the PIP joint (i.e., volar plate avulsion) of the fingers and perhaps the MCP joint of the thumb.1,27 Often the ligaments are stronger than the bone, and a physeal fracture occurs. The most common ligament injuries occur at the PIP joint and may simply be called a sprain. Avulsion fractures are common with ligament injuries as well, and radiographs should be obtained.

Sprain of the Proximal Interphalangeal Joint Clinical Presentation and History It is often hard to ascertain the exact mechanism of injury or to know if there was a frank dislocation and spontaneous reduction of the PIP joint from the child’s history. After a hyperextension injury to the PIP joint, the child may initially cry but then may resume physical activity until the next day, when the area of injury may become more swollen and be associated with significant pain and more loss of active and passive motion. The amount of swelling is typically commensurate with the amount of injury. Ecchymosis may appear along the volar surface of the PIP flexion crease, and the PIP joint will be diffusely tender.

Physical Examination and Testing Initial examination should include testing for intact flexor and extensor tendons. Both collateral ligaments and the volar plate should be tested with passive stress maneuvers.

In small children, only the collateral ligaments may be testable. In the cooperative patient, the power of an extensor or flexor helps to determine whether it is intact. It is helpful at the PIP joint to determine whether the dorsal surface is more or less tender than the volar surface. An avulsion of the central tendon slip may be difficult to determine. If PIP joint active extension is strong, the central slip is likely intact. An avulsion of the central slip (boutonnière injury) is easy to miss in the child.

Treatment Options If there is more tenderness on the volar surface than on the extensor surface (indicating a volar plate injury) of the PIP joint, and the collateral ligaments are stable, one can either cast the PIP joint in slight flexion for 3 weeks or start immediate motion.1,23 It is always safe to put someone in a cast—then all parties (coach, players, and parents) understand the circumstances. If stability is reasonable, buddy taping the injured digit to an adjacent finger is a reasonable option. Local splints on the actual digit do not last or do much in the young patient.

Author’s Preferred Method Most often in this population, the digit is immobilized by a cast for 3 weeks in slight PIP joint flexion to allow for some healing of the volar plate, and then motion is begun. Immobilization is mandatory if there is a history of an ulnar dislocation at the PIP joint in the small finger combined with a finding of radial collateral ligament instability. Especially in the small finger, instability resulting from laxity of the radial collateral ligament can lead to chronic repetitive dislocations and eventually secondary degenerative arthritis. The digit is guarded for 6 weeks after injury.

Sprain of the Metacarpophalangeal Joint The most common injury may occur on the ulnar side of the MCP joint of the thumb or in the radial collateral ligament of the MCP joint of the little finger.7 Usually, a Salter-Harris fracture occurs instead of a collateral ligament avulsion.28 Surgery is rarely indicated in the pediatric population.1 This is fortunate because repair of an MCP joint collateral ligament is fraught with difficulties in the index, middle, ring, and little fingers. If the repair is too tight, substantial MCP joint flexion is lost, and if too loose, repair is of little benefit.

Anatomy and Biomechanics The proximal attachment of the MCP joint collateral ligament is dorsal and proximal to the center (axis) of rotation. Therefore, the collateral ligament is somewhat lax in extension and tight in flexion (see Fig. 20B2-1). There is also a condylar flare in the final 40 degrees of MCP joint flexion that further stretches the collateral ligament. This is especially relevant at the MCP joint and accounts for the side-to-side MCP joint motion that exists in extension and

Wrist and Hand 1415

the minimal passive side-to-side motion in the final phases of flexion.

Gamekeeper’s Thumb (Tear of the Ulnar Collateral Ligament) Clinical Presentation and History Many times, it is difficult to understand how the injury occurred. One just knows that the patient fell or was involved in some collision in sports. Specific stress testing of the ulnar collateral ligament should be carried out with any hint of clinical suspicion. A tear of the ulnar ­collateral ligament is often accompanied by a tear of the volar plate.28

Physical Examination and Testing On physical examination, one should ascertain whether there is more tenderness over the proximal or distal attachment of this ulnar collateral ligament. Passive hyperextension of the MCP joint is almost always painful. The volar plate is tender. Be sure to test the laxity of both the ulnar and radial collateral ligaments and compare this finding to the opposite MCP joint. Complete tears of the ulnar collateral ligament can be associated with volar migration of the proximal phalanx relative to the metacarpal head. Test the MCP joint by stressing it radially and ulnarly in extension and in flexion to get a complete picture of this ligament. Sometimes, it can be difficult to tell how significantly this ligament is torn because many people have some natural laxity of this joint. Hence, comparison with the opposite thumb is critically important. Also test for passive dorsal and volar translation of the proximal phalanx at this joint. Always include testing for the presence of flexor pollicis longus and extensor pollicis longus function.

collateral ligament is torn and significantly displaced (as indicated by displacement of a small fleck of bone at the distal end of the ligament), or (2) obvious severe laxity of this joint when stressed. Stener’s lesion is an avulsion of the distal end of the ulnar collateral ligament that flips out from under the adductor aponeurosis.29,31 The interposition of this aponeurosis then prevents the collateral ligament from healing.

Author’s Preferred Method Generally, closed treatment in a cast for 6 weeks is sufficient in children. The cast application is not easy but is critical to a good outcome. The essence of the cast is to use a thumb spica to deviate the MCP joint toward the side of collateral ligament injury. If this injury involved the ulnar collateral ligament, ulnar deviation is important without placing the thumb in a position of adduction. It may be necessary to change the cast if the area is swollen and one anticipates reduction of the swelling so that the thumb is held tightly during the course of treatment. It should also be noted that for several months after such an injury and casting, passive hyperextension of the MCP joint may be painful. This will eventually resolve and should not be confused as an indicator of poor healing. Some motion, especially in flexion, may be lost as a result of this injury and immobilization. Generally, function is excellent.

LIGAMENTOUS INJURIES OF THE HAND: DISLOCATIONS Dislocation of the Metacarpophalangeal Joint

Radiographic Examination

Clinical Presentation and History

Anteroposterior and lateral radiographs of the thumb should be obtained. If the MCP joint actively will not come into complete extension, train the central beam on the MCP joint with the proximal phalanx parallel to the plate so that a small flexion contracture will not make the joint space look falsely narrow. Classically, a stress anteroposterior radiograph of the MCP joint can be done while the examining physician passively deviates the MCP joint radially and ulnarly. Lead-lined glove use is recommended, although they may seem bulky or clumsy to use. It can be tricky to hold and deviate a thumb, especially in a child. Also look for small avulsed “flecks” of bone as an indicator of the end of the collateral ligament and note, if present, how far the end is from the insertion or origin of the ligament.29 Clinical examination alone is often sufficient to ascertain laxity. Occasionally, ligamentous laxity is confused with motion through the fracture plane of a SalterHarris I fracture.

Many times after a fall and hyperextension of the MCP joint, any dislocation may well reduce back into its original position. This can occur in the thumb or the digits. The thumb and the index finger appear to be the most commonly involved. If it does not spontaneously reduce, the patient typically presents to an emergency department.25 The MCP joint motion will be minimal. If an accompanying volar laceration is present, typically transverse in the palm over the MCP joint, an open MCP joint dislocation must be considered. These skin edges should be débrided before closure. It is easy for a practitioner to assume that this is a simple laceration instead of realizing that this wound goes down to an open MCP joint, which should be irrigated before skin closure.

Treatment Options Pediatric gamekeeper’s injuries can be treated by open or closed methods.28,30 There are circumstances in which open treatment would be preferred: (1) the distal side of the ulnar

Radiographic Testing Plain radiographs will reveal the dislocation. If only anteroposterior and lateral radiographs of the hand are obtained, the MCP joint will have a narrowed (or overlapped) appearance because the joint is dislocated. A 45-degree oblique view will reveal the dislocation clearly, which will be dorsal. If the thumb is involved, ­ anteroposterior and lateral radiographs are sufficient to make the diagnosis.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1416 DeLee & Drez’s� O

Treatment Options Closed treatment should always be attempted if the injury is closed.25 The first person, often a coach, trainer, or emergency department personnel, has the best chance of accomplishing reduction. Generally, a surgeon becomes involved when someone in the emergency department is unable to reduce the joint. The volar plate may prevent reduction. With hyperextension of the MCP joint, the volar plate tears proximally and can become lodged between the proximal phalanx and the metacarpal head. With adequate anesthesia, and with the wrist in palmar flexion (to relax the flexor tendons), the injury is re-created by further hyperextension of the MCP joint. Then, with the MCP joint hyperextended, a thumb is used to apply pressure at the dorsal base of the proximal phalanx to push it distally, up and over the metacarpal head (Fig. 20B2-18). Care must be taken to avoid fracture of the metacarpal head. A postreduction radiograph is mandatory. If, after several attempts, reduction is not obtained, open reduction with release of the volar plate is indicated.

Open Reduction of Metacarpophalangeal Joint Dislocation There are two methods of approach: volar and dorsal. The volar approach at first would appear more direct. The metacarpal head is buttonholed through various structures. The flexor tendons with one common digital nerve will be on one side of the metacarpal head and the other common digital nerve on the opposite side. The danger is laceration of a common digital nerve during the volar approach. The most vulnerable is the radial neural element because it lies on the superficial surface of the lumbrical. The offending agent (i.e., the volar plate) will not be visible because it will be behind (dorsal to) the metacarpal head. Sometimes, it is possible to grab the proximal edge of the volar plate with a small skin hook and bring it distally over (anterior to) the metacarpal head, which would then allow relocation of the proximal phalanx. Therefore, in a blind manner, one edge (radial or ulnar) of the volar plate must be found and released longitudinally with a pair of scissors to bring it distally over the buttonholed metacarpal head.

A

Author’s Preferred Method If the dislocation was accompanied by a volar wound, I prefer a volar approach. Otherwise, the safest method is to go dorsally because the approach will bring one directly down to the volar plate, which can be split longitudinally, thus allowing reduction and removal of any bony debris within the MCP joint. Surgical Method: Dorsal Approach to the Metacarpophalangeal Joint

A curvilinear approach is made over the MCP joint. One can either (1) split longitudinally the extensor hood, thus preserving its stability by leaving the sagittal bands intact or (2) release the ulnar sagittal band. ­Because of the tendency of the extensor hood to migrate ulnarly during ordinary MCP joint flexion, the radial sagittal fibers should be preserved. Once the ulnar sagittal bands are released, the extensor hood is pulled radially. The dorsal capsule may be intact. Incise this longitudinally. The volar plate will be visible, draped dorsally over the distal portion of the metacarpal head. I use a knife to longitudinally release the volar plate. The digital nerves are safe and out of the way. As the release (from proximal to distal) of the volar plate is done, the distal end of the metacarpal will become visible. Be cautious not to damage the articular cartilage of the metacarpal head with the knife during the release. The dorsal wound (to include the dorsal capsule and extensor hood) is closed, and the digit is placed in the intrinsic positive position in a compression splint. ­Motion is begun in 3 to 4 days, and the splint is removed at 1 week. For the MCP joint of the thumb, the approach is done dorsally by splitting the extensor pollicis longus tendon and the dorsal capsule of the MCP joint longitudinally. Motion is begun 1 week after surgery. Some loss of MCP joint thumb flexion, if it occurs, is better tolerated than in the other digits because the function of the thumb is to be a pinching post.

B

Figure 20B2-18   A and B, Maneuver of relocation of metacarpophalangeal (MCP) joint dorsal dislocation. This is often irreducible because of the interposition of the volar plate. The maneuver: (1) the MCP joint is hyperextended, (2) the proximal portion of the proximal phalanx is pushed distally with a thumb, (3) the digit is passively flexed, and (4) reduction is radiographically confirmed.

Wrist and Hand 1417

Proximal Phalangeal Joint Dislocation Relevant Anatomy and Biomechanics The PIP joint is stabilized by the volar plate, which is whitish in color and quite stout. In the child, it is attached distally to the epiphysis. Additionally, there are other stabilizing elements to the PIP joint: both collateral ligaments, the dorsal joint capsule, and the central tendon slip. The central tendon slip attaches distally to the dorsal epiphysis of the middle phalanx. The central tendon slip and both lateral bands look like one continuous tendinous shroud over the PIP joint.

miss the diagnosis and undertreat this like a simple sprain (Fig. 20B2-20).

Physical Examination and Testing Palpate to see whether the PIP joint is stable and reduced. Test both collateral ligaments. In children, instability of a joint can be confused with a Salter-Harris type fracture involving the epiphysis. It is important to test both of the flexor tendons.

Radiographic Imaging

Injury to the PIP joint is common; in fact, the most common injury in a digit is an avulsion of the volar plate with a small fleck of bone.7,27 PIP dislocations are classified by the relative position of the middle phalanx to the proximal phalanx, typically dorsal, and rarely volar. Associated fracture is also important to recognize along with the direction of dislocation.

Anteroposterior and lateral radiographs with good orientation are needed of the involved digit. Oblique views can be helpful. It is difficult but important to have an excellent lateral view of the PIP and DIP joints. Radiographs tend to be taken early before much physical stressing and examination, which is appropriate to identify associated fracture. However, without physician input, monitoring, or positioning, the quality and usefulness of initial radiographs tends to be poor. It is worthwhile to accompany the patient back to the radiology suite to obtain good radiographs as well as the initial survey of the patient and the initial films.

Clinical Presentation and History

Treatment Options

Volar Dislocation

Dorsal Proximal Interphalangeal Dislocation

If the dislocation is purely volar, the central tendon slip has been avulsed. This injury is rare in children and mostly has occurred in adults (Fig. 20B2-19).

If seen early, reduction of a dorsal dislocation of the PIP joint is relatively easy. Good anesthesia, such as a metacarpal block, is helpful. Typically, the metacarpal block is placed in the palm at the level of the MCP joint. To reduce the joint, the injury is re-created by manually hyperextending the PIP joint. While in a hyperextended position, the examiner’s thumb pushes the dorsal base of the middle phalanx distally. Passive PIP joint flexion will become easy once the middle phalanx is reduced. Postreduction anteroposterior and lateral radiographs are obtained to confirm reduction.

Classification

Dorsal Dislocation The most common direction of PIP joint dislocation is dorsal. This involves disruption of the volar plate. If there is a concomitant deviation of the distal side of the dislocation (e.g., a dislocation of the PIP joint of the little finger) either radially or ulnarly, the volar plate injury will be accompanied by a collateral ligament injury. The most common PIP joint dislocation occurs dorsally and ulnarly and involves the little finger. It may spontaneously reduce, or the injured person, or coach, may pull on it and reduce the joint. This may cause the health care professional to

Figure 20B2-19   Boutonnière avulsion fracture with volar dislocation.

Volar Proximal Interphalangeal Dislocation If the dislocation is volar, relocation is usually simple. With the digit flexed at the area of the PIP joint, simple volar pressure in a distal direction on the flexed volar base

Figure 20B2-20   Dorsal dislocation of the proximal interphalangeal joint.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1418 DeLee & Drez’s� O

Sling

A

B

Figure 20B2-21   Dynamic extension block splint for dorsal middle joint dislocation. A, Close-up of extension block splint. Note the sling of material around the proximal phalanx keeping it up against the splint during active extension. B, The splint does not prevent full active and passive flexion.

of the middle phalanx (P-2) will cause the joint to relocate. At that moment, the PIP joint can be extended. It would be rare for the proximal end of the middle phalanx (P-2) to become lodged in the volar recess of the distal condyle of the proximal phalanx (P-1). If this relocation is difficult, think of the potential for the distal condyle of the proximal phalanx to puncture (buttonhole) through this extensor shroud, creating a traumatic rent between the central tendon slip and the lateral band.

Postreduction Care Options 1. Early motion: one could tape the digit to the adjacent digit and begin active flexion and extension. The fear of early motion is failure of the volar plate to heal, causing a swan-neck (hyperextension) deformity of the PIP joint. 2. Early motion: with a PIP joint extension block splint for 3 weeks (Fig. 20B2-21) 3. Casting in slight PIP joint flexion for 3 weeks, allowing for active and passive DIP joint flexion and extension (see Fig. 20B2-17)

Author’s Preferred Method Acute Dorsal Dislocation

In a young person, I cast the PIP joint in slight (5 degrees) flexion for 3 weeks, allowing active and passive motion of the DIP joint while in the cast. The cast includes only one digit. This approach is especially critical with a dorsal ulnar acute (first time) dislocation of the PIP joint of the little finger. Without casting for 3 weeks, the radial collateral ligament and volar plate may fail to heal sufficiently, allowing recurrent instability. It is important to instruct the patient to put the adjacent digits through a full passive range of motion, especially at the DIP joints.

Acute (Boutonnière) Volar Dislocation The basis of this method is the observation that the amount of foreshortening (retraction) of the central tendon slip from its distal insertion on the base of the middle phalanx (P-2) that still gives a good functional and cosmetic result is extremely small. Making the diagnosis of a boutonnière injury, specifically, avulsion of the central tendon slip, can be difficult in the child. However, almost by definition, a straight volar PIP joint dislocation is accompanied by a complete tear of the central tendon slip. If one judges that the central tendon slip has truly been avulsed by testing the power of active PIP joint extension (under local anesthesia), I place a K-wire across the PIP joint for 5 weeks and cast the digit in a short arm boxer’s cast (Fig. 20B2-22). The pin is pulled at 5 weeks; the boxer’s cast is reapplied and subsequently is removed at 6 weeks. The objective is to allow healing of the central tendon slip to the proximal base of the middle phalanx. Although in the cast, an important point is to allow active and passive motion of the DIP joint to prevent flexor tendon and lateral band adhesions in the zone of injury. The cast or splint should be short enough to allow active and passive flexion and extension of the DIP joint.

Dislocation of the Distal Interphalangeal Joint Relevant Anatomy and Biomechanics The DIP joint is stabilized by the radial and ulnar collateral ligaments; the volar plate, which tends to be a bit lax and volarly by the flexor tendon and dorsally by the distal extensor mechanism. Dorsal dislocations are rare and may be accompanied by a volar open skin wound.32,33 Reduction can be difficult. A volar dislocation is ­ accompanied by a concomitant tear or avulsion of the distal extensor

Wrist and Hand 1419

B

A

C

Figure 20B2-22   A, Classic boutonnière posture of proximal interphalangeal (PIP) joint flexion and distal interphalangeal (DIP) joint hyperextension. B, Pinned PIP joint. C, Boxer’s cast allows active and passive DIP joint flexion and extension.

mechanism (mallet injury) from the dorsal base of the distal phalanx.32,34 It is important with any dislocation or fracture-dislocation that the articular surfaces are congruent after reduction. If the joint is subluxated, final range of motion will be compromised.

transverse rent in the volar skin. The articular surface of the distal end of the middle phalanx may protrude through the skin. Usually the flexor tendon will still be attached and dislocated radially or ulnarly (Fig. 20B2-23).

Clinical Presentation and History

Treatment Options and Author’s Preferred Method

Usually, the cause of a dislocation is a fall on an outstretched hand. If there is no associated fracture, the most common direction for a distal joint dislocation is dorsal. Dorsal DIP joint dislocation is still rare. It is often associated with a

Closed DIP joint dislocation is rare in children, and most are associated Salter-Harris fractures involving the physis of the distal phalanx.9 Closed DIP dislocations may spontaneously reduce. If there is an open dislocation, the joint must

Figure 20B2-23   Open dorsal dislocation. Open dislocation with the condyle of the middle phalanx protruding through skin and lateral radiograph showing dorsal dislocation.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1420 DeLee & Drez’s� O

be irrigated and washed out before closure. The option exists whether or not to débride the wound before closure. The amount of redundant skin is minimal, so débridement and subsequent closure may cause a DIP joint flexion contracture. Care should be taken during the débridement of the transverse wound that the flexor tendon and digital nerves are not injured. It is unnecessary to repair the volar plate avulsion or to pin the distal joint after it has been relocated. Postoperative antibiotics are given, and active and passive motion is begun when the wound is stable (between 3 and 7 days).

TENDON INJURIES Mallet (“Baseball Finger”) Injury Clinical Presentation and History The classic injury of an avulsion with or without a piece of bone of the distal extensor mechanism of a digit occurs when some object hits the end of a digit causing it to unexpectedly flex. The extensor mechanism (the confluence of the terminal ends of the lateral bands) cannot relax quickly enough to avoid injury, and it may avulse from its insertion on the epiphysis of the distal phalanx.

Physical Examination and Testing If the distal extensor mechanism avulses and retracts from the DIP joint, the area over the middle phalanx (P-2) may be swollen, tender, and slightly erythematous. It can mimic an infection. The patient will be unable to actively extend the DIP joint in complete injury. Examine for ­ passive

extension of the DIP joint as well. In chronic injury, passive motion may also be limited.35 Test the power of active extension at the DIP joint. In a true acute mallet injury, active extension is greatly diminished. The amount of extensor lag might be somewhat subtle. Over time, if untreated, this extensor lag can become worse.

Imaging Anteroposterior and lateral radiographs of the digit are obtained. The true extensor lag posture may not be appreciated if the joint has been splinted. Pressure volarly will also reduce the deformity. It may be difficult to get an accurate true lateral of the DIP joint but is helpful in assessing for articular incongruity and subtle fracture.

Treatment Options This is a frustrating injury to treat, and there are many regimens described in the literature.9,14,34-36 Many approaches to this problem end up with a significant lag. What is difficult is to connect the initial extensor lag and the final result. Many times the patient is in a splint when he or she appears for examination. One hates to flex the digit very far for fear of undoing the effect of the initial splinting. When a digit is taken out of a splint and the extension is measured, it may be deceptively good and not representative of the true extensor lag. Many regimens call for splinting in DIP joint extension for 8 weeks. That does not seem sufficient. Some commercially available splints do not hyperextend the DIP joint well. It is better not to immobilize the PIP joint. Tendon avulsions take longer to heal than bony ones.

Author’s Preferred Method After many failures and many regimens, we have resorted to pinning acute (<2½ weeks) mallet injuries with a buried K-wire placed longitudinally down the distal phalanx (P-3) across a hyperextended DIP joint for 8 weeks. The splint is

made from a conforming hyperextended, Alumafoam splint and is placed dorsally (Fig. 20B2-24) from the tip of the finger to the PIP joint. The splint can be removed (as long as the K-wire is in place) during shower activities after the first week.

Swan Neck Posture

A

C

B

D

Figure 20B2-24   Radiographs of mallet avulsion injury (A and B) and photographs of splints (C and D).

Wrist and Hand 1421

Author’s Preferred Method—cont’d

A

B

C

D

Figure 20B2-25   A-D, Tenodermodesis operative technique for failed mallet repair.

At 1 week, no contamination of the pin should occur because the fenestration point in the skin should be healed. Passive motion is allowed at the MCP and PIP joints. At 8 weeks, the K-wire is pulled, and the dorsal splint is placed back over the DIP joint for a total immobilization time of 10 weeks. The splint is discontinued, and active and passive motion in hand therapy is begun. A resulting extensor lag of less than 22 degrees is deemed acceptable. Rarely are the results ­perfect

and often are unpredictable. Women who wear much finger jewelry hate loss of extension; men hate loss of flexion. Chronic Mallet Injury

Tenodermodesis has been helpful in rescuing a chronic mallet.35 There are two major requirements: (1) the patient must have regained full passive range of motion, and (2) this

A

B

C

D

Figure 20B2-26   Tenodermodesis operative technique for failed mallet repair. A and B, Operative photographs. C and D, Splint technique. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1422 DeLee & Drez’s� O

Author’s Preferred Method—cont’d should not be done before 3 months after an avulsion of the distal extensor mechanism. It takes 3 months for a good pseudotendon to be created. The idea is to resect over the DIP joint an elliptical piece of full-thickness skin and underlying pseudotendon. This effectively shortens the ­ extensor mechanism over the DIP joint, where it cannot become ­adherent. To determine the resection, a transverse line is drawn dorsally transversely across the DIP joint from midaxis to midaxis (Fig. 20B2-25). The DIP joint is then passively flexed, and a straight transverse line is drawn across the joint. A symmetric, elliptical line is drawn proximally and distally to this transverse line so that the lines meet just dorsal to midline radially and ulnarly. The tricky part is how to judge

Boutonnière Deformity and Injury Anatomy and Biomechanics The central tendon slip attaches to the epiphysis at the base of the middle phalanx (P-2).37 It extends the PIP joint. However, it is interconnected by a series of oblique fibers to the lateral bands that pass radially and ulnar to this tendon and extend the DIP joint. The central tendon slip and lateral bands do not have to be powerful. They are simply positioning elements. They put the digit into extension in order for them to be able to flex. The two systems are interconnected to each other by a series of crossing fibers so that injury to one has an effect on the other (Fig. 20B2-27). The central tendon slip is simply an extension of the extensor tendon found on the dorsal hand. When the extensor tendon crosses the MCP joint, it is called the central tendon slip. The central tendon slip is the most powerful extender of the PIP joint. If this joint is put passively into full extension, Central Tend on Slip nd

the amount of resection that should be done. If too much is done, the surgeon might not be able to bring the opposing skin edges together, and a too small full-thickness skin resection would not make as good a correction. You are “borrowing from Peter to pay Paul.” This operation gains some extension at the expense of flexion. The DIP joint is passively flexed, and the resection along the elliptical is done. Passive extension of the DIP joint should bring this wound together. A 0.035-inch K-wire is passed retrograde through the distal phalanx and across a hyperextended DIP joint. This is cut off subcutaneously. The postoperative regimen is the same as for an acute mallet. The K-wire is kept in for 8 weeks and is pulled, and the splint remains for a total of 10 weeks after surgery (Fig. 20B2-26).

the lateral bands will lie predominantly slightly dorsal to the axis of PIP joint rotation and will act as a weak extender of that joint. If the central tendon slip avulses from its insertion on the dorsal base of the middle phalanx, the middle phalanx will no longer lie (because of the pull of the flexor tendons) in a fully extended position, and the DIP joint will be reactively pulled into a hyperextended position.36 When the central tendon slip avulses, it retracts because of the natural contractility of a muscle that is no longer at its resting length. Through the remaining cross-fiber interconnections, the retracted central tendon slip now pulls on the lateral bands, causing the hyperextension of the DIP joint as the PIP joint collapses into flexion because of the unopposed pull of the flexor tendons (Fig. 20B2-28).

Clinical Presentation and History Typically, an avulsion of the central tendon slip occurs when an objects hits the end of a digit and causes acute flexion of the PIP joint. If the blow is unexpected, the child may not be able to relax the digit quickly enough,

1

al Ba Later

Central Tend on Slip

3

2 l Band Latera

A

4

Funtional tendinous interconnections between two extensor mechanisms 1 Central tendon slip pulls off bone and retracts 2 Through the connection to lateral band retracts it 3 The lateral band, in turn, extends the DIP joint

B Figure 20B2-27   Digital extensor mechanism. A, The proximal interphalageal joint is extended by the central tendon slip (an extension of the hand’s dorsal extensor tendon). B, The X is a functional representation of the fibrous interconnections between the two systems.

4 With no central tendon connection, P-2 flexes, completing the full boutonnière deformity Figure 20B2-28   Pathoanatomy of boutonnière deformity. The sequence: rupture of the central tendon slip, which then simultaneously pulls on the lateral bands, pulling the distal interphalangeal joint into extension as the middle phalanx without central slip connection collapses into some flexion.

Wrist and Hand 1423

and the central slip will be pulled from its distal insertion with or without a piece of bone attached to it. The joint will be swollen and in a deceptively slight posture of PIP joint flexion. Likewise, the distal joint will be slightly ­hyperextended.

Physical Examination and Testing The PIP joint will be diffusely tender. Occasionally, one can determine that the dorsal aspect of the PIP joint is more tender than the volar surface. This might be the first clue to the injury. Second, the power of PIP joint extension will be diminished. In the younger child, distinguishing pain from actual weakness as the source of loss of extension power is difficult. An injury to the volar plate, as well as to the central tendon slip, often leads to mild to moderate contracture of the middle joint. However, in a volar plate injury, the passive excursion of the distal joint is not limited in flexion, whereas it is moderately limited in a typical boutonnière injury.

Imaging Accurate anteroposterior and lateral radiographs of a digit is obtained. Sometimes the central tendon slip will be avulsed with a small fleck of bone from the epiphysis. The injury is easy to understand when this avulsion fracture is found. Commonly, in the child, the boutonnière injury may be accompanied by a fracture pulled off from the epiphysis. The amount of displacement of the tendon avulsion is easily comprehended if a bit of bone lies at the end of the central tendon slip.

Treatment Options Nonoperative Treatment If acute, the object is simple: to keep the PIP joint in full extension during the healing period (6 weeks). The implementation, however, is difficult. If the PIP joint falls into flexion, the distance between the avulsed end of the central tendon slip and the insertion point on the epiphysis dorsally is more displaced, and the result will be compromised. It is also important that the DIP joint be allowed to flex actively and passively while the PIP joint is kept in extension. This allows for some excursion of both lateral bands by the injured central tendon slip and also allows excursion of the profundus tendon to prevent adhesions. It is almost impossible to keep the PIP joint in extension while the DIP joint is exercised.

Operative Treatment There are several possible approaches: (1) closed treatment and casting or splinting the PIP joint in complete extension for 6 weeks; (2) closed treatment and pinning of the PIP joint in complete extension; (3) open repair and reinsertion of the central tendon slip into the dorsal base of the middle phalanx (P-2); and (4) ORIF of a bony avulsion fracture fragment if it is significantly displaced. This latter approach would be mandatory if the remaining portion of the middle phalanx were volarly dislocated. It is very tempting, if this injury is caused by an object such as a knife, especially over the PIP joint or the insertion point at the base of the middle phalanx, to openly repair this central tendon slip interruption.

Author’s Preferred Method Acute Boutonnière Injury

The typical injury is closed. If the avulsion is acute (<2 weeks), the PIP joint is placed in extension, and a 0.035-inch K-wire is used to cross-pin this middle (PIP) joint. The pin is left in for 5 weeks, but the digit is casted for 6 weeks. The cast is made in such a way to allow active and passive flexion of the DIP joint through this period of immobilization (see Fig. 20B2-22). Chronic Boutonnière Injury

It the injury is more than 2 weeks old, no attempt is made to immobilize this injury. It is critical to restore passive motion before any operative intervention is made. The patient is started on a mobilization regimen to attempt to regain ­passive motion. This includes passive PIP joint extension and passive DIP joint flexion, which are both difficult to restore. ­Often, regaining passive motion is sufficient to make the patient functionally happy without any further reconstructive surgery. If these passive motions are obtained and yet active motion is not reasonable, at 3 months, reconstruction is done. There are many procedures described in the literature. It is important to emphasize that no open operation

should be done before 3 months in order to allow a good pseudotendon to form. What works best is the simple procedure described next. Again, if passive motion is obtained in extension and flexion, a curvilinear incision is made over the dorsal PIP joint. The PIP joint is then pinned in complete extension with an oblique 0.035-inch K-wire. This pin is cut off subcutaneously. Two parallel incisions are made longitudinally on the radial and ulnar portions of the central tendon slip about 1 cm in length (in the adult and adjusted accordingly in the child). The width of the central tendon slip is judged by imagining the width of its insertion on the dorsum of the middle phalanx because there is no obvious interval between the central tendon slip and the two lateral bands. From this central section, a 1-cm resection (in the adult; adjusted for size of the child) is made from the central tendon slip, leaving a stub of distal pseudotendon or tendinous attachment to the base of the middle phalanx. The proximal end of the resection is advanced and sutured with figure-of-eight permanent woven sutures (Ethibond) to the distal stub (Fig. 20B2-29). This repair is basically done over the articular surface of the PIP joint to minimize tenodesis of the central tendon slip. The skin is closed. At 1 week, the compression Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1424 DeLee & Drez’s� O

Author’s Preferred Method—cont’d splint is changed to a cast that allows (1) immobilization of only the affected digit and (2) active and passive motion of the DIP joint. After 5 weeks, the pin transfixing the PIP joint is removed, and the digit is recasted in complete PIP joint extension for an additional week, still allowing for active and passive DIP joint motion. After 6 weeks, the cast is removed, and motion is begun in hand therapy if it is age appropriate. This has worked well, but this is a significant time of immobilization for a digit. It is important to emphasize that throughout this whole process (except for the first week postoperatively), the DIP joint is kept in active and passive motion. Again, a critical element to re-emphasize is gaining excellent passive range of motion of the digit before the surgery.

Figure 20B2-29   Operative boutonnière reconstruction.

Jersey Finger: Avulsion of the Profundus Tendon Anatomy and Biomechanics When flexion is predominantly carried out by the sublimis tendons at the PIP joint (which is what happens when someone grabs quickly to snatch a flag from an opponent’s belt), the ring finger is the longest finger (Fig. 20B2-30). Hence, if all four digits (index, middle, ring and little) are participating in the act of snatching the flag, the ring finger is most likely to be injured.10 The classic injury is not only the ring finger but also the ring finger on the nondominant hand.

Classification Classically, this injury occurs in the nondominant ring finger when a person playing flag football attempts to grab the flag from the opponent’s waist. Instead, the person misses the flag and grabs the opponent unexpectedly, causing an overwhelming force to be transmitted to the tip of the ring finger. The classification system depends on the anatomic element avulsed. The least injury occurs with a fracture of the distal phalanx distal to the insertion of the profundus tendon. This heals quite readily. If the flexor profundus is avulsed from the distal phalanx and only the vincula brevis is torn, the tendon will retract to the PIP joint. If both the vincula brevis and longus are torn, the tendon will retract to the palm. If the tendon avulses with a significant piece of bone, this may not pass through the A-4 flexor pulley. The further the tendon retracts, the more critical is the timing of repair (Fig. 20B2-31).38

Evaluation

A

B

Figure 20B2-30   Prominence in palm of flexor digitorum superficialis on quick flexion. A, Quick grab of a “flag” does not allow full flexion to occur. As a result, the ring finger is the most prominent and, unlike the fully flexed hand (B), bears a large and unequal load when the opponent is grabbed inadvertently.

If there is no bony injury, the injury may easily be missed and thought to simply be a sprain of a digit. The digit, especially at the soft tissue tuft of the distal phalanx, will be very swollen. Significant ecchymosis is often noted over the distal phalanx. Especially in younger children who may not be cooperative, attempting to diagnosis a true avulsion of the tendon can be difficult. Fortunately, this injury mostly occurs in the teens and 20s. One may not be able to tell the difference between pain inhibition and actual anatomic interruption. The digit could be anesthetized and testing done. However, just asking the patient to flex the distal joint of the involved digit once despite pain may help make the diagnosis. You need only feel a flicker of active flexion to tell that the tendon is intact. The best way to test this is to hold the affected digit at the level of the middle

Wrist and Hand 1425 Forme Fruste Fracture

Lateral band

A

B

C

D

FDP

A-4 pulley

Figure 20B2-31   The anatomy of the distal joint shown is concomitant fracture P-3. A, Typical posture of flexor digitorum profundus (FDP) avulsion of the ring finger. B, Forme fruste fracture only distal to tendon insertions. C, Pure avulsion of the profundus tendon without fracture. D, FDP avulsion fracture, displaced and unable to go through the A-4 pulley.

phalanx in some MCP and PIP joint flexion. While holding the middle phalanx with one hand, the patient is asked to flex once the distal joint while it is palpated with the examiner’s other hand. With some experience, the examiner is able to tell whether there is a flicker of active distal phalangeal motion when the tendon is intact. If avulsed, one may also palpate for the end of the tendon to determine whether it lies in the digit or more proximally at the A-1 pulley at the distal flexion crease in the palm. The palm may not only be mildly tender at that point but also a bit prominent. Often, one cannot tell where the distal end of the tendon lies. Magnetic resonance imaging has been used in a difficult situation to show not only the location of the end of the tendon but also where the rupture occurred.39

Imaging The most radiographic information is gleaned from a good lateral radiograph of the digit. Look for a bony fragment, which may indicate where the distal end of the tendon lies. If the fracture fragment is visible, the diagnosis is more easily made.

Treatment If completely torn and retracted, many use 3 weeks as the cutoff for repair.38 Otherwise the tendon not only becomes dysvascular but also shortens (Fig. 20B2-32). The ­exception appears to occur especially with an associated avulsion fracture that is limited in its retraction by the A-4 annular pulley. This may be reinserted as late as 3 months.38 Figure 20B2-32   Diagram of the tendon contracture phenomenon. The surrounding synovium tethers the flexor tendon in the relaxed position, not allowing it to re-expand to regain its length

The surrounding synovium tethers the flexor tendon in the relaxed position not allowing it to re-expand to regain its length

�rthopaedic ����������� S �ports ������ � Medicine ������� 1426 DeLee & Drez’s� O

Author’s Preferred Method If the tendon is retracted into the palm, and it has been less than 3 weeks from the date of injury, it is retrieved and rethreaded through the A-1 and A-2 pulleys. Some attach the tendon transiently by a suture to a small Silastic vascular loop, which has been threaded retrograde from the finger to the palm by way of the flexor sheath. Before threading, one end is incompletely split for about ½ inch with a knife. When this is retrograded back into the palm, a suture that has been placed in the flexor tendon is passed through the slit. The vascular loop is pulled distally, bringing the suture attached to the flexor tendon into the finger. This allows for the tendon to be pulled through the A-1 and A-2 pulleys into the finger. Once it is pulled into the finger distal to the A-2 pulley, there are various methods to bring this through the A-4 pulley. Resection of 50% of its distal end makes it much easier to pass through the flexor tendon. A tendon pullout wire kit (made by Ethicon) consists of (1) two Keith needles attached at opposite ends of a 4-0 monofilament stainless steel wire, (2) a wire loop with a curved cutting needle attached, and (3) a button (Fig. 20B2-33). With one end of the double-ended Keith needles passed through the pull-out wire loop (to allow for later removal), the Keith needles are woven with a Bunnell-type stitch into the distal end of the profundus tendon. With care, the two Keith needles are passed through the remaining A-4 pulley and are used to assist in bringing the flexor tendon through the pulley. With patience, the tendon can usually be brought through this pulley. A dental chisel is then used to raise the soft tissues with perhaps a little bone (periosteal osteal flap) from the volar surface of the distal phalanx, and two drill holes are placed into the distal phalanx using a 0.035-inch K-wire. They should exit over the base of the nail plate. A small 1 × 1 cm cotton pledget several layers thick cut from a 4 × 4 inch sterile gauze is formed and held by a small hemostat. The two Keith needles are threaded through the holes

in the distal phalanx, up through the nail plate, through the small gauze padding, and through the plastic button. The button is oriented so that its flat surface lies toward the nail plate and the curved surface is superficial to allow for the suture to be tied over the button more easily. When the tendon is reinserted into the distal phalanx, the button is used to anchor the wire while the tendon heals (Fig. 20B2-34). A pullout wire, which was previously assembled, is passed proximally and volarly through the skin flap in such an orientation as to be able to directly pull the remaining wire out of the finger (at 4 to 6 weeks) when the button is cut loose from the wire. The wound is closed accurately and carefully so that motion can begin 2 to 3 days after surgery. A circumferential dressing of Vaseline gauze is placed around the affected digit and then covered with a circumferential Kling-type dressing. In preparation for the postoperative rehabilitation, a 4-0 suture is placed through the soft tissue tuft of the digit, taken up through the undersurface of the nail plate, looped over the nail plate, passed back in a volar direction through another portion of the nail plate and through the soft tissues and tied to itself for placement of a rubber band (Fig. 20B2-35). An extension block splint is made in the operating room placing the wrist in 40 degrees of palmar flexion, blocking the MCP joints in 45 degrees of flexion but not impeding full extension of the PIP and DIP joints. A circumferential wrap of four layers of 4 ×15 inch plaster splint is wrapped around the distal forearm at the wrist level to keep the hand from sinking back into the extension block splint, which would then not allow full extension of the PIP and DIP joints. Two serial rubber bands are attached to the suture loop in the nail plate, tensioned, and attached by large safety pin to the more proximal portion of the splint. The patient, therefore, leaves the operating room with the final splint that will be used through the rehabilitation period.

2 doubleended Keith needles 1 button held in cardboard

Wire loop on curved needle Full view of button shown

Figure 20B2-33   Tendon pullout wire kit manufactured by Ethicon.

Wrist and Hand 1427

Author’s Preferred Method—cont’d

A

B

C

D

Figure 20B2-34   A, Preoperative posture is a give-away sign of flexor digitorum profundus (FDP) avulsion. B, Intraoperative photograph of readvancement during repair of the FDP. C, Postoperative photograph with tendon pullout technique. D, Active flexion result.

Day 2 Active Extension

Assisted Extension

Circumferential dressing removed at 2 days

A

Day 2 Passive Flexion

B

Figure 20B2-35   Kleinert extension block splint. A, Extension block splint allowing for full extension of the proximal interphalangeal joint. B, Series to show rehabilitation at 2 days, consisting of active extension against rubber band, with relax of band and passive flexion as wound allows.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1428 DeLee & Drez’s� O

Postoperative Care In the method of Dr. Harold Kleinert, when the wound will tolerate motion, preferably 2 days after surgery, the circumferential dressing is removed, and active extension and passive flexion of the digit is begun. Great emphasis is placed on quickly obtaining full active PIP joint extension in the first week. No passive extension is allowed. At the end of 3 weeks, the rubber band is cut free of the digit, and in the same splint, active flexion is begun. After 4 weeks, the extension block splint is removed, and the digit is guarded until 4 months after surgery while working on motion. The patient is warned that a rupture of the repair would take multiple operations over the next 2 years to make up for this unfortunate event. In 30 years of hand surgery experience, only two flexor tendon ruptures have occurred in my practice. The goal is to have a digit that flexes within 1 inch of the midpalmar crease. This is entirely possible because this is a zone I injury.

FINGERTIP INJURIES Amputations of Fingertip In young children, it is common to have the end of a digit crushed between objects such as a door and door jamb. Often the bony injury is not great, but the nail and nail bed may suffer an injury if the tip is not entirely amputated. This section is divided into two parts: (1) injury to the nail plate and nail bed and (2) complete amputation of some portion of the finger.

Nail Plate and Nail Bed Injury The nail plate, commonly called the fingernail, should be valued. It can act as a great biologic dressing over an injured nail bed. It is not a foreign material, so reinsertion is well accepted. It can act as an external splint to reduce and hold underlying bony and soft tissue injuries. It can separate the dorsal and volar portions of an injured nail fold. It is deleterious to allow a dead space to exist under it that can accumulate serous or bloody fluids, which can then act as a culture medium. Avoid placing foreign bodies

Proximal Avulsion

A

x2

B

Distal Avulsion

C

x2

D

Figure 20B2-36   Suture pattern of nail avulsion. A, Avulsion of the base of the nail. B, Reduced nail is sutured in the fold with two sutures. C, Distal separation (avulsion) of the nail plate.   D, Reduced nail clip back at the distal edge, closed with two sutures.

such as ­ dissolvable sutures under it. The important concept is to save this important biologic structure. It should be valued and used during repair whenever possible.

Evaluation Observe what portion of the nail plate has been avulsed. One must ascertain whether the detachment is distal or proximal, or both and whether the nail plate itself is fractured. If it is avulsed proximally at its base, it must be determined whether the plate is lying superficial to the eponychium (i.e., the epidermis at the base of the nail) or within the nail fold. If it is lying superficial to the eponychium, an underlying dead space is created that can harbor or promote bacterial growth. Often, because of bleeding, it is hard to tell, but a radiograph of the digit will help ascertain whether there is any underlying bony injury. It is popular to detach the nail plate entirely to determine whether there is an underlying injury to the nail bed; if found, the nail bed is repaired by dissolvable sutures. Stabilization of a bony injury by K-wire fixation might be necessary, depending on the stability of the injury after suturing (Fig. 20B2-36).40

Author’s Preferred Method The whole philosophy of repair hinges on the concept of using the nail plate in the repair as a biologic cover of the subungual tissues and as help in the reduction and stabilization of an underlying bony injury. The subungual tissues are well fixed to the underlying bone and are relatively immobile. Débridement of subungual tissue is difficult because the remaining tissue cannot easily be mobilized to cover the resulting deficit. Trying to suture subungual tissue is often frustrating and sometimes results in more devascularization of already compromised tissues. Therefore, if this tissue is visible because of avulsion of part of the nail plate and appears viable, it is simply irrigated and laid back over the

­ istal phalanx. The nail is reduced over that area. This red duction includes reinsertion of the base of the nail into the nail fold. This serves to compress and hold the nail bed tissues in place. The reduced nail plate is then stabilized by sutures to hold it in place. Otherwise, the reduced nail may migrate out of the eponychial fold. Generally, if the nail is avulsed proximally, after thorough irrigation, the nail plate will be reinserted into the eponychial fold. This takes patience in a child. It is then stabilized by one or two sutures at its base (Fig. 20B2-37). Holding the nail reduced and introducing sutures can be tricky. The needle needs to be stout and is bent to 180 degrees (Fig. 20B2-38).

Wrist and Hand 1429

Author’s Preferred Method—cont’d The needle is introduced at the base of the nail perpendicular to the nail plate (and bone). Once the nail plate is penetrated, the tip of the needle will come in contact with the underlying distal phalanx. The needle is backed off of the bone slightly without coming out of the nail plate, then reoriented more proximally so that it passes tangential to the bone and emerges proximal to the eponychial edge of the skin. Avoid trying to pass the needle with one continuous motion because it may rip through the base of the nail plate and hold nothing. Usually, two sutures at the base of the nail plate are necessary to accomplish this. If the distal end of the nail plate is disconnected from the nail bed, it is trimmed of dirty nail and sutured to the edge of the skin (hyponychium) distally with two sutures. To eliminate dead space, this is then compressed with a compression dressing. The dressing is then followed by a cast until the soft tissues are stabilized, which generally takes 2 to 3 weeks. Most nail bed injuries can be handled this way. I have found this method most rewarding. It eliminates suturing of the nail bed, avoiding unnecessary exposure of this tissue, desiccation and shrinkage of this area, and dressing problems resulting from oozing of serous material from this raw uncovered subungual tissue. Occasionally, there is a significant deficit of subungual tissue that might need a nail bed graft from an adjacent digit. The key to a nail bed graft is that it must be exceedingly thin (Fig. 20B2-39).

1 —Insert perpendicular to nail plate

Bend to 180°

2

2 —Back off, rotate needle, and drive through, skiving bone Figure 20B2-37   Technique of needle insertion through the nail plate. A cutting needle on a 4-0 suture is bent to about 180 degrees. The tip is then inserted perpendicular to the nail plate until it hits bone, then partially backed off, rotated, and driven more proximally, emerging through skin.

B

A

C

Figure 20B2-38   A-C, Use of the nail plate in an open nail bed injury. Do not throw away a nail plate. Nail-to-nail suturing is not easy to accomplish but does provide good biologic dressing. If there is an underlying fracture, a K-wire through the bone reduces the nail plate. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1430 DeLee & Drez’s� O

Author’s Preferred Method—cont’d Donor Digit

Recipient Digit

B

A Donor Finger 1 yr f/u

C

Normal Finger

Thumb 1 yr

Normal Thumb

D

Figure 20B2-39   A-D, Completed nail bed graft.

R eferences Please see www.expertconsult.com

S E C T ION C

Wrist Arthroscopy Steven M. Topper

Once the techniques of wrist arthroscopy were mastered, as in larger joints, it quickly became the gold standard diagnostic tool1-5 and a critical therapeutic tool in the management of wrist disorders.6-15 This section highlights the indications, techniques, and outcomes, with special attention to the athlete as the patient.

INDICATIONS Arthroscopy of the wrist is indicated for (1) evaluation of chronic wrist pain, (2) treatment of cartilage (triangular fibrocartilage complex [TFCC]) and ligament tears, (3) resection of synovitis and joint-based ganglia, (4) ­ visualization for

reduction and fixation of intra-articular fractures and acute carpal dislocations, (5) management of ligament attenuation and resultant dynamic instability, (6) treatment of ulnocarpal impaction syndrome or ulnar styloid impaction syndrome (wafer procedure or ulnar styloid recession), (7) loose body removal or débridement of chondral lesions, and (8) partial or complete ostectomy for arthritis.

EQUIPMENT Because the wrist is a significantly smaller and more complex joint than the knee or shoulder, traction and small instruments are required. Traction can be applied overhead

Wrist and Hand 1431

or by several commercially available devices designed specifically for the wrist (Fig. 20C-1). Regardless of the specific traction apparatus chosen, it is preferable to have circumferential access to the joint and the ability to flex, extend, and radially and ulnarly deviate the wrist as well as pronate and supinate the forearm. Scope size selections vary based on the manufacturer, between 2.3 and 2.7 mm in diameter. Short barrel scopes, which facilitate working distance and surgeon comfort, are also preferable. Small joint arthroscopy tools that closely mimic the selection available for larger joints are also available (e.g., probes, biters, graspers, shavers, burs). Inflow and outflow can be done by gravity or a pump. Although this is sometimes dictated by the equipment, it is otherwise the surgeon’s choice. If a powered device is used, lower flow and rate settings are necessary because the wrist is a joint with a 5- to 7-mL capacity.

PORTALS Wrist arthroscopy portals exist as safe pathways to the joint. The wrist is the waist (thinnest segment) of the upper extremity. Despite being an area of compact anatomy, ­surprisingly, 11 portals are available to the surgeon (Fig. 20C-2). These portals are named based on their relationship to the extensor tendon compartments at the radiocarpal level and based on their relationship to the underlying joint at the midcarpal and distal radioulnar joint (DRUJ) levels.

and ulnarly; therefore, it is helpful to localize the portal with a needle while visualizing with the scope in the 3-4 portal. After the skin incision is made, it is necessary to dissect to the joint capsule bluntly, which also increases safety. The 1-2 portal is primarily used when performing an arthroscopic radial styloidectomy.

3-4 Portal The 3-4 portal is the primary portal for initial visualization. It is used for most of the diagnostic arthroscopy of the radiocarpal joint and for instrumenting the radial side of the wrist. It is a good place to start because it is based off of Lister’s tubercle, which is almost always palpable. The portal is placed between the EPL and the tendons of the fourth dorsal compartment 1 cm distal to Lister’s tubercle. While inserting instruments and needles, it is important to remember that the joint tilts volarly 10 to 12 degrees. Initially in the 1980s, longitudinal skin incisions were advised to avoid injury to subcutaneous longitudinal structures. It is unsafe to plunge a No. 11 blade directly into the wrist joint no matter how the blade is oriented. Therefore, the practice of a dermal incision only must be mastered. When that is accomplished, it makes more sense to place the incisions in Langer’s lines, which are transverse at the wrist.

1-2 Portal The 1-2 portal is rarely used because of the proximity of the deep branch of the radial artery. A cadaveric study has shown that a trocar in this portal lies a mean distance of 3 mm from both the deep branch of the radial artery and the sensory branch of the radial nerve.16 The portal is placed between the extensor carpi radialis longus (ECRL) and the tendons of the first dorsal compartment; it is bounded distally by the extensor pollicis longus (EPL). When establishing this portal, safety is increased by erring proximally

Wrist Arthroscopy Tower

STT

MCU

MCR

6-U

1-2

6-R

3-4

DRLU-1

4-5 DRLU-2 Figure 20C-1   Wrist arthroscopy traction tower. (Redrawn from Miller WD: MRI-Arthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 219.)

Figure 20C-2   Wrist arthroscopy portals. (Redrawn from Henry M: Arthroscopic treatment of acute scapholunate and lunotriquetral ligament injuries. Atlas Hand Clin 9(2):190, 2004.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1432 DeLee & Drez’s� O

B

A

C

Figure 20C-3   Arthroscopic view through 3-4 portal. A, Radioscaphocapitate ligament and long radiolunate ligaments. B, Scapholunate interosseous ligament and ligament of Testut. C, Lunate facet of radius and short radiolunate ligament.

This affords a more pleasing postoperative scar and also facilitates movement of the scope or instruments, which is predominantly in a transverse radioulnar plane rather than proximal distal. With the scope in this portal, the following structures are visualized (Fig. 20C-3). Starting radially, the capsular reflection is seen near the dorsal ridge of the scaphoid; next, the articular surfaces of the proximal pole of the scaphoid and the scaphoid facet of the radius are evaluated. The volar radiocarpal ligaments are next visualized. These ligaments are intracapsular and are therefore best seen from the arthroscopic perspective.17 Starting radially, the first ligament encountered is the radioscaphocapitate (RSC), next is the long radiolunate (LRL). The sulcus between these two ligaments is easily visualized and will be discussed further relative to the volar portal.18 Moving radially, the radioscapholunate (RSL), formerly known as the ligament of Testut, is seen. The ligament of Testut was thought to be a critical structural ligament based on its proximity to the scapholunate interosseous ligament (SLIOL). Definitive histologic studies have since demonstrated that it is simply a neurovascular

conduit of no structural significance19 (Fig. 20C-4). At this point, the scope is withdrawn slightly so that the lens can be moved distally (effected by dropping the hand holding the camera) and the SLIOL can be visualized all the way to the capsular reflection separating the radiocarpal joint from the midcarpal joint. This ligament has histologic and structural differences along its course. The dorsal one third and volar one third are composed of structural type 1 collagen, with the dorsal one third being the stronger of the two segments (Fig. 20C-5). The central one third is composed of fibrocartilage and therefore is not structurally significant.20 Moving into the radiolunate joint, the cartilage on the proximal pole of the lunate and the lunate facet of the radius are examined. The sagittal ridge, which

SLId

ST

SLIpx T L S SRL SLip Radius

Figure 20C-4   Illustration of the ligament of Testut. S, scaphoid; L, lunate; T, triquetrum. (Redrawn from Berger RA: The gross and histologic anatomy of the scapholunate interosseous ligament. J Hand Surg [Am] 21:172, 1996.)

RSL

LRL

Figure 20C-5   Illustration of the scapholunate interosseous ligament. ST, scaphotriquetral; SLId, scapholunate dorsal one third; SLipx, scapholunate central one third; SLip, scapholunate volar one third; LRL, long radiolunate; RSL, radio scapholunate or ligament of Testut; SRL, short radiolunate. (Redrawn from Berger RA: The gross and histologic anatomy of the scapholunate interosseous ligament. J Hand Surg [Am] 21:172, 1996.)

Wrist and Hand 1433

A

B

Figure 20C-6   Arthroscopic view through the 4-5 portal. A, Ulnar aspect lunate, triangular fibrocartilage complex, and prestyloid recess. B, Lunatotriquetral interosseous ligament and volar ulnocarpal ligaments.

separates the scaphoid and lunate fossae of the radius, often has chondromalacia-like surface changes that are of little or no clinical significance. Covering almost the entire length of the volar aspect of the lunate facet of the radius is the short radiolunate ligament (SRL). This is the stoutest of the volar radiocarpal ligaments and the last one to rupture in a perilunar dislocation.21At this point, the scope lens is turned toward the ulnar side of the wrist, where the TFCC, lunotriquetral (LT) interosseous ligament, and volar ulnocarpal ligaments can be visualized. The volar ulnocarpal ligaments include the ulnotriquetral, ulnolunate, and deep to these, the ulnocapitate, which is often covered by synovium and difficult to visualize. Fortunately, the ulnocapitate ligament is rarely involved with isolated pathology. It can be injured in a perilunar dislocation, but plenty of other arthroscopic clues will shed light on the nature of this devastating wrist injury. In most wrists, it is possible to visualize the proximal pole of the triquetrum with the scope in the 3-4 portal. In small or tight wrists, this may be difficult, requiring the surgeon to switch the scope to a more ulnar portal to complete the diagnostic portion of the radiocarpal and ulnocarpal arthroscopy.

4-5 Portal

noted that this portal is generally distal to the 4-5 portal. Needle localization is helpful to establish this portal. The 6-R portal provides excellent visualization of the prestyloid recess. This structure is differentiated from an ulnar avulsion of the TFCC by its smooth contours and its synovial lining. In most wrists, it is also possible to visualize the pisotriquetral recess, which lies just to the ulnar side of the ulnotriquetral ligament.

6-U Portal The 6-U portal lies just ulnar to the ECU tendon in close proximity to the dorsal sensory branch of the ulnar nerve. The nerve passes dorsal to the flexor carpi ulnaris and pierces the deep fascia. It becomes subcutaneous on the medial aspect of the forearm at a mean distance of 5 cm from the proximal edge of the pisiform. The nerve gives off an average of five branches with diameters between 0.7 and 2.2 mm that cross dorsally just distal to the ulnar head.22 This portal’s main utility is in instrumenting an ulnar avulsion–type TFCC tear or suture plication of the volar ulnocarpal ligaments.23 Because of the risk for injury to the sensory branch of the ulnar nerve, needle localization and blunt dissection down to the capsule are necessary.

The 4-5 portal lies between the extensor digitorum communis (EDC) and the extensor digiti minimi (EDM). This portal is useful for diagnostic examination of the far ulnar side of the wrist. It is also useful for visualization when instrumenting the far ulnar and radial sides of the wrist. The structures accessible for instrumentation through this portal include the lunate, LT interosseous ligament, radial portion of the triquetrum, volar ulnocarpal ligaments, and radial margin of the TFCC. This portal has great utility because it is a relatively central portal (Fig. 20C-6).

6-R Portal The 6-R portal lies just to the radial side of the extensor carpi ulnaris (ECU) and is between this tendon and the EDM. This portal is of great utility because it generally provides the best access for instrumenting the TFCC (Fig. 20C-7). Because of the 22-degree radioulnar slope of the radius and the desire not to injure the TFCC, it should be

Figure 20C-7   Instrumentation of the triangular fibrocartilage complex through the 6-R portal.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1434 DeLee & Drez’s� O

A

B

C

Figure 20C-8   Midcarpal diagnostic views. A, Scaphoid trapezium trapezoid (STT) joint. B, Scapholunate (SL) interval. C, Capitohamate proximal pole and lunotriquetral (LT) interval.

Volar Portal As facility with wrist arthroscopy has increased, the advantages of minimally invasive surgery have driven the desire to do more and more through the scope. To better visualize the dorsal and far volar aspects of the joint, a volar radiocarpal portal has been developed. It finds utility in visualizing the volar aspect of the SLIOL, the dorsal capsule, and the dorsal rim of the radius. It is established with a switching stick technique. The scope in the 3-4 portal is driven into the sulcus between the RSC and LRL ligaments. The scope is removed from the scope sheath, and a switching stick is placed in the canula. This switching stick is then popped through the joint capsule and should come to lie subcutaneous just radial to the flexor carpi radialis tendon (FCR). A dermal incision is made, and the scope canula is advanced over the switching stick to emanate between the RSC and LRL ligaments in a volar to dorsal direction. Initial descriptions of a volar ulnocarpal portal, volar midcarpal portal, and volar scaphoid trapezium trapezoid (STT) portal have also been made.24-26

Radial Midcarpal Portal The radial midcarpal portal is located 1 cm distal to the 3-4 portal. In this location, there is a palpable sulcus. The scope in this portal allows for visualization of the entire midcarpal joint from the STT joint radially to the triquetrohamate joint on the ulnar side (Fig. 20C-8). The distal extensions of the RSC and ulnocarpal ligaments with the potential space of Poirier in between are often difficult to visualize because of overlying synovium. Evaluation of intercarpal instability is best done from the midcarpal joint. Geissler and Freeland10 described a useful arthroscopic classification system for grading degrees of intercarpal instability (Table 20C-1).

Ulnar Midcarpal Portal The ulnar midcarpal portal lies 1 cm distal to the 4-5 portal. This portal is easily localized with a needle or the light from the scope when it is in the radial midcarpal portal. Generally, this portal is used to instrument the midcarpal joint. If needed, problems on the ulnar side of the wrist are visualized through the radial midcarpal portal and

i­nstrumented through the ulnar midcarpal portal, and vice versa for problems on the radial side of the wrist.

Scaphoid Trapezium Trapezoid Portal The STT portal is used when it is necessary to instrument the STT joint, such as in arthroscopic distal scaphoid recession for STT arthritis (Fig. 20C-9). It is located at the ulnar margin of the EPL adjacent to the ECRL. This portal is easily localized with a needle visualizing through the radial midcarpal portal because sometimes the tendons are not palpable owing to subcutaneous adipose tissue. The radial artery is radial to the EPL tendon, but there are terminal sensory branches of the radial nerve in close proximity to this portal. Jeopardy to these branches is minimized by dermal incision only and blunt dissection to the joint capsule.

Distal Radioulnar Joint Portals The DRUJ joint is visualized and instrumented through two portals. The proximal DRUJ portal is established first. It lies just proximal to the sigmoid notch of the radius

  TABLE 20C-1  Ligament Instability by Arthroscopic Dynamic Evaluation Grade

Description

I

Attenuation or hemorrhage of interosseous ligament as seen from radiocarpal space. There is no incongruity of carpal alignment in midcarpal space. Attenuation or hemorrhage of interosseous ligament as seen from radiocarpal space. Incongruity or step-off of carpal space. There may be a slight gap (less than the width of a probe) between the carpal bones. Incongruity or step-off of carpal alignment as seen from both radiocarpal and midcarpal space. Probe may be passed through the gap between carpal bones. Incongruity or step-off of carpal alignment as seen from both radiocarpal and midcarpal space. There is gross instability with manipulation. A 2.7-mm arthroscope may be passed through the gap between the carpal bones.

II

III IV

From Geissler WB, Freeland AE, Savoie FH, et al: Intracarpal soft tissue lesions associated with an intra-articular fracture of the distal end of the radius. J Bone Joint Surg Am 78:357-365, 1996.

Wrist and Hand 1435 Ulnar carpal ligament ECU in sheath

Central TFC

Figure 20C-9   Instrumenting the scaphoid trapezium trapezoid (STT) joint through the STT portal.

between it and the ulnar metaphysis. The distal portal is located between the distal surface of the ulnar head and the TFCC. To avoid injury to the TFCC, this portal is localized with a needle while visualizing through the proximal portal. These portals are primarily used for evaluation of DRUJ arthritis and for loose body removal.

OPERATIVE WRIST ARTHROSCOPY Triangular Fibrocartilage Complex Tears The advances in wrist arthroscopy operative technique and instrumentation have made the necessity for open surgery for TFCC pathology rare. The TFCC is critical to the stability of the DRUJ and the ulnar side of the carpus, and it provides a continuous gliding surface from the radiocarpal joint to the ulnocarpal joint for the proximal carpal row27-29 (Fig. 20C-10). It arises from the articular cartilage in the area of the sigmoid notch of the radius and inserts into the base of the ulnar styloid and volarly into the ulnocarpal ligament complex in a confluence of fibrous tissue that gives rise to the ulnolunate and ulnotriquetral ligaments.18 Dorsally, the fibers of the ECU subsheath crisscross deep to the ECU tendon and blend with the dorsal rim of the TFCC. This illustrates the important dynamic stabilizing effect of the ECU tendon on the ulnar carpus and DRUJ.30 At the origin of the TFCC from the sigmoid notch of the radius, there are dorsal and volar fibrous thickenings called the dorsal and volar radioulnar ligaments. They function to limit pronation and supination as well as axial migration. In full supination, the volar ligaments are taut, and in full pronation, the dorsal ligaments are taut. Additionally, the ulna translates dorsally with pronation and volarly with supination within the sigmoid notch.29,31 The DRUJ is most stable to translational motion at the extremes of pronation and supination. Confluent with, and emanating from, the dorsal and volar radioulnar ligaments is a peripheral thickening of the TFCC known as the meniscus homologue or the V ligament. These ligaments coalesce to the apex of the V into a structure called the ligamentum subcreutum, which inserts at the base of the ulnar styloid and is critical to

Ulna

Radius Normal TFC complex

Figure 20C-10   Illustration of the triangular fibrocartilage complex. (Redrawn from Cooney WP, Linschied RL, Dobyns JH: The wrist: Diagnosis and operative treatment. St. Louis, Mosby Yearbook, 1998.)

DRUJ stability.19,30,32 The articular disk (central portion) of the TFCC is composed of fibrocartilage and is avascular and aneural. Within the substance of the disk, there are interwoven sheets of obliquely oriented collagen fibers that are thought to primarily withstand compressive forces. The thickness of the articular disk is inversely proportional to ulnar variance.33 Additionally, ulnar variance has an implication on load transmission through the TFCC and distal ulna. In a wrist with neutral ulnar variation, about 20% of the axial load across the wrist is transmitted through the TFCC and distal ulna. When the ulnar variance is altered from −2 to +2.5, it alters the load borne through the TFCC and distal ulna from 8% to 40%.29 There are also changes in dynamic load transmission due to a screw home–type mechanism between the two forearm bones.34,35 In supination, the radius moves distally on the ulna, creating a relative negative ulnar variance. In pronation, the radius moves proximally, creating a relative positive ulnar variance. This has implications on dynamic ulnocarpal impaction syndrome.34 The blood supply to the TFCC arises from the ulnar and the anterior interosseous arteries. These vessels supply the peripheral 10% to 40% of the disk. The central section and radial origin are avascular. The percentage of the periphery of the TFCC that is vascular diminishes with age. 36,37

Classification TFCC tears are divided into two broad categories, traumatic (type 1) and degenerative (type 2).38 Traumatic lesions are subdivided based on location (Fig. 20C-11). A type 1A tear is in the central articular disk, and although this is an aneural area, mechanical symptoms can be produced as the unstable flap gets caught in the joint, causing abnormal tension on the innervated periphery. A type 1B tear represents a detachment of the TFCC from its insertion on the ulnar styloid. This can be associated with an avulsion fracture involving the base of the

�rthopaedic ����������� S �ports ������ � Medicine ������� 1436 DeLee & Drez’s� O

1C

1A

1B

1D

the wrist in ulnar deviation. The wrist is then flexed and extended by the examiner with a gentle rolling motion. Pain provocation or a painful snap indicates a positive test. The TFCC (or ulnocarpal) compression test starts with the forearm in neutral rotation. The wrist is then dorsiflexed maximally as the examiner rotates the forearm into pronation. Reproduction of symptoms is a positive test. The stability of the DRUJ should also be assessed. This is done by shucking the ulna while holding the radius stable. A positive test is discernable as increased motion compared with the opposite side. Another way to test this is with the “piano key” sign. The patient is simply asked to push on the examination table with the flat of the hand. Dorsal to palmar translation of the ulna is observed and compared with the opposite side. It is also necessary to assess for LT ligament instability, which is done with an LT ballottement test.39

Radiographic Examination

Figure 20C-11   Illustration of the Palmer classification of triangular fibrocartilage complex tears. (From Miller MD: MRIArthroscopy Correlative Atlas. Philadelphia, WB Saunders, 1997, p 217.)

ulnar styloid. Because of the intimate association with the ECU subsheath, ECU subluxation can also be seen, as can demonstrable clinical instability of the DRUJ. Type 1C tears are volar and involve the origin of the volar ulnocarpal ligaments. This injury is rare and can be associated with volar subluxation of the ulnocarpal bone relative to the distal end of the ulna. Type 1D tears represent a radial detachment of the TFCC from the sigmoid notch of the radius. By definition, this tear involves either or both of the radioulnar ligaments (dorsal and volar). This injury is also rare and is high energy in nature, so it is often associated with a clinically unstable DRUJ. Tears of the central articular disk near the sigmoid notch that do not involve one or both of the radioulnar ligaments are treated as type 1A tears.

History and Clinical Examination A patient with an acute TFCC tear presents with ulnarsided wrist pain that is usually associated with painful snaps, clicks, or pops. There is often a history of injury that involves axial load on a pronated wrist, and symptoms are generally provoked by activities that involve wrist rotation. Examination findings include pain with palpation over the TFCC and often the ECU because tendinosis of this tendon commonly coexists. The TFCC snap test is performed with the forearm in neutral rotation and

Radiographic examination should include plain films taken in neutral rotation. These films are assessed for carpal alignment, ulnar styloid morphology, DRUJ degenerative joint disease, and particularly ulnar variance because there is a higher incidence of TFCC tears in ulnar-­positive wrists.40 In the past, arthrography was a popular way to demonstrate dye leakage between compartments in the wrist. Recent evidence has demonstrated a high incidence of similar findings in the opposite, asymptomatic side.41,42 It has also been demonstrated that there are clinically insignificant congenital communications and age-related degenerative perforations in the TFCC and intercarpal ligaments, all of which make the finding of dye leakage between compartments of questionable value.37,43-45 In light of the fact that it is also an invasive test, arthrography has little use in current practice.4,46-48 Computed tomography is generally reserved for evaluation of chronic instability of the DRUJ and has little role in evaluation of acute TFCC tears. Magnetic resonance imaging (MRI) is the gold standard for imaging evaluation of TFCC pathology. 49-54 The T2-weighted image in the coronal plane has the greatest diagnostic value and has been shown in a number of studies to have a 90% accuracy rate for central and radial-sided tears.55-58 The sensitivity of MRI for ulnar-sided avulsions is not as good.59,60

Treatment Options Type 1A Tear The symptoms from a type 1A tear are mechanical and generally resolve with resection of the unstable flap. Some controversy exists over a central disk tear that involves the area of the TFCC that is confluent with the cartilage at the edge of the sigmoid notch of the radius. Most authors agree that if this tear does not involve the dorsal and volar radioulnar ligaments, it is not associated with DRUJ instability and should be treated as a type 1A tear. Excision of up to two thirds of the central disk does not adversely affect load transmission or stability.61,62 During resection, the dorsal and volar radioulnar ligaments, as well as the remainder of the V ligament, are carefully preserved.

Wrist and Hand 1437

Author’s Preferred Method Wrist arthroscopy is performed under regional or general anesthesia and with a tourniquet to control bleeding. I prefer a commercially available traction tower, which allows for 10 to 12 lb of inline traction, circumferential access, and the ability to position the wrist as well as control pronation and supination. The joint is distended with 7 mL of saline before establishing portals. The scope is initially placed in the 3-4 portal, and the diagnostic portion of the procedure is accomplished. The 6-R portal is then established, and a small 2.0 to 3.0 full radius suction shaver is introduced to clear synovitis, which is generally present with a symptomatic tear of the TFCC. It is preferable to separate the visualization portal and instrumentation portal as far apart as possible to avoid “sword fighting.” In some tight or very large wrists, it will be necessary to visualize the TFCC through the 4-5 portal. The torn portion and general tension (stability) of the TFCC are then examined with a probe. The unstable

A

flap is then resected with a combination of small joint biters or a suction punch. A small joint banana blade can also be helpful for the far ulnar portions of the tear. For the past 7 years, I have performed most type 1A tear resections with an electrothermal device (Fig. 20C-12). I prefer a monopolar device because of the feedback temperature control and the depth of penetration (4 mm). Bipolar probes work by boiling the arthroscopy fluid between the two tips on the probe, which I believe is dangerous in a joint with a 7-mL capacity. It is occasionally necessary to switch the scope to the 6-R portal and instrument through the 3-4 or 4-5 portal to get a better angle on the ulnar side of the tear. After completion of the radial and ulnocarpal portion of the arthroscopy, the diagnostic portion is continued with midcarpal arthroscopy, paying particular attention to the stability of the LT interval. Good to excellent outcome can be expected in 85% of patients.12,63-65

B

Figure 20C-12   Type 1A triangular fibrocartilage complex tear. A, Before resection. B, Resection with electrothermal ablative device.

Postoperative Care

The wrist is immobilized in neutral rotation with a sugartong splint for 2 weeks. At that point, the wounds are checked, and an Orthoplast sugar-tong splint is fashioned to be worn for comfort for up to 4 more weeks. A range of motion program is started after 2 weeks, followed by strengthening once the motion is 80% of the contralateral side. Criteria for Return to Sports

A high-performance athlete could return to sport as soon as 3 weeks after surgery if he or she is able to wear a sugartong splint while playing. Otherwise, it is generally advisable to wait until motion and strength are at least at 80% of the contralateral side at 5 to 6 weeks.

Type 1B Tear Type 1B tears can be difficult to diagnose because the MRI findings are not reliable. These patients typically present with ulnar-sided wrist pain and a history of a significant

injury. Clinically, DRUJ instability is often demonstrable, and the patient may have concomitant ECU tendon instability or tendinosis. In a patient in whom the diagnosis is in question and who has failed a trial of nonoperative care, a diagnostic arthroscopy can be revealing. The pathognomonic finding is an abnormally loose TFCC. Normal tension of the TFCC has been likened to a drumhead or trampoline. The tautness of the TFCC is examined with a probe, and if it is more like a feather pillow than a trampoline, peripheral detachment should be suspected.66 Up to 50% of these lesions are associated with an unstable ECU tendon that emanates from attenuation or disruption of the ECU subsheath. This structure is intimately associated with the dorsal-ulnar rim of the TFCC and can be involved in a continuum of injury, as postulated by Melone and Nathan.67 In fact, ECU tendinosis in a young person on the MRI is a clue to significant dorsal ulnar pathology of the TFCC. In this case, open repair or reconstruction of the ECU subsheath, in combination with TFCC repair, is necessary to resolve the problem. 68 Arthroscopic suture repair of a type 1B tear is effective and yields results that are

�rthopaedic ����������� S �ports ������ � Medicine ������� 1438 DeLee & Drez’s� O

Author’s Preferred Method A standard wrist arthroscopy setup is used. The tear is visualized through the 3-4 portal and instrumented through the 6-R portal. Once synovectomy and probe-assisted examination of the TFCC are completed, the tear is débrided with a full radius shaver to remove scar tissue and promote capillary proliferation and healing. Next, the transverse 6-R portal is extended distally 1.5 cm by converting it to a small hockey stick–type incision. The subcutaneous tissues are bluntly dissected until the ECU sheath is identified. This sheath is divided longitudinally over a 1.5-cm length, which allows for safe mobilization of the ECU and ulnar retraction without concern for postoperative instability. Several microsuture passers are available with an appropriate diameter and curve to perforate the TFCC without destroying it. They often include a straight passer that comes equipped with a wire loop to retrieve the suture that has been passed through

A

the TFCC. The curved passer is placed first just distal to the ulnar head. While visualizing, the needle is passed by the surgeon, dropping the hand so that the needle perforates the TFCC with 2- to 3-mm bite. Next, the straight passer is brought in through the region of the previous 6-R portal, which is above the TFCC. The wire loop is deployed and placed over the tip of the curved needle (Fig. 20C-13). Next, a 2-0 absorbable (2-0 PDS) suture is passed through the curved needle and captured by the wire loop. One to three sutures are sufficient to repair most type 1B tears. The sutures are not tied until all sutures are placed. Once all sutures are placed, the wrist is positioned in supination, and then the sutures are sequentially tied. The wrist is kept in this position during closure and dressing until the sugar-tong splint is on. This procedure yields 85% to 90% good to excellent results at long-term follow-up.70

B

C

Figure 20C-13   Arthroscopic repair of type 1B tear. A, Débridement of synovitis and unstable scar tissue. B, Passing PDS suture with a microsuture passer. C, Repair after sutures are tied.

comparable with open repair.69,70 The sutures can be placed with either an outside-in or an inside-out technique. Most authors think that the outside-in technique allows for better control of suture knot placement to avoid constriction of the sensory branch of the ulnar nerve and ECU tendon. Postoperative Care

The wounds are checked 2 weeks after surgery, and the splint is converted to a long arm cast with the forearm in full supination or an Orthoplast splint depending on the patient. At 6 weeks, active and gentle passive range of motion therapy is started. Once motion is 80% of the opposite side, a strengthening program is started, focusing on grip and wrist flexors and extensors. Activities of daily living are allowed at this point. Once the strength returns to 80% of the other side, sports, tool use, and heavy lifting are allowed. Criteria for Return to Sports

A high-performance athlete could be returned to sports as soon as 3 weeks after surgery if he or she is able to wear a padded long arm cast while playing. Otherwise, it is generally advisable to wait until their motion and strength are at least at 90% to 100% of the contralateral side 8 to 12 weeks after surgery.

Type 1C Tear Type 1C tears are relatively rare. When present, they can be associated with a debilitating carpal supination deformity. Arthroscopically, the diagnosis is confirmed by demonstration of loss of tension in the volar rim of the TFCC and volar ulnocarpal ligaments. An actual tear can be visualized acutely, but often by the time these patients are arthroscopically examined, the tear is obscured by synovitis and scar tissue. There are two approaches to managing these tears: suture imbrication of the ulnotriquetral and ulnolunate ligaments, or advancement of the volar ulnocarpal tissues to a suture anchor placed in the triquetrum. The imbrication technique is done with arthroscopic assistance and visualization through a 1-cm incision placed just volar to the ECU tendon in the area of the triquetral “snuff box.” The soft tissues are dissected bluntly to expose the volar ulnar wrist joint capsule and protect the sensory branch of the ulnar nerve and the ulnar neurovascular bundle. Similar to the type 1B repair, sutures are passed on one side of the defect in the volar ulnocarpal ligaments and retrieved with the straight passer on the other side. They are then tied over the capsule (Fig. 20C-14).

Wrist and Hand 1439

Type 1D Tear A radial detachment of the TFCC from the sigmoid notch by definition involves the dorsal or volar radioulnar ligaments and is a highly unstable injury. It is rare to encounter this lesion in isolation and much more common to see it in conjunction with a distal radius fracture. Excellent results with open repair have been reported.71 Arthroscopic repair techniques have been described by several authors and appear to yield equivalent results.72,73

Author’s Preferred Method

Figure 20C-14   Suture imbrication of ulnocarpal ligaments. (Redrawn from Moskal MJ, Savoie FH, Field LD: Arthroscopic capsulodesis of the lunotriquetral joint. Clin Sports Med 20:141-153, 2001.)

Author’s Preferred Method Sutures placed with the previously described imbrication technique are generally in line with the fibers of the ligament, so I worry about them holding. For this reason, I prefer advancing the proximal end of the tear with a horizontal mattress suture to an anchor in the triquetrum. The same incision is used as well as the same careful dissection. The triquetrum is palpable in this location. It is necessary to keep the anchor out of the pisotriquetral joint, so for surgeons unfamiliar with the anatomy in this location, an image intensifier is useful. Once the anchor is placed in the triquetrum, one limb of the suture is passed with a horizontal mattress technique and arthroscopic visualization through the proximal aspect of the ulnotriquetral and ulnolunate ligaments just distal to the TFCC. This is then tied to the other limb because these tissues are drawn distally, closing the defect and retensioning the volar ulnocarpal capsule and volar rim of the TFCC.

Postoperative Care

The patient is initially immobilized in a sugar-tong splint with the forearm in slight pronation to take tension off the repair and also increase comfort. The wounds are checked at 2 weeks, and the immobilization is converted to a long arm cast or Orthoplast sugar-tong splint in the same position, depending on the patient. Immobilization is discontinued at 6 weeks, and a therapy program similar to that for a type 1B tear is started. Criteria for Return to Sports

The recommendations are identical to those for a type 1B tear.

A standard wrist arthroscopy setup is used. The tear is visualized through the 3-4 portal. A 2- to 3-mm bur is introduced through the 4-5 portal, and the rim of the sigmoid notch of the radius (original attachment site of the TFCC) is decorticated to expose bleeding bone. Next, a long 6.2-inch K-wire is directed from the ulnar side of the wrist through a small incision just volar to the ECU and distal to the TFCC. This K-wire is then passed from the sigmoid notch to emanate on the radial aspect of the radius. A counterincision is made over the tip of the K-wire, and soft tissues, such as the first dorsal compartment tendons, the sensory branch of the radial nerve, and the cephalic vein, are mobilized and retracted. A second K-wire is placed in a similar fashion 8 to 10 mm volar to the first. Initially, image intensification is helpful during this portion of the procedure. Once the holes are drilled, the K-wires are removed. Next, a Tuohy needle is introduced into the joint through the ulnar incision. This needle is passed through the TFCC and then into the dorsal hole. It is passed through the ­radius to emanate from the radial incision. A twisting motion facilitates needle passage through bone. A 2-0 PDS suture is passed into the tip of the needle and is then withdrawn into the ulnocarpal joint. With the suture still in the tip of the needle, the needle is moved volarly to perforate the TFCC in this location and then is advanced into the second volar hole to deliver the suture back out through the radius and out the radial incision. This effects a horizontal mattress ­repair of the TFCC. The Tuohy needle has been designed so that it is blunt and difficult to cut suture with. The suture is then tied over bone in the radial incision while appropriate tension on the TFCC is visualized arthroscopically. Jantea has described an alignment guide for hole placement in the radius that some surgeons find beneficial.73

Postoperative Care

The patient is initially immobilized in a sugar-tong splint with the forearm in neutral rotation. The wounds are checked at 2 weeks, and the immobilization is converted to a long arm cast in the same position, depending on the patient. Immobilization is discontinued at 6 weeks, and a therapy program similar to that for a type 1B tear is started. Criteria for Return to Sports

The recommendations are identical to those for a type 1B tear except that a stable DRUJ should be confirmed clinically before competitive play.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1440 DeLee & Drez’s� O

Type 2 Tear and Ulnocarpal Impaction Syndrome Degenerative tears of the TFCC result from chronic overloading of the ulnocarpal joint. This primarily occurs as a result of a spectrum of pathology from ulnocarpal impaction syndrome associated with positive ulnar variance. Hulten74 demonstrated a much higher incidence of degenerative TFCC tears in wrists with positive ulnar variation (73%) than in those with negative ulnar variation (17%). This observation was further substantiated by Palmer and Werner,29 who demonstrated that altering the ulnar variation from neutral to +2.5 mm in a cadaveric model altered the load borne by the ulnocarpal joint from 18% to 40% of total load across the wrist. The impaction may also be dynamic in ulnar neutral to negative wrists as a result of power grip or activities requiring forceful pronation.34 Other secondary causes that can lead to ulnocarpal joint overloading include distal radius fracture malunion, distal radius physeal arrest, and proximal radial migration following radial head resection. Classification

The progressive degenerative changes seen with ulnocarpal impaction syndrome are subdivided into five categories: Type 2A: wear of the TFCC without perforation or adjacent chondromalacia Type 2B: wear of the TFCC associated with adjacent chondromalacia of the lunate or ulnar head Type 2C: perforation of the TFCC with lunate chondromalacia Type 2D: TFCC perforation with lunate or ulnar head chondromalacia and perforation of the LT ligament; no static instability pattern such as volar intercalated ligament instability (VISI) Type 2E: TFCC perforation with arthritis of the lunate and ulnar head; perforation of the LT ligament that may be associated with VISI History and Clinical Examination

Patients present with insidious onset of ulnar-sided wrist pain, often without a history of injury. If they do have a history of a previous distal radius fracture, childhood radius fracture, or Essex-Lopresti–type injury, this should raise suspicion for secondary ulnocarpal impaction syndrome,

which is often managed open to correct the underlying problem such as radius malunion with excessive dorsal tilt.75 The physical examination reveals tenderness over the ulnocarpal joint. Patients have a positive ulnocarpal impaction test and often have a positive TFCC snap. The LT joint should be examined for instability with provocative maneuvers such as the shuck test.76 Differential measurement of grip strength may be less in pronation than in supination. Radiographic Examination

Plain films taken in neutral rotation are assessed for ulnar variation. Positive ulnar variation correlates strongly with ulnocarpal impaction syndrome. Arthritis involving the ulnocarpal joint to include the distal ulna, DRUJ, and proximal pole of the lunate, as well as VISI, is indicative of advanced stages. More subtle findings include degenerative cysts in the proximal pole of the lunate and the ulnar head. MRI is sensitive at revealing the early stages of ulnocarpal impaction syndrome. Imaeda and associates50 noted focal abnormal signal intensity of the ulnar aspect of the lunate in 87% of wrists, of the radial aspects of the triquetrum in 43%, and of the radial aspects of the ulnar head in 10% in patients with clinical ulnocarpal impaction syndrome. Other imaging modalities, such as bone scan, arthrography, and ultrasound, are generally not helpful. Methods of Treatment

The goal of treatment is to decompress the ulnocarpal articulation and remove any degenerative tissue that is causing mechanical symptoms. This can be accomplished with an open diaphyseal ulnar shortening, which affords the added benefit of tightening the volar ulnocarpal ligaments, which is beneficial in cases with coexistent LT instability. Another approach is a partial ulnar head recession (the wafer procedure) that was originally described by Feldon and colleagues.77,78 This procedure involves a resection of the distal 2 to 4 mm of the ulnar head while leaving the portion of the ulnar head that articulates with the sigmoid notch of the radius and the TFCC insertion at the ulnar styloid intact. When there is a degenerative hole in the TFCC, this procedure can be accomplished arthroscopically.79 The other options are salvage in nature such as Darrach’s resection of the distal ulna or ulnar head replacement. These salvage procedures are reserved for cases with significant DRUJ arthritis.

Author’s Preferred Method Types 2A and 2B

Type 2C

The role of arthroscopy is mainly diagnostic. Arthroscopic examination allows for a careful assessment of chondromalacia, TFCC, and stability of the DRUJ and LT intervals. ­Degenerative fronds of cartilage and synovitis are also ­débrided arthroscopically. The ulnocarpal joint is then decompressed with an open diaphyseal ulnar shortening. Patients at risk for nonunion based on their general health, smokers, and patients who have return-to-activity time constraints are alternatively managed with an open wafer procedure.

The existing central perforation in the TFCC in this type can be exploited to allow for arthroscopic management. The central perforation is débrided back to a stable rim with the same techniques as described for a type 1A tear. While viewing through the 3-4 portal, a 3-mm bur is introduced through the 6-R portal. The ulnar head exposed through the hole in the TFCC is then recessed with the bur (Fig. 20C-15). As the procedure progresses, pronation and supination of the forearm allow the surgeon to expose new areas of the

Wrist and Hand 1441

Author’s Preferred Method—cont’d ulnar head for recession. The ulnar margin of the sigmoid notch of the radius serves as a reference for the resection level. This process generally results in a 3- to 4-mm recession of the radial two thirds of the ulnar head. It can be difficult to reach the far ulnar aspect of the ulnar head, and

often fluoroscopic examination reveals a small nib of bone remaining just to the radial side of the sulcus before the ulnar styloid. This can be accessed through the distal DRUJ portal with the bur or a small osteotome if it tents the TFCC with pronation or supination.

A

B

C

D

Figure 20C-15   Arthroscopic ulnar head recession. A, Degenerative tear of the triangular fibrocartilage complex. B, After triangular fibrocartilage complex débridement back to stable rim demonstrating prominent ulnar head associated with impaction. C, Working through a hole in the triangular fibrocartilage complex to recess the ulnar head with a 2.9-mm bur. D, After ulnar head recession.

Postoperative Care

An ulnar gutter splint is worn for 2 weeks, and then the wounds are checked. An Orthoplast ulnar gutter splint is then fashioned and worn for an additional 2 weeks intermittently for comfort as an active range of motion program is started. A graduated strengthening program is initiated 4 weeks after surgery, and most patients complete this program 2 to 4 weeks later. Criteria for Return to Sports

Because structural healing is not an issue with this procedure, an athlete can return to sports as soon as comfort allows. Athletes could potentially be returned sooner with a protective padded sugar-tong splint.

Types 2D and 2E When there is radiographic evidence of arthritis (type 2E), a salvage procedure such as Darrach’s resection of the distal ulna or ulnar head replacement, should be considered. Type 2D lesions are amenable to arthroscopic management. These lesions are managed exactly like the type 2C lesions with one exception, which is a careful assessment of the LT interval. By definition, a type 2D lesion involves at least a perforation of the LT interosseous ligament, which is visualized from the ulnocarpal perspective. This is further assessed from the midcarpal perspective. If there is significant translational motion or step-off at the LT interval, it is unstable and better managed with an open diaphyseal ulnar

�rthopaedic ����������� S �ports ������ � Medicine ������� 1442 DeLee & Drez’s� O

shortening, which tightens the volar ulnocarpal ligaments. If not, a simple débridement of the torn portion of the LT interosseous ligament is added to the previously described procedure for a type 2C lesion. Whipple80 advocated a multiple pinning procedure to induce a fibrous pseudoarthrosis of an unstable intercarpal interval. This procedure is performed with the assistance of image intensification and involves placing three to five small K-wires across the interval, which has been reduced under arthroscopic visualization. To allow the fibrous tissue induced to form by the multiple K-wires to mature, a cast is worn for 8 weeks after surgery. This procedure can be considered an alternative or addition to diaphyseal ulnar shortening in patients with a persistently unstable LT interval after diaphyseal shortening or in patients who are at high risk for nonunion.

Ligament Tears The proximal carpal row is known as an intercalated segment because the bones are almost entirely covered by articular cartilage and have no direct attachment to muscle-tendon units. The stability of these bones is dependent on their geometry and ligamentous connections. Disruption or attenuation of these ligaments leads to predictable instability patterns that remain a vexing problem because of the small size of the ligaments, their variable histologic composition, and their intra-articular location. Additionally, the tolerances are tight, making accurate reconstruction of these ligaments an extreme technical challenge. The various approaches that have been developed all have the same goal of restoring stability and alignment to prevent abnormal articular loading and the inevitable arthritis and joint destruction that follow. Initially, arthroscopy played a diagnostic role, but as instrumentation and technique have advanced, it is playing more of a therapeutic role.81

Classification The Geissler (see Table 20C-1) classification system is a useful tool for arthroscopic grading of degrees of instability because of its defined and reproducible parameters. The judgment is made from the midcarpal perspective.

History and Clinical Examination Patients present with wrist pain that can usually be localized as predominantly radial or ulnar sided. They often have a history of remote trauma, but do not always remember this. Their complaints are of activity-related pain with mechanical symptoms. Myriad provocative examination maneuvers were designed to illicit instability and provoke symptoms. These maneuvers are examiner dependent and a learned art. The specific tests and methods are discussed elsewhere in this chapter, but it is important to remember to take advantage of comparison to the opposite side while this art is being learned.

Radiographic Examination The radiographic evaluation of intercarpal instability is discussed elsewhere in this chapter. It is important to remember that arthroscopy remains a valuable diagnostic

tool to help determine whether perforations picked up on an arthrogram or MRI are congenital, degenerative, or clinically significant tears.82

Treatment Options Management of intercarpal instability secondary to ligament attenuation or disruption is fraught with problems. Because the degree of instability and the chronicity increases, the chances of achieving a good outcome decreases. Traditionally, the arthroscopic approach involves débridement of the torn ligament and, in cases with demonstrable instability, the attempt to create a fibrous pseudarthrosis to stabilize the interval. Débridement alone is effective for grade 1 lesions.14 It likely effects a localized denervation of the damaged ligament and, when combined with synovectomy, effectively diminishes symptoms. In cases of demonstrable instability, such as grade 3-4 lesions, more is required to attempt to correct existing or potential malalignment that can lead to arthritis. The technique described by Whipple83 involves débridement of the interosseous space and placement of multiple small K-wires with the interval anatomically aligned. The patients then wear a cast for 6 to 8 weeks to allow for establishment of a stable fibrous pseudarthrosis at the interval. In this series, this approach was effective for patients who had symptoms for less than 3 months and had a scapholunate gap of less than 3 mm. The results were substantially less than placebo effect for those with more chronicity and greater instability. My personal experience has demonstrated that this approach is more effective at the LT interval than the scapholunate interval. When a K-wire leaves the scaphoid and touches the lunate, the interval gaps open 1 to 2 mm before the K-wire enters the bone. At an interval where a 2-mm gap is normal and a 3-mm gap is not, this is a problem. Some have tried a variable pitch screw placed arthroscopically to circumvent this issue and improve compression. To my knowledge, no published reports using this approach exist. The availability of small electrothermal probes has prompted some investigators to apply this technology to the troublesome minor attenuations (grade 2 lesions).84 In these cases in which there is a degree of instability, débridement alone appears insufficient, and arthroscopic multiple pinning or open linkage, stabilization, or salvage procedures seem like overkill. This technology employs radiofrequency energy to denature heat-labile cross-links in collagen by resistive ohmic heating. This results in shortening of the collagen fibril as the triple helix uncoils and shrinks. Through the reparative process, the heat-labile cross-links are reestablished, resulting in remodeling of the collagen fibril in a shortened posture. During the reparative phase, the thermally modified tissues are weaker and susceptible to elongation if left unprotected. This has caused a disturbing recurrence rate when applied to the shoulder, where excessive immobilization is avoided to prevent unacceptable postoperative stiffness.85 Geissler84 reported on 19 patients with interosseous ligament shrinkage and noted good to excellent results on the Mayo wrist score in those with a grade 2 instability. The results achieved were not as good in those with grade 3 instability. The advantage of the application of

Wrist and Hand 1443

Author’s Preferred Method Grade 1 lesions are treated with synovectomy and débridement of the torn portion of the ligament. I often combine this with a posterior interosseous nerve neurectomy if the patient responded to a diagnostic injection of the posterior interosseous nerve preoperatively. Remember in this case there is no structural instability, so the point of this operation is to relieve pain. Grade 2 lesions are managed in an identical fashion with the addition of electrothermal shrinkage of the unstable interval. I use a monopolar probe with the previously mentioned temperature and power settings. The focus is on the dorsal one third and volar one third portions of the ligament where the structural type one collagen exists. One to two passes are used, and a striping technique rather than a

A

painting technique is employed wherever space allows (Fig. 20C-16). It is important to carefully monitor temperature to prevent inadvertent chondrolysis. This is done using a probe with a temperature feedback loop or careful monitoring of the temperature of the outflow fluid. Grade 3 and 4 lesions are documented arthroscopically and managed with an open stabilization or linkage procedure as discussed elsewhere in this chapter. This author has had success managing ­selected patients with midcarpal instability with thermal tissue shrinkage. Proper patients for this procedure include those with relatively minor instability and in young patients in whom open ligament reconstruction or limited arthrodesis is associated with unacceptable postoperative consequences.

B

C

Figure 20C-16   Arthroscopic picture of electrothermal shrinkage of scapholunate interosseous ligament. A, Attenuated scapholunate interosseous ligament (convex) associated with occult dorsal carpal ganglion consistent with dorsal wrist syndrome. B, Electrothermal capsulorrhaphy of the scapholunate interosseous ligament after ganglion stalk and dorsal capsular resection. Note the exposed extensor tendons. C, After electrothermal capsulorrhaphy, the scapholunate interosseous ligament has a yellow-brown tinge and is concave.

this technology in the wrist compared with the shoulder is that a short arm cast is well tolerated, allowing for proper healing of the tissues during the reparative process. Generally, monopolar probes are favored over bipolar probes to reduce the risk for chondrolysis in a small capacity joint. Those that have experience applying radiofrequency energy to the wrist recommend a temperature setting of 68 degrees and a wattage setting of 30.84

a protective padded sugar-tong splint. It is advisable to protect grade 2 patients until the thermally treated tissue is fully remodeled. Patients do not return to competition until their range of motion and strength are 80% to 90% of the opposite side and for at least 3 months after surgery.

Postoperative Care

Occasionally, an athlete with rheumatoid arthritis may be plagued with synovitis that does not respond to medical management. Alternatively, it may be advisable to consider surgical treatment early to avoid the side effects and associated performance degradation that can occur with certain medications. When compared with arthrotomy, arthroscopic synovectomy allows for better visualization, better access to articular recesses, and an outpatient approach that minimizes capsular disruption and scarring. The contraindication to arthroscopic synovectomy of the wrist is overlying dorsal extensor compartment tenosynovitis because establishing portals may put these weakened and vascularly compromised tendons at risk for disruption. The results of arthroscopic synovectomy reliably yield improved motion and grip strength as well as diminished rehabilitation time compared with open ­synovectomy.6,86,87

Grade 1 patients wear an Orthoplast short arm splint until the swelling and inflammation subside. Range of motion exercises are started on the fifth postoperative day. A graduated strengthening program is started 2 weeks after surgery if necessary. Grade 2 patients are immobilized for the first 6 weeks after surgery. They are then converted to a removable Orthoplast splint while active range of motion exercises are started. At 8 weeks, a graduated strengthening program is started if necessary.

Criteria for Return to Sports For grade 1 patients, structural healing is not an issue, so an athlete can be returned to sports as soon as comfort allows. They could potentially be returned sooner with

Synovectomy

�rthopaedic ����������� S �ports ������ � Medicine ������� 1444 DeLee & Drez’s� O

History and Clinical Examination The time since diagnosis and medications should be recorded. Drugs that could potentially compromise the immune system should be taken into consideration when assessing the risks for anesthesia and surgery. Clinically, these patients present with generalized achy wrist pain and swelling. Range of motion may be limited, and there is usually a palpable boggy synovitis.

Radiographic Examination Plain film radiographs are used to access joint space preservation and carpal alignment. Patients who have significant loss of cartilage space or ulnocarpal subluxation are unlikely to benefit significantly from arthroscopic synovectomy. If it is difficult to clinically differentiate joint synovitis from overlying tenosynovitis, an MRI can be helpful in making this distinction.

Author’s Preferred Method

some people. As pressure builds, the fluid can herniate out of the joint in the weak area between the dorsal radiocarpal and dorsal intercarpal ligaments. In this location, it is seen as foreign, which induces a fibrous response that walls it off, creating a cyst that communicates with the joint. Because the cyst communicates with the joint, it can vary in size and sometimes spontaneously resolve. Aspiration is effective only 40% of the time.91 Obvious or occult cysts can be quite painful if they put pressure on the posterior interosseous nerve. When nonoperative measures fail, it is reasonable to consider surgical resection. Because the patient is trading a bump for a scar, minimizing the scar with an arthroscopic approach makes particularly good sense. Additionally, some of these cysts can be associated with underlying dynamic scapholunate instability as a spectrum of “dorsal wrist syndrome.”90 Arthroscopic resection allows for a careful evaluation of the stability of the scapholunate interosseous interval.

History and Clinical Examination

A sugar-tong splint is worn with the forearm in neutral rotation for 2 weeks. At that point, the wounds are checked, and an active range of motion program is begun. If necessary, a graduated strengthening program can be started as soon as the motion is satisfactory to allow for sufficient muscle-tendon unit excursion.

Most patients present with a bump on the back of their wrist that had an insidious onset. Examination should assess baseline parameters such as range of motion and grip strength. If the lesion transluminates, it is a fluidfilled cyst. The stability of the scapholunate ligament should also be accessed with Watson’s test. This test is performed by placing the wrist in full ulnar deviation. The examiner then attempts to block flexion of the scaphoid by placing the thumb on its distal pole as the wrist is brought into full radial deviation. A positive test is characterized by pain and a palpable clunk that represents the scaphoid reducing from a subluxated position when the digital pressure is released from the scaphoid tuberosity. A positive test implies attenuation or disruption of the scapholunate interosseous ligament. For patients who present with insidious onset of dorsal wrist pain and no palpable mass, a dorsal wrist syndrome test can reveal either dynamic scapholunate instability or an occult dorsal carpal ganglion.92 This test is performed with the patient’s elbow on the examination table. The wrist is maximally flexed, and the fingers are held straight. The examiner puts downward pressure on the fingers against the patient’s resistance. If this reproduces the pain in the dorsal scapholunate area, it represents irritation of the dorsal wrist capsule or posterior interosseous nerve, which can be indicative of an occult ganglion or dynamic scapholunate instability, or both.

Criteria for Return to Sports

Radiographic Examination

Because structural healing is not an issue with this procedure, an athlete can return to sports as soon as comfort allows. They could potentially be returned sooner with a protective padded sugar-tong splint.

A five-view motion series is helpful in assessing for intercarpal instability. This series is taken in neutral rotation and includes the following radiographs: posteroanterior, lateral, posteroanterior in full ulnar deviation, posteroanterior in full radial deviation, and posteroanterior with a tightly clenched fist. The images are examined for excessive intercarpal angles or interosseous diastasis (see Chapter 20A1). MRI is diagnostic for those patients with an occult ganglion and also allows for assessment of the intercarpal ligaments.

A standard arthroscopy setup is used. The procedure begins with visualization through the 3-4 portal. If visualization allows, I prefer to instrument through the 6-R portal. If not, then it is advisable to begin instrumenting through the 4-5 portal until enough of the inflamed synovium is cleared to improve visualization. Upon completion of the ulnocarpal synovectomy, the scope is switched to the 6-R portal, and the radiocarpal joint is instrumented through the 3-4 portal. A similar approach is used in the midcarpal joint, switching the visualization portal with the instrumentation portal to effect as near a complete synovectomy as possible. Finally, it is necessary to perform a synovectomy of the DRUJ as well. It is advisable to clean out the proximal portion of the joint first so that visualization is improved and inadvertent injury to the TFCC can be avoided when cleaning out the distal aspect of the joint.

Postoperative Care

Dorsal Wrist Ganglion Cysts The pathogenesis of dorsal wrist ganglia remains a subject of debate.88-90 For some reason, the joint-lubricating fluid principally composed of hyaluronic acid is ­overproduced in

Wrist and Hand 1445

Treatment Options

Criteria for Return to Sports

Open ganglionectomy involves excision of the ganglion with a window of joint capsule. Although this is an effective procedure, it puts the SLIOL at risk. The SLIOL is immediately deep to the joint capsule in close proximity to the origin or stalk of the cyst. The ligament cannot be visualized from the outside-in perspective and is adherent to the capsule in this location. Complications such as hypertrophic painful scars and loss of wrist flexion have been also been reported with an open approach. With the arthroscopic approach, up to 42% of patients have additional intraarticular pathology such as perforation or attenuation of the SLIOL identified.93 Additionally, the recurrence rate of less than 1% compares very favorably with a 2% to 40% recurrence rate reported for open ganglionectomy. Finally, capsular disruption and subsequent scarring are minimized with the arthroscopic approach, and the portals if made transversely are generally imperceptible once fully healed.

Patients are allowed to return to sports as soon as comfort allows. A padded splint may be necessary to provide adequate comfort in the first 2 weeks after ­surgery.

Postoperative Care A soft tissue dressing is applied. The wounds are checked at 2 weeks, and activity as tolerated is allowed.

Ulnar Styloid Impaction Syndrome Ulnar styloid impaction syndrome is a condition that results from impaction between the triquetrum and an excessively long ulnar styloid. It is in the differential diagnosis of ulnar-sided wrist pain. Once the diagnosis has been established, this condition reliably responds to ulnar styloid recession.94,95 If left untreated, shear forces produced by the impaction can cause attenuation or disruption of the LT ligament, leading to LT instability. Conversely, patients who had a previous LT arthrodesis are susceptible to the condition because the normal rotational motion of the LT interval, which allows the triquetrum to get out of the way of the styloid with wrist ulnar deviation, dorsiflexion, and forearm supination, has been eliminated.

Author’s Preferred Method A standard wrist arthroscopy setup is used. It is advisable to be careful not to puncture the cyst when instilling saline in the joint because this may obscure the anatomy of the stalk. The procedure is started with the scope in the 3-4 portal. The stalk can often be visualized by careful examination of the dorsal distal scapholunate area. This is seen by the surgeon dropping the hand that holds the scope while the ligament is followed dorsally to the capsular reflection. After the diagnostic portion of the scope is completed, the 6-R portal is established, and the scope is switched to that portal. In most wrists, the dorsal scapholunate area can be visualized after synovectomy with the scope in this

Figure 20C-17   Arthroscopic view of stalk of ganglion cyst emanating from the dorsal scapholunate ligament.

­ ortal. Occasionally, in a tight or very large wrist, it is necp essary to visualize with the scope in the 4-5 portal. About two thirds of the time, an identifiable small, pearl-like stalk is seen (Fig. 20C-17). When such a stalk is not seen, the origin is assumed to be in the dorsal capsular area ­ adjacent to the SLIOL. A full-radius shaver is introduced into the 3-4 portal, and the stalk, if present, as well as a 1 cm window of capsule in this area, is resected (Fig. 20C-18). There are two indicators that sufficient resection has been accomplished. First, the extensor tendons are visualized; and second, the ­ arthroscopy fluid flows freely into the extensor tendon bursa.

Figure 20C-18   Arthroscopic view of extensor tendons viewed through dorsal capsular resection.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1446 DeLee & Drez’s� O

A

B

Figure 20C-19   Ulnar styloid impaction test. A, The examiner begins by dorsiflexing the wrist with the forearm in neutral rotation. B, The forearm is then rotated into supination, which brings the triquetrum into the impaction position with the tip of the ulnar styloid.

History and Clinical Examination Patients present with ulnar-sided wrist pain of insidious onset. Clinically, they appear similar to patients with ulnocarpal impaction syndrome. This can be differentiated with an ulnocarpal impaction test compared with an ulnar styloid impaction test. The ulnocarpal impaction test is performed with the patient’s elbow resting on the examination table. The examiner dorsiflexes the wrist and applies pressure as the forearm is rotated into pronation (Fig. 20C-19). Pronation through the screw-home mechanism of the radius causes a relative lengthening of the ulna at the wrist, revealing impaction between the ulnar head and proximal lunate. The ulnar styloid impaction test is performed in the same way, except that the forearm is rotated into supination, which causes a relative ulnar shortening but also rotates the ulnar carpus toward the styloid, where impaction between the triquetrum and ulnar styloid is revealed. A diagnostic injection of lidocaine, followed by a repeat ulnar styloid impaction test, can be diagnostic.

on the tip of the ulnar styloid and in the proximal pole of the triquetrum.

Treatment Options The original description of this condition advised open recession of the ulnar styloid, with careful preservation of the base of the styloid where the V ligament inserts. Subsequently, it has been demonstrated that arthroscopic recession is safe and effective and facilitates avoiding unnecessary capsular incisions as well as a real-time evaluation of the sufficiency of the recession.9

C

B

A

Radiographic Examination Patients have an excessively long ulnar styloid. This is assessed by the ulnar styloid process index (USPI), which controls for radiographic magnification variations as well as individual bone size and ulnar variance (Fig. 20C-20). The ulnar styloid process index is calculated by subtracting the ulnar variance from the length of the ulnar styloid and dividing by the width of the ulnar head on the neutral rotation posteroanterior view. The average USPI is 0.21 ± 0.07. Anything over this is considered excessively long. Often, these patients have a blunted ulnar styloid tip that opposes a kissing lesion on the proximal triquetrum. MRI reveals increased signal

U.S.P.I . =

C–B A

Figure 20C-20   Illustration of the ulnar styloid process index. (Redrawn from Topper SM, Wood MB, Ruby LK: Ulnar styloid impaction syndrome. J Hand Surg [Am] 22:699-704, 1997.)

Wrist and Hand 1447

Author’s Preferred Method A standard wrist arthroscopy setup is used, and diagnostic arthroscopy is performed. Chondromalacia on the proximal pole of the triquetrum is a common finding, but it is necessary to evaluate the stability of the LT interval from the midcarpal perspective. With the arthroscope in the 4-5 portal, a 3.5-mm bur is introduced into the 6-U portal. By palpation, the bur is placed onto the tip of the ulnar styloid and confirmed with fluoroscopy. Alternatively, the soft tissue on the tip of the ulnar styloid can be cleared with an electrothermal ablative device, which allows for direct visualization. The tip of the ulnar styloid is then ­recessed until there is no impaction on the triquetrum with ulnar deviation and supination. This can be confirmed on ­fluoroscopy.

Author’s Preferred Method Occasionally, it is necessary to dilate a portal to remove a large loose body. This is easily done with the trocar from a 4-mm scope cannula. It is also helpful to use a toothed grasper so that the loose body is not inadvertently released and trapped between the wrist joint capsule and the extensor retinaculum. If that occurs, a larger incision and open dissection will be necessary to safely retrieve the loose body while avoiding damage to the extensor tendons.

Postoperative Care A soft dressing is used. The wounds are kept clean and dry for 2 weeks. Activity as tolerated is allowed. After the sutures are removed, hand therapy is generally not necessary.

Postoperative Care

Criteria for Return to Sports

A short arm splint is worn for 2 weeks. Then the wounds are checked, and the sutures are removed. Postoperative therapy is generally not necessary.

Patients are allowed to return to sports as soon as comfort allows. A padded splint may be necessary to provide adequate comfort in the first 2 weeks after surgery.

Criteria for Return to Sports

Chondral Lesions

Patients are allowed to return to sports as soon as comfort allows. A padded Orthoplast sugar-tong splint may be necessary to provide adequate comfort in the first 2 to 3 weeks after surgery.

Loose bodies in the wrist are a relatively rare occurrence compared with the elbow or knee. They do occur, however, and are assumed to result from direct injury or slough of cartilage flaps from chronic chondromalacia.

Focal chondral lesions in the wrist are rarely the result of isolated blunt trauma. They are more frequently associated with underlying conditions, such as Kienböck’s disease, interosseous ligament laxity, and ulnocarpal impaction syndrome. Viegas97,98 demonstrated an increased incidence of chondromalacia on the proximal pole of the hamate associated with a type 2 lunate, which has a medial hamate facet. He also described a lesion that mimics osteochondritis dissecans involving the scaphoid.99 Several other authors have implicated chondromalacia as a cause of occult wrist pain.2,100,101

History and Clinical Examination

Classification

Patients with loose bodies in the wrist complain of pain and mechanical symptoms such as locking and giving way episodes.96 The onset may be related to trauma but more often has an insidious onset. The examination may reveal swelling, focal pain, and catching or locking while manipulating the wrist through a provocative range of motion.

Outerbridge102 described a classification system for chondromalacia of the patella that applies well to the wrist. Grade 1 describes surface softening of the hyaline cartilage. Grade 2 represents fissures and surface fibrillation. In Grade 3, there are in substance defects of varying depth. Grade 4 describes a full-thickness defect with exposed bone.

Radiographic Examination

History and Clinical Examination

Ossified loose bodies can be seen on plain film radiographs, but this is often difficult because of the superimposition of the carpal bones. Magnetic resonance arthrography is the test of choice when loose bodies are suspected.

The history in these patients can often be elusive because it is rare to have a patient present with a history of isolated blunt trauma that raises the suspicion of focal chondromalacia. More often, these patients present with chronic wrist pain that may have been worked up extensively by other surgeons. The pain tends to be global rather than focal. It is often associated with mild swelling and a history of exacerbated swelling. Careful palpation can often localize the lesion. Provocative examination is focused on identifying an underlying or associated cause.

Loose Bodies

Treatment Options Arthroscopic removal is the treatment of choice and is performed with a standard arthroscopic setup.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1448 DeLee & Drez’s� O

Radiographic Examination These lesions can be difficult to visualize with plain film radiography. Arthrograms are not helpful. Bone scan is a reasonable screening examination that demonstrates increased osteoblastic activity that may or may not be associated with a chondral defect. MRI will effectively demonstrate associated synovitis. The MPGRV view is the most helpful in identifying articular cartilage defects, but with the current resolution capabilities, a negative study does not rule out the possibility of a cartilage defect that can produce symptoms.

Treatment Options If a focal chondral defect is associated with an underlying cause, it is essential to correct that as part of the overall treatment plan. For example, if chondromalacia of the proximal pole of the lunate is associated with ulnocarpal impaction syndrome, the ulnocarpal articulation must be

decompressed in addition to dealing with the chondral defect. Treatment of grade 1 to 3 lesions involves débridement of surface fibrillation and surrounding synovitis. Electrothermal chondroplasty results in chondrolysis and is not recommended. For grade 4 lesions, abrasion chondroplasty remains the only viable arthroscopic alternative. The background basic science and clinical work related to this approach have been accomplished in the knee. 103-105 Exposure of bleeding subchondral bone fills the defect with a clot. More often than not, this transforms into a fibrocartilaginous patch, which at least binds to the adjacent hyaline cartilage and appears to prevent extension of the defect. The ability of “scar cartilage” to bear load and hold up over time remains controversial; however, the fact that the wrist is not a weight-bearing joint is advantageous. There are no hard and fast rules on defect size and the appropriateness of abrasion chondroplasty. Depending on the patient and the size of the defect, carpectomy, limited arthrodesis, or an osteoarticular transfer system (OATS) procedure may be more appropriate.

Author’s Preferred Method Grade 1 to 3 lesions are managed with débridement of fibrillated cartilage and the surrounding synovitis. Often as part of the work-up of chronic wrist pain, a diagnostic injection of the posterior interosseous nerve is done. If this relieves the patient’s pain, I often add a posterior interosseous neurectomy to the débridement. This is accomplished through a 2-cm incision directly over the fourth dorsal compartment just proximal to the level of Lister’s tubercle. The extensor

retinaculum is divided longitudinally over a 1.5-cm length, and the fourth dorsal compartment tendons are retracted ulnarly. Blunt dissection on the floor of the compartment reveals the posterior interosseous nerve. A 1.5-cm section of the nerve is removed and sent to the laboratory for pathologic confirmation and documentation. At this level, the posterior interosseous nerve transmits pure sensory fibers from the dorsal wrist capsule. This neurectomy does not

A

B

C

Figure 20C-21   Arthroscopic images of abrasion chondroplasty.

Wrist and Hand 1449

Author’s Preferred Method—cont’d result in a Charcot joint. Grade 4 lesions are managed by débridement of fibrillated cartilage and surrounding synovitis. Unstable cartilage flaps around the periphery of the lesion are trimmed to stable adherent cartilage with a small arthroscopy knife. I have found that drilling the lesion with

Postoperative Care A short arm splint is worn for the first 2 weeks. At that point, wounds are checked and sutures are removed. For grade 1-3 lesions, a graduated motion and strengthening program is started over the next 2 weeks. For grade 4 lesions, early active wrist motion without strengthening or power grip activities is started over the next month. At 6 weeks, a graduated strengthening program is started.

Criteria for Return to Sports Patients are allowed to return to sports as soon as comfort allows for grade 1-3 lesions. A padded splint may be necessary to provide adequate comfort in the first 2 weeks after surgery. It is not advisable to return grade 4 lesions to sports until the initial healing and remodeling period is over at 3 months.

OSTECTOMY FOR ARTHRITIS “Ectomy” procedures can be accomplished arthroscopically and offer the potential benefits of less capsular scarring, leading to quicker recovery and better postoperative motion.106,107 These potential benefits must be weighed against longer operative times and the risk for inadvertent injury to adjacent articular cartilage and extra-articular structures.

Radial Styloidectomy

a 4.5 K-wire is an effective way to expose bleeding cancellous bone (Fig. 20C-21). It is quicker and more efficient than a bur and therefore exposes the bone to less thermal damage. Additionally, the chondral picks that are available for the knee are too large for most of these lesions.

(>60 degrees) will be evident on the neutral rotation lateral. In cases of SNAC, the scaphoid nonunion will be evident. Certain malunions of a radial styloid fracture are also amenable to treatment with radial styloidectomy.

Author’s Preferred Method A standard wrist arthroscopy setup is used. After synovectomy, the long radiolunate ligament is identified. While visualizing through the 3-4 portal, a 3-mm bur is introduced through the 1-2 portal. Starting at the radial border of the long radiolunate ligament, the radial styloid is ­resected to the point that it is level with the remainder of the scaphoid facet of the radius. The capsular attachments on the far radial aspect of the styloid are thick, and it is necessary to débride this capsule to expose that portion of the styloid. A retractor such as a small Hohmann or Woodson placed percutaneously is helpful to facilitate this portion of the procedure. Occasionally, I use a small osteotome placed percutaneously to get portions of the styloid that are not efficiently accessible with the bur. Adequate resection is confirmed with fluoroscopy.

Postoperative Care

Radial styloidectomy is indicated as a pain-relieving procedure in certain patients with stage 1 scapholunate advanced collapse (SLAC) or early scaphoid nonunion advanced collapse (SNAC) who have maintained reasonable wrist motion and strength and wish to avoid a more involved or complicated salvage procedure.

A short arm splint is worn for the first 2 weeks after surgery. The wounds are then checked, and the sutures are removed. At that point, early active motion is encouraged, and if necessary, a graduated strengthening program is started.

History and Clinical Examination

Criteria for Return to Sports

These patients present with a remote history of injury and slowly progressive radial-sided wrist pain. The examination is characterized by a positive radial carpal grind, which is performed with the wrist in full radial deviation while the ­examiner flexes and extends the joint. The patient will have pain with palpation of the radial styloid and may also have swelling or an associated ganglion cyst. If the underlying cause is a scapholunate ligament disruption, Watson’s test is also positive.

Patients are allowed to return to sports as soon as comfort allows. A padded splint may be necessary to provide adequate comfort in the first 2 to 4 weeks after surgery.

Radiographic Examination Plain film radiographs reveal a loss of cartilage space between the distal pole of the scaphoid and the radial styloid. In cases of SLAC, an elevated scapholunate angle

Proximal Row Carpectomy Proximal row carpectomy is a motion-preserving salvage procedure indicated for stage 1-2 SLAC wrist and in certain cases of SNAC wrist. For this procedure to be a viable option, the cartilage on the proximal pole of the capitate and lunate facet of the radius must be healthy, and this is confirmed arthroscopically. The classification, history, examination findings, and radiographic evaluation are discussed elsewhere in this chapter.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1450 DeLee & Drez’s� O

Author’s Preferred Method A standard wrist arthroscopy setup is used. After synovectomy and removal of loose debris, the scope is switched to the 6-R or 4-5 portal, and a 4-mm bur is introduced into the 3-4 portal. The proximal pole of the scaphoid is then removed. This creates working space for removal of the lunate. The proximal pole of the capitate is protected with a Freer elevator placed through the radial midcarpal portal, and the lunate facet of the radius is protected by careful technique. It is also helpful to morcellize the bone with small osteotomes and remove the larger fragments with a toothed grasper or a pituitary rongeur. Once the lunate is removed, the bur is introduced into the 6-R portal, and the scope is switched to a more radial portal, and then the triquetrum is removed. In most cases, it is necessary to perform a radial styloidectomy as well as to prevent radial impingement.

Postoperative Care It is not necessary to pin the capitate to the lunate facet of the radius. The capitate is simply allowed to descend into the lunate facet, and the wrist is splinted for 4 weeks. At that point, a gentle range of motion program is started with intermittent splint wear. After 8 weeks, the splint is discontinued, and a graduated strengthening program is started.

Criteria for Return to Sports This is a salvage procedure, so it is unlikely that any athlete who substantially uses the upper extremity will return to effective competition. For the weekend athlete, such as a club golfer or tennis player, it is advisable not to return until the initial healing and remodeling period is complete 3 to 6 months after surgery.

C

r i t i c a l

P

o i n t s

The wrist is a significantly smaller and more complex joint than the knee or shoulder, so traction, small joint scopes, and instruments are required. l It is unsafe to plunge a No. 11 blade directly into the wrist joint; therefore, the practice of a dermal incision only must be mastered. l When learning the myriad of provocative examination maneuvers for the wrist, it is important to remember to use the contralateral side for comparison. l Retraction is a vital facilitator of open surgery and, where possible, should be used in small joint arthroscopy to protect vital neurovascular structures that are in proximity. l Unlike the shoulder, the wrist can be immobilized postoperatively, making the use of electrothermal technology more feasible. In a joint that does not have enough room to accomplish suture placation of attenuated tissue, this technology may find a role. l

S U G G E S T E D

R E A D I N G S

Abrams RA, Petersen M, Botte MJ: Arthroscopic portals of the wrist: An anatomic study. J Hand Surg [Am] 19:940-944, 1994. Berger RA, Landsmeer JM: The palmar radiocarpal ligaments: A study of adult and fetal human wrist joints. J Hand Surg [Am] 15:847-854, 1990. Cooney WP: Evaluation of chronic wrist pain by arthrography, arthroscopy, and arthrotomy. J Hand Surg [Am] 18:815-822, 1993. Geissler WB, Freeland AE, Weiss AP, Chow JC: Techniques of wrist arthroscopy. Instr Course Lect 49:225-237, 2000. Koman LA, Poehling GG, Toby EB, Kammire G: Chronic wrist pain: Indications for wrist arthroscopy. Arthroscopy 6:116-119, 1990. Osterman AL, Raphael J: Arthroscopic resection of dorsal ganglion of the wrist. Hand Clin 1995, 11:7-12. Palmer AK: The distal radioulnar joint: Anatomy, biomechanics, and triangular ­fibrocartilage complex abnormalities. Hand Clin 3:31-40, 1987. Palmer AK: Triangular fibrocartilage complex lesions: A classification. J Hand Surg [Am] 14:594-606, 1989. Topper SM, Nagle D, Savoie FH, Jordan B: Application of electrothermal technology in arthroscopy of the hand and wrist. Hand Surg Q Summer:9-14, 2002. Whipple TL: The role of arthroscopy in the treatment of wrist injuries in the ­athlete. Clin Sports Med 1998, 17:623-634.

R eferences Please see www.expertconsult.com

C H A P TE R  

21

Hip, Pelvis, and Thigh S e c t i o n

A

Hip and Pelvis Agam Shah and Brian Busconi Historically, injuries involving the hip, pelvis, and thigh have been difficult to diagnose, challenging to treat, and associated with poor outcomes. There are several factors that contribute to these conditions. First, injuries within this region occur far less frequently than injuries in other areas of the body. Consequently, the investigative skills, including proper history taking and physical examination of hip and pelvis pathology, have lagged behind more traditional injuries. The problem becomes compounded when considering the variety of soft tissue, bone, and ligamentous structures that are located in this region. Injuries to these tissues often arise from similar histories and are difficult to distinguish on physical examination. Lastly, treatment modalities for these injuries have been limited for several reasons, including difficulty accessing the hip joint, disagreement about the diagnosis, and poor evidence-based reports of treatment outcomes. More recently, however, there has been a renewal of interest in this area as diagnostic testing has improved and newer treatment methods have been developed. As with most orthopaedic injuries, accurate diagnosis is made by obtaining an appropriate history, injury-­specific physical examination, and diagnostic imaging. Injury specifics that should be obtained during the history are presented in Box 21A-1. The location of pain is extremely helpful in developing a differential diagnosis (Fig. 21A-1). The physical examination of the hip and pelvis should be carried out systematically. Bilateral range of motion and strength testing should be documented. A thorough neurovascular examination should be performed. Specialized maneuvers to rule out specific conditions are helpful ­(discussed later.) Nonmusculoskeletal pathology should also be considered and ruled out, including hernia and lumbar disk herniation. Plain radiographs of the area in question should be the first imaging modality, with a minimum of two orthogonal views. Further imaging should be diagnosis specific or used to qualify and quantify the magnitude of damage. The spectrum of treatment for hip, pelvis, and thigh injuries has changed. The physician-patient relationship has broadened to incorporate coaches, trainers, and therapists. Treatment is more activity specific, with a focus on return to play as well as prevention of injury. As these injuries are becoming more common, stepwise protocols are

Box 21A-1 I nformation Obtained from Patient History Character and location of pain Mechanism of injury (be specific if possible) Duration of symptoms Activity-related pain (Does it subside with rest?) Pain related to bowel or bladder activity or ingestion of food Pain related to menses Treatment history (injections, physical therapy, other physicians)

being formulated and improved upon to allow athletes to successfully return to play. Intra-articular hip pathology is a relatively new frontier in sports medicine. Enhanced imaging techniques and the increasing popularity of the magnetic resonance imaging (MRI) arthrogram have vastly improved the precision with which intra-articular lesions are being detected. Moreover, the ability to treat these lesions has also improved. Advances in hip arthroscopy, including instrumentation and technique, have allowed surgeons to successfully treat patients with pathology that was previously considered inaccessible.

JOINT ANATOMY AND BIOMECHANICS Pelvis The human pelvis consists of three different parts: the sacrum, the coccyx, and the innominate bone. The sacrum lies in the posterior central portion of the pelvis and is composed of the five terminal vertebrae. The coccyx is distal to and continuous with the sacrum. Three rudi­ mentary vertebrae are fused to form the coccyx, which serves as the bony floor of the pelvic outlet. An innominate bone is located on either side of the sacrum and coccyx. The innominate bone has three parts: the ilium, 1451

1452 DeLee & Drez’s Orthopaedic Sports Medicine

Box 21A-2 Functions of the Pelvis

1. Protects the internal viscera, including reproductive organs, bladder, and lower gastrointestinal tract 2. Serves as an attachment point for multiple muscles of the trunk, back, and lower extremities 3. Allows for transfer of load between the lower ­appendicular skeleton and axial skeleton 4. Plays a critical role in balanced ambulation

Figure 21A-1  Location can help guide diagnosis. 1, Intra-articular pain—osteoarthritis, labral pathology. 2, Pain from lateral hip structures—trochanteric bursitis. 3, Pain from symphyseal structures—osteitis pubis. 4, Pain from abdominal structures—athletic hernias.

the ischium, and the pubis. The ilium is located posterolaterally and articulates with the sacrum at the sacroiliac joint. The ilium gives rise to a broad expansion termed the iliac crest. The anterior and posterior borders of the iliac crest are bounded by the anterior superior iliac spine and posterior superior iliac spine, respectively. The ischium is located posteroinferiorly and is bordered by the ilium superiorly and the pubis anteriorly. The ischium has two bony prominences: the ischial tuberosity and ischial spine. The sacrospinous and sacrotuberous ligaments arise from these bony processes and serve as a strong attachment site between the sacrum and the ischium. The pubis forms the third part of the innominate bone and is found in the anterior aspect of the pelvis. The pubis is attached to its contralateral counterpart anteriorly at the pubic ­symphysis. The posterior aspect of the pubis is continuous with the ilium superiorly and the ischium inferiorly. All three bones meet at the center of the acetabulum. The acetabulum is a dome-shaped structure that approximates a hemisphere. It is situated distally and laterally relative to the center of the pelvis and is covered by articular cartilage. On average, the acetabulum is 35 degrees abducted and 20 degrees forward flexed. Functions of the pelvis are listed in Box 21A-2.

Femur The proximal femur comprises the bony elements of the thigh and articulates proximally with the pelvis at the hip joint. The femur is a diaphyseal structure that broadens proximally to form the trochanteric ridge. The ridge expands laterally to form the greater trochanter and medially to form the lesser trochanter. The femoral neck arises proximal to the trochanteric ridge. It ascends medially and is

continuous with the femoral head. The neck forms an average angle with the shaft of about 130 degrees (neck-shaft angle) and is found to have an average anteversion angle of 15 degrees. Deviations from normal anatomic geometry can lead to improper mechanics and injury, such as stress fracture.1,2 The femoral head forms the most proximal portion of the femur. Roughly, it forms two thirds of a sphere and is covered with articular cartilage. A small defect in the cartilage in the center of the femoral head, termed the fovea capitis, allows for the attachment of the ligamentum teres. The artery of the ligamentum teres, the obturator artery, inferior gluteal artery, lateral femoral circumflex, and intramedullary arteries all contribute to the blood supply of the femoral head, but most of the blood supply comes from the lateral ascending intracapsular branches of the medial femoral circumflex artery. Trauma or occlusion of these branches may lead to avascular necrosis of the femoral head.

Hip Joint The hip joint is a highly constrained ball-and-socket joint (Table 21A-1; Figs. 21A-2 and 21A-3). The hip joint is limited predominantly by its bony configuration. Additionally, the joint is surrounded by three strong ligaments: the pubofemoral ligament inferiorly, the ischiofemoral ligament posteriorly, and the iliofemoral ligament anteriorly (the ligament of Bigelow.) These three ligaments integrate with the surrounding connective tissue to form the hip joint capsule. Because of the anatomic configuration of these ligaments, the capsule becomes taught with extension and internal rotation and loose with flexion and external rotation. The capsular attachments are at the intertrochanteric line on the femur and, by a few millimeters, superior to the acetabular rim. The labrum, ligamentum teres, and fat pad are all intraarticular structures within the hip joint. The labrum is a fibrocartilaginous structure that is attached to the acetabular margin. It is continuous with the transverse acetabular ligament inferiorly. The labrum effectively deepens the hip socket and contributes to the normalization of joint reactive   TABLE 21A-1  Average Hip Range of Motion Flexion Abduction Adduction Internal rotation External rotation

115 degrees 50 degrees 30 degrees 45 degrees 45 degrees

Hip, Pelvis, and Thigh 1453

Most investigators would agree that limited weight-­bearing places the least amount of force across the hip joint.1

Muscles Please refer to Table 21A-2 and Figure 21A-4 for more details.

Neurovascular Structures

Figure 21A-2  Intra-articular structures of the hip. (From Busconi B: Anatomy. In McCarthy J [ed]: Early Hip Disorders: Advances in Detection and Minimally Invasive Treatment. New York, Springer-Verlag, 2003, p 52.)

forces across the hip joint. Although the labrum’s function remains controversial, studies show that the labrum does not participate in direct load transmission.3 The ­ligamentum teres is a strong connective tissue structure that is found between the inferior portion of the acetabular fossa and the fovea capitis of the femoral head. The artery of the ligamentum teres lies within this structure. There is a small fat pad within the acetabular fossa, and when combined with the ligamentum teres, it forms the pulvinar. Intra-articular pathology can exist at the level of the articular cartilage, the bony anatomy, the labrum, and the ligamentum teres. During activities of daily living, the hip joint will experience a joint reactive force 1.6 times body weight.4 The force is found to be greatest when rising from a seated position. Strenuous activity has been shown to produce joint reactive forces exceeding 3 to 6 times body weight.4 Abductor tension has been found to cause the join reactive force to be 1.5 times body weight on the ipsilateral side during non–weight-bearing standing and ambulation.5

The hip joint and most of the surrounding structures derive innervation from the lumbosacral plexus. The L2-S1 nerve roots combine to form several exiting nerves, including the sciatic, femoral, and obturator nerve. These nerves cross the hip joint and contribute to the innervation of the hip capsule, with a large contribution from the L3 root. Irritation of the capsule will produce referred pain to the L3 dermatome, including the medial thigh. The obturator nerve also provides innervation to the knee joint, and patients may sometimes experience referred knee pain with intra-articular hip pathology. The anterior aspect of the hip has several important neurovascular structures (Fig. 21A-5). The femoral triangle, bordered by the inguinal ligament, sartorius, and pectineus, contains three structures: lymphatic vessels, the femoral vein, and the femoral artery. The lymph vessels drain the anterior thigh, portions of the groin, and portions of the lower leg. The femoral vein drains directly into the external iliac vessel. The femoral artery is a continuation of the external iliac artery as it passes under the inguinal ligament. Major branches of the femoral artery include the medial and lateral femoral circumflex and femoral profunda arteries. Lateral to the femoral triangle, the femoral nerve exits deep to the inguinal ligament to supply the anterior muscles of the thigh. Formed by the L2-L4 nerve roots in the lumbar plexus, the femoral nerve courses along the iliacus muscle and alongside the psoas, descends under the inguinal ligament, innervates the anterior thigh musculature, and provides cutaneous innervation to the lower leg through its terminal branch of the saphenous nerve.

Figure 21A-3  Ligaments surrounding the hip joint. (From Busconi B: Anatomy. In McCarthy J [ed]: Early Hip Disorders: Advances in Detection and Minimally Invasive Treatment. New York, Springer-Verlag, 2003, p 52.)

1454 DeLee & Drez’s Orthopaedic Sports Medicine   TABLE 21A-2  Muscles Surrounding the Hip Joint Name of Muscle

Origin

Insertion

Nerve Supply

Action

Distinguishing Characteristics

Anterior superior iliac spine

Anteromedial surface of proximal tibia

Femoral nerve

Flexes, abducts, ­laterally rotates thigh at hip joint

Longest muscle in body; one third of pes anserinus

Tensor fascia lata

Iliac crest

Iliotibial band, then Gerdy’s tubercle

Superior gluteal nerve

Can be responsible for ­external snapping hip

Gluteus medius

Outer surface of ilium

Greater trochanter

Superior gluteal nerve

Assists in extending the knee joint and abducting the hip Abducts hip joint

Surface of ilium, sacrum, and coccyx

Gluteal tuberosity of femur and iliotibial band

Inferior gluteal nerve

Extends and ­externally rotates the hip and extends the knee

Termed the “pelvic deltoid” by Henry; largest muscle in body

Gracilis

Inferior ramus

Obturator nerve

Body of pubis

Abducts hip and flexes knee Adducts hip

One third of pes anserinus

Adductor longus

Anteromedial surface of proximal tibia Linea aspera of femur

Lumbar plexus

Flexes hip and ­extends knee Flexion, adduction, and external rotation of hip joint

Only quadriceps muscle that crosses the hip joint Associated with internal snapping hip syndrome

Superficial Anterior Group

Sartorius

Lateral Group

Posterior Group

Gluteus maximus

Medial Group

Obturator nerve

Injury to this muscle leads to Trendelenburg gait

Deep Muscles Anterior Group

Rectus femoris

Anterior inferior iliac spine and ilium Iliac fossa and T12 to L5 vertebral bodies

Quadriceps tendon into patella Lesser trochanter

Surface of ilium

Greater trochanter

Superior gluteal nerve

Abduction of hip joint

Piriformis

Anterior sacrum

Greater trochanter

Obturator internus

Greater trochanter

Quadratus femoris

Obturator membrane and spine of ischium Ischial tuberosity

1st and 2nd sacral nerves Sacral plexus

Superior gemellus Inferior gemellus

Externally rotate the hip joint Externally rotate the hip joint Externally rotate the hip joint Externally rotate the hip joint Externally rotate the hip joint

Iliopsoas

Lateral Group

Gluteus minimus Posterior Group

Medial Group

Adductor brevis Adductor magnus

Femoral nerve

Spine of ischium

Quadrate tubercle of femur Greater trochanter

Sacral plexus

Ischial tuberosity

Greater trochanter

Sacral plexus

Inferior ramus of pubis Inferior ramus of pubis and ischial tuberosity

Linea aspera of femur Posterior femur and adductor tubercle

Obturator nerve Obturator and sciatic nerve

The lateral femoral cutaneous nerve of the thigh exits under the inguinal ligament near the anterior superior iliac spine. This nerve originates from the L2-L3 nerve roots and provides cutaneous innervation to the lateral aspect of the thigh. Posterior neurovascular structures that need to be considered include the sciatic nerve, inferior gluteal nerve and artery, and pudendal nerve.6,7 The sciatic nerve is by roots L4-S3. The nerve courses laterally through the pelvis, exits at the greater sciatic foramen, travels distally deep to the piriformis and superficial to the remaining short external rotators, and continues into the thigh to provide innervation to the muscles of the posterior thigh and leg.

Sacral plexus

Adduct the hip joint Adduct the hip joint

Most proximal of the short external rotators Most distal of the short ­external rotators

Muscle has dual innervation

The nerve divides along the course of the posterior thigh into the common peroneal trunk and the tibial trunk. The common peroneal trunk lies more laterally and is more likely to be injured during approaches to the hip and thigh. The inferior gluteal nerve, a branch of the sacral plexus, leaves the pelvis through the lower part of the greater sciatic foramen, below the piriformis, and terminates by innervating the gluteus maximus. The inferior gluteal artery, a branch of the internal iliac artery, travels alongside the inferior gluteal nerve. This artery can be injured during approaches to the hip and can be a source of significant bleeding. The pudendal nerve arises from the roots S2-S4 and exits the pelvis from the greater sciatic foramen. The

Hip, Pelvis, and Thigh 1455

Figure 21A-4  Muscle insertions and attachments around the pelvis. (From McCarthy J, Busconi B, Owens B: Assessment of the painful hip. In McCarthy J [ed]: Early Hip Disorders: Advances in Detection and Minimally Invasive Treatment. New York, Springer-Verlag, 2003, p 3.)

nerve then reenters the pelvis through the lesser sciatic foramen and courses along the floor of the pelvis to provide innervation to the perineum. Bicyclists have been reported to sustain compression neuropathies of this nerve.7

ADULT INJURIES Please see Box 21A-3 for an outline of specific injuries.

Trochanteric Bursitis

Soft Tissue Injuries Bursitis Bursae are connective tissue sacs filled with fluid. They serve to cushion the places where tendons, ligaments, and muscles move over bones and help prevent or decrease Superficial circumflex iliac a. and v. Anterior superior iliac spine Lateral femoral cutaneous n. Gluteus medius m. Femoral n. Iliopsoas m. Tensor fasciae latae m. (retracted) Sartorius m. (cut)

friction between surfaces that move in opposite directions. Bursitis is caused by repetitive soft tissue friction with nearby muscle or by trauma with chronic inflammation to the bursa. Because the symptoms are exacerbated by activity and are location specific, it can be difficult to differentiate this condition from other musculoskeletal processes such as tendinitis.

The trochanteric bursa lies on the lateral aspect of the greater trochanter and serves to protect the underlying bone from movement of the iliotibial band and gluteus maximus muscle. This bursa can become inflamed from a variety of different sources, including trauma and ­repetitive motion. Trochanteric bursitis occurs more often Figure 21A-5  Anterior neurovascular structures. (From Busconi B: Anatomy. In McCarthy J [ed]: Early Hip Disorders: Advances in Detection and Minimally Invasive Treatment. New York, Springer-Verlag, 2003, p 53.)

Superficial epigastric a. and v.

Femoral sheath Superficial external pudendal a. and v. Pectineus m. Deep external pudendal a. and v. Lateral and medial circumflex femoral a. Deep femoral a. Femoral a. and v.

1456 DeLee & Drez’s Orthopaedic Sports Medicine

Box 21A-3 Outline of Specific Hip Injuries Soft tissue injuries Bursitis Trochanteric Ischial Iliopsoas Iliopectineal Snapping hip syndrome Contusions Iliac crest Quadriceps Groin Myositis ossificans Strains Adductor Iliopsoas External oblique Hamstring Quadriceps Sacroiliac sprain Hernias Inguinal hernia Femoral hernia Sports hernia (athletic pubalgia) Bone injuries Traumatic fractures Dislocation Stress fractures Pelvic Sacral Femoral neck Osteitis pubis Osteonecrosis Degenerative joint disease Nerve entrapment injuries Sciatic Obturator Pudendal Ilioinguinal Femoral Lateral femoral cutaneous Intra-articular pathology Labral tears Femoral acetabular impingement Loose bodies Chondral injuries Ruptured ligamentum teres Synovial disease

in females than males.8 In athletes, causes include a tight iliotibial band, which may be a precursor to external snapping hip syndrome.9 Athletes with wide pelvises or excessive foot pronation are predisposed to trochanteric bursitis.8 Patients with leg-length discrepancy or runners that train on baked surfaces or road cambers are also at increased risk.8 Repetitive trauma, such as the continuously falling ice skater, may cause chronic bursitis.10 Also, prior surgical incisions located over the ­trochanter should alert the physician to bursitis as a possible diagnosis.

Evaluation Clinical Presentation and History

The clinical symptoms are varied. All patients complain of aching over the trochanteric area and lateral thigh. This pain may be acute at the onset, or it may build up gradually over time, sometimes lasting for many months or even years. In chronic cases, discomfort may be vague, and it may be difficult to describe its exact location. Some patients may have difficulty walking or walk with a limp. The pain experienced with trochanteric bursitis can radiate distally and must be distinguished from that associated with lumbosacral spine disease. Other neuromuscular symptoms must be ruled out through history and physical examination. Physical Examination and Imaging

On physical examination, the patient will have direct tenderness to palpation over the greater trochanter. Pain is exaggerated by resisted abduction and external rotation. Pain is also elicited by asking the patient to stand from a squatted position with the hips internally rotated. Because trochanteric bursitis can be secondary to other hip pathology, a thorough examination should be performed to rule out other possible primary hip pathology. Imaging for trochanteric bursitis is limited. Plain radiographs are most often normal but may sometimes show an irregular lateral greater trochanteric border that is consistent with chronic bursitis. MRI shows increased fluid signal in the area of the iliotibial band (Fig. 21A-6). Treatment

Treatment of trochanteric bursitis should proceed in a stepwise fashion. Despite the high prevalence of this condition, there are no controlled studies in regard to treatment

Figure 21A-6  Magnetic resonance image showing edema surrounding the iliotibial band.

Hip, Pelvis, and Thigh 1457

­ rotocols. Any offending activities should be removed p from the patient’s training program. A physical therapy program focused on gluteal strengthening, iliotibial band stretching, and proper body mechanics should be implemented.10 Oral anti-inflammatories and steroid injection should be considered.11 Athletes may return to play once symptoms have resolved. Surgical intervention should be considered for patients with refractory symptoms. Bursectomies, bone débridement, and tendon release haven been attempted.11 One study reported excellent success rates and patients returning to full activity after longitudinal release of the iliotibial band and excision of the ­trochanteric bursa.12

Ischial Bursitis Also known as ischial gluteal bursitis, ischial bursitis is an inflammation of the bursa that lies between the ischial tuberosity and the hamstring origin. Injury can result from prolonged sitting, direct trauma to the area, or repetitive microtrauma to the area from activities such as cycling or running.10 Hamstring injuries can lead to hematoma and scar formation, which can then cause irritation of the bursa. Evaluation

Ischial bursitis causes pain at the level of the tuberosity and radiates down the posterior aspect of the thigh. Pain is exacerbated by hip flexion or sitting. The pain can be confused with sciatica or hamstring syndrome, a compressive neuropathy of the sciatic nerve caused by compression of the musculotendinous portion of the hamstrings.13 Treatment

Treatment of ischial bursitis is similar to that of other types of bursitis. Rest, avoidance of aggravating activities, and oral anti-inflammatories have been shown to accelerate the healing process. Steroid injection into the tuberosity has been advocated but should be approached with caution because it may place the hamstring tendon origin at risk.9 Operative treatment with removal of the bursa and

A

B

removal of bone spurs have been suggested, but success rates are yet unpublished.14

Iliopsoas Bursitis The iliopsoas bursa lies between the musculotendinous portion of the iliopsoas muscle and the pelvic brim. It is the largest synovial bursa in humans and can extend into the hip joint. The three main causes of iliopsoas bursitis are rheumatoid arthritis, acute trauma, and overuse injury.15 It is thought that aggravation of this bursa occurs from rapid lengthening and contracting of the iliopsoas muscle over the pelvic brim. It occurs in track and field athletes, uphill runners, and rowers.15 Some investigators consider this entity to be a precursor to snapping hip syndrome (see later).15 Evaluation

Patients with iliopsoas bursitis present with inguinal pain and an audible or palpable snap that can be reproduced by aggravating activities. Physical examination findings include tenderness to palpation, tenderness in the femoral triangle, weak external rotation strength in hip flexion, a positive Thomas test, and a positive snapping hip sign. Plain radiographs are usually normal. Dynamic ultrasound may reveal snapping of the iliopsoas tendon over the pelvic ridge.16 MRI shows fluid within the bursa.17 Bursography under fluoroscopy with injection of contrast, lidocaine, and steroid is an excellent method for diagnosis as well as treatment (Fig. 21A-7).18 Treatment

Treatment of iliopsoas bursitis should begin with conservative measures, including a program of rest, physical therapy for iliopsoas stretching, and oral anti-inflammatories. If these measures fail, steroid injection should be attempted under fluoroscopy or ultrasound. Steroid injection has been shown to provide permanent relief in 50% of the population and 2 to 8 months of relief otherwise.18 If ­conservative measures fail, surgery should be considered. Surgical treatments include partial release and lengthening of the

C

Figure 21A-7  A to C, Bursography to help diagnose and treat iliopsoas bursitis.

1458 DeLee & Drez’s Orthopaedic Sports Medicine

iliopsoas tendon, with 50% to 90% success rates.15 Complications of surgery include hip flexion weakness, hematoma, skin paresthesia, and recurrence of symptoms.19

Iliopectineal Bursitis The iliopectineal bursa can exist as a separate bursa or a continuation of the iliopsoas bursa coursing over the iliopectineal eminence. Inflammation of this bursa causes anterior hip pain that can be severe enough to cause a limp. The symptoms are relieved by flexion and external rotation of the hip. The pain may be related to snapping of the iliopsoas tendon over the iliopectineal eminence.20 Stretching of the iliopsoas tendon by hip extension will usually worsen the discomfort. Treatment consists of rest, oral anti-inflammatories, iliopsoas stretching, and possible steroid injection.15 With recalcitrant cases, operative excision of the offending bursa and lengthening of the iliopsoas tendon has been described, but only as a final resort.15

Snapping Hip Syndrome Classification The snapping hip syndrome, also described as coxa saltans, is an entity that can be caused by several different pathologic processes. All patients with snapping hip syndrome have a reproducible audible or palpable “snap” during particular movements of the hip joint. The etiology of coxa saltans has been broadly categorized into external snapping hip syndrome, internal snapping hip syndrome, and intraarticular snapping hip syndrome. Both internal and external snapping hip syndrome usually occur in patients who are in their early 20s and late teens.20 Of the three categories, external snapping hip is the most common and is most often caused by thickening of the iliotibial band or gluteus maximus snapping over the greater trochanter.21 The thickening is believed to be caused by microtrauma to the tendons. Patients with external snapping hip will also have inflammation within the trochanteric bursa. Internal snapping hip is most often caused by the iliopsoas snapping over the iliopectineal eminence of the pelvis or over the femoral head.21 Ballet dancers have a proclivity for this syndrome.22 Intra-articular causes of snapping hip can be labral tears, loose bodies, or osteochondral injuries. Most patients with intra-articular causes have a history of trauma.

Evaluation The diagnosis of snapping hip syndrome is made by history and physical examination. Patients describe a snap that usually can be reproduced. The history is generally insidious, with onset of pain before snapping over the region of the snap. Localization of the pain can help distinguish external from internal causes. Physical examination helps confirms the diagnosis and may give insight as to the underlying cause. Patients with external snapping hip have the greatest amount of tenderness over the greater trochanter, and the pain can be reproduced with repetitive flexion and extension.21 The snap is best reproduced while the patient is standing. Ober’s test is helpful in establishing a diagnosis and is performed with the patient in the lateral

decubitus position with the affected hip up. With the knee flexed and the hip maximally extended, the hip is brought from a maximally abducted position to an adducted position. If the hip can be adducted beyond midline, the test is negative. If the patient’s hip cannot be adducted beyond the midline, the test is considered positive and indicative of a contracted iliotibial band.23 Internal and intra-articular causes of snapping hip cause pain along the inguinal crease and medial thigh. In the case of internal snapping hip, the snap can be reproduced with the patient supine and flexing and extending the hip. The snapping occurs when the iliopsoas is suddenly forced under tension over the iliopectineal eminence or the femoral head. The snap can be made more obvious if the hip is ranged from an abducted, externally rotated, and flexed position to an adducted, internally rotated, and extended position. This motion moves the iliopsoas tendon from a position lateral to the femoral head to a position medial to the femoral head and is associated with a very loud clunk. Gentle pressure over the femoral head prohibits the snap from occurring.21 Intra-articular injuries should be considered when external and internal causes have been ruled out. Patients generally have pain in the groin area and a history of trauma. Often patients have clicks on ranging of the hip, but the clicks are not necessarily reproducible. Most likely causes include loose bodies, labral tears, and osteochondral lesions.24

Imaging Imaging for patients with snapping hip syndrome can be helpful. Plain radiographs are often normal but may help identify intra-articular loose bodies or bone spurs on the greater trochanter. Coxa vara is a common finding in patients with snapping hip.25 Dynamic ultrasound may show internal or external snapping, but sensitivity is technician dependent. Bursography is a useful means of distinguishing between intra-articular and internal causes.20 The iliopsoas bursa is injected with contrast dye, and the hip is ranged under fluoroscopy. In the case of internal snapping hip syndrome, a sudden jerk of the iliopsoas muscle over the bursa is diagnostic.26 Along with contrast dye, the bursa may be injected with lidocaine and steroid to provide concomitant treatment. MRI with and without contrast is an excellent methodology for diagnosing intra-articular lesions. (See the sections on intra-articular pathology for specific lesions.)

Treatment Options Nonoperative

Treatment for snapping hip is pathology specific. Conservative management for internal and external causes is implemented first. Intra-articular causes are mechanical and generally are not amenable to conservative management, including oral anti-inflammatories, physical therapy, and avoidance of aggravating activities. Injections of corticosteroid can be attempted. The complete resolution of symptoms may take up to 1 year.21 Surgical treatment should be considered if conservative management does not provide symptom relief.

Hip, Pelvis, and Thigh 1459

Operative

Several different methods of surgical management have been described for treating snapping hip syndrome. The cause of the snap must be addressed. The goal of treating external snapping hip is to release tension of the iliotibial band as it crosses over the greater trochanter.27 This can be accomplished by a Z-plasty lengthening techinique27 or an elliptical incision with bursa removal.28 Both procedures have shown mixed results, with patients complaining of continued pain over the incision and return of the snapping sensation secondary to scarring. Internal snapping hip syndrome can be surgically addressed through surgical release of the iliopsoas tendon or fractional lengthening of the iliopsoas tendon.29 In this procedure, partial tenotomies of the posterolateral aspect of the iliopsoas tendon are performed at 2-cm intervals beginning 1 cm proximal to the insertion on the lesser trochanter.21 Postoperative management of these patients is patient specific with limits on active flexion. There are no range of motion or weight-bearing restrictions. Return to play is generally allowed after 3 to 6 months. Complications from the operation include motor and sensory changes, bursa swelling, hematoma formation, and infection. Endoscopic tendon release has recently been described with excellent short-term results, but no comparative data.30 Treatment of intra-articular causes of snapping hip is lesion specific and amenable to arthroscopic management.

Contusions Contusions are the most common injury to the hip, thigh, and pelvis. Collisions with other athletes or falls to the ground are the most common cause. Contusions can be superficial and limited to the subcutaneous tissue, or they can be deep and involve the bone, muscle, and ligaments. When associated with muscle involvement, contusions can result in a slow bleed with significant hematoma formation. Depending on the depth of contusion and extent of injury, symptoms may occur immediately or after a 24- to 48-hour delay. Often, particularly with superficial contusions, treatment is of short duration and is based on patient’s symptoms. Patients have a rapid return to play. With the diagnosis of deep contusion, however, the treatment period may be extended. Aside from icing, treatment should be delayed 48 hours to ensure that all bleeding has stopped. After 48 hours, treatment with anti-inflammatories, heat, massage, and physical therapy is implemented. Rehabilitation should be aimed at maintaining flexibility and muscle mass. Range of motion should be monitored because patients are prone to developing myositis ossificans after deep contusions involving the muscle.

Iliac Crest Contusion The most common contusion is the contusion to the iliac crest, also know as the “hip pointer.” Patients present with pain over the iliac crest and may have difficulty with ambulation. Patients may present with a fluctuant mass over the area, resulting from hematoma. Plain radiographs are essential to rule out fracture or apophyseal avulsion

in the skeletally immature patient. If left untreated, this condition can lead to periostitis or the formation of bone exostosis. Treatment

Treatment of an iliac crest contusion should proceed in a stepwise fashion, beginning with ice, elevation, and direct compression. Modalities should be aimed at decreased inflammation with anti-inflammatory medications starting at 48 hours. Crutches are prescribed if necessary. If a large hematoma is present, consider immediate aspiration followed by ice and compression. Physical therapy for range of motion and strengthening may be required, depending on the duration of symptoms and amount of deconditioning. Steroid injection to the iliac crest may accelerate the rehabilitation. The patient should be allowed to return to play once full range of motion and full strength have returned. Padding over the injured area will help prevent recurrence or exacerbation.

Quadriceps Contusion Quadriceps contusions generally result from a direct injury to the anterior thigh. Patients will present with pain in the middle of the thigh or localized to the area of maximal impact during injury. Patients may have difficulty ambulating or flexing and extending the knee. Hematoma formation in the muscle can result from trauma to the microvasculature in the muscle. Hematoma development, in turn, can lead to myositis ossificans. Therefore, the major goals of treatment of quadriceps contusion are maximizing flexibility and strength and minimizing hematoma formation.31 Treatment

Once a quadriceps contusion has been diagnosed, treatment should begin with immobilizing the ipsilateral knee in flexion. This serves the purpose of minimizing blood pooling and preventing quadriceps muscle contractures.31 Ice and compressive dressing should be initiated immediately. Depending on the extent of injury, the patient may need to be monitored for compartment syndrome.32 Modalities should be aimed at decreased inflammation, with anti-inflammatory medications starting at 48 hours. Crutches are prescribed if necessary for ambulation. Physical therapy for range of motion and strengthening may be required, depending on the duration of symptoms and amount of deconditioning. Return to play should only be considered once a patient has full range of motion and full strength. Complications of treatment include reinjury, stiffness, and myositis ossificans. Premature return to play can lead to reinjury or quadriceps muscle strain. Myositis ossificans can be a devastating complication of quadriceps contusion, and clinicians must be wary about the occurrence of such a complication (See “Myositis Ossificans”).33,34

Groin Contusions Groin contusions occur usually from a direct blow to the inner thigh and are often seen in soccer players and cyclists. Aside from standard contusion treatment of ice,

1460 DeLee & Drez’s Orthopaedic Sports Medicine

anti-inflammatories, physical therapy, and gradual return to play, the treating clinician should be aware of the possibility of developing vascular complications such as phlebitis and thrombosis. Ultrasound is a useful noninvasive method of diagnosis of vascular complications.

Myositis Ossificans Myositis ossificans is heterotopic ossification in an area of muscle, soft tissue, or disrupted periosteum. Although this disease process can occur without a history of trauma, athletes can usually describe a sentinel event that causes a hematoma formation. The hematoma organizes, and calcium deposits are formed by the body. In a process that is not entirely understood, osteoblasts invade the formed hematoma and begin to make bony spicules.35 This process tends to occur near joints and at tendon origins, but can occur anywhere along the course of a muscle. The process can start as soon as 1 week after injury and can be detected on plain films a minimum of 3 weeks after injury. Patients present with a rapid enlargement within the soft tissues, decreased range of motion, and significant pain 1 to 2 weeks after injury. The patient has swelling and warmth at the site as well as an increased erythrocyte sedimentation rate and serum alkaline phosphatase.36 Any treatment modalities implemented should not promote hematoma formation. Massage and manipulation should be avoided. Therapy should consist of active stretching and strengthening. Passive range of motion should be delayed for at least 3 to 6 months.33 For patients with refractory loss of range of motion, surgery should be considered. Surgery, if warranted, should be delayed for at least 9 to 12 months to allow the lesion to mature. Bone scan can help ascertain lesion maturity. Despite surgical removal, patients and clinicians should be wary that the lesion may recur.

Strains Muscle strains occur as the result of forceful contractions of a stretched muscle. The most frequent cause of this type of injury is eccentric contraction, or muscle contraction during elongation.37 The injury involves partial tears at the myotendinous junction. Chronic strains can occur from repetitive microtrauma to the myotendinous junction. Recently, it has been recognized that muscles that cross two joints are more likely to sustain a strain injury than muscles that cross a single joint. Investigators hypothesize two major reasons why this happens. First, muscles

that cross two joints have been found to be less flexible and experience a greater strain per applied force when compared with single-joint muscles. Second, muscles that cross two joints contain on proportionate average more type II fast-twitch fibers. These fibers can produce violent contractions leading to injury.38 Classification

Muscle strains are classified into three categories: grade I, II, and III. Grade I strains are classically thought of as “muscle pulls” and generally involve less than 5% of the musculotendinous junction. Grade II is considered to be a more significant tear of the musculotendinous unit, but continuity of some portion of the musculotendinous unit does remain. Grade III strains involves complete tear and discontinuity between the muscle origin and insertion.38 Evaluation

Diagnosis is made through careful history and physical examination. The injury mechanism is often one of eccentric contracture of the muscle group involved. In the case of complete rupture, palpation reveals a defect at the musculotendinous junction. If a defect is palpable at the bonetendon interface, the clinician should be alerted to the possibility of tendon avulsion. Imaging, specifically MRI, shows a quantifiable area of injury, but is not predictive of the level of disability.38 Treatment

Treatment of muscle strain injuries should proceed in a stepwise fashion. Table 21A-3 outlines a five-phase protocol proposed by Nuccion, Hunter, and Finerman in the previous edition, which is a modification of Metzmaker and Pappas’ descriptive protocol for treating avulsion fractures.39,40 It is imperative that the regimen be followed sequentially. The program emphasizes a reduction of pain and inflammation, followed by gradual range of motion exercises, and finally strength, flexibility, and endurance. Isokinetic testing is used to assess strength and balance and is helpful in determining when a patient can return to play. Premature return to play can lead to reinjury and further damage.

Adductor Strain The adductor muscle group contains six different muscles: pectineus, gracilis, obturator externus, adductor brevis, adductor longus, and adductor magnus. Violent

  TABLE 21A-3  Rehabilitation Guidelines for Muscle Injuries Phase

Goals

Treatment

Time Frame

I II

Reduce pain, inflammation, and bleeding Regain range of motion

48 to 72 hr 72 hr to 1 wk

III IV V

Increase strength, flexibility, and endurance Increase strength and coordination Return to competition

Rest, ice, and compression; crutches PRN Passive range of motion, heat, ultrasound, electrical muscle stimulation Isometrics, well-leg cycling Isotonic and isokinetic exercises Sport-specific training

1 to 3 wk 3 to 4 wk 4 to 6 wk

From Nuccion S, Hunter D, Finerman G: Hip and pelvis: Adult. In DeLee J, Drez D, Miller M (eds): Orthopaedic Sports Medicine: Principles and Practice, 2nd ed. Philadelphia, WB Saunders, 1994, p 1449.

Hip, Pelvis, and Thigh 1461

contraction of any of these muscles can lead to adductor strain. Despite being a single-joint muscle, the adductor longus is the most commonly strained of the six. The longus has the least mechanical advantage on adduction of the hip and, therefore, undergoes the most strain. Soccer and hockey players are most susceptible to adductor strains. Poor adductor strength and conditioning are risk factors for adductor strains.41 Some authors have concluded that decreased preinjury flexibility leads to muscle strain, but some recent data have suggested that the two are unrelated.42 Evaluation

Adductor strains generally present with a history of forced external rotation of the abducted hip. Patients have pain along the medial border of the thigh. Rarely, a defect is palpable. Passive abduction and resisted active adduction reproduce the symptoms. Plain radiographs are generally negative. MRI shows edema at the myotendinous junction. Treatment

Treatment for adductor strains should follow the fivephase protocol outlined in Table 21A-3. Surgery should be reserved for patients with symptoms that persist longer than 6 months despite appropriate conservative management.43 Adductor tenotomy has been described for chronic adductor strains. Akermark and Johansson performed adductor tenotomy on 16 patients, with improvements in all patients.44 Only 10 athletes, however, were able to return to their previous level of competition. Postoperatively, the patients had decreased strength compared with the normal side. No complications were reported.

Iliopsoas Strain Iliopsoas strain is an uncommon lesion. Injury occurs when there is forceful and sudden resistance to hip flexion (e.g., when a soccer player collides with another while kicking a ball). Because the pain from an iliopsoas strain radiates into the groin, it is difficult to distinguish this entity from other anterior or intra-articular injuries. Evaluation

Patients describe a sharp pain in the groin, which is increased with resisted hip flexion or passive external rotation. The patient will not have pain with resisted knee flexion, helping to distinguish this entity from other muscle strains such as gracilis and sartorius. Plain films may show complete avulsion of the lesser trochanter; a lesion more common in the skeletally immature, but will otherwise be normal. MRI of the hip shows fluid edema surrounding the iliopsoas tendon as well as intra-substance edema. Treatment

Treatment of these injuries should follow the treatment outlined in Table 21A-3 for general muscle strains. Caution should be taken, however, with patients suspected of complete muscle rupture. Complete muscle rupture should be treated with rest and activity modification for at least 4 to 6 weeks. Return to play should ensue once the patient has full motion and at least 90% of strength.

External Oblique Strain The external oblique muscle is a broad flat muscle that runs superficially in the trunk. The muscle originates from the lower eight ribs and descends distally and medially to insert onto the iliac crest, pubic crest, inguinal ligament, xiphoid process, and linea alba. Injury to this muscle occurs by force when the muscle is under stretch, that is, when the trunk is rotated to the contralateral side. Hockey players and football players are the athletes most susceptible to these injuries during forceful unexpected contact from another athlete or object. The injury is generally a result of partial tearing of the muscle from the iliac crest. Patients will complain of difficulty with trunk rotation and straightening. Some patients may have some bruising along the superior border of the iliac crest with an area of local tenderness. Treatment of this lesion should follow the treatment plan outlined in Table 21A-3. Taping of the abdominal wall has been reported to assist athletes to return to play more rapidly.

Hamstring Strain The hamstring muscles as a group include the semitendinosus, semimembranosus, hamstring portion of the adductor magnus, and long and short heads of the biceps femoris. As a group, these muscles originate from the posterior pelvis and insert distal to the knee joint. These muscles serve as powerful hip extensors and knee flexors. Strains to this muscle group can occur anywhere along the length of a very long musculotendinous junction. Athletes with poor flexibility and strength imbalance are more prone to hamstring injuries.45 Activities like sprinting and water-skiing are common mechanisms of injury.46 Avulsion injuries usually result from severe hip flexion while the knee is maintained in a position of full ­extension. Evaluation

Hamstring strains are seen in the acute setting when the athlete experiences sudden onset of pain in the posterior portion of the thigh during strenuous exercise. Often, an audible pop is heard. Patients complain of immediate pain along the posterior thigh and are forced to discontinue activity. On physical examination, the patient will have ecchymosis that may track down the leg and possibly a visual lump that represents retracted lump of muscle. A defect may be palpable. Examination of patients with suspected acute tears should be performed with the patient lying on the side or stomach with the knee flexed. Extension of the knee may cause spasm or cramping to occur. Palpation at the ischial tuberosity may reveal bony avulsion. Plain radiographs may reveal avulsion fractures or soft tissue collection, but are generally unrevealing. MRI can be used to diagnose hamstring strain and shows edema centered around the musculotendinous units. MRI should be reserved for when the diagnosis is questioned because most hamstrings strains are readily diagnosed by history and physical examination. Treatment

Treatment of hamstring strain should follow the treatment plan outlined in Table 21A-3. Patience is needed with larger strains, and athletes should be cautioned

1462 DeLee & Drez’s Orthopaedic Sports Medicine

against returning to play too early. Rehabilitation of antagonistic muscles, such as the quadriceps, is essential because many injuries recur secondary to muscle imbalance.45 Operative treatment is rarely needed for hamstring strains. It should only be considered for complete tears at the origin or insertion site of the musculotendinous unit. In competitive-level athletes, repair of the hamstring tendon to the ischial tuberosity has been advocated for tears with greater than 2 cm of retraction. Distal injuries generally occur with high-energy injuries to the knee joint that also disrupt the stabilizing ligaments of the knee. The hamstring injuries are addressed at the same time as the ligaments.

Quadriceps Strain The quadriceps muscle group consists of four different muscles. Three of the muscles originate from different sites of the femur: the vastus lateralis, vastus intermedius, and vastus medialis. The fourth muscle, the rectus femoris, originates from the anterior inferior iliac spine of the pelvis. All four muscles receive innervation from the femoral nerve and travel distally in the anterior thigh and insert onto the patella as a common quadriceps tendon. The tendon then envelops the patella and distally reconstitutes as the patella tendon, which courses distally to insert onto the tibial tubercle. Of the four muscles, the rectus femoris is the only muscle that crosses both the hip joint and the knee joint and is therefore more susceptible to strain injury. It is uncommon to see isolated injuries of the vastus lateralis, vastus intermedius, or vastus medialis. Injury to this group of muscles usually occurs during kicking-type activities such as soccer.37 Patients with acute pain remember a specific injury that forced the discontinuation of activity. They may present with a limp and have pain in the anterior thigh. Examination of the thigh may reveal a large palpable defect in the case of a grade III strain. Pain is elicited with resisted hip flexion and knee extension. Sometimes, large ecchymotic areas are present. The more chronic injury presents as an indolent pain in the front of the hip that is exacerbated with kicking activities. Acute injuries tend to occur distally near the patella, and more chronic injuries present with proximal symptoms. Like most muscle strains, plain radiographs are generally unrevealing. In the skeletally immature population, avulsion injuries at the anterior inferior iliac crest may occur. Treatment of quadriceps strains should follow the treatment plan outlined in Table 21A-3.

Sacroiliac Sprains The sacroiliac ligaments are among the strongest ­ligaments in the body; therefore, injury to these ligaments usually requires a violent force. There exists some variety as to the exact mechanism of injury. The ligament structure may be sprained during a torsional load that is transmitted from the axial spine to the pelvis. Simultaneous contraction of the hamstrings and abdominal musculature may cause a sprain in the sagittal plane of the ligament. Also, a violent blow to the buttocks may cause an axially oriented sprain.

Evaluation Clinical Presentation, History, Physical Examination, and Testing

Diagnosis of sacroiliac ligament sprain can be difficult when considering other possible causes of low back pain with possible radiation. Appropriate history is essential. Neuromuscular weakness should alert the treating clinician to consider other diagnoses. Patients with sacroiliac sprain complain of pain that originates in the lower back and radiates into the buttock and occasionally the groin. If there is no history of injury, a clinician must suspect other causes such as lumbar arthritis or spinal radiculopathy. On examination, some patients have tenderness with direct palpation, but this may not always be positive. Lateral compression through pressure along the iliac crests of the pelvis may also reproduce the pain. The most provocative maneuvers of sacroiliac sprain are those that involve flexion of the trunk with the knee extended. Gaenslen’s test result is considered positive if there is a painful response and involves tensioning the sacroiliac ligament in the saggital plain. It is performed by maximally flexing the contra­ lateral hip and hyperextending the hip on the injured side. Imaging

Imaging modalities for this condition should begin with plain radiographs. Although these are often negative, plain films may reveal other sources of pathology. If diagnosis is not evident, MRI reveals some fluid edema within the sacroiliac joint and ligament complex. Computed tomography (CT) and ultrasound have no role in diagnosis. Treatment

Treatment of sacroiliac sprain is primarily accomplished by rest, heating, activity modifications, and anti-inflammatory medications. Lateral compression bracing or taping may help provide relief in the acute phase. Crutch ambulation may be necessary. Steroid injection into the area may be considered, with published reports of moderate relief.47 Like other ligamentous sprains, return to play should only be considered after a stepwise increment in activity results in pain-free progression. A minimum of 4 to 6 weeks is necessary for sprains to heal.

Hernias Traditionally, hernias exist as a spectrum of pathologic processes that involve the extrusion of contents from the abdominal cavity through a defect in abdominal musculature. Certain hernias can cause groin and pelvic pain, and a sports medicine clinician should be aware of how to diagnose and properly treat these entities. Unfortunately, hernias often go undiagnosed for long periods of time because of poor clinician recognition of symptoms and signs. There are three different hernias that may present as hip or ­pelvic pain: the inguinal hernia, the femoral hernia, and the sports hernia.

Inguinal Hernia Inguinal hernias are the most common form of hernia. There are two types of inguinal hernias: direct and indirect. Indirect hernia, the more common of the subtypes, is thought to be congenital in nature. The hernia sac forms

Hip, Pelvis, and Thigh 1463

from the remains of the processus vaginalis that protrudes through the deep inguinal ring. With this type of hernia, the neck of the hernia lies at the level of the deep inguinal ring, and the body lies within the inguinal canal or scrotum and can easily be palpated. The indirect hernia is 20 times more common in males than in females and one third of the time can be bilateral.48 The direct hernia results from musculature defects in the abdominal wall medial to the deep inguinal ring. The hernia sac presents as a bulge medial to the deep inguinal ring along the posterior inguinal wall. The direct hernia is rare in women and is rarely bilateral. Evaluation and Treatment

Inguinal hernias can present as groin pain that is located proximal to or along the inguinal ligament. The pain may radiate into the scrotum. Tenderness can be elicited by direct palpation (in the case of the direct type) and by palpating the inguinal canal (indirect type). Asking the patient to cough may facilitate palpating the hernia. The pain can be exacerbated by a Valsalva maneuver. The diagnosis is made by history and physical examination alone. In the setting of acute pain, differential diagnoses that must be considered include epididymitis, scrotal abscess, testicular torsion, and varicocele. For treatment, general surgery consultation is required. Hernia repairs can be performed through open and laparoscopic techniques with high success rates.49 Recovery from hernia repair can take 2 to 4 weeks, with return to sports after 6 to 8 weeks.

Femoral Hernia Femoral hernias occur when a hernia sac descends through the femoral sheath and enters the thigh. The femoral canal is a term used for the small medial compartment for the lymph vessels in the anterior groin. The superior aspect of this ½-inch-long canal is bordered superiorly by the femoral ring, a condensation of connective tissue and surrounding fascial layers. The femoral septum serves to plug the ring contents from moving in and out of the abdominal cavity. Failure of the femoral septum results in a femoral hernia. The neck of a femoral hernia can be palpated in the femoral canal, which lies lateral and distal to the pubic tubercle. Evaluation and Treatment

Femoral hernias occur more commonly in women than in men. Patients usually complain of activity-related anterior hip pain, and the examiner must have a high index of suspicion in order to make the diagnosis. On physical examination, the patient will have a tender mass within the femoral canal. A femoral hernia can be distinguished from an inguinal hernia by the fact that the sac lies below and lateral to the pubic tubercle, whereas the inguinal hernia lies above and medial to it. Like treatment for inguinal hernias, general surgery consultation is recommended. Recovery from hernia repair can take 2 to 4 weeks, with return to sports after 6 to 8 weeks.

Sports Hernia The sports hernia is a relatively new pathology. Also termed Gilmore’s groin or athletic pubalgia, there is a growing interest and recognition of this condition. Before the

diagnosis of sports hernia, athletes were diagnosed as having chronic groin strains and were treated as such, often with career-ending pain. The pathology of sports hernia is a broad spectrum of injuries involving the inguinal ligament, conjoined tendon, transversalis fascia, internal oblique muscle, external oblique muscle, and rectus abdominis insertion. Sports hernias are postulated to develop from a muscle imbalance that leads to weakening and possible tearing of the structures of the pelvis floor. The muscles involved are thought to be the adductor muscle group of the thigh and the abdominal musculature, where the adductor group is far more powerful than the abdominal musculature. Evaluation Clinical Presentation and History

Patients with athletic pubalgia complain of exertional pain in the pubic area that is aggravated by specific activities. The pain may radiate into the adductor region or the testicles. The pathology is noted to occur more frequently in males than in females. History may reveal a specific aggravating injury. The proposed mechanism of injury is described as a hyperextension injury pivoting around the pubic symphysis, seen frequently in hockey and soccer players.50 There is an indolent progression of the extent of pain that resolves with cessation of activity. Often these patients have seen various specialists, including general surgeons to rule out traditional hernias, urologists to rule out gonadal pathology, gynecologists to rule out ovarian and uterine pathology, and other orthopaedists to rule out other musculoskeletal pathology. Physical Examination and Testing

Physical examination findings for athletic pubalgia are subtle. Patients have tenderness along the conjoined ­tendon or inguinal ring. Patients may also be tender to palpation at the pubic tubercle, symphysis, and adductor origin. A small defect may be palpable within the abdominal musculature near the insertion on the pubis, but this is unlikely. Neurologic examination is normal, but patients may have tight hamstrings or limited hip motion. Provocative testing includes pain with resisted sit-ups and with resisted adduction with the hip externally rotated.50 Imaging

The role of imaging for athletic pubalgia is limited. Imaging is performed to rule out other causes of pain, but there are no modalities specific for athletic pubalgia. Plain radiographs are negative. Three-phase bone scan may produce some increased uptake in the area of the symphysis pubis that may be confused with osteitis pubis. MRI may show rectus tear or symphyseal edema. Herniography is positive only in the case of a large defect.51 Treatment

Treatment for athletic pubalgia should start with conservative management. Proposed treatment should consider whether the patient is preseason, postseason, or midseason and if this is a first-time or recurrent episode. If the athlete has an acute first-time episode and is preseason or midseason, treatment should constitute a 2-week period of rest, ice, and anti-inflammatories. Physical

1464 DeLee & Drez’s Orthopaedic Sports Medicine

therapy should be instituted after this 2-week period. The ­athlete can return to play when he or she is pain free and when range of motion and strength have returned to preinjury levels. Recurrent episodes in the preseason or midseason athlete need to be treated with rest for at least 4 weeks. Recovery may be facilitated with steroid injection. If the patient has symptoms for longer than 6 weeks despite conservative management, operative treatment should be considered. Surgery for a sports hernia is similar to that for a traditional hernia, and a general surgeon is often required to assist with the procedure.50 The repair involves reattachment of the distal edge of the rectus abdominis and imbrication of the transversalis fascia. Some surgeons use a reinforcing mesh.52 There are also successful reports of laparoscopic techniques of hernia repair.52 Recovery from the repair takes about 4 to 6 weeks. Restrictions on strenuous activities continue for 6 weeks total, and patients can usually return to play after 8 weeks. The success rate for surgical treatment has been documented to be as high as 95%.51

Bone Injuries Traumatic Fractures Traumatic fractures around the hip, thigh, and pelvis can be devastating. Unlike hip fractures in the elderly population, fractures in this region of the body require high-energy transmission to the bone. Because the force necessary to disrupt the cortical bone is so tremendous, the clinician should be aware of secondary soft tissue damage. In the case of pelvic fractures, a fascial degloving injury, Morel-Lavelle lesion, should be excluded. Compartment syndrome is rare but can occur in the thigh after femur fracture. Fractures in the hip, thigh, and pelvis can be categorized into four broad categories: pelvic fractures, acetabular fractures, femoral head and neck fractures, and femoral shaft fractures. Pelvic fractures can further be categorized into anteroposterior compression, lateral compression, and vertical shear injuries. Most pelvic fractures are treated nonoperatively with bed rest or limited weight-­bearing ambulation. Acetabular fractures, depending on the location of fracture lines, are traditionally classified into 10 different fracture types using the Judet-Letournel classification.53 Measurements of congruity and involvement of the weight-bearing joint surface help guide clinicians in determining the appropriate treatment. Fractures of the femoral head and neck can involve the intra-articular head (Pipkin classification),54 the femoral neck (Garden classification),55 and the intertrochanteric region ­(number of parts classification). Fractures involving the femoral neck are at special risk because of possible disruption of the blood supply to the femoral head. Even after anatomic fixation of femoral neck fractures, the possibility of avascular necrosis exists as the blood supply to the femoral head remains compromised. Fractures involving the intertrochanteric region, the subtrochanteric region, and the shaft of the femur should all be rigidly fixed with close anatomic alignment. Athletes may take up to 6 months to 1 year to recover from fractures involving this region of the body.

Hip Dislocations The stability of the hip joint is a summation of the congruent bony restraints, the strength of the surrounding ligaments and hip capsule, the dynamic stability provided by opposing muscle tone, and the negative pressure with the joint under tension. The force necessary to cause a hip dislocation must be violent enough to overcome these stabilizing forces, rendering a patient completely disabled. Eighty-five percent of the dislocations are posteriorly directed and occur from a posteriorly directed force with the knee and hip flexed.56 Fifteen percent of dislocations are directed anteriorly and occur from an axial-directed force with the knee and hip extended.56

Evaluation and Treatment Patients with posterior hip dislocations present after traumatic injury unable to bear weight, with the hip flexed, adducted, and internally rotated. Patients with anteriorly dislocated hips present with the hip flexed, abducted, and externally rotated. Radiographs of these patients should include a standard trauma series as well as anteroposterior and lateral radiographs of the affected hip. The clinician should suspect and check for other injuries. Treatment of a hip dislocation requires reduction under paralytic sedation. After reduction is performed, the hip should be placed through a range of motion to check for stability. CT should be performed to diagnose any loose fragments from either a femoral head fracture or acetabular wall fracture. Avascular necrosis may occur after dislocation injury and patients need to be followed with radiographs for at least 2 years.57 For patients who have no associated fractures, the rehabilitation process starts with bed rest for 48 hours. Some investigators recommend the use of a knee immobilizer or abduction pillow. Partial weight-bearing with physical therapy for gentle range of motion is instituted after 2 days. Patients can be advanced to weight-bearing as tolerated with strengthening at 2 weeks. Return to full activities may take up to 6 to 12 weeks.

Stress Fractures Stress fractures can occur in both the femur and the pelvis. Unlike fractures that occur from trauma, stress fractures occur as a result of repetitive submaximal loading of bone. The rate of bone resorption exceeds that of bone formation, leading to the accumulation of microfractures. The accumulations lead to the clinical entity of stress fractures. The rate and incidence of stress fracture formation is determined by the strength of the native bone and the degree of the repetitive force applied. For example, running on hard surfaces applies more axial force to the hip than running on cushioned surfaces. Also, the quality of the bone is critical in the formation of stress fracture. Osteopenic bone is at greatly increased risk for stress fractures. Patients with stress fractures have pain that is aggravated by activity and that subsides with rest. Plain radiographs may be normal until the fracture is already remodeling, at which point callus may be seen. Three-phase bone scan and MRI are both highly sensitive tests for stress fractures. Bone scan often

Hip, Pelvis, and Thigh 1465

shows a discrete line or area of increased uptake in the region of the stress fracture.58 MRI shows bony edema at the area of the fracture and is more specific than bone scan. MRI also assesses for other pathology in the surrounding anatomy.59

Pelvic Rami Stress Fractures Stress fractures of the pelvic rami, like other stress fractures, develop from overuse activity. Runners or joggers are more likely to be affected.60 Patients present with pain localized to the inguinal, perineal, or medial thigh region. The pain is exacerbated with activity and relieved with rest. Tenderness can be elicited by asking the patient to stand unsupported on the affected leg.61 Treatment of a ramus stress fracture involves a stepwise protocol. First the offending activity is identified and eliminated. Training is discontinued for at least 1 week. After 1 week, physical therapy for flexibility and nonimpact strengthening is initiated. Impact activities are gradually started after 6 weeks, with a progressive return to offending activities.

Sacral Stress Fractures The sacrum serves as a keystone in helping transmit axial forces from the appendicular skeleton to the axial skeleton. Stress fractures of the sacrum can occur from repetitive axial loading as well as repetitive standing from a seated position. Patients will complain of lower back pain that is exacerbated with the offending activity and relieved with non–weight-bearing. Clinicians must have a high index of suspicion to make this diagnosis. MRI or bone scan is used to confirm the diagnosis. Treatment of sacral stress fractures includes activity modification and protected weight-bearing as long as the patient remains symptomatic, and return to play may take as long as 3 months.

will have full range of motion but most likely will have tenderness at the extremes of motion, particularly internal rotation. The neurovascular examination will be normal. Imaging

Imaging is essential in confirming the diagnosis of femoral neck stress fracture. Plain radiographs may be normal. There may be a visible fracture line, a visible break in the trabeculae, or callus formation. It is important to note the neck-shaft angle because patients with coxa vara are at higher risk for stress fractures. Radiographs of the contra­ lateral hip may show a subclinical stress fracture because this process can be bilateral.63 Traditionally, bone scan was the established method of diagnosing and following stress fractures until they were fully healed. Recently, however, MRI has proved more sensitive and specific at detecting femoral neck fractures (Fig. 21A-8).64 Classification

Femoral neck stress fractures can be classified into two different groups of fractures: tension-side fractures and compression-side fractures. The classification scheme is based on location of the fracture and dictates treatment. Tensionside stress fractures occur along the superior lateral cortex of the neck such that a weight-bearing axial force causes distraction at the fracture site. These fractures are more often seen in an older population Compression-side stress fractures are located on the inferior medial cortex. Axial forces cause compression along the fracture line. Radiographs reveal bone formation along the medial cortex. Younger athletes are more prone to compression-type stress fractures.65

Femoral Neck Stress Fractures Femoral neck stress fractures are the result of intense impact-loading training. The well-oriented trabeculae of the femoral neck undergo microfracture because of a mismatch in bone turnover. These fractures occur in two populations: the elderly population with osteoporotic bone and the young active athlete. In the young ­population, the aggravating activity is easily identified. Often there is a history of initiation of new activity or increasing the intensity of an existing activity. Classically, these injuries occur in long-distance runners, ballet dancers, and military personnel.62 Evaluation Clinical Presentation and Physical Examination

Femoral neck stress fractures are difficult to diagnose. They usually present with the complaint of groin pain that is activity related. The pain may radiate to the thigh and knee. There is no history of trauma, and the symptoms have a history of gradual onset. Often, there is a history of a recent change in activity, duration, or frequency. Pain during the nighttime is common. On clinical examination, the patient

Figure 21A-8  Magnetic resonance image of compressionsided stress fracture of femoral neck (arrow).

1466 DeLee & Drez’s Orthopaedic Sports Medicine Treatment

The treatment of femoral neck stress fractures is dependent on the type of fracture. Patients with nondisplaced compression-type fractures should be treated with protected weight-bearing for 3 to 6 months. Sequential radiographs should be performed to determine healing progress. At the end of treatment, bone scan can be performed to confirm complete healing.58 If, during the treatment, any displacement or diastasis occurs at the fracture site, the fracture is unstable and should be operatively fixed without delay. Displaced or tension-side stress fractures should be fixed without delay upon diagnosis. Any delay in treatment for a displaced fracture may increase the risk for nonunion or avascular necrosis.66 At the time of surgery, it is important to anatomically reduce the fracture. Three cannulated screws have been deemed adequate for fixation. The complication rate for displaced fractures is noted to be as high as 60%.55 Surgical treatment for nondisplaced fracture has an approximately 90% success rate.55

Osteitis Pubis Osteitis pubis is a general description of pathology involving the inflammation of the pubic symphysis with secondary pain. Osteitis pubis in the athlete should not be confused with postsurgical osteitis pubis, which resolves with time. Causative mechanisms are not well understood in the athletic population, but investigators believe that the underlying feature involves overuse of the adductors and gracilis muscles.67 Possible causes are hypothesized to include microstrains at the origins of these muscles, avascular necrosis of the symphysis, osteochondritis dissecans at the symphysis, or fatigue fracture. Some believe that an imbalance between abdominal wall musculature and hip adductor strength can be a risk factor.67

Evaluation Clinical Presentation, History, and Physical Examination

Osteitis pubis presents clinically as pain at the level of the symphysis pubis that may radiate into the groin, medial thigh, or abdomen. The history may reveal that the patient is involved in kicking and running activities such as soccer, football, and long-distance running.68 Any repetitive activity that may cause microtrauma to the adductor origin may be the culprit. The pain more often has an insidious onset but can be acute secondary to an injury. Physical examination reveals tenderness along the symphysis. Pain can be elicited by passive abduction and resisted adduction. Other clinical disorders can be ruled out by the history and physical ­examination. Imaging

Imaging for osteitis pubis consists of plain radiographs and bone scan. Although plain radiographs can be negative early in the disease process, Harris and Murray have described three common features of athletic osteitis pubis: bone resorption at the pubis symphysis, widening at symphysis, and rarefaction along the pubic rami.50 Three-phase bone scan shows a broad area of bone uptake centered around the pubis (Fig. 21A-9).

Figure 21A-9  Bone scan of osteitis pubis.

Treatment Options Nonoperative

Conservative treatment for osteitis pubis has reported success rates of 90% to 95%.69 Treatment should include anti-inflammatory medications for at least 2 weeks. Therapeutic modalities, including cryotherapy, ultrasound, and electrical stimulation, can be helpful. A progressive rehabilitation program, including core strengthening, adductor stretching, and balance control, should be instituted. Some investigators advocate steroid injection at the point of maximal tenderness.68 Patients should be made aware that complete recovery may take 2 to 3 months. Operative

Surgical treatment of osteitis pubis should be limited to those patients who fail conservative treatment. Four different procedures have been described in the literature: curettage of the symphysis, wedge resection, wide resection, and arthrodesis.70 Reports suggest that most athletes respond well to curettage alone without resection.70 Full recovery from curettage can take 3 to 6 months. ­Arthrodesis should be reserved for those that fail curettage, and patients may take up to 1 year to recover from this procedure.69,70

Osteonecrosis of the Femoral Head Osteonecrosis of the femoral head is a common pathologic outcome of several conditions (Box 21A-4). All conditions listed cause the process of decreased blood flow leading to necrosis of the femoral head. Despite the known potential causes of avascular necrosis, idiopathic avascular necrosis constitutes 10% to 20% of all causes.71 Unfortunately, for most patients, the diagnosis is made at the later stages of the disorder, and treatment options are limited.

Evaluation Clinical Presentation, Physical Examination, and Testing

Avascular necrosis of the femoral head can occur at any age but most often occurs between the ages of 20 and 50 years. The most common chief complaint in patients with avascular necrosis is pain, insidious in onset and localized to the

Hip, Pelvis, and Thigh 1467

Box 21A-4   Causes of Osteonecrosis

• Trauma • Corticosteroid use • Alcohol abuse • Smoking • Sickle cell anemia • Coagulopathies • Systemic lupus erythematosus • Hypercholesterolemia • Organ transplantation • Gaucher’s disease • Caisson disease • Radiation therapy • Arterial disorders • Intramedullary hemorrhages • Chronic pancreatitis • Hypertriglyceridemia • Idiopathic

groin with possible radiation to the buttock or knee. Physical examination reveals pain with hip motion, particularly with internal rotation. Range of motion may be limited. The patient may walk with a limp. Bilateral examination should be performed because there is a reported incidence of both hips being involved of 40% to 80%.72 Imaging

Imaging should begin with plain radiographs, which include anteroposterior and lateral views of the hip. Findings of changes of the femoral head have been classified by several systems, the most popular being the Arlet-Ficat staging system (Table 21A-4).73 MRI is the most accurate method of diagnosing osteonecrosis of the femoral head. Sensitivity has been reported to be as high as 88% to 100%.74 It is extremely helpful in diagnosing patients with Ficat stage I avascular necrosis.

core ­decompression in patients with Ficat stage I has been reported at 70% to 95%. Core decompression is reported to be only 50% successful in patients with Ficat stage II necrosis and only minimally successful in stages III and IV.75 In symptomatic patients with stage II or III disease, there have been some reports of success with bone grafting using vascularized fibula.76 Total hip arthroplasty remains the treatment of choice for stage IV disease.77

Degenerative Joint Disease of the Hip Degenerative joint disease of the hip is the final common pathway of many disease entities, including idiopathic hip osteoarthritis. Other causes include traumatic arthritis, congenital hip dysplasia, systemic inflammatory diseases, and avascular necrosis (Box 21A-5). Patients present with an insidious onset of pain localized to the groin that may radiate to the buttocks, thigh, and knee. Physical examination may show limited range of motion, leg-length discrepancies, and antalgic gait. Plain radiographs can show a variety of different findings, the most common of which is joint space narrowing, osteophyte formation, subchondral cysts, and sclerotic bony margins. Treatment of degenerative joint disease should begin with anti-inflammatory medications, physical therapy, and activity modifications. If conservative treatment does not relieve the patient’s pain, surgical treatment with total hip arthroplasty is recommended and has been well documented to have very successful results.

Nerve Entrapment Injuries Several major nerves in the hip, thigh, and pelvis are of concern to the orthopaedic clinician. Injury to these nerves can be a source of severe disability and can be frustrating to the patient and treating clinician. Making the diagnosis can be difficult as symptoms are based on exact location of nerve lesion. Often, the patient has no motor involvement, with only complaints of pain. Treatment is symptom dependent and begins with activity modification, but may eventually require surgical release of the nerve.

Treatment Treatment of avascular necrosis is dependent on the stage at which it is clinically detected. In early stages of the disease, conservative methods, including limited weightbearing and weight-bearing with assist, prevent head deformation and limit pain. Surgical treatment for early stages includes core decompression. The success rate of

  TABLE 21A-4  Arlet-Ficat Staging System for Avascular Necrosis of Femoral Head Stage

Findings

I

No radiographic finding, but ­suspicious clinical presentation Subchondral sclerosis and cysts. Remodeling of femoral head without overall change in shape Crescent sign: partial collapse of necrotic ­segment Joint space narrowing, osteophytes, deformed femoral head

II III IV

Box 21A-5   Common Causes of Hip Arthritis

• Osteoarthritis • Trauma • Rheumatoid arthritis • Infection • Reiter’s arthropathy • Psoriatic arthropathy • Avascular necrosis • Gout • Pseudogout • Ankylosing spondylitis • Hemophilia • Paget’s disease • Legg-Calvé-Perthes disease • Slipped capital femoral epiphysis • Developmental dysplasia of the hip

1468 DeLee & Drez’s Orthopaedic Sports Medicine

Sciatic Nerve Entrapment The sciatic nerve arises as a branch of the sacral plexus and has contributions from the L4-S3 nerve roots. Once formed, the nerve exits the pelvis through the greater sciatic foramen. Traveling deep to the piriformis muscle, the nerve courses distally and laterally, remaining superficial to the ischial tuberosity and remainder of the short external rotators. The nerve then continues distally within the posterior compartment of the thigh. The term sciatica is a general phrase used to describe symptoms of nerve injury that occurs anywhere along its course and may cause debilitating symptoms. Intrapelvic and ischial spine compression are common sites of the lesion. Often a single injury or repetitive trauma, as seen in cyclists, can be implicated. Symptoms may vary from radiating posterior thigh pain to complete paralysis. A comprehensive neurologic examination should be performed to document progression of muscular strength and neurologic deficits. Straight-leg raising and nerve-stretch testing may be positive. Electromyographic and nerve conduction studies can help confirm the diagnosis. Treatment for sciatic injury involves activity modification, directed physical therapy, anti-inflammatories, and possible steroid injections to the nerve roots. In refractory cases, if exact site of entrapment is known, then surgical decompression can be successful.

Piriformis Syndrome Piriformis syndrome is a specific type of sciatic nerve entrapment that occurs as the nerve courses deep to the piriformis muscle. Piriformis syndrome has three different identified causes: variations in normal anatomy of the piriformis muscle (i.e., slit muscle belly encompassing the nerve), repetitive sports activity that causes hypertrophy of the surrounding muscle, and acute trauma to the ischial spine and gluteal area.78 Postural problems and leg-length discrepancies have been implicated in exacerbating the symptoms.78 Evaluation

Patients present with pain in the buttocks with radiation down the posterior thigh and possibly even the leg. Muscle strength is often normal. Physical examination can be helpful in excluding other causes of pain but is nonspecific for piriformis syndrome. Deep palpation reveals tenderness near sciatic notch and also may reveal a gluteal mass. Pain may be elicited by resisting abduction and external rotation of the flexed hip. Similarly, stretching the ­piriformis by flexing, adducting, and internally rotating the hip may be painful. Tenderness with palpation of the piriformis during rectal examination is a characteristic finding.79 Imaging for piriformis syndrome is helpful in ruling out other possible diagnoses. MRI may show compression or edema of the sciatic nerve as it travels through the sciatic notch. Electromyographic and nerve conduction velocity studies may be helpful.80 Treatment

Piriformis syndrome should first be treated conservatively. A routine of activity modification, physical therapy with piriformis stretching, ultrasound treatment, and oral­

anti-inflammatories should be implemented. A maximum of three steroid injections can be given into the area of maximal tenderness.81,82 Surgical treatment is recommended for refractory cases lasting longer than 6 months and involves release of the piriformis muscle. During the procedure, the surgeon must be aware of possible anatomic variants of the nerve and muscle. The athlete can return to play 8 to 12 weeks after an open procedure. Success rates of more than 90% have been documented.83 Recent reports have been made of endoscopic piriformis release with equal results.84

Obturator Nerve Entrapment The obturator nerve is a branch of the lumbar plexus and has contributions from the L2-L4 nerve roots. It enters the pelvis from the medial border of the psoas muscle. After crossing the anterior border of the sacroiliac joint, the nerve passes through the upper part of the obturator foramen, known as the obturator canal, and exits the pelvis. The nerve splits into anterior and posterior divisions. The anterior division provides innervation to the hip joint, most of the adductor muscles, and cutaneous portions of the distal two thirds of the medial thigh. The posterior division of the obturator nerve provides motor innervation to the adductor brevis, obturator externus, and proximal portions of the adductor magnus.

Evaluation Entrapment of the obturator nerve can occur anywhere along the course of the nerve. The most common place of entrapment is at the ischium within the obturator canal by a persistent fascial band.85 The nerve may also be compressed by inflammation of the surrounding adductor musculature in the thigh, hematoma, obturator hernia, or even fracture caused by trauma.86 Patients present with pain localized to the medial thigh that may radiate to the groin or knee. The pain is often exacerbated with activity.85 On physical examination, the patient may or may not demonstrate adductor weakness. To confirm the diagnosis, electromyographic and nerve conduction velocity studies should be performed.

Treatment Treatment of obturator entrapment should begin with conservative measures. For patients with acute symptoms, the clinician must rule out instigating causes, including adductor avulsion and muscle strain with hematoma. Anti-inflammatories and physical therapy with stretching, deep tissue massage, and ultrasound have been helpful in relieving symptoms. A series of up to three fluoroscopyguided steroid injections can be tried. If symptoms persist, surgery should be considered.85 Surgical treatment involves medial approach to the groin, blunt dissection, and release of the anterior division through the obturator canal.85 The surgeon must be careful with treatment of this nerve because nerve palsy can be a devastating complication to the athlete. The success rate of nerve release is reported as high as 80%, with return to play in as little as 1 month.

Hip, Pelvis, and Thigh 1469

Pudendal Nerve Entrapment

Femoral Nerve Entrapment

The pudendal nerve is a branch of the sacral plexus with contributions from the S2-S4 nerve roots. The major trunk passes posterior to the sacrospinous ligament and into the ischiorectal fossa. The nerve then enters the pudendal canal, also known as Alcock’s canal, and emerges below the pubic bone to provide terminal branches. Motor branches of the nerve provide innervation to the perineal muscles. The pudendal nerve also carries parasympathetic and sympathetic fibers to the perineal area. Pudendal nerve compression can occur anywhere along the course of the nerve. The pudendal canal is a common location of nerve compression.87 Microtrauma to the canal affects pudendal nerve excursion within the canal and can provoke symptoms. Lastly, the nerve can be compressed by external pressure on the symphysis pubis. Patients with pudendal nerve compression complain of pain in the perineum with potential numbness in the shaft of the penis and perineum. Some patients present with impotence. History usually reveals activity that involves prolonged compression of the perineum, such as long­distance cycling.88 The injury may be caused iatrogenically with the use of a fracture table traction post.87 Treatment of this injury is discontinuing the offending activity. In the case of cyclists, seat modification has been helpful. Urology consultation should be considered if surgical release is required.

The femoral nerve arises as a branch of the lumbar plexus with contributions from the L2-L4 nerve roots. The nerve enters the pelvis from the lateral border of the psoas and passes deep to the iliacus muscle. The nerve then travels distally and enters the thigh by passing under the inguinal ligament. The nerve then divides and innervates the anterior muscles of the thigh and provides cutaneous innervation for the anterior thigh and lower leg. Entrapment of the femoral nerve can occur from multiple causes, including trauma, tumor, and childbirth.89 The two most common causes of femoral nerve entrapment in the athlete are iliopsoas strain with compressing hematoma formation and anomalous slip of iliacus muscle.89 Patients complain of difficulty walking downstairs and paresthesias over the anterior thigh and anterior medial portions of the leg. Examination reveals quadriceps weakness and diminished knee-jerk reflex. Stepwise treatment of this lesion should include ice, ultrasound, and physical therapy. Surgical decompression is rarely necessary and is only successful when a definitive lesion can be located.89

Ilioinguinal Nerve Entrapment The ilioinguinal nerve arises as a branch of the lumbar plexus and has contributions from the L1-L2 nerve roots. The nerve pierces through the psoas muscle and spirals distally and anteriorly, deep to the abdominal musculature. The nerve sends contributions to the abdominal musculature as well as cutaneous contributions over the iliac crest. The remainder of the nerve enters the inguinal canal and then divides to provide cutaneous innervation for the groin, proximal medial thigh, base of the penis, and upper part of the scrotum in men or the mons pubis in women. Ilioinguinal nerve entrapment can occur from scarring from prior surgical incisions, iatrogenic damage such as iliac crest harvest, and hypertrophied abdominal muscles as seen in bodybuilders and pregnant women.89 Patients complain of pain in the right inguinal region radiating into the genitals and sensory abnormalities. Physical examination reveals tenderness to palpation 2 to 3 cm medial to and below the anterior superior iliac spine.90 If ilioinguinal nerve entrapment is not surgically caused, conservative methods of treatment should be followed. A program of ice, ultrasound, anti-inflammatories, rest, and physical therapy for abdominal stretching often resolves the entrapment. Surgical treatment through open neurectomy is rarely needed but has been successful.91 In patients with surgically related causes, a period of 6 months to 1 year should be allowed for symptom resolution. If there is no resolution, surgical release of the nerve with scar formation should be performed. The results of surgery are dependent on the amount of preoperative nerve damage.89

Lateral Femoral Cutaneous Nerve Entrapment The lateral femoral cutaneous nerve (LFCN) arises as a branch of the lumbar plexus with contributions from the L2-L3 nerve roots. The nerve courses retroperitoneally and emerges just medial to the anterior superior iliac spine. The nerve then passes under the inguinal ligament and crosses over the sartorius muscle and travels distally to provide sensory innervation to the lateral thigh. Entrapment of the LFCN, also known as meralgia paresthetica, can be caused by constricted clothes, rapid weight loss, systemic disease, and iliac crest harvest.92,93 Gymnasts who use the uneven bars have been reported to have this injury because of suspected repetitive trauma to the anterior superior iliac spine.94 Patients complain of anterolateral thigh numbness and paresthesias. Patients may have a positive Tinel’s sign located 1 cm medial and inferior to the anterior superior iliac spine.95 Nerve conduction velocities demonstrate prolonged nerve latency.95 LFCN entrapment treated conservatively has been reported to resolve in 90% of patients.96 Patients should remove offending activities. Local steroid injections have been reported to facilitate symptom resolution. LFCN entrapment that fails conservative treatment can be treated surgically. Two surgical methods have been described. The first method involves exploration and neurolysis of the LFCN. Success rates for this operation are reported to be 30% to 80%.97 The second procedure involves LFCN transection. Reported success rates of transection have been as high as 90%, but this operation leaves the patient with hypoesthesia of the anterolateral thigh.97

Intra-articular Pathology The diagnosis of intra-articular pathology has recently generated excitement among sports medicine clinicians. Improved imaging techniques, such as MRI arthrogram, have allowed for more definitive diagnosis. In addition, the

1470 DeLee & Drez’s Orthopaedic Sports Medicine

rising popularity of hip arthroscopy has incited development of sophisticated equipment and techniques that allow the surgeon to access and treat intra-articular lesions that otherwise would have gone untreated.

Labral Tears The acetabular labrum is a ring of fibrocartilage that lines the border of the acetabulum and is continuous with the transverse acetabular ligament. Labral tears have been reported as the most common form of intra-articular path­ ology in the hip. Five different causes of labral tears have been identified: femoroacetabular impingement (see later for full discussion), hip dysplasia, trauma, capsular laxity, and joint degeneration.98 Patients with dysplastic hips are often found to have a large floppy labrum that is more susceptible to tearing, particularly with increased translational motion as seen in dysplastic hips. Trauma, such as hip dislocation or subluxation, can be responsible for labral tears as well as loose bodies. Patients with capsular laxity and hip hypermobility, such as patients with EhlersDanlos syndrome, also have increased translational forces across the hip joint. Patients with joint degeneration are likely to tear the labrum owing to repetitive nonconcentric motion of the hip with acetabular edge loading. History, physical examination, and diagnostic techniques can help diagnose labral tears and possibly elucidate the underlying cause.

Evaluation Clinical Presentation and Physical Examination

Patients with labral tears complain of pain in the groin with certain motions. The pain can be exacerbated by sudden twisting or pivoting motion. They may also have a clicking or catching sensation, known as intra-articular snapping hip. Physical examination may reveal decreased range of motion of the affected hip. Dependent on the location of the labral tear, certain provocative tests can be performed. Ranging the hip from a fully flexed, externally rotated, and abducted position to a position of extension, internal rotation, and adduction causes pain if an anterior-based labral tear is present.99 Posterior tears can be painful if the hip is brought from a flexed, adducted, and internally rotated position to one of abduction, external rotation, and extension.100

using a 3-tesla magnet report sensitivity as high as MRI [AQ9] arthrogram, but there is a minimal amount of published data at this time.102 MRI arthrogram should be the diagnostic test of choice. There have been reports of greater than 92% sensitivity in discovering labral tears.103 Furthermore, intra-articular injection of lidocaine and steroid, along with the contrast dye, at the time of arthrography can be diagnostic and therapeutic.

Treatment Attempts at conservative treatment of labral tears have been made, but with poor results. Physical therapy for periarticular strengthening and limited weight-bearing may allow for temporary resolution of symptoms, but the symptoms usually return after the patient returns to full activity.101 If the patient is in midseason or preseason, steroid injection into the joint may be beneficial, but symptoms usually return. Permanent treatment is best accomplished through surgery.101 Hip arthroscopy can be successful at removing any unstable portions of the labrum and removing some synovitis (Fig. 21A-10).104-106 Postoperative recovery usually takes 6 to 8 weeks. There are no weight-bearing restrictions, and patients should start physical therapy for range of motion and strengthening within 1 week of surgery. Return to play is allowed once the patient has full motion and full strength. Femoroacetabular impingement can also be addressed at the same time as the labral tear. If bony work is performed, the rehabilitation of the patient is greatly influenced, and the patient may take longer to return to play. There have been recent reports of arthroscopic labral repair.98 Repair should be considered for the patient with full-thickness labral tear involving the capsulolabral junction. Postoperatively, the patient must limit weight-bearing for 4 weeks. Also, flexion and abduction are limited for 4 to 6 weeks. Short-term results have been very positive, but no longterm results have been published to date.98 For patients with degenerative labral tears, surgical relief may be temporary and is dependent on the amount of preoperative

Imaging

Imaging modalities to diagnose labral tears have recently made substantial improvements. Plain radiographs do not demonstrate signs of labral tears but are ­essential in ­diagnosing underlying pathology. Evidence of hip dysplasia, trauma, joint degeneration, and femoroacetabular ­impingement can be found on plain films.101 Computed tomographic scanning is also helpful in determining an underlying cause, but does not show direct evidence of labral tear. Dynamic ultrasound may show evidence of snapping hip, but is not specific for labral tear. MRI using 1.5-tesla magnets allows for excellent soft tissue definition around the hip and can often diagnose the underlying cause, but it has not been found to be highly specific or sensitive for the diagnosis of labral tears. Recent studies

Figure 21A-10  Arthroscopic photo of acetabular labral tear.

Hip, Pelvis, and Thigh 1471

joint damage. For permanent pain relief, arthroplasty may be necessary.104

Femoroacetabular Impingement Femoroacetabular impingement (FAI) has recently gained increased recognition as a potential cause of hip pain in athletes. Impingement was classically defined as articular incongruity between the femoral head and acetabulum that led to static overload of the hip joint and eventual degenerative arthritis. Better understanding of FAI suggests that there is an abnormal dynamic contact between proximal femur and acetabulum that results in damage to femoral neck, acetabular rim, hip labrum, and articular cartilage. Causes of FAI include Legg-Calvé-Perthes disease, slipped capital femoral epiphysis, trauma, and idiopathic causes.107,108 Historically, patients with FAI presented with a vague diagnosis of hip pain that eventually progressed to fulminant osteoarthritis of the hip joint.109 With recent advances, pathologies caused by FAI are being detected and treated earlier.

Classification FAI can be classified into three types of impingement: cam impingement, pincer impingement, and a combination of cam and pincer impingement. Cam impingement is a femur-based pathology (Figs. 21A-11 and 21A-12).110-112 It can occur when there is insufficient offset at the head and neck junction or when the head is “out of round,” essentially making the head too large for the acetabulum at the extremes of motion. With this type of anatomic configuration, standard hip motion causes abnormal loading of the anterior acetabular cartilage and labrum and increased stress at the acetabular edge.112 Pincer impingement is an acetabulum-based pathology. Anatomically, the acetab­ulum is found to have poor anteversion or may be too deep, essentially encompassing the femoral head and limiting the amount of hip motion. The femoral neck continuously loads the anterior acetabular rim, with consequent damage to the cartilage and labrum.111 The combination of both cam and pincer mechanics can also be seen and can accelerate the occurrence of damage caused by normal motion.

Figure 21A-12  Pincer impingement. The arrow depicts an acetabulum that is too deep and blocks range of motion of the femoral head because of impingement on the femoral neck.

Evaluation Clinical Presentation, Physical Examination, and Testing

The clinical presentation of FAI can be varied.109 With minimal symptoms, patients complain of hip stiffness or start-up pain and limited range of motion. Some patients may present with the clinical finding of labral tear. Older patients may present with fulminant osteoarthritis. Patients may have pain with weight-bearing activities. Most often, patients complain of groin pain with activities that involve hip flexion. Physical examination reveals limited flexion that can be less than 90 degrees. Patients generally have more passive external rotation than internal rotation. An anterior impingement test is positive if a patient has pain with passive flexion, adduction, and internal rotation.109 Imaging

Imaging is essential in the diagnosis of FAI. Plain films should include an anteroposterior pelvis and true lateral x-rays with the hip in 15 degrees of internal rotation. Acetabular retroversion can be diagnosed on the anteroposterior pelvis by noting more coverage of the anterior rim of the acetabulum when compared with the posterior rim. Inadequate femoral head and neck offset can be diagnosed on the true lateral. A severe example of this leads to the “pistol grip” deformity. Three-dimensional computed tomographic scans can show asphericity within the femoral head and an offset imbalance. A computed tomographic scan that includes images of the distal femur can assist in calculating femoroacetabular version mismatch. MRI is ideal for determining cam impingement versus pincer impingement. The α angle can be measured on the oblique coronal images. The angle is formed by a line along the neck-shaft axis intersecting with a line drawn from the femoral head center to the femoral head asphericity point (Fig. 21A-13). Reports have shown that angles greater than 55 degrees coincide with cam-type FAI.113 If labral tear is suspected, MRI arthrogram should be performed.

Treatment Bump Figure 21A-11  Cam impingement. The arrow depicts an “out-of-round” femoral head that impinges on the acetabular margin at extremes of hip motion.

Patients with symptomatic FAI should be treated surgically. Because FAI is a mechanical problem, nonsurgical management does not allow for resolution of symptoms. If a patient presents with fulminant osteoarthritis, total joint arthroplasty should be considered. If the patient presents

1472 DeLee & Drez’s Orthopaedic Sports Medicine

Loose Bodies Loose bodies can occur within the hip joint as a result of trauma, FAI, labral tears, or synovial chondromatosis. Once formed, loose bodies can be devastating to joint surfaces due to third body wear. Patients complain of pain with motion localized to the groin and radiating to the buttocks or medial thigh. Examination may reveal limited motion or an audible snap. Diagnosis is confirmed by imaging. Plain radiographs and MRI and computed tomographic scans should be ordered appropriately, depending on the suspected mechanism of formation of loose body. MRI arthrography can significantly increase sensitivity of detecting loose bodies compared with plain MRI. Once diagnosed, removal of third bodies can be performed through open or arthroscopic procedures (Fig. 21A-14).

Chondral Injuries

Figure 21A-13  Measuring the α angle (arrow) using oblique coronal magnetic resonance images. The line drawn along the neck-shaft axis bisects the line drawn from head center to femoral head asphericity point.

without fulminant arthritis, the goals of operative treatment include resecting the impinging tissue, appropriately treating damaged intra-articular lesions, restoring motion, and minimizing trauma to surrounding structures. Surgery can be performed arthroscopically or by an open technique. Arthroscopic surgery has the advantages of being minimally invasive. The constant flow of arthroscopic fluid minimizes the risk for infection. Also, labral tears can be managed successfully. The disadvantages of arthroscopic surgery include the technical difficulty of hip arthroscopy and availability of equipment.114 Surgeon experience may limit the ability to perform a proper resection. Also, arthroscopic treatment of FAI requires the use of dynamic traction, which may not be available, in order to access the lateral femoral head. The advantage of open treatment includes that it allows for easy access to all hip structures and facilitates intraoperative ranging of the hip at extremes of motion.115 The disadvantages of open surgery include the risk for avascular necrosis after surgical capsulotomy of the hip joint. Also, the recovery time may be prolonged secondary to the highly invasive nature of the operation. Despite the method of surgery, patients should remain on protected weight-bearing precautions for at least 4 to 6 weeks. Return to full activity may begin as early as 3 months. Patients should be cautioned against premature return to activity because of the documented risk for femoral neck fracture. The results of treatment have been varied in the literature and depend on the procedure ­performed.114,115 Pain relief is dependent on the amount of preexisting damage. Younger patients have higher success rates than older patients, and the complication rate is highly dependent on the complexity of the procedure performed.

Chondral injuries on the femoral head or the acetabulum can be caused by trauma, loose bodies, labral tears, and FAI. Often these lesions coexist with labral tears. They occur as a large spectrum of injuries and can be classified using the Outerbridge classification system (Table 21A-5). Grade 0 to I injuries must be monitored and treated symptomatically. Débridement of grade II to III injuries has good results. Options for grade IV lesions include débridement and microfracture. Investigators report success with microfracture, but overall results depend on amount of preoperative damage.116

Ruptured Ligamentum Teres The ligamentum teres is an intra-articular structure that can be damaged by acute trauma or by chronic synovitis and inflammatory processes within the hip joint. Three types of ligamentum teres lesions have been described.117 The first type corresponds to complete rupture and is usually associated with a subluxation or sudden twisting injury. The second type of lesion is a partial rupture and is also usually secondary to trauma. The third type is degenerative ­fraying

Figure 21A-14  Arthroscopic removal of loose bodies within the hip joint.

Hip, Pelvis, and Thigh 1473

  TABLE 21A-5 Outerbridge Classification Scheme

Box 21A-6   Indications for Hip Arthroscopy

for Chondral Injuries Grade

Description

0 I II III

Normal cartilage Softening and swelling of the cartilage <1.5 cm defect with fissuring, no bone exposed >1.5 cm defect with fissuring and delamination, no bone exposed Any size defect with subchondral bone exposure

IV

and is caused by chronic inflammation of the hip joint. Any of these three lesion types can cause mechanical hip pain. The patient may have a snapping sensation. Diagnosis may be confirmed with MRI arthrogram. If intra-articular steroid injection is unsuccessful in treating the symptoms, then hip arthroscopy with removal of any loose fragments should be performed (Fig. 21A-15). There are some reports that conclude that ligamentum teres–deficient hips are more lax and more susceptible to labral trauma.117

Synovial Disease There are several causes of synovial inflammation: systemic processes, synovial chondromatosis, pigmented villonodular synovitis (PVNS) and idiopathic osteoarthritis. Systemic processes include rheumatoid arthritis, infectious arthritis, and a multitude of autoimmune and seronegative arthropathies. Such disease processes should be treated accordingly. Synovial chondromatosis and PVNS are both detectable by MRI and are amenable to surgical treatment with débridement of the hip joint and synovectomy.

Hip Arthroscopy Hip arthroscopy is a rapidly growing field among orthopaedic surgeons. With the increase in popularity, more pathologic lesions are being identified and treated. When first introduced, instrumenting the hip joint with arthroscopic equipment was challenging. The depth of the joint, highly constrained configuration of the femoral head within the

Figure 21A-15  Partial rupture of ligamentum teres.

• Labral tears • Femoroacetabular impingement • Chondral lesions • Septic arthritis • Loose bodies • Pigmented villonodular synovitis • Synovial chondromatosis • Ruptured ligamentum teres acetabulum, dense surrounding tissue envelope, and thick joint capsule were obstacles that surgeons had to overcome. With new instrumentation, positioning equipment, and surgical techniques, hip arthroscopy has allowed surgeons to access the hip joint with minimal neurovascular risk and lower morbidity than conventional open arthrotomy. There are several indications for hip arthroscopy (Box 21A-6). The surgeon must assess the patient’s limitations and ability to tolerate positioning and operation. Patients with external causes for restricted range of motion or inability to withstand hip traction, that is, osteoporotic bone, are not candidates for hip arthroscopy.118 The surgical technique of performing hip arthroscopy varies with surgeon preference. The patient may be positioned in lateral decubitus or supine. Lateral decubitus positioning is best performed with special traction equipment that may not be available to the surgeon. Supine hip arthroscopy can be performed with the patient on a standard fracture table with the hip abducted. Newer, specialized tables are available that allow for dynamic traction that is particularly useful in the care of FAI. Regardless of the patient position, traction across the hip joint is required to allow for the introduction of instrumentation. The traction should be in line with the femoral neck and requires the hip to be slightly abducted and flexed. The amount of traction varies with each patient, but enough force should be applied to distract the hip at least 8 mm. To achieve this traction, the foot is placed in a leg holder, and a perineal post or countertraction towel role is positioned within the perineum (Fig. 21A-16). Both areas need to be well padded and secured because traction forces may reach as high as 60 pounds and are a potential source of operative compli­ cations. The patient should not be in traction for more than 2 hours because of increased complication rates associated with traction.119 Image intensification should be available to ensure that adequate traction is applied and to assist with operations.120 Before making the incision, the surgeon must ensure the appropriate instrumentation is available. Compared with conventional arthroscopy, hip arthroscopy requires ­longer, stouter instruments. A 30- and 70-degree scope, long ­trocars, long spinal needles, and a variety of arthroscopic instruments that the surgeon may need to perform the procedure should be available (Fig. 21A-17). The actual performance of the procedure varies with different surgeons, but standard portals have been established.121 The anterior paratrochanteric portal is located 1 to 2 cm anterior and superior to the tip of the greater trochanter.

1474 DeLee & Drez’s Orthopaedic Sports Medicine

Figure 21A-16  Distraction setup in the lateral decubitus position for hip arthroscopy.

This portal is least likely to injure any neurovascular structures and can be used to visualize the femoral head, loose bodies, ligamentum teres, and acetabular lesions. The posterior paratrochanteric portal is 1 to 2 cm proximal and posterior to the tip of the greater trochanter. This portal is frequently used as the working portal for shavers, ­radiofrequency, and other arthroscopic instruments. Caution must be used when making this portal because it may cause damage to the superior gluteal nerve or sciatic nerve. The anterior portal is placed at a point intersected by a line connecting the tips of the greater trochanters and a line drawn distally from the anterior superior iliac spine. This portal allows for visualization of the medial aspect of the femoral head and medial labrum. Caution is used when making this portal because there is a risk for injuring the LFCN and femoral nerve.121 Postoperative rehabilitation is shorter than with open procedures. Most cases are performed as day surgical procedures without need for overnight stay. Weight-bearing status is dependent on the procedure performed. Physical therapy should be started within 1 week of the procedure. Progression of passive, active-assisted, and active range of motion is monitored. After motion has been regained, progressive resistance exercises are begun, with return to full activities within 2 to 3 months after the procedure.118-120

Figure 21A-17  Hip arthroscopy instrumentation.

The overall complication rate of hip arthroscopy has been reported to be as low as 1.3%.119 Direct neurovascular injuries can be avoided with appropriate portal placement. Traction injuries can occur from prolonged traction and aggressive force application. Traction should last no longer than 2 hours and only the minimum traction force necessary should be used. Pressure necrosis injuries to the perineum, scrotum, and foot have been well documented. These injuries can be avoided by minimizing time of traction and using an appropriate amount of padding. There have been reports of fluid extravasation into the thigh or abdominal cavity, leading to compartment syndrome.119 Capsular incisions should be minimized to prevent extravasation and arthroscopic fluid use should be monitored. A spectrum of iatrogenic joint damage may occur with hip arthroscopy and includes cartilage and labral damage.

PEDIATRIC INJURIES The skeletally immature hip and pelvis can sustain injuries that are unique to the pediatric population. Open growth plates are an area of weakness in the bone that are subject to injury. Vascular changes within the proximal femur can lead to Legg-Calvé-Perthes disease. The clinician should be aware that although the athlete may be skeletally immature, any injury that can occur in the adult may also occur in children.

Avulsion Fractures Avulsion fractures and muscle strain injuries involve a similar pathologic mechanism.122 Both are the result of a violent muscle contracture during eccentric loading. Unlike muscle strains, in which the injury occurs at the musculotendinous junction, the injury occurs at the site of origin or insertion of the muscle (Fig. 21A-18). The point of failure usually occurs at the physis rather than the tendon-bone interface. The injury generally occurs in young males.123 It has been postulated that increased androgen activity leads to relatively stronger muscles with weak physeal attachments that lead to avulsion fractures.40 The diagnosis is easily made by clinical examination and radiographs. Once the diagnosis is made, the initial fracture displacement is unlikely to increase. Four common sites of avulsion fractures around the hip and pelvis have been identified: anterior superior iliac spine, anterior inferior iliac spine, ischium, and lesser trochanter. Treatment of avulsion fractures is best done through conservative measures. Surgical treatment has not been proved to provide superior results. Metzmaker and Pappas40 have proposes a five-stage treatment program for avulsion fractures. The first stage involves resting the involved muscle with proper positioning, ice, and analgesics. This stage is started immediately and lasts 1 to 7 days. The second stage involves the initiation of physical therapy, including hip range of motion exercises, ultrasound, and electrotherapy. During the second stage, 7 to 20 days, isometric, stretching, and Thera-Band exercises for hip joint muscles and light endurance exercises (jogging) are used. The third phase is initiated when full motion has returned. During the third phase, a guided resistance exercise program is added to the treatment. The fourth phase involves the introduction

Hip, Pelvis, and Thigh 1475

the same area. The treatment for avulsions of the anterior inferior iliac spine should follow the Metzmaker and Pappas regimen outlined earlier.

Ischial Tuberosity Avulsion Fracture The ischial tuberosity undergoes a late ossification at the age of 25 years. Multiple hamstring origins involve the ischial tuberosity, and avulsion injury can result in loss of function of one or more hamstring tendons. Avulsion fracture occurs as the hamstring contracts with the hip and pelvis flexed and the knee in extension. The injury is common among gymnasts and track and field ­athletes.40,123 The treatment for an avulsion fracture of the ischial tuberosity should follow the Metzmaker and Pappas regimen outlined earlier. There have been reports of unsuccessful conservative management. Some clinicians have tried surgery for patients who have failed ­conservative management. Surgical attempts to debulk callus and use of internal fixation have been reported with successful results.124

Avulsion of the Lesser Trochanter

Figure 21A-18  Magnetic resonance image showing adductor avulsion injury.

of integration of stretching, strengthening, and patterned motions and is initiated when the patient has full range of motion and 50% of his or her strength. The fifth phase is the return to competitive activity and is gradually instituted. The entire process may take as long as 6 to 10 weeks.

Anterior Superior Iliac Spine Avulsion The anterior superior iliac spine undergoes ossification along with the anterior iliac crest at the age of 12 to 15 years. The sartorius muscle originates from this bony landmark, and violent contraction during running or jumping results in an avulsion fracture. Patients have pain with resisted flexion, abduction, and external rotation of the hip. Plain radiographs confirm the fracture of the anterior superior iliac spine. Treatment should follow the Metzmaker and Pappas regimen outlined earlier.

Anterior Inferior Iliac Spine The anterior inferior iliac spine ossifies earlier than the anterior superior iliac spine, and fractures usually occur in a slightly younger population. The rectus femoris muscle originates from the anterior inferior iliac spine, and ­avulsion can occur during forceful contraction of this muscle. Straight leg kicking has been a proposed mechanism for this injury.123 It is important to obtain contralateral ­radiographs to confirm avulsion injury and rule out os acetabulum, a nonpathologic ossicle that can occur in

The lesser trochanter undergoes ossification at about the age of 18 years. The common tendon of the iliopsoas muscle inserts onto the lesser trochanter and is a powerful hip flexor. The mechanism of injury for avulsion is thought to be forceful flexion of the hip with the hip maximally extended and internally rotated. Treatment of this lesion should follow the Metzmaker and Pappas regimen outlined earlier.

Iliac Apophysitis Iliac apophysitis is considered an overuse injury that has been described in the adolescent athlete. Patients complain of pain along the iliac crest, which is tender to palpation. There is no specific injury, and patients describe a gradual onset of symptoms. History may reveal a recent change in training or increase in intensity. Plain radiographs are often normal. As with most other apophysitis injuries, treatment consists of activity modification, cold therapy, anti-inflammatory medication, and physical therapy for muscle-stretching exercises. Return to full activity may be possible as early as 6 weeks. The patient must be aware of possible recurrence of symptoms.

Slipped Capital Femoral Epiphysis Slipped capital femoral epiphysis (SCFE) is a disorder that is exclusive to the skeletally immature. The proximal femoral physis fails because of a shearing mechanism. The femur is displaced anteriorly and is externally rotated relative to the femoral head. The disorder can be classified as chronic or acute as well as stable and unstable.125 A patient with an acute slip has symptoms for less than 3 weeks, whereas chronic slips can cause symptoms for months. Furthermore, with stable slips, patients may bear weight, whereas patients with unstable slips cannot bear any weight on the affected side. The incidence of avascular necrosis has been reported higher with unstable slips.126

1476 DeLee & Drez’s Orthopaedic Sports Medicine

The disorder can be bilateral in up to half of patients.127 SCFE tends to occur in males more than females, and affected patients tend to have a high body mass index.127 Clinicians should be aware of possible systemic illness such as thyroid and renal abnormalities that may cause atypical SCFE.

Evaluation Patients with SCFE complain of groin, thigh, and knee pain. Patients may also report hip stiffness. Proper history taking helps determine whether the injury is acute or chronic and stable or unstable. Physical examination may reveal limited range of motion, with the most common finding being obligatory external rotation with hip flexion. In acute slips, any motion of the hip may be painful. Diagnosis is confirmed with plain radiographs. Anteroposterior and frog-leg lateral radiographs of the pelvis can be used to quantify the amount of slip (Fig. 21A-19). A four-grade system exists for SCFE depending on the radiographic findings (Table 21A-6). MRI or bone scan is rarely needed to confirm the diagnosis.

Treatment Once SCFE has been diagnosed, treatment requires surgery. The first goal is to prevent further slippage. Most slips are amenable to single-screw fixation crossing the physis at a right angle. Grade III or IV slips may sometimes self-reduce during the positioning on the fracture table. In the uncomplicated slip, postoperative treatment includes limiting weight-bearing for 6 weeks in the stable slip and 12 weeks in the unstable slip. Controversy exists in terms of treatment of slips greater than 50% that do not reduce spontaneously. Investigators debate the need for, timing of, and type of osteotomy to prevent impingement and arthritis.128-130

Complications Complications of SCFE treatment include avascular necrosis and chondrolysis. Avascular necrosis of the femoral head can occur as a result of the initial injury or instrumentation of the epiphysis, or during the corrective osteotomy.

  TABLE 21A-6  Grading System for Slipped Capital Femoral Epiphysis Grade

Slip Percentage

I II III IV

<25 >25 and <50 >50 and <75 >75

Chondrolysis may be caused by technical errors during screw placement.131 Care must be taken to ensure the screw threads do not penetrate the articular cartilage of the femoral head. Postoperatively, the patient should be encouraged to range the hip and stimulate cartilage nutrition.

Legg-Calvé-Perthes Disease Idiopathic osteonecrosis of the femoral head in a skeletally immature patient, or Legg-Calvé-Perthes disease, is characterized by necrosis of the ossific nucleus of the femoral head secondary to occlusion of the arterial or venous blood supply. After infarction, healing occurs by creeping substitution and resorption of the dead bone. The resulting deformity of the hip joint may be extensive and can exceed the remodeling and healing capacity of the developing ­epiphysis.

Evaluation Patients with Legg-Calvé-Perthes disease present with pain in the groin, thigh, or knee region. Onset usually is insidious, and the patient may be symptomatic before presentation. Most children have a Trendelenburg gait. Internal rotation, abduction, and extension of the hip are limited. Plain radiographs are essential for diagnosis and can be predictive of outcome. Several classification systems have been proposed, including the Caterall, Herring, Salter, and Waldenstrom systems.132-134 All these classification systems used different graduated radiologic criteria to assist in predicting outcome. “Head at risk” signs have also been proposed to help predict outcome and include Gage’s sign (a radiolucent wedge defect on the lateral side of the epiphysis), calcification lateral to the epiphysis, lateral subluxation of the femoral head, a horizontal physis, and metaphyseal cysts.

Treatment Nonoperative

Figure 21A-19  Radiograph of slipped capital femoral epiphysis.

Treatment is based on containment of the avascular femoral head within the acetabulum to allow healing and remodeling of the affected cartilage.133 Restricted activity and limited weight-bearing can help control the ­symptoms and prevent further deformation of the femoral head. The most important factor for healing is age and amount of femoral head involvement; younger patients with little deformity have a better chance of normal cartilage development.135 During the treatment period, it is essential that range of motion of the hip is preserved and aggressive physical therapy is warranted. Treatment is continued

Hip, Pelvis, and Thigh 1477

until plain radiography shows reossification of the ossific nucleus and lateral coverage for the femoral head, and the patient is symptom free.136 Long-term risks in athletes with Legg-Calvé-Perthes disease include the risk for early arthritis secondary to the amount of residual femoral head collapse. Patients are prone to overuse syndromes because of joint incongruity.

Operative Surgical treatment during the acute disease can help to control symptoms and to contain the hip. Hip arthroscopy has been shown to be therapeutic for patients with LeggCalvé-Perthes disease.137 Removal of loose bodies and débridement of cartilage flaps provide most of the therapeutic benefit. Other operations include a wide spectrum of femoral osteotomies, acetabular osteotomies, and shelf procedures, with the goal of providing containment for the femoral head. C

r i t i c a l

P

o i n t s

l All hip and groin pain in the athlete is not a “groin strain.” Consider all possible causes of pain, including nerve, soft tissue, bone, and joint. l Hip pathology is location specific. Differential diagnosis must be tailored to the location of the patient’s pain and history of symptoms. l Therapy should be advanced in a stepwise fashion, with coaches, trainers, and therapists involved in the rehabilitation program. Premature return to play could lead to recurrent injury or chronic problems.

l A patient with the chief complaint of knee pain may indeed have an intra-articular hip disorder radiating to the knee. This is more often encountered in pediatric pathology, such as SCFE. l Currently, MRI arthrogram should be the study of choice after plain films to diagnose intra-articular pathology. Three-tesla MRI may be the next advancement in imaging technology. l At the time of MRI arthrogram, injection of lidocaine and steroid can be both diagnostic and therapeutic and can be done simultaneously with injection of contrast agent.

S U G G E S T E D

R E A D I N G S

Bharam S: Labral tears, extra-articular injuries, and hip arthroscopy in the athlete. Clin Sports Med 25(2):279-292, 2006. Byrd JWT: Operative Hip Arthroscopy. New York, Thieme, 1998. Kocher MS, Tucker R: Pediatric athlete hip disorders. Clin Sports Med 25(2): 241-253, 2006. McCarthy J: Early Hip Disorders. New York, Springer-Verlag, 2003. McCrory P, Bell S: Nerve entrapment syndromes as a cause of pain in the hip, groin and buttock. Sports Med 27(4):261-274, 1999. Metzmaker JN, Pappas AM: Avulsion fractures of the pelvis. Am J Sports Med 13:349, 1985.

R E F E R E N C E S Please see www.expertconsult.com

S e c t i o n   B

The Thigh Scott  Waterman

FEMORAL SHAFT STRESS FRACTURES Stress fractures are a common cause of pain in the athlete, occurring in up to 37% of runners.1 A stress fracture is caused by a repetitive force that in isolation is insufficient to cause an acute fracture but when cumulatively applied leads to fatigue failure of the involved bone.2 Clinical stress fractures result from repetitive mechanical stresses and loading, resulting in microfracture from the imbalance between repairing and remodeling. Upon initial training, osteoclastic resorption proceeds at a faster rate than osteoblastic formation, resulting in bone that is susceptible to microfracture.3,4 Originally this entity was diagnosed in military recruits, but with increased awareness and more rigorous training, stress fractures are increasingly being diagnosed in athletes,

especially runners and dancers.5-7 Most stress fractures occur in the tibia (33%), whereas the metatarsals (20%), navicular (20%), and femur (11%) are less frequently involved.8 Although proximal femoral stress fractures are commonly seen in clinical practice, femoral shaft accounts for 53% of all femoral stress fractures.9 Stress fractures often pose a diagnostic dilemma secondary to difficulty isolating the cause and source of the pain. Femoral stress fractures require a high index of suspicion for an early diagnosis, halting fracture progression and possible displacement.

Relevant Anatomy and Biomechanics Significant physiologic stresses are placed on the femur with physical activity. Anatomically, stress fractures can occur along the entire length of the femur.10,11 Oh and Harris12

1478 DeLee & Drez’s Orthopaedic Sports Medicine

reported that the greatest stress is located at the junction of the proximal and middle thirds of the femur, with forces up to 1200 pounds per square inch.13 In the sagittal plane, Tuan and associates noted that maximal stress occurs at the posterior proximal shaft.14 Multiple studies have shown that this area is the location of most stress fractures.11,15 Studies on military recruits, however, reported that 51% of femoral stress fractures occur in the distal third.16 The medial aspect of the femur at the junction of the proximal and middle thirds of the shaft serves as the origination of the vastus medialis and the insertion point for the adductor brevis. Repetitive loading of the proximal medial femoral shaft may result in a stress fracture. The lateral aspect of the femur is protected by contraction of the iliotibial band and vastus lateralis, which decreases tensile strain.11 The stress transferred to the bone is increased as muscles fatigue, resulting in stress concentration at the proximal to middle third of the femur.14 The cortex eventually weakens to the point of microfracture. Under subsequent repetitive cycling, the femur is unable to remodel, resulting in a stress riser and the development of a clinical fracture.

Classification Several classification systems have been proposed for diaphyseal femoral stress fractures to include biologic, scintigraphic, and magnetic resonance imaging (MRI) grading systems. Classification systems do not use plain radiographs because they lag behind symptoms for 2 to 3 weeks and are reported to miss 20% to 40% of symptomatic stress fractures.17-19 With lack of radiographic findings, advanced diagnostic imaging is commonly required. Multiple authors have classified stress fractures into high and low risk based on location.20-22 Low-risk fractures involve femoral shaft, ribs, ulnar shaft, medial tibia, and first through fourth metatarsals. High-risk stress fracture locations include the femoral neck, the patella, the anterior tibial diaphysis, the medial malleolus, the talus, the navicular, the proximal fifth metatarsal, and the first metatarsal phalangeal sesamoids.21 Low-risk fractures usually go on to unite with weight-bearing and activity modifications. High-risk stress fractures have an increased risk for progression, nonunion, and recurrence.6,20,21,23 The clinical grade of the fracture is relevant because it can help guide the duration and manner of treatment. The MRI grading system presented by Roub and colleagues requires both radiographs and short T1 inversion recovery (STIR) MRIs (Table 21B-1).19 STIR images are used because this allows for visualization of bone marrow edema while suppressing fat signal. Grade 1 stress injuries have normal radiographs with positive (increased signal intensity) STIR images. Grade 2 injuries have normal radiographs with positive STIR and T2-weighted images. Grade 3 stress injuries have periosteal reaction on radiographs with positive STIR, T1-, and T2-weighted images without a cortical break. Grade 4 injury radiographs show injury or periosteal reaction with positive cortical break on T1- and T2-weighted images. Arendt and colleagues reported that grade 1 and 2 injuries have an average recovery time of 3.3 and 5.5 weeks, respectively, whereas grade 3 and 4 injuries have a recovery time of 11.4 and 14.3 weeks, respectively.5

  TABLE 21B-1  Radiographic Grading of Stress Injuries Grade

Radiographic Findings

MRI Findings

1 2

Normal Normal

3

Periosteal reaction

4

Fracture line or periosteal reaction

Positive STIR image Positive STIR image, plus positive T2-weighted image Positive T1- and T2-weighted images; STIR without cortical break Fracture line on T1- or T2-weighted images

MRI, magnetic resonance imaging; STIR, short T1 inversion recovery.

Evaluation Clinical Presentation and History To help discern a cause, a detailed medical history must be obtained. This should include past medical problems, diet, workout schedule, changes in training regimens, onset of pain, exacerbating and alleviating factors, menstrual irregularities, and past injuries. The pain usually has an insidious onset, which initially worsens with physical activity. This often begins after a change in training regimen. A training error is the most common cause and risk factor for developing a stress fracture.7,23 The pain is frequently attributed to muscle soreness following activity, delaying clinical presentation for days to weeks after onset of symptoms. As the fracture progresses, activity is usually limited secondary to pain with a notable decline in performance. The location of pain must be ascertained to guide further history gathering, physical examination, and imaging studies. Stress fractures must be differentiated from spine, pelvis, and knee etiologies because these can also refer pain to the thigh. Pain secondary to femoral shaft stress fractures commonly presents at the fracture site or in the ipsilateral knee. Osteopenia must be considered a causative factor of stress fractures; therefore, dietary and menstrual histories need to be obtained. Eating disorders are more common in certain sports in which low weight, aesthetics, and weight categories are emphasized: distance running, gymnastics, dancing, and wrestling, respectively.22,24 Women with the “female athlete triad” (eating disorder, menstrual irregularities, and osteopenia) are at high risk for stress fractures secondary to their hypoestrogenic state.25 With peak bone mass occurring between 25 and 30 years of age and greatest acquisition between 11 and 14 years of age, delayed menses and low estrogen levels lead to decreased bone mass.25,26 Current increased levels of training in high school–aged athletes and inadequate dietary intake can be detrimental. Brukner and Bennell22 found the age of menarche in women to be an independent risk factor for the development of stress fractures, with the risk increasing by a factor of 4 for each additional year of age at menarche.

Physical Examination and Testing Physical examination can vary significantly depending on the stage at which the patient presents. During initial examination, there is usually a paucity of examination ­findings.

Hip, Pelvis, and Thigh 1479

Box 21B-1   Typical Findings in Femoral Shaft Stress Fractures Localized tenderness to palpation Pain with bending or torsional stresses Positive fist test Positive fulcrum test Positive hop test

Palpation may reveal a focal area of tenderness and, in more severe cases, swelling. Muscle mass and strength are usually similar to the contralateral side. Ranging the patient’s leg with logrolling and straight leg raising may elicit pain from a proximal shaft or neck fracture.27 Pain with bending and torsion can indicate a femoral shaft stress fracture. Distal femoral shaft and supracondylar stress fracture pain can be reproduced with both passive and active knee motion. Gait analysis may reveal an antalgic gait. The patient should be examined for possible risk factors, including leg-length discrepancy, rotational malalignments of the hip, excessive genu varum and genu valgum, and overpronation or cavus foot deformities.28 Three tests have been described for femoral stress fractures. Milgrom and associates29 described the fist test, in which the examiner applies pressure to the anterior thigh starting distally and working proximally on a seated patient. A positive test occurs when there is asymmetric pain. The fulcrum test refines the fist test by applying an anteriorly directed force with a hand on the posterior thigh in combination with a posteriorly directed force on the distal femur.15 This test is highly sensitive and aids in early ­diagnosis and follow-up.15,30 Monteleone31 described the hop test, in which the patient hops on the affected leg, reproducing pain. About 70% of patients with a positive hop test have a femoral stress fracture (Box 21B-1).31

Imaging Although relatively insensitive, plain radiographs are usually the first imaging study obtained. If obtained within 1 week of injury, only 10% to 50% of radiographs are positive.32,33 Two to 6 weeks after injury, radiographs may reveal periosteal reaction, callus formation, fracture, or lucency (Fig. 21B-1). Because these are late findings, other imaging studies are valuable for diagnostic purposes. Since the 1970s, bone scintigraphy has been the gold ­standard for diagnosis. This is a very sensitive test that is usually positive within 72 hours of injury.19 All three phases of a technetium-99m bone scan will be positive with a stress fracture (Fig. 21B-2). Roub and associates19 were able to diagnose femoral stress fractures in 20% to 40% of those with negative radiographs. However, bone scans have a 24% false-positive rate and poor anatomic resolution, and they expose the patient to a significant radiation dose.17,34,35 Computed tomography (CT) is infrequently used to make the diagnosis, but it is the best imaging modality for subperiosteal and endosteal bone formation and cortical detail. As a secondary imaging modality, CT can be

Figure 21B-1  Anteroposterior radiograph with periosteal reaction (arrow) on proximal medial femur. (From Kang L, Belcher D, Hulstyn MJ: Stress fractures of the femoral shaft in women’s college lacrosse: A report of seven cases and a review of the literature. Br J Sports Med 39:902-926, 2005.)

used to visualize a vertical fracture if present on multiple sequential cuts. MRI has become more widely used because of its high sensitivity and increased specificity compared with bone scans.1,8,36 MRI allows the visualization of bone marrow edema, especially on STIR sequence, which is often the first response to stress.37,38 MRI provides anatomic detail, soft tissue edema, and degree of injury, which are difficult to differentiate using other imaging modalities.24 Therefore, it is now becoming the imaging modality of choice and is the basis for the most common grading ­classifications.5

Treatment Options Once the diagnosis is made, femoral stress fracture treatment has three potential arms: prevention, nonoperative treatment, and operative treatment. Most of these stress fractures can be managed nonoperatively. Early treatment is important to prevent displacement, which portends poor results.39,40 Prevention requires the coaches, athletes, and possibly parents to understand the impact of overtraining, the importance of a balanced diet, and risk factors (Box 21B-2).

1480 DeLee & Drez’s Orthopaedic Sports Medicine

Operative Treatment Operative treatment is rarely required for femoral shaft stress fractures.6,10 Operative treatment is indicated only for failure of nonoperative treatment, fracture ­displacement, and nonunion. Intramedullary fixation is indicated in these rare circumstances.

Weighing the Evidence The literature on femoral stress fractures is replete with case reports and case series. There are few class I data on treatment, and much is inferred from work on stress ­fractures elsewhere. Most of the literature discusses diagnostic imaging and the classification systems.

Author’s Preferred Method

Figure 21B-2  Bone scan with increased activity in right proximal medial femur. (From Kang L, Belcher D, Hulstyn MJ: Stress fractures of the femoral shaft in women’s college lacrosse: A report of seven cases and a review of the literature. Br J Sports Med 39:902-926, 2005.)

Nonoperative Treatment Conservative treatment of femoral shaft stress fractures should start with protected weight-bearing for 1 to 3 weeks. During this time, pain should guide treatment and progression. The patient who remains asymptomatic is allowed to start noncontact activities such as deep-water jogging, swimming, or cycling to maintain conditioning.9,11 At the same time, a multidisciplinary approach should be taken to evaluate and treat causative factors, muscle weakness, hormonal imbalance, eating disorders, shoewear, training regimen, and anatomic alignment.28 Following pain-free weight-bearing and radiographic evidence of healing, a gradual return to training should be implemented. If the patient remains pain free after training, the patient may gradually return to competition. Arendt and Griffith reported that patients with early stress fractures were able to return to full activity after 4 to 6 weeks, whereas more severe stress fractures healed significantly later (12 to 14 weeks).5 Therefore, grading should be done at the time of diagnosis to help with patient counseling and guide length of treatment. Box 21B-2   Treatment Options in Femoral Shaft Stress Fracture Protected weight-bearing; change in activity Intramedullary fixation

After the diagnosis is made, protected weight-bearing is initiated, and the fracture is graded on MRI. Crutch use is initiated, with progression to full weight-bearing as long as the patient remains pain free. During this time, upper extremity strength training and non–weight-bearing conditioning are allowed, assuming the patient is pain free. When the patient is able to bear weight fully without pain, lower extremity conditioning is progressed to allow cycling and then aqua jogging or running and elliptical training. To return to running, the patient needs to be able to ­cycle hard for 35 to 40 minutes without pain. Light running should be initiated on soft ground or turf, with gradual progression back to the normal running surface and activity following a return-to-running protocol. During this progression, the cause of the stress fracture needs to be elucidated. This should include evaluation of the training regimen, especially to discover training errors. If this is the only risk factor noted, the patient should be allowed to progress as described previously. If the patient is found to have dietary abnormalities, a history of other stress fractures, multiple contemporaneous stress fractures, or any of the components of the female triad, the patient is in a high-risk group and should be further evaluated with a dual-energy x-ray absorptiometry scan. Other ­considerations should include evaluation of shoewear, orthotics, and gait. If the fracture displaces or there is failure of nonoperative therapy, intramedullary nail fixation should be considered to help return the patient back to activity. After surgery, the patient should start the nonoperative therapy protocol.

Postoperative Prescription, Outcome Measurement, and Potential Complications After femoral nailing, recovery should follow the protocol described previously.24 The patient should be allowed to bear weight as tolerated. Although a rare operative injury may be seen, the literature on this subject contains mostly class III and IV  data. Although only anecdotal, most authors believe that surgical treatment has a benign course as long as no displacement occurs, and most patients return to regular activity.

Hip, Pelvis, and Thigh 1481

  TABLE 21B-2  Return-to-Running Protocol Week

Activity

1 2 3

Walking Walking, jogging Running

Total Miles per Week 3-5 5 5

The main complication of femoral diaphyseal stress fractures is propagation of the fracture progressing to ­displacement.24,41 The incidence is not reported in the literature, nor is the risk for malunion, nonunion, or ­avascular necrosis.

Criteria for Return to Play On completion of the nonoperative treatment discussed earlier, a return-to-running program should be initiated to ensure an appropriate return to activity (Table 21B-2).41 This should be incorporated into the last phase of rehabilitation. After completion of the running program, a gradual return to regular training can be initiated.

Special Populations Only 3% to 9% of femoral stress fractures occur in children because pediatric femurs are more resistant to fatigue and have a larger osteogenic response.6,42,43 Children with femoral stress fractures usually present ­similarly to adults. The risk factors and location of fractures are similar to their adult counterparts.44,45 The only additional theoretical factor implicated in children is a change in the relative strength and flexibility, leading to asymmetric stress distribution.7,42 The radiographic appearance in children is slightly different from that in adults, with more abundant periosteal reaction and callus formation 1 to 3 weeks after the onset of symptoms.42,46,47 These patients should be treated similarly to adults with protected weight-bearing and a gradual return to activity as pain allows. MRI or bone scans should be obtained to help diagnose and differentiate from eosinophilic granuloma, osteoid osteoma, osteogenic sarcoma, and Ewing’s sarcoma.48,49

QUADRICEPS CONTUSIONS AND MYOSITIS OSSIFICANS Muscle contusions are the second most common athletic injury after muscle strains.50 These occur frequently in contact sports, such as football, rugby, and soccer, as a result of a blunt direct blow to the anterior or lateral thigh. ­Numerous studies have noted an average length of disability of 19 to 45 days.51,52 Ryan and colleagues reported a 9% incidence of myositis ossificans after quadriceps ­contusions.52

Relevant Anatomy and Biomechanics The quadriceps is composed of the rectus femoris and vastus musculature (lateralis, intermedius, and medialis), which lie anterior to the femur. The rectus femoris is the only biarticular muscle of this group, crossing both the hip and knee. The quadriceps musculature originates from the proximal femur. These muscles work together to extend the knee through the patella and patellar tendon. Quadriceps contusions result from a blow to the anterior or lateral thigh. The most common muscles contused are the vastus intermedius and lateralis. Although it was once believed that most muscle damage occurred deep, animal studies have revealed that the injury occurs primarily at the muscle surface, with deeper involvement as injury severity worsens.53 Animal models have shown that muscle contusion results in microscopic muscle breakage and damage, which results in edema and decreased muscle contractility.54 Beiner50 noted that contracted muscles are able to absorb 10% more energy than noncontracted muscles. In other animal studies, muscle recovery progresses through a characteristic series of events.55,56 Within 24 hours of injury, edema and hematoma formation occur at the sight of injury, resulting in swelling and limited knee flexion. As swelling starts to recede, intense proliferation of myoblasts and fibroblasts ensues. The wound bed undergoes granulation formation, which matures into dense collagenous scar tissue. Most athletes sustain mild injuries resulting in minimal symptoms following the contusion and continue to play. In severe contusions, the quadriceps can be partially or completely ruptured.

Classification C

r i t i c a l

P

o i n t s

l History often involves a recent change in activity. l Physical examination is often nonspecific. l Plain radiographs may miss up to half of femoral stress fractures. l STIR sequence MRIs need to be obtained to evaluate for bone marrow edema. l Immediate protective weight-bearing is instituted after diagnosis. l A multidisciplinary approach is taken to evaluate the cause of the stress fracture. l A slow return to activity follows a return-to-running protocol

Jackson and Feagin51 established a classification system based on knee flexion. They describe the muscle contusion as mild, moderate, or severe. Forty-eight hours after injury, mild contusions are able to flex greater than 90  degrees, moderate from 45 to 90 degrees, and severe less than 45 degrees (Table 21B-3). This classification helps guide counseling and treatment, and resolution occurs on days   TABLE 21B-3  Classification of Quadriceps Contusion Grade

Range of Motion

Mild Moderate Severe

>90 degrees 45-90 degrees <45 degrees

1482 DeLee & Drez’s Orthopaedic Sports Medicine

6.5, 56, and 72, respectively. There is no corresponding MRI classification system. Injuries classified as moderate to severe contusions were significantly more likely to develop myositis ossificans than were mild c­ ontusions.51,52

Evaluation Clinical Presentation and History The athlete usually presents within 24 hours of blunt isolated traumatic contusion with limited knee range of motion, thigh pain, and swelling.52 If the contusion is mild, the athlete can usually continue to compete or train. In moderate and severe cases, patients are unable to perform at their normal level secondary to pain and stiffness. If delayed in presentation, quadriceps contusions are likely to result in longer recovery time and a delayed return to competition.51,52,57

Physical Examination and Testing The physical examination should start with observation and palpation because the affected thigh is usually swollen with localized tenderness and edema (Box 21B-3). In severe cases, a knee effusion can be present. Alonso and colleagues57 noted that increased muscle firmness compared with the contralateral leg indicated a longer recovery time and the need for more physical therapy. They also reported that the location of contusion and degree of tenderness do not affect prognosis.57 The most important physical examination finding is range of motion at the knee because this is the main basis of the classification system and has prognostic implications for speed of recovery.52,57

Imaging Initial plain radiographs are often obtained to rule out bony pathology. Radiographs may have a role 2 to 4 weeks after injury if a firm mass is palpated to evaluate for developing myositis ossificans.58,59 Some report that ultrasound and MRI help the clinician evaluate the extent of the injury because of their high sensitivity and specificity for edema and hemorrhage.57,60 Although intuitive, Armfield and associates61 anecdotally noted that the greater the cross-sectional area of edema, the slower the recovery.

Box 21B-4   Treatment Options in Quadriceps Contusion Immobilization followed by progressive stretching and strengthening Surgical excision of myositis ossificans Compartment fasciotomy Hematoma evacuation

i­ mmobilization in an extended position, they noted that the last motion to return was knee flexion.51 Currently, most protocols recommend initial bracing at 120 degrees of knee flexion to place the quadriceps on passive stretch for 1 day in mild to moderate cases and 2 days for severe cases.52,62 As part of this initial period, bed rest, compressive dressings, and cryotherapy should be included to reduce swelling and hematoma formation. After this initial period, early flexion exercises should be started. Using their protocol, Ryan and colleagues52 reduced the time to recovery for the moderate and severe groups from 56 to 19 days and from 72 to 21 days, respectively. Following return to pain-free motion, patients are allowed to return to noncontact sport-specific training. Return to contact sports is allowed after the thigh firmness has resolved and the muscle is not tender to palpation. If the patient is involved in a contact sport, a thigh pad with ring should be recommended to minimize recurrence (Fig. 21B-3).52,62 Nonsteroidal anti-inflammatory drugs (NSAIDs) are often prescribed early to help decrease pain and prevent myositis ossificans.62,63 Consideration may also be given to low-dose radiation therapy for recurrent or severe contusions.63 Ryan and colleagues52 noted that risk factors for developing myositis ossificans are a delay of treatment greater than 72 hours and flexion less than 120 degrees on presentation. With the emphasis on obtaining early

Treatment Options Nonoperative Treatment As with most of the other conditions affecting the thigh, quadriceps contusions are treated conservatively (Box 21B-4). Although early studies recommended Box 21B-3   Typical Findings in Quadriceps Contusion Anterior thigh swelling, with or without knee effusion Ecchymosis over anterior thigh Pain with range of motion of knee

Figure 21B-3  Thigh pad with ring. (From Aronen JG, Garrick JG, Chronister RD, McDevitt ER: Quadriceps contusions: Clinical results of immediate immobilization in 120 degrees of knee flexion. Clin J Sport Med 16:383-387, 2006.)

Hip, Pelvis, and Thigh 1483

­ exion, Aronen and associates noted a decrease in myositis fl ossificans.62

Operative Treatment Operative management of quadriceps contusions is limited. Some authors have reported hematoma evacuation through aspiration or open decompression.64 However, there is no literature supporting its use. Attention must also be paid to thigh compartment pressures following this type of injury. Multiple case series have reported performing fasciotomies for quadriceps compartment syndrome, but not all agree with this treatment.64-66 None of the studies in which anterior fasciotomies were performed revealed any necrotic muscle or subsequent loss of motion.64-67 Robinson and coworkers68 treated six patients nonoperatively with anterior compartment pressures greater than 50 mm Hg and noted no loss of motion or muscle weakness at 1-year follow-up.

Weighing the Evidence Jackson and Feagin’s range of motion classification system should help guide counseling and prognosis.51 Ryan and colleagues’52 and Aronen and associates’62 studies, although they had no control groups, revealed that immediate immobilization in flexion with early flexion exercise improves symptoms and hastens return to play52,62 and minimizes heterotopic ossification. There are no controlled studies to help critically evaluate treatment options. Most of the literature on this subject is limited to case series with only anecdotal treatment protocols.

Author’s Preferred Method Treatment of these injuries should be initiated as soon as the quadriceps contusion is diagnosed. This should initially be treated with a mild compressive dressing and ice. The knee should be passively flexed and immobilized with an Ace bandage on the field or in a hinged knee brace locked in flexion as tolerated. The athlete should then be encouraged to remain at bed rest with the leg elevated to decrease swelling and hematoma formation. If pain is severe enough, the patient should be hospitalized for observation and pain control. NSAIDs, typically Indocin, are initiated for 7 to 10 days during this period to help relieve discomfort and prevent heterotopic ossification. After 24 hours of rest, pain-free active-assisted flexion exercises should be initiated with the goal of greater than 120 degrees of flexion. After return of full motion and strength, noncontact training is initiated, usually after 10 to 14 days. If the athlete remains asymptomatic, he or she is allowed to return to training and competition 3 to 4 weeks after injury. In contact sports, a thigh pad should be worn through the rest of the athletic season to prevent recurrence. If return of motion is still limited 2 to 3 weeks after ­injury, other modalities, including heat and ultrasound, should be incorporated into the treatment protocol.

Postoperative Prescription, Outcome Measurement, and Potential Complications If a hematoma is drained or decompressed, a compressive dressing and bed rest are implemented to prevent recurrence. After decompression, the patient should be started on the nonoperative protocol described previously. Early range of motion needs to be initiated. The complications of quadriceps contusion are myositis ossificans and anterior thigh compartment syndrome. Although the cause of myositis ossificans is not fully understood, Jackson and Feagin51 noted an increased incidence with severe contusions. This results from muscular and soft tissue disruption leading to heterotopic ossification and occurs in 9% to 14% of contusions.69-71 Myositis ossificans is usually suspected with persistent soft tissue swelling and pain that does not resolve after 4 to 5 days.50 Patients may also report increased induration and loss of motion 2 to 3 weeks after the injury (Box 21B-5).55 Ryan and colleagues reported that myositis ossificans can be prevented by obtaining early range of motion greater than 120 degrees; however, they reported an incidence of myositis ossificans of about 10% in all three grades.52 Elevated sedimentation rate and serum alkaline phosphatase, although nonspecific, can indicate myositis ossificans.53 NSAIDs or bisphosphonates, when started early, are believed to prevent myositis ossificans.55 There have been no controlled studies on NSAIDs and prevention of myositis ossificans; however, prevention has been inferred from studies on prevention of heterotopic bone.72,73 Radiographs do not show heterotopic ossification for 2 to 4 weeks after the injury (Fig. 21B-4).74 After 6 to 8 weeks, radiographs reveal a lacy pattern of new bone formation with the periphery corticallized.74 Although bone maturation may take up to 1 year, by 4 to 6 months, lamellar bone has usually formed, and resorption may start to occur.75 On radiographs, ­myositis ossificans has a calcified edge with a radiolucent center.76,77 Bone scans can be used to both detect and evaluate the maturity of the myositis ossificans. Bone scan may turn positive 4 to 6 weeks before radiographic ossification.78 Triple-phase bone scans may also be used to assess the maturity of the heterotopic ossifications because mature bone appears negative on the pool and flow phases, and positive only on the delayed phase.78,79 In immature bone, all three phases are positive. Recovery of normal activity and strength is possible in the presence of myositis ossificans. Patients with myositis ossificans should be treated as described previously with early motion and strength training. Although rarely indicated, surgical excision should be performed to remove the myositis ossificans only if it is prominent, predisposes the area to further injury, or limits range of motion.50 Excision should not be performed before bone maturity, usually 6 to 12 months, to prevent significant recurrence.55,75 The presence of myositis ossificans is a harbinger of slower rehabilitation, and patients should be instructed that it may take months to return to preinjury state (Boxes 21B-5 and 21B-6).71,80,81 Anterior thigh compartment syndrome is also a rare complication of sports-related quadriceps contusion, with only 21 cases reported in the literature.82 Compartment syndrome is defined by Mubarak and associates83 as “high

1484 DeLee & Drez’s Orthopaedic Sports Medicine

Box 21B-5   Typical Findings in Myositis Ossificans

Box 21B-6   Treatment Options in Myositis Ossificans

Persistent soft tissue swelling Decreased motion 2 to 3 weeks after injury

Early motion and strength training NSAIDs and bisphosphonates Surgical excision

pressure within a closed fascial space that reduces blood perfusion below the level necessary for tissue viability.” Compartment syndrome is thought to be less common in the thigh secondary to the large potential space. This condition is a clinical diagnosis of pain out of proportion to clinical findings. Many different absolute pressures have been put forward to help diagnose compartment syndrome, and compartment pressures should be used only as an adjunct to clinical examination.84 Compartment pressures of 30 mm Hg, 45 mm Hg, and less than 30 mm Hg from diastolic pressure are considered compartment syndromes in different studies.85-87 Most authors recommend emergent fasciotomies to alleviate pressure, with delayed skin closure versus skin grafting. However, in the case of thigh compartment syndrome for a sports-related contusion, surgery may not be needed. Several authors have reported nonoperative treatment of thigh compartment syndrome with compartment pressures greater 50 mm Hg, noting no adverse sequelae with return to normal strength and range of motion.64-66,68

Criteria for Return to Play Patients are allowed to return to noncontact training after full range of motion and nearly symmetrical strength are obtained.60,62 When full function is recovered, the patient is allowed to return to competition with a hollow thigh pad to protect the contused area (Box 21B-7).52

Special Populations Adolescents with quadriceps contusion present, similar to adults, after blunt trauma to the thigh, usually while participating in contact sports. After this trauma, a hematoma and muscular edema develop, resulting in thigh swelling.88 Patients with mild to moderate quadriceps contusion may continue to play, although severe contusion may present with knee effusion and pain with knee movement. Physical examination reveals thigh ecchymosis, swelling, and pain with motion of the thigh. Treatment is initiated with immobilization of the knee in 120 degrees of flexion and 24  to 48 hours of rest, ice, compression, and elevation. This should be followed by progressive stretching and strengthening as detailed previously.88 C

r i t i c a l

P

o i n t s

l Patients with moderate to severe quadriceps contusion usually present within 24 hours of the inciting event. l Range of motion should be obtained to classify severity of contusion and determine prognosis.­ l Initial radiographs are useful to rule out bony injury; however, myositis ossificans cannot be visualized for 2 to 4 weeks. l If diagnosed early, the knee should be immobilized in 120 degrees of knee flexion for 24 hours. l Pain-free, active range of motion should be initiated after an initial rest period with the goal of greater than 120 degrees of knee flexion. l Surgery is indicated only for extreme cases of thigh compartment syndrome and rarely for heterotopic ossification resection.

Box 21B-7   Return to Play Ossificans Figure 21B-4  Myositis ossificans of the quadriceps. (From Quek ST, Unger A, Cassar-Pullicino VN, Roberts SNJ: A self limiting tumour. Ann Rheum Dis 59:252-256, 2000.)

in

Full range of knee motion Near symmetrical strength of motion Wear thigh pad

Myositis

Hip, Pelvis, and Thigh 1485

MUSCLE STRAINS AND RUPTURE Relevant Anatomy and Biomechanics Muscle strains are the most common athletic injury, rep­ resenting 30% to 50% of all injuries.50,89,90 Most strains occur at the myotendinous junction in fast-twitch type 2 muscle fibers of biarticular muscles undergoing an eccentric contraction.91 Although most strains occur at this interface, muscle strains can occur anywhere along the length of the muscle.92,93 While following a professional soccer team, Volpi and colleagues94 noted that 32% of strains involved the quadriceps, 28% hamstring, 19% adductor, and 12% gastrocnemius. The thigh contains a large cross-sectional area of muscles with three main muscle groups: hamstrings, quadriceps, and adductors. The hamstring consists of the biceps femoris, semitendinosus, and semimembranosus. The semitendinosus, semimembranosus, and long head of the biceps are biarticular and are innervated by the tibial portion of the sciatic nerve. The short head of the biceps is monarticular and is innervated by the common peroneal nerve. These muscles work together to extend the hip, flex the knee, and externally rotate the hip and knee.61 There is significant overlap of the myotendinous junctions of the hamstrings (Fig. 21B-5). The most common cause of hamstring strains is bending forward while sprinting, such as diving for a ball or coming out of the blocks in track.95 The quadriceps consists of the rectus femoris and the vastus musculature (medialis, lateralis, and intermedius) (Fig. 21B-6). The only biarticular muscle is the ­rectus femoris, but all receive innervation from the femoral nerve. The indirect head of the rectus femoris gives rise

Figure 21B-6  Cross-sectional anatomy of thigh. 1, Vastus lateralis; q, quadratus femoris; m, vastus medialis; i, vastus intermedius. (From Armfield DR, Kim DH, Towers JD, et al: Sports-related muscle injury in the lower extremity. Clin Sports Med 25:803-842, 2006.)

to a central tendon in the proximal thigh, allowing strains to occur proximal to the musculotendinous junction. The primary function of these muscles is knee extension.61 The most common cause of quadriceps strains is kicking while ­running.95 The adductor muscles consist of three muscular structures: superficial (pectineus, gracilis, and adductor longus), middle (adductor brevis), and deep (adductor magnus).96 The adductor musculature receives innervation from the obturator nerve, except for the pectineus, which is innervated by the femoral nerve. The primary actions of these muscles are to adduct the thigh and stabilize the lower extremity in the closed kinetic chain. The most common cause of adductor strain is a sudden change in direction, such as a cutting maneuver to avoid being tackled (see Figs. 21B-5 and 21B-6).95

Classification The clinical classification system depends on the severity of the muscle injury: mild, moderate, or severe (Table 21B-4). Mild (grade I) sprains involve tearing of a few muscle fibers with mild pain and minimal loss of strength. Moderate (grade II) sprains involve increased tearing of muscle fibers with some strength loss. Severe (grade III) sprains include tearing of the entire muscle with complete loss of strength.97,98 More commonly in adolescents and a small subset of adults, the tendons of origin or insertion may be avulsed and are classified as a grade IIIB (see Table 21B-4).

Semitendinosus muscle

Biceps femoris muscle Semimembranosus muscle

  TABLE 21B-4  Strain Classification Grade

Findings

I

Small disruption of structural integrity at ­musculotendinous junction Partial tear, some musculotendinous fibers remain intact Complete rupture of musculotendinous unit Avulsion fracture at tendon’s origin or insertion site

II IIIA IIIB Figure 21B-5  Overlap of the hamstring tendons.

1486 DeLee & Drez’s Orthopaedic Sports Medicine

Figure 21B-7   Biceps placed on stretch with the taking-off-shoes test. (From Zeren B, Oztekin HH: A new self-diagnostic test for biceps femoris muscle strains. Clin J Sport Med 16[2]:166-169, 2006.)

Hamstring Strain Evaluation Clinical Presentation and History Patients with hamstring strain present acutely or chronically. They usually report a sudden pain in the posterior thigh while sprinting or jumping. Many studies have shown that semimembranosus and semitendinosus strains occur in late swing phase as the hamstrings eccentrically contract to slow knee extension. Biceps femoris strains more commonly occur in early toe-off. The biceps ­femoris is the most commonly strained hamstring muscle. Patients with grade I injuries usually do not seek medical attention, but those with grade II and III strains report acutely with pain, loss of strength, transient sciatica, and posterior thigh tenderness.99 The patient should be asked about adequacy of warm-up, muscle fatigue, preseason training, age, type of sport, and history of prior hamstring injury.100-102 Athletes participating in sports that require short bursts of sprinting, such as football, track, and soccer, are more likely to sustain hamstring strains.89,101,103,104 Lack of conditioning, inflexibility, and inadequacy of warm-up are highly associated with strains.101,102,105 Woods and coworkers106 noted that 62% of hamstring strains occur in competition, compared with only 38% during training. This incidence of strains was noted to increase at the ends of halves of play Box 21B-8   Typical Findings Strains

in

and after prolonged training.106 The patient who presents late usually notes only mild pain with daily activity, which is accentuated with training, as well as muscle tightness and decreased performance and strength. Previous hamstring injuries should also be investigated because there is a 34% recurrence rate of hamstring strains.107

Physical Examination and Testing The involved area is usually slightly swollen, tender, and possibly ecchymotic. This pain can be accentuated with 90 degrees of hip flexion or resisted knee flexion.108 The patient should be placed in a prone position with the knee

Hamstring

Swollen posterior thigh Localized tenderness usually over musculotendinous ­junction Pain worsens with hip flexion and knee extension Positive taking-off-shoes test

Figure 21B-8  Cross-sectional anatomy of the distal thigh. bf, biceps femoris; g, gracilis; s, sartorius; sm, semimembranosus. (From Armfield DR, Kim DH, Towers JD, et al: Sports-related muscle injury in the lower extremity. Clin Sports Med 25:803-842, 2006.)

Hip, Pelvis, and Thigh 1487

Figure 21B-9  Central tendon rupture (straight arrow) with adjacent hematoma (curved arrow). (From Connell DA, Schneider-Kolsky ME, Hoving JL, et al: Longitudinal study comparing sonographic and MRI assessments of acute and healing hamstring injuries. AJR Am J Roentgenol 183:975-984, 2004.)

extended to place the muscle on slight stretch. The most proximal portion of the tenderness needs to be ascertained because this has a direct prognostic effect. The more proximal the injury to the ischial tuberosity, the slower the return to previous activity.109 Askling and colleagues109 noted that this is as sensitive a prognostic factor as MRI findings during the first 3 weeks after injury. Although present in only more severe injuries, palpable knots and defects should be recognized because they also portend a slower recovery. With serial examination, the severity and progression to recovery can be determined. Zeren and coworkers110 reported a taking-off-shoes test. During this test, the patient is required to stand and remove both shoes using the feet only. To accomplish this examination, the leg is externally rotated 90 degrees, and the knee is flexed to 20 to 30 degrees, placing the biceps femoris parallel to the posterior thigh and resulting in maximal contraction (Fig. 21B-7). The patient notes a sharp pain in a positive test. These authors report this test to be 100% sensitive and specific in diagnosing hamstring strains (Box 21B-8).

Imaging Diagnostic imaging is not needed for evaluation of most muscle strains because these are treated conservatively.111 Verrall and associates108 noted that even MRI misses 14% of clinical hamstring strains. Plain radiographs and CT have a limited role in evaluating muscle strains but are Box 21B-9   Treatment Options in Hamstring Strains RICE followed by progressive stretching and ­strengthening Surgical myotendinous junction and avulsion repair Compartment fasciotomy Hematoma evacuation

often employed to rule out bony pathology and evaluate for ischial avulsion fractures.61,112 Ultrasound and MRI are the main modalities for evaluation. Ultrasound is popular at some institutions secondary to its accessibility, cost, and portability.92 The cross-sectional area of muscle injury confirmed by ultrasound had a poorer prognostic value.92 Connell and associates92 reported that in the acute injury, ultrasound is as sensitive as and more cost-effective than MRI. Although widely acknowledged to be operator dependent, ultrasound does have some other drawbacks. These include the inability to assess deep structures secondary to sound wave dissipation, poorer reproducibility, and the decreased sensitivity with subacute and chronic injuries.61,108 In comparison with ultrasound, MRI is a modality with good spatial resolution, tissue contrast, and reproducibility (Fig. 21B-8).61 Connell and associates92 reported that the length of injury on MRI has prognostic implications with larger involvement portending a slower return to play. MRI is a better modality to image subacute or chronic injuries and for serial examination. Verrall and coworkers108 reported that patients with MRI-confirmed hamstring strains had a poorer prognosis with a slower return to competition. Although MRI has been used to assess extent of injury and possibly the need for surgical repair, the information obtained from the MRI rarely changes the treatment (Fig. 21B-9).111 Therefore, the cost precludes routine imaging of athletes with this injury.

Treatment Options Nonoperative Treatment Rest, ice, compression, and elevation (RICE) are the mainstay of initial hamstring strain treatment (Box 21B-9). This regimen helps decrease hemorrhage and the inflammatory process.103 After the initial 3 to 7 days, most authors recommend progressive range of motion exercises.101,103,113,114 Although controversial, NSAIDs are frequently initiated during this period. Levine and colleagues115 recently reported a faster return to competition after steroid injections for severe strains.101 This has not been studied in a blinded, controlled trial. As swelling improves, a second phase should be initiated when pain-free full motion is obtained, including submaximal isometric contractions and stretching.101,113,116,117 When strength has returned with concentric contractions, eccentric exercises should be initiated. Eccentric contraction should be avoided in the initial phases because this creates more stress and could potentiate the injury. After starting eccentric contractions, hamstring strength and flexibility should start returning to preinjury condition, and the patient may start pain-free jogging exercises.103 Following pain-free sport-specific activities, the patient is allowed to return to competition.118 With initial recovery completed, interventions should be made to prevent recurrent strains. Much of the recent literature indicates that a history of hamstring injuries, inflexibility, agonist-antagonist strength imbalance, and increasing age are risk factors for recurrence.95,101,118 Croisier and colleagues119 reported a risk factor for ­hamstring injury of 15% with strength imbalance, which decreases to 3% when muscular forces are balanced.

1488 DeLee & Drez’s Orthopaedic Sports Medicine

Operative Treatment Surgical intervention is indicated only in failure of nonoperative management and complete proximal or distal attachment rupture. For a displaced avulsion fracture greater than 2 cm, operative repair is indicated to avoid long-term disability.113,120 Chakravarthy and coworkers,121 in their series of four patients, noted improved outcome with surgical repair of the hamstring origin, with all patients regaining full strength and returning to previous level of competition. Injuries to the distal hamstring insertions are associated with severe knee injuries and posterior lateral corner injuries, and are covered in that section.113 Recently, Schilders and associates122 described a tenotomy for treatment of a chronic partial semitendinosus rupture that failed nonoperative management.

Postoperative Prescription, Outcomes Measurement, and Potential Complications After surgical repair of the hamstrings, Chakravarthy and colleagues121 recommend 10 to 14 days of splinting in 90 degrees of knee flexion. This should be ­followed with isotonic prone knee flexion from 90 degrees to full flexion and passive extension as tolerated. A thermoplastic splint should be modified as needed to allow a safe motion arc. The patient should then be started on a progressive stretching and strengthening program starting in phase II of the previously described ­protocol. Cross and ­associates123 recommended that the knee be ­immobilized for 8 weeks after repair of the avulsion. Although most papers discussing the treatment of hamstring strains are class III or IV data, the identification of risk factors and prevention of hamstring strains has been

Author’s Preferred Method After the diagnosis is made, a five-stage program is initiated. This regimen requires intermittent monitoring of range of motion, pain, and muscle testing. The progression can be expedited or slowed depending on the individual patient (Table 21B-5). Phase I starts with the standard RICE regimen and clinical examination to determine strain severity. A brief period of NSAIDs is initiated in all hamstring strains to help ­control pain and alleviate swelling unless contraindicated. Pain-free passive or active-assisted range of motion should be initiated early to maintain motion. If severe, the patient may be placed on a short period of crutch ambulation, which should be discontinued when the patient is able to tolerate weightbearing. During this phase, the proximal extent of the pain, as well as any palpable knots or defects, is determined to guide rehabilitation and establish prognosis. Progression through to phase V should be dictated by pain, range of motion, and functional testing. Phase II should continue to progress and restore full active range of motion. Once obtained, pain-free prone submaximal isometric contractions, static stretching, and pool therapy can be initiated. Electrical stimulation may be initiated to help prevent atrophy and control pain. Pain-free swimming and upper body conditioning should be initiated to maintain conditioning. Phase III allows the patient to start pain-free cycling and the progression to prone isotonic contractions. Care must be taken to ensure that the machines are set on the concentric mode because eccentric contractions may place undue stress, resulting in recurrent strains. Concentric exercises are initiated first with progression to eccentric contractions as tolerated. Static stretching is also slowly progressed to active stretching. During phase IV, proprioceptive neuromuscular facilitation and eccentric isokinetic contractions are initiated. When the patient is able to tolerate slower isokinetic speeds, sport-specific training can commence. This should include a progression from light jogging through to sprinting. When able to complete sport-specific training, the patient may be allowed to return to competition.

For mild strains, symptoms are usually only pain or spasm with activity, and these patients may return to competition within 3 to 7 days. Grade I strains should be advanced to phase III to prevent deconditioning and excessive exclusion from play. Moderate and severe strains, consisting of partial or complete rupture, result in pain and possibly spasms with activity. The patient usually feels a tearing sensation or pop in the posterior thigh, which results in a significant decrease in performance or inability to continue competition. In these patients, return to sport is usually delayed 3 to 6 weeks to allow for muscular healing and return of flexibility. Surgery is indicated only for those with complete avulsion fractures with displacement greater than 2 cm. This should include a posterior approach to the ischium with anchoring of the tendon to its anatomic origin. Intraoperatively, a safe arc of motion should be assessed. After surgery, the patient is splinted at 90 degrees of knee flexion for 10 to 14 days. This is followed with prone isometric knee flexion from 90 degrees to full flexion with passive extension. The patient should be placed in a thermoplastic splint while not in a therapy session. The patient is then started on phase II of the above treatment protocol.   TABLE 21B-5  Rehabilitation Protocol for Hamstring Strains Phase

Activity

I II

RICE, NSAIDs, passive ROM Ice, passive stretching, NSAIDs, electrical ­stimulation, isometrics, conditioning Ice, active stretching, ± NSAIDs, ± electrical stimulation, isotonics ± isokinetics, ­conditioning Ice, stretching, isokinetics, running, ­sport-specific training Return to sports, preventive exercises

III IV V

NSAIDs, nonsteroidal anti-inflammatory drugs; RICE, rest, ice, ­compression, and elevation; ROM, range of motion.

Hip, Pelvis, and Thigh 1489

Box 21B-10   Return to Play in Hamstring Strains Isokinetic strength within 15% to 20% of contralateral side Sports-specific training that includes interval sprinting and figure-eight running Flexibility that is nearly symmetrical Wear thigh sleeve on return to competition

more rigorously evaluated. Proske and coworkers124 placed an Australian football club on an eccentric strengthening program that reduced the hamstring injury incidence from 16 to 2 over a 2-year period. Verrall and associates118 followed another Australian football club for four seasons while implementing a functional training program and tracked the incidence of injuries, the convalescence period, and the number of games missed. They noted that there was a fourfold decrease in competition injuries as well as a decrease in the total players injured and games missed secondary to injury. Complications of hamstring strains include recurrence, residual weakness, myositis ossificans, and compartment syndrome. Traditional treatment of hamstring strains has resulted in nearly a one-third reinjury rate.118 The incidences of myositis ossificans, compartment syndrome, and residual weakness from hamstring strains are unknown.

Criteria for Return to Play The main goal in treating athletes with hamstring strains is to return them to play as soon as possible while preventing recurrence. The patient should progress through the previously described phases of rehabilitation followed by functional exercises. Once the patient has graduated through these steps, isokinetic and functional testing should be conducted. Isokinetic strength should be within 15% to 20% of the contralateral side. The quadriceps and hamstrings

strength should be evaluated to determine whether there is an imbalance of greater than 40%. ­Flexibility should also be evaluated as a risk factor for recurrence. If muscular imbalance or inflexibility is noted, the athlete should be held out of competition until these are corrected. If isokinetic testing is unavailable or in conjunction with strength testing, sport-specific testing should be conducted to include three 30-yard sprints as well as figure-eight running drills. If the patient is able to complete these at full speed and without pain, they may be returned to competition (Box 21B-10).

Special Populations Hamstring injuries in skeletally immature patients are uncommon. Secondary to the imbalance between stronger musculotendinous junctions and weaker open apophyses, adolescents are more likely to sustain avulsion fractures (Fig. 21B-10).88,125 A combination of hip hyperflexion with knee extension commonly results in apophyseal avulsion fractures.99,120 Although grade I strains usually continue with activity, grade II and III present for clinical evaluation secondary to pain and decreased sporting ability. As with adults, most adolescents with hamstring strains report an acute onset of posterior thigh pain and occasionally an audible pop with activity, especially sprinting.88 Physical examination reveals swelling, ecchymosis, and pain with knee flexion. The entire length of the hamstring musculature, as well as the hamstring’s tendinous origin off the ischium, should be examined for a palpable defect. If concerned about an avulsion fracture, radiographs should be obtained to document displacement. If plain radiographs are not diagnostic and a complete tear or avulsion fracture is suspected, an MRI should be obtained. Acute injuries are more likely to be noted on STIR or T2-weighted imaging. Treatment is similar to that for adults. See the previous discussion for further details. Surgical intervention is indicated only for avulsion fractures with greater than 2 cm of displacement because this can result in chronic weakness and pain. These injuries should be repaired back to their anatomic tendinous origins. C

Figure 21B-10  Ischial avulsion fracture. (From Akova B, Okay E: Avulsion of the ischial tuberosity in a young soccer player: Six years follow-up. J Sci Med Sport 1:27-30, 2002.)

r i t i c a l

P

o i n t s

l Hamstring strains usually occur during repeated episodes of sprinting during late swing and early toe-off. l A prior history of hamstring injuries, lack of conditioning, increasing age, inflexibility, and inadequacy of warm-up increase incidence of strains. l After a hamstring strain, there is a 34% risk for reinjury. l Classification by physical examination and imaging modalities should be completed within 2 weeks of injury to help aid in treatment and establish prognosis. l MRI or ultrasound should be used to evaluate the injury if diagnosis is unclear. l RICE is the mainstay of initial treatment. l A slow progression of stretching is achieved, starting with isotonic and progressing to isokinetic and isometric. l Surgery is indicated only for proximal or distal attachment rupture. l Return to play should be allowed after sport-specific ­training and regaining of strength and flexibility.

1490 DeLee & Drez’s Orthopaedic Sports Medicine

Adductor Strain Evaluation Clinical Presentation and History Adductor strains are the most common groin injuries in athletes.126,127 Patients usually present with acute groin pain after either kicking sports or sports with sudden changes in direction, such as soccer, hockey, rugby, martial arts, and football. Adductor strain incidence in European hockey and soccer players is 13% to 43%.27,128-130 The eccentric contraction occurs with the hip in external rotation, and abduction is hypothesized to cause the adductor strain. The adductor longus is the most commonly injured muscle.127,131 The patient usually complains of medial thigh or groin pain, which is accentuated with hip abduction. The complaint could be from minor pain with activity to weakness and significant pain limiting or precluding training. History should also include aggravating or alleviating factors to help differentiate from the other causes of groin pain: osteitis pubis, athletic pubalgia, hernias, and stress fractures. A thorough history is important because groin pain in 27% to 90% of athletes has more than one cause.132,133 Inflexibility, previous injury, and strength imbalance between adductors and abductors have been implicated as risk factors for adductor strains. Ekstrand and colleagues129 noted that soccer players with decreased abduction range of motion sustained more strains. Other studies refute these findings, noting that flexibility made no difference in the incidence of adductor strains.134,135 Strength imbalances between opposing muscle groups has been noted to be associated with a increased risk for strains of the weaker muscle group.134 Seward and coworkers136 and Tyler and associates134 report recurrence rates of 32% and 44%, respectively. The highest risk factor for adductor strain is an adductor-to-abductor strength of 80% or less.137 ­Multiple other studies have reinforced these findings, noting a decreased adductor strength as a risk factor for recurrent strains.95,138

Physical Examination and Testing The involved area is usually slightly swollen and tender. This pain can be accentuated with resisted hip adduction. The area most commonly affected is the musculotendinous junction of the adductor longus. Therefore, palpation from the proximal insertion of the muscle belly on the pubic rami distally should elicit the maximal site of tenderness. With some resistance, the strain grade can be determined, and any muscular defects can be palpated. Mild (grade I) strains result in pain with minimal strength loss, moderate (grade II) strains result in strength loss and pain, and severe (grade III) strains result in complete functional and motor loss (Fig. 21B-11). Adductor strains must be differentiated from other sources of groin pain. Tenderness to palpation at the pubis may result from osteitis pubis, athletic pubalgia, or muscular origin avulsion injury. Osteitis pubis results in tenderness over the pubic tubercle or inguinal ring and can often be differentiated because symptoms

Figure 21B-11  Adductor retraction.

­exacerbate with sit-ups.139,140 Athletic pubalgia may have tenderness over the pubic tubercle or conjoined tendon or at a dilated superficial inguinal ring.91 This pain can be reproduced with sit-ups or Valsalva maneuver. ­Inspection for an inguinal hernia should also be ­performed (Box 21B-11).

Imaging Diagnostic imaging is not needed for evaluation of most muscle strains because these are treated conservatively.111 A significant amount of information has been inferred from studies on hamstring strain. Plain ­radiographs are often used to evaluate for other causes, such as ­muscular origin avulsions, osteitis pubis, and stress fractures (Fig. 21B-12). Bone scans have been used to differentiate muscular avulsion injuries and early osteitis pubis from other conditions but have been supplanted by MRI because of the high sensitivity but low specificity of bone scans. Ultrasound and MRI are the main modalities for ­evaluation. Ultrasound has been used to evaluate groin pain and differentiate strain from other conditions, including hernia.141 In a recent review, Bianchi and colleagues142 anecdotally preferred MRI because of difficulty differentiating tendons at their origin with ultrasound. MRI, with higher sensitivity and specificity compared with other imaging modalities, is the imaging method of choice when examining chronic groin pain of unknown etiology, grading strains, and evaluating suspected grade III strains (Fig. 21B-13).143

Box 21B-11   Typical Findings in Adductor Strains Localized tenderness to palpation over medial thigh or groin Pain with resisted abduction Ecchymosis, thigh swelling With or without palpable defect

Hip, Pelvis, and Thigh 1491

Figure 21B-12  Radiograph of osteitis pubis.

As with hamstring injuries, the cost of the MRI precludes its routine use for evaluation of adductor strains. Some have suggested that involvement of greater than 50% of crosssectional area, tissue fluid collection, and deep muscle tears are predictive of a slower return to sport.89,144

Treatment Options

This should include pain-free isometric adduction initially with bent knee and progressing to full knee extension. During this time, the athletes may continue to participate in upper body and core weight-training. When able to concentrically adduct against gravity, the patient may be progressed to resisted adduction and stretching. During this phase, swimming and biking can be initiated. With symmetrical passive range of motion and adductor strength that is 75% of ipsilateral abductors, the patient may resume sport-­specific training and continue to increase resistance on adductor exercises. The patient is allowed to slowly pro­gress back to training once sport-specific exercises are completed with the goal of symmetrical active range of motion and equal abduction-adduction strength.137 With chronic adductor strains, Holmich and colleagues138 demonstrated that a physical therapy program consisting of massage, stretching, and modalities was ineffective. They reported that an 8- to 12-week strengthening program consisting of ­progressive resistive adduction and abduction exercises, balance training, abdominal strengthening, and sport­specific training was more effective. Adductor strain prevention programs should be instituted in the off-season to prevent strains during the season. Tyler and colleagues showed that those with an adduction-to-abduction ratio of less than 0.8 have a higher risk for sustaining an adductor strain. With implementation of a program in the National Hockey League, Tyler and colleagues137 were able to reduce the incidence from 8% to 3%.

Nonoperative Treatment As with other muscle strains, RICE, followed by a progressive therapy program, is the mainstay of initial adductor strain treatment (Box 21B-12). NSAIDs are frequently initiated during this period to decrease ­swelling and ­inflammation. After the initial 1 to 2 days, most authors recommend progressive pain-free range of motion exercises.137,145

Operative Treatment Surgical intervention is rarely indicated for adductor strains. For complete avulsion injuries, operative repair may be indicated.126,146 In strains lasting longer than 6 months after completing the therapy regimen, adductor tenotomy may be indicated. This should only follow further work-up to rule out other etiologies. Operative repair is considered an end-stage procedure because only 63% of patients are able to return to previous athletic level.147 Van Der Donckt and associates148 noted that adductor tenotomy and hernia repair returned 90% of patients to ­previous activity.

Box 21B-12   Treatment Options in Adductor Strains Figure 21B-13  Cross-sectional anatomy of the thigh. ab, adductor brevis; al, adductor longus; am, adductor magnus; g, gracilis, s, sartorius. (From Armfield DR, Kim DH, Towers JD, et al: Sports-related muscle injury in the lower extremity. Clin Sports Med 25:803-842, 2006.)

RICE followed by progressive stretching and strengthening Surgical myotendinous junction and avulsion repair Compartment fasciotomy Hematoma evacuation

1492 DeLee & Drez’s Orthopaedic Sports Medicine

Postoperative Protocol, Outcome Measurement, and Potential Complications If surgically repaired, limited knee range of motion is allowed initially. There is a paucity of literature supporting a postsurgical protocol.150 Straw and coworkers recommended that the knee be immobilized for 6 weeks

after surgery, followed by a progressive strengthening and training program.151 We recommend an initiation of gentle passive range of motion program starting 7 to 10 days postoperatively followed with knee immobilizer for 3 to 4 weeks. Following this the patient is started on phase II of the protocol in Table 21B-6.

Author’s Preferred Method After the diagnosis is made, a three-stage program is initiated. 137,149 This regimen requires intermittent monitoring of range of motion, pain, and muscle testing. The initial phase begins with RICE and NSAIDs to help control pain. After 24 to 48 hours, passive range of motion and submaximal isometric contractions are initiated. During this initial phase, trunk and upper extremity weight-training and conditioning may continue. Once able to adduct against ­gravity without pain, the patient may progress to phase II. During this phase, the patient is allowed to start bicycling and swimming for conditioning as well as isotonic adduction. The patient is encouraged to stretch, and start single-leg squats to develop muscular control. When the ­adduction-to-abduction strength ratio is equal to 75%, the patient is allowed to progress to phase III. This phase allows increased resistance, speed, and volume of resisted training. The patient is then allowed to start sport-specific training to improve adductor strength and modification of kicking or skating technique. When the adduction-to-abduction strength ratio is at least 90% to 100% and the adduction strength is symmetrical, resumption of sports training and competition is allowed. Following resumption of competition, an adduction strain prevention program should be initiated. Before training or competition, a warm-up program consists of cycling and stretching. A strengthening program focusing on adductor musculature should be initiated during the off-­season and maintained throughout the season. This should be reinforced with sport-specific exercises and ­proper

  TABLE 21B-6  Rehabilitation Protocol for Adductors Phase

Activity

I

RICE, NSAIDs, passive ROM, ± electrical stimulation, isometric adduction RICE, stretching, NSAIDs, ± electrical stimulation, ­ isotonic adduction, bicycling, swimming Ice, stretching, sport-specific training, preventive exercises, ± NSAIDs, isotonics ± isokinetics, ­conditioning, return to sports Return to sports, preventive exercises

II III IV

NSAIDs, nonsteroidal anti-inflammatory drugs; RICE, rest, ice, compression, and elevation; ROM, range of motion.

s­ kating, running, and kicking technique (Table  21B-6; Box 21B-13). Surgical intervention is considered only in complete avulsion with displacement greater than 2 cm or with ­failure of nonoperative management. There are multiple reports of adductor tenotomy for chronic adductor strains. Although this alleviates pain in most patients, only 63% were able to return to previous level of competition.

Box 21B-13   Adductor Strain Prevention Program Warm-up Bike Adductor muscle stretching Sumo squats Side lunges Kneeling pelvic tilts Strengthening program Ball squeezes (legs bent to legs straight) Different ball sizes Concentric adduction with weight against gravity Adduction while standing with a cable column or elastic resistance Seated adduction machine Standing with involved foot on sliding board and moving in the sagittal plane Bilateral adduction on sliding board and moving in the frontal plane (i.e., bilateral adduction simultaneously) Unilateral lunges with reciprocal arm movements Sports-specific training On-ice kneeling adductor pull-togethers Standing resisted stride lengths with a cable column to simulate skating Slide skating Cable column crossover pulls Clinical goals: adduction strength at least 80% of the ­abduction strength From Tyler TF, Nicholas SJ, Campbell RJ, McHugh MP: The ­effectiveness of a preseason exercise program to prevent adductor muscle strains in ­professional ice hockey players. Am J Sports Med 30(5):680-683, 2002.

Hip, Pelvis, and Thigh 1493

As with other strains, the main outcome measurements have been incidence and days missed from competition. Tyler and colleagues implemented an adductor strengthening program in the National Hockey League and noted greater than a fourfold decrease in ­adductor strain ­incidence. Ekstrand and associates129 also reported a 75% lower incidence of adductor injury with a ­stretching ­regimen and supervised warm-ups. Holmich and ­colleagues138 also noted that a significantly larger ­portion of athletes participating in an active exercise group were able to return to competition compared with those with passive therapy program. With limited data in the literature, the only outcome measurements reported in the literature are the postoperative range of motion, strength compared with the contralateral side, and return to previous level of activity. The few case reports and series in semiprofessional and professional athletes showed that all returned to previous levels of competition with nearly symmetrical strength.150,151 Possible complications of adductor strains include myositis ossificans, compartment syndrome, calcific tendinitis, and chronic groin pain. There is only one case report of myositis ossificans in the adductor musculature. Myositis ossificans most commonly occurs after direct trauma to an area. Risk factors include passive, forceful stretching; ­premature return to competition; reinjury of the same area; and predisposition to heterotopic bone formation.152 Myositis ossificans is usually suspected with persistent soft tissue swelling and pain that do not resolve after 4 to 5 days.50 Patients may also report increased induration and loss of motion 2 to 3 weeks after the injury. Radiographs do not show heterotopic ossification for 2 to 4 weeks after the injury. After 6 to 8 weeks, radiographs reveal a lacy pattern of new bone formation with the periphery corticalized. Although bone maturation may take up to 1 year, by 4 to 6 months, lamellar bone has usually formed, and resorption may start to occur.75 Recovery of normal activity and strength is possible in the presence of myositis ossificans. Patients with myositis ossificans should be treated as described earlier with early motion and strength ­training. Although rarely indicated, surgical excision should be done only to remove the myositis ossificans if it is prominent, predisposing the area to further injury or limits in range of motion.50 Excision should usually be done 6 to 12 months after bone maturity to prevent ­significant ­recurrence.75 There is one case report in the literature of thigh ­compartment syndrome following a quadriceps strain. The patient underwent thigh fasciotomies and was noted to have an uncomplicated return to sports. The final complication is residual weakness and loss of motion. The incidence of this complication is not reported for either nonoperative or operative treatment.

Criteria for Return to Play The patient should be allowed to return to competition after achieving symmetrical hip range of motion and ­adductor strength within 20% of the contralateral side. Although most patients return in 2 to 3 weeks, severe cases may take 8 to 12 weeks. On return to competition, the patient should continue the adductor rehabilitation ­program and wear a compressive thigh sleeve (Box 21B-14).

Box 21B-14   Return Strains  

to

Play

in

Adductor

Nearly symmetrical hip range of motion Adductor strength within 20% of contralateral side Sports-specific functional testing

Special Populations The incidence of adductor strain in adolescents is not reported in the literature. As in adults, most of these injuries occur with rapid change in direction and repetitive kicking, especially in an externally rotated abducted position.64 These patients usually present with acute onset of groin or medial thigh pain during activity.88 Although most of these occur at the adductor longus musculotendinous junction, complete ruptures are rare and usually represent an origin avulsion fracture.153,154 These avulsion fractures may be quite painful but usually do not result in muscle retraction or palpable defect as in adults.154 Radiographs in adolescents often help to differentiate soft tissue injuries from bony causes of groin pain. The rest of the physical examination and treatment are similar to those in adults. Avulsion fractures can be treated nonoperatively. Adolescents may return to competition with symmetrical flexibility, strength similar to the contralateral side, and completion of functional testing.88 These patients should be fitted with a circumferential thigh wrap or girdle.

C

r i t i c a l

P

o i n t s

l Adductor strains usually occur with rapid change in ­direction or kicking. l Adductor longus is the most common adductor strained. l History and physical examination should rule out other causes of groin pain to include osteitis pubis, athletic pu­­ balgia, hernias, and stress fractures. l Treatment should include a short RICE period, followed by progressive stretching and strengthening starting with isometric and progressing to isokinetic exercises. l Abductor-adductor strength imbalance greater than 80% increases the incidence of adductor strains.

Quadriceps Strain Evaluation Clinical Presentation and History As with other strains, patients report acute onset of pain in the anterior thigh during competition. Quadriceps strains have been reported in running, jumping, and kicking sports such as soccer, basketball, and football. The rectus femoris is the most commonly strained muscle of this group.155 The eccentric contraction that occurs during phases of kicking (back swing, ball contact, and ground contact before back swing) and deceleration before heel-strike in sprinting are the typical strain etiologies.151 Most patients present

1494 DeLee & Drez’s Orthopaedic Sports Medicine

with pain while running or kicking, loss of knee flexion, or an anterior thigh mass.155-159 Although most quadriceps strains occur in the distal thigh at the myotendinous junction, Garrett and colleagues158 described strains occurring at the central tendon in the proximal thigh. Orchard160 reported risk factors for Australian football players as dry field, previous quadriceps or hamstring injury, height less than 182 cm, and kicking sports. This study did not note an increase in quadriceps strain with age, as in adductor or hamstring strains.

Physical Examination and Testing The anterior thigh is usually slightly swollen and tender and may have a palpable mass or defect. This pain can be accentuated with resisted knee extension. The area most commonly affected is the musculotendinous junction of the rectus femoris. Range of motion should be examined for any loss of flexion with the hip flexed and extended. If diminished with hip flexion, the quadriceps femoris is involved. If the loss of flexion is worse with the hip extended, the vastus musculature (vastus medialis, vastus intermedius, or vastus lateralis) is involved because these muscles do not cross the hip joint. Thigh circumference should be measured to evaluate for swelling, hematoma, or defect. For chronic rectus strains, a palpable mass and pain with flexion may be present. However, strength and range of motion are usually symmetrical to the contralateral side. As with other strains, strain grading should be conducted to help counsel the patient. Mild (grade I) strains result in pain with minimal strength loss. Moderate (grade II) strains result in strength loss and pain. Severe (grade III) strains result in complete functional and motor loss. Differentiation between moderate and severe strains may be difficult in the first 7 to 10 days, and repeat examinations may be needed. After swelling diminishes, contraction of the muscles with complete tears reveals muscle retraction or palpable defect (Fig. 21B-14; Box 21B-15).

Figure 21B-14  Quadriceps rupture with retraction.

Box 21B-15   Typical Findings in Quadriceps Strains Localized tenderness to palpation over anterior thigh Pain worsens with hip extension Ecchymosis, anterior thigh swelling With or without palpable defect

Imaging Diagnostic imaging is usually obtained only to rule out other injuries. This allows for evaluation of avulsion injuries and femoral fractures. More advanced ­diagnostic ­imaging—ultrasound and MRI—are required to fully evaluate the soft tissue. Finlay and Friedman141 reported that ultrasound is a highly sensitive and specific method for evaluating acute quadriceps injuries. In muscle tears, dynamic ultrasound with the knee in flexion and extension helps differentiate hematoma from muscle tear. Although portable, economic, and accessible, ultrasound is operator dependent. Many institutions consider MRI to be the standard of imaging for muscle strain injuries (Fig. 21B-15).161-163 Cross and associates93 reported that MRI can estimate the size of the quadriceps strain and predict duration of rehabilitation. The cross-sectional capabilities allow determination of the involved muscle. The rectus quadratus has a slower recovery time than the vastus musculature.55,95 Injuries involving greater than 15% of rectus cross-sectional area resulted in a statistically increased rehabilitation interval, 14.6 days versus 8.9 days. Strains greater than 13  cm in length also resulted in a doubling of recovery time. The contrast enhancement around the central tendon, the so-called acute bulls-eye lesion, also revealed a significantly worse prognosis (26.8 days versus 9.2 days) (Fig. 21B-16).93,156 Involvement of the central tendon heralds a significantly longer rehabilitation interval. MRI also offers better diagnostic imaging of chronic injuries. Cross and associates93 believe this imaging modality should be used in the evaluation of professional athletes to confirm diagnosis and improve prognosis estimates. For recreational athletes, MRI is indicated if the symptoms do not

Figure 21B-15  Magnetic resonance image of rectus femoris strain. (From Cross TM, Gibbs N, Houang MT, Cameron M: Acute quadriceps muscle strains: Magnetic resonance imaging features and prognosis. Am J Sports Med 32:710-719, 2004.)

Hip, Pelvis, and Thigh 1495

treated two National Football League kickers with a nonoperative treatment regimen, noting that both patients were able to return to full practice after 3 to 12 weeks.

Operative Treatment

Figure 21B-16  Acute bulls-eye lesion. (From Cross TM, Gibbs N, Houang MT, Cameron M: Acute quadriceps muscle strains: Magnetic resonance imaging features and prognosis. Am J Sports Med 32:710-719, 2004.)

improve after 2 to 3 weeks of rehabilitation or for evaluation of chronic (>8 weeks) strains to evaluate for completeness of strain and complications.92

Treatment Options Nonoperative Treatment Most quadriceps strains can be managed nonoperatively (Box 21B-16). Determination of the degree of the injury should be attempted by clinical examination. After the degree of involvement is determined and the diagnosis is made, initial treatment should include RICE and passive, pain-free stretching for 24 to 48 hours.55,95 Following this phase, isometric exercises are initiated progressively, increasing within pain-free arc of motion. As strength and range of motion continue to improve, the patient is transitioned to isotonic training with increasing resistance as tolerated.55 With full pain-free range of motion, isotonic exercises with progressive increasing weights are allowed.95 When improving strength, isokinetic exercises should focus on low resistance and high speeds. A progressive running and kicking program should then be initiated, starting with slow jog and advancing to sprinting. When sprinting is tolerated, kicking may be started with a light ball over short distances and advancing to a normal-weight ball over longer distances.93 Sport-specific training should then be initiated. The patient may return to play after completion of this training but should continue to work on a quadriceps rehabilitation program involving stretching, isokinetic strengthening, and conditioning. Although many have recommended surgical treatment for proximal avulsion fractures, Hsu and coworkers150 Box 21B-16   Treatment Options in Quadriceps Strains RICE followed by progressive stretching and ­strengthening Surgical myotendinous junction and avulsion repair Compartment fasciotomy Hematoma evacuation

Surgical intervention is indicated only for total or near-total muscle tears, persistent pain with nonoperative treatment, reattachment of avulsion fractures, and large hematoma evacuation.55,150,151 There are limited studies on surgical treatment and outcomes of quadriceps strains, and most are extrapolated from treatment of hamstring strains. Straw and coworkers151 treated a semiprofessional soccer player with weakness and pain from a chronic proximal rectus avulsion with surgical repair and noted that the patient underwent full recovery and return to previous level of competition.

Postoperative Prescription, Outcome Measurement, and Potential Complications If surgically repaired, limited knee range of motion is allowed initially. There is a paucity of literature supporting a postsurgical protocol. Straw and coworkers recommended that the knee be immobilized for 6 weeks after surgery, followed by a progressive strengthening and ­training program. We recommend an initiation of a gentle passive range of motion program starting 7 to 10 days after surgery followed by a knee immobilizer for 3 to 4 weeks. The patient is then started on phase II of the previously described protocol. With limited data in the literature, the only outcome measurements reported in the literature are the postoperative range of motion, strength compared with the contralateral side, and return to previous level of activity. The few case reports and series in semiprofessional and professional athletes showed that all returned to previous levels of competition with nearly symmetrical strength.150,151 The potential complications of quadriceps strains include myositis ossificans, compartment syndrome, and residual weakness. The incidence of myositis ossificans is unknown after quadriceps strains and results from muscular and soft tissue disruption leading to heterotopic ossification.69-71 Myositis ossificans is usually suspected with persistent soft tissue swelling and pain that do not resolve after 4 to 5 days.50 Patients may also report increased induration and loss of motion 2 to 3 weeks after the injury. Radiographs do not show heterotopic ossification for 2 to 4 weeks after the injury. After 6 to 8 weeks, radiographs reveal a lacy pattern of new bone formation with the periphery corticalized. Although bone maturation may take up to 1 year, by 4 to 6 months, lamellar bone has usually formed, and ­resorption may start to occur.75 Recovery of normal activity and strength is possible in the presence of myositis ossificans. Patients with myositis ossificans should be treated as described earlier with early motion and strength training. Although rarely indicated, surgical excision should be done only to remove the myositis ossificans if it is prominent, predisposing the area to further injury or limits in range of motion.50 Excision should usually be done 6 to 12  months before bone maturity to prevent significant recurrence.75

1496 DeLee & Drez’s Orthopaedic Sports Medicine

Author’s Preferred Method The treatment regimen is similar to that described previously for adductor and hamstring strains (Table 21B-7). ­After diagnosis of the injury and degree determination, a RICE program should be initiated for 24 to 48 hours depending on the severity of the symptoms. If the patient has an antalgic gait, crutch weight-bearing is encouraged until no limp is present. NSAIDs should be initiated early to help reduce pain and swelling. Although rare, compartment syndrome has been seen after severe quadriceps strains.164,165 Therefore, patients should be instructed on this complication and to return if symptoms occur. First-degree quadriceps strain, usually with only activityrelated pain, will present with only mildly diminished knee range of motion. These should be treated as described previously but can be advanced more rapidly through the phases described next. Second- and third-degree strains present with more pain and limited motion. Patients with secondand third-degree strains should be progressed stepwise through the following phases. Following the acute phase, gentle stretching and isometric exercises in a pain-free arc of motion should be started. The isometric contractions should be conducted in multiple degrees of knee flexion. When the patient can perform a straight leg raise at 0, 20, and 40 degrees of knee flexion, isotonic exercises can be started. Isotonic contractions should start at a weight the leg can lift. As the knee progressively obtains full extension, the weight may slowly increase. A rapid increase in weight or motion may lead to recurrence or increased strain. During these early phases, swimming and upper body weight-training should be initiated to maintain conditioning. Cycling should not be initiated until the hip and knee can be flexed to 90 degrees. As the patient enters phase III, range of motion is nearly normal, and strength continues to improve. Isometric concentric contractions should be initiated with low resistance and high speed, progressing to lower speed with higher resistance. When the patient is able to tolerate higher resistance, eccentric contractions should be initiated.

There is one case report in the literature of thigh compartment syndrome following a quadriceps strain. The patient underwent thigh fasciotomies and was noted to have an uncomplicated return to sports. The final complication is residual weakness and loss of motion. The incidence of this complication is not reported for either nonoperative or operative treatment.

Phase IV introduces sport-specific training to the regimen along with continued strength training. This allows for a slow return to running. The first phase should include a warm-up and stretching followed with a slow jog. ­Progression to increased speed and distance should follow the ­patient’s pain. Ice and compressive dressing are used following each workout. Passive stretching can also be initiated during this phase to restore symmetrical range of motion. Phase V follows completion of sport-specific training and allows the athlete to return to full training and competition. Patients should also continue to alleviate risk factors for recurrence, including strengthening hamstrings and quadriceps, maintaining conditioning, and improving technique. Surgery is reserved for those with complete or nearly complete muscle tears. Avulsion injuries are treated similarly to other quadriceps strains, and surgery is reserved for those who fail this treatment. During this repair, the tendon and muscle and its epimysium are sutured together.166 Gentle range of motion is initiated 7 to 10 days after surgery; the patient then begins at phase II of rehabilitation.

  TABLE 21B-7  Rehabilitation Protocol for Quadriceps Strain Phase

Activity

I II

RICE, NSAIDs, ± crutch weight-bearing Ice, NSAIDs, electrical stimulation, isometrics, isotonics, active stretching Ice, NSAIDs, electrical stimulation, isokinetics (concentric then eccentric), conditioning, active stretch Ice, isokinetics, sport-specific training, running, passive stretching Return to sports, preventive exercises, risk factor alleviation

III IV V

NSAIDs, nonsteroidal anti-inflammatory drugs; RICE, rest, ice, compression, and elevation; ROM, range of motion.

should include repetitions of straight sprints and sprints requiring change of direction such as figure-eights. Sportspecific functional testing should also be completed. Most patients are able to return to play within 2 to 3 weeks. Players can be returned to competition while continuing a quadriceps stretching and strengthening rehabilitation program, including warm-up before exercise and wearing a thigh sleeve (Box 21B-17).

Criteria for Return to Play Return to competition may be allowed after achieving symmetrical range of motion to the contralateral side, isokinetic quadriceps strength within 15% of the ­contralateral side, and completion of a functional training program. Return to play should include observed sprinting and isokinetic testing to ensure that the athlete is not returning to sport too early and increasing the risk for aggravating or worsening the quadriceps strain. This

Box 21B-17   Return to Play in Quadriceps Strains Isokinetic strength within 15% of contralateral side Completion of functional training program Sports-specific functional testing to include interval sprinting and figure-eight runs

Hip, Pelvis, and Thigh 1497

Special Populations

Evaluation

The incidence of quadriceps strain in adolescents is not reported in the literature. Although most quadriceps strains occur at the distal musculotendinous junction, Hughes and colleagues156 reported that 30% of the athletes with central tendon injuries were adolescents. Third-degree tears are uncommon in adolescents, with only a few case reports.167-169 All quadriceps strains are less common in skeletally ­immature patients; avulsion fractures are more common.88 The physical examination, imaging, and treatment are similar to those in adults.

Clinical Presentation and History

C

r i t i c a l

P

o i n t s

l Most

quadriceps strains occur while running and k­ icking. l Patients usually present with pain on knee flexion and thigh swelling. l Involvement of the central tendon and greater than 15% cross-sectional area increase the recovery time. l Most quadriceps strains can be managed nonoperatively with a short period of RICE, followed by progressive strengthening and stretching. l Surgery is indicated only for complete or nearly complete tears and failure of nonoperative treatment. l Before return to competition, patients should undergo isokinetic testing as well as functional testing to reduce recurrence rates.

ADDUCTOR CANAL SYNDROME Relevant Anatomy and Biomechanics The adductor canal is an aponeurotic passageway on the anterior and medial thigh that transmits the femoral artery, the femoral vein, and the saphenous nerve.170-172 The femoral artery exits the pelvis through the femoral triangle, giving off the profundus femoris and continuing distally as the superficial femoral artery.171 The saphenous nerve is a large cutaneous branch of the femoral nerve that has sensory function only. The adductor canal is bordered medially by the sartorius, anterolaterally by the vastus medialis, and posteromedially by the adductor longus. Although rare, compression of the superficial femoral artery and saphenous nerve has been reported. Surgical exploration in selected cases noted compression by the adductor magnus and fibrous tissue, and intimal tears.171,173,174

Classification Adductor canal syndrome consists of two distinct causes. The first is described as an occlusion or kinking of the superficial femoral artery in the distal adductor canal. The second cause is entrapment of the saphenous nerve.

Adductor canal syndrome has been reported in populations ranging from young and active to older and sedentary patients. The two distinct causes have different presentations. Patients with compression of the superficial femoral artery complain of gradual onset of lower extremity claudication or activity-related pain.171 These symptoms worsen with activity and improve with rest.170 Patients with this condition often do not recall an inciting event or associated trauma. The other presentation is with knee pain or dysesthesia in the saphenous nerve distribution. This condition has an insidious onset, and patients do not recall a specific injury or event initiating their ­symptoms. Patients can complain of worsening symptoms with activity that are alleviated with rest.

Physical Examination and Testing With claudication symptoms, a thorough physical examination should be completed with the help of a vascular surgeon. Physical examination of the affected leg can reveal a cooler affected extremity, decreased or absent pulses, and decreased leg hair.171 An ankle-brachial index should be obtained to complete the assessment. Doppler examination reveals decreased or absent arterial flow distal to the distal adductor canal. For saphenous nerve entrapment, tenderness to palpation can be elicited with palpation over the adductor canal. Decreased sensation over the saphenous nerve distribution can be noted (Box 21B-18).172

Imaging Imaging is rarely indicated for evaluation of this condition. Arteriography, arterial pulsatile waveform, and Doppler ultrasound have been used to evaluate claudication symptoms,173 usually revealing compression of the superficial femoral artery in the distal adductor canal.

Treatment Options Nonoperative Treatment Initial treatment requires differentiation of the etiologies (Box 21B-19). Saphenous nerve compression can be differentiated and treated by injection of local anesthetic and steroid mixture into the adductor canal (Fig. 21B-17). Nonoperative therapy for superficial femoral artery compression is not indicated. Box 21B-18   Typical Findings in Adductor Canal Syndrome Tenderness to palpation of distal adductor canal Diminished pulses and decreased ankle-brachial index Decreased sensation over the saphenous nerve distribution

1498 DeLee & Drez’s Orthopaedic Sports Medicine

Box 21B-19   Treatment Options in Adductor Strains

Anterior Vastus medialis

Injection for saphenous nerve compression Surgical release and repair of superficial femoral artery

Saphenous nerve Fibrous sheath Sartorius

Operative Treatment

Saphenous vein

After compression of the superficial femoral artery is confirmed, surgical exploration should be performed. Surgical exploration should include examination for any focal area of compression, usually by the adductor magnus or fibrous tissue. If no external compression is noted, an arteriotomy is recommended to examine for intimal tears, which can be treated with segment excision and end-to-end anastomosis as compared with interpositional grafting.

Weighing the Evidence Limited literature is written on adductor canal syndrome. All the information is class III or IV data.

Author’s Preferred Method Initial evaluation should include differentiating between saphenous nerve entrapment and superficial femoral artery occlusion. This can usually be done based on history and physical examination, including ankle-brachial index. If still unsure, an injection with local anesthetic and steroid mixture can rule out saphenous nerve compression. If the patient exhibits obvious claudication symptoms, positive ankle-brachial index, or diminished vascular perfusion distally, arteriography should be performed. Using this information, surgical exploration should be performed with the assistance of a vascular surgeon. External compression should initially be evaluated. The most common cause is the tendinous insertion of the adductor magnus. If external compression is noted, inspection of the arterial lumen should be completed for possible intimal tear. This can be treated with excision with end-to-end anastomosis as compared with interpositional grafting.

Postoperative Prescription, Outcomes Measurement, and Potential Complications There is no general consensus on the postoperative treatment of adductor canal syndrome. Patients treated with lysis of the tendinous insertion of the adductor magnus or fibrous bands can usually return to play soon after incisional healing. Those who undergo end-to-end ­anastomosis or interpositional grafting should delay return to play per the vascular surgeon’s protocol. This is usually 3 to 4 weeks after end-to-end repair and 6 to 8 weeks after graft ­interposition.

Gracilis Medial Add. longus Add. magnus

Femur

Fem. artery Fem. vein

Figure 21B-17  Cross-sectional anatomy of adductor canal for injection. (From Romanoff ME, Cory PC, Kalenak A, et al: Saphenous nerve entrapment at the adductor canal. Am J Sports Med 17:478-481, 1989.)

With limited data, the only outcomes measured in the case reports and case series reported are reperfusion of the distal leg for superficial femoral artery repair or release. In patients with saphenous nerve compression after injection, Romanoff and colleagues172 reported symptom resolution in 80%, no change in 13%, and worsening of symptoms in 7%. Potential complications of superficial femoral artery compression include a dysvascular limb, worsening claudication symptoms, and activity limitation. Potential complications of saphenous nerve compression include persistent or worsening numbness and dysesthesia. Romanoff and colleagues172 noted that the longer the saphenous compression, the poorer the results of injection. This is similar to other compression neuropathies.

Criteria for Return to Play Patients with saphenous nerve compression may be allowed to continue sporting activities through their treatment. Patients with superficial femoral artery compression should be able to return to training as dictated by their surgical treatment. This return to play should be determined in conjunction with the vascular surgeon and usually is 7 to 10 days for lysis of external compression, 3 to 4 weeks for end-to-end anastomosis, and 6 to 8 weeks for interpositional grafting (Box 21B-20). Box 21B-20   Return to Play in Adductor Canal Syndrome Isokinetic strength within 15% of contralateral side Completion of functional training program Sports-specific functional testing to include interval sprinting and figure-eight runs

Hip, Pelvis, and Thigh 1499

Special Populations

S U G G E S T E D

Adductor canal syndrome has not been reported in any adolescent population. If this does occur, pediatric patients should be treated as described for adults. C

r i t i c a l

P

o i n t s

l Patients with saphenous nerve compression report knee pain or dysesthesia or numbness in the saphenous nerve distribution. No motor symptoms are noted. l Saphenous nerve compression can be treated with injection of local anesthetic and steroid. l Superficial femoral artery occlusion is the result of external compression by tendinous insertion of adductor magnus or fibrous bands or internal compression by an intimal tear. l Superficial artery occlusion results in leg pain or claudication symptoms. l Superficial artery occlusion should be treated with surgical exploration with lysis of external compression or ­surgical repair.

R E A D I N G S

Armfield DR, Kim DH, Towers JD, et al: Sports-related muscle injury in the lower extremity. Clin Sports Med 25(4):803-842, 2006. Aronen JG, Garrick JG, Chronister RD, McDevitt ER: Quadriceps contusions: Clinical results of immediate immobilization in 120 degrees of knee flexion. Clin J Sport Med 16(5):383-387, 2006. DeFranco MJ, Recht M, Schils J, Parker RD: Stress fractures of the femur in ­athletes. Clin Sports Med 25(1):89-103, 2006. Hershman E, Lombardo J, Bergfeld J: Femoral shaft stress fractures in athletes. Am J Sports Med 9:111-119, 1990. Hughes CT, Hasselman CT, Best TM, et al: Incomplete, intrasubstance strain ­injuries of the rectus femoris muscle. Am J Sports Med 23(4):500-506, 1995. Morelli V, Weaver V: Groin injuries and groin pain in athletes: Part 1. Prim Care 32(1):163-183, 2005. Nicholas SJ, Tyler TF: Adductor muscle strains in sport. Sports Med 32(5): 339-344, 2002. Orchard J: Intrinsic and extrinsic risk factors for muscle strain in Australian ­Football. Am J Sports Med 29(3):300-303, 2001. Ryan JB, Wheeler JH, Hopkinson WJ, et al: Quadriceps contusions: West Point update. Am J Sports Med 9:299-304, 1991. Verta MJ Jr, Vitello J, Fuller J: Adductor canal compression syndrome. Arch Surg 119(3):345-346, 1984.

R eferences Please see www.expertconsult.com

S e c t i o n

C

Physical Activity and Sports Participation after Total Hip Arthroplasty Shawn M. Brubaker and Thomas E. Brown Modern hip arthroplasty with cement fixation has evolved dramatically since its inception and later popularization in the late 1950s and early 1960s by Sir John Charnley. Total hip arthroplasty (THA) was first approved for use in the United States in the late 1960s, and the first Charnley low-friction prosthesis was implanted in the United States at the Mayo Clinic by Mark B. Coventry in 1969. THA has since become a very common procedure, with more than 200,000 implantations performed in 2003 and with demographic studies suggesting a continuing exponential increase of THA in the future.1,2 Classic indications for THA include groin pain and lateral hip pain with weight-bearing and restricted range of motion, with physical examination showing hip pain of

intra-articular origin and radiographs demonstrating coxarthrosis. Additional indications include inability to ­control hip pain with nonsurgical measures along with inability to perform daily activities and absence of, or acceptable, ­medical comorbidities. Many of the problems of THA, such as durable fixation of the prosthesis, bearing wear, and instability, have been at least partially resolved with modified surgical approaches, larger bearing couples, biologic fixation, and decreased wear rates. Likewise, the population of the United States continues to increase in life expectancy as well as activity level at an overall older age. As outcomes of THA improve, orthopaedic surgeons are being increasingly questioned about utilization in younger

1500 DeLee & Drez’s Orthopaedic Sports Medicine

patients and about recommendations regarding ­activity level, including returning to sports activity after hip re­placement. There are differing opinions among orthopaedic surgeons as to what are acceptable activities following total hip replacement. As indications for THA have been expanded to include very young patients with post-traumatic arthritis and avascular necrosis, as well as older athletes who want to continue to train and compete, there has been a respective allowance for increased activity and even return to recreational sports. There are no long-term studies indicating outcomes of THA after return to athletic activity for most sports; however, general guidelines based on available information can be used to guide sports ­medicine ­professionals as they work with these highly motivated patient populations.

ETIOLOGY The etiology of degenerative hip disease can generally be divided into two categories.3 The first is secondary osteoarthritis of the hip, which is attributable to either a developmental or structural abnormality such as develop­ mental dysplasia of the hip. Secondary osteoarthritis of the hip may also be attributed to a pathologic metabolic process that leads to destruction of the joint, such as rheumatoid arthritis, or to a structural abnormality such as Legg-Calvé-Perthes disease. The second is primary osteoarthritis of the hip, which is a diagnosis of exclusion that is applied when the source of the degenerative process cannot be attributed to either a recognizable structural or metabolic ­abnormality. Rheumatoid arthritis, seronegative spondyloarthopathies, traumatic coxarthrosis, and septic arthritis are causes of secondary osteoarthritis in which destruction of the joint surface occurs directly by the insulting process. Slipped capital femoral epiphysis, developmental dysplasia of the hip, Paget’s disease, Legg-Calvé-Perthes disease, and osteonecr­osis of other origins are all examples of processes that alter the normal anatomy of the hip articulation and ultimately lead to osteoarthritis of the joint by overloading the available hyaline cartilage surfaces to failure. Failure of the cartilage initiates the degenerative process, which leads to the typical physical and radiographic findings seen in patients who require THA to alleviate their pain and improve their ­function. Primary osteoarthritis is a diagnosis of exclusion, and as such, studies on its “etiology” have become epidemiologic in nature. Attempts to identify common characteristics in patients who lack structural abnormalities in the presence of coxarthrosis have common findings. Two findings show that a strong genetic component plays a role in the development of primary osteoarthritis. American Caucasians of Western European descent have a 3% to 6% rate of primary coxarthrosis, which is 10 times higher than African ­Americans and up to 27 times higher than Asian Americans.4 This is despite of the fact that these three populations have essentially similar rates of secondary osteoarthritis of the hip. Studies of relatives of patients who have undergone THA showed that 27%5 to 50%6 have findings specifically suggestive of coxarthrosis. These findings

­ enerally suggest that this population has an underlying g cartilage defect. Additional studies on the epidemiology of primary osteoarthritis of the hip have suggested that high body weight or body mass index, occupational factors such as heavy labor, elite sports participation, and femoral anteversion may all contribute to the degeneration of the hip in whites or patients with a genetic predisposition. Of special interest, studies done in Sweden reported increased risk for osteoarthritis in patients involved in track and field, racket sports, and soccer, with professional soccer players demonstrating the highest risk overall.7

ANATOMY AND BIOMECHANICS A full review of the anatomy and biomechanics of the hip is beyond the scope of this chapter; however, ­pertinent anatomic and biomechanical factors are discussed in detail. Proximal femoral osteology includes the head, neck, greater and lesser trochanters, and proximal ­femoral diaph­ysis. The diameter of the femoral head averages 46 mm, with a range of 35 to 58 mm, and the average ­neck-shaft angle averages 125 ± 7 degrees.8 ­Femoral version is determined by the angle formed between the plane of the femoral epicondyles and plane of the ­femoral neck. It averages 13 ± 7 degrees in patients without coxarthrosis but averages 20 ± 9 degrees in patients with coxarthrosis.9 It should also be noted that most of the proximal femur, including the greater trochanter, has a posterior bow that intersects the anterior bow of the diaphyseal femur at the level of the lesser trochanter. This proximal posterior bow has implications for placing the femoral stem in the correct position within the femur in the sagittal plane. The morphology of the acetabulum is the result of fusion between the ischium, the ilium, and the pubis. This provides three major areas of bony support for the femoral head. The anterior and posterior columns are composed of the pubis and ischium, respectively, with their corresponding acetabular walls.10 The thin medial wall provides only minimal support, whereas the acetabular tectum or dome is the area that provides direct cephalad coverage of the femoral head at the confluence of the posterior column into the anterior column. The densest bone in the acetabulum is located posteriorly and superiorly.11 The acetabulum faces anteriorly with an average anteversion of 17 ± 6 degrees. The acetabulum also faces caudally, with an inclination adequate to give a center-edge angle greater than 25 degrees in the average white American population, usually about 45 degrees.12 There are 21 muscles that span the hip joint. Anterolateral and posterior approaches to the hip are most commonly used today for hip arthroplasty; consequently, the medial muscles are uncommonly encountered. The gluteus maximus, tensor fasciae latae, and iliotibial band compose the superficial muscle layer. Posterior, lateral and anterolateral approaches to the hip traverse this layer in varying fashions. The sartorius is encountered during anterior approaches to the hip. The next muscular layer is composed of the gluteus medius and minimus anteriorly and laterally, covering the hip capsule on their way to insert on the greater trochanter and its overlying fascia. Posteriorly,

Hip, Pelvis, and Thigh 1501

the short external rotators cover the hip capsule and insert posteriorly, proximally to distally, into the medial aspect of the trochanteric crest. The piriformis, superior gemellus, obturator internus, inferior gemellus, and quadratus femoris form this layer of muscles. All “posterior” approaches to the hip involve taking down the short external rotators and posterior capsule while leaving the abductors intact, whereas lateral and anterolateral approaches all take down a portion of the gluteus medius and minimus, preferably in a fashion to allow accurate and sturdy reattachment. This is typically done by releasing the anterior third of the attachment of the gluteus medius occasionally with a flake of bone to improve healing potential after reattachment.13 The gluteus minimus is released as necessary to allow adequate visualization of the anterior capsule to allow for capsulotomy. Knowledge of blood supply and innervation about the hip is useful for understanding safe surgical intervals for approaches to the hip as well as healing potential of surgical procedures. The hip has random cutaneous blood supply, which allows more variation in skin incisions, particularly when compared with the medial-to-lateral supply of blood to the skin overlying the knee. This becomes more important in the knee as compared with the hip when an incision is already present. Blood supply to the acetabulum has received more attention with the increasing popularity of bone-preserving procedures. The proximal femur has received the most attention, particularly in relation to femoral neck fractures. Pertinent arterial and neural anatomy is outlined here. The blood supply to the femoral head is primarily from the ascending cervical branches of the femoral neck.14 The lateral ascending cervical branch is the most prominent and arises from the deep branch of the medial femoral circumflex, which may arise from either the femoral artery directly or from the profunda femoris and courses between the iliopsoas and the pectineus muscles. The ascending branch of the medial femoral circumflex courses deep to the insertions of the short external rotators on its way to supplying the cervical branches. The second and much smaller source of blood to the femoral head is from the foveal artery, which runs in the ligamentum teres. This artery arises from the acetabular branch of the obturator artery, which originates from the external iliac artery. It courses through the cephalad aspect of the obturator foramen and deep to the transverse acetabular ligament into the pulvinar of the cotyloid fossa where it enters the ligamentum teres. The piriformis muscle and tendon is the landmark that defines the location of relevant neurovascular structures during posterior approaches to the hip. The sciatic nerve and the inferior gluteal artery and nerve, which supply the gluteus maximus, exit below the piriformis in most cases. Occasionally, the sciatic nerve can course through or partially through the piriformis muscle. The superior gluteal artery exits the greater sciatic notch with the superior gluteal nerve above the piriformis and courses anterolaterally 4 to 6 cm above the superior edge of the acetabulum, ­supplying the gluteus medius and minimus muscles.15 Other arteries of importance during the posterior approach to the hip include the branch to the quadratus femoris, which also arises from the ascending branch of the medial

femoral circumflex artery. In some instances, a portion of the gluteus maximus tendon insertion to the posterolateral aspect of the proximal femur needs to be released to allow adequate anterior mobility of femur for acetabular visualization. The first perforating branch of the profunda femoris artery pierces the lateral femoral retinaculum just deep to this point, and care should be taken to prevent it from retracting through the retinaculum into the posterior compartment of the leg following inadvertent transection. Also worthy of mention is that the sciatic nerve runs deep to the gluteus maximus tendon in this region but superficial to the quadratus femoris. When using an anterolateral approach, the primary neurovascular concern is the superior gluteal artery and nerve, which course 4 to 6 cm above the superior edge of the acetabulum. Some authors have recommended that dissection into the gluteus medius in this approach proceed no higher that 4 to 5 cm superior to the proximal edge of the greater trochanter to prevent additional injury to the already split abductors.16 During an anterior approach to the hip, the superior and transverse branches of the lateral femoral circumflex artery may be encountered and require ligation. One could potentially encounter the femoral artery and nerve, but only if dissection was carried far out of the appropriate plane. Many surgeons prefer to place screws through the acetabular component to augment initial fixation until bone ingrowth occurs. The placement of acetabular screws requires the surgeon to be familiar with neurovascular anatomy around the pelvis as well as the bony anatomy of the acetabulum to prevent iatrogenic damage. Wasielewski and colleagues developed an acetabular quadrant system that is commonly used to guide surgeons in the placement of acetabular screws.17 To orient the quadrants, a line is drawn from the anterior superior iliac spine through the center of the acetabulum for a lateral position. The line generally carries through the center of the ischium and is called line A. A second line is drawn 90 degrees, or perpendicular, to the first line and is called line B. Line A divides the quadrant system into posterior and anterior, and line B divides the quadrant system into superior and inferior, respectively. The safest quadrant is posterior and superior, with the greatest risk to the sciatic nerve and with long screws oriented toward the sciatic notch. The posteroinferior quadrant is also considered safe for screws that are less than 20 mm in length. Screw placement into either anterior quadrant is not recommended because of the proximity of neurovascular structures. Motion, stability, and force transmission across the hip are some of the basic biomechanical functions to consider during hip replacement. The nonarthritic hip loses range of motion as it ages. About 0.5 to 1 degree of range of motion in any direction is lost per decade of life after full skeletal maturity. Roach and Miles determined that patients 40 to 59 years of age have an average of 120 degrees of flexion, 18 degrees of extension, 42 degrees of abduction, 31 degrees of internal rotation, and 32 degrees of external rotation.18 Consequently, the goal of implant design should allow for restoration of this range of motion. Most commercially available primary total hip implants allow for this range of motion. Head-to-neck ratio is important to understand in this regard. Small heads, such as the 22-mm head that has

1502 DeLee & Drez’s Orthopaedic Sports Medicine

been used in the past for low-friction ­arthroplasty, have a relatively decreased range of motion before the femoral neck comes into contact with the acetabular component. A large head, such as 44 mm, has relatively more range of motion before the neck impinges if both femoral necks are the same size. Consequently a smaller head could have increased range of motion if the neck were made smaller, hence the importance of head-to-neck ratio. Constrained or “captured” head articulations have as little as 70 degrees range of motion, whereas newer designs have greater ranges of motion, some of which nearly approximate average ranges of motion.19 The recent interest in metal-on-metal articulations is in part due to the ability to use a very large head, which improves range of motion before impingement occurs. Stability is one of the most important factors for successful hip replacement. The nonpathologic hip is highly constrained because of fully congruent bony articulations and stout capsule and overlying ligaments. With smaller than anatomic prosthetic femoral heads and resultant prosthetic neck–acetabular impingement, instability of THA has brought great attention to component design, implantation orientation, surgical technique, and soft tissue tensioning. Assuming the component orientation is correct, soft tissue tension is reestablished; there are, theoretically, two ways to increase stability of the femoral head. The first is to increase the size of the head. This increases the head-to-neck ratio and greatly reduces the likelihood that the femoral neck will impinge on the acetabular component. Enlarging the femoral head also increases the distance the head would have to travel out of the acetabulum to clear the edge and actually dislocate, the “jump height” after the prosthetic neck impinges against the edge of the acetabular component or the bone of the pelvis. The second is to constrain the prosthetic head with a head-capturing acetabular component. This method, however, results in loss of range of motion because the acetabular component has to cover the head beyond its equator so that the dimensional circumference is less than that of the prosthetic head in order to “capture” it and not allow excursion out of the prosthetic socket. This type of mechanism can be subject to high torque load during episodes of impingement, which predisposes it to failure. Proper soft tissue tension not only increases the stability of the replaced hip but also greatly affects the load transmission across the hip. This has stimulated a significant amount of interest in restoring the lateral offset in the hip with high-offset femoral components if necessary.20 A highoffset femoral component is one that has a decreased neckshaft angle with a lengthened neck. In general terms, if the offset of the hip is increased (restored), the abductors have a longer moment (lever) arm to pull against, and less force is required from the muscle to generate the same amount of axial torque around the hip joint. This increased offset also lengthens the abductors, which intrinsically increases their efficiency by lengthening the muscles. Conversely, the longer moment arm and decreased force exerted by the abductor results in less force transmission across the hip joint by the hip musculature. Forces across the hip joint can be quite high: walking generates forces equaling 3 times body weight across the hip, whereas catching oneself from trip and fall may generate forces up to 8 times body weight.21 Anatomic and biomechanical considerations play a large role in selecting the appropriate implant for a patient who

wishes to return to sports. In our opinion, stability of the implant is of paramount importance, followed by wear characteristics of the bearing surface and durable fixation of the implant to the bone by ingrowth.

CLASSIFICATION There are no predominant classifications for hip osteoarthritis currently in use for clinical application. This is due to the lack of correlation between the severity of the radiographic findings and the patient’s clinical presentation, including physical examination. Many attempts have been made to establish a correlative radiographic classification to aid in research attempts to more thoroughly understand osteoarthritis, but a common effective classification has failed to emerge. One of the earliest classifications was a grading system developed by Kellgren and Lawrence.22 This classification is based solely on radiographs and has been the most extensively utilized to report on hip arthritis. Hip arthritis is characterized as absent (grade 0), doubtful (grade 1), minimal (grade 2), moderate (grade 3), or severe (grade 4). However, a more useful classification for the clinician is one that indicates when a patient is in need of total joint arthroplasty after having exhausted all other medical management. The American College of Rheumatology has developed criteria for the diagnosis of osteoarthritis of the hip (Box 21C-1).23 This classification is part of an algorithm that directs nonoperative treatment of the arthritic hip, which is particularly helpful to the physician who wishes to delay THA because of high daily activity. The algorithm is discussed under the treatment section.

EVALUATION Clinical Presentation and History Most people with osteoarthritis of the hip complain of groin pain of insidious onset. The pain waxes and wanes with activity level and load or demand placed on the joint. Not uncommonly, the patient has recently experienced some minor trauma to the hip such as a fall or a missed step that causes a sharp increase in pain. Changes or increases in activity level may also exacerbate preexisting hip pain. In particular, pain with weight-bearing that subsides with rest is a strong indicator of intra-articular hip pathology. Stiffness after periods of decreased activity such as sleep and sitting and increased pain throughout the day are hallmarks of osteoarthritis of the hip. It is important to delineate the temporal relationship of the patients’ activities Box 21C-1  American College of ­Rheumatology Classification Criteria for ­Osteoarthritis of the Hip Hip pain and at least two of the following three items: • Erythrocyte sedimentation rate < 20 mm/hr • Radiographic femoral or acetabular osteophytes • Radiographic joint space narrowing

Hip, Pelvis, and Thigh 1503

to their pain both in daily ­activity and over monthly and even yearly time spans. Other complaints about the hip joint are common. Lateral or greater trochanteric pain is common, and trochanteric bursitis and abductor tendinitis are thought to be related to osteoarthritis because changes in the ­efficiency and strength of the abductor musculature directly affect intra-articular loading of the hip joint. Patients may also complain of medial knee pain and occasionally have no hip pain whatsoever. This presentation is uncommon, but anyone with medial knee pain should have radiographs of the hip simultaneously to rule out hip pathology as a contributing source. Another common presentation is pain centered in the gluteal or low lumbar area. Pain perceived in these areas is only rarely referred from the hip joint, and careful evaluation of the lumbar spine and sacroiliac joints is warranted.24

Physical Examination and Testing Physical examination of the hip begins with inspection of the hip for skin condition and previous incisions as well as any gross morphologic abnormality. Palpation of the anterior, lateral, and posterior aspects of the hip can aid in determining precisely where the patient’s pain originates. Range of motion testing is important to record and understand because patients with significant stiffness may need alterations in the planned surgical approach to allow for adequate visualization when the hip has very limited range of motion. A proportionately greater loss of range of motion is common with osteoarthritis of the hip. Next, muscle strength testing and deep tendon reflexes should be evaluated. Vascular status of the leg can be ascertained in part by visualizing the skin for abnormalities, palpating the patient’s pulses and looking for calcifications in the vessels on plain radiographs. Comparative leg lengths should also be evaluated both clinically and radiographically. It is important to ask whether the patient has noticed a leglength discrepancy or any tilt in the pelvis, either at presentation or in the past. Several provocative tests or measures are available to determine the cause of the patient’s pain. First, the patient’s gait should be evaluated for abnormal gait pattern such as antalgic or Trendelenburg-type patterns. The Stinchfield test is useful in differentiating low back pain from intra-articular hip pain.25 The patient lays supine, and the affected lower extremity is held in full extension and elevated several inches off of the examination table with and without resistance to hip flexion. Anterior hip pain with resistance to flexion is specific for intra-articular hip pain. This maneuver can be simultaneously compared with a straight leg raise examination for radicular irritation. Pain at the end of range of motion is also commonly tested as part of the physical examination. The FABERE test is done with hip flexion, abduction, and external rotation of the extremity. This position is occasionally called the figure-four position and places the hip to the limit of its range of motion in several planes and simultaneously compresses the joint. The Thomas test evaluates the extremities for hip flexion contracture and is performed by bringing both knees to the chest and, while maintaining the contralateral hip in full flexion, extending the affected hip as much as possible; the angle between the coronal axis of the spine

and of the femur in maximal attainable extension is the flexion contracture of the affected hip. Femoroacetabular impingement is evaluated by fully flexing the hip, internally rotating and adducting the hip simultaneously, to oppose the femoral neck against the anterior wall of the acetabulum. This test, combined with radiographs, gives information about some of the causes of pain and osteoarthritis, such as labral tearing, acetabular retroversion, and anterolateral femoral neck prominence. Occasionally, despite a well-performed history and physical examination, the source of the pain may remain obscure. This is particularly true if the patient has multiple arthritic joints. Referred pain from the spine can confuse the clinical picture enough that the physician may wish to perform additional testing to determine the source and location of the pain. Fluoroscopically guided intra-articular injection of anesthetic such as lidocaine or bupivacaine (Marcaine), along with temporal recording of symptoms after the intervention, is very useful in determining the origin of the pain. For the patient who has hip and lumbar spine pathology, the injection can give the physician an indication as to what percentage of the patient’s hip pain is truly intra-articular and how much is referred from the lumbar spine or elsewhere. This same approach can be used to rule out other potential pain generators, such as the sacroiliac joints, the facet joints, and very occasionally the knee. The goal is to offer the patient a surgery that will resolve the hip pain to a known degree.

Imaging Standard radiographs for evaluation of the hip joint include an anteroposterior view of the pelvis with as much femur as possible, combined with a frog-leg lateral view of the affected hip. A surgical lateral view of the hip may also be obtained in patients with very limited range of motion in rotation. This view may also help delineate any acetabular abnormalities that may be present. The false profile view of the pelvis, although helpful for surgeons considering joint-sparing procedures of the pelvis, generally provides little additional useful information for the patient requiring hip arthroplasty. Obturator oblique and iliac oblique views can provide information about the quantity of bone present in the anterior and posterior columns, respectively, should there be any concern on the anteroposterior view. Patients who present with relatively normal radiographs and a clear physical examination of intra-articular hip pain and patients who have unilateral avascular necrosis should undergo magnetic resonance imaging of the pelvis to evaluate for nonradiographically visible intra-articular hip pathology and for contralateral avascular necrosis of the femoral head.

TREATMENT OPTIONS Nonoperative Nonoperative treatment options for osteoarthritis of the hip may be limited once the patient’s hip pain has become great enough to warrant referral to an orthopaedic surgeon for hip pain. However, the surgeon should verify

1504 DeLee & Drez’s Orthopaedic Sports Medicine

that all nonoperative measures have been exhausted before ­offering surgical intervention to the patient. Nonoperative treatment options can generally be divided into pharmacologic and nonpharmacologic interventions. Nonpharmacologic interventions can be subdivided into physical therapy based and non–physical therapy based. Weight loss, patient education, patient self-help programs, and other health care professional support options are non–physical therapy–based measures that can be undertaken. It is important to evaluate the patient’s mental status in the history and physical examination because patient care outcomes are heavily influenced by factors such as depression and sleep deprivation. Substance use and abuse should also be evaluated. Physical therapy, including pool therapy and occupational therapy, is an important nonoperative treatment that can either delay the need for hip replacement or optimize the patient physically before undergoing surgery. Physical therapy should focus on range of motion exercises and muscle strengthening as well as proper instruction on using an assistive ambulation device such as a walking stick to help unload the painful hip. Occupational therapy may have a smaller role, but focus would be on equipment that can aid in the activities of daily living such as a hand­actuated grasper that prevents the patient from flexing the hip as is required to bend over and pick an item from the floor. Otherwise, occupational therapy instruction should focus on energy conservation and techniques to unload the hip as much as possible. Pharmacologic nonoperative treatment begins with a non-narcotic analgesic such as acetaminophen and a low-dose nonsteroidal anti-inflammatory drug (NSAID). Should this regimen fail to provide relief, the surgeon may consider using a full-dose NSAID such as ibuprofen. We recommend that this regimen, if implemented, be done under the supervision of the patient’s primary medical physician because the side-effect risk profile is significant. Narcotic analgesics may also be used, including propoxyphene, codeine, hydrocodone, and oxycodone. Parenteral pharmacologic therapy for hip osteoarthritis includes injection of local anesthetic medication combined with a steroid into the greater trochanteric bursa if it is determined to be inflamed on physical examination. Fluoroscopically guided intra-articular injection of steroid may also provide the patient with great relief for a variable period and can be used in certain patients to delay the need for surgery. As previously mentioned, the American College of Rheumatology provides an algorithm for nonsurgical management of hip osteoarthritis (Box 21C-2; Fig. 21C-1).26

Box 21C-2 Medical Management of Patients with Osteoarthritis of the Hip Nonpharmacologic therapy Patient education Self-management programs (e.g., Arthritis Self-Help Course) Health professional social support through telephone contact Weight loss (if overweight) Physical therapy Range of motion exercises Strengthening exercises Assistive devices for ambulation Occupational therapy Joint protection and energy conservation Assistive devices for activities of daily living and ­instrumental activities of daily living Aerobic aquatic exercise programs Pharmacologic therapy Nonopioid analgesics (e.g., acetaminophen) Nonsteroidal anti-inflammatory drugs Opioid analgesics (e.g., propoxyphene, codeine, ­oxycodone)

in regard to level of activity and how that level of activity will affect the longevity of the chosen implant as well as patient satisfaction with the procedure.27 Primary consideration should be given to bone fixation, bearing type, and activity level.

Weighing the Evidence Review of the evidence first focuses on surgical approach and durable prosthesis fixation, and then on bearing types and activity levels and their relationship to bearing wear and implant longevity. Special consideration is given to stratifying athletic activity levels and the resulting durability of the prosthesis in the face of increase demand. The first consideration in THA for the high-demand patient is to consider what surgical approach to use. Nonpharmacologic modalities and acetaminophen (up to 1 g qid)

If response inadequate, use alternative analgesic, low-dose ibuprofen (up to 400 mg qid) or nonacetylated salicylates

Operative Operative intervention of the osteoarthritic hip ­primarily focuses on total joint arthroplasty. There may be ­occasion to consider hip arthroscopy in the athlete who has ­minimal degenerative changes in the hip on plain radiographs but significant soft tissue findings on magnetic resonance imaging, such as a labral tear or cartilage flap. The primary concern for the patient wishing to return to an active lifestyle and even competitive sports is the durability and stability of the implant chosen for the task. It is also important for the surgeon and the patient to have an understanding before surgery of what the expectation of the patient is

If response inadequate, use full-dose nonsteroidal anti-inflammatory drug (with misoprostol if patient has risk factors for UGI bleeding or ulcer disease)

If response inadequate, consider referral for joint surgery (osteotomy, total joint arthroplasty) Figure 21C-1  Guidelines for the medical management of patients with symptomatic osteoarthritis of the hip. UGI, upper gastrointestinal (tract).

Hip, Pelvis, and Thigh 1505

The posterior approach and the lateral or anterolateral approaches are the most commonly used for THA, and their relative benefits and liabilities are important to consider. The posterior approach to the hip spares the abductors but takes down the short external rotators and the posterior capsule. Historically, the posterior approach was plagued with a significantly higher dislocation rate than that of the other approaches. A large study of 10,500 hips showed a 5.8% dislocation rate associated with the posterior approach.28 More recent studies have shown the rate of dislocation from the posterior approach to approximate those of lateral and anterolateral approaches when the posterior capsule is not split through its posterior transverse aspect and when it is used in conjunction with the short external rotator to create an “augmented” deep posterior closure. One study showed a reduction from 4.3% to 0.7%,29 and another showed a reduction from 6.2% to 0.8%.30 The benefit of this approach is the preservation of the hip abductors, which are of paramount importance in reestablishing normal biomechanics of the hip during arthroplasty. There is also higher risk for damage to the sciatic nerve compared with other approaches. The lateral and anterolateral approaches offer many benefits also, but with a different set of liabilities. The rate of dislocation is historically lower because acetabular preparation and component insertion can often be done without violating the posterior capsule and the short external rotators. This approach offers more direct visualization of the acetabulum and potentially easier acetabular component positioning and insertion. Masonis and Bourne’s metaanalysis of more than 13,000 hips showed that the rate of dislocation was 3.23% for a posterior approach.31 Lateral and anterolateral approaches showed 0.55% and 2.18% rates of dislocation, respectively. The disadvantage to the anterolateral approach is the violation of the hip abductors. Patients may potentially be left with abductor weakness and Trendelenburg gait. This same meta-analysis studied the rates of limp in a subset of the patient population with a 4% to 20% limp rate in patients who underwent an anterolateral or lateral approach and a 0% to 16% rate of limp in patients who underwent a posterior approach to the hip. Recently, minimally invasive or minimal incision surgery has been conceived and has gained popularity with patient populations and orthopaedic surgeons alike. New approaches have been devised and other approaches modified to allow for smaller incisions. Of note is that minimally invasive or less invasive surgery has only been shown to improve patient recovery in the postoperative period. Long-term results are not yet available, and short-term to midterm outcomes show no advantage for patients who undergo minimally invasive hip surgery. There are several reports of component mal­ position and fracture as complications of these approaches for THA. An example of a newly conceived approach is the two-incision approach described by Berger.32 Many modifications of existing approaches have also been described, with some less commonly used approaches gaining increased attention, such as the “anterior supine intermuscular” as marketed by industry. This approach uses the distal portion of the Smith-Petersen approach with the purported advantage of not transecting any muscle or tendon to access the hip. The next area of consideration is fixation of the implant to bone in a durable manner. Historical experience with

cemented acetabular components, including the original Charnley low-friction arthroplasty, showed good initial results for the first 10 years. However, after the first decade, the rates of loosening were unacceptably high despite attempts at improved cement techniques.33 This led to the advent of porous-coated acetabular components with the hope that bony ingrowth into the component would occur and durable fixation would be the result. The first generation of porous-coated acetabular components was plagued with design issues that limited their longevity to about 70% survivorship at midterm. The second-generation acetabular components have excellent midterm longevity of greater than 90% at 10 or more years of follow-up.34 The fixation of femoral stems has been a little more controversial owing to less disparity in outcomes between press-fit and cemented femoral stems. Long-term outcomes are available for the Charnley flat-back polished prosthesis, which had a revision rate of only 3% at 30-year follow-up.35 This same femoral prosthesis had a 10% rate of revision at 30 years when used in patients younger than 30 years. Other reports have also shown decreased cemented femoral component longevity when used in younger, more active patients for osteoarthritis.36 Although substantive data are lacking, many surgeons advocate cemented femoral fixation for older, thinner, and less active patients with osteopenic bone. The next consideration is the bearing couple. There are four U.S. Food and Drug Administration–approved bearing couples generally in use in the United States: metal on polyethylene, ceramic on polyethylene, ceramic on ceramic, and metal on metal. Other bearing types, such as metal on ceramic and diamond surfaces, are also being studied. The primary concern of the bearing couple is the rate of wear and the quantity and quality of the generated debris. Bearing couples can be placed into two different categories: hard on soft, such as metal on polyethylene, and hard on hard, such as metal on metal. Hard-on-soft bearings include metal on polyethylene and ceramic on polyethylene. There are multiple types of femoral head hard-bearing surfaces available, but the most commonly used in the United States is cobalt-chromium. Metal on polyethylene is the bearing couple in longest use and is the baseline to which all other bearings are compared. Cobalt-chromium on highly cross-linked polyethylene has an initial wear or “run-in” of 100 μm in the form of creep and a steady-state wear of 10 to 20 μm per year. There are currently conflicting reports about whether ceramic on polyethylene decreases the wear of the “soft” surface. Ceramic has the theoretical advantage of less second body wear from the head owing to its surface ­hardness that resists scratches and other surface damage to which chrome cobalt is susceptible. A recent study showed slightly improved wear rates with oxidized zirconium on polyethylene when the two couples were compared.37 The past or initial experience with metal-on-metal bearings was poor. However, there are metal-on-metal articulations that were found to be functioning well at 20- to 30-year follow-up. This finding renewed interest in this bearing couple. Improved manufacturing techniques and tolerances with excellent short-term and early midterm results have fueled interest in this bearing. The wear rate is 5 to 10 times less than hard-on-soft bearings. There is an

1506 DeLee & Drez’s Orthopaedic Sports Medicine

initial run-in period that represents true wear as the bearing “seats” of 25 μm. This is followed by a linear steadystate wear of 5 μm per year on average. Metal-on-metal bearings have their own set of concerns that other bearing couples do not have. The number of particles generated by this bearing is 100 to 200 times greater than that of ­hard-­on-soft bearings despite the fact the linear and volumetric wear is significantly lower. These particles are also biologically active ions that are toxic to cells and cause chromosomal changes. Carcinogenicity is a concern, but a causal relationship has yet to be established. Hard-on-hard bearings have been reported to squeak and click, although current manufacturing tolerances have decreased this possibility. There are historical reports of metal-on-metal bearings seizing because of insufficient equatorial clearance from antiquated design and manufacturing tolerances. Increased head size, combined with precise manufacturing tolerances, has improved lubrication characteristics. The larger head size produces greater surface area and higher angular or sliding velocities within the joint, which helps maintain the integrity of the fluid film.38 Ceramic-on-ceramic bearings have the overall lowest rate of wear compared with other bearing couples, including metal-on-metal bearings. The initial experience with the first generation of alumina ceramic bearings was poor owing to the brittle nature of the material. Large ceramic grain and pore size predisposed the material to high wear rates and fracture. The latest generation of alumina bearings has incorporated zirconia into the material. Zirconia is another ceramic material that has greater fracture toughness but is softer and consequently has higher wear rates when compared to alumina. This most recent generation of ceramic incorporates both materials in appropriate proportions to maximize material toughness and fracture resistance without losing the excellent wear characteristics of alumina. Improvements in component design have also improved the durability of ceramic bearings.39 Ceramic liner fracture is associated with impingement and protective designs that protect the neck of the femoral component from contacting the edge of the ceramic acetabular liner and careful attention to intraoperative component positioning have decreased fracture rates. Ceramic bearings demonstrate a wear rate that is 3 to 5 times less than a metal-on-metal bearing. The average run in wear is 1 μm, and subsequent steady-state wear is 0 to 3 μm per year. The particle debris that is generated is quantitatively similar to that of a metal bearing. Both types of hard-on-hard bearings have significantly smaller particles and consequently higher particle counts than do hard-on-soft bearings. Ceramic bearings in particular have a bimodal distribution of particle sizes.40 Average sizes are either smaller than 0.02 μm or larger than 0.2 μm. The smaller sizes of particles are generated by continuous full-contact bearing wear, and the larger particles present when micro-separation occurs at the bearing interface, presumably contact loading wear as opposed to adhesive wear. Ceramic particles appear to have less biologic reactivity compared with metal particles. Hip resurfacing procedures are gaining popularity among hip surgeons performing arthroplasty procedures. There are several reasons behind the resurgence of interest in this procedure. One is preservation of femoral bone stock as compared with THA. Another is the use of large metal head

femoral components that allow physiologic behavior of the hip with a theoretical decreased risk for hip dislocation. Preservation of the femoral neck may benefit patients for several reasons. The first is that stress transfer across the proximal femur should mimic nonpathologic physiologic behavior. In a hemi-resurfaced hip, this may decrease the peak compressive loads that are presented to the acetabular cartilage. Second, less stress-shielding of the proximal femur should take place in the resurfaced hip than in a totally replaced hip. The concept of femoral head resurfacing is more than 50 years old, and many different materials have been used during the history of this procedure. The acetabular portion of resurfacing has had the more tortuous evolution. Initially, metal acetabular components were used with an unacceptably high failure rate, as also noted with metal-on-metal THA. Subsequently, metal-on-metal hip resurfacing was abandoned in favor of resurfacing the femoral head with metal and cementing a polyethylene component into the acetabulum. This resurfacing combination demonstrated extremely high volumetric wear rates and intense osteolytic reactions. Improved tribology, as noted with metal-onmetal total hip replacement, is being carried over to surface replacement, and the use of thin metal acetabular resurfacing components with ingrowth surfaces has resolved the historical issues of the acetabular component. Total hip resurfacing has its own inherent limitations. The procedure does not allow for more than 1 cm of leg lengthening and offers essentially no ability to alter the offset of the femur.41 Acetabular component fixation cannot be augmented with screws, and any significant bony loss preventing a stable press-fit precludes its use. Bone loss in the femoral head from cyst formation or avascular necrosis may compromise the durability of cement fixation of the femoral component. Other relative contraindications include body mass index above 35, age greater than 60 years, female gender, and tall patients. Current concerns focus on complications related to the femoral component, specifically loosening, preservation of blood supply, and femoral neck fracture. A 5-year study using hybrid fixation, press-fit ingrowth acetabular components, and a cemented femoral component showed 96% survivorship.42 Notching of the femoral neck and varus placement of the femoral resurfacing component have been shown to increase the risk for femoral neck fracture and ultimately decreased survivorship of the reconstruction. At this point, there are no outcome studies regarding return to athletic activities for patients who have undergone hip resurfacing. To date, there are two surveys reporting on patient return to athletic activities. The first report was on 43 patients who underwent 51 resurfacing procedures.43 Before surgery, 65% of patients were able to participate in athletic activity; after surgery, the percentage involved in sports increased to 92%. The authors concluded that the procedure allows patients to participate in sporting activities who were unable to participate previously. The impact level of the activities was not reported. The second survey was performed in Switzerland; of 112 patients who underwent surface replacement, 110 returned to multiple sporting activities. Fifty-one percent of the patients returned to downhill skiing, 12% to high-impact sports, and 22% to contact sports. The authors concluded that patients were capable of returning to very high levels

Hip, Pelvis, and Thigh 1507

  TABLE 21C-1  Classification of Sports Based on Level of Impact Low Impact

Potentially Low Impact

Intermediate Impact

High Impact

Stationary cycling Calisthenics Golf Stationary skiing Swimming Walking Ballroom dancing Water aerobics

Bowling Fencing Rowing Isokinetic weights Sailing Speed walking Cross-country skiing Table tennis Jazz and ballet Bicycling

Free weights Hiking Horseback riding Ice skating Rock climbing Low-impact aerobics Tennis In-line skating Downhill skiing

Baseball, softball Basketball Volleyball Football Racquetball, handball Jogging, running Lacrosse Soccer Water-skiing Karate

of activity after resurfacing and that additional studies are required to determine the rates of loosening and revision in this highly active patient population. Currently, there are no reports to validate the reasoning that patients can safely resume high-risk activities or sports and expect longterm survivorship of the implants. Outcome reports at this time only have 2- to 5-year follow-up with the current generation of resurfacing implants. Midterm and long-term outcomes of patients participating in high levels of athletic activity could potentially be disastrously poor. There are no prospective controlled studies on survivorship of total joint arthroplasty in patients desiring to return to athletic activities. Many recommendations have been made as to what activities are acceptable. There has also been a significant evolution from the early 1980s until now as to what activities are acceptable following total hip replacement. One of the first studies on return to sport was produced from data out of the Swedish Registry. Visuri and Honkanen reported on spontaneous patient return to recreational exercise.44 The average patient age of 539 patients surveyed was 64 years, and the mean follow-up time was 4.2 years following surgery. Regular walking increased from 2% preoperatively to 55% postoperatively. Cycling increased from 7% to 29%. Swimming increased from 13% to 30%, and skiing from 0% to 9%. The authors concluded that swimming and cycling were valuable recreational activities because joint reaction forces were minimized in these two sports. Another early (1983) survey of patients returning to sporting activity after THA also came from Europe. Dubs and colleagues reported on 110 cemented THAs in male patients averaging 55 years old and reported a paradoxical rate of femoral stem loosening of 14.3% in sedentary patients and only 1.6% in patients involved in sporting activity.45 Acetabular liner wear rate, however, was noted to be 4 times greater in the patients that returned to sports. Previous to this report, the recommendation after THA was severe limitation or termination of all athletic activity. The authors concluded that there was no need to strictly prohibit return to sports but recommended viscoelastic heel inserts to attenuate peak joint loading during walking and running. In 1987, Ritter and Meding reported on patients in the United States who were sent questionnaires regarding their activity level after total hip replacement.46 Information was gathered regarding preoperative and postoperative athletic activities. Postoperatively, all patients had a decreased level of involvement in various athletic activities except for cycling. Most patients returned to an active sport but at less intense

levels than before surgery. The authors concluded that low-impact activities such as walking, golf, and bowling would have no adverse influence on the outcome of total hip replacement. One of the earliest (1991) reports on the fate of total hip resurfacing and THA demonstrated twice the revision rate for patients who were highly active or involved in sports.47 This became apparent after 10 years in patients with the diagnosis of osteoarthritis. Other patients who underwent hip arthroplasty for other causes such as avascular necrosis had early failure of fixation apparent at 6 years. Loosening in patients who underwent hip resurfacing procedures primarily occurred in patients involved in high-impact activities. The first survey of surgeons regarding acceptable sporting activities was performed by McGrory and colleagues at the Mayo Clinic.20 They used a questionnaire that was distributed to 28 surgeons, including 15 senior residents and fellows at their institution, to evaluate 28 common sports. Recommended sports after total hip or knee arthroplasty included sailing, lap swimming, scuba diving, cycling, ­golfing, and bowling. Sports not recommended included running, water-skiing, football, baseball, basketball, hockey, handball, karate, soccer, and racquetball. Importantly, this report was the first to classify a large number of sports into categories. Recommended, intermediate, and not recommended categories were developed based on what sports were permitted by the surgeon. The second such type of report was performed by Healy, Iorio, and Lemos.48 They surveyed the Hip Society, the Knee Society, and the American Shoulder and Elbow Surgeons Society and reviewed 42 sports that were either allowed, allowed with experience, or not recommended, or as a last category, sports for which no conclusion was reached by the sur­ vey. The Hip Society allowed stationary cycling, croquet, ballroom dancing, golf, horseshoes, shooting, shuffleboard, swimming, doubles tennis, and walking. Sports allowed with experience included low-impact aerobics, road cycling, bowling, canoeing, hiking, horseback riding, and cross-country skiing. No conclusion was reached at that time (1999) on jazz dancing, square dancing, fencing, ice skating, roller skating or in-line skating, rowing, speed walking, downhill skiing, stationary skiing, weight-lifting, and weight machines. All other sports presented in the ­survey were not recommended nor allowed and were highimpact, contact, or extreme sports. Clifford and Mallon took the concept one step further and specifically classified sports based on their level of impact and then made general recommendations regarding each level (Table 21C-1).49

1508 DeLee & Drez’s Orthopaedic Sports Medicine   TABLE 21C-2  Classification of Sports Based on

­Recommended Activity after Total Hip Replacement Allowed Athletic Activity

Allowed with Experience

Not Allowed or Undecided

Golf Swimming Doubles tennis StairMaster, elliptical walking, treadmill Speed walking Hiking Stationary skiing Bowling Road and stationary cycling Low-impact aerobics Rowing Dancing Weight machines

Downhill skiing Cross-country skiing Weightlifting Ice skating Rollerblading Pilates

Racquetball Squash Jogging All contact sports Football Soccer Basketball High-impact aerobics Baseball Softball Snowboarding Martial arts Singles tennis

Low-impact and potentially low-impact sports were recommended for most patients. Although these activities were considered low impact, concern for increased rates of wear remains. Intermediate-impact loading sports were considered appropriate for selected patients only. Patients must be in excellent physical condition with preactivity evaluation, close monitoring while involved, and specific guidelines for extent of involvement in these sports. High-impact loading sports were not recommended because of the risk for injury and increased likelihood of revision surgery. Klein and associates recently published consensus guidelines for return to athletic activity after total hip replacement.50 They surveyed the members of the American Association of Hip and Knee Surgeons as well as ­members of the Hip

Society. At the time of the survey, there were 99 members of the Hip Society and 727 members of the American Association of Hip and Knee Surgeons. Ninety-two members of the Hip Society and 522 members of the American Association of Hip and Knee Surgeons responded, making this survey by far the largest performed to date. The survey asked the members of each organization to evaluate 37 specific sporting activities and determine whether they were allowed, allowed with experience, not allowed, or undecided. This survey directed participants to make recommendations based on the use of standard metal-­ on-polyethylene THA (Table 21C-2). Also surveyed was the amount of time required to return to sports postoperatively. Thirty-two percent of the ­surgeons surveyed allowed return to sports at 1 to 3 months, but 59 required 3 to 6 months of recovery to pass prior to allowing the patient to return to athletic ­activity. Based on these reports and their temporal relationship to each other, there has been a gradually increasing tolerance over time to permit patients who have undergone THA and hip resurfacing procedures to return to sporting activities and in many cases to activities that are relatively higher impact and therefore higher risk. The lack of any midterm or long-term studies on the survival of hip arthroplasty under these high strain conditions is concerning. Surgeons need to understand that the results of allowing patients to return to high levels of sporting activities are unknown and potentially catastrophic. Historically, there are good data to prove that increased demand decreases the durability of the hip reconstruction.51 Surgeons do not have the ability to directly control how patients will either use or abuse their implants after surgery, but patient education should be paramount, and patients should be fully informed about the probable consequences of “abusing” their implants.

Authors’ Preferred Method Owing to lack of any outcome studies available on athletic and sports activity after total hip replacement, we have elected to use other outcomes as previously mentioned to guide our treatment of patients wishing to remain active. We use a minimally invasive posterior approach to the hip. Our experience with this approach, as well as with the anterolateral and lateral approaches, suggests that patients regain their abductor function more quickly, reliably, and completely when the abductors are not split nor their insertion taken down. Biologic fixation is used on the femoral and acetabular components. We prefer to use a metaphyseal bony ingrowth surface on a triple-taper stem to maximally load the proximal bone and prevent stress shielding. Subsidence of the femoral component is less of a concern if a triple-taper component is used compared with a double-taper component. Some commonly used triple-taper component designs allow for placement on the femoral stem in a proximal femur with stenotic or small canals while simultaneously providing good to excellent metaphyseal “fill.” Hydroxyapatite or similar derivatives sprayed onto the femoral stem are not

used in our patients because this coating adds unnecessary cost to an expensive, high-demand implant and does not significantly alter the reliability of bony ingrowth. A bony ingrowth acetabular component is also used. We expect rim loading of the acetabular component on press-fit insertion with excellent stability. The bearing surface is critical for two reasons: wear and stability. Our choice, to minimize one and maximize the other, respectively, is to use a metal-onmetal articulation with a large metal head and, as previously mentioned, a resurfacing-type nonmodular acetabular component. Large metal head hip systems have been developed to allow for a 6- to 7-mm discrepancy between the outer diameter of the femoral head and outer diameter of the ­acetabular component. For example, a 58-mm acetabular shell will allow for a 51-mm femoral head component in the Anatomic Surface Replacement (ASR) system manufactured by DePuy (Warsaw, Ind). This bearing selection has the benefit of low wear rates provided by hard-on-hard bearing surfaces as well as excellent stability provided by the extremely large head and high femoral head–to–femoral neck

Hip, Pelvis, and Thigh 1509

ratio. Biomechanically, the jump distance required for dislocation may nearly be the entire width of the classic 28-mm head. Of note, the acetabular component in this particular system is hemispherical geometrically, although not a complete hemisphere. It has 168 degrees of arc coverage. This complements the high head-to-neck ratio and potentially allows for greater than anatomic range of motion, making dislocation due to implant-specific impingement and “lever out” with subsequent dislocation nearly impossible. Bony impingement between anterior or posterior columns and the greater trochanter can then potentially become a source of mechanical impingement. Our approach to preventing boneagainst-bone impingement between the proximal femur and the ­pelvis is to restore the patient center of rotation with accurate placement of the acetabular component and slightly medializing if the bony pelvis, in conjunction with the preoperative plan, dictates. This allows the use of a high offset stem, which increases the length of the moment arm around the hip and decreases joint reaction forces. This approach may also decrease wear rates of the bearing couple. The patient is given hip precautions for a posterior approach for the first 6 weeks after surgery with appropriate physical therapy. After 6 weeks, the patient starts a more intensive graduated program that focuses on restoration of hip flexion and abduction strength and endurance. The patient is allowed to resume noncompetitive athletic activities after 3 months and competitive noncontact, low- to medium-impact sports as described by Clifford and Mallon49 after 4 months. Surgical Technique

Our THA for the active patient is done through an incision that ranges between 7 and 12 cm and is centered over the middle and posterior thirds of the greater trochanter from anterior to posterior (Fig. 21C-2). About one third of the incision is positioned superior to the cephalad edge of the greater trochanter, and two thirds is placed caudad to the tip of the greater trochanter. The incision is gently curved posteromedially but less so than the classic Moore approach. We place the concave or posterior side of the incision just anterior to the “soft spot” in thinner patients that correlates with the location of the tendinous portion of the piriformis. The proximal portion of the incision curves posteriorly ­midway between a directly vertical incision and the Moore incision. The fascia lata and the superficial gluteal fascia are incised and split in line with the skin incision (Fig. 21C-3). If the posterior edge of the gluteus medius is palpable through the fascia and muscle of the gluteus maximus, the division of the maximus is carried down directly over the posterior edge of the medius with the leg flexed between 30 and 45 degrees. The objective is to maximize the view of the acetabulum without fighting either the medius anteriorly or the maximus posteriorly. The muscular plane between the maximus and the short external rotators is developed posteriorly as well as between the gluteus medius and the anterior portion of the gluteus maximus and tensor fasciae latae (Fig. 21C-4). A Charnley-style retractor is used to hold the superficial muscle, adipose, and cutaneous layers away from the working area of the procedure. The greater trochanteric

Figure 21C-2  The patient is positioned in the right lateral decubitus position with the left hip up. A planned posterior approach surgical incision of 8 to 10 cm is marked after preparation and drape.

bursa is resected, and the short external rotators are exposed from gluteal sling to the piriformis. The piriformis tendon is delineated superiorly, and a Cobb elevator is used to lift the superior capsular insertion of the gluteus minimus. The dissection is carried far enough anteriorly to allow the placement of a Homan-style retractor, allowing excellent visualization of the capsule superiorly. The capsule and short external rotators are then taken down as one tissue sleeve. The capsule is transected vertically starting superiorly 1 cm proximal to the superior edge of the acetabulum and about 1 cm anterior to the superior edge of the piriformis tendon. This is carried distally to the insertion of the piriformis tendon and taken posteriorly around the femoral neck and

Figure 21C-3  Sharp dissection is performed through the skin, and electrocautery is used to split subcutaneous fat in line with the incision. A Cobb elevator is then used to clear off the fascia of overlying tissue for 1 to 2 cm on each side of the planned division.

Continued

1510 DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method—Cont’d

Figure 21C-4  The fascia of the gluteus maximus and the iliotibial band is divided, and the deep tissue is exposed. The greater trochanter is seen in the top left of the incision with the short external rotators coursing toward it. Of note is the sciatic nerve in the bottom of the incision as it travels caudad beneath the piriformis and above the remaining short external rotators.

Figure 21C-6  The center of the femoral canal is located with a canal finder, and femoral preparation is performed with a combination of reamers and broaches depending on the hip system used. The photograph demonstrates the anteversion of the final broach compared with the anatomic version of the femoral neck.

distally to the posterior aspect of the tranverse acetabular ligament (Fig. 21C-5). The hip is then dislocated posteriorly, and a femoral neck cutting template is used to mark the level and angle of the femoral neck resection based on preoperative templating. A bone saw is used to cut the femoral neck, and the remaining proximal femur is freed of soft tissue attachments and excised. The anterior capsule is then lifted off the remaining anterior femoral neck with a Cobb elevator for a distance of 1 to 2 cm. The femoral canal is prepared first. A ­combination of osteotomes, awls, canal finders, reamers, and broaches is

used depending on the femoral component selected for implantation. We open the femoral canal just posterior to the insertion of piriformis. This allows the preparation of the femoral canal to accept a component positioned in about 20 to 30 degrees of anteversion. If the native femoral neck version is less than 20 degrees, bone is removed from the posterolateral neck as far down as the intertrochanteric ridge to allow positioning of the lateral shoulder of the implant in a maximally posterior and lateral position (Fig. 21C-6). This technique permits the calcar portion of the implant to rest against the femoral neck without broaches or reamers removing excessive bone from this area. This is particularly important when using an implant with neutral version (side nonspecific). The acetabulum is prepared after the femur. This order of preparation allows the addition of anteversion to the acetabular component should the femoral component be anteverted less than that of our target anteversion because of ­anatomic restrictions. A small, 5-mm defect is created directly anterior in the anterior capsule. A long-handled Hohmanntype retractor is placed anteriorly. The proximal femur is retracted anteriorly against the femoral broach (left in place) to protect the greater trochanter as well as the abductors from direct pressure. Another Hohmann-type retractor is placed directly superiorly to allow visualization of the superior portion of the acetabulum. The inferior limb of the posterior capsular reflection is completed along with transection of the transverse acetabular ligament. A third, offset-type, Hohmann retractor is placed posteriorly. Excellent visualization of the acetabulum is obtained. The labrum is excised in its entirety, and the capsule is elevated off the walls and superior edge of the acetabulum as necessary (3 to 10 mm) for accurate reaming. A reamer is then placed into the acetabulum aimed directly medially and slightly ­posteriorly

Figure 21C-5  The short external rotators have been released from the posterior aspect of the proximal femur along with the capsule to expose head and neck of the femur. The templated femoral cut is marked next as measured from the lesser trochanter.

Hip, Pelvis, and Thigh 1511

Figure 21C-7  The acetabulum is centrally reamed to the floor of the cotyloid fossa to determine depth. Sequential reaming is then carried out until bleeding bone is visible throughout the acetabulum. One to 2 mm of press-fit is generally recommended, but this may vary depending on the acetabular system in use.

Figure 21C-8  This photograph demonstrates flexion, adduction, and internal rotation of the leg with trial or final components in place to verify adequate stability of the prosthetically replaced hip in the position of instability. If an approach other than the posterior approach is used, stability must be verified.

to medialize to the medial-most portion of the cotyloid fossa. We use a smaller reamer to initiate this process, usually between 44 and 48 mm. When the floor of the cotyloid fossa is encountered, positional reaming is then carried out. This is done by placing the reamer in the center of the acetabulum horizontally as well as vertically, with appropriate corrections for abnormal anatomy. The reamer is held in 45 degrees of abduction and 20 to 30 degrees of anteversion, with care taken to prevent the femur or areas of sclerotic bone from forcing the reamer into an eccentric position within the acetabulum. Size-progressive reaming is carried out until the reamer is covered anteriorly, posteriorly, and superiorly. When the reamer is covered, if preoperative templating permits, the reamer is pushed in the direction of reaming about 2 to 4 mm to slightly “tunnel” the acetabulum. This technique ensures excellent rim fit and loading of the component and allows for the use of a solid-back or nonmodular acetabular component. We generally underream the actual component size by 1 to 2 mm, or stated otherwise, we use a 1- to 2-mm press-fit of the acetabular component. Short arc hemispherical components such as the ASR previously mentioned may require underreaming by 2 mm to obtain a good press-fit. True hemispherical components are underreamed by 1 mm, and “dual geometry” or rim-flared acetabular components are reamed line to line because the rim is 1 to 2 mm wider in diameter than the rest of the shell (Fig. 21C-7). Trial components are placed, and length and stability of the construct are assessed. Leg length is measured before dislocation and restored with modular neck lengths, as is offset. Stability of the implant is tested and is deemed stable if the joint does not dislocate with the leg first positioned at 90 degrees of flexion and 30 degrees of adduction, and

s­ econd at 90 degrees of flexion and 60 or more degrees of internal rotation (Fig. 21C-8). We are commonly able to obtain 80 degrees of internal rotation in this position before trochanter-to-ilium impingement. The hip is also flexed to 120 degrees, as well as maximally internally rotated at 45 degrees of hip flexion; the knee and hip are also maximally extended and the hip is externally rotated to verify anterior stability. With proper restoration of leg length and offset, the hip can become very difficult to dislocate intraoperatively and may require the use of a bone hook to “pull” the femoral head out of the acetabulum. When testing of the trial components is completed, press-fit components are impacted into the testing position, and a second trial is carried out because the positions of the component may not precisely match the trial component position. Necklength modularity is used to fine-tune the length of the leg (Fig. 21C-9). An augmented posterior repair is performed with No. 5 Ti-Cron sutures. The suture is passed twice through the piriformis and the superior portion of the capsular reflection, and another suture is taken through the inferior capsule and corresponding short external rotator in similar fashion. Two 2-mm drill tunnels are placed through the posterior aspect of the greater trochanter, with 1-cm bone bridges in all ­directions, and the suture passer is used to pass the inferior Ti-Cron through the inferior drill hole and the superior Ti-Cron through the superior drill hole, respectively. The leg is abducted and externally rotated, the repair is approximated to the bone, and the suture is tied over the bone bridge. The gluteal and iliotibial fascia is reapproximated using interrupted figure-eight sutures, and the skin is closed in the surgeon’s preferred fashion. Postoperative radiographs are obtained to verify implant positioning (Fig. 21C-10). Continued

1512 DeLee & Drez’s Orthopaedic Sports Medicine

Authors’ Preferred Method—Cont’d

Figure 21C-9  The femoral stem and the acetabular socket are already impacted into position. The photograph demonstrates a large metal head, for metal-on-metal articulation, that is about to be impacted onto the trunion of the femoral stem.

C

r i t i c a l

P

o i n t s

l Determine preoperative athletic activity level and ­athletic competence in the patients’ sports areas of i­nterest. l Evaluate and document expectations of both the ­surgeon and the patient in regard to postoperative athletic activity level; if a discrepancy exists, reconciliation of the ­discrepancy is mandatory before surgery. l Fully disclose to the patient that neither midterm nor long-term durability outcomes have been established for total hip replacements in athletically active patients. l Hip resurfacing has strict inclusion and exclusion criteria and is not recommended at this time for very active athletes. l Strong consideration should be given to using an ­approach to the hip that does not violate the abductors. l Alternative bearing surfaces should also be considered because of their markedly improved wear characteristics. l It is imperative that extreme stability of the hip is achieved intraoperatively to reduce risk for dislocation in both the early and late postoperative periods. l Postoperative routine radiographic surveillance should be performed yearly while the patient remains athletically active.

Figure 21C-10  Anteroposterior pelvic radiograph 1 year after surgery. This 38-year-old man, who underwent total left hip replacement with a large metal head, for metal-on-metal articulation, is athletically active in low- and medium-impact sports as permitted by his physician.

S U G G E S T E D

R E A D I N G S

Clifford PE, Mallon WJ: Sports after total joint replacement. Clin Sports Med 24(1):175-186, 2005. Healy WL, Iorio R, Lemos MJ: Athletic activity after joint replacement. Am J Sports Med 29(3):377-388, 2001. Klein GR, Levine BR, Hozack WJ, et al: Return to athletic activity after total hip arthroplasty: Consensus guidelines based on a survey of the Hip Society and American Association of Hip and Knee Surgeons. J Arthroplasty 22(2):171-175, 2007. Marti B, Knobloch M, Tschopp A, et al: Is excessive running predictive of degenerative hip disease? Controlled study of former elite athletes. BMJ 299(6691):91-93, 1989. Masonis JL, Bourne RB: Surgical approach, abductor function, and total hip arthroplasty dislocation. Clin Orthop 405:46-53, 2002. McGrory BJ, Stuart MJ, Sim FH: Participation in sports after hip and knee arthroplasty: Review of literature and survey of surgeon preferences. Mayo Clin Proc 70:342, 1995. Pellicci PM, Bostrom M, Poss R: Posterior approach to total hip replacement using enhanced posterior soft tissue repair. Clin Orthop 355:224-228, 1998. Ritter MA, Meding JB: Total hip arthroplasty: Can the patient play sports again? Orthopedics 10(10):1447-1452, 1987. Visuri T, Honkanen R: Total hip replacement: Its influence on spontaneous recreation exercise habits. Arch Phys Med Rehabil 61(7):325-328, 1980. White RE Jr, Forness TJ, Allman JK, Junick DW: Effect of posterior capsular repair on early dislocation in primary total hip replacement. Clin Orthop 393:163-167, 2001.

R eferences Please see www.expertconsult.com

C H A P TER�  

 �� 22

Patella S e c t i o n

A

Patellar and Quadriceps Tendinopathies and Ruptures Michael A. Rauh and Richard D. Parker

The patellofemoral joint encounters one of the highest joint reaction forces found within the human body. Essential to creating these forces are competent patellar and quadriceps tendons. Together these tendons serve to provide lower leg motion through the extensor mechanism. It is important to have a firm understanding of the anatomy and biomechanics of this area to provide a rationale for sound treatment. Afflictions of the extensor mechanism can involve either one of these tendons in the form of tendinosis, tendinitis, and even rupture. The treating physician must have a sound understanding of the pathophysiology of these entities because correct diagnosis influences treatment options and rates of success. Tendinopathies can also occur in the presence of systemic diseases and external hormonal supplementation. As a result, it is important to have a firm grasp on the usual or anticipated history and physical examination findings as well as a preparation for the unusual and unexpected. Advances in imaging techniques have assisted the orthopaedic clinician in the evaluation and treatment of these various entities. An understanding of the techniques and anticipated radiologic findings given these methods is essential. It is when pathology is encountered that our notions of “normal” are tested and accurate diagnoses are more difficult to make. The scope of disease in the extensor mechanism can affect adults as well as children. Yet, it is important to understand the injuries that are likely to be found in certain age groups and populations. Treatment options in each of these groups also depend on an individual’s age and diagnosis. Situations can be envisioned in which operative as well as nonoperative treatment plans can be recommended. The wise physician must understand the options and indications for both. Recent studies have assisted in our understanding of this disease process. Randomized, prospective, clinically controlled trials have been outlined. Despite this, the age-old question, “When can I return to play?” continues to be asked. Again, it is incumbent on the treating physician to

have a firm understanding of the pathophysiology in each situation and the effects of time on healing. As well, there are always external influences such as the player, teammates, coaches, colleagues, and media. The number one priority must always be a consideration for the safety of the patient.

RELEVANT ANATOMY AND BIOMECHANICS Anterior knee pain is one of the most common clinical complaints in all age groups and can affect as many as 25% of all athletes. Symptoms can be acute in onset or develop in a setting of chronic pain and may present as either pain, instability, or a combination of both. Issues related to patellar instability are discussed elsewhere in this text. However, pain and disability that occur as a result of tendinitis, tendinosis, and ruptures of the patellar or quadriceps tendons will be discussed here. The extensor mechanism is composed of the quadriceps, patella, and patellar tendon or “patellar ligament” as some prefer. Because this structure actually connects two bones, the patella to the tibia, it properly should be described as the patellar ligament. Pathology and disability can be derived from any of these structures as well as the corresponding articular structures such as the medial and lateral trochlear ridges and intervening groove. In particular, the patella is the largest sesamoid bone in the body and possesses the thickest articular cartilage in the body (up to 6.5 mm). It has seven facets and is described elsewhere by the Wiberg classification (Fig. 22A-1). The patella has static stabilizers, including the quadriceps and patellar tendons, osseous anatomy created by the congruence of the patella within the trochlear groove, medial and lateral retinacular structures, and associated ligaments, including the medial patellofemoral (MPFL) and patellomeniscal ligaments. Desio and associates in 1998 described the MPFL and its influence on the restraint from lateral subluxation at 20 degrees of flexion.1 1513

�rthopaedic ����������� S �ports ������ � Medicine ������� 1514 DeLee & Drez’s� O L Type I

Type II

M

L

M Type II/III

Vastus lateralis muscle

Rectus femoris tendon

Type IV Patella

Type III

Jagerhut

Figure 22A-1   Wiberg classification. (Redrawn from Scott WN (ed): Surgery of the Knee, 4th ed. Philadelphia, Churchill Livingstone, 2006.)

The patella is also surrounded by dynamic stabilizers, which typically work in concert during functional activities to aid in maintaining physiologic function. These structures include the rectus femoris, vastus lateralis, vastus ­intermedius, and vastus medialis, including the vastus medialis obliquus. All are supplied by the femoral nerve, which is composed of the posterior divisions of the second, third, and fourth lumbar spinal nerves. The direct head of the rectus femoris originates at the anterior inferior iliac spine of the ilium, whereas the reflected head begins superior to the acetabulum.2,3 It therefore crosses two joints and can serve to flex the hip and extend the knee. The remaining three muscles of the ­quadriceps femoris include the vastus lateralis, which originates from the lateral lip of the linea aspera and lateral surface of the greater trochanter; the vastus intermedius—from the anterior aspect of the femoral shaft; and the vastus medialis—from the medial lip of the linea aspera along with the distal part of the ­intertrochanteric line. The quadriceps tendon results as a confluence of these individual tendons and both inserts on the proximal pole of the patella, and additionally surrounds the patella on three sides as it progresses from proximal to distal. The tendon has a considerable vascular supply; however, it does have an avascular area in the deep part of the tendon, which is reported to measure about 1.5 × 3 cm.4 Arteries from the descending branches of the lateral femoral circumflex, the descending geniculate, and the medial and lateral superior geniculate arteries provide the tendon with vascular ­nourishment.2,3,5 The patellar tendon is a continuation of the quadriceps tendon beyond the distal pole of the patella and inserts on the tibial tuberosity.3,5,6 The blood supply of the patellar tendon is not as spectacular as that of the quadriceps. It is supplied by vessels from the infrapatellar fat pad as well as the inferior medial and lateral geniculate arteries (Fig. 22A-2). The femoral artery courses through the anterior thigh deep to the rectus femoris and medial to the vastus medialis in Hunter’s canal. The femoral and its main branch, the profunda femoris, provide blood to the thigh. Distally, the femoral artery passes through the adductor hiatus 8 to 10 cm proximal to the knee to become the popliteal artery. Superior and inferior medial and lateral geniculate arteries

Rectus femoris muscle

Lateral rectinaculum Iliotibial tract Patellar tendon

Vastus medialis muscle Medial retinaculum Sartorius tendon Gracilis tendon Semitendinosus tendon Tibial tuberosity

Figure 22A-2   Normal anatomy of the extensor mechanism of the knee. (Redrawn from Matava MJ: Patellar tendon ruptures. J Am Acad Orthop Surg 4[6]:287-296, 1996.)

arise from the popliteal artery to supply the knee joint and associated structures.2 Overall, the extensor mechanism remains stable owing to the function of these static and dynamic stabilizers. It is only after injury to one of these structures occurs that the true value of the extensor mechanism can be appreciated. Consider the situation that follows injury or disruption to the patellar or quadriceps tendon. The patient is physically unable to ambulate on this extremity because function has been eliminated. Surgical intervention is considered necessary to restore function.

Biomechanics The knee consists of two joints: the femorotibial and patellofemoral joints. A detailed discussion of the biomechanics of the knee is beyond the scope of this section; however, it must be understood that the patella serves to increase the extensor moment arm by transmitting the longitudinal contractile force at a greater distance from the knee axis of rotation (Fig. 22A-3).7 The efficiency of the extensor mechanism increases 1.5 times by its presence. Patellofemoral contact initiates at 10 degrees of flexion and shifts distally to proximally with greater degrees of flexion (Fig. 22A-4).7 Weakness of quadriceps musculature can lead to in­ creased stresses and strains throughout the tendons of the extensor mechanism. Ascending stairs can increase forces within the patellar tendon about 3.2 times body weight. The greatest forces found within the patellar tendon occur at about 60 degrees of knee flexion.5,8,9 Huberti and colleagues described a concept referred to as the “extensor mechanism force ratio”.10 This is the ratio of the force found in the patellar tendon (distal) divided by the quadriceps tendon force (proximal)5,10 and is greater than 1.0 when the knee is in less than 45 degrees of flexion. With smaller degrees of flexion, the distal pole of the patella is articulating with the trochlear groove. The quadriceps tendon thus has a mechanical advantage at this point. Conversely, with knee flexion angles greater than

Patella 1515 O d P

P'

– Fe

Fe –

O d P'

Fe

O

Fe +

P'

O d P

Fe P'

Figure 22A-3   Patellar biomechanics.

45 degrees, the patellar articulation with the trochlear is considerably more proximal and allows for the patellar tendon to have the mechanical advantage.5,10 This, in turn, relates the position of knee flexion to the likelihood of tendon failure. At a position of knee flexion less than 45 degrees, the quadriceps tendon is more likely to be injured. In positions greater than 45 degrees, the patellar tendon is at higher risk for tensile failure.

Tendon Structure Tendon is a complex material consisting of collagen fibrils embedded in a matrix of proteoglycans. Both quadriceps and patellar tendons are mainly composed of water in

the extracellular matrix; however, the predominant cell type found within tendons is the fibroblast. This cell is arranged in the spaces between the parallel collagen bundles9 (Fig. 22A-5). Tendon is composed mostly of type I collagen and contains a high concentration of glycine, proline, and hydroxyproline.9 The secondary structure of collagen is related to the arrangement of each chain in a left-handed configuration; the tertiary structure refers to three collagen chains being combined into a collagen molecule; and the quaternary structure is related to the organization of collagen molecules into a stable, low-energy biologic unit based on a regular association of the basic and acidic amino acids of adjacent molecules. This quarter-stagger arrangement of adjacent collagen molecules results in oppositely charged amino acids being aligned.9 Thus, a great deal of energy is required to separate these molecules, accounting for the strength of this structure (Fig. 22A-6). Tendon possesses one of the highest tensile strengths of any soft tissue in the human body for two reasons. First, tendon is composed of collagen, which is one of the strongest fibrous proteins; and second, tendon collagen fibers are arranged parallel to the direction of the tensile force.9 The elastic modulus of human tendon ranges from 1200 to 1800 MPa (megapascals), whereas the ultimate tensile strength ranges from 50 to 105 MPa, and the ultimate strain ranges from 9% to 35%.9

PATHOPHYSIOLOGY OF TENDON INJURY The specific details of tendon injury continue to be investigated. Tendons can become injured as a result of direct trauma with laceration or contusion, or indirect trauma through tensile overload. However, it is generally accepted that healthy tendons do not rupture. Thus, in the situation in which there is tensile overload of the extensor

Medial

Lateral

Medial

0° 30°

120°

Lateral

90°

60°

90° 60° 30°

120° Figure 22A-4   Patellar contact areas. (Redrawn from Aglietti P, Insall JN, Walker PS, et al: A new patella prosthesis. Clin Orthop 107:175, 1975.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1516 DeLee & Drez’s� O

­ echanism, the forces will result in a transverse fracture m of the patella.11 Research into the pathophysiology of tendinopathy continues to proceed at a rapid pace, yet it continues to be one of the more difficult challenges in sports medicine. Its pathogenesis consists of mechanisms including repetitive chronic overloading, ischemia with reperfusion injury, microtrauma, hypoxia, and hyperthermia. The injury mechanisms can be correlated with the skeletal maturity of the individual, anatomic site involved, vascularity of the area, and magnitude of the applied forces. Generally, the development can be related to internal and external factors. Most tendons are able to withstand tensile forces larger than those exerted by the muscles or sustained by the bones. Therefore, these types of injuries often result in avulsion fractures or tendon disruption at the musculotendinous junction.9 Mid-substance tendon ruptures are less commonly seen; however, there is a common requirement of preexisting pathology. Kannus and Jozsa in 1991 studied the specimens obtained from the biopsy of spontaneously ruptured tendons in 891 patients.12 Most were either Achilles or biceps tendons. Age- and sex-matched control specimens were obtained for comparison, and no healthy structures were seen in spontaneously ruptured tendon.12 Characteristic histopathologic patterns in the ruptured tendons included hypoxic degenerative tendinopathy, mucoid degeneration, tendolipomatosis, and calcifying tendinopathy, either alone or in combination.12,13 These changes were found in 34% of the control tendons, but significantly less frequently (P < .001). Mechanical testing performed on patellar tendon has shown tensile strains to be significantly less mid-­substance than at the respective insertion sites on the patella and tibial tuberosity.9,14 Woo and colleagues demonstrated that at peak load just before tendon failure, the end-region strain at the insertion site is 3 to 4 times that seen in the mid-substance.5,9,14 Thus, healthy tendon rarely fails in its mid-substance. If this clinical entity is encountered, the physician must immediately consider metabolic derangements (Box 22A-1). Metabolic abnormalities have also been shown to influence the physiologic status and biomechanical functioning of tendons. These conditions can be innate, induced, and iatrogenic. Conditions such as diabetes mellitus can limit the blood supply to a given area through the associated vasculopathy and limit the repair ability of a structure after injury. Metabolic conditions such as gout, renal failure, hypothyroidism, and chondrocalcinosis can also lead to tendinopathy and resultant ruptures.15-22 Local and systemic corticosteroid injections have been shown to limit the inflammatory phase of healing.23 Certainly, ruptures have followed the administration of these agents.24 Although there is no clinical test available for human growth hormones, the orthopaedic surgeon might encounter a higher rate of tendon injury and ruptures in elite athletes who choose to administer these drugs illicitly. Other drugs have been associated with tendinopathy and ruptures. Fluoroquinolone antibiotics (e.g., levofloxacin, ciprofloxacin) have been shown to alter the extracellular matrix in tendons and can influence healing of an injured tendon.25,26 Ciprofloxacin also induces

Figure 22A-5   Histologic features of a normal patellar tendon with the typical collagen crimp pattern.

interleukin-1β–mediated matrix metalloproteinase-3 (MMP-3) release.27,28 MMPs are a family of proteolytic enzymes that have the ability to degrade the components of the extracellular matrix network and facilitate tissue remodeling.29-31 Fluoroquinolones inhibit tenocyte metabolism, reducing cell proliferation and collagen and matrix synthesis, a mechanism that may induce tendinopathy.25,32 It is currently believed that cyclic tensile loading of tendons is required to maintain normal tendon health.13 This repetitive mechanical loading of a tendon leads to a cellular matrix response that can either be adequate, leading to the adaptation of the tissue, or can be inadequate, leading to a transient weakness in a tendon. This process is referred to as mechanobiology or mechanotransduction, and it implies that when there is stretch of a tendon, there is a biologic response of cells. Mechanical signals in excess of this set point stimulate an anabolic response, whereas mechanical stimulus below this point stimulates a catabolic response. If there is continued loading in the face of an already weakened tendon, there is compromise of the tendon. This process can lead to an accumulation injury and inhibit the healing capacity of a tendon, leading to overuse injury.13 The adaptive and reparative ability of tendon can be exceeded when the tendon is strained repeatedly to 4% to 8% of the original length.5 Repetitive strain can result in microscopic or macroscopic injuries to the collagen fibrils, noncollagenous matrix, and microvasculature, resulting in inflammation, edema, and pain.5 Definitions of tendon pathology have included the terms tendinitis and tendinosis. Tendinitis infers the ­ presence of inflammatory cells and could be obtained from a histologic evaluation of the tendon involved. Early pathologic alterations that occur in the presence of repetitive microtrauma to the patellar tendon include inflammatory cell ­invasion, resultant tissue edema, and fibrin exudation in the paratenon.5 This is currently referred to as paratenonitis. Maffulli has recommended the use of the term ­tendinopathy as a generic descriptor of the clinical conditions described

Patella 1517 Microfibril quaternary structure

Collagen molecule

300 nm 1.5 nm diameter

10.4 nm 2 (i)

Tertiary structure

t (ii) 1 (ii)

1.74 nm Glycine Hydroxyproline HO Secondary structure

A

0.87 nm

Proline

Tendon

Microfibril

Sub fibril

Fascicle

Fibril

Collagen

Fibroblasts Crimp

B

Fascicular membrane

Figure 22A-6   Schematic drawing of the structural organization of collagen into the microfibril. (Redrawn from Woo SL, An KN, Arnoczky SP, et al: Anatomy, biology, and biomechanics of tendon and ligament. In Buckwalter JA, Einhorn TA, Simon SR [eds]: Orthopaedic Basic Science. Rosemont, Ill, American Academy of Orthopaedic Surgeons, 2000, pp 581-616, Figs. 4 and 5.)

in Table 22A-1. He believes that the terms ­ tendinosis and tendinitis should be used only after ­ histopathologic ­examination.27,28 If the microtrauma continues to overwhelm the reparative capability of the tendon itself, a situation exists whereby the inflammation becomes chronic. This process results in fibrosis and thickening of the paratenon and chronic peritendinitis.5 The development of tendinosis is thought to result from a continuum of chronic peritendinitis. Kannus and Jozsa showed the characteristic histopatho­ logic pattern found in a degenerative tendon to be one of mucoid degeneration, tendolipomatosis, and calcifying tendinopathy, either alone or in combination.12,13 Living tissues are usually in homeostatic balance. This balance can involve both inhibitory and stimulatory growth and survival signals.33 The pathway toward tendinosis was elucidated recently; however, the stimulus leading to these

Box 22A-1 C  onditions Associated with Occult Tendinopathy

• Hyperparathyroidism • Calcium pyrophosphate deposition • Diabetes mellitus • Steroid-induced tendinopathy • Fluoroquinolone-induced tendinopathy • Osteomalacia • Chronic renal insufficiency • Gout • Uremia • Systemic lupus erythematosus • Rheumatoid arthritis

�rthopaedic ����������� S �ports ������ � Medicine ������� 1518 DeLee & Drez’s� O

  TABLE 22A-1  Classification of Tendon Disorders New

Old

Definition

Histologic Findings

Clinical Signs and Symptoms

Paratenonitis

Tenosynovitis Tenovaginitis Peritendinitis Tendinitis

Inflammation of only the paratenon whether or not lined by synovium Paratenon inflammation associated with intratendinous degeneration

Cardinal inflammatory signs: warmth, swelling, pain, crepitation, local tenderness, dysfunction Same as above, often with palpable tendon nodule, swelling, and inflammatory signs

Tendinosis

Tendinitis

Intratendinous degeneration due to atrophy (aging, microtrauma, vascular compromise)

Tendinosis

Tendon strain or tear

Symptomatic overload of the tendon with vascular disruption and inflammatory repair response

Inflammatory cells in paratenon or peritendinous areolar tissue Same as above, with loss of tendon, collagen fiber disorientation, scattered vascular ingrowth, but no prominent intratendinous inflammation Noninflammatory intratendinous collagen degeneration with fiber disorientation, hypocellularity, scattered vascular ingrowth, occasional local necrosis, or calcification Three recognized subgroups: each displays variable histologic characteristics from purely inflammation with acute hemorrhage and tear to inflammation superimposed on preexisting degeneration, to calcification and tendinosis changes in chronic conditions. In the chronic stage, it may be (1) interstitial microinjury, (2) central tendon necrosis, (3) frank partial rupture, or (4) acute complete rupture

Paratenonitis with tendinosis

changes has yet to be identified. Yuan and colleagues in 2002 and Cook and associates in 2004 presented good evidence that the earliest identifiable morphologic changes in tendinosis occur in the tenocytes, not the collagen fibers.13,34,35 It is noteworthy that on microscopic evaluation of tendons undergoing degenerative changes, there is a paucity of inflammatory cells and a common finding of tenocyte morphologic and density changes along with accumulation of glycosaminoglycans and collagen fiber thinning and disarray, with or without neurovascular proliferation.13,35-40 Other causes of tendinosis include the presence of programmed cell death or apoptosis. Yuan and colleagues in 2002 demonstrated the occurrence of excessive apoptosis in ruptured human rotator cuff specimens.34 Lian and associates in 2007 performed a similar study on biopsy specimens from patellar tendon in patients with patellar tendinopathy diagnosed clinically and with typical magnetic resonance image findings. They used immunohistochemical methods using a polyclonal antibody, which recognizes active caspase-3, a feature found in apoptotic cells.36 They found that the samples that had tendinopathy displayed increased cellularity compared with controls and also showed a higher number of apoptotic cells.36 The influence and reason behind the initiation of apoptosis remains to be ­elucidated.13 The exact details of the development of tendinosis continue to become clearer. Continued research into the basic science of this process will aid in our understanding of the process and should assist in developing treatment strategies focused on the pathophysiology of tendinosis.

Often palpable tendon nodule that may be asymptomatic but may also be point tender; swelling of tendon sheath is absent.

Symptoms are inflammatory and proportional to vascular disruption, hematoma, or atrophy-related cell necrosis. Symptom duration defines each subgroup: A: Acute (<2 wk) B: Subacute (4-6 wk) C: Chronic (>6 wk)

EVALUATION OF QUADRICEPS AND PATELLAR TENDINOSIS Clinical Presentation and History Patellar tendinosis, or so-called jumper’s knee, results from microtears of the patellar ligament followed by a chronic inflammatory response. It is an injury that follows excessive use and is commonly seen in athletes who participate in sports that involve jumping, kicking, or leaping, such as volleyball and basketball.41 Thus, it is important to obtain a history of athletic activity from each patient. Lian and colleagues studied the prevalence of jumper’s knee among elite athletes and demonstrated the likelihood of developing tendinopathy while participating in various sports.42 Cyclists had a zero prevalence, whereas male basketball and volleyball players had prevalences of 32% and 44%, respectively. Players routinely demonstrated symptoms lasting longer than 2 years, and affected athletes had significant pain and functional losses.42 Ferretti and associates demonstrated a linear relationship between training volume and the prevalence of tendinopathy among volleyball players.43,44 They demonstrated a higher prevalence of tendinopathy among players who trained on a harder floor type. Symptoms may occur in younger patients who are encountering a rapid phase of growth. In these cases, there may be a relative discrepancy in length of a given tendon to the adjacent bony structures. This occurs when the ligament does not lengthen as fast as the bone is growing.41

Patella 1519

Figure 22A-7   Ultrasound of knee and patellar tendon. Arrow points to a hypoechoic region of the patellar tendon.

Commonly, patients complain of pain in the area of the patellar tendon. Traditionally, pain in tendinopathy has been attributed to inflammation. However, in 1999, Khan and colleagues reported that chronically painful Achilles and patellar tendons had no evidence of inflammation, and many tendons with intratendinous lesions detected on magnetic resonance imaging (MRI) or ultrasound were not painful.27,45 Many different causes have been proposed to explain the generation of pain in the degenerative tendon. Theories include high concentrations of the neurotransmitter glutamate, elevations of prostaglandin E246-49 and substance P,50,51 and an opioid system.52

Physical Examination and Testing On examination, a patient may have tenderness over the patellar ligament and have signs of inflammation in the area, such as redness, swelling, warmth, and crepitation. Pain is often centered on the distal patellar pole and the proximal part of the patellar tendon. There may be a feeling of “bogginess” centered over the tendon itself.5 The patient may also have pain with resisted use of the quadriceps and with full passive flexion of the knee.41 In 1973, Blazina and colleagues described four stages of jumper’s knee or patellar tendinopathy. In stage 1, there is pain only after activity. In stage 2, pain is present at the beginning of activity and disappears after a warm-up but may reappear with fatigue. In stage 3, pain is constant at rest and with activity. In stage 4, there is complete rupture of the patellar tendon.5,53

Imaging Imaging is not routinely necessary for the treatment of patellar or quadriceps tendinosis. However, plain radiographs should be obtained in the initial evaluation of all patients with patellar tendinitis. One should be alert for the presence of traction osteophytes at the distal pole of the patella, tendon calcification, and even decreased bone mineral density at the distal pole of the patella.5 Other

Figure 22A-8   Magnetic resonance image of knee showing thickening of patellar tendon.

modalities such as ultrasound and MRI are used when nonoperative treatments have failed to produce anticipated improvements or when planning surgical intervention. Ultrasound evaluations can demonstrate a hypoechoic signal within the fibers of the patellar ligament.41,54 In 2007, Warden and associates performed MRI and grayscale and color Doppler ultrasound on 30 patients with clinically diagnosed patellar tendinopathy and 33 activity-matched, asymptomatic participants.55 It was thought that ultrasonography was more accurate than MRI in confirming clinically diagnosed patellar tendinopathy. They added that grayscale ultrasound and color Doppler ultrasound together may represent the best combination for confirming clinically diagnosed patellar tendinopathy because the grayscale ultrasound had the greatest sensitivity, whereas a positive color Doppler test result indicated a strong likelihood that an individual was symptomatic (Fig. 22A-7).55 On sagittal MRI, there may be evidence of thickening of the patellar tendon, especially on the posterior and central or medial aspect (Fig. 22A-8).56,57 MRI can demonstrate a focus of abnormal signal intensity in the posterior aspect of the proximal patellar tendon (Fig. 22A-9). MRI can also be helpful in the planning of treatment when there is an absence of abnormality on T2-weighted images. This suggests that nonoperative treatments may be more likely to be effective.5

Treatment Options Nonoperative Various modalities have been used in the management of acute and chronic tendon disorders. However, there are few randomized prospective controlled studies demonstrating the efficacy of these treatments. The mainstay of current treatment includes decreasing or stopping the inciting activity, initiation of a period of active rest, use of nonsteroidal anti-inflammatory medications, and use of cryotherapy. Stretching and isometric strengthening of the quadriceps should be started immediately.41 ­However,

�rthopaedic ����������� S �ports ������ � Medicine ������� 1520 DeLee & Drez’s� O

Operative

Figure 22A-9   Magnetic resonance image of knee showing high signal intensity within the tendon.

isotonic or isokinetic exercises for strengthening of the quadriceps should begin only when the pain and tenderness have improved. Other modalities that continue to be investigated include extracorporeal shock-wave therapy, pulsed magnetic fields, direct current applied to tendons, laser therapy, radiofrequency coblation, administration of cytokines and growth factors, gene therapy, bone morphogenic protein-12, gene transfers, and tissue engineering with mesenchymal stem cells.27 In 2000, Panni and coworkers performed a clinical cohort study of 42 patients with patellar tendinopathy. After 6 months of nonoperative treatment as outlined ­previously, 33 patients (79%) showed symptomatic improvement and were able to return to sports.58

When nonoperative means have failed to produce improvement in symptoms and imaging studies have revealed evidence of intratendinous degenerative changes, operative intervention may be considered. One option is to split the ligament longitudinally and excise the ­ gelatinous material between normal fibers.41,54 The remainder of the ligament should be closed with ­ absorbable suture. In 1978, Roels and associates ­presented a case series of 10 patients who had this type of surgical intervention, all of whom were able to return to participation in sports.41,59 In the latter half of the study by Panni and coworkers in 2000, the remaining 9 patients of the 42 did not improve with nonoperative treatment.58 All 9 patients had Blazina stage 3 tendinopathy. Operative treatment consisted of removal of the degenerated areas of the tendon, multiple longitudinal tenotomies, and drilling of the lower pole of the patella at the site of tendon attachment. At a mean 4.8 years, clinical results were good to excellent in all patients. Of the group treated nonoperatively, results were better in the patients who had stage 2 tendinopathy than in those with stage 3.58 In 2006, Shelbourne and colleagues performed a similar study and looked at 16 elite athletes with 22 symptomatic and MRI-documented instances of patellar tendinitis.60 These patients had all failed nonoperative treatment and underwent tenonectomy of the necrotic portion in conjunction with stimulation of the remaining tendon by making multiple longitudinal cuts in the tendon. By a mean of 8.1 months, subjective improvement was noted in all 16 athletes. An ability to return to the same sport at a prior level of intensity was accomplished by 14 of 16 patients (87.5%).60 Arthroscopic patellar tenotomy has been advocated based on a retrospective outcome study.61 See Figure 22A-10 for a treatment algorithm.

Clinical diagnosis of patellar tendinopathy

3 months nonoperative treatment

Clinical improvement

No improvement

Imaging: US / MRI

Surgery

No abnormalities

Abnormal findings

Figure 22A-10   Treatment algorithm for patellar tendinopathy. US, ultrasound; MRI, magnetic resonance imaging.

Patella 1521

Weighing the Evidence In 2006, Bahr and colleagues performed a randomized controlled trial on surgical treatment versus eccentric training for patellar tendinopathy.62 Thirty-five patients (40 knees) with grade III patellar tendinopathy were randomized to surgical or eccentric strength training. There was no advantage demonstrated for surgical treatment compared with eccentric strength training. As a result and in line with previous recommendations, Bahr and colleagues advised that eccentric training should be tried for 12 weeks before an open tenotomy is considered.62 Certainly, an attempt at nonoperative management should be made before surgical intervention.

Authors’ Preferred Method When individuals present to us with a history and physi­ cal examination consistent with patellar tendinitis, our first step is to obtain plain radiographs of the knee. We look at the underlying personality of the knee, that is, the presence or absence of osteoarthritis, patellar malalignment and malpositioning, and spur formation. Usually, however, the radiographs do not change our first step of treatment. We also give consideration to obtaining an MRI of the knee or a standard technetium-99m bone scan. These images simply help us to determine prognosis and likelihood of recovery. Patients with pain and obvious spur formation at the distal pole of the patella are less likely to recover after nonoperative intervention. Individuals without gross edema within the tendon are at a higher likelihood of improving with nonoperative ­intervention. After the initiation of physical therapy, a clinical re-­evaluation at 6 weeks is routine. This appointment is made for patient reassurance and evaluation of compliance with the nonoperative treatment regimen. Discussions and communications with the physical therapist assist in making subsequent recommendations. Improvement at 6 weeks is a good indicator of future clinical success; however, failure to improve offers no insight. A full 3 months of therapy should be performed as outlined earlier in an effort to improve the clinical picture. If nonoperative interventions fail to produce clinical improvement, a discussion should occur that is focused on the surgical options and the risks and benefits of each procedure. Our preferred treatment of refractory patellar tendinitis is to perform a midline incision centered over the proximal third of the patellar tendon. We essentially perform a patellar tendon harvest as if we are obtaining a graft for an anterior cruciate ligament reconstruction with an autologous bone–patellar tendon–bone graft. Based on the location of tendinopathy, we may or may not include the opposite bone-tendon interface. If it is proximal only, we focus the bone-tendon harvest there and remove only that middle-third tendon distal enough to include the area of involvement.

CLASSIFICATION OF QUADRICEPS AND PATELLAR TENDON RUPTURES Ruptures of the patellar and quadriceps tendons occur with relative infrequency. However, they are devastating injuries that require surgical intervention in an effort to restore function. Patellar tendon ruptures are thought to occur less frequently than those in quadriceps tendons.63 Patellar tendon ruptures are usually seen in patients younger than 40 years,64 whereas quadriceps ruptures are more common in patients older than 40 years and are often associated with underlying medical conditions. Galen is credited with first describing a patient with a ruptured extensor mechanism. This injury was sustained as a result of a “wrestling match.”64 McBurney, in 1887, was the first to publish a single case in the American literature.64 He described a 50-year-old man who was struck by the edge of a heavy box just above the patella. The patient’s ruptured quadriceps tendon was sutured successfully with catgut and silver wire.64 Spontaneous ruptures tend to occur in elderly patients with minimal trauma as a result of degenerative changes related to aging.65 Additionally, preexisting medical problems such as chronic steroid use,66 rheumatoid arthritis, and diabetes mellitus may act as predisposing factors.67 Anabolic steroid use is associated with an increased risk for tendon ruptures.68-70 Recently, an association be­ tween androstenediol supplements and tendon rupture was presented.70 Patellar tendon rupture is the third most common cause of disruption of the extensor mechanism of the knee, after patellar fracture and quadriceps tendon rupture.66 ­Zernicke and coworkers in 1977 estimated that a force of 17.5 times body weight is required to cause rupture in healthy patients.66,71 This finding is consistent with the earlier statement that normal tendons do not rupture. Quadriceps ruptures have been demonstrated to occur in relationship to total knee arthroplasty. In brief, these injuries have been associated with impingement of the tendon by the femoral prosthesis, excessive resection of the patella during resurfacing leading to intrinsic weakening of the tendon, and tissue ischemia as a result of iatrogenic sacrifice of the lateral geniculate artery during release procedures.72-74 Additionally, Ritter and colleagues noted that the likelihood of a quadriceps tendon rupture increased with extensive lateral releases.75 However, the authors stated that the rupture was not related to the decreased vascularity of the patella and surrounding tendons, but proposed it to be a result of the “release.” A recent case of quadriceps rupture in the setting of total knee arthroplasty was attributed to the physical erosion of the patellar implant through the tendon and not presumed vascular insufficiency following balancing.76 Further detail is beyond the intent and scope of this textbook. Prompt diagnoses of patellar and quadriceps tendon disruptions are important to obtain because of the consequences of neglected injuries. Accentuated scar tissue formation, tendon shortening and contraction, and atrophy of associated musculature are all problems encountered with neglected injuries.63,64,66,76,77

�rthopaedic ����������� S �ports ������ � Medicine ������� 1522 DeLee & Drez’s� O

  TABLE 22A-2  Classification of Patellar Tendon or Quadriceps Tendon Ruptures Ruptured Tendon

Age Seen (yr)

Causes of Injury

Diagnosis

Treatment

Patella

Usually <40

Traumatic; underlying tendinopathy

Early primary reconstruction; if delayed diagnosis, can perform delayed reconstruction

Quadriceps

Usually >40

Traumatic; underlying tendinopathy

Pain: inability to perform a ­ straight-leg raise Radiographs: patella alta Ultrasound: gap Magnetic resonance imaging: increased signal intensity within the tendon indicating gap formation Pain: inability to perform a ­ straight-leg raise Radiographs: patella baja Ultrasound: gap Magnetic resonance imaging: increased signal intensity within the tendon indicating gap formation

There is no widely accepted classification system for patellar tendon or quadriceps tendon ruptures. ­Clinically, it is helpful to group them based on the location, ­configuration, and chronicity of the rupture (Table 22A-2).

EVALUATION OF QUADRICEPS AND PATELLAR TENDON RUPTURES Clinical Presentation and History Patellar and quadriceps tendon ruptures can at times be difficult to diagnose. Siwek and Rao in 1981 performed a retrospective analysis of 36 quadriceps tendon and 36 patellar tendon ruptures.64 They found that 38% of these injuries were misdiagnosed initially.64,78 Diagnosis may be more difficult when the injury is accompanied by hemarthrosis, which can mask the presence of a suprapatellar gap.77 As such, the orthopaedic physician should have a high index of suspicion with injuries around the knee. Despite this, diagnostic failure rates of 10% to 50% have been reported, and delays in diagnosis have ranged from days to months.64,79-81 Bottoni and associates reported that the mechanism of injury is that of deceleration, whereby the knee is in a ­ semiflexed position with a strong contraction against a planted or obstructed foot or lower leg.78 Patients with patellar tendon rupture often present with a popping or tearing sensation.78 The pain associated with a quadriceps tendon rupture is described as an immediate, intense tearing sensation at the time of rupture.77 Immobilization of the extremity in extension provides pain relief.

Early primary reconstruction; if delayed diagnosis, can perform delayed reconstruction

an intact retinaculum. Knee aspiration with an intra­articular local anesthetic can relieve pain and allow for an adequate physical examination and diagnosis.77 A suprapatellar gap or palpable depression just superior to the patella is pathognomonic for quadriceps tendon rupture (Fig. 22A-11).77

Imaging Various imaging modalities may be of use in the diagnosis of patellar and quadriceps tendon ruptures. Plain anteroposterior and lateral radiographs should be obtained initially to evaluate for fracture, osteochondral injuries, or patellar dislocations. Associated findings on radiographs may include the obliteration of the respective quadriceps or patellar tendon shadows, calcific densities at the proximal or distal patellar poles, and patellar height alterations due to the disruption of the tethering effect of the noninjured tendon. The Insall-Salvati ratio has been used traditionally as a method to determine patellar height. This measurement carries higher intraobserver and interobserver error rates. The Insall-Salvati ratio is the patellar tendon length divided by the greatest diagonal length of the bony patella and numerically represented by the number 1. This measurement does not have high reproducibility compared with other systems of measurement (Fig. 22A-12).

Physical Examination and Testing Physical examination of a quadriceps tendon rupture presents with the triad of pain, inability to actively extend the knee, and a suprapatellar gap on palpation.77,80-82 Patients are unable to actively extend the knee but may be able to maintain extension against gravity through

Figure 22A-11   Patient with a rupture.

Patella 1523

Figure 22A-12   Insall-Salvati ratio. Figure 22A-14   Lateral radiograph of patellar tendon rupture.

Clinically, the Blackburne-Peel method of measuring patellar height yields a lower interobserver error and lacks ratio changes by flexion of the knee from 30 to 50 degrees of flexion.83,84 The Blackburne-Peel system measures a perpendicular distance from a line drawn along the tibial plateau to the inferior articular margin of the patella. This length is compared with the length of the articular surface of the patella. Normal values for males and females are 0.805 and 0.806, respectively (Fig. 22A-13). With disruption of the patellar tendon, there is proximal migration of the patella and a resulting higher Insall­Salvati ratio (>1) and higher Blackburne-Peel measurement (>0.805) (Fig. 22A-14). With disruption of the quadriceps tendon, there is a distal migration of the patella and a resulting lower InsallSalvati ratio (<1) and lower Blackburne-Peel measurement (<0.805). There may also be an associated anterior tilt of the patella away from the trochlear groove (Fig. 22A-15). High-resolution ultrasonography has been shown to be an effective means of evaluating both the patellar and quadriceps tendons in situations of acute and chronic tendon injuries.63,77

Figure 22A-13   Blackburne-Peel method of measuring patellar height.

MRI is the most sensitive test available to evaluate and visualize the injured quadriceps and patellar tendon.63,77 It also provides accurate information about location of injury, providing for a rational approach to reconstruction. Routine use of MRI is limited only by its cost (Fig. 22A-16).

Treatment Options Nonoperative Incomplete ruptures of quadriceps and patellar tendons may be treated nonsurgically. Accurate diagnosis is essential, and immobilization in extension for 3 to 6 weeks is recommended.77,78 Gradual knee flexion following this period of immobilization is advocated after the patient achieves good quadriceps muscle control and is able to perform a straight-leg raise without discomfort.77 The physician and patient are to expect the sequelae of weakness, extension lag, and quadriceps atrophy.

Operative Complete ruptures of quadriceps and patellar tendons require operative intervention. A delay in surgery can lead to associated complications and problems noted earlier, which may lead to poorer outcomes. Surgical intervention

Figure 22A-15   Lateral radiograph of quadriceps tendon rupture.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1524 DeLee & Drez’s� O

A

B

Figure 22A-16   A and B, Magnetic resonance images of patellar tendon disruption.

for acute ruptures involves a simple end-to-end repair with or without a reinforcing cerclage of nonabsorbable suture, tape, or wire.63 Although less common, mid-substance tears can usually be repaired through reapproximation with heavy nonabsorbable suture.78 More commonly, the disruption occurs at the patella. In this situation, a series of Krackow-type sutures85 should be placed in either the distal quadriceps or proximal patellar tendon, depending on the rupture type. In the case of a patellar tendon rupture, a small groove or trough may be made at the nose of the patella to allow for the tendon to nestle into a bleeding bony bed. The Krackow sutures can then be passed through drill holes in the patella. An anterior cruciate ligament guide may be used to facilitate drill and suture passage through the patella. Augmentation with hamstring tendons may also be performed in the case of patellar tendon disruption. Neglected and recurrent ruptures are understandably more difficult to manage. In the case of a neglected ­patellar tendon, the operating surgeon must be aware of ­patellar mobility, height, and likelihood to return to its usual position in the trochlea. Augmentation with fascia

lata or hamstring tendons has been done (Figs. 22A-17 and 22A-18).63

Weighing the Evidence Most patients who undergo a primary repair of patellar or quadriceps rupture relatively soon after the injury can expect good to excellent results.63,77 In independent reviews, Matava63 and Ilan and associates77 showed that there does not appear to be a relationship among the configuration of the rupture, method of repair, and clinical outcome; however, the timing of the repair did appear to correlate with clinical outcome. Most studies indicate a 1-week timeframe as optimal for performing the repair. No large series have evaluated the outcome of chronic neglected patellar tendon disruptions. Alternatively, Konrath and coworkers looked at 51 quadriceps tendon rupture repairs in 39 patients.79 Most patients were satisfied and were able to return to their previous state of employment. Interestingly, they found no correlation between the length of time from tendon rupture to surgical repair and final strength, functional score, or activity score.77,79

Authors’ Preferred Method Primary Patellar and Quadriceps Tendon Repairs

Figure 22A-17   Quadriceps tendon rupture repair.

We prefer to use a standard anterior incision to allow full exposure of the extensor mechanism. Commonly, the ruptures occur adjacent to the patella. This allows for the use of a running, locking, Krackow-type suture in the residual tendon with free sutures emanating from the tendon end. A bony trough is created in either the distal or proximal pole of the patella to allow the tendon stump to lie adjacent to a bleeding bony bed. The nonabsorbable suture (No. 2 FiberWire, Arthrex, Naples, Fla) is passed through the patella by using drill holes in a longitudinal fashion through the patella.

Patella 1525

A

B

Figure 22A-18   Recurrent quadriceps tendon rupture. A, Recurrent quadriceps rupture. B, Codivilla method of quadriceps tendon lengthening and repair.

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS The most common complications following patellar and quadriceps tendon repairs and reconstructions are diminished quadriceps strength and loss of knee flexion. In particular, patients have difficulty obtaining full knee flexion. Other complications that may be encountered through surgical intervention relate to the risks of surgery itself— infection, bleeding, recurrent rupture, wound dehiscence, and deep venous thrombosis—and to the anesthetic risks— myocardial infarction and stroke.

C l

r i t i c a l

P

S U G G E S T E D

R E A D I N G S

Bahr R, Fossan B, Loken S, Engebretsen L: Surgical treatment compared with eccentric training for patellar tendinopathy (jumper’s knee): A randomized, controlled trial. J Bone Joint Surg Am 88:1689-1698, 2006. Ilan DI, Tejwani N, Keschner M, Leibman M: Quadriceps tendon rupture. J Am Acad Orthop Surg 11:192-200, 2003. Lian OB, Engebretsen L, Bahr R: Prevalence of jumper’s knee among elite athletes from different sports: A cross-sectional study. Am J Sports Med 33:561-567, 2005. Lian OB, Scott A, Engebretsen L, et al: Excessive apoptosis in patellar tendinopathy in athletes. Am J Sports Med 10:1-7, 2007. Matava MJ: Patellar tendon ruptures. J Am Acad Orthop Surg 4:287-296, 1996. Panni AS, Tartarone M, Maffulli N: Patellar tendinopathy in athletes: Outcomes of nonoperative and operative management. Am J Sports Med 28:392-397, 2000. Shalaby M, Almekinders LC: Patellar tendonitis: The significance of magnetic resonance imaging findings. Am J Sports Med 27:345-349, 1999. Sharma P, Maffulli N: Tendon injury and tendinopathy: Healing and repair. J Bone Joint Surg Am 87:187-202, 2005. Siwek CW, Rao JP: Ruptures of the extensor mechanism of the knee joint. J Bone Joint Surg Am 63(6):932-937, 1981. Teitz CC, Garrett WE, Miniaci A, et al: Tendon problems in athletic individuals. J Bone Joint Surg Am 79(1):138-152, 1997.

o i n t s

 atellar tendinitis routinely responds to nonoperative P modalities and treatments. l Extensor mechanism disruptions can be difficult to diagnose. A high clinical suspicion must be maintained. l Early repair in the setting of patellar tendon rupture leads uniformly to good results. l Monitored rehabilitation is as important as the surgical intervention.

R eferences Please see www.expertconsult.com

1526 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

B

Osteochondroses Vasilios Moutzouros and Richard D. Parker

Osteochondrosis results from disordered endochondral ossification of a previously normal developing epiphysis.1 This process can occur throughout the maturation of the axial and appendicular skeleton. In the skeletally immature patient, patellofemoral tendinopathies should be referred to as extra-articular osteochondroses. The symptomatology that directs a clinician to make the diagnosis of tendinopathy is in reality a change in the bone and cartilage of the apophysis. The inflammation that is typically seen with overuse injuries in adults is found in a different location in an immature patient. Young athletes present with complaints of pain localized in a specific region of the knee, classically at the origin or insertion of a tendinous structure. Although in mature individuals the advanced radiographic findings are consistently within the tendon, the young athlete shows changes in the bone and cartilage alone. The area of the apophysis is the “weak link” in the young athlete. Therefore, the stresses of repetitive activities lead to changes within this area of development as opposed to the tendon itself. In the extensor mechanism complex, a variety of osteochondroses are found. Changes can lead to symptoms at the level of the tibial tubercle, infrapatellar region, or superior pole of the patella. The eponyms used for these osteochondroses are well known. Much has been written in regard to Osgood-Schlatter (Fig. 22B-1) and Sinding-Larsen– Johansson disease (Fig. 22B-2), which affect the tibial tubercle and distal pole of the patella, respectively. These entities are found in individuals undergoing rapid growth who participate in athletic activities. The incidence in nonathletic

adolescents has been quoted as 4.5%, as opposed to 21% in the active adolescent population.2 The onset of symptoms of both conditions directly correlates with the rapid growth spurt, which is earlier in girls (ages 10 to 13 years) when compared with boys (ages 12 to 14 years). Osgood-Schlatter disease occurs bilaterally in 20% to 30% of patients.3 Recently, Tyler and McCarthy reported on osteochondroses of the superior pole of the patella.4 Histologic samples from two adolescent female patients with activity-related pain at the insertion of the quadriceps tendon showed findings consistent with those seen in Osgood-Schlatter and Sinding-Larsen–Johansson disease. Extra-articular osteochondroses of the patella should not be confused with osteochondritis dissecans (OCD) of the patella. OCD is a lesion that affects the subchondral bone that leads to subchondral delamination or ­ sequestration (Figs. 22B-3 and 22B-4).5 The condition was first described in two patients by Sir James Paget, and given the name OCD by Koenig.6 At the time, Koenig believed that these lesions developed as a result of trauma and ­deteriorated because of subsequent inflammation. Although he later ­acknowledged that his theory was flawed, the name ­ continued even though the inflammatory process is not the cardinal cause of the abnormality. Even though the direct cause of OCD is unclear, the most universally accepted ­hypothesis is that it results from repeated trauma and ­ ischemia in convex articular areas that experience stress.5,7-10 This condition most often affects the knee joint: 85% of lesions are found in the medial femoral condyle, 14% in the ­lateral femoral condyle, and fewer than 1% in the patella and trochlear

Figure 22B-1   Osgood-Schlatter disease of the tibial apophysis.

Figure 22B-2   Sinding-Larsen–Johansson disease.

Patella 1527

areas.5 Patellar OCD was first reported in the 1930s, with repeated studies showing the rarity of the condition.8,10-15 As our appreciation of this condition has developed, new treatment strategies have surfaced to prevent the advancement of patellofemoral degeneration.

Osteochondrosis of the tibial tubercle was first recognized in the early 1900s. Two separate clinicians reported on a phenomenon in which active adolescents were complaining of pain with running and jumping centered over their tibial tubercle. Both Osgood and Schlatter were credited with identifying this condition and thus were attached to its title. They described the disease as occurring in children who are undergoing rapid growth who place stress on the developing tubercle through patellar tendon force. Through this force, others theorized that the maturing tubercle is partially or completely avulsed. The authors distinguished this entity from a tibial tubercle avulsion fracture and stressed that the condition is due to repetitive loading of the area.16 This is the classic description of traction apophysitis (see Fig. 22B-1). Osgood-Schlatter disease was originally reported to be found more often in boys than in girls.2,17 With the increasing number of young female athletes, the condition is now being seen at a similar rate to young males. Young girls present with symptoms earlier, between the ages of 10 and 13 years, as opposed to the boys, who develop the condition between the ages of 12 and 14 years.18 The prevalence of Osgood-Schlatter disease was reportedly 21% of a group of athletic adolescents, compared with only 4.5% of nonathletic individuals.2 Running, basketball, and hockey are the most common sports in which boys develop symptoms, whereas gymnastics, volleyball, and figure skating are the most common for girls. The disease is found bilaterally in 20% to 30% of individuals. Often, enlargement of the tibial tubercle is found from repeated repair attempts.

The development of the tibial tubercle has been investigated to better understand the location in which this ­condition occurs. Ehrenborg described four stages of development of the tibial tubercle19,20: the cartilaginous, apophyseal, epiphyseal, and bony stages, in that order. The cartilaginous stage is from birth to 8 to 10 years of age and consists of a cartilaginous tongue-like mass. The apophyseal stage begins when one or more centers of ossification are identified in the tibial tuberosity. The epiphyseal stage quickly follows and is heralded when the ossification center from the proximal tibia and tibial tuberosity coalesce. The final stage is seen when the closure of the physis of the tuberosity is complete, the bony stage. Ehrenborg believed that Osgood-Schlatter disease results from repetitive traumatic avulsion of the patellar tendon from the tibial ­tubercle. Histologic and advanced radiographic studies have been performed to better describe this process of maturation in the setting of Osgood-Schlatter disease.16,21 Hirano and colleagues21 evaluated the progression of Osgood-Schlatter with magnetic resonance imaging (MRI). They described five stages in the progression and resolution of this condition. The normal stage (1), with no MRI changes, was seen in the earliest of symptoms of Osgood-Schlatter, mild pain and minimal swelling. The early stage (2) showed low signal intensity at the secondary ossification center, which did not translate to changes on plain radiographs. Cartilaginous damage of the tibial tuberosity and tearing of the secondary ossification center with an open shell separation were seen in the progressive stage (3) on MRI. The knee in this stage also showed swelling at the insertion of the patellar tendon. The terminal stage (4) showed signs of healing and resolution of swelling. It was in this stage that ossicles were identified and pulled superiorly. The final stage (5), healing, showed almost normal radiographic findings with prominence of the tibial tubercle. Histologically, Ogden and Southwick described three zones within the maturing tibial tubercle.16 The proximal region is similar to the proximal tibial physis and is

Figure 22B-3   Osteochondritis dissecans defect of the patella treated with osteochondral transplantation.

Figure 22B-4   Osteochondritis dissecans of the trochlea (grade I).

OSGOOD-SCHLATTER DISEASE

1528 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

c­ omposed of columnar cartilage. The midzone is composed of fibrocartilage. The distal zone is a transition of fibrous tissue into the tibial perichondrium. As the individual matures, the growth plate begins to close from proximal to distal.

Etiology The cause of Osgood-Schlatter disease has been investigated for some time without a definitive answer. The condition is universally found in adolescents who are undergoing a rapid period of growth. During this time, an imbalance of growth between the bone and the muscle unit may lead to a decrease in overall flexibility. With this loss, the apophysis becomes more susceptible to overuse injury.22 Authors have evaluated a variety of anatomic factors that may predispose to the condition. Patella alta has been associated with Osgood-Schlatter disease, but no definitive predisposition has been established.2 A single study evaluated radiographs in patients with and without OsgoodSchlatter–associated patella alta. Further investigation of these data showed some difficulties with assessment of the patellar height on plain radiographs.23 A later investigation attempting to link patella alta, Osgood-Schlatter, and tibial tubercle avulsion fractures showed no association among the three.24 Ogden and colleagues presented no relationship between preexisting Osgood-Schlatter and acute tibial tubercle avulsion fractures. They differentiated the two entities by commenting that avulsion fractures occur more commonly in the mature athlete, just before full closure of the tubercle.24 One study that has shown a significant anatomic variant in these patients focused on the patellar angle.25 The patellar angle is the angle between the articular surface and the inferior pole of the patella. Individuals with the condition showed a smaller patellar angle, which the authors associated with a need for greater quadriceps force to perform similar activities as others.25 Although the maturation of the tibial tubercle predisposes to this condition, overuse activities are required to develop the syndrome. Repetitive microtrauma has been seen as an integral part of the disease process because individuals who are not active develop the condition at an exponentially lower rate.2 A single event such as a forceful jump or direct contact with repeated kneeling can aggravate the syndrome. It is believed that repetitive, submaximal stresses acting on the immature patellar tendon–tibial tubercle junction, leading to minor avulsion and attempts at repair, are what define Osgood-Schlatter.16 This extensive description of causality attempts to highlight the requirement of a developing tubercle and active adolescent.

Diagnosis The clinical presentation of Osgood-Schlatter disease is straightforward. Patients complain of activity-related pain focused over the tibial tubercle and distal patellar tendon. With questioning, adolescents will report discomfort that worsens with running, jumping, or kneeling. Typically, patients will not halt their activities because of discomfort. The history may include a gradual onset of pain that is accompanied by swelling over the area. Basketball, volleyball, gymnastics, and soccer are often the activities ­associated

with this syndrome. The common threads in these activities include running and jumping that loads the knee in flexion, leading to an eccentric quadriceps contraction. Tenderness, swelling, and prominence of the tibial tubercle are often found on physical examination. Patients may walk with an antalgic gait, which is often noticed by the parent rather than the individual. The tendon may also show tenderness with palpation primarily in the distal half. Bone irregularities are often palpated in chronic situations. Acute cases may present with an extensor lag, confusing the diagnosis and raising concern for a tibial avulsion fracture. Passive range of motion of the knee is full, and knee effusions are not seen. Quadriceps and hamstring tightness is not uncommon, with some investigators postulating that it is part of the cause. The differential diagnosis of Osgood-Schlatter disease includes avulsion fracture of the tibial tuberosity, patellofemoral stress syndrome, pes anserinus bursitis, Sinding-Larsen–Johansson disease, and infection. Plain radiographs of the affected knee are ordered for evaluation of the tibial tuberosity. The lateral radiograph of the knee is most useful in assessing the extensor mechanism (see Fig. 22B-1). The apophysis can show separation or fragmentation of the tibial tubercle in the Osgood­Schlatter patient. Enlargement of the tubercle may also be seen. Soft tissue swelling can also be appreciated along with thickening of the patellar tendon.2 More important, radiographs assist in excluding unlikely diagnoses such as neoplasm or infection. Ultrasound has been studied and used in some institutions for diagnosis of Osgood-Sclatter.26 Because of the high level of operator dependence, this modality is not widely recommended. MRI is appropriate for cases in which the extent of soft tissue swelling or tubercle prominence is unusual.

Natural History A handful of studies have followed the natural history of Osgood-Schlatter disease.2,27,28 Krause and colleagues reported on the natural history of Osgood-Schlatter in 50 patients. Half the patients received some form of nonoperative treatment, and the rest did not. Most patients seen in adulthood for follow-up had no residual symptoms, regardless of treatment. The authors concluded that two groups of patients can be identified when looking back at the presentation of the disease and how it progressed. One group of patients had early radiographic evidence of fragmentation with ossicle formation or an abnormality in the ossification of the tubercle. These patients were found to continue to have symptoms when kneeling at their adult follow-up. The second group showed no radiographic abnormalities and was characterized as the isolated soft tissue swelling cohort. These patients had no symptoms clinically, and few had any radiographic evidence of prior Osgood-Schlatter disease. There was also an association of fragmentation and residual pain in the segment of the treatment group placed in plaster immobilization. These individuals were more likely to have tubercle fragmentation and pain with kneeling at time of follow-up. The group of natural history investigations all showed no increased incidence of patellar instability, anterior knee pain, or premature proximal tibial growth arrest in patients with Osgood-Schlatter.2,27,28

Patella 1529

Authors’ Preferred Method The treatment of Osgood-Schlatter disease is guided by the severity of symptoms. As evidenced by the natural history evaluations, this syndrome is generally a self-limited condition.27 Improvement can be gradual, which should be stressed to the patient and family. A period of 12 to 18 months can pass before the complete resolution of symptoms. This correlates with the duration needed for closure of the epiphysis. Although most patients continue with their athletic endeavors, the severity of pain in some individuals leads to a change in position or even in sport. Nonoperative management includes activity restriction or adjustment and proper padding of the tubercle to avoid painful symptoms. Severe pain should lead the clinician to recommend a short time of absolute sport restriction until the symptoms improve. This may include a brief period of immobilization for those unable to perform their activities of daily living. Pain that does not resolve with restrictions should be investigated further for other causes. Nonimpact activities, such as swimming and cycling, can be implemented to maintain a patient’s cardiovascular fitness. Hamstring and quadriceps flexibility exercises will maintain knee range of motion and may quicken recovery. Injections over the site of discomfort are contraindicated. For those patients whose symptoms do not improve, surgical intervention can be successful. The two most common procedures performed are ossicle excision and tibial tubercle prominence resection.28,29 Ossicle excision is performed by creating a longitudinal split in the patellar tendon and enucleating the fragment. The success rate of this procedure has been reported at 93%. It is most useful in patients who have a distinct separate ossicle in the proximal aspect of the tibial tubercle that is painful with direct contact.28 Tubercle prominence resection has also shown good results in recalcitrant cases of OsgoodSchlatter.30,31 These patients have symptoms for more than 1 year and are well past growth plate closure. The greatest success is in individuals who have pain with activities and particularly with kneeling. Flowers and Bhadreshwar reported an 85% success rate following this procedure.30 Rehabilitation following these procedures includes the placement of a hinged knee brace locked in extension and full weight-bearing. Early knee range of motion is allowed, but weight-bearing in extension is mandatory for the first 3 to 4 weeks if tendon healing to bone is needed. Motion of the knee should be limited if there is any concern that the bone-tendon unit is at risk. Straight-leg raises can be initiated immediately after surgery. Once tendon healing is complete, usually between 4 and 6 weeks, active range of motion and strengthening are allowed.32

SINDING-LARSEN–JOHANSSON DISEASE Osteochondrosis of the inferior pole of the patella was first described in the early 1920s by two independent investigators. The two studies described calcification and ­ossification at the inferior pole of the patella, presumably due to ­persistent traction at a location in which growth is

progressing. Similar to Osgood-Schlatter disease, this syndrome affects the maturing adolescent. Sinding-Larsen– Johansson disease occurs predominantly in boys, with no recent data showing a change due to an increase in young female athletic activity. The condition typically affects girls at a younger age (10 to 13 years) than boys, which is believed to be due to their earlier skeletal maturity.33 The presentation of Sinding-Larsen–Johansson disease universally is found in an active adolescent. The individual reports activity-related pain over the anterior knee. Running and jumping sports become difficult to participate in. Severe cases can lead to pain with ascending or descending stairs. There may be a history of intermittent symptoms that improve with rest from activities. Patients can often directly point to the area of concern. The examination shows point tenderness at the patella–patellar tendon junction along with variable amounts of swelling. Calcification at this location is found in chronic cases of Sinding-Larsen– Johansson disease. The remainder of the knee examination is typically normal. A palpable gap, though, in the location of the patella–patellar tendon junction, should raise the suspicion of a patellar sleeve fracture. This step-off, along with the inability to perform a straight-leg raise, is the classic presentation.34 Full evaluation for patellar alta or baja and limitations in knee or hip range of motion that can lead to anterior knee pain need to be recognized. Assessment for tibial tubercle pain and prominence should be performed to differentiate this condition from Osgood-Schlatter. Although rare, the diagnosis of both these entities has been reported.35,36 Traditionally, Sinding-Larsen–Johansson disease is seen in slightly younger individuals because the maturation of the inferior pole of the patella occurs before that of the tibial tuberosity. Radiographic assessment includes anteroposterior, lateral (see Fig. 22B-2), and Merchant views of both knees in the adolescent. Calcification or ossification of the inferior pole of the patella can often be identified. Other entities, such as patellar sleeve fractures, Osgood-Schlatter, bipartite patella, patella alta, and patella baja, can be evaluated with these radiographs. Medlar and Lyne developed a radiographic staging system for Sinding-Larsen–Johannson disease.33 They divided the findings into five stages. Stage I shows normal radiographs. Stage II has irregular calcification at the inferior pole of the patella. Stage III shows coalescence of the calcification. Stage IV is divided into IVa, which is incorporation of the calcification into the patella (normal radiograph), and IVb, which is coalescence of the calcific mass separate from the patella. The authors thought that these are the changes that occur through the progression of the disease. Ultrasound has also been used as a diagnostic tool for this condition, but although successful, has not been accepted into widespread use.26 The differential diagnosis of Sinding-Larsen–Johansson disease includes a number of entities. Again, it is critical to appreciate the findings consistent with a patellar sleeve fracture because its treatment approach is quite different.37 Slightly older patients may be diagnosed with patellar tendinitis, or “jumper’s knee,” which is a chronic proximal patellar tendinitis associated with repetitive eccentric ­quadriceps loading. Other diagnoses include bipartite patella, Osgood-Schlatter, and patellar stress fractures, which are quite uncommon.

1530 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Much like Osgood-Schlatter disease, Sinding-Larsen– Johansson disease is a self-limited condition. The resolution of symptoms follows the full maturation of the inferior pole of the patella. This takes a period of 12 to 18 months. The management of this syndrome is focused on decreasing pain and inflammation. Activity modification, along with the use of ice, anti-inflammatory medication, and other modalities, is recommended. Physical therapy is employed with an eccentric quadriceps loading program and lower extremity stretching.38 A knee sleeve can be used both for compression and protection of the area as the adolescent returns to athletic activities. Patients are allowed to participate in sports to a limited degree until symptoms begin to improve. Late sequelae of the disease are few, with one report of a fracture through a previously united Sinding-Larsen– Johansson ossicle.39

SUPERIOR POLE OSTEOCHONDROSIS Superior pole osteochondrosis has been reported in the ­literature on only a handful of occasions.4,40-42 The condition can be distinguished from bipartite changes in the patella by both radiographic appearance and clinical symptomatology. In accordance with Sinding-Larsen– ­Johansson and Osgood-Schlatter disease, this syndrome is a traction apophysitis. The condition is found more often in young males when reviewing the case reports published. The largest case series, by Batten and Menelaus, presented the condition in six boys.40 The young men ranged in ages from 10 to 11 years, and all were active. The location of their discomfort was the anterior portion of the knee, not specifically the proximal pole. Upon radiographic investigation, the proximal pole of the patella showed fragmentation similar to that seen in Sinding-Larsen–Johansson disease. Grogan and colleagues reported on seven cases of proximal pole fragmentation which they believed was an avulsion injury due to direct trauma. They found these cases to be similar in appearance and presentation to avulsion injuries to the distal pole of the patella and tibial tuberosity.37 Tyler and McCarthy assessed the histologic change that occurs in these patients and found osteonecrosis with reparative changes.4 These findings are similar to those seen in Sinding-Larsen–Johansson and OsgoodSchlatter disease.43 Osteochondrosis of the proximal pole of the patella presents with anterior knee pain. Unlike Osgood-Schlatter and Sinding-Larsen–Johansson disease, the patient may not be able to localize the pain specifically.40 Pain can be reproduced with palpation over this area. Individuals may also have symptoms of other extra-articular osteochondroses.40 Radiographic changes can be at different stages depending on the chronicity of symptoms. The treatment includes the use of ice, anti-inflammatory medication, a knee sleeve, and some period of activity modification. There are no reports of operative interventions because of this self-limiting process. Although our understanding of this condition is limited because of its rarity, the principles of nonoperative management lead to the resolution of this syndrome.

OSTEOCHONDRITIS DISSECANS OF THE PATELLA OCD is a disease process that can occur in a variety of joints. First described by Sir James Paget, OCD continues to challenge investigators on its true cause. Multiple theories have been proposed and range from ischemia to endochondral ossification defects. Although Paget first described the condition, Koenig labeled it OCD to reflect his theory that the disease process was composed of an inflammatory process that dissects the affected area.6 Although his hypothesis was in error, the name has continued to be used. The condition was first recognized in the knee but subsequently has been described in a number of joints, including the ankle, elbow, and hip. Initially described in the adult population, OCD has been found in maturing adolescents, with a healing potential that varies a great degree from the mature population. OCD of the patella was first described by Rombold in 1936. Although quite uncommon, it must remain in the differential diagnosis of anterior knee pain.7,8,13,44 Similar to OCD in other areas, the differentiation of juvenile, adolescent, and adult types of lesions is necessary to follow an appropriate treatment algorithm.45 The juvenile type of OCD is characterized by patients with wide open growth plates. The adolescent category involves patients who are in the process of growth plate closure. Adult lesions are found in patients with fully closed growth plates. The peak prevalence of juvenile OCD is during the preadolescent period, 10 to 12 years of age, and the adolescent and adult segments vary in range depending on the process of maturation. The incidence of OCD lesions of the knee is higher in males than females.46 The condition can be found bilaterally in 20% to 25% of patients. However, the lesions often do not correspond in size.47 The lateral posterior portion of the medial femoral condyle is the classic location of OCD of the knee and makes up 70% to 80% of all OCD lesions of the knee joint.5 Lateral femoral condyle defects are found in the inferocentral region and compose 15% to 20% of cases.48 Patellar lesions are quite rare, making up 5% to 10% of knee lesions.7 The defect is most often found in the distal half of the patella.7,10 Desai and colleagues reported that more than 85% of lesions are found in the inferomedial area of the patella.7 The rate of bilaterality ranges from 16% to 25%; this is most often found in young men in their second and third decades.10

Etiology The true cause of OCD is unknown. The number of theories in the literature has been narrowed to just a few. Most center on an initial traumatic event, followed by repetitive microtrauma that leads to subchondral delamination and sequestration on convex articular surfaces.5,46 In evaluating the pathology of these lesions, ischemia is an overwhelming finding.5,9,49 A number of investigators have associated macrotrauma from sport-related injuries to OCD lesions of the knee. Although there is no controversy that trauma has an association, there is a large proportion of patients who have no history of trauma with OCD lesions. Patellar subluxation or dislocation was associated with OCD lesions of the patella in a number of studies.7,8,50 Another ­associated

Patella 1531

finding is a flattened articular surface of the patella, as described by Bruns and colleagues.51 Although this study and others have described associations, no distinct cause of OCD dissecans of the patella can be delineated. Whether a significant traumatic event with repetitive microtrauma is the initiating factor, the progression of an OCD lesion has been described. The trauma, whether macro or micro, occurs at a vulnerable site on the convex surface of the joint. Subchondral avascularity occurs in the area along with change in the overlying cartilage. The body responds with attempts at healing by revascularization. The success of this process is substantial in the juvenile and adolescent cases. Adult patients unfortunately do not experience the amount of creeping substitution that is seen in younger individuals. Because of the lack of revascularization and healing, fissuring of the cartilage and subchondral fractures result. The modulus mismatch that is created with these defects is what causes the fracture and further leads to cartilage change. The persistent stress seen by these lesions accentuates the deterioration. Once fissuring and fracture occur, synovial fluid can intrude behind the fragment, limiting any chance of healing. The nonunion that is created can lead to fragment instability and loose body creation.

Diagnosis Patients with OCD lesions in the knee often present with nonspecific, poorly localized knee pain that is related to activity. Historical questioning on the onset of symptoms, previous episodes of knee pain, and association with activities is important to guide the diagnosis. An association with recent trauma is often made by the patient but is frequently incidental. Reports of intermittent knee effusions, mechanical symptoms, or stiffness are common. Lesions can be seen in both adult and adolescent patients. The classic range has been reported as 12 to 40 years of age.7,11,14 There have been no reports of a familial predisposition. Medical conditions such as chronic renal disease or malignancy have been associated with the development of OCD. These cases typically follow the natural history of avascular necrosis without hope for resolution. Bilateral lesions do occur; therefore, a previous history of an OCD lesion should raise suspicion for one on the opposite side. Physical examination will vary depending on the status of the lesion. A knee effusion can be seen in cases of unstable lesions. Limitations in range of motion can occur in patients with a loose body but are uncommon. Tenderness can be appreciated over the compartment for which the individual is experiencing symptoms, similar to those with meniscal pathology. Patellar apprehension and translational testing create discomfort in the setting of patellar

OCD. Crepitus can also be appreciated in most of these patients.12 Chronic symptoms can lead to quadriceps atrophy and difficulty with straight-leg raising. OCD lesions of the knee can affect an individual’s gait. Wilson has described patients with medial femoral condyle lesions walking with an external rotation gait.52 Pain has been theorized to occur in these patients when the tibial spine impinges against the lesion. Although not seen in all patients, discomfort with forced internal rotation of the tibia should raise suspicion for a medial femoral condyle lesion. Defects in other locations, including in the patella, can present with an antalgic gait. No connection between patellar alignment and OCD has been reported in the literature. Standard imaging of the knee is ordered when there is suspicion for OCD. The tunnel view is best for lesions of the femoral condyle. The difficulty of diagnosis of patellar OCD has been reported.11,14 In respect to the patella, skyline and lateral views are the most useful in diagnosis. The affected area typically shows an area of lucency with subchondral sclerosis in chronic lesions. A history and physical examination that suggest this defect should be further investigated if radiographs are negative. Computed tomography (CT) or MRI is ordered in these situations. Both modalities provide further detail, with MRI delineating fluid in the area of the lesion and CT showing regions of sclerosis. Healing of defects can also be evaluated using these measures. Fluid shown behind a lesion on MRI is suggestive of delayed union. Finally, cartilage changes can be seen on MRI, and new sequencing techniques are becoming more reliable in identifying them.

Natural History The progression of OCD lesions is dependant on the age and maturity of the patient. Individuals with open physes have the greatest potential to heal these defects. Adolescent patients with closing growth plates can heal their lesions to a variable degree.46,48,53-55 Although the likelihood of healing in an adolescent approaches 50%, the possibility of healing without intervention in an adult is minimal. The mature patient often develops some degree of degenerative changes in relation to the lesion. Chronic symptoms, stability and size of the fragment, and quality of the overlying articular cartilage have significant effects on the prognosis of the patient. Large defects in a weight-bearing portion of the knee predict a poor long-term outcome. In patellar OCD, disability can be profound. Stair climbing and rising from a seated position can cause debilitating pain. Although suggested, patellar OCD has not been associated with a diminished opportunity for healing when compared with other areas of the knee.

Authors’ Preferred Method Traditionally, patellar OCD lesions are discovered when they produce significant anterior knee pain and mechanical symptoms. Because of this, defects can often be loose or completely detached from their original bed. In these situations, nonoperative management is not appropriate. Lesions

that are diagnosed earlier in their natural history progression can be treated with a nonoperative approach. Individuals should be instructed to refrain from impact activities that place an ­ increased amount of load on the patellofemoral joint. A variable course of activity modifications can lead to Continued

1532 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d healing of a patellar lesion. As ­mentioned earlier, the likelihood of success is directly related to the immaturity of the patient. Juvenile individuals can heal these lesions to a much greater degree than their adult counterparts. Reports have even shown immature patients healing their patellar defects after 8 weeks of cast immobilization. The patient and family need to be instructed that the period for healing can be prolonged, and that healing may take up to 12 months in some individuals. If a lesion is discovered incidentally, an immature patient should be informed of the condition and observed. Activity modifications in this subset is unnecessary and implemented only if the lesion becomes symptomatic.14 Adult individuals with an incidental finding should be followed with serial radiographs to ensure the stability of the lesion and health of the compartment. Surgical interventions are often required for patellar OCD. The indications for operative management are no different from those for lesions in other areas. Symptomatic, unstable, or detached lesions need some form of intervention. A nonoperative attempt at treatment is undertaken for stable lesions in immature patients. When the decision to operate is made, a number of different techniques can be used to treat the condition. The size and location of the lesion, along with the type of intervention needed, determine whether an arthro­ scopic approach can be used. Previously, the most common technique for patellar OCD was fragment excision.14 With the advancement of technology, clinicians are becoming more comfortable with fragment repair techniques. Fragment excision or removal of a loose body is a common intervention for many OCD lesions. A small fragment in a non–weight-bearing portion of the knee joint can be appropriately treated in this fashion. Typically, lesions in the patella are in weight-bearing portions of the articulation.7,11 Excision is appropriate for small lesions or fragments without appropriate subchondral bone. Chronic lesions may appear sclerotic on evaluation. A sclerotic fragment or bed creates a poor environment for healing and typically leads to fragment excision. When this modality is chosen, an ­arthroscopic approach can be used. The fragment, if loose, can be found in the gutters of the knee joint or in the notch area. Once removed, the bed of the lesion can be débrided of any fibrous tissue. Drilling or microfracture of the ­region can be performed to elicit a vascular response. When micro­ fracture fails, osteochondral transplantation (see Fig. 22B-3) or autologous chondrocyte implantation can be a consideration. Numerous reports have shown that excision with base débridement can relieve pain and return patients to their activities of daily living.8,10,12 No studies have investigated this condition and commented on return to athletic participation. The rehabilitation of this procedure requires the patient to be protected from placing weight on the lesion for 6 weeks. The patient is on crutches and ambulating with partial weight in full knee extension. Range of motion is allowed, but the degree of flexion accepted varies among clinicians. After the initial 6 weeks, full motion is allowed, and strengthening begins once all motion is recovered. Lesions that remain in their anatomic location (see Fig. 22B-4) can be treated with in situ drilling, fixation, or grafting with fixation. In situ drilling is reserved for patients who

have failed nonoperative management. The lesions in these situations are well maintained without evidence of significant instability. The overlying cartilage is intact. Drilling is performed for lesions that cannot be opened for appropriate débridement of the base without causing significant trauma. These closed, stable lesions can be approached in an antegrade or retrograde fashion. Retrograde techniques preserve the articular cartilage surface and are preferred by many clinicians.56,57 Drilling can be performed using a K-wire driver or drill. An anterior cruciate ligament guide can be used to ensure appropriate tracking. The drill is advanced to subchondral articular margin. The amount of passes performed is dependent on the size of the lesion, with tracks being about 3 to 5 mm apart. An all-arthroscopic approach minimizes scarring, allows for an earlier reestablishment of motion, and leads to a shorter rehabilitation course.7 With a closed stable lesion, full range of motion and closed chain exercises are begun within a few days. Patients remain on crutches with weight-bearing in extension for 4 weeks. Thereafter, strengthening and graduated return to activities allows for the lesion to heal. Radiographic follow-up should be guided by symptoms using either plain radiographs or MRI. Lesions that are in appropriate position but unstable can be fixed in situ. At surgery, arthroscopic probing of these defects will show a closed, unstable lesion. The articular surface is ballotable, with the size of the lesion being outlined by invagination of the cartilage. Symptomatic patients can be assisted with simple fixation of the lesions. Again, a retrograde or antegrade approach can be used for fixation. Matava and Brown reported a technique in which bioabsorbable pins are used in a retrograde fashion to fix patellar OCD lesions.13 One limb of a ring forceps is introduced arthroscopically to stabilize the lesion. The opposite limb is placed on the skin overlying the patella and clamped to provide compression on the area. An incision is made within the ring, and the pins are introduced in a retrograde fashion. Sekiya and colleagues presented a case report with a similar retrograde technique to allow for fixation without compromising the articular surface.58 Stabilization of the defect allows for mechanical stability, with the hope of neovascularization leading to healing. Antegrade techniques have also been successful and are often needed when there is a lack of subchondral bone involved in the lesion. A variety of devices can be used for fixation, from darts to small metallic screws. Metallic devices often need to be removed owing to settling of the defect. Bioabsorbable products have also had difficulties, with inconsistent degradation leading to fragmentation of the device.59,60 Outcomes of in situ fixation are related to the size of the lesion and the maturity of the patient. The postoperative rehabilitation is similar to in situ drilling, although it may be progressed at a slower rate. In situ bone grafting with fixation is employed for lesions that are unstable and that provide a window for access to the subchondral bone. An ongoing or chronic lesion can have a significant loss of subchondral bone. In these cases in which the lesion is open and loss is observed, autologous bone graft can be placed in the defect. Appropriate débridement of ­fibrous tissue at the base is first performed, ­followed by

Patella 1533

Authors’ Preferred Method—cont’d placement of the graft. The roof of the lesion is then closed back onto the base and fixed in an antegrade or retrograde fashion. Typically, antegrade techniques are preferred because the lack of bone on the roof of the lesion does not allow for appropriate purchase of a retrograde device. Antegrade placement, however, must be completed with the device countersunk to the level of the subchondral bone to prevent articular abrasion. In the case of a closed lesion with bone loss, retrograde placement of bone graft can be performed. Postoperative care requires the patient to be protected for a longer time. Weight-bearing in extension with crutches for 6 to 8 weeks is recommended. Advancement in strengthening and return to athletic activities is graduated thereafter. Open reduction with internal fixation is reserved for large, unstable lesions that are partially or completely detached. The fragment must have potential for healing without significant evidence of sclerosis. The base of the lesion also must show some evidence of healing potential after débridement of any fibrous tissue. Lesions without macroscopic evidence of healthy articular cartilage should not be replaced. Cartilage shell lesions without appropriate subchondral bone should also be discarded. The fragment should have its subchondral backing débrided along with the base of the lesion. After, the base should be bone-grafted if necessary and

C

r i t i c a l

l Osgood-Schlatter

P

o i n t s

disease is typically self-limited. However, sometimes surgery is necessary for ossicle excision. l Sinding-Larsen–Johansson disease is osteochondrosis of the inferior pole of the patella and is self-limited and rarely requires surgery. l Osteochondrosis of the superior pole of the patella is rare and requires symptomatic treatment. l OCD of the patella accounts for 5% to 10% of all OCD lesions about the knee. l Operative treatment of patellar OCD ranges from removal and débridement to repair to cartilage restoration.

evaluated for bleeding potential. The fragment can then be fixed to its original position using a variety of devices. Both retrograde and antegrade techniques are appropriate for fixation. Kocher and colleagues reported an 85% rate of success with operative fixation in juvenile patients with OCD of the knee.61 Although patients had mostly condylar lesions, the significance of this study was found when investigating lesions that were fully detached. All six fully detached lesions healed with fixation. Clinicians therefore should reconsider fragment excision or removal carefully in this subset of patients for large lesions. The rehabilitation protocol for open reduction and internal fixation is similar to that of in situ fixation as mentioned earlier. Two salvage techniques for problematic OCD lesions of the patella are osteochondral grafting and autologous chondrocyte implantation. These interventions are performed in patients who have large symptomatic unstable lesions that are unsalvageable. Only recently have reports been published using these modalities in the knee.62-64 No long-term studies have been presented investigating the outcomes of these procedures. Although interesting in approach, chondrocyte implantation or mosaicplasty for patellofemoral chondral injuries has been disappointing when compared to condylar interventions.

S U G G E S T E D

R E A D I N G S

Federico DJ, Lynch JK, Jokl P: Osteochondritis dissecans of the knee: A historical review of etiology and treatment. Arthroscopy 6(3):190-197, 1990. Ogden JA, Southwick WO: Osgood-Schlatter’s disease and tibial tuberosity development. Clin Orthop 116:180-189, 1976. Paletta GA Jr: Juvenile osteochondritis dissecans of the knee. Tech Sports Med 6:268, 1998. Schenck RC Jr, Goodnight JM: Osteochondritis dissecans. J Bone Joint Surg Am 78(3):439-456, 1996. Sekiya LC, Fontbote CA, Harner CD: Arthroscopically assisted retrograde fixation of patellar osteochondritis dissecans using fluoroscopic guidance: A case report and technical note. Arthroscopy 19(7):E1-17, 2003.

R eferences Please see www.expertconsult.com

1534 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

C

Subluxation and Dislocation 1. Patellofemoral Instability: Acute Dislocation of the Patella Timothy Steiner and Richard D. Parker

Acute dislocation of the patella is a common sporting injury encountered on the field, in the emergency department, and in the orthopaedic clinic. Often dismissed as a minor injury, patellar dislocations require a lengthy recovery and rehabilitation period, and most patients note limitation in strenuous activity at 6 months.1 The outcome is often less than optimal despite treatment method and can be complicated by pain, instability, recurrent dislocation, chondromalacia, and ultimately patellofemoral arthritis. The true incidence of acute patellar dislocation is not known because of the lack of prospective, populationbased studies. Only two studies of a captured population exist, and both report on the same managed care patient base of nearly 400,000 members.1,2 Fithian and associates recently published a prospective cohort study on this population, examining the epidemiology of patients with acute patellar dislocations.1,2 The overall annual risk for a first-time dislocation was 5.8 per 100,000 members, with 61% of injuries occurring during sports. There was an increasingly higher incidence in younger and female patients. The annual risk for those with prior subluxation or dislocation was 3.8 per 100,000 members, with a statistically higher proportion of older and female patients (Table 22C1-1). There is a clear distinction in these two groups of acute patellar dislocations: those with normal anatomy who ­suffer

  TABLE 22C1-1  Risk for Acute Patellar Dislocation from Study of a Captured Managed Care Population Patient Age (yr)

Female

Male

Both

33 7 1

25 10 1

29 9 1 5.8

6 12 0

12 11 2 3.8

First-Time Dislocation

10-17 18-29 30+ Overall

History of Prior Subluxation/Dislocation

10-17 18-29 30+ Overall

18 11 3

Adapted from Fithian DC, Paxton EW, Stone ML, et al: Epidemiology and natural history of acute patellar dislocation. Am J Sports Med 32(5): 1114-1121, 2004.

a macrotraumatic event and those with a microtraumatic event with predisposing anatomy, malalignment, and a history of prior subluxation. This distinction was recognized by Gallie nearly a century ago as dislocation “as a result of direct violence or unusual muscular action” or dislocation “without any obvious causative traumatism.”3 Over time, the recognition, evaluation, treatment, and outcomes of these distinct patient populations has evolved. The purpose of this section is to discuss the evaluation, treatment options, and outcomes of the first-time acute patellar dislocation. Extensive discussion of anatomic predisposition and reconstructive procedures is reserved for the section on recurrent patellar dislocation.

ANATOMY A number of anatomic risk factors have been associated with acute patellar dislocation (Box 22C1-1). Although both bony architecture and soft tissue restraints contribute to stability of the patella, the importance of the medial soft tissue structures of the knee is paramount in the discussion of acute dislocation. Warren and Marshall first described the structures on the medial side of the knee in detail. Their anatomic dissection described a consistent three-layered pattern, with fibers of the medial patellofemoral ligament (MPFL) transversely oriented in the second layer.4 Merely a thickening in the medial retinaculum, early anatomic dissections found only variable existence of the MPFL. In 1981, Reider identified a distinct ligament in only 35% of knees.5 More recently, the importance of the medial soft tissue structures has been clarified, causing investigators to look more closely Box 22C1-1  Risk Factors for Acute Patellar Dislocation Patella alta Increased Q angle Femoral anteversion Systemic hypermobility Trochlear dysplasia or hypoplasia Lateralized tibial tubercle Lateral patellar tilt External tibial torsion Genu valgum Vastus medialis obliquus hypoplasia

Patella 1535

at the anatomy of the MPFL. Conlan’s landmark work on the medial patellar restraints found a distinct ligament in 29 of 33 cadaveric specimens, and fine wispy fibers were found in four.6 Since then, most studies have shown the presence of a distinct structure in 100% of specimens.7-13 The anatomy of the MPFL has since been well studied. It originates on the medial epicondyle with fibers attaching both to the adductor tubercle and the superficial portion of the medial collateral ligament (Fig. 22C1-1). It inserts on the proximal two thirds of the patella, with fibers inserting both directly onto the patella and indirectly through fibers of the vastus medialis obliquus portion of the quadriceps tendon. The length of the ligament ranges from 4.5 to 6.4 cm, and its width is slightly greater at its patellar insertion than its femoral origin.12,13 Tensile testing of the MPFL has shown the mean failure load is 208 Newtons.14 ­ Histopathologic examination has shown the presence of nerve fibers, but not mechanoreceptors, within the MPFL.13 The MPFL is almost universally disrupted in cases of acute lateral patellar dislocation. Vainionpaa and colleagues presented operative findings in a prospective study on operative treatment for acute dislocation of the patella.15 Although the MPFL was not identified, complete disruption of the medial retinaculum was found in 54 of 55 knees. Sallay and associates examined the MPFL during open exploration of the medial knee structures in 16 patients with acute dislocation of the patella.16 Fifteen of 16 had a complete tear of the MPFL off the femoral insertion at the adductor tubercle. Nomura studied the MPFL in 67 knees operatively treated for acute or chronic dislocation and devised a classification system to describe the location of injury (Fig. 22C1-2).17 Seventeen of the

18 acute dislocations had complete disruption of the MPFL; 7 were avulsions from the adductor tubercle, and 10 were intrasubstance tears. In a second study, Nomura reported on acute dislocation in 28 knees, where all had disruption of the MPFL with an equal distribution of avulsion and intrasubstance tears.18 Four early studies defined the role of the MPFL in patellar stability. In 1993, Conlan and associates6 published the first biomechanical study demonstrating the importance of MPFL in preventing lateral displacement of the patella with the knee in full extension. Their sequential sectioning showed that the MPFL contributed 53% of the restraining force, whereas the patellomeniscal ligament contributed an additional 22%. Subsequent work by Desio and colleagues showed the MPFL to be the primary restraint at 20% of knee flexion, contributing 60% of the restraining force, with the patellomeniscal ligament, patellotibial ligament, and medial retinaculum contributing lesser amounts. Interestingly, their study also found that the lateral retinaculum provided restraint against lateral patellar dislocation, a finding that seems counterintuitive and important in the discussion of lateral release.7 Hautamaa and coworkers sequentially sectioned the medial structures at 30 degrees of flexion and measured displacement of the patella under a constant load. Isolated sectioning of the MPFL resulted in a 50% increase in lateral translation, and repair of the MPFL restored translation to that of the normal knee.9 These studies all examined patellar stability in only one position of knee flexion, whereas Nomura’s work evaluated displacement under a constant load through a range of knee flexion from 20 to 120 degrees. He concluded that the MPFL resisted lateral patellar translation from 20 to

Figure 22C1-1   Anatomy of the medial patellofemoral ligament, originating near the adductor tubercle and inserting on the proximal two thirds of the patella. (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

1536 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 22C1-2   Classification of medial petellofemoral ligament disruption in acute patellar dislocation. Partial injury (A), avulsion from the origin near the adductor tubercle (B). (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

A

B 90 degrees, but had minimal contributions in higher degrees of flexion due to the bony constraint of the femoral trochlea.10 A number of other anatomic factors have been associated with patellar dislocation, including skeletal malalignment, trochlear dysplasia, patella alta, and quadriceps dysplasia. These factors become increasingly important when evaluating the patient with recurrent dislocations and therefore will be discussed in the following section.

CLINICAL PRESENTATION Most acute patellar dislocations occur during sport. Two Kaiser studies evaluating a captured managed care population have shown that sporting injuries account for 61% to 72% of acute patellar dislocations.1,2 A military study out of Finland showed that 89% of dislocations occurred during military exercises or sports. Of the 11% of ­injuries sustained during leisure time, most occurred while dancing.19

Acute dislocation of the patella can occur either by a direct blow to the knee or indirectly, as the body rotates around a planted foot. The player will usually fall down secondary to pain and disruption of the normal extensor mechanism alignment. The player may sense that something is out of place, but often the patella will dislocate and spontaneously reduce during the injury. If the patella remains dislocated, it may be palpable over the lateral femur, and the medial femoral condyle will appear prominent (Fig. 22C1-3). In lateral dislocations, the patella will often spontaneously reduce as the knee is extended. If not, the physician or trainer may need to place slight pressure on the patella as the knee is extended; this usually leads to a palpable “clunk” and improvement in the athlete’s discomfort as the patella reduces. The indirect mechanism of injury is more common than a direct blow, encompassing 66% to 82% of dislocations.1,19 This mechanism is noncontact and occurs with

Patella 1537 Figure 22C1-3   Clinical appearance of a knee with a lateral dislocation of the patella. (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

the knee in slight flexion and valgus as the tibia externally rotates relative to the femur. It can occur on a planted foot as the femur and body rotate internally, such as the hind leg of a baseball player swinging hard at a pitch. Alternatively, the free foot can be forced into external rotation, such as a soccer player whose instep kick is met with excessive resistance, or a snow skier whose ski acts as a lever arm (Fig. 22C1-4).

A

Causing as few as 7% of dislocations,1 the direct blow is less common and can either be to the medial or lateral side of the knee. A direct blow to the lateral side of the knee can force the knee into valgus, which predisposes the patella to dislocation during quadriceps contraction. A direct blow to the medial side of the knee can force the patella out directly, especially when the knee is in an at-risk position of 20 to 30 degrees of flexion.

B

Figure 22C1-4   Mechanisms of acute patellar dislocation. A noncontact dislocation occurs by external rotation of the lower leg relative to the body (A), whereas contact injuries are caused by a direct blow to the medial side of the knee (B). (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

1538 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

PHYSICAL EXAMINATION

ASSOCIATED INJURIES

Most athletes present for examination in the subacute phase after the patella has either spontaneously reduced or has been reduced on the field by a provider. In this subacute phase, the examination is often clouded by swelling and effusion, especially if osteochondral fracture is present. Care must be taken to ensure that the patella is not persistently subluxated or dislocated in these cases. Aspiration of a tense joint effusion may be therapeutic, and the appearance of the fluid may aid in diagnosis. Hemarthrosis is present with more severe injuries to the knee, most commonly anterior cruciate ligament tears, followed by meniscal tears and patellar dislocations.20-24 A larger effusion may represent a more extensive injury to the medial retinacular structures or a concomitant osteochondral injury. Lipohemarthrosis indicates fracture, and when there is an appropriate history of patellar dislocation, osteochondral fracture must be presumed. After aspiration, local anesthetic may be injected to augment pain relief for improved physical examination and radiographic ­assessment.25 With any knee injury, a complete examination of the lower extremity is warranted. Observation of skeletal alignment may show genu valgum or rotational abnormalities that are predisposing factors. In the acute phase of a painful, swollen knee, gait examination often provides little useful information. A ligamentous examination is necessary to rule out concomitant injury to the cruciate and collateral ligaments, and the joint line should be examined for tenderness secondary to meniscal injury. A complete neurovascular examination should document any deficit. Finally, generalized hyperlaxity should be noted by examining finger metacarpophalangeal hyperextension, thumb to forearm apposition, and elbow hyperextension; such findings may predispose to patellar dislocation but also have been shown to reduce the incidence of associated osteochondral injury.26 The knee should then be palpated for areas of maximal tenderness. Although a palpable defect over the medial retinaculum is possible, tenderness alone is the norm. Pain may be localized at the origin, at the insertion, or along of the course of the MPFL. Tenderness at the medial patellar border or along the lateral femoral condyle may suggest osteochondral injury. Rarely, a loose osteochondral fragment may be palpated within the joint. Tenderness or asymmetry at the distal portion of the vastus medialis obliquus may suggest significant disruption of its tendinous insertion. The classic examination finding for instability following acute patellar dislocation is the apprehension test, which is performed by placing a laterally directed force on the patella with the knee in 20 to 30 degrees of flexion. A positive finding occurs when the patient has a sense of pain and impending dislocation (Fig. 22C1-5). In addition to apprehension, there may be increased translation of the patella when compared with the uninjured knee. Recently, Tanner and associates tested a modification of the apprehension test on 10 cadaveric specimens, showing that the distal-lateral displacement is more sensitive than direct lateral displacement in detecting disruption of the MPFL.27

The most common findings associated with acute dislocation of the patella are chondral and osteochondral injuries. The incidence of such injury has been extensively studied. In a recent publication, Stefancin and Parker systematically reviewed the literature on first-time patella dislocation. In their compilation of 70 articles, the incidence of osteochondral fracture confirmed by open surgery, arthroscopy, or MRI (magnetic resonance imaging) ranged from 0% to 73%, with an overall incidence of 24% (Table 22C1-2).25 Stanitski found that only 32% of chondral injuries and 29% of loose bodies noted at arthroscopy were identified on preoperative plain radiographs.28 The high incidence of chondral injury and potential for delayed diagnosis should heighten the clinician’s index of suspicion. Osteochondral injuries from patellar dislocation have a characteristic pattern. The medial border of the patella can be avulsed as the retinaculum and MPFL are stretched. More commonly, there is injury to the medial patellar facet and the lateral femoral condyle (Fig. 22C1-6). Osteochondral fragments may remain attached, may become loose in the joint, or may be retained in the peripatellar retinacular tissue. Nomura recently published a series of acute dislocations with operative documentation of chondral injury, which included both cracking and full-thickness defects in the articular surface. Twenty-four percent of knees had cracks alone, 19% had a cartilage defect without cracks, and 57% had cartilage defect with cracks. Of the cartilage-injured knees, 100% had involvement of the patella, whereas only 31% had involvement of the lateral femoral condyle. The main site of chondral injury was the medial facet of the patella, whereas cracking alone most commonly involved the central dome (Table 22C1-3).29 Whether this chondral damage occurs during dislocation or reduction is unclear. In 1943, Milgram suggested that quadriceps contraction during reduction produced shear forces on the patella and lateral femoral condyle. Recently, a two-stage mechanism was proposed based on the incidence of osteochondral injury and bone contusion patterns seen on MRI.30 The first stage involves shearing of both patellar and trochlear articular surfaces during dislocation. During reduction, the patella first impacts the nonarticular portion of the lateral femoral condyle and then slides over the lip into the trochlea. The convex shape of the patella makes it susceptible to shear injury during both dislocation and reduction, whereas the trochlea is protected during reduction by its concave shape. Because the injury to the lateral femoral condyle occurs during the dislocation, the location of the osteochondral defect correlates to the degree of knee flexion at the time of injury.

IMAGING A complete evaluation of the suspected patellar dislocation must include basic imaging studies. Plain radiographs of the knee should be obtained, and MRI should be considered to evaluate damage to the articular surfaces and the medial retinacular structures. The plain radiograph series should include standing anterior-posterior, standing 45-degree flexion weight-bearing, lateral, and axial views. Because

Patella 1539 Figure 22C1-5   Examination for patellar instability. The patient experiences a sensation of the patella dislocating as a lateral force is applied to the medial border of the patella with the knee slightly flexed. This has been termed apprehension. (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

A

B

most dislocations are reduced before presenting for evaluation, radiographs rarely show the patella in the dislocated position. However, they will show osteochondral fractures, persistent subluxation, and abnormal skeletal morphology. In the case of acute patellar dislocation, the lateral and axial views provide the most essential information. The lateral view provides information about the height of the patella, trochlear depth, and patellar tilt. Patella alta is a known risk factor for patellar dislocation and can be determined on the lateral radiograph by numerous methods, including the Insall-Salvati ratio,31 modified Insall-Salvati ratio,32 Blackburne-Peel ratio,33 Caton-Deschamps ratio,34 and Blumensaat line (Fig. 22C1-7). The Blackburne-Peel ratio, which is based on consistent bony landmarks, is the most reproducible and has the most moderate results for classification into patella alta and baja.35,36 The Blumensaat line is least reliable and shows the weakest correlation with the other indices.37 In addition, the ratio indices are independent of knee flexion between 30 and 50 degrees, but the

Blumensaat method introduces a potential source of error by requiring exactly 30 degrees of flexion.35 Trochlear depth and patellar tilt can also be determined from the lateral radiograph. For accurate interpretation of these factors, the lateral view must be a “true” lateral with the posterior borders of the femoral condyles overlapping. Dysplasia of the proximal trochlea, which is best evaluated on the lateral radiograph, is a critical anatomic factor in the evaluation of patellar dislocation. The axial view shows more distal trochlear morphology because of the angle of the x-ray beam.38 Patellar tilt can also be evaluated on the lateral radiograph. In full extension, evaluating the degree of tilt in the lateral view is highly sensitive in confirming prior patellar dislocation.39 A detailed discussion of these factors is reserved for the following section on recurrent patellar dislocation. Although both the lateral and axial views show trochlear morphology, the axial view of the patellofemoral joint also provides information about any persistent dislocation

1540 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 22C1-2  Incidence of Osteochondral Fracture in Acute Patellar Dislocation Study

Mode of Evaluation

Osteochondral ­Fracture (%)

Ahmad et al (2000)76 Atkin et al (2000)1 Boring & O’Donoghue (1978)73 Buchner et al (2005)61 Cash & Hughston (1988)54 Cofield & Bryan (1977)52 Elias et al (2002)51 Harilainen & Sandelin (1993)74 Hawkins et al (1986)60 Jensen & Roosen (1985)68 Lance et al (1993)82 Maenpaa & Lehto (1995)75 McManus et al (1979)83 Nietosvarra et al (1994)84 Nikku et al (1997)63 Nikku et al (2005)62 Nomura et al (2003)29 Nomura et al (2005)78 Sallay et al (1996)16 Stanitski & Paletta (1998)28 Vainionpaa et al (1986)40 Vainionpaa et al (1990)15 Vironlainen et al (1993)85 Visuri & Maenpaa (2002)19

Open Radiography and MRI Open Radiography and MRI Arthroscopy Radiography MRI Open Radiography and arthroscopy Open MRI Open Open Arthroscopy Arthroscopy Arthroscopy Arthroscopy and open Open Open Arthroscopy Open Open Arthroscopy Open

0 19 0 29 28 15 15 43 52 22 73 12 11 39 22 21 72 60 68 58 14 11 46 16

MRI, magnetic resonance imaging. Adapted from Stefancin JJ, Parker RD: First-time traumatic patellar dislocation: A systematic review. Clin Orthop 455:93-101, 2007.

or subluxation. In one series, Atkin noted persistent subluxation, as indicated by lateral patellar overhang, in 97% of acute patellar dislocations.1 Using similar criteria, Vainionpaa showed without statistical significance that lateral displacement of the patella occurred in most acute dislocations.40 This position of patellar subluxation has been attributed to the loss of the medial retinacular integrity.41

In addition to patellar overhang, the sulcus angle can be determined on the axial view. In 1964, Brattstrom first quantified the difference in sulcus angle in lateral patellar instability, reporting nearly 10 degrees of difference when compared with normal subjects. Although widely used in clinical practice, the sulcus angle is only one of myriad of skeletal factors related to dislocation. There are a number of techniques described for taking the axial radiograph. Some involve standing, but most are taken in the supine position with knees flexed 20 to 45 degrees. Sulcus angles measured on cadavers using the methods of Merchant,42 Hughston,43 and Laurin44 have been proved statistically similar.45

  TABLE 22C1-3  Arthroscopic Appearance of Cartilage Lesions following Acute Patellar Dislocation Site Medial facet Medial facet and central dome Central dome Medial facet, central dome, and lateral facet Central dome and lateral facet Figure 22C1-6   The shear pattern that creates the typical osteochondral injury pattern to the lateral trochlea and the medial patella. (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

Cartilage Defect (%)

Cartilage ­Cracking (%)

64 25

10 27

7 4

50 3

0

10

Adapted from Nomura E, Inoue M, Kurimura M: Chondral and osteochondral injuries associated with acute patellar dislocation. Arthroscopy 19(7): 717-721, 2003.

Patella 1541

A

D

B

E

Ultrasound examination has recently been proposed for evaluation of acute patellar dislocation because of its low cost, safety, and availability. O’Reilly reported on 10 patients evaluated with ultrasound followed by surgical correlation. All 10 had complete correlation between sonographic and surgical findings in the medial retinacular complex. Bony avulsions from the patella and adductor tubercle were also well visualized, but one osteochondral fracture of the patella was missed with sonography.46 The inability of ultrasound to completely evaluate chondral injuries limits its usefulness in acute patellar dislocations. With increasing availability, lower cost, and better images, MRI has become the imaging modality of choice in the evaluation of acute patellar dislocations. Although

C

Figure 22C1-7   Methods of determining patellar height. Insall-Salvati ratio (A), the modified Insall-Salvati ratio (B), Blackburne-Peel ratio (C), the CatonDeschamps ratio (D), and the Blumensaat line (E). (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

not appreciable on plain radiographs, chondral, retinacular, and ligament injuries are well visualized on MRI (Table 22C1-4). In 1993, Quinn and associates were the first to describe the magnetic resonance appearance of the medial retinacular injury following an acute dislocation as part of “a triad of findings that included focal impaction injuries involving the lateral femoral condyle, osteochondral injuries of the medial patellar facet, and injuries of the medial retinacular ligament.”47 Although difficult to see on coronal and sagittal images, tears of the medial retinaculum and MPFL are visible on axial images as a loss of continuity of retinacular tissue with associated edema. The retinaculum and MPFL are seen at the same depth, but the MPFL is located more proximal,

1542 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 22C1-4  Summary of Magnetic Resonance Imaging Findings Following Acute Patellar Dislocation

LFC MP ­Contusion ­Contusion

OC Injury

Medial LP Retinacular MPFL Loose Body MCL Injury ­Subluxation Injury Injury

VMO Edema

Study

No. of Cases Effusion

Kirsch et al (1993)41 Virolainen et al (1993)85 Quinn et al (1993)47 Lance et al (1993)82 Spritzer et al (1997)50 Sanders et al (2006)30 Elias et al (2002)51

26

100

81

19

58

31

—

92

96

—

—

25

100

100

32

—

—

32

64

100

—

—

9

—

100

100

78

—

—

—

100

—

—

22

95

82

41

73

27

18

50

82

—

—

20

100

100

60

—

—

—

—

60

90

80

—

—

—

—

—

—

—

—

—

100

85

82

55

80

61

70

15

11

15

84

49

45

LFC, lateral femoral condyle; LP, lateral patellar; MCL, medial collateral ligament; MP, medial patellar; MPFL, medial patellofemoral ligament; OC, osteochondral; VMO, vastus medialis obliquus. Adapted from Elias DA, White LM, Fithian DC: Acute lateral patellar dislocation at MR imaging: Injury patterns of medial patellar soft-tissue restraints and osteochondral injuries of the inferomedial patella. Radiology 225(3):736-743, 2002.

adjacent to the distal aspect of the vastus medialis muscle.48 Some investigators have been careful to distinguish between these structures, but one should appreciate the continuum of injury on the medial side of the knee following dislocation. Several studies have correlated MRI findings of MPFL disruption with surgical exploration (Table 22C1-5). In 1996, Sallay evaluated 23 knees with MRI after ­documented patellar dislocation; 20 had avulsion of the MPFL from the adductor tubercle, one had a tear from the patella, and two had sprain without tear. Surgical exploration confirmed tears in all patients with MRI-proven tear, and two patients with sprain on MRI demonstrated partially healed avulsions from the adductor tubercle.16 Nomura classified MPFL tears into avulsion type and substantial type at open exploration and compared them to preoperative MRI findings. MPFL injury was accurately diagnosed on MRI in 92% of knees but was classified correctly only 81% of the time.49 In a similar surgical correlation study by Sanders, MRI was 85% sensitive and 70% accurate in detecting disruption of the MPFL.48 MRI is also useful in evaluating osteochondral injury, which encompasses a spectrum of injury from the subchondral edema of a bone contusion to frank osteochondral fracture. The incidence of osteochondral injury was previously presented, and discussion in this section is

  TABLE 22C1-5  Location of Medial Patellofemoral Ligament Injury

Adductor tubercle Midsubstance Patella Incomplete tear

Sallay et al (1996)16

Sanders et al (2001)48

Kirsch et al (1993)41

87% 0% 4% 9%

50% 0% 7% 43%

19% 42% 0% 38%

limited to the magnetic resonance appearance of these injuries. Contusion of the femur is commonly seen over the anterior aspect of the lateral femoral condyle and is usually greater in the anterior-posterior direction.50 This pattern is distinctly different from the bone contusion commonly seen in the distal weight-bearing portion of the condyle following anterior cruciate ligament injury (Fig. 22C1-8). The lateral lip of the overlying trochlear cartilage may be impacted or fractured; the location of this correlates with the degree of knee flexion at the time of injury and the patellar height. Bone contusion of the patella predictably occurs in the inferior-medial portion. In addition, a concave impaction fracture of the inferiormedial patella, analogous to a Hill-Sachs lesion in the shoulder, may occur. This MRI finding, first described by Elias, should be present on at least two consecutive image slices to avoid a misdiagnosis due to partial volume averaging.51 These MRI findings following acute patellar dislocation are useful to the clinician in tailoring a treatment plan because the presence of osteochondral injury, loose bodies, or MPFL injury may point the surgeon toward operative treatment.

NONOPERATIVE TREATMENT Nonsurgical treatment consists of immobilization followed by a period of structured rehabilitation. Immobilization allows for healing of the soft tissues, especially the supporting structures on the medial side of the knee. Traditionally, immobilization has been 3 to 6 weeks in a cylinder cast, but today, brace treatment with early mobilization has become the norm. Some have proposed rapid rehabilitation without immobilization to avoid harmful effects on tissue strength and articular cartilage integrity. Unfortunately, there are no randomized prospective studies on the type or duration of immobilization for acute patellar dislocation. Most early studies on nonoperative treatment are ­retrospective and observational and tend to focus on

Patella 1543

A

C

B

D

Figure 22C1-8   A comparison of the bone contusion pattern seen on magnetic resonance imaging following acute patellar dislocation and anterior cruciate ligament tear. In acute patellar dislocation, the contusion is seen on the outer border of the lateral femoral condyle (A), the inferomedial border of the patella (B), and the medial patellofemoral ligament is torn (C). In anterior cruciate ligament tear, the contusion is seen in the weight-bearing portions of the lateral femoral condyle and lateral tibial plateau (D). (Reprinted with the permission of The Cleveland Center for Medical Art & Photography © 2008. All Rights Reserved.)

r­ ecurrence rates rather than clinical outcome. In 1977, Cofield and Bryan evaluated 48 acute patellar dislocations initially treated nonoperatively with an average of 3.5 weeks of immobilization in a cylinder cast. Recurrent dislocation was reported in 44%, and 27% had surgical treatment within 5 years because of recurrent dislocations, subluxations, or persistent pain and effusion. Including those who underwent operation, an overall failure rate of 33% was reported, with 52% of athletes unable to return to sport.52 In another early series, Larson and Lauridsen followed 79 acute patellar dislocations for an average of nearly 6 years. They reported no clinical difference in treatment by a ­plaster cast or an elastic bandage, and a tendency to subluxation or dislocation in 53%.53 Cash and Hughston subsequently reported a series of 103 patients, divided retrospectively into two groups based on a presumed predisposition to dislocation derived from the presence of a congenital abnormality of the extensor mechanism, patellar hypermobility, or history of previous subluxation or dislocation in the opposite knee.54 ­Seventyfour knees were treated nonoperatively with a plaster splint for up to 6 weeks followed by physical therapy. At an ­ average of 8 years, the recurrence rate was 43% with 52% good to excellent results in the group predisposed to

­ islocation, whereas the recurrence rate was only 20% with d 75% good to excellent results in the group with a normal contralateral knee history and examination. No significant relationship was found between the length of immobilization and recurrence rate. It is important to note that 29 knees were treated with open or arthroscopic procedures, eliminating more severe injuries from the nonoperative treatment group. In 1997, Maenpaa and Lehto published the only longterm study on nonoperative treatment of acute patellar dislocation.55 One hundred patients were followed for an average of 13 years after treatment with a plaster cast, posterior splint, or bandage and splint for 2 to 4 weeks, and then underwent rehabilitation. The recurrence rate was 44% overall, yielding 0.17 redislocations per follow-up year (Table 22C1-6); an additional 19% without recurrence had continued symptoms of pain and instability necessitating operative treatment. The mean Kujala score at follow-up was 80, with significantly lower scores in those older than 30 years of age. Citing deleterious changes in ligaments and cartilage following immobilization,56,57 Garth and coworkers studied nonoperative treatment with immediate functional ­rehabilitation instead of traditional immobilization.58 Sixty-nine knees in 58 athletes were placed in a sleeve with

1544 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 22C1-6  Results of Nonoperative Treatment Following Acute Patellar Dislocation Plaster Cast (n = 60) Loss of extension of >5 degrees Loss of flexion of >5 degrees Retropatellar crepitation Positive apprehension test Redislocation frequency Late problems Kujala score

Posterior Splint (n = 17)

Bandage/Brace (n = 23)

15%

6%

13%

27%

6%

17%

67%

53%

52%

53%

53%

48%

38%

47%

57%

47% 80

21% 82

32% 74

Adapted from Maenpaa H, Lehto MU: Patellar dislocation: The long-term results of nonoperative management in 100 patients. Am J Sports Med 25(2): 213-217, 1997.

a ­ lateral buttress, and patients began supervised physical therapy immediately following evaluation of their injury. Therapies consisted of initial straight-leg raises followed by stationary bicycle for passive and active motion, and isotonic and isometric quadriceps strengthening. Return to full activities was allowed when tenderness subsided and isotonic quadriceps strength was symmetric, taking between 3 and 8 weeks. Two thirds of the knees had good to excellent results following a first-time acute dislocation, compared with only 50% of those with a recurrent dislocation. Overall, 73% were satisfied with their knees, but 16% were not and eventually elected to have surgical stabilization. Whether immobilized or not, those with acute patellar dislocation can expect a lengthy rehabilitation period before return to sport. Atkin and associates prospectively studied the recovery during the first 6 months following injury in 74 patients.1 Only 16% returned to sport by 6 weeks. At 6 months, 69% had returned to sport, despite more than half having continued difficulty with kneeling and squatting.

OPERATIVE TREATMENT If deemed necessary, operative treatment of acute patellar dislocation should be used to reestablish patellofemoral stability and treat any osteochondral injury that may

be present. In restoring stability, the approach should be to “repair, reconstruct, release, or realign.” After careful and complete evaluation of the injured knee, any combination of these principles may be needed to tailor a treatment plan to the individual. Repair and reconstruction should be directed at identifiable soft tissue injuries on the medial side of the knee. Release or lengthening of the lateral retinaculum should be aimed at restoring soft tissue balance of the patellofemoral joint, taking care not to destabilize the patella.59 Realignment procedures should be used to address clear underlying anatomic malalignment. Unfortunately, no one procedure or combination of these procedures has produced reproducible and uniformly excellent results. In a systematic review of the literature on primary patellar dislocations, Stefancin and Parker identified only five studies comparing nonoperative and operative treatment head to head, only two of which are randomized, prospective studies (Table 22C1-7).25 In a nonrandomized study, Hawkins evaluated 27 patients with either nonoperative treatment with a cylinder cast for an average of 3 weeks or operative treatment with arthrotomy, excision of osteochondral fragments, repair of the medial retinaculum, advancement of the vastus medialis, and lateral release followed by a cylinder cast for 5 to 6 weeks.60 Fifteen percent of patients treated nonoperatively sustained a redislocation in the follow-up period, whereas there were none in the operative group. More important, the study showed that 40% to 70% of patients can anticipate residual symptoms of anterior knee pain, and 20% to 30% will experience feelings of instability despite treatment method. Cash and Hughston similarly showed no dislocations in their operative treatment group, which underwent acute medial ligament repair without lateral release or realignment, compared with 36% of those treated nonoperatively.54 Recently, Buchner and colleagues retrospectively evaluated primary patellar dislocations in 126 patients after an average of 8 years.61 The four treatment groups consisted of nonoperative; arthroscopy only; open surgery with reconstruction of the medial retinaculum, realignment, and lateral release; and open surgery with fixation of a loose osteochondral fragment. The overall recurrence rate was 26%, with no significant difference between treatment groups. In addition, there was no statistical difference between treatment groups in activity level, function, pain, or subjective evaluation. Nikku and associates have published the only two prospective, randomized trials comparing nonoperative to

  TABLE 22C1-7  Results of Randomized Clinical Trials Comparing Operative and Nonoperative Treatment of Acute Patellar Dislocation

Knees Study (2005)61

Buchner et al Cash & Hughston (1988)54 Hawkins et al (1986)60 Nikku et al (2005)62 Nikku et al (1997)63

Good to Excellent Results

Redislocation Rate

Follow-up (yr)

Nonoperative

Operative

Nonoperative

Operative

Nonoperative

Operative

8.1 8.0 2.8 2.1 7.0

63 74 20 55 57

63 29 7 70 70

67% 58% 71% 81%

76% 82% 70% 67%

27% 36% 14% 27% 39%

25% 10% 0% 31% 17%

Adapted from Stefancin JJ, Parker RD: First-time traumatic patellar dislocation: A systematic review. Clin Orthop 455:93-101, 2007.

Patella 1545

operative treatment of primary patellar dislocation.62,63 Their operative group was treated with repair, duplication, or augmentation of the medial retinaculum with or without lateral release, and a subset of patients with subluxation only on examination under anesthesia were treated with isolated lateral release. Postoperative care was identical in both treatment groups. In their initial report of 125 patients with an average follow-up of 2 years, there was no significant difference in the patients’ subjective opinion, recurrent episodes of instability or subluxation, and redislocation between treatment groups. The only statistical difference was a slightly higher Hughston Visual Analog Scale (VAS) score in the nonoperative group.63 In 2005, Nikku re-evaluated 127 patients from the same study group at a mean follow-up of 7 years.62 No statistical significance was shown between treatment groups with respect to subjective Kujala, Hughston, and Tegner scores, or incidence of instability or redislocation. Risk factors for poor subjective outcome and recurrent instability were identified: lower Kujala scores were associated with female gender, loose bodies, and an initial history of instability in the contralateral knee, and a high recurrence rate was associated with young age and initial contralateral instability. Late operation was performed in nearly half of the study participants, signifying the frequency of long-term problems associated with acute patellar dislocation, despite the initial treatment. In general, operative indications for treatment of the acute dislocation of the patella in the adult are controversial. The most commonly cited indications for initial operative treatment are osteochondral loose bodies, palpable defects in the vastus medialis insertion, obvious tear in the medial retinaculum, and persistent asymmetric subluxation. With better imaging and more awareness of its importance, tear or avulsion of the MPFL should be added on this list.

Arthroscopy Arthroscopy of the knee following acute patellar dislocation has become more widely accepted. Arthroscopy is a minimally invasive procedure that allows direct visualization of the articular surfaces and can be done alone or in combination with open procedures. However, as MRI techniques have improved, the use of arthroscopy is becoming more therapeutic than diagnostic. Chondral and osteochondral injuries can be documented, excised, or fixed with the use of arthroscopic equipment. Recently, minimally invasive medial retinacular repair by suture anchor to the patella has been suggested by Fukushima, but the technique involved is technically difficult and does not address lesions at the adductor tubercle.64

Lateral Release Release of the lateral retinaculum has been used alone or in combination with other procedures in the treatment of acute patellar dislocation. Open lateral release was popularized by Merchant and Mercer in 1974.65 Arthroscopic lateral release was first described by Chen and Ramanathan in 1984 and has become a commonly performed, but controversial, adjunct to arthroscopy of the knee in acute

patellar dislocation.66 The concept of creating a “balanced laxity” in the patellofemoral joint is based on equalizing the soft tissue tension on both sides of the patella, that is, cutting the lateral side and allowing it to scar in a lengthened position balances out the stretched or torn medial retinaculum and patellofemoral ligament. Although this has been shown to decrease patellar tilt, it has not been shown to decrease lateral subluxation. Although it is often cited as treatment for recurrent lateral dislocations, there is little published on the use of lateral release in acute dislocations. At two-year follow-up, Dainer and colleagues showed no recurrent dislocations and 93% good to excellent results with arthroscopy alone, but a 27% recurrence rate and only 73% good to excellent results with concomitant lateral release.67 Jensen and Roosen concluded in a nonrandomized trial that lateral capsulotomy offered no advantage in preventing chondromalacia following acute patellar dislocation.68 More recently, Panni and coworkers showed that lateral release was less favorable in treating instability than pain. Threeyear results showed 72% satisfactory outcomes, dropping to 50% at 8-year follow-up with only 60% return to sports.69 There is considerable controversy in the concept of balancing the patella with a lateral release because it may contribute to lateral patellar dislocation instead of helping prevent it. Desio and associates showed in a biomechanical cadaveric model that the intact lateral retinaculum actually prevents lateral patellar displacement, contributing 10% of the restraining force. In a recent biomechanical study, Christoforakis and colleagues found that the force required to displace the patella 10 mm following lateral release was reduced by 16% to 19% from 0 to 20 degrees of knee flexion.70 Clinically, several studies have confirmed these findings, reporting recurrent lateral dislocations almost exclusively in groups treated with lateral release.15,67,68 In addition to recurrent lateral problems, iatrogenic medial subluxation and dislocation following lateral release have been observed by several authors.59,71 Hughston and Deese reported on 54 patients with worsening symptoms following arthroscopic lateral release and found that 50% demonstrated medial subluxation or dislocation that was not present preoperatively.59 Partially based on this issue, a survey of the International Patellofemoral Study Group was conducted to determine the current views regarding lateral release. Results were published in 2004, showing that only 7% of respondents would consider a lateral release in a first-time lateral patellar dislocation with a tight lateral retinaculum, and 37% would consider a history of lateral patellar dislocation as a contraindication to lateral release.72 In light of these findings, one should be aware of the potential complications and use lateral release with caution in cases of acute lateral patellar dislocation.

Medial Retinacular Repair Disruption or stretching of the medial retinaculum and MPFL always accompanies lateral patellar dislocation. Therefore, the mainstay of early surgical treatment in the acute first-time patellar dislocation is repair or reefing of the medial soft tissue structures, often accompanied by ­lateral release. In 1978, Boring reported on ­immediate

1546 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

s­ urgical repair of the acute patellar dislocation in 17 patients, of which 8 had repair or reefing of the medial retinaculum. In all of the surgical groups, there was no recurrent dislocation, but 12 knees were described as painful at follow-up. Fourteen patients were satisfied with their outcome, and there was no difference in outcome between medial retinacular reefing and medial transfer of the patellar tendon.73 Cash and Hughston also reported on a group treated with acute medial repair. Although the details of their repair are not described, they reported no recurrent dislocations and no subsequent operative reconstruction in this treatment group. Good to excellent subjective results were reported in 91% of those with contralateral congenital extensor mechanism abnormality and in 80% of those with a normal contralateral knee.54 Vainionpaa and colleagues prospectively reviewed 55 acute dislocations treated with medial retinacular repair, augmented with lateral release in 37 for tight lateral retinaculum with lateralization of the patella. Dislocation recurred in only 9%. Despite 80% good to excellent results, the incidence of discomfort, snapping, and giving way noticeably increased over the 2-year follow-up period. One patient had loss of motion requiring manipulation at 12 weeks.15 In 1993, Harilainen and Sandelin prospectively studied 53 patients treated with medial retinacular suturing or reefing and lateral capsular release. At an average of 6.5 years of follow-up, the redislocation rate was 17%, with all occurring in women.74 Maenpaa and Lehto performed medial reefing in 270 patellar dislocations, with concomitant release of the lateral retinaculum in 235. The overall subjective result was good to excellent in 70% with a recurrence rate of 17%. In accordance with other authors’ findings, the subjective results and recurrence rates were less favorable with a prior history of instability and a nontraumatic mechanism.75 In 2005, Buchner retrospectively studied operative treatment with medial retinacular repair combined with a lateral release, with an average 8 years of follow-up. The redislocation rate was 27% and statistically similar to nonoperative and arthroscopy groups. At follow-up, the Tegner activity score was 4.5, the Lysholm score was 85, and subjective success was realized in 80%. All were statistically similar to nonoperative and arthroscopy groups.61 Also in 2005, Nikku reported on an operative group of 63 patients treated with repair of medial retinaculum or augmentation of the MPFL, with a 7-year follow-up. Fifty-four had a concomitant lateral release. Although 67% had episodes of postoperative instability, recurrent dislocation occurred in only 31%, demonstrating no statistical difference from the nonoperative group. Good to excellent results were found in 67%.62

Medial Patellofemoral Ligament Repair and Augmentation Repair or reefing of the medial retinaculum often does not completely address the medial-sided pathology after acute patellar dislocation. The importance of the MPFL for stability of the patella has been proved in cadaveric models, and several studies have shown a high incidence of avulsion of the MPFL from the adductor tubercle. Repair of retinacular injury does not restore continuity of the medial

restraints in cases of MPFL avulsion from the adductor tubercle. Continuity can be restored by direct repair, augmented repair, or reconstruction with graft at the site of injury. Reconstruction is usually reserved for cases of recurrent dislocation and will be discussed in depth in the following section. In 1996, Sallay and associates were the first to report on direct repair of the MPFL. Twelve patients were evaluated at 2-year follow-up; there were no redislocations, although four had episodes of sharp pain that may have represented subluxation. Good to excellent results were found in only 58% with an average Lysholm score of 81. After 4 weeks of immobilization, two patients suffered loss of motion requiring manipulation under anesthesia.16 Ahmad and coworkers also reported on direct repair of the MPFL. A limited series of eight acute patella dislocations were treated with direct repair of the MPFL to the adductor tubercle along with lateral release and repair of the vastus medialis obliquus (VMO) to the adductor magnus tendon. At 3 years’ follow-up, the average Kujala score was 92, and patient satisfaction was 96%. Despite these impressive results, athletes were able to return to only 86% of their preinjury level.76 Although several augmentation procedures were previously described, Gallie and Lemesurier were the first to report a “reconstruction” of the medial structures with graft in 1924. They described a procedure “anchoring” the patella in place with a strip of fascia to an isometric point on the medial femoral condyle. The procedure was performed on seven patients with no recurrence and “perfect functional results.”3 Since then, several authors have reported on augmentation procedures with less than perfect results. In 1993, Avikainen and associates performed augmentation of the MPFL with the distal 8 cm of adductor magnus tendon in 14 patients, 10 with acute dislocation. There was one recurrence at 2.5 years. There were 86% good results, but no excellent results.77 Recently, Nomura published the 5-year results of an augmented repair of the MPFL with a slip of the medial retinaculum. Despite recurrent subluxation in four of five knees, excellent results were reported in three of five, with an average Kujala score of 97.78 Although there have been favorable results with both MPFL repair and augmentation of the MPFL, there are no randomized, prospective studies comparing them to each other or to nonoperative treatment. Further investigation is needed before these procedures can be recommended in the case of acute patellar dislocation.

REHABILITATION A structured rehabilitation program is essential following acute patellar dislocation, despite treatment method. The program should be supervised by a certified physical therapist or athletic trainer familiar with these injuries. Initial goals should be to advance weight-bearing and regain both active and passive range of motion. Subsequently, closed kinetic chain strength training and proprioceptive exercises should begin, followed by functional and sportspecific training. The ultimate goal should be a stable, functional, and asymptomatic knee with strength equal to the uninjured side.

Patella 1547

Return to play should be allowed only when the following criteria have been met: Subjectively, there should be no pain, swelling, or sensation of instability. Objectively, there should be no effusion, no tenderness or apprehension, and a full range of motion without pain. Quadriceps strength should be at least 80% of the contralateral side, as determined by functional testing. Once running and cutting can be performed without symptoms, the athlete is allowed unrestricted return to competition. The benefit of patellar bracing after dislocation is unclear. Effective bracing should augment the static and dynamic stabilizers of the patella.79 Dynamic MRI studies have shown improvement in the position of the laterally subluxated patella with bracing; whether this clinically extrapolates to a rapidly moving athlete is unclear.80,81 Unfortunately, no randomized, prospective studies evaluating patellar bracing for instability exist.

Authors’ Preferred Method Treatment of acute patellar instability requires a thorough history, physical examination, radiographic and MRI ­examination, and differentiation between instability and disability. Usually a large osteochondral fracture and instability are the predominant issues, and care must be taken to educate the patient regarding expectations and outcomes. An individualized treatment plan should be developed for each patient. In other words, care must be taken not to treat each patient the same. Rehabilitation is the usual treatment for the acute patellar dislocation. Immobilization in extension is usually the initial treatment for the first 10 to 14 days. Rehabilitation as outlined previously is then initiated. Rehabilitation is chosen both to strengthen and increase endurance in the operative candidate unless a large osteochondral fracture is present, in which case operative treatment is initiated. The operative patient undergoes an examination under anesthesia and diagnostic arthroscopy as part of the surgical procedure. It is imperative to rule out other forms of instability, such as anterior cruciate instability. In addition, chondral or osteochondral loose bodies or changes on the patella or trochlear groove are important considerations in the treatment algorithm. If an associated meniscal injury is encountered, it should be recognized and treated as well. Patellofemoral tracking can be assessed as well. Based on the previous discussion, decisions are made regarding proximal and distal realignment.

C

r i t i c a l

P

o i n t s

l

 wo distinct groups of patella dislocations exist: those T with normal anatomy and a traumatic event, and those with predisposing anatomy and a history of subluxation without a traumatic event. l The medial patellofemoral ligament is the main restraint to lateral patellar subluxation. The medial retinaculum, medial patellotibial ligament, and lateral retinaculum are secondary restraints. l The medial patellofemoral ligament is nearly universally disrupted in lateral patellar dislocation. l The mechanism of dislocation can be a direct blow or an indirect rotation of the body on a valgus, flexed knee relative to a fixed foot. The indirect mechanism is more ­common. l Clinical examination should include palpation for defect in the medial retinaculum and evaluation of apprehension of the patella to a lateral force. l Imaging should include plain radiographs and MRI to evaluate for evidence of chondral and medial retinacular injury. l Head-to-head studies show no difference in outcome when comparing conservative with early surgical treatment in acute lateral patellar dislocations. l Surgical treatments include arthroscopic débridement, medial retinacular repair, medial patellofemoral ligament repair, and augmentation. Care should be taken to evaluate the location of the medial lesion because retinacular repair will not address avulsion of the medial patellofemoral ligament from the adductor tubercle. Lateral release should be done only in select cases. l Structured rehabilitation is essential for optimal recovery despite treatment method.

S U G G E S T E D

R E A D I N G S

Amis AA, Firer P, Mountney J, et al: Anatomy and biomechanics of the medial patellofemoral ligament. Knee 10(3):215-220, 2003. Fithian DC, Paxton EW, Post WR, Panni AS: Lateral retinacular release: A survey of the International Patellofemoral Study Group. Arthroscopy 20(5):463-468, 2004. Fithian DC, Paxton EW, Stone ML, et al: Epidemiology and natural history of acute patellar dislocation. Am J Sports Med 32(5):1114-1121, 2004. Insall J, Salvati E: Patella position in the normal knee joint. Radiology 101(1):101104, 1971. Stanitski CL: Articular hypermobility and chondral injury in patients with acute patellar dislocation. Am J Sports Med 23(2):146-150, 1995. Stefancin JJ, Parker RD: First-time traumatic patellar dislocation: A systematic review. Clin Orthop 455:93-101, 2007.

R eferences Please see www.expertconsult.com

1548 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

C

Subluxation and Dislocation 2. Patellofemoral Instability: Recurrent Dislocation of the Patella Timothy Steiner and Richard D. Parker

Patellofemoral instability is a complex topic. Significant advances have occurred in our knowledge about the anatomy and function of the patellofemoral joint, which has resulted in a better understanding of the evaluation and treatment of patellar instability.

ANATOMY Stability of the patella depends on the complex balance of forces imposed by static restraints and dynamic stabilizers around the knee. The static restraints are twofold: the bony architecture of the patellofemoral joint and the surrounding soft tissue structures, primarily those of the medial retinaculum and medial patellofemoral ligament (Fig. 22C2-1). The dynamic restraints consist of muscular forces that act on the patella, especially the quadriceps (Fig. 22C2-2). The importance of each structure has been well studied, but only recently has the relative contribution of each been defined. Senavongse and Amis reported the biomechanical contributions of the articular, retinacular, and muscular stabilizers of the patella.1 When compared with an intact knee,

Figure 22C2-1   Static restraints of the knee. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

flattening of the lateral facet of the femoral trochlea had the greatest effect on patellar stability, reducing the force necessary for displacement by 70% at 20 degrees of knee flexion. Sectioning of the medial retinacular structures, including the medial patellofemoral ligament, reduced the displacement force by 49% in full extension, with lessening effects during knee flexion. Relaxing the vastus medialis had a smaller but more consistent effect on stability, reducing the force by 30% throughout the midrange of

Figure 22C2-2   Dynamic restraints of the knee. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Patella 1549

A

B

C

Figure 22C2-3   Anteversion of the femur. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

knee motion. The patella was found to be most unstable at 20 degrees of flexion, both in the intact knee and in all test groups. These quantitative findings help define the relative contributions of each factor, which warrant individual discussion. Although one may tend to focus on the knee and its surrounding structures, the entire lower extremity, from hip to foot, should be considered in the patient with recurrent patellar dislocation. For ease of discussion, lower extremity anatomy in this chapter is separated into its bony and soft tissue elements.

Bony Anatomy The bony anatomy of the femur, tibia, and patella contribute to the stability of the patella. Abnormalities of the femur include increased anteversion of the femoral neck, torsion of the femoral shaft, hypoplasia of the lateral femoral condyle, and dysplasia of the femoral trochlea. Tibial abnormalities include increased external tibial torsion and variable location of the tibial tuberosity. Morphology of the patella also plays an important role in its stability. Before discussing the individual bones, the overall alignment of the leg should be addressed. Abnormalities in alignment, genu valgum and genu varum, are dictated by the combined morphology of the femur and tibia. These abnormalities cause the quadriceps pull to be out of alignment with the trochlear groove, leading to a sidedirected vector on the patella. In genu valgum, this vector is directed laterally, which can lead to lateral dislocation if medial restraining forces are overcome.2 The anatomy of the entire femur is important in patellofemoral mechanics. At the proximal end, the version of the femur is a measure of the anterior or posterior projection of the femoral neck and head. In the transverse plane, this is defined as the angle between a line through the long axis of the femoral neck and a line through the center of the femoral condyles (Fig. 22C2-3). Normal femoral anteversion varies, with anatomic, biplane radiographic and computed tomography (CT) data ranging from 6 to 48 degrees, with most between 7 and 20 degrees.3 An increase in femoral anteversion causes the distal femur and

Figure 22C2-4   Torsion of femur and/or tibia. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

­ atellofemoral articulation to face medial when the femop ral head lies neutral in the acetabulum. A decrease in version, or retroversion, causes the opposite effect. Torsional deformity can also be present anywhere along the length of the femur, leading to a similar effect on the alignment of the patellofemoral joint or an amplification of abnormalities caused by increased anteversion of the femoral neck. Abnormalities in femoral rotation are often seen in combination with external tibial torsion, causing the patellas to point medially toward each other when the feet are facing forward (Fig. 22C2-4). These torsional abnormalities, alone or in combination, tend to displace the patella laterally by a side-directed force similar to that seen in genu valgum. Although distinctly different, this may appear as an increased quadriceps angle when viewed in the coronal plane only. The bony anatomy of the distal femur has a very different but equally significant role in stability of the patella (Fig. 22C2-5). Recognized by Albee nearly a century ago, the medial and lateral portions of the femoral trochlea act as a buttresses to displacement of the patella.4 Any ­reduction in the size of this buttress confers lessening resistance

1550 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 22C2-6   Articular surface of patella. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Figure 22C2-5   Trochlear depth and patellar shape contribute to stability. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

toward subluxation or dislocation of the patella. Seen alone or in combination, this reduction in size may present in different ways. Trochlear dysplasia presents proximally as a shallow trochlear floor relative to the medial and lateral condyles, whereas a hypoplastic lateral condyle presents as an isolated deficiency in the size of the lateral condyle, as depicted by the height of the lateral side seen on an axial radiograph. At 45 degrees of flexion, the height of the lateral condyle should be 1 cm more anterior than the medial condyle. One must be careful not to be fooled by “illusory dysplasia,” or the appearance of a lateral condylar deficiency secondary to excessive rotational deformity of the femur.5 Multiple studies have demonstrated that the bony stability of the femoral trochlea is the most important stabilizer of the patella,1,6,7 and sufficient trochlear restraint may compensate for other anatomic deficiencies. The patella is a sesamoid bone of irregular shape that lies within the extensor mechanism of the knee. Its anatomy is well described, with a convex anterior surface and a complex, cartilage-covered posterior articular surface. The inferior pole is nonarticular, and the remaining superior pole is divided into medial and lateral facets divided by a central ridge. An additional odd facet at the most medial edge of the patella may be found. The lateral facet is longer and more sloped, and the subchondral bone density is highest at its proximal facet.8 These anatomic differences are likely adaptations derived to match the larger forces imparted by the larger, wider, and more prominent lateral femoral condyle. Wiberg devised a classification system in the axial plane, and Grelsamer devised one in the sagittal plane.9,10 A correlation exists between Wiberg type and lateral patellofemoral ligament width, suggesting a

­ evelopmental manifestation of muscular forces around d the patella.11 The contact area of the patella changes throughout the range of knee flexion (Fig. 22C2-6). As knee flexion increases, the contact area moves from the distal to the proximal pole of the patella, and the odd facet contacts only in deep flexion, where the lateral and odd facets separately contact the femoral condyles.12 Unfortunately, the complex shape of the patella and matching contours of the femoral trochlea are not seen on plain radiographs because the varying thickness of articular cartilage is radiolucent. The bony anatomy of the tibia is also important in the stability of the patella. As previously discussed, external tibial torsion often accompanies abnormalities in femoral anteversion. The tibial tuberosity is the distal attachment of the extensor mechanism, and its location on the tibia affects the alignment of the patellar tendon. The combination of external tibial torsion and lateral placement of the tibial tuberosity can cause a deleterious change in the quadriceps angle. One must not forget to include the bony anatomy of the foot. Although distant from the knee, deformities in the foot cause changes in the alignment of the lower extremity. Pronation of the foot is easily seen on physical examination as valgus in the heel. Excessive valgus at the subtalar joint causes obligatory internal rotation of the tibia, whereas varus causes external rotation.

Soft Tissue Anatomy The soft tissue elements that contribute to patellar stability include muscles, tendons, and ligaments that act as static restraints and dynamic stabilizers of the patella. The quadriceps muscle is intimately involved because of its direct attachment to the patella by way of the quadriceps tendon. The ligamentous structures are particularly important because of their role as checkreins in the static stability of the patella, on both the medial and lateral sides. These ligamentous structures arise within the retinacular tissue and include patellofemoral and patellotibial ligaments. The quadriceps muscle is the most important dynamic stabilizer of the patella. It is named for its four portions: rectus femoris, vastus lateralis, vastus intermedius, and vastus medialis (Fig. 22C2-7). Only the rectus femoris crosses both the hip and knee joints, with its origin at the ­anterior inferior iliac spine of the pelvis. The other ­muscles ­originate

Patella 1551

on the proximal shaft of the femur. All four muscles insert in a layered arrangement into the proximal half of the patella through the quadriceps tendon, with the rectus femoris inserting most anterior and the vastus intermedius most posterior. All but the vastus intermedius continue as a tendinous expansion over the top of the patella to become part of the patellar tendon.13,14 The vastus medialis and lateralis also provide connection to the tibia through the attachment of their investing fascia to the medial and lateral patellar retinacula, respectively. Each portion of the quadriceps muscle imparts a different force vector based on the angle of its tendinous insertion. In 1968, Lieb and Perry measured the obliquity of the quadriceps muscles with respect to the long axis of the femur in the frontal plane13: the vastus lateralis was 12 to 15 degrees lateral, and the rectus femoris was 7 to 10 degrees lateral. They described the vastus medialis as consisting of two distinct parts, which they called the vastus medialis longus and vastus medialis oblique (VMO), with a marked difference in the angle of their fibers. The angle of fibers in the oblique portion measured 50 to 55 degrees medially, but only 15 to 18 degrees medially in the longus portion. They also reported distinctly different anatomy, with oblique fibers originating off the intermuscular septum and adductor tubercle and inserting on the proximal third of the medial patellar border. Farahmand and colleagues reported similar angles of 15 and 47 degrees for the long and oblique portions, respectively,7 and Andrikoula and associates recently reported 45 degrees from the rectus for the vastus medialis as a whole.15 Despite these numbers, some estimate the oblique portion to have an angle as high as 55 to 65 degrees.16 Special attention has been paid to the oblique portion of the vastus medialis and also of the vastus lateralis. ­Hallisey

Figure 22C2-7   Quadriceps musculature. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

and coworkers described the vastus lateralis oblique with distinct anatomy from the long portion. The oblique originates off the lateral intermuscular septum with three distinct patterns of insertion. The mean angle of insertion was statistically different by sex at 48 degrees in men and 38 degrees in women.17 Andrikoula recently reported 26 degrees with respect to the rectus femoris, whose own vector lies lateral to the femur.15 These angles are significantly higher than the 12 to 15 degrees for the entire vastus lateralis previously reported by Lieb and Perry.13 The high angle of insertion of the medial and lateral oblique muscle bellies results in a large portion of the muscular force directed perpendicular to the long axis of the thigh. Although not proved, these perpendicular forces may serve as dynamic stabilizers of the patella in addition to their role as knee extensors. Recent work by Panagiotopoulos and colleagues has shown that the distal attachment of the VMO may also dynamically tension the medial patellofemoral ligament, providing additional medial support to the patella, confirming previous work by Sallay and ­associates.18,19 Although the quadriceps muscles provide extension to the knee and dynamic balance to the patella, the remaining soft tissue structures provide static restraint. Both the medial and lateral retinacula contain intrinsic fibers and thickenings that have been described as distinct ligamentous structures named the patellofemoral, patellotibial, and patellomeniscal ligaments (Fig. 22C2-8). Additionally, a thickening on the lateral side runs transversely from the iliotibial band to the patella and is named the iliopatellar band (Fig. 22C2-9). Kaplan first described the epicondylopatellar ligaments, which are now called patellofemoral ligaments.20 Warren and

Figure 22C2-8   Medial patellofemoral ligaments. MPFL, medial patellofemoral ligament; MPML, medial patellomeniscal ligament; MPTL, medial patellotibial ligament. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

1552 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 22C2-9   Lateral thickening of iliotibial (IT) band to the patella called the iliopatellar band. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Marshall subsequently described the anatomy of the medial side of the knee in a three-layered pattern; the medial patellofemoral ligament (MPFL) resides in the second layer, deep to the vastus medialis.21 Although Reider and colleagues found a distinct ligament in only 35% of knees, most authors report a nearly universal existence of the structure.11 The MPFL originates near the adductor tubercle, with additional fibers arising from the superficial portion of the medial collateral ligament, and it inserts on the proximal two thirds of the medial border of the patella, with additional fibers inserting on the quadriceps tendon. The MPFL is 5 cm in length and averages 1.9 cm in width, becoming wider at its insertion than its origin.15,22 The tensile strength of the ligament has been found to be 208 Newtons.23 The anatomy of the patellotibial and patellomeniscal ligaments is less well studied. On the medial side, they reside deep to the patellofemoral ligaments in the third layer.21 The medial patellotibial ligament (MPTL) originates at the inferomedial border of the patella and inserts on the anteromedial tibia. Its insertion is 1.5 cm below the joint line and 1.5 cm medial to the patellar tendon, lying 20 to 25 degrees oblique to the patellar tendon.18 The medial patellomeniscal ligament (MPML) originates near its patellotibial counterpart on the inferomedial border of the patella. It inserts broadly on the anterior horn of the medial meniscus, with fibers fanning out posteriorly.15 Both the MPML and MPTL are relatively thin and only 3 to 6 cm. No studies have quantified their tensile strength.18 The lateral retinaculum is composed of only two layers. The superficial oblique retinaculum spans the iliotibial band to the patella, blends with the quadriceps expansion, and becomes thinner in its distal portion. The deep transverse retinaculum is thicker but ends at the inferior border of the patella. It consists of the lateral patellofemoral, patellotibial, and iliopatellar ligaments. In distinction from the medial side, no patellomeniscal ligament exists; rather, fibers of the patellotibial ligament insert on both the proximal tibia near Gerdy’s tubercle and on the anterior horn of the lateral meniscus.14,24 The size and existence of the lateral patellofemoral ligament (LPFL) is variable. Reider and associates identified the ligament in only 13 of 48 specimens but noted a correlation between ligament width and Wiberg shape of the patella.11 Like the MPFL, recent studies have shown nearly universal existence of the LPFL.25 Although no lateral sectioning studies exist, several authors have ­ recognized

medial patellar instability following a lateral release, which severs the LPFL.26-28 Two studies have examined the tension of the LPFL during knee motion. Luo showed a significant increase in LPFL tension at 30 degrees of flexion during passive knee motion in vitro, whereas Ishibashi showed maximal tension at 120 degrees of flexion in vivo in patients with lateral patellar instability.29,30 Most patellar dislocations are lateral, and biomechanical studies have concentrated on the importance of the medial ligaments in patellar stability (Fig. 22C2-10). Conlan published the first biomechanical study that focused on the medial supporting structures of the knee. Using sequential sectioning and a laterally directed load with the knee in full extension, he demonstrated that the MPFL contributed 53% of the restraining force to lateral displacement; the

Figure 22C2-10   Most dislocations of the patella are lateral because of the sum of the vectors directing the patella laterally. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Patella 1553

  TABLE 22C2-1  Results of the Sectioning Studies Percentage of Patellar Stability Contribution Determined by Study Structure Contributing to Static Patellar Stability

Conlan et al (1993)187

Desio et al (1998)159

Hautmaa et al (1998)9188

Panagiotopoulos et al (2006)18

Medial patellofemoral ligament Medial patellotibial ligament Medial patellomeniscal ligament Medial retinaculum Lateral retinaculum

53

60

50

50

5

3

25

13

22

13

25

24

11 —

3 10

5 —

13 —

MPML was a secondary restraint, contributing 22%. In a similar work, Desio later confirmed these findings at 20 degrees of knee flexion, with the MPFL and the MPML contributing 60% and 13%, respectively. Interestingly, the lateral retinaculum was also found to contribute 10% of the restraining force to lateral displacement. Conversely, Hautmaa measured displacement of the patella under constant load as the medial structures were sectioned in two different sequences. Sectioning of the MPFL caused a 50% increase in displacement in both cases, whereas sectioning the patellotibial and patellomeniscal ligaments together increased displacement by 25%. Recent studies have confirmed the results of these early works. Panagiotopoulos measured displacement under a constant load in a sequential sectioning study to determine the relative contribution of each structure to static patellar stability. The MPFL contributed 50%, the MPML contributed 24%, and the MPTL and medial retinaculum each contributed 13%.18 The results from all the sectioning studies mentioned are summarized and referenced in Table 22C2-1.

CLINICAL PRESENTATION The patient with recurrent patellar dislocation may or may not have had a classic episode of acute patellar dislocation. Often, the patient complains that the knee “gives way” and commonly reports weakness of the leg. Occasionally, the

  TABLE 22C2-2  Most Common Symptoms of Patellofemoral Subluxation Symptom

Patients Reporting (%)

Pain going down stairs Pain on flexion Weakness Giving way Swelling Pain going up stairs Locking

76 75 73 61 60 54 50

From Henry JH: Conservative treatment of patellofemoral subluxation. Clin Sports Med 8(2):261-278, 1989.

patient reports feeling the kneecap “slide out of place” or “pop” into or out of place. The femur is often in a position of internal rotation on a fixed tibia, regardless of whether episodes occur during daily activities or sporting events. Many patients with recurrent instability complain of pain. In a study of 465 patients with the diagnosis of patellofemoral subluxation, pain represented the top two most commonly reported symptoms (Table 22C2-2). A careful history is essential to separate pain related to instability from other patellofemoral problems.

PHYSICAL EXAMINATION A careful and complete physical examination is necessary for accurate diagnosis and proper clinical decision making. Examination should not be limited to the affected knee but should include examination of the entire lower extremity of both sides. Information should be gathered by examining the patient in the upright, sitting, and supine positions while barefoot and dressed in shorts. The physical examination should begin with observation. Obesity, posture, and body habitus should be noted. Skin should be examined on the extremities; prior surgical and traumatic scars should be noted, as should any evidence of vasomotor abnormality. Any muscular asymmetry of the thigh or calf should be confirmed with circumferential measurements at a standard distance above and below the knee. The size and location of the vastus medialis muscle belly should be noted because a lower insertion is considered more efficient in resisting lateral subluxation.31

Quadriceps Angle Because the quadriceps angle, or Q angle, can be measured in the standing, sitting, or supine positions, it is discussed first. The Q angle is a commonly used measurement in the evaluation and treatment of patellofemoral disorders (Fig. 22C2-11). Because it is easily measured, it is often used in clinical practice. However, it is important to stress that the Q angle should not be used as the sole factor in surgical decision making in recurrent patellar subluxation.32 The Q angle was first described in 1847 by Cruveilhier as the angle between the vector of the quadriceps tendon and the vector of the patellar tendon.33 Clinically, these

1554 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the Q angle at varying degrees of knee flexion and found decreasing values as flexion increased,37 a finding similar to that of Sojbjerg and coworkers.38 In general, the Q angle decreases from supine to standing to sitting examination. It is widely accepted that females have higher Q angles than males, an observation that has been attributed to the wider pelvis, shorter femur, and increased genu valgum seen in females.39,40 However, Horton demonstrated that males actually have a longer distance between greater trochanters and that there is no relationship between hip width and Q angle. Further analysis showed that although the ratio of femur length to hip width was greater in females, it did not correlate to Q angle. When measuring the Q angle with the leg extended, the value may be erroneously low if the patella lies superior and lateral to the trochlear groove. This should be considered during the examination, especially in those with patella alta, a J sign, or obvious lateral displacement of the patella. On the contrary, the Q angle may be falsely elevated by excessive femoral anteversion, femoral torsion, external tibial torsion, and genu valgum.

Standing Examination

Figure 22C2-11   Standing Q angle. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

vectors are simplified as lines from the anterior superior iliac spine to the midpatella and from the midpatella to the tibial tubercle. The logic behind the Q angle is best summarized with a two-dimensional free body diagram of the patella. The vertically oriented resultant vectors balance each other out, whereas the horizontally oriented resultant vectors add up to a large laterally directed vector (see Fig. 22C2-10). The theoretical problem then becomes clear: a larger Q angle results in a larger laterally directed force that must be resisted by the bony and soft tissue restraints previously discussed. When the restraints are overcome, dislocation or instability may result. Although a larger Q angle theoretically causes more lateral force, the clinical significance is unclear. Aglietti and associates demonstrated no significant difference in the Q angle of patients with patellar subluxation compared with healthy controls,34 and Fairbank and coworkers showed no difference between adolescents with and without knee pain.35 Because the Q angle can be measured in the standing, supine, or sitting position, a complete assessment should include all three. Insall’s original description of 50 normal knees with an average Q angle of 14 degrees did not specify the position of measurement. A review of the literature demonstrates a range from 8 to 16 degrees for males in the supine position and from 15 to 19 degrees in females. Woodland showed that the standing Q angle was statistically higher than supine for both male and female subjects, although only by roughly 1 degree.36 Q angles in the standing position range from 11 to 20 degrees in males and 15 to 23 degrees in females. Johnson and colleagues examined

Standing alignment should be viewed from the front, back, and side. In the sagittal plane, recurvatum may indicate systemic hypermobility, and lack of full extension may indicate fat pad impingement or flexion contracture. Coronal plane observations may include leg-length discrepancy, patella alta or baja, and genu varum or valgum. A sharp varus angulation in the proximal tibia may be present and is termed the bayonet sign. Excessive femoral anteversion and internal tibial torsion may give the appearance of “squinting,” or inward pointing, of the patella. The patella should be observed with the feet in a comfortable position and with the feet facing forward; those with excessive external tibial torsion may show squinting only in the latter position. If suspected, abnormalities in femoral anteversion should be measured by observing maximal prone internal and external hip rotation as well as rotation of the leg at the position of maximal prominence of the greater trochanter.41 Similarly, transmalleolar axis and thigh-foot angle should be used to confirm excessive tibial torsion, although the latter does not set apart abnormalities of the foot. These rotational abnormalities may be part of a larger constellation of findings called miserable malalignment, in which squinting of the patella from excessive femoral anteversion and outward tibial rotation are accompanied by tibia vara, patella alta, and an increased quadriceps angle.42 Finally, the feet should be examined in the weight­bearing position. From behind, hindfoot alignment can be observed; valgus may indicate foot pronation, a compensatory position that allows a plantigrade foot in subjects with genu varum or tibia vara. Foot pronation causes obligate internal tibial rotation, with little change noted at the knee joint.43,44 Although controversial in the pathogenesis of patellofemoral pain, this combination of findings may alter patellofemoral mechanics.45,46 The conclusion of the standing examination should include an observation of the patient’s gait. Observation of the whole body should note antalgia or limping. A quadriceps avoidance gait with reduced knee flexion in stance

Patella 1555

A

B

Figure 22C2-12   A, Normal patella position on physical examination. B, Patella alta; note the patella tilting upward. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

may be present.47 A Trendelenburg gait with a drop in the contralateral pelvis during stance phase indicates gluteus medius weakness; this change in pelvic obliquity tightens the ipsilateral iliotibial band, a common source of knee pain. Then, attention should focus on the patella during gait. Catching, jumping, or any other abnormal movements should be noted as the patella enters and leaves the trochlea near full extension.

Sitting Examination The patient should next be examined in the sitting position, with the knees flexed 90 degrees over the table. The position of the patella and the tibial tubercle should be observed. Patella alta or baja is easily observed from the side (Fig. 22C2-12). Normally, the patella should sit at the distal end of the femur with the proximal pole in line with the anterior portion of the femur. In patella alta, the patella will lie more proximal with its anterior surface tipped up. If there is also patellar tilt, the knees will have the so-called grasshopper-eyes appearance. The position of the tibial tubercle should be observed in relation to the center of the patella. In flexion, the patella is centered in the trochlea, and the tibia derotates (Fig. 22C2-13). Any deviation of the tubercle laterally from the center of the patella is called the tubercle sulcus angle (Fig. 22C2-14). In normal subjects, this angle averages 4 degrees,37 and an abnormal angle has been postulated to be greater than 10 degrees.48 Because the patella is centered in the trochlea at 90 degrees of knee flexion, the tubercle sulcus angle may correlate with the radiographic measurement of tibial tubercle–to-trochlear groove (TT-TG) distance.

Figure 22C2-13   The patellar tendon and tibial tuberosity should line up with the knee flexed 90 degrees. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

1556 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Next, the musculature surrounding the knee should be examined. The VMO belly is important in dynamic stabilization of the patella. Lieb and Perry concluded that the only function of this muscle was for alignment of the patella.13 Electromyographic studies subsequently showed that the VMO was most active in the last 30 degrees of extension, a time in which the patella is unconstrained by the trochlea.49,50 The size and location of insertion of the VMO are of importance. A well-developed and symmetrical VMO is ideal, with muscle bulk arising near the adductor tubercle and inserting on the proximal third of the patella. A more distal insertion on the patella provides even more resistance to lateral movement of the patella.31 Finally, tracking of the patella should be observed in the sitting position. The term J sign refers to an abnormal tracking pattern in which the patella sits lateral to the femoral sulcus in full extension; the movement of the patella appears in the shape of an upside-down J as the knee goes from flexion into full extension. Conversely, the patella starts laterally and makes an abrupt shift medially as it enters the femoral trochlea at the initiation of ­flexion. Although tracking can be seen during a weightbearing squat, tracking during both active and passive motion can be observed in the sitting position. An exact cause of this pathologic motion has not been defined, but Post has postulated that VMO deficiency may be to blame if the J sign is observed during active motion, and underlying bone morphology and soft tissue imbalance may be the cause during passive motion.51 The presence of a J sign during active or passive motion is clearly abnormal, and the J sign did not appear in a study of 210 asymptomatic men and women.37

Figure 22C2-14   Tubercle sulcus angle. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Supine Examination The patient should be asked to lie comfortably on the examination table. The knee is first observed for evidence of effusion. A large effusion is often visible as a swelling in the suprapatellar pouch and an obliteration of the normal dimples adjacent to the patellar tendon at the joint line. Unless the patella has recently dislocated, a large effusion is usually absent. It should be noted that a large effusion may accompany an acute patellar dislocation, but it should not be tense because of the necessary disruption of the retinaculum. A ballotable patella without obvious visual clues may indicate a moderate effusion. If not readily observed, the joint should be milked for evidence of a small effusion. If present, even a moderate effusion can cause reflex inhibition of the quadriceps muscle. Inhibited by only 20 mL of intra-articular saline, the vastus medialis has been shown to be even more sensitive to effusion than the rectus femoris or the vastus lateralis.52 Care should be taken to differentiate an intra-articular effusion from extra-articular swelling, which presents as immobile, thickened soft tissue. Any abnormalities in skin temperature and color around the knee should also be assessed as possible indicators of diffuse synovitis or infection. Structures around the knee should be systematically palpated (Fig. 22C2-15). First, the peripatellar structures should be evaluated. Tenderness at the proximal and distal poles of the patella may indicate insertional tendinitis, and tenderness within the quadriceps or patellar tendon may indicate more diffuse tendinitis. Pain at the medial border of the patella may represent injury to the MPFL, which should be palpated along its length to its origin at the adductor tubercle. The insertion of the nearby VMO should also be palpated for pain or defect. In patients with prior operative scars, the operative area should be examined for neuroma. These areas are highly innervated,53,54 and a diagnostic injection of local anesthetic can be used to confirm a suspected diagnosis. The medial articular facet can be palpated by displacing the patella medially with the knee in 30 degrees of ­flexion; the patella tilts laterally,

Figure 22C2-15   Palpation of the peripatellar structures is important. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Patella 1557

Figure 22C2-16   Lateral pull test; with contraction, the patella should move a bit laterally. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

uncovering the medial facet.55 Pain on the lateral border of the patella is commonly found in excessive lateral pressure syndrome and should be differentiated from pain on the lateral femoral condyle as a result of osteochondral fracture after dislocation. Finally, the medial and lateral joint lines should be evaluated for tenderness that may represent meniscal tear, arthrosis, or tear of the patellomeniscal ligament along its course to insertion on the anterior horn of the medial meniscus.55 Active and passive range of motion should be evaluated next. Any deficit or asymmetry should be measured for future reference. During motion, the patella should be observed for abnormal tracking and palpated for patellofemoral crepitation. A resisted straight-leg raise should be performed to rule out disruption or injury to the extensor mechanism. Evaluation of patellar mobility is an essential part of the examination and should include passive testing for patellar tilt, glide, apprehension, and compression with comparison to the contralateral leg. The position of the knee is important in evaluating these criteria because the patella becomes captured in the femoral trochlea by 30 degrees of flexion. If this difference in stability is not found, the possibility of underlying trochlear dysplasia or patella alta should be considered. Before passive stability testing, the movement of the patella under isometric quadriceps contraction should be observed in full extension. During this lateral pull test, the position of the patella should be observed before, during, and after contraction; normally, the patella should move straight superiorly or superiorly and laterally in equal proportions, and any excessive lateral movement indicates an abnormal pull of the quadriceps tendon48 (Figs. 22C2-16 and 22C2-17).

Figure 22C2-17   The resultant lateral vector. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Patellar tilt is used to determine whether the medial or lateral soft tissue restraints are excessively tight and should be performed with the knee extended and the quadriceps relaxed (Fig. 22C2-18). The examiner pushes down on the medial patella while lifting the lateral patella, taking care not to dislocate the patella. Normally, the patella should correct to neutral with the anterior surface of the patella parallel to the examination table. Decreased tilt, or

Figure 22C2-18   Passive patellar tilt. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

1558 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

i­nability to reach the neutral position, is indicative of tight lateral structures. Patellar glide demonstrates the integrity of the medial and lateral patellar restraints by passively translating the patella to each side with the quadriceps relaxed (Fig. 22C2-19). Lateral glide is a test of the medial structures, and medial glide is a test of the lateral structures. Although Kolowich and colleagues described this test with the knee in 20 to 30 degrees of flexion, examination in full extension isolates the soft tissue restraints by removing the bony contact of the patella in the trochlea.48 The degree of glide is graded in four levels, determined by the amount of translation relative to the width of the patella divided into quadrants. A deficient medial or lateral restraint is defined as glide of three or four quadrants in the respective direction.48,51 Grading of both patellar tilt and glide is highly variable among observers,56 which has led to the use of instrumented measurement of patellar displacement.57-59 Skalley and associates attempted to quantify normal medial and lateral motion with a handheld force gauge and concluded that medial and lateral displacement was limited by ligamentous restraints only with manually produced displacement more reproducible.60 Fithian and associates used a similar device to quantify patellar balance as the difference between instrumented medial and lateral displacement with good interobserver and intraobserver reliability.57 Despite the benefits of objective translational data, these instruments are not widely used in clinical practice. Translation of the patella that causes a feeling of impending dislocation is called apprehension.61,62 The apprehension test is the classic examination in acute patellar dislocations and is often positive in patients with recurrent dislocations. The patella is translated laterally with the knee flexed 20 to 30 degrees and the quadriceps relaxed. In a positive test, the patient feels the patella about to dislocate and contracts the quadriceps involuntarily. The patient may look apprehensive about a recurrent dislocation.61 Although pain ­usually accompanies the apprehension, the latter is considered the major element of a positive test.

The apprehension test may also be used for medial instability, which is most commonly an iatrogenic injury following a lateral release.26-28 Several other tests for medial subluxation should be used to evaluate patients with failed patellofemoral surgery. Fulkerson has described a test that attempts to reproduce the reduction of a medially subluxated patella.63 The patella is manually displaced medially as the knee is flexed; the test is positive if sudden relocation of the patella into the trochlear groove reproduces the symptoms. The gravity subluxation test is also used for evaluation of iatrogenic medial dislocation from previous lateral release.64 The patella is displaced medially while the patient is in the lateral decubitus position; an inability to actively reduce the patella against gravity by muscular contraction demonstrates incompetence of the lateral supporting structures. Compression of the patella against the trochlea may cause pain in patients with articular lesions, which may be the result of a previous patellar dislocation. Compression should be done with the knee in fixed positions of flexion, taking care to compress only the patella and not the surrounding soft tissue structures. The degree of knee flexion may predict the location of articular damage on the patella because the contact area moves proximally on the patella as flexion increases. Suspicion of an articular cartilage lesion should be confirmed with imaging studies for operative planning. The supine examination should be concluded with evaluation of lower extremity flexibility. Tightness is most commonly encountered in the hamstrings, where excessive tightness requires more quadriceps force for extension of the knee during gait; the result is increased pressure across the patellofemoral joint. Hamstring flexibility is best assessed by measuring the popliteal angle. In the supine position, the hip is flexed 90 degrees, with the contralateral leg extended. The maximal extension of the knee in this position is a measure of hamstring tightness. Gastrocnemius and soleus tightness should also be assessed with the patient in the supine position. The flexibility of both muscles can be judged by ankle dorsiflexion with the knee extended. With the knee flexed, the gastrocnemius is relaxed, and the soleus is isolated. In both positions, the ankle should dorsiflex 15 to 20 degrees past neutral. Limitation causes a compensatory increase in subtalar pronation, increasing internal tibial rotation during gait. Iliotibial band tightness has been correlated with lateral knee pain and should be assessed with Ober’s test.65 In the lateral decubitus position with the affected knee up, the leg abducted, brought into full extension, and then adducted toward the table. Tightness or pain may be elicited. Comparison should be made with the opposite side.

IMAGING Radiographs

Figure 22C2-19   Patellar glide. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Imaging of the patellofemoral joint should begin with a complete series of plain radiographs in all cases of recurrent patellofemoral dislocation. Radiographs should include standing anteroposterior, 45-degree standing posteroanterior, 30-degree flexion lateral, and axial views. If any

Patella 1559

malalignment is detected on clinical examination, a standing, full-length alignment film should also be obtained.

The Anteroposterior Radiograph In recurrent patellar dislocation, the standing anteroposterior and 45-degree posteroanterior views offer little useful information. Patella fractures and bipartite patella are most easily seen on these views. Additionally, joint space narrowing, subchondral sclerosis, and osteophytes associated with tibiofemoral arthritis are easily seen. With the knee in neutral rotation, medial and lateral position of the patella can be quantified, but subluxation is best quantified on the axial view. Although patella alta and baja can be seen, they are best quantified on the 30-degree lateral view.

The Lateral Radiograph The 30-degree flexion lateral view provides a wealth of information about the patellofemoral joint, including morphology of the femur as well as size, shape, and position of the patella. To provide the most useful and accurate information, the radiograph must be a true lateral, showing overlap of the distal and posterior cortices of the medial and lateral femoral condyles.

A

B

D

E

The height of the patella is most easily seen on the lateral radiograph. Because patella alta is a known risk factor in recurrent patellar dislocation, height of the patella should always be determined. A number of techniques have been described using the lateral radiograph (Fig. 22C2-20). The method of Blumensaat determines the height of the patella based on the projection of the roof of the intercondylar fossa on a 30-degree flexion lateral (see Fig. 22C2-20F).66 The anterior projection of this line should intersect the inferior pole of the patella; if the line passes below, patella alta is present. This method is hindered by anatomic and positional variation. The knee must be in exactly 30 degrees of flexion for accurate determination of patellar height. In addition, Brattström pointed out a second potential problem with this method after demonstrating a wide variation in the angle of Blumensaat’s line. On the lateral radiographs of 100 patients, he found that the angle between Blumensaat’s line and the longitudinal axis of the femur ranged from 27 and 60 degrees, showing that a 10-degree variation in the angle would equal a 10-mm difference in patellar height.67 Even when modified to normalize the angle of the line, Blumensaat’s method shows poor correlation with other patellar indices.68 The method of Labelle and Laurin requires the knee be flexed 90 degrees for the lateral radiograph, a position that

C

F

Figure 22C2-20   A, Insall-Salvati ratio. B, Grelsamer’s modification of the Insall-Salvati ratio. C, Blackburne-Peel measurement. D, Caton-Deschamps measurement. E, Labelle-Laurin measurement of patellar position. F, Blumenstaat’s measurement of patellar position. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

1560 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

is used clinically to determine patella alta (see Fig. 22C220E). Using this method, patella alta is determined by the projection of the anterior cortex of the femur in relation to the proximal pole of the patella. In their original study, this line passed above the patella in 97% of knees. Based on these findings, patella alta is diagnosed by intersection of this anterior femoral line with any portion of the patella.69 Patellar height ratios are popular among clinicians because they are independent of knee flexion angle, easily recalled, and simple to use. The Insall-Salvati ratio was described in 1971 and has remained the most identifiable index for patellar height. It is based on the ratio of the length of the patellar tendon divided by the greatest length of the patella (see Fig. 22C2-20A). The normal ratio defined by the authors was 1.0, plus or minus 20%. Therefore, patella alta is defined as an Insall-Salvati ratio greater than 1.2.70 Unfortunately, difficulty in determining the exact insertion of the patellar tendon and abnormal morphology of the nonarticular portion of the patella may falsely alter this ratio. In addition, patellar tendon length varies between sexes in the normal population.71 For these reasons, a modified Insall-Salvati ratio was proposed by Grelsamer and Meadows in 1992.72 This modified ratio is defined as the distance from the inferior articular surface of the patella to the patellar tendon insertion divided by the length of the articular surface (see Fig. 22C2-20B). Using this method, patella alta is defined as a ratio greater than 2.0, a point at which only 3% of controls would be falsely identified as patella alta.72 In a recent study, the traditional Insall-Salvati ratio was found to be more reproducible; there was agreement in classification in 67% of radiographs using the traditional method, compared with 47% agreement using the modified method proposed by Grelsamer.73 Blackburne and Peel described a ratio that is independent of the length of the patellar tendon, once again citing the difficulty in measuring the true length of the patellar tendon (see Fig. 22C2-20C). Their ratio is based on the anterior projection of the tibial plateau and is defined as the ratio of the length of the perpendicular line from the lower end of the patellar articular surface to the tibial plateau line divided by the length of the articular surface. In their study of 171 normal knees and 58 with recurrent subluxation of the patella, they defined a ratio of 0.8 to be normal and a ratio greater than 1.0 to describe patella alta.74 Caton and Deschamps also devised a ratio to address the difficulty in measuring the length of the patellar tendon.75 Their ratio is defined as the ratio of the distance from the inferior articular surface of the patella to the anterosuperior border of the tibia divided by the length of the articular surface of the patella (see Fig. 22C2-20D). Patella alta is defined as a ratio greater than 1.2. A summary of the indices of patellar height is provided in Table 22C2-3. Several recent studies have compared the different techniques of measuring patellar height. In 1996, Berg looked at the Insall-Salvati, the modified Insall-Salvati, the ­Blackburne-Peel, and Caton-Deschamps indices using three blinded observers and radiographs of 15 asympto­ matic knees. The Blackburne-Peel ratio was found to be the most reproducible, having the lowest standard error and least intraobserver error. Interestingly, the authors

  TABLE 22C2-3  Indices of Patellar Height Measurement

Normal

Alta

Baja

Insall-Salvati70

1.0 — 0.8 <1.2 Visual

>1.2 >2.0 >1.0 >1.2 Visual

<0.8 — <0.5 <0.6 —

Modified Insall-Salvati72 Blackburne-Peel74 Caton-Deschamps75 Labelle-Laurin69

also found that mild osteoarthritis decreased measurement variability by 24% and proved that indices were unaffected by knee flexion angles between 30 and 50 degrees.76 In a similar study, Seil and colleagues evaluated the same indices in addition to the Labelle-Laurin method and also looked at the classification of each knee as normal, alta, or baja.77 Despite low interobserver error and high correlation coefficients, the classification into normal, alta, or baja varied in 68% of patients. The Insall-Salvati ratio showed the lowest number of normal patella and the Labelle-­Laurin showed the highest number of patella alta. Seil and colleagues recommended the Blackburne-Peel ratio because it showed the most intermediate classification results and had the least interobserver variability.77 Like patella alta, trochlear dysplasia is a known risk factor for recurrent patellar dislocation, and the morphology of the femoral trochlea is essential information obtained from the lateral radiograph. Although the axial view is often used to quantify the depth of the trochlea, it only provides information about one fraction of the trochlea, and dysplasia in the proximal portion may be missed. According to Dejour and coworkers, the 30-degree axial view allows diagnosis of only 65% of dysplasias.78 Malghem and Maldague first described the anatomy of the femoral trochlea on the lateral view, citing the importance of visualizing the depth of the trochlea over its entire length.79,80 On the true lateral radiograph, three anterior lines are seen: the most anterior is a projection of the medial femoral condyle, the middle is a projection of the lateral femoral condyle, and the remaining line is a projection of the floor of the trochlea (Fig. 22C2-21). In a normal knee, the line of the lateral femoral condyle should terminate proximal to that of the medial condyle. In the knee with a shallow trochlea, its projection will prematurely cross that of the condyles in their proximal portion. Dysplasia can be quantified or classified based on this appearance. The depth of the trochlea is defined as

Figure 22C2-21   Lateral view of the knee. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Patella 1561

Figure 22C2-22   Trochlear depth. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

the ­ distance between the projections of the trochlea and ­condyles at a position 1 cm distal to the upper extent of the trochlea; this is quantified as the average of the distance from the trochlea to the medial and lateral condyles. In their original study of 218 knees, Malghem and Maldague showed a depth of less than 5 mm in 78% to 90% of patients with treated or symptomatic instability but only 20% of ­controls.80 In a landmark publication, Dejour and associates subsequently defined depth with two separate measures in a radiographic study of the factors of patellar instability.78 The first is called the trochlear bump and is defined as the distance between the projection of the anterior femoral cortex and the projection of the trochlea, which can be anterior positive or posterior negative. The “bump” was greater than +3 mm in 85% of patients with objective patellar instability. The second measure is the trochlear depth, which is defined as the depth of the trochlea along a line 15 degrees from the perpendicular to the tangent of the posterior femoral cortex (Fig. 22C2-22). A depth of less than 4 mm was found in 85% of patients with objective patellar instability and in only 3% of controls. In addition to quantifying dysplasia, several classifications exist. In conjunction with Dejour, Galland and colleagues devised a classification system based on the location of the crossing sign.81 At some point, the line of the trochlear floor crosses the anterior projection of the lateral femoral condyle, a place at which the trochlea becomes flat. There are varying degrees of dysplasia with a positive crossing sign (Fig. 22C2-23). In type I, the line of the trochlea crosses the medial and lateral condyle lines in its proximal portion, and the dysplasia is minor. In type II, the medial condyle is smaller, and the line of the trochlea crosses that of the medial condyle first in its midportion followed by the lateral condyle. In type III, the condyles are normally symmetrical, but the line of the trochlea crosses the condyles in a more distal position. Dejour believes the crossing sign is of “fundamental diagnostic value” because it is present in 96% of patients with objective patellar instability.78 Because it is not an objective measurement, there is some variability between observers in grading degrees of dysplasia. In a radiographic study of 68 knees, Remy and coworkers showed that the crossing sign was reliable for diagnosis of dysplasia, with only a 3% chance of falsely diagnosing a dysplastic trochlea as normal.82 However, there was poor interobserver agreement for identification of degree of dysplasia.

Figure 22C2-23   Galland’s three types of dysplasia. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

The Axial Radiograph The axial radiograph has long been used for evaluation of patients with patellar instability. In 1964, Brattström first found a relationship between femoral sulcus angle and patellar dislocation by demonstrating a 10-degree increase in sulcus angle in the knees of patients with documented patellar dislocation compared with 100 controls (Table 22C2-4).5 Since that time, a number of radiographic techniques have been developed to evaluate the patellofemoral articulation in patients with patellar instability, including views with intrinsically and extrinsically applied stress. The traditional “sunrise” or “skyline” views should not be used for evaluation of patellar instability; the knee flexion needed for these radiographs is too great, the patella will be captured within the trochlear groove, and dysplasia of the proximal trochlea will be missed. Additional axial view techniques have been described that show the patella in the proximal trochlea and should be used when

  TABLE 22C2-4  Brattström’s Relationship between Femoral Sulcus Angle and Patellar Dislocation

Controls Right patellar dislocation Left patellar dislocation

Females

Males

141.8 152.5 150.1

141.9 149.9 151.1

1562 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 22C2-24   Axial view, called Merchant’s view of the patella.

evaluating a patient with patellar instability. In the 1950s, Hughston began using a modification technique, separately described by Settegast and Jaroschy, in which the patient is prone with the knees flexed 60 degrees and beam angled cephalad 45 degrees to the horizontal, allowing visualization of the patella and the proximal trochlea.62,83-85 However, the prone position causes compression of the patella and the orientation of the x-ray cassette distortion of the image. The axial views subsequently described by ­Merchant and colleagues86 and Laurin and associates87,88 are most ­ commonly used, and objective measurements have been developed to aid in clinical interpretation of these radiographs. Merchant described a technique in which the patient is supine with the knees flexed 45 degrees over the end of the table (Fig. 22C2-24).86 The x-ray beam is angled caudad 30 degrees from the horizontal, and thus, 30 degrees from the shaft of the femur; the difference in knee flexion and beam angle allows for positioning of the x-ray cassette. In this study, Merchant proposed a new method of evaluating the congruence of the patellofemoral joint. On this axial radiograph, the sulcus angle is bisected, and a second line is drawn to the lowest point on the articular ridge of the patella (Fig. 22C2-25). The congruence angle is designated positive if it lies lateral to the bisector and negative if medial to the bisector. In 100 normal subjects, the congruence angle was −6 degrees with a standard deviation of 11. Merchant then defined a cutoff angle of +16 degrees based on the 95th percentile of normal subjects. The average congruence angle in 25 subjects with recurrent patellar dislocation was +23 degrees, nearly 30 degrees different from controls. Aglietti and associates confirmed this difference in congruence angle between populations using the method described by Merchant.34,89 In their series, the average congruence angle among controls was −8 degrees but was +16 in those with recurrent patellar dislocation. In another group of patients with patellar subluxation, Moller and coworkers found the congruence angle often to be within normal limits.90 However, they showed a significant difference when compared with the opposite, asympto­ matic knee and suggested that a side-to-side comparison may be more useful than an absolute number. Laurin and colleagues proposed a similar technique for the axial radiograph, but placed the x-ray tube between

the legs, sending the beam caudad (Fig. 22C2-26).88 This technique requires only 20 degrees of flexion and allows the patient to hold the x-ray cassette closer to the knee. Although not cited in their work, a lower knee flexion angle allows visualization of a more proximal portion of the trochlea. Laurin defined the lateral patellofemoral angle as the angle formed by a line referencing the medial and lateral femoral condyles and a line referencing the lateral patellar facet, which does not vary by Wiberg type. Laurin and associates demonstrated a difference in the lateral patellofemoral angle between control subjects whose angles opened laterally and subjects with patellar subluxation whose angles were zero or opened medially.87 There was no difference between groups with the knees flexed 60 or 90 degrees and the patella captured in the trochlea. In a subsequent work, Laurin and associates described lateral patellar displacement in addition to the lateral patellofemoral angle on the axial radiograph.87 ­ Displacement

Figure 22C2-25   Merchant’s view measurements. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

Patella 1563 Figure 22C2-26   A and B, Laurin’s view. (Reprinted with the permission of The Cleveland Clinic Center for Medical Art & Photography © 2008.)

A

B

is determined by position of the patella relative to a line originating at the medial femoral articular margin that is perpendicular to the femoral reference line used in determination of the patellofemoral angle. In 100 control subjects, the medial margin of the patella was no more than 1 mm lateral to this line but was displaced significantly ­lateral in 54% of patients with patellar subluxation. The sulcus angle is commonly used to diagnose trochlear dysplasia. It was described by Brattström as the angle subtended by two lines originating at the lowest point in the trochlea to the highest points of the medial and lateral condyles.5 The sulcus angle can be determined on any axial

radiograph but must be compared with normative data for the radiographic technique used. The larger the angle from the beam to the long axis of the femur, the more distal part of the trochlea does the radiographic projection represent; comparison across techniques may give a false interpretation of dysplasia. This concept was confirmed in a ­ magnetic resonance imaging study by Kujala and coworkers that showed sulcus angle variation by knee flexion angle in women with recurrent patellar dislocation and asymptomatic controls.91 Normal sulcus angles have been reported to be 142 degrees using Brattström’s technique,5 138 to 139 degrees using Merchant’s technique,86,92-94 and

1564 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 22C2-5  Sulcus Angle and Dejour-type Dysplasia Dysplasia Type

Davies et al (2000)95

Steiner et al (2006)96

I II III

143.1 149.0 171.2

145.4 148.1 153.8

142 degrees using ­ Laurin’s technique.94 Davies and colleagues95 and Steiner and associates96 compared sulcus angle to type of dysplasia as defined by Dejour on the lateral radiograph and showed increasing sulcus angle with worsening degree of ­dysplasia (Table 22C2-5). The plain radiograph fails to show the contour of the articular cartilage, and van Huyssteen and coworkers have shown that the cartilaginous sulcus angle is significantly higher than the bony angle in subjects with and without trochlear ­dysplasia.97

Stress Radiographs Stress radiographs of the patellofemoral joint attempt to capture the patella in an abnormally subluxated or dislocated position, usually on an axial view. The stress may be intrinsic, such as that from quadriceps contraction or weight-bearing, or it may be extrinsic, such as that applied from an external device. These views are especially ­useful in evaluating the competence of the soft tissue structures not directly seen on radiographs, especially in cases in which bony anatomy appears normal. Lateral movement of the patella during quadriceps contraction has been demonstrated clinically in patients with recurrent patellar dislocation,98 which is a premise for axial radiographs under the same conditions. Al-Habbal and colleagues evaluated patellofemoral indices using the Merchant axial view both with and without quadriceps contraction.99 There was no difference between groups with respect to congruence angle. More important, differences seen in the congruence angle existed both with and without contraction, and the authors concluded that contraction of the quadriceps did not add the diagnostic yield of axial radiographs. To evaluate the patellofemoral joint with quadriceps contraction under physiologic conditions, Toft devised a weight-bearing axial radiograph technique with the knee at 30, 60, and 90 degrees of flexion.100 Turner and Burns subsequently suggested a similar technique at 40 degrees of flexion.101 However, no controlled study has examined patellofemoral indices using the weight-bearing radiograph. External rotation of the leg during an axial radiograph has been proposed to simulate the position of the leg during episodes of patellar subluxation. Miremad and associates first suggested this method, using a cushion and thigh strap to statically produce this position.102 Malghem and Maldague used a dynamic method of manually holding the leg externally rotated while centering the leg within the x-ray beam with the quadriceps relaxed.103 Subluxation was determined by a congruence angle greater than +16 degrees, which was demonstrated in 100% of patients with patellar instability on the lateral rotation view but in only 26% on the routine view.3

Fukui and colleagues combined lateral rotation of the leg with and without quadriceps contraction in a recently published technique.104 The manual application of valgus and external rotation without quadriceps contraction produced an increase in congruence angle and a decrease in lateral patellofemoral angle in both symptomatic and control knees. Contraction of the quadriceps reduced the patella into the trochlear groove in all of the normal knees but in only 70% of the symptomatic knees. The remaining 30% demonstrated worsening displacement, suggesting that subluxation occurs in daily activities; all of these eventually underwent realignment surgery. Instrumented quantification of patellar mobility has been proposed to provide an objective component to the physical examination.57,60 Similarly, stress radiographs that apply direct pressure provide an objective component to radiographic evaluation of patellar instability and best determine the competence of the patellofemoral ligamentous restraints. Teitge and colleagues proposed a method of applying a constant 16-pound load to both the medial and lateral side of the patella while taking an axial radiograph using the method of Merchant; radiographs were compared with those of the unloaded knee and the contralateral knee.105 Using 2 standard deviations to achieve 95% confidence, a 3.7-mm difference in displacement compared with the contralateral, asymptomatic knee was determined to be abnormal in those with lateral instability; a 3.5-mm difference was considered abnormal for those with medial instability. Teitge concluded that stress radiographs are more sensitive than other radiographic measures in assessing unidirectional patellar instability and suggested that they can be used to document improvement after operative intervention.

Computed Tomography The use of CT has become more widespread in the evaluation of patellofemoral disorders. The cross-sectional nature of the scan allows the patellofemoral joint to be examined with the knee in full extension, a position that conventional axial radiographs cannot image; imaging in extension is essential because most patellar instability takes place in the first 30 degrees of flexion, before the patella becomes restrained by the trochlea. Early computed tomographic studies in full extension demonstrated this, showing that the patella rests proximal to the trochlear groove and that only 13% of patellas are centered with the quadriceps relaxed, decreasing to 4% with quadriceps contraction.106 Subsequent work demonstrated linear increases in patellofemoral parameters from quadriceps relaxation to contraction, obviating the need for CT with both.107,108 Measurements of congruence angle, lateral patellofemoral angle, patellar tilt, and others have been studied extensively on CT, just as they have on conventional axial radiographs. However, Delgado-Martinez and colleagues demonstrated that the only measurement that can be reliably reproduced on CT in full extension is lateral patellar tilt,109 and Biedert and Warnke have shown no correlation of clinically determined Q angle with lateral patellar tilt or ­displacement.110 Using patellofemoral measurement parameters on CT with knee flexion from 0 to 30 degrees, Schutzer and associates established criteria for normal patellar ­tracking

Patella 1565

as less than 8 degrees of tilt and a congruence angle of greater than 0 degrees at 10 degrees of knee flexion.111,112 Three patterns of malalignment were established: in type I, patients had subluxation without tilt, in type II, subluxation with tilt, and in type III, tilt without subluxation. Unfortunately, this study was not limited to patients with instability, and a clear correlation between malalignment and symptoms cannot be established. In contrast, Inoue and associates evaluated only patients with symptomatic patellar subluxation and asymptomatic controls using CT in full extension combined with axial radiographs in 30 and 45 degrees of flexion.113 The difference in the lateral patellofemoral angle between groups was greatest on computed tomographic images in full extension, leading to the conclusion that CT is more accurate in detecting patellar subluxation than conventional axial radiographs.113 Dejour and associates also noted a highly significant difference in lateral patellar tilt in patients with objective patellar instability and established a pathologic threshold of 20 degrees.78 Dejour has provided the most objective evidence of patellar instability related to malalignment on CT with the TT-TG measurement. This measurement is obtained by superimposing two computed tomographic images, which best represent the trochlear groove and tibial tuberosity. The distance between the center of the groove and tuberosity in a line parallel to the posterior femoral condyles represents the TT-TG distance. In Dejour’s original study of 143 knees with patellar instability, there was a significant difference in the TT-TG distance compared with asymptomatic controls, and a threshold value of 20 mm was established as pathologic.78 Because of difficulty identifying anatomic structures on superimposed computed tomographic images, Koëter and associates developed a modified technique of measuring the TT-TG distance, which considerably reduces measurement error.114 Measuring the TT-TG distance on conventional axial radiography had been proposed but was shown to have an unacceptable degree of measurement error.115,116 Rotational abnormalities of the tibia and femur are easily evaluated with CT and have been suggested as possible causes of patellar instability. Studies have associated increased femoral anteversion with increased patellofemoral contact pressure and anterior knee pain, but not patellofemoral instability.117,118 In addition, two works have shown no correlation of sulcus angle, congruence angle, or lateral patellofemoral angle with increased femoral anteversion.117,119 Conversely, increased external tibial torsion has been associated with instability.120-123 In a rotational study of 1672 knees, Turner demonstrated that only patellofemoral instability and Osgood-Schlatter disease were associated with increased lateral torsion, with instability averaging 25 degrees.123 Other studies have performed derotational osteotomy for tibial torsion averaging 30 and 40 degrees.120,122 If abnormalities are suggested by clinical examination, a computed tomographic rotational study should be considered. Kinematic CT acquiring up to one image per second has become possible with the advent of spiral CT scanners and improved computer systems.124 In a trial of 20 patients with anterior knee pain, cinematic viewing of the patellofemoral articulation allowed successful visualization

  TABLE 22C2-6  Magnetic Resonance Imaging–Based Quantitative Predictors of Instability

Lateral patellar tilt Trochlear groove depth Insall-Salvati ratio Patellar nose Patellar morphology ratio

Cutoff

Sensitivity

Specificity

> 11 < 5 mm

92.7 85.7

63.3 71.7

> 1.2 < 9 mm < 1.2

78.0 66.1 44.1

67.6 84.5 86.9

From Escala JS, Mellado JM, Olona M, et al: Objective patellar instability: MR-based quantitative assessment of potentially associated anatomical features. Knee Surg Sports Traumatol Arthrosc 14(3):264-272, 2006.

of patellar tracking from 0 to 45 degrees of knee flexion and proved more useful than single images.125 Currently, there are no published data on the use of kinematic CT in patients with patellar instability, but it is an imaging modality likely to be seen in the near future.

Magnetic Resonance Imaging The use of magnetic resonance imaging (MRI) of the knee has become increasingly popular as machines have become more readily available and cost has decreased. In addition, scans have become more useful as resolution has improved and more specific sequences have been developed. The same measurements used on plain radiographs and CT can be applied to MRI without exposure to radiation. Escala and colleagues have shown that several parameters have a high sensitivity in predicting objective patellar instability: lateral patellar tilt greater than 11 degrees, trochlear groove shallower than 5 mm at the level of the roman arch, and Insall-Salvati ratio greater than 1.2 (Table 22C2-6).126 In addition, MRI offers the distinct advantage of visualizing the articular cartilage, meniscus, and ligaments. The classic MRI following acute patellar dislocation includes a triad of findings: impaction injury to the lateral femoral condyle, osteochondral injury to the medial patellar facet, and disruption of the medial retinaculum.127 Signal change within the superior portion of Hoffa’s fat pad has also been described.128 In acute patellar dislocation, the MPFL is universally disrupted with obvious surrounding soft tissue edema.19,129,130 In the setting of recurrent dislocation, these findings may be absent, and the ligament may have healed in an elongated position. Shellock and colleagues were the first to describe the use of kinematic MRI in the evaluation of patellofemoral tracking, compiling a series of nine axial images of the patella from 0 to 32 degrees of flexion in a cinematic loop.131 The images were able to show unloaded patellofemoral tracking through the first 30 degrees of flexion. Kujala and associates similarly studied the patellofemoral joint on MRI from 0 to 30 degrees of flexion but included quadriceps contraction as an additional variable.132 With quadriceps contraction in asymptomatic knees, the lateral patellar tilt decreased by 4 degrees, whereas changes in patellar displacement were unpredictable. Both these early studies relied on passive knee flexion that may not accurately simulate daily or sporting activities.

1566 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 22C2-7 Comparison of Imaging Modalities

Invasiveness Range of motion Availability Active joint motion Evaluation Examination time Side effects

Arthroscopy

Axial Radiographs

Computed Tomography

Kinematic Magnetic ­Resonance Imaging

Yes Full flexion to full extension +++ Passive Qualitative 60 min Postsurgical

No Any angle > 30 degrees

No 60 degrees of flexion to full extension ++ Active or passive Quantitative 30 min Radiation exposure

No 45 degrees of flexion to full extension + Active or passive Quantitative or qualitative 20 min None

+++ Passive Quantitative 15 min Radiation exposure

Adapted from Muhle C, Brossmann J, Heller M: Kinematic CT and MR imaging of the patellofemoral joint. Eur Radiol 9(3):508-518, 1999.

In 1991, Shellock and associates introduced a new pulse sequence that allowed imaging of the patellofemoral joint during active motion, a way to quantify the effect of both static and dynamic patellar stabilizers.133 Brossmann and colleagues subsequently studied 13 patients with patellar maltracking and 15 controls with motion-triggered cinematic MR imaging.134 During active motion scans, there were significant differences in all measurement parameters between groups; these differences were not universally present during passive motion scans. Subsequent studies have confirmed these findings.135 Kinematic MRI provides useful information but is currently limited by availability. A summary of imaging modalities is presented in Table 22C2-7.

NONOPERATIVE TREATMENT Nonoperative treatment of chronic patellar dislocations with immobilization followed by rehabilitation has produced dismal results, with nearly half of patients having recurrent dislocations or continued symptoms. Immobilization following recurrent episodes of patellar dislocation is of little benefit, although it may be used in the short-term for patient comfort. An attempt at rehabilitation should be reserved for the patient who experiences only occasional dislocation and displays no obvious anatomic or radiographic abnormalities that may predispose to recurrent episodes. Those with dislocation in activities of daily living or gross anatomic deficiencies will likely need operative treatment. In some patient populations, a trial of rehabilitation is warranted to show commitment to getting well because surgical treatment often requires an extended period of postoperative rehabilitation for success. Although numerous prospective studies exist on the nonoperative treatment of acute patellar dislocation with rehabilitation, no studies exist on its use in recurrent patellar dislocation. If chosen, rehabilitation may be augmented by the use of a patellar orthosis if tolerated by the patient. Palumbo first proposed a brace with a pad to prevent lateral patellar translation, which proved subjectively beneficial in 39 patients with subluxation. Since that time, several kinematic MRI studies have examined the position of the patella with and without the use of a brace. Shellock and colleagues demonstrated qualitative improvement in patellar position in 76% of subjects and cited bony dysplasia and

patella alta as ­reasons for failure.136 In a subsequent study also using qualitative measures, they showed a similar rate of improvement but cited obesity and patella alta as reasons for failure.137 Only one study has used quantitative measures to evaluate patellar position on kinematic MRI; Muhle and associates138 showed no significant difference in patellar tilt, bisect offset, and lateral patellar displacement with and without a patellar brace during active motion. Given these results, bracing should be applied only to patients who meet criteria for rehabilitation: occasional dislocators without gross anatomic abnormalities.

OPERATIVE TREATMENT Operative treatment for recurrent dislocation of the patella should be directed at the underlying pathology that can be determined from careful evaluation of history, physical examination, and imaging studies. There is a tendency for less experienced patellofemoral surgeons to perform the same operation, anterior medialization of the tibial tuberosity with lateral release, for all cases of patellar instability. However, surgical treatment should be customized to the causative pathology found in each patient. According to Andrish, the first principle in the operative treatment of patellar instability is to individualize, customize, and normalize to correct the offending pathoanatomy and not to create a secondary pathoanatomy to compensate for the primary pathoanatomy.139 This philosophy is clearly reflected in the recent literature as increasing reports of MPFL reconstruction and trochlear osteotomy have been devised to address loss of medial retinacular integrity and trochlear dysplasia, respectively.

Distal Realignment Procedures Most surgeons immediately think of extensor mechanism reconstruction procedures when dealing with recurrent patellar instability. Distal realignment techniques can be divided into those that transfer the bony insertion of the patellar tendon and those that involve soft tissue transfer, with the former reserved for patients who have reached skeletal maturity. Modern bony techniques are based on the description by Roux in 1888, which included transposition of the patellar tendon insertion, tightening of the medial structures, and lateral retinacular release.140 Shortly thereafter, Goldthwait described a soft tissue realignment

Patella 1567

that involved transposition of the lateral half of the patellar tendon under the remaining medial half with attachment to the soft tissue of the pes anserine.141 In 1938, Hauser reported a procedure that became synonymous with extensor mechanism realignment142; this procedure involves a distal and medial transfer of the tibial tuberosity onto the medial surface of the tibia with concomitant release of the lateral retinaculum and imbrication of the medial retinaculum. By today’s standards, the degree of transposition is considered extreme and fraught with complications. The placement of the tibial tuberosity on the sloping medial border of the tibia not only medializes the patella but also places it in a position posterior to its original location, increasing patellofemoral contact forces. Short-term results were favorable in cases of recurrent patellar dislocation, with up to 70% good to excellent results.143 However, suspicion of increased contact pressures was confirmed with long-term results that showed progression to arthritis in the same number of patients at 7 to 18 years.144-146 More recently, Barbari and coworkers showed that only 29% of 54 patients were stable and free from pain at 8 years of follow-up, with a high incidence of both patellofemoral and tibiofemoral arthritis.147 The Elmslie-Trillat procedure allows medialization without posterior transfer of the tibial tuberosity in combination with lateral release and medial capsular reefing. At a short-term follow-up of 3 years, Cox reported a 7% rate of recurrent dislocation but 73% good to excellent results.148 At 26-years’ follow-up in a subset of the same patients, only 7% had sustained a recurrent dislocation, and good to excellent results remained in 54%. Incidence of arthritis was not quantified by radiograph, but visual analog scale pain scores averaged 48.149 Modifications of this procedure and its subsequent outcomes are too numerous to ­mention. Anteromedial tibial tuberosity transfer was first described in 1983 by Fulkerson.150 In this procedure, an osteotomy of variable obliquity allows the degree of anterior and medial transfer to be independently adjusted to address the patient’s individual pathology. Biomechanical studies have demonstrated beneficial effects of the procedure. Molina showed that the most predictable way of increasing contact area and decreasing patellofemoral stress is transfer of the tuberosity 1 cm anterior and 0.5 to 1 cm medial.151 Recent studies have looked at contact pressures following anteromedialization procedures. On the trochlear side, Beck and coworkers showed that contact pressures are decreased and shift medially with anteromedial transfer.152 On the patellar side, Ramappa and colleagues showed that the shift in force to the lateral facet was better corrected by medialization than anteromedialization, although both corrected maltracking and reduced the overall contact force.153 In addition to these findings, Fulkerson has pointed out that anteromedial transfer also unloads the distal pole of the patella, a common source of pain.154 Although the intent of the anteromedialization procedure is to reduce an abnormal Q angle and correct lateral tracking of the patella, some authors believe there are untoward effects. Arendt and associates have shown that medialization externally rotates the tibia and alters patellar rotation, rather than stabilizing the patella.155 Huberti

and Hayes showed that decreasing the Q angle not only resulted in decreased lateral facet load but also was always associated with increased loads elsewhere in the joint.156 Alteration of these loads following tuberosity transfer may lead to early medial facet arthrosis. Studies on the clinical implications of these biomechanical results in the specific patient population with recurrent patellar dislocation are few and of short-term follow-up only. Bellemans and colleagues evaluated 29 patients with chronic knee pain associated with patellar subluxation.157 Kujala scores significantly improved from 43 to 89, and Lysholm scores increased from 62 to 92 at a mean of 32 months following anteromedialization. In 2005, Dantas and coworkers reviewed a more specific group of 24 knees with recurrent patellar dislocation treated by anteromedialization with lateral retinacular release.158 At a mean 52 months’ follow-up, the average Lysholm score improved from 63 to 98, with no recurrent dislocations.158 Longer term studies are needed to evaluate the true value of the procedure.

Lateral Retinacular Release or Lengthening The lateral retinaculum is an important structure that contributes to both lateral and medial stability of the patella.130,132 Although often performed in combination with other procedures, isolated release of the lateral retinaculum has been described in the treatment of lateral instability.160-163 Multiple studies have demonstrated that isolated lateral release is effective in reducing patellar tilt; however, the clinical implications in instability are unclear.164,165 There are no randomized, prospective studies comparing isolated lateral release to other methods of treatment. In 2007, Ricchetti reviewed the lower level evidence comparing isolated lateral release and that performed in combination with medial soft tissue realignment.166 Groups treated with isolated lateral release showed significantly greater odds of recurrent instability and dislocation. The controversy surrounding lateral retinacular release is reflected in Fithian’s survey of the International Patellofemoral Study Group on the role of the lateral release.167 In this survey completed by a group of surgeons with unparalleled interest and experience in the treatment of the patellofemoral joint, 37% of respondents said that lateral patellar dislocation is actually a contraindication to lateral release. Only 11% of respondents considered recurrent lateral patellar dislocation as a clinical scenario for which they would consider lateral release. If chosen, lateral release should be done with caution; multiple authors have shown iatrogenic medial subluxation of the patella as a complication of this procedure, suggesting that lateral release may worsen cases of lateral patellar dislocation.26,28,168,169 If a tight lateral retinaculum is indeed the offending pathology, a lengthening should be performed as described by Larsen and colleagues.170 In this technique, the oblique fibers of the superficial lateral retinaculum are incised just lateral to the patella to the depth of the deep transverse layer. A plane is developed between the two layers to the lateral epicondyle, where the deep layer is released. The two layers can then be sutured together in a length appropriate for soft tissue balance.

1568 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Lateral retinacular release should be performed as an adjunct to other procedures chosen to address the primary offending pathology. In this situation, most authors recommend that it be performed to address residual patellar tilt or limited medial translation.171 The release should be performed as the final procedure, preferably suturing it together in a lengthened position that allows the most normal intraoperative patellar tracking.

Proximal Realignment Procedures The goal of proximal realignment surgery is to reestablish a dynamic balance of forces around the patella. In 1979, Insall and associates described the “tube” realignment for the treatment of chondromalacia patella.172 The procedure consists of release of the medial and lateral retinacular tissue, which are sewn together over the quadriceps proximal to the patella. Modifications of this procedure involving a lateral release with a lateral and 1-cm distal advancement of the vastus medialis were later described in the treatment of patellar dislocation. Scuderi and colleagues reported on 60 knees treated with this procedure for recurrent dislocation of the patella refractory to conservative management.173 At 3.5 years’ follow-up, there were good to excellent results, with significantly better results achieved in males and those younger than 20 years of age. There was only one recurrent dislocation, but 11 patients required further operation or manipulation. Zeichen and associates also reported midterm results of the procedure with 6 years’ follow-up on 36 patients.174 Only one patient sustained a recurrent dislocation, and good to excellent results were achieved in 63% of patients. Andrish pointed out that distal advancement of the vastus medialis may reduce the obliquity of the muscle vector and reduce its mechanical effectiveness.139 To increase the obliquity, the posteromedial portion of the tendon should be advanced, as described by Ahmad and coworkers.175

Trochleoplasty and Trochlear Osteotomy The importance of trochlear dysplasia in recurrent patellar dislocation is well accepted. Classic studies by Brattström and Dejour and associates demonstrated a 10-degree increase in femoral sulcus angle in patients with patellar instability and a “crossing sign” in 96% of patients with objective patellar instability, respectively.5,78 Procedures to treat the dysplastic trochlea were described long before these radiographic parameters popularized the concept of trochlear dysplasia; investigators proposed gouging of the femoral trochlea in the late 19th century, and Albee first proposed the lateral femoral osteotomy in 1915.4,176 Bony procedures that address the dysplastic trochlea can be divided into two main categories: those that deepen the trochlea (trochleoplasty) and those that elevate the anterior portion of the femoral condyles (trochlear osteotomy). Numerous variations in these techniques and retrospective case series of their results have been published. However, there are no prospective, randomized studies in the literature that support the use of these techniques. Albee’s lateral femoral osteotomy addresses dysplasia by elevating the lateral condyle to provide a buttress to lateral motion of the patella.4 Weiker and Black reported

the results of this procedure in six knees that had undergone an average of 3.8 procedures before presentation.177 At 8 years’ follow-up, functional status was improved in four knees, despite a high perioperative complication rate. Inability to regain motion in three of six knees resulted in one closed manipulation under anesthesia and two open lyses of adhesions.177 In contrast, Koëter and colleagues reported no loss of motion at a mean 4 years’ follow-up in a recent study of 19 knees undergoing the same procedure for objective patellar instability with isolated trochlear dysplasia.178 The authors cited meticulous surgical technique in combination with postoperative CPM for maintaining range of motion. Functional improvement was reported in 16 of 19 knees, and pain relief during activities was reported in 12 knees. Progressive osteoarthritis was noted in 2 knees, a possible complication recognized by both Koëter and Weiker. The risk for patellofemoral pain and arthrosis from increased contact pressure in the lateral aspect of the patellofemoral joint has been recognized as a limitation of the Albee procedure. Kuroda published a biomechanical study of cadaveric knees that evaluated the patellofemoral contact pressures following elevation of the lateral condyle by 3, 6, and 10 mm.179 No significant change was noted at 3 mm, but significant increases in pressure were noted at 45 degrees of flexion with 6 mm of elevation and at 15 and 45 degrees with 10 mm of elevation. Unfortunately, the clinical significance of this is unknown. Masse first devised a procedure to address dysplasia by deepening the trochlea rather than elevating the femoral condyles.180 The initial procedure involved simple impaction of the prominent cartilage but was subsequently modified to remove subchondral bone before impaction. In Dejour and colleagues’ modification of this technique, the subchondral bone of the trochlea is removed with a bur and the overlying cartilage is first incised, then impacted and fixed. Verdonk evaluated 13 knees treated with Dejour’s procedure at a mean follow-up of 18 months. Using the Larsen-Lauridsen score, 59% had fair or poor results, but an equal number had good or very good results on subjective scoring. Five patients suffered postoperative arthrofibrosis. The study is limited by a short follow-up period and a population that included both patients with instability and pain. Donell recently published results of the same procedure at a mean follow-up of 3 years. Radiographically, the trochlear boss was reduced from 8 to 0 mm, and the Kujala score improved from 48 to 75. One patient required reoperation for arthrofibrosis and six had bothersome patellofemoral crepitus. In an effort to improve on the Dejour technique, Bereiter and Gautier described a technique in which an osteochondral wedge is removed from the femoral trochlea, allowing for deepening of the underlying cancellous bone and replacement of the cartilage in a deepened position.182 Schöttle and colleagues evaluated 19 knees at a mean 3 years’ follow-up using this procedure for the treatment of recurrent patellar instability with trochlear dysplasia.183 No dislocations recurred, and good to excellent results were achieved in 16 of 19 knees, despite persistent apprehension in 2 knees. Two knees had poor results that were attributed to degenerative changes of the trochlea noted during surgery. Radiographic parameters were also

Patella 1569

e­ valuated before and after surgery, showing an increase in the mean depth of the trochlea from 1 to 7 mm, a decrease in the TT-TG distance from 20 to 10 mm, and an improvement in the patellar inclination angle from 22 to 8 degrees. In a similar study with 8 years’ follow-up, von Knoch and coworkers showed less promising results.184 Although pain improved in half the patients, it worsened in one third. There were no redislocations and one continued subluxation, and all patients but one were able to return to a higher level of sporting activity. Radiographic findings showed that the trochlear depth increased from 0 to 5 and that the trochlear bump decreased from 4 to 0. Radiographic findings of grade 2 or worse osteoarthritis in the patellofemoral joint were absent preoperatively but found in 30% ­postoperatively. The results of these studies evaluating both trochlear deepening and condylar elevating procedures show that dysplasia can clearly be reduced on radiographic follow-up and that stability can likely be restored. However, the risks for short-term postoperative stiffness and long-term patellofemoral arthrosis from an incongruent patellofemoral articulation and increased contact forces remain legitimate concerns.

Medial Patellofemoral Ligament Reconstruction In the past several decades, restoring soft tissue restraints has become an important surgical principle in the treatment of unstable joints.185 Examples of this include anterior cruciate ligament reconstruction of the knee, anterior talofibular ligament reconstruction of the ankle, Bankart repair of the shoulder, and ulnar collateral ligament repair of the elbow. The principle of reconstruction of the MPFL of the knee is no different. In addition to the patellofemoral articulation, the MPFL is an important passive stabilizer of the patella that serves as a “check rein” to lateral patellar motion.186 Historically, emphasis has been placed on realignment surgery, but Davis and Fithian state that there exists no evidence that any amount of malalignment will cause dislocation unless passive stabilizers are damaged.185 Studies have shown that the MPFL is universally disrupted in patellar dislocation and that its integrity is of primary importance in lateral stability of the patella.19,129,159,187-190 In cases of recurrent instability of the patella, the MPFL may be torn, stretched, or healed in an elongated position, allowing excessive translation and subsequent episodes of subluxation or dislocation. Instrumented physical examination and stress radiographs have demonstrated this increased lateral laxity compared with that in controls and in contralateral knees in patients with patellar ­instability.57,105 Although first used in the operative treatment of acute patellar dislocation, primary repair of the medial retinaculum for recurrent patellar dislocation was suggested by Sargent and Teipner in 1971.191 The technique involved shortening of the retinaculum with suture repair to the roughened bone of the proximal patella in eight patients with recurrent patellar dislocation; there were no recurrent dislocations in the 16-month follow-up period. Although Fithian and Meier described a detailed modification of Sargent’s technique, no other literature exists on

the results of isolated primary repair in recurrent patellar dislocation.192 Since the identification of the MPFL as the primary restraint, several investigators have looked at primary repair of the ligament at its origin on the adductor tubercle19,175,193 and at its insertion on the patella in acute dislocation.191 It is important to point out that if primary repair is the chosen method of treatment, the entire MPFL must be imaged and explored because injury away from the proposed site of repair will liken the possibility of failure. In the early 1900s, a number of techniques were published for reconstruction of the “proximal transverse retinaculum,” including fascial transplantation, transposition of the hamstring tendons, and transposition of the medial half of the patellar tendon.194 In 1924, Gallie published the premise for modern techniques of MPFL reconstruction. The technique used bone tunnels and involved “tethering the patella itself, by strand of tendon or fascia, to the internal condyle of the femur at the point which forms the centre of the circle through the arc of which the patella moves on flexion and extension of the knee.”195 Nearly a century ago, this technique fulfilled all of the modern principles of ligament reconstruction as suggested by Teitge and Torga-Spak: selection of a sufficiently strong and stiff graft, isometric graft placement, correct tension, adequate fixation, and no condylar impingement.196 Since confirmation of the MPFL as the primary restraint in lateral patellar translation in the early 1990s, much attention has been focused on reconstruction of this structure in an attempt to restore stability, as evidenced by more than 50 recent publications in the English literature of various reconstructive techniques. Before modern reconstruction techniques using free grafts, tenodesis procedures as described by Galeazzi were sometimes performed for recurrent patellar dislocation.197 These procedures were often used in children to avoid growth disturbance of the proximal tibia from distal realignment procedures.198-200 Avikainen and coworkers proposed augmentation of medial retinacular plication with tenodesis of the distal adductor magnus tendon to the medial border of the patella in adults; they reported 12 good, but no excellent, results and 1 recurrent dislocation in 14 patients.194 Augmentation is still used by some in cases in which direct repair can be done after acute dislocation.201 Despite favorable results, attention has turned toward more anatomic procedures of MPFL reconstruction for recurrent cases of instability. Soon after the biomechanical reports proving the ­importance of the MPFL were published, techniques of reconstruction for recurrent dislocation began to surface. In 1992, Ellera Gomes proposed a surgical technique using a polyester graft placed in an isometric position. Eightythree percent of the 30 patients showed significant improvement in their symptoms at 39 months’ follow-up.202 Results were significantly worse results in patients with concomitant chondral pathology, but good to excellent results were found in 96% of those without. Nomura and associates also published a technique using a polyester Leeds-Keio artificial ligament covered by a medial retinacular slip.203 At 6 years’ follow-up, 96% of 27 patients had good to excellent results. Mild stiffness and recurrent apprehension were reported as occasional ­complications in both studies.

1570 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Implantation of artificial material is ­controversial, but the histologic changes in these polyester grafts have been well studied; 6 years after implantation, small but regularly aligned collagen fibers were found encasing the polyester, with a near-normal bimodal pattern of collagen size seen at 8 years’ follow-up.204,205 Despite the apparent success of artificial graft materials, soft tissue autograft remains the gold standard for all ligament reconstruction procedures. A number of donor sites have been reported, including patellar tendon,206 quadriceps tendon,96,196,207 adductor magnus tendon,96,196 medial retinaculum,208 and most commonly, hamstring tendon.209-217 Although the tensile strength of all these graft choices far exceed the 208-Newton tensile strength of the native MPFL,218 the graft must not be expected to hold the patella in place once engaged in a normal trochlea. Rather, it should guide the patella into the trochlea in the first 10 to 30 degrees of flexion, allowing bony constraint to stabilize the patella thereafter.139 A number of technical notes have been published on reconstruction of the MPFL, all of which describe slight variations in the method of restoring the tether between the origin and insertion of the native ligament.196,207,209,213,216,219 Methods of fixation include sutures, spiked washers, staples, bone tunnels, bone anchors, and interference screws. In a recent study, Mountney and colleagues showed that compared with suture anchors, blind tunnels, and sutures alone, a through-tunnel technique of graft fixation was the only fixation method statistically equivalent to the strength of the native MPFL.218 The MPFL has a broad, thin structure, and the exact origin and insertion are difficult to define. Nevertheless, some authors believe that isometric placement of the graft in reconstruction is essential for proper function and longevity of the graft as evidenced by some of the early reconstruction techniques.196,202,216 In a recent study of 11 cadaveric knees, Steensen and colleagues showed that the most isometric points in the native MPFL are the inferior portion of its patellar attachment and the superior portion of its femoral attachment.220 In cases in which the native origin and insertion are no longer identifiable after rupture, these points can be estimated at positions 23 mm from the superior pole of the patella and at the most superior aspect of the anterior portion of the medial epicondyle. Variation in the location and length of the graft can greatly alter the compressive forces at the medial aspect of the patellofemoral joint. Elias and Cosgarea showed that a combination of a graft that is 3 mm short and positioned 5 mm proximal on the femur results in significantly higher compressive forces on the medial patellar cartilage from 30 to 90 degrees of flexion.221 The clinical significance of these findings is unknown. In an in vitro, head-to-head study, Ostermeier and colleagues examined MPFL loading and stabilization of the patella after medial tibial tuberosity transfer versus ligament reconstruction in a MPFL-deficient knee under physiologic loading.222 Maximal loading of the MPFL occurred in full extension, when bony constraint is absent. Despite its traditional use in the treatment of patellar dislocations, tibial tuberosity transfer showed no significant relief of MPFL loading and did not stabilize the patella during lateral loading. MPFL reconstruction alone did restore ­stability to the

patella under load; ligament load was reduced, likely owing to difference in cross-sectional area between the native MPFL and the semitendinosis graft. Although no randomized, prospective trials exist on the outcome of isolated MPFL reconstruction in vivo, a number of retrospective studies have shown very promising results at midterm follow-up. A summary of these results is presented in Table 22C-8. Good to excellent results in these reports of isolated MPFL reconstruction range from 80% to 96%, with almost nonexistent cases of recurrent dislocation. Reports of MPFL reconstruction in combination with realignment procedures have also been published. Mikashima and colleagues retrospectively reviewed 40 patients treated with an Elmslie-Trillat procedure with or without concomitant MPFL reconstruction. There was no residual apprehension in cases with ligament reconstruction, compared with 30% in those without at 2-year followup without a significant difference in Kujala scores.223 MPFL reconstruction has also been used to treat offending causes of recurrent dislocations. Steiner and colleagues retrospectively reviewed 34 cases of MPFL reconstruction in patients with isolated trochlear dysplasia and recurrent patellar instability; those with other risk factors, including patella alta, large Q angle, and rotational deformities, were excluded.96 Good to excellent results were found in 91% of patients at 5 years’ follow-up, and knee scores were comparable to results of other studies of MPFL reconstruction in patients without dysplasia (see Table 22C2-8). Schöttle published a case report of MPFL reconstruction in a patient with rotational deformities of the femur and tibia and showed a correction in the position of the patella and no recurrent dislocations in a 10-month followup.224 Although results of these studies are promising, care should be taken to correct the offending pathology first and foremost.

Arthroscopic Treatment In the current era of minimally invasive surgery, some authors have attempted fully arthroscopic stabilization for recurrent dislocation of the patella. Procedures that have been advertised as minimally invasive include those using mini open medial capsular reefing in combination with arthroscopic lateral release225,226 and arthroscopically assisted procedures.227 However, only a few studies have reported fully arthroscopic stabilization of the patella. In 1986, Yamamoto first described a completely arthroscopic technique for repair of the medial retinaculum in acute patellar dislocations,228 but Henry and Pflum were the first to report an arthroscopic patellar realignment and stabilization for recurrent dislocations and subluxations.229 Their technique involved plication of the medial retinaculum using spinal needles to pass sutures, which were tied percutaneously. They experienced no recurrent dislocation over a 6-year period in an unspecified number of patients. Halbrecht described a similar technique with intra-articular, rather than percutaneous, knot tying with 2-year follow-up in 29 subjects; 93% subjectively felt significantly better, and the mean Lysholm score improved from 42 to 79.230 Objective radiographic measures showed significant improvement in congruence angle, lateral patellofemoral angle, and lateral translation. Hašpl and

Patella 1571

  TABLE 22C2-8  Results of Medial Patellofemoral Ligament Reconstruction Knees

Graft Type

Lateral Retinacular Release

Follow-up (yr)

Results

6

Hamstring

−

7.4

Deie et al (2005)233

46

Hamstring

NR*

5.0/10.0

Drez et al (2001)211

15

Hamstring

+

2.6

Ellera Gomes (1992)202

30

Polyester

+

3.3

Ellera Gomes et al (2004)212

16

Hamstring

+

5.0

Fernandez et al (2005)214

30

Hamstring

±

3.2

Mikishima et al (2006)215

24

Hamstring

NR*

2.0

Nomura et al (2000)203

27

Polyester

±

5.9

Nomura et al (2005)201

5

Retinaculum

±

5.9

Nomura & Inoue (2006)234

12

Hamstring

±

4.2

Schottle et al (2005)217

15

Hamstring

NR*

4.0

Steiner et al (2006)96

34

Adductor, quadriceps

−

5.5

Redislocations: 0/6 Subluxation: 0/6 Apprehension: 2/6 Kujala mean: 96.3 Redislocation: 0/46 Subluxation: 4/46 Apprehension: 4/46 Kujala mean: >90 Redislocation: 0/15 Subluxation: 1/15 Apprehension: 0/15 Kujala mean: 88.6 Fulkerson mean: 93.0 Tegner mean: 6.7 Good to excellent: 13/15 Redislocation: 0/30 Subluxation: 1/30 Apprehension: 0/30 Redislocation: 0/16 Subluxation: 0/16 Apprehension: 1/16 Good to excellent: 15/16 Redislocation: 0/30 Subluxation: 0/30 Apprehension: 0/30 Good to excellent: 29/30 Redislocation: 0/24 Subluxation: 0/24 Apprehension: 1/24 Kujala mean: 95.2 Redislocation: 1/27 Subluxation: 1/27 Apprehension: 2/27 Good to excellent: 26/27 Redislocation: 0/5 Subluxation: 4/5 Apprehension: 1/5 Kujala mean: 97.6 Good to excellent: 4/5 Redislocation: 0/12 Subluxation: 0/12 Apprehension: 0/12 Good to excellent: 10/12 Kujala mean: 96.0 Redislocation: NR Subluxation: NR Apprehension: 3/15 Kujala mean: 85.7 Good to excellent: 13/15 Redislocation: 0/34 Subluxation: 0/34 Apprehension: NR Kujala mean: 90.7 Lysholm mean: 92.1 Tegner mean: 5.1 Good to excellent: 31/34

Study Deie et al

(2003)210

*Not registered.

colleagues ­ subsequently described a technique in which the medial retinaculum was reefed from a working cannula that remained extra-articular on the medial side of the knee.231 At 13 months’ follow-up, 17 patients with a history of patellar instability had experienced no recurrence.

Ali recently reported a 4-year follow-up of 38 knees treated by all arthroscopic lateral release with medial plication. All the aforementioned arthroscopic techniques included a concomitant lateral retinacular release, the risks of which have been previously discussed.

1572 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Treatment Treatment of chronic patellar instability requires a thorough history, physical examination, radiographic and MRI examination, and differentiation between instability and disability. Regardless of whether disability or instability is the predominant issue, care must be undertaken to educate the patient regarding expectations and outcomes. An individualized treatment plan should be developed for each patient. In other words, care must be taken not to treat each patient the same. Rehabilitation is chosen both to strengthen and increase endurance in the operative candidate or for treatment in the patient who has never had rehabilitation or who cannot comply with operative treatment. Operative treatment is the mainstay of treatment for the symptomatic recurrent patellar instability patient. Each patient undergoes an examination under anesthesia and diagnostic arthroscopy as part of the surgical procedure. It is imperative to rule out other forms of instability, such as anterior cruciate instability. In addition, chondral or osteochondral loose bodies or changes on the patella or trochlear groove are important consideration in the algorithm of treatment. If an associated meniscal injury is encountered, it should be recognized and treated as well. Patellofemoral tracking can be assessed as well. Based on the discussion in the previous sections, decisions are made regarding proximal and distal realignment, reconstruction of the MPFL, and lateral retinaculum lengthening or release.

C

r i t i c a l

P

o i n t s

bony congruity of the patellofemoral joint is the most significant contributor to patellar stability. l The medial patellofemoral ligament is the primary soft tissue restraint to lateral patellar dislocation, followed by the patellotibial ligament.

l Patella alta, trochlear dysplasia, increased patellar tilt, and increased tibial tuberosity–to–trochlear groove distance are radiographic findings that are associated with objective patellar instability. l Instrumented stress radiographs are simple to perform and provide objective radiographic evaluation of patellar mobility. l A course of nonoperative treatment with physical therapy should be attempted before surgical treatment. l Surgical treatment should be directed at the offending pathology determined through history, physical examination, and imaging studies. l Pain, recurrence, and patellofemoral arthrosis are possible complications of any procedure following multiple patellar dislocations.

S U G G E S T E D

R E A D I N G S

Amis AA, Firer P, Mountney J, et al: Anatomy and biomechanics of the medial patellofemoral ligament. Knee 10(3):215-220, 2003. Dejour H, Walch G, Nove-Josserand L, Guier C: Factors of patellar instability: An anatomic radiographic study. Knee Surg Sports Traumatol Arthrosc 2(1):19-26, 1994. Fulkerson JP, Gossling HR: Anatomy of the knee joint lateral retinaculum. Clin Orthop 153:183-188, 1980. Hughston JC, Deese M: Medial subluxation of the patella as a complication of lateral retinacular release. Am J Sports Med 16(4):383-388, 1988. Kolowich PA, Paulos LE, Rosenberg TD, Farnsworth S: Lateral release of the patella: Indications and contraindications. Am J Sports Med 18(4):359-365, 1990. Kuroda R, Kambic H, Valdevit A, Andrish J: Distribution of patellofemoral joint pressures after femoral trochlear osteotomy. Knee Surg Sports Traumatol Arthrosc 10(1):33-37, 2002. Steiner TM, Torga-Spak R, Teitge RA: Medial patellofemoral ligament reconstruction in patients with lateral patellar instability and trochlear dysplasia. Am J Sports Med 34(8):1254-1261, 2006.

l The

R eferences Please see www.expertconsult.com

S e c t i o n

D

Patellar Fractures Agbecko Ocloo and Richard D. Parker

Patella fractures are relatively uncommon injuries constituting about 1% of all skeletal injuries.1 They are usually the result of significant trauma and less commonly through participating in sporting events. The mechanism of injury is either a direct blow to the patella or an indirect force applied to the patella through the extensor mechanism. The various fracture patterns are usually representative

of the injury mechanism. These fractures can be broadly divided into displaced and nondisplaced fractures. Both nonoperative and operative management may be employed with good outcomes depending on the age and activity level of the patient, the fracture pattern, and the amount of displacement. Multiple techniques have been described for the surgical treatment of displaced fractures.

Patella 1573

The salient points of surgical treatment include anatomic reduction and rigid internal fixation that will allow early knee motion and rehabilitation.

SG

ANATOMY AND BIOMECHANICS The patella is the largest sesamoid bone in the body. It ossifies from a single center which usually makes its appearance in the second or third year, but may be delayed until the sixth year. More rarely, the bone is developed by two centers, placed side by side. Ossification is completed about the age of puberty. Incomplete ossification results in a bipartite patella. The patella is a flat triangular bone with a base or superior border which is thick and gives attachment to the quadriceps muscle. The medial and lateral borders are thinner and converge below. They give attachments to the vastus lateralis and medialis. The apex is pointed and gives attachment to the patella ligament. The posterior articular surface of the patella is divided by a vertical ridge and then again into thirds by two horizontal ridges. The lateral facet is larger than the medial. The lower facets articulate first with the trochlear groove in early flexion followed by the middle and then the upper facets. In full flexion, the most medial aspect of the patellar articular surface, designated the crescentic or odd facet, is the main contact point. The patella articulates with the trochlear groove of the distal femur and undergoes approximately 7 cm of excursion from extension to full flexion. The patellofemoral contact is initiated at about 20 degrees of flexion. The forces generated across the patellofemoral joint are tremendous, ranging from half of body weight for normal walking to nearly eight times body weight for jumping from a small height.2 Because of these forces, the articular surface of the patella is the thickest in the body. The average thickness is more than 1 cm. The blood supply to the patella is from a vascular anastomotic ring lying in the thin layer of loose connective tissue covering the rectus expansion (Fig. 22D-1). The main vessels contributing to this anastomotic ring are the supreme genicular, the medial superior genicular, medial inferior genicular, lateral superior genicular arteries and the anterior tibial recurrent artery. Nutrient vessels pass obliquely into the anterior surface of the patella from this complex network. Disruption of this supply by injury and subsequent surgical dissection can result in avascular necrosis. Rates of 3.5% to 24% have been reported after patellar fracture.3

CLASSIFICATION Fracture patterns of the patella are classified by their configuration and are usually consistent with the mechanism of injury. Indirect forces usually produce a nondisplaced or minimally displaced transverse fracture of the central or distal third or more uncommonly, a vertical fracture. Blunt injury to the patella either from a direct blow or from a fall onto the flexed knee, produces a comminuted stellate patella fracture. However, vertical or transverse patterns can also be produced. Do not forget that stress fractures can occur, too. Fractures are further designated displaced or ­nondisplaced,

LSG

MSG

LIG

MIG

ATR

Figure 22D-1   Schematic arterial blood supply to the patella. LSG, lateral superior geniculate; SG, superior geniculate; MSG, middle superior geniculate; MIG, middle inferior geniculate; ATR, anterior tibial recurrent; LIG, lateral inferior geniculate.

which is defined as more than 1 to 2 mm of articular separation or 3 to 4 mm of fracture separation.

EVALUATION Clinical Presentation and History Patellar fractures result either from an indirect force applied through the strong contraction of the quadriceps mechanism against a flexed knee or from a direct force, such as a fall onto a flexed knee or blunt trauma to the anterior patella. The subcutaneous location of the patella places it at risk for injury from direct impact. Traumatic separation of a bipartite patella can occur.4 This group of patients has a dull ache or pain in the knee prior to the traumatic episode. Reports of patella fractures after anterior cruciate ligament (ACL) reconstruction using the central third bonepatella tendon-bone autograft is being reported in the ­literature.5 Patients with patella fractures usually present with a painful, swollen knee after either direct trauma to the knee or a fall when an attempt was made to stop suddenly or “catch oneself.” Weight bearing is painful and depending on the competence of the medial and lateral retinacula, the patient may be able to extend the knee.

1574 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

TREATMENT OPTIONS Nonoperative Initial treatment of patellar fractures should include splinting in extension and application of ice. The indications for nonoperative treatment are:

Figure 22D-2   Significantly displaced patellar fracture with disruption of the extensor retinaculum.

Physical Examination and Testing Localized tenderness and a hemarthrosis are typically present. It is important to examine the skin around the knee for abrasions and lacerations. In their presence intra-articular communications with the skin wound need to be ruled out. Open fractures will require immediate surgical débridement. With a displaced fracture, a gap between the two fracture fragments can be palpated. The integrity of the extensor mechanism needs to be evaluated. On occasion, a patient may be limited by pain instead of extensor mechanism disruption. For further evaluation of this, an aspiration of the hemarthrosis under sterile conditions followed by injection of intra-articular lidocaine can be performed.

Imaging Most patellar fractures can easily be diagnosed with standard radiographs (Fig. 22D-2). Radiographic evaluation includes anteroposterior, lateral and Merchant views. The anteroposterior view is used to assess for fragmentation, but visualization can be difficult owing to overlap of the distal femur. The lateral view best reveals the degree of fragmenting of the fracture or separation of fragments. Some vertical and osteochondral fractures are best seen on tangential or Merchant views. Osteochondral, marginal, and especially chondral injuries can more accurately be evaluated with magnetic resonance imaging. Other diagnostic studies, including computed tomography, bone scans, and conventional tomography, have been described but provide little additional clinical value. Bone scans have been reported to be helpful in evaluating patella stress fractures in athletes.6 A bipartite patella can be confused with an acute fracture and must be considered when radiographs reveal a small fragment separated from the main portion of the patella. In most cases these secondary ossification centers are located in the superolateral pole. The separation is minimal and the borders are usually smooth. Contralateral radiographs are of assistance if the variant is also present.

1. Undisplaced fractures with intact articular surface. There should be minimal displacement of fragments (<2 to 3 mm) and minimal displacement of the articular surface (<1 to 2 mm) (Fig. 22D-3). 2. Preserved extensor mechanism. The retinacula on either side of the patella should not be torn as evidenced by the patient’s ability to maintain the extended knee against gravity. 3. Elderly patients who are poor surgical candidates and debilitated patients with poor bone quality. Nonoperative treatment consists of casting in extension for up to 6 weeks. After initial splinting, the fracture is immobilized in a cylinder cast from the ankle to the groin. Weight bearing is allowed as tolerated. Quadriceps exercises are encouraged during casting to limit atrophy. Brostroum1 reported 90% good to excellent results in patients treated nonoperatively. A loss of terminal extension or persistent extension lag is not uncommon, but this is rarely a problem in the elderly population. A study on nonoperative treatment of patella fractures displaced by more than 1 cm in elderly patients who were not surgical candidates revealed satisfactory outcomes in 9 of 12 patients.7

Operative When operative treatment is considered, the goal should be an anatomic reduction and rigid internal fixation that allows early motion. The technique of patella fixation is a topic of considerable controversy. Anatomic restoration of the fragments with stabilization by screws or wires gives the strongest construct. Skin integrity is an important consideration. Abrasions and lacerations should be closely scrutinized for joint communication. Open fractures should be treated with expeditious irrigation and débridement followed by operative fixation if indicated. Timely surgical intervention is recommended if skin abrasions or superficial lacerations are present. Once the lacerations become superficially infected, surgical repair should be delayed 7 to 10 days. Postoperative swelling can also present problems because of the lack of soft tissue coverage over the patella. Skin and subcutaneous necrosis due to swelling can result in a need for skin grafting or flap coverage. Meticulous skin-handling techniques and postoperative elevation should be employed to help prevent these complications. Both transverse and longitudinal incisions have been used for exposure and repair of the patella and retinacula. A midline longitudinal incision allows excellent exposure and it does not compromise the skin if future knee operations are required. The retinaculum injury usually allows excellent visualization of the articular surface and subsequent reduction.

Patella 1575

A

B

Figure 22D-3   A, Nondisplaced fracture amenable to nonoperative treatment. B, Healed at 6 weeks.

The lateral retinaculum can be incised longitudinally to allow for better exposure if necessary.8 Digital palpation of the chondral surface during reduction and subsequent to fixation will ensure anatomic articular surface alignment. Stabilization of the bony fragments can be accomplished by cerclage wire, tension band techniques, or interfragmentary lag screw fixation. Berger first described circumferential cerclage wiring to reduce the fragments. Indirect reduction is obtained as compression is applied to the wire. Muller and associates9 as part of the AO group advocated the tension band wiring technique whereby two Kirschner wires are placed longitudinally through the patella. Around these wires, a heavy-gauge wire is wrapped in a figure-eight fashion and tightened over the anterior surface of the patella. The Kirschner wires provide an anchor against which the figure-eight wire is tensioned. Biomechanical evaluation has found that this configuration produces a compression effect on the articular surface with knee flexion.10 Lag screw fixation of larger fragments can be used alone or in conjunction with tension band techniques. Two large fragments can be rigidly secured with parallel lag screws, or several smaller fragments can be stabilized with lag screws to allow tension banding of the remaining construct. Another technique reported by Berg11 incorporates the use of cannulated screws through which the wire is passed and then secured in a figure-eight fashion. The advantage of this technique is the combination of the compression of the lag screws with the tension band technique while maintaining a low profile. Because of the subcutaneous position of the patella, hardware placed to stabilize fractures often requires subsequent removal after successful stabilization. To avoid hardware complications, heavy-gauge suture has been advocated instead of the wire. Techniques of tension band using braided heavy polyester sutures has shown promising results comparable to using wire.12 Alternative techniques have been described to obtain stable fixation. The use of biodegradable wires and screw have not compared favorably with conventional ­ fixation.

If comparable fixation can be obtained, the need to remove hardware can be obviated. External patellar fixation was originally described by Malgaigne but abandoned. Liang and Wu13 reported good or excellent results in 26 of 27 patients treated with an external compression fixator attached by transverse percutaneous pins. No osteomyelitis developed, and 80% of the patients regained knee motion equal to that on the uninjured side. Fractures involving the distal pole of the patella can be treated with internal fixation of the fragment or by distal pole excision and reattachment of the patella ligament. Matjaz and associates compared osteosynthesis with a basket plate against distal pole resection.14 Internal fixation allowed for early mobilization, weight bearing and overall better results than with distal pole excision. When the fracture is multifragmented, all attempts should be made to conserve the patella. After patellectomy, extensor lag, weakness, malalignmnent, and restricted motion are common problems. Quadriceps strength is reduced by 20% to 60%. Tibiofemoral forces can increase up to 250% leading to early degeneration of the tibiofemoral joint.15 Primary reconstruction of the patella is recommended in multifragmentary fractures. If the outcome is poor, patellectomy should be considered early since it gives a better outcome than if delayed.16

Authors’ Preferred Method The authors’ preferred treatment is based on the configuration of the fracture patterns. If at all possible we prefer to save as much of the patella as possible. We prefer rigid strong fixation and favor tension band wiring whenever possible. We are aggressive in terms of surgical fixation due to the desire for early motion and rehabilitation.

1576 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

POSTOPERATIVE PRESCRIPTION AND POTENTIAL COMPLICATIONS For undisplaced fractures, the knee is immobilized in a locked extension brace and the patient can partial weight bear on crutches for six weeks. Quadriceps exercise is continued during this period, after which the knee is mobilized to regain its range of motion. During internal fixation, the knee is flexed to determine the degree of flexion that does not potentially disrupt the construct. A hinged brace is applied and the patient weight bears as tolerable. The knee is mobilized through the predetermined range twice a day. Both quadriceps and hamstring muscles are continuously strengthened throughout the recovery period. Significant complications may occur after fractures of the patella. Patella baja can occur especially after prolonged immobilization. Early range of motion of the knee reduces its incidence.17 The occurrence of arthrofibrosis can also be reduced with early mobilization. Skin necrosis is an uncommon but disastrous complication and it is better prevented. Careful examination of the skin is required, and when in doubt of viability, to delay surgery for up to ten days is prudent. Skin necrosis is managed by covering the defect with a local medial gastrocnemius flap or a free muscle flap. Surgical site infections may occur in open injuries or as a result of skin necrosis. Wound débridement with appropriate soft tissue coverage and intravenous antibiotics often proves successful in eradicating the infection. A stable construct in the milieu of an infection should be left in situ. If, however, the fixation is unstable, all hardware should be removed until the infection is eradicated. Hardware failure is most often consequent to poor surgical technique. Failure to achieve an anatomic reduction is setting the stage for a non-union and subsequent implant breakage.

Figure 22D-4   Lateral radiography of the knee showing a displaced sleeve fracture.

motorcycle crashes.20 Sleeve fractures are the most common fracture pattern followed by transverse fractures.20 Clinical presentation and physical signs are similar to the adult patient. Because the distal bony fragment in sleeve fractures is often so small, the correct diagnosis may be missed or delayed (Fig. 22D-4). In a sleeve fracture of the patella, the cartilaginous injury is much more extensive than the bony avulsion shown by radiography. A patellar sleeve fracture in children differs from fracture of the lower pole of the patella in adults. Initial radiographs maybe misleading, and the fracture is not readily apparent on radiographs. Magnetic resonance imaging is very sensitive in identifying the cartilaginous injury (Fig. 22D-5). The treatment protocol for children is the same as that for adults. Restoration of the extensor mechanism and realignment of the articular surface are essential.21

CRITERIA FOR RETURN TO PLAY Most athletes are able to return to play after the fracture has consolidated and regained about 90% of quadriceps strength. Most will be able to return within 6 to 9 months. Despite improved surgical techniques and rehabilitation protocols, athletes seem to have a difficult time returning to their preinjury level of competition.18 It is much difficult if parts of the patella have been excised.

PATELLA FRACTURE IN CHILDREN Fractures of the patella are rare in children and account for less than 1% of all pediatric fractures.19 The mechanism of injury is similar to the adult population. Most fractures occur from indirect forces produced by violent and sudden contraction of the quadriceps against resistance as occurs in sporting endeavors. Ray and Hendrix reviewed 185 cases and found the mechanism of injury was predominantly motor vehicle and

Figure 22D-5   T2-weighted sagittal plane image confirms a sleeve fracture of the distal fragment with a cartilage area surrounding it.

Patella 1577

Some authors recommend nonoperative treatment for undisplaced fractures, but malalignment22 and heterotopic ossification have been reported. Open surgical reduction should be considered first for fractures of the patella in children, especially for sleeve fractures. The retinaculum, if torn, should be repaired after fracture fixation, and the articular surface should be realigned. The prognosis is generally good, and growth disturbance are very rare. All attempts are made to save the patella, even in the case of displaced, comminuted ­fractures. The authors take a very aggressive approach to restoring articular surface anatomy in young people and fixate with rigid fixation to allow for early motion and rehabilitation.

S U G G E S T E D

R E A D I N G S

Berg EE: Open reduction and internal fixation of displaced transverse patella fractures with figure-eight wiring through parallel cannulated compression screws. J Orthop Trauma 11:573-576, 1997. Houghton GR: Ackroyd CE: Sleeve fractures of the patella in children: A report of three cases. J Bone Joint Surg Br 61:165-168, 1979. Hughes SA, Stott PM, Hearnden AJ: A new and effective tension-band braided polyester suture technique for transverse patellar fracture fixation. Injury 38: 212-222, 2007. Rocket JF, Freeman BL 3d: Stress fracture of the patella: Confirmation by triple phase bone imaging. Clin Nucl Med 15:873-875, 1990. Scapinelli R: Blood supply of the human patella: Its relation to ischemic necrosis after fracture. J Bone Joint Surg Br 49:563-570, 1969.

R eferences Please see www.expertconsult.com

C

r i t i c a l

P

o i n t s

l The patella is the largest sesamoid bone in the body. l The patella has a robust blood supply from the superior, middle, and inferior geniculate arteries anastomosing and entering anteriorly. l Fractures occur from injuries indirectly or directly. Direct injuries are likely to be comminuted. l Open reduction internal fixation (ORIF) is reserved for unstable displaced fractures. l The goals of ORIF of patellar fractures are anatomic reduction, secure fixation, and early motion. l Patellar fractures are rare in children but require operative treatment if displaced.

C H A P TER�  

 �� 23

Knee S e c t i o n

A

Relevant Biomechanics of the Knee Bruce D. Beynnon, Robert J. Johnson, and Lauren Brown

The knee joint is the largest and most complex joint in the human body. The joint capsule and ligaments, which provide structural stability to the knee, are particularly vulnerable to injury by large moments that can be created through the forces acting along the long lever arms of the lower limb. Thus, it is not surprising that the knee is one of the most frequently injured joints. An injury to the knee, such as disruption of the anterior cruciate ligament (ACL), can result in an extensive disability because this injury may alter normal knee kinematics and therefore locomotion. An extensive background in biomechanics or mathematics is not required to understand the fundamental mechanical principles governing the knee joint. Knowledge of knee biomechanics provides an essential framework for understanding the consequences of injury and joint disorders; it aids in the intelligent planning of surgical procedures, serves as the basis for developing objective rehabilitation programs, and describes the effects of different types of orthoses on the knee joint. The knee joint comprises three independent articulations, one between the patella and femur and the remaining two between the lateral and medial tibial and femoral condyles. The patellofemoral articulation consists of the patella, which has a multifaceted dorsal surface that articulates with the femoral trochlear groove. The tibiofemoral articulations consist of femoral condyles with saddle-shaped tibial condyles and interposing menisci. The posterior aspect of the femoral condyles is spherical, whereas the anterior aspect of the femoral condyles is more flat. Thus, in extension, the flat portion of the femoral condyles is in contact with the tibia, and in flexion, the spherical portion of the femoral condyles is in contact with the tibia. To the untrained observer, the knee joint may appear to function as a simple pinned hinge (ginglymus), with flexion-extension rotation the only apparent motion between the femur and tibia. The motion characteristics of the knee joint are extremely complex, however, requiring a full 6 degrees of freedom (3 translations and 3 rotations) to completely describe the coupled, or simultaneous, joint motions (Fig. 23A-1). An example of coupled motion is

demonstrated with flexion rotation of the knee from the extended position. With this rotation, there is a coupled posterior movement of the femoral contact regions on the tibial surface in the sagittal plane and an internal rotation of the tibia relative to the femur in the transverse plane. By use of the Eulerian-based coordinate system described by Hefzy and Grood,1 the translations and rotations can be described in anatomically referenced directions (see Fig. 23A-1). Although many different types of coordinate systems have been used to describe three-dimensional knee motion, this system is appealing because it allows joint

Adduction Anteriorposterior Abduction Mediallateral Tibial Axial

External

Flexion

Extension

Joint translation vector

Internal

Figure 23A-1   Coordinate system for knee joint rotations and translations. Flexion-extension rotation is about the fixed femoral axis. Internal-external rotation is about a fixed tibial axis. Abduction-adduction is about an axis that is perpendicular to the femoral and tibial axes. The joint translations occur along each of the three coordinate axes. (From Hefzy MS, Grood ES: Review of knee models. Appl Mech Rev 41:1-13, 1988.)

1579

�rthopaedic ����������� S �ports ������ � Medicine ������� 1580 DeLee & Drez’s� O

DESCRIPTION OF BIOMECHANICAL TECHNIQUES The American Society of Biomechanics has defined the term biomechanics as the study of the structure and function of biologic systems using the methods of mechanics. Specifically for the knee joint, this involves modeling and experimental investigation techniques. Each technique is based on fundamental principles and terminology that require definition. These may be grouped into three classes of study: 1. Models of the knee joint 2. Experimental study of the entire knee by the following techniques: a. Flexibility approach b. Stiffness approach 3. Experimental study of an individual ligament by the following techniques: a. Anatomic observation b. Force measurement c. Strain measurement The following discussion provides an overview of the different study techniques.

MODELING OF THE KNEE JOINT Hefzy and Grood1 have presented a detailed review of different modeling techniques and applications. Perhaps the most commonly described physical knee model is the crossed four-bar linkage called the cruciate linkage.5,8-11 This approach has been used to study the interaction of the cruciates with the tibiofemoral joint (Fig. 23A-2). The model consists of two crossed rods representing the cruciates, and two connecting bars representing the tibial and femoral attachments of these ligaments. This approach has been used to describe the shape of both the tibial and femoral condyles, the path of the instantaneous center of

Fe m ur

ur Fem L AC Tibia

IC

IC PC

L

L AC

L PC

rotation to be expressed in terms familiar to the clinician. Grood and Noyes2 have applied the three-dimensional coordinate system to the interpretation of various clinical examination techniques and have developed a “bumper model” of the knee joint. This model is useful in describing the soft tissue restraints to anteroposterior translation and internal-external rotation of the knee joint. In addition, the model can be applied to demonstrate the types of tibiofemoral subluxations that may result when different soft tissue structures are disrupted. Application of this approach may aid in the examination of injuries to the knee ligaments and capsular structures. This section assumes a working knowledge of the biomechanical terms essential to the description of knee function. For an introduction to basic knee biomechanics, the reader is encouraged to review the work of Frankel,3 Frankel and Burstein,4 and Mow and Hayes,5 along with the definition of biomechanical terms as they apply to the knee presented by Noyes and coworkers6 and Bonnarens and Drez.7 A review of experimental and model studies of the tibiofemoral and the patellofemoral joints with associated contact morphometry studies is presented in this section.

Tibia

Figure 23A-2   The four-bar cruciate linkage model. The model includes two crossed bars, which represent the anterior and posterior cruciate ligaments (ACL, PCL). The remaining two bars represent the tibial and femoral attachments of the ligaments. IC, instantaneous center of joint rotation. (From Hefzy MS, Grood ES: Review of knee models. Appl Mech Rev 41:1-13, 1988.)

knee joint rotation, and the posterior migration of the ­tibiofemoral contact point that occurs with knee flexion. The four-bar approach is based on rigid interconnecting cruciate linkages that are not allowed to elongate. Because the cruciates elongate and twist during normal joint articulation,12-14 this technique may be inadequate for modeling the detailed interaction of the cruciates with the tibiofemoral joint. Hollister and colleagues15 and Churchill and associates16 described the knee in passive and active flexion, respectively, in three dimensions using a compound-hinge model. Flexion is described about the transepicondylar axis, and internal-external rotation is described about an axis parallel to the longitudinal axis of the tibia (Fig. 23A-3). The model accounts for three-dimensional kinematics while allowing the axes to remain fixed in bone. Previous models use the concept of an instantaneous center of rotation, which accounts for femoral rollback by using a different center of rotation with each flexion angle.17-19 Crowninshield and colleagues20 published one of the first theoretical investigations of the knee by means of a mathematical model. The cruciate, collateral, and capsular ligaments were represented by 13 elements. Input parameters included both in vitro and in vivo measurements of the attachment sites and dimensions of the knee ligaments. The effect of a given ligament was then investigated by eliminating an element in the model and comparing this with an actual test in which the ligament was cut. Wismans and associates21 developed a three-­dimensional tibiofemoral joint model that incorporated the nonlinear characteristics of tibiofemoral geometry and ligamentous material properties. This model was applied to investigate the biomechanics of the major knee ligaments, tibiofemoral compressive loading, and joint load-displacement response, and it determined the relative position of the femur with respect to the tibia as a function of different applied jointloading conditions. In later studies, Blankevoort and ­colleagues applied similar techniques to study the effects of

Knee 1581 z

y

Transepicondylar axis

x z

x

y

Tibial shaft axis

Figure 23A-3   The compound-hinge model describes the three-dimensional kinematics of the knee using two axes: the transepicondylar axis and an axis parallel to the longitudinal axis of the tibia.

different ACL replacement positions on tibiofemoral joint biomechanics,22 to investigate the recruitment of the knee ligaments,23 to study tibiofemoral contact in three dimensions,24 and to examine ligament-bone interaction.25 Coughlin and coworkers26 characterized the 6 degrees of freedom of the tibiofemoral joint in terms of 2 rotations about the femoral epicondylar (FE) axis, and an axis parallel to the anatomic tibial axis. The FE axis is defined as the line passing through the spherical centers of the medial and lateral condyles. In the midsagittal plane, the patella follows a circular arc with a constant radius of curvature. These axes remain nearly perpendicular with the patella tracking along a circular arc that can be described relative to the FE axis during flexion from 0 to 90 degrees. The authors suggest this model has clinical application in identifying the FE axis intraoperatively.26

EXPERIMENTAL STUDIES OF THE ENTIRE KNEE Flexibility Approach This approach involves observing or measuring the displacement due to an applied joint load; a ligament is then cut, and the procedure is repeated. The relative difference

in displacement is then used to establish the importance of that ligament. This method is analogous to clinical laxity examinations (i.e., Lachman’s test), in which the clinician applies a load and estimates the resulting joint displacement. This technique is useful in evaluating the sensitivity of knee laxity tests to injuries created in human cadavers. A drawback of this technique is that the difference in the behavior of a knee joint before and after excision of a ligament does not necessarily indicate that the cutting of the ligament was responsible.

Stiffness Approach This approach establishes the role of a particular ligament by applying a predetermined displacement while simultaneously measuring the load applied to the knee joint, cutting the structure under investigation, and repeating the test while documenting the decrease in load that results. This methodology is useful for determining what motions are resisted by each ligament and the relative importance of the structure. A comparison between methodologies reveals that the stiffness approach produces results independent of the order in which the ligaments are sectioned and therefore is a more direct approach, whereas the results from the flexibility approach are governed by the order of ligament cutting. Thus, the outcomes of these latter studies may be difficult to interpret.

EXPERIMENTAL STUDIES OF INDIVIDUAL LIGAMENTS Anatomic Observation The clinical approach is to infer a ligament’s function from its anatomy. The important information that can be obtained is anatomic location of the ligament attachment points, orientations of major fiber bundles, and size (length and crosssectional area). Important early observations were made by Hughston and coworkers,27,28 Slocum and colleagues,29 and Kennedy and associates.30 Careful dissection techniques have been employed by a number of workers to improve our qualitative understanding of the knee ligaments.31-34 More recently, the cross-sectional areas of the ACL, the posterior cruciate ligament (PCL), and the meniscofemoral ligaments were reported at four flexion angles.35

Force Measurement of Ligaments Measurement of ligament or tendon force continues to be one of the biggest challenges in orthopaedic biomechanics. To meet this challenge, Salmons36 introduced the buckle transducer. The buckle transducer is a structure containing a beam over which the ligament is looped to create bending of the buckle frame. The beam is attached separately, and therefore transverse cutting of the ligament is not required. Lengthwise incisions are required, however, disrupting many ligament fibers and altering the normal function of the ligament by dissociating one group of fiber bundles from another.32,37,38 Salmons36 and other ­investigators39-41

�rthopaedic ����������� S �ports ������ � Medicine ������� 1582 DeLee & Drez’s� O

have applied this technique to various knee ligaments. This approach is limited to in vitro applications in which a knee ligament is instrumented with the transducer and then the knee joint is loaded, and the output from the buckle transducer is then directly recorded. After the tissue is tested in situ, it is dissected free from one of its bony attachments so that a known load can be applied to the tissue-buckle system and a calibration of the gauge–soft tissue system can be made. By use of this approach, the load can be determined for the tissue in situ. This technique cannot be used for in vivo testing. Markolf and colleagues42-45 have presented a direct approach to the measurement of ACL and PCL force in cadaveric knees. This technique involves isolating the tibial attachment of the cruciate ligament by creating a bone plug with a coring cutter and attaching a load sensor to the external portion of the bone plug. This approach facilitates the measurement of resultant cruciate ligament force. The authors demonstrated that passive flexion-extension motion of the knee from 10 degrees to full flexion does not load the ACL, whereas loading of the quadriceps musculature (simulating active extension of the lower limb against gravity) developed ACL loading between the limits of 0 and 45 degrees. Markolf and coworkers46 demonstrated in later studies using human cadaver knees that hamstring loading has a greater effect on cruciate force than does quadriceps loading. PCL force increases significantly with hamstring loading beyond 30 degrees of flexion. Although direct force measurements of the cruciate ligaments are desirable, the current state-of-the-art force transducers allow the associated errors to fall within relatively large error bounds. Fleming and colleagues47 reported mean errors ranging from 20% to 29% for the arthroscopically implantable force probe (AIFP), if it is calibrated after implantation. They noted that similar limitations have been reported for other force transducers that operate on the same principle, although the errors associated with other arthroscopic force transducers have not been reported. In addition, Fleming and colleagues48 determined that the AIFP output is specimen dependent. When the force transducer was removed and reimplanted into the same location, the results were not repeatable, with errors ranging from 4% to 109%. On reimplantation into another location in the same ACL, the percentage errors ranged from 2% to 203%. These findings highlight a need for a more repeatable transducer that will yield relative measurements of ACL stress in vivo. Because stress is directly related to force or strain, several investigators have chosen to measure strain, rather than force, in the ligaments. Fleming and coworkers49,50 have shown the accuracy associated with strain measurements in the ACL to be on the order of 0.2% and 0.1% with use of the Hall-effect strain transducer and the differential variable reluctance transducer (DVRT), respectively.

Strain Measurement of Ligaments Several investigators have measured ligament displacement, enabling the calculation of strain pattern, to understand the effect of knee joint position and muscle activity on ligament biomechanics.51-56 Most of this work has been carried out in vitro, and the results are conflicting.

Edwards,53 Kennedy,54 Brown,52 Berns,57 Hull,58 and their associates used mercury-filled strain gauges to measure the length of ligaments at various angles of knee flexion. Henning and colleagues59 have constructed a device to measure displacement in the ACL in vivo. Butler and associates60 and Woo and colleagues61 have developed optical techniques for mapping surface strains in various tissues. Butler and coworkers60,62 used high-speed cameras to record the movements of surface markers and measured both mid-substance and insertion-site deformations of soft tissues. These techniques are ideal methods for monitoring surface strains, particularly during highrate tests, but are not useful for out-of-plane movements or for ligaments such as the cruciates that cannot be directly viewed. They also suffer from the theoretical disadvantage that the tissue of interest has to be exposed and therefore is not in a physiologic state. Other workers have calculated strain by measurement of the change of ligament attachment length under various applied joint loadings. For example, Wang and colleagues56 measured the three-dimensional coordinates of pins stuck in a cadaveric joint at the palpated origin and insertion points of the major knee ligaments. They recorded the relationship between torque and angular rotation of the femur relative to the tibia. After excision of certain ligaments, the tests were repeated to determine the contribution of these elements to torsional restraint. In the most extensive and elegant studies, Sidles and associates63 used a three-dimensional digitizer to compute ligament length patterns. In a slightly different approach, Trent and colleagues64 used pins embedded in the ligament attachments and measured the displacement of one pin relative to the other. In addition, in this study they located the instant centers of transverse joint rotation. Warren and coworkers65 also used pins placed at ligament origins but measured displacements with a radiographic technique. The pins or other markers used as locators of ligament origin generally estimate average ligament strain. This technique may produce confusing results owing both to the difficulty in choosing the center of a ligament insertion and to the changes in strain from place to place within a ligament. For example, Covey and colleagues66 demonstrated that the fiber anatomy across the PCL leads to differences in strain measurement among four geographically distinct areas of the ligament. Previous work at the University of  Vermont has focused on the measurement of ACL displacement in the in vitro environment by use of the Hall-effect strain transducer and, more recently, a DVRT, which allow calculation of strain.51,55,67,68 This technique has been applied to the measurement of ACL strain in vivo.12,13,50,67,69-71

LIGAMENT BIOMECHANICS The primary function of the knee ligaments is to stabilize the knee, to control normal kinematics, and to prevent abnormal displacements and rotations that may damage articular surfaces. Ligaments are the most important static stabilizers and are primarily composed of type I collagen, the constituent that provides resistance to a tensile load developed along the length of the ligament, with lesser and varying amounts of elastic and reticulin fibers. ­ Cellular

Knee 1583

e­ lements, ground substance, vascular channels, and nerves are also present. Collagen fibers and their orientation within the tissue are responsible for the primary biomechanical behavior of each of these structures. The fibers of the large distinct ligaments are almost all arranged in parallel bundles, making them ideal for withstanding tensile loads, whereas capsular structures have a less consistent orientation, making them more compliant and not as strong in resisting axial loading. The fibers within ligaments do not act uniformly across a ligament during loading; using excursion filaments implanted in four distinct fiber regions of the PCL, Covey and associates66 demonstrated the differential behavior of the various fiber regions. The ligament insertion sites are designed to reduce the chance of failure by distributing the stresses at the bone­ligament interface in a gradual fashion. This is accomplished by the collagen fibers passing from the ligament into the bone through four distinct zones: (1) ligament substance, (2) fibrocartilaginous matrix, (3) mineralized fibrocartilage, and (4) bone itself.72 Despite the transitions, Noyes and colleagues73 demonstrated that some strain concentration occurs near the ligamentous insertion sites. Later, Sidles and coworkers74 developed an analytic model of the ACL tibial insertion. They demonstrated that for typical ACL insertion geometry, the transverse pressures are similar to the tensile stress along the ligament.

C

C

The knee ligaments can best control motion of the bones relative to each other if the motion takes place along the direction of the ligament fibers. For example, when the knee is loaded in valgus, the medial collateral ligament develops a tensile stress in combination with a compressive force across the lateral compartment of the knee, and a resistance to medial joint opening is provided. Acting alone, ligaments cannot restrain the relative rotation associated with applied torques because the ligament would simply rotate about its bony insertion sites. A second force, usually developed through cartilage-to-cartilage compression, is required. For example, as the knee is loaded with an internal torque, a transverse rotation causes the femoral condyles to ride up the tibial spines (Fig. 23A-4). This combination creates a compressive force across the tibiofemoral contact regions and an oppositely directed tensile force along the cruciate and collateral ligaments. This example may help demonstrate the mechanism by which the ACL interacts with tibiofemoral articular compression to resist an applied internal rotation to the knee joint. The ability of a ligament to resist applied tensile loading may best be described through examination of the load-elongation curve produced during tensile failure testing of an ACL (Fig. 23A-5). As a tensile load is applied, the ligament elongates; the slope of the measured load­displacement relationship represents the stiffness of the ligament. The steeper the slope of this curve, the stiffer the ligament. In the unloaded state, the ligament fibers are under minimal tension, and the collagen fibers have a wavy pattern. As a tensile load is applied, the wavy pattern begins to straighten. Initially, little load is required to elongate the ligament. This is characterized by the relatively flat “toe” region of the curve. The change from the toe to the linear portion of the curve represents the change in stiffness that an examiner perceives during a clinical laxity examination when a ligament’s end point is reached. As the tensile load continues to increase, all the collagen fibers are straightened, and the curve becomes nearly linear. This region of the curve characterizes the elastic deformation of the ligament until the yield point is reached. At this point, there

T

Yield Point

Maximal load

Injury

Load

Physiologic loading

Clinical test Figure 23A-4   Internal rotation of the tibia relative to the femur. The internal rotation causes the femoral condyles to ride up on the tibial spine, producing tension in the cruciate ligaments and a compressive force across the articular surfaces. C, compressive force produced between the tibiofemoral articular surfaces; T, tensile load developed along the anterior cruciate ligament.

0

1

2

3 4 5 6 7 Joint displacement (mm)

8

Figure 23A-5   Load-elongation curve for the tensile failure of the anterior cruciate ligament. (From Cabaud HB: Biomechanics of the anterior cruciate ligament. Clin Orthop 172:26, 1988.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1584 DeLee & Drez’s� O

is a sudden loss in the ability of the ligament to transmit load because some fibers within the ligament fail. If loading ­continues, a maximal or ultimate failure load is reached, and a sudden drop in load is recorded when many or all fibers fail, representing total failure of the ligament. The area under the load-deformation curve represents the amount of energy absorbed by a ligament during failure testing. Noyes and associates demonstrated that the characteristics of the ACL’s load-displacement curve are dramatically affected by variables such as age,75 strain rate,76 and duration of immobilization (disuse).77 Young adults have a yield point that can be as much as 3 times greater than that of an older person.75 In addition, Chandrashekar and coworkers78 revealed sex-based differences in the material properties of the ACL. The female ACL withstood 8.3% lower strain at failure when evaluated in a tensile failure test, 14.3% lower stress, and 22.5% lower modulus of elasticity. Recent studies have revealed sex-based differences in joint laxity; Hsu and associates79 found that the application of simultaneous tibial and valgus torque revealed 25% lower torsional joint stiffness in female knees as well as rotary joint laxity 30% higher than that for male knees.79 Noyes and colleagues76 also demonstrated the sensitivity of the ACL’s load-displacement response to strain rate. ACLs that failed rapidly (0.6 second) demonstrated a 20% increase in load to failure above those that failed at a speed two orders of magnitude slower (60 seconds). The energy stored just before ligament failure was 30% greater in preparations tested at the high strain rate in comparison with those that failed at the slow rate. In addition, any ligament that has been immobilized for even short periods will demonstrate a reduction in ligament strength.77 Most orthopaedic surgeons who operatively restore the function of the ACL perform an intra-articular reconstruction with autograft material.80 Noyes and associates73 characterized the relative strength of the various ligament replacement materials, demonstrating that a 14-mm wide bone–patella tendon–bone preparation was 168% as strong as the normal ACL, the strength of all other autogenous replacements being less in comparison with the normal ACL. Woo and coworkers81 have demonstrated that the normal tensile strength of the ACL may be as high as 2500 N, rather than the original 1725 N standard presented by Noyes and associates.73 This has led some surgeons to use combinations of autogenous graft material in an effort to increase the strength of the ACL replacement. Butler82 used the primate model to demonstrate that maintaining a vascular supply to an ACL graft produces no material property differences in comparison with a similar free graft 1 year after implantation. Papannagari and associates83 demonstrated that bone– patellar tendon–bone autograft reconstruction does not restore normal knee kinematics under physiologic loading conditions. Three months after surgery, subjects displayed an additional 2.9-mm anterior translation of the tibia relative to the femur at full extension and 2.2-mm anterior translation at 15 degrees of flexion compared with the contralateral knee under weight-bearing conditions. Tashman and colleagues84 developed a three-dimensional system to accurately assess dynamic joint motion, revealing abnormal kinematics in ACL-reconstructed knees during ­ weightbearing motion. A combination of radiographic targets

inserted into bone to eliminate skin artifact for biplane imaging (radiographic stereophotogrammetric analysis [RSA]) and subject-specific bone modeling revealed an additional 4-degree external rotation and 3-degree adduction in the tibiofemoral joint of the ACL-reconstructed knee compared with the contralateral normal limb during dynamic weight-bearing motion.84 Zantop and coworkers85 assessed the role of the anteromedial (AMB) and posterolateral bundles (PLB) of the ACL in tibial translation and rotary laxity. AMB transection significantly increased anterior tibial translation at 60 and 90 degrees, whereas isolated PLB transection resulted in increased tibial translation at 30 degrees of flexion in addition to increased combined internal-external rotation in response to internal-external rotary load. From these results, the investigators concluded that ACL reconstruction should include both bundles in order to restore normal translational and rotational kinematics. Mae and colleagues86 examined the force sharing of anteromedial and posterolateral grafts in “anatomic” two-bundle reconstruction in response to 134-N anterior tibial loading. They observed that this reconstruction technique yields grafts that share force similarly to the two bundles of the normal ACL.86

Function of the Cruciates in Joint Stability The concept of primary and secondary knee stabilizers was introduced by Butler and colleagues,87 who applied the stiffness approach to human cadaver specimens. They demonstrated that the ACL is a primary restraint to anterior translation of the tibia relative to the femur, providing an average restraint of 87.2% to the applied load at 30 degrees. With the knee at 90 degrees, this figure was 85.1%. After ACL transection, the remaining intact ligamentous structures provided little restraint to anterior subluxation, leading Butler to describe the function of the remaining soft tissues as secondary restraints to this particular motion. The remaining ligament and capsular structures each contributed less than 3% to the total restraining force resulting from an applied anterior shear load. Butler and colleagues87 demonstrated that the PCL is the primary restraint to posterior translation of the tibia relative to the femur, providing 94% of the restraining force at 90 and 30 degrees of knee flexion. None of the remaining ligamentous and capsular secondary structures contributed more than 2% of the total restraining force to an applied posterior shear load. Markolf and colleagues88 compared posterior tibial translation after isolated transection of the posterolateral band of the PCL, finding increased laxity after transection at 0 and 10 degrees of flexion. Sectioning of the PLB had no significant effect at higher flexion up to 90 degrees. Force measurements of the anterolateral band remained unchanged with PMB sectioning, leading the authors to conclude that the ALB is the primary restraint to posterior tibial translation. Fukubayashi and associates89 used the flexibility approach to investigate the coupled behavior between anteroposterior shear loading and internal-external tibiofemoral rotation in human cadavers. They showed that the ACL produces an internal tibial rotation with anterior shear load applied across the tibiofemoral joint, whereas the PCL

Knee 1585

produces an external tibial rotation with applied posterior shear loading. The magnitude of tibial ­ rotation, coupled with applied anteroposterior shear loading, decreased after transection of either cruciate ligament. Gollehon and coworkers90 also applied the flexibility approach to human cadaver specimens in an effort to investigate the role of the cruciates, lateral collateral ligament (LCL), and deep complex (arcuate ligament and popliteus tendon) in joint stability. This investigation showed that the PCL is the principal structure resisting posterior translation of the tibia relative to the femur. Isolated transection of the PCL did not affect varus or external rotation of the knee. Transection of the PCL, LCL, and deep complex was then performed to investigate the effects of a combined injury. This created a significant increase in varus rotation, posterior translation, and external rotation at all angles of knee flexion, suggesting that subjects with such a combined injury may have a functionally impaired joint. Combined sectioning of the ACL and the posterolateral structures produced a significant increase in internal-external rotation of the tibia, indicating that patients with this combined injury may also have compromised knee function.90 Similar methods were used by Markolf and coworkers, and their coupled-motion results agree with those of Gollehon and colleagues. Markolf also measured force in the cruciates and found that the cruciates become loadbearing structures with the application of varus rotation after transection of the LCL and deep complex.91 The same group measured force in the ACL and the PCL in the intact cadaveric knee under combined loading conditions.45,92 They found that force in the ACL increases most when anterior tibial force is combined with internal tibial torque when the knee is near full extension. Force in the ACL is also increased when anterior tibial force is combined with a valgus moment flexion angle greater than 10 degrees. The combination of posterior tibial force, varus moment, and internal torque produced the greatest forces in the PCL. In addition, they reported that the forces in the ACL were higher than the forces in the PCL in forced hyperflexion.43,44,91 More recently, Withrow and coworkers93 confirmed the results of Markolf and colleagues regarding the effect of valgus loading on the ACL. Peak strain in the anteromedial aspect of the ACL increased 30% with the addition of valgus loading in combination with impulsive compression loading, leading the authors to suggest that minimizing valgus loading, or abduction, during impulsive compression should reduce ACL strain. Grood and associates94 applied the flexibility approach with human cadaver specimens to investigate the role of the PCL and posterolateral structures (LCL, arcuate ligament, and popliteus tendon) in joint stability. Isolated sectioning of the PCL revealed that the amount of posterior tibial translation, measured relative to the femur, was twice as much at 90 degrees in comparison with that at 30 degrees of knee flexion. This occurred without abnormal axial tibial rotation and varus-valgus rotation. The concurrent increase in posterior laxity with flexion of the knee was attributed to slackening of the posterior portion of the joint capsule, which provides a secondary restraint to posterior translation. The authors concluded that clinical examination of the PCL should be performed at 90 degrees of flexion, at which the secondary restraints are

less ­ effective in blocking posterior tibial translation.94 At this knee angle, the clinician can gain a full appreciation of the PCL contribution to joint laxity. Removal of the posterolateral complex while leaving the PCL intact produced an increase in both external tibial rotation and varus rotation. The increase in external rotation was greatest at 30 degrees of flexion, at which it was 2 times larger in comparison with that measured at 90 degrees. This demonstrated that the posterolateral complex provides the primary restraint to external rotation with the knee at 30 degrees. Therefore, the authors recommend clinical examination of the posterolateral complex with the external rotation examination while the knee is between 20 and 40 degrees of flexion.94 A significant increase in external tibial rotation with the knee flexed to 90 degrees required transection of both the PCL and the posterolateral complex. This finding suggests that in a clinical examination demonstrating a significant increase in external tibial rotation with the knee at 90 degrees of flexion, deficiencies in both the PCL and posterolateral complex may exist. Several investigators have measured ACL displacement patterns, enabling calculation of strain pattern, to understand the effect of knee joint position and muscle activity on ligament biomechanics.38,51,54-56,57,63-65,95 Most of this work has been carried out in vitro, and the results are conflicting. The results of in vitro studies do not capture the effects of muscle activity or the effects of body weight, soft tissues, and secondary stabilizers in the knee. Recent in vivo studies have employed magnetic resonance imaging and three-dimensional computer modeling techniques to observe morphologic changes in ligaments, such as elongation, rotation, and twist.14,96 Li and coworkers96 demonstrated the elongation and rotation the ACL undergoes during weight-bearing flexion. The ACL length decreased by 10% at 90 degrees flexion compared with full extension. At 30 degrees of flexion, the ACL exhibited a 20-degree internal rotation. At lower flexion angles, the ACL oriented 60 degrees vertically and 10 degrees laterally, leading the authors to suggest that the ACL may have a greater role in weight-bearing activities at lower flexion angles. Li and associates14 used the same technique to demonstrate the reciprocal behavior of the ACL and PCL in vivo during weight-bearing flexion. The AMB of the ACL displayed a relatively constant length from full extension to 90 degrees of flexion, whereas the PLB shortened. Both bundles of the PCL elongated during flexion, leading the authors to highlight the reciprocal behavior of the ACL and PCL under weight-bearing flexion, rather than of the two bands of the ACL.14 Beynnon and colleagues12,13,67,69-71,97 have analyzed the in vivo strain biomechanics of the ACL in patients who were candidates for arthroscopic meniscectomy. This work involved arthroscopic implantation of the Hall-effect strain transducer or the DVRT into the AMB of the ACL after the routine surgical procedure was complete. These subjects had normal ACLs and consented to have their surgery performed under local anesthesia, allowing them full control of the lower limb musculature. The objective of these studies was to provide invaluable data for the clinical management of patients who had ACL ruptures. It was revealed that anterior shear loads of 150 N applied at 30 degrees of flexion (Lachman’s test) produced more

�rthopaedic ����������� S �ports ������ � Medicine ������� 1586 DeLee & Drez’s� O

strain within the normal AMB than did shear testing at 90 degrees (anterior drawer test).13,97 It was possible to predict AMB strain from anterior tibial translation at 30 degrees of flexion, but not at 90 degrees of flexion.97 Henning and associates59 directly measured the displacement pattern of the anteromedial aspect of the ACL in vivo under anterior shear load conditions and demonstrated that Lachman’s test produced greater elongation of the AMB in comparison with the anterior drawer test. These in vivo results13 are in agreement with previously published studies that used either instrumented knee laxity testing or clinical impressions to assess the behavior of the ACL under clinical examination conditions and to confirm that Lachman’s test is the clinical examination of choice to evaluate the integrity of the ACL.98-103 The in vivo study revealed that there was no significant change in ACL strain during isometric quadriceps contraction when the knee was maintained at a flexion angle of 90 degrees.13 At this flexion angle, across all study patients, the ACL remained unstrained (or slack) as quadriceps activity increased. Isometric quadriceps strengthening should therefore be safe in the ACL-injured or reconstructed knee if the flexion angle is maintained at 60 and 90 degrees. At 15 and 30 degrees of knee flexion, isometric quadriceps activity produces a large increase in AMB strain and should be carefully controlled, especially during the early stages of rehabilitation after reconstruction in which soft tissue fixation may be tenuous.13,104 Nisell and coworkers105 developed a two-dimensional model of the knee. They predicted that isometric quadriceps extension against a fixed resistance produced an anterior-directed shear force on the tibia, a loading that strains an ACL replacement with the knee positioned between 0 and 60 degrees. An isometric quadriceps extension effort between 60 degrees and full flexion produced posterior-directed forces on the tibia that would strain the PCL or its replacement and not the ACL. The in vivo ACL strain study13 indicates that the knee flexion angle at which isometric quadriceps activity produces an increase in ACL strain and may become unsafe for the injured or reconstructed ACL is somewhere between 45 and 50 degrees and remains to be delineated. The model predictions presented by Nisell and coworkers105 suggest that isometric quadriceps extension efforts at knee angles between 60 and 0 degrees may become unsafe for a newly reconstructed ACL, whereas this activity would be safe for a PCL reconstruction. Isometric quadriceps extension with the knee positioned between 60 degrees and full flexion may be unsafe for a PCL reconstruction, whereas this activity would be safe for an ACL reconstruction. Fleming and coworkers106 used in vivo strain measurement to demonstrate that the gastrocnemius muscle acts as an antagonist to the ACL. Gastrocnemius contraction produced greater strain on the ACL at 5 and 15 degrees of flexion than at 30 and 45 degrees. The authors proposed that rehabilitation design should take into account knee flexor torque supported by the gastrocnemius when it is desirable to minimize strain on the healing ACL graft.106 In vivo strain measurement within the ACL when a seated subject performed an isotonic quadriceps contraction (active range of motion [AROM]) consistently revealed ACL strain between 10 and 48 degrees, and an unstrained region between 48 and 110 degrees of flexion

ACL strain (%) 6

(N=8)

4 2

Flexion

0 −2 0

20

40 60 80 Knee flexion angle (deg)

100

Figure 23A-6   In vivo normal anterior cruciate ligament (ACL) strain for the seated subject extending the lower leg against gravity (active range of motion). (From Beynnon BD, Johnson RJ, Fleming BC, et al: The strain behavior of the anterior cruciate ligament during squatting and active flexion-extension. A comparison of an open and a closed kinetic chain exercise. Am J Sports Med 25:823-829, 1997.)

(Fig. 23A-6).13 AROM rehabilitation programs may now be prescribed with these two flexion angle regions adapted to the clinician’s requirements. In the unstrained region, quadriceps activity associated with AROM did not produce significantly different ACL strain values in comparison with the same knee motion without contraction of the leg musculature (flexion-extension motion of the subject’s knee performed by an investigator and termed passive range of motion [PROM]).13 This suggests that AROM between the limits of 50 and 100 degrees may be performed safely immediately after ACL reconstruction. The AROM activity may then move to flexion angles nearer full extension when the reconstruction and fixation will tolerate greater levels of strain.13 The maximal AROM strain values were greater (ranging between 4.1% and 1.5%) than the maximal PROM strain values.13 Application of a 10-pound boot to a subject’s foot during the AROM activity increased the ACL strain values in comparison with the same activity without a weighted boot. Our group also evaluated strain in the ACL during open and closed kinetic chain exercises and found no significant differences in the two exercises evaluated.67 This finding suggests that the specific closed kinetic chain exercise evaluated (squatting with and without resistance) is not necessarily “safer” than the open kinetic chain exercise tested (active flexion-extension). The results conflict with the results of some cadaveric studies, which conclude that closed kinetic chain activities are safer for the ACL than open chain kinetic activities.42,107 In a separate study, our group reported strains of the same magnitude (about 2.7%) during stair climbing.70 In addition, we investigated strain in the ACL during steady-state cycling and found the mean peak strain to be half of that experienced during closed kinetic chain exercises (squatting) or open kinetic chain exercises (active flexion-extension).71 The results imply that cycling may be a safe method of rehabilitating the knee musculature without damaging a healing ACL. More recently, Fleming and coworkers108 evaluated strain in the ACL during flexor-extensor exercises against resistance torque with and without a compressive load applied at the foot. Application of compressive load did not reduce the peak strain measurement of the ACL. However, ACL strain did not increase with an increase in resistance

Knee 1587

while the compressive load was applied.108 The same group later demonstrated that one-legged closed kinetic chain exercises did not produce more strain on the ACL than two-legged exercises.109 Peak ACL strain values at 30, 50, and 70 degrees of flexion were similar during four exercises: single-leg step-up and step-down, lunge, and onelegged sit-to-stand. The strain values were greatest at 30 degrees of flexion across all exercises. These results led the investigators to suggest that closed kinetic chain exercises can be used with increased resistance to rehabilitate muscles without placing additional strain on the healing ACL graft.108,109 Grood and associates110 have demonstrated in an in vitro model that leg extension exercises (in the range of 0 to 30 degrees) produce loadings that are potentially destructive to the repaired or reconstructed ACL. Further investigation is required to determine whether the AROM strain values between extension and 48 degrees are large enough to produce permanent elongation of the reconstructed tissue or failure of the fixation construct. Our findings illustrate that both muscle activity and knee position determine ACL strain at rest and with joint motion.13 It appears that for AROM, the ACL is strained between the limits of full extension and 48 degrees.13 These findings are consistent with Henning’s in vivo study of two patients with injured ligaments59 and with the findings of Markolf and ­colleagues.42 A ranked comparison of the different activities evaluated in subjects with normal ACLs, ordered from high to low risk on the basis of peak strain values, is presented in Table 23A-1. These in vivo data may be used in the development of rehabilitation programs after ACL reconstruction. In vivo strain measurement within the ACL for PROM between 110 degrees and full extension revealed that the ACL is strained as the joint is brought into extension, and measurement remains at or below the zero strain level between the limits of 11.5 and 110 degrees of flexion when

distal leg support loading is used (see Fig. 23A-6).13 Therefore, continuous passive motion of the knee within these limits should be safe for the reconstructed ACL immediately after surgery when the leg is supported throughout flexion-extension motion without applied varus or valgus loading, internal or external torques, or anterior shear forces. The limits of near extension (0 to 10 degrees), however, can cause small magnitudes of strain (1% or less).13 We think this should not be viewed as a constraint to bracing a patient’s knee in the fully extended position (0 degrees) or to the use of continuous passive motion during a rehabilitation program. Our in vivo PROM investigations12,13 and previous in vitro studies38,51,68 have shown that the ACL is not “isometric.” That is, the fiber length does vary as the knee passes through a range of motion (see Fig. 23A-6).13 The change in length of fibers is least in the AMB14,38; thus, most intraarticular reconstruction procedures now attempt to reattach an ACL graft to the attachment sites of this portion of the ligament. In vivo strain gauge analysis immediately after the fixation of an ACL graft has allowed us to determine whether the graft strain pattern is either similar to the normal ACL or unacceptable.12 Devices known as isometers have been devised to assist surgeons in identifying optimal ­attachment sites for an ACL reconstruction. Therefore, the in vivo ACL data13 for PROM of the joint may serve as important standards by which to accept or to reject isometer measurements of potential reconstruction tunnel placement sites. For PROM, it was observed that the difference between mean peak and mean minimal AMB strain values was 4.2% (range, 3.0% to 7.2%).13 If this difference is assumed to occur uniformly over the AMB length, and the mean length of the AMB is equal to 36 mm,111 there would be an average change in AMB length of 1.5 mm (range, 1.1 to 2.6 mm). The range of isometer displacement guidelines should be used with two considerations in performing

  TABLE 23A-1  Rank Comparison of Peak Anterior Cruciate Ligament Strain Values during Commonly Prescribed Rehabilitation Activities* Rehabilitation Activity

Peak Strain (%)

No. of Subjects

Isometric quadriceps contraction at 15 degrees (30 Nm of extension torque) Squatting with sport cord Active flexion-extension of the knee with 45-N weight boot Lachman’s test (150 N of anterior shear load; 30 degrees of flexion) Squatting Active flexion-extension (no weight boot) of the knee Simultaneous quadriceps and hamstrings contraction at 15 degrees Isometric quadriceps contraction at 30 degrees (30 Nm of extension torque) Stair climbing Weight-bearing at 20 degrees of knee flexion Anterior drawer (150 N of anterior shear load; 90 degrees of flexion) Stationary bicycling Isometric hamstrings contraction at 15 degrees (to −10 Nm of flexion torque) Simultaneous quadriceps and hamstrings contraction at 30 degrees Passive flexion-extension of the knee Isometric quadriceps contraction at 60 degrees (30 Nm of extension torque) Isometric quadriceps contraction at 90 degrees (30 Nm of extension torque) Simultaneous quadriceps and hamstrings contraction at 60 degrees Simultaneous quadriceps and hamstrings contraction at 90 degrees Isometric hamstrings contraction at 30, 60, and 90 degrees (–10 Nm of flexion torque)

4.4 (0.6) 4.0 (1.7) 3.8 (0.5) 3.7 (0.8) 3.6 (1.3) 2.8 (0.8) 2.8 (0.9) 2.7 (0.5) 2.7 (1.2) 2.1 (1.7) 1.8 (0.9) 1.7 (1.9) 0.6 (0.9) 0.4 (0.5) 0.1 (0.9) 0.0 0.0 0.0 0.0 0.0

 8  8  9 10  8 18  8 18  5 11 10  8  8  8 10  8 18  8  8  8

*Mean (±1 SD).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1588 DeLee & Drez’s� O

measurements for an ACL reconstruction tunnel placement. First, with repeated PROM of the knee, the surgeon should strive to reproduce the PROM pattern presented in Fig. 23A-6.13 This would require the isometer to measure elongation between 1.1 and 2.6 mm as the knee is brought from 50 degrees (when it should be at a minimum) out to extension. Likewise, as the knee is flexed from 50 degrees to nearly full flexion, the isometer pattern should demonstrate a slight elongation. Second, our data represent useful criteria when they are used with an isometer measurement system that has load-displacement behavior similar to the normal ACL but not with the highly compliant isometer systems. Our data show that measurements of ACL strain after ACL reconstruction do not correlate with elongation predicted by one type of isometer.69 It is important to recognize the limitations inherent in isometry systems. Care must be taken in interpreting isometer findings because the measurement is being made in an ACL-deficient knee that may have abnormal kinematics. Because the ACL is not present, a measurement system with normal ACL load-displacement behavior would provide the unique ability to restore normal tibiofemoral joint kinematics. In these circumstances, the isometry measurement will make a potential tunnel placement prediction based on conditions to which the ACL substitute will be exposed once it is implanted.50 This is not possible with current commercially available isometer systems. The cruciate ligaments serve several functions as passive stabilizers of the knee. The cruciates guide the knee joint through normal kinematics as demonstrated by the four-bar linkage model. The anterior and posterior cruciates are the primary restraints to corresponding anterior and posterior translation of the tibia relative to the femur and have a reciprocal relationship during weightbearing flexion. The coupled internal and external tibial rotation that occurs with corresponding anterior and posterior shear loading is controlled in part by the cruciate ligaments and should be considered a significant aspect of the clinical examination. In addition, the cruciates act as secondary restraints to varus-valgus motion of the knee joint. Surgical reconstruction of the anterior cruciate should reproduce the in vivo normal ACL strain biomechanics.

Medial and Lateral Collateral Ligaments and Their Function in Joint Stability Using the flexibility approach, Warren and associates65 assessed the restraining action of the medial collateral ligament (MCL) complex in human cadaver specimens. They demonstrated that sectioning of the superficial long fibers of the MCL complex produced a significant increase in valgus rotation of the tibiofemoral joint in experiments performed at 0 and 45 degrees of knee flexion. Sectioning the posterior oblique or deep medial portions of the MCL complex had no significant effect on increasing valgus knee angulation. These findings were confirmed by the work of Seering and coworkers,112 who employed the stiffness approach in the study of two human cadaver specimens. They reported that combined superficial and deep portions of the MCL provided 71% of the resistive valgus restraint in one specimen and 55% in another. Grood and colleagues113 also

applied the stiffness approach to investigate the medial ligament complex and presented results that support the findings of the Warren65 and Seering112 groups. More recently, Ellis and associates114 demonstrated in cadaveric knees that ACL deficiency led to an increase in MCL insertion site and contact forces during anterior tibial loading and had no effect during valgus loading, indicating that the ACL does not play a role in valgus restraint. These results led the investigators to suggest that increased valgus laxity during clinical examination of an ACL-deficient knee would indicate MCL compromise.114 In addition, Grood and coworkers113 demonstrated that the long superficial portion of the MCL complex provided 57% of the valgus restraint at 5 degrees, which increased to 78% at 25 degrees of flexion. The variable restraint behavior with valgus loading was attributed to the restraint provided by the posterior medial capsule, which decreased as the knee was brought from an extended into a flexed position. The same research group115 studied six intact and MCL-deficient cadaveric knees and found that there is a coupled external rotation associated with abduction in an MCL-deficient knee at extension, 15 degrees of flexion, and 30 degrees of flexion. In contrast, the intact knees studied had a coupled internal rotation associated with abduction. This finding, that is, the presence of a coupled external rotation as opposed to a coupled internal rotation, may be used in physical examination for diagnosis of isolated MCL injuries. Grood and coworkers113 applied the stiffness approach to investigate the LCL complex. They demonstrated that in response to varus stress, this complex limits lateral opening of the joint. In response to varus loading, the LCL was found to provide 55% of the total restraint at 5 degrees and 69% at 25 degrees of knee flexion. An increase in the contribution of the LCL to the total varus restraint resulted as the knee was brought from an extended to a flexed position. This change was attributed to a decrease in resistive support provided by the posterior portion of the lateral capsule as the knee was flexed. With the knee joint in full extension, the investigators demonstrated that the secondary restraints (including the cruciate ligaments and the posterior portion of the joint capsule) block opening of the knee joint after the collateral ligaments have been cut.113 Simulating the forces applied by the dynamic stabilizers (iliotibial tract and biceps muscles) revealed their important contribution to varus stability of the knee in vivo.113 The contribution of the dynamic stabilizers to overall laxity of the knee is difficult to assess because the actual ­muscle force magnitudes for a specific activity are unknown. In a later investigation, Gollehon and associates90 applied the flexibility approach to study the contribution of the LCL and deep ligament complex (popliteus tendon and arcuate ligament) to joint laxity. They demonstrated that the LCL and deep ligament complex function together as the principal structures resisting varus and external rotation of the tibia.90 Höher and colleagues116 conducted a cadaver study and concluded that the LCL and popliteus carry most of the force in PCL-deficient knees under a posterior load at high flexion angles. Further, they loaded the popliteus in tension to simulate muscle activation and found that the force in the popliteus complex was significantly greater than the force in the LCL at all flexion angles tested

Knee 1589

(0 to 90 degrees), in both intact cadaveric knees and PCLdeficient knees. Further research is required to assess the interaction of the popliteus complex when the other surrounding muscles are simulated.

MENISCAL BIOMECHANICS Function of the Meniscus in Load Transmission Meniscal injury is thought by some investigators to be the most common injury sustained by athletes.117 The menisci were originally thought to be vestigial structures that served no significant function for the tibiofemoral joint.118 The meniscus was thought to be an expendable structure; this perspective prompted many orthopaedists to treat meniscal tears by complete removal.119-123 As recently as 1971, Smillie123 recommended complete removal of the meniscus even if the posterior horn was the only structure suspected of being damaged at the time of anterior arthrotomy. Conversely, as early as 1948, Fairbank124 had suggested the load-transmission function of the meniscus and postulated that complete meniscectomy frequently resulted in tibiofemoral joint space narrowing, flattening of the femoral condyles, and osteophyte formation. In long-term followup studies performed in the late 1960s and 1970s, several investigators confirmed Fairbank’s observations, reporting a high incidence of unsatisfactory results after complete meniscectomy.125-127 It was not until the mid-1970s that several biomechanical studies confirmed the clinical observations by measuring the load-transmission function of the meniscus.128-136 These investigations predicted that between 30% and 99% of the load transmitted across the tibiofemoral joint passes through the menisci during weight-bearing activities. Maquet and colleagues130 used a contrast injection radiography technique to measure contact area in human cadaver specimens subjected to a physiologic compressive load. This study revealed a posterior translation of the medial and lateral contact areas as the knee was brought from an extended to a flexed position along with a decrease in contact surface area with knee flexion. The contact area also decreased significantly with meniscectomy, leading Maquet to postulate that the menisci transmitted a significant proportion of tibiofemoral compressive load. Walker and Erkman136 employed a methacrylate-­casting technique to measure contact area in both loaded and unloaded human cadaver specimens. With no compressive joint load, they demonstrated that tibiofemoral contact occurred predominantly through the menisci and that a substantial increase in tibiofemoral articular cartilage contact occurred when the compressive joint load was increased to 1500 N. The investigators also revealed that the meniscal contact was primarily along the lateral and medial periphery with the knee extended, moving from an anterior to a posterior region with knee flexion. Seedhom and Hargreaves131,132 reported that 70% to 99% of the tibiofemoral compressive load is transmitted through the normal menisci and that all of the load is transmitted through the posterior horns of the menisci with joint flexion past 75 degrees. These ­investigators also

revealed that partial removal of the meniscus decreased the compressive stress transmission of the joint less compared with removal of the entire structure, provided that the circumferential continuity of the meniscus was maintained. More recently, Zielinska and Donahue137 employed a three-dimensional finite element model to quantify changes in contact pressure in response to varying degrees of medial meniscectomy. The maximal contact pressure and contact area were linearly correlated with the proportion of the meniscus removed. The investigators revealed that removal of 60% of the medial meniscus increases the contact pressure on the remaining meniscus by 65%, and by 55% on the medial tibial plateau.137 Krause and coworkers129 reported an increase in stress across the knee joint of about 3 times in the canine model and 2½ times in human cadaver knees after removal of both menisci. The investigators also measured the circumferential displacement of the medial meniscus with an applied axial compressive load, demonstrating the presence of “hoop,” or tangential stress, acting at the outside fibers of the meniscus. This observation led Johnson and Pope to demonstrate how the meniscus absorbs energy by undergoing circumferential elongation as a load is developed across the knee joint.138 As the joint compresses, fibers elongate. Thus, the meniscus absorbs energy and reduces the impulsive shock loading that would otherwise be developed across the articular cartilage and subchondral bone. Bylski-Austrow and colleagues139 used radiographic techniques to measure the displacement of the meniscus as nine cadaveric knees were axially loaded in compression from 250 to 1000 N. They did not observe any radial displacement due to pure joint compression. The authors suggest that all of the radial displacement occurs before the axial load reaches 250 N. Ahmed and Burke140 directly measured the tibiofemoral pressure distribution using a microindentation transducer. They demonstrated that the medial and lateral menisci transmit at least 50% of the compressive load imposed on the tibiofemoral joint in the flexion range between 0 and 90 degrees. Removal of the medial meniscus caused a reduction in the contact area that ranged between 50% and 70%, the latter reduction occurring at greater axial load. Because articular contact stress is inversely proportional to contact area, a 50% decrease in the contact area would cause a twofold increase in contact stress. Allen and colleagues141 determined the resultant load acting on the meniscus in 10 intact and ACL-deficient human cadaver knees using an instrumented testing system. They found that the application of a 134-N anterior tibial load on an ACL-deficient knee significantly increases the resultant force acting on the medial meniscus compared with an intact knee at all flexion angles tested (0, 15, 30, 60, and 90 degrees). On the basis of the results, they suggest that ACL reconstruction will contribute to the goal of preserving meniscal integrity. These biomechanical investigations provide a basis for the concept of partial meniscectomy, which has been made possible with the technology provided by modern arthroscopic surgery. There can be no doubt that partial meniscectomy provides better results compared with total excision of that structure.142

�rthopaedic ����������� S �ports ������ � Medicine ������� 1590 DeLee & Drez’s� O

Function of the Meniscus in Joint Stability The menisci have been shown to provide increased geometric conformity to the tibiofemoral joint (thereby optimizing contact stress) and to share efficiently in the transmission of the tibiofemoral compressive load. Johnson and associates126 and Tapper and Hoover127 have performed postoperative clinical examinations in patients undergoing meniscectomy. Both studies revealed an increased varusvalgus and anteroposterior laxity in 10% to 25% of these patients, leading the investigators to conclude that the menisci provide some ability to stabilize the knee joint in connection with knee ligaments and bony geometry. The clinical follow-up study presented by the authors suggested a relationship between an increase in joint laxity after complete meniscectomy and sequelae of marginal spur formation and even degenerative changes.126 Bylski-Austrow and coworkers139 subjected nine human cadaver knees to physiologic loads at three flexion angles and measured displacements of the menisci radiographically. The tibias were either loaded in the anteroposterior direction with 100 to 150 N or torqued to 10 to 15 Nm. In all cases, a 1000-N compressive load was applied. Internal rotation caused the lateral meniscus to move 3 to 7 mm farther posteriorly than the medial meniscus moved anteriorly. Similarly, external rotation caused the medial meniscus to move posteriorly and the lateral meniscus to move a greater distance anteriorly. In all cases, the menisci “stayed with the femur” as the tibia was moved. The authors suggest that increased or decreased meniscal displacement, caused by ligament injury, meniscal repair, or meniscal replacement, might increase the risk for meniscal injury. Shefelbine and associates143 performed an in vivo study of human knees using magnetic resonance imaging to demonstrate that meniscal translation was not affected by ACL deficiency, but bone kinematics were altered. With a 125-N compressive load applied at the foot, the femur in ACL-deficient knees translated, on average, 4.3 mm further anteriorly from 0 to 45 degrees flexion than observed in the healthy knees. At full extension, the contact area centroid was shifted posteriorly relative to the tibia in the ACL-deficient knee. Translation of the medial meniscus did not differ between ACL-deficient and normal knees, leading the authors to suggest that altered bone kinematics subsequent to ACL injury, coupled with lack of compensation in meniscal translation, may increase the risk for secondary meniscal injury. Von Eisenhart-Rothe and colleagues144 revealed similar findings: magnetic resonance imaging revealed posterior translation of the medial femoral condyle relative to the tibia in ACL-deficient knees during isometric contraction of flexor and extensor muscles. Meniscal translation was the same across healthy and ACL-deficient knees. Wang and Walker145 evaluated the effects of transverse plane rotary laxity before and after removal of the menisci. In this work, two different types of rotary laxity have been defined. The first was primary laxity: the joint rotation that occurred between the limits of 0.5 Nm of applied internal-external torque, representing the “looseness” of the joint before a significant resistance to the applied torque was encountered. The second type was termed secondary laxity and was defined as the joint rotation that occurred

between the limits of 0.5 and 5 Nm of applied torque. The investigators measured primary and secondary rotary laxity in human cadaver knees before and after removal of both menisci. They revealed a 14% increase in primary rotary laxity and a 2% increase in secondary laxity. Even though meniscectomy did not produce a significant increase in secondary laxity, they concluded that the menisci serve as restraints to the rotation associated with primary laxity by acting as “space-filling buffers” between the tibiofemoral articular cartilage. Hsieh and Walker146 used an in vitro knee-testing device to evaluate the anteroposterior load-displacement response of human cadaver knees before and after bicompartmental meniscectomy. An evaluation of one specimen, with and without a compressive load across the tibiofemoral joint, demonstrated that a dual meniscectomy produced only a minimal effect on the anteroposterior tibial displacement at 0 and 30 degrees. A similar test procedure was followed in another specimen in which the cruciates were initially sectioned, followed by a dual meniscectomy. This finding demonstrated that in the absence of the cruciate ligaments, the menisci serve an important role in providing resistance to tibiofemoral anteroposterior translation. Allen and colleagues141 conducted a similar study to measure anteroposterior translation in intact cadaveric knees, ACL-deficient knees, and ACL-deficient knees that underwent a medial meniscectomy. Their findings agreed with those of Hsieh and Walker: the anteroposterior laxity of the intact knee was significantly different from the anteroposterior laxity of both groups of ACL-deficient knees. Further, they found that the coupled internal rotation associated with the anterior tibial load was less for both groups of ACL-deficient knees than for the intact knees. In both loading scenarios, the ACL-deficient knee that underwent a meniscectomy was the most lax, followed by the ACL-deficient knee, indicating that both the ACL and the meniscus are important in preventing anteroposterior laxity in the knee. Beynnon and associates147 conducted in vivo testing to assess tibial movement relative to the femur during transition from non–weight-bearing to weight-bearing subsequent to ACL injury. Knees with ACL insufficiency demonstrated an average anterior tibial translation of 3.4 mm compared with 0.8 mm observed in the healthy, contralateral knee. The authors hypothesized that further translation is prevented by the posterior horn of the medial meniscus, which would experience greater strain during transition to weight-bearing after ACL injury.147 Hollis and colleagues148 conducted similar testing to Allen and coworkers141 in nine human cadaver knees. They loaded the tibia from 0 to 38 N in the anteroposterior direction while applying a 200-N axial force along the axis of the tibia and measured meniscal strains in nine cadaveric knees with intact ACLs, sectioned ACLs, and reconstructed ACLs. Their results showed that meniscal strains increase after the ACL has been sectioned. After ACL reconstruction, however, the meniscal strains return to levels observed in the ACL-intact state, suggesting that ACL reconstruction reduces the likelihood of meniscal damage. Conversely, Pearsall and coworkers149 evaluated lateral and medial meniscal strain using DVRT strain gauges in eight cadaveric knees with intact PCLs, sectioned PCLs, and reconstructed PCLs. Strain in both

Knee 1591

menisci increased at 60 and 90 degrees of flexion in knees with sectioned PCLs. Similar to the results of Hollis and associates,148 PCL reconstruction reduced meniscal strain to PCL-intact levels.149 Markolf and colleagues44 evaluated the effect of meniscectomy on anteroposterior, varus-valgus, and rotary knee laxity in human cadavers with an instrumented laxity testing device. They observed that bicompartmental removal of the menisci increased anteroposterior joint laxity between 45 and 90 degrees of knee flexion, whereas there was only a minor increase in laxity in the rotary and varus-valgus planes. In a later study, these researchers demonstrated that although bicompartmental meniscectomy made the unloaded knee looser, in the knees with a developed tibiofemoral compressive load, laxity measurements were little affected.107 The same group used the instrumented laxity testing device to evaluate anteroposterior load-­displacement and varus-valgus torque-rotation responses of a subject’s knee joint in vivo.39 They demonstrated that medial meniscectomy alone does not create a measurable increase in varus-valgus laxity, whereas a trend of increased anteroposterior laxity was observed. A significant increase in anteroposterior laxity was observed in subjects with a combined medial ­meniscectomy and torn ACL.39 Levy and associates150 used human cadavers to investigate the effects of isolated medial meniscectomy and to study the effects of medial meniscectomy in the ACLdeficient knee. They demonstrated that an isolated medial meniscectomy did not produce a significant change in the anteroposterior load-displacement response of the knee. This finding is corroborated by the previous works presented by Hsieh and Walker146 and Bargar and associates.39 In the ACL-deficient knee without a compressive joint load, Levy and coworkers150 demonstrated that resection of the meniscus caused a significant increase in the anterior displacement of the tibia relative to the femur at 30, 60, and 90 degrees of knee flexion. This observation led these researchers to suggest that in the ACL-deficient knee, the posterior horn of the meniscus acts as a wedge between the tibiofemoral articular surfaces, resisting anterior excursion of the tibia relative to the femur. In a later study, this observation was confirmed by the same group.151 In this work, the medial ligament structures in an ACL-deficient knee with an intact meniscus were sectioned, revealing an increase in anterior tibial displacement relative to the femur in comparison with the knee in which only the ACL was cut. In the ACL-deficient knee with an intact medial ligament complex, the mechanism of anterior tibial restraint was demonstrated to be the wedging apart of tibiofemoral articular surfaces by the meniscus—a distraction resisted by the intact medial ligament complex and capsular structures—and the development of a tibiofemoral compressive load.151 The authors hypothesized that this may be one of the mechanisms that produces posterior horn tears of the menisci in an ACL-deficient knee.151 Tienen and coworkers152 demonstrated the relative immobility of the posterior horn of the medial meniscus using cadaveric knees. In the absence of tibial torque, the anterior horn moved further posteriorly and laterally than did the posterior horn during flexion. Application of external torque revealed constrained posterior horn displacement during the first 30 degrees of flexion. Watanabe

and associates153 demonstrated that anterior tibial translation increases after two-thirds and complete resection of the posterior horn of the medial meniscus. Additionally, applied varus torque increased external tibial rotation by 2.2 and 6.7 degrees after two-thirds and complete resection of the posterior horn, respectively.153 In a later study, Levy and coworkers154 investigated the effect of lateral meniscectomy on the motion of the human knee joint without compressive joint loading. They determined that isolated lateral meniscectomy did not produce a significant change in the anteroposterior load-­displacement behavior of the knee. In addition, these investigators revealed that the lateral meniscus does not act as a restraint to anterior translation of the tibia relative to the femur, leading the researchers to suggest that this structure may not behave like the medial meniscus in providing an effective posterior wedge to anterior translation.154 It is important that the results of these investigations of the meniscus be applied to events that occur without a compressive joint load, such as the swing phase of gait, and not to activities that include compressive joint loading. The menisci have also been thought to assist with joint lubrication, to provide resistance to extreme joint flexion or extension, and to aid in the damping of impulsive loads transmitted across the tibiofemoral joint. These functions are difficult to characterize biomechanically or to describe with clinical impressions, however.

PATELLOFEMORAL JOINT BIOMECHANICS The patellofemoral joint consists of the patella with a multifaceted dorsal surface that articulates with the femoral trochlear groove. It is a key component of the knee extensor mechanism. In 1977, Ficat and Hungerford155 characterized the patellofemoral joint as the “forgotten compartment of the knee.” A study of patellofemoral joint biomechanics is necessary to understand the pathologic processes, to develop rational treatment regimens, and to understand the effects that various rehabilitation programs have on this joint. For example, an abnormally high compressive patellofemoral joint reaction (PFJR) force produces abnormally high stress across the articular cartilage and is thought to be one of the initiating factors of alterations in articular cartilage metabolism, chondromalacia, and subsequent osteoarthritis156-159; in addition, morphometric abnormalities in the trochlear groove or the dorsal articular surface of the patella in combination with high lateral forces at the patellofemoral articulation have been thought to cause lateral subluxation or dislocation of the patella.155,159-161

Patellofemoral Contact Area In the normal knee, the patellofemoral contact area is optimally designed to respond to the increase in PFJR load developed with knee flexion through a corresponding increase in contact area. This helps distribute the contact force while minimizing patellofemoral contact stress. Goodfellow and colleagues162 used the dye method to measure patellofemoral contact area in human cadaver

�rthopaedic ����������� S �ports ������ � Medicine ������� 1592 DeLee & Drez’s� O S 90

A

M

L 45

20 B I 20 ˚

45˚

90˚

S A

M

L

B I 135˚ Figure 23A-7   Patellofemoral contact regions at different knee flexion angles. (From Goodfellow J, Hungerford DS, Zindel M: Patellofemoral joint mechanics and pathology. J Bone Joint Surg Br 58:287-290, 1976.)

5.0 4.0 3.0 2.0 1.0 0

20 30 60

90 120

Flexion angle (degrees)

2500

5.0 Contact force (N)

Contact pressure (MPa)

Contact area (cm2)

knees subjected to simulated weight-bearing conditions. Area measurements were made at 20, 45, 90, and 135 degrees of knee flexion and are presented in Figure 23A-7. Movement of the knee from full extension to 90 degrees revealed that the contact area on the dorsal aspect of the patella moves in a continuous zone from the inferior to the superior pole of the patella. Continued flexion of the knee to 135 degrees developed two separate contact regions, one on the “odd medial facet” and the other on the lateral

aspect of the patella (see Figure 23A-7). Singerman and colleagues163,164 calculated the center of pressure from a 6-degrees-of-freedom patellar transducer in human cadaver knees and reported that the center of pressure translates superiorly and medially as the knee is flexed to 90 degrees. At flexion angles greater than 85 degrees, the results were somewhat variable, but the center of ­ pressure always moved inferiorly with extension.163 Huberti and Hayes165 used pressure-sensitive film to measure the increase of patellofemoral contact area that occurs concurrently with knee flexion (Fig. 23A-8). At a flexion angle of 10 degrees, contact between the dorsal surface of the patella and the trochlea is initiated. The length of the patellar tendon controls when patellar-trochlear contact occurs. In patients in whom the patellar tendon is too long, patella alta may be present, and flexion of the knee greater than 10 degrees may be required to seat the patella adequately in the trochlear groove. Von Eisenhart-Rothe and colleagues166 employed an open magnetic resonance system coupled with threedimensional image postprocessing to evaluate kinematics and contact areas in the knee compartment in vivo. Patella tilt decreases during flexion from 30 to 90 degrees, coupled with an increase in lateral patellar shift. The femur rotates externally and translates posteriorly relative to the tibia in the same flexion range. These movements result in a significant increase in contact area.166 With knee movement between extension and 90 degrees, the patella was found to be the only component of the extensor mechanism that contacts the femur, holding the quadriceps tendon away from the femur. With knee motion between 90 and 135 degrees, the quadriceps tendon contacts the femur.167 Once the quadriceps tendon contacts the femur, the compressive PFJR force is divided between contact of the broad band of the quadriceps tendon with the femur and patellofemoral contact. The interaction between the patellofemoral contact area and PFJR force can be demonstrated with the squatting activity. During this activity, as knee flexion increases, the PFJR force initially increases, whereas the patellofemoral contact area available for distributing the contact force also increases, effectively distributing the articular contact stress. Besier and colleagues168 demonstrated that patellofemoral contact area increased on average by 24% during weightbearing knee flexion in both female and male subjects. The opposite situation may occur with knee ­ extension during

4.0 3.0 2.0 1.0 0

20 30 60

90 120

Flexion angle (degrees)

2000 1500 1000 500 0

20 30 60

90 120

Flexion angle (degrees)

Figure 23A-8   Experimental measurement of patellofemoral contact made in human cadaver specimens for the squatting activity with a normal Q angle. Values between 90 and 120 degrees have been extrapolated. Left, Contact area; middle, contact pressure; right, contact force. (From Huberti HH, Hayes WC: Patellofemoral contact pressures: The influence of Q-angle and tibiofemoral contact. J Bone Joint Surg Am 66:715-724, 1984.)

Knee 1593

weight-training programs that apply a weight to the distal aspect of the tibia with the athlete in a seated position. For this activity, the patellofemoral contact area decreases as the PFJR force increases; therefore, the PFJR stress may become high even if light weights are applied to the distal aspect of the tibia. This example may help explain why isotonic or isokinetic exercises through a full range of motion are not advised in the treatment of patellofemoral pain syndromes. Quadriceps exercises extending the knee only through the last 15 to 20 degrees of extension are more likely to be tolerated, as demonstrated by the decrease in PFJR force in Figure 23A-8.

Patellofemoral Force Transmission The patella transmits force from the quadriceps muscle group to the patellar tendon while developing a large PFJR force. This serves to stabilize the knee against gravity when the joint is in a flexed position and assists in the forward propulsion of the body as the knee is extended during gait. Therefore, the loads developed along the patellar tendon and the PFJR force are a function of both quadriceps force and knee flexion angle. A sagittal plane analysis can be used to demonstrate this. This employs application of statics to describe the forces and moments required to maintain the knee joint in equilibrium. For example, with use of this technique, the quadriceps force (FQuads), the PFJR force, and the patellar tendon force (FPT) may be related at chosen knee flexion angles. Figure 23A-9 is a simplified sagittal plane static representation of the relation between the PFJR and the quadriceps muscle forces. The mass of the upper body (W), assumed to act at the hip joint, is supported by the FQuads developed by the quadriceps muscle groups. The vertical line below the center of mass at the subject’s hip joint represents the force vector due to upper body weight, and this falls well behind the flexion axis of the knee. The distance from the center of mass force vector to the flexion axis of the knee is defined as the moment arm (c). The moment arm is relatively small with the knee near extension. Therefore, the support mechanism provided by the FQuads and the developed PFJR are relatively small. In the right portion of Figure 23A-9, the knee is in a position of greater flexion with an associated increase in

the moment arm (c′). To maintain the knee in static equilibrium, the new force (FQuads) generated by the quadriceps must increase significantly. As a result of the increased quadriceps force, the PFJR must also be larger. This model may help explain the mechanism by which both PFJR and FQuads increase during squatting activities. In the earlier force analysis studies, the patellatrochlea articulation was represented as a frictionless pulley.155,156,169-175 This assumption was justified on the basis of the low coefficient of friction between the patellofemoral articular surfaces. With this approach, the forces developed by the quadriceps muscle group were assumed to be equal to the force developed along the patellar tendon throughout the full range of knee motion, with the direction of the PFJR force defined as the bisector of the angle between the quadriceps and the patellar tendon force vectors. Employing the mechanics principle of static ­equilibrium, and with the assumption that the patella-­trochlea articulation behaves like a frictionless pulley, Reilly and Martens174 predicted a compressive PFJR force of 0.5 times body weight for level walking. For ascending and descending stairs, the PFJR was estimated to reach 3.3 times body weight.174 Analysis of the squatting activity revealed that a maximal PFJR of 2.9 times body weight occurred at 90 degrees of flexion.174 Active extension of the lower leg with a 9-kg boot while the femur was orientated in a horizontal position produced a peak PFJR at 36 degrees of flexion.174 Maquet176,177 questioned the frictionless pulley assumption and demonstrated with a lateral vector diagram of the patellofemoral articulation that the forces in the quadriceps mechanism and patellar tendon can differ and can also vary as a function of knee flexion angle. Several investigators have confirmed Maquet’s findings.178-183 In later work performed by Huberti,178 Van Eijden,183 Buff,181 Ahmed,179 Singerman,163 and their coworkers, the combined tibia, femur, and patella were evaluated by use of both experimental and theoretical techniques. Because the force values FPT and FQuads are unequal, these researchers have chosen to report results by calculating the ratio between the two force values (FPT:FQuads) at selected knee flexion angles. Huberti’s group simulated the squatting activity in human cadaver specimens while measuring FQuads with a tensile load cell and the FPT with Figure 23A-9   Static model of the patellofemoral joint reaction force (FPFJR) at two positions of knee flexion. With the knee in the flexed position (right), the values of FQuads and FPFJR are large, supporting the weight (W) of the upper body acting through a large moment arm (c′). The FQuads and FPFJR values are much less with the knee in a more extended position (left), in which the moment arm (c) is smaller.

W

W FQuads FQuads

Moment arm (c)

θ

FPFJR θ Moment arm (c’)

FPFJR

�rthopaedic ����������� S �ports ������ � Medicine ������� 1594 DeLee & Drez’s� O Force ratio: FPT FQuads

FQuads

FQuads

1.5 FPT

1.0

FPT

.5 FPT > FQuads 0

0

30

60 90 Knee flexion angle

120

Figure 23A-10   The predicted force ratio FPT:FQuads for the knee positioned between 0 and 120 degrees of flexion. Between 0 degrees and 45 degrees, the force developed in the patellar tendon is greater than that developed by the quadriceps musculature, whereas from 45 to 120 degrees, the patellar tendon force is less. (From Huberti HH, Hayes WC, Stone JL, Shybut GT: Force ratios in the quadriceps and the ligamentous patellae. J Orthop Res 2:49, 1984.)

a buckle transducer.178 They demonstrated that for knee flexion between 0 and 45 degrees, the FPT developed was greater than FQuads (Fig. 23A-10). With continued knee flexion to 120 degrees, the FPT was consistently less in comparison with FQuads. The authors suggested that not only does the patella function as a pulley that changes the magnitude and direction of forces in the quadriceps and patellar tendon, but also the patella has two distinct mechanical functions.178 In the first and more classically described function, the anteroposterior thickness of the patella can be attributed to increasing the effective moment arm of the quadriceps muscles and patellar ligament, whereas in the second, the patella acts as a lever (Fig. 23A-11). Therefore, the parameters that define the proximal and distal lever arms of the patella have a direct effect on the balance of forces in the quadriceps and patellar tendon. The researchers reasoned that the parameters were the length of the patella, the location of Figure 23A-12   The predicted force ratio FPFJR: FQuads (patellofemoral joint reaction force/quadriceps force) for the knee positioned between 0 and 120 degrees of flexion. Between 0 and 70 degrees, FPFJR is less than that developed by quadriceps contraction, whereas from 70 to 120 degrees, FPFJR is equal to FQuads. (From van Eijden TM, Kouwenhoven E, Verburg J, Weijs WA: A mathematical model of the patellofemoral joint. J Biomechan 19:219-288, 1986.)

FPT < FQuads

Figure 23A-11   The mechanical function of the patella as a lever and as a spacer to increase the patellar tendon moment arm. Left, With the knee near the extended position, the levering action of the patellar mechanism produces greater force values in the patellar tendon (FPT) in comparison with those developed by quadriceps contraction (FQuads). Right, With the knee in the flexed position, the levering action of the patella is decreased, and the force values developed in the patellar tendon are less than those developed by the quadriceps. (From Huberti HH, Hayes WC Stone JL, Shybut GT: Force ratios in the quadriceps tendon and ligamentous patellae. J Orthop Res 2:49, 1984.)

the patellofemoral contact area, and the angle between the quadriceps tendon and patellar tendon.178 In a parallel experimental investigation of the squatting activity, Huberti and Hayes165 estimated that the compressive PFJR force reached a maximal value of 6.5 times body weight. With knee flexion to 120 degrees, tendofemoral contact supported one third of the compressive PFJR force.165 Van Eijden and associates183 developed a mathematical model of the patellofemoral articulation and verified model predictions with experimental findings. Predictions of the FPT:FQuads ratio were similar to the experimental findings presented by Huberti,178 Ahmed,179 Buff,181 and their coworkers. Van Eijden’s group demonstrated that the PFJR is about 50% of the quadriceps force at full extension and increases to 100% of the quadriceps force with the knee positioned between 70 and 120 degrees of flexion (Fig. 23A-12).183 1.5

Force Ratio: FPFJR

1

FQuads

0.5

0 0

20

40 60 80 Knee flexion angle

100

120

Knee 1595

These studies have important implications for knee rehabilitation programs that are designed to minimize patellar tendon forces, as in patellar tendinitis. In these programs, application of the large isokinetic or isometric knee moments with the knee in positions between ­extension and 45 degrees should be avoided. In this flexion range, ­ quadriceps activity actually produces forces of greater magnitude in the patellar tendon. This has been demonstrated by the work of Huberti and colleagues,178 who showed that the FPT:FQuads ratio is greater than 1.0 for knee positions between extension and 45 degrees (see Fig. 23A-10). It may be advisable to restrict rehabilitation programs for patellar tendinitis to flexion angles between 45 and 120 degrees, in which the FPT:FQuads ratio is less than 1.0. This constraint will prevent an amplification of FPT. This restriction would not apply to normal gait, in which the bending moments and therefore FQuads are not high. Owing to the changing relationship between developed FQuads and resulting FPT as the knee courses from extension through full flexion, the effectiveness of the quadriceps in developing an extension moment becomes substantially smaller at larger knee flexion angles, and this will also prevent the amplification of FPT. These studies also have important implications in the rehabilitation and surgical treatment of patellofemoral pain syndromes. Rehabilitation programs designed to minimize the PFJR but not the FQuads should avoid large isokinetic, isotonic, or isometric moments with the knee positioned between 60 and 120 degrees of flexion. In this range, the predicted PFJR force is equal to the FQuads (see Fig. 23A12).183 With the requirement of minimizing the PFJR force, it may be advisable to restrict knee rehabilitation to range between the limits of extension, when the PFJR is about 50% of the FQuads, and 40 degrees, when the PFJR is 90% of the FQuads.183 Maquet158 has investigated the surgical treatment of patellofemoral pain, ­demonstrating that by increasing the extensor moment arm by a 2-cm elevation of the tibial tubercle, there is a 50% reduction in the PFJR force when the knee is flexed to 45 degrees. Ferguson and associates184 investigated the effect of anterior displacement of the tibial tubercle on patellofemoral contact stress. In this study, the patella-trochlea interfaces of human cadaver specimens were instrumented with miniature force sensors to monitor the patellofemoral contact stress. They revealed that anterior displacement of the tibial tubercle decreased the patellofemoral contact stress between 0 and 90 degrees of flexion.184 The largest decrease in contact stress was achieved with a 12.5-mm elevation of the tubercle; further elevation produced only a minimal decrease in contact stress.184 This demonstrates the importance of the anteroposterior position of the patellar tendon and its role in controlling the extensor moment arm. In addition, the proximal-distal location of the patellofemoral contact point is critical to the function of the patella as a lever (as explained earlier). In the frontal plane, the axis of the FQuads forms an angle with the patellar tendon. This has been defined as the Q angle and is measured as the intersection of the center line of the patellar tendon and the line from the center of the patella to the anterior superior iliac spine.185 The normal Q angle is reported to range between 10 and 15 degrees with the knee in full extension.186,187 With knee flexion,

the Q angle decreases because there is a coupled internal rotation of the tibia relative to the femur.155 Contraction of the quadriceps creates a bowstring effect that displaces the patella in a lateral direction, producing a contact force against the lateral margin of the femoral trochlear groove. Abnormal tracking of the patella, which allows lateral subluxation of only a few millimeters, markedly decreases the contact area, greatly increasing the local stress (force per unit area) (Fig. 23A-13). This may contribute to patellofemoral pain and degeneration of the patellar articular cartilage (chondromalacia). Other anatomic conditions can also contribute to abnormal patellar tracking. These include hypoplasia of the trochlear groove, abnormal patellar articular configuration, underdevelopment of the vastus medialis, transverse plane rotational malalignment of the proximal tibia relative to the distal femur, and abnormally high Q angle. Huberti and Hayes165 studied the effect of different Q angles by simulating the squatting activity in human cadaver specimens while measuring patellofemoral contact pressure with pressure-sensitive film. They demonstrated that either an increase or a decrease in Q angle developed an increased peak patellofemoral pressure and the associated unpredictable patterns of cartilage loading. Cox188 has presented a retrospective study of the Roux-Elmslie-Trillat procedure for realignment of the knee extensor mechanism and prevention of recurrent subluxation of the patella. An evaluation of 116 patients observed for at least 1 year demonstrated this procedure to be a satisfactory method for the prevention of lateral subluxation, with recurrence in only 7% of the cases. Careful attention to the medial transfer of the tibial tuberosity without a posterior displacement was emphasized as the key to successful long-term results.188 Procedures resulting in some posterior transfer of the tibial tuberosity, such as that described by Hauser, decrease the patellar tendon moment arm and consequently increase

R5

R5

RM

RL

R5

Figure 23A-13   Patellofemoral joint reaction forces for the normal knee (left). The joint reaction force (R5) is resisted by the lateral (RL) and medial (RM) components. In the knee with a lateralized patella (right), the joint reaction force is resisted by the lateral component only (R5). From Maquet P: Mechanics and osteoarthritis of the patellofemoral joint. Clin Orthop 144:70, 1979.

1596 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the patellofemoral contact stress. Fulkerson and Hungerford189 have reviewed the clinical and radiologic outcomes of the Hauser procedure and have presented evidence of progressive knee joint degeneration.

SUMMARY Future research endeavors in biomechanics should continue to perform in vivo strain measurement of the soft tissues surrounding the knee and establish new in vivo measurement techniques, such as pressure or force sensors. Sex-based differences in knee kinematics and intrinsic biomechanical properties of the soft tissues that span the knee should be explored. In addition, the continued development of an analytic model that includes both patellofemoral and tibiofemoral articulations and the dynamic forces produced by the foot-floor reaction and muscle contraction will permit the study of injury mechanisms, will allow the investigation of soft tissue reconstruction procedures, and will permit research concerning commonly prescribed rehabilitation activities. Application of the in vivo experimental techniques and analytic models should strive to establish the relationship between the biomechanical behavior of a graft and the resulting biologic properties. Technology, such as an implantable telemetered load sensor, should be designed to allow an optimal match between a rehabilitation regimen and the biologicmechanical behavior of the graft. Biomechanics research efforts should strive to establish intraoperative techniques and measurements that can accurately provide the surgeon with the ability to reestablish normal joint kinematics during a soft tissue reconstruction procedure. Future ­clinical biomechanical investigations of surgical procedures should include prospective, randomized, well-controlled, longterm studies that use standardized outcomes to assess the relative effectiveness of the many different soft tissue reconstruction techniques.

S U G G E S T E D

R E A D I N G S

Beynnon BD, Fleming BC: Anterior cruciate ligament strain in vivo: A review of previous work. J Biomech 31:519-525, 1998. Beynnon BD, Fleming BC, Labovitch R: Chronic anterior-cruciate ligament ­deficiency is associated with increased anterior translation of the tibia during the transition from non-weightbearing to weightbearing. J Orthop Res 20:332-337, 2002. Chandrashekar N, Mansouri H, Slaughterbeck J, et al: Sex-based differences in the tensile properties of the human anterior cruciate ligament. J Biomech 39:29432950, 2006. Coughlin KM, Incavo SJ, Beynnon BD: Tibial axis and patellar position relative to the femoral epicondylar axis during squatting. J Arthroplasty 18:1048-1055, 2003. Heijne A, Fleming BC, Renstrom P: Strain on the anterior cruciate ligament during closed kinetic chain exercises. Med Sci Sports Exerc 36:935-941, 2004. Levy MI, Torzilli PA, Warren RF: The effect of medial meniscectomy on anteriorposterior motion of the knee. J Bone Joint Surg Am 64:883-888, 1982. Markolf KL, Gorek JF, Kabo M, et al: Direct measurement of resultant forces in the anterior cruciate ligament: An in vitro study performed with a new experimental technique. J Bone Joint Surg Am 72:557-567, 1990. Shefelbine SJ, Ma CB, Lee KY, et al: MRI analysis of in vivo meniscal and tibiofemoral kinematics in ACL-deficient and normal knees. J Orthop Res 24:12081217, 2006. Tashman S, Collon D, Anderson K, et al: Abnormal rotational knee motion during running after anterior cruciate ligament reconstruction. Am J Sports Med 32:975-983, 2004. Woo SLY, Hollis JM, Adams DJ, et al: Tensile properties of the human femur­anterior cruciate ligament-tibia complex: The effects of specimen age and orientations. Am J Sports Med 19:217-225, 1991.

R eferences Please see www.expertconsult.com

S e c t i o n

B

Meniscal Injuries Stephen F. Brockmeier and Scott A. Rodeo

HISTORICAL PERSPECTIVE The human menisci, structures that were once believed to represent “functionless remnants of intra-articular leg muscles”1 have been the subject of voluminous publication and intense investigation over recent decades. Although most of our appreciation of the role of the meniscus in

knee function and performance has been elucidated more recently, a number of early reports served as the backbone to the modern comprehension of meniscal function, injury, and treatment. The earliest of these was published in 1936 by King, who used a canine model to study meniscal healing. In this report, he observed consistent, complete healing of defects

Knee 1597

created peripherally at the meniscosynovial junction, with little to no healing response noted in tears created within the meniscal substance itself. Additionally, King noted severe degeneration of the articular cartilage in animals in which all or a portion of the meniscus had been removed. These investigations led him to postulate a chondroprotective role for the meniscal cartilages and that a suitable vascular supply need exist for meniscal tears to heal.2,3 In 1948, Fairbank presented further evidence demonstrating the significant chondroprotective role of the meniscus. In a radiographic study of 107 patients evaluated after total meniscectomy, he observed three characteristic radiographic findings, now termed Fairbank’s changes: (1) formation of an anteroposterior ridge projecting downward from the margin of the femoral condyle, (2) flattening of the marginal half of the femoral articular surface, and (3) joint space narrowing. These changes were evident at variable times after surgery (as early as 5 months) and were seen to progress with time. Fairbank concluded that these changes resulted from a “loss of the weight-bearing function of the meniscus,” and that “meniscectomy is not wholly innocuous.”4 Subsequent investigations over the ensuing three decades evaluating the long-term clinical and radiographic outcomes after total meniscectomy confirmed Fairbank’s observations. Tapper and Hoover, Johnson and Kettelkamp, and others reported a high incidence of poor outcomes and the development of degenerative arthrosis in a large percentage of patients after meniscal excision.5-7 Despite this, total meniscectomy remained the treatment of choice for meniscal injuries, representing one of the most common procedures performed by orthopaedic surgeons into the 1970s. As the understanding of the consequences of meniscal deficiency increased, however, further study of meniscal structure, function, vascularity, and the development of alternate treatment options, including partial meniscectomy and meniscal repair, led to the modern algorithm for the management of meniscal injuries.

ANATOMY, STRUCTURE, BIOMECHANICS, AND FUNCTION OF THE MENISCUS Anatomy The menisci are paired, semilunar, fibrocartilaginous structures interposed between the femoral condyles and tibial plateaus. When viewed in cross section, they are wedge shaped, with a concave superior surface and a flat or convex inferior surface. This anatomy allows for congruency, with conformation to the rounded femoral condyles above and the relatively flat or concave tibial plateau below. Formed from the mesenchymal cells of the intermediate layer of the three chondrogenic layers that become the knee joint, the menisci become discernable structures at about the 8th week of embryologic development. At that time, they are highly cellular, with vascularity throughout their substance. The collagen composition of the menisci becomes similar to the adult structure at about 3 years of age owing to the effects of weight-bearing. Anatomic

meniscal variants exist and are significantly more common on the lateral side. The incidence of discoid lateral meniscus has been reported to be from 1.5% to 16.6%; in many, however, this variant is asymptomatic and discovered as an incidental finding.7,8 There are numerous macroscopic anatomic differences between the medial and lateral menisci (Fig. 23B-1). The medial meniscus is C-shaped, whereas the lateral meniscus has a more circular appearance. The medial meniscus is larger than the lateral meniscus in the sagittal plane, with its anterior attachment located more anterior than the anterior horn of the lateral meniscus and its posterior attachment located just posterior to the posterior horn of the lateral meniscus. The posterior horn of the medial meniscus is larger than the anterior horn when viewed in sagittal cross section. Conversely, the sagittal profiles of anterior and posterior horns of the lateral meniscus are similar in size. Although the medial meniscus covers about 50% of the surface area of the medial tibial plateau, the lateral meniscus covers 70% of the lateral plateau. Clinically, these anatomic differences are important considerations, both in the management of meniscal injuries and in the performance of meniscal allograft replacement. The anterior and posterior attachments of both menisci are firmly anchored to subchondral bone through insertional fibers. The anterior horns are attached to each other by the intermeniscal ligament in most patients. The medial meniscus is also solidly attached throughout its body to the medial joint capsule through the coronary and meniscotibial ligaments. The deep portion of the medial collateral ligament is a discrete medial thickening of the capsule and is intimately attached to the midbody of the medial meniscus. Despite these constraints, the posterior horn of the medial meniscus is still able to translate up to 5 mm with knee flexion to accommodate femoral rollback.9,10 In contrast, the lateral meniscus is more loosely attached to the surrounding capsule. This allows for a normal increase in excursion of the lateral meniscus when compared with the medial side. A translation of up to 11 mm is considered normal for a lateral meniscus.10 This increased mobility is thought to account for the lower incidence of lateral meniscal injuries. Posterolaterally, the popliteus tendon traverses the lateral meniscus at the popliteal hiatus. In an anatomic study performed at our institution, Simonian reported on the anatomy of the lateral meniscus and its attachments to the capsule and popliteus tendon, describing distinct superior and inferior popliteomeniscal fasciculi, which connect the lateral meniscus in this area to the popliteus and joint capsule (Fig. 23B-2).11 Disruption of these attachments led to gross instability of the lateral meniscus, and subsequent repair of the meniscus to the capsule led to restoration of normal meniscal stability and function. Accessory meniscofemoral ligaments exist in up to two thirds of patients. A ligament extending from the posterior horn of the lateral meniscus to the medial femoral condyle anterior to the posterior cruciate ligament (PCL) is known as the ligament of Humphrey; a ligament that attaches the posterior horn of the lateral meniscus to the medial femoral condyle posterior to the PCL is known as the ligament of Wrisberg. A variation in the posterior attachment of the lateral meniscus, known as the Wrisberg ligament variant, has been noted in a small number of patients with a discoid

1598 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 23B-1   Axial anatomy of the menisci when viewed from above. Schematic (A) and cadaveric (B) dissection, demonstrating the differences in the shape of the medial and lateral menisci as well as the positioning of the respective anterior and posterior horn attachments. (Schematic redrawn from Pagnani MJ, Warren RF, Arnoczky SP, et al: Anatomy of the knee. In Nicholas JA, Hershman EB [eds]: The Lower Extremity and Spine in Sports Medicine, 2nd ed. St Louis, Mosby, 1995, pp 581-614.)

Transverse intermeniscal ligament

Anterior cruciate ligament

Lateral meniscus Medial collateral ligament

Ligament of Wrisberg Medial meniscus

A

Posterior cruciate ligament

B

lateral meniscus. In these patients, the posterior capsular attachments to the lateral meniscus are absent, and the posterior meniscofemoral ligament is the sole stabilizing structure, allowing for excessive motion and an increased rate of meniscal instability and tears. Patients typically present in the first two decades with mechanical symptoms (the so-called snapping knee); treatment includes stabilizing the posterior attachment by repair to the capsule as well as addressing any concomitant intrasubstance tears. The microstructure of the meniscus reflects its functional role within the knee. Composed largely of water (70%), its dry weight consists mostly of type I collagen, with smaller amounts of type II, III, V, and VI collagen, noncollagenous proteins, proteoglycans, and elastin.7,9,11 Fibrochondrocytes are the characteristic cells found in the meniscus. There are two types of meniscal fibrochondrocytes: oval or fusiform fibrochondrocytes, which are noted superficially, and round fibrochondrocytes noted in the deeper matrix. Scanning electron microscopy of the meniscus has demonstrated most collagen fibers oriented in a circumferential fashion along the length of the meniscus, with interposed radial fibers located in larger density in the superficial areas but also noted in the deep substance. The superficial layer

Figure 23B-2   Anatomy of the popliteomeniscal fasciculi of the lateral meniscus. Sagittal magnetic resonance imaging demonstrating the superior and inferior fasciculi (red carets).

Knee 1599

Radial fibers

Mesh network fibers

Circumferential fibers

Figure 23B-3   Microstructure of the menisci. (From Bullough PG, Munuera L, Murphy J, et al: The strength of the menisci of the knee as it relates to their fine structure. J Bone Joint Surg Br 52: 564-567, 1970.)

contains a more mesh-like organization of radial fibers, with a preponderance of circumferential fibers noted deep to this in the substance of the meniscus (Fig. 23B-3). In the deep substance, radial fibers act to tie the circumferential fibers together, providing resistance to longitudinal splitting. This ultrastructure endows the meniscus with its tensile strength, allowing the dispersion of compressive loads as axial loads are converted to circumferentially directed stresses (“hoop stresses”). Although proteoglycans represent only a small fraction of the meniscal dry weight, the interaction of these molecules with water is responsible for the tremendous shock-absorbing capacity of the meniscus. Composed of chondroitin 6-sulfate (60%), chondroitin 4-sulfate, dermatan sulfate, and keratin sulfate, the proteoglyan component provides for absorption of loads in a sponge-like fashion by taking in or releasing stored water. The collagen and proteoglycan content of the meniscus increases with age until maturity and then remains relatively constant. The water content tends to increase with degenerative changes owing to a disruption of the meniscal collagen ultrastructure.7,12 An understanding of the vascular supply to the meniscus is essential because blood supply or the lack thereof has critical implications with regard to meniscal healing. As stated, prenatally and at birth, the meniscus is extremely vascular throughout its substance. By 9 months of age, the inner one third has become avascular, and by 10 years of age, meniscal vascularity mirrors the adult meniscus. The meniscus receives its blood supply from the superior and inferior geniculate arteries, with a contribution by the middle geniculate artery providing some blood supply to the anterior and posterior horns. Arnoczky and Warren investigated the microvasculature of the menisci in adults.13 They demonstrated a perimeniscal capillary plexus originating in the capsular and synovial tissues that penetrated only a small peripheral cuff of meniscus (Fig. 23B-4). On average, the outer 10% to 30% of medial meniscus and the outer 10% to 25% of lateral meniscus was vascularized.13 In addition, the area surrounding the popliteus tendon in the posterolateral meniscus was observed to be absent of vascularity. A fringe of synovial tissue was seen to extend over the peripheral

Figure 23B-4   The microvasculature of the adult meniscus.

few millimeters of the menisci, but did not penetrate their substance. These layers appeared to contribute no blood supply to the normal meniscus but were observed to play a role in meniscal healing when stimulated during meniscal injury or repair.13,14 Thus, apart from the peripheral few millimeters and the anterior and posterior horns, most of the adult meniscus is essentially avascular. Most of the meniscus appears to receive nutrition through diffusion. It has been proposed that the intermittent mechanical compression with applied loads acts to improve synovial fluid diffusion through the meniscal substance in a similar fashion to articular cartilage.9 The neuroanatomy of the menisci appears to mirror the vascular anatomy. Neural elements are concentrated in the peripheral menisci and at the insertion sites of the anterior and posterior horns. Sensory fibers, as well as mechanoreceptors, have been found in these areas, which likely play a role in pain production and proprioceptive feedback during joint motion. The central portions of the menisci appear to be essentially devoid of nerve fibers.7,9 Neurosensory mapping of the menisci was performed by Dye and associates.15 They found that probing of the central areas of meniscus in awake, nonanesthetized patients resulted in little or no pain, whereas stimulation of the more peripheral areas and the meniscocapsular junction produced more significant discomfort, providing further evidence for the presence of pain fibers in the peripheral meniscus.

Function The menisci are essential to normal knee biomechanics and function, serving several roles within the joint. These include load transmission, improving joint congruency, reducing joint contact stresses, shock absorption, providing passive stability especially in the anterior cruciate ligament (ACL)-deficient knee, and providing some lubrication and nutrition within the synovial cavity.7-9,16-27 The forces across the tibiofemoral joint range from 2 to 5 times body weight with normal ambulation. The menisci protect the underlying articular cartilage from feeling the entirety of this load by load sharing, increasing contact area, and dispersing axial stress. In full extension, the

1600 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

menisci transmit about 50% to 70% of the load (medial versus lateral). This increases to 85% when the knee is in 90 degrees of flexion.16 The radii of curvature of the femoral condyles do not mirror the corresponding tibial articular surfaces. The menisci improve joint congruency by bridging this anatomic gap. In the absence of a meniscus, the contact area between the femoral and tibial articular surfaces is a focused central area, which experiences high point stress. The menisci significantly increase this contact area by a factor or 2 or greater and, more importantly, decrease the peak contact stresses in the central area of the articular surface by 100% to 200%.17,19,20 Partial resection of the meniscus has been shown to lead to a proportionate increase in contact stress; a segmental resection or resection of 75% of the posterior horn can increase contact stresses to a level similar to those measured after total meniscectomy.17,18 Because of notable anatomic differences, contact stresses in the lateral compartment are altered to a greater extent after meniscal excision, accounting for the observed increased risk for progressive arthrosis after lateral ­meniscectomy.20,21 Closely connected to the load transmission role of the menisci is their role in shock absorption. Meniscal cartilage has about half the stiffness of articular cartilage. With the application of a sudden load, the meniscus acts to disperse forces by converting axial load to centripetal or horizontal stress through the elongation of the circumferentially oriented fibers within the meniscal substance. In a similar fashion to articular cartilage, the biphasic structure of meniscal cartilage also assists in the absorption of compressive load by movement of water through the matrix. This function of the meniscus is similarly compromised by any disruption of meniscal structure. Partial meniscectomy is preferable to total meniscectomy as some of the circumferential fibers remain intact; total meniscectomy results in at least a 20% reduction in this shock absorption function.22 The menisci also play a key role in enhancing joint stability. The primary anteroposterior knee stabilizers are the cruciate ligaments. However, in the absence of a ­functional ACL, the medial meniscus has been shown to be a ­secondary stabilizer to anterior tibial translation (Fig. 23B-5).23,25,26 Acting like a wedge to prevent excessive anterior translation of the tibia, the posterior horn of the medial meniscus

experiences increased stresses in the ACL-deficient knee, accounting for the increased incidence of medial meniscal injuries in this scenario. In a classic biomechanical study, Levy and colleagues demonstrated that medial meniscectomy leads to a 58% increase in anterior translation in an ACL-deficient knee with the knee flexed.23 In contrast, the lateral meniscus has not been shown to play a significant role in stability.24 Although the posterior horn of the medial meniscus can aid in stability in the ACL-deficient state, Shoemaker and associates reported that normal in vivo stresses, such as would be experienced with pivoting athletics, would likely overwhelm the resistive capacity of the meniscus, leading to clinical instability.26 The menisci also serve a nutritional and homeostatic role within the joint. During knee motion, the menisci shift within a set limit of excursion. This motion may help to circulate synovial fluid and aid in nutrition of the underlying articular cartilage. In addition, the menisci also serve as reservoirs of synovial fluid, which is expressed as the knee is loaded.

EPIDEMIOLOGY The incidence of acute meniscal tears has been reported to be about 60 to 70 per 100,000; an estimated 850,000 meniscal procedures are performed in the United States annually.9,28-30 Epidemiologic data have highlighted variations in the incidence of specific tear types in different age groups, genders, levels of activity, types of activity, and comorbid states.5,28-35 Meniscal tears are more common in men than in women with a ratio of 2.5:1 to 4:1.9,30 The most common activities leading to meniscal injury are those that require cutting or pivoting, especially with concomitant knee flexion, such as soccer, basketball, wrestling, football, gymnastics, and skiing. Medial meniscal tears are more common than lateral meniscal tears in all age groups and with most activity types.30-32,34,35 The incidence of meniscal injuries peaks in men in the third decade, whereas in women, the incidence remains relatively constant beginning in the second decade. In general, in both men and women younger than 30 years, vertical bucket handle tears of the periphery are more frequent, often resulting from a single traumatic event. In patients older than 30 years,

Figure 23B-5   Schematic representation of the role of the medial meniscus as a stabilizer to anterior tibial translation in the anterior cruciate deficient knee. ACL, anterior cruciate ligament.

Ruptured ACL

Intact ACL Medial Meniscus

Medial Meniscus

Knee 1601

degenerative tears become more common. These tears tend to have more complex configurations, often with a horizontal component. With the advent of magnetic resonance imaging (MRI), a noteworthy prevalence of asymptomatic meniscal tears discovered incidentally has been reported in the literature.36,37 In one series of young patients (average age, 35 years), a 5.6% overall incidence of frank meniscal tears was reported. Notable signal abnormalities within the posterior horn of the medial meniscus were even more frequently observed in 24% of knees.36 The prevalence of asymptomatic meniscal abnormalities increases with increasing patient age. In another series, 76% of asymptomatic patients with an average age of 65 years were noted to have tears of either the medial or lateral meniscus by MRI.37 Because of the anatomic considerations previously mentioned, the incidence of meniscal tears is greatly influenced by the status of the ACL. The reported incidence of concomitant meniscal tears in patients with acute ACL injuries has been reported to be about 60% to 70%.38 In patients with acute ACL ruptures, lateral meniscal injuries are observed more commonly than medial meniscal injuries.38,39 This is presumably due to the initial subluxation event with ACL rupture, leading to the pathognomonic bone bruise pattern and direct injury to the lateral ­meniscus as it is compressed between the lateral femoral condyle and tibial plateau. In a prospective series of young patients with acute complete ACL injuries, the rate of lateral meniscal tears was reported to be 57%, whereas the incidence of medial meniscal tears was 36%.38 As previously illustrated, in the setting of chronic ACL deficiency, medial meniscal tears become significantly more common.

CLINICAL EVALUATION OF MENISCAL INJURIES Clinical Presentation and History Any delay in the diagnosis and treatment of meniscal injuries can have a significant short- and long-term impact, given the mainly young, active patient population affected. A thorough history, physical examination, and appropriate diagnostic imaging tests will lead to an accurate diagnosis in most cases. Essential information that must be determined from the patient history includes the onset of symptoms, mechanism of injury, duration of symptoms, current symptomatology, and any activities or positions that exacerbate the symptoms. Patient age, activity level, and a history of any previous injuries to the knee or ipsilateral extremity are also important historical points. Acute, isolated meniscal injuries usually occur during sports, with a twisting or hyperflexion event. Patients often report mild swelling that presents a few days after the inciting injury, in contrast to the large, immediate effusion that is often evidenced with other intra-articular ligamentous injuries. Patients most commonly complain of pain located medially or laterally at the joint line; mechanical symptoms such as clicking, catching, or locking should alert the clinician to the potential presence of an unstable bucket handle or flap tear. True locking of the knee is defined

as a mechanical block to full extension due to a displaced bucket handle meniscal tear with incarceration. This can either be episodic with reduction of the fragment by patient manipulation or motion of the knee or ­persistent, which is an ­indication for immediate surgical intervention. Degenerative meniscal tears tend to have a more insidious, atraumatic onset, presenting with vague joint line pain, mild swelling, and occasional mechanical symptoms. Patients tend to be older and less active, often with some concomitant osteoarthritic changes.

Physical Examination and Testing The physical examination of a patient with a suspected meniscal injury should begin with an assessment of the patient’s stance and gait. The contralateral extremity should also be examined to allow for comparison. An examination of the entire extremity with attention to limb alignment is critical. The knee examination should begin with inspection to assess for an effusion as well as any quadriceps atrophy or joint line swelling, as would be seen with a meniscal cyst. This can be done in either the seated or supine position. Manipulation or “milking” of the suprapatellar pouch in an inferior direction can help to identify a small effusion, which may otherwise be difficult to detect. Gentle passive range of motion should be performed to assess for any asymmetry or loss of extension, as might be seen with a displaced bucket handle tear. Discomfort experienced with extremes of flexion is a nonspecific sign but is often noted in the presence of a meniscal injury involving the posterior horn medially or laterally. Next, palpation of the femoral origins of the collateral ligaments, the patellofemoral region, and the medial and lateral joint line are performed to assess for point tenderness. In our experience, tenderness at the medial or lateral joint line is best elucidated in the supine position with the hip externally rotated and the knee at 90 degrees of flexion. In multiple series, the presence of joint line tenderness to palpation has been observed to be the most accurate and sensitive clinical sign of a meniscal tear.39-42 However, in the presence of an ACL injury, the presence or absence of point tenderness at the joint line becomes a significantly less accurate predictor of the presence or absence of concomitant meniscal injury.43 Numerous specialized provocative tests have been described in the diagnosis of meniscal tears. These include McMurray’s test, the Apley grind test, variations of weight-bearing squatting maneuvers, and others.40,41,43-45 ­McMurray’s test is classically performed with the patient in the supine position. The knee is placed in hyperflexion; for evaluation of the medial meniscus, the knee is loaded and, with combined varus and internal rotation forces applied, brought into extension. The joint line is palpated for a click, representing subluxation of an unstable meniscal flap. The lateral meniscus is similarly tested by applying valgus and external rotation and extending the knee. Although a palpable click is considered a positive ­McMurray’s sign, patient discomfort or reproduction of symptoms with this maneuver can also be helpful in diagnosis. Apley’s test is performed in the prone position with flexion of the knee. Internal and external rotations are performed with both

1602 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

distraction and then compression or “grinding” of the knee to differentiate meniscal pathology from collateral injuries. Recent investigations of these maneuvers have reported low sensitivities but relatively high specificity for meniscal injury. The squat test is another useful provocative maneuver. It is both useful for diagnosing a meniscus-based disorder and helpful in distinguishing meniscal pathology from a patellofemoral problem. The patient descends from a standing posture to a deep squat. With a meniscal injury, the patient reports pain located posteriorly or at the medial or lateral joint line that increases at the lowest point of the squat; with patellofemoral pain, the patient complains of pain throughout the squatting motion, which is located anteriorly. The knee examination should also include a complete assessment of ligamentous stability, with specific attention to evaluation of ACL integrity by a Lachman maneuver and pivot-shift testing. The collateral ligaments should be evaluated by assessing varus and valgus stability with the knee at 0 and 30 degrees of flexion. A complete neurovascular examination of the lower extremities completes the physical examination. Despite the lack of a true pathognomonic sign for the diagnosis of a meniscal injury, the clinical examination remains extremely useful. An accumulation of findings such as the presence of an effusion, tenderness to joint line palpation, and positive provocative maneuvers will help the clinician to hone his diagnosis and choose an appropriate course of action.

Imaging In the setting of an isolated meniscal tear, the combination of history and a comprehensive physical examination should lead to an accurate diagnosis in more than 90% of patients. Further diagnostic testing in the form of an imaging study is frequently confirmatory in this situation. However, in the presence of concomitant ligamentous or chondral disorders, accurate diagnosis with regard to the presence of a meniscal disorder is often more difficult, requiring further investigation. Plain radiographs, arthrography, and MRI have all been employed in the diagnosis of meniscal injuries. In our opinion, plain radiographs should be routinely obtained in any patient with a suspected meniscal injury. A standard series consisting of a weight-bearing anteroposterior, lateral, and 45-degree posteroanterior flexion view, as well as a Merchant patellar view, will help to either identify or rule out degenerative joint disease, loose bodies, fractures, and osteochondritis dissecans lesions. The 45-degree flexed posteroanterior view can be particularly helpful in alerting the clinician to early joint space narrowing that may not be evident in the fully extended view. In any patient presenting with an abnormality or asymmetry in clinical limb alignment, a full-length standing long cassette anteroposterior radiograph from hip to ankle of both extremities should also be obtained. This will facilitate a full assessment of knee mechanical alignment. In any patient with a suspected internal derangement of the knee, further imaging evaluation by MRI is helpful in assessing the soft tissue structures, including the ligaments,

menisci, capsule, and articular cartilage. Arthrography was the imaging study of choice for diagnosing meniscal tears before the advent of MRI; currently, it is mostly of ­historical interest but should be considered if MRI cannot be performed. MRI has many advantages, including its noninvasive nature, its lack of ionizing radiation, and its ability to image the knee in multiple planes. It also allows for the evaluation of concomitant abnormalities such as ligamentous injuries and chondral lesions. Its limitations include cost and potential for technical error associated with either under-reading or over-reading the presence of pathology. In general, modern MRI scans have been reported to have an extremely high accuracy with regard to detection of meniscal abnormality.46,47 In the evaluation of a suspected meniscal injury, we routinely obtain an MRI series that includes both fat-suppressed and diffusion-weighted fast spin-echo (cartilage sensitive) axial, coronal, and sagittal images (Fig. 23B-6). Sagittal meniscal windows are also obtained. A comprehensive, systematic approach to reading an MRI is critical. Sagittal images provide visualization of most acute, vertically oriented meniscal tears. Radial tears and horizontal, degenerative tears are also well visualized on the coronal images; axial images, if obtained correctly, can further identify radial and flap tears and can help correlate the findings noted in sagittal and coronal cuts. With careful attention, displaced flaps can be noted either in the joint space, peripherally, or in the intercondylar notch region. The classic double PCL sign refers to the appearance of a second low signal structure paralleling the PCL that is a bucket handle tear of the meniscus that has been displaced into the notch. The normal appearance of the menisci on both fatsuppressed and fast spin-echo images is that of uniformly low signal intensity. Areas of higher signal within the substance of the meniscus can be frequently seen and can lead to over-reading of a meniscal lesion. A grading system has been described for classifying meniscal tears as viewed on MRI (Fig. 23B-7).48 Grade I signal change is a generalized intrasubstance increase in signal with no communication to the articular surface. Grade II is a linear area of increased signal that also does not exit inferiorly or superiorly into the articular space. Grade III signal is a linear area that abuts the free edge of the meniscus, exiting either superiorly or inferiorly, and communicates with the joint. The presence of grade III signal is consistent with a true meniscal tear and, if visualized on two or more MRI images, has a sensitivity of greater than 90%.48 Care must be taken, however, because a number of normal intra-articular structures, such as the meniscofemoral ligaments of Humphrey or Wrisberg, the intermeniscal ligament, and the tendon of the popliteus are in proximity to the meniscus and can mimic a meniscal tear.

Arthroscopy As stated, the diagnosis of a meniscal tear is often evident after a thorough history and examination; imaging is often confirmatory and helpful in planning treatment by further clarifying the extent of meniscal injury, tear configuration, and the presence or absence of concomitant injuries. Arthroscopy remains the gold standard in the diagnosis of

Knee 1603 Figure 23B-6   Magnetic resonance imaging evaluation of the menisci. A, Sagittal view demonstrating a peripheral vertical meniscal tear. B, Coronal view demonstrating a complex meniscal tear. C, Double posterior cruciate ligament sign due to a displaced bucket handle tear. D, Coronal view demonstrating an absent posterior horn due to a displaced bucket handle tear.

A

B

C

D

meniscal pathology as well as other intra-articular derangements. It affords direct visualization, inspection, and probing of the meniscus throughout its entirety.

CLASSIFICATION OF MENISCAL TEARS In general terms, a classification is valuable only if it assists in guiding prognosis and treatment or facilitates communication between clinicians in the discussion of the particular entity. Meniscal tears have been classified based on etiology, tear pattern, vascularity, and location. Etiologically, meniscal tears are either acute, due to abnormal forces (trauma) acting on a normal meniscus, or degenerative, resulting from normal forces acting on an abnormal meniscus. Although specific varieties of acute, traumatic tears may be amenable for repair, most degenerative tears are irreparable because of poor tissue quality and require partial meniscectomy. Descriptive terms such as vertical, longitudinal, bucket handle, oblique, parrot beak, radial, horizontal, and complex have been used to describe tear configurations (Fig. 23B-8). In younger patients, vertical, longitudinal and oblique tear patterns are most pervasive; however, in patients older than 40 years, complex tears become more frequent. Vertical/longitudinal meniscal tears are also commonly called bucket handle tears when they are complete and unstable. Typically associated with injury to the ACL, they most commonly originate posteriorly and vary in length. Generally, complete tears larger than 1 cm in length are unstable and will displace into the joint, leading to mechanical

symptoms and a block to extension or locking of the knee. Bucket handle tears occur more commonly in the medial meniscus, likely owing to its more firm peripheral attachments. Incomplete vertical/longitudinal tears, as well as complete tears less than 1 cm in length, are often encountered during arthroscopy. These variants are

Figure 23B-7   Grade III signal in a sagittal fat-suppressed image demonstrates linear signal, which communicates with joint space through the inferior surface of the meniscus.

1604 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Vertical / longitudinal

Oblique

Transverse (Radial)

Degenerative

Horizontal

Figure 23B-8   Descriptive classification of meniscal tears. (Redrawn from Ciccotti MG, Shields CL, El Attrache NS: Meniscectomy. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery. Baltimore, Williams & Wilkins, 1994, pp 591-613.)

f­ requently stable and asymptomatic and thus require no ­intervention.49,50 Oblique tears, alternatively described as flap or parrot beak tears, most frequently occur at the junction of the posterior horn and the body of the meniscus. These injuries tend to be highly symptomatic owing to mobility of the flap, which can catch within the joint or flip under the intact surrounding meniscus, causing local irritation of the synovium, peripheral meniscus, and capsule. Repair of oblique tears is rarely possible, but patients tend to experience significant relief from a local partial excision to a stable rim. Radial tears can occur both medially and laterally and can be either isolated or in conjunction with other tear components. An isolated radial tear can be relatively asymptomatic but can propagate, leading to a more symptomatic, unstable configuration. Horizontal tears are thought to result from sheer force leading to a parallel disruption between the superior and inferior substance of the meniscus. These increase in frequency with age, occurring more often as the tissue quality becomes less healthy and pliable; repair is infrequently possible. Complex tear configurations occur in multiple planes and often include a horizontal component. These most commonly occur in the posterior horn region, are degenerative in nature, and are never amenable to repair. In evaluating the suitability of a meniscal tear for repair, the most important criterion is the vascular supply. In this light, meniscal tears, specifically vertical tears, have been classified based on their proximity to the periphery. A “red-red” tear is located in the peripheral third of the meniscal substance, has a functional blood supply on both sides of the tear, and obviously has the best prognosis for healing. A “red-white” tear is more centrally located, has a functional peripheral blood supply, but the central surface of the tear is devoid of vascularity. These lesions should theoretically have sufficient blood supply to effectively heal by fibrovascular scar formation, but the prognosis is more guarded. Tears that occur in the “white-white” region are confined to the avascular central portion (inner two thirds) of the meniscus. These have no functional blood supply and theoretically no healing potential, unless they are provided with healing potential through vascular access channels, abrasion of the synovial fringe tissue, or augmentation with exogenous fibrin clot.51-57 Cooper and associates51 provided a comprehensive classification of meniscal lesions using a zone classification, to

allow for a more consistent clinical documentation of tear pattern and location and to facilitate outcomes comparison (Fig. 23B-9). They divided each meniscus into three radial zones: zones A (posterior), B, and C for the medial meniscus and zones D, E, and F (posterior) for the lateral meniscus. Each meniscus is then divided into four circumferential zones: 0 for the meniscosynovial junction, 1 for the outer third, 2 for the middle third, and 3 for the inner, central third.

TREATMENT OPTIONS Nonoperative A patient presenting on referral with complaints related to their knee and an MRI demonstrating an abnormality of the meniscus before a thorough clinical evaluation has become a common clinical scenario. Given the significantly high incidence of meniscal abnormalities seen on MRI evaluation of asymptomatic individuals, it is first imperative that it be confirmed that the patient’s symptoms are truly due to meniscal pathology. Once the diagnosis of a

Posterior A

F

Lateral 0

Medial 1

2

3

3

2

1

0

E

B

D

C Anterior

Figure 23B-9   Classification of meniscal tears according to Cooper and colleagues. (Redrawn from Cooper DE, Arnoczky SP, Warren RF: Arthroscopic meniscal repair. Clin Sports Med 9:589-607, 1990.)

Knee 1605

symptomatic meniscal tear has been made, a nonoperative or an operative approach can be taken. To recommend a nonoperative approach to a patient with a meniscal tear, the clinician must have an understanding of meniscal healing and the natural history of meniscal tears. Although evidence exists that suggests spontaneous healing in some stable, isolated, peripheral tears of the meniscus,58 it is safe to say that, in general, most meniscal tears will not heal or spontaneously resolve without intervention. The literature to date is deficient with regard to the natural history of patients presenting with symptomatic meniscal tears treated nonoperatively. Although limited evidence suggests that patients may experience some symptomatic resolution,28 it has been our experience that most unstable and symptomatic meniscal tears respond poorly to a nonsurgical approach, especially if the patient is experiencing mechanical symptoms. A 6- to 12-week period of rest, icing, anti-inflammatories, and activity modification with avoidance of pivoting activities may help modulate symptoms in patients who are poor surgical candidates or who wish to avoid surgery. An injection of corticosteroid can be quite effective in relieving the secondary synovial inflammation associated with a symptomatic meniscal tear. However, owing to the theoretical deleterious effects and potential impedance to healing associated with intraarticular steroid, we do not routinely use this modality.59 There also exists with a nonoperative approach a moderate risk for propagation of the existing tear and the conversion from a previously reparable tear configuration to one that is not amenable to repair.

Operative Indications Based on known evidence, the indications for surgical intervention in a patient with a meniscal tear are as follows: • Symptoms of meniscal injury that include persistent pain or limitations of activities of daily living, work, or athletics • Mechanical symptoms • Positive findings on examination, including effusion, joint line tenderness, restriction of motion, and the presence of provocative signs • Absence of evidence for other causes of the patient’s knee symptoms • Confirmatory evidence on imaging studies, including plain films and MRI, which demonstrate a meniscal tear with a configuration amenable to surgical repair or resection • A failure of nonoperative treatment (in select cases) Traditionally, many authors have recommended management with an initial, attempted period of nonsurgical management before operative intervention. In the presence of mechanical symptoms or an unstable tear configuration, nonoperative treatment is often ineffective and carries with it a theoretical risk for further injury to the already compromised meniscus. In these patients, it is reasonable to pursue operative arthroscopic treatment to address the known pathology without a dogmatic period of “conservative” management. After it has been determined that a patient requires operative intervention, the next judgment that must be

made is whether the tear is amenable to surgical repair or is more appropriately treated with resection. The decision about which meniscal tears are suitable for repair is based on a number of factors, including patient age, activity level, tear location, tear configuration, duration of the tear, the presence of associated injuries, and most importantly the capacity of the tissue to heal (vascularity). Although no consensus has been reached as to the absolute indications for meniscal repair, the ideal scenario for meniscal repair is in a young, active patient with an unstable vertical tear, measuring greater than 10 mm in length, and located in the peripheral 3 to 4 mm (peripheral one third) of the meniscus. Successful healing and outcomes are relatively predictable in this setting.11,60-69 With tears located slightly more central in the red-white area, or between 3 and 5 mm from the meniscosynovial junction, the decision to attempt meniscal repair should be based on the individual scenario because success is less predictable.60,62-64 In younger, active patients with good tissue quality, tears in this transitional zone should have adequate vascularity, and thus, an attempt at meniscal repair is sound as the benefits of meniscal preservation outweigh the potential risk for failure of the repair. The indications for a meniscal repair have been broadened as techniques have improved to include certain vertical white-white zone tears in young, active patients, as well as some more complex configurations, including radial and flap tears in certain situations, in order to preserve the native meniscus.68-71 In these ­situations, adjuncts such as trephination, extension of the tear into the vascularized zone, or the use of fibrin clot may be employed.52-57 Careful attention should be paid to meniscal preparation, repair fixation, and postoperative rehabilitation. Circumstances that render a meniscal tear irreparable include poor tissue quality, such as in a degenerative tear, horizontal or complex tear configurations, unaddressed instability due to ACL deficiency, and advanced degenerative changes. Radial or flap tears are rarely reparable, but attempted repair with healing adjuncts may be warranted in young patients.57,68,69

Concomitant Anterior Cruciate Ligament and Meniscus Injuries The interrelation of ACL injuries and meniscal tears has been extensively studied.25-27,49,50,60,62,66,70-77 This constellation of injuries represents a large subset of meniscal pathology and requires separate discussion. As previously stated, there is a high incidence of meniscal tears that occur with an acute ACL rupture, with lateral meniscal injuries being more common. Furthermore, in patients with chronic ACL insufficiency, the medial meniscus acts as a secondary stabilizer to anterior tibial translation. Because of the increased strain on the posterior portion of the medial meniscus in the ACL-deficient knee, there is a known consequence of medial meniscal tears in this population.25-27 The treatment of meniscal injuries is directly affected by the presence of associated instability with ACL injury. Significantly higher rates of meniscal healing have been reported in the presence of concomitant ACL injury and following combined meniscal repair and ACL reconstruction.49,50,60,62,66,70-77 Shelbourne reported on his experience

1606 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

with aggressive nontreatment of certain lateral ­ meniscal tears noted on diagnostic arthroscopy before ACL reconstruction. He found that certain stable tear variants, specifically posterior horn avulsion tears, stable radial or flap tears, and peripheral vertical tears extending less than 1 cm beyond the popliteus, remained asymptomatic when either treated with abrasion or trephination or left in situ.49,76 Yagashita reported reasonable rates of healing without suture repair, noted on second-look arthroscopy, in patients undergoing ACL reconstruction with stable, peripheral tears of the menisci measuring smaller than 1.5 cm. Tears of the lateral meniscus had better outcomes than similar tears on the medial side.50 Series evaluating success rates of meniscal repair have also demonstrated improved rates of healing in patients undergoing concomitant ACL reconstruction. Cannon reported a 90% rate of successful healing, verified by either arthroscopy or arthrography, in patients undergoing inside-out meniscal repair in conjunction with ACL reconstruction compared with a 50% success rate in those undergoing isolated meniscal repair in an ACL stable knee.72 Tenuta and Arciero reported similar success rates with 90% success in concomitant meniscal repair and ACL reconstruction compared with 57% healed in isolated meniscal repairs.60 Several rationales have been suggested to explain the increased healing rates in meniscal repair performed in conjunction with ACL reconstruction. Combined acute ACL and meniscal injuries often occur in younger, active individuals as a result of trauma sustained during pivoting sports. The tear characteristics in this population, with a higher ratio of reparable tears as well as likely superior tissue quality, might account for the improved outcomes of meniscal repair. The rehabilitation program used in these combined injuries may also play a role. Finally, it has been proposed that the large intraarticular hemarthrosis produced by ACL disruption and by intra-articular drilling during surgery may improve access to marrow-derived factors and cells that improve the local milieu and assist in successful healing of the torn meniscus. Although the improved outcomes might attest to this final theory, its validity has yet to be completely proved.

Surgical Setup Most arthroscopic meniscal operations can be performed on an ambulatory basis under general, regional, or local anesthesia. A tourniquet is usually not necessary in isolated meniscal surgery. Either pump modulated or gravity flow systems can be used. Pump systems offer improved flow dynamics and superior visualization, but extravasation of fluid can occur, especially in acute knee injuries in which capsular disruption may have occurred. The patient is positioned supine. A leg holder or a lateral post is useful to provide a suitable fulcrum for placement of a valgus stress in order to better visualize the medial ­tibiofemoral compartment and posterior horn of the medial meniscus. In general, the lateral post offers an easier set-up and is sufficient for an isolated arthroscopic meniscectomy. A leg holder provides excellent, circumferential access to the limb and can be helpful for inside-out ­ meniscal

Figure 23B-10   View of the posterior horn of the medial meniscus and the meniscotibial attachments from the posterior compartment of the knee.

repairs or concomitant ligamentous reconstructions. Care must be taken with either device to avoid iatrogenic injury to the medial collateral ligament with overzealous stress application. Surgical instrumentation should include a cannula system with inflow and outflow through separate cannulae. Most meniscal work can be performed using a 30-degree arthroscope; however, a 70-degree scope can be useful. Manual instruments, including straight, up-going, and side-biting duckbill punches, as well as a motorized shaver, facilitate meniscal resection and contouring. Instrumentation used for meniscal repair is surgeon and technique dependent. A standardized, systematic approach should be developed for diagnostic knee arthroscopy in order to accurately and reproducibly identify normal structures and to identify and treat all pathology. A probe can be used to palpate the meniscus superiorly and inferiorly and, if an unstable tear is present, to displace it into the joint and into view. Care should be taken to comprehensively evaluate the periphery of the meniscus as well as the posterior horn and root. The arthroscope can be placed into the posteromedial or posterolateral compartment of the knee (either between the PCL and the medial femoral condyle or between the ACL and the lateral femoral condyle), affording an excellent view of the periphery of the posteromedial and posterolateral menisci (Fig. 23B-10). The use of a 70-degree arthroscope is helpful in viewing this area. Using a systematic approach, arthroscopic evaluation should provide a definitive means of identifying meniscal pathology. Knowledge of a wide variety of portals and surgical approaches is essential to the appropriate diagnosis and management of meniscal and other intra-articular disorders.

Meniscectomy After the determination has been made that a meniscal tear is irreparable, the surgeon should take a deliberate approach in performing a partial meniscal resection. First, by ­probing

Knee 1607

and visualization, the extent of meniscal injury and the tear configuration should be elucidated. A thorough evaluation should be performed to plan a resection that removes all unstable and irreparable portions but preserves as much residual meniscus as is appropriate. The type of resection is highly dependent on the tear geometry. The guiding principle in meniscal resection is to remove all unstable segments while preserving as much meniscus as possible, especially at the periphery. A systematic approach is paramount. All mobile fragments, especially those that can be displaced into the articular surface or flip under or over the residual meniscus, must be resected (Fig. 23B-11). A combination of manual instruments such as a meniscal punch or biter with a motorized shaver is helpful in removing tissue in a calculated fashion. We find an upgoing punch placed through an anteromedial portal, while viewing from the anterolateral portal, to be most useful for the posterior horn of the medial meniscus. A straight biter placed anteromedially appears to work best for the lateral meniscus. It is often helpful to view the meniscus alternately through the anteromedial portal and work through the anterolateral portal, especially to access the midbody and anterior portions of the medial meniscus at a right angle. During resection, care should be focused on both the meniscal resection and protection of the articular cartilage surfaces of the femur and tibia. Appropriately positioning the limb with application of controlled valgus stress to access the medial side and placing the limb in the “figure of four” position for the lateral side assists in visualization. The use of curved instruments and a curved shaver is also helpful in avoiding iatrogenic chondral injury. As the resection is carried out, the probe should be used ­frequently to determine the mobility of the residual meniscal rim. After a satisfactory resection has been completed, a motorized shaver is effective for fine-tuning the residual rim to smooth rough, frayed edges and remove free debris. Although some authors have reported a role for radiofrequency probes in meniscectomy and meniscal contouring,78,79 the potential for cellular injury and collateral damage to surrounding ligaments and articular ­ cartilage makes these devices unappealing for this objective. The final product should have as smooth a contour as possible, without any uneven, acute edges that ­ theoretically could propagate a new tear.

A

Meniscal Repair Meniscal repair was first reported in the literature in 188580 but was rarely performed until the modern era. As the significance of the meniscus and its roles within the knee have been elucidated during the past 30 years, techniques of meniscal preservation and repair have been developed. The earliest series of open meniscal repairs were reported by Dehaven in the early 1980s, with long-term followup over the ensuing two decades.66,81,82 After an initial ­evaluation with the arthroscope to determine suitability for repair with his technique, the meniscus was approached and exposed through a capsular incision made posterior to the medial or lateral collateral ligament. This facilitated direct repair of the tear to the peripheral meniscal rim and capsule with horizontal or vertical mattress sutures. High short- and long-term success rates were reported for repairs of vertical tears, especially in patients with ACL-stable knees.81,82 Advances in arthroscopic techniques have made open meniscal repair a procedure of mostly historical interest. Current options for meniscal repair include outside-in, inside-out, or all-inside arthroscopic repair techniques. Each is discussed later. Several fundamental principles of repair remain constant regardless of repair technique. After a determination that the tear configuration and location are suitable for meniscal repair, the first step is to obtain adequate visualization of the tear and access for instrumentation required to perform a repair. Next, the meniscal bed is prepared to stimulate a healing response. A small shaver (3.5 mm) can be used to abrade both sides of the tear to remove frayed edges and to débride the rim to stimulate a vascular response. Alternately, a rasp can be used to further prepare the meniscal bed and to abrade the synovial fringe superiorly and inferiorly to create bleeding. Abrasion of the synovial fringe is intended to stimulate the formation of a vascular pannus, which will migrate into the tear site and help produce a reparative response. After the tear site has been prepared, the tear must be adequately reduced and fixed with either sutures or appropriate implants. Finally, an appropriate rehabilitation protocol must be followed to protect the meniscus, promote nutrition of the meniscal and chondral tissue, and restore joint function.

B

Figure 23B-11   Complex meniscal tear with an inverted flap. A, Flap reduced with probe. B, After partial meniscectomy. Note the targeted resection of compromised tissue with contouring of the residual meniscus, leaving behind as much tissue as possible.

1608 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Biologic Enhancement Techniques Certain techniques have been used either alone or in combination with meniscal repair to stimulate a biologic healing response. These include synovial abrasion, trephination, and fibrin clot augmentation. In the treatment of incomplete tears or small stable tears, synovial abrasion or trephination can be used without formal repair with good success.52-54 Trephination theoretically provides vascular access channels to allow the proliferation of reparative, fibrovascular tissue into the tear site. Under arthroscopic visualization, multiple small, horizontal channels can be created using a spinal needle or a small-caliber trephine from outside-in through the periphery. The placement of a fixed number of perforations will theoretically provide neovascularization. However, there is a risk for disruption of the structural integrity of the meniscus with large vascular access channels. This technique can be used in isolation or combined with synovial abrasion or formal meniscal repair, or both. Another healing adjunct that deserves attention is the use of fibrin clot augmentation, which was first reported by Arnoczky and associates in an animal study.55 The fibrin clot appeared to act as a chemotactic and mitogenic stimulus as well as a scaffold for the reparative process, with filling of meniscal defects through the proliferation of fibrous connective tissue that modulated to fibrocartilaginous tissue. Since this publication, exogenous fibrin clot has been used as an adjunct to meniscal repair in tears with poor vascularity. The technique involves the creation of a fibrin clot by the gentle agitation of a sample of the patient’s blood to promote the formation of a semisolid clot. The fibrin clot is then placed at the tear site either by direct injection or fixed in place with the suture used for the meniscal repair. Improved rates of healing have been reported with fibrin clot adjuncts in tears with uncertain vascularity.56,57 Similar refined, more stable substrates, rich in platelets and certain growth factors, are under investigation for use in meniscal repairs.

Inside-Out Meniscal Repair The inside-out technique for meniscal repair was initially popularized in the United States by Scott and colleagues62 in the 1980s as an improvement on open repairs. Initial reports demonstrated excellent outcomes with high rates of healing and a low rate of complications. In this technique, double-limbed sutures are passed under direct arthroscopic visualization and retrieved through a counterincision with dissection to the joint capsule. Proper technique with planned incision position and careful dissection to the capsule are necessary to avoid risk to the neurovascular structures. The technique allows safe access to repair most of the meniscus; caution should be taken in far posterior tears because the risk for neurovascular injury increases. Far anterior tears are also difficult to access with this technique. Although all arthroscopic techniques have come into vogue, the inside-out repair remains the gold standard for repair of most tear variants and the method of choice for many surgeons. The patient should be positioned supine, with the operative extremity at 90 degrees of knee flexion and prepared

free to allow circumferential access. A leg holder is helpful; alternatively, a lateral post can be used. After complete diagnostic arthroscopy, the tear is visualized, and the defect is prepared for repair. The arthroscope is placed in the ipsilateral anterior portal (i.e., anteromedial for a medial tear, anterolateral for a lateral tear) for visualization, and the contralateral portal is used for the placement of sutures in order to direct the needles medially or laterally as opposed to posteriorly toward neurovascular structures. Zone­specific cannulas are available for use in this technique. Single- and double-barreled, straight, and curved cannulas can be used to access various portions of either meniscus and facilitate reduction of the tear and suture passage. A double-barreled cannula allows for faster and easier suture passage. However, we find the single-barreled cannula to be preferable for improved coaptation of the repair site, more accurate suture placement, and placement of vertical mattress suture configurations. A successful and safe inside-out repair hinges on precise exposure of the posteromedial or posterolateral joint capsule (Fig. 23B-12). Structures at risk include the peroneal nerve laterally and the saphenous nerve and its branches medially. Careful capsular exposure is crucial to avoid the tying of sutures around these structures and significant morbidity due to neural injury or neuroma formation. The incision is about 3 to 4 cm in length and is positioned posterior to the midaxis at the joint line, with three fourths of the incision distal to the joint line and one fourth above. Medially, the infrapatellar branch of the saphenous nerve can be encountered in the superficial tissue in the distal aspect of the incision; care should be taken to avoid iatrogenic injury to this structure. The medial dissection is carried down to the sartorial fascia, which is incised longitudinally at the anterior edge of the sartorius. Blunt finger dissection can then be carried out deep to the sartorius, gracilis, and semitendinosus, which are retracted posteriorly. The medial head of the gastrocnemius is then easily palpated and teased from the underlying posteromedial capsule. A Henning or popliteal retractor is placed deep to the gastrocnemius, fully exposing the joint capsule anteriorly and protecting the saphenous nerve, which remains posterior to the plane of dissection throughout. The lateral approach is similar, with the main difference being the relative lack of mobility of the lateral gastrocnemius tendon. After skin incision, sharp dissection is carried down to the iliotibial band and bicep femoris. The interval between these structures is developed through a longitudinal fascial incision, thus guaranteeing avoidance of the peroneal nerve located posterior to the biceps tendon. The biceps is retracted posteriorly and deep to this interval; the lateral collateral ligament and lateral head of the gastrocnemius are visualized. The lateral gastrocnemius must be freed from the posterolateral capsule, a process that is facilitated by beginning dissection distally at the level of the muscle and mobilizing the tendon moving proximally. A popliteal retractor can then be placed deep to the lateral gastrocnemius adjacent to the capsule. After the counterincision and approach are completed and the popliteal retractor is in place, suture passage can commence (Fig. 23B-13). Special double-limbed 2-0 nonabsorbable sutures attached to long straight needles are most commonly used. The surgeon supports the ­extremity

Knee 1609 Gastrocnemius m. Peroneal n. Plantaris n. Biceps tendon

Posterior join capsule

LCL

Gastrocnemius m.

PT

Lat. collateral ligament

A

B

Gastrocnemius m. Semimembranosus m. Semitendinosus m. Sartorius m.

Gracilis m.

Gracilis m. Semimembranosus tendon Semitendinosus tendon Gastrocnemius m. Medial collateral ligament

Sartorius m. Joint capsule

C

D

Figure 23B-12   Counterincision and exposure for lateral ��(A and B) and ��������� medial (C and D)����������������������������������� ���������������������������������� inside-out meniscal repair. (Redrawn from Noyes FR, Barber-Westin SD, Rankin M: Meniscal transplantation in symptomatic patients less than fifty years old. J Bone Joint Surg Am 87[Suppl]:149-165, 2005.)

and passes sutures while the assistant is responsible for maintaining the position of the retractor and retrieving the passed needles. After reduction of the tear using the cannula, one limb of the suture is passed through the central and peripheral sides of the tear and then the capsule until it hits the retractor. It is retrieved by the assistant using a needle holder. The second limb of the suture is then passed, preferably in a vertical mattress fashion, either through the meniscal tissue again or through the meniscocapsular junction. It is retrieved, the needles are cut off, and the sutures are tagged with a clamp. Multiple sutures should be passed at 3- to 5-mm increments from one end of the tear to another. Sutures on both the superior and deep aspects of the meniscus are helpful in order to most

accurately reduce and most rigidly fix the meniscal tear. Sutures are then individually tied, with the knee flexed 15 to 20 degrees, under direct visualization over the ­posterior capsule. A few technical notes should be emphasized. First and foremost, the ease and safety of this technique hinges on accurate placement of the counterincision, dissection, and retractor placement. If done properly, the passed needles should continually enter view and hit the back side of the retractor as they emerge from the capsule. If difficulty is encountered, the mistake is most commonly a retractor that is placed too superiorly. Second, the initial pair of sutures should be placed in the midportion of the tear through the superior aspect of the meniscus because this

1610 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E

F

Figure 23B-13   Inside-out meniscal repair. A, Schematic: vertical mattress suture placement into the superior surface. B, Schematic: vertical mattress suture placement into the inferior surface. C to F, Arthroscopic images demonstrating suture placement and final construct. (Schematics from Rubman MH, Noyes FR, Barber-Westin SD: Arthroscopic repair of meniscal tears that extend into the avascular zone: A review of 198 single and complex tears. Am J Sports Med 26:87-95, 1998.)

allows for the most accurate reduction of most tears. One trick that is helpful in the case of a difficult reduction is to pierce the inner flap with the needle and use it as a lever to reduce the inner flap to the outer fragment before passage peripherally. After two pairs of sutures have been passed superiorly, inferior sutures (tibial side) should be passed to augment the repair integrity. Whenever possible, sutures should be passed in a vertical mattress configuration because this construct has been consistently shown to have superior biomechanical strength.83-85 Sutures should be tied with the knee in only slight flexion because tying in greater degrees of flexion can malreduce the meniscus and coapt the posterior capsule, theoretically altering meniscal excursion and limiting knee flexion.

Outside-In Meniscal Repair The outside-in repair technique was developed by Warren as a method to decrease the risks for neurovascular injury associated with inside-out repair.65,86 This technique is particularly useful for repairing tears of the anterior portion of the meniscus but can also be used for fixation of meniscal allografts and for fixing fibrin clot into a tear defect to augment healing.65,86 There are a number of advantages associated with this technique. Suture passage does not require a rigid cannula to pass through the joint space, thus eliminating the risk for iatrogenic chondral injury as well as improving visualization because no instrumentation is

placed between the arthroscope and the tear. A posterior counterincision with a meticulous dissection is not necessary for this technique. A variety of suture configurations can be easily created, including vertical mattresses. Finally, owing to its versatility, this technique has also been applied to more difficult tear configurations, such as flap and radial tears, with reported success.57 The major disadvantage of this technique is that it offers poor access to tears located in the posterior regions of the meniscus because passage of needles in this area endangers surrounding structures. There also exists the theoretical potential for chondral irritation by suture material if a mulberry knot construct is used, although this has not been reported. An outside-in repair requires only 18-gauge spinal needles, a suture grasping device, and suture material. After tear preparation is completed, the meniscus is viewed with the scope and a spinal needle is introduced percutaneously through the meniscus, bridging the tear and exiting either through the superior or inferior surface centrally. Needle placement and reduction of the tear may be facilitated by the use of a probe placed through the anteromedial portal to provide counterpressure. A second needle is then placed, exiting at an appropriate position in reference to the first. At this point, one 0-0 or 2-0 polydioxanone (PDS) suture is shuttled through each needle and retrieved with the suture grasper out the accessory portal. Final fixation can then be performed in a few ways. In the traditional technique, the ends of both sutures are tied into mulberry knots and then pulled back into the joint, each knot exerting a reduction force across the tear (Fig. 23B-14). The sutures are then tied to each other over the capsule through a small counterincision. Alternatively, the sutures can be tied to each other using a standard square knot and pulled through the meniscus to create a standard mattress stitch. This can be made easier by the placement of a dilating knot that traverses the meniscus before the joining knot. Nonabsorbable braided suture, although not easily passed through the spinal needles, can be used to replace the PDS by shuttling it through in this fashion. Finally, a wire snare device can be placed through the second spinal needle or a suture retrieval instrument can be passed in place of the second needle and can be used to shuttle the suture around creating a mattress stitch that is then tied over the capsule through a small incision. This technique is excellent for the treatment of anterior meniscal tears, with clinical success rates of 90%.65,86

Arthroscopic or “All-Inside” Meniscal Repair Although most reparable meniscal tears can be effectively addressed using one of the above techniques, a number of all-inside arthroscopic techniques using sutures or specially designed implants have been developed over the recent two decades. Advantages of all-inside repairs include no requirement for counterincisions, large or small, minimal risk for neurovascular complications, and with the introduction of novel instrumentation and implants, decreased technical demands and reduced operative times. The initial generation of all-inside meniscal repairs mimicked arthroscopic repair techniques elsewhere in the body. Suture repairs of meniscal tears were performed with the use of shuttle devices and arthroscopic knot tying.

Knee 1611

Capsule Sutures Femur

Meniscus Tibia

A

B

C

D

Figure 23B-14   Outside-in meniscal repair. A, Diagram depicting suture placement through the superior or inferior margin of the meniscus. B, PDS suture placement through spinal needle. C, Suture shuttled through working portal. D, Repair construct with mulberry knots.

A ­ posterior portal was necessary for tears involving the posterior horn. These methods were difficult to master, technically demanding, and extremely cumbersome. Because of these shortcomings, a number of specially designed implants have been developed to facilitate allinside meniscal repair. These include rigid, biodegradable implants, such as the meniscal arrow, and newer suturebased implants, which incorporate a deployable rigid device attached to a suture construct, such as the T-Fix (Smith & Nephew Endoscopy, Andover, Mass) device and two popular, contemporary devices, the RapidLoc (Mitek, Westwood, Mass) and the Fas-T-Fix (Smith & Nephew Endoscopy). The meniscal arrow was one of a few similar rigid meniscal repair devices that became available in the mid1990s. It consists of a biodegradable (poly-l-lactic acid, or PLLA), dart-shaped implant that is deployed through a specially designed cannula. After tear preparation, the tear was reduced, and a pilot hole was made using a punch device perpendicular to the tear. With the cannula held in place, the device is then impacted into the pilot hole, fixing the tear edges to each other. The device has barbs distally for fixation to the rim fragment and a widened head to compress the central fragment. Initial enthusiasm for this implant was high owing to its ease of application, entirely arthroscopic placement, and lack of the need for tying arthroscopic knots. Furthermore, a number of early series suggested comparable healing rates

and outcomes when compared with inside-out repairs.87-90 However, a high number of reported complications, such as chondral injury due to implant abrasion (Fig. 23B-15) and chronic, refractory posterior joint line pain and tenderness due to the implants, diminished this fervor. Additionally, longer term clinical studies began to reveal a significant

Figure 23B-15   Severe chondral injury to the femoral condyle due to prominent meniscal arrows. (From Anderson K, Marx RG, Hannafin J, Warren RF: Chondral injury following meniscal repair with a biodegradable implant. Arthroscopy 16:749-753, 2000.)

1612 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E

F

Figure 23B-16   Meniscal repair using the Fas-T-Fix device. Probing of vertical tear, red-white zone, medial meniscus (A, B), Fas-TFix device (C), placement of the first limb of the device using the introducer (D), Repositioning for second limb placement (E), and final suture construct (F).

deterioration in outcomes with high failure rates and an unacceptably high frequency of complications.91-94 An evaluation of the longer follow-up series suggested that, in many of the patients who appeared to have had good outcomes over the short term, complete biologic healing of their tears had not occurred. Over time, implant degradation and migration led to a substantial frequency of re-tears and unacceptable morbidity secondary to chondral injury and implant-generated discomfort. Although rigid all-inside repair devices remain in use in some centers, these poor outcomes and high risk for complications, combined with the introduction of newer implants with technical improvements, has led to a steady decline in their overall usage. The newest generation of all-inside devices incorporates many of these technical advantages, with improvements aimed to enhance repair stability and avoid untoward implant-related complications. One such device is the RapidLoc, which incorporates a PLLA “backstop” that is

connected by 2-0 Ethibond suture to a tensionable “tophat” also made of PLLA or PDS. Each device is placed using an introducer with a shield used to protect the chondral surfaces during implantation. Once the “back-stop” is deployed, the tear can be reduced and compressed by tensioning the top-hat against the central superior surface. Early outcome studies have reported acceptable outcomes using this implant95-97; however, because of the ­ intra­articular position of the PLLA top-hat, the risk for ­chondral injury similar to that seen with the meniscal arrows remains and has been reported in the literature.98,99 The Fas-T-Fix device offers both biomechanical and clinical improvements to previous all-inside devices (Fig. 23B-16). This implant consists of two sequentially deployed 5-mm polymer bar anchors that are connected by a pre-tied 0-0 Ethibond suture. Available in both straight and curved configurations, the delivery system is introduced into the tibiofemoral joint with a protective sheath to avoid chondral injury. When the implant is in position and the meniscus is

Knee 1613

reduced, the sheath is removed, and one bar is deployed by penetrating through the peripheral meniscus and capsule. The introducer is withdrawn and then “reloaded” intraarticularly with the triggering device. A second position on the meniscus or above the meniscus on the capsule is chosen, and the second bar is deployed. The introducer is then removed, and a combined knot-pusher and suture-cutting device is threaded over the suture. The pre-tied locking knot is advanced until flush with the meniscus and an appropriate amount of tension can be placed on the suture construct. Excess suture is then cut, yielding the final construct. This implant offers a number of advantages over previous all-inside devices. First, only suture material is present in the joint; thus, the potential for chondral injury is low, comparable to other suture repairs. The instrumentation allows for the safe placement of multiple suture constructs, including vertical mattress and oblique sutures. Most of the body and posterior horns are easily accessible, and using the straight or curved introducers, sutures can be placed into both the superior and inferior surfaces of the meniscus. There are only a few clinical series to date evaluating this implant, but the initial outcomes have been promising.100,101 Furthermore, in a number of biomechanical studies, the Fas-T-Fix device has demonstrated equivalent strength to vertical mattress sutures and superior strength when compared with other all-inside devices.102-106

Special Circumstances Tears Involving Meniscal Attachments Unstable tears that involve the anterior or posterior meniscal root necessitate repair. The most common clinical scenario is a complete posterior detachment involving the posterior horn and posterior root. Root avulsions are more common in association with ligamentous injuries, especially high-energy multiligamentous knee injuries. However, posterior root injuries seen in association with concomitant ACL injury are frequently not unstable and can be left alone.49,50,76 Repair of an anterior or posterior root avulsion can be performed arthroscopically. After the bone at the anatomic insertion site has been prepared, nonabsorbable sutures are passed through the anterior or posterior root tissue. Two bone tunnels can be created using the ACL targeting guide or freehand. The sutures are then retrieved using a suture retriever and tied over a suitable bone bridge anteriorly. The remainder of the meniscal tear is then repaired using one of the above described techniques.

Fascicular Tears of the Lateral Meniscus Because of the unique anatomy of the lateral meniscus and the intimate association between the popliteus and the lateral meniscus posterolaterally, a distinct injury variant may occur in this area. Simonian and colleagues11 described the anatomy of this region and identified two distinct superior and inferior popliteomeniscal fasciculi, which when disrupted, will lead to meniscal instability and mechanical symptoms. This injury occurs in the periphery where the vascularity should tolerate an attempted repair, either by means of an inside-out repair, or using an all-inside

t­ echnique with an implant such as the Fas-T-Fix. Care should be taken to avoid incarceration of the popliteus by sutures when performing this type of repair.

Meniscal Cysts Meniscal cysts may occur periodically in association with a meniscal tear. Historically, lateral meniscal cysts were reported to have a three- to sevenfold higher prevalence than medial meniscal cysts; however, recent surveillance MRI evidence suggests that the rates are almost ­equivalent.107,108 There is some debate as to the pathogenesis of meniscal cysts. Myxoid deterioration of fibrocartilaginous meniscal tissue due to trauma or degeneration has been proposed. However, the current prevailing theory is that most meniscal cysts develop as a result of the expulsion of synovial fluid through certain meniscal tear variants, specifically horizontal tears, which act as one-way valves. In support of this hypothesis is the fact that most reported meniscal cysts have been found adjacent to horizontal meniscal tears. Meniscal cysts tend to be located anteriorly and become symptomatic due to mass effect. Patients generally present with a vague history of joint line pain and swelling. Mechanical symptoms are rare because of the type of tear associated. The cyst is often point tender and can be palpable if it is larger than a few centimeters in diameter. MRI and ultrasound are both effective diagnostic studies for the diagnosis of meniscal cysts; the differential diagnosis includes the more common popliteal or Baker’s cyst, ganglion cyst, bursitis, tenosynovitis of the pes anserine tendons, exostosis, loose bodies, and certain soft tissue sarcomas. Most meniscal cysts can be effectively treated intraarticularly. The meniscal tear is débrided, and the cyst can be accessed and decompressed through the tear. By removing the torn portion of meniscus and opening up the communication with the cyst, the one-way valve effect is resolved. In the uncommon situation of a cyst with no direct association with the meniscal lesion or an extremely large collection, open cystectomy is an option. With either approach, cyst recurrence is rare, and outcomes have been consistently favorable.109-112

Meniscal Variants and Discoid Meniscus The discoid meniscus is the most common anatomic meniscal variant. First described in 1889 by Young113 in a cadaveric specimen, the incidence of this anomaly has been estimated to be around 5% in the general population, with a higher prevalence among Asian populations.8,114-119 Most discoid menisci occur on the lateral side; however, discoid medial menisci have been sporadically reported in the literature.8,116,120 The incidence of bilateral discoid lateral menisci has been poorly defined but is thought to be as high as 20%. Although initially theorized to result from incomplete resorption of the central meniscus during an embryologic arrest in development, the current opinion regarding etiology is that discoid menisci arise as congenital anatomic variants. Watanabe and associates presented the most widely used classification system for this entity, describing three types of discoid lateral menisci based on morphology and tibial

1614 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

attachments, as noted on arthroscopy (Fig. 23B-17).121 Type I , the most common type in most series, is a complete discoid meniscus (covers the entire tibial plateau) with intact peripheral attachments. Type II is an incomplete discoid meniscus with intact attachments. This heterogenous group has inconsistent morphology, covering a variable percentage of the tibial plateau. Type III, also known as the Wrisberg ligament type, is characterized by absent normal posterior attachments with only the meniscofemoral ligament of Wrisberg acting as a posterior stabilizer. This anatomic abnormality results in a significant increase in meniscal mobility and can manifest clinically as instability, classically described as the “snapping-knee syndrome.” This is the least frequently reported type of discoid meniscus, with a range of 0% to 33% reported in the literature.117,122-125 Klingele and associates reported a 28.1% overall incidence of peripheral rim instability detected on arthroscopy in a series of more than 100 consecutive patients. In their series, the peripheral detachment was located most commonly in the anterior third of the meniscus, with posterior rim anomalies detected about one third of the time; peripheral instability was most common in younger patients with complete discoid morphology.118 Jordan has proposed a revised classification, which divides discoid menisci based on morphology, the presence of peripheral stability, the presence of a tear, and the presence or lack of symptoms.126 This system allows for a more accurate description of discoid meniscal type and advocates treatment based on the meniscal anatomy and pathology. In general, complete and incomplete discoid menisci with normal peripheral attachments tend to be asymptomatic. However, because of tissue variability and abnormal knee kinematics with high sheer stresses, discoid menisci are at an increased risk for the development of tears. Discoid menisci become clinically relevant owing to instability secondary to absent or abnormal peripheral attachments or in the event of a tear. Patient presentation can be variable. The classic presentation of the symptomatic discoid lateral meniscus is due to mechanical symptoms with a snapping or popping knee. This is more commonly seen in patients with a type III discoid meniscus. Stable discoid menisci become symptomatic in the event of a tear. Patients often present with mild, vague lateral joint line pain and swelling with or without an inciting event. Mechanical symptoms can be present but

are less common in this group. Examination will reveal joint line tenderness, a mild effusion, and in the case of a displaced tear or an unstable variant, palpable or audible clicking or popping or a block to extension. Radiographs should be obtained, which may reveal widening of the ­lateral joint space, flattening of the lateral femoral condyle, concave deformation of the tibial plateau, meniscal calcification, and tibial spine hypoplasia. Concomitant osteochondritis dissecans of the lateral femoral condyle has also been reported.127,128 MRI is ­confirmatory, demonstrating ­continuity of the anterior and posterior horns of the lateral meniscus (absent bow-tie) in three or more consecutive 5-mm cuts (Fig. 23B-18). Intrasubstance tears and displaced flaps can been well ­ visualized. However, unstable type III variants, especially those with relatively normal morphology, are more difficult to detect on MRI.127,129 This diagnosis should be made clinically and confirmed during diagnostic ­arthroscopy. Treatment options for the discoid lateral meniscus include observation, partial meniscectomy or saucerization, repair or reattachment of an unstable peripheral rim (with or without concomitant saucerization), and total meniscectomy. Appropriate treatment is determined based on the type of meniscal variant, the age of the patient, the presence or absence of a tear, and the severity and duration of symptoms. Asymptomatic discoid menisci are frequently identified incidentally during MRI evaluation or at arthroscopy and should be treated with observation. For symptomatic, stable, complete or incomplete discoid menisci (types I and II), the treatment of choice is arthroscopic “saucerization” (see Fig. 23B-18).124,125,130-135 This technique involves partial meniscal resection with removal of the abnormal central tissue, as well as any torn or degenerative meniscal tissue, while retaining a peripheral rim resembling a normal meniscus. Ideally, a rim width of 6 to 8 mm should be preserved in order to more closely reproduce meniscal anatomy and function and to avoid re-tear. Horizontal cleavage tears of the central substance, which are most commonly encountered, can be addressed with this technique. Arthroscopic saucerization can be technically challenging because visualization and manipulation of instruments is often difficult owing to the thickened meniscus and decreased joint space. Appropriate positioning and accurately planned portal placement are essential. Smaller instruments can be useful because many

Figure 23B-17   Watanabe classification for discoid lateral menisci. (From Kocher MS, Klingele K, Rassman SO: Meniscal disorders: Normal, discoid, and cysts. Orthop Clin North Am 34:329-340, 2003.)

Knee 1615

patients are children and have smaller knees. Saucerization is often aided by the presence of a flap or full-­thickness tear to allow for an entry point and a place to begin resection and contouring. As the procedure is carried out, a periodic assessment should be made to verify peripheral rim stability. Significant meniscal instability will require ­stabilization. The ideal treatment of a type III discoid lateral meniscus with an unstable rim is a combined saucerization with repair of the peripheral, reshaped meniscus to the capsule. This can be achieved using any of the previously described techniques for meniscal repair. After the outer edge of the meniscal rim and the inner border of the peripheral capsular tissue have been prepared with abrasion using a motorized shaver, a stable repair can be achieved using an inside-out technique posteriorly or an outside-in technique for anterior-third rim instability. Multiple sutures are often required because these variants tend to be particularly unstable. With the advent of improved devices and instrumentation, the role of all-inside repairs may increase, but at this point, most authors continue to advocate insideout repair in this subset of patients. Total meniscectomy had been the treatment of choice for discoid menisci for many years. Although it is no longer considered an appropriate intervention, in some clinical situations, meniscal preservation is not feasible. Long-term studies have reported variable clinical and radiographic outcomes in patients who have undergone total meniscal resection for this diagnosis.136-140 The summation of available evidence reveals fair to poor long-term clinical outcomes in patients after total meniscectomy. ­Radiographic

follow-up has demonstrated an unacceptable rate of changes consistent with degeneration and arthrosis of the involved compartment. Although a large series evaluating the long-term results of arthroscopic saucerization with or without peripheral repair has yet to be published, the results in a number of small series with shorter follow-up reveal improved clinical and radiographic results.124,125,130-135 In patients who have undergone total meniscectomy, close surveillance is essential to identify signs or symptoms associated with meniscal deficiency early so that the option of meniscal transplantation may be explored.

WEIGHING THE EVIDENCE: OUTCOMES OF MENISCECTOMY, PARTIAL MENISCECTOMY, AND MENISCAL REPAIR There is an abundance of basic science and clinical data that can assist the clinician in determining prognosis and appropriate management in patients with meniscal injuries. Long-term studies have evaluated the results 15 to 30 years after total meniscectomy in young patients,141,142 adding to the classic reports of King, Fairbank, and others.2-6 In general, patient subjective outcomes were fair to poor, with most reporting significant limitations in activities. Radiographic assessment revealed a high prevalence of gonarthrosis in these populations. The aggregate of these data and the evidence reported in discoid patients attest to unacceptable outcomes after total meniscal resection and provide the basis for the role of meniscal replacement.

A

B

C

Figure 23B-18   Discoid lateral meniscus. A, Diagnostic magnetic resonance imaging demonstrating lack of “bow-tie” appearance of the lateral meniscus in three successive 5-mm cuts. B, Arthroscopic view of complete discoid lateral meniscus. C, Arthroscopic view of completed saucerization procedure.

1616 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

As would be expected, better long-term outcomes have been reported after partial meniscectomy. Burks and colleagues evaluated a series of 146 patients treated with arthroscopic partial meniscectomy with 15-year outcomes. They reported 88% good to excellent clinical outcomes in patients with ACL-intact knees. Factors including age, gender, and medial versus lateral involvement had no effect on clinical outcomes. They did note mild radiographic deterioration when compared with the contralateral, normal knee, especially in those with concurrent ACL deficiency and in patients with combined varus alignment and partial medial meniscectomy.143 Schimmer and colleagues presented a series of 119 patients treated with arthroscopic partial meniscectomy who were evaluated an average of 12 years after surgery.144 They reported excellent clinical outcomes in 78.1% of patients, with better outcomes in patients with isolated meniscal injuries and no chondral damage on arthroscopy. Shelbourne and Dickens assessed a series of 135 patients radiographically a mean of 12 years after arthroscopic partial meniscectomy and noted only minimal joint space narrowing.145 Although the clinical outcomes after partial meniscectomy appear to be more favorable than after total meniscectomy, in the young, active patient, every effort should be made to preserve the meniscus. There is great inconsistency when comparing one partial meniscal excision to another. General variables such as patient age and activity level, as well as anatomic variables such as mechanical alignment, ligament insufficiency, and condition of the articular cartilage, are all important considerations. Also of considerable importance is tear location and configuration because these will dictate the amount of meniscus that requires resection. In a biomechanical study, Lee and associates evaluated the difference in stress concentrated on the tibiofemoral articular surface associated with increasing resection of meniscal tissue.18 As would be anticipated, they noted a steady increase in the stress as more of the posterior horn of the medial meniscus was resected. They further reported that a segmental resection of meniscus, as is often necessary for radial and flap tears that extend peripherally, was biomechanically equivalent to a total meniscectomy. In this circumstance, the full-thickness disruption of the circumferential fibers led to structural alterations to the function of the meniscus, which approximated a total meniscal excision. Based on these data, every consideration should be given to meniscal preservation. In the management of an irreparable meniscal tear, the clinician should have dual goals in mind, to resect as much tissue as is required to effectively resolve the patient’s symptoms while at the same time retaining as much of the meniscal structure as possible to preserve meniscal function. Success rates after meniscal repair are dependent on many factors. A number of outcome instruments have been employed to assess for successful meniscal repair. These include clinical evaluation, subjective assessment tools, arthrography, MRI, and second-look arthroscopy. Success or failure can be concluded based on one or many of these with quite a bit of variability. Patient age, meniscal tissue quality, and tear configuration are important criteria as well. The presence or absence of associated injuries, especially ACL injury, has been noted to have an immense

impact on outcomes after meniscal repair. Finally, the technique used for repair has a significant bearing on likelihood of a successful outcome. The results of inside-out meniscal repair have been extensively reported. Scott and associates published a series of 178 inside-out repairs in 167 patients.62 They documented a 92% clinical success rate with resolution of symptoms, with 80% of patients returning to their previous level of activities. All patients were evaluated with arthrography or arthroscopy; 61.8% of repairs had completely healed, with additional 16.9% healing incompletely, but stable. Combined ACL reconstruction and meniscal repair, as well as a narrow peripheral rim (<2 mm), were criteria that correlated with a high likelihood of complete healing. Cannon published a similar series of 164 patients who had an inside-out repair and were evaluated with second-look arthroscopy or arthrography. 61 He reported an 87% clinical success rate and a 71% anatomic success rate (completely or incompletely healed but stable); rim width of less than 2 mm and short tear length positively correlated with complete healing in this series. Johnson and associates published a long-term clinical and radiographic assessment of a series of 50 isolated inside-out meniscal repairs, reporting a clinical success rate of 76% (based on the absence of pain, a normal examination, and the lack of further knee surgery).63 Radiographs revealed no ­significant increase in arthritic changes when compared with the normal contralateral extremity. The impact of impaired vascularity on healing and the clinical outcomes has been extensively investigated by Noyes in a series of publications. In 1998, Rubman and colleagues evaluated a series of 198 meniscal tears, with a major segment of the tear located in the central avascular zone.69 These patients were treated with inside-out meniscal repair with the placement of multiple sutures, spaced 3 to 4 mm apart in vertical, oblique, or horizontal ­ mattress ­ configurations, based on the morphology of the tear. In this group, 80% of patients were noted to be clinically asymptomatic 42 months after surgery. Followup arthroscopy in 91 of these patients revealed 25% of the tears to be completely healed and 38% partially healed.69 In a follow-up series, Noyes and Barber-Westin reported a clinical success rate of 75% in 71 meniscal repairs performed in patients younger than 20 years of age with tears extending into the avascular zone.68 Noyes and BarberWestin also evaluated a group of patients 40 years of age or older with meniscal tears extending into the avascular area that were treated with repair. They noted an 87% clinical success rate in this cohort, with 26 of 30 patients asymptomatic for tibiofemoral joint symptoms an average of 33 months after surgery.70 Despite the expected lower rates of complete healing in these “avascular” tears, the high rate of ­clinical success led the authors to conclude that repair of certain white-white zone meniscal tears is favorable in select patients, especially those who are younger and highly active. Voloshin and associates evaluated a small cohort of patients who underwent attempted repeat meniscal repair after failure of an initial attempt.67 They noted a 72% survival rate with improvement in knee outcome scores in this group. The apparent discrepancy between the clinical and anatomic success rates in the literature bears emphasis.

Knee 1617

In general, about 85% of patients undergoing meniscal repair should be anticipated to realize a good or excellent long-term clinical outcome with symptom resolution and return to activities. However, in a subset of these patients, objective evaluation using arthrography, MRI, or second-look arthroscopy reveals either partial, incomplete healing of the tear or a complete lack of healing. In a series from our institution, van Trommel and colleagues reported that noncontrast MRI was more accurate than arthrography in discriminating partial from complete healing.146 Although the significance of this inconsistency between clinical symptoms and tear healing is not known, it would seem logical that the likelihood of reinjury would

be higher in patients in whom only partial repair has occurred. Based on the accumulated evidence, the literature favors an attempt at meniscal repair because the benefits of a potentially functional meniscus outweigh the risks for failure and revision surgery. The indications for repair can be extended, especially in younger and more active individuals, to more complex tears and those that involve the avascular zone. The jury is still out, however, on the longterm success rates of newer suture-based all-inside meniscal repair implants. Early results are encouraging, but it will take time to effectively determine equivalence to the gold standard inside-out repairs.

Authors’ Preferred Method We find that the initial diagnosis of a meniscal injury can be made in most cases with a thorough history and physical examination. We pay careful attention to the patient’s age, activity level, occupation, and athletic endeavors, as well as a detailed injury history and current symptoms. Our standardized, comprehensive physical examination was detailed previously. Findings such as an effusion, joint line tenderness, posterior pain with hyperflexion or squatting, and mechanical symptoms are all indicative of a meniscal injury. We also pay close attention to evidence of concomitant ligamentous abnormalities as well as the mechanical alignment of the limb in these patients. We routinely obtain an MRI in all patients with a suspected meniscal disorder; the utility of this study outweighs its cost in our opinion. We find an MRI to be helpful as both a confirmatory study and in preoperative planning. Using specified sequences, MRI aids in the delineation of the tear location and configuration, helps predict tear reparability, and can assist in ruling out articular chondral injury, which may present in a similar fashion to a meniscal tear. Our principal goal is to choose the appropriate intervention in each patient, with a detailed plan made before entering the operating room. In young patients, we make every attempt at meniscal preservation. In the setting of a potentially reparable meniscal tear, prompt diagnosis and early surgical intervention are critical. In our institution, meniscal surgery is almost always performed on an outpatient basis with a regional anesthetic. Perioperative antibiotic prophylaxis is administered before incision. A pneumatic thigh tourniquet is placed, but rarely inflated for isolated meniscal surgery. The entire extremity is prepared and draped free, and a leg holder or lateral post is used to facilitate exposure. After a ligamentous examination under anesthesia, a comprehensive diagnostic arthroscopy is carried out. For most procedures, we find that only two anterior portals are necessary; a separate outflow cannula placed laterally or medially through the quadriceps can carry with it unnecessary morbidity and is usually not required. A methodical evaluation of the patellofemoral, medial, and lateral compartments is carried out, with careful evaluation of the chondral surfaces, ligaments, and menisci. We evaluate each

meniscus both visually and with tactile probing throughout its entirety in order to detect any tear or disruption. If a tear is encountered, its location, vascularity, and configuration are evaluated, and a decision is made as to the requirement for repair or resection. In general, truly stable vertical tears measuring less than 10 mm in length and partial-thickness tears are treated with abrasion or trephination, or both, to stimulate healing. Longer, unstable vertical tears are treated with a formal repair. We have traditionally used an inside-out technique for repair of tears involving the posterior or body segments and an outside-in technique for anterior tears. Recently, we have transitioned to using an all-inside approach with the Fas-T-Fix system for many vertical tears. In some scenarios, such as very young patients with large bucket handle tears or unstable Wrisberg-type discoid menisci, we still implement an inside-out repair or a hybrid repair using inside-out sutures combined with Fas-T-Fix implants for the more posterior aspects of the tear. For anterior tears, an outside-in repair remains the best approach. In young, active patients, we attempt to repair bucket handle tears in the white-white zone that involve a large central fragment, especially if they are undergoing concomitant ACL reconstruction. We consider repair in more complex tear configurations on a case-by-case basis, especially in young patients. In general, radial, oblique, horizontal, and degenerative tear configurations are infrequently reparable and are ­treated with partial meniscectomy. We approach meniscal resection with extreme care to remove all damaged and ­nonfunctional meniscus and retain as much meniscal ­ architecture and ­function as possible. Resection is accomplished using straight and up-going meniscal punches placed through the ipsilateral and contralateral anterior portals. The residual meniscus is contoured using a small-diameter (3.5-mm) ­motorized shaver. A curved shaver is often helpful to access the posterior meniscal tissue, while avoiding scuffing of the articular cartilage of the femoral condyle. We do not routinely use radiofrequency devices for meniscal surgery because of the potential for collateral damage to the articular cartilage and healthy meniscal tissue.

1618 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

POSTOPERATIVE MANAGEMENT AND REHABILITATION Historically, the postoperative rehabilitation protocols used after meniscal repair often employed immobilization or extended periods of protected weight-bearing to theoretically protect the repair and shield it from disruptive forces during the initial stages of healing. Because of a lack of convincing clinical evidence, a number of authors have advocated more aggressive, “accelerated” rehabilitation protocols that emphasize early motion, diminished weight-bearing limitations, and an early return to pivoting sports.147,148 The reported clinical outcomes and incidence of reinjury with these accelerated protocols were equivalent to and in some cases superior to the standard rehabilitation programs. Our published rehabilitation guidelines for patients after meniscal repair consist of three phases.149 In designing a rehabilitation strategy that promotes an optimal environment for healing, we take into account factors such as tear location, tear configuration, type of repair, and any concomitant procedures to construct an appropriate approach that allows for a successful outcome and a timely resumption of unrestricted activity. Range of motion is encouraged immediately in every patient, with limitations of flexion in some repair types (radial or complex tear configurations). Weight-bearing is typically progressive, with early weightbearing allowed after repairs of vertical or bucket handle tears because compressive loads act to close these tear types. Weight-bearing after repairs of complex or radial tears, as well as after meniscal allograft transplantation, is often limited to toe-touch ambulation for the first 4 weeks because compressive loads act to distract these repair types and can lead to early failure. Communication between the surgeon and the therapist is critical, especially in the early phases of rehabilitation. The first phase of our rehabilitation guidelines begins in the recovery room and continues until 6 weeks after surgery. Patients are placed in a double-upright knee brace that is locked in extension during ambulation. Passive and active-assisted range of motion exercises are instituted immediately, with the goal of accomplishing full extension and progressive flexion to 90 degrees. Flexion may be limited to 90 degrees in repairs of the posterior horn because flexion has been noted to cause the posterior horn of the meniscus to displace from the capsule. Weight-bearing is progressed as dictated by repair type; patients are typically instructed to discard their crutches when a nonantalgic gait is demonstrated. Quadriceps strengthening is also initiated in the recovery room with isometric exercises and straight leg raises and continues throughout this phase. By the end of phase one, the patient should have range of motion of 0 to 90 degrees, pain-free ambulation, and improved quadriceps strengthening as demonstrated by the ability to perform a straight leg raise without an extensor lag. Phase two, which spans weeks 6 to 14, focuses on restoring normal range of motion of the involved knee and improving muscle strength and proprioception to the level needed to perform daily activities. The brace is discontinued during phase two, after the patient has demonstrated effective quadriceps control for protection. Pool therapy is often useful in this phase, as is cycling.

The third and final phase spans weeks 14 to 22. During this period, exercises are progressed, with the goals of optimizing functional capabilities and preparing the patient for a safe return to sporting activities. Running is initiated early in this period along with agility and sportspecific exercises. Advanced strengthening using isokinetic and plyometric training are also introduced.

Criteria for Return to Play Upon successful completion of the third phase of rehabilitation, patients are cleared for unrestricted sports activities. There is no consensus in the literature about when a patient is definitively ready to return to competition. Objective criteria that we find useful include the singleleg hop test and the crossover hop test. A patient should demonstrate less than a 15% deficit in these tests, as well as in isokinetic testing, when compared with the nonoperative leg before returning to play. Full, symmetrical range of motion, lack of an effusion, and absence of meniscal symptoms are other important criteria. Most patients are able to return to unrestricted sporting activities by 6 months, if not earlier; elite athletes are often able to resume activity even earlier because of greater preoperative muscle strength and conditioning.

Complications of Meniscal Surgery The overall complication rate of knee arthroscopy has been estimated to be about 2.5%.150 General complications can include hemarthrosis, infection, thromboembolic disease, neurovascular injury, and complications related to anesthesia. The rate of infection after knee arthroscopy is less than 1 per 1000 cases. Risk factors include longer surgical times, multiple procedures, prior surgery, and the use of intra-articular, intraoperative corticosteroids. The rate of deep venous thrombosis has been reported to be as high as 18% after knee arthroscopy. Fatal pulmonary embolism after knee arthroscopy is a rare but serious complication. Risk factors for the development of thromboembolic disease include advanced age (>40 years), obesity, smoking, a history of previous thrombosis, and others. Routine use of anticoagulant medications such as aspirin during the perioperative period is sound, especially in high-risk patients. The complication rate of meniscal surgery has been reported to be 1.2%.150 Complications specifically related to these procedures include repair failure, implant-related complications, arthrofibrosis, neurovascular injury, and iatrogenic chondral injury. Rates of failure of meniscal repair range from 5% to 30%. This inconsistency is due to variations such as tear location, repair type, patient age and tissue quality, and the use of differing measuring tools to quantify success and failure. A survey of the literature reveals good to excellent clinical outcomes of meniscal repair in the 90% range; however, anatomic assessments of the repair using second-look arthroscopy or arthrography have demonstrated complete healing in a smaller percentage, especially in repairs involving the avascular zone.60-63,68-70 MRI evaluation after meniscal repair has also revealed a high prevalence of asymptomatic patients

Knee 1619

with evidence of a persistent defect.151 Thus, there appears to be a subset of patients, with either incomplete or no healing after meniscal repair, who maintain a good clinical outcome. Implant-related complications vary from one technique to another, with the highest rates of persistent joint line discomfort and chondral injury attributed to the use of rigid implants, such as the meniscal arrow. The risk for arthrofibrosis can be minimized by technical considerations, such as preventing suture entrapment of the soft tissues of the posterior capsule by appropriate placement and knee positioning and by following a structured rehabilitation protocol as described. Neurovascular injury is a rare but serious complication. The saphenous nerve and its branches and the common peroneal nerve are most at risk. Meticulous surgical technique with appropriate needle or implant placement and accurate retractor positioning minimize this risk. Iatrogenic chondral injury can also be avoided by precise surgical technique and the use of small-diameter and curved instruments to access remote structures.

Meniscal Deficiency and Meniscal Allograft Transplantation In some cases, meniscal preservation is not feasible, despite the best of efforts. Meniscal deficiency, due to either total meniscectomy or segmental meniscal excision, can lead to progressive deterioration of the articular cartilage with resultant arthrosis, pain, and limitation of activity and function. Meniscal replacement with allograft tissue has been developed as a viable surgical option for this clinical situation.

Patient Evaluation A thorough and carefully planned approach is essential in the diagnosis and management of meniscal deficiency. These are some of the more complex patients that an orthopaedic sports specialist will see because they tend to

have had multiple previous knee injuries, have undergone multiple procedures, and frequently present with other abnormalities about the knee that need consideration when planning treatment. Patients frequently present with vague joint line pain with activities referred to the involved compartment. Swelling, crepitation, and episodes of giving way may also be reported. Physical examination should include evaluation of gait, alignment, range of motion, the presence or absence of an effusion, ligamentous stability, and the skin and soft tissues, paying attention to previous surgical incisions. Joint line tenderness to palpation and the presence of provocative signs are highly suggestive findings. Radiographic studies are extremely useful in these patients because they have often undergone multiple previous procedures and can have concomitant issues that need to be addressed. Radiographs, including weight-bearing anteroposterior, lateral, 45-degree flexion posteroanterior, and a Merchant view, should be routinely obtained to evaluate for joint space narrowing, osteophytes, and other signs of joint arthrosis. Long-cassette hip-to-ankle films of both extremities aid in determining the mechanical alignment. An MRI should be obtained in all patients with this entity (Fig. 23B-19) because it can provide indispensable information regarding the extent of meniscal excision, amount of residual meniscal tissue, and status of the subchondral bone and articular cartilage.

Indications Meniscal transplantation is indicated in patients younger than 50 years with meniscal deficiency, who present with pain localized to the affected compartment with activities of daily living or sports. Normal axial alignment and joint stability are also required and, if abnormal, must be addressed. Combined meniscal and ACL deficiency is frequently encountered; successful meniscal transplantation necessitates a stable knee; thus, ACL reconstruction should be performed in tandem. Long-standing meniscal deficiency can be seen in concert with, and in many cases can lead to, progressive varus or valgus malalignment. In

Figure 23B-19   Preoperative sagittal and coronal magnetic resonance imaging (MRI) slices in a patient with meniscal deficiency status after total lateral meniscectomy. Preoperative MRI evaluation can assist in surgical planning by quantifying the amount of residual meniscus and evaluating the condition of the chondral surfaces.

1620 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

this scenario, a corrective osteotomy should be performed before meniscal replacement in a staged fashion. The most commonly encountered contraindication to meniscal transplantation is the presence of advanced arthritic disease, such as Outerbridge grade III or IV cartilage changes, flattening of the femoral condyle, and marked osteophyte formation. Focal chondral defects may be addressed simultaneously to meniscal transplantation with a resurfacing procedure, such as osteochondral autograft transfer or autologous chondrocyte implantation. However, in the presence of more diffuse, severe chondral disease, the failure rate of meniscal transplantation is unacceptably high. Flattening of the femoral condyle leads to abnormal stress distribution on the transplanted meniscus and graft extrusion. Other contraindications to meniscal replacement include inflammatory arthritis, a history of knee infections, obesity, and skeletal immaturity. There are a number of nonsurgical alternatives in the symptomatic management of postmeniscectomy patients. Options in these patients include unloading braces, activity modification with encouragement of nonimpact activities, anti-inflammatories and other oral agents, and injectable viscosupplements. Meniscal transplantation is indicated in patients who have failed a sufficient attempt at nonsurgical management.

Meniscal Allograft Basic Science, Procurement, Processing, and Sizing The molecular and structural changes that occur within a meniscal allograft over time have been investigated. ­Arnoczky evaluated cryopreserved meniscal allografts in a canine model, revealing normal gross appearance, effective peripheral healing of the allograft to the capsule by fibrovascular scar, and normal cellular histology and metabolic activity at 3 months.152 In another canine study, the same author demonstrated repopulation of fresh frozen allograft menisci with host-derived cells originating from the synovium. He also observed alterations in the material properties of the allograft meniscus, which raised some concerns regarding structural integrity and long-term function.153 Jackson and associates reported peripheral healing, revascularization, cellularity, and incorporation of allograft menisci in sheep, but also noted biochemical changes in the extracellular matrix that called into question the longterm function of meniscal allografts.154,155 Rodeo and colleagues evaluated biopsies of human meniscal allografts by histology an average of 16 months after implantation. They noted allograft repopulation by cells that appeared to be derived from the synovium and matrix remodeling. They further noted a mild immune response in some samples obtained during second-look arthroscopy and the presence of histocompatability antigens in the meniscal allograft at the time of transplantation, indicating the potential for an immune response that could be elicited by the graft.156 The American Association of Tissue Banks has defined a protocol for donor screening, allograft procurement, and graft processing. Allograft meniscus procurement occurs within 12 hours of death, or within 24 hours, provided the body has been stored at 4° C. The tissue is harvested, with or without bone, under aseptic conditions after stringent donor screening and testing have been performed.

After the tissue has been harvested, there are four means of tissue processing: fresh, cryopreserved, fresh frozen, and freezedried or lyophilized. Fresh grafts are harvested within 12 hours of death and maintained at 4° C in lactated Ringer’s solution. Fresh allografts maintain a high percentage of donor cell viability but can be kept for only a short period (about 1 week). This time constraint presents significant logistical difficulties with regard to proper graft sizing, serologic testing, and implantation, limiting the clinical utility of fresh meniscal allografts. Freeze-dried or lyophilized allografts are processed by dehydration and freezing in a vacuum, which allows for unlimited storage. However, because of the alterations in the biomechanical properties induced by lyophilization, this processing technique is also rarely used. Most meniscal allografts implanted today are either cryopreserved or fresh frozen. Cryopreserved allografts undergo controlled freezing in a cryoprotectant solution (glycerolbased medium), which helps to preserve some donor cell viability. The increased expense associated with this technique may not be justified because it is not known how long the transplanted cells survive after transplantation. Furthermore, histologic samples have shown early graft repopulation by host cells.152-156 Fresh frozen allografts are processed by a rapid cooling to −80° C and stored at this temperature. This method, although deleterious to donor cell viability, does not adversely affect the biomechanical properties of the graft and, more importantly, has not been demonstrated to affect allograft ­survival and clinical outcomes. Although the risk for disease transmission due to allograft tissue is minute, some tissue banks use protocols for secondary sterilization of allografts. Gamma radiation has been used historically, but the required dose of irradiation necessary to eliminate virus contaminants such as HIV and hepatitis C is beyond the threshold whereby the mechanical properties of the allograft tissue is compromised. In an early report, Noyes noted poor clinical outcomes in a series of patients after implantation of irradiated meniscal allografts.157 Ethylene oxide sterilization has also been employed but can lead to synovitis with recurrent effusions in some patients. Newer secondary sterilization methods have been developed and are used by specific tissue banks. Although enthusiasm is high, there are limited data to attest to the efficacy of these techniques. Accumulated biomechanical data and experimental evidence in animal trials attest to a chondroprotective role for meniscal replacement.158-160 Although the tolerance of the knee to allograft size mismatch has not been firmly established, many authors have postulated that a 5% or smaller millimeter discrepancy may result in alterations in knee mechanics and failure. In a recent study, Dienst and associates reported the biomechanical consequences of size mismatch due to oversized or undersized grafts, finding that accurate allograft sizing has significant effects on the contact mechanics of the knee.161 They reported that a size mismatch of greater than 10% led to unacceptable alterations in stress placed on the articular cartilage or the graft. A number of protocols have been described for accurate allograft sizing, using radiographs, computed tomography, or MRI.162-165 The most commonly used technique is that of Pollard and colleagues, who use bony landmarks on orthogonal plain radiographs to determine allograft length and width.162 Although seemingly more suitable,

Knee 1621

e­ valuations of the utility of MRI for allograft sizing have revealed a tendency for underestimating meniscal size with accuracy not significantly superior to plain radiographs.163 This is an area in which further investigation is warranted.

Surgical Technique Meticulous preoperative planning is critical in approaching these patients because they frequently present with multiple abnormalities that may need to be addressed at the same time or in a staged fashion. Allograft preparation is performed in concert with the initial arthroscopic preparation and can vary based on technique. Although using an all soft tissue allograft may have benefit with regard to potential disease transmission, we use a composite allograft with preserved bony attachments at the anterior and posterior horns because biomechanical investigation has emphasized the importance of the fixation at these sites for meniscal

allograft function.160 For medial meniscal allografts, we employ a bone plug technique because a portion of the tibial footprint of the ACL occupies the space between the horns on the medial side; for lateral allografts, a bone bridge or “slot” technique is preferred because of the proximity of the insertion sites on the lateral side and the absence of an at-risk intervening structure (Fig. 23B-20). The use of a bone bridge has theoretical advantages, including maintaining the anatomic distance between the horns and preserving circumferential hoop stresses. However, to create a trough medially and avoid injury to the ACL footprint, the trough needs to be medialized, which risks an overly medialized graft placement and altered mechanics. The bone bridge technique may be used for medial meniscal transplantation that is performed concomitantly to ACL reconstruction. In our institution, meniscal allograft transplantation is performed under regional anesthesia with the patient in a

A

C

E

B

D

F

Figure 23B-20   Medial (A, C) and lateral (B, D) meniscal allografts. Medial allografts are prepared with cylindrical bone plugs, whereas lateral meniscal allografts are prepared with a slot that is broader on its inferior surface (demonstrated in D). Second-look arthroscopy of lateral meniscal allograft (E, F).

1622 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

position similar to an inside-out meniscal repair. ­Diagnostic arthroscopy is carried out first in order to ­ evaluate the magnitude of meniscal deficiency and confirm the condition of the chondral surfaces. After the decision has been made to proceed with transplantation, the residual meniscal tissue is resected back to a 2-mm peripheral rim and the anterior and posterior horn regions are débrided to bone. A small parapatellar arthrotomy is then created either medially or laterally through which preparation and graft passage can be achieved. A posteromedial or posterolateral counterincision is also created, with dissection to the capsule for later inside-out meniscal repair. Next, the bony preparation is begun. On the medial side, transosseous tunnels are created at the anterior and posterior horns over a guidewire placed with a tibial ACL guide. On the lateral side, a trough is created from anterior to posterior using the surgeon’s preferred technique. Described techniques include various shaped slots or keyholes. Our current approach, called the dovetail technique, creates a trapezoidal trough with a wider base using a reamer followed by cutting osteotomes to create a slot that closely matches the allograft bone block. Care must be taken to ensure proper depth and to protect the posterior neurovascular structures by maintaining a posterior cortical rim. A minor, selective medial or lateral notchplasty of the lower, interior femoral condyle can facilitate visualization for these preparatory steps as well as graft passage. After the recipient site and the allograft have been suitably prepared, graft passage is carried out. For a medial allograft, heavy, nonabsorbable sutures are passed through each bone plug, incorporating the tissue of the horn insertion. These sutures are used to dock the bone plugs into each tunnel and can be tied to each other over an anterior bone bridge for fixation. A mattress suture placed at the intersection of the posterior horn and body using meniscal repair needles is passed through the peripheral rim and capsule and can be used to provisionally reduce the meniscus under the femoral condyle and ensure anatomic placement. At this point, the knee is brought from flexion to extension to assess for appropriate positioning of the allograft during motion. Upon satisfactory allograft placement and fixation of the horns, multiple meniscal sutures are placed using an inside-out repair technique. As firm circumferential repair is critical for healing, we like to place an abundance of sutures in vertical and horizontal mattress configurations to rigidly fix the meniscal periphery. The anterior sutures can often be safely and easily placed using a curved needle under direct visualization through the anterior arthrotomy. Passage of a lateral allograft is done in a similar fashion. Sutures can be placed through the bone bridge and retrieved through drill holes created in the slot to assist in both graft reduction and fixation of the bone bridge. As an alternative, fixation of the bone bridge can be achieved using an interference screw.166 The bone bridge is reduced, and an initial meniscal suture is used to reduce the allograft as described for the medial side. Circumferential sutures are then placed to firmly fix the allograft. The literature has not reached a consensus regarding the postoperative rehabilitation after meniscal transplantation. We have developed an approach that protects the graft in the early phases of healing and promotes early motion to prevent contracture and promote cartilage and

meniscal nutrition. Patients are placed in a brace and kept at toe-touch weight-bearing for the first 4 weeks. Motion is initiated immediately by the therapist as well as at home with a continuous passive motion device, with flexion limited to 90 degrees for the first 3 weeks and an emphasis on achieving full extension. Weight-bearing is progressed to full from week 4 to week 6, and flexion is advanced at week 4. Our approach to strengthening, stretching, agility, and sports-specific activities in these patients mirrors the rehabilitation approach to patients after a complex meniscal repair, with a return to running at 6 months. Although we have had a few patients resume high-load activities successfully, a return to activities involving cutting, jumping, and pivoting is not recommended after meniscal transplantation.149

Outcomes Milachowski performed the first allograft meniscal transplantation in 1984 and published the first outcomes in 20 patients.167 In their series, they noted encouraging results at 14 months, with fresh frozen grafts being superior to freeze-dried grafts.167 Since this early report, there have been numerous series presented in the literature.157,167-182 Short-term outcome series have demonstrated relatively consistent improvement in pain and function,167-171, 175-177,179-181 with functional limitations and some residual disability in many patients, especially as they are followed out long term.171-173,178,181,182 The outcomes after isolated meniscal allograft transplantation appear to be favorable179,180 in the absence of other knee abnormalities, as do combined ACL reconstruction and meniscal allograft transplantation.175,176 Although the literature has demonstrated that meniscal transplantation is relatively successful in alleviating pain and improving knee function, the effects of meniscal replacement on the natural history of meniscal deficiency and the onset of significant arthrosis have been more difficult to assess. Recent evidence has demonstrated effective prevention of the progression of arthritis in most patients and suggests a role for meniscal replacement in select patients with more advanced chondral degeneration.181 Similar to meniscal repair, good clinical outcomes may be seen despite anatomic failure of allograft healing. The long-term fate in this circumstance is guarded. In interpreting this evidence, it must be reinforced that meniscal allograft transplantation is truly a salvage procedure. It represents an option for young, active, sympto­ matic patients who do not have many satisfactory alternatives. There is a reasonable success rate; however, the procedure is technically challenging and carries with it a high risk for complications and failure. Hopefully, technical advances and the advent of other alternatives for meniscal replacement will help to improve the long-term outlook for these patients.

AREAS OF FUTURE INTEREST Although significant progress has been made over recent decades in the understanding of meniscal structure and function and in the surgical treatment of meniscal injuries, there remains substantial room for development. Contemporary investigation has focused on two main areas: the

Knee 1623

improvement of surgical techniques and implants and the development of novel cellular and tissue-based ­applications to augment meniscal repair and improve on current options for meniscal replacement. As previously highlighted, there has been a great recent interest in the development of practical, effective, and less demanding means for arthroscopic meniscal repair. Although the consistent outcomes of inside-out repair have set a standard, all-inside arthroscopic repair offers many potential advantages. However, one need look no further than the early generation of rigid implants to demonstrate that enthusiasm for new devices and techniques must be mitigated until there exists sufficient evidence to confirm equivalent or superior efficacy and safety. The newest generation of all-inside devices, specifically suturebased implants such as the Fas-T-Fix, appear to have great promise. Further investigation over the coming years will demonstrate whether this technique will replace inside-out repair as the standard for meniscal repair. The area of the greatest future interest in the treatment of meniscal disorders is in the implementation of cell-based therapies and tissue engineering. The use of autologous fibrin clot by Arnoczky and Warren was an early attempt at cellular modulation of meniscal repair. Investigation continues into the use of cells, including fibrochondrocytes, chondrocytes, mesenchymal stem cells, scaffolds, and growth factors to augment the biology and improve the healing of meniscal tears in the avascular zones.182,183 Perhaps the most remarkable area of current and future interest involves the development of a feasible synthetic meniscal replacement. In recent years, researchers have attempted to engineer potential off-the-shelf biologic meniscal substitutes containing a porous scaffold impregnated with cellular components to stimulate host incorporation, degradation, and repopulation. Preclinical experiments using animal models and investigational trials in humans have investigated the efficacy of tissue-engineered meniscal substitutes, including small intestinal submucosa scaffolds,184 a collagen-based meniscal implant,185-188 a polyurethane scaffold,189 and a hydrogel meniscus composed of cross-linked polyvinyl alcohol and polyvinyl pyrrolidone.190 At the present time, there remain challenges that need to be surmounted in the development of a ­tissueengineered meniscus. A suitable implant must closely match the biologic, structural, and material properties of the normal meniscus, must accurately approximate meniscal size and shape, must lack immunogenicity, and must be technically feasible to implant. Despite these obstacles, it is probable that this area of biologic therapy and tissue engineering represents the future direction for the treatment of meniscal injuries. C l Knowledge

r i t i c a l

P

o i n t s

of the anatomy, microstructure, and vascular supply of the menisci is critical in understanding the functions of the meniscus within the knee, such as load transmission, improving joint congruency, shock absorption, providing passive stability in the ACL-deficient knee, lubrication, and nutrition.

l Accurate diagnosis of most meniscal injuries can be made based on patient history and physical examination findings. Radiographs and MRI are valuable in confirming the diagnosis and identifying concomitant pathology. l Although most meniscal tears are not amenable to repair, every effort should be made in meniscal preservation, especially in young, active patients. l Important factors in determining the suitability for attempted meniscal repair include tear location, tear configuration, patient age and activity level, and the presence of concomitant knee pathology, specifically ACL injury. l In irreparable tears, the surgeon’s goal should be to remove all nonviable meniscal tissue while retaining as much meniscal architecture as possible because biomechanical data suggest that segmental meniscal resection approximates total meniscectomy. l Clinical outcomes for inside-out meniscal repair approach 90% with regard to symptom resolution; however, anatomic rates of complete healing are poorer. l Although inside-out repair remains the standard for the treatment of reparable tears involving the posterior horn or body, newer generation suture-based all-inside techniques show promise. Long-term study of these techniques is warranted. l Meniscal allograft transplantation is an option in patients with symptomatic meniscal deficiency; however, it is technically difficult, and results are dependant on many factors, especially concomitant chondral disease.

S U G G E S T E D

R E A D I N G S

Arnoczky SP, Warren RF: Microvasculature of the human meniscus. Am J Sports Med 10(2):90-95, 1982. Burks RT, Metcalf MH, Metcalf RW: Fifteen-year follow-up of arthroscopic partial meniscectomy. Arthroscopy 13:673-679, 1997. Dienst M, Greis PE, Ellis BJ, et al: Effect of lateral meniscal allograft sizing on contact mechanics of the lateral tibial plateau: An experimental study in human cadaveric knee joints. Am J Sports Med 35(1):34-42, 2007. King D: The healing of semilunar cartilages. J Bone Joint Surg 18:333-342, 1936. Lee SL, Aadalen KJ, Malaviya P, et al: Tibiofemoral contact mechanics after serial medial meniscectomies in the human cadaveric knee. Am J Sports Med 34(8):1334-1344, 2006. Noyes FR, Barber-Westin SD, Rankin M: Meniscal transplantation in symptomatic patients less than fifty years old. J Bone Joint Surg Am 86:1392-1404, 2004. Rubman MH, Noyes FR, Barber-Westin SD: Arthroscopic repair of meniscal tears that extend into the avascular zone: A review of 198 single and complex tears. Am J Sports Med 26(1):87-95, 1998. Scott GA, Jolly BL, Henning CE: Combined posterior incision and arthroscopic intra-articular repair of the meniscus: An examination of factors affecting healing. J Bone Joint Surg Am 68:847-861, 1986. Shoemaker SC, Markolf KL: The role of the meniscus in the anterior-posterior stability of the loaded anterior cruciate-deficient knee: Effects of partial versus total excision. J Bone Joint Surg Am 68:71-79, 1986. Tenuta JJ, Arciero RA: Arthroscopic evaluation of meniscal repairs: Factors that effect healing. Am J Sports Med 22(6):797-802, 1994.

R eferences Please see www.expertconsult.com

1624 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

C

Medial Ligament Injuries 1. Medial Collateral Ligament Injuries in Adults Manuj Singhal, Jayesh Patel, and Darren Johnson

Medial ligament injuries of the knee are often assumed to be only medial collateral ligament (MCL) injuries. However, the medial ligament includes not only the MCL but also posteromedial structures that play a vital role in the stability of the knee. The management of the MCL has evolved over the past 30 years. Most isolated MCL injuries are treated conservatively, with a rare role for surgical intervention. However, the treatment of MCL with anterior cruciate ligament (ACL) injury and the timing of ACL reconstruction continue to be controversial. This chapter sections describes the complex anatomy of the medial knee including the often forgotten posteromedial corner, evaluation of the knee, treatment of medial ligament injuries, and the role of rehabilitation.

ANATOMY AND BIOMECHANICS The anatomy of the medial stabilizing structures of the knee extends from the medial aspect of the patellar tendon to the medial edge of the posterior cruciate ligament.

Both static and dynamic structures act in coordination to provide resistance to valgus and external rotation stresses. Dynamic medial stabilizers include the pes anserinus, semimembranosus, medial head of the gastrocnemius, and vastus medialis. Superficial MCL, deep MCL, and posterior oblique ligament are the major static stabilizers. In their classic paper, Warren and Marshall divided the medial capsuloligamentous structures of the knee into three layers because they found the ligaments in their dissections to be condensations, not discrete structures.1 Layer I is a superficial fascia that invests the sartorius and quadriceps proximally and blends with the pes anserinus and tibial periosteum distally (Fig. 23C1-1). Layer II is defined by the parallel fibers of the superficial medial ligament. This ligament originates from the medial femoral epicondyle with an average length of 10 to 12 cm and has a broad insertion site on the anteromedial metaphyseal area of the tibia 5 to 7 cm below the joint line. In addition to the parallel anterior fibers of the superficial medial ligament, there are obliquely and posteriorly oriented fibers ­joining the femur

Sartorius Vastus medialis

vm

s Medial patellar retinaculum

Great saphenous vein

Saphenous nerve

Gastrocnemius, medial head

A

B

Figure 23C1-1   A and B, Medial aspect of the knee: layer I. vm, vastus medialis; s, sartorius muscle. (From Scott WN [ed]: Insall and Scott Surgery of the Knee, 4th ed. New York, Churchill Livingstone, 2005, p 41.)

Knee 1625

to the tibia (Fig. 23C1-2). Layer III comprises the capsular layer that thickens and forms the deep MCL or “middle capsular ligament” beneath the superficial medial ligament that also originates from the medial femoral epicondyle. The deep ligament extends from the femoral condyle to the meniscus (meniscofemoral) and from the meniscus to the tibia (meniscotibial). Anteriorly, the superficial and deep ligaments are discrete structures, but posteriorly, layers II and III merge to form the posteromedial corner (Fig. 23C1-3). Warren and Marshall’s work placed little emphasis on the posteromedial corner. Muller recognized the importance of the structures posterior to the medial ligament and stated, “One thing is certain: despite its close topographic relationship to the MCL, the posteromedial corner is fundamentally different in nature and function from the tibial collateral ligament itself.”2 Hughston and Eilers described a discrete anatomic structure in the posteromedial capsule called the posterior oblique ligament (POL), a thickening of the capsular ligament attached proximally to the adductor tubercle of the femur and distally to the tibia and posterior aspect of the capsule.3 This anatomically separate structure fans out into three arms or extensions distally: (1) the tibial arm attaches close to the posterior edge of the tibial articular surface; (2) superior or capsular arm, which is continuous with the posterior capsule and blends with the oblique popliteal ligament; and (3) a poorly defined superficial or distal arm, which attaches to the semimembranosus ­tendon and tibia (Fig. 23C1-4). The POL is firmly attached to the posterior horn of the medial meniscus, and one half to two thirds of the fibers run directly from the femur to the tibia. In addition to the POL, the posteromedial corner includes the semimembranosus expansion, the oblique popliteal ligament, and the posteromedial horn of the

medial meniscus.5 The semimembranosus has five extensions into the posteromedial corner and capsule: (1) the pars reflexa passing beneath the MCL and inserting on the tibia; (2) the posteromedial tibial insertion; (3) the oblique popliteal ligament insertion; (4) expansion to the POL; and (5) popliteal aponeurosis expansion (Fig. 23C1-5).2,5 The superficial and deep MCLs, along with the posteromedial structures, work in coordination to provide stability against valgus and external rotation stresses. The attachment of the MCL places it near the center of rotation of the knee. As the knee flexes, the anterior, parallel fibers are tensioned because the femoral attachment site is rotated upward with flexion, and the posterior fibers are relaxed (Fig. 23C1-6). Conversely, the POL is tense in extension and loose in flexion. Biomechanical studies have shown that the superficial MCL (parallel fibers) is the primary stabilizer to valgus loading.6,7 Grood and associates showed that at 25 degrees of flexion, the superficial MCL provided 78% of the restraint to valgus stress, and at 5 degrees, the contribution was 57%.6 As the knee extends, the posterior capsule and POL tighten, providing increasing restraint to valgus stress. At 5 degrees, the POL provides 18% medial stability and at 25 degrees only 4%. Similarly, the deep ligament accounted for 8% at 5 degrees and 4% at 25 degrees. The cruciate ligaments provided the remaining stabilizing force with increasing amounts as the knee was extended. This study shows the increased awareness the physician should exhibit with increased laxity in extension owing to the possible rupture of the cruciates and posteromedial complex. Warren and colleagues also showed the importance of the superficial MCL in stabilizing against valgus and external rotation stresses.7 They sequentially sectioned the deep ligament, POL, and superficial ligament and found that the superficial ligament was the primary stabilizer against

Adductor tubercle Semimembranosus

Patellofemoral ligament Anterior joint capsule

p

Posterior oblique a

Semitendinosus

Superficial medial collateral ligament Sartorius (cut) Gracilis

A

B

Figure 23C1-2   A and B, Medial aspect of the knee: layer II. a, anterior parallel fibers; p, posterior oblique fibers of the superficial medial collateral ligament. (From Scott WN [ed]: Insall and Scott Surgery of the Knee, 4th ed. New York, Churchill Livingstone, 2005, p 43.)

1626 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Sartorius (cut) Vastus medialis

Gracilis (cut)

Adductor magnus

a

Semitendinosus (cut)

Medial patellar retinaculum (cut)

Capsule

d

c

Semimembranosus

a

Deep medial collateral ligament

Superficial medial collateral ligament

Gastrocnemius, medial head Popliteus

B

A

s

d

Figure 23C1-3   A and B, Medial aspect of the knee: layer III. a, anterior parallel fibers of superficial medial collateral layer 2;   d, fibers of deep medial collateral ligament; c, capsule; s, superficial medial collateral ligament that has been reflected. (From Scott WN [ed]: Insall and Scott Surgery of the Knee, 4th ed. New York, Churchill Livingstone, 2005, p 46.)

v­ algus and external rotation stresses. After the deep ligament and POL had been cut, sectioning the superficial MCL increased the medial joint opening 5 to 7 mm and increased external rotation 200% to 300%. Sectioning studies by Warren and colleagues7 and Grood and associates6 evaluated the static stabilizers but did not address the dynamic stabilizers of the knee, such as the semimembranosus, pes anserinus, and vastus medialis. Various studies have attempted to look at the role of the musculature in injuries to the MCL. Practically speaking, normal muscle activity protects the knee against injury. The quadriceps is regarded as the most important for protection against valgus stress. Through the retinaculum, quadriceps contraction causes the anteromedial and anterolateral parts of the joint capsule to become tense. For example, Goldfuss and coworkers showed that contraction of the hamstrings and quadriceps increased stiffness to valgus stress by 48%.8

In addition, Pope and colleagues looked at the effect of valgus stiffness with sartorius and vastus medialis contraction.9 Tests were done with the muscles quiescent and with contraction of the sartorius or vastus medialis. The stiffness of the medial complex increased by 108% with contraction of the pes anserinus and 164% with the vastus medialis. Along with the quadriceps and pes anserinus, it has been theorized that the semimembranosus provides dynamizing stabilization to the posteromedial corner. Previous studies have shown that the POL and posteromedial corner are lax in flexion.2,3,7 However, when the knee flexes, the semimembranosus contracts, tensing the three arms of the POL and providing both a kinetic and static stabilizing effect. In addition, contraction of the semimembranosus tenses the tibial arm of the POL, retracting the posterior horn of the medial meniscus, preventing impingement of the meniscus as the knee flexes (Fig. 23C1-7).

Knee 1627 Capsular arm, POL

Gastrocnemius bursa Oblique popliteal ligament

Tibial arm, POL

Capsular arm, semimembranosus

Tibial collateral ligament

Semimembranosus Direct arm, semimembranosus

Semimembranosus tendon

Insertion that blends with the POL

4 1

Pars reflexa

2

3

Arcuate complex

5 Inferior arm, semimembranosus Superficial arm, posterior oblique ligament

Anterior arm, semimembranosus

Figure 23C1-4   The posterior oblique ligament (POL). Note the more posterior origin, expansile insertions, and relationship with the posteromedial capsule and semimembranosus expansions. (Redrawn from Sims WF, Jacobson KE: The posteromedial corner of the knee: Medial-sided injury patterns revisited. Am J Sports Med 32:337-345, 2004.)

The anatomy of the medial side of the knee is complex and involves multiple structures that work synergistically to provide static and dynamic stability. These functional and biomechanical studies show that valgus and external rotation instability is not solely dependent on the MCL.

Direct insertion

Oblique popliteal ligament Popliteal muscle

Insertion that blends with the popliteal fascia

Figure 23C1-5   The semimembranosus expansions. The five insertions are (1) pars reflexa, (2) direct posteromedial tibial insertion, (3) oblique popliteal ligament insertion, (4) expansion to posterior oblique ligament (POL), and (5) popliteus aponeurosis expansion. Note the investment into the POL. (Redrawn from Sims WF, Jacobson KE: The posteromedial corner of the knee: Medial-sided injury patterns revisited. Am J Sports Med 32:337-345, 2004.)

CLASSIFICATION Extension

See Table 23C1-1.

EVALUATION Similar to other orthopaedic injuries, the physician must rely on a thorough history, detailed physical examination, and diagnostic studies to make a diagnosis.

B

B

C

90° flexion C

History The history from the patient will depend on whether the injury is witnessed by the physician on the sidelines or in clinic. Most of these injuries present in the office setting. It is important to ask when the patient was hurt and how. The mechanism of injury should be elicited as thoroughly as possible. Typically the mechanism of injury is caused by a blow to the lateral aspect of the leg or lower thigh, such as a clipping injury in football or a noncontact injury from cutting, pivoting, or twisting. Also skiers are more prone to medial-side injuries with 60% of all ­skiing knee injuries affecting the MCL.10,11 In addition, it is important to ask the patients about pain, onset of swelling, ­ability

A

A

Figure 23C1-6   Diagram showing the long fibers of the superficial medial collateral ligament. Points A and B are at the anterior border of the long fibers. Point C is 5 mm posterior to point B. (Redrawn from Warren LF, Marshall JL, Girgis F: The prime static stabilizer of the medial side of knee. J Bone Joint Surg Am 56:665, 1974.)

1628 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

POL

A

Semimembranosus Area of attachment to adductor tubercle Tibial collateral ligament

Posterior oblique ligament

Semimembranosus

Mid-third capsular ligament Area of attachment to adductor tubercle Tibial collateral ligament

Medial meniscus

Mid-third capsular ligament

Meniscotibial portion of posterior oblique and oblique popliteal ligaments

B

Posterior retraction of meniscus with contraction of the semimembranosus

Figure 23C1-7   A, Bird’s-eye view of the proposed dynamizing action of the semimembranosus. The large arrow represents tension created in the posterior meniscocapsular complex by the semimembranosus. Note the ability of the semimembranosus to tension the posterior oblique ligament (POL) and aid in posterior meniscal retraction, represented by the small arrow. B, Intracapsular orientation of the posteromedial corner structures showing the proposed dynamizing action (arrow) of the semimembranosus. Note the relationship of the semimembranosus capsular expansion, the POL, and the posteromedial meniscus. (Redrawn from Sims WF, Jacobson KE: The posteromedial corner of the knee: Medial-sided injury patterns revisited. Am J Sports Med 32:337-345, 2004.)

to ­ ambulate, the sensation of a “pop,” and presence of a deformity necessitating a reduction, such as patellar or knee dislocation. In addition, past history or previous surgeries should be elicited.

Physical Examination Ideally, the examination of the knee should occur at the time of injury before the onset of muscle spasm. However, most of these injuries are examined in the office

setting after some time has elapsed from the injury. A thorough knee examination includes observation of the patient’s gait, documentation of the neurovascular status, palpation of the knee for tenderness, swelling, ecchymosis, and assessment of stability. The physician should follow some basic principles: (1) make the patient as relaxed as possible to assess the ligaments and muscles, (2) perform the physical examination as gently as possible, and (3) examine the uninjured knee before ­assessing the injured knee.

Knee 1629

   TABLE 23C1-1  Grading System for Medial Collateral Injuries6 Grade

Description

I II III

0- to 5-mm opening, with 2 mm being physiologic 6- to 10-mm opening >10 mm opening with valgus stress at 30 degrees of flexion

The patient’s gait should be observed as the patient walks into the room or at some point during the examination. Gait may be misleading because patients with a complete MCL tear may walk with a barely perceptible limp. Hughston and colleagues found that 50% of athletes with grade III injuries could walk into the office unassisted and reported that a complete disruption of the medial compartment can occur “without subsequent significant pain, effusion, or disability for walking.”12 However, patients with an MCL tear may exhibit a vaulting-type gait in which the quadriceps is activated to allow for medial stabilization. This differs from the patient with an ACL or meniscus tear who may walk with a bent knee gait because of an effusion. Similar to any orthopaedic injury, the neurovascular status of the limb should be assessed. Pedal pulses should be palpated along with assessing sensation over the dorsum, plantar, and first web space of the foot. Compartments should be felt to rule out compartment syndrome, and the ability to dorsiflex and plantarflex the ankle and great toe should be assessed. On the skin, the physician should look for edema, effusion, and ecchymosis to help localize the site of injury. It is important to differentiate between localized edema and an intra-articular effusion. Isolated MCL injuries usually have localized swelling, and intra-articular pathology, such as an ACL or peripheral meniscal tear, may indicate a hemarthrosis. Severe medial complex injuries with an ACL tear frequently have no evidence of effusion because the capsular rent is large enough to allow extravasation of fluid. In addition, on examination, if there is hemarthrosis, the examiner should exclude other causes such as a torn cruciate, patellar dislocation, osteochondral fracture, and peripheral meniscal tear. Along with assessment of swelling, palpation of the anatomic sites of attachment can provide clues to the diagnosis. The entire course of the MCL should be palpated from proximal to distal. Pain at the medial femoral epicondyle signifies injury at the femoral insertion of the MCL. With tibial-sided injuries, patients have pain along the proximal tibia below the pes anserinus adjacent to the tibial tubercle. Mid-substance tears exhibit pain at the joint line, which also occurs with meniscal tears, posing a diagnostic dilemma. Hughston and colleagues showed that point tenderness can accurately identify the location of injury in 78% of cases, and localized edema can identify a tear in the medial side 64% of the time.12 A valgus injury that disrupts the MCL can also result in lateral meniscus tears or osteochondral fracture to the lateral condyle or lateral tibial plateau, so palpation of the lateral joint line should also be performed. Valgus stress testing at 30 degrees of knee flexion is still the gold standard for assessing damage to the MCL. This

test should be performed with the foot in external rotation because increased instability will be noted if the knee moves from internal to external rotation. For larger patients, 30 degrees of flexion can be achieved by dropping the foot a few inches over the table. The examiner then grasps the ankle and applies a valgus stress with the other hand resting on the fibular head to assess the amount of opening and the quality of the end point compared with the uninjured side. The laxity of the MCL can be recorded based on a grading system or the amount of opening. Based on the Noyes classification, 5 to 8 mm of medial opening signifies a significant collateral ligament injury with “impairment of the ligament’s restraining effect.”6 The grading system has three grades: (1) stress examination produces little to no opening with pain along the line of the collateral ligament; (2) some opening to stress occurs but with a firm end point; (3) there is significant opening of the joint with no end point. After assessing the degree of opening, a repeat valgus stress should be performed with the examiner palpating the medial meniscus to assess if it subluxates in and out of the joint, indicative of injury to the meniscotibial ligament.13 In addition to valgus testing in flexion, opening of the medial joint should be assessed with the knee in full extension. The cruciates, POL, posteromedial capsule, and MCL all contribute to stability in full extension. Asymmetric joint opening compared with the contralateral side should alert the physician to the possibility of a combined MCL injury with a cruciate tear or posteromedial ­complex injury. The ACL should be assessed with Lachman’s test because the pivot shift is difficult to ­perform owing to guarding and the loss of the pivot axis with abduction instability. In ­addition, the posterior cruciate ligament and lateral ligament should be examined. Along with cruciate injury, patellar instability and tearing of the vastus medialis obliquus are associated with laxity in full extension. Hunter and colleagues14 found 18 of 40 laterally displaceable patellas on stress radiographs in patients with medial-sided ­injuries and a 9% to 21% incidence of damage to the extensor mechanism with medial ligament injury. In ­addition to ­valgus testing at 30 and 0 degrees, Slocum’s modified ­anterior drawer test and anterior drawer in external rotation should be tested to assess for medial-sided injuries (Table 23C1-2).

Imaging Radiography, arthrography, magnetic resonance imaging (MRI), and arthroscopy can provide information in knee injuries. All knees should receive radiography with anteroposterior, lateral, and sunrise views. These radiographs should be evaluated for occult fractures, lateral capsular sign (Segond’s fracture), ligamentous avulsions, old ­PellegriniStieda lesions (old MCL injury) (Fig. 23C1-8), and loose bodies. In adolescent and pediatric patients, stress radiographs help differentiate between physeal injuries versus ligamentous injuries. MRI without contrast is the imaging study of choice for evaluating MCL tears because it is less invasive and provides detail for lesions of the medial meniscus, the superficial MCL, POL, posteromedial complex, and semimembranosus tendon (Fig. 23C1-9). In addition, MRI is beneficial in assessing injuries to anterior and posterior

1630 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������    TABLE 23C1-2  Methods for Examining the Medial Collateral Ligament Examination

Technique

Valgus stress at 0 and 30 degrees

Illustration

Grading

Significance

Valgus force applied to tibia while stabilizing the femur. This should be done at 0 and 30 degrees of flexion and should be compared with the opposite leg.

Grade I: 0- to 5-mm opening, firm end point Grade II: 5- to 10-mm opening, firm end point Grade III: 10- to 15-mm opening, soft end point

Slocum’s modified anterior drawer test

Valgus force in 15 degrees of external rotation and 80 degrees of flexion

This test is positive if there is a noticeably increased prominence of the medial condyle compared with the other side.

Anterior drawer test in external rotation

Anterior drawer test at 90 degrees of knee flexion with an external rotation applied to proximal tibia

This test is positive if there is a noticeably increased anterior translation of the medial condyle.

Opening at 30 degrees occurs from isolated medial collateral ligament (MCL) injuries. Valgus stress at 0 degrees is associated with other ligament tears (anterior cruciate ligament, posterior collateral ligament, or posterior oblique ligament). The disruption of the deep MCL allows the meniscus to move freely and allows the medial tibial plateau to rotate anteriorly, leading to an increased prominence of the medial tibial condyle. A disruption of the MCL alone should not lead to an increased anteromedial translation. An increased anteromedial translation indicates an anteromedial rotatory instability that involves an injury of the posteromedial structures.

cruciate ligaments, meniscus, and osteochondral structures. Loredo and associates showed that intra-articular contrast may help highlight and better define the structures of the posteromedial complex but still concluded the assessment of the posteromedial complex was difficult.15 They found that the posteromedial complex was best visualized on the coronal and axial images. Indelicato and Linton stated that MRI can provide advantages in four circumstance: (1) when the status of the ACL remains uncertain despite physical examination, (2) when the status of the ­meniscus is in

­ uestion, (3) when surgical repair of the MCL is indicated q and localization of the tear will help limit the exposure, and (4) when an unexplainable effusion occurs during rehabilitation.16 However, MRI does not always provide concrete diagnosis, and the clinical examination becomes the deciding factor. Examination under anesthesia is another tool the physician can use to assess the injury pattern in patients who present late, or in patients in whom the office examination and MRI do not provide a ­diagnosis. Norwood and coworkers found on examination under anesthesia that

Figure 23C1-8   A and B, PellegriniStieda lesion. (Reproduced with permission from Pavlov H: Radiology for the orthopedic surgeon. Contemp Orthop 6:85, 1993.)

A

B

Knee 1631

Figure 23C1-9   Magnetic resonance image showing medial collateral ligament tear.

18% of patients had anterolateral rotatory instability that was not suspected preoperatively.17 In addition to MRI, arthrograms can be used to evaluate meniscal disease and capsular tearing with extravasation of contrast material. Kimori and colleagues found arthrography to be more useful than arthroscopy in diagnosing tears of the meniscotibial and meniscofemoral ligaments.18 With the increased use of MRI, arthroscopy is used infrequently as a diagnostic tool. ACL and meniscal tears may be identified on MRI. Also, it is rare to find an intrasubstance medial meniscal tear in an isolated MCL rupture because meniscocapsular separation occurs, and the fulcrum to load the medial compartment and tear the medial meniscus is lost.

TREATMENT OPTIONS Treatment of the MCL and medial-sided knee injuries can be divided into operative and nonoperative approaches. Numerous factors, including the timing, severity, location, and associated injuries such as an ACL tear, need to be taken into account when formulating a treatment plan. The MCL has the greatest capacity to heal of any of the four major knee ligaments because of its anatomic and biologic properties.19,20 As a result of multiple biomechanical, clinical, and functional studies, the trend has been toward a conservative, nonsurgical method for most MCL injuries. Recently, posteromedial corner injuries have been recognized as a separate entity from MCL injuries and may need to be addressed more aggressively because of rotational laxity and instability.5 Isolated tears of the MCL, grade I and II, do well with nonoperative management. Routinely, partial tears are treated with temporary immobilization and protected weight-bearing with crutches. Once the swelling subsides, range of motion, resistive exercises, and progressive weightbearing are initiated. Nonsteroidal anti-inflammatory drugs can be used to help with pain and swelling. Studies have shown no deleterious effect of nonsteroidal drugs on ligament healing.21 Numerous authors have shown excellent results with nonoperative treatment of grade I and II MCL tears.22-25 Ellsasser and associates looked at 74 knees

in professional football players and achieved a 98% success rate with a nonoperative protocol.22 They had strict inclusion criteria to ensure isolated MCL injury: (1) up to grade II laxity with a firm end point in flexion, (2) no instability to valgus stress in extension, (3) no significant rotatory or anteroposterior subluxation, (4) no significant effusion, and (5) normal stress radiographs. In this series, patients were treated with crutches, no brace, and progressive weightbearing. Ellsasser thought, based on his experience, that by 1 week patients should progress to full extension, no effusion, and decreased tenderness. The players returned to football in 3 to 8 weeks. The only failure occurred in a patient with an osteochondral fracture that was found later. Derscheid and Garrick performed a prospective study looking at 51 grade I and grade II MCL injuries in college football players.23 They used a nonoperative rehabilitation protocol with a knee immobilizer initially. Players with a grade I injury returned to full participation at an average of 10.6 days, and players with grade II injury returned at an average of 19.5 days. At long-term follow-up, these patients showed slight increases in medial instability. Injured knees had a higher incidence of reinjury than control knees, but this was not statistically significant. Bassett and associates24 and Hastings25 studied the use of cast brace in treating isolated MCL ruptures. Both studies found early return to athletics with the use of the cast brace. Nonoperative treatment varies from casting to functional bracing to no bracing, and good outcomes occur with all three forms of treatment. Management of grade III injuries remains much more controversial. Grade III injuries not only involve complete disruption of its fibers but also are frequently associated with additional ligamentous injuries. Fetto and Marshall found an 80% incidence of concomitant ligamentous injuries with a grade III MCL tear, with 95% of the associated injuries being an ACL tear.26 Early authors recommended primary repair for grade III injuries. O’Donoghue stressed the importance of immediate repair of complete tears of the MCL.27 Hughston and Barrett supported primary repair of all medial structures, including the superficial MCL and POL.28 They believed that repair and advancement of the posterior oblique ligament was key to restoring medial stability. Their results were good to excellent in 77% to 94% of patients. Muller reported 65% good and 31% excellent results in repair of isolated grade III MCL injuries.2 He repaired the superficial MCL avulsion with screw and washers and intrasubstance tears with a combination of approximation and tension-relieving sutures. In addition to Hughston, O’Donoghue, and Muller, Collins29 and Kannus30 have written that surgical intervention is necessary for complete ruptures of the MCL. Even though early authors demonstrated good results with surgical repair of the MCL, recent literature has focused on the nonoperative management of grade III MCL injuries. Fetto and Marshall were among the first to assess outcomes after nonoperative treatment of grade III MCL injuries.26 They studied 265 MCL injuries and found that patients with grade II injuries did much better than grade III injuries (97% compared with 73%). Initially, in their study, all grade III injuries received operative intervention. However, there were some patients with grade III injuries that were not operated owing to skin lesions and infection.

1632 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

At follow-up, patients with operative treatment of isolated MCL ruptures had no improved outcome compared with the nonsurgical group. This incidental finding led the way for more prospective studies to investigate the role of nonoperative treatment in isolated grade III MCL injuries. Indelicato prospectively compared operative to nonoperative treatment of isolated third-degree ruptures.31 All patients underwent examination under anesthesia and arthroscopy to rule out any other pathology, such as ACL and meniscal tears. He found objectively stable knees in 15 of 16 patients treated operatively and in 17 of 20 patients treated nonoperatively. Both groups followed a rigid rehabilitation protocol including casting at 30 degrees of flexion for 2 weeks and then 4 weeks longer in a cast brace with hinges that allowed motion from 30 to 90 degrees. Subjective scores were higher in the nonsurgical group, with good to excellent results of 90% in the nonsurgical group and 88% in the surgically repaired group, suggesting there was no benefit to surgical intervention. Indelicato also showed that patients treated with early motion returned to football 3 weeks earlier than immobilized patients. In a subsequent study by Indelicato and associates, they showed that conservative approach in complete MCL ruptures was successful in collegiate football players.32 All players were managed with a functional rehabilitation program, and 71% had good to excellent results. Similar to Indelicato, both Reider and colleagues33 and Jones and associates34 found excellent outcomes in athletes with isolated grade III medial ligament injuries treated conservatively and agreed that nonoperative treatment of these lesions is justified. Reider and colleagues looked at 35 athletes who were treated with early functional rehabilitation for isolated grade III tears.33 Of these, 19 patients returned to full, unlimited activity in fewer than 8 weeks. At an average follow-up of 5.3 years, outcomes based on subjective and objective measurements were comparable to earlier investigations employing a surgical repair. In 1985, Jones reported his results on 24 high school football players who returned to competition at an average of 34 days.34 Management consisted of 1 week of immobilization followed by gradual range of motion and strengthening. The knee was tested weekly with valgus stress, and instability was reduced to grade 0 or 1 by 29 days. No increased incidence of reinjury was found the following spring. Even though Indelicato, Fetto and Marshall, Jones, and Reider found excellent results with nonoperative treatment, Kannus studied 27 patients with grade III lesions at an average 9 years of follow-up.30 Patients were found to have poor outcomes (Lysholm score of 66) and degenerative changes on radiographs. Kannus concluded that early surgical repair would prevent deterioration. A careful review of the patients showed that 16 of 27 had greater than 2+ Lachman score, and 10 of 27 had anterolateral instability. This study did not show that nonoperative treatment has poor outcomes, but associated injuries need to be addressed, such as ACL, to prevent poor long-term outcome. Combined injury to the MCL and ACL represents a completely different entity than isolated MCL injury. The ACL is a primary restraint to anterior displacement and acts as a secondary stabilizer to valgus stress, especially in full extension. Conversely, the MCL is the primary restraint to valgus stress at 30 degrees of flexion. ­Therefore, injury

to the MCL and ACL results in both anterior and valgus instability and can significantly compromise knee function. Even though the apparent consensus that solitary MCL rupture can be treated nonoperatively, the optimal treatment for a concurrent ACL and MCL injury remains controversial. Two controversial studies regarding the management of combined ACL and MCL injury have been addressed in the literature. The first issue pertains to the various surgical options available for managing these injuries. Three principal surgical options exist: (1) surgical repair of both ligaments, (2) ACL reconstruction and nonoperative MCL management, and recently (3) operative management of MCL with nonoperative ACL treatment. ACL reconstruction with nonoperative management of the MCL remains the most popular option. The second controversial issue regarding combined injuries is whether early or late ACL reconstruction provides better functional and long-term results. Early authors recommended surgical intervention for both ligamentous structures in concomitant ACL and MCL ruptures.26,36,37 Fetto and Marshall had 79% unsatisfactory outcomes in patients treated operatively for ACL and MCL tears.26 Even though studies have shown operative repair of all ligaments results in stable, functional knees, there was a high incidence of knee motion postoperative complications.38-40 Other authors have stated that isolated operative MCL repair and nonoperative ACL reconstruction leads to good results. Hughston and colleagues reported that 94% of their patients with combined ACL and MCL injury and treated with only MCL reconstruction returned to their preinjury levels of athletic performance.41 They stated that the key to obtaining excellent results was reconstruction of the POL and posteromedial structures. Noyes and Barber-Westin criticized Hughston and Barrett’s method of reporting results and stated that the results may have been overly optimistic. However, Hughston continued to report good results at 22 years of follow-up.35 In addition to Hughston, Shirakura and associates reported excellent results in 14 patients with combined lesions but reconstruction of the MCL only; however, they did not report anteroposterior instability.42 Conversely, Frohlke and coworkers reported poor results with solitary MCL repair.43 They performed arthroscopically guided repair of the MCL, which led to functional stability in 68% of knees, but clinical testing of all 22 knees showed abnormal or severe abnormal examination. Most authors, however, suggest that nonoperative treatment of the MCL with reconstruction of the ACL provides good to excellent results. Shelbourne and Porter demonstrated good to excellent results in 68 patients with ACL reconstruction and nonsurgical management of an MCL tear.44 They also showed that these patients achieved a greater range of motion and more rapid strength gains than patients with surgical repair of both ligaments. Similarly, Noyes and Barber-Westin demonstrated a higher incidence of motion problems when MCL and ACL were treated operatively, and they recommend arthroscopic reconstruction of the ACL with nonoperative management of the MCL after recovery of range of motion and muscle function.45 In a prospective randomized study, Halinen and associates treated 47 consecutive patients with combined ACL and grade III MCL injuries.46 All patients

Knee 1633

underwent early ACL reconstruction within 3 weeks of injury. The MCL was treated operatively in 23 patients and nonoperatively in 24 patients. All patients were available for follow-up at a mean of 27 months. The nonoperative treatment of the MCL led to results similar to those obtained with operative treatment, with respect to subjective function, ­ postoperative stability, range of motion, muscle power, return to activities, and Lysholm score. Halinen and ­ colleagues concluded that MCL ruptures did not need to be treated operatively when the ACL was reconstructed early. In a retrospective study, Millett and colleagues reported on 19 patients with a complete ACL injury and a minimal grade II MCL tear who underwent early ACL reconstruction and nonoperative treatment of the MCL.47 At 2-year follow-up, subjective evaluation showed a Lysholm score of 94.5 and Tegner activity score of 8.4. Clinical examination revealed good range of motion and strength. No patient experienced graft failure or required subsequent surgery. The second controversial issue regarding combined ACL and MCL injury is whether early or late ACL ­reconstruction provides optimal return of function and long-term results. Based on animal studies, MCL healing is adversely affected

by ACL insufficiency.48 Therefore, it has been proposed that early ACL reconstruction will improve healing of the MCL. Both Halinen and colleagues46 and Millett and associates47 showed good subjective scores and minimal loss of motion complications with early ACL reconstruction (within 3 weeks). Conversely, Petersen and Laprell demonstrated poorer results with early ACL reconstruction compared with late ACL reconstruction in combined injuries.49 All patients underwent nonoperative treatment of MCL injury, and early ACL reconstruction was performed within 3 weeks and late ACL reconstruction after a minimum of 10 weeks. The late reconstruction group had a lower rate of loss of motion and higher Lysholm scores compared with the early reconstruction group. The literature supports nonoperative treatment of the MCL tear with surgical reconstruction of the ACL. This is the trend that most surgeons are currently using. However, early versus late reconstruction continues to be a subject of debate, with studies supporting both sides. Other factors such as preoperative and postoperative rehabilitation protocol along with bracing may need to be further analyzed to help better assess whether early or late reconstruction is more beneficial.

Authors’ Preferred Method Before proceeding with a treatment plan, it is essential to know the extent of injury. Initially we perform a ­thorough history and physical examination. With MCL injuries, we assess the grade of injury of the MCL and also any ­associated ligamentous, meniscal, posteromedial corner, or ­ patellar ­ injuries. We obtain radiographs as a routine ­diagnostic tool to rule out fracture or any signs of chronic

­ edial insufficiency (Pellegrini-Stieda lesion) and chronic m ACL deficiency (deep femoral notch sign, peaked tibial spines, cupula lesion). The use of MRI is dependent on the grade of the MCL lesion and associated injury. Isolated grade I or II injuries can be diagnosed on clinical ­examination and do not require MRI. However, in a grade I or II injury with an indeterminate cruciate examination and Figure 23C1-10   Algorithm for treatment of mcl injuries. ACL, anterior cruciate ligament; MCL, medial collateral ligament; PT, physical therapy.

Suspected MCL Injury? Examine

Isolated Grade I or II MCL

Isolated Grade III or MCL with associated injuries

MRI

Rehab

Isolated Grade III MCL

ACL/MCL

Femoral Avulsion Isolated Grade III

Rehab, PT, regain motion

Full ROM

Tibial Avulsion

MCL Repair/Reconstr

MCL unstable ACL Reconstruction

Continued

1634 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

Figure 23C1-11   Coronal magnetic resonance image shows complete avulsion of the superficial and deep medial collateral ligament with an unattached medial meniscus.

effusion, we ­order an MRI. In contrast, we perform MRI on all grade III injuries because the site of ­ involvement, tibia or femur, is important in our decision making. In ­addition, most grade III lesions are associated with concomitant ­ligamentous ­injuries. Figure 23C1-10 presents our ­treatment algorithm. We treat isolated grade I and II MCL injuries conservatively. In the first 48 hours, we encourage rest, ice, compression, and elevation to help reduce swelling. In addition, we place all patients in a hinged knee brace and provide crutches for protected weight-bearing. If patients have significant pain and valgus laxity, initially we lock the brace in extension. Once the swelling subsides and pain is improved, we encourage aggressive range of motion exercises and straight leg raises with quadriceps-setting exercises. Once the patient

Figure 23C1-12   Arthroscopy confirms gross laxity of the medial compartment with complete disruption of medial support structures and a free-floating meniscus.

has regained full range of motion and ambulation without a limp, crutches and the brace can be discontinued. Stationary bicycle and progressive resistive exercises are instituted as tolerated. Once full range of motion and 80% strength of the opposite side have been achieved, closed-chain kinetic exercises and jogging are allowed. In athletes, once they have achieved 75% of the maximal running speed, sport-specific training is allowed. Return to sports is permitted after the patient has strength, agility, and proprioception equal to the other side. We recommend a functional brace for contact or high-risk sports. In grade I sprains, patients usually return to sports in 10 to 14 days; because immobilization is temporary, patients ­regain strength and motion quickly. However, return to play after grade II sprains is much more variable. With grade II sprains, the period of immobilization can be up to 3 weeks to allow the pain to dissipate. Therefore, patients can lose more strength and motion with increased time of immobilization compared with grade I sprains. Patients are allowed to return when they have equal strength, and there is no pain with valgus stress. The treatment of grade III MCL sprains has significantly evolved over the past 20 years. The general consensus has been to treat isolated grade III injuries conservatively. We believe that the treatment of grade III injury is dependent on not only the specific location of the MCL rupture but also the degree of laxity on physical examination as well as the quantity of the arthroscopic drive-through sign. A case example is that of a 16-year-old high school football player who sustained a contact MCL and ACL injury treated operatively in a staged fashion. Figures 23C1-11 to 23C1-14 highlight the treatment plan. This allowed full return to contact sports 1 year from injury. Although most femoralsided tears can be treated successfully conservatively, complete tibial-sided avulsions of the deep and superficial MCL, although rare, often heal with residual laxity. In athletes who

Figure 23C1-13   Open surgery confirms complete avulsion of the medial collateral ligament from the tibia with a free-floating, unattached medial meniscus between the articular cartilage of the medial femoral condyle and tibial plateau.

Knee 1635

Authors’

preferred method—cont’d

Figure 23C1-14   Magnetic resonance image showing tibial-sided avulsions.

­ articipate in level I sports, frequently we favor operative p repair of these tibial-sided complete avulsions that display retraction of the deep or superficial MCL on MRI (Fig. 23C1-15).50 Figures 23C1-16 and 23C1-17 highlight a case example of a Division I football player with a tibial-sided complete MCL avulsion with gross laxity and an impressive arthroscopic drive-through sign treated surgically. Our rehabilitation protocol for grade III lesions is placement in a long-leg hinged knee brace locked in extension with weight-bearing as tolerated on crutches for 2 weeks. In the first 4 weeks, our goal is to have the patient attain nearly full range of motion and normal gait pattern with full

Figure 23C1-15   Arthroscopy 3 months after fixation at the time of primary anterior cruciate ligament reconstruction shows complete healing of meniscus and elimination of the drive-through sign.

Figure 23C1-16   Coronal magnetic resonance image shows tibial-sided avulsion of the medial collateral ligament with retraction and a contrecoup bipolar bone bruise lesion laterally, which suggest a high-energy injury pattern.

weight-bearing in a hinged knee brace, and begin quadriceps and hamstring strengthening. In contrast, patients who undergo a repair of the MCL follow a different protocol. Postoperatively, a hinged knee brace is locked from 30 to 90 degrees for 3 weeks, followed by unlimited motion. Weight-bearing is limited for 3 weeks with crutches and then ­progressed to full weight by 4 to 6 weeks. Bracing is discontinued at 6 weeks, and nonimpact conditioning is allowed, with running started by 3 months.

Figure 23C1-17   Arthroscopy confirms a drive-through sign with liftoff of the medial meniscus from the tibia requiring open repair of medial structures.

1636 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS The treatment of MCL injuries has evolved into nonoperative management. Therefore, rehabilitation is pivotal and is the primary modality for treatment. There is no one perfect rehabilitation protocol that will work with every athlete. When reviewing the literature, there is no apparent consensus on the most efficacious rehabilitation protocol, and protocols are usually based on surgeon preference and experience. Steadman,51 Bergfeld,52 O’Connor,53 Cox,54 and Wilk and colleagues55 have had excellent success with their individual protocols for treatment of MCL injuries. To effectively treat MCL injuries, the grade of the injury must be determined because the parameters of rehabilitation are based on the degree of injury. Look at Table 23C1-3 for general principles to follow in ­rehabilitation of MCL. Isolated grade I sprains are treated with rest, ice, compression, and elevation for the first couple of days to help reduce swelling. Patients are allowed weight-bearing as tolerated with use of an assistive device if there is pain with walking. The only exception is patients with significant valgus deformity because they will place more stress on the MCL, affecting healing. In these patients, it may be safer to allow them partial weight-bearing for a couple of weeks. With grade I MCL tears, immobilization in a brace is rarely required, and if patient compliance is of concern, a short leg hinged brace is used to control valgus and rotational stresses. Range of motion is begun immediately to prevent arthrofibrosis and stiffness. In addition, quadriceps strengthening and closed chain exercises are started. Once the patient regains full range of motion, resistive exercises are begun along with sport-specific drills. Isolated grade II injuries are treated similarly to grade I injuries with rest, ice, elevation, and compression. Because grade II injuries involve a greater degree of damage to the ligament with increased valgus instability, a long-leg hinged brace is usually needed. Patients are allowed to progressively bear weight as tolerated in the brace; however, if the patient is having significant pain, the brace can be locked in extension until the pain subsides, usually in 1 week. Assistive devices are used until the patient has a

   TABLE 23C1-3  Principles for Rehabilitation of the   Medial Collateral Ligament Phase

Goals

Criteria for Progression

Maximal protection phase

Early protected range of motion (ROM) Decrease effusion and pain Prevent quadriceps atrophy Full painless ROM Restore strength Ambulation without crutches Increase strength and power

No increase in instability No increase in swelling Minimal tenderness Passive ROM at least 10-100 degrees No instability No swelling or tenderness Full painless ROM

Moderate protection phase Minimal protection phase

nonantalgic gait. Active range of motion exercises are started ­ immediately. During the early period, quadriceps strengthening is done in a non–weight-bearing fashion with straight leg raises, quadriceps-setting exercises, and electrical stimulation. Once the patient has achieved full range of motion and functional strength, proprioceptive and agility drills can be initiated. Isolated grade III injuries usually involve disruption of both the superficial and deep fibers. Therefore, the rehabilitation process is slower, and a longer period of immobilization is required. The treatment of grade III injuries can be divided into stages. In the first phase (about 4 weeks), the patient should wear a brace locked in extension and progressively increase weight-bearing to attain a normal gait pattern. Also, the patient needs to perform range of motion exercises with strengthening of quadriceps and hamstrings. In phase II, 4 to 6 weeks, the patient continues to attain full range of motion, unlock the brace, and achieve quadriceps and hamstring strengthening. After 6 weeks, the brace can be discontinued if the patient has a nonantalgic gait and has regained quadriceps strength for daily ambulation. Phase III starts after 6 weeks and includes squatting, light jogging with agility drills, and continued strengthening to return to sports. After surgical repair of an isolated MCL, the patient is locked in 30 degrees and allowed toe-touch weight-bearing for 3 weeks. The patient is encouraged to continue range of motion from 30 to 90 degrees. The patient also continues strengthening of the quadriceps and hamstring in brace. After 3 weeks, the patient is allowed to progress to full weight-bearing with full-time brace wear to continue to protect the repair. The brace can be worn unlocked to allow free range of motion as well as valgus and rotational stability. From 3 to 6 weeks, the goal is to restore full range of motion along with continued strengthening with closed kinetic chain exercises. After 6 weeks, the patient continues to progressively increase activities with resistive and sportspecific exercises. Combined injuries of the MCL and ACL require additional steps to be taken compared with the rehabilitation of isolated MCL tears. Reviewing the literature as stated previously on ACL and MCL injuries, conservative treatment of MCL followed by surgical reconstruction of ACL is the favored management. Initially the protocol focuses on the severity of the MCL injury. For example, a grade I MCL injury with an ACL injury will proceed with the protocol presented earlier for grade I injuries. The patient will quickly regain range of motion and functional strength, and then the surgeon can proceed with reconstruction of the ACL. Conversely, the patient with a grade III injury with an ACL injury will take much longer owing to the slower protocol for type III injuries. Regaining range of motion and functional strength training may take 8 to 10 weeks. Therefore, ACL reconstruction with a type III injury will take longer to proceed. Once the ACL is reconstructed with conservative treatment of the MCL, the rehabilitation protocol follows one for an ACL. After a combined ACL reconstruction and medial-sided repair, the knee is braced in full extension, and a standard ACL protocol is followed. In combined ACL and MCL injury, it is important to remember that ACL rehabilitation takes precedence over medial-sided repair.

Knee 1637

Outcomes were thoroughly discussed in the treatment section. However, nonoperative treatment for grade I and II sprains is well accepted in the literature. Ellsasser and colleagues showed in 74 professional football players with isolated grade I and II MCL sprains return to play in 3 to 8 weeks and excellent results in 98%.22 Similarly, ­Derscheid and Garrick demonstrated in 51 college football players with grade I and II sprains return to play at 10.6 days and 19.5 days, respectively.23 Indelicato, Reider, and Jones and their associates found excellent results with conservative management of grade III sprains with return to sports by 8 weeks.32-34 Complications of MCL ruptures are rare in the literature. Failure to diagnose associated ligament injuries, such as ACL, can lead to long-term instability and degenerative problems. In addition, missed associated meniscal tears and articular cartilage defects can lead to continued pain. Atrophy and arthrofibrosis are rare complications given the aggressive rehabilitation protocols with early motion and strengthening. Infection is a rare complication with surgical reconstruction. Residual pain can occur after grade I sprains, usually near the femoral origin, possibly because of a small neurovascular bundle.2 Treatment consists of an injection or anti-inflammatory medication. In addition to pain, patients with femoral-sided lesions are more prone to have loss of motion and associated stiffness.

CRITERIA FOR RETURN TO PLAY See Box 23C1-1.

Box 23C1-1 Return to Play

• Full range of motion • No instability • Muscle strength 85% of contralateral side • Proprioception ability satisfactory • No tenderness over medial collateral ligament • No effusion • Quadriceps strength; torque/body weight • Lateral knee brace (if necessary)

C

r i t i c a l

P

o i n t s

l The superficial MCL is the primary static stabilizer against valgus and external rotation stress. l Posteromedial corner structures include the posterior horn of the medial meniscus, POL, semimembranosus expansions, meniscotibial ligaments, and oblique popliteal ligament.

l The posteromedial corner provides static and dynamic restraint to anteromedial rotatory instability. Concomitant injuries, such as ACL, PCL, meniscus tear, and patellar injuries, with MCL injury should be ruled out. l Valgus laxity at 30 degrees occurs with isolated MCL injury. l Valgus laxity at 0 degrees occurs with combined MCL and ACL or posteromedial complex injury. l Grade I or II MCL injury is treated conservatively with rehabilitation. l Isolated grade III injury is treated conservatively. Combined grade III ACL and MCL injury is treated with rehabilitation of the MCL and then reconstruction of the ACL if the MCL heals. Watch for nonhealing of tibialsided MCL lesions. If nonhealing, repair the tibial-sided MCL and reconstruct the ACL.

S U G G E S T E D

R E A D I N G S

Fanelli GC, Harris JD: Surgical treatment of acute medial collateral ligament and posteromedial corner injuries of the knee. Sports Med Arthrosc Rev 14:78-83, 2006. Halinen J, Lindahl J, Hirvensalo E: Operative and nonoperative treatments of medial collateral ligament rupture with early anterior cruciate ligament reconstruction. Am J Sports Med 34:1134-1140, 2006. Hughston JC, Eilers AF: The role of the posterior oblique ligament in repairs of acute medial (collateral) ligament tears of the knee. J Bone Joint Surg Am 55: 923-940, 1973. Indelicato PA, Hermansdorfer J, Huegel M: Nonoperative management of complete tears of the medial collateral ligament of the knee in intercollegiate players. Clin Orthop 256:174-177, 1990. Jacobson KE, Chi FS: Evaluation and treatment of medial collateral ligament and medial-sided injuries of the knee. Sports Med Arthrosc Rev 14:58-66, 2006. Noyes FR, Barber-Westin SD: The treatment of acute combined ruptures of the anterior cruciate and medial collateral ligament of the knee. Am J Sports Med 23:380-389, 1995. Petersen W, Laprell H: Combined injuries of the medial collateral ligament and the anterior cruciate ligament: Early ACL reconstruction versus late ACL reconstruction. Arch Orthop Trauma Surg 119:258-262, 1999. Sims WF, Jacobson KE: The posteromedial corner of the knee: Medial-sided injury patterns revisited. Am J Sports Med 32:337-345, 2004. Warren LF, Marshall JL: The supporting structures and layers on the medial side of the knee. J Bone Joint Surg Am 61:56-62, 1979. Wilk KE, Andrews JR, Clancy WG: Nonoperative and postoperative rehabilitation of the collateral ligaments of the knee. Oper Tech Sports Med 4:192-201, 1996.

R eferences Please see www.expertconsult.com

1638 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

S e c t i o n

C

Medial Ligament Injuries 2. Pediatric Medial Knee Injuries Jayesh K. Patel, Manuj Singhal, and Darren Johnson

Knee injuries in children are rather complex given the broad spectrum of patients in regard to age and maturity. In this section, we describe the basic anatomy of the pediatric knee, focusing on, in particular, the medial aspect of the knee and the medial collateral ligament (MCL). Much has been written regarding adult MCL injuries, but few studies focus on pediatric knee injuries. As the world of organized sports incorporates younger and younger children, knee injuries are become more prevalent and complex. We focus on studies and literature pertaining to only the pediatric population.

ANATOMY The growth plates of the distal femur and the proximal tibia, as well as the ligamentous structures, are all vulnerable to injury. The ligaments and cartilaginous structure of the knee start to develop at about the eighth week of embryonic development.1 The distal femoral epiphysis is the largest and most rapidly growing growth plate in the body. It grows an average of 10 to 11 mm per year. It is the first secondary ossification center to appear in the body and is present at birth in fullterm newborns.1,2 The distal femur contributes roughly 40% of the length of the entire leg. It fuses at about 14 years of age in girls and at 16 years of age in boys. The distal femoral epiphysis is the last epiphysis to fuse in adults.2 The rapid growth of the distal femoral epiphysis increases its vulnerability to injury. The proximal tibial epiphysis is seen at about 2 months of age. There is an additional secondary epiphysis that develops in the tibial tubercle between the ages of 9 and 14 years.1 The proximal tibia grows at rate of about 6 to 7 mm per year.3 By the age of 16 years in girls and 18 years in boys, the proximal tibia is fused. It contributes about BOX 23C2-1 Layers of the Medial Side of the Knee 1. Layer I—sartorius fascia 2. Layer II—superficial medial collateral ligament, ­posteromedial capsule 3. Layer III—deep medial collateral ligament, capsule of knee joint

27% of the entire length of the leg.2 The distal femur is twice as likely to be injured compared with the proximal tibia. The medial structures of the knee are divided into three distinct layers. Layer I includes the deep sartorius fascia. Layer II includes the superficial MCL and the ligaments of the posteromedial capsule. Layer III is the deep MCL and the knee capsule (Box 23C2-1; see also Chapter 23C1).1,2,4 The superficial MCL, layer II, originates over a large region of the medial femoral epiphysis that extends to the physis. It finally inserts on the metaphyseal region of the proximal tibia. The deep MCL also originates at the distal femoral epiphysis and extends to the epiphyseal ­perichondrium of the proximal tibia.1,2 Both the distal femoral physis and the proximal tibia physis are extracapsular and extrasynovial.1

HISTORY AND PHYSICAL EXAMINATION The most important tool in diagnosing any injury is the history and physical examination. This holds true especially in the pediatric population, even though sometimes it is tough to obtain. The history should include multiple questions, including but not restricted to precipitating and exacerbating events and whether the mechanism of injury was low or high energy (e.g., motor vehicle crash versus twisting the knee playing soccer). A key question is the presence or absence of an effusion. The presence of effusion can narrow the differential diagnosis (Box 23C2-2). The five main causes of an effusion in the knee are fracture, BOX 23C2-2 differential Diagnosis of Medial-Sided Knee Injuries

• Fracture—proximal tibia, distal femur • Medial collateral ligament injury • Medial meniscal tear • Discoid meniscus • Legg-Calvé-Perthes disease • Slipped capital femoral epiphysis • Functional knee • Tumor

Knee 1639

meniscal pathology, patella dislocation, ligamentous injury (anterior cruciate ligament [ACL] tear), and cartilage damage. It is important to note whether the effusion happened within 24 hours or was delayed in its presentation. The presence of locking, popping, or catching prompts diagnostic questions leading to intra-articular pathology. The parents are asked about limping and swelling of the limb. The patient’s age and the absence or beginning of menses are important regarding physeal injuries. The accuracy of a history is difficult in young patients because they frequently forget the mechanism of injury, have difficulty describing their symptoms, and may exhibit a lack of cooperation or inconsistencies with verbal questioning.5 ­ Harvell and associates found that the accuracy of preoperative ­clinical ­diagnosis in ­preadolescent children was only 55%, and that it increased to 70% in adolescent patients.6 Functional knee pain should also be in the differential diagnosis. This entity develops to allow children to deal with environmental stressors and emotional outburst.5 The next important step is the physical examination. This should always begin with analysis of the patient’s gait. The differentiation between an antalgic and a Trendelenburg gait can help narrow the differential. The examination of the hip is important in children. All patients between 5 and 15 years of age presenting with knee pain should have their hip evaluated.5 Many pathologic conditions of the hip can present as medial-sided knee pain, including slipped capital femoral epiphysis or Legg-Calvé-Perthes disease. It is important to observe for the presence of a lateral thrust or asymmetry between foot progression angles. In children, examination of the normal knee is helpful. First, it places the patient at ease before causing pain, and second, it gives the physician the ability to compare findings, especially if the patient has increased physiologic laxity. With children, establishing a good rapport and understanding helps facilitate the physical examination. Once a full examination of the normal knee has been performed and full range of motion and ligamentous and neurovascular examination recorded, the physician can focus on the injured knee. This begins with visual inspection, making note of lacerations, abrasions, ecchymosis, and swelling that may help to point toward the injury. The presence of an effusion is important in determining intra-articular pathology. As stated earlier, an effusion can narrow the differential diagnosis to include meniscal injury, ligamentous injury, fracture, patella dislocation, or cartilage damage. The presence of quadriceps atrophy occurs with long-term immobilization or altered gait. Palpation of the entire knee in a systematic way is crucial. Joint line tenderness has been shown to correlate with meniscal pathology.20 Reproduction of pain or popping with flexion and extension maneuvers with rotation, such as McMurray’s test and Apley’s grind test, is suggestive of meniscal injuries. Numerous studies have shown that in young patients, these tests are not particularly accurate.1,7,19,21 Ligamentous examination should follow basic range of motion evaluation. The most commonly described is Lachman’s test, looking for ACL injury. Lachman’s test is the most sensitive test for the presence of anterior laxity.22 The posterior drawer test is used to determine the competency of the posterior cruciate ligament (PCL). Valgus

stress to the knee at 30 degrees of flexion tests the competency of the MCL. If there is laxity to valgus stress at full extension, one should be alert to the fact there might be a multiligamentous injury. Varus stress at 30 degrees of flexion tests the strength of the lateral collateral ligament; laxity with varus stress at full extension should warn to a possible posterior collateral ligament or ACL injury. After a basic physical examination and history are performed, imaging techniques can be used to further evaluate the knee. These examinations include radiography, magnetic resonance imaging (MRI), and computed tomography (CT), if needed.

INJURIES TO THE MEDIAL COLLATERAL LIGAMENT Injuries to the knee in children result in fractures or physeal injuries more commonly than ligament disruptions because of the weakness of the growth plate relative to the strength of the ligaments in a skeletally immature patient. Injuries to the MCL are divided into three grades according to the American Medical Association Standard Nomenclature of Athletic Injuries. Grade I shows no knee instability with no medial opening on stress radiographs. Grade II includes partial anatomic discontinuity and mild functional instability. The injury presents with mild (0 to 5 mm) or moderate (6 to 10 mm) displacement with valgus stress to the injured knee. Radiographs present with some medial opening. Grade III demonstrates gross instability with displacement of greater than 10 mm and no end point with valgus stress.9,10 Isolated MCL injuries are rare, and therefore the literature is scarce. Most ligamentous injuries are associated with a physeal fracture.11,12 Injuries to the MCL are usually found in patients with high-energy trauma. Clanton reviewed 932 patients with knee injuries. Only 9 of these patients had ligamentous injuries with open physes, all younger than 14 years. Five children were hit by automobiles; one was in a motorcycle crash; one in a go-cart crash; one fell out of a moving vehicle; and one fell from a merry-go-round. Five patients had injuries to the MCL: two were torn from the meniscofemoral portion, and three were torn from meniscotibial portion of the MCL.12 Kannus reviewed 33 patients who sustained knee ligament injuries in skeletally immature patients. They found 13 patients with grade II MCL strains, 1 patient with grade III MCL strains, 4 with combined ACL and grade II MCL, and 1 with combined ACL and grade III MCL injuries.10 The extent of injury can be correlated to a careful history and examination. The examination should include a full knee examination and should be directed to the amount of valgus instability. The test is performed at 30 degrees of flexion, and a valgus strain is applied. This test is also performed at full extension to rule out other ligamentous instabilities that can occur with MCL tears. A medial opening between 5 and 10 mm suggests an incomplete injury, grade II. A medial opening greater than 1 cm with no end point points to a grade III injury or complete disruption of the MCL. One must also assess the competency of the ACL and PCL by performing Lachman’s test and the posterior drawer test (Box 23C2-3; see Chapter 23C1).

1640 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

BOX 23C2-3 Signs and Symptoms of Medial ­Collateral Ligament Tear  enderness to palpation along insertion or origin of • T medial collateral ligament • Pain with valgus stress • Opening to valgus stress at 30 degrees of flexion

Diagnostic Testing The two most important modalities are plain and stress radiographs. If more imaging is need, MRI may be warranted to evaluate ligamentous and meniscal pathology. The first step would be to obtain radiographs to assess for fractures—anteroposterior and lateral views of the knee. If a child presents with chronic knee pain, anteroposterior and frog-leg lateral views of the ipsilateral hip may be warranted to rule out hip pathology. The next modality of choice is MRI. This can be employed if intra-articular or ligamentous injury is suspected. In children, however, clinical examination is more sensitive in determining the extent of the knee injury.13 Kocher and colleagues found a lower sensitivity (61.7% versus 78.3%) and specificity (90.2% versus 95.5%) with the diagnostic performance of MRI studies in children younger than 12 years than in adolescents 12 to 16 years of age, respectively.14 This correlates to prior studies that have shown lower values of sensitivity and specificity for the diagnosis of meniscal injuries (sensitivity, 50% versus 100%; specificity, 46% versus 95%) and ACL injuries (sensitivity, 64% versus 78%; specificity, 94% versus 100%) in children compared with adults. The data clearly show that clinical examination and history are the keys to diagnosing injuries of the knee in children.

Treatment Most MCL injuries in children can be treated nonoperatively. Deciding which ones to operate on is difficult. Treatment principles of MCL injuries are the same as in adults (see Chapter 23C1). Grade I MCL strains are all treated nonoperatively. Grade II MCL strains are also treated nonoperatively, but they tend to have a longer recovery period. The controversy revolves around grade III MCL tears. Numerous studies have shown nonoperative treatment to be effective. Jones and associates retrospectively evaluated 24 high school football athletes with isolated grade III MCL injuries.15 They found 22 cases in which knee stability was achieved at an average of 29 days. The athletes returned to play competitive football at a mean of 34 days. Kannus and Jarvinen reviewed 32 patients with grade II or III injuries to the ligaments of the knee and concluded that grade II injuries in adolescents can be treated nonoperatively but that grade III injuries show long-term functional instability and some early posttraumatic osteoarthritis.10 The study had many deficiencies, including combining multiligamentous injuries with the isolated MCL results. This could explain the long-term instability found in the knees with grade III injuries. Most authors agree that nonoperative treatment is the standard of care for grade III injuries.

Authors’ Preferred Method We start treating grade I and II MCL using a hinged knee sleeve for 2 to 3 weeks, with activity modification and physical therapy focusing on quadriceps and hamstring strengthening. Grade III MCL injuries are also treated ­nonoperatively. They are placed in a hinged knee brace locked at 30 degrees of flexion with non–weight-bearing for 2 weeks. Patients are allowed to bear weight 2 to 4 weeks after injury and progress as tolerated. They begin passive range of motion and physical therapy at this time. Patients are allowed to return to athletic competition when they have no pain or instability. They are also assessed based on their physical examination. If laxity with valgus stress remains, the athlete is held from activities until examination mirrors the opposite knee (Box 23C2-4; see Chapter 23C1).

BOX 23C2-4 Treatment of Medial Collateral Ligament Injury Grade I—hinged knee sleeve, physical therapy Grade II—hinged knee sleeve, protected immobilization for 2 to 4 weeks, physical therapy Grade III—hinged knee brace locked at 2 to 4 weeks, non-weight-bearing for 2 to 4 weeks, passive range of motion, physical therapy

FRACTURES OF THE MEDIAL KNEE Fractures to the physeal aspects of the knee are more common than ligamentous injuries in children. Physeal fractures are relatively uncommon, accounting for only 1% of all pediatric fractures.11 They can be divided into two groups: (1) distal femoral epiphyseal fractures, which account for about 5% to 15% of all physeal fractures16,24 and (2) proximal tibial epiphyseal fractures, which account for less than 2% of all physeal fractures (Box 23C2-5).16 The rate and magnitude of the injuring force are the determining factors for whether there will be a ligamentous injury or a physeal fracture. High-magnitude forces at low velocity result in physeal injuries, and low-magnitude forces at high velocity result in ligament injuries.17 Physeal fractures are divided according to the Salter-Harris classification, types I thru IV (Fig. 23C2-1). Salter-Harris I injuries are complete separation of the epiphysis from the

BOX 23C2-5 Anatomic Location of Fractures around the Knee

• Distal femoral epiphysis or physis fracture • Proximal tibial epiphysis or physis fracture • Patella sleeve fracture • Tibial tubercle fractures • Tibial spine fractures • Osteochondral fractures

Knee 1641

Type I

Type II

Type III

Type IV

Type V

Figure 23C2-1   Classification of epiphyseal fractures according to the Salter-Harris system. In type I fractures, the fracture line traverses the physis, staying entirely within it. In type II injuries, the fracture line traverses the growth plate for a variable length and then exits obliquely through the metaphysis. Type III fractures also begin in the physis but exit through the epiphysis toward the joint. Type IV fractures involve a vertical split of the epiphysis, physis, and metaphysis. Type V fractures are crush injuries to the physeal plate. (Redrawn from Edwards P, Grana W: Physeal fractures about the knee. J Am Acad Orthop Surg 3:63-69, 1995.)

metaphysis without bone fracture. Type II injuries have a line of separation that extends from the epiphyseal plate out through a portion of the metaphysis; this produces the Thurston-Holland fragment. Most of these injuries occur in children older than 10 years. These are the most common type of injury seen in physeal fractures. Type III injuries are intra-articular; the fracture line begins from the joint surface into the physis than exits perpendicular to the physis. Type IV injuries are also intra-articular. The injury begins at the joint surface through the physis and exits through the metaphysis, resulting in a complete split. Type V injuries are rare but involve a crushing injury to the epiphysis and physis.8

Distal Femoral Epiphyseal Fractures History and Physical Examination Patients with distal femoral epiphysis fractures usually present with limited range of motion, inability to bear weight, large effusion of the knee, and tenderness along the physis (Box 23C2-6). Obvious deformity of the knee can be detected on visual inspection. With anterior displacement, the patella can be prominent, and anterior skin dimpling can be seen. With posterior displacement, the distal metaphysis becomes prominent at or above the patella. Crepitus with range of motion can also be felt.

BOX 23C2-6 Signs and Symptoms of Fractures around the Knee

• Inability to bear weight • Tenderness to palpation over growth plate or epiphysis • Knee effusion, hemarthrosis • Pain with range of motion • Crepitation • Visual deformity of knee

Imaging The first step in evaluating for a physeal injury after the history and physical examination is obtaining plain radiographs. Standard anteroposterior and lateral radiographs are required to establish a diagnosis (Fig. 23C2-2). If there remains a concern for fracture, oblique radiographs and radiographs of the contralateral side can be helpful. When there is apparent laxity to the knee and plain radiographs are normal, stress radiographs can be obtained to rule out physeal separation compared with ligamentous injury. MRI or CT can also be obtained to rule out nondisplaced injuries (Fig. 23C2-3). MRI can also be obtained to evaluate intra-articular pathology.

Treatment Treatment is based on the amount of displacement and type of fracture pattern. The main goal in treatment of physeal injuries is to prevent growth disturbances and avoid damage to the physis. Salter-Harris I and II injuries that are nondisplaced can be treated nonoperatively. They require immobilization in either a long leg cast or hip spica cast for a minimum of 6 weeks.2 The duration of the cast and the method for immobilization are determined by the age and size of the patient. Displaced fractures need to be reduced to anatomic alignment. The acceptable amount of displacement is less than 2 mm.2 Reductions generally require general anesthesia. If reduction cannot be obtained or the fracture is unstable, operative fixation may be warranted. Smooth transphyseal pins are used for type I and II injuries. A large metaphyseal fracture fragment in a type II injury can be stabilized with either smooth pins or 4.5-mm or larger cannulated screws. Some physicians advocate bending or burying the pins to avoid bacterial contamination and the risk for a ­septic joint. Salter-Harris III and IV injuries that are stable and nondisplaced can also be treated nonoperatively using a cast. Weekly follow-up is indicated for the first 2 to 3 weeks to observe for any displacement of the fracture in the cast.

1642 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 23C2-2   Antero­ posterior (A) and lateral (B) radiographs of a 12-year-old girl with knee pain after a motor vehicle crash. These radiographs demonstrate a Salter-Harris III fracture of the distal femur.

B

A

Fractures with displacement or that are irreducible with closed means require operative fixation. The goal in treatment is to restore joint congruity and align the physis. Percutaneous pins or cannulated screws can be used to obtain fixation (Fig. 23C2-4). These are all intra-articular fractures, so weight-bearing is limited until fracture healing is observed on plain radiographs (Table 23C2-1).

Proximal Tibial Epiphyseal Fractures History and Physical Examination

Treatment Treatment once again is similar to that for distal femoral epiphyses. The goal for these epiphyseal fractures is anatomic reduction to prevent growth disturbances. Nondisplaced and stable Salter-Harris I and II injuries can be treated nonoperatively with the use of cast immobilization for a minimum of 6 weeks.2 Displaced fractures require reduction under anesthesia. Most of the injuries are hyperextension type; thus, flexion usually achieves reduction of the fracture. These injuries are usually immobilized

Fractures of the proximal tibial epiphysis are less common than those of the distal femoral epiphysis. Almost half of proximal tibial injuries occur in sporting ­activities.17 These injuries are usually the result of hyperextension and valgus forces. Because of the close proximity of the popliteal artery, vascular injuries are a concern with these types of injuries. Neurovascular injury occurs in up to 10% of proximal tibial epiphyseal injuries.3 Many believe that proximal tibial fractures are equivalent to knee dislocations in adults. Physical examination should consist of a thorough evaluation of the neurovascular status of the extremity, including the dorsalis pedis and posterior tibial arterial pulses and function of the peroneal and posterior tibial nerves. Close monitoring for compartment syndrome should also be considered. If there is any abnormality in the vascular status of the limb, a vascular surgery consultation is needed for evaluation for an arteriography.

Imaging Basic imaging includes plain radiographs, anteroposterior and lateral views. Stress radiographs may also be indicated (Fig. 23C2-5). See the earlier section on distal femoral epiphyseal fractures for further imaging studies.

Figure 23C2-3   Computed tomographic scan of the knee demonstrating a Salter-Harris III fracture of the distal femur. The patient has a fracture line that starts intra-articularly and extends into the metaphysis.

Knee 1643 Figure 23C2-4   Anteroposterior (A) and lateral (B) radiographs after operative fixation with two cannulated screws of the Salter-Harris III distal femur fracture.

A

B

in about 20 to 30 degrees of flexion, which helps reduce the risks for displacement and vascular compromise.22 If reduction cannot be sustained or the fracture is irreducible by closed means, operative fixation is required. Smooth, crossed transphyseal pins are used for stable fixation. A Salter-Harris type II injury with a large metaphyseal fragment can be stabilized with cannulated screws. Treatment of Salter-Harris III and IV injuries is the same as for distal femoral epiphyseal fractures of the same type. Toetouch is permitted in the cast once the patient’s symptoms allow (see Table 23C2-1).

Complications The risk for growth deformity remains the number one concern after epiphyseal fractures. This risk is lowest with type I and II injuries and increases with severity of the fracture. There is almost a 100% chance of growth disturbance with type V injuries. In fractures that involve the distal femoral epiphysis, shortening and angular deformity can be common (Box 23C2-7). Leg-length discrepancies of less than 2 cm can be treated nonoperatively. If more than

Return to Sports The time to return to athletic activities is determined by the type of fracture. Patients with Salter-Harris I and II injuries can usually return to sports in 3 to 4 months. Athletes with Salter-Harris III and IV injuries take longer to return to sports, roughly 4 to 6 months.2

   TABLE 23C2-1  Treatment Options Type

Nondisplaced

Displaced

I

Cast immobilization for at least 6 wk Cast immobilization for at least 6 wk Cast immobilization for 6 wk

1. Closed reduction with cast immobilization 2. ORIF: smooth pins; cast for 4 wk 1. Closed reduction with cast immobilization 2. ORIF: smooth pins; cast for 4 wk 1. Closed reduction with cast immobilization 2. ORIF: cannulated screws; cast for 4 wk 1. Closed reduction with cast immobilization 2. ORIF: cannulated screws; cast for 4 wk

II III

IV

Cast immobilization for 6 wk

ORIF, open reduction with internal fixation.

Figure 23C2-5   Physeal separation of medial proximal tibia with stress radiographs without joint space widening. (From Edwards P, Grana W: Physeal fractures about the knee. J Am Acad Orthop Surg 3:63-69, 1995.)

1644 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

BOX 23C2-7 Complications of Knee Fractures

• Leg-length discrepancies • Angular deformity—valgus • Vascular compromise—popliteal artery injury 2 cm, referral to a pediatric orthopaedist for epiphysiodesis or lengthening would be prudent. Angular deformity, especially valgus deformity, can occur after physeal injuries. Valgus deformity after proximal tibial epiphyseal ­fractures can spontaneously correct up to 3 years after injury if enough growth remains in the child.17 Bony bridges can lead to angular deformities; they can be evaluated using MRI, which is the imaging modality of choice to evaluate for osseous bridge lesions. These lesions can be removed if they occupy less than 50% of the growth plate. Patients should be followed clinically at 6-month intervals to observe leg-length discrepancy and angular deformity until skeletal maturity. They should be followed for up to 2 years after initial injury.2

C

r i t i c a l

P

o i n t s

l MCL injuries in children are treated similarly to those in adults. l Isolated grade I and II MCL strains are treated nonoperatively. l Grade III MCL tears are also treated nonoperatively, but some surgeons advocate operative reconstruction.

l Fractures of the epiphysis or physis are more common than ligamentous injuries in children. l Stress radiographs may be indicated to rule out physeal injuries. l Fracture treatment revolves around anatomic reduction and stabilization with cast immobilization or open reduction with internal fixation. l Salter-Harris I and II injuries require 3 to 4 months before return to sports. l Salter-Harris III and IV injuries require 4 to 6 months before return to sports. l Leg-length discrepancies and angular deformity are ­complications of physeal fractures.

S U G G E S T E D

R E A D I N G S

Abel M: Orthopedic Knowledge Update: Pediatrics 3. Rosemont, Ill, American Academy of Orthopedic Surgeons, 2006, pp 281-289. Bertin K, Goble E: Ligament injuries associated with physeal fractures about the knee. Clin Orthop 177:188-195, 1983. Birch J: Instructional Course Lectures: Pediatrics. Rosemont, Ill, American Academy of Orthopedic Surgeons, 2006, pp 121-129. Johnson D, Mair S: Clinical Sports Medicine. Philadelphia, Mosby-Elsevier, 2006, pp 639-650. Jones R, Henley B, Frances P: Nonoperative management of isolated grade III collateral ligament injury in high school football players. Clin Orthop 213:137-140, 1986. Kannus P, Jarvinen M: Knee ligament injuries in adolescents: Eight year follow up of conservative management. J Bone Joint Surg Br 70:772-776, 1988. Salter R, Harris W: Injuries involving the epiphyseal plate. J Bone Joint Surg 45:587-622, 1963. Zionts L: Fractures around the knee in children. J Am Acad Orthop Surg 10: 345-355, 2002.

R eferences Please see www.expertconsult.com

S e c t i o n

D

Anterior Cruciate Ligament Injuries 1. Anterior Cruciate Ligament Injuries in the Adult Nicholas J. Honkamp, Wei Shen, Nnamdi Okeke, Mario Ferretti, and Freddie H. Fu

ANATOMY AND BIOMECHANICS Anatomy The anterior cruciate ligament (ACL) originates on the medial wall of the lateral femoral condyle. It courses anteriorly and medially across the knee joint and inserts into the tibial articular surface. It consists of two functional bundles, the anteromedial (AM) bundle and the ­posterolateral

(PL) bundle, named for their tibial insertion sites.1-3 The primary role of the ACL is to provide primary anteroposterior stability and secondary rotatory stability to the knee joint. Microscopically, the ligament is primarily composed of longitudinally arranged collagen fibrils. The diameter of these fibrils ranges from 20 to 170 μm with the diameter being largest in the distal region decreasing proximally. The percentage of total cross-sectional area occupied by collagen fibers remains significantly unchanged along the

Knee 1645

A

B

Figure 23D1-1   Sagittal plane (A) and oblique coronal plane (B) magnetic resonance imaging of the knee show the two bundles of the anterior cruciate ligament, the anteromedial (AM) and the posterolateral (PL) bundles.

length of the ligament. Multiple type III collagen–positive fibrils form a collagen fiber that is bundled together and ensheathed by a thin layer of connective tissue named the endotendineum. The bundled fibers and endotendineum make up what is called the subfascicular unit. Subfasciculi are collected in another connective tissue layer called the epitendineum, a much thicker layer than the endotendineum. This unit is named the fasciculus, the primary collagen unit of the ligament. The ligament is surrounded by the paratenon, which blends in with the epitendineum.4 The ends of the ligament deviate from this architecture and instead resemble fibrocartilage. This portion of the ligament includes oval-shaped cells, surrounded by a metachromatic extracellular matrix, that lay amongst the collagen fibrils.5 The blood supply to the ACL is primarily the middle genicular artery, a branch of the popliteal artery. It pierces

Figure 23D1-2   An arthroscopic view shows the two functional bundles of the anterior cruciate ligament, the anteromedial (AM) bundle and the posterolateral (PL) bundle. LFC, lateral femoral condyle.

the posterior capsule at the level of the intercondylar notch and courses along the posterior surface of the ACL within the synovial membrane surrounding the ligament. This synovial membrane originates at the posterior inlet of the intercondylar notch of the femur and extends distally to the tibial attachment of the ACL. Along the dorsal surface of the ACL, the middle genicular artery gives off ligamentous branches. The largest ligamentous branch, the tibial intercondylar artery, reaches the ACL at its proximal end and bifurcates just proximal to the tibial spine to supply both tibial condyles.6 The ligamentous branches form a periligamentous plexus, indicated as the source of profuse effusion and hemarthrosis typically seen after injury to the ACL. Blood vessels from the plexus penetrate the ligament horizontally and anastomose with a longitudinally oriented interligamentous network. Other sources of blood supply to the ACL are the inferior medial and lateral genicular arteries originating from the posterior surface of the popliteal artery.7 The innervation of the ACL comes from the posterior articular nerve, a branch of the tibial nerve in the popliteal fossa. It follows the path of the middle genicular artery piercing the posterior capsule and forms the popliteal plexus that tracks along the synovial lining and periligamentous vessels of the ligament.6 Although these appear to be primarily vasomotor, there are fibers in the intrafascicular spaces of the ligament that are similar in size to pain fibers.5,8 Furthermore, mechanoreceptors have been described on the surface of the ACL with their long axes running parallel with the ligament. The receptors themselves are housed at the insertion sites, primarily the femoral insertion site. These receptors may have proprioceptive qualities; however, there remains some uncertainty as to their exact function.9,10 There have been several fetal, cadaveric, arthroscopic, and radiographic studies detailing the gross anatomy of the ACL (Figs. 23D1-1 to 23D1-3).1-3,11-14 Fu and coworkers11 have shown that during early fetal development, the ACL is observed to consist of two distinct bundles, the AM and PL bundles. Fetal histologic examination of the ACL observed a septum of connective tissue separating the two bundles

1646 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

LFC

AM PL

A

B

Figure 23D1-3   A, Fetal knee joint displaying two distinguishable bundles of the anterior cruciate ligament (ACL), the anteromedial (AM) bundle and the posterolateral (PL) bundle. B, On histologic examination of the fetal ACL, a septum (arrow)   that separates the AM and PL bundles can be observed. LFC, lateral femoral condyle.

(see Fig. 23D1-3). The position and length of the bundles vary with changing angles of knee flexion and extension (Fig. 23D1-4). The selective recruitment of ACL fibers is partially the result of the changing length of the fibers with tension. As the fibers are recruited, the ligament has been shown to elongate by up to 3 mm with extension.15 From 0 to 30 degrees of flexion, the AM bundle shortens from its baseline length. With continued flexion from 30 to 70 degrees, the AM bundle lengthens back to its baseline length. Beyond 70 degrees of flexion, the bundle continues to elongate, beyond the baseline length, until it reaches maximal strain at about 120 degrees of flexion. The PL bundle is at maximal length and maximal strain when the knee is at full extension. As the knee is flexed, the PL bundle shortens, achieving minimal strain at about 120 degrees. At full extension, the bundles are parallel, and the femoral attachments sites are oriented vertically. With 90 degrees of knee flexion, the femoral PL attachment moves anteriorly, aligning the attachment sites horizontally. This motion creates

a situation in which the ACL bundles cross each other as the knee goes from extension into flexion. As mentioned earlier, the ACL has attachments on both the femur and the tibia. The specific locations of these attachments also play an important role in the selective recruitment of the collagen bands. This is due to the large area of each attachment site that allows various portions of the ligament to tighten at various degrees of knee flexion. The femoral attachment site is located on the posteromedial surface of the intercondylar notch on the lateral femoral condyle. The attachment site is circular, spanning an area of about 113 mm2. The tibial attachment site is located about 15 mm behind the anterior border of the tibial articular surface, medial to the attachment of the anterior horn of the lateral meniscus. This attachment is more oval, covering an area of about 136 mm2.2 For comparison, the cross-sectional area of the ACL mid-substance averages about 40 mm2. The femoral attachment sites of both bundles are often accompanied by bony landmarks. Arthroscopic and cadaveric studies (unpublished data) have shown two bony landmarks known as the cruciate ridge and the bundle ridge.11 The cruciate ridge runs proximal to distal, and it has been observed that no ACL fibers are anterior to this ridge. The bundle ridge runs anterior to posterior between the femoral insertion of the AM and PL bundles. This ridge is seen with careful dissection of the femoral ACL insertion site that maintains the bony anatomy of the intercondylar notch of the femoral condyle (Fig. 23D1-5). These two bony ridges can be used as useful landmarks in ACL reconstruction.

Biomechanics

Figure 23D1-4   In full extension, the femoral insertion sites of the anteromedial (AM, red) and posterolateral (PL, yellow) bundles of the anterior cruciate ligament are vertical. In 90 degrees of knee flexion, the insertion sites are horizontal. The fibers of the two bundles are parallel in extension and cross each other as the knee is flexed.

The main function of the ACL is restraint of anteroposterior translation of the tibia relative to the femur.16,17 It also acts as a secondary restraint to tibial rotation and valgus or varus stress. Age is an important factor in the strength of the ACL; typically, older ACLs fail with lower loads than do younger ACLs.18,19 With passive range knee extension, the ACL experiences forces of about 100 N, whereas walking produces about 400 N of force. Activities involving

Knee 1647

Figure 23D1-5   Arthroscopic view of the intercondylar notch. The cruciate ridge is above the femoral insertion sites of the anteromedial (AM) and posterolateral (PL) bundles, whereas the bundle ridge separates them.

acceleration, deceleration, or cutting maneuvers can produce up to 1700 N of force on the ACL.20 The ACL has a maximal tensile load of 2160 ± 157 N and a stiffness of 242 ± 28 N/mm.18 It is able to withstand strain of roughly 20% before failing. Furthermore, the ACL works in concert with many other anatomic structures in and around the knee joint that complement the function of the ACL to pro­ vide joint stability and limit pathologic knee motion.21-23 Thus, the ACL must experience an abnormal load to exceed failure capacity and sustain injury. Important variables that influence ACL strain are the position of the knee and the dynamic interaction of muscle activity. As shown by Beynnon and colleagues, increasing knee extension increases strain on the ACL.24 Restoration of anteroposterior translational stability alone does not correlate with subjective evaluations of knee stability. This suggests that rotational stability is an important function of the ACL and is a key to adequate knee stability.25 The cadaveric study of 10 knees by Gabriel and associates26 was an analysis of a combined rotatory load of 10 Nm (Newton-meter) valgus and 5 Nm internal tibial torque at 15 and 30 degrees of flexion. For the PL bundle, an in situ force of 21 N was recorded at 15 degrees and 14 N at 30 degrees. For the AM bundle, the in situ forces were 30 N and 35 N, respectively. This shows that at these angles, both the AM and PL bundles contribute not only to anteroposterior ­stability but also to rotational stability of the knee. Cadaveric biomechanical studies have shown that singlebundle ACL reconstruction is most successful in restoring anteroposterior knee stability but is insufficient in controlling combined rotatory loads of internal tibial and valgus torque.27 Yagi and associates28 performed a study comparing a single-bundle reconstruction with the femoral tunnel placed at the 11- or 1-o’clock position with anatomic double-bundle ACL reconstruction. This study concluded that

double-bundle ACL reconstruction is better able to resist anterior tibial translation at full extension and 30 degrees of flexion compared with the single-bundle technique. Furthermore, when a rotatory torque was applied at 15 and 30 degrees of flexion, the double-bundle ACL reconstruction had a response closer to that of the intact ACL compared with the single-bundle technique. Yamamoto and coworkers29 also did a comparison of the double-bundle ACL reconstruction to single-bundle reconstruction, with the femoral tunnel being placed at about the 10-o’clock position for the right knee. They reported that the doublebundle anatomic reconstruction better restored the anterior tibial translation at 60 and 90 degrees of flexion when compared with the single-bundle technique. Mae and associates30 duplicated similar results in their work using a two-socket quadrupled hamstring graft when tested by use of robotics. Another study performed by Tashman and coworkers31 looked at the in vivo kinematics of normal knees and knees subjected to single-bundle reconstruction. Subjects with a normal ACL were compared with a group of patients who underwent single-bundle ACL reconstruction to evaluate anteroposterior translation and knee rotation during downhill jogging. It was discovered that patients who underwent single-bundle ACL reconstruction had fully restored anteroposterior translation but increased tibial rotation as compared with subjects with a normal ACL during treadmill running. These studies indicate that double-bundle ACL reconstruction may be the best reparative technique for restoring the normal kinematics of the knee joint.

BASIC SCIENCE OF THE ANTERIOR CRUCIATE LIGAMENT Biologic Response to Anterior Cruciate Ligament Injury The ACL functions in unison with other anatomic structures in the knee to limit anterior translation and maintain knee joint stability. However, when injury does occur, it is often found that the complementary structures are also damaged or are insufficient to maintain the function of the lost ligament.21-23,32,33 Furthermore, when the ACL sustains injury or becomes deficient, other structures within the knee joint are at risk for injury, such as the menisci and the chondral surfaces.34-36 Typically, injury to extra-articular ligaments leads to the formation of local hematoma, which organizes into a fibrinogen mesh where inflammatory cells settle to mediate the natural inflammatory response. As the inflammation wanes, granulation tissue forms and reorganizes into fibrous tissue. The formation of fibrous scar tissue restores function to the ligament.37 The ACL, however, is intraarticular. The ACL is encased in only a thin envelope of synovial lining, unlike other extra-articular ligaments such as the medial collateral ligament (MCL) that have strong soft tissue encasement. When the ACL sustains injury, its synovial lining is compromised as well. Bleeding from this injury dissipates throughout the joint space and is unable to organize into fibrous tissue. Thus, the formation of fibrous scar tissue never occurs, and the ligament

1648 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

remains ­functionally incompetent. When the synovial lining remains intact, which often occurs in partial ligament injuries, blood clot formation occurs that initiates scar ­formation.38,39 Other factors that may be influencing the lack of healing seen in the ACL injuries include the cytokine profile of the intra-articular space. After injury, proinflammatory cytokines such as interleukin-1 and tumor necrosis factor-α are elevated, whereas protective anti-inflammatory cytokines like interleukin receptor antagonist protein are decreased. Such a hostile cytokine environment is thought to contribute to the poor healing observed in the ACL and may have long-term implications in the development of osteoarthritis.40 Despite current knowledge of the healing differences between intra-articular and extra-articular ligaments, much remains unclear. In comparing extra-articular ligaments such as the MCL with intra-articular ligaments such as the anterior cruciate, the cruciate ligament fibroblasts have a significantly higher production of extracellular matrix and collagenous proteins than the collateral ligament fibroblasts.41,42 However, under conditions of inflammation, the fibroblasts of the cruciate ligament exhibit lower mobility than those of the medial collateral.43 Furthermore, migration of the cells in the cruciate ligament is slower in comparison to the MCL.44 More work in vivo and in vitro is clearly needed in understanding and characterizing the healing response in ligament tissue.

Biology of Anterior Cruciate Ligament Graft Reconstruction The poor healing of the ACL, combined with the risk for subsequent injury to the other supporting structures, has treatment implications. With the ligament unable to heal on its own, reconstructive surgery is often performed wherein tendon grafts are transplanted to replace the deficient ligaments. After transplantation, biologic modifications occur to incorporate the grafts into the host knee. Initially, the tendon graft is subjected to the process of inflammation and avascular necrosis. After the donor fibroblasts have died, revascularization of the remaining graft tissue occurs. Inflammation and revascularization is seen within 20 days after implantation but takes 3 to 6 months for revascularization to finalize. Next, migration and repopulation of the graft by host fibroblasts occur. This is reported to be complete within 4 to 6 weeks at which time no donor fibroblasts are detectable. Lastly, there is gradual remodeling of the graft and remodification of the collagenous structure as the collagen realigns longitudinally along the graft in a similar fashion to the normal cruciate ligament.45,46 This collagen remodeling and realignment of the graft can be seen on histologic evaluation at 6 weeks after transplantation and can continue for up to 6 months.37 The transplanted graft has the ability to develop and adapt to mechanical demands. However, it never fully resembles the structure of the original graft tendon tissue or the native cruciate ligament it replaced. The ultrastructure of the ligament, the size of the fibrils, and the concentrations of the extracellular matrix differ significantly from the original tissue.37,47

EPIDEMIOLOGY Injury of the ACL accounts for roughly 40% to 50% of all ligamentous knee injuries. Injury to the cruciate ligament is even more common in the young athletic population. In the general population, 70% of injuries that occur are secondary to sports activity.48 Among the activities related to ACL injury, skiing and soccer are ranked as the highestrisk activities for cruciate injury. Women display a higher propensity to ACL injury compared with men in most sports, including basketball, volleyball, and soccer.49 Before advances in reconstructive surgery, many of these injuries would limit sports. Currently, cruciate ­ ligament reconstruction is allowing for the return to sports activity for most patients. Although results vary, many return to sports activity after about 6 months when given appropriate postoperative therapy. Recognition of the high prevalence of ACL injury in the sporting population will assist in prompt diagnosis and treatment as well as reduce time lost to injury.

CLASSIFICATION There is no standardized classification system widely used in the evaluation of ACL injuries. We prefer to divide patients into those with multiple ligamentous injuries, including an ACL injury, and those with isolated ACL injuries. For those with isolated ACL injuries, we attempt to determine whether the injury is a partial versus a complete tear. We subdivide partial injuries by which bundles are involved, as we commonly perform ACL single-bundle reconstruction and augmentation surgery in those cases in which one bundle remains functionally intact and the other bundle is torn. Although a further subclassification, which includes any associated chondral or meniscal pathology, is noteworthy, we do not formally separate these injuries into different sub-categories.

EVALUATION Clinical Presentation and History A thorough patient history is the initial step to diagnose and treat ACL injuries. Obtaining a complete history of the injury will direct the physical examination toward a complete and accurate diagnosis. Mechanism of injury, initial symptoms, previous injuries, time since injury, and any late sequelae, including reinjuries, are all pertinent information to obtain while taking a presenting patient’s history. Mechanism of injury can be crucial in the proper diagnosis of injury. Unfortunately, most of the time, patients are unable to recall precisely the events or mechanisms of injury.50,51 When this situation arises, it is often helpful to inquire about witness accounts and, if possible, video footage of the injurious event. Often in sports injuries, a coach, personal trainer, or teammate may be able to give an account of the events. Among the more common mechanisms of injury are low-energy injuries that occur during athletic activities. These types of injuries include direct contact injuries and those secondary to indirect noncontact mechanisms such

Knee 1649

as sudden deceleration or rotational maneuvers. Direct contact injuries often result in hyperextension or valgus stress on the knee, resulting in cruciate ligament injury. Of the two types, noncontact injury is the most common, with only about one third of patients detailing a contact mechanism of injury.52 High-impact or high-energy injuries such as motor vehicle crashes less commonly cause cruciate ligament injuries. With high-energy mechanisms, patients often present with concurrent injuries, including musculoskeletal, pelvic, abdominal, spine, and head injuries. In such cases, a thorough history and physical examination is highly important to assess and treat these associated injuries. The initial symptoms and sensations at the time of injury are also very important in determining a diagnosis. Patients are often able to recall common sensations such as popping or tearing at the time of injury; these represent some of the most common initial complaints of cruciate ligament injury.51,53 Other symptoms that a patient may complain about include the inability to bear weight on the injured leg and instability or the sensation of the knee “giving out.” Attention should be given to the ability to continue with competitive activity because many athletes are unable to participate after sustaining an acute injury.54 Post-traumatic swelling of the knee joint is another indicator of ligament injury. Swelling is the physical manifestation of hemarthrosis following disruption of the ligament’s blood supply. This event can be seen within 12 hours after injury.51,55 All of the aforementioned symptoms are indicative of injury to the ACL; however, they are not exclusive to the cruciate ligament because they may occur with injury to other anatomic structures in the knee such as the MCL, the menisci, the patella, or the posterior cruciate ligament (PCL). Conversely, absence of symptoms or signs such as a hemarthrosis does not exclude a diagnosis of ACL injury. In documenting a complete history, activities of daily living are important pieces of patient information to collect. Such facts include activity level, job requirements, sports activity, and future plans of activity. These factors weigh heavily in making treatment decisions. Surgically, such factors may dictate the choice of graft, postoperative rehabilitation speed, as well as surgical timing.

Physical Examination and Testing Physical examination is of monumental importance in diagnosing ACL injuries. Although patient history is useful in indicating ligament injury, the physical examination should be able to establish a definite diagnosis of injury to the cruciate ligament in most cases. Timing of the physical examination must be considered when evaluating an injury. Examinations performed immediately after an injury (before the onset of swelling, pain, and reflex muscle splinting) are more accurate than after the injury response has been initiated. Typically, this can occur only when the examining clinician is present at the time of the injury. If the examination is delayed and the initial symptoms have manifested, the patient may be difficult to evaluate, decreasing the accuracy of the examination. In these situations, it is best to repeat the examination in a few days. The initial step in examination is observation of the knee, including the presence of malalignment or swelling.

Malalignment can be indicative of a fracture or a sign of knee dislocation, both of which may require urgent medical attention. Depending on the time frame of the examination, an effusion may be detectable. This often appears about 4 hours after injury and may not be present with immediate examination of an injured knee. Therefore, a lack of an effusion is not an indication to exclude cruciate ligament injury. Furthermore, the severity of the injury may affect the presence of an effusion as well. With more severe injuries that injure a greater portion of the knee and the surrounding tissues, the consequential effusion may have several pathways to diffuse from the primary site of injury. After inspection, manual palpation of both knee joints is performed. Accurate assessment begins with palpation of the unaffected knee provided the injury is unilateral. Doing so will provide the examiner with a referential baseline with which to compare the injured limb. It also has the secondary benefit of familiarizing the patient with the examination procedure and thus relaxing the patient and preventing guarding, which can limit examination findings. The ACL is itself unable to be palpated, but palpation still plays an important role in cruciate ligament injury evaluation. Palpation is useful in detecting the presence of an effusion that may have been missed on inspection. It also serves to quantify the degree of effusion if present. Palpation is also a good examination tool to detect injury to surrounding knee structures. Medial and lateral joint line tenderness may indicate concomitant meniscal or chondral injury. Other structures to palpate include the medial and lateral collateral ligaments and their insertion sights. Functional testing of the knee should then be performed. This includes both active and passive range of motion testing to check for loss of motion. Various factors may cause loss of motion, including pain in the knee, a large effusion, an incompetent extensor mechanism, or a mechanical block. Effusions may be aspirated to alleviate pain and improve motion. Examining the aspirate can provide further clinical clues by confirming the presence of hemarthrosis, which can be indicative of ligament injury, or the presence of fat globules in the fluid, which indicates a fracture within the knee. If there is a mechanical block, the differential diagnosis should include meniscal tears, ruptured ACL obstruction, or loose body. Use of other diagnostic tools, such as radiographs or magnetic resonance imaging (MRI), may be needed to fully discern the cause of the obstruction. Stability testing of the knee should follow next. Standard practice should include testing of not only anterior stability but also posterior, varus, valgus, and rotational stability. Anterior stability testing usually employs the use of the Lachman test. The Lachman test is performed while the knee is flexed at 20 to 30 degrees (often bolstered with a thigh support). In this position, a manual anterior force is applied to the proximal tibia while the distal femur is stabilized with the opposite hand. The anterior laxity is assessed in the degree of anterior translation of the tibia relative to the femur and in the firmness of the end point at which translation is halted. From patient to patient, there is natural variance of normal laxity within the joint. Therefore, in assessing anterior stability of an injured individual, there is no absolute degree indicative of disease; rather,

1650 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

a comparison is made between the injured and the contralateral normal knee. The degree of translation is categorized in grades of laxity. Grade I laxity describes 1 to 5 mm of increased anterior translation. Grade II laxity is 6 to 10 mm, and grade III is more than 10 mm of translation when compared with the opposite, uninjured knee.56 In addition to the Lachman test, arthrometers have been employed to provide objective instrumented laxity measures of ACL laxity. The KT-1000 (MEDmetric, San Diego, CA) is the mostly commonly cited device. Although not replacing the function of the Lachman test, arthrometric examination does provide some advantages over the manual examination, including the assessment of large patients when the manual examination can be limited owing to lack of control of a large limb. In similar fashion to the Lachman test, KT-1000 testing is performed with thigh support placing the limb in 20 to 30 degrees of knee flexion. A footrest is included with lateral supports to prevent external rotation of the tibia, maintaining a neutral rotation of both limbs (an issue to be aware of when performing the Lachman test). The device is aligned with the joint line, and positioned with a sensor pad on the patella and tibial tubercle. Two Velcro straps attached to the arthrometer are used to fasten the device in position vertically along the lower limb. Before zero point calibration through trial runs, the patient is asked to relax the thigh muscles; it is best to have the patient resting in the supine position. A vertical anterior force is applied with the use of a handle, and measurements of translation are made with 15, 20, and 30 pounds (67, 89, and 134 N, respectively) of force on the tibia. At each force level, the device emits an audible tone denoting the force threshold for measurement. Additional testing with the KT-1000 includes the maximal manual pull test. This is performed by applying an anterior force on the proximal posterior tibia, much like the Lachman and recording the maximal measurement displayed on the arthrometer. This device is used on both the injured and normal knee for comparison. Accuracy is a key concern when employing the use of either the Lachman test or the KT-1000 to assess laxity. With the Lachman test, the knee must be maintained in a neutral alignment while performing the examination. It has been shown that internal or external rotation decreases the degree of anterior translation, yielding a false-negative reading and potentially masking true injury.20 Other factors to consider include the presence of a large effusion or muscle splinting, which limits anterior translation as well. PCL incompetence often can mislead an examiner to overestimate anterior translation. This occurs when an insufficient PCL allows for posterior sag of the tibia. If unnoticed, the tibia would appear to translate anteriorly a greater distance, giving a false-positive result. Accuracy of the KT-1000 is affected by a variety of issues. Among these include user error, such as malalignment of the arthrometer along the tibia or joint line and improper application of anterior force. Relaxation of the quadriceps is also important in ensuring accuracy of arthrometric measures. Patient apprehension or pain may make relaxation of the thigh muscle impossible when performing the KT-1000 examination; however, it is essential in obtaining correct information about the state of the ACL.

The anterior drawer test is another examination technique used to evaluate anterior translation of the tibia. The knee is placed in 90 degrees of flexion, and the foot is held in place throughout the examination. Next, manual application of an anterior force is performed on the posterior proximal tibia. Measurement of the observed translation is graded like that of the Lachman test. The Lachman test is a more sensitive physical examination test than the anterior drawer test and therefore is used more frequently in the clinical examination. The pivot shift test is another clinical examination test to measure instability in the knee secondary to ACL injury. The test begins with the knee in full extension, and the patient is asked to relax the musculature of the limb being tested. A valgus stress is placed on the tibia, while an axial load and internal rotation are simultaneously applied. The knee is then slowly flexed with these applied forces. During this motion, the lateral side of the plateau subluxates to a greater extent than the medial side.57 With further flexion, the lateral tibia reduces, producing the pivot shift. This test is graded on the degree of subluxation and reduction of the lateral compartment of the knee, with grade 0 having no detectable shift, grade I having the tibia in a smooth glide during reduction, grade II having an abrupt reduction, and grade III having the tibia momentarily lock in the subluxated position before reduction. The sensitivity of this examination has shown to be highly dependent on patient apprehension, with its highest accuracy performed with the patient under anesthesia.54 Accuracy of the examination is influenced by injury to other structures in the knee as well, such as meniscal injuries and MCL injuries, which tend to decrease a detectable shift, leading to a misdiagnosis of ACL injury.

Imaging Radiographic studies of patients presenting for evaluation of ACL injury are often complementary in the primary diagnosis of ligament injury. A thorough history and physical examination of the knee can usually provide all the information to accurately diagnose ACL deficiency. However, there are benefits to performing radiographic imaging studies that can affect treatment and rehabilitation. Plain radiographic imaging plays a primary role in the exclusion of associated injuries in the evaluation of the ACL. Such associated injuries include lateral capsular avulsions (known as Segond’s fractures)58 and tibial eminence avulsion fractures seen in younger patients or those with osteopenia. Plain films can also alert the physician to the presence of loose bodies, proximal and distal knee fractures, degenerative disease, and osteophyte formation in chronic ACL-deficient knees.59,60 MRI is a highly useful tool for confirming the diagnosis of ACL disease. It is highly specific and sensitive and is able to provide information on the other intra-articular structures in the knee as well as evaluate both bundles of the native ACL. Fu and coworkers have published studies detailing the visualization of both bundles using MRI.11-13 They demonstrated use of viewing planes that follow the natural course of the ligament to improve the visualization of the AM and PL bundles (see Fig. 23D1-1B). The ability to visualize the individual bundles provides important

Knee 1651

information for surgical reconstruction of an incompetent cruciate ligament, especially when the double bundle reconstruction technique is employed for treatment. The presence of a chondral injury or bone bruise on MRI is highly indicative of ACL injury. About 80% of ligament injuries are accompanied by a bone bruise.61,62 These lesions are commonly located in the lateral tibial plateau and lateral femoral condyle. They are the result of abnormal impaction of the articular surfaces after ACL injury as the lateral compartment subluxates anteriorly. The presence of a bone bruise may also affect rehabilitation, leading to prolonged recovery of range of motion and stability, as was shown by the work of Johnson and coworkers.63

TREATMENT CONSIDERATIONS Gender Issues Female athletes have a fourfold to sixfold greater ­incidence of ACL injuries compared with male athletes participating in the same cutting and landing sports.64 When combined with an exponential increase in high school and collegiate sports participation among females in the past 30 years,65,66 the number of ACL injuries in female ­athletes has ­skyrocketed. The reasons for this gender disparity in ACL injuries are likely multifactorial. Several theories have been published to explain this disparity, and they can generally be divided into anatomic, hormonal, neuromuscular, and biomechanical differences. Anatomic differences include static standing knee alignment, notch width differences, joint laxity, foot alignment, and body mass index. Women have a relatively wider pelvis, which can lead to an increased Q angle. Some authors have linked an increased Q angle to increased ACL injury rates,67-69 whereas others have not.70 Because Q angle is generally a measurement of the patellofemoral joint and not knee valgus, other factors may be contributing to these mixed results.71 A smaller femoral notch width, even when corrected for bone width (so-called notch width index), has been show by some authors to increase the risk for ACL injury independent of gender,72-75 whereas others have noted no difference in notch width between genders nor an association between notch width and injury.70 Increased joint laxity is more commonly found in female athletes, and this increased laxity affects both sagittal plane (ACL) and coronal plane (MCL) motion.74,76,77 Such coupled anterior and valgus loading of the knee is a common mechanism for an ACL injury. Increased joint laxity in women can also be found in the foot, whereas increased foot pronation (“navicular drop”) is a predictor of anterior tibial translation, which may further strain the ACL.78,79 Increased ACL injury risk appears in girls typically around age 12 years, which coincides with a natural increase in body mass index (BMI). Uhorchak and associates74 and Buehler-Yund80 both found body mass index to be a significant risk factor for knee injury. The influence of hormonal changes during the female menstrual cycle and its potential effects on ACL injuries has generated significant controversy. Although some

s­ tudies have confirmed increased ACL injury risk during the ovulatory or postovulatory (early luteal) phases,81,82 others have not found any relationship between cycle phase and injury risk.83 A recent meta-analysis of nine prospective cohort studies on the topic found that six of the studies failed to show any relationship between cycle phase and anterior knee laxity, whereas the remaining three showed increased knee laxity during the ovulatory and early luteal phases (days 10 to 14 in a standard 28-day cycle).84 Oral contraceptives may have a role in decreasing the risk during this period.81 Neuromuscular and biomechanical differences also exist between male and female athletes. ACL injuries occur most commonly during periods of high dynamic loading of the knee joint, when muscular forces are unable to ­sufficiently dampen joint loads such that the passive ligamentous restraints are subjected to threshold failure loads. The gender difference in neuromuscular and subsequent biomechanical forces across the female knee joint significantly affects dynamic knee stability.85-87 Female athletes show increased activation of the quadriceps relative to the hamstrings (Q/H ratio)88,89 as well as decreased ratio of firing of medial to lateral quadriceps and hamstrings.88,90 Combined, these imbalances cause anterior tibial translation and valgus, which directly loads the ACL. The neuromuscular differences in female athletes have prompted research into neuromuscular interventional training programs in females in an effort to decrease the risk for knee injuries in general and ACL injuries in particular. A recent meta-analysis by Hewett and coworkers70 examined six published interventions targeted toward ACL injury prevention in female athletes. Four of the six significantly reduced knee injury incidence, and three of the six significantly reduced ACL injury incidence in female athletes. In evaluating those studies that successfully reduced ACL injuries among female athletes, plyometric training, combined with biomechanical analysis, and technique training were common components. Furthermore, training sessions should be performed more than 1 time per week, and the duration of training should be a minimum of 6 weeks in length.

The Older Patient The health benefits of improved physical fitness are readily accepted. This has led to a recent increase in the activity level of patients older than 40 years. More sports-related injuries, including ACL injuries, are being seen in this group. Traditionally, patients older than 40 years were treated nonoperatively after ACL injury.91 It has been demonstrated, however, that a significant portion of these patients have episodes of instability despite activity modification and progressive degenerative changes.91,92 Clearly, patients older than 35 years do benefit from reconstruction of the ACL and can expect results comparable with those in groups of younger patients.93-96 The ACL deficiency must be addressed in the early stages after injury, however, before chronic degenerative changes occur. Results of ACL reconstruction in older patients with long-term, chronic ACL deficiency are not as predictable. Patients older than 40 years willing to modify their activities can do well with an ACL-deficient knee, but we recommend

1652 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������    TABLE 23D1-1  Clinical Studies of Partial Anterior Cruciate Ligament (ACL) Tears Study

N

Follow-Up (yr)

Age (yr)

Isolated Partial Tear

Exclusion Criteria

Lysholm Score

Activity Rate of Return

ACL Insufficient during Follow-up

McDaniel, 1976101 Odensten et al, 1985104 Kannus & Jarvinen, 198792

9

1.3

20

6

>75% tear

N/A

N/A

17%

21

5.8

25.5

6

N/A

95(84-100)

N/A

14%

41

8

32

10

N/A

90(17-100)

N/A

N/A

Sandberg & Balkfors, 1987105 Noyes et al, 1989103 Buckley et al, 198999

29

3

26

29

N/A

94 (mean)

21%

62%

32

5

21.4

15

>75% tear

N/A

21%

38%

25

4.1

25

12

>75% tear

60%

Fruensgaard & Johannesen, 1989100 Sommerlath et al, 1992106 Bak et al, 199798 Messner & Maletius, 1999102

41

1.5

29

18

N/A

21

12

29

0

60% G/E 44% (Feagin-Blake score) 92(69-100) stable 50% knees; 84(45-100) unstable knees 93 32%

56

5.3

27

56

>75% tear

86(52-100)

30%

23%

22

240

29

0

>50% tear

95(71-100)

14/22 after rehab, 7/22 at 12 yr f/u 5/22 at 20 yr f/u

N/A

ACL reconstruction for those patients who wish to remain active, particularly if they wish to remain involved with high-risk activities, and for those patients who are “physiologically” young. This subset of patients has done well in our experience after ACL reconstruction; however, this group represents a skewed population, consisting of the most active and motivated of all middle-aged patients. Older patients do present with both medical and nonmedical issues that may not be seen in the younger population and must be taken into account in planning an ACL reconstruction. Medically, there may be associated health concerns requiring appropriate evaluation and coordination with primary care physicians or internists. There are often more stringent personal or professional responsibilities, which may have an impact on the timing of surgery and the ability to rehabilitate the knee. In an effort to reduce operative morbidity and enhance recovery, the use of allograft has gained popularity. Recent follow-up ­studies have shown allograft ACL reconstruction to be ­comparable to autograft in this older ­population.94,96

Partial Tears What constitutes the diagnosis of a partial ACL tear is controversial. Some have chosen to define a partial tear on the basis of the physical examination findings, whereas others have based the diagnosis on findings at arthroscopy. The incidence of partial tears ranges from 10% to 28% of all ACL injuries.97 In our experience, the incidence tends to

Radiographic Follow-up

15% of grade II ACL injuries had degenerative joint disease

44% 9%

3 of 12 had degenerative joint disease progression; 9 of 12 were stable

be toward the lower end of this range. Although the natural history of complete ACL ruptures has been well defined, patients with partial ACL tears have a less predictable clinical course. A review of the literature on the natural history of nonoperatively treated, isolated, arthroscopically confirmed partial ACL tears is shown in Table 23D1-1.92,98-106 Critical differences such as the degree of chondral and meniscal pathology, the type of rehabilitation, presence of symptomatic instability, and the lack of long-term radiographic follow-up make broad conclusions difficult to determine. The diagnosis of partial ACL tears can be challenging. Findings on clinical examination, including Lachman and anterior drawer testing, can be subtle. We use a combination of history, physical examination findings, KT-2000 arthrometer testing, and findings on MRI. The two-bundle concept of the ACL anatomy is well documented.107 Partial ACL tears involving the PL bundle, which has a large contribution to rotatory stability, often manifest as increases in pivot shift testing; similarly, partial ACL tears involving the AM bundle, which has a large contribution to sagittal plane stability, often manifest as increases in Lachman or anterior drawer testing.108 Liu and colleagues have shown, with KT-2000 arthrometer testing of partial tears, that mild to moderate injuries (one half to full tear of a single bundle) produce only small changes in the anterior tibial translation at different force levels.109 Additionally, a normal KT does not preclude the presence of a partial ACL injury.

Knee 1653

   TABLE 23D1-2  Mediolateral Distribution of Meniscal   Tears in the Acutely Injured Anterior Cruciate   Ligament (ACL)

Author DeHaven McDaniel Noyes Woods Indelicato Cerabona McCarroll Henning Warren Hirshman Sgaglione Shelbourne Shelbourne Sherman Paletta (skiers) Paletta (nonskiers) Keene Sgaglione Spindler Ihara Overall

Incidence of ­Meniscal Tears in Acute ACL Injuries (%)

No. of Injured Menisci

Medial (%)

Lateral (%)

65 82 72 50 77 46 75 NA 65 53 60 NA 67 45 41

56 10 40 64 40 50 18 76 — 127 43 32 286 27 33

37 50 37 53 65 56 67 55 50 39 44 13 42 70 24

63 50 63 47 35 44 33 45 50 61 56 87 58 30 76

63

54

52

48

81 73 68 NA

57 24 50 40 1127 (total)

40 46 40 25 44

60 54 60 75 56

Medial vs. Lateral Distribution

From Bellabarba C, Bush-Joseph CA, Bach BR Jr: Patterns of meniscal injury in the anterior cruciate–deficient knee: A review of the literature. Am J Orthop 26:18-23, 1997.

   TABLE 23D1-3  Mediolateral Distribution of Meniscal Tears in Chronic Anterior Cruciate Ligament (ACL) Insufficiency

Author Warren McDaniel Noyes Woods Indelicato Fowler Warren Aglietti Kornblatt McCarroll Finsterbush Henning Hirshman Irvine Keene* Keene† Sgaglione Satku Overall

Incidence of Medial vs. Lateral Meniscal Tears Distribution in Chronic ACL No. of Injured Injuries (%) Menisci Medial (%) Lateral (%) 98 86 92 88 91 73 — — 82 100 65 — 76 86 — 89 100 58

107 27 17 135 65 38 34 110 36 6 23 111 119 127 54 02 22 61 1184 (total)

87 81 59 64 69 50 76 84 75 67 74 71 90 44 59 58 59 74 70

13 19 41 36 31 50 24 16 25 33 26 29 10 56 41 42 41 26 30

*Less than 12 months after injury. †More than 12 months after injury. From Bellabarba C, Bush-Joseph CA, Bach BR Jr: Patterns of meniscal injury in the anterior cruciate–deficient knee: A review of the literature. Am J Orthop 26:18-23, 1997.

Associated Injuries We find that reviewing the MRI with an experienced musculoskeletal radiologist is often helpful in correctly diagnosing a partial ACL tear. Standard MRI cuts have variable sensitivity in diagnosing partial ACL tears, ranging from 0.4 to 0.75.110 The addition of oblique sagittal and coronal MRI has been shown to increase the diagnostic accuracy.111 The finding of a residual straight and tight ACL fiber seen on at least one image, a focal increase in ACL signal intensity, and the absence of a bone bruise are signs suggestive of a partial ACL tear.112,113 When the diagnosis is confirmed, the treatment of ­partial ACL tears is still difficult. Age, activity level, degree of laxity, associated injuries, and the presence of symptomatic instability are all important factors to consider. Patients compliant with a postinjury rehabilitation protocol emphasizing hamstring strengthening, brace wear, and activity modification may respond favorably to nonoperative treatment. Other patients not willing or able to mentally and physically cope with such a program may be better served with reconstructive surgery. As our knowledge of ACL double-bundle anatomy has increased, we have performed single-bundle reconstructions in those patients found to have single-bundle ACL tears (AM or PL bundle) with the remaining bundle functionally intact. In our last 360 patients operated on for ACL injury, we have performed 16 single-bundle augmentation reconstructions.

The association of ACL tears with injuries to other structures of the knee has long been recognized.114-116 O’Donoghue115,116 coined the phrase “the unhappy triad” in referring to the association of ACL injury with MCL and medial meniscal tears. More recently, it has been noted that lateral meniscal tears are more commonly seen in association with combined ACL and MCL injuries.117-119 Bellabarba and coworkers120 performed an extensive review of meniscal injuries associated with acute and chronic ACL insufficiency (Tables 23D1-2 and 23D1-3). They found a 41% to 81% incidence of meniscal tears in acute ACL injuries; 56% were lateral tears, and 44% were medial tears. In chronic ACL-deficient knees, the rate of associated meniscal injury ranged from 58% to 100%. In this population, medial meniscal tears were more common, representing 70% of all meniscal injuries. Special consideration should be given in evaluating the meniscus in patients with concurrent ACL tears because this combination frequently changes management and prognosis. Prospective studies have shown that MRI is capable of both high sensitivity and specificity (>90% each) in the detection of meniscal tears in ACL stable knees.121 However, the accuracy of meniscal tears by MRI drops significantly in the context of a concurrent ACL tear (Table 23D1-4).121-124 Many of these tears involve the peripheral portion of the posterior body and horn of the lateral

1654 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������    TABLE 23D1-4  Sensitivity and Specificity of Magnetic Resonance Imaging in Context of Anterior Cruciate Ligament (ACL) Tear

Sensitivity in MM Tears Sensitivity in LM Tears Study

ACL Intact

ACL Torn ACL Intact

ACL Torn

DeSmet & Graf, 1993121 Jee et al, 2004122 Rubin et al, 1998124 Justice & Quinn, 1995123

0.97

0.88

0.94

0.69*

N/A 0.98 0.97

1 0.84* 0.92

N/A 0.9 0.84

0.62* 0.83 0.77

   TABLE 23D1-5  Treatment Factors Associated with Meniscal Tears

Factors Associated with Healed Meniscal Tears

Factors Associated with Nonhealing Meniscal Tears

Lateral meniscal tears Partial-thickness tears

Medial meniscal tears Ability to lock meniscus arthroscopically Rim width > 4 mm Radial tears

Rim width < 4 mm Concurrent anterior cruciate ligament reconstruction Vertical longitudinal tears

Meniscal body tears

*P < .05 LM, lateral meniscus; MM, medial meniscus.

meniscus; surgeons should pay particular attention to this area during the diagnostic arthroscopic examination. Because the importance of the meniscus in knee stability,125,126 load transmission,127,128 and the prevention of long-term arthrosis has been proved,129,130 the need for meniscal preservation is essential. If meniscal repair is not possible, partial meniscectomy should be chosen with care taken to leave an intact outer rim of meniscus. Meniscal repairs done in conjunction with ACL reconstruction have a higher rate of healing.131,132 Reasons for this include the beneficial effects of a hemarthrosis and fibrin clot formation which aid in repair,133,134 the protective effect of a competent ACL on decreasing the mechanical forces imposed on the meniscus,130 and the nondegenerative nature of the meniscal tears seen in associated acute ACL injuries (versus the degenerative meniscal changes seen in meniscal tears with intact ACLs).135 Longitudinal tears less than 10 mm or partial-thickness tears of the lateral meniscus have been shown to have a high rate of healing and a low propensity for development of symptoms; these tears have been shown to do well with conservative treatment and concurrent ACL reconstruction.136,137 The same has not been true of the medial meniscus tears, which have been shown both clinically and on second-look arthroscopic examinations to have a high propensity for nonhealing or progression with conservative treatment.136-138 Therefore, we aggressively treat all medial meniscal tears at the time of ACL reconstruction. Partial lateral meniscus tears we treat conservatively, with full-thickness lateral tears treated surgically depending on multiple factors (Table 23D1-5). MCL injuries are often found in association with ACL injury. Ninety percent of all knee ligament injuries in young and active individuals are ACL, MCL, or combined ACL and MCL injuries.139 Isolated MCL injuries are generally regarded to heal with nonoperative treatment.140,141 The treatment, however, of combined high-grade MCL and complete ACL injuries has been controversial. Many patients present with low-grade MCL strains and complete ACL injuries and are treated with ACL reconstruction only. In the presence of a severe MCL injury, concern exists that the reconstructed ACL may see increased forces and that the nonreconstructed MCL will not heal properly in such an environment. However, evidence favoring the nonoperative treatment of MCL injuries in combined ACL and MCL injuries comes from animal studies142,143; ­ however,

animal studies do show that repaired MCLs may have higher ultimate strengths than unrepaired MCLs.144,145 Adequate healing of the nonoperatively treated MCL in the context of ACL reconstruction, however, has been shown in multiple retrospective studies.146-149 Additionally, prospective studies have documented the good outcomes in patients treated with ACL reconstruction and nonoperative bracing of the MCL.150,151 The one randomized, prospective study comparing ACL reconstruction with operative versus nonoperative treatment of combined ACL and grade III MCL injuries found no difference between the two groups.152 Concern exists regarding the risk for arthrofibrosis in the setting of a multiligament (ACL and MCL) injured knee in which acute operative treatment is undertaken. We do not delay surgery a set time after injury to avoid arthrofibrosis; instead, we routinely wait until the patient has attained full extension and has achieved flexion to 120 degrees and until the majority of the acute hemarthrosis has resolved (typically 5 to 10 days).

Natural History For fair comparisons to be drawn on the effects of ACL reconstruction versus conservative treatment in ACL injuries, knowledge of the natural history of the ACL-deficient knee is required. Unfortunately, no large prospective trials on the natural history of ACL-deficient knees have been adequately done. With the available studies done at present, there is a general consensus among practicing orthopaedic surgeons that chronic ACL deficiency can lead to chronic functional instability, which may increase the risk for meniscal or chondral injury. It is clear from the literature that the primary complaint of ACL-deficient patients is recurrent episodes of givingway.15,153 It is such recurrent episodes that lead to less than 20% of patients returning to their preinjury level of activity.153-155 These recurrent episodes of instability also place these patients at increased risk for new and recurrent meniscal and chondral injuries, with such injuries leading to the development of further intra-articular damage153,154,156,157 and ultimately osteoarthritis of the knee.158,159 Although some studies have shown no increased rates of osteoarthritis in conservatively treated ACL-deficient knees as compared with uninjured knees,160,161 studies with greater than 10 years of follow-up have shown greatly increased rates in ACL-deficient knees.154,159,162 It appears that at least 10 to 15 years are needed for such radiographic changes to occur. Levy and associates estimated the incidence of

Knee 1655

meniscal tears in patients with unreconstructed ACL injuries at 40% by 1 year, 60% by 5 years, and 80% by 10 years after the initial ACL disruption.163 This is likely accelerated in young, active patients who attempt to return to high-level activity. Nebelung and associates showed that 35 years after ACL injury treated conservatively in a group of former East German Olympic athletes, 18 of 19 patients required at least partial meniscectomy, and 10 of those 19 patients had already undergone total knee replacement secondary to severe osteoarthritis.164 However, a subgroup of patients is able to compensate for their ACL deficiency and do well with nonoperative treatment. Multiple studies have attempted to define factors that would prospectively identify this subgroup of patients.165-168 Such an identification system would both spare those patients from surgery who would do well with nonoperative treatment and identify those patients who, treated conservatively, are at risk for further reinjury, which ACL reconstructive surgery could help avoid. Such factors have included the number of preinjury International Knee Documentation Committee (IKDC) level I or II activities (cutting and jumping activities), KT-1000 arthrometer manual maximal injured minus normal displacement difference exceeding 5 to 7 mm, inability to perform a onelegged hop test, and an inability to regain normal gait parameters by 40 days after injury.165-168 Although age at the time of injury has not been consistently proved an accurate predictor in identification of this subgroup, age by itself is correlated with increased activity levels that include cutting and jumping activities. Thus, older patients have been shown to be more likely to successfully adapt their activities to avoid recurrent instability episodes.91

TREATMENT OPTIONS Nonoperative Treatment As stated in the previous section, ACL reconstruction is recommended in patients who either are young, are active in high-level sports involving cutting or pivoting, or whose physical examination reveals greatly increased knee laxity. The goal of ACL reconstruction is to return functional stability to the knee to provide for a return to full activities as well as to prevent any further injury to meniscal or chondral surfaces that can lead to early-onset osteoarthritis. Advances in the surgical technique, anesthetic and pain management, and postoperative rehabilitation have reduced morbidity and increased functional outcomes in ACL reconstruction. With these advances, the use of conservative treatment for ACL injuries has fallen out of favor for patients in the aforementioned groups. Older and more sedentary patients may do well with conservative treatment that includes aggressive quadriceps and hamstring strengthening.91 However, even in this subgroup, some patients may continue to experience recurrent episodes of functional instability. Thus, the most important part of this decision process involves a thorough discussion with the patient regarding reasonable functional outcomes with each treatment and the activity modifications and chance of success that accompany operative versus conservative treatment.

Operative Management Early surgical treatment of ACL injury involved attempts at primary repair.3,169 Early reports were thought to be promising in returning stability and increasing patient function170,171; however, intermediate (5 years) and longterm follow-up (15 years or more) studies documented poor subjective, objective physical examination, and radiographic results.172-174 Attempts at improving these results focused on augmentation procedures, both intra-articular and extra-articular, that commonly used hamstring tendons or the iliotibial band. Initial follow-up studies showed improvements in knee stability, return of function, and a decreased rate of revision surgery.175,176 Although these results were consistently better than primary repair, longer term follow-up studies again showed deteriorating results as compared with more modern autogenous reconstruction techniques that began to appear.172,175,177 Thus, both primary repair and augmentation procedures fell from favor. With continued medical and technologic advances, the use of prosthetic ligament reconstruction devices became popular in the 1980s. Carbon fiber,178 polylactic acid (PLA)–coated carbon fiber,179 and polytetrafluoroethylene (PTFE)180 were all introduced during this period. The most popular device, the Kennedy ligament augmentation device (LAD)181 introduced in 1980, was a flat 6-mm ­diamond-braided polypropylene device. Justification for its popularity centered on its ability to provide protection and load-sharing to the biologic reconstructive tissue during the time of its transient phase of weakness and degeneration in the early postoperative period.181 Other touted advantages for augmentation and stand-alone LADs included decreased donor site morbidity, faster rehabilitation, and stronger structural grafts.182 Early results of LADs used in primary repair and augmentation cases were encouraging, particularly in purely soft tissue ACL graft reconstructions.183,184 However, further studies showed complication rates ranging up to 63%, including persistent effusions and reactive synovitis,185 delayed maturation of the autogenous graft,186 and infection.182,185,187 Because of these complications and the lack of any definitive data showing significant improvements with the use of LADs,172,186,188,189 their overall use has greatly decreased. In addition, advances in autograft and allograft tissues and improvements in graft placement and fixation have rendered the previous theoretical advantages of LADs obsolete. Advances in ACL reconstructive surgery have continued to occur with improvements in arthroscopic technology, equipment, and surgeon skill. A gradual transition has occurred from open reconstructive procedures, to an arthroscopic two-incision technique, to an arthroscopic one-incision technique. Multiple advances have led to an increase in our understanding and success in ACL reconstruction surgery. These include a better understanding of the appropriate timing of surgery as well as graft selection and subsequent harvesting. Additional biomechanical and animal research has also provided more detailed explanations of the anatomic structure of the ACL and which has led to improvements in ACL tunnel placement, tensioning, and fixation. Each of these areas is discussed in detail in the sections to follow.

1656 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Timing of Surgery There has been ample debate surrounding the ideal timing of ACL reconstruction surgery. Beginning in the early 1980s, ACL reconstructions were often done acutely within the first week after injury. As results of postoperative complications including arthrofibrosis began to appear in the literature, a common belief was that a delay in surgery would help minimize this complication. Arthrofibrosis is the most common postoperative complication after ACL reconstruction,190 and a loss of motion (particularly terminal extension) can be more debilitating than instability. A prospective cohort study by Kocher and colleagues191 on the determinants of patient satisfaction after ACL ­reconstruction showed that stiffness is one of the most common subjective complaints given by patients with poor postoperative outcomes. Use of the available literature to help resolve this issue, however, is difficult because of varying definitions surrounding acute versus chronic reconstruction and loss of motion parameters. Additionally, most studies are retrospective and suffer from bias and confounding variables that make accurate conclusions difficult to determine. Thus, studies have found increased rates of arthrofibrosis from early ACL reconstruction,192-195 whereas others have found early reconstruction to be safe.196,197 More recently, the determination of whether it is safe to proceed with ACL reconstruction has shifted from a measurement of the absolute time interval after injury to a measurement of the amount of inflammation and swelling present in the knee. Articles by both Shelbourne and colleagues198 and Mayr and associates190 have supported this view, as patients with excellent preoperative range of motion, minimal swelling, good leg control, and an appropriate mental state have had good postoperative outcomes regardless of the time period after injury. Our own experience supports this as we routinely use range-of-motion, swelling, knee irritability and pain, and quadriceps control as our determining factors for the readiness of the knee for surgery. To facilitate this process, we occasionally aspirate larger knee effusions, use compression knee braces, and start physical therapy and electrical stimulation of the

quadriceps muscle preoperatively. Aggressive ­postoperative rehabilitation has also been shown to decrease the rates of arthrofibrosis after surgery.193,194

Graft Selection The optimal graft material for ACL reconstruction remains an area of active debate. Although advances in ACL reconstruction have improved the success rate up to 90% in some cases with regard to stability and patient satisfaction,199 the ideal graft material to obtain these results has varied from surgeon to surgeon. The ideal graft should have structural properties similar to the native ACL that are present at implantation and persist throughout the “ligamentization” process of graft incorporation. Additionally, the graft should allow secure fixation, good biologic incorporation, and minimal donor site morbidity.200 Autograft ACL graft options include bone–patellar tendon–bone (BPTB), quadriceps tendon, and quadrupled semitendinosus and gracilis hamstring (HS) tendon. Allograft options include quadriceps, Achilles, tibialis anterior or posterior, BPTB, and HS. Because of its relative ease of harvest, its comparable structural properties to that of the native ACL,201,202 rigid fixation,202 bone-to-bone healing,37,201,202 and favorable track record,203-205 the autograft BPTB has historically been the graft of choice and is considered the gold standard against which other grafts are compared. Multiple factors are involved in a surgeon’s ACL reconstruction graft choice and include the biomechanical properties, biologic incorporation, associated donor site complications, graft tensioning issues, graft fixation options, and clinical outcome.

Biomechanical Properties The biomechanical properties of the common ACL grafts, including autograft BPTB, are compared in Table 23D1-6.19,20,202,206,207 A few important points deserve mentioning when evaluating biomechanical data on ACL grafts in the literature. Both native ACLs and potential graft choices should always be tested with both the graft and the subjected forces aligned in the anatomic position.

   TABLE 23D1-6  Graft Choices for Anterior Cruciate Ligament (ACL) Reconstruction Surgery Graft

Tensile Load (N)

Stiffness (N/mm)

Cross-sectional Area

Native ACL

2160

242

44

BPTB (10 mm)

2977

620

35

Quadruple hamstring Quad Tendon

4090

776

53

2352

463

62

Biologic Healing

Morbidity

Fixation

Outcomes/ Return to Play

Autograft Bone to bone (6 wk)

Extremity Interference weakness, screw kneeling pain Soft tissue healing Flexion weakness Variable (10-12 wk) Combination Similar to BPTB Interference bone to bone and screw soft tissue

4-6 mo Slight subjective laxity, 6 mo Limited data

Allograft BPTB (10 mm)

Similar to BPTB Similar to BPTB autograft autograft

BPTB, bone–patellar tendon–bone.

Similar to BPTB Bone to bone autograft (6 wk)

None

Interference screw

>6 mo, limited data

Knee 1657

Woo and colleagues208 showed that testing in the anatomic position greatly affected the ultimate load and stiffness data obtained; additionally, these same researchers found a substantial age-related decrease in the properties of the native ACL,208 which may also apply to cryopreserved allograft tissue. Because of such differences in the donor age, graft size, and specific biomechanical testing used, it is very difficult to compare studies. An additional factor in previous studies dealing with allografts was the use of sterilizing radiation (>3.0 mrad) or the use of ethylene oxide, both of which significantly weakened the grafts.209-211 Current cryopreservation techniques quoted in more recent studies do not have this weakening effect.210,212,213

Graft Healing Biologic graft healing encompasses both the graft attachment site healing as well as the healing process of ligamentization or graft revascularization and incorporation. Attachment site healing in grafts containing bone, particularly autografts, closely resembles fracture healing with graft bone–to–host bone healing occurring within 6 weeks.214 Purely soft tissue grafts typically take 8 to 12 weeks to heal into host bone.215 The process of graft revascularization and ­incorporation proceeds through well-defined phases starting with an in­ flammatory phase during which the graft undergoes degeneration.215-218 The donor fibroblasts undergo cell death, and the remaining biologic material serves as a scaffold for host cell fibroblast migration, which occurs with host ­revascularization during the second phase of ­incorporation. Graft strength and stiffness greatly decrease (up to 80%) during this phase, which lasts from about day 20 to 3 to 6 months after surgery.217 The final phase involves collagen maturation as the graft approaches but does not reach its original strength at implantation. Although both allografts and autografts proceed through the same phases of incorporation, allografts are thought to proceed at a slower rate,218,219 leading to a potentially increased rupture rate.220,221 There are no randomized, controlled trials comparing the outcomes of ACL reconstructions using allografts versus autografts. Allografts have been proposed to have less perioperative and postoperative morbidity than autografts.222 However, a study by Saddemi and associates did not find any difference in perioperative or postoperative morbidity between the two grafts.223,224 Although numerous articles have been published concerning the results of ACL reconstruction with allografts, most are retrospective case series with varying surgical techniques and graft types, which makes comparisons to autografts difficult.225-230 A small number of these studies support the notion of delayed allograft incorporation leading to laxity and failure,65,220,221 but most studies have shown no increased rate of graft rupture or increased laxity as compared with autograft.225-232

Donor Site Complications and Graft Harvest Although donor site complications are infrequently reported overall, most of the complications arise from autograft BPTB grafts. These include patellar fractures,233 patellar tendon

ruptures,234 localized numbness, and tendonitis.214 Generally, the dimensions of the bone blocks are 25 mm in length by 10 mm in width. Although larger bone blocks significantly increase the stress across the remaining patella,235 there are reportedly no differences between trapezoidal, square, or circular bone plugs.235,236 The effect of bone grafting of the defect is a difficult issue to resolve because of the statistical numbers needed237 but has not been shown to be of any statistical benefit.238,239 Patellar tendon rupture is rare,240 but care should be taken to identify the middle third of the tendon when harvesting. Closure of the patellar tendon after harvest may cause shortening of the tendon.241 We prefer to close the paratenon, which may help in healing of the tendon defect and avoid scarring to the overlying skin.239,242 Anterior knee pain after BPTB harvest has been reported to occur in up to 50% of cases,243 but a direct correlation to BPTB harvest is being refuted. The source of this pain may be multifactorial; the incidence of postoperative knee pain has been decreasing in more recent studies because of earlier rehabilitation, avoidance of immobilization, and emphasis on recovery of motion and strength.201,244,245 Autograft HS harvest may injure a superficial branch of the saphenous nerve.223 Problems associated with quadriceps tendon autograft harvest are more infrequent, resulting in part from fewer nervous structures in the incision area and the denser bone present in the proximal patella.214 Allograft use has increased during the past 20 years.246 Purported reasons for this increase include decreased morbidity and operative time, preservation of the extensor and flexor mechanisms, availability of specific graft sizes, improved cosmesis, and a reliable source of graft material in multiligament injuries or revision cases in which autograft choices are limited.221,226,227,247 In contrast, concerns have been raised about slow graft incorporation (see previous section), tunnel enlargement,230 and the risk for disease transmission.214 The risk for disease transmission remains a serious concern. There have been two cases of disease transmission of hepatitis C in 1991 and one case of human immunodeficiency virus (HIV) transmission in 1985, all from BPTB allografts.248,249 Many of these problems have been decreased or eliminated because of improved donor screening and testing procedures employing polymerase chain reaction (PCR) testing, which has significantly decreased the window of vulnerability between host infection and the detection of antibodies during screening procedures. The American Association of Tissue Banks frequently revises their recommended guidelines and currently recommends screens for HIV, hepatitis B and C, syphilis, and human T-cell lymphotropic virus, as well as blood and tissue cultures for bacterial infection. Despite this, two separate patients in 2000 receiving BPTB allografts from a common donor both developed Pseudomonas aeruginosa septic joint infections, with bacterial testing showing both strains to have an identical genotype.250 Additionally, a patient died in 2001 of Clostridium sordellii septic shock after receiving an infected osteochondral knee allograft.251

Graft Tension Appropriate graft tensioning during ACL reconstruction surgery remains a difficult quantifiable task. The concept that adequate tension is necessary to restore adequate

1658 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

anteroposterior stability at the time of ACL reconstruction, whereas too much tension may lead to graft stretching, fixation failure, and capture of the knee is easy to understand. However, there are multiple variables that affect graft tensioning, including the knee flexion angle and rotational position of the knee during tensioning and the specific graft type used. These variables make drawing conclusions across multiple studies difficult. Cadaveric and finite element modeling studies on the effects of initial graft tension on knee stability have been done using both HS252 and BPTB.253 These recommend tensioning the grafts between 40 and 60 N of force near full extension. Another conflicting variable is that most surgeons manually tension their grafts, and this applied force can vary significantly254 such that reproducibility becomes a significant issue. Prospective, randomized controlled trials evaluating the potential effects of different tensioning levels and their clinical and functional effects have been done. Three studies255-257 were completed on patients receiving BPTB grafts, and one study258 was completed on patients receiving four-strand HS grafts. In the studies by van Kampen and associates256 and Yoshiya and colleagues257 using BPTB grafts, they did not find any significant differences on clinical or functional outcomes between patients whose grafts were tensioned at 20 versus 40 N or 25 versus 50 N, respectively. Nicholas and coworkers255 compared groups tensioned at 45 versus 90 N and found that, in the 45 N group, 23% of the subjects had side-to side differences in knee laxity greater than 5 mm, compared with 0% in the 90 N group. No other clinical or functional differences were found. Yasuda and colleagues258 in a similar study using four-strand HS grafts, found a significant correlation between higher initial tension applied at the time of fixation and normal anteroposterior knee laxity measurements at the time of final follow-up. Again, no other differences in clinical or functional results were found. Questions have been raised regarding the fact that the difference in the two tension levels was not of sufficient magnitude to create differences in knee laxity measurements at baseline in those BPTB studies showing no difference in laxity.259 It is also possible that the loss in tension is the result of friction between the bone block and the tibial tunnel in BPTB grafts,259,260 from wrapping of the graft around an interference screw in soft tissue grafts260 and from postoperative cyclic loading.260,261 A lack of preloading the graft before implantation, which would help eliminate the natural viscoelasticity of soft tissue grafts, has also been suggested as a reason for loss of graft tension. A primate study by Graf and colleagues262 showed that preconditioning can reduce acute tension loss in BPTB grafts. However, a randomized, controlled trial by Ejerhed and colleagues263 compared preconditioning with no preconditioning in BPTB grafts and found no differences in knee laxity, clinical outcome, or activity level. Similarly, Nurmi and associates,264 in an experimental study, questioned the reasonableness of preconditioning soft tissue grafts in ACL reconstruction. Studies by Schatzmann and colleagues265 and Arnold and coworkers261 showed that a large number of cycles (>100 cycles) or high tension (75 to 80 N) were needed to reach a steady viscoelastic state, perhaps explaining why no effect of preconditioning was found in previous studies.

Recommendations regarding tension are still not universally agreed on. From the aforementioned studies, it appears that application of higher tensions (up to 90 N) with the knee near or at full extension may reduce anteroposterior knee laxity. However, the effect that such tensioning has on the tibiofemoral contact stresses is unknown and requires further biomechanical and clinical study. The effects of preconditioning on ultimate ligament tension also requires further study.

Initial Fixation and Strength Rigid fixation of the ACL graft at the time of implantation is considered one of the most important factors determining the long-term success of ACL reconstruction. The strength of the initial graft fixation is the weak leak during the initial 6- to 12 week period during which healing of the graft to host bone occurs.214 Biomechanical testing of ligament reconstruction fixation devices has been widely performed in the laboratory using various materials, including testing machines and different fixation devices. This allows better comparisons across different fixation devices and even across different mechanical studies. The most commonly reported biomechanical measure has been ultimate load to failure. Although it is an important measure of the ultimate load the graft can withstand during a catastrophic event such as a fall, it does not give information about how the graft construct (graft plus fixation) responds to the more common submaximal repetitive loading cycles that are experienced during aggressive postoperative rehabilitation programs.266 Thus, cyclical loading and stiffness profiles have become more common biomechanical measures reported to address this question. Most loads seen by an ACL graft during early rehabilitation are likely 200 N or less, with a maximum between 400 and 500 N.267,268 Although the literature surrounding graft fixation constructs is extensive, more attention has been focused on the more problematic fixation on the tibial side. Reasons for increased tibial fixation problems include the decreased bone quality of the tibial metaphysis, the increased force on the ACL graft in the parallel tibial tunnel compared with the nonparallel femoral tunnel, difficulty in securing and tensioning a four-strand hamstring graft, and the fact that tibial inference screw fixation is inserted counter to the direction of tension on the graft.269-271 It is also helpful to be aware of certain variables and differences that exist across various biomechanical publications when evaluating the literature. Particularly as it pertains to soft tissue fixation in the tibia, these cross-study variables include differences in specimen bone density, the geometry of implanted hardware or screws, hardware material differences, screw length and width differences, and the manner in which tunnels are drilled. Finally, owing to the scarcity of human bone and tissue, many studies are done using nonhuman tissues such as porcine or bovine specimens, which may have noticeable differences when compared with human bone and tissue. More studies are commonly reporting specimen bone mineral density (BMD) because increased BMD has been linked to an increase in fixation strength.269,272,273 BMD is higher in animal (particularly porcine) than in human

Knee 1659

   TABLE 23D1-7  Bone–Patellar Tendon-Bone Biomechanical Fixation Studies Study

Fixation Type

Specimen

Test

Scheffler et al, 2002279

Stainless steel interference screw (8 × 25 mm)

Human

Caborn et al, 1997286

Biointerference screw (7 × 25 mm)

Human

Titanium alloy interference screw (7 × 25 mm) 2 staples with bone block in tibial groove Stainless steel interference screw (9 mm)

Human

Femur-tibia complex; anterior tibial displacement Femur with tensile load 20 mm/min

Biodegradable interference screw (9 mm)

Human

Interference screw (9 × 25 mm, outside-in technique) Interference screw (7 × 25 mm, endoscopic technique) Suture tied over buttons Staple 6.5 mm screw

Human

Gerich et al, 1997290 Johnson & van Dyk, 1996287

Steiner et al, 1994289

Kurosaka et al, 1987291

Load to Failure (N)

Stiffness (N/mm)

384

66

Cyclic Testing

Mode of Failure

Stepwise increase 20 N; laxity increase 3.4 mm

Tibial graft pullout with (n = 3) and without (n = 5) bone block fracture Femoral fixation; ligament-bone separation Fracture tibial bone block; ligament-bone separation Slippage of bone block in tibial groove

552.5

N/A

558

N/A

Human

Tibia

588

86

N/A

Human

Femoral cortex removed; force in line with tunnel Femoral cortex removed; force in line with tunnel Femur-tibia complex

436

N/A

Tendon/cortical bone graft pull-out from femur

565

N/A

Failure associated with cortical and cancellous bone of graft Bone plug slippage past interference screw; usually tibial

423

46

N/A

Human

Femur-tibia complex

588

33

N/A

Human

Tibia/femur not specified

248.2

12.8

N/A

128.5 214.8

10.8 23.5

N/A N/A

Human Human

tibia,273-275 such that more optimistic results may be obtained when different fixation methods are tested on porcine specimens only. Screw geometry, specifically screw diameter and length, is also related to fixation strength. Although increasing screw size reportedly increases fixation strength,276 screw length has shown an even more important positive correlation to fixation strength.276,277 The screw material, metal or bioabsorbable, can also be responsible for cross-study differences, with biomechanical data on bioabsorbable screws at least equal to the data on metal screws.272,278,279 The ratio of the screw diameter to that of the tunnel diameter also varies across studies. Although intuitively a better match would seem to improve fixation, studies have shown no significant differences when the gap between the two is equal to or less than 2 mm.280,281 Finally, using tibial dilators to compact the soft cancellous bone in the proximal tibia during creation of the tibial tunnel has been variably done in clinical and biomechanical studies. Contrary to intuitive thinking, this has not been shown to have a significant effect on fixation strength.275,282-284 The most commonly used grafts are autograft BPTB and HS grafts and, to a lesser extent, their allograft counterparts. Their mechanical fixation to host bone can be categorized as either direct fixation (interference screws, staples, spiked washers), which compresses the graft against the host bone, or indirect fixation (cross-pin, screw

Avulsion fracture at tendon insertion

and post, EndoButton), which suspends the graft within a bony tunnel.266 For BPTB grafts, the most commonly performed and reported fixation is direct fixation using interference screws on both the tibial and femoral sides (Table 23D1-7). Because of its direct fixation of bone against bone,266 its aperture or juxta-articular fixation,285 and its favorable biomechanical profile,286-289 it has been the workhorse of BPTB graft fixation. Owing to concerns about screw divergence and graft injury during insertion, other options have been tested, including staple290,291 and press fit292,293 techniques. Less biomechanically secure constructs include suture post and button fixation (see Table 23D1-7).279,291,294 Soft tissue graft fixation on the femoral side has multiple options, with no clearly superior option (Table 23D1-8). Because of their ease of insertion with no additional incisions, both the EndoButton and interference screws are popular choices. Kousa and colleagues295 found that the bone mulch screw had the most favorable biomechanical profile, whereas To and associates296 favored cross-pin fixation. Scheffler and coworkers279 showed favorable results in their femur-graft-tibia model with bioscrew interference fixation. Soft tissue graft fixation on the more problematic tibial side also has multiple options (Table 23D1-9). Although multiple fixation devices had ultimate fixation strengths above 500 N, the WasherLoc and Intrafix devices had the

1660 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������    TABLE 23D1-8  Anterior Cruciate Ligament Soft Tissue Graft Biomechanical Fixation Studies: Femoral Side Study

Fixation Type

Specimen

Load to Failure (LFT) (N)

Test

Stiffness (N/mm)

Cyclic Testing (mm)

1086, 781

79

After 1500 Cycles (50-200 N), Displacement in mm 3.9 mm

1112, 925 589, 565 546, 534 868, 768 794, 842 430

115 66 68 77 96 23

2.2 4 3.9 3.7 3.2 N/A

312 1126 242

25 225

N/A N/A

LTF, LTF after Cyclic Loading Kousa et al, 2003295

To et al, 1999296

Caborn et al, 1998278

Endobutton CL

Porcine

Bone mulch screw Bioscrew RCI screw Rigid fix SmartScrew ACL Endobutton CL Human Mitek anchor Cross-pin RCI interference screw (7 mm)

Human

Femur; force along axis of drill hole

Femur; pull in line of tunnel

Femur

Bioabsorbable interference screw

Mode of Failure

Suture loop knot failure Migration of anchor Pin failure Graft pulled out around screw (13/16 specimens) Graft and screw pulled out from femoral tunnel (3/16)

341

   TABLE 23D1-9  Soft Tissue Anterior Cruciate Ligament Graft Biomechanical Fixation Studies: Tibial Side Study

Fixation Type

Specimen

Kousa et al, 2003272

Magen et al, 1999273

Load to Failure (LTF) (N)

Tibia: pull in line of tunnel

Coleridge & Amis, 2004297

Caborn et al, 2004274

Test

RCI screw Delta screw Intrafix bicortical screw WasherLoc Intrafix Tapered bioabsorbable screw (35 mm)

Bovine

Human

Metal interference screw (9 × 25 mm) WasherLoc Tandem washers

Cyclic Testing (mm)

Mode of Failure

N/A

N/A N/A N/A N/A N/A N/A N/A

491 641 543 770 946 796

N/A N/A N/A N/A N/A 49.2

1000 cycles 70-220 N slippage (mm) 1.3 1.15 0.69 1.17 0.88 N/A

647

64.5

N/A

N/A N/A

975,917

87

After 1500 cycles (50-200 N); displacement in mm 3.2

769,675 1321, 1309 612,567 471,423

69 223 91 61

4.2 1.5 4.1 4.7

N/A N/A N/A N/A

665,694 Yield load in place of LTF (N)

115

N/A N/A

350

340

3.8 Stepwise increase by 50 N; displacement at 250 and 500 N mm) 1.80, 3.67

905 768

506 318

0.55, 1.95 0.30, 0.86

N/A N/A

LTF, LTF after cyclic loading

WasherLoc Spiked washers Intrafix Bioscrew Titanium interference screw (8 × 25 mm) Bioscrew (8 × 25 mm)

Tibia: pull in line of tunnel

Stiffness (N/mm)

Tibia: pull in line of tunnel Porcine

Human

Tibia: pull in line of tunnel

N/A

N/A

Knee 1661

more favorable data when comparing stiffness and displacement during cyclic loading.272-274,297

CLINICAL OUTCOMES The choice of ACL graft and technique available to the orthopaedic surgeon consists of dozens of choices with hundreds of articles expressing opinions and outcome data. Most of these articles are retrospective case series of a single graft, with several potential sources of bias and no adequate comparison group. There are articles describing comparisons between the two most common grafts, autograft BPTB and HS, but each article is statistically limited in its ability to draw strong conclusions because of limited sample size.65,203-205,298-309 In an effort to increase statistical power and provide orthopaedic surgeons with statistically stronger data on which to make their decisions, meta-analyses have been done.199,200,310 A meta-analysis is a technique to statistically combine or integrate the results of several independent clinical trials to increase statistical power. In addition to meta-analyses, an emphasis has been placed on performing more randomized controlled trials (RCTs) of ACL reconstruction using BPTB versus HS grafts. Such RCTs help to limit any confounding variables or biases that limit the ability of non-RCTs to draw statistically powerful and meaningful conclusions regarding graft choice and operative technique. In a similar vein, meta-analyses199,200,310 and systematic review311 of the available RCTs have more recently been completed. A systematic review presents data on several studies in a tabular form, which allows the reader to draw comparisons and conclusions regarding the data present in multiple studies. A large meta-analysis of the available articles published on ACL reconstruction with BPTB versus HS grafts was done by Freedman and colleagues199 in 2003. Their meta-analysis inclusion criteria were broad, including retrospective and uncontrolled studies as well as studies involving only one type of reconstruction with varying rehabilitation protocols. It reported on 1348 patients culled from 21 and 13 studies, respectively, involving BPTB or HS ACL reconstructions. They found that BPTB ACL reconstruction was associated with a statistically significant decreased rate of failure and laxity and provided patients with a more stable knee. Hamstring ACL reconstruction was found to have a significantly decreased incidence of anterior knee pain and rate of arthrofibrosis requiring manipulation or lysis of adhesions. Yunes and associates310 performed a more restricted meta-analysis involving only prospective, semirandomized studies. It consisted of four studies comprising 424 patients. Their findings were somewhat similar to those of Freedman and colleagues199 in that BPTB reconstructions were found to give a statistically more stable knee with regard to KT-2000 and pivot shift objective testing. Additionally, they found that BPTB had an 18% increased chance of returning to preinjury levels than HS grafts. Remaining variables, including range of motion, failure and complication rates, and Lachman testing, showed no statistical ­difference between the two groups. No data were obtained on anterior knee pain or quadriceps weakness because of limitations in the available studies.

Multiple RCTs, comparing BPTB and HS ACL reconstructions, have been published since 2000. Spindler and colleagues311 and Goldblatt and coworkers200 performed a systematic review and meta-analysis, respectively, on these trials in an effort to improve our understanding. Spindler and colleagues311 included nine RCTs in their systematic review. Because this was not a meta-analysis, data were presented in tabular form only. They found slightly increased laxity in HS reconstructions in three of seven studies, but no differences in graft failure, functional scores, or activity levels between the two groups. They also found more kneeling pain in BPTB groups in all four of the studies that reported this outcome; however, only one of nine studies reported an increased incidence of anterior knee pain in the BPTB groups. Goldblatt and coworkers200 performed a comprehensive meta-analysis of randomized or controlled trials, which included 11 studies comprising 1039 patients. Inclusion criteria were identical rehabilitation protocols within each study, a minimal 2-year follow-up, and the presence of both subjective and objective outcomes data. Outcomes favoring BPTB (P <.05) over HS reconstructions included a normal Lachman test, normal pivot shift test, KT examination showing less than 3 mm, and less flexion loss. Outcomes favoring HS reconstructions (P <.05) included decreased incidence of patellofemoral crepitance, less extension loss, and less kneeling pain. Objective outcome rating scores, anterior knee pain, and graft failure were not statistically different between the two groups (Fig. 23D1-6). With the increased use of ­interference fixation, they also found no difference in failure rates between grafts fixed with interference fixation and those fixed without it. They were unable, owing to the data in the individual studies, to compare BPTB and HS grafts fixed with interference fixation. These studies lay an excellent foundation for future work in evaluating ACL reconstructions. They allow the practicing orthopaedic surgeon, who typically does not possess the time or expertise to systematically review the literature, to individually examine the available data and judge how best to incorporate this into his or her practice. These studies also help assign priority to certain outcome variables that may be important to individual patients during preoperative discussions with their surgeon regarding graft selection. They also emphasize the need for standardized, validated patient outcome questionnaires that will allow comparisons across multiple studies because many of these analyses were limited by missing data. A final important variable not addressed in these analyses is that of surgical technique, namely, tunnel position. Tunnel position continues to be an area of confusion, particularly because many of the descriptions of tunnel position are given with the knee extended, whereas most arthroscopic knee surgery is performed with the knee flexed. Additionally, prospective studies on proper tunnel position are difficult to perform because placement of a tunnel anywhere other than the most optimal position (as assessed by the surgeon) would be unethical. Prospective studies by both Good and associates312 and Khalfayan and colleagues313 have shown a direct relationship between tunnel position and clinical outcomes, with worse ­ outcomes

1662 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Summary of Results Outcomes

Studies N

Lachman >=2 Giving-way KT > 3 mm Flexion Loss > 5 deg Rupture Decreased Activity Pivot-Shift > 0 Lachman > 0 Complications Meniscus Surgery Pivot-shift >= 2 KT > 5 mm IKDC B+C+D Swelling IKDC C+D IKDC D Anterior knee pain Extension loss > 0 deg Extension loss > 5 deg

6 3 3 4 6 4 8 7 5 4 7 3 4 3 5 4 5 3 5

Favors HT

Favors BPTB

P value .22 .37 .01 .04 .46 .12 .09 .06 .79 .75 .83 .84 .71 .97 .68 .68 .12 .13 .06

624 286 182 446 597 340 761 713 530 451 672 182 450 286 516 450 367 186 406 0.2

0.5

1 2 Relative Risk

5

Figure 23D1-6   Summary of risk ratios for various outcomes measures: hamstring versus patellar tendon autograft in anterior cruciate ligament reconstruction. BPTB, Bone patellar tendon bone; HT, Hamstring tendon; IKDC, International Knee Documentation Committee. Redrawn from Goldblatt JP, Fitzsimmons SE, Balk E et al: Reconstruction of the anterior cruciate ligament: Meta-analysis of patellar tendon versus hamstring tendon autograft. Arthroscopy 21[7]:791-803, 2005.

in patients with tunnels greater than 2 mm from the anatomic insertion site as defined by radiographs. ­Significant room for improvement in tunnel position likely exists. Kohn and coworkers,314 in an analysis of cadaveric knees that had undergone BPTB reconstructions, showed only 50% excellent and 75% acceptable tunnel position on the femoral and tibial sides. Malpositioned tunnel placement continues to be one of the most common reasons for revision ACL reconstruction.315,316 Thus, although single-bundle reconstruction techniques are capable of providing good or excellent outcomes in up to 90% of our patients,205,302 significant room for improvement exists. Despite recent data that moving the femoral tunnel down the face of the femur to a more horizontal position better restores both anteroposterior and rotational stability to the knee,317 recent authors have

noted persistent instability with functional testing and degenerative radiographic changes after single-­bundle reconstruction.31,318-321 Tashman and associates have used a stereoradiographic system to evaluate the three­dimensional kinematics of reconstructed and intact knee. Their findings suggested that single-bundle ACL reconstruction failed to restore normal rotational knee kinematics during dynamic ­loading.31 Newer techniques involving re-creation of both functional bundles of the ACL, the AM and PL, have recently been published; clinical as well as biomechanical studies involving double-bundle ACL reconstruction have shown promise in restoring more normal stability to the knee.28,29,108,322-326 This anatomic double-bundle reconstruction may provide us with a technique to further improve the success rates of ACL reconstruction.

Authors’ Preferred Method Positioning and Arthroscopic Setup

The correct limb to be operated is identified and marked by the surgeon. The patient is placed in a supine position on the operative table. The contralateral leg is placed in an abducted position with the knee and hip slightly flexed in a well-������� padded leg holder, and the leg is secured in position with an elastic bandage (Fig. 23D1-7). The operative knee is examined under anesthesia with Lachman, anterior and posterior drawer, varus-valgus, and pivot shift testing. The examination allows assessment of the extent of the ACL injury. If a partial tear of the

ACL is considered, the arthroscopic examination will confirm the diagnosis. After examination, a pneumatic tourniquet is applied around the upper thigh of the operative leg. The limb is exsanguinated by elevation for 3 minutes, and the tourniquet is insufflated to 300 to 400 mmHg depending of the size of the patient. The foot of the table is flexed, and the operative leg is placed in the leg holder. The use of the leg holder allows the knee to be positioned at 90 degrees of flexion with a range of motion from full extension to 120 degrees of flexion (see Fig. 23D1-7). The operative leg is aseptically prepared and draped.

Knee 1663

Authors’ Preferred Method—cont’d needle is inserted medially and distally to the AM portal just above the meniscus. The spinal needle needs to reach the medial wall of the lateral femoral condyle, at the origin of the footprint of the PL bundle. Once the needle is correctly placed, the AAM portal is made with a No. 11 blade. The AAM portal allows better visualization of the tibial and femoral ACL footprints and will also be used as a working portal during the surgical procedure. A diagnostic arthroscopy is completed, including inspection of the suprapatellar pouch, patellofemoral joint, medial and lateral gutters, and medial and lateral compartments. Special attention is given to the menisci and chondral surfaces. Usually, full-thickness longitudinal and bucket handle meniscal tears in the red-red and red-white zones are repaired using an inside-out technique. Partial-thickness and stable tears less than 10 mm in length are not surgically treated. Outerbridge grade II and III chondral lesions may be treated with the use of a cartilage thermal device. Unstable cartilage lesions and Outerbridge grade IV lesions may be treated by débridement and microfracture. Figure 23D1-7   A knee holder is used to keep the operative knee stable during the surgery. It also allows good range of motion of the knee during the surgery.

Diagnostic Arthroscopy

Arthroscopy is performed to diagnose and treat any associated injuries. The correct portal positions are critical to obtain optimal intra-articular visualization and to manage the arthroscopic instrumentation. We use the standard anterolateral (AL) and AM portals, as well as an accessory anteromedial portal (AAM) (Fig. 23D1-8). The AL portal is close to the inferior border of the patella. The AM portal is placed medial to the patellar tendon, slightly lower than the AL portal. To establish the AAM portal, the arthroscope is placed into the standard AM portal, and an 18-gauge spinal

Figure 23D1-8   Surgical incisions for anterior cruciate ligament surgery: anterolateral portal (LP), anteromedial portal (MP), accessory medial portal (AMP), and tibial incision.

Insertion Site Marking

The ACL rupture pattern is evaluated during the diagnostic arthroscopy, as are the native footprints of the AM and PL bundles on the tibial plateau (Fig. 23D1-9) and lateral wall of the intercondylar notch (Figs. 23D1-10 and 23D1-11). When identified, the tibial and femoral footprints of each bundle are marked by a thermal device (ArthroCare Corporation, Sunnyvale, Calif). The PL tibial footprint is located in the center of a triangle formed by the posterior root of the lateral meniscus, the PCL, and the AM bundle of the ACL (see Fig. 23D1-9). The AM tibial footprint is located anteromedial to the PL tibial footprint. The tibial ACL stumps are left intact to preserve their proprioceptive and vascular

Figure 23D1-9   The oval anteromedial (AM) and circular posterolateral (PL) tibial insertions are marked with a thermal device. Tunnel guidewires have been placed in the center of the AM and PL bundle tibial insertion sites. The center of the PL bundle is 3 to 5 mm from the posterior aspect of the triangle formed by the posterior cruciate ligament (PCL), lateral meniscus posterior root, and AM bundle insertion site. LFC, lateral femoral condyle. Continued

1664 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d rupture pattern and bony anatomy are easier to visualize in acute than in chronic ACL injuries. Thus, a thorough understanding of the double-bundle ACL anatomy is essential to identify and mark the correct insertional footprints. Graft Preparation

Figure 23D1-10   With the knee in 90 degrees of flexion, the posterolateral (PL) bundle insertion is marked on the femoral side with a thermal device. Care is taken to preserve the superior border of the ACL fibers, which serve as the upper limit for tunnel placement. AM, anteromedial bundle.

contributions. On the femoral side, the ACL bundle origins are located under a bony landmark called resident’s ridge. It is important to note that in most cases, there is a clear change in the topography between the AM and PL femoral insertions, forming a bony landmark between the AM and PL bundle femoral origins called the cruciate ridge. The cruciate ridge and the change in the ACL femoral topography make location of the AM and PL bundles on the femur easier (see Fig. 23D1-5) We do not perform a notchplasty because it destroys the bony landmarks and the topography of the ACL femoral anatomy. It is important to note that the

Figure 23D1-11   Similar to the marking of the posterolateral (PL) bundle insertion, the anteromedial (AM) insertion on the femoral side is marked with a thermal device. The anterior cruciate ligament remnant is preserved to serve as the reference for tunnel placement.

While the diagnostic arthroscopy and insertion site marking are being performed, the grafts are being prepared on the back table by an assistant. Typically, we use two separate tibialis anterior or tibialis posterior tendon allografts. However, HS autograft may also be used. The allografts are removed from the −80° C freezer and thawed in a warm saline solution. Usually, the allograft is 24 to 30 cm in length, and we fold each tendon graft to obtain 12 to 15 cm doublestranded grafts (Fig. 23D1-12). First, the tendon allografts are trimmed, and the diameters of the double-stranded grafts are adjusted. Typically, the tendon grafts are trimmed such that the diameter of the double-stranded AM graft is 8 mm and that of the double-stranded PL graft is 7 mm. However, the diameters of the graft may be adjusted depending on the size of the patient. The ends of the grafts are sutured using a baseball stitch with No. 2 Ti-Cron sutures. An EndoButton CL (Smith & Nephew, Andover, Mass) is used to loop each graft and obtain a double-stranded graft. The length of the EndoButton loop is chosen according to the measured length of the femoral tunnels. It is important to know the available EndoButton loop lengths (15 to 40 mm in 5 mm increments). The exact length of the EndoButton loop is chosen after measuring the femoral tunnel length. Tunnel Placement

The PL femoral tunnel is drilled first. A Steadman awl is used to create a small hole in the center of the PL bundle femoral insertion to facilitate the placement of a 3.2-mm guidewire that is inserted through the AAM portal (Fig. 23D1-13). The tip of the guidewire is placed on the small hole and malleted into position. Once the tip of the guidewire is placed in the

Figure 23D1-12   Grafts were prepared on the back table by an assistant.

Knee 1665

Authors’ Preferred Method—cont’d

Figure 23D1-13   A Steadman awl is used to mark the desired femoral anteromedial (AM) and posterolateral (PL) positions for guide pins. This also makes it easier for the guide pins to be inserted into the chosen spot.

correct position, the femoral PL tunnel is drilled with an acorn drill that is inserted over the guidewire. It is important to note that during the entire procedure to create a PL femoral tunnel, the knee is positioned at 90 degrees of flexion, which brings the PL bundle footprint anteriorly. The PL femoral tunnel is drilled to a depth of 25 mm (Fig. 23D1-14). The far cortex is breached with a 4.5-mm EndoButton drill, and the depth gauge is used to measure the distance to the far cortex. When the total length to the far cortex is longer than 38 mm, the PL femoral tunnel is drilled to a depth of 30 mm. To obtain the correct length of the EndoButton loop, 6 mm is added to the difference between the total length (TL) and the length of the graft femoral tunnel (GL). Thus, the formula is length of the EndoButton loop = (TL − GL) + 6. For example, when a 38-mm tunnel is drilled to the far cortex and the graft femoral tunnel is drilled to 30 mm, the

Figure 23D1-14   Posterolateral (PL) femoral tunnel is drilled through the accessory anteromedial portal.

f­ ollowing calculation is performed: EndoButton loop length = (38 − 30) + 6. In this example, the EndoButton length is 14 mm. Because 14 mm is not available, the 15-mm EndoButton loop is chosen. Always the approximation is done to the higher number to ensure that the EndoButton will be able to flip and achieve appropriate cortical fixation. To create the two tibial tunnels, a 4-cm skin incision is made over the AM surface of the tibia at the level of the tibial tubercle. The PL tibial tunnel is the first to be drilled. The elbow ACL tibial drill guide is set at 45 degrees, and the tip of the drill guide is placed intra-articularly on the tibial footprint of the PL bundle previously marked. On the tibial cortex, an osteoperiosteal flap is detached, and the tibial drill starts just anterior to the superficial MCL fibers. Once the tibial drill guide is set, a 3.2-mm guidewire is passed into the stump of the PL tibial footprint (Fig. 23D1-15). The AM

Figure 23D1-15   Different views of the tibial guide pins from lateral portal (left) and medial (right) portal. Both portals provide good views on the tibial side. However, the medial portal is superior in observing the lateral wall of the intercondylar notch. Continued

1666 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

Figure 23D1-16   Overhead and lateral pictures demonstrate the orientation of anteromedial (AM) and posterolateral (PL) tibial tunnel dilators. The tibial tunnels were both drilled with the drill guide set at 45 degrees.

tibial ­tunnel is drilled with the elbow ACL tibial drill guide set at 45 degrees, and the tip of the drill guide is placed on the tibial footprint of the AM tunnel previously marked. On the tibial cortex, the starting point for the AM bundle is placed anterior, lateral, and proximal from the PL starting point, under the osteoperiosteal flap detached previously (Fig. 23D1-16). After the elbow ACL tibial drill guide is placed in the desired position, a 3.2-mm guidewire is passed into the stump of the AM tibial footprint (see Fig. 23D1-15). When both tibial guidewires are in place, the two tibial tunnels are drilled using a cannulated drill. Usually, the diameter of PL and AM tunnels are 7 and 8 mm, respectively. The femoral AM tunnel is the last tunnel to be drilled. Dur­ing the AM tunnel drilling, it is important to flex and hold the knee to about 120 degrees. A transtibial technique as used for a single-bundle ACL reconstruction is our first choice to create the AM femoral tunnel. However, in some cases, the transtibial technique cannot reach the center of the AM bundle previously marked. Then, we can try to reach the center of the AM femoral footprint through the PL tibial tunnel. Using this approach, the AM graft needs to be trimmed to a 7-mm diameter. In cases in which the center of the AM femoral footprint is not reached through either the PL or AM tibial tunnels, we use the AAM portal to reach the center of AM bundle (Fig. 23D1-17). Using this approach, the total length of the femoral tunnel is shorter, and caution is needed to avoid compromising the cortical integrity of the lateral femoral condyle with the 8-mm drill. When the tip of the guidewire is placed in a correct position, the guidewire is malleted into position, and an acorn drill is inserted over the guidewire. The AM femoral tunnel is drilled to a depth of 35 to 40 mm when it is done through a transtibial approach. In cases in which the AM femoral tunnel is performed through the AAM portal, it is drilled to a depth of 25 to 30 mm (Fig. 23D1-18). The far cortex of the AM femoral tunnel is breached with a 4.5-mm EndoButton

drill, and the depth gauge is used to measure the distance to the far cortex. The length of the EndoButton loop is chosen in a similar fashion to the PL tunnel drilling. Graft Placement and Fixation The first graft to be passed is the PL graft (Fig. 23D1-19). A beath pin with a long-looped suture attached to the eyelet is passed through the AAM portal and out through the PL femoral tunnel. The looped suture is visualized within

Figure 23D1-17   Creation of the anteromedial (AM) femoral tunnel through either the accessory medial portal (AMP) or the tibial posterolateral (PL) tunnel results in anatomic placement of the AM bundle. Attempts at creating the AM femoral tunnel through the AM tibial tunnel may sometimes result in placement of the AM graft superior to the upper limit of the native AM bundle (gray area). Thus, the femoral tunnel should be drilled independent of the tibial tunnels.

Knee 1667

Authors’ Preferred Method—cont’d

Figure 23D1-18   The anteromedial (AM) and posterolateral (PL) femoral tunnels have been created in the anatomic position. LFC, lateral femoral condyle.

Figure 23D1-20   Final result of anatomic double-bundle anterior cruciate ligament (ACL) reconstruction. Note that the upper border of the ACL remnant can still be visualized after passing the grafts. AM, anteromedial bundle; PL, posterolateral bundle.

the joint and retrieved with an arthroscopic suture grasper through the PL tibial tunnel. The Ti-Cron sutures of the graft are passed through the looped suture, and the graft is passed in a retrograde fashion up the tibia and femur. After the graft is successfully passed, the EndoButton is flipped for femoral fixation. The AM graft is passed using the trans­ tibial technique when the tunnel was created transtibially or through the AAM portal when the tunnel was created through the AAM portal, using the same technique as in the PL bundle graft passage. The EndoButton is flipped in a similar fashion to establish AM bundle femoral fixation. When the AM femoral tunnel is drilled through the PL

tibial tunnel, it is ­ important to pass the suture of the AM graft through the PL tibial tunnel and retrieve it through the AM tibial tunnel before performing the PL graft passage (Fig. 23D1-20). After passing the grafts, the knee is moved in full range of motion, and the grafts are observed under scope to exclude the possibility of ACL or PCL impingement (Fig. 23D1-21). Preconditioning of the grafts is performed by flexing and extending the knee through a range of motion from 0 to 120 degrees about 20 to 30 times. On the tibial side, we prefer to use a Calaxo bioabsorbable screw (Smith & Nephew, Andover, Mass). A 7 × 25-mm screw is used in the PL tibial tunnel, and an 8 × 25-mm screw is used in the AM

Figure 23D1-19   The posterolateral (PL) graft is passed first, followed by the anteromedial (AM) graft. When the knee is in flexion, the AM and the PL bundles show a crossing pattern.

Figure 23D1-21   The posterior cruciate ligament (PCL) triangle that is formed by the anterior cruciate ligament (ACL), the PCL, and the roof of the intercondylar notch is shown. PCL impingement is rarely seen even without notchplasty. Continued

1668 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method tibial tunnel. The graft fixation is performed with the knee in 0 degrees of flexion for the PL bundle graft and at 60 degrees of flexion for the AM graft (Fig. 23D1-22). ­During fixation, maximal tension is applied to the sutures, and the screws are properly placed. The remaining graft left protruding from the tibial tunnel is removed. Arthroscopic inspection is performed to observe the intra-articular graft tension. The use of the AAM portal to drill AM femoral tunnel may lead to a short femoral tunnel, and the far cortex may be compromised by the use of an 8-mm drill, making EndoButton fixation impossible. In this case, a lateral incision is made, and postfixation of the EndoButton loop is performed using a 4.5-mm diameter screw.

Closure

The osteoperiosteal flaps overlying the tibial drill holes are closed with a 0-0 absorbable suture. The subcutaneous layer is closed with an interrupted 2-0 absorbable suture, and the skin with a running subcuticular 3-0 absorbable suture. We do not inject Bupivacaine intra-articularly, to avoid possible cell toxicity. The peripheral nerve block typically provides anesthesia for about 24 hours. We do not use drains. The wound is covered by Steri-Strips, dry sterile gauze, Kerlix roll, a cryotherapy pad, and finally a Cryocuff. We put the knee in a hinged knee brace, locked in full extension.

Figure 23D1-22   Anteromedial bundle graft passage and fixation at 60 degrees of knee flexion. The posterolateral bundle has been passed first and fixed in full extension.

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Rehabilitation The optimal rehabilitation program after ACL reconstruction has changed considerably over the past 20 years because of advancements in surgical techniques, graft selection, and fixation methods, as well as an improved understanding of the biology and biomechanics of the knee. Accelerated rehabilitation programs, which permit early range of motion, immediate weight-bearing, and early return to sport, have become the accepted standard and have helped the patient to return to a normal and complete level of function in the shortest time possible without compromising the integrity of the surgically reconstructed knee. However, prospective randomized controlled trials are still needed to prove this trend, as well as further research into the biology of graft healing, the appropriate limits of graft strain, and the effects of functional activities on graft stability.

Open and Closed Kinetic Chain Exercise One of the major goals of postoperative rehabilitation is to restore the range of motion of the knee joint, without compromising the integrity and function of the ACL graft. Generally, in the early rehabilitation program, closed kinematic chain (CKC) exercises are safer than the openor kinematic chain (OKC) exercises because research has suggested that CKC exercises apply less anteriorly directed forces on the tibia,327-329 increase tibiofemoral compressive forces,330,331 increase co-contraction of the hamstrings,328,332 mimic functional activities more closely than OKC exercises,329,333 and reduce the incidence of patellofemoral complications, especially at low knee flexion angles.329,333 A variety of different definitions of CKC and OKC exercises are used in the literature. Briefly, CKC exercises are defined as those in which the foot is in contact with a solid surface (Fig. 23D1-23). Ground reaction force is ­transmitted to all of the joints in the lower extremity, and muscles spanning all of the joints of the lower extremity are used. Examples of CKC exercises are the squat and leg press. In contrast, OKC exercises are defined as those in which the foot is not in contact with a solid surface

Knee 1669

Figure 23D1-23   Leg press machine. Patient is doing a closed chain exercise.

(Fig. 23D1-24). Thus, one segment of the limb is stabilized while the other segment moves freely, and only the muscles spanning the knee are required to perform the exercise. An example of OKC exercise is the leg extension machine. However, many activities cannot be clearly classified as CKC or OKC. Daily activities like walking, stair climbing, and jumping are combinations of OKC and CKC movements. Understanding the mechanics of different activities and their impact on the healing ACL graft is the key to the success of ACL surgery rehabilitation. Several studies have reported that OKC exercises generate significantly greater anterior tibial translations in the ACL-deficient knee than CKC exercises. Jonsson and colleagues reported a 1.9-mm average increase of anteriorly directed tibial displacement in the ACL-deficient knee during the active knee extension exercise (OKC) when the knee was near extension (15 to 10 degrees), whereas no increase was recorded during the step-up exercise (CKC).334 Using analytic models, Escamilla and associates determined that

Figure 23D1-24   Straight leg raise. Patient is doing an open chain exercise.

the mean peak force on the ACL was 158 N during the OKC exercise when the knee was at 15 degrees of flexion, whereas it was not loaded during the CKC exercises.329 However, incorporating accurate three-dimensional geometry of the knee and the material properties of different tissues into the analytic model is needed in the future to accurately mimic the loading environment on the ACL graft. Attention has also been paid to the direct measurements of ACL strains when OKC and CKC exercises are performed (Table 23D1-10) because excessive strains could permanently stretch out or fail the tissue. Using the differential variable reluctance transducer, Beynnon and associates52 showed that consistent distinction between CKC and OKC activities was not found, although both exercises resulted in increased strain values in the ACL, with squatting showing slightly higher strain values than knee extension. This finding suggests that certain CKC activities, such as squatting, may not be as safe as we believed, particularly at low flexion angles. However, CKC exercises did show an advantage in that increasing resistance did not lead to an increased strain in the ACL, whereas it did ­during OKC exercises.

   TABLE 23D1-10  Rank Comparison of Peak Anterior Cruciate Ligament Strain Measured during Commonly Prescribed Rehabilitation Exercises* Rehabilitation Exercise Isometric quadriceps contraction at 15 degrees (30 Nm of extension torque) (OKC) Squatting with sport cord (CKC) Active flexion-extension of the knee with 45-N weight boot (OKC) Lachman test (150 N of anterior shear load; 30-degree flexion) Squatting (CKC) Active flexion-extension (no weight boot) of the knee (OKC) Simultaneous quadriceps and hamstring contraction at 15 degrees (OKC) Isometric quadriceps contraction at 30 degrees (30-Nm extension torque) (OKC) Stair climbing (CKC) Leg press at 20-degree flexion (40% body weight) (CKC) Lunge (CKC) Stationary bicycling (CKC) Isometric hamstring contraction at 15 degrees (to 10 Nm flexion torque) (OKC) Simultaneous quadriceps and hamstring contraction at 30 degrees (OKC) Isometric quadriceps contraction at 60 degrees (30-Nm extension torque) (OKC) Isometric quadriceps contraction at 90 degrees (30-Nm extension torque) (OKC) Simultaneous quadriceps and hamstring contraction at 60 degrees, 90 degrees (OKC) Isometric hamstring contraction at 30, 60, and 90 degrees (10 Nm flexion torque) (OKC)

Peak Strain (%) 4.4 (0.6) 4.0 (0.6) 3.8 (0.5) 3.7 (0.8) 3.6 (0.5) 2.8 (0.8) 2.8 (0.9) 2.7 (0.5) 2.7 (1.2) 2.1 (0.5) 1.9 (0.5) 1.7 (0.7) 0.6 (0.9) 0.4 (0.5) 0.0 0.0 0.0 0.0

*Mean ± standard error. CKC, closed kinetic chain; OKC, open kinetic chain. From Fleming BC, Oksendahl H, Beynnon BD: Open- or closed-kinetic chain exercises after anterior cruciate ligament reconstruction? Exerc Sport Sci Rev 33(3):134-140, 2005.

1670 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The effects of OKC and CKC exercises on functional outcome have been evaluated in three independent prospective randomized clinical trials.5,10,14 Bynum and coworkers found that the mean side-to-side difference in knee laxity of the OKC group (3.3 mm) was significantly greater than that of the CKC group (1.6 mm) 19 months after surgery. The CKC group also returned to sport earlier than the OKC group. However, the two rehabilitation protocols have some differences in the levels of resistance and the progression of exercise, which may account for the better outcomes of the CKC group. Mikkelsen and associates335 randomized the patients into two rehabilitation programs. One group used CKC exercises for 24 weeks, and the other used the same CKC rehabilitation program with the addition of OKC exercises (isokinetic quad strengthening) from weeks 6 to 24. Interesting, significantly higher quadriceps torque was observed in the later group. The authors suggested that certain OKC exercises may be beneficial. However, the improvement may simply be due to the addition of exercises, independent of the type of exercise added. Nonetheless, the addition of the OKC exercises in this time frame did not produce a negative outcome. In another study, Morrissey and Hooper found that there are no clinically significant differences in the functional improvement resulting from the choice of OKC and CKC exercises in the early period of rehabilitation.333 However, this report may be limited because of the short rehabilitation program (2 to 6 weeks after surgery). In summary, the CKC exercises appear to protect the graft and help restore knee functions by different mechanisms. CKC exercises generate low anterior shear force and tibial displacement through most of the flexion range, increase tibiofemoral compressive forces, and reduce the incidence of patellofemoral complications. Additionally, in vivo strain data supported the notion that the ACL is a primary restraint to anteroposterior translation of the knee, and that knee hamstring co-contractions reduce ACL strains relative to isolated contractions of the quadriceps and gastrocnemius muscles. However, in the prospective randomized clinical trials that were designed to evaluate the effectiveness of OKC and CKC exercises in helping graft healing and restoring knee functions, contradictory conclusions were drawn, suggesting that the difference of OKC and CKC exercises may not be clinically significant and that additional prospective randomized clinical trials must be performed to determine the optimal time and combination to introduce these exercises. The key of ACL surgery rehabilitation is to use activities that minimize graft strain and put the ACL at the lowest risk for development of laxity.

REHABILITATION CONSIDERATIONS Pain and Effusion Pain and swelling are common after any surgical procedure. They cause reflex inhibition of muscle activity and therefore should be controlled appropriately to facilitate early range of motion and strengthening activities. The RICE principle, including rest, ice, compression, and ­ elevation,

remains the standard of care in reducing pain and swelling. Narcotic and anti-inflammatory pain medications are commonly prescribed in the acute postoperative setting. Muscle activities like quad sets and ankle pumps can help to reduce swelling by improving venous return. Electrical muscle stimulation of the quadriceps is also used to promote muscle activity before the return of voluntary muscle control.

Cryotherapy Cryotherapy is a common treatment modality after orthopedic surgery procedures. The forms of cryotherapy include ice packs, ice baths, and continuous flow cooling devices. The beneficial effects of cryotherapy are obtained through lowering joint temperature.336,337 Low temperature can help to lower the metabolism, reduce inflammation, and induce vasoconstriction, which, in turn, contributes to less tissue swelling, less pain, and less hemarthrosis.336,338 A recent meta-analysis including seven randomized clinical trials showed that postoperative drainage (P = .23) and range of motion (P = .25) were not significantly different between cryotherapy and control groups. However, cryotherapy was associated with significantly lower postoperative pain (P = .02).339 It was considered an inexpensive, easy to use way to reduce pain, inflammation, and effusion after knee surgery. Complications such as superficial frostbite and neurapraxia are rarely seen and can be prevented by avoiding prolonged placement of the cold source directly on the skin.

Motion Loss of motion is one of the most common complications after ACL reconstruction. The most common causes of this complication include arthrofibrosis, cyclops syndrome, and inappropriate graft placement or tensioning.340 It may lead to symptoms like anterior knee pain, abnormal gait, muscle atrophy, and early degenerative changess of the joint.340-342 Usually, the loss of extension is more commonly seen and more poorly tolerated than the loss of flexion.341 With the evolvement of rehabilitation concepts and the improvement of surgical technique, range of motion problems after ACL reconstruction have been minimized. Aggressive postoperative rehabilitation protocols are being used to restore the knee motion. The goal is to achieve full extension right after the surgery and regain 10 degrees of flexion per day. By 7 to 10 days after surgery, the knee should achieve 90 degrees of flexion. Bracing in slight hyperextension has been proposed as an easy way to ensure full knee extension.343 Early passive and active range of motion is augmented by the use of a continuous passive motion machine. Prevention is the key to achieving range of motion. It is suggested that control of pain and swelling, early reactivation of the quadriceps musculature, patellar mobilization, and early return to weight-bearing contribute to the return of motion. In comparison, postoperative immobilization may slow down the process and increase the need for manipulations to regain motion.344 It is worth noting that graft positioning is an important factor in regaining range of motion. If tunnel placement of the ACL graft is incorrect, it may limit knee flexion. With

Knee 1671

the growing popularity of double-bundle ACL reconstruction, some argue that it may cause ACL and PCL impingement and lead to loss of motion.345 This emphasizes the importance of correct tunnel positioning. If the tunnels are positioned anatomically and the normal anatomy is closely restored, range of motion should not be a concern in performing this type of surgery.326

Weight-Bearing Weight-bearing was prohibited in earlier rehabilitation protocols because of concern over graft laxity and failure. With the development of surgical technique and the knowledge that immediate rehabilitation produces no adverse reactions, a trend toward immediate weight-­bearing in the postoperative setting has evolved.346 Additionally, early weight-bearing may help to improve cartilage nutrition, reduce disuse osteopenia, and hasten quadriceps recovery. In a prospective randomized trial performed by Tyler and colleagues,347 immediate weight-bearing was compared with delayed weight-bearing for 2 weeks after ACL reconstruction with a central third BPTB autograft. At a mean follow-up of 7.3 months, no differences were observed between the two rehabilitation protocols in terms of range of motion, vastus medialis oblique function, and anteroposterior knee laxity (clinical examination and KT-1000). However, the immediate weight-bearing group showed a decreased incidence of anterior knee pain. The findings indicate that immediate weight-bearing after ACL reconstruction may be beneficial by lowering the incidence of anterior knee pain, whereas it does not apply excessive loads on the graft or its fixation. Accelerated ­rehabilitation protocols usually begin with tolerated weight-­bearing immediately after surgery with a progression to full ­weightbearing without crutches by 10 to 14 days.

Muscle Training Issues The early initiation of muscle training is very important in the prevention of muscle atrophy and weakness. Muscle activation and strengthening, including voluntary exercises, electrical muscle stimulation, and biofeedback, should be started before surgery as well as immediately after surgery. Electrical stimulation can help to initiate muscle activation when reflex inhibition can not be overcome in patients who are suffering severe pain and swelling. Biofeedback is helpful in enhancing the force of muscle contraction. Quadriceps muscle strength is correlated with good outcomes after ACL reconstruction. Strengthening of the quadriceps is the focus of many rehabilitation programs. However, achieving the appropriate hamstring-­to-­quadriceps ratio may provide even better protection for the ACL. The role of the hamstring muscles is to flex the knee joint, increase joint compression, and pull the tibia backward through a posterior shear force at tibial flexion angles greater than 20 degrees. Thus, hamstring contraction decreases ACL strain.348,349 Baratta and colleagues350 suggested that with reduced hamstring antagonist activity, the risk for injury is increased. However, no correlation could be found between hamstring strength and functional tests.351 Little attention has been paid to the gastrocnemius muscle, although some authors have demonstrated its functional importance for knee stability.328,352

Endurance training should be included in the rehabilitation program because fatigue affects not only the muscle contraction strength but also the electromechanical response time and rate of muscle force generation.353,354 Fatigue is also associated with decreased knee proprioception and increased joint laxity compared with baseline values.355,356 It has been shown to alter motor control strategies in recreational athletes, which may increase anterior tibial shear force, strain on the ACL, and risk for injury in both female and male subjects.357

Electrical Muscle Stimulation and Biofeedback Electrical muscle stimulation is used as an adjunct to voluntary exercises in an effort to recover muscle strength after ACL reconstruction. The effectiveness of this method is ­controversial in the literature. Sisk and colleagues358 claimed there was no significant difference in muscle strength as a result of electrical stimulation. Halkjaer-Kristensen and Ingemann-Hansen359 noted that isometric exercise and electrical stimulation were both ineffective in preventing muscular atrophy. However, Morrissey and colleagues360 reported that application of electrical stimulation alone is more effective than voluntary exercises. This controversy is due in part to the use of different electrical stimulation protocols, including parameters like frequency, intensity and impulse width, duration, and the number of electrodes and their placement. Snyder-Mackler and colleagues361 performed a randomized controlled trial of rehabilitation after ACL reconstruction with either a semitendinosus tendon combined with a ligament augmentation device or a central third BPTB preparation. Patients were randomized to undergo rehabilitation with neuromuscular electrical stimulation and volitional exercises or with volitional exercises alone. Eight weeks after surgery, patients were evaluated for their gait and thigh muscle strength. The incorporation of neuromuscular electrical stimulation into volitional exercises had resulted in more near-normal gait parameters and stronger quadriceps muscles. Later, Snyder-Mackler and colleagues362 reported the results of a multicenter randomized controlled trial of rehabilitation after ACL reconstruction with different graft materials and surgical techniques. The patients were assigned to treatments with high-intensity neuromuscular electrical stimulation, high-level volitional exercises, low-intensity neuromuscular electrical stimulation, or combined high- and low-intensity neuromuscular electrical stimulation. The 4-week follow-up results revealed that high-intensity neuromuscular electrical stimulation combined with volitional exercises was better at restoring extensor strength compared with volitional exercises alone. Biofeedback is a principle that is widely used in muscle rehabilitation and strengthening programs. The principle of electromyographic biofeedback is that the patient has a visual or auditory representation of the quality of muscle contraction. The patient then tries to enhance the level of the visual or auditory output by contracting the muscle group in or over which the electromyogram electrodes have been placed. Biofeedback is more effective than electrical stimulation in facilitating recovery of peak torque of quadriceps after ACL reconstruction.363 The ­combination

1672 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

of biofeedback and exercises is better than exercise alone in recovery of quadriceps function after ACL reconstruction,364 and better than exercise alone in quadriceps strengthening in normal subjects.365

Proprioception Proprioception is defined as the culmination of all neural inputs originating from joints, tendons, muscles, and associated deep tissue proprioceptors. These inputs are projected to the central nervous system for processing and ultimately result in the regulation of reflexes and motor control.366 Mechanoreceptors are specialized nerves located in skin, joints, tendon, ligament, and skeletal muscle. They serve as transducers to convert mechanical signals into afferent nerve signals, providing position sense and conscious awareness by initiating reflexes to stabilize joints and maintain stance. ACL deficiency results in an unstable knee. However, several authors have demonstrated that restoring mechanical stability alone (ACL reconstruction) does not guarantee functional stability.367,368 Anatomic and histologic studies have demonstrated the presence of proprioceptive mechanoreceptors in the fibers of the ACL.367,369 In addition, with forced increases in anterior tibial translation, muscle responses, including increases in hamstring activation and inhibition of quadriceps activity, were observed.370,371 It has been suggested that the ACL serves a sensory and proprioceptive role, in addition to its role as a mechanical stabilizer.367,368 Lephart and coworkers9 suggested that altered proprioception may reduce the effectiveness of protecting the knee and may predispose the ACL to repetitive microtrauma and ultimately failure. Proprioception was also shown to be positively related to the function of the ACL-injured knee.372 After ACL reconstruction, patients continue to have deficits in proprioception and neuromuscular joint ­control for at least 6 months and as long as 1 year after surgery.373,374 Thus, it is important to incorporate beginning, intermediate, and advanced proprioceptive training exercises throughout the postoperative rehabilitation protocol.375 Studies376,377 have shown proprioceptive deficits in both limbs of patients with unilateral ACL deficiency, indicating that clinicians should not use the contralateral limb as a control when assessing proprioceptive parameters. Another important issue of proprioception is the difference between genders. Females possess greater deficits in proprioception after injury or ACL reconstruction. This is further discussed in under “Gender Issues.”

Bracing Two forms of braces, rehabilitation braces and functional braces, are used to protect the graft in the rehabilitation of ACL reconstruction. Rehabilitation braces are used in the early postoperative period, and functional braces are used when the patient returns to strenuous activity. There appears to be a consensus that the use of a rehabilitation brace results in fewer problems than no bracing during the early stage of rehabilitation, including less pain, less swelling, and lower prevalence of hemarthrosis and wound drainage. Three independent randomized ­ controlled

t­ rials378-380 suggested, however, that rehabilitation bracing does not have a long-term effect on clinical outcome, range of knee motion, subjective outcome, anteroposterior knee laxity, activity level, or function. In addition, rehabilitation bracing was found to be helpful in preventing potential loss of extension range of motion. Two randomized controlled trials343,381 showed that the application of a rehabilitation brace in full extension or hyperextension during the early postoperative stage was effective in recovering full extension of the operative knee and that the stability of the graft was not compromised. The role of functional knee bracing in ACL reconstructions is controversial. Two randomized controlled trials382,383 did not show evidence of benefit by using functional braces after ACL reconstruction. However, biomechanical studies of functional ACL bracing suggest that some functional knee braces do increase mechanical stability under low loading conditions.52,384 In a recent cohort study,385 ACL reconstructed skiers without a functional brace were estimated to be almost 3 times more likely to sustain a subsequent knee injury.

Rehabilitation Protocol after Anterior Cruciate Ligament Reconstruction The rehabilitation protocol for ACL reconstruction has changed dramatically during the past several years. Instead of conservative rehabilitation with limitation of range of motion, delayed weight-bearing (8 to 10 weeks), and delayed return to sports (9 to 12 months), current ACL reconstruction rehabilitation protocols emphasize immediate range of motion, immediate weight-bearing, and earlier return to sports (4 to 6 months). Although there may be slight differences between protocols used in different practices, most are based on the same basic principles. Because of improvements in surgical technique and accelerated rehabilitation protocols, most recent ACL reconstruction outcome studies have achieved a 90% success rate in terms of restoring knee stability, patient satisfaction, and return to full athletic activity.199 The optimal graft material remains controversial. Autografts including the patellar, hamstring, and quadriceps tendons, and allografts including the quadriceps, patellar, Achilles, hamstring, anterior and posterior tibialis tendons and the fascia lata, are the main options. Although each graft choice has its unique features, no evidence has been provided that we need different rehabilitation protocols to match different graft choices. Comparison studies of allografts versus autografts226,386 and BPTB graft versus HS graft302,387,388 have demonstrated few differences in outcomes with similar accelerated rehabilitation protocols. Our rehabilitation protocol for ACL reconstruction is listed as an example of current trend. This protocol is used in all of our ACL reconstruction patients, regardless of the choice of graft and single-bundle versus double-bundle technique (Box 23D1-1).

Rehabilitation of Associated Injuries Meniscal damage frequently occurs at the time of an ACL injury. The frequent association of repairable tears with a torn ACL adds concerns about the influence of an

Knee 1673

Box 23D1-1 Anterior Cruciate Ligament Reconstruction and Rehabilitation Protocol Stage 1 Begin immediately after surgery through about 6 weeks. Goals 1. Protect graft fixation. 2. Minimize effects of immobilization. 3. Control inflammation. 4. Achieve full extension and flexion range of motion. 5. Educate patient on rehabilitation progression. Therapeutic Exercises 1. Heel slides 2. Quadriceps sets 3. Patellar immobilization 4. Non–weight-bearing gastroc-soleus stretches; hamstring stretches begin at 4 weeks 5. Straight leg raises in all planes with brace in full extension until quadriceps strength is sufficient to prevent extension lag 6. Quadriceps isometrics at 60 and 90 degrees Stage 2 Begin 6 weeks after surgery and extend to about 8 weeks. Goals 1. Restore normal gait. 2. Maintain full extension, progress with flexion range of motion. 3. Protect graft fixation. Therapeutic Exercises 1. Wall slides 0 to 45 degrees 2. Four-way hip machine 3. Stationary bike 4. Closed-chain terminal extension with resistance tubing or weight machine 5. Toe raises 6. Balance exercises 7. Hamstring curls 8. Aquatic therapy with emphasis on normalization of gait 9. Continue hamstring stretches, progress to weight-bearing gastroc-soleus stretches Stage 3 Begin 8 weeks after surgery and extend through about 6 months.

a­ ccelerated rehabilitation program on meniscal healing rates. Barber and Click suggested that no modification of an ACL reconstruction accelerated rehabilitation program is needed for meniscus repairs performed in conjunction with the reconstruction.131 Mariani and colleagues also showed that no deleterious effects were observed in patients undergoing ACL reconstruction and concomitant meniscus repair.389 Bone bruises and chondral lesions are also often seen with ACL injury. Little has been reported on the rehabilitation of these concomitant injuries.

Goals 1. Achieve full range of motion. 2. Improve strength, endurance, and proprioception of the lower extremity to prepare for functional activities. Therapeutic Exercises 1. Continued flexibility exercises as appropriate for patient 2. Stairmaster (begin with short steps, avoid hyperextension) 3. Nordic Track 4. Advanced closed-chain strengthening (one-leg squats, leg press 0 to 50 degrees, step-ups begin at 2 inches, etc. 5. Progress proprioception activities (slide board, use of ball with balance activities, etc.) 6. Progress aquatic therapy to include pool running, swimming (no breaststroke) Stage 4 Begin 6 months after surgery and extend through about 9 months. Goal 1. Achieve progress in strength, power, and proprioception to prepare for return to functional activities. Therapeutic Exercises 1. Continue to progress with flexibility and strengthening program. 2. Initiate plyometric program as appropriate for patient’s functional goals. 3. Achieve functional progression, including walking and jogging progression; forward and backward running at half, three fourths, and full speed; cutting; crossover; and carioca exercises, among others. 4. Initiate sport-specific drills as appropriate for patient. Stage 5 Begin 9 months after surgery. Goals 1. Safe return to athletics 2. Maintenance of strength, endurance, proprioception 3. Patient education with regard to possible limitations Therapeutic Exercises 1. Gradual return to sports participation 2. Maintenance program for strength and endurance

Functional Training During the past several years, restoring proprioception, dynamic stability, and neuromuscular control has been gradually incorporated into the rehabilitation programs of ACL reconstruction patients. They have a critical effect on the prevention of knee reinjuries. In addition, these drills are usually fun and take some of the focus away from the knee, thereby helping to maintain the patient’s motivation during physical therapy.

1674 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Basic proprioceptive exercises such as joint repositioning, CKC weight-shifting, and minisquats are usually started during the second postoperative week, when the pain and swelling associated with surgery subsides and the patient regains quadriceps control. As proprioception is advanced, drills to encourage agonist-antagonist muscle coactivation during functional activities should be incorporated. Dynamic stabilization drills, including single-leg stance on flat ground and unstable surfaces, cone-stepping, and lateral lunge drills, usually begin during the first 2 to 3 weeks. These drills help to enhance dynamic stability and facilitate gait normalization.390 Plyometric jumping drills are also used to facilitate dynamic stabilization and neuromuscular control of the knee joint. By producing a maximal concentric contraction following a rapid eccentric loading on the muscle, plyometric training is used to train the lower extremity to avoid injury by producing and dissipating forces.391,392 Another aspect of rehabilitation regarding neuromuscular control is the enhancement of muscle endurance. Bicycle, elliptical machines, stair climbing, and slide boards are highly repetitive and low-resistance drills that are safe to be used for long durations to strengthen the muscles and to train the muscles to perform against fatigue. When performed toward the end of a treatment session, these drills can challenge the neuromuscular control of the knee joint when the dynamic stabilizers have been adequately fatigued.390 Sport-specific drills should be added slowly at the late stage of rehabilitation to ensure the safety of the graft. The intention of sport-specific training is to simulate the functional activities associated with sports and facilitate the return to the previous level of sports activities. It also helps to prevent injury by training the neuromuscular system to perform in a reflexive pattern. It is worth noting that neuromuscular control at the knee is supported by the entire lower extremity and the core muscles. Therefore, exercises to enhance total body strength, flexibility, endurance, and neuromuscular control must not be neglected.

Functional Testing To measure the functional status of the knee, a variety of tests have been used. Noyes and colleagues393 developed a battery of functional tests consisting of the single hop for distance, the triple hop for distance, the crossover hop for distance, and the 6-min timed hop. Besides these commonly used functional tests, the vertical jump, the crossover hop for distance, and the figure-eight hop have also been proposed. Because no single test can evaluate the dynamic function of the knee adequately, a series of functional tests is usually used in combination to test knee function. Papannagari and associates394 measured the kinematics of reconstructed and the intact contralateral knees 3 months after surgery using a dual-orthogonal fluoroscopic system while the subjects performed a single-leg ­weight-bearing lunge. There was a significant increase in

anterior translation of the reconstructed knee compared with the intact knee at full extension and 15 degrees of flexion. Meanwhile, the anterior laxity of the reconstructed knee as measured with the arthrometer was similar to that of the intact contralateral knee. This suggested that future ACL reconstruction should aim at restoring the function of the knee under physiologic loading conditions. Researchers have noted persistent instability with functional testing and degenerative radiographic changes after single-bundle ACL reconstruction.31,167 Single-bundle ACL surgery and biomechanical studies of ACL ­reconstruction have focused on restoring anterior stability in response to anterior tibial loads. However, knee joint kinematics are not restored under functional loading conditions and are at least partly responsible for the development of degenerative changes.31,395-397 Double-bundle ACL reconstruction more closely restores the normal anatomy of ACL and may be the future of ACL reconstruction.398 Its efficacy of restoring normal knee kinematics needs to be evaluated by functional tests. In addition, current functional tests are typically performed under nonfatigued test conditions in both the clinical and scientific settings. The ability of these tests to assess whether a patient has regained lower extremity function after ACL reconstruction is therefore limited. Augustsson and Thomee399 developed a single-leg hop test performed under standardized, fatigued conditions and have examined its reliability. They suggested that functional testing should be performed both under nonfatigued and fatigued test conditions to evaluate the ACL reconstructed knee comprehensively.400

Criteria for Return to Play The decision of when to permit a patient to return fully to unrestricted activities and sports is empirical in most cases because there is poor correlation of functional testing, clinical testing, and subjective testing methods in evaluating a patient after ACL reconstruction (Table 23D1-11). In a retrospective study, Glasgow and colleagues401 found no difference in sagittal translation or graft failure in patients who returned to sports activities before or after 6 months after surgery. Shelbourne and Davis.402 observed that some of their patients participated in sporting activities against their advice 3 months after surgery. Measurements of sagittal translation before and after sports participation revealed increased translation in only 2% of the patients. Although unnecessary delay of returning to unrestricted activities should be avoided, a premature return is dangerous and can jeopardize the graft. The use of multiple criteria, including the return of range of motion, muscle strength and balance, static stability as measured by KT1000, and dynamic stability as measured by functional testing, is necessary in determining the clearance for a patient to return to full activity. Unfortunately, many practicing orthopaedists simply make their decision based on time instead of these criteria.

Knee 1675

   TABLE 23D1-11  Review of Recent Literature for Returning to Play after Anterior Cruciate Ligament Reconstruction Study

Method

Weight-Bearing

Brace

Exercises

Running, Drills

Return to Play

Aglietti et al, 1997403 Anderson et al, 2001203

Hamstring vs. patellar tendon Hamstring vs. patellar tendon

PWB day 3, FWB at 8 wk 25% × 1 week, 50% week 2, full week 3

7 days

4 mo

7-8 mo

Day 1

12 wk

6-7 mo

Aune et al, 2001299 Beynnon et al, 2002301

Hamstring vs. patellar tendon Two-strand hamstring vs. patellar tendon

Full immediately

Knee immobilized 4 wk Knee immobilized after surgery; functional at 3 wks None

CCE and bike at 2 wk 7 days

6 wk

6 mo

Run at half speed 12 wk, full speed 4 mo

6-8 mo

Ejerhed et al, 2003303

Hamstring vs. patellar tendon

Full immediately

Running at 3 mo

6 mo

Running at 10 wk

9 mo

As tolerated

6-8 mo

TTWB × 3 wk

Hinged brace locked at 10 degrees flexion for 1 wk, then motion allowed on day 7 in brace, used as necessary at 5 wk None CCE immediately, full extension with load at 6 wk None CCE × 6 mo, bike 3 wk Postoperative brace CPM × 1 wk, early in full extension for hydrotherapy PT, 10 degrees of extension ext lock for hamstring None OCE 4 wk

8-10 wk

4 mo

Conserve: PWB at 2 wk, full at 4 wk

Hinged brace for 12 wk

Conserve: ROM at 1 wk, isokinetic at 8 wk

Conserve: sports activity at 6 mo

Conserve: 12 m  o

Accelerated: ROM immediate, CCE at 2-3 wk CCE 7-10 days

Accelerated: swimming 5-6 wk

Accelerated: 6-9 m  o

Jogging at 6 wk

6 mo

TTWB × 3 wk, then WBAT

Feller & Webster, Hamstring vs. WBAT 2003305 patellar tendon Gobbi et al, Quadrupled bone– WBAT 2003404 semitendinosus vs. patellar tendon Howell & Taylor, Hamstring 1996405 Majima et al, Hamstring, 2002406 accelerated vs. conservative

Accelerated: WBAT immediate

Pinczewski et al, 2002407 Rose et al, 2004408 Shaieb et al, 2002308

Hamstring vs. patella tendon Hamstring vs. patellar tendon Hamstring vs. patellar tendon

WBAT

None

Full at day 1 WBAT at 1 wk

Hinged postoperative brace for 6 wk Not stated

Shelbourne & Davis, 1999402

Patellar tendon

WBAT

Functional brace

CPM × 4 days, Drills at 7 wk, CCE at 3 wk running at 12 wk ROM at 1 week, Running at 2 mo CCE and bike at 2-3 wk Active-assisted 5-6 wk flexion at 7-10 days

6 mo 5-6 mo 2-6 mo

CCE, closed chain exercises; CPM, continuous positive motion; FWB, full weight-bearing; OCE, open chain exercises; PT, physical therapy; PWB, partial ­ weightbearing; ROM, range of motion; TTWB, toe touch weight bearing; WBAT, weight bearing as tolerated. From Cascio BM, Culp L, Cosgarea AJ: Return to play after anterior cruciate ligament reconstruction. Clin Sports Med 23(3):395-408, 2004.

C l

r i t i c a l

P

o i n t s

 he ACL is an intra-articular ligament that consists of T two functional bundles, the AM bundle and the PL bundle, named for their tibial insertion sites. l The insertion site orientation and the position and the length of the ACL bundles vary with changing angles of knee flexion and extension. In full extension, both bundles are in parallel and act as anteroposterior stabilizers of the knee joint. In flexion, the bundles cross one another, and the PL bundle functions in rotational stabilization of the knee. l Biomechanical studies have shown that single-bundle reconstruction cannot restore normal knee kinematics. Double-bundle ACL reconstruction closely restores normal ACL anatomy and knee kinematics.

l Females have an increased risk for ACL rupture beginning at about age 12 years. The reasons are multifactorial and may include a decreased notch width index, hormonal differences, and altered neuromuscular firing patterns that help create increased anterior and valgus moments about the knee. l There is no rigid upper age limit on ACL reconstruction. Symptoms such as recurrent instability, active lifestyle, and the ability to comply with postoperative guidelines are important aspects to consider. l Partial ACL tears, often of a single ACL bundle, are being diagnosed with increasing frequency. A combination of physical examination and MRI findings is helpful in making the diagnosis. l The MCL and lateral meniscus are commonly injured concurrently with an ACL tear. Medial meniscal injuries are more common in chronic ACL tears.

1676 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� l The natural history of an ACL-deficient knee in an active or young patient is often one of progressive meniscal and chondral degeneration, which may lead to early-onset arthrosis. l Multiple ACL graft and fixation options exist for ACL reconstruction. Both HS and BPTB autografts have performed well in follow-up studies. l Although the results of ACL reconstruction are generally good, there is no evidence to date that ACL reconstruction can prevent knee arthrosis. As more research is completed, the use of an anatomic double-bundle ACL reconstruction may help improve ACL outcomes.

S U G G E S T E D

Buoncristiani AM, Tjoumakaris FP, Starman JS, et al: Anatomic double-bundle anterior cruciate ligament reconstruction. Arthroscopy 22(9):1000-1006, 2006. Chhabra A, Starman JS, Ferretti M, et al: Anatomic, radiographic, biomechanical, and kinematic evaluation of the ACL and its two functional bundles. J Bone Joint Surg Am 88(Suppl 4):2-10, 2006. Fu FH, Bennett CH, Lattermann C, Ma B: Current trends in anterior cruciate ligament reconstruction. Part 1: Biology and biomechanics of reconstruction. Am J Sports Med 27:821-830, 1999. Girgis FG, Marshall JL, Monajem A: The cruciate ligaments of the knee joint: Anatomical, functional and experimental analysis. Clin Orthop 106:216-231, 1975. Goldblatt JP, Fitzsimmons SE, Balk E, et al: Reconstruction of the anterior cruciate ligament: Meta-analysis of patellar tendon versus hamstring tendon autograft. Arthroscopy 21(7):791-803, 2005. Hewett T, Ford K, Myer G: Anterior cruciate ligament injuries in female athletes. Part 2: A meta-analysis of neuromuscular interventions aimed at injury prevention. Am J Sports Med 34:490-498, 2006. Spindler KP, Kuhn JE, Freedman KB, et al: Anterior cruciate ligament reconstruction autograft choice: bone-tendon-bone versus hamstring: does it really matter? A systematic review. Am J Sports Med 32(8):1986-1995, 2004.

R E A D I N G S

Arnoczky SP: Anatomy of the anterior cruciate ligament. Clin Orthop 172:19-25, 1983. Beynnon, BD, Johnson RJ, Abate JA, et al: Treatment of anterior cruciate ligament injuries: Part 1. Am J Sports Med 33:1579-1602, 2005. Beynnon, BD, Johnson RJ, Abate JA, et al: Treatment of anterior cruciate ligament injuries: Part 2. Am J Sports Med 33:1751-1767, 2005.

R e f erences Please see www.expertconsult.com

S e c t i o n

D

Anterior Cruciate Ligament Injuries 2. Anterior Cruciate Ligament Injuries in the Child Nicholas J. Honkamp, Wei Shen, and Freddie H. Fu

The past 2 decades have seen a significant increase in the number of anterior cruciate ligament (ACL) tears in adolescents. Most of these injuries involve either a midsubstance ACL tear or a tibial spine avulsion fracture that contains the ACL attachment. Femoral ACL avulsion fractures are rare. Central to this increase is the explosion of athletic participation in both male and female adolescents in sports commonly associated with ACL injuries such as soccer, basketball, skiing, and football. Coincident with this increased participation level is the improved diagnostic capabilities of the physician using magnetic resonance imaging (MRI), instrumented laxity measuring devices, and arthroscopy. Although the treatment of tibial spine avulsion fractures is generally agreed on, the management of midsubstance ACL tears is still a matter of debate. The potential for skeletal growth postoperatively presents unique problems when surgical treatment is considered. Nonoperative management with bracing and rehabilitation has been the traditional approach, but more recent reports document poor functional results and a high incidence of further injury to the menisci and joint surfaces. Extra-articular reconstructive procedures are nonanatomic, and their long-term

results are unknown. For these reasons, intra-articular reconstruction has been proposed. ACL reconstruction with transtibial and transfemoral tunnels, transtibial and over the top femoral placement, and nontunnel techniques has been reported. Recent studies simulating these procedures in growing animals have had mixed results, with some finding few growth disturbances and others frequent and severe angular and length deformities. These studies suggest that tunnel size, graft choice, tunnel fill, and graft tension all play a role in skeletal growth after reconstruction. The current review seeks to critically appraise the literature to date and construct an analysis of the current techniques and their reported results.

INCIDENCE Clanton reported the earliest review of the literature in 1979, citing less than 1% incidence of knee ligament injuries in all knee injuries reported in patients younger than 14 years.1 DeLee and Curtiss in 1983 reported an incidence of 1% in children aged 14 years or younger with knee injuries.2 More recent reports puts the figure at

knee 1677

3.3% to 3.4%.3,4 Interestingly, Kellenberger and associates5 reported an 80% incidence of tibial spine avulsions in children aged 12 years or younger, compared with a 90% incidence of intrasubstance ACL tears in children 12 years or older. Thus, an age of about 12 to 14 years may serve as a rough defining zone between possible avulsion fractures and intrasubstance ACL tears. As children age into late adolescence, the relative weakness of the proximal tibial physes relative to the ACL diminishes because the closing physes no longer represent the weak link in the knee joint support system.

MECHANISM OF INJURY, HISTORY, AND PHYSICAL EXAMINATION ACL injuries in children, like those of an adult, are usually caused by a quadriceps active, noncontact valgus injury. A fall off a bike with the use of a planted foot to help break the fall is a frequent mechanism.6-8 Like adults, most children present with a knee hemarthroses in the acute case, or recurrent effusions and instability in the chronic ACLdeficient patient. Often, an audible sound can be heard with an ACL injury. A reluctance to bear weight or return to activities and loss of range of motion are also common. Children represent a unique population in that congenital or physiologic states should always be considered in the differential diagnosis. Thus, the practice of examining the contralateral limb in an adult patient takes on special importance in the child. What may present as an ACLdeficient knee may in actuality be physiologic laxity or congenital absence of the ACL. Examining other joints in the lower or upper extremities may also help confirm excessive laxity or ligament absence. The standard tests of knee laxity, including the Lachman, anterior and posterior drawer, and pivot shift tests, should all be elicited. Often, guarding and swelling are present to a degree that ligamentous examination may be difficult or impossible. Re-examination in 1 to 2 weeks may then be helpful. Additional attention in the physical examination should be paid to patellar instability or subluxation represented by a positive apprehension test, increased Q angle, hindfoot valgus, or foot pronation. Referred pain from the hip may occur with congenital pelvic deformities or slipped capital femoral epiphysis. Infection should always be included in the differential diagnosis in a child as well. Plain film radiography should always be obtained before more advanced imaging. Anteroposterior, lateral, and patellar views should be obtained to look for periarticular fractures, avulsions, osteochondral damage, and congenital or developmental abnormalities.

Assessment of Skeletal Maturity Numerous methods have been used to assess skeletal maturity in adolescents. The most common, devised by ­Tanner,9 physiologic signs of development, including the presence of menarche, breast tissue, and pubic hair in females and the development of the ­ testes and penis and pubic hair in males. Other methods to assess skeletal ­maturity include anteroposterior and ­ lateral ­ radiographs

of the knee, assessment of bone age using radiographs of the hand, Risser staging of iliac crest ossification, ­presence or absence of an adolescent growth spurt, and comparison of the patient’s height to that of older siblings and parents.3

TIBIAL EMINENCE FRACTURES Fractures of the tibial eminence occur because of a chondroepiphyseal avulsion of the ACL insertion on the anteromedial tibial eminence.10,11 Meyers and McKeever described a classification system for avulsion fractures based on the elevation and rotation of the eminence fragment.7 Type I fractures have minimal or no displacement, and type II fractures have anterior hinging of one third to one half of the tibial eminence. Type IIIA fractures are completely displaced from the fracture bed, and type IIIB fractures demonstrate rotational malalignment (Fig. 23D2-1). Zaricznyj added a type IV to Meyers and McKeever’s fracture classification in 1977.12 He described seven comminuted tibial eminence fractures that he described as type IV. Additional knee injuries should also be sought during the evaluation of a child with a tibial eminence fracture, including meniscal tears, meniscal entrapment within the fracture (most commonly medial), and associated collateral ligament injuries.

Treatment Type I fractures are universally treated nonoperatively with immobilization owing to their nondisplaced nature. Type II fractures can often be successfully treated with closed reduction and immobilization with or without associated aspiration of the hemarthrosis to allow for improved extension. Although some authors believe that immobilization in full extension helps reduce the avulsion fragment through direct compression from the lateral femoral condyle,8,13,14 it has been shown that the ACL functionally lengthens in the last 20 degrees of extension.15 Thus, data support treatment of type II fractures with knee immobilization in approximately 15 to 20 degrees of flexion for 4 to 6 weeks, which places the least strain on the ACL–­avulsion fragment complex.16-19 In whatever position the knee is ultimately immobilized, fragment reduction should be confirmed with radiographic methods. Often, soft tissue interposition consisting of menisci or cartilage prevents the eminence fragment from anatomically seating in its bed. Kocher and colleagues20 found that 65% of type III fractures and 26% of type II fractures had entrapment of the anterior horn of the meniscus, most commonly medial. This has been confirmed by other authors as well, particularly in type III fractures.21,22 Additionally, plain radiographs may underestimate the extent of the fracture when cartilage composes a large proportion of its volume.23 Because of these findings, many authors have advocated direct inspection through open or arthroscopic means of all type III fractures and any type II fracture that does not reduce anatomically on radiographs.20,23 With either method, the fracture fragment and its associated bony bed should be débrided of any clot or debris, and any associated meniscochondral entrapment should be released. Fracture fixation has been commonly obtained

1678 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 23D2-1   Meyers and McKeever classification of tibial eminence fractures in children. A, Type I is a nondisplaced fracture. B, Type II is a displaced fracture that is hinged posteriorly. C: Type III fractures are completely displaced.

with suture,22,24,25 wire,26 or screw fixation.6,8,14,27,28 Comminuted type IV fractures may be best managed with suture fixation through the soft tissue insertion of the ACL fibers. Transepiphyseal fixation is not recommended because of the risk for anterior growth arrest and hyperextension deformity.29 Follow-up studies of type III fractures using cannulated screw fixation have all found good functional outcomes despite persistent laxity. Kocher and coworkers28 reviewed their results in six patients at minimal 2-year follow-up. They found mean postoperative Lysholm and Tegner scores of 99.5 and 8.7, respectively. One patient had a grade A Lachman (normal) test, three had grade B (nearly normal), and two had grade C (abnormal). Instrumented knee laxity showed side-to-side differences of greater than 3 mm in four of six patients. Smith and colleagues30 found subjective instability symptoms in only 2 of 13 patients, but 87% had a positive Lachman test. Reynders and associates31 found similar results of good subjective stability with documented objective knee laxity in their 26 patients treated with an intrafocal screw and spiked washer. Davies and coworkers32 used a cannulated Acutrak screw in their four pediatric cases, with all patients returning to their preinjury exercise status.

Larger studies comprising all three types of tibial eminence fractures treated with a variety of fixation methods (closed reduction with or without aspiration, arthroscopic reduction and casting, suture fixation, screw fixation) have been done by Janarv and colleagues6 and Willis and associates14 Similarly, Janarv and colleagues6 found that 38% of their 61 cases had pathological knee laxity that was not related to subjective knee function. Willis and associates14 found clinical signs of anterior instability in 64% and instrumented laxity in 74% of their 50 patients. However, most of these patients had no subjective complaints. Persistent anterior knee laxity despite anatomic or near anatomic reduction and healing of the tibial spine avulsion fractures in children is likely related to some interstitial stretch injury to the ACL at the time of the fracture. In a primate study, Noyes and colleagues10 found frequent elongation and disruption of ligament architecture despite gross ligament continuity in experimentally produced tibial spine fractures at both slow and fast loading rates. Another potential complication of either operative or nonoperative treatment is loss of terminal knee extension.7,8 This can be caused by either hypertrophy of the tibial spine secondary to hyperemia or residual displacement and can lead to a bony block to full extension.

Authors’ Preferred Method The goal of any treatment approach should be anatomic reduction of the fracture fragment. Type I fractures may be treated with cast treatment near full extension for 4 to 6 weeks. Type II fractures should undergo attempted closed

reduction in extension under anesthesia with radiographic evaluation. Because of the high rate of meniscal entrapment or other debris causing a block to anatomic reduction, we advise arthroscopic or open treatment of any type II fracture

knee 1679

Authors’ Preferred Method—cont’d

A

B

Figure 23D2-2   Anteroposterior (A) and lateral (B) radiographs of a type III tibial eminence fracture.

Figure 23D2-3   Sagittal magnetic resonance imaging section showing a type III tibial eminence fracture.

MIDSUBSTANCE ANTERIOR CRUCIATE LIGAMENT TEARS Nonsurgical Treatment Because of the perceived rarity of the injury, as well as fear of damaging open physes, much of the initial published literature discussed nonoperative treatment. The distal femur and proximal tibia is a key growth center in the ­longitudinal

that does not anatomically reduce in extension. We recommend open or arthroscopic treatment of all type III fractures (Figs. 23D2-2 and 23D2-3). Fixation may be obtained with any cannulated screw device or with heavy suture (1-0 or 2-0 nonabsorbable) placed through the base of the ACL attachment. The suture technique is especially useful in cases in which the bony ­component is small or fragmented. We bring the sutures out in a transepiphyseal fashion and tie them over a bony bridge. We often use an ACL-aiming guide for creation of our epiphyseal tunnels (Fig. 23D2-4). Care should be taken to ensure that all hardware and bone tunnels be placed in an ­extraphyseal location and that the reduction is confirmed ­visually and radiographically. Range of motion and strengthening exercises may then be started on a gradual basis, usually 2 to 4 weeks after surgery, depending on the strength of fixation. Return to activities is allowed between 3 and 4 months after fixation if there is evidence of radiographic healing and clinical stability of the knee.

Figure 23D2-4   The use of an anterior cruciate ligament– aiming guide to help direct the epiphyseal placement of the tibial tunnels for fixation.

growth of the leg, accounting for about 65% of longitudinal growth of the leg. Concern over causing premature growth plate damage and subsequent closure or angular deformity dictated that early treatment be nonoperative. Bracing, quadriceps and hamstring strengthening, and activity modification were cornerstones of early treatment. Chick and Jackson,33 in 1978, presented one of the first series on partial or complete ACL ruptures managed nonoperatively in young adults. In their series of 30 patients

1680 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

with an average age of 20 years, symptoms of knee instability, intermittent effusion and stiffness, and mild sclerotic changes on radiographs were noted. A similar series by Fetto and Marshall34 in 1979 involving more than 200 young adult patients with ACL tears noted “a progressive deterioration and dysfunction” of the knee joint. Hawkins and associates,35 in 1986, described 40 ACLdeficient patients, with an average age of 20 years, treated nonoperatively. Thirty-five of 40 rated their knee as fair or poor, and 12 of 40 considered it a liability. These studies on young adults demonstrate that recurrent instability due to nonoperative treatment of ACL injuries can often lead to increased incidence of meniscal and articular cartilage damage. Attention to the long-term outcomes of nonoperative ACL injury began to focus on the adolescent population in the 1980s and early 1990s. Numerous studies documented low satisfaction rates, low knee function scores, high rates of meniscal pathology, and poor return to sport and recreational activities. In a 1988 paper, McCarroll and associates4 reported on 16 patients with torn ACLs, with an average age of 13 years, treated conservatively with bracing and rehabilitation. Nine of 16 gave up their sport because of continued instability. Of the 7 patients who continued playing their sport, all had recurrent “giving way” episodes, and 3 had serious reinjuries to their affected knee. Kannus and Jarvinen36 reported on 32 adolescent patients with partial or complete knee ligament injuries treated by cast immobilization. Partial ligament injuries had good to excellent results. However, the 4 patients with complete ACL ruptures and longer term followup had decreased knee scores, definitive radiographic evidence of post-traumatic arthritis, and decreased strength. Angel and Hall37 reported on 22 patients with ACL tears treated conservatively. In their average 51-month ­follow-up, 41% had meniscal pathology, and no patients with complete ACL tears were able to return to sports. Graf and associates38 and Mizuta and Kubota39 both reported on adolescents with complete ACL tears treated with bracing and hamstring and quadriceps strengthening. All their patients ultimately failed bracing with recurrent instability and pain. Sixteen of their 20 patients treated strictly nonoperatively suffered meniscal tears. Although the numerous authors cited previously found poor results with nonoperative treatment, one recent study did find acceptable results with delayed nonoperative treatment. Woods and colleagues compared 13 patients (average age, 13.8 years) who underwent delayed surgical reconstructive treatment of the ACL-deficient knee once they had reached skeletal maturity to a group of skeletally mature adolescents who underwent the same arthroscopically assisted ACL reconstruction with a bone–patellar tendon–bone (BPTB) graft. They showed no significant differences between the groups in additional intra-articular knee injuries. Before reconstruction, the skeletally immature patients were given specific rehabilitation exercises, strictly forbidden from participating in vigorous team sports involving cutting or twisting activities, excused from gym class, and given a noncustom (“off the shelf”) knee brace.

Surgical Treatment Primary Anterior Cruciate Ligament Repair Similar results were also found when primary repair of a torn ACL was attempted in an adolescent. Two series2,40 dealing strictly with adolescents found that 7 of 11 patients had unstable knees, and the authors did not advocate further use of this procedure. An additional study41 including both pediatric and adult ACL injuries treated with primary repair or a synthetic ligament repair found lower levels of activity and increased episodes of instability. Difficulties relating to primary repair stem from interruption of the blood supply as well as exposure to the knee joint synovial fluid. Tearing of the ACL severs its main blood supply from the middle genicular artery. Additionally, this injury often damages the ligament’s synovial sheath, which further destroys its blood supply and exposes it to the harsh and limited nutritional supply of the synovium and joint fluid. The adolescent ACL injury thus represents a difficult problem. This population is not amenable to dependable activity modification and subjects their knee joints to high functional levels. Benefits gained by not violating the physeal growth plates and limiting growth arrest and angular deformities are mitigated by the increased rate of meniscal and cartilage damage that can similarly curtail future functional capabilities through the development of early degenerative joint disease. Additionally, the psychosocial drawbacks of limited recreational and sporting activities in the adolescent managed with activity modification must be considered.

Extra-articular Reconstruction As nonoperative and primary repair procedures reported suboptimal results, extra-articular reconstruction became more widely reported. These procedures, while avoiding the physes, provide some increased stability. The types of reconstructions have varied, as have their results. Various modifications of an iliotibial (IT) band tenodesis have been performed.4,38 These procedures take a portion of the IT band left attached distally, secure it to the lateral femur proximally, place it from posterior to anterior through the femoral notch, and tack it to the proximal tibia distally. Less than half of the reported patients in these series treated with this procedure have remained free of instability. Micheli and colleagues42 used both an intra-articular and extra-articular repair with the IT band with improved results in 10 prepubescent patients. Kocher and associates43 used a combined intra-articular and extraarticular IT band graft in 42 patients in Tanner stage 1 or 2 and reported excellent midterm results at a mean of 5.3 years. Similarly, some surgeons44,45 have devised intra- and extra-articular reconstructions using the gracilis and semitendinosus tendons on the medial aspect of the knee, or reconstructions through the epiphyseal portion of bone.46 These reconstructions avoid violating the physes by placing the grafts in various positions, including through epiphyseal drill holes above the physes, grooved troughs on the tibia and femur, or threaded positions under the menisci.

knee 1681

Although these studies have shown some positive results, the data are hindered by low numbers and varied outcomes measures. In addition, varied skeletal ages and follow-up periods make comparisons difficult. For instance, some patients returned to lesser levels of competitive sport, whereas others returned with bracing or other precautions not specified.

Transphyseal Anterior Cruciate Ligament Reconstructions—Animal Studies Although transphyseal drill holes are the gold standard of adult ACL reconstruction because of their anatomic placement of ligament grafts and their superior fixation, drilling across an open physis is still controversial. In an attempt to more accurately assess the risk for transphyseal drilling leading to premature physeal closure in adolescents, numerous animal studies have been reported. Earlier animal studies demonstrated that a hole drilled across an open growth plate will fill with bone and subsequently result in a physeal bar formation and at least partial physeal fusion.47 A smooth pin placed across an active growth plate and left in place will prevent the formation of a physeal bar. This is the rationale for the frequent use of smooth pins in the treatment of various conditions involving the fractures through the growth plates of children, including the distal tibia and distal humerus. Stadelmaier and associates47 used skeletally immature dogs to show that a semitendinosus graft placed across the femoral and tibial growth plates also prevented the formation of a bony bridge. Factors such as the size and location of the drill hole have been shown to be important in determining the extent of physeal damage. Makela and colleagues48 used New Zealand white rabbits to show that drilling the equivalent of 7% or more of the cross-sectional physeal growth plate resulted in significantly higher rates of shortening and angular deformities. Janarv and associates,49 using similar animals, confirmed these results and concluded that drilling 7% to 9% of the physeal cross-sectional area resulted in growth disturbance, whereas no retardation was seen in injuries of 4% to 5%. Guzzanti and coworkers50 used white rabbits to show that physeal drilling of 3% to 4% of the cross-sectional area of the femur and tibia caused 3 of 21 tibias to develop shortening or valgus angulation, whereas no leg-length discrepancies (LLDs) were found in any of the 21 femurs analyzed. Although an untensioned graft, like those used in the aforementioned studies, may help prevent physeal bar formation, human ACL reconstructions involve use of a tensioned graft to provide normal ligamentous tension and adequate postoperative stability. The Heuter-Volkmann principle asserts that although certain amounts of compression and tension can stimulate physeal bone growth, supraphysiologic forces (like a tensioned graft) may cause physeal bar formation and premature physeal closure. Studies by Houle and colleagues51 and Edwards and ­coworkers52 suggest a relationship between tensioned ACL grafts across either dog or rabbit tibial and femoral physes and significant shortening and angular deformities. Again, the tibia seemed more sensitive to the effects of physeal drilling.

In summary, the animal data do offer some support to limiting physeal damage to less than 7% to 8% of the crosssectional area of the physes. It is estimated that an 8-mm drill hole used in an ACL repair in a 12-year-old patient destroys 3% to 4% of the physis. Drill holes placed more perpendicular to the physes damage less cross-sectional area and are preferred. Placement of tensioned grafts, particularly across the tibia, appears to increase the chance of angular deformities.

Transphyseal Anterior Cruciate Ligament Reconstructions—Human Studies Some authors have used similar-sized drills with 8-mm drill holes to place allograft or autograft ACL reconstructions in a transphyseal position across the tibia, while placing the graft behind and posterior to the lateral femoral condyle in an over-the-top position on the femur. Studies by Lo and colleagues,53 Andrew and associates,54 and Bisson and coworkers55 used this procedure and reported moderate success in 22 patients. Generally, these patients were 13 years or older and had follow-ups ranging from 39 to 88 months. Three patients had LLDs greater than 5 mm, with 18 of 22 patients rating their outcome as good or excellent. There were two re-tears, and 3 patients were unable to return to their preinjury activities. The failure of nonoperative treatment, combined with the mixed results and lack of long-term follow-up of extraarticular procedures, has led several authors to recommend operative treatment. Particularly in the older (13 to 14 years and older) adolescent, drilling of the tibia and femur has gained in popularity as more reported studies are published. Lipscomb and colleagues56 reported in 1986 the technique of ACL reconstruction using a transphyseal tibial tunnel and a femoral tunnel placed distal to the femoral physes in the distal femoral epiphysis. They reported on 24 patients aged 12 to 15 years treated with the aforementioned reconstruction supplemented with an IT band tenodesis. Sixteen patients reported their knee as “normal,” and 8 as “improved.” Twenty-three of 24 patients had good or excellent objective results. However, 5 patients had LLDs of 6 to 10 mm, and 2 patients had LLDs of 13 mm and 20 mm. In McCarroll’s landmark 1988 series,4 he also included 14 patients treated with autologous BPTB reconstructions using tibial and femoral transphyseal drill holes. All 14 returned to their previous sports. There were no growth abnormalities reported, and the average passive anterior drawer sign was 1.7 mm. Two additional studies by Edwards and colleagues57 and Pressman and coworkers58 showed encouraging results with transphyseal tibial and femoral tunnels. Pressman58 compared this intra-articular, transphyseal technique with nonoperative and primary repair techniques. The transphyseal technique demonstrated significantly better objective knee outcomes scores, and no LLD greater than 1 cm. Edwards57 evaluated 20 patients, average age 13.7 years, with hamstring or BPTB transphyseal autografts. Nineteen patients returned to their previous sport, and 15 rated their function as excellent. Two patients with poor results suffered re-tears within 1 year of operation, whereas

1682 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

2 additional patients with poor results had excessive anterior translation and stretching of their graft. Additionally, 3 patients had greater than 5 mm LLD, all of which were asymptomatic. As more studies with ACL reconstruction using transphyseal drilling were published, more attention began to be focused on accurate measuring of LLD, rates of ACL reinjury, and size and placement of the graft. McCarroll and colleagues59 published a follow-up series in 1994 including 60 patients, average age 14.2 years, with a mean follow-up of 4.2 years. Again, transphyseal tibial and femoral tunnels were used (diameter not reported). No LLDs or angular deformities were reported using plain radiographs and manual measurements. Average postoperative growth measured 2.3 cm. Three patients tore the ACL graft longer than 2 years after surgery, whereas 4 patients tore their opposite ACL during the follow-up period. Aronowitz and associates60 published a series on 19 patients treated with Achilles tendon allograft placed through both physes. Average age was 13.4 years, and average follow-up was 25 months. The authors used 9- to 10-mm drill holes, and measured leg lengths postoperatively with scanograms. All patients were satisfied with their repair, and 16 of 19 returned to their sport. Average leg length discrepancy was −1.2 mm on the

­ perative side. In 2002, Fuchs and colleagues61 reported o on 10 patients treated with BPTB graft. Average age was 13.2 years. At an average of 40 months of followup, 9 rated their result as excellent and 1 as good. Average postoperative growth was 10 cm, and none had a significant LLD by clinical examination or by patient complaint. Overall, transphyseal reconstruction using hamstring, patellar tendon, or Achilles grafts offers an anatomic reconstruction with good clinical results and a low incidence of LLD. Overwhelmingly, these studies have focused on children 13 to 14 years old and older. Care was taken to place only the soft tissue graft across the physes, and drill holes less than 8 to 10 mm were used. Taking these precautions, growth disturbances can still occur. A case report of a 14-year-old boy with 9-mm transphyseal tunnels demonstrated a significant distal femur valgus deformity.62 It was attributed to a cannulated screw placed across the distal femoral physes as well as bone plugs placed across both physes. Kocher and colleagues, in a study based on a survey of experts in the management of pediatric ACL injuries, reported 15 patients who had growth disturbances including varus and valgus deformities, tibial recurvatum, and LLDs.63 Physeal hardware complications were the biggest contributing factor.

Authors’ Preferred Method It is our practice to manage pediatric patients with wide open physes with hamstring autografts placed in a transphyseal tibial location and over-the-top position on the femur. We use a more centrally placed tibial tunnel, typically 6 to 7 mm in diameter. We place only soft tissue across the physes, use suture and post fixation, and ensure that our post fixation is well away from the physes (Fig. 23D2-5).

A

In those patients with closing growth plates and less than 2-year growth remaining, we will typically use a hamstring autograft reconstruction utilizing transphyseal tibial and femoral tunnels. Fixation options include screw and post, EndoButton, staples, or interference screws away from the physes. We use our standard postoperative ACL rehabilitation protocol.

B

Figure 23D2-5   Postoperative anteroposterior (A) and lateral (B) radiographs of a transtibial or over-the-top femur anterior cruciate ligament reconstruction on a skeletally immature patient using staple fixation placed well away from the physes.

knee 1683

C

r i t i c a l

P

S U G G E S T E D

o i n t s

l The increased number of pediatric ACL injuries reflects the increased participation seen in youth sports. l Most injuries represent midsubstance ACL tears or tibial avulsion fractures. Femoral avulsion fractures of the ACL attachment are rare. The cut-off age for midsubstance versus avulsion fractures is about 12 to 14 years of age. l Physical examination of a pediatric patient with a suspected ACL injury should focus on ligamentous instability, patellar instability, and referred pain from the hip. Comparison to the contralateral extremity is critical to rule out ligamentous laxity or congenital absence of the ACL. l Meyers and McKeever described a classification system for avulsion fractures based on the elevation and rotation of the eminence fragment. Type I fractures have minimal or no displacement, and type II fractures have anterior hinging of one third to one half of the tibial eminence. Type IIIA fractures are completely displaced from the fracture bed, and type IIIB fractures demonstrate rotational malalignment. l Type I fractures can be managed with cast immobilization in 20 degrees of flexion. Type II fractures can be managed with cast immobilization if an anatomic reduction can be maintained. Type III fractures are generally treated ­operatively. l Treatment of pediatric midsubstance ACL tears is controversial. Nonoperative treatment, however, has led to recurrent instability, pain, and new meniscal and chondral injuries in a high percentage of patients. l Operative treatment of pediatric ACL tears, however, is also controversial. Options include extra-articular reconstructions, intra-articular reconstructions, and combined intra-articular and extra-articular reconstructions. No specific technique has demonstrated superiority. l Recently, the most popular techniques have included transphyseal tibial tunnels with an over-the-top femoral placement and transphyseal tibial and femoral tunnels with soft tissue grafts in patients nearing skeletal maturity.

R E A D I N G S

Aronowitz ER, Ganley TJ, Goode JR, et al: Anterior cruciate ligament reconstruction in adolescents with open physes. Am J Sports Med 28(2):168-175, 2000. Bales CP, Guettler JH, Moorman CT 3rd: Anterior cruciate ligament injuries in children with open physes: Evolving strategies of treatment. Am J Sports Med 32(8):1978-1985, 2004. Fehnel DJ, Johnson R: Anterior cruciate injuries in the skeletally immature athlete: A review of treatment outcomes. Sports Med 29(1):51-63, 2000. Graf BK, Lange RH, Fujisaki CK, et al: Anterior cruciate ligament tears in skeletally immature patients: Meniscal pathology at presentation and after attempted conservative treatment. Arthroscopy 8(2):229-233, 1992. Kocher MS, Foreman ES, Micheli LJ: Laxity and functional outcome after arthroscopic reduction and internal fixation of displaced tibial spine fractures in children. Arthroscopy 19(10):1085-1090, 2003. Kocher MS, Garg S, Micheli LJ: Physeal sparing reconstruction of the anterior cruciate ligament in skeletally immature prepubescent children and adolescents. J Bone Joint Surg Am 87(11):2371-2379, 2005. Kocher MS, Micheli LJ, Gerbino P, et al: Tibial eminence fractures in children: Prevalence of meniscal entrapment. Am J Sports Med 31(3):404-407, 2003. Kocher MS, Saxon HS, Hovis WD, Hawkins RJ: Management and complications of anterior cruciate ligament injuries in skeletally immature patients: Survey of the Herodicus Society and the ACL Study Group. J Pediatr Orthop 22(4): 452-457, 2002. Mah JY, Adili A, Otsuka NY, Ogilvie R: Follow-up study of arthroscopic reduction and fixation of type III tibial-eminence fractures. J Pediatr Orthop 18(4): 475-477, 1998. Meyers MH, McKeever FM: Fracture of the intercondylar eminence of the tibia. J Bone Joint Surg Am 52(8):1677-1684, 1970.

R e f erences Please see www.expertconsult.com

S e c t i o n

E

Posterior Cruciate Ligament Injuries 1. Posterior Cruciate Ligament Injuries in the Adult Nicholas J. Honkamp, Anil S. Ranawat, and Christopher D. Harner

The past decade has seen a renewed interest in the posterior cruciate ligament (PCL) as numerous studies have reported on its anatomy, biomechanics, and various reconstruction techniques. Despite this, our knowledge of the PCL still lags behind that of other knee ligamentous structures, particularly the anterior cruciate ligament (ACL). The more infrequent nature of PCL injury, lack of familiarity with its injury

and treatment, increased surgical risks, and poor results of early surgical treatment are all factors that have led to this disparity in knowledge. However, with advances in basic science knowledge pertaining to the PCL and improved awareness and imaging modalities, PCL injuries are being more accurately diagnosed and treated. This has led to more focus on the various treatment options and their results.

1684 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Likewise, multiligament injuries are also being more increasingly recognized, particularly injuries to the PCL and posterolateral corner (PLC). These injuries can occur from athletics but are more commonly related to trauma. Accurate recognition of all injured ligaments is critical because the prognoses of multiligament injuries are more than those of isolated PCL injuries. This chapter reviews the relevant anatomy, biomechanics, physical examination and radiographic diagnostics, treatment, and outcomes data on PCL injury.

23 mm

19 mm

Level of adductor tubercle

RELEVANT ANATOMY To accurately diagnose and treat PCL injuries, a thorough knowledge of the anatomy and biomechanics of the PCL is mandatory. Although easily visualized during arthroscopy, the PCL is technically an extra-articular structure. Synovium that reflects from the posterior capsule surrounds its anterior, medial, and lateral sides, whereas the posterior border of the PCL is intimately associated with the capsule and periosteum.1,2 The PCL averages in length between 32 and 38 mm and has a cross-sectional area at its midsubstance of 31.2 mm2, which is about 1.5 times larger than the ACL.1,3,4 Knowledge of the intra-articular length is critical for the selection of an appropriate graft in reconstruction. The femoral and tibial insertion sites are about 3 times larger than the cross-sectional area at the midsubstance level of the ligament.4,5 The fibers of the PCL attach in a lateral to medial direction on the tibia at a fovea about 1.0 to 1.5 cm below the joint line (Fig. 23E1-1). On the femoral side, the PCL fibers attach adjacent to the anterior cartilage margin of the medial femoral condyle in an anteroposterior direction on the femur (Figs. 23E1-2 and 23E1-3). There are three main components to the PCL: the anterolateral (AL) bundle, the posteromedial (PM) bundle, and the meniscofemoral ligaments.4,6 These components each have unique bony insertions as well as anatomic and biomechanical properties (Figs. 23E1-4 and 23E1-5).4 The AL bundle is about 2 times larger in cross-sectional area than the PM bundle, and its stiffness and ultimate strength are about 150% of the PM bundle.1,2,4 Tension in the two bundles varies depending on the degree of knee flexion. With the knee in extension, the PM bundle is aligned in a proximal to distal direction and is taut. The PM bundle

Anterior cruciate

32 mm 9 mm Figure 23E1-2   The broad complex origin of the posterior cruciate ligament is in the form of a semicircle on the medial femoral condyle.

slackens as the knee flexes, with its fibers passing between the medial femoral condyle sidewall and the AL bundle.4,7 With deep knee flexion, the PM bundle moves anterior and away from the tibial plateau such that they become taut again in deep knee flexion (Fig. 23E1-6).2,7 The AL bundle is slack in the extended knee and thus appears curved when viewed on magnetic resonance imaging (MRI) in the sagittal plane. With knee flexion, this bundle becomes taut.4,7 The varying width of the ligament, its complex insertional site anatomy, and its nonisometric tensioning patterns all combine to make isometric reconstruction of the ligament difficult.

Posterior cruciate ligament in extension

Posterior cruciate 13 mm

Figure 23E1-1   The posterior cruciate ligament inserts in a central fovea on the tibia about 1.5 cm below the joint line.

Figure 23E1-3   The fibers of the posterior cruciate ligament attach in an anterior-to-posterior direction on the femur and a medial-to-lateral direction on the tibia. (Redrawn from Girgis FG, Marshall JL, Monajem A: The cruciate ligaments of the knee joint: Anatomical, functional and experimental analysis. Clin Orthop 106:216-231, 1975.)

knee 1685

AL PM

AL PM

Figure 23E1-4   The anatomic position of the anterolateral (AL) and posteromedial (PM) bundles of the posterior cruciate ligament. (From Harner CD, Höher J: Current concepts: Evaluation and treatment of the posterior cruciate ligament injuries. Am J Sports Med 26:471-482, 1998.)

The anterior and posterior meniscofemoral ligaments (MFLs) of Humphrey and Wrisberg, respectively, are the third component of the PCL complex. These ligaments arise from the posterior horn of the lateral meniscus and sandwich the PCL anteriorly and posteriorly, respectively (Fig. 23E1-7). A recent review of the literature found that 93% of knees had at least one MFL and about half had both.8 Because their attachment is to the mobile lateral meniscus, it is possible for the PCL to be ruptured but the MFLs to be intact. Because of their anatomic locations, their relative strength, and their ability to resist posterior drawer forces, they may act as a splint to keep a ruptured PCL in position while it heals.7,9 Although not intimately associated anatomically with the PCL, the PLC structures of the knee are commonly injured together with the PCL. Thus, knowledge of their anatomy is relevant to the PCL. Various names have been attributed to structures comprising the PLC, greatly contributing to the confusion surrounding this area of the knee. Two recent cadaveric studies have helped define this anatomic area.10,11 The main constituents of the PLC are the popliteus ligament, the popliteofibular ligament (PFL), and the lateral collateral ligament (LCL).12-14 The arcuate and fabellofibular ligaments are variably present.15,16 The LCL and capsule are tight in full knee extension and slacken as the knee flexes. In contrast, the popliteofibular ligament complex (popliteus and PFL) is isometric and stabilizes the knee at all angles of flexion.17 These structures are ideally situated to resist tibial external rotation and, to a lesser degree, posterior tibial translation, whereas the LCL primarily restricts varus.12-14

Posteromedial bundle attaches to side wall of notch

Anterolateral bundle attaches to roof of notch

Artificial split between bundles

Posterior-oblique fibers

Posteromedial bundle overlays anterolateral Figure 23E1-5   Posterolateral view of the left knee showing double-bundle architecture of native posterior cruciate ligament. (From Amis AA, Gupte CM, Bull AMJ, Edwards A: Anatomy of the posterior cruciate ligament and the meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc 14[3]:257-263, 2006.)

1686 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 23E1-6   Sagittal cross section of the left knee. A, In full extension, the anterolateral bundle (dashed arrow) is slack, whereas the posteromedial bundle (solid arrow) is taut. B, In 90 degrees of flexion, the anterolateral bundle is taut, and the posteromedial bundle is slack. (From Amis AA, Gupte CM, Bull AMJ, Edwards A: Anatomy of the posterior cruciate ligament and the meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc 14[3]:257-263, 2006.)

Blood Supply and Innervation The vascular supply of the knee and the cruciate ligaments has been well described.18,19 The popliteal artery gives rise to five branches that supply blood to the knee joint: the superior and inferior geniculate arteries (both with medial and lateral branches) and the middle geniculate artery (Fig. 23E1-8). The middle geniculate artery penetrates the posterior capsule of the knee, providing the major blood supply to the cruciate ligaments, synovial membrane, and posterior capsule itself. Furthermore, the synovial sleeve covering the PCL is well vascularized and is a major contributor to the blood supply of the ligament.18,19 The distal portion of the PCL also receives a portion of its blood supply from capsular vessels originating from the inferior geniculate and popliteal arteries.

The PCL and its synovial sleeve are supplied by nerve fibers from the popliteal plexus. This plexus receives contributions from the posterior articular nerve (prominent branch of posterior tibial nerve) and from the terminal portions of the obturator nerve.20 Golgi tendon organlike structures have been observed near ligament origins beneath the synovial sheath and are thought to have a proprioceptive function in the knee.20,21 Katonis and associates22 have identified Ruffini’s corpuscles (pressure receptors), Vater-Pacini corpuscles (velocity receptors), and free nerve endings (pain receptors) in a histologic study of mechanoreceptors in the PCL. These studies indicate that disruption of the PCL alters not only the knee kinematics but also the afferent signals to the central nervous ­system.23

Anterior cruciate ligament Anterior meniscofemoral ligament (ligament of Humphry)

Posterior meniscofemoral ligament (ligament of Wrisberg)

Posterior horn of lateral meniscus

Posterior cruciate ligament

Figure 23E1-7   The meniscofemoral ligaments (Humphry’s and Wrisberg’s) arise from the posterior horn of the lateral meniscus and insert anteriorly and posteriorly to the posterior cruciate ligament, respectively, on the medial femoral condyle.

knee 1687

Popliteal artery

Superior medial genicular artery

Superior lateral genicular artery

Middle geniculate artery Inferior lateral genicular artery Inferior medial genicular artery

Figure 23E1-8   The vascular supply to the knee joint is through the superior and inferior medial and lateral geniculate arteries. The middle geniculate artery pierces the posterior capsule and is the main vascular supply to the cruciate ligaments.

Ligament Biomechanics The tensile strength of the PCL has varied across multiple investigators from 739 to 1627 N.4,24-26 This variation is explained partially by the fact that the individual PCL fibers act at different flexion angles and at different directions such that a uniaxial test of the whole ligament is not an accurate estimate of its strength. Thus, these numbers are likely underestimations.27 Additionally, age has been shown to be a significant factor in the strength of ligamentous tissue.28,29 As stated earlier, the PCL is composed of three functional components, the AL and PM bundles and the MFLs of Wrisberg and Humphrey.1 The AL bundle is considerably larger in cross-sectional area (about 43 mm2 versus 10 mm2) and strength (1620 N versus 258 N) than the PM bundle.30 The mean strength of the MFLs averages 300 N, roughly equivalent to the PM bundle (Table 23E1-1). The PLC structures have garnered increased attention during the past decade as their anatomy has been clearly defined and their intimate relationship with the PCL in    TABLE 23E1-1  Biomechanical Size and Strength of

the Three Major Components of the Posterior Cruciate Ligament Mean Strength Anterolateral bundle Posteromedial bundle Meniscofemoral ligaments

1620 N 258 N 300 N (each)

Cross-Sectional Area mm2

43 10 mm2 Variable

controlling knee movements has been elucidated. As the knee moves from extension to flexion, there is a coupled external tibial rotation owing to the greater mobility of the lateral compartment.31,32 Although the PCL functions as the dominant restraint to posterior tibial translation with increasing flexion, none of its three functional components are the primary restraint to posterior tibial translation at full extension or to external tibial rotation at any knee flexion angle.33 The importance of these secondary ­stabilizers, notably the PLC, in helping to control posterior tibial translation as well as functioning as the primary restraint to external tibial rotation has been a major research milestone during the past decade. Perhaps the most notable structure among the PLC is the popliteus complex. The popliteus complex consists of the popliteus muscle-tendon unit and the ligamentous connections from the tendon to the fibula, tibia, and meniscus, known as the popliteofibular ligament and popliteotibial and popliteomeniscal fascicles, respectively.10-15 As these structures cradle the posterolateral aspect of the knee, they tighten in full extension (providing the main restraint to posterior tibial translation at extension) and tibial external rotation at all flexion angles. The popliteus muscle itself provides both static and dynamic stability through its tendinous ­insertion into the lateral femoral condyle and its muscular pull, respectively.14,34,35 It is important to note that although the LCL is part of the PLC, the LCL functions independently of the PLC. The LCL provides varus stability and does not assist the PLC structures in preventing posterior tibial translation. Thus, varus and posterior or posterolateral instability should be evaluated separately.36,37 Finally, hamstring and quadriceps muscle actions provide dynamic stability to help reduce loads on the PCL.38,39 The synergistic relationship between the PCL and PLC in limiting posterior translation and external rotation is well established. Combined sectioning of both structures results in significantly increased laxity as compared with sectioning of either structure alone (Table 23E1-2).5,12,14 Studies have shown that isolated sectioning of the PLC or the PCL increased posterior tibial translation up to

   TABLE 23E1-2 Kinematic Changes in Response to

Isolated and Combined Injury of the Posterior Cruciate Ligament and Posterolateral Structures under a Posterior Drawer Test* Injury

PCL

PCL-PLS

PLS

Isolated

Posterior translation (90 degrees) External rotation (30 degrees) Varus

2+ 1+ 1+

0 1+ 1+

Combined (PCL and PLS)

Posterior translation (90 degrees) External rotation (30 degrees) Varus

3+ 2+ 2+

*Values indicate the amount of knee laxity as described by knee classification systems. PCL, posterior cruciate ligament; PLS, posterolateral structures. From Boynton MD, Tietjens BR: Long-term followup of the untreated isolated posterior cruciate ligament-deficient knee. Am J Sports Med 24:306-310, 1996.

1688 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

12 mm at full extension and 90 degrees of flexion, respectively. When both structures were sectioned, laxity at each position increased up to 25 mm.12,13 Similarly, the absence of one structure significantly increases the stress in the remaining structure. Vogrin and colleagues showed that the in situ force in the PCL increases 2 to 6 times that of the normal knee in response to an external rotation torque in the presence of PLC deficiency.40 The converse (increased stress seen in the PLC structures in the PCLdeficient knee in response to posterior drawer force) is also true. Thus, failure to recognize and reconstruct both structures in the setting of a combined PLC and PCL injury can result in persistent joint laxity and early failure when one structure is reconstructed.41,42 A biomechanical study by Sekiya and associates42 proved that by addressing both structures in a combined injury, more normal knee kinematics was restored. Because of its larger area and strength, the historical approach to PCL reconstruction has centered on replacing the larger anterolateral bundle. Multiple studies, however, have shown that this reconstructive technique fails to reproduce normal knee biomechanics.43-46 In single-­bundle PCL reconstructions, the anterolateral bundle is commonly fixed at 90 degrees of knee flexion, where the PCL is the primary restraint to posterior tibial ­translation. Doing so, however, only restores knee kinematics at middle to high flexion angles but leaves residual laxity at knee flexion angles near extension.47 Conversely, fixation of a singlebundle AL graft near full extension increases the likelihood of graft failure and loss of flexion due to overconstraining the knee joint.47,48 It has been shown that adjusting the insertion site of the femoral bundle more strongly ­influences graft tension and overall posterior tibial translation than adjusting the tibial insertion site.49-53 This is due to the rotational movement of the femoral condyle relative to the more fixed tibial insertion site during knee motion.52,53 Grood and colleagues52 and Sidles and associates53 have noted that the shallow-deep (proximal-distal) location of the tunnel in the femoral notch has the greatest effect on the tension in the PCL bundle grafts. The most ideal position for a single-bundle femoral tunnel remains unanswered. Multiple authors have looked for an isometric point, but few fibers of the native PCL are isometric.54 Efforts to place the graft in this location have not restored stability, particularly at higher flexion angles.50,55 Other authors have advocated a nonisometric femoral attachment located shallow (distal) within the femoral notch when viewed at 90 degrees of flexion.50,54,56 This has shown improved stability but still does not replicate the native PCL biomechanics.50 To more closely reproduce the anatomy and tensioning patterns of the intact PCL, double-bundle PCL reconstructions were developed. The results of these are discussed later.

the incidence in trauma or sporting activity cohorts is much higher. Fanelli and associates,58,59 in a prospective analysis on patients presenting to a regional trauma center with acute hemarthrosis of the knee, found that 56.5% of these patients were trauma victims, whereas 32.9% were sports related. In this population, isolated PCL ruptures were rare, with 96.5% of the PCL injuries involving multiligamentous injuries. In a more sports focused cohort, Schulz and colleagues60 found an athletic cause in 40% of their 587 patients with PCL insufficiency. Thus, motor vehicle crashes and sports-related trauma are the most commonly cited causes of PCL injury.58-63 The exact incidence and distribution of PCL injuries among specific sports is less well known. Sports involving contact, such as football, baseball, skiing, and soccer, are more frequently associated with PCL injury.60,64-68 The retrospective sports-specific incidence of PCL injury for hockey, soccer, handball, wrestling, and rugby has been found to range between 1% and 4%.64,66,69,70 Ligamentous disruption of the PCL, unlike injury to the ACL, is most often the result of external forces. The classic “dashboard” injury (i.e., extrinsic force) occurs from a posteriorly directed force on the anterior aspect of the proximal tibia with the knee in a flexed position (Fig. 23E1-9).71 Similarly, an athlete who falls on a flexed knee with the foot in plantar flexion is at risk for a PCL injury.43 In contrast, if the foot is in dorsiflexion, the force is transmitted more through the patella and distal femur, protecting the PCL from injury (Fig. 23E1-10). Noncontact injuries to the PCL have also been reported in the literature. The most common mechanism for isolated PCL injuries in athletes was forced hyperflexion of the knee as reported by Fowler and Messieh.65 These injuries often resulted in only partial tearing of the PCL, leaving the posteromedial fibers intact. Another mechanism is knee hyperextension, which is often combined with either a varus or valgus force resulting in a combined ligamentous injury with a much more guarded prognosis.

EVALUATION Clinical Presentation and History The incidence of PCL injuries varies significantly across different study populations. The reported incidence in the general population has been found to be 3%.57 However,

Figure 23E1-9   Posterior cruciate ligament injuries are most frequently the result of a blow to the front of the flexed knee.

knee 1689

B

A

Figure 23E1-10    A, Falling on a flexed knee with the foot in a dorsiflexed position spares injury to the posterior cruciate ligament (PCL) by transmitting the force to the patellofemoral joint. B, Landing with the foot plantar flexed injures the PCL as the posteriorly directed force is applied to the tibial tubercle. C, Hyperflexion of the knee without a direct blow is a common mechanism of PCL injury in athletes.

C

Injuries to the PCL can be classified according to severity (grade I to grade III), timing (acute versus chronic), and associated injuries (isolated versus combined). These variables have significant implications for outcomes of patients and thus are an important consideration in making treatment decisions. Isolated injuries to the PCL can be classified into partial (grade I or grade II) and complete (grade III) tears. In most cases, this is done clinically and corresponds to the laxity in the PCL, as measured by the step-off between the medial tibial plateau and the medial femoral condyle with the posterior drawer test. Stress radiography has also been show to corroborate these findings.72 Isolated grade III or complete PCL tears can occur, but they are frequently associated with other ligament injuries and, in particular, injury to the PLC.36 Distinguishing between the isolated and combined PCL injuries is critical because the prognosis for and treatment of these injuries are vastly different. Isolated injuries, in general, may be treated nonoperatively and have a good to excellent prognosis.65,67,73,74 Combined ligament injuries involving the PCL have a more guarded prognosis, however. Superior results may be possible in this group with early surgical intervention rather than with conservative treatment.36,63 Although the distinction between acute and chronic injury is somewhat arbitrary, it has important ramifications, especially in the treatment of combined ligamentous injuries. These cases frequently involve injury to the PLC, and surgical treatment in the acute phase (i.e., before 3 weeks) allows primary repair of the PLC. After 3 weeks, significant scar formation limits the success of primary repair and typically commits the surgeon to waiting a full 3 months to allow completion of the healing response before surgical reconstruction can be implemented. Furthermore, chronic injuries may become associated with significant pericapsular stretching (secondary to posterior and posterolateral tibial subluxation), resulting in secondary rotational instability or the development of arthrosis. In these cases, it

may be difficult to determine the extent of the initial injury as well as to devise the optimal treatment plan. Finally, combined PCL injuries are commonly associated with fractures as well as with injuries to vessels, nerves, and other soft tissue structures. An occult knee dislocation should be suspected when examination reveals disruption of both the ACL and the PCL or any three-­ligament injury. Although the integrity of the popliteal vessel and peroneal nerve must be assessed in any significant knee injury, their function must be particularly scrutinized in this situation because the incidence of injury ranges from 15% to 49%.36,71,75-79 If a knee dislocation is suspected, a low threshold for appropriate vascular studies and monitoring is recommended.

Physical Examination and Testing Patients with injuries to the PCL may present for evaluation in a variety of different scenarios. Injuries may range from a seemingly benign fall on the athletic field to severe trauma after a motor vehicle crash. Evaluation of the injured knee begins with obtaining a detailed history, trying to delineate the mechanism of injury, its severity, and possible associated injuries. The more acutely traumatized the knee, the more difficult it will be to examine. Unlike patients with isolated ACL injuries, those with acute, ­isolated PCL injuries do not typically report hearing or feeling a “pop.” Although many suspect a knee injury, patients do not typically relate a sense of instability. They may note mild to moderate swelling, accompanying stiffness, and occasionally mild knee pain. The pain may be located posteriorly, and they may lack 10 to 20 degrees of knee flexion.63 Conversely, patients with combined ligament injuries are rarely asymptomatic. After the initial swelling from the acute injury has resolved, individuals with PCL and PLC injuries may complain of pain and instability of the knee. This is especially true in patients

1690 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

with varus alignment, who may develop varus recurvatum thrust during gait.80 Dysesthesias or weakness of the foot dorsiflexors and evertors may also be present if an associated injury occurred to the peroneal nerve. Chronic injuries to the PCL and PLC can cause disability ranging from almost no functional impairments to severe limitations during activities of daily living.65,67,81-85 Furthermore, patients who are symptomatic frequently report pain as their predominant complaint.81-83,85-87 Other patients with chronic PCL tears, however, may complain more of disability than instability, with walking on inclines or ramps being most problematic.43 In general, patients with chronic, isolated injuries to either the PCL or PLC tend to function at higher levels than do patients with combined ligamentous injuries.81,88 The resultant posterior subluxation, increased adduction moment, and medial concentration of the joint reaction force may contribute to the development of arthrosis. Whether the severity corresponds to the degree of abnormal translation is still controversial.81,84 A thorough knee examination is essential and should follow the sequence of observation, evaluation of range of motion, palpation, and ligamentous examination followed by specialized testing. In the acute setting, contusions about the anterior tibia and popliteal ecchymosis may be noted. Assessment for meniscal damage or other ligamentous injury aside from the PCL and PLC should routinely be performed. Special care must be undertaken in evaluating the ACL in the setting of a PCL-insufficient knee. Although less frequent, the posteromedial corner (PMC) can also be injured, and its integrity should also be documented. The noninvolved knee must be examined first to determine the normal relationship of the tibia to the femur because the tibia will be subluxated posteriorly in the injured knee. After this is corrected in the injured knee, standard anterior drawer and Lachman’s tests can be performed. Significant translation of more than 10 mm in the sagittal plane suggests injury to both cruciate ligaments.36 Despite increased awareness of PCL, PLC, and PMC injuries, they are still frequently not recognized at the initial evaluation.

Posterior Drawer Test The most accurate clinical test for assessment of PCL integrity is the posterior drawer test (Fig. 23E1-11).43,61 The patient is placed supine, and the knee is flexed to 90 degrees while a posteriorly directed force is placed on the proximal tibia. This test can be performed with the tibia in neutral, external, and internal rotation. In cases of isolated PCL tears, there is a decrease in posterior tibial translation with internal tibial rotation. The MCL and posterior oblique ligament contribute to this secondary restraint.89,90 The extent of translation is evaluated by noting the change in the distance of step-off between the medial tibial plateau and the medial femoral condyle. Equally important during this test is to assess the quality of the end point. The plateau is normally positioned about 1 cm anterior to the condyle but can vary, making examination of the contralateral knee essential. PCL injury can be graded with respect to the amount of laxity determined by this test. Grade I is consistent with excessive posterior translation but maintenance of an

Figure 23E1-11   Assessing the tibial step-off before performing the posterior drawer examination. (Adapted from Miller MD, Harner CD, Koshiwaguchi S: Acute posterior cruciate ligament injuries. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery, vol 1. Baltimore, Williams & Wilkins, 1994.)

anterior step-off. Grade II is classified as a 5- to 10-mm translation corresponding to the plateau’s being displaced flush to the level of but not posterior to the condyle. Both of these grades represent partial tears of the PCL. More than 10 mm of translation constitutes a grade III injury, with the plateau displaced posterior to the condyle, and is consistent with a complete tear of the PCL. The degree of sagittal translation should also be assessed with the knee flexed 30 degrees. A slight increase in translation at 30 degrees and not at 90 degrees of flexion may indicate a PLC injury; increased sagittal translation at both 30 degrees and 90 degrees, with maximal translation at 90 degrees of knee flexion, is consistent with a PCL injury (see Table 23E1-2).

Posterior Sag Test (Godfrey’s Test) The posterior sag test may provide additional information in evaluation of the PCL. The hip and knee are flexed to 90 degrees. With a complete PCL tear, the pull of gravity will displace the tibia posterior to the femur while the examiner supports the weight of the limb by the foot (Fig. 23E1-12).

knee 1691

External Rotation of the Tibia (Dial Test)

Figure 23E1-12   A positive Godfrey’s test. (Adapted from Miller MD, Harner CD, Koshiwaguchi S: Acute posterior cruciate ligament injuries. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery, vol 1. Baltimore, Williams & Wilkins, 1994.)

Quadriceps Active Test The quadriceps active test can also aid in the diagnosis of complete ruptures. For this test, the knee is placed at 60 degrees of flexion. While the examiner holds pressure on the foot, the patient is asked to contract the quadriceps isometrically. In the presence of a complete tear of the PCL (grade III), the patient will achieve dynamic reduction of the posteriorly displaced tibia (Fig. 23E1-13).

Proper evaluation of the PLC can be difficult with the tibia subluxated posteriorly as in a grade III PCL injury, so reducing the tibia to neutral is essential before testing for PLC injury.61,91-93 Testing is best performed with the patient positioned prone or supine, while an external rotation force is applied to both feet with the knee positioned at 30 degrees and then 90 degrees of flexion. The degree of external rotation is measured by comparing the medial border of the foot with the axis of the femur. Because wide variability of external rotation is possible at these positions, it is essential to compare the results with the contralateral side.94,95 More than a 10-degree difference is considered abnormal.96 The popliteus complex portion of the PLC is the primary restraint to external rotation at all degrees of knee flexion, but its effect is maximal at 30 degrees. An increase of 10 degrees or more of external rotation at 30 degrees of knee flexion, but not at 90 degrees, is considered diagnostic of an isolated PLC injury.97 Conversely, the PCL is the secondary restraint to external rotation when the knee is at 90 degrees of flexion.12,52,95 Thus, an isolated PCL injury should have increased external rotation at only 90 degrees, whereas increased external rotation at both 30 and 90 degrees of knee flexion suggests a combined PCL and PLC injury (see Table 23E1-2). The recognition of this posterolateral instability component is essential clinically because it may significantly affect the treatment of associated ligamentous instability.

Posteromedial Pivot Test The integrity of the posteromedial corner can be tested using the PM pivot test.98 This test evaluates the integrity of the PCL, MCL, and posterior oblique ligament. A positive test result occurs when the knee shifts anteriorly as it is extended to about 20 degrees while a varus, compression, and internal rotation stress is applied to the tibia.

Figure 23E1-13   The quadriceps active drawer test. In the presence of a complete tear of the posterior cruciate ligament, dynamic reduction of the posteriorly displaced tibia will be achieved.

1692 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

varus opening at full extension, a combined injury of the PLC, PCL, and ACL may be present.12-14,95,96,100

Gait and Limb Alignment The evaluation of gait and limb alignment is particularly important for those with chronic injury of the PLC. Compared with the medial side of the knee, the articular anatomy of the lateral side is inherently less stable.31,32,94 In patients with chronic PLC insufficiency, the dynamic stabilizers of the lateral knee are not functioning optimally. This may lead to excessive posterolateral rotation and varus opening (or thrust) in the stance phase of gait.101 Therefore, it is critical to evaluate lower limb alignment in patients with chronic PLC instability. In the presence of varus alignment and a lateral thrust, a valgus (opening or closing) proximal tibial osteotomy is necessary to correct the alignment because soft tissue reconstructions alone will fail.102,103 Similarly, a patient with chronic PCL deficiency and a primary varus alignment will be more susceptible to posterolateral injury. Thus, to plan successful treatment of the patient with multiple ligamentous knee injuries, the examiner must determine the extent of injury to all ligaments involved, namely, the PCL, the PMC, the popliteus complex, and the LCL, as well as the primary alignment of the limb.

Imaging Studies Radiography Figure 23E1-14   The reverse pivot shift test starts with the knee in flexion. The posteriorly subluxed lateral tibial plateau reduces as the knee is brought into extension. (Adapted from Miller MD, Harner CD, Koshiwaguchi S: Acute posterior cruciate ligament injuries. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery, vol 1. Baltimore, Williams & Wilkins, 1994.)

Reverse Pivot Shift Test This test can be performed by passively extending the knee from 90 degrees of flexion with the foot externally rotated and a valgus force applied to the tibia. A positive result is observed when the posteriorly subluxated lateral tibial plateau abruptly reduces at 20 to 30 degrees of flexion (Fig. 23E1-14).99 Other tests, such as the external rotation recurvatum, the posterolateral drawer, and the posterolateral Lachman’s test, can further aid in diagnosis of the extent of injury to the PCL and PLC but are more difficult to perform reproducibly and to interpret.

Collateral Ligament Examination Varus and valgus stress tests are important in assessment of the integrity of the LCL portion of the PLC. These should be performed with the knee in full extension and in 30 degrees of flexion. Isolated PCL injury does not significantly affect varus or valgus stability. Increased varus opening at 30 degrees of knee flexion indicates an injury to the LCL and possibly the popliteus complex. Additional slight increased opening at full extension is consistent with injuries to both of these structures. If there is a large degree of

Our standard knee series includes bilateral standing anteroposterior and flexion 45 degrees weight-bearing, as well as Merchant’s patellar and lateral radiographs. These films should be carefully scrutinized for subtle posterior tibial subluxation (which may be the only radiographic finding in isolated PCL injuries), avulsion fractures, joint space narrowing, and slope of the proximal tibia. The plain radiographs must also be closely evaluated for any evidence of avulsion fractures involving the PCL, fibular head, and Gerdy’s tubercle. These bony avulsion injuries, when recognized acutely, may be repaired primarily with superior results compared with late reconstruction.104 Hall and Hochman105 described a medial Segond’s fracture that represents a medial capsular avulsion in PCL injuries. Furthermore, failure to recognize fractures of the tibial tubercle can be a particular problem. The unopposed pull of the hamstrings causes posterior tibial subluxation, which can become fixed within a short time, requiring open reduction. In addition, long-leg cassette views are critical to evaluate overall lower extremity alignment, particularly varus, in chronic or revision cases. Stress radiographs and contralateral views, although not routine, may be helpful in some situations.106 Hewett and associates106 retrospectively evaluated 21 patients with partial or complete PCL tears with both stress radiology and KT-1000 measurements. They concluded that stress radiographs were more accurate than KT-1000 measurements in diagnosing PCL tears. Most have concluded that greater than 8 mm posterior displacement on stress radiographs indicates a complete PCL tear, whereas greater than 12 mm indicates injury to secondary structures.72,106,107

knee 1693

We agree with Schulz and colleagues,72,107 however, that stress radiography may be influenced by multiple factors, including tibial rotation and tissue compliance, such that physical examination by an experienced clinician is as ­sensitive as stress radiography.

Magnetic Resonance Imaging MRI has become the diagnostic study of choice in evaluation of the knee with a presumed PCL injury. This study is 96% to 100% sensitive in detecting acute tears of the PCL and can also determine the precise location of the tear, with implications for treatment.95,108-110 For example, the femoral “peel-off” injury is particularly amenable to primary repair.111,112 The normal PCL appears dark on T1- and T2-weighted sequences and is curvilinear in appearance. However, chronic tears of the PCL can heal and assume the aforementioned curvilinear appearance; thus, MRIs for chronic PCL tears are much less sensitive, and the appearance of a normal shape of the ligament should not be used as a criterion for a normal PCL.63,113,114 There is some data to support the use of MRI as a prognostic sign of PCL healing.114 MRI is also necessary to assess the menisci, articular surfaces, and other ligaments of the knee, which also have relevance to treatment and prognosis.115 Bone bruises are common in PCL injuries; however, their location can vary within the knee, unlike the uniform appearance of ACL bone bruises.116 MRI of the PLC of the knee has consistently improved. With the help of different imaging techniques and the addition of a coronal oblique technique, the sensitivity of visualizing and accurately diagnosing injuries to the PLC can exceed 80%.117,118

Bone Scanning A bone scan may prove helpful in the evaluation and management of chronic PCL injuries. Patients with these injuries are predisposed to early medial and patellofemoral compartment chondrosis.43,73,119 In the setting of an isolated PCL-deficient knee with medial or patellofemoral compartment pain and normal radiographs, a bone scan to assess these compartments may be helpful. If there is increased uptake, surgical intervention may be beneficial.87 If there is no increased uptake, a continued nonoperative approach is our treatment of choice.

TREATMENT OPTIONS Controversy still exists with respect to the indications for nonoperative versus surgical intervention, techniques of reconstruction, and methods of rehabilitation for the PCL-injured knee. The relatively infrequent occurrence of this injury has unfortunately led to clinical studies with small sample sizes and short-term follow-up. The limited understanding of the PCL and associated injuries has additionally resulted in studies that are frequently a collection of differing patterns of PCL injury—acute, chronic, isolated, combined, partial, and complete—and also lack welldefined indications for surgical management. Most series do not include control groups, combine primary repair and multiple reconstructive techniques, and use different

outcome measures. Because of this, it has been difficult to compare results of different operative techniques and approaches. The following section focuses on a review of the ­pertinent literature followed by a discussion of current treatment recommendations. We conclude with a ­presentation of our approach to the management of the PCL-injured knee.

Review of Nonoperative Treatment and Natural History of Isolated Posterior Cruciate Ligament Deficiency Among orthopaedic surgeons, the treatment of isolated PCL injuries continues to be an area of active debate. This debate will likely continue until prospective, randomized trials are done to compare different methods of treatment. Currently, most studies are retrospective in nature and use various outcome measurements, which makes comparisons difficult. To date, there have been four published prospective studies on nonoperative management for isolated PCL injuries (Table 23E1-3). Many authors have found good results with nonoperative treatment of isolated PCL injuries that would indicate a more benign natural history. Dandy and Pusey83 treated 20 patients with persistent knee symptoms due to unrecognized, isolated PCL injury. At 7.2 years of follow-up, 18 of the 20 had a good functional result. The authors thought that the subjective results, objective evaluation findings, and functional capacity without operative intervention were adequate and did not warrant surgical reconstruction or repair. Parolie and Bergfeld67 observed 25 patients with isolated PCL tears resulting from sporting injuries. At a minimum of 2 years of follow-up, 80% were satisfied with their knee function, and 68% returned to their previous level of activity. They also noted that those with unsatisfactory results tended to have diminished quadriceps strength compared with those with successful results. Torg and colleagues87 observed 14 patients with isolated PCL injuries and 29 patients with combined PCL injuries for more than 6 years and concluded that the patients with isolated injuries remained without symptoms and did not require subsequent reconstruction. Other authors have found similar outcomes, but did find some deterioration in outcome with time. Boynton and Tietjens120 observed 30 patients with isolated PCL­deficient knees for an average of 13.2 years and found that the prognosis varies. Eight (21%) of the original 38 patients had surgery for subsequent meniscal tears. Among the 30 patients with isolated PCL-deficient knees with normal menisci, 24 (81%) had at least occasional pain, and 17 (56%) had at least occasional swelling. Keller and colleagues73 observed 40 patients with isolated PCL injuries for an average of 6 years. In this shorter interval, 90% still complained of pain, 65% stated that the knee limited their activity, and 43% reported difficulty with walking. Four prospective studies have been reported on the natural history of isolated PCL injuries. In 1987, Fowler and Messieh65 published their results on 13 patients followed for a mean of 2.6 years. Subjective and functional ratings were all good; when assessed objectively, however, only 3 rated good and 10 fair. Shino and associates68 ­followed

Study

Design

No. of Patients

Laxity

Follow-Up Outcome Age (yr) (yr) Assessment

Functional Score

Shelbourne Pro et al, 1999121

68

Grade I, 25; grade I.5, 13; grade II, 30

25.2

5.4

83.4 (Lysholm) 84.2 (Noyes) 5.7 (Tegner)

Boynton & Tietjens, 1996120

Retro

30

28.7

13.2

N/A

76/100 (used own scale)

Torg et al, 198687 Keller et al, 199373

Retro

14

Grade I, 5; grade II, 15; grade III, 10 N/A

29

5.7

N/A

Retro

40

33

6

N/A

Shino et al, 199568

Pro

13

Grade I, 24; grade II, 13; grade III, 3 All grade II or III

N/A

4.25

N/A

5 Excellent; 7 good; 1 fair; 1 poor Noyes grade I average, 81; grade II average, 72; grade III, 58 IKDC: 3 normal, 5 nearly normal, 5 abnormal

Fowler & Messieh, 198765

Pro

13

N/A

22

2.6

Hughston criteria: 3 good, 0 fair, 10 poor

Dandy & Pusey, 198283

Retro

20

N/A

31

7.2

Parolie & Bergfeld, 198667

Retro

25

N/A

24.7

6.2

Hughston criteria: 0 good, 8 fair, 12 poor 80% satisfied (scale not specified)

Correlation to DJD

Correlation to Functional Score

Comments

No correlation with laxity, time Positive correlation with quadriceps strength and improved scores Positive Trend (P = .08) correlation for time, positive with time correlation with laxity (P = .037) and decreased scores N/A No correlation with laxity Positive Positive correlation correlation with between laxity and time time to decreased scores No correlation No correlation with time

50% same level, 63/68 had same or 33% lower less PCL laxity at level, 17% follow-up changed sports

Hughston criteria: 13 good

N/A

N/A

All returned to “previous activity”

Hughston criteria: 5 good, 11 fair, 4 poor

No correlation

N/A

N/A

N/A

No correlation

Positive correlation between improved scores and quadriceps strength

68% same level, N/A 16% lower level, 16% no return to sport

DJD, degenerative joint disease; IKDC, International Knee Documentation Committee; Pro, prospective; Retro, retrospective.

Positive trend with time (P = .07)

Return to Sport

26% same level, Prognosis 37% lower extremely varied; level, 37% 8 had meniscal changed sports injuries N/A N/A

11/13 returned to same level

36/40 had knee pain, 26/40 had decreased activity level All 22 original subjects underwent arthroscopy: 5 had lateral meniscus tears, 11 had chondral damage All 13 subjects underwent arthroscopy: 2 had meniscal tears, 4 had chondral lesions 18/20 had good functional result

1694 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 23E1-3  Natural History of Isolated Posterior Cruciate Ligament Injuries Treated Nonoperatively

knee 1695

15 patients for an average of 4.25 years. Eleven of 15 returned to sports at their preinjury level, and 14 patients remained symptom free. Shelbourne and colleagues121 followed 133 patients with isolated PCL injuries for an average of 5.4 years. More than half of these patients were able to return to their preinjury level of activity, and no correlation was found between grade of laxity and radiographic change. Shelbourne and Muthukaruppan122 performed a longer term follow-up of this patient cohort with a modified Noyes survey and the International Knee Documentation Committee (IKDC) subjective knee survey at a mean time of 7.8 and 8.8 years, respectively. Of the 146 patients who had at least four modified Noyes subjective surveys, 40% scored consistently excellent, 10% consistently good, 6% consistently fair, and 2% consistently poor. Furthermore, 16% of these patients had consistently improving scores, 12% had consistently decreasing scores, and 14% had inconsistent scores. Of the 67 patients who scored less than 85 points in the first 2 years after injury, only 34 had a score of less than 85 points at their most recent survey. In summary, numerous investigators have shown that isolated acute PCL injuries (grade I to grade III) do ­relatively well with conservative treatment. Most patients in these studies, however, had grade II PCL laxity or less. The relatively benign course of these injuries is most likely due to the integrity of the secondary restraints and various portions of the PCL remaining intact.36 Accordingly, Fontboté and associates123 have shown that patients with grade II PCL laxity demonstrate minimal biomechanical and neuromuscular differences despite significant clinical laxity. Although these studies varied with respect to injury mechanism, extent of the PCL injury, physical therapy protocol, and evaluation of outcome and time of follow-up, they do suggest that most patients with PCL-injured knees will do relatively well with conservative care. Even though nonoperative treatment appears to give relatively good results, these patients do not have a normal outcome in all series reported, especially for grade III injuries. The natural history of the PCL-deficient knee leads to increased contact pressures in both the medial and patellofemoral compartments.124-126 The pathomechanics of the PCL-injured knee is unique in comparison to other ligamentous knee injuries (Fig. 23E1-15). As a result of the excessive posterior tibial translation, abnormal wear occurs, and pain rather than instability becomes the major symptomatic issue. We agree with Logan and associates126 that a PCL injury is analogous to a medial meniscus resection in that the PCL injury leads to a fixed anterior subluxation of the medial femoral condyle, which overloads the medial knee compartment. Patients with PCL-deficient knees treated nonoperatively have a high incidence of acute chondral injury and late chondrosis43,73,119,120 (involving the medial femoral condyle and patellofemoral joint) as well as acute and chronic meniscal tears.68,120,127 It has been shown that acutely, 52% of 61 acute PCL-injured patients had chondral injury noted at arthroscopy,128 whereas patients with chronic PCL deficient knees have a variable progression of articular degeneration and symptoms over time. Geissler and Whipple127 studied groups of both acute and chronic PCL-deficient patients. In the acute group,

2) Tibiofemoral contact shifts anteriorly

X

• Posterior horn medial meniscus unloaded • Increase wear of articular cartilage

3) Force in the PLS

1) Posterior tibial translation Figure 23E1-15   Pathomechanics of articular wear secondary to altered tibiofemoral contact forces in the chronically posterior cruciate ligament-deficient knee. PLS, posterior lateral structures.

12% had ­ chondral defects, and 27% had meniscal tears. In the chronic group, 49% had chondral defects (most commonly medial) and 36% had meniscal tears (most commonly medial). These findings are likely to be the result of increased contact pressure that occurs after PCL ­disruption.124-126 The problem is that investigators have been unable to consistently identify prognostic factors to help predict outcomes of patients. Surprisingly, objective instability and time from injury have correlated poorly with final outcome and radiographic degenerative changes in most studies. Therefore, we, like Shelbourne and associates, believe that in patients with isolated grade II laxity of the PCL, a PCL reconstruction that improves laxity only to grade I may not achieve any better result than nonoperative treatment.122,129 However, in order to prevent the progressive decline associated with the natural history of many grade III injuries, we have adopted a more aggressive approach for these injury ­patterns.

Review of the Literature on Operative Treatment Avulsion Fractures Involving the Posterior Cruciate Ligament Avulsion fractures of the PCL are relatively rare injuries. When isolated and nondisplaced, these fractures have been effectively managed with a brief period of immobilization. Most would agree, however, that displaced avulsions should undergo operative management.74,130-132 Although several reports do not differentiate between isolated and combined injury patterns, surgical management of tibial avulsion fractures has been fairly successful. In independent studies, both Lee133 and McMaster74 were the first to report good results with this type of treatment. After treating 13 avulsion fractures, Trickey132 noted better results in the surgical compared with the nonsurgical group. Torisu130 treated 21 patients with tibial avulsion fractures with cast immobilization for nondisplaced fractures and early internal fixation for displaced fractures. At an average of 4 years of follow-up, all of the patients had good or excellent results.

1696 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Isolated Posterior Cruciate Ligament Injuries As shown previously, there is an emerging consensus that isolated grade III PCL injuries are not as benign in the long term as previously thought.88 Therefore, many surgeons have elected to proceed with PCL reconstruction in patients with isolated grade III (tibia subluxated posterior to femoral condyles, >10 mm displacement) injuries. In most of the studies published on these patients,42,71,134-141 the indications for treatment are persistent knee instability with continued knee pain. The initial results of contemporary reconstruction of the PCL were first described by Clancy and associates43 in 1983. Since that time, multiple studies have been published on the results of PCL reconstruction. This section focuses on those studies involving isolated PCL injuries that underwent reconstruction using contemporary techniques. There are multiple series published on PCL reconstruction done on patients with isolated high-grade (majority grade III) PCL injuries (Table 23E1-4).43,134-143 Several important points are notable from these studies. In ­studies by Sekiya and associates,142 Wang and coworkers,141 and Mariani and colleagues,139 acute reconstructions had significantly better outcome measures than chronic ­reconstructions, whereas the study by Deehan and associates136 did not find such a correlation. This conclusion may relate to more articular degeneration that is present in patients with chronic posterior tibial subluxation, which effectively unloads the medial meniscus, as mentioned earlier. Second, no specific graft type has shown superiority in PCL reconstructions135,140,143,144; both autograft and allograft hamstring, patellar tendon, Achilles tendon, and quad tendon have all been used. Third, most patients continued to have residual posterior laxity when compared with their contralateral leg, with many patients improving only one grade in laxity. The somewhat inconsistent results and residual laxity noted after arthroscopic single-bundle PCL reconstruction have been attributed to several potential technical issues. These issues include the use of a transtibial versus tibial inlay method, single- versus double-bundle reconstruction, and the various graft fixation options. Most PCL reconstruction techniques employ a transtibial technique. In posterior cruciate reconstructions using a transtibial technique, the PCL graft must make an acute turn at the posterior opening of the tibial tunnel. This so-called killer turn has been suspected of leading to graft abrasion with subsequent thinning of the graft and eventual graft rupture or excessive laxity.145-147 Thus, the residual posterior knee laxity observed clinically after traditional transtibial PCL reconstruction techniques may be related to this acute turn. Therefore, the tibial inlay technique was introduced by Jakob and Ruegsegger148 and by Berg149 to overcome this perceived disadvantage. The main attraction of the tibial inlay technique is the direct fixation that can be achieved at the tibial attachment site, as well as obviating the killer turn and allowing graft tendon length adjustment. Biomechanical cadaveric studies have been performed to test this hypothesis. Bergfeld and associates150 showed that the inlay technique resulted in less posterior tibial translation with less graft degradation than did the ­ transtibial

technique. This increased rate of graft thinning and degradation was also seen in a later cyclic loading study by Markolf and associates.151 However, the authors of the later subsequent studies152-155 did not find any significant increase in the in situ graft forces, graft laxity, or graft rupture in comparing either technique. Margheritini and coworkers153 argued that the evaluation of cyclic wear in the nonviable grafts used during these cadaveric studies does not account for the biologic remodeling that occurs in vivo, and thus makes such findings clinically less applicable. Retrospective studies comparing transtibial versus tibial inlay138,140 patients did not show any significant differences in subjective outcome or knee laxity measurements (Table 23E1-5). Thus, although the tibial inlay technique may have some biomechanical advantages when tested in a cadaveric model, these advantages have yet to be identified in clinical studies. Another controversy surrounding PCL reconstruction is the debate between single- and double-bundle techniques. Double-bundle PCL reconstructions were introduced to more closely reproduce the anatomy and biomechanical properties of the intact PCL. Biomechanical studies in our laboratory and others have shown that the two bundles demonstrate some reciprocal tightening during knee range of motion.156,157 Additionally, both bundles are active in reducing posterior tibial translation and external tibial rotation, reinforcing the notion that both are required for normal knee kinematics.157 Multiple biomechanical ­studies have been performed comparing single- versus doublebundle PCL reconstruction (Table 23E1-6).45,46,54,158,159 Four of the five studies have shown that double-­bundle PCL reconstructions achieve similar or improved knee biomechanics when compared with single-bundle reconstructions.45,46,54,159 In reviewing these studies, as well as single-bundle biomechanical studies,46,48,50,55,56 a few important points are worth noting. Graft fixation and tensioning patterns differ across studies, which may affect results. Also, despite efforts at replicating the anatomic femoral insertion sites of the PCL bundles, considerable variation exists in the placement of tunnels at these sites across the four studies. The deep-shallow (proximal-distal) attachment location on the femur appears to greatly determine the flexion angle where the graft will be functional. More shallow grafts tense as the knee is flexed, whereas deeper grafts slacken as the knee is flexed. Double-bundle grafts with both grafts placed in deeper (proximal) locations within the notch risk loss of stability in deeper knee flexion angles. A combination of shallow and deep locations produces a reciprocal tightening pattern across the full range of motion. Two clinical studies have compared the outcomes of single- versus double-bundle PCL reconstruction (see Table 23E1-6). Houe and Jorgensen160 used a patellar tendon graft with one femoral tunnel versus a hamstring graft using two femoral tunnels. They evaluated 16 patients at a mean follow-up of 35 months and found no significant differences in Lysholm score, activity level, or graft laxity between the two reconstruction types. Wang and associates161 found similar results in their 35 patients using hamstring grafts in either a single- or double-­bundle configuration. They found no significant difference in ligament laxity, functional score, or radiographic changes between

   TABLE 23E1-4  Results of Isolated Posterior Cruciate Ligament Reconstructions Study

No. of Patients Age (yr)

Followup (yr)

Sekiya et al, 2005142 Retrospective

21

38

Chan et al, 2006134 Prospective

20

29

Chen et al, 2002135 Retrospective

Graft Type

5.9

5 acute, 16 chronic; all grade III

Achilles allograft

3.3

Unknown; average Quadrupled time 4 mo; all hamstring grade III autograft

A (quad A: 29; tendon): 22; B:27 B (hamstring tendon): 27

A: 2.5; B: 2.2

Mariani et al, 1997139 Retrospective

24

26

2.2

Jung et al, 2004137 Retrospective

12

29

Wang et al, 2003141 Retrospective

30

Deehan et al, 2003136 Prospective

27

Ahn et al, 2005143 Retrospective

Group I Group (hamstring I: 30 autograft): Group 18; group II: 31 II (Achilles allograft): 18

Surgical Tech

Fixation

Subjective ­ utcome O

Transtibial

Tibia: screw and IKDC knee washer; femur: function: 57% metal IS N/NN, 43% A/SA; IKDC activity level: 62% N/NN, 38% A/SA

Transtibial

Tibia: bio IS + screw and washer; femur: bio IS + washer

A:12 acute, 10 A: quad Transtibial chronic; B: 16 tendon acute, 11 chronic; autograft; all grade III B: quadrupled hamstring autograft All chronic Patellar Transtibial tendon autograft

Tibia: A: suture and post; B: screw and washer; femur: A: metal IS; B: screw and washer Tibia and femur: metal IS

4.3

Unknown; average Patellar time, 5.4 mo; tendon range, 1-10 mo autograft

Tibial inlay

Femur: IS; tibia: screw and washer

32

3.3

13 acute, 17 chronic; all grade III

Mixed

Transtibial

Femur: IS; tibia: screw and post

27

3.3

All chronic (16 patients 4-12 mo after injury, 11 patients > 1 yr); grade II and III injuries All chronic; group I: 11 grade II, 7 grade III; group II, 10 grade II, 8 grade III

Hamstring autograft

Transtibial

Femur: IS; tibia: IS

Hamstring autograft and Achilles allograft

Transtibial

Femur: IS and screw and washer; tibia: IS and screw and washer

Group I: 2.9 Group II: 2.3

*Instrumented laxity as determined by stress radiographs. A, abnormal; bio, biologic; IS, interference screw; KT, KT-1000 testing; N, normal; NN, nearly normal, SA, severely abnormal.

Instrumented Laxity*

Posterior Drawer

Miscellaneous

KT posterior IKDC Acute/subacute drawer: 4.5 mm, grading group had KT side-side acute/ significantly difference: subacute: better IKDC and 1.96 mm 75% KT-1000 than N/NN; chronic group chronic: 40% N/NN Lysholm: 93; Average Grade I, 16; 18/20 showed Tegner: 6.3; postoperative grade II, 3; no radiographic IKDC activity KT posterior grade III, 1 deterioration; level: 85% drawer: 3.8 mm 3 patients had N/NN stiffness Lysholm: A: Average N/A 1 patient in 90.63; B: 91.44; postoperative each group had IKDC overall KT posterior stiffness; 86% rating: A: 18 drawer: group A and N/NN, 4 A/SA; A: 3.72 mm, 92% group B B: 22 N/NN, 5 B: 4.11 mm had normal A/SA radiographs Lysholm: 94; KT side-side N/A Significant differences: 6, Tegner: 5.4; correlation mm; 13, IKDC rating: 19 0-2 between poor 3.5 mm; 3, 6-10 N/NN, 5 A/SA results and more mm; 2, >10 mm chronic injuries OAK score: 92.5; Stress x-rays: 3.4 N/A 7 excellent, 4 mm side-side good; IKDC: difference; 11/11 were KT side-side N/NN difference, 1.8 mm Lysholm: 92 (24 N/A Grade I: 16; Significant excellent/good, grade II: 12; correlation 6 fair/poor); grade III: 3 between poor Tegner: 4.5 results and more chronic injuries Lysholm: 94; KT side-side Grade 0/I, No correlation IKDC: 25 N/ difference 23; grade between time NN, 2 A/SA < 2mm: 17 II, 1 from injury to patients; 3-4 surgery and mm, 6 patients outcome

Lysholm group I: Telos stress 90; group 2, 85; radiograph IKDC group I: posterior 16 N/NN, 2 displacement: A; group 2: 14 group I, 2.2 N/NN, 3 A, mm; group II, 1 SA 2.0 mm

N/A

No difference in outcome between the two groups

knee 1697

Chronicity/Grade

Study

No. of ­Patients

Age (yr)

Follow-up (yr)

Chronicity/ Grade

Graft Type

Seon & Song, Group A A: 29.1; 2006140 (transtibial): B: 29.4 Retrospective 21; group B (tibial inlay): 22

A: 2.6; B: 3.0

All chronic; all grade II or greater

MacGillivray, Group I 2006138 (transtibial):13 Retrospective Group II (tibial inlay):7

Group I: 6.3; group II: 4.8

All chronic; Mixed group I: 5 grade II, 8 grade III; group II: 3 grade II, 4 grade III

Group I: 29; group II: 31

A: hamstring autograft; B: patellar tendon autograft

Surgical Technique

Fixation

Subjective Outcome

Instrumented Laxity*

Posterior Drawer

Miscellaneous

A: transtibial; Tibia: A: bio Lysholm: A: Telos A: 19 normal/ No significant B: tibial IS; B: screw 91.3; B: 92.8; side-side grade I; differences inlay and washer; Tegner: A: difference: 2 grade II; B: 20 between the femur: A: 5.6; B: 6.1 A: 3.7 mm; normal/grade I; two anchor B: 3.3 mm 2 grade II screw; B: bio IS screw Group I: Tibia: I: IS; Lysholm: KT posterior Group I: Neither method transtibial; II: screw/ group I: 81; drawer: 3 grade I, restored group II: washer group II: group I: 6 grade II, anteroposterior tibial inlay Femur: 76; Tegner: 5.9 mm; 4 grade III; stability to the I: IS, II: IS group I: 6; group II: group II: knee group II: 6 5.5 mm 3 grade I, 3 grade II, 1 grade III

*Instrumented laxity as determined by stress radiographs. A, abnormal; bio, biologic; IS, interference screw; KT, KT-1000 testing; N, normal; NN, nearly normal, SA, severely abnormal.

1698 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

   TABLE 23E1-5  Results of Isolated Posterior Cruciate Ligament Reconstruction: Transtibial versus Inlay

knee 1699

   TABLE 23E1-6  Biomechanical Results of Single-Bundle versus Double-Bundle Posterior Cruciate Ligament Reconstructions Study

Graft Type and Configuration

Manneo et al, Patellar tendon; Y type (2 200054 tibial ends, 1 femoral end)

Graft Locations

Tensioning

Single-bundle All graft fixed at set reconstruction: proximal flexion angle to restore shallow (S1), distal normal posterior tibial shallow (S2), distal deep translation (D); double-bundle reconstruction: S1 and S2, S1 and D

Results S1 and S2 restored laxity within 2 mm of normal. D did not control laxity above 45 degrees of flexion. Both S1/S2 and S1/D double-bundle reconstructions controlled laxity with both bundles of S1/S2 configuration tight in flexion and the bundles in the S1/D configuration tight in a reciprocal fashion

Race & Amis, Patellar tendon; Y type, 18 Single-bundle 199846 mm graft for double-bundle reconstruction: isometric grafts; single 10-mm graft single bundle, single AL for single bundle bundle; double-bundle reconstruction: anatomic AL and PM bundles

All grafts fixed at 60 degrees of flexion

Isometric single bundle graft overconstrained knee in extension and underconstrained in flexion. AL singlebundle graft lax from 90 to 130 degrees of flexion. Only double-bundle graft restored normal knee laxity

Harner et al, 200045

Increased posterior tibial translation at all flexion angles in single-bundle reconstruction, with in situ forces up to 44 N lower than intact ligament. No difference in posterior translation or in situ forces in double-bundle reconstructions compared with intact knee

10-mm Achilles for AL bundle; 7-8 mm doubled semitendinosus for PM bundle; one or two femoral tunnels, one tibial tunnel

Single-bundle reconstruction: AL bundle; double-bundle reconstruction: anatomic AL and PM bundles

AL graft fixed with 134-N posterior load at 60 degrees of flexion. PM graft fixed at full extension

Bergfeld et al, Half Achilles for AL bundle; 2005158 half Achilles split into Y type graft for double bundle; one or two femoral tunnels, tibial inlay

Single-bundle reconstruction: AL bundle; double-bundle reconstruction: anatomic AL and PM bundles

AL graft fixed with Both the single- and double-bundle 40-N anterior drawer techniques closely reproduced the force at 90 degrees of stability as compared with the intact flexion. For doubleknee. Trend for double-bundle construct bundle reconstruction, to overtighten knee from 30 to 60 AL and PM bundles degrees, and for single-bundle construct tensioned at 90 and to allow more translation between 10 and 30 degrees, respectively 90 degrees

Markolf et al, Y-shaped patellar tendon Single-bundle 2006159 graft; 11-mm AL and 8-mm reconstruction: AL PM bundles bundle; double-bundle reconstruction: PM bundles with wide and narrow bone bridge between the AL and PM bundles

AL graft tensioned at Double-bundle graft with PM tunnel with 90 degrees of flexion to wide bone bridge that is tensioned to restore laxity of intact 10 N best restored knee laxity to that of knee within ±1 mm. the normal knee at the expense of slightly Each different PM increased graft forces bundle tensioned at 30 degrees of flexion with either 10 or 30 N

AL, anterolateral; PM, posteromedial.

the two groups at a minimum of 2 years’ follow-up. An additional study by Nyland and coworkers162 found that, in patients with complete PCL deficiency and grade I and II posterolateral corner instability, a double-bundle PCL reconstruction alone restored knee stability better than a single-bundle reconstruction. In a similar effort to maintain the normal double-­bundle anatomy of the PCL, an augmentation procedure can be performed when only one bundle of the PCL is injured. In our experience, this can occur in up to one third of our acute or chronic PCL cases. Often the posteromedial bundle of the PCL remains relatively intact, whereas the anterolateral bundle is torn. In this case, we perform a single-bundle augmentation of the AL bundle. Various techniques of augmentation, typically with retainment of the PM bundle and reconstruction of the AL bundle, have been described elsewhere with good results.163-166 In addition, there is evidence that the retained intact ligament may help avoid graft abrasion at the killer turn in transtibial techniques, and the original fibers and augmented fibers may heal together, forming one ligament.167

Another potential source of laxity is the choice of graft and subsequent fixation. Most of the fixation techniques used for PCL reconstruction were originally developed for ACL reconstruction and were adapted for use in PCL reconstruction.144 Because of the different insertional geometries and size of the two ligaments, biomechanical differences exist between the ACL and PCL,168 and requirements for graft fixation in PCL reconstruction may vary significantly from those of the ACL. There is little in the published literature on the biomechanics of PCL fixation devices.169,170 A recent survey of knee surgeons in the Herodicus Society membership found the Achilles tendon allograft to be the most popular graft choice in both acute and chronic reconstructions.171 The same study found that inference screw fixation was used by nearly 70% of surgeons for femoral fixation, whereas interference screw, screw and post, and other devices were used for tibial ­fixation.171 A final potential source of laxity is the position of the knee at the time of fixation. Most authors have recommended fixation at 90 degrees of knee flexion with an

1700 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

applied anterior drawer force applied to the knee.36,47,48 This position reflects the dominant role of the PCL in knee stability at this flexion angle as well as the concern in AL single-bundle reconstructions that tensioning the graft at angles closer to full extension risks overconstraining the knee. Accordingly, 55% of those surveyed in the Herodicus Society tensioned their grafts near 90 degrees of knee flexion.171 Definitive answers to all of the aforementioned questions concerning PCL reconstruction outcomes, however, will likely not be answered conclusively until prospective, randomized biomechanical and clinical trials comparing various techniques are ­conducted.

Combined Posterior Cruciate Ligament Injuries Although PCL injuries can occur in isolation, an increasing number are being recognized as occurring as part of a combined ligament injury pattern and have been treated operatively.172-175 A common injury pattern, which will be the focus of this section, involves damage to the PCL and the structures of the PLC.175,176 In cases of combined posterior and posterolateral instabilities, laboratory results have shown that isolated reconstruction of the PCL only leads to excessive stresses on the PCL graft with subsequent failure.34,39 Biomechanical and clinical studies have shown that the addition of a PLC reconstruction improves the stability of the knee in the setting of a concurrent PCL reconstruction.42,177,178 Various techniques of both PCL and posterolateral reconstruction exist, as detailed in previous sections of this book. Fanelli and associates44 and Noyes and colleagues103 presented some of the earliest reconstructive results of combined PCL and PLC injuries. Noyes and colleagues103 described a proximal advancement of the LCL and PLC in patients with definitive but lax lateral and ­posterolateral structures with good results. Fanelli and associates44 treated 21 patients with the use of a biceps tenodesis procedure and concurrent PCL allograft or autograft reconstruction. The results showed improvement in preoperative and postoperative Lysholm and Tegner scores, from 51.8 to 90.9 and 2.2 to 5.1, respectively, at minimal 24-month follow-up. In a larger study of 41 patients with PCL and PLC injuries followed for 2 to 10 years, Fanelli and Edson179 found similarly good results. Using Achilles allograft for the PCL reconstruction and a combination of biceps tenodesis and PLC advancement for the PLC reconstruction, they found postoperative Lysholm and Tegner scores of 91.7 and 4.9, respectively. Reflecting the complex injury pattern and the significant surgical reconstruction required for such injuries, however, other studies have found inferior results.180,181 Khanduja and colleagues180 in a follow-up study of 19 patients with 2 to 9 year followup, found that their Lysholm and Tegner scores increased from 41.2 to 76.5 and 2.6 to 6.4, respectively, from preoperatively to postoperatively. They used mainly Achilles allograft for their PCL reconstruction and a Larson-type tenodesis for their PLC reconstruction. They noted that although knee function significantly improved, patients did not have complete knee stability. Similarly, Wang and associates181 followed 25 patients for an average of

2.3 years. Their technique included various allografts for the PCL and the use of iliotibial and biceps tenodesis reconstructions of the popliteus tendon and popliteofibular ligament with or without LCL advancement. They noted 68% satisfactory and 32% unsatisfactory results using the Lysholm scores. Complete restoration of ligament stability was observed in 44% of knees.

Weighing the Evidence Despite some inconsistencies in the literature, most surgeons agree that the treatment of peel-off injuries of the PCL should consist of acute primary repair. Direct repair or reconstruction is also the preferred technique of most orthopaedic surgeons for the treatment of combined PCL injuries (particularly PCL-PLC and PCL-ACL-PLC). In such multiligamentous injury cases, reconstructions are typically done using AL single-bundle reconstruction techniques. Isolated injuries continue to generate a fair amount of debate. However, most surgeons treat isolated, acute grade I and II injuries conservatively with limited activities with or without extension bracing for 4 to 6 weeks, ­followed by muscle (particularly quadriceps) strengthening. Additionally, many surgeons now recommend conservative treatment of acute, isolated grade III PCL injuries that consists of 4 weeks of full-time extension bracing to prevent posterior tibial subluxation and allow healing of the PCL. However, because operative techniques have advanced, some surgeons may elect to proceed with primary reconstruction of the PCL. This is particularly true in the young or athletic patient. Multiple viable options exist concerning specific graft choices, fixation techniques, and inlay versus transtibial graft placement. With the available literature, there is no clear consensus on these topics. Currently, more reconstructions are done using single-bundle PCL ­reconstruction techniques, although evidence suggesting superior biomechanical results with double-bundle reconstruction may favor this technique in the future. More clinical studies of double-bundle PCL reconstruction and, more specifically, randomized trials comparing single- versus double-bundle PCL reconstruction are needed. For chronic PCL injuries, clear treatment recommendations are lacking. Most surgeons would elect to treat chronic grade I and II PCL injuries with conservative treatment consisting of quadriceps strengthening and activity modification. In patients who fail such conservative treatment and demonstrate minimal to no chondral damage, PCL reconstruction using either single- or double-bundle techniques is generally recommended. In chronic cases with varus knee alignment and chondral damage in the medial and patellofemoral compartments, interest is gaining in biplanar tibial osteotomies. Biplanar osteotomies in these patients convert the varus knee into neutral or slight valgus alignment while increasing the tibial slope to help reduce the chronic posterior tibial sag. For chronic multiligamentous injuries, surgery to reconstruct the PCL is generally recommended. In such cases, the magnitude of the injury and any previous surgeries often makes stiffness a bigger issue. In such cases, the need for concomitant ACL or collateral reconstruction, or both, needs to be assessed on a case-by-case basis.

knee 1701

Authors’ Preferred Method The treatment algorithm for PCL injuries is multifactorial and steadily evolving. Treatment must be tailored to specific characteristics of the patient and injury. Although the timing, grade, location, and extent of associated injuries are important prognostic factors that must be considered, the patient’s age, occupation, comorbidities, and expectations regarding return to play are equally important in the decision-making process. Algorithms have been developed to aid the physician in treating these difficult problems, but it is important to consider the specific characteristics of each patient before pursuing a specific treatment (Figs. 23E1-16 and 23E1-17). Isolated injuries, in general, may be treated nonoperatively with an excellent prognosis.65,67,73,82,83,87,121,182 Combined injuries have a more guarded prognosis. Better results are possible in this group with early surgical intervention rather than with conservative treatment.36,80,86,92 Isolated Acute Posterior Cruciate Ligament Injury

Like most surgeons, we believe that acute, isolated grade I and grade II PCL injuries usually do not require surgical intervention.65,67,73,82,83,87,121 The outcome of these injuries is most likely related to the remaining integrity of the PCL (single-bundle injury), other secondary restraints (meniscofemoral ligaments), and the intrinsic healing capabilities of the PCL. Partial injuries of the PCL are more likely to heal than the ACL owing to its large size and better blood supply.36,121 For any PCL injury, the specific injury pattern of the PCL should be identified, which includes isolated AL

or PM bundle injuries versus both bundle injuries as well as damage to the meniscofemoral ligaments. With grade I and grade II injuries, the PLC structures are usually intact, and the tibia does not significantly sag at any knee flexion angle. The natural posterior slope of the tibia is protective at preventing further posterior subluxation. The initial treatment is focused on maintaining reduction of the tibia on the femur with intermittent extension bracing for 2 to 4 weeks. We protect their weight-bearing, advance range of motion as tolerated, and focus on quadriceps strengthening. In our experience, most patients are able to return to sports within 2 to 6 weeks after the injury. In acute grade III injuries, the rehabilitative course is not as predictable and frequently requires a longer time. This has pushed our focus to more early surgical interventions in younger athletic patients. A high index of suspicion is necessary to rule out any significant associated PLC or posteromedial side injuries. We recommend 2 to 4 weeks of immobilization in full extension. This will minimize the posterior-­displacing effect of both gravity and the hamstrings on the tibia and will allow minor PLC injuries, if present, to heal with less stress.183,184 After the period of immobilization, the mainstay of rehabilitation is quadriceps strengthening to counteract posterior tibial subluxation.183,185,186 Quadriceps sets, straight leg raises, mini flexion squats, and partial weight-bearing with crutches are initiated as measures to help reduce the tibia on the femur.4,36,65,67,68,102,121,122,187-192 About 1 month after the injury, motion exercises are started, and weight-bearing is progressed. Functional exercises follow, such as biking and stair climbing as well as leg presses and knee extensions. After

Acute PCL Injury

“Combined” • PLC (+/- LCL) • MCL and medial side injury • ACL (+/- collaterals) (knee dislocation)

“Isolated”

Grade I or II

III

Nonoperative • quadriceps therapy • 2-4 week extension immobilization • gradual return to activity

Is patient young, athletic or avulsion injury?

No

Nonoperative • 4 weeks of full extension • Avoid posterior tibial subluxation • quadriceps therapy • limited activity

Operative • Surgery peformed < 2 weeks, acute repair of collateral injuries • Single bundle reconstruction for dislocated knee • Single bundle augmentation for grade II PCL

Yes Operative • ORIF avulsion • Single bundle augmentation • Possible, double-bundle reconstruction

Figure 23E1-16   Treatment algorithm for acute injuries of the posterior cruciate ligament (PCL). ACL, anterior cruciate ligament; LCL, lateral collateral ligament; MCL, medial collateral ligament; ORIF, open reduction, internal fixation. Continued

1702 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method— �c ���� o n t ’� d � Chronic PCL Injury

“Combined” • PLC (+/- LCL) • MCL and medial side injury • ACL (+/- collaterals) (knee dislocation)

“Isolated”

Grade I or II

III Malalignment?

Nonoperative • quadriceps therapy • activity modification

Symptomatic pain or instability? Yes

No Nonoperative • quadriceps therapy • activity modification Operative • Double-bundle reconstruction • Occasionally, singlebundle augmentation

Yes

No

Malalignment? No

Yes

Operative • Biplanar osteotomy • Staged reconstructions for combined injuries

Operative • Reconstruct all injured components, especially PLC • Double-bundle reconstruction • Single-bundle augmentation if Grade II PCL

Figure 23E1-17    Treatment algorithm for chronic injuries of the posterior cruciate ligament (PCL). ACL, anterior cruciate ligament; LCL, lateral collateral ligament; MCL, medial collateral ligament.

grade III injuries, it takes up to 3 months before the athlete is able to return to sport. PCL functional braces may be used but have not been effective in our experience. Unfortunately, some patients do not ­respond to nonoperative therapy. These athletes become increasingly symptomatic and are unable to return to sport without some type of surgical intervention. Because of the failure of nonoperative treatment, we are more inclined to operate on young, athletic patients with acute isolated grade III PCL injuries.73,82,120 A PCL avulsion injury is a unique isolated PCL injury with its own treatment algorithm. Although most PCL injuries requiring surgical treatment are best suited for ­reconstruction versus repair, PCL avulsion injuries are best treated with primary repair typically resulting in favorable outcome.74,104,130-132,193 PCL avulsion most commonly ­occurs on the femoral side of the ligament, but it can also occur on the tibial side, where it is usually associated with a large bone fragment. One must be careful in this setting because of the potential for postoperative stiffness. With these types of injuries, we have found that it is best to delay surgery for 1 to 2 weeks to allow swelling to resolve and to keep the knee immobilized in full extension. The femoral and tibial avulsions are approached through an anteromedial arthrotomy and a standard or modified posterior knee ­approach, respectively.194 They can then be fixed with either screws or sutures through drill holes, depending on the presence and size of the bone fragment.37,45,111,195,196 Likewise, if

an avulsion injury to the ACL is also present, it is treated in a similar fashion.196,197 Isolated Chronic Posterior Cruciate Ligament Injury

Most chronic, isolated grade I and grade II tears will similarly respond well to physical therapy. On occasion, some grade II injuries will develop persistent symptoms of ­intermittent swelling and pain. In these situations, we find it helpful to proceed with both radiographic and scintigraphic evaluation of the joint to assess the status of the joint ­ compartments. If the studies, most notably the bone scan, are consistent with significant medial uptake with osseous malalignment, we recommend an osteotomy. Reconstructions in these settings are usually avoided because current techniques have not been consistent in restoring the knee to normal function.89,158,151-155,198-200 In our experience, patients with isolated, chronic, grade III deficiencies have a mixed postinjury course. Reconstruction is recommended for all patients who become symptomatic despite maximizing physical therapy intervention, especially if the compartments of the knee are still well preserved. In these cases, current surgical techniques can potentially provide enough stability to diminish the persistent posterior instability or pain from the medial or patellofemoral compartment. In our opinion, patients with persistent

knee 1703

Authors’ Preferred Method— �c ���� o n t ’� d � symptoms after adequate therapy for an isolated grade III PCL injury are likely to have occult concomitant ligamentous injury, particularly involving the PLC.12,41,81,95,201,202 Depending on the severity, this particular combination of injuries can lead to disability ranging from minimal ­functional alterations to profound limitations in daily activities.16,80,203-207 We therefore recommend a PLC reconstruction in conjunction with the PCL reconstruction. When a coexistent PLC injury is overlooked, surgical treatment of the PCL may have a higher risk for failure.40,41 In addition, for chronic injuries, it is essential to assess the entire limb for asymmetric varus alignment and, even more important, the presence of a dynamic varus thrust with gait.81,92,208-212 Soft tissue reconstructions alone for ligamentous deficiencies with malalignment will more likely fail over time.102,208,213 Although there are no long-term studies documenting its effectiveness, many authors believe the single most reliable procedure for correcting varus malalignment is a high tibial osteotomy.102,208,213 Corrections can also be biplanar to manipulate and accentuate the native posterior slope in the PCL-deficient knee.189,214,215 Another emerging subset of PCL-injured patients ­includes those with chronic PCL injuries and associated medial compartment arthrosis. As detailed earlier, patients with chronic PCL injuries often have chronically subluxated tibias, which effectively unloads the medial meniscus and functionally results in a postmeniscectomy knee.36,216 Often these patients have early articular cartilage wear (despite an intact meniscus) and have begun to drift into varus ­alignment with or without an associated PLC deficiency. These patients often complain more of pain in the medial and patellofemoral compartments than instability symptoms. Their PLC ­ injuries may have been diagnosed and treated conservatively, or overlooked because the patient was able to cope adequately with their PCL insufficiency. It is often with activities that these patients will experience their worst symptoms of pain and knee swelling. Patients with early to moderate medial and ­patellofemoral arthrosis typically do not respond well to isolated PCL ­reconstruction. During the past 5 years, we have ­ begun ­addressing this problem by correcting the alignment of the knee that contributes to the pathology. We have used a ­ biplanar osteotomy to reduce contact forces in the ­medial compartment by decreasing varus and increasing the ­ posterior slope of the tibia. Radiographic studies have ­previously ­ reported the relationship between tibial slope and tibial ­ translation.217 Several investigators have noted a ­similar ­concept for ACL with varus.218,219 In those cases, the ­posterior tibial slope is decreased with a combined ­osteotomy and ACL ­ reconstruction. Cadaveric studies have demonstrated that increasing posterior tibial slope shifts the resting position of the tibia anteriorly, thereby offsetting posterior tibial sag associated with PCL deficiency.189,215,220 In a ­similar fashion, a slight correction into valgus will help correct any associated varus or PLC deficiency. Such ­ anteromedial opening wedge high tibial osteotomies have been described for the ­treatment of the PCL-deficient knee.221

Combined Posterior Cruciate Ligament Injury

Most patients with combined injuries benefit from surgery.58,81,85-87,91,206,222-224 These patients are at high risk for persistent and progressive functional instability, and surgical treatment has given a more predictable outcome. Early and accurate diagnoses of all concomitant ligamentous injuries is essential. Numerous injury patterns exist, most commonly, PCL injury with an associated PLC injury. It is important to differentiate PLC injuries that include the LCL from those that do not include the LCL. Other common injury patterns include PCL injuries with concomitant PLC and ACL injuries. These injuries fall under knee dislocations. The appropriate vascular work-up is essential before any reconstructive procedure. Finally, the least common pattern is PCL injury with an MCL or posteromedial injury. In this acute setting, we recommend combined PCL reconstruction and simultaneous repair or reconstruction of all associated, complete ligamentous injuries followed by early motion. If peripheral meniscal tears, capsular avulsions, and ostechondral injuries are encountered, they are also repaired primarily. This ­treatment approach subjects the patient to fewer operations, decreases concern for late instability, and limits the possibility of postoperative stiffness. Combined PCL and PLC injury is one of the most complex treatment problems encountered in the management of knee ligament injuries. When both the PCL and PLC are ruptured, substantial posterior translation, external rotation, and varus opening can be present at differing angles of knee flexion.81 This combination creates a complex surgical dilemma.100 With several patterns of injury possible, it is difficult to have one surgical plan. It is essential that this injury be appropriately identified because if it is mistaken for an isolated PCL injury and treated nonsurgically, posterior and ­ posterolateral instability will invariably persist.36,37,41,47,58,59,225 Treatment of the acute PLC injury is generally more successful than that of the chronic injury; therefore, acute surgical intervention is recommended for combined PCL and PLC injury.36,63,81,86,88,91,95,187,202,205,206 The timing for surgical treatment of the injured PLC is critical; acute repairs consistently give more favorable results than does reconstruction of chronic injuries.63,81,88,95,187,202,208 Primary repair has also met with better results if undertaken within the first 2 weeks.84,92 Primary repairs are best done for avulsion injuries. Although midsubstance repairs of the LCL can be performed, we tend to supplement the repair of intrasubstance tears with a graft reconstruction. This is done because of less consistent healing of the LCL as well as the limited success in treating chronic insufficiency of the LCL if it were to occur. Attempts at surgical repair beyond the acute time frame (3 to 4 weeks) are frequently disappointing both in localizing discrete anatomic structures and in finding any sturdy tissue to repair. Accordingly, surgical ­ options for chronic injuries are reconstructions rather than repairs. Many reconstructive techniques have been ­described, but none has consistently shown better results than acute ­repair. Continued

1704 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method— �c ���� o n t ’� d � The treatment of chronic, combined injuries differs from that of the acute disruptions. Beyond 2 weeks from ­injury, pericapsular scarring becomes significant, and primary ­repairs are not possible, particularly for the collateral ligaments and avulsion injuries of the cruciates. This occurrence typically commits the surgeon to waiting up to 3 months to allow completion of the healing response before effective ­ treatment can be implemented. In addition, these chronic injuries may become associated with significant capsular stretching, leading to a more extensive rotational instability pattern, persistent subluxation, or the ­ development of ­ arthrosis. This could cause substantial difficulty in ­determining the extent of the injury as well as the optimal treatment plan. Surgical intervention for these injuries in the chronic setting requires simultaneous reconstruction of the PCL and PLC. As with any chronic grade III PCL injuries, the limb must be assessed for malalignment and a varus thrust gait. Any PCL and PLC reconstruction in this setting will have a significant risk for failure because of chronic repetitive stretching of the reconstruction with time.81,205,207,226 If there is significant deformity, a high tibial biplanar osteotomy may be performed in conjunction with PLC reconstruction, ­although we favor performing a staged operation because the osteotomy alone may alleviate the patient’s symptoms, avoiding further surgical intervention.92,95 Unlike the PLC, the medial side traditionally has a ­greater opportunity to heal conservatively. However, in the PCLdeficient knee, we have a low threshold to surgically address the medial side if significant instability exists. Avulsions or intrasubstance tears of the medial collateral ligament may be directly repaired and are best performed acutely when the quality of tissue is robust.227-231 Usually, significant (grade III) medial-sided injuries can be localized to the femoral or tibial sides with the use of radiographic imaging. These can be treated with a posterior oblique ligament advancement or on rare occasion graft reconstruction.229,230,232-235 The specific technical details for the repair or reconstruction of complex, combined injuries are beyond the scope of this chapter. Instead, we focus our discussion mainly on the reconstruction of the PCL, including single-bundle, ­double-bundle, and augmentation techniques. We also discuss the treatment of associated PLC injuries because they are the most commonly disrupted structures in combined PCL injuries.12,41,44,59,80,81,88,91,92,95,96,103,206 We have also found them to be injured to varying degrees in most chronic PCL injuries. Rationale

Our surgical approach is dictated by identifying all the ­injured structures and surgically addressing each structure based on our understanding of the basic science, insertional anatomy, and the patient’s characteristics or preferences. To ensure optimal surgical outcome, the surgeon must be familiar with and capable of performing a repair, reconstruction, or augmentation of the PCL and any other associated cruciate, collateral, and capsular structures. For any PCL injury, specific injury patterns including the AL bundle or PM bundle, or

both, must be recognized. This is typically determined by the information obtained from the preoperative examination, the imaging studies, the examination under anesthesia (EUA) and finally arthroscopic evaluation. The EUA is most valuable because it determines the grade of PCL injury and identifies other potentially injured ligamentous structures. Surgery in these cases should be performed in a semielective setting with a skilled operating room staff. In addition, ­ because of the proximity of the vessels to the tibial PCL graft placement, a vascular surgeon should be immediately available. Current reconstructive options include arthroscopic ­single-bundle or double-bundle techniques as well as arthroscopic and open tibial inlay procedures.36,63,88,187,188,208,236 Various combination of these techniques have also been employed. Most recently, we have begun augmentation ­procedures, in which we attempt to preserve any intact PCL structures and augment only damaged tissue. We ­ believe, however, that none of these options can effectively reproduce all the components of the PCL complex. The ­single-bundle technique was developed to reconstruct the anterolateral bundle because of its larger size and greater biomechanical properties.4,6,7,33,46,48,50,205,237 In an attempt to place the graft in the anatomic position of the native ­ anterolateral bundle, a single tibial and femoral tunnel is used. We ­occasionally use this technique in performing acute PCL reconstructions, especially if it is part of a combined injury pattern.196 Because biomechanical data suggest that the ­addition of a second bundle significantly decreases posterior tibial translation, the double-bundle technique has become our preferred ­ procedure over the traditional single-bundle technique.45,46,238-240 Although the tibial inlay procedure is favored by some surgeons, we think this approach is technically demanding and requires a prone or lateral decubitus position, adding operative time and becoming particularly burdensome in attempting repair of a combined ligament ­injury.149,208 At this time, tibial fixation variation has not been shown to affect the behavior of the graft significantly, and so the theoretical benefits may not outweigh the technical demands of this technique.149,152-155,241,242 A variety of tissues and fixation devices have been used for reconstruction. As with the ACL, we feel that proper placement is more critical than graft type or fixation technique. Autologous tissues typically used today include ipsilateral or contralateral bone–patellar tendon–bone, hamstring, iliotibial band, and central quadriceps tendon grafts. Bone– ­patellar tendon–bone, Achilles tendon, and soft tissue (anterior tibialis) are the most commonly used allograft tissues. We now favor soft tissue tendon anterior tibialis allograft because of its high tensile strength, ease of passage and fixation, and lack of donor site morbidity. Additional benefits of this graft include its exceptional size and length, making it versatile compared with other graft options. Multiple methods of fixation exist, including EndoButton, (Smith & Nephew Endoscopy, Andover, Mass), cortical screws and washers, and staples. No single technique is universally ­accepted. The treating surgeon should be familiar with ­several of these options so that the final choice can depend on the surgical situation.

knee 1705

Authors’ Preferred Method— �c ���� o n t ’� d � The rationale for our present techniques is based on the anatomy and biomechanics of the PCL. Not all PCL injuries are the same. Different injury patterns will dictate different techniques. As stated earlier, EUA, MRI, and arthroscopic findings, combined with individual patient characteristics, are all critical for a successful reconstruction. Currently, we use three different arthroscopic transtibial techniques, which are all variations based on PCL insertion site anatomy. Our indications for AL single-bundle reconstruction is an acutely injured knee or a multiligamentous knee. The indications for augmentation are evolving but include mostly acute but some chronic cases, in which there is preservation of function in one of the remaining bundles of PCL. Finally, our indications for double-bundle reconstructions are chronic grade III with either PLC or PM-sided injuries, revision, and rarely acute injuries. Stated another way, for most acute injuries, a singlebundle (AL) reconstruction or augmentation is performed (see Fig. 23E1-16). For chronic injuries, either a ­ single­bundle augmentation or a double-bundle ­reconstruction with or without collateral surgery is performed. If there is significant malalignment, an opening wedge biplanar high tibial ­osteotomy is performed (see Fig. 23E1-17). Operative Technique: Single-Bundle Reconstruction, Single-Bundle Augmentation, and Double-Bundle Reconstruction

When performing PCL surgery, our patient is positioned supine in 90 degrees of flexion without use of a tourniquet. A detailed examination under anesthesia is performed first. After an EUA is performed and the imaging and office notes reviewed, standard arthroscopic assessment of the knee joint is performed to confirm the extent of the injury and to assist with the repair or reconstructive procedure whenever possible. If the operation is performed in the acute setting, the potential for fluid extravasation should be carefully considered. In this case, we recommend using gravity flow rather than a fluid pump, and the surgeon should frequently assess the calf pressure. If increased calf pressure is noted during the case, the arthroscopic procedure should be immediately abandoned. If warranted, we have a low threshold to extend any of our lateral tibial incisions and perform a fasciotomy of any involved compartments in the leg. First, our attention is turned to the ACL and the tibiofemoral compartments to assess any secondary signs of associated injury patterns. Only after this is done is our attention turned to the PCL and its injury pattern. After confirmation of the PCL injury pattern, a decision is made if an augmentation procedure is possible, remembering that tissue may appear normal but can have significant interstitial injury. MRI data can also be helpful in making this decision. Once the exact procedure is determined, the remnants of the PCL insertions are débrided with the use of a posteromedial as well as standard anterior portals. A 70-degree scope is critical to accurately view the tibial insertion, and care must be taken to avoid injury to the closely situated neurovascular structures. A tibial tunnel is then created from the anteromedial tibia and directed posteriorly to the native PCL tibial attachment (Fig. 23E1-18). The correct position is critical

and should be checked with intraoperative radiographs ­after the guidewire is placed. If a single-bundle or augmentation procedure is performed, care is taken to place the single guidewire in the appropriate insertion site. If a double-­bundle reconstruction is performed, two guidewires are placed and confirmed radiographically, with the AL guidewire being more lateral and distal and with the PM guidewire more ­medial and proximal (Fig. 23E1-19). The starting point for the two tibial tunnels can either be divergent (one medial, one lateral) or stacked (Fig. 23E1-20). We use both fluo­ roscopy and arthroscopic guidance to drill the tibial tunnels, and we usually start with power and finish by hand. After the tibial tunnels are completed, attention is focused on creating the femoral tunnels. The lateral portal is enlarged, and the knee is hyperflexed to drill the femoral tunnel (Fig. 23E1-21). The femoral insertion site anatomy is identified, and the appropriate tunnel is marked for a single-bundle reconstruction, singlebundle augmentation, or double-bundle reconstruction. If an augmentation procedure is being performed, care is ­taken to preserve any remaining functional PCL tissue. Most commonly, the PM and MFLs are preserved and augmented with a single-bundle AL reconstruction (Fig. 23E1-22). We most commonly use anterior tibialis allograft tendons. This tissue is looped over a 40-50 Endoloop (EndoButton may or may not be removed depending on fixation) to make a two-stranded construct with whipstitches in both free ends with 2 nonabsorbable suture. These grafts are then passed anterograde through the tibial tunnel and subsequently retrograde into the femur. One or two grafts are used depending on whether an augmentation, single-bundle ­reconstruction, or double-bundle reconstruction is being performed. The grafts are first fixed on the femoral side with either EndoButtons or posts through an Endoloop and then cycled repeatedly. Depending on the reconstruction, the anterolateral bundle is tensioned at 90 degrees and then fixed with a

Figure 23E1-18   Posterior cruciate ligament tibial guidewire placement and drilling. (Adapted from Miller MD, Harner CD, Koshiwaguchi S: Acute posterior cruciate ligament injuries. In Fu FH, Harner CD, Vince KG [eds]: Knee Surgery, vol 1. Baltimore, Williams & Wilkins, 1994.) Continued

1706 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method— �c ���� o n t ’� d �

AL PM

PM AL

Figure 23E1-21   External view of femoral drilling with camera in medial portal and drill in lateral portal with the knee hyperflexed.

Posterior Cruciate Ligament and Postero­lateral ­Corner: Acute and Chronic Figure 23E1-19   The ideal femoral and tibial tunnel positions for a double-bundle posterior cruciate ligament reconstruction. AL, anterolateral bundle; PM, posteromedial bundle.

post and a washer on the tibia, and the posteromedial bundle is tensioned at 30 degrees of flexion and secured in a similar fashion over a post (Fig. 23E1-23). If additional collateral or meniscal surgery is being performed, tibial fixation is withheld until these procedures are done.

Figure 23E1-20   A double-bundle posterior cruciate ligament reconstruction with one stacked tibial tunnel.

The treatment of combined acute PCL and PLC injuries is centered on recognizing the injury pattern, as well as its extent and grade. This is followed by immediate, direct anatomic repair of all PLC ligamentous injuries, preferably within the first 2 weeks. Depending on the quality of the tissue or type of injury, repair or reconstructive techniques may be used. First, attention is turned to the PCL reconstruction. If there is functional PCL tissue remaining and the injury grade is only grade II, we favor an augmentation technique. In most cases, however, we perform double­bundle reconstructions because these are commonly grade III injuries. Double-bundle reconstructions begin with proper identification of the AL and PM tibial insertion sites. After the tibial tunnels have been completed, attention turns to the femoral tunnel insertion sites. They are marked with an awl, and then guidewires are placed through the anterolateral portals and are drilled and dilated with the knee in hyper­flexion (Fig. 23E1-24). As with single-bundle ­reconstructions, soft tissue anterior tibialis allograft tendons are looped over a 40-50 Endoloop, making a two-stranded construct with whipstitches in both free ends. For double-bundle reconstructions, two grafts are then passed anterograde through the tibial tunnel and subsequently retrograde into the femur. Our preferred graft is an anterior tibialis allograft, but if an Achilles tendon allograft is used, it is used for the anterolateral tunnel while an anterior tibialis tendon is used for the posteromedial tunnel (Fig. 23E1-25). The grafts are first fixed on the femoral side with either EndoButtons or posts through an Endoloop and then cycled repeatedly. Final ­tibial fixation is deferred until the PLC is addressed.

knee 1707

Authors’ Preferred Method— �c ���� o n t ’� d �

B

A

Figure 23E1-22   Positioning of femoral tunnels for anterolateral (AL) bundle augmentation. A, Injury to the AL bundle with an intact posteromedial bundle (PM) bundle and meniscofemoral ligaments. B, AL bundle reconstruction with tibialis anterior allograft and preservation of intact PM bundle.

Our approach to PLC begins with a lateral “hockey-stick” incision paralleling the posterior edge of the iliotibial band, which is then split, exposing the deep structures of the LCL anteriorly and the lateral head of the gastrocnemius muscle and underlying popliteus complex more posteriorly. ­Special

A

attention is given to identifying the injured structures. When the posterolateral structures, including the LCL, are avulsed off their femoral attachments with preservation of the popliteus tendon, direct repair of these structures by internal fixation, or suture anchors is recommended. On

B

Figure 23E1-23   Radiographs after single bundle anterolateral augmentation. A, Anteroposterior radiograph. B, Lateral radiograph. Continued

1708 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method— �c ���� o n t ’� d �

A

B

Figure 23E1-24   Positioning of femoral tunnels for double-bundle reconstruction. A, Femoral tunnel position for anterolateral and posteromedial bundles. Note that the anterolateral bundle is more anterior. B, Double-bundle reconstruction with tibialis anterior allografts.

­ ccasion, there is interstitial tearing of the LCL, mandato ing concomitant reconstruction (Fig. 23E1-26).243,244 For this, either bone–patellar tendon–bone or Achilles tendon allograft is used. The LCL can be detached and elevated from its distal insertion, and the allograft bone block is then fixed vertically into the fibular head by interference screw fixation distally, respecting the proper insertional anatomy. Fluoroscopic guidance is used to properly position this tunnel.10 A blind femoral tunnel is then made, and the Achilles tendon allograft is tensioned proximally over a post, interference screw, or suture anchors. The native LCL can

then be tensioned both proximal and distal to the graft and then directly sutured to the midsubstance of the Achilles ­allograft to reinforce the reconstruction. The extent of injury to the popliteus and, more important, its attachments to the fibula through the popliteofibular ligament must then be visualized. The popliteofibular ligament is now recognized as a significant component of the popliteus complex, particularly as a static stabilizer.14,95 We, therefore, believe that this step is the most crucial to the overall success or failure of the procedure. When this tendon is avulsed off its tibial or femoral insertion, ­appropriate

1

2

1 2

A

B

Figure 23E1-25   Graft placement. A, The tibialis anterior or Achilles tendon allograft for the anterolateral bundle (inset, 1) and a second tibialis anterior for the posteromedial bundle (inset, 2) are passed in anterograde fashion through the tibial tunnel. B, Grafts are then fixed to corresponding femoral tunnels. (Adapted From Petrie RS, Harner CD: Double bundle posterior cruciate ligament reconstruction technique: University of Pittsburgh approach. Op Tech Sports Med 7:118-126, 1999.)

knee 1709

Authors’ Preferred Method— �c ���� o n t ’� d �

A

B

Figure 23E1-26   Lateral collateral ligament (LCL) reconstruction with bone–patellar tendon–bone or Achilles tendon allograft. A, The torn or stretched LCL is elevated from its fibular insertion, and the allograft is fixed in a tunnel in the proximal fibula using an interference screw. The tensioned graft is then fixed to the lateral epicondyle using multiple suture anchors or with a post on the medial side. B, The native LCL is tensioned and sutured to the graft. (Adapted from Cole BJ, Harner CD: The multiple ligament injured knee. Clin Sports Med 18:241-262, 1999.)

tension with anatomic restoration of the popliteofibular ligament is possible. The popliteus can be easily repaired with internal fixation, suture anchors, or a blind tunnel to its femoral insertion. The popliteofibular ligament can be repaired to its fibular attachments with similar fixation. ­ Tension is applied with the knee in 20 to 30 degrees of flexion during the final fixation. If the popliteus tendon tissue cannot be repaired by this approach, reconstruction is indicated as with a chronic injury. Final PCL tibial fixation is then performed. The reconstructed anterolateral bundle is tensioned at 90 degrees and then fixed with a post and a washer on the tibia. Subsequently, the posteromedial bundle is tensioned at 30 degrees of flexion and then secured in a similar fashion over a post (Fig. 23E1-27). Chronic cases of posterolateral instability demonstrate tissue redundancy and excessive scarring posterior to the LCL, and identification of the particular structures of the popliteus complex is difficult. Many techniques have been recommended, including arcuate ligament advancement, ­biceps tenodesis, and popliteofibular ligament reconstruction with allograft or autograft tissue, but no consensus ­exists on which is the best procedure.11,86,100,103,203,245,246 In this situation, we currently recommend anatomic reconstruction of the popliteofibular ligament and, if necessary, the LCL. The PCL is reconstructed using a double-bundle technique and then attention is then focused on the reconstruction of the PLC. With use of the same approach to the lateral knee, the LCL is identified; if it is part of the injury pattern, it is ­reconstructed as described previously. If the LCL is intact, our attention turns to the PFL. A PFL reconstruction is then performed with anterior tibialis allograft with either

soft tissue fixation around the biceps tendon or more rigid fixation with a second fibular tunnel. An oblique anteriorto-­posterior tunnel oriented similarly to the course of the ligament is then created in the proximal fibula. A proximal, blind femoral tunnel is then created at the ­ anatomic ­insertion site of the popliteus tendon anterior to the femoral epicondyle in its anatomic “saddle.” The popliteus is ­ mobilized and then advanced with a prepared anterior ­tibialis allograft into the blind femoral tunnel. The anterior tibialis graft is then passed under the iliotibial band and LCL into the proximal, posterior fibular tunnel opening and out the distal, ­anterior opening. The knee is then placed in 30 ­degrees of flexion, in which final posterolateral stabilization is ­performed. ­Fixation at the femoral side is stabilized over a post or button tied over the medial cortex through a separate skin incision. The knee is maintained in 20 to 30 degrees of flexion with neutral to slight internal rotation of the foot as the graft is tensioned and secured in the fibular tunnel with an interference screw and the remaining portion of the graft is sutured to the iliotibial band. This is then followed by final PCL tibial fixation as described previously (Fig. 23E1-28). Postoperatively, the patients follow PLC protocol modification of the PCL reconstruction protocol. Immediately after the operation, the injured limb is placed in a well­padded hinged knee brace locked in extension. This should be performed with care to avoid any posterior translation of the tibia while it is applied. Immobilization in extension enables the knee joint to remain reduced and minimizes the effects of gravity and hamstring forces that create posterior tibial sag, thus allowing the collateral ligament repair or ­reconstruction a chance to heal.36,247,248 Continued

1710 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method— �c ���� o n t ’� d �

A

B

Figure 23E1-27   Radiographs after double-bundle posterior cruciate ligament (PCL) reconstruction with repair of femoralsided PLC. A, Anteroposterior radiograph. B, Lateral radiograph.

A

B

Figure 23E1-28   Radiographs after double-bundle posterior cruciate ligament (PCL) reconstruction with popliteofibular ligament reconstruction for chronic grade III PCL and posterolateral corner. A, Anteroposterior radiograph. B, Lateral radiograph.

knee 1711

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Postoperative Prescription The PCL rehabilitation protocol is, in general, slower than that for isolated ACL reconstruction. Reconstructions of PCL and PLC combined injuries are progressed even slower with crutch use continued for 12 versus 8 weeks to allow the lateral soft tissue repair or reconstruction to heal without undue tension. Our postoperative PCL rehabilitation program is broken down into four phases: (1) 0 to 4 weeks, (2) 1 to 3 months, (3) 3 to 9 months, and (4) 9 to 12 months.36,88 Supervised physical therapy takes place for about 3 to 5 months after surgery.

Phase 1 A hinged knee brace is maintained in full extension for the first postoperative week and is then unlocked for range of motion exercises. These exercises are performed with the assistance of a physical therapist who applies an anterior drawer force to the proximal tibia as the patient flexes the knee. The anterior force is important in preventing posterior tibial sag. During this time, the patient is allowed to bear weight as tolerated on the limb with the brace locked in full extension. Crutches should be used for ambulation during the first 6 to 8 weeks. For combined PCL and PLC reconstruction, the brace is locked in full extension at all times for a total of 4 weeks, and crutches are required for 12 weeks. Quadriceps exercises are the mainstay of rehabilitation and are begun in the form of quadriceps sets and straight leg raises starting the first postoperative day. Therapeutic exercises include wall slides (0 to 45 degrees), which are closed chain to take advantage of native sagittal slope of the tibia and its tendency to keep the tibia anterior.189 Active hamstring exercises are avoided because of the potential for the muscles to subluxate the tibia posteriorly and stress the reconstruction. Range of motion is gradually increased but not to exceed 90 degrees during this phase.

Phase 2 This phase begins 4 weeks after surgery and lasts 8 weeks. The goals are to allow the healing of the soft tissue reconstruction to bone, which takes 6 to 8 weeks, and the return of normal motion and gait. The brace is unlocked between 4 and 6 weeks after surgery for controlled gait training only. It is then unlocked for all activities during the 6- to 8-week period. Crutches and the brace are removed after 8 weeks if the patient exhibits good quadriceps strength and control, full knee extension, knee flexion of 90 to 100 degrees, and normal gait pattern. If the PLC was also reconstructed, the brace and crutches are continued for a total of 12 weeks after surgery. At 4 weeks, therapeutic exercises still include wall slides (0 to 45 degrees), which are gradually progressed to mini squats. At 8 weeks, a stationary bike is added with the heel forward on the pedal and the seat slightly higher than normal. Pool therapy is

also begun focusing on normal heel-to-toe gait pattern pool. Finally, balance and proprioception exercises such as single leg stances are introduced.

Phase 3 Phase 3 extends from 3 to 9 months after surgery. The patient is expected to achieve full pain-free range of motion, normal gait, and good quadriceps strength and should have no patellofemoral complaints. Obtaining the last 10 to 15 degrees of flexion may take up to 5 months. Exercises are advanced to jogging in the pool and walking on the treadmill. Closed chain kinetic exercises are continued throughout this period to improve functional strength and proprioception. Quadriceps strength and hamstring flexibility need to be maximized and maintained.

Phase 4 This period extends from about 9 to 12 months after surgery. The goal during this time is the gradual return to work and athletic participation as well as the maintenance of strength and endurance. This may involve sports-specific training, work hardening, or job restructuring as needed. Education is essential to provide the patient with a clear understanding of the possible limitations. Therapeutic exercises include cross-country ski machines, slide board, running and cutting skills, and jumping and plyometrics.

Outcomes Measurements There has been a growing interest in patient-orientated outcomes and evidence-based medicine (EBM). EBM is the conscientious, explicit, and judicious use of the current best evidence in making decisions about the care of the individual patient.249 The World Health Organization released the International Classification of Function, a model of functioning that provided a framework for identifying meaningful clinical outcomes.250 This has led to the International Knee Documentation Committee (IKDC) and the American Orthopedic Society for Sports Medicine to develop the IKDC Subjective Knee Form, a kneespecific health-related quality-of-life instrument that is appropriate for measuring symptoms, function, and sports activity for individuals with a variety of knee conditions, including ligamentous and meniscal injuries, patellofemoral pain, articular cartilage lesions, and arthritis.251 This instrument has been demonstrated to be reliable, valid, and responsive.252-254 This instrument is a common currency form and does not differentiate by age or sex. Recently, a Cochrane Collaboration of Systematic Reviews of 2005 did a meta-analysis of all randomized and quasi-randomized clinical papers on PCL treatment.255 Although 286 studies were found, none fit the inclusion criteria for randomized controlled studies. The study found only numerous relevant observational studies. It concluded that the lack of randomized controlled trials reflects the relative infrequency of these injuries and possibly the absence of a culture to perform multicenter randomized controlled trials. The study also concluded, based on observational data, that isolated PCL injuries may be treated conservatively with a good prognosis, and PCL

1712 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

injuries with other ­multiligament injuries are more likely to be treated surgically but have a more guarded prognosis because these are more extensive injuries.

Potential Complications Complications in the management of PCL injuries can be related to the initial injury, nonoperative treatment, intraoperative complications, or postoperative management. Complications related to the initial injury are usually related to multiligamentous knee injuries and not isolated PCL injuries. Neurovascular injuries occur in about 30% of knee dislocations243,256 (range, 15% to 49%).36,71,75-79 Complications from nonoperative management can include residual laxity, stiffness, knee pain, degenerative joint disease of the medial compartment and patellofemoral compartments, heterotopic ossification, and reflex sympathetic dystrophy.257-259 All these complications are more likely in multiligamentous knee injuries than in isolated PCL injuries. In addition, any one of these can also develop in operatively managed PCL injuries. Intraoperative complications of PCL surgery represent a unique component of PCL injuries and include neurovascular injuries, medial femoral condyle osteonecrosis, compartment syndrome, and tourniquet complications. Although compartment syndrome and tourniquet problems do happen, they are usually considered more likely in multiligamentous knee injuries. Neurovascular injuries, specifically popliteal artery injuries, are relatively unique to PCL surgery because of the proximity of the popliteal artery. Popliteal artery and tibial nerve injuries, although uncommon, are possible with both the inlay and transtibial techniques. In a recent cadaveric study, the popliteal artery is about 29.1 mm from the midportion of the PCL and 9.7 mm from the proximal PCL fovea.260 In another study, it was shown that the mean sagittal and coronal distances were 7.6 and 7.2 mm, respectively, and these distances increase with greater knee flexion angles up to 100 degrees.261 Other cadaveric studies have shown the safe posteromedial approach for a tibial inlay. In one study, the closest any screw was to the popliteal artery was 21.1 mm.262 There is anatomic variation of the popliteal artery, and it has been found to pass medial to or within the medial head of the gastrocnemius,263 so it is recommended that the popliteus be subperiosteally dissected off the tibia. Injuries have been reported when the surgeon strays out of the appropriate plane when doing inlay surgery. For transtibial techniques, neurovascular injuries have been documented in case reports as either lacerations or thrombus formation.264,265 Fanelli has described the posteromedial incision as both a working portal and safety incision.265a This incision allows for adequate visualization and ­ protection of neurovascular structures, and helps reduce compartment syndrome by acting as an outflow portal. In addition to the popliteal and tibial nerve, if PLC surgery is being performed, the peroneal nerve is at risk and should always be identified and protected. Other potential intraoperative complications include medial femoral condyle osteonecrosis and tibial fracture. Osteonecrosis is likely related to drilling femoral tunnels too close to the articular surface and damaging the single nutrient artery to the condyle.266,267

Common complications diagnosed in the postoperative course of PCL injuries include arthrofibrosis, anterior knee pain, and residual laxity. Most motion loss for PCL reconstruction is in flexion as opposed to extension. Causes of flexion loss include suprapatellar adhesions; improper tunnel placement; improper graft tensioning; multiple concurrent ligament procedures, especially open medial sided surgery; poor physical therapy compliance; and the nonisometric nature of PCL reconstructions.258 These can be treated with a manipulation or arthroscopic lysis of adhesion. Manipulation rates may be as high as 10% to 15% in the multiple-injury knee, whereas other series show a mean 10-degree terminal flexion loss.44,268,269 On the opposite spectrum, residual laxity is also quite common. In many cases, PCL reconstructions gradually loosen to grade I or II laxity. The reasons for this are multifactorial as well and include malalignment, missed concomitant injuries, and technical errors. Although most patients can tolerate some degree of residual laxity, reconstructed knees with grade III instability are failures and need to be revised. In a recent review of PCL surgical failures, the modes of failure were documented.270 The most common problem was PLC deficiency (40%), improper graft tunnel placement (33%), associated varus malalignment (31%), and primary suture repair (25%). Residual laxity may also pre­ sent as anterior knee pain as a result of the posterior sag and increased patellofemoral forces. Other causes of anterior knee pain include harvest morbidities, symptomatic hardware, and postoperative synovitis.

Criteria for Return to Play Using our four-phase PCL rehabilitation program, our criteria for return to play are quite predictable. Patients must exhibit full, pain-free range of motion, understanding that it is not unusual for flexion to be lacking 10 to 15 degrees up to 5 months after surgery. They must also have a normal gait pattern with normal to near-normal quadriceps control and appropriate hamstring flexibility. In addition, they should have no patellofemoral or soft ­tissue complaints. Finally, they need appropriate endurance and proprioception from sport-specific training. This usually takes 9 to 12 months depending on associated injuries. We do not require brace therapy for return to play, but occasionally, athletes are more comfortable with a functional brace when they first return to sport. C l

r i t i c a l

P

o i n t s

 nderstand the insertional anatomy of the three major comU ponents of the PCL: AL bundle, PM bundle, and MFLs. l Understand the anatomy of the posterolateral corner complex, including the LCL, popliteus, popliteofibular ligament, biceps femoris, and iliotibial band. l Understand the biomechanics of the PCL and PLC in terms of their synergistic actions. The PCL is the primary stabilizer to posterior tibial translation at 90 degrees of knee flexion and is the secondary stabilizer to external rotation. The PLC is the primary stabilizer to tibial external rotation at 30 degrees of knee flexion and is the secondary stabilizer to posterior tibial translation.

knee 1713

l Accurately determine what knee structures are injured by physical examination. In particular, be able to evaluate and document injuries to the PLC, LCL, and PCL. l Understand the natural history of isolated PCL injuries (grades I to III) as well as combined PCL and PLC injuries, including PCL deficiency and associated patellofemoral and medial compartment arthrosis. l Understand the various techniques of PCL reconstruction, including single- versus double-bundle reconstruction, tibial versus transtibial inlay, PCL augmentation, and graft choices, fixation types, and tensioning patterns.

S U G G E S T E D

Harner CD, Vogrin TM, Höher J, et al: Biomechanical analysis of a posterior cruciate ligament reconstruction: Deficiency of the posterolateral structures as a cause of graft failure. Am J Sports Med 28(1):32-39, 2000. LaPrade RF, Ly TV, Wentorf FA, Engebretsen L. et al: The posterolateral attachments of the knee: A qualitative and quantitative morphologic analysis of the fibular collateral ligament, popliteus tendon, popliteofibular ligament, and lateral gastrocnemius tendon. Am J Sports Med 31(6):854-860, 2003. Mariani PP, Becker R, Rihn J, Margheritini F: Surgical treatment of posterior cruciate ligament and posterolateral corner injuries: An anatomical, biomechanical and clinical review. Knee 10(4):311-324, 2003. Shelbourne K, Muthukaruppan Y: Subjective results of nonoperatively treated, acute, isolated posterior cruciate ligament injuries. Arthroscopy 21:457-461, 2005. Vogrin TM, Hoher J, Aroen A, et al: Effects of sectioning the posterolateral structures on knee kinematics and in situ forces in the posterior cruciate ligament. Knee Surg Sports Traumatol Arthrosc 8(2):93-98, 2000. Woo SL, Vogrin TM, Abramowitch SD: Healing and repair of ligament injuries in the knee. J Am Acad Orthop Surg 8(6):364-372, 2000.

R E A D I N G S

Amis AA, Bull, AMJ, Gupte, CM, et al: Biomechanics of the PCL and related structures: Posterolateral, posteromedial and meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc 11(5):271-281, 2003. Amis AA, Gupte CM, Bull AMJ, Edwards A: Anatomy of the posterior cruciate ligament and the meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc 14(3):257-263, 2006. Fontbote C, Sell TC, Laudner KG, et al: Neuromuscular and biomechanical adaptations of patients with isolated deficiency of the posterior cruciate ligament. Am J Sports Med 33:982-989, 2005. Harner CD, Baek GN, Vogrin TM, et al: Quantitative analysis of human cruciate ligament insertions. Arthroscopy 15(7):741-749, 1999.

R e f erences Please see www.expertconsult.com

S e c t i o n

E

Posterior Cruciate Ligament Injuries 2. Posterior Cruciate Ligament Injuries in the Child Nicholas J. Honkamp, Anil Ranawat, and Christopher D. Harner

Knee injuries in children are generally less frequent than in adults. However, with the progressive increase in the number of young people participating in organized sports at younger ages, physicians are seeing knee injuries occurring more frequently.1,2 It has been estimated that more than 30 million children and adolescents in the United States now participate in organized athletics.3 Sports participation is beginning at earlier ages with greater frequency of participation and higher intensity. Furthermore, this increased intensity and participation of elite or highly skilled preadolescent athletes in competitive sporting programs have also been implicated in the growing frequency of knee injuries.4 The level of commitment and the ­ intensity of ­ training

required to reach this level of athletic prowess put the skeletally immature knee at considerable risk. In pediatric athletes, the knee and ankle are at greatest risk for injury.3,5 Cutting sports such as football, soccer, basketball, and volleyball represent the highest-risk sports for knee injury.6,7 Causes other than sporting events include traffic accidents and recreational activities, with more injuries occurring during the afternoon when children have more free time.2 Acute ligamentous injury to the knee and, in particular, to the cruciate ligaments in children is being diagnosed more frequently.8,9 This observation has been attributed to a better understanding of and greater clinical suspicion

1714 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

for ligamentous tears as well as to our ability to diagnose these injuries with improved imaging and arthroscopic techniques.8,10

ANATOMY OF THE YOUNG ATHLETE AND ASSOCIATED KNEE INJURIES Children present unique challenges related to their different physical and physiologic characteristics; children are not merely small adults. Such differences include their larger heads in relation to their bodies, less developed motor coordination, increased incidence of falls and accidents including traffic accidents, and their intense physical activity during play and sports.2,3 There are important physiologic and biomechanical differences that relate specifically to the pediatric knee.2 The presence of open cartilaginous growth plates, increased bone porosity and pliability, unique musculotendinous apophyseal insertions, and growing articular cartilage lead to a different spectrum of injuries in children and adolescents than in adults. In addition, the child is lighter and has a lower center of gravity, shorter lever arms, and decreased muscle strength, thereby considerably reducing the magnitude of forces generated across the lower extremities. These factors, combined with the relatively greater strength of the ligaments compared with the physes, generally protect the pediatric knee from ligamentous injury. Unfortunately, however, growing articular cartilage has been shown to be more susceptible to injury, in both clinical and biomechanical studies, than mature cartilage.11 Knee injuries in children may involve the ligaments, the extensor mechanism, the menisci, the articular cartilage and subchondral bone, the epiphysis, and adjacent structures. The age of the athlete and the anatomic insertion of the ligament (i.e., metaphyseal versus epiphyseal) will determine whether physeal or ligamentous injury occurs (Fig. 23E2-1). Physeal and ligamentous injuries occur with relatively equal frequency in children between 7 and 11 years of age, whereas younger children are more likely to sustain metaphyseal fractures.12 Teenagers sustain ligament injuries with low-energy trauma and physeal fractures with high-energy trauma.12,13 In general, the physis is more likely to be injured during times of rapid growth (i.e., peak height velocity during puberty). Multiple reports have been published dealing with the evaluation and diagnosis of acute knee injuries with associated effusions in children and adolescents.10,14 In general, most (75% or more) involve injuries to the anterior cruciate ligament (ACL), medial or lateral menisci, and osteochondral surfaces. Less frequent causes include collateral ligament and posterior cruciate ligament (PCL) injuries.2,10,14

POSTERIOR CRUCIATE LIGAMENT INJURY The PCL is widely recognized as the primary restraint to posterior tibial translation and is a secondary restraint to external rotation. Although ligamentous disruptions of the ACL are being reported with increasing frequency in the

Femoral epiphyseal plate Fibular collateral ligament Lateral capsular ligament

Fibular epiphyseal plate

Tibial collateral ligament

Medial capsular ligament Tibial epiphyseal plate

Figure 23E2-1   The attachment of the cruciate ligaments occurs within the epiphysis of the femur and the tibia. The medial collateral ligament is the only ligament to cross the tibial physeal plate. Because of the ligaments’ relationship to the physeal plates and their relative strength, stress concentrates at the growth plates, producing physeal injury rather than ligament failure.

pediatric population,8,15 PCL injuries in this age group are extremely uncommon. A thorough review of the literature reveals only sporadic case reports of PCL injuries in the skeletally immature knee.3,16-28 Most of the reports of PCL injuries in children involve avulsions from either the tibial or the femoral attachments, with femoral avulsions being more common. The femoral attachment of the PCL has been reported to be the weakest at the chondro-osseous junction.23,27 Because this attachment site has not yet ossified, avulsion from the femur may not be appreciated on plain radiographs. This has important ramifications for treatment because the physician must be aware of this pattern of avulsion injury (with adjacent periosteum or perichondrium), which is amenable to a successful primary repair if recognized early.

Mechanism of Injury Three mechanisms have been proposed for ­ ligamentous disruption of the PCL: (1) direct pretibial trauma, (2) hyperflexion, and (3) hyperextension (Fig. 23E2-2). Pretibial trauma (i.e., a posteriorly directed blow on the anterior aspect of the proximal tibia) is a commonly cited cause of PCL injury. For example, an athlete falling on a flexed knee with the foot in plantar flexion is at risk for tearing the PCL. If the foot is dorsiflexed, however, the force is transmitted proximally through the patella and the distal femur, thereby protecting the PCL from injury.29 Noncontact injuries, such as forced hyperflexion, have been reported to be the most common isolated PCL injuries in athletes.30 These injuries often result in only partial tearing of the PCL, with the posteromedial fibers remaining intact. Hyperextension injuries are often combined with varus or valgus forces to result in multiligament disruptions, which have a much more guarded prognosis.

knee 1715

A

B

C

Figure 23E2-2   Three mechanisms of posterior cruciate ligament injury. A, Direct pretibial trauma. B, Forced hyperflexion.   C, Hyperextension. (From Miller MD, Harner CD, Koshiwaguchi S: Acute posterior cruciate ligament injuries. In Fu FH, Harner CD, Vince KG (eds): Knee Surgery, vol 1. Baltimore, Williams & Wilkins, 1994.)

Classification PCL injuries in both children and adults can be classified according to severity (grades I to III), timing (acute versus chronic), and presence of associated injuries (isolated versus combined).31 These variables have significant implications for patient outcome and thus are important to consider when making treatment decisions. Isolated injuries to the PCL can be classified as partial (grades I or II) or complete (grade III) tears. In most cases, this is done clinically and corresponds to the laxity in the PCL, as measured by the step-off between the medial tibial plateau and the medial femoral condyle. Isolated grade III injuries or complete PCL tears can occur, but they are frequently associated with other ligament injuries, in particular injury to the posterolateral structures. Distinguishing between isolated and combined PCL injuries is critical because the prognosis and treatment of these injuries are vastly different (see Chapter 23E1 on adult PCL injuries).

Evaluation An accurate and detailed history and physical examination are essential to arriving at the correct diagnosis and formulating an appropriate treatment plan. Those physicians who deal frequently with children know the difficulty of accurately diagnosing pediatric musculoskeletal conditions. Young patients may not remember the mechanism of injury or may have difficulty describing it and the details of their associated joint complaints. Their symptoms are frequently vague and generalized. In general, they may limit the accuracy of the knee history and physical examination through a lack of cooperation or inconsistent history and physical examination findings.32-35 Children younger than 6 to 7 years of age are usually unable to localize pain reliably. Their manifestation of a knee disorder may be a limp or refusal to walk.11 Coaches or parents may provide helpful information, but often the injury occurs when the child is unsupervised. Thus, the accuracy of preoperative knee diagnoses varies with the patient age, with the lowest percentages occurring in the preadolescent age group (18% to 55%) and improved ­percentages

occurring in the adolescent age group (44% to 70%).36-39 This makes a good physical examination even more important, which can be difficult if the child is frightened or in pain. It is very important to gain the child’s trust before proceeding with the examination and to remember that pain from hip injuries may be referred to the medial aspect of the knee as a result of the common sensory supply by the obturator nerve. Evaluation of the injured knee begins with obtaining a detailed history, trying to delineate the mechanism of injury, its severity, and possible associated injuries. Patients with PCL injuries may present for evaluation in a variety of different scenarios. Injuries may range from a seemingly benign fall on the athletic field to severe trauma caused by a motor vehicle crash. The more acutely traumatized the knee, the more difficult it will be to examine it. Unlike patients with isolated ACL injuries, those with acute isolated PCL injuries do not typically report hearing or feeling a “pop.” Although many suspect a knee injury, patients do not typically relate a sense of instability. They may note mild to moderate swelling, accompanying stiffness, and occasionally mild knee pain. The examination of the acutely injured knee begins with evaluation of the neurovascular status, followed by observation of the knee for its resting position and the presence and location of any ecchymosis. The examiner must differentiate between intra-articular effusion and extra-articular swelling. Acute hemarthrosis after trauma to the skeletally immature knee alerts the clinician to a significant intraarticular injury. As mentioned earlier, the most likely diagnoses in such cases include ACL, meniscal, or osteochondral injuries. All important anatomic structures of the knee should be palpated sequentially, leaving those areas most likely to be tender (according to the suspected diagnosis) until last. Stress maneuvers to test ligament integrity in the mediolateral and anteroposterior planes may be performed gently. Comparison with the noninjured extremity is crucial because of the physiologic laxity present in many children. If there is significant tenderness over a growth plate, it may be prudent to obtain radiographs before stressing the knee,

1716 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

which might displace a fracture. Finally, range of motion of both the knee and the hip (to rule out primary hip disease) is checked near the end of the examination once the patient’s trust has been gained and the rest of the examination has been completed. In any child with ligamentous instability of the knee, congenital absence of the cruciates should be considered. The most common associated causes include proximal femoral focal deficiency, fibular hemimelia, congenital dislocation of the knee, and ball-and-socket ankle. Congenital absence of the cruciates may be an isolated finding40; however, radiographs may show flattening of the tibial eminence or a shallow intercondylar notch.40,41 Finally, hemophilia must be also be considered in the differential diagnosis of a child presenting with hemarthrosis and a history of no or minimal trauma.42

Imaging Imaging studies play an important role in the diagnosis of PCL injuries in children. Plain radiographs may detect avulsion fractures of the femoral and tibial attachments of the PCL. As previously mentioned, the attachment site may not yet have ossified; thus, avulsion of the PCL (especially from the femur) may not be appreciated on plain films. The only clue may be a slight irregularity at the femoral or tibial attachment, which could easily be missed unless the radiographs are carefully scrutinized.43 Before the widespread availability and use of magnetic resonance imaging (MRI), knee arthroscopy was used as a diagnostic and therapeutic tool.10,38,44,45 Unfortunately, multiple authors reported that a significant number of arthroscopies yielded normal results, thus exposing patients to unnecessary surgery.36,38 MRI is extremely useful for evaluating PCL injuries in the skeletally immature knee. It can accurately differentiate between intrasubstance and “peel-off” injuries and can determine whether there is any associated chondral or meniscal disease.31 However, MRI should not be considered a substitute for a thorough history and physical examination. Two recent studies have compared the radiologist’s interpretation of MRI diagnosis and clinical diagnosis of the treating physician (without knowledge of the MRI result) to that of the findings at the time of the diagnostic arthroscopy.46,47 Stanitski47 found that the clinical examination was more accurate than the MRI, whereas Kocher and colleagues46 found no significant differences between clinical examination and the findings on MRI with respect to agreement with the arthroscopic findings. Luhmann and coworkers48 found that the more accurate diagnoses were obtained when the treating physician who performed the physical examination also reviewed the MRI personally. Thus, a history and physical examination, as well as judicious use and personal review of MRI results, are the best method for arriving at a correct diagnosis.

Natural History Owing to the rarity of PCL injury in children and the limited short-term follow-up that has been reported in the literature, the consequences of PCL deficiency in

the ­skeletally immature knee (with or without treatment) remain unknown. Unfortunately, the literature regarding adult PCL injuries is not extremely helpful because most of the studies are retrospective, combining both isolated and multiligamentous injuries, and do not stratify the outcome according to the degree of instability. There is a general consensus that PCL injuries (especially grade III) are not as benign as previously thought over the long term.31 Patients with chronic PCL injuries experience a variable progression of articular degeneration and symptoms over time. A high incidence of late chondrosis29,49 (involving the medial femoral condyle and the patellofemoral joint) and meniscal tears50 has been noted in adult patients treated nonoperatively. These findings are likely the result of the increased contact pressure that occurs in these compartments after PCL disruption.51,52 The fact that growing articular cartilage is more susceptible to injury than mature articular cartilage is particularly worrisome.11 The problem to date has been that investigators have been unable to identify prognostic factors consistently to help predict patient outcome after PCL injury. It should also be noted that the long-term outcome after nonsurgical treatment of ACL injuries in the skeletally immature has been poor with respect to return to sports and long-term sequelae.53-55

Treatment Rationale The treatment of acute PCL injuries in children is ­dependent on both the pattern of ligamentous injury and whether there is any associated meniscal or chondral disease. The full extent of the injury must be determined before formulation of a treatment plan because the site of injury (avulsion or midsubstance), its grade (partial or complete), and the presence of associated injuries greatly influence the treatment algorithm. Furthermore, both the child and the parents must be actively involved in the decision-making process. The expectations of both the patient and the family, as well as the maturity of the patient and his or her commitment to rehabilitation, must be taken into ­ consideration before embarking on any surgical ­intervention.

Avulsion Injuries PCL avulsions (soft tissue or bony) from the femur or the tibia should be repaired primarily. If the child suffered a hyperflexion injury, an avulsion should be suspected; this can be confirmed with MRI of the knee. Arthroscopy and examination with the patient under anesthesia remain the most accurate means of determining the extent of the child’s injury, however, because partial ligament tears can be difficult to distinguish, even with MRI, in the skeletally immature knee.11 Furthermore, arthroscopy will confirm the location of PCL disruption and the feasibility of repair as well as assist in evaluating the menisci and the chondral surfaces for injury. Meniscal repair should be attempted whenever possible, and PCL avulsions from the femur or the tibia should be repaired primarily with transosseous (intraepiphyseal) sutures through drill holes or, for bony

knee 1717

Bunnell suture in PCL

PCL

A

B

C

Figure 23E2-3   A, Posterior cruciate ligament (PCL) avulsion from the femur repaired with a Bunnell-type stitch through the femoral epiphysis. B, PCL avulsion from the tibia repaired with a Bunnell-type stitch through the tibial epiphysis. C, PCL bony avulsion repaired with an intraepiphyseal screw.

avulsions, with either screw or transosseous suture fixation (Fig. 23E2-3).

Midsubstance Injuries Isolated midsubstance cruciate tears are generally not repaired in children because the outcome has not been proved to be any better than in adults.16 Permanent plastic deformation and early degeneration of the ligament may contribute to failure of the repair. Although transphyseal ACL reconstructions with soft tissue grafts (i.e., hamstring) are being performed at some centers,56-58 PCL reconstruction in the skeletally immature knee may be contraindicated. Animal studies have shown that the tibial physis can be very sensitive to transphyseal drilling.59,60 The tibial tunnel in PCL reconstruction crosses the physis peripherally, in comparison with the more central location of an ACL tunnel, thereby theoretically posing a greater risk for physeal injury or closure. Conservative treatment is currently recommended for skeletally immature patients with an isolated grade III midsubstance PCL tear. A recent study has suggested that interstitial tears of the PCL may have some propensity to heal with closed management, particularly those with less than 8 mm of posterior displacement on stress radiographs.61 Initial treatment should include immobilization of the knee in extension with anterior translation to reduce posterior sag. Restoration of range of motion and quadriceps and hamstring strength starts at 4 to 6 weeks. The patients must be followed annually. If functional instability or pain develops, or if radiographs or bone scans are notable for early signs of arthrosis, reconstruction of the PCL can be performed after growth is completed.

Multiligament Injuries In the presence of multiligament injuries, the collateral ligaments are repaired surgically along with any associated meniscal disease. In this scenario, attempting to repair a midsubstance tear of the PCL with suturing may be indicated if there is significant growth remaining. If the resultant posterior laxity is clinically significant, a PCL reconstruction can be performed in the future. In cases in which physeal growth is limited (<1 cm), an acute reconstruction can be performed to stabilize the knee. If a nonoperative approach is preferred by the treating surgeon, the knee can be immobilized in a cast in full extension to reduce posterior sag. It is critical to obtain radiographs immediately after immobilization and at regular intervals to ensure that the knee remains reduced. Restoration of range of motion and quadriceps and hamstring strength will begin after 4 to 6 weeks. Late reconstruction can be performed in the future after physeal closure if the patient is symptomatic. C l

r i t i c a l

P

o i n t s

 hildren are not merely small adults. Their unique C ­anatomy—the presence of open cartilaginous growth plates, increased bone porosity and pliability, unique musculotendinous apophyseal insertions, and growing ­articular cartilage—leads to a different set of injury patterns not typically seen in the adult. l The femoral attachment of the PCL has been reported to be the weakest link in PCL anatomy. Because this attachment site has not yet ossified, avulsion from the femur

1718 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� may not be appreciated on plain radiographs. Thus, if a PCL injury is suspected in a child, additional imaging such as an MRI scan should be sought. l Three mechanisms have been proposed for ligamentous disruption of the PCL: (1) direct pretibial trauma, (2) hyperflexion, and (3) hyperextension. l It is critical to grade PCL injuries and to distinguish between isolated and combined PCL injuries because the prognosis and treatment of these injuries are vastly different. l Evaluation of child with a knee injury can be difficult. Parental assistance with the history, the presence of an effusion, and a thorough physical examination including that of the ipsilateral hip and knee are all important. l Treatment of pediatric PCL injuries can be difficult and should be directed by both the severity of the PCL injury and the presence of additional knee injuries. l Avulsion fractures should be operatively treated and generally have a good prognosis. Treatment of midsubstance PCL injuries is controversial, but nonoperative treatment is generally preferred at this time because of the presence of open physes.

S U G G E S T E D

R E A D I N G S

Kocher MS, DiCanzio J, Zurawski D, et al: Diagnostic performance of ­ clinical examination and selective magnetic resonance imaging in the evaluation of ­intraarticular knee disorders in children and adolescents. Am J Sports Med 29(3):292-296, 2001. Luhmann SJ, Schootman M, Gordon JE, Wright RW: Magnetic resonance imaging of the knee in children and adolescents. Its role in clinical decision-making. J Bone Joint Surg Am 87(3):497-502, 2005. Matelic TM, Aronsson DD: Acute hemarthrosis of the knee in children. Am J Sports Med 23(6):668-671, 1995. Stanitski CL, Harvell JC, Fu F: Observations on acute knee hemarthrosis in children and adolescents. J Pediatr Orthop 13(4):506-510, 1993. Sullivan JA: Ligamentous injuries of the knee in children. Clin Orthop 255:44-50, 1990. Vahasarja V, Kinnuen P, Serlo W: Arthroscopy of the acute traumatic knee in children: Prospective study of 138 cases. Acta Orthop Scand 64(5):580-582, 1993.

R e f erences Please see www.expertconsult.com

S e c t i o n

F

Lateral and Posterolateral Injuries of the Knee Andrew J. Schorfhaar, Jeffrey J. Mair, Gary B. Fetzer, Brett W. Wolters, and Robert F. LaPrade

Injuries involving the lateral and posterolateral aspect of the knee are much less common than injuries to the medial knee structures and cruciate ligaments. The incidence of lateral and posterolateral injuries is difficult to accurately determine because many of these injuries remain undetected at the time of initial evaluation and treatment. In recent years, extensive research has been published that has increased our understanding of the anatomy, biomechanics, and early surgical treatment outcomes of injuries to the posterolateral corner (PLC) of the knee. For this reason, posterolateral rotatory instability is becoming more widely recognized as a cause for residual symptoms and failure of anterior cruciate ligament (ACL) or posterior cruciate ligament (PCL) reconstruction grafts. In order to appropriately recognize these injuries, a thorough understanding of the anatomy and biomechanical function is important. In this chapter, we discuss the interrelationship of the PLC anatomy and biomechanics with surrounding structures of the knee. We review clinical presentation and pertinent

diagnostic tools to aid in the classification of these injuries. In addition, we review treatment options, postoperative management, and early data regarding outcomes of injuries to the PLC.

ANATOMY The anatomy of the lateral and posterolateral aspects of the knee is complex and variable.1-8 In addition to the complex anatomy, confusion has been created because of the reported variability of occurrence of the different structures and competing nomenclature used in the literature (Fig. 23F-1).5 Knowledge of this anatomy is an important precursor to understanding the respective biomechanical function and significance, to be able to identify and interpret structures on magnetic resonance imaging (MRI), and to repair or reconstruct these structures at the time of surgery. Each of the important structures of the lateral and posterolateral knee complex is discussed below.

knee 1719

Lateral gastrocnemius tendon

Fibular collateral ligament

Popliteus tendon

A

Popliteofibular ligament

B Figure 23F-1   Clinical photograph (A) and schematic drawing (B) demonstrating the course and relationship of the fibular collateral ligament, popliteus muscle-tendon complex, and popliteofibular ligament on the lateral side of the knee. (Adapted from LaPrade RF, Ly TV, Wentorf FA, Engebretsen L: The posterolateral attachments of the knee: A qualitative and quantitative morphologic analysis of the fibular collateral ligament, popliteus tendon, popliteofibular ligament, and lateral gastrocnemius. Am J Sports Med 31[6]:854-860, 2003.)

Iliotibial Band (Tract) The iliotibial band (ITB) extends proximally from its main distal insertion on Gerdy’s tubercle on the proximal tibia.3,5,6,9-13 The ITB is composed of four main structures.5,6,10 The main component of the ITB is the superficial layer. It covers a large portion of the lateral knee and is the first layer encountered after dissecting through the subcutaneous tissue of the lateral knee. An anterior expansion of the superficial layer is known as the iliopatellar band.6,10 These latter fibers attach to the lateral border of the patella and are important in patellofemoral tracking. The ITB also has a distinct deep layer that attaches the medial aspect of the superficial layer to the distal aspect of the lateral intermuscular septum of the thigh. The fourth component of the ITB is the capsulo-osseous layer.5,6,10 It begins proximally from the region of the distal lateral intermuscular septum and blends with a confluence of fascia from the short head of the biceps femoris muscle and the lateral gastrocnemius tendon. This capsulo-osseous layer and portions of the deep layer extend distally to attach to the anterolateral aspect of the tibia just posterior and proximal to Gerdy’s tubercle. The entire ITB complex has been described to act as an anterolateral sling to the knee.5,6,10 The clinical significance of this complex is that it is considered a secondary stabilizer of the lateral side of the knee, preventing increasing varus opening of the knee; however, marked laxity of the knee can be present with an

intact ITB.5,6,10,14 Although the superficial layer of the ITB is seldom injured with posterolateral knee injuries, it serves as an important surgical reference for many other structures of the lateral and PLC of the knee.15

Fibular (Lateral) Collateral Ligament The fibular collateral ligament (FCL) is the primary static stabilizer to varus opening of the knee (Fig. 23F-2).13,16-18 The main proximal attachment site is located in a small bony depression slightly proximal and posterior to the lateral femoral epicondyle, with some fibers extending anterior over the lateral epicondyle in a fan-like fashion (see Fig. 23F-1).5,6,19 The average femoral attachment site is 1.4 mm proximal and 3.1 mm posterior to the lateral femoral epicondyle.19 The distal insertion is onto the head of the fibula at the apex of a superior- and lateral-facing Vshaped plateau.6,19,20 Through its course distally, it passes medial (deep) to the superficial layer of the ITB and the anterior arm of the long head of the biceps. The direct arm of the long head of the biceps femoris overlaps the fibular attachment of the fibular collateral ligament, which inserts at the distal rim of this plateau.20 A fibular collateral ligament–biceps femoris bursa has been described to exist between the fibular collateral ligament and the anterior arm of the long head of the biceps femoris (Fig. 23F-3A).7 Cadaveric studies have found that the average total length

1720 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

of the fibular collateral ligament between attachment sites is about 70 mm.7,19,20 The FCL attaches an average of 8.2 mm posterior to the anterior aspect of the fibular head and 28.4 mm distal to the tip of the fibular styloid.19 Clinically, the FCL is a vital stabilizer to varus stress, and knowledge of the proximal and distal bony insertions is important during repair or reconstruction.5,15

Popliteus Muscle-Tendon Complex and Popliteofibular Ligament

Figure 23F-2   Fibular collateral avulsion off the femoral origin with suture in the fibular collateral ligament remnant (lateral view, left knee).

A

C

The popliteus muscle-tendon complex has many components that provide both static and dynamic stability to the posterolateral aspect of the knee.5,6,8,19,21 The anatomy and biomechanics of the popliteus are complex and are well described in the literature.5,6,8,15,19-21 Distally, the popliteus muscle originates from a broad muscular insertion at the posteromedial surface of the proximal tibial metaphysis. Proximally, the main tendinous attachment is at the proximal half and anterior one fifth of the popliteus sulcus of the femur, about 18.5 mm anterior and inferior to the femoral insertion of the fibular collateral ligament (Fig. 23-4).6,8,19 The popliteus tendon traverses deep to the

B

Figure 23F-3   Surgical approach. A, The iliotibial band and long head of the biceps are landmarks as the biceps bursa is seen between the hemostat ends. B, Midsubstance of the fibular collateral ligament is tagged, and a tug test is performed to determine the origin and course of the fibular collateral ligament. C, A cannulated aiming guide is placed anterior and distal to Gerdy’s tubercle to the posterior popliteal sulcus. The trocar is placed within the fibular tunnel at the attachment site of the fibular collateral ligament.

knee 1721

LGT origin

18.5 mm

FCL-femur Lateral epicondyle Popliteus sulcus

PLT

to the apex and down-slope of the posteromedial fibular styloid process, and the anterior division attaches to the down-slope of the anteromedial aspect of the fibular styloid process.8,19 The posterior division has generally more substance and biomechanically provides more static stability to the PLC than the anterior division.5 Clinically, the PFL has been found to be an important stabilizer of external rotation of the knee.6,15 Distal to the musculotendinous junction of the popliteus muscle, a broad aponeurotic attachment extends to attach to the posterolateral joint capsule, coronary ligament, and posterior horn of the lateral meniscus.5,6,23 Clinically, this attachment reinforces and stabilizes the lateral meniscus and posterior knee capsule.

Long and Short Heads of the Biceps Femoris Fibular styloid FCL-fibula

Figure 23F-4   Schematic drawing of the lateral aspect of the knee demonstrating the distance between the femoral attachment sites of the popliteus tendon (PLT) and the fibular collateral ligament (FCL). (Adapted from LaPrade RF, Ly TV, Wentorf FA, Engebretsen L: The posterolateral attachments of the knee: A qualitative and quantitative morphologic analysis of the fibular collateral ligament, popliteus tendon, popliteofibular ligament, and lateral gastrocnemius tendon. Am J Sports Med 2003;31:854-860.)

FCL to its insertion on the lateral femoral condyle. The popliteus tendon does not fully engage into the popliteal sulcus until the knee is flexed to 112 degrees.19 Additionally, there is a small sinusoidal indentation of the articular cartilage of the lateral femoral condyle made by the popliteus tendon when the knee is in full extension (sulcus sartorius of Fürst).8 As the tendon courses distally, three popliteomeniscal fascicles (anteroinferior, posterosuperior, and posteroinferior) extend to the neighboring capsule and lateral meniscus (Fig. 23F-5).6,8 These fascicles are described to form the boundaries of the popliteal hiatus and are believed to provide stability to motion of the lateral meniscus.6,8,15,22 It is thought that the anteroinferior fascicle is the most important attachment of the popliteus complex to the lateral meniscus.22 Distal to the popliteomeniscal fascicles, the popliteofibular ligament (PFL) originates from the popliteus muscle-tendon complex. The PFL branches from the musculotendinous junction and attaches on the posterior aspect of the fibular styoid.4,6,8,19,21 There are two divisions of the PFL, an anterior division and a posterior division. These two divisions form an inverted Y, providing a fibular fixation point for the lateral and inferior quadrant of the ­ popliteus tendon. The posterior division attaches

The long head of the biceps femoris has six components and has been extensively described in the literature.5,6,7,24,25 About 1 cm proximal to the fibular head, the common tendon of the long head of the biceps femoris divides into two tendinous components: a direct arm and an anterior arm. The most important of these is the direct arm attachment to the posterolateral aspect of the fibular head. The anterior arm has two distal attachment sites. A portion of it inserts onto the fibular head distally and laterally to the direct arm attachment. However, most of the tendon continues anterior and distal, where it crosses lateral (superficial) to the fibular collateral ligament, and terminates as an anterior aponeurosis over the anterior compartment of the leg. The location at which the anterior arm crosses the FCL is an important surgical reference point because a small horizontal incision 1 cm proximal to the fibular head (location of the fibular collateral ligament–biceps femoris bursa) allows for direct access to the FCL and its fibular attachment.5 There are also four fascial components: an anterior aponeurosis, a lateral aponeurosis, a reflected arm, and a distal fascial expansion. The anterior aponeurosis is a continuation of the anterior arm as previously described. The lateral aponeurosis expansion extends from the anterior edge of the anterior arm and attaches to the posterior and lateral aspect of the distal FCL. The reflected arm extends from the tendon of the long head, crossing over the short head and to the posterior edge of the ITB (superficial layer). The distal fascial expansion extends from the posterior aspect of the long head of the biceps femoris tendon to the lateral gastrocnemius. The short head of the biceps femoris also has six major distal attachments at the knee.5,6,15,24 The most proximal attachment is a muscular attachment to the common tendon of the biceps femoris. There are two tendinous insertions: a direct arm and an anterior arm. The direct arm inserts onto the superior surface of the fibular head lateral to the tip of the styloid process and medial to the insertion of the direct arm of the long head of the biceps femoris. The anterior arm courses anterior, just proximal to the fibular styloid attachment, passes medial (deep) to the FCL, and inserts on the tibia about 1 cm posterior to Gerdy’s tubercle, along with the meniscotibial portion of the midthird lateral capsular ligament. The capsular arm is a broad capsular attachment that extends from the main tendon of

1722 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

2 1

3 5

7 6

A

4

B

C

Figure 23F-5   Lateral dissection (right knee). A, Rectangle indicates area enlarged in drawing. B, Posterolateral corner structures: 1, popliteus tendon; 2, fibular collateral ligament; 3, sulcus sartorius; 4, meniscotibial ligament; 5, inferior popliteomeniscal fascicle; 6, popliteofibular fascicle (anterior limb); 7, popliteofibular fascicle (posterior limb). C: Arthroscopic view of popliteal hiatus: *, lateral meniscus; white arrow, anteroinferior popliteomeniscal fascicle; blue arrow, popliteus tendon; double black arrow, posterosuperior popliteomeniscal fascicle; single black arrow, popliteofibular ligament. (A and B, Adapted from Stäubli HU, Birrer S: The popliteus tendon and its fascicles at the popliteal hiatus: Gross anatomy and functional arthroscopic evaluation with and without anterior cruciate ligament deficiency. Arthroscopy 6:209-220, 1990.)

the short biceps to the posterolateral aspect of the knee capsule, a bony fabella or cartilaginous fabella analogue, and to the tip of the fibular styloid. This capsular arm is one of the main components of the biceps femoris complex.5 Some reports in the literature have previously referred to this structure as part of the arcuate ligament.1,5,8 The distal edge of the capsular arm of the short biceps is the fabellofibular ligament.6 There also is a fascial confluence with the capsulo-osseous layer lateral to the ITB, which contributes to the anterolateral sling. This confluence extends distally to attach on the proximal and lateral tibia, just posterior to Gerdy’s tubercle. The final distal attachment of the short biceps is a lateral aponeurotic expansion off the short head of the biceps muscle attached to the posteromedial aspect of the fibular collateral ligament. Clinically, the long and short heads of the biceps femoris provide both static and dynamic stability to the lateral and PLC of the knee.5,13,16,17 In posterolateral knee injuries, the most commonly injured structures of the long biceps are the two tendinous components.5 It is important to reconstitute their attachment to restore their dynamic contribution to posterolateral stability. With respect to the short head of the biceps femoris, the capsular arm, the anterior arm, and the direct arm are most important clinically.5 The capsular arm is important because of its broad fascial attachment throughout the posterolateral knee. The anterior arm of the short head of the biceps femoris is frequently found to be avulsed with Segond’s fracture or Segond’s soft tissue injury.26 It has also been noted clinically that the distal attachments of the biceps femoris complex to the fibula are frequently avulsed off in posterolateral knee injuries.5

Additional Structures The mid-third lateral capsular ligament is a thickening of the lateral capsule. It has been described to be similar to the deep medial collateral complex.6,15 There are two components of this capsular thickening: the meniscofemoral component and the meniscotibial component. The meniscofemoral component extends from the lateral meniscus to an area just posterior to the lateral femoral epicondyle, and the meniscotibial component extends from the lateral meniscus to the tibia, just posterior to Gerdy’s tubercle. The meniscotibial component is more commonly injured and is associated with a bony avulsion (Segond’s fracture).25,26 The fabellofibular ligament, by definition, is the distal edge of the capsular arm of the short head of the biceps femoris.5,6 The fabellofibular ligament inserts along the lateral edge of the fabella (or fabella analog) and attaches just lateral to the fibular styloid.6 The clinical significance of the fabellofibular ligament is that it appears to be important for providing stability of the knee near full extension. However, no biomechanical studies have been performed to evaluate this hypothesis.5 The coronary ligament of the lateral meniscus is the meniscotibial portion of the posterior joint capsule.6,8,23 It anchors the posterior horn of the lateral meniscus between the tibial PCL attachment site and the posteroinferior popliteomeniscal fascicle. The coronary ligament is reinforced along its entire course by the aponeurotic attachment of the popliteus muscle (previously discussed). Clinically, it is important to provide restraint to hyperextension and posterolateral rotation of the tibia.5

knee 1723 10

PCL + LCL + deep (n-9) Posterior translation (mm)

Varus rotation

30˚

LCL + deep (n-10)

20˚

10˚

PCL + LCL + deep (n-8)

8 LCL + deep (n-10) 6 4 2 0

Intact (n-17) 0˚

0˚

15˚

A

External rotation

40˚

30˚ 45˚ 60˚ Flexion angle

75˚

90˚

B

LCL + deep (n-10)

30˚

20˚

0˚

0˚

15˚

30˚ 45˚ 60˚ Flexion angle

75˚

90˚

PCL + LCL + deep (n-6)

10˚

C

Intact (n-18) –2

Intact (n-17)

0˚

15˚

30˚ 45˚ 60˚ Flexion angle

75˚

90˚

The lateral gastrocnemius tendon forms at the far lateral edge of the lateral gastrocnemius muscle belly.5,6 Through its course proximally, it is strongly adherent to the posterior knee capsule. It has also been found to be consistently attached to the fabella, or fabella analogue.6 Its femoral attachment is at or near the supracondylar process of the distal femur, about 13.8 mm posterior to the FCL femoral attachment.19 Clinically, the lateral gastrocnemius is rarely injured in posterolateral knee injuries, so it can serve as a useful reference point for the FCL attachment on the femur during surgical ­dissection.5,27 The arcuate ligament complex is a variable combination of structures that is not a single distinct ligament, but several previously named structures that combine to form an arched (arcuate) appearance. There is considerable confusion in the literature describing this complex, and some authors have recommended the term arcuate ligament not be used, rather referring to the individual names of the structures.5 However, this Y-shaped complex possesses medial and lateral limbs, which cross over the popliteus muscle at the musculotendinous junction.6 The lateral limb consists of the PFL, fabellofibular ligament, and capsular arm of the short head of the biceps femoris. The medial limb is formed by the oblique popliteal ligament (ligament of Winslow). The arcuate ligament complex reinforces the other PLC structures of the knee.

Figure 23F-6   Cutting studies showing motion versus knee flexion angle relative to contribution of the posterolateral structures and posterior cruciate ligament with respect to varus rotation (A), posterior translation (B), and external rotation stability (C). Lateral collateral ligament (LCL), posterior cruciate ligament (PCL), and popliteus-arcuate ligament complex (deep). (Adapted from Gollehon DL, Torzilli PA, Warren RF: The role of the posterolateral and cruciate ligaments in the stability of the human knee: A biomechanical study. J Bone Joint Surg Am 69: 233-242, 1987.)

The inferior lateral genicular artery originates off the popliteal artery, passes laterally along the posterior joint capsule, and then passes posterior to the PFL and anterior to the fabellofibular ligament attachment sites on the fibula.1,3,6,8 Clinically, it is important to identify this artery along its course so that it can be coagulated properly (if necessary) to minimize postoperative bleeding.5 The lateral meniscus also contributes to lateral stability by creating a concave surface on the otherwise convex lateral tibial plateau.

FUNCTION AND BIOMECHANICS Most of the biomechanical stability data characterizing the relative importance of individual anatomic structures of the posterolateral knee in providing stability are derived from sequential cutting of the structures during motion testing.13,16 The posterolateral and lateral structures act in combination with the cruciate ligaments to provide overall static and dynamic stability to the lateral knee (Fig. 23F-6). These structures function primarily to resist varus rotations and posterolateral tibial rotation. They also serve as important secondary stabilizers to anterior and posterior tibial translation. In addition, there are biomechanical data to support a small contribution to internal rotation stability of the tibia. The effects of these various structures in relation to abnormal motions are described next.

1724 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Varus Stability From the data of numerous cutting studies, the fibular collateral ligament has been found to be the primary static restraint to varus instability.13,16-18 Isolated sectioning of the FCL results in a significant increase in varus rotation at any knee flexion angle. However, as long as the FCL is intact, an increase in varus rotational instability has not been demonstrated.5,16 The PCL also serves as a secondary restraint to varus rotation. Further, sectioning of the PCL and posterolateral structures in addition to the FCL results in increased varus rotation of the knee at all flexion angles, with the maximal increase at 60 degrees.16,27 However, sectioning of the PCL does not affect varus stability of the knee if the FCL is intact.13,27 The popliteus tendon complex (including the PFL), the long and short heads of the biceps femoris, ITB, and lateral gastrocnemius have also all been shown to contribute to varus stability when the FCL is sectioned.13,16,27,28

Anteroposterior Stability The structures of the lateral side and PLC of the knee do not provide a significant role in preventing anterior tibial translation in knees with an intact ACL.13,16,27-29 Veltri and colleagues reported no significant change in anterior translation of the tibia with an applied force at any knee flexion (ACL-intact knees).27,29 However, in ACLdeficient knees, the posterolateral structures assume an important role as a secondary restraint to anterior tibial translation.27,28,30 In sectioning studies of the PLC, Nielsen and Helmig found that anterior translation was not increased until the ACL was sectioned.28 They reported the increased tibial translation was most notable in the initial 40 degrees of knee flexion. Wroble and associates published similar results and noted that this increased tibial translation should be measurable clinically when performing Lachman’s test.30 The PCL is the primary stabilizer to posterior translation of the tibia.13,27 It is also important to recognize that the relative contribution of restraint varies with knee flexion angle, with the greatest amount of posterior translation at about 90 degrees of knee flexion. With an intact PCL, the posterolateral knee structures play a slight primary role in preventing posterior tibial translation, especially near full knee extension.13,16,27 Gollehon and coworkers reported a slight increase in posterior translation at all knee flexion angles.13 Grood and colleagues found an increase in posterior tibial translation from 0 to 45 degrees of knee flexion.16 Veltri and associates found isolated sectioning of the FCL and posterolateral structures resulted in increased posterior translation, with maximal increases between 30 and 45 degrees of knee flexion.27 In the PCL-deficient knee, the posterolateral structures assume a more important role. Authors have demonstrated that combined sectioning of the PCL and posterolateral structures resulted in significant increased posterior tibial translation compared with isolated PCL or posterolateral sectioning at all knee flexion angles.13,16 In the PCL-deficient knee, the popliteus muscle-tendon complex has been found to provide a significant restraint

to posterior tibial translation.21,31 In summary, the structures of the lateral and PLC are an important secondary stabilizer of anterior and posterior tibial translation. In cruciate-deficient knees, these structures assume a more important biomechanical function. With concomitant cruciate ligament and PLC injuries, some authors have recommended that both injuries be addressed concurrently, rather than staged (or ignored), to protect the posterolateral structures from being stretched out or failure of cruciate ligament reconstruction.5,32,33

Internal and External Rotational Stability The posterolateral structures assume a small role in preventing internal rotation in the knee with an intact ACL; however, they are a secondary restraint in the ACL­deficient knee. Lipke and colleagues reported that isolated sectioning of the FCL and posterolateral structures resulted in no significant increases in tibial internal rotation at knee flexion angles of 0 to 40 degrees until the ACL was sectioned.34 Wroble and associates demonstrated that, in the ACL-deficient knee, the posterolateral structures provided a secondary restraint in preventing internal rotation near full extension.30 There are no biomechanical data to suggest changes in internal rotation in PCL-deficient knees.35,36 Structures of the lateral and posterolateral knee are the primary restraints to external tibial rotation.13,16-18,28,29,31,34-36 The FCL, popliteus muscle-tendon unit, and the PFL have been found to be the primary stabilizers. Lipke and coworkers found significant increases in external rotation after sectioning the posterolateral structures at knee flexion angles between 0 and 40 degrees.34 Nielsen and colleagues reported maximal external rotation at 35 degrees of knee flexion after sectioning the FCL and the posterolateral structures.35 In another study, Nielsen and Helmig reported increased external rotation for isolated popliteus tendon sectioning between 20 and 130 degrees of knee flexion, with a maximal amount of increased external rotation at 30 degrees of knee flexion with the FCL, popliteus tendon, and PLC structures cut.28 Similarly, other authors have demonstrated increased external tibial rotation in cutting studies of the posterolateral knee.13,16,27 The clinical significance of these biomechanical studies is that the dial test outcomes at 30 and 90 degrees of knee flexion are based on these studies. Both the ACL and the PCL have been found to be important stabilizers to external tibial rotation in biomechanical studies.13,16,30 Wroble and coworkers reported that in an ACL-deficient knee with the FCL and posterolateral structures sectioned, a significant increase in external rotation was observed, with maximal rotation occurring at 90 degrees of knee flexion.30 In the same study, they reported that isolated ACL sectioning resulted in a clinically insignificant increase in external rotation of the tibia at all flexion angles. Nielsen and Helmig reported that isolated sectioning of the PCL resulted in no ­ significant increase in external rotation of the knee.31 However, ­ combined sectioning of the PCL and posterolateral structures resulted in increased external rotation at all knee flexion angles, with maximal increase at 90 degrees.13,16,35,36 The conclusions from

knee 1725

these ­biomechanical studies support that the PCL acts as a secondary restraint to external rotation with PLC injuries at higher knee flexion angles.

Intra-articular Effects Patients with PLC injuries have increased contact pressures within the knee that may result in a higher incidence of knee osteoarthritis.37,38 In a clinical series, Kannus noted that patients with PLC injuries had a higher incidence of osteoarthritis.37 Biomechanical data to support this clinical observation were established by Skyhar and coworkers.38 They reported significant increases in contact pressures within the patellofemoral and medial compartments of the knee with sectioning of the PCL, and the highest increases were observed in knees with additional combined sectioning of the posterolateral structures.

CLASSIFICATION Classification of lateral and posterolateral knee injuries are determined by physical examination findings.5 The classification scheme is based on the posterolateral drawer test, external rotation stability at 30 and 90 degrees, and varus stress testing (Table 23F-1).

HISTORY AND CLINICAL EVALUATION

   TABLE 23F-1  Classification of Posterolateral Knee Injuries Grade

Description

I

Minor sprains result in minimal increases in varus translation, external rotation at 30 degrees and 90 degrees, and posterolateral rotation at 90 degrees Demonstrate: increase in varus of 6 to 10 mm (compared with the contralateral side) with end point, and an amount of increase in posterolateral rotation that is no more than one grade higher compared with the contralateral side Concomitant posterior cruciate ligament tear: should not result in a significant increase in posterior translation at 90 degrees Demonstrate: increase in varus opening of the knee at 30 degrees of greater than 1 cm compared with the contralateral knee, significant increase of external rotation at 30 degrees compared with the contralateral side (10-15 degrees), and a one- to two-grade increase (or more) in the posterolateral drawer test at 90 degrees compared with the contralateral side Concomitant anterior cruciate ligament tear: increased anterior translation on Lachman’s test Concomitant posterior cruciate ligament tear: increased amount of posterior translation on the posterior drawer test (in neutral) at 90 degrees

II

III

Adapted from LaPrade RF: Posterolateral knee injuries: Anatomy, evaluation, and treatment. New York, Thieme, 2006.

History About 40% of patients with posterolateral instability are injured during athletic participation.39 Falls and motor vehicle crashes are also common mechanisms of knee injury.40-43 The typical mechanisms for posterolateral injury include a direct blow to the anteromedial knee, contact or noncontact hyperextension, and a varus noncontact force (Fig. 23F-7).43 These forces apply stress to the PLC structures that normally provide resistance to varus rotation, posterolateral tibial rotation, and posterior tibial translation near full extension. However, many of these same forces can be responsible for injury to either the ACL or PCL. In a large study population of posterolateral knee injuries, LaPrade and Terry43 showed that only 28% were found to have isolated grade III PLC injuries, with the remainder of the cases occurring with concomitant ligamentous injuries around the knee.43 Thus, it is more common to injure the PLC structures in conjunction with other ligaments, such as the ACL or the PCL, than it is to sustain an isolated injury. The clinical presentation of posterolateral instability can be variable (Box 23F-1). Mechanical alignment, degree of laxity, and even activity level contribute to this variability. Pain along the posterolateral aspect of the knee in either acute isolated or combined ligament injuries is a common complaint.43-45 Swelling about the knee is often minimal, even in the setting of acute injury.43 Patients with posterolateral knee injuries frequently present with a varus thrust gait with ambulation and feel most unstable as the knee approaches full extension. Activities such as climbing and descending stairs, pivoting and

twisting exercises, and even level walking can produce functional instability.5 Initial complaints may also include paresthesias, dysesthesias, or distal motor weakness due to a neurapraxia or complete injury to the common peroneal nerve.5 It has been noted that 15% to 29% of posterolateral knee injuries have associated peroneal nerve injury.29,43,46 Depending on the mechanism, vascular injury, as well as fractures about the knee, may also be a source of knee symptoms.47

Physical Examination A thorough physical examination of the involved and uninvolved extremities is essential for posterolateral injuries. This includes an evaluation of gait, ligamentous stability, and neurovascular status. Tibiofemoral alignment and patellofemoral stability should also be assessed. As previously noted, other knee stabilizers such as the ACL and PCL are commonly injured in conjunction with posterolateral structures, and their competency should not be omitted from the examination. The gait examination is performed for identification of a varus thrust pattern that can be seen in chronic posterolateral injuries. Although this gait pattern is also commonly seen in medial compartment arthrosis, the varus thrust simply implies an opening of the lateral compartment of the knee at foot strike and the varus joint subluxation that follows.11 Some patients may present with a flexed knee gait to compensate for this thrust.

1726 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Biceps femoris

Varus thrust at foot strike

Torn FCL

A

B

Figure 23F-7   Mechanism of injury. A, The mechanism of injury of a posterolateral corner injury is classically a blow to the anteromedial aspect of the knee with the foot planted. B, Lateral compartment gapping (two-headed arrow) that can occur with a varus thrust (horizontal arrow) at foot strike. As a result of the injury, weight-bearing through the medial compartment is frequently seen. (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York, Thieme, 2006, pp 76, 77.)

External Rotation Recurvatum Test

Box 23F-1 Clinical Presentation of Posterolateral Knee Injuries

This test assesses for increased genu recurvatum and relative genu varus48 and is performed with the patient supine, with the knee and hip fully extended. With the patient’s quadriceps relaxed, the great toe and foot are lifted simultaneously (Fig. 23F-8). Measurement of this test is expressed in negative degrees of knee motion or by increased heel height off the table when compared with the contralateral knee.17,48 In a positive test, the involved knee falls into a position of recurvatum and varus. An increase in external rotation can be seen by lateral rotation of the tibial tubercle. This result usually signifies a combined cruciate and posterolateral injury. Sensitivity of this test is variable, from 33% to 94%.42,49

A

• Lateral knee and joint line pain • Swelling and ecchymosis • Difficulty with pivoting and twisting, stairs • Numbness and weakness along peroneal nerve distribution

• Varus thrust gait pattern • Instability near full extension

B

Figure 23F-8   External rotation recurvatum test. A, Initial starting point for external rotation recurvatum test. B, Increased recurvatum as seen with a combined anterior cruciate ligament and posterolateral corner injury.

knee 1727

B

A

Figure 23F-9   Varus stress stressing (left knee). A, Alignment of knee without stress at 30 degrees of knee flexion. B, Increased varus laxity seen at 30 degrees of knee flexion consistent with a posterolateral corner injury.

Varus Stress Test In the supine position, the varus stress test is performed at 0 and 30 degrees of knee flexion (Fig. 23F-9). A varus stress is applied to the knee through the foot and ankle while the other hand of the examiner provides a counter force and gauge of lateral joint line opening. When compared with the contralateral side, increased lateral joint line opening of between 0 and 5 mm (grade I), 6 and 10 mm (grade II), and more than 1 cm (grade III) is noted (Table 23F-2).50 A positive test at 0 degrees is often indicative of severe injury to the fibular collateral ligament, mid-third lateral capsule, popliteus tendon, and possibly the superficial layer of the iliotibial band43 and cruciate ligaments. In contrast, increased laxity at 30 degrees correlates with complete tearing of the fibular collateral ligament and possibly other varus stabilizers of the posterolateral complex.43 The fibular collateral ligament is the primary restraint to varus stress, and biomechanical sectioning studies demonstrate that increases in varus opening will not be present unless there is concurrent injury to this structure.13,51

Dial Test The dial test, also called the posterolateral rotation test, is important for assessing external tibial rotation in relation to the femur (Fig. 23F-10). Performed either supine or prone, the femur is stabilized while the tibia, ankle, and foot are externally rotated and compared with the

   TABLE 23F-2  Classification of Varus Instability of the Knee Grade

Amount of Varus Opening

Severity of Injury

I II III

0-5 mm 6-10 mm >1 cm

Mild Moderate Severe

From American Medical Association: Standard nomenclature of athletic injuries. Chicago, American Medical Association, 1966.

c­ ontralateral side at both 30 degrees and 90 degrees of knee flexion. While observing the tibial tubercle, an increase of 10 to 15 degrees of external rotation at 30 degrees of flexion, compared with the uninvolved side, is indicative of an injury to the PLC.16 The knee is then flexed to 90 degrees, and the test is repeated. In an isolated injury to the posterolateral knee, external rotation will now decrease, compared with 30 degrees, due in part to an intact PCL.16 However, if external rotation increases at 90 degrees, a combined injury to the PLC and PCL is suspected.16

Posterolateral Drawer Test The posterolateral drawer test assesses posterolateral stability by comparing the amount of external tibial rotation and posterior tibial translation, relative to the lateral ­ femoral condyle (Fig. 23F-11).48 The knee is flexed to 80 degrees and the hip to 45 degrees. The patient’s foot is stabilized by the examiner in the seated position. A posterior drawer test is then performed in external, neutral, and internal rotation, while observing the relationship of the anterior tibia with respect to the distal aspect of femoral condyles. Increased posterior translation in external rotation signifies injury to the posterolateral and popliteus complex. Conversely, a positive test in neutral or internal rotation is consistent with injury to the PCL. Sensitivity for PLC injuries with this test ranges from 70% to 75%.42,44,46

Reverse Pivot Shift Test This test can be considered a dynamic variant of the posterolateral drawer test,52 and several methods have been described (see Fig. 23F-11). Most commonly, the involved knee is flexed from 60 to 90 degrees. While applying a valgus stress to the knee with the foot in external rotation, the knee is slowly extended. In the patient with posterolateral instability, the tibia is subluxated posteriorly while in the initial flexed position. With extension, the tibia reduces with a tactile shift. This action is thought to result from the iliotibial band changing from a flexor to an extensor at 25 to 30 degrees of knee flexion.11 This test should be

1728 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Visualize tibial tubercle rotation

Externally rotate ankle

A

B

Figure 23F-10   A and B, Dial test. Increased external rotation clinically is assessed with the dial test demonstrating isolated posterolateral corner injuries. An external rotation force is applied to the knee while the knee is placed in 30 degrees of flexion. The amount of external rotation is qualitatively measured by observing for differences in rotation of the tibial tubercles between the injured (arrow) and normal contralateral limb. (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York: Thieme, 2006, p 82.)

distinguished from the true pivot shift test, in which the knee is started in extension, and the anterior subluxation of the tibia is reduced as the knee is taken into flexion. The reverse pivot shift test has been shown to have a high falsepositive rate, and up to 35% of uninjured patients can have a positive test when examined under anesthesia.53

Finally, a complete examination should include motor and sensory testing, especially along the common peroneal nerve distribution.

Posterior Tibial Translation

Radiographs

Posterior tibial translation can also be used in the examination process. Performed at both 30 and 90 degrees of knee flexion, the position of the anterior tibia relative to the femoral condyles is assessed following a posteriorly directed force to the proximal tibia. In isolated PCL injuries, there is a loss of the normal anterior tibial step-off at 90 degrees of flexion.16 Conversely, isolated PLC injuries elicit increases in posterior translation only at 30 degrees. Combined PCL and PLC injuries result in increased posterior tibial translation at both 30 and 90 degrees of knee flexion, which is usually greater than or equal to 12 mm.

Radiographic evaluation for cases of posterolateral instability should include anteroposterior weight-bearing views in both flexion and extension, a lateral of the injured knee, and patellofemoral views of both knees.55 Although radiographs are often normal in cases of posterolateral instability, ­several entities may be evident. Medial compartment arthrosis, Segond’s fracture, an arcuate avulsion fracture from the fibular head, and avulsion fractures of Gerdy’s tubercle are several examples.26,43,49 Other associated injuries, such as tibial plateau fractures, tibiofemoral dislocation, PCL avulsions, or tibial spine fractures, may also be evident. In a patient with chronic posterolateral instability, patellofemoral or tibiofemoral degenerative changes can be observed. Commonly, the medial compartment is most often affected, showing joint space narrowing, tibial osteophytes, and subchondral sclerosis of the tibial plateau. In this situation, standing alignment radiographs are used to evaluate for varus malalignment.11 Finally, stress radiography can be a valuable tool for determining the degree and direction of ligament laxity (Fig. 23F-12).56,57 Bilateral valgus and varus stress radiographs emphasize side-toside variations in medial or lateral joint space widening, whereas radiographs with applied anterior and posterior forces to the proximal tibia can quantitate cruciate ligament integrity.

Other Tests Other tests exist for examining the posterolateral knee, as do combinations of several previously described techniques. Shelbourne and colleagues described a similar test to the reverse pivot shift, coined the dynamic shift test.54 With the hip flexed to 90 degrees, the flexed knee is slowly extended. The combination of gravity and hamstring pull keep the tibia subluxed posteriorly until about 20 degrees. When the tibia reduces, a clunk, or dynamic shift, is elicited. In addition to specific tests for posterolateral knee injury, it is important to include Lachman’s test to assess for increased anterior tibial translation and, thus, integrity of the ACL.

IMAGING

knee 1729

hip and knee at 70˚angle

Tibial subluxation posterolaterally

A

hip and knee at 30˚angle

B

C

Tibia self-reduces at this point

D

Figure 23F-11   A and B, Reverse pivot shift test. A, The knee is flexed to between 70 and 90 degrees, and the foot is externally rotated. This results in a posterolateral subluxation of the tibia on the femur. B, The knee is then extended with a reduction of the posterolaterally subluxed tibia on the femur at about 30 degrees. C and D, Posterolateral drawer test. The knee is flexed to 90 degrees and the foot is externally rotated 15 degrees. A gentle posterolateral rotation force is applied to the knee and the amount of posterolateral rotation on the femur is qualitatively measured compared with the normal contralateral knee. C, Neutral position.   D, Posterolateral drawer applied demonstrating increased posterolateral rotation.

1730 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 23F-12   Radiograph of normal knee (A) with normal lateral compartment opening with varus stress. Radiograph of contralateral abnormal knee (B) with increased lateral compartment opening with varus stress at 30 degrees of knee flexion demonstrating a posterolateral corner injury.

Magnetic Resonance Imaging MRI is useful in both the acute and chronic settings for evaluation of the posterolateral knee, especially given the complex anatomy of the lateral and posterolateral knee (Figs. 23F-13 to 23F-15).26,58-60 In the acutely injured and painful extremity, MRI may be the most comprehensive imaging tool for assessing the entire knee joint. Information obtained from the MRI is essential for preoperative planning because structures such as the fibular collateral ligament, popliteus muscle-tendon complex, biceps tendon complex, and lateral capsular attachments can be predictably identified.59 Associated injuries to the meniscus, articular cartilage, and cruciate ligaments can also be identified.

A

In addition to the standard coronal, sagittal, and axial imaging sequences, thin-slice (2-mm) proton density coronal oblique images that include the complete fibular head and styloid are extremely valuable for evaluating the fibular collateral ligament and popliteus tendon.26 High-quality MRI (1.5 Tesla or higher) is preferred because less powerful magnets make delineation of injury difficult.

Arthroscopy The intra-articular assessment of the PLC can be an adjunct for knee evaluation, and its utility has been shown in a prospective case series.61 The posterolateral ­structures

B

Figure 23F-13   Magnetic resonance imaging appearance of the superficial and deep layers of the iliotibial band and fabellofibular ligament. A, Normal superficial and deep layers (coronal view, right knee). B, Avulsion of iliotibial band off Gerdy’s tubercle (coronal view, left knee).

knee 1731

A

B

Figure 23F-14   A, Magnetic resonance imaging appearance of a normal fabellofibular ligament (white arrow) and normal popliteofibular ligament (black arrow). B, Abnormal appearance of an chronically, attenuated popliteofibular ligament (white arrow).

whose injury can be identified by arthroscopy include the meniscofemoral and meniscotibial portions of the midthird lateral capsular ligament, the femoral attachment of the popliteus tendon, the popliteomeniscal fascicles, and the coronary ligament to the posterior ­ lateral meniscus.

Excessive lateral compartment laxity can be elicited by noting a “drive-through” sign when a varus stress is applied to the knee (Fig. 23F-16).61 This is represented by more than 1 cm of lateral joint line opening.61 In addition, arthroscopy allows further visualization of meniscal and articular cartilage as well as the cruciate ligaments. However, the examiner must be aware and judicious with fluid management to avoid the possibility of fluid extravasation in the acutely injured or capsular-deficient knee.

TREATMENT Grades I and II Posterolateral Knee Injuries Isolated Injury A

B

C

D

Figure 23F-15   Magnetic resonance imaging appearance of the fibular collateral ligament and popliteus insertion. A, Coronal view of normal fibular collateral ligament (right knee). B, Coronal view of normal femoral insertion of the popliteus (right knee). C, Coronal view of tearing of the popliteus tendon origin (black arrow) and fibular collateral ligament (left knee). D, Sagittal view of a popliteus muscle belly edema from an acute popliteus musculotendinosus junction injury (right knee).

Isolated grade I and II PLC injuries are frequently missed in the acute setting, making natural history studies difficult to perform. Fortunately, isolated acute grade I and II knee injuries are almost always treated nonsurgically,62 especially if they have initial valgus alignment.46 Patients are initially immobilized in full knee extension for 3 weeks to allow the injured tissues to heal, then range of motion is gradually introduced. Mild injuries are allowed to bear weight in extension, whereas more severe injuries may require 6 weeks of protected weight-bearing. A functional rehabilitation program is started 6 weeks after injury, emphasizing endurance exercises and quadriceps strengthening, whereas aggressive hamstring work is restricted. Patients with chronic symptomatic posterolateral instability after grade I or II injuries may be treated with a medial compartment unloader brace to determine whether they have improvement in subjective symptoms. Additionally, bilateral varus stress radiographs can quantify the amount of varus laxity present. If patients exhibit varus laxity and subjective improvement in symptoms with a medial compartment unloader brace, they may benefit from a posterolateral knee reconstructive procedure or a proximal tibial osteotomy if they are in varus alignment.5

1732 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 23F-16   Arthroscopic views of lateral compartment. A, Drive-through sign in lateral compartment in a patient with an anterior cruciate ligament, posterior cruciate ligament, and posterolateral corner injury. B, Popliteus tendon avulsion off of the femur (arrow).

Concomitant Cruciate Ligament Injury

Concomitant Cruciate Ligament Injury

Combined ACL or PCL tears with grade I or II PLC injuries need to be critically evaluated for the amount of instability present. LaPrade has recommended that the cruciate ligament tear causing instability be reconstructed and that grade I or II PLC injuries be treated nonoperatively.5 Stress radiographs are helpful in assessing the amount of instability. Increased varus instability places a large amount of stress on the ACL reconstruction graft, whereas increased varus laxity combined with posterolateral rotatory instability places undue stress on a PCL reconstructive graft. This has been known to cause chronic laxity and even failure of the reconstructed cruciate ligament grafts.17,32,33,63

Combined ACL or PCL injuries and grade III PLC injuries are probably best treated surgically in the acute setting after the swelling has subsided. It is recommended to leave a 6- to 7-cm skin bridge between the anterior incision for a cruciate ligament reconstruction procedure and the PLC approach. All the following repairs and reconstructions, whether performed acutely or chronically, use a common surgical approach (see “Authors’ Preferred Method”).

Grade III Posterolateral Knee Injuries

In the acute setting, the treatment of femoral-based PLC knee injuries may include the popliteus tendon or FCL femoral recess procedures as originally described by Hughston.5,15,25 Recent modifications have been noted.5,64 After the surgical exposure of the PLC, the avulsed ends of the FCL or the popliteus tendon are whip-stitched with 2-0 nonabsorbable suture. After sufficient length of the FCL or popliteus is verified, a cruciate ligament guide is used to place an eyelet-tipped pin through the exact attachment site of the FCL or popliteus tendon (Fig. 23F-17). The pin exits medially on the femur proximal to the medial epicondyle and adductor tubercle region; this angulation avoids passing the suture or graft through the intercondylar notch. A horizontal incision is made proximal to the course of the medial patellofemoral ligament and along the distal border of the vastus medialis obliquus muscle. The vastus medialis obliquus muscle is retracted proximally to allow the passing sutures and surgical button to be tied deep to the muscle fibers. A 5- or 6-mm cannulated reamer is placed over the eyelet-tipped pin laterally and reamed to a depth of 1 cm. The passing sutures are placed in the eyelet-tipped pin and are pulled out medially. With the knee in full extension, care is taken to ensure that the avulsed structure is pulled into the tunnel. Finally, the

Isolated Injury In contrast to grade II PLC injuries, grade III PLC injuries have a low probability of healing nonoperatively, and surgical treatment is often needed to allow the best outcome.37 Ligamentous repairs and reconstructions aim to provide varus and posterolateral stability. Interestingly, acute repairs have been shown to lead to improved outcomes when compared with the results of chronic reconstructive procedures.46 In fact, the International Society of Arthroscopy, Knee Surgery, and Orthopaedic Sports Medicine (ISAKOS) has recommended that these surgical procedures be performed within the first 2 weeks after a PLC injury. Significant scar planes develop within 3 weeks of injury that may complicate the surgical exposure of the iliotibial band, biceps femoris, and common peroneal nerve. Additionally, 3 weeks after injury, the posterolateral knee structures do not hold sutures well, which makes rehabilitation techniques using early range of motion protocols more difficult. It is important to note that multitrauma patients and those with severe skin problems may benefit from delayed surgery.

Repair Techniques Advancement or Recession of the Posterolateral Corner

knee 1733 Eyelet-tipped pin enters at popliteus tendon attachment site and exits medially (proximal to medial epicondyle and adductor tubercle region)

Cannulated reamer (to depth of 1 cm)

Figure 23F-18   Lateral capsular repair to bone with suture anchors just distal to articular margin (lateral view, right knee).

Passing sutures through end of avulsed popliteus tendon Figure 23F-17   Popliteus tendon recess procedure. The torn popliteus is whip-stitched. A cannulated reamer is used to drill over an eyelet-tipped guide pin. The guide pin is used to pull the sutures medially and then they are tied over a button on the medial femur. Finally, the popliteus tendon is recessed into the femoral socket. A similar procedure is performed for the fibular collateral ligament. (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York, Thieme, 2006, p 157.)

The coronary ligament is accessed through the second fascial incision. The lateral gastrocnemius is retracted ­posteriorly; this approach protects the neurovascular bundle and allows visualization of the coronary ligament (Fig. 23F-19). Multiple sutures are passed through the popliteal aponeurosis into the substance of the lateral meniscus and down into the distal tibial attachment site of the coronary ligament. The sutures are then tied with the knee in flexion. The results of repairs of intrasubstance tears of the popliteus and FCL have faired poorly; therefore, these tears are traditionally treated with reconstruction procedures.

Reconstruction Techniques: Overview

passing sutures are fed through a button medially and tied down over the cortex, pulling the avulsed structure into the recess hole. Additionally, suture anchors can be used to reattach the lateral gastrocnemius tendon and portions of the posterior and lateral capsule to their anatomic femoral attachment sites (Fig. 23F-18). Significant gapping occurs when attempts are made to repair the FCL or popliteus to the femur with suture anchors, and thus it is recommended instead to use the recess procedure for these structures. Interestingly, a recent nonrandomized study comparing acute PLC repair to reconstruction found a higher failure rate among patients who underwent a repair.65

PLC reconstructive techniques have recently been broken down into nonanatomic and anatomic techniques. All these techniques may be used to reconstruct an acute or a chronic PLC-deficient knee. The nonanatomic techniques include biceps femoris tenodesis,66,67 Larson technique,68 and Stannard’s modified two-tailed technique.69 The reconstructions all involve creating restraints from either the fibular head or the PLC of the tibia to the ­lateral femoral epicondylar region. Re-creating the ligament length relationships of the lateral knee in proper isometric positions attempts to create a mechanical advantage and thus resist varus and posterolateral tibial rotations. Recently, an anatomic reconstruction has also been described by LaPrade.19

Intrasubstance Posterolateral Corner Repair

Reconstruction Techniques: Nonanatomic

Intrasubstance injuries to the popliteomeniscal fascicles and coronary ligament to the lateral meniscus may be repaired directly. Direct suture repair of the anteroinferior and posterosuperior popliteomeniscal fascicles is made to the popliteus tendon through the mid-third capsular arthrotomy. The structures are repaired in a direct horizontal mattress technique with nonabsorbable 0-0 suture; this technique allows early range of motion rehabilitation protocols.

Biceps Tenodesis Clancy popularized the biceps tenodesis procedure to repair and augment intrasubstance injuries to the FCL and the PFL; however, the surgeon must ensure that the long head of the biceps tendon is completely intact (Fig. 23F-20). The tenodesis involves transferring the biceps femoris tendon to the anterior aspect of the lateral femoral

1734 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 23F-19   Coronary ligament repair of the lateral meniscus performed between the lateral gastrocnemius tendon and soleus muscle (lateral view, right knee). PCL, posterior cruciate ligament; PFL, popliteofibular ligament. (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York, Thieme, 2006, p 159.)

Biceps femoris

PCL

Coronary ligament PFL

Lateral gastrocnemius Popliteus

Common peroneal nerve Soleus

epicondyle. This procedure attempts to replace the FCL and the PFL through its attachments to the posterolateral capsule complex. Wascher and colleagues determined that a point of fixation 1 cm anterior to the FCL attachment is needed to significantly reduce external tibial rotation and varus laxity.70 After the standard surgical approach, the peroneal nerve is dissected free from the posterior margin of the biceps, and the biceps tendon is released from the lateral gastrocnemius muscle tendon unit. A vessel loop is placed around the nerve, and any attachments of the nerve to the biceps muscle tendon unit are freed. The nerve is traced distally into its entry point in the anterior compartment musculature. If there is constriction of the nerve at this location, it must be released. The inferior aspect of the iliotibial band is freed from its intermuscular septal attachments to allow the biceps tendon and muscle belly to be brought up beneath it. The lateral epicondyle is exposed by incising the iliotibial band and dissected free of soft tissue, exposing the superior aspect of the FCL. A 1-cm wide and 3-cm long trough is made at the upper portion of the lateral femoral condyle. A 3.2-mm hole is drilled just superior to the lateral femoral epicondyle and is directed medially and slightly proximally into the medial femoral condyle, avoiding any tunnels that may have been created for cruciate ligament reconstruction. A 6.5-mm screw with a spiked soft tissue washer is placed at that location. The distal 7 to 8 cm of biceps muscle is resected away from the tendon; otherwise, fixation of the tendon to the lateral femoral condylar trough is impossible because of the interposition of the muscle. The biceps tendon with its posterolateral capsular attachments is brought

proximally and looped over the fixation device; then the knee is passed through a range of motion ­ensuring that the point of fixation does not interfere with an adequate range of motion. The graft is secured by advancing the screw and washer to the cortex. In an attempt to avoid sacrificing the important biceps tendon function, variations of the biceps tenodesis have been developed using only a portion of the biceps tendon. Generally, a central slip of the tendon that involves about 50% of the cross-sectional diameter is created and left attached at the fibular head. The free end is routed beneath the posterior-most fibers of the biceps tendon at the fibular head, and then advanced to the point of attachment on the femur; this creates a band of tissue approximately in the position of the PFL that can resist posterolateral rotations as well as varus rotations. Importantly, procedures that do not tenodese the total biceps tendon do not provide for retensioning of the posterolateral capsular complex.

Modified Two-Tailed Technique Stannard has described the modified two-tailed technique using a tibialis anterior or posterior allograft (Fig. 23F-21).65 The three critical components of the deep layer of the PLC are reconstructed: the popliteus, the popliteofibular ligament, and the FCL. After a standard surgical approach, a 5-mm hole is drilled from anterior to posterior through the lateral tibia. The posterior hole approximates the position of the popliteus as it crosses along the posterior tibia. The tibial tunnel is tapped with a 7-mm tap. The allograft is sized to 5 mm and passed into the tunnel

knee 1735

B

A

Lateral gastrocnemius muscle

Trough Fibular collateral ligament

Resected 2-inch short head biceps muscle

Short head Long head biceps muscle biceps tendon

C

Lateral gastrocnemius tendon

Common biceps tendon

D Trough

Fibular collateral ligament Common biceps tendon

E

F

Figure 23F-20   Posterolateral reconstruction using biceps femoris tendon (Clancy). A, Common peroneal nerve is carefully dissected; biceps femoris muscle is freed from attachments to the lateral gastrocnemius. B, Femoral origin of the fibular collateral ligament. C, Insertions of the common biceps tendon into the arcuate complex posteriorly and insertion around the fibular collateral ligament. D, Trough is made in the upper third of the lateral femoral epicondyle. E, Biceps femoris tendon is brought anteriorly to the trough in the epicondyle. F, AO cancellous screw and washer are placed inferior to the tendon and tightened. (Adapted from Clancy WG Jr: Repair and reconstruction of the posterior cruciate ligament. In Chapman MW [ed]: Operative Orthopaedics. Philadelphia, JB Lippincott, 1988.)

1736 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Passing sutures

FCL Popliteus tendon graft

Figure 23F-21   The modified two-tailed reconstruction of the posterolateral corner with tibialis anterior autograft. Tibial fixation is accomplished with a bioabsorbable screw. Femoral fixation with a 6.5 cancellous screw and ligament washer. (Adapted from Stannard JP, Brown SL, Farris RC, et al: The posterolateral corner of the knee: Repair versus reconstruction. Am J Sports Med 33[6]:881-888, 2005.)

from posterior to anterior. The graft is fixed with a 7-mm bioabsorbable ligament screw. A second 5-mm hole is then drilled through the proximal fibula from anterolateral to posteromedial. The isometric point on the lateral femoral condyle is then located. The “isometric point” is located just superior to where the FCL and popliteus cross on the lateral femoral condyle. A 4.5-mm bicortical screw going from lateral to medial is placed in the lateral femoral condyle with a spiked ligament washer. The surrounding bone is decorticated to facilitate allograft healing between the attachments of the FCL and popliteus. The graft is then passed from the posterior tibia up and around the screw in the lateral femoral condyle, back down to and through the fibular tunnel, and then back up to the screw and washer. The graft is tensioned with the foot internally rotated and the knee flexed 40 to 60 degrees. The free end of the graft is anchored primarily by the spiked ligament washer but is supplemented with a 2-0 suture.

Reconstruction Techniques: Anatomic Reconstruction of the Popliteus Acute reconstruction of the popliteus tendon can be performed with an autograft or allograft hamstring tendon (Fig. 23F-22). Harvesting of the hamstring tendons has been previously described.5 The semitendinosus tendon is preferred because it is the thicker of the two hamstring tendons; additionally, harvesting the semitendinosus is less likely to injure the saphenous nerve. After the semitendinosus tendon is harvested and tubularized with a 2-0 or 5-0 suture, a 25- to 30-mm deep femoral tunnel is drilled into the anatomic origin of the popliteus.71 A lateral interference screw or a medial surgical button is used

Gerdy’s tubercle

Figure 23F-22   Acute popliteus tendon reconstruction with a hamstring graft. The hamstring graft is pulled into the femoral socket and fixed with a bioabsorbable screw, wrapped around the musculotendinous junction of the popliteus complex, and pulled anteriorly through the tibial tunnel. The graft is fixed in place with a screw and staple. FCL, fibular collateral ligament. (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York, Thieme, 2006, p 161.)

to fix the graft to the femur. Next, the posterolateral sulcus of the tibia is identified at the musculotendinous junction of the popliteus. An eyelet-tipped guide pin is drilled from ­anterior to posterior in the tibia, using a cruciate ligament guide, exiting at the posterolateral sulcus. The entry point for the anterior tibia is placed just medial and distal to Gerdy’s tubercle. A posterior retractor is placed to protect the posterior neurovascular bundle during the procedure. A 7- to 8-mm reamer is used to fashion the tibial tunnel from anterior to posterior. The graft is then passed distally through the popliteus hiatus between the gastrocnemius and soleus muscles; then the graft is pulled anteriorly from its posterior entry point. The knee is placed in 60 degrees of flexion and neutral rotation while the graft is fixed to the tibia with an interference screw and a bone staple for backup fixation.

Reconstruction of the Fibular Collateral Ligament An FCL reconstruction with autogenous and allograft hamstring tendons has been described.5 The FCL graft is tubularized with a 2-0 or 5-0 suture. An eyelet-tipped pin is then inserted into the anatomic femoral attachment of the FCL. A 7-mm reamer is used to drill a 20- to 25-mm tunnel. The tunnel is then tapped, and the graft is passed into the femur and fixed with an interference screw. A second tunnel is then drilled starting at the lateral aspect of the fibular head at the attachment site of the FCL and exiting through the posterior aspect of the fibular styloid. Care is taken to avoid the attachment site of the popliteofibular ligament. A 6- or 7-mm cannulated reamer is used to drill the tunnels. The graft is then passed under the superficial iliotibial band and the lateral aponeurosis to the long head

knee 1737

of the biceps femoris. The graft is then passed through the fibular tunnel and pulled back upon itself up to the lateral epicondyle (Fig. 23F-23). The knee is placed in 30 degrees of knee flexion and neutral rotation while a valgus force is applied to decrease any varus opening. An interference screw is then placed into the fibular head to secure the graft. The graft is then tied back on itself by passing it through a split in the biceps at the lateral fibular location with 0-0 Vicryl suture. If both the FCL and popliteus are torn in their midsubstance, a direct anatomic reconstruction with an Achilles tendon allograft is performed similar to the technique for the anatomic chronic reconstruction described later in the chapter.

Figure 23F-23   Fibular collateral ligament hamstring autograft reconstruction. The graft is fixed in the femoral tunnel with a bioabsorbable screw. The graft passes under the superficial iliotibial band and the lateral aponeurotic layer of the long head of the biceps femoris. The fibular tunnel is placed through the anatomic attachment site laterally and exits posteriorly to avoid the popliteofibular ligament insertion. The graft is tied upon itself, after fibular tunnel fixation is performed with a bioabsorbable screw (left knee).

Authors’ Preferred Method Anatomic Posterolateral Corner Reconstruction

Preoperative planning. Preoperative planning for reconstruction of the PLC is essential (Table 23F-3). The senior author (RFL) does not use a leg holder. Instead, a sandbag is taped at the foot of the bed to allow for about 70 degrees of knee flexion. A sandbag is also commonly placed under the patient’s hip so that the knee will balance at 70 degrees and not require an assistant to actively hold it in that ­position. This allows the assistant to concentrate on appropriate

   TABLE 23F-3  Authors’ Preferred Surgical Technique Preparation

Instrumentation

Setup

Transtibial or transfemoral anterior and posterior cruciate ligament guide system Beath passing pins with eyelets Cannulated 7- and 9-mm reamers, bioabsorbable interference screws Adsen tipped hemostat for peroneal nerve neurolysis and fine dissection Soft tissue staple fixation system

No leg holder necessary Sandbag taped to bed to allow for 70 degrees of knee flexion Bump under hip to allow for neutral position of knee Standard orthopaedic soft tissue retractors Achilles tendon allograft ≥23 cm long with calcaneus bone plug

From LaPrade FR: Posterolateral knee injuries: Anatomy, evaluation, and treatment. New York, Thieme, 2006.

r­ etraction throughout the procedure. The patient is placed supine with the operative leg prepared and draped free. The equipment necessary for the posterolateral reconstruction includes standard cruciate ligament reconstruction instruments. Cannulated drill guides are used for guide pin positioning during tunnel placement. Eyelet-tipped ­passing pins are also used to pass sutures and to ream over during tunnel placement. Cannulated metal and bioabsorbable screws are used for graft fixation, as are small bone staples when secondary fixation is deemed necessary. The Achilles tendon allograft that is used should be at least 23 cm long to ensure that it will be long enough to pass through the course of the reconstructive technique. Surgical exposure of the PLC. A lateral hockey-stick skin incision is used unless previous skin incisions need to be incorporated (Fig. 23F-24A). The incision is drawn with a surgical marker while the knee is in 70 degrees of flexion. It starts 7 to 8 cm proximal to the knee joint at the lateral intermuscular septum, traverses centrally over Gerdy’s tubercle, and extends distally 3 to 4 cm over the anterior compartment of the leg. The knee is then extended to verify that the ­incision is roughly a straight line in full extension. After the skin incision is made, the subcutaneous tissues are dissected to expose the superficial layer of the iliotibial band. A posteriorly based skin flap is developed using meticulous dissection along this layer until the long head of the biceps femoris is reached. At this point, a common peroneal neurolysis is performed. The common peroneal nerve is typically identified posterior and medial to the long head of the biceps (see Fig. 23F-24B). It is imperative to know the location of the nerve throughout the case and to ensure that it is free of any Continued

1738 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

A

B

Figure 23F-24   Posterolateral corner approach. A, Marking of surgical incision. B, Peroneal nerve neurolysis performed posterior to the long head of the biceps femoris.

tension. In chronic injuries, the nerve is usually identifiable by direct palpation. If there is scar tissue present that makes identification of the nerve more difficult, the nerve may need to be identified proximally where it crosses posteriorly under the long head of the biceps tendon. After the neurolysis has been completed, a small Penrose drain is placed around it to allow it to be gently retracted from the surgical field. The superficial layer of the iliotibial band is then incised in line with its fibers from the supracondylar process of the femur proximally to Gerdy’s tubercle distally. This fascial incision is retracted to expose the femoral attachments of the fibular collateral ligament, the popliteus tendon, and the mid-third lateral capsular ligament (see Fig. 23F-3B). Next, a vertical arthrotomy is made through the meniscofemoral

portion of the mid-third lateral capsular ligament. This incision is about 1 cm anterior to, and parallel with, the fibular shaft when the knee is flexed to 70 degrees. This incision makes accessible the popliteus origin on the femur, the popliteomeniscal fascicles, and the lateral meniscus. A second fascial incision is made just posterior and parallel to the long head of the biceps femoris concurrent with the peroneal nerve neurolysis. The interval between the lateral head of the gastrocnemius and soleus is then developed by blunt dissection, providing access to the posteromedial aspect of the fibular styloid and the posterolateral aspect of the tibia (Fig. 23F-25). Palpation of the posterolateral aspect of the tibial plateau through this interval allows for identification of the posterior popliteal sulcus, located at

Figure 23F-25   Surgical approach to the posterolateral corner. Through the interval between the lateral gastrocnemius and soleus muscle, one can Biceps femoris palpate tears of the popliteofibular ligament (PFL), the popliteus tendon, and the posterior cruciate ligament (retracted) (PCL) at its attachment on the PCL facet on the posterior aspect of the tibia (lateral view, right knee). FCL, fibular collateral ligament. (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York, Thieme, 2006, p 148.) FCL PCL

PFL

Lateral gastrocnemius (retracted) Popliteus Soleus Common peroneal nerve (retracted)

knee 1739

Authors’ Preferred Method—cont’d

A

B

Figure 23F-26   A, Split Achilles tendon allograft. B, Tubularization of split Achilles tendon allograft.

the ­ musculotendinous junction of the popliteus. The popliteofibular ligament’s attachment site on the posteromedial down-slope of the fibular styloid is usually identified through this second fascial incision.19 If necessary, a third fascial incision may be made between the posterior border of the iliotibial tract and the anterior aspect of the short head of the biceps femoris. Graft preparation. Tendon grafts are prepared by splitting a calcaneus and Achilles tendon allograft into two equal portions parallel with the fibers of the tendon (Fig. 23F-26). The tendons must be at least 23 cm in length to complete the reconstruction in most patients. The bone blocks of the grafts are fashioned to fit 9 × 20-mm tunnels for the femoral attachments. Two drill holes are made in the bone blocks to hold sutures necessary for graft passage. The tendinous ends of the grafts are tubularized using a whipstitch. Fixation technique. Four bone tunnels are used in the reconstruction: two femoral, one tibial, and one fibular ­tunnel (Fig. 23F-27). A 7-mm tunnel is made through the fibular head from the attachment site of the fibular collateral ligament, on

the lateral aspect of the fibular head, to the ­attachment site of the popliteofibular ligament on the posteromedial downslope of the fibular styloid. The attachment site of the fibular collateral ligament is found by entering the bursa between the long head of the biceps femoris and the fibular collateral ligament (see Fig. 23F-3A). The tibial tunnel is created using a cannulated aiming guide placed on the posterior popliteal sulcus at the level of the musculotendinous junction of the popliteus. A guidewire is drilled in the anteroposterior direction from the distal medial aspect of Gerdy’s tubercle to the posterior popliteal tibial sulcus (see Fig. 23F-3C). It is important to leave a bony roof under the articular cartilage in the tibial tunnel. A 9-mm reamer is then passed over the guidewire from anterior to posterior to prepare the tunnel. The anatomic femoral attachments of the popliteus and fibular collateral ligament are once again identified. Two large Beath pins are then drilled parallel across the femur from the attachment sites, exiting proximal and medial to the medial epicondyle and adductor tubercle. It is important to measure the distance between the two Beath pins to verify

FCL- femur

FCL- femur

PLT- femur PFL/PLTtibia

PLT- femur

PFL/PLT -tibia FCL- fibula

PFL- fibula

A

B

FCL- fibula

Figure 23F-27   Fibular, tibial, and femoral tunnel placement for an anatomic posterolateral knee reconstruction. A, Lateral view. B, Posterior view. FCL, fibular collateral ligament; PLT, popliteus tendon; PFL, popliteofibular ligament. (Adapted from LaPrade RF, Johansen S, Wentorf FA, et al: An analysis of an anatomical posterolateral knee reconstruction: An in vitro biomechanical study and development of a surgical technique. Am J Sports Med 32[6]:1405-1414, 2004.) Continued

1740 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

FCL PLT

PLT PFL

FCL

A

B

Figure 23F-28   The anatomic posterolateral knee reconstruction procedure. A, Lateral view, right knee. B, Posterior view, right knee. PLT, popliteus tendon; FCL, fibular collateral ligament; PFL, popliteofibular ligament. (Adapted from LaPrade RF, Johansen S, Wentorf FA, et al: An analysis of an anatomical posterolateral knee reconstruction: An in vitro biomechanical study and development of a surgical technique. Am J Sports Med 32[6]:1405-1414, 2004.)

that the attachment sites have been correctly identified. This distance should be about 18.5 mm. With the aid of the Beath pins, the two tunnels are then reamed, and the bone plugs are pulled into their respective holes in the femur by suture passage. The bone plugs are then secured with cannulated titanium interference screws. The graft fixed in the popliteus attachment on the femur is used to reconstruct the popliteus tendon. The free end of the graft is passed distally and medially through the popliteal

A

hiatus, reaching the posterolateral aspect of the lateral tibial plateau. The graft is passed through the tibial tunnel from posterior to anterior. The second graft, from the fibular collateral ligament ­attachment on the femur, is used to reconstruct the fibular collateral ligament and the popliteofibular ligament. It is passed deep to the superficial layer of the ITB and anterior arm of the long head of the biceps femoris, following the path of the native fibular collateral ligament. The graft is then passed through the fibular head from lateral to posteromedial. The knee is flexed to 30 degrees, and a slight valgus stress is applied to reduce any lateral compartment gapping. The graft is then pulled tight and fixed to the fibular head with a bioabsorbable interference screw, thus reconstructing the fibular collateral ligament. The remaining portion of the fibular collateral ligament graft is then passed from posterior to anterior through the tibial tunnel, reconstructing the popliteofibular ligament. With the knee flexed to 60 degrees and in neutral rotation, the popliteofibular ligament and popliteus tendon are secured using a bioabsorbable interference screw placed in the anterior tibial tunnel. These grafts are then reinforced using a staple over the free tendon ends on the anterior tibia. An examination under anesthesia is then performed to confirm that the grafts are securely fixed and functioning to prevent any varus, external rotation, or posterolateral rotation of the tibia on the femur (Figs. 23F-28 and 23F-29). When stability of the knee has been verified, wound closure is performed. The anterior arm of the long head of the biceps femoris is reattached to the fibula, attempting to reconstitute the biceps bursa. The lateral capsular ­arthrotomy is closed with horizontal mattress 0-0 absorbable sutures. The iliotibial band incision is also closed with 0-0 absorbable sutures. Copious irrigation is performed, and the subcutaneous tissue is closed with either 0-0 or 2-0 absorbable sutures. The skin is closed with a subcuticular nonabsorbable suture with a pullout stitch. Steri-Strips are loosely applied, ­followed by nonadherent dressings, circumferential cast padding, and a knee immobilizer (Box 23F-2).

B

Figure 23F-29   Anatomic posterolateral reconstruction procedure. A, Lateral view, left knee. B, Posterior view, left knee. FCL, fibular collateral ligament; PFL, popliteofibular ligament. (Adapted from LaPrade RF, Johansen S, Wentorf FA, et al: An analysis of an anatomical posterolateral knee reconstruction: An in vitro biomechanical study and development of a surgical technique. Am J Sports Med 32[6]:1405-1414, 2004.)

knee 1741

Authors’ Preferred Method—cont’d Box 23F-2 Surgical Pearls and Pitfalls Surgical Pearls

• A 1-cm horizontal incision through the anterior arm of

the long head of the biceps facilitates identification of the fibular collateral ligament. • A suture placed into the fibular collateral ligament, through the biceps bursa, facilitates identification of its course and potentially its femoral attachment site. • Placement of the transfemoral eyelet pins should aim through the anatomic attachment sites of the popliteus tendon and fibular collateral ligament laterally to slightly proximal and anterior to the adductor tubercle on the ­anteromedial aspect of the knee. • The entry site anteriorly for the tibial reconstructive tunnel should be just distal and medial to Gerdy’s tubercle along its flat spot; lateral to this, the tibia downslopes at the anterior compartment and it is difficult to place in a proper tunnel and obtain fixation. • The initial part of the surgical procedure may be performed without a tourniquet; the initial surgical approach, identification of the fibular collateral ligament through the biceps bursa, the common peroneal nerve ­neurolysis,

and ­ identification and reaming of the fibular head and tibial tunnels may be done with minimal bleeding. Surgical Pitfalls not place the fibular head tunnel too proximal. It may not preserve a good cortical rim of bone at the superior aspect of the fibular tunnel. • Drilling the tibial tunnel too distal at its exit site on the posterolateral tibia could result in a horizontal popliteofibular ligament graft. The tibial tunnel posteriorly should exit 8 to 10 mm proximal from the exit site of the fibular head tunnel. Therefore, the fibular head tunnel should be reamed first, to serve as a reference guide, before drilling the tibial tunnel. • Placing the transfemoral eyelet pins too parallel to the joint may result in the passing sutures going through the intercondylar notch or the guide pins hitting a posterior cruciate ligament femoral tunnel. • Not tubularizing the reconstructive grafts accurately at their ends may result in them bunching up when attempting to pass them through the fibular or tibial tunnels.

• Do

Adapted from LaPrade RF: Posterolateral knee injuries: Anatomy, evaluation, and treatment. New York, Thieme, 2006.

POSTOPERATIVE REHABILITATION, OUTCOMES, AND COMPLICATIONS Postoperative Rehabilitation The postoperative protocol for these injuries can vary depending on the surgical technique, whether the injury is acute or chronic, the presence of associated ligamentous injuries, and the overall health status of the patient. In the acute setting for repair or reconstruction of the PLC, the authors’ preferred rehabilitation protocols are similar.5 The patients are kept non–weight-bearing for 6 weeks. An immobilizer is used to maintain full extension for 1 to 2 weeks, followed by gentle increases in range of motion. Straight leg raises and quadriceps exercises are instituted while the immobilizer is in place (Fig. 23F-30). Open chain hamstring exercises are avoided for 4 months to eliminate hamstring contraction and subsequent stress to the repair site. After 6 weeks, patients are allowed a slow and progressive increase in weight-bearing, weaning from crutches, and light work on an exercise bike. Closed chain leg presses are started, using one fourth of their body weight and limiting knee flexion to 70 degrees. At 3 months, increased activity and jogging are allowed, along with increasing weight with leg presses and increased resistance on the exercise bike. Four months after surgery, patients undergo knee stability evaluation and functional capacity testing for determination of return to play or full activity. In the situation of acute,

combined PLC and cruciate injury, the surgical rehabilitation protocol is generally guided by the PLC injury for the first 6 weeks, followed by attention to specific needs based on the type of cruciate reconstruction. In the chronic posterolateral injury, the authors prefer to reconstruct the fibular collateral ligament, popliteus tendon, and the popliteofibular ligament.5,11 It is imperative that during the early postoperative rehabilitation period, extra stress is not placed on these grafts by knee motions that could lead to stretching and subsequent failure. Thus, the rehabilitation process focuses on developing strong

Figure 23F-30   Nonoperative management. Patient in knee immobilizer performing straight leg raise.

1742 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������    TABLE 23F-4  Postoperative Rehabilitation Protocol for

the Chronic Posterolateral Knee Reconstruction Procedure Recovery Time

Rehabilitation Process

Early postoperative period

Knee immobilizer at all times, other than working on range of motion (ROM) No weight-bearing for 6 wk Avoid tibial external rotation, no hamstring exercise for 4 mo Elevation for control of swelling Quad sets hourly and straight leg raises 4-5 times/ day in immobilizer Gentle ROM 4 times/day out of immobilizer Goal: 90 degrees of knee flexion by 2 wk Increase ROM and quad strengthening Achieve full knee extension and maintain Increase flexion past 90 degrees Initiate weight-bearing, wean from crutches when limp has resolved Begin low-impact closed chain exercise Continue quad sets and straight leg raises Exercise bike once flexion of 105 to 110 degrees obtained; start with 5 min daily with low resistance and advance as tolerated Goals: full range of motion, normal gait pattern Increase in functional strength program Goals: improved quad strength/function, increased endurance, improved coordination/proprioception Walking program: 20 to 30 min daily Biking: increase resistance as tolerated, 3-5 times/ wk for 20 min Step-ups: place operative knee on step and step up, increase repetitions as tolerated Continue maintenance exercise program 3-5 times/wk Strive for maximal strength to operative extremity No competition or pivot sports until cleared by surgeon

Weeks 1 to 2

Weeks 3 to 6 Weeks 7 to 12

Weeks 13 to 16 Months 4 to 6

Months 7 and later

musculature support around the knee to protect the reconstruction during early recovery as well as during normal activity in the future (Table 23F-4).5

Potential Complications As with any operation, PLC surgery has potential complications. Risk for infection and wound breakdown are worrisome complications, especially in the setting of an acute injury or open knee dislocation.5 Surgical planning includes time for soft tissue recovery and resolution of swelling before operative intervention. In the chronic instability patient, previous surgical scars may be present, and surgical approaches should be altered to incorporate these incisions to minimize skin devascularization. Common peroneal injury can occur if careful surgical technique and dissection are not employed. However, between 15% and 29% of posterolateral knee injuries have associated peroneal nerve injury.29,43,46 Thus, it is important to perform a thorough motor and sensory examination of this nerve before surgical intervention so that proper documentation of nerve status is completed preoperatively and a baseline is established. Postoperative arthrofibrosis may occur, especially in the setting of acute PLC surgery and concomitant medial

Figure 23F-31   Anteroposterior standing varus thrust radiograph demonstrating varus laxity after posterolateral corner repair (right knee). (Adapted from LaPrade RF: Treatment of posterolateral knee injuries. In Gumpert E [ed]: Posterolateral Knee Injuries: Anatomy, Evaluation, and Treatment. New York, Thieme, 2006, p 88.)

c­ ollateral ligament injury.5 Deep venous thrombosis, chro­ nic pain, and symptomatic hardware may also occur. Additionally, recurrent laxity or instability has been reported (Fig. 23F-31). This can hopefully be avoided by proper preoperative evaluation and diagnosis of associated ligamentous injuries, proper graft and tunnel placement, and appropriate immobilization and rehabilitation protocols following surgical repair.

Return to Play The timing for return to play varies depending on the acute or chronic nature of the original injury and subsequent repair versus reconstruction (Box 23F-3). Surgical technique and associated ligamentous injuries also factor into the length of convalescence. Following acute surgical repair and reconstruction of the PLC, some high-level intercollegiate athletes have returned to full participation 4 months after surgery.5 All these patients had a return of normal knee range of motion, strength, and stability before return to play. In the chronic or combined ligament reconstruction setting, time for return to play may not occur until 6 or 9 months after surgery.5,47,72 Box 23F-3 Return to Play Criteria

•��� Near maximal strength compared with uninvolved limb •��� Full range of motion •��� No instability •��� No pain or swelling with activity •��� No limp •��� Complete sport-specific or functional tests

knee 1743

Although the timing for return to play is based on multi­ ple factors, the criteria for return to play can be generalized (see Box 23F-3). First, the patient should have regained normal range of motion and strength compared with the uninvolved side. Maximal strength of the quadriceps and hamstring complexes is essential. Well-placed and functional grafts can stretch out over time if one relies on graft integrity for knee stability rather than having appropriately rehabilitated musculature. In addition, the patient should be without a limp, resultant swelling or pain, or symptoms of instability with activity and sport participation. These criteria can be accurately assessed by performing functional or sport-specific tests that challenge the patient through maneuvers and drills with similar duration and intensity that their particular sport requires.

EVIDENCE-BASED MEDICINE Incidence Injuries to the posterolateral aspect of the knee occur less frequently than injuries to the cruciate ligaments or medial aspect of the knee. The overall incidence of knee ligament injuries is about 1 per 1000 patients per year.73 DeLee and associates retrospectively reviewed 735 knee ligament injuries from 1971 through 1977.42 They found that 12 patients (1.6%) had acute isolated posterolateral knee injuries, whereas 32 (44%) other patients had posterolateral injuries with associated cruciate ligament tears. This resulted in a total incidence of 5.8%. In a consecutive MRI series of 481 knees, Miller and coworkers reported a 6% incidence in 30 patients with posterolateral knee injuries.74 The difficulty in diagnosing posterolateral injuries has led authors to believe that the overall incidence of these injuries has been undetected or underreported.41,42,46,75 Fanelli and Edson reported on a consecutive group of 222 patients with a knee hemarthrosis in a tertiary trauma center, with 28.4% having posterolateral injuries.76 Within this group of 222 patients, 85 had PCL tears, of which 53 (62%) had concurrent posterolateral injuries. Also, 148 of the 222 patients had ACL tears, and 18 (12%) had concurrent posterolateral injuries. LaPrade and colleagues reported on a group of 100 consecutive ACL reconstructions with an 11% incidence of concurrent posterolateral instability.25 In summary, the incidence of posterolateral injuries appears to be between 5.8% and 11%, with up to 28.4% in a tertiary trauma setting. This is in agreement with the belief of most authors that the true incidence of posterolateral knee injuries has been underreported. Hopefully, improved awareness of these injuries and better clinical acumen in diagnosing them will provide future studies with more accurate data.

Cause of Graft Failure Undetected or untreated injuries to the PLC have been identified as a cause of failure of anterior and PCL reconstructions.77-79 Noyes and colleagues reported on a consecutive series of 41 patients with ACL tears, genu varus alignment, and posterolateral instability.79 Of the

41 patients, 15 had a combined total of 19 previous ACL reconstructions that failed. They believed that most of the failures were due to unrecognized posterolateral instability. Noyes and Barber-Westin also reported that between 1990 and 1996, 18 (30%) of 57 patients who underwent revision ACL reconstruction with bone–patellar tendon–bone autografts had untreated or unrecognized posterolateral injuries, apparently contributing to graft failure from the index reconstruction.77 LaPrade and coworkers used a selective ligament cutting technique to determine whether untreated injuries of the posterolateral structures of the knee contribute to increased force on an ACL reconstruction graft, thus potentially leading to graft failure.17 The force on the ACL graft during varus loading at both 0 and 30 degrees of knee flexion was significantly higher after cutting the FCL. The increase in graft force remained significant with additional sequential cutting of the popliteofibular ligament and popliteus tendon. Harner and colleagues evaluated a PCL reconstruction in isolated and combined injury models using a roboticuniversal force-moment sensor testing system.32 In the isolated injury model, reconstruction reduced posterior tibial translation to within 1.5 mm of the intact knee at 30 degrees and 2.4 mm at 90 degrees of knee flexion with a 134-N posterior tibial load applied to the knee. In the combined injury model, posterior tibial translation was increased by 6.0 mm at 30 degrees and 4.6 mm at 90 degrees for the PCL reconstruction compared with the intact knee. Also, external rotation increased up to 14 degrees, and varus rotation increased up to 7 degrees. In situ forces in the PCL graft also increased significantly (22% to 150%) for all loading conditions when the posterolateral structures were cut. Another study by LaPrade and colleagues analyzed untreated posterolateral knee injuries with PCL reconstruction.33 A significant increase in force on the PCL graft was seen when the popliteofibular ligament, popliteus tendon, and fibular collateral ligament were cut compared with intact posterolateral structures when both a varus moment and external rotation torque at 30, 60, and 90 degrees of knee flexion was applied.

Functional Limitations Many authors believe that most patients with PCL tears and functional instability with activity have associated posterolateral rotatory instability of the knee.76 Kannus observed that patients with posterolateral injuries have a higher incidence of osteoarthritis over time.37 Twenty-three patients were followed who had been treated nonoperatively for a grade II or III sprain of the posterolateral complex. At an average of 8 years after the injury, the 11 patients with a grade II sprain had an excellent or good result as assessed with standardized scales. Nine were asymptomatic; however, all had residual laxity. The 12 patients with a grade III sprain had much worse results, with average scores of either fair or poor. Six of the 12 patients with a grade III injury had post-traumatic arthritis on radiographic evaluation, whereas none of the patients with grade II injuries had any radiographic evidence of arthritic changes. A biomechanical study by Skyhar and colleagues used Fuji pressuresensitive film to observe the articular contact pressures in

1744 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

10 cadaveric knees before and after sequential sectioning of the PCL and the posterolateral complex (the posterolateral capsule, the popliteus muscle and tendon, and the fibular collateral ligament).38 Patellofemoral pressures and quadriceps load were most significantly elevated after combined sectioning of the PCL and the posterolateral complex. Medial compartment pressure was significantly elevated after sectioning of the PCL. They concluded that patients with combined PCL and posterolateral injuries need to be informed about the increased risk for osteoarthritis of these compartments if their injuries were not treated.

Treatment Options For grade I and grade II injuries, nonoperative treatment is almost always recommended.62 Baker and associates observed 13 patients treated nonoperatively who were believed to have isolated grade I PLC injuries.46 All 13 were noted to have returned to their full preinjury level of activity. It is well documented that grade III posterolateral injuries have a low likelihood of healing, and operative treatment is necessary to ensure the best clinical outcome possible.7,37 Furthermore, the results of an acute repair are much better than repair or reconstruction of chronic injuries.42,44,78 However, Stannard and colleagues recently reported a nonrandomized study comparing acute PLC repair to acute reconstruction.69 Fifty-six patients with 57 PLC injuries had a minimum 24-month follow-up. Acute primary repair was performed on 35 patients, with 22 successful outcomes and 13 (37%) failures. Primary reconstructions were performed on 22 patients, with 20 successful outcomes and 2 (9%) failures. They concluded that repair followed by early motion rehabilitation was significantly inferior compared with reconstruction using the modified two-tailed technique.

SPECIAL CONSIDERATIONS Skeletally Immature Patients The basic anatomy and biomechanics of the lateral and posterolateral corners of the knee is described earlier in this chapter. The knee capsule, ligaments, and cartilaginous structure of the knee first appear as an undifferenti­ ated blastema cell mass at about 5 weeks’ gestation. By week 7 of embryonic development, the individual structures can be identified. PLC knee injuries are rare in skeletally immature populations; however, physeal fractures are more common. Stress radiographs can be extremely helpful in this ­population.

Chronic Reconstructions The first step in assessing chronic PLC instability involves assessing the patient’s alignment. It has been well documented that failure to correct concomitant genu varum results in failed operative repair or reconstruction of chronic posterolateral injuries.25,79 Varus alignment and the resulting lateral thrust with foot strike place excessive

tension on the repaired or reconstructed lateral structures. Thus, it is vital to correct any genu varus alignment of the lower extremity before any soft tissue reconstructions of the PLC of the knee. Once again, a medial compartment unloader brace may be beneficial in determining who will have improved stability with a proximal tibial osteotomy.

Proximal Tibial Opening Wedge Osteotomy for Genu Varus and Chronic Posterior Cruciate Ligament Instability Preoperatively, long-leg anteroposterior films are obtained to determine the mechanical axis of the lower extremity. If the mechanical axis falls medial to the medial tibial spine, a proximal tibial osteotomy is recommended. If concurrent ACL instability is noted, it is recommended that the sagittal slope be tilted anteriorly to address this laxity pattern (Fig. 23F-32). Conversely, for patients with a concomitant PCL tear and chronic PLC instability, an increase in the posterior tibial slope is recommended. In patients with genu recurvatum, it is thought that increasing the sagittal slope of the tibia posteriorly increases stability. In pure chronic grade III PLC injuries, a pure valgus opening wedge osteotomy is recommended.

Proximal Tibial Osteotomy Technique A standard surgical incision is made vertically on the tibia medially between the tibial tubercle and the posterior border of the tibia to a point just distal to the midportion of the tibial tubercle. The incision is made directly down to bone to avoid skin flap formation. A small periosteal elevator is used to perform a subperiosteal dissection anteriorly under the deep infrapatellar bursa and patellar tendon just proximal to the tibial tubercle. A small Z-retractor is then placed into this location to identify where to make the proximal tibial anterior cut. Posteriorly, a subperiosteal dissection of the tibial collateral ligament and popliteus off the proximal tibia is performed. A radiolucent retractor is placed posteriorly to protect the neurovascular bundle during the subsequent portions of the procedure. Fluoroscopy is used to confirm the placement of two guidewires parallel to the joint line starting just distal to the flare of the proximal medial tibia (Fig. 23F-33A). The slope of the guidewires should attempt to replicate the desired sagittal slope of the proximal tibia after the osteotomy is completed. Initially, an oscillating saw performs the osteotomy. Secondarily, osteotomes are used to complete the osteotomy anteriorly and posteriorly (see Fig. 23F-33B). The lateral-most 1 cm of the tibia should be preserved as the osteotomy is hinged from this location. A spreader device is then slowly inserted into the defect, and the osteotomy is gradually opened to the desired correction. Care is taken to prevent propagation of the osteotomy either into the lateral cortex or the intra-articular portion of the lateral compartment. If the lateral cortex becomes compromised, a bone staple is placed laterally to provide stability. A plate is then used to secure the osteotomy. With concomitant ACL insufficiency, the plate is placed as far posterior as needed to decrease the slope of the proximal tibia so that the anterior tibial cortex acts as a hinge with the lateral cortex. Conversely, PCL insufficiency requires the

knee 1745

A

B

C

D

F

E Figure 23F-32   Chronic combined posterior cruciate ligament and posterolateral corner ligament injury with varus malalignment. A, Preoperative clinical photograph of knee at 90 degrees of flexion without stress. B, Preoperative clinical photograph of a 3+ posterior drawer. C, Preoperative bent knee stress radiograph demonstrating increased posterior sag consistent with a posterior cruciate ligament injury. D, Preoperative varus stress radiograph consistent with a posterolateral corner ligament injury. E, Preoperative longleg radiograph demonstrating varus malalignment. F, Preoperative lateral radiograph demonstrating the patient’s natural posterior tibial slope.

plate to be placed more anteriorly. Standard screw fixation is performed, filling the most proximal and distal holes first, then filling the central holes. An allograft or autograft cancellous bone graft is placed into the osteotomy site. Excess bone graft is removed to minimize the chance of drainage postoperatively (see Fig. 23F-33C and D). The periosteum is closed over the plate followed by subcutaneous

and then skin tissue. Postoperatively, quadriceps sets are frequently performed in a knee immobilizer. The immobilizer is removed 4 times daily to work on range of motion. By 2 weeks after surgery, a goal of 90 degrees of range of motion is usually met. The patient is kept non–weightbearing in a knee immobilizer for 8 weeks. If consolidation of trabecular bone on the lateral portion of the osteotomy

1746 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

D

E

F

Figure 23F-33   Opening wedge proximal tibial osteotomy. A, Fluoroscopic (anteroposterior) view demonstrating that the guide pin is parallel to the joint line before performing the proximal tibial osteotomy. B, Completed proximal tibial osteotomy with cut placed strategically to increase the posterior slope. C, Proximal tibial opening wedge osteotomy with spreader device in place. D and E, Anteroposterior and lateral fluoroscopic image demonstrating completed proximal tibial osteotomy with anteriorly placed osteotomy plate and bone graft. Posterior tibial slope has been increased in order to decrease the posterior drawer. F, Immediate postoperative clinical photograph demonstrating minimal posterior drawer after sagittal plane correction with opening wedge osteotomy.

is seen, the patient is gradually allowed to increase their weight-bearing from 25% initially, increasing by 25% weekly, to full by 3 months after surgery. After 6 months, standing long-leg radiographs are obtained to confirm corrective coronal and sagittal plane alignment and to determine whether a second-stage reconstruction is necessary. It is important to note that the osteotomy site must be

completely healed before proceeding to the second-stage reconstruction, which is typically 8 to 9 months after the osteotomy. The hardware is removed at the time of cruciate and PLC reconstructions. Multiple anatomic and nonanatomic techniques have been described; the choice of technique should be based on the injury pattern and surgeon preference.

knee 1747

C

r i t i c a l

P

S U G G E S T E D

o i n t s

l

I mportant structures include the FCL, popliteus muscletendon complex, and PFL. l The PLC structures provide varus, external rotational, and posterior stability to the knee. l PLC injuries are underdiagnosed; therefore, high suspicion and thorough examination are critical. l PLC injuries are often associated with concomitant injuries (e.g., ACL, PCL injuries). l Untreated PLC injuries result in increased failure rates of ACL and PCL reconstruction and poorer outcomes. l Malalignment must be addressed before or in conjunction with PLC reconstruction. l Acute treatment has improved outcomes compared with chronic treatment. l Reconstruction procedures have improved outcomes compared with primary repair procedures.

R E A D I N G S

Harner CD, Vogrin TM, Hoher J, et al: Biomechanical analysis of a posterior cruciate ligament reconstruction: Deficiency of the posterolateral structures as a cause of graft failure. Am J Sports Med 28:32-39, 2000. Hughston JC, Andrews JR, Cross MJ, et al: Classification of knee ligament instabilities. Part II: The lateral compartment. J Bone Joint Surg Am 58:173-179, 1976. Kannus P: Non-operative treatment of grade II and III sprains of the lateral ligament compartment of the knee. Am J Sports Med 17:83-88, 1989. LaPrade RF: Arthroscopic evaluation of the lateral compartment of knees with grade 3 posterolateral complex knee injuries. Am J Sports Med 25:596-602, 1997. LaPrade RF: Posterolateral knee injuries: Anatomy, evaluation, and treatment. New York, Thieme, 2006. LaPrade RF, Gilbert TJ, Bollom TS, et al: The magnetic resonance imaging appearance of individual structures of the posterolateral knee: A prospective study of normal knees and knees with surgically verified grade III injuries. Am J Sports Med 28:191-199, 2000. Noyes FR, Barber-Westin SD: Surgical restoration to treat chronic deficiency of the posterolateral complex and cruciate ligaments of the knee joint. Am J Sports Med 24:415-426, 1996. Skyhar MJ, Warren RF, Ortiz GJ, et al: The effects of sectioning of the posterior cruciate ligament and the posterolateral complex on the articular contact pressures within the knee. J Bone Joint Surg Am 75:694-699, 1993.

R e f erences Please see www.expertconsult.com

S e c t i o n

G

Multiple Ligament Knee Injuries Gregory C. Fanelli, Justin D. Harris, Daniel J. Tomaszewski, John T. Riehl, Craig J. Edson, and Kristin N. Reinheimer

Knee dislocations are true orthopaedic emergencies. They can result from high-energy trauma in addition to lowenergy mechanisms. The observed incidence of these injuries is relatively low (less than 1 in 100,000 of all hospital admissions).1 However, this number is likely under­representative of the true incidence because many knee injuries spontaneously reduce and do not demonstrate radiographic evidence of dislocation at initial presentation. Most knee dislocations involve tears of the central pivot, including both the anterior cruciate ligament (ACL) and the posterior cruciate ligament (PCL) along with one or both of the collateral ligaments. In addition to ligamentous injuries, significant capsular and meniscal injury can be present. Fractures and extraneous trauma are not uncommon. Vascular and nervous injuries occur relatively frequently in the multiple ligament–injured knee. They must be evaluated promptly and comprehensively. A detailed neurovascular examination is essential both before and after reduction. Historically, nonoperative treatment consisting of immobilization was deemed acceptable in most instances. With the advent of current arthroscopic techniques of multiple ligament reconstruction, nonoperative treatment is frequently limited to patients with very low functional demands.

RELEVANT ANATOMY AND BIOMECHANICS The knee is primarily a ginglymus (hinge) joint that allows some rotation of the tibia on the femur. Normal range of motion is from 0 degrees of extension to roughly 140 degrees of flexion. Internal and external rotation are typically 10 degrees in either direction, and external rotation allows “locking” of the knee joint in full extension. Stability of the knee is maintained in part by the bony articulation between the femoral condyles and the tibial plateau. Medial and lateral menisci increase the contact surfaces and thus increase static stability to the joint.

Osseous Anatomy The osseous anatomy of the knee consists of the primary articulations of the distal femur, the proximal tibia, and the patella. The distal femur is divided into medial and lateral condyles. The size of the condyles is asymmetric, with the medial condyle projecting more distally and the lateral condyle projecting more anteriorly. The condyles are separated by the trochlear groove, which makes up the patellofemoral articulation.2

1748 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The tibial plateau has an approximate 10-degree posterior slope. The medial plateau is slightly concave, whereas the lateral plateau has a more rounded appearance. Although the tibial plateau is somewhat flattened relative to the curved distal femur, congruency is maintained within the knee as the menisci help to increase conformity within the tibiofemoral articulation. The tibial spines separate the medial and lateral plateaus and serve as attachments for the menisci and cruciate ligaments. The patella is the largest sesamoid bone in the body and serves as a fulcrum for the extensor mechanism as well as providing a protective surface for the anterior aspect of the knee.

Ligamentous Anatomy The most significant ligamentous stabilizers of the knee include the ACL, PCL, medial collateral ligament (MCL), and lateral collateral ligament (LCL). In addition, the posteromedial and posterolateral corners (PLCs) of the knee are important structures that also contribute to knee ­stability. The ACL originates on the posteromedial lateral femoral condyle and courses anteriorly and distally to insert anterior and between the intercondylar eminences on the tibial plateau. The ACL consists of two anatomic bundles. The anteromedial bundle is taut in flexion, whereas the posterolateral bundle is more convex and tight in extension.3 The ACL is typically 30 to 38 mm in length and 10 to 12 mm width. It is an intra-articular structure, yet it has its own synovial membrane. It receives its blood supply from the middle geniculate artery and is innervated by the posterior articular nerve.4 The primary function of the ACL is to resist anterior translation of the tibia relative to the distal femur, but it also serves as a secondary adjunct to varus and valgus stability in full extension. The PCL has a broad femoral origin on the posterolateral aspect of the medial femoral condyle and inserts centrally on the posterior tibial plateau. It is an intra-articular structure but is also encompassed by its own synovial sheath. The anterolateral bundle of the PCL is tight in flexion, whereas the posteromedial bundle receives more tension in extension.5 These bundles are supplemented by the posterior meniscofemoral ligaments. The average length of the PCL is 38 mm, and its width is 13 mm.6 The vascularity of the PCL is supplied by the middle geniculate artery, and it is innervated by nerve fibers from the pop­liteal plexus from the tibial and obturator nerves.7 The PCL resists posterior translation of the tibia and is a secondary restraint to tibial external rotation. The MCL and the posteromedial corner are the primary restraints to valgus stress in the knee. The anatomy of the medial side of the knee has been described by Warren and Marshall8 in terms of layers. The most superficial layer is the sartorial fascia. The second layer consists of the superficial MCL. The deep MCL and the medial joint capsule are found in layer three. Alternatively, the medial side of the knee can be divided into thirds from anterior to posterior. The anterior third consists of capsular ligaments covered by the extensor retinaculum. The middle third contains the superficial and deep

MCL. The ­posteromedial corner occupies the ­posterior third and includes the posterior oblique ligament, the oblique popliteal ligament, and the termination of the ­semimembranosus. The superficial MCL is the primary restraint to valgus stress of the knee at 30 degrees of knee flexion. Its origin is on the medial epicondyle of the distal femur, and it inserts just posterior to the insertion of the pes anserinus. The posterior oblique ligament, semimembranosus, and oblique popliteal ligament resist valgus stress in full extension as well as anteromedial rotatory instability. The lateral side of the knee has also been divided into layers.9 The most superficial layer consists of the iliotibial band and the biceps femoris. The peroneal nerve lies deep to the biceps at the level of the distal femoral condyle. The middle layer consists of the patellar retinaculum anteriorly and the patellofemoral ligaments posteriorly. The deep layer, layer three, consists of the LCL, popliteal tendon, popliteofibular ligament, fabellofibular ligament, arcuate ligament, and lateral joint capsule. The LCL is the primary restraint to varus stress with the knee in 30 degrees of flexion. It originates on the lateral femoral condyle, just superior and posterior to the axis of motion. It attaches on the fibular head. The remaining structures in layer III make up the PLC. The PLC provides static support to resist posterior translation of the tibia as well as external rotation and varus angulation.

Neurovascular Anatomy A detailed understanding of the neurovascular anatomy of the knee is critical if one is going to treat patients with knee injuries. The neurovascular bundle within the pop­ liteal fossa is at great risk in knee dislocations because of a few anatomic features. The popliteal fossa is bordered by the biceps femoris tendon at its superior lateral border, the semimembranosus muscle superomedially, and the two heads of the gastrocnemius muscle inferiorly. Within the popliteal fossa, the popliteal artery and vein are separated by a thin layer of fat from the underlying posterior joint capsule. Crossing through the popliteal fossa, from superficial to deep, are the posterior tibial nerve, the popliteal vein, and popliteal artery. With the knee in full extension, the popliteal fascia is tensioned, making palpation of the popliteal artery difficult. Proximally, the popliteal artery emerges from the adductor hiatus and is tethered to this fibrous tunnel. Distally, the popliteal artery is also relatively immobile as it enters another fibrous canal deep to the soleus. These two somewhat immobile points leave the popliteal artery vulnerable to injury when the knee is dislocated. Superior, inferior, and middle geniculate branches branch off of the popliteal artery but are unable to maintain adequate collateral circulation in the event of a vascular injury. As the sciatic nerve emerges into the popliteal space from between and deep to the long head of the biceps and the semitendinosus muscles, it divides into the tibial and common peroneal nerves at a variable level. After dividing, the common peroneal nerve continues along the lower edge of the biceps toward the fibula, crossing superficially to the lateral head of the gastrocnemius.

knee 1749

The tibial nerve continues down the middle of the popliteal fossa and gives off muscular branches to the plantaris and gastrocnemius muscles. The common peroneal nerve then courses distally around the fibular head to innervate the anterior and lateral compartments of the lower leg.

CLASSIFICATION Numerous classification systems exist to describe knee dislocations. The most commonly used method is by describing the direction of displacement of the proximal tibia relative to the distal femur. However, this system does not account for spontaneously reduced dislocations and may fail to recognize other variations of the multiple ligament–injured knee. Additional characteristics that may help to classify a knee dislocation include the mechanism of injury, the presence or absence of open wounds, the degree of displacement, and the status of the neurovascular structures. In practice, all these characteristics are helpful in the classification of the multiple ligament–injured knee and can assist in determining ­optimal treatment. The directional classification of knee dislocations is based on the position of the tibia relative to the distal femur. Anterior dislocations occur following a hyperextension injury greater than 30 degrees and are the most common directional dislocation. Posterior knee dislocations occur in 25% of all knee dislocations and typically result from a posteriorly directed force applied to the proximal tibia. Lateral, medial, and rotatory dislocations have also been described.10 High-energy dislocations typically follow motor vehicle collisions and falls from height. Low-energy injuries typically refer to those that occur during athletic activities.11 An ultralow-energy dislocation has been described in morbidly obese patients who sustain severe ligamentous injury following seemingly trivial trauma.12 An anatomic classification system was created by Schenck (Table 23G-1), in which these injuries are classified based on the specific structures about the knee that are compromised. This system describes the ligamentous injury pattern and uses the letter C to designate a circulatory injury, whereas the letter N indicates neurologic injury. It has been used by some authors to direct treatment and predict outcome.13

   TABLE 23G-1  Schenck Anatomic Classification of Knee Dislocations Type

Description

KD-I KD-II

Dislocation with intact posterior cruciate ligament Dislocation with complete bicruciate disruption, collaterals intact Complete bicruciate disruption with one collateral disrupted Complete bicruciate dislocation with both collaterals disrupted Dislocation with periarticular fracture Associated vascular injury Associated neurologic injury

KD-III KD-IV KD-V C N

EVALUATION Clinical Presentation and History Knee dislocations, resulting in multiple ligament injuries, are seen in as few as 0.001% of all patients receiving attention for orthopaedic injury.1,14,15 Despite a relatively low annual incidence, a thorough evaluation and comprehensive physical examination are essential because the diagnosis is often missed in the setting of multi-injury trauma. Two common mechanisms of injury have been identified. Sporting injuries often represent a lower energy mechanism, and patients can often present with an isolated knee injury. In contrast, high-energy trauma involving motor vehicles can present a very different picture. Many times, a multiple ligament–injured knee can be overlooked as extremity trauma takes a back seat to more pressing, life-threatening injuries. There is also a subset of patients whose knee dislocations may spontaneously reduce before formal orthopaedic evaluation. The direction of the force vector applied to the knee determines the ultimate position of the dislocation as well as the structures that become injured. Anterior dislocation of the proximal tibia relative to the distal femur typically results from hyperextension of the knee. A posteriorly directed force, such as seen in a dashboard injury, results in posterior dislocation of the tibia. Pure varus or valgus stresses lend to lateral or medial dislocations, respectively. Combined forces can lead to rotational dislocation. In any case, patients with a plausible mechanism and knee pain should have a complete and through knee examination. A high index of suspicion for neurovascular insult is essential.

Physical Examination and Testing Evaluation of the knee with multiple ligament involvement mandates a systematic approach in order to accurately identify all potential injuries. A comprehensive musculoskeletal and neurovascular examination supplemented with appropriate ancillary studies helps the physician formulate a treatment plan. A knee dislocation represents the most dramatic example of the multiple ligament–injured knee. Obvious deformity may be present and a grossly dislocated knee is unlikely to escape diagnosis. However, dislocations that have spontaneously reduced may present more subtly. Complete disruption of two or more knee ligaments should alert the clinician to the possibility of a spontaneously reduced knee dislocation.16 Without proper evaluation and treatment, considerable adverse sequelae and morbidity may result. A thorough history, including mechanism and position of the limb at the time of injury may provide clues to possible ligamentous involvement. Any manipulation of the limb before the patient’s arrival in the emergency department should be noted. A detailed secondary survey should be conducted to assess the other extremities for signs of trauma. The contralateral lower extremity can also be used as a control to compare side-to-side differences with the injured knee with regard to ligament integrity and neurovascular status.

1750 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Examination begins with simple inspection (Fig. 23G-1). Resting limb posture should be noted. The presence of lacerations, deformity, skin dimpling, swelling, ecchymosis, and peripheral skin color changes can all provide valuable clues to the mechanism and severity of injury and can potentially affect management. For instance, a “dimple sign” about the anteromedial surface of the knee has been associated with the medial femoral condyle being incarcerated within the medial joint capsule. This injury may be irreducible by closed means and may necessitate immediate open reduction in the operating room to prevent skin necrosis. Likewise, a traumatic arthrotomy, if present, should be identified quickly because open knee dislocations portend a particularly poor outcome. Complete neurovascular examination is the most essential aspect of the initial evaluation. A detailed sensory and motor examination is often difficult to perform in a polytrauma patient. Integrity of the tibial and peroneal nerves should be evaluated and documented. This serves mostly as a baseline for comparison when serial neurovascular examinations prove necessary. More important is the continuity and function of the popliteal artery. The incidence of vascular injury in patients with knee dislocations has been estimated to be anywhere from 16% to 64%.14,17 The popliteal artery is particularly vulnerable because it is relatively tethered between the adductor hiatus and the gastrocnemius-soleus arch. Active hemorrhage, expanding hematoma, and a bruit in the popliteal fossa are all signs of a vascular injury. Two discrete injury patterns to the popliteal artery have been described. Stretching of the popliteal artery, as is often seen with knee hyperextension and anterior dislocation, results in intimal injury. This patient may have a well-perfused limb with intact peripheral pulses. The only sign of injury may be a reduced palpable pulse or Doppler signal compared with the contralateral limb. This is a worrisome injury. The vessel can develop late thrombosis, which can threaten the viability of the entire lower limb if not found promptly. Faithful serial vascular examinations are mandatory. The other described injury pattern involves direct contusion or transection of the artery as is seen with

posterior knee dislocations.10 This injury should be evaluated emergently by a vascular surgeon. Under no circumstances do the presence of pulses rule out arterial injury. The remainder of the physical examination involves evaluation of all extremities with particular attention to the affected knee. Gross deformity or bony crepitus should be noted. Testing the integrity of the major knee ligaments is often difficult in the acute trauma setting. A comprehensive examination generally requires conscious sedation or general anesthesia. A stabilized Lachman’s test, in which the examiner places his or her thigh under the affected knee, allows for a reasonably comfortable and accurate evaluation of anterior and posterior end points. Gross varus or valgus laxity in full extension implies disruption of a collateral ligament, one or both cruciates, and associated capsular injury. Varus or valgus laxity in 30 degrees of flexion better isolates the lateral and medial collateral ligaments, respectively (Fig. 23G-2).

Figure 23G-1   Patient with bilateral multiple ligament– injured knees. Contusions indicate the possibility of knee ligament injury.

Figure 23G-2   Severe valgus laxity in full extension indicating combined posterior and anterior cruciate ligament medial-sided complete tears.

Associated Injuries As stated, knee dislocations often occur in the setting of high-energy trauma. As such, a comprehensive primary and secondary trauma survey are necessary to rule out internal injury as well as injury to the remainder of the axial and appendicular skeleton. Any or all of the four major ligaments about the knee as well as the posteromedial and PLCs can be compromised. Additionally, vascular and neurologic injuries are common. Furthermore, osseous injuries to the distal femur, proximal tibia, patella, and proximal fibula are occasionally associated with knee dislocations. Meniscal tears and osteochondral injuries can also occur.

knee 1751

Vascular Injuries Vascular injury following a knee dislocation is not uncommon. The estimated incidence in the literature is anywhere from 16% to 64%,14,17 with most studies citing a 32% incidence.10 As mentioned, popliteal artery trauma can occur by direct rupture or intimal tear. Both can lead to disastrous consequences if there is a delay in diagnosis or treatment. Because the popliteal artery is an end artery to the leg, with minimal collateral circulation provided by the geniculate system, any compromise to the point of prolonged obstruction often leads to ischemia and eventual amputation. Furthermore, the popliteal vein is responsible for most of the venous outflow from the knee. Injury to this structure also compromises the viability of the lower limb. As stated previously, one cannot assume the absence of vascular injury simply because pulses are present. Serial vascular examinations are mandatory because intimal flaps can often present as delayed thrombus formation. Additionally, the absence of pulses implies an arterial injury and cannot be attributed to vascular spasm. Failure to recognize an arterial injury can lead to disastrous outcomes. The diagnosis of vascular injury is a clinical one. Any signs of limb ischemia should be taken seriously. The physical examination includes assessment of the pulses by palpation with comparison with the contralateral limb. Doppler ultrasound and ankle-brachial indices (ABIs) are indicated if pulse status is questionable. Distal perfusion, as well as motor and sensory function, should be documented. All these variables need to be reassessed on a frequent basis to evaluate for signs of vascular deterioration. If at any point asymmetry exists between the two lower extremities, urgent vascular studies are indicated. Difficulty ensuring vascularity clinically has always been well documented in numerous reports of knee dislocations.10,17-20 Misdiagnosis of vascularity and subsequent delay in arterial repair based on palpable peripheral pulses or capillary refill has been reported.18 Consequently, numerous authors advocate liberal or mandatory angiographic studies in patients with documented knee dislocations regardless of physical examination findings.21-24 Proponents of this philosophy have emphasized the risks for a missed arterial injury, including muscle ischemia and potential limb amputation. Late vascular compromise has been cited despite normal physical findings following closed reduction.19 The role of routine arteriography has undergone scrutiny in the recent literature and is quite controversial. Several authors have proposed using angiography only in patients who present with diminished or absent pulses and closely observing those with normal, symmetrical pulses.25-28 A recent prospective study compared the use of arteriography in patients with hard clinical signs of vascular injury with serial physical examinations in those without such signs. Physical examination alone yielded a 100% negative predictive value. The authors concluded that an invasive vascular study is not warranted in the absence of hard signs of vascular injury.28 Klineberg and associates29 retrospectively reviewed 57 knee dislocations and discovered that no patients who presented with normal vascular status on physical examination had a vascular injury as determined by

angiography or by clinical follow-up assessment. Stannard and coworkers30 prospectively evaluated 134 consecutive patients who had sustained a multiligamentous knee injury. Only patients with abnormal physical examination findings underwent arteriography. Ten patients had an abnormal physical examination, and arteriography confirmed nine vascular injuries. No cases of vascular injury were found in patients with normal examination findings. The value of the ABI has recently been evaluated as a means to diagnose arterial injury after knee dislocation. In a study by Mills and colleagues,31 38 patients were prospectively evaluated with ABIs after knee dislocations. Only patients whose ABI was lower than 0.90 underwent arteriography. Of the 38 patients, 11 (29%) had an ABI lower than 0.90. All 11 had an arterial injury requiring surgical intervention. None of the patients with an ABI greater than 0.90 experienced vascular compromise. The sensitivity, specificity, and positive predictive value of an ABI lower than 0.90 were 100%. Duplex ultrasonography has also been proposed as a safer, cheaper, and less invasive method of evaluating the popliteal vasculature. Results have been favorable showing a 98% accuracy of detecting extremity vascular trauma.32 Advocates state that ultrasonography offers yet another safeguard against missing a potentially disastrous injury.33 Opponents argue that ultrasounds are operator dependent, cannot account for distorted anatomy surrounding the knee following dislocation, and present an intrinsic delay because a technician must be present to complete the study.28 Less controversy exists regarding management of patients with hard signs of limb ischemia. Arteriography is not indicated in cases with an obviously ischemic limb because there is a danger in delaying revascularization. Many authors contend that arteriographic studies supply little additional information because the location of the lesion is readily predictable.10 Arteriography may be useful if more than one level of injury exists. This can often be accomplished with an intraoperative angiogram before surgical exploration. Angiography is clearly indicated in patients with abnormal or asymmetric vascularity following a knee dislocation in the absence of cold ischemia. It is also useful in those that have had a change in their vascular status following serial physical examinations. Any patient diagnosed with a vascular injury needs emergent vascular surgery consultation, and intervention. Delay in diagnosis or treatment jeopardizes limb viability.

Nerve Injuries Nerve injury can be quite common following dislocation of the knee. The orthopaedic literature estimates the incidence of nerve injury to be anywhere from 20% to 30%.1,1820,34,35 Although tibial nerve injuries have been reported,36 most nerve trauma involves the peroneal nerve. A common theme in peroneal nerve injuries is concomitant injury to the LCL and PLC. The tibial and peroneal nerves are less tightly tethered than the popliteal artery, which explains why they may be less prone to injury. Most injuries are the result of a stretch neurapraxia rather than laceration or frank transection. Recovery of nerve function is unpredictable with most series reporting no recovery in more than

1752 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

50% of injuries.14,34,37,38 It is essential to distinguish nerve injury from stocking paresthesias that may be indicative of a developing compartment syndrome. Multiple anatomic factors contribute to the propensity of peroneal nerve injury during knee dislocation. Studies have shown that the peroneal nerve has only 0.5 cm of excursion at the fibular head during knee motion.39 Additionally, there is a significantly smaller ratio of epineural tissue to axonal tissue in the peroneal nerve compared with other peripheral nerves making it prone to stretch injuries (Fig. 23G-3).40 Most series report that peroneal nerve continuity is usually maintained after knee dislocations. Reports of traumatic rupture, however, have been described.39,41,42 A diffuse zone of injury is typically encountered at the time of exploration, which correlates with the poor results seen after observation of complete nerve palsies.14,34,37,38 Evaluation of a potential nerve injury mandates a detailed history and physical examination. Motor function, paresthesias, and sensory loss must be documented initially and with serial examinations. Common peroneal nerve injury is the most frequent nerve injury seen following knee dislocations, followed by tibial nerve injury. Motor improvement can often be seen initially because motor grades are typically reduced after dislocations secondary to pain. Delayed neurologic deterioration is often related to swelling, hematoma, or direct compression from a splint or a cast. Electromyography (EMG) and nerve conduction studies can often be beneficial in determining the status of the motor axons in the peroneal nerve. Electromyographic changes that include fibrillation potentials, positive sharp waves, and absence of activity on voluntary effort indicate axon disruption. These changes do not typically appear for 2 to 4 weeks after injury, so EMG is not indicated in the acute setting. A neuropraxic lesion is present if there are no signs of denervation on EMG more than 3 weeks from the initial injury. Serial electromyography can be useful in following nerve recovery and regeneration. There are few reports on long-term treatment of peroneal nerve injury associated with knee dislocation. Management options include observation, neurolysis, primary repair, and

Figure 23G-3   Forced varus dislocation causing severe stretch and attenuation of peroneal nerve. The injury was permanent and required tendon transfers to restore ankle dorsiflexion.

neuroma excision and grafting. Goitz and Tomaino43 presented an excellent review of the management of peroneal nerve injuries associated with knee dislocations. Based on the current literature, the authors recommended observation of incomplete nerve palsies with reasonable anticipation of recovery. Partial or complete ruptures should be referred to an appropriate microsurgeon within 3 months. In cases in which a tension-free primary repair of a complete transection can be performed, consideration should be given to acute anastomosis. The patient can then be kept in a knee immobilizer for 3 weeks, at which time ligamentous reconstruction can be undertaken. This will allow for an uninterrupted postoperative rehabilitation. If the nerve is thought to be in continuity, electrodiagnostic studies should be performed at 6 weeks and 3 months. If no evidence of reinnervation is found by 3 months, consideration should be given to open exploration and neurolysis. If intraoperative nerve action potentials are absent over a small portion of the nerve, consideration can be given to excision of the affected portion with interposition of a cable graft.

Osseous Injuries Bony injuries about the knee may be seen in as many as 60% of all knee dislocations (Fig. 23G-4).20 Characteristic ligamentous and tendinous avulsions are common. These include Segond’s fractures, fibular head avulsions, and cruciate avulsions. These should be treated primarily as ligamentous injuries as opposed to major fractures that are seen in true fracture-dislocations of the knee. Major fractures of the tibial plateau and distal femur are not uncommonly associated with multiligamentous injury.

Figure 23G-4   Segmental tibia fracture accompanying ipsilateral knee dislocation.

knee 1753

Figure 23G-5   Significant lateral joint line opening on anteroposterior radiograph secondary to anterior and posterior cruciate ligament lateral-sided tears sustained during knee dislocation.

A comprehensive treatment plan is essential because fracture-dislocations certainly add an element of complexity to their treatment. It is essential to recognize that these high-energy fractures result in marked joint instability and are associated with a high risk for soft tissue and neurovascular compromise. Spanning external fixation is often a useful adjunct to assist in temporary stabilization of these injuries in anticipation of delayed bony and ligamentous reconstruction.

Imaging Imaging begins with anteroposterior and lateral radiographs in orthogonal planes before any manipulation. Although a frank dislocation is relatively easy to identify, a spontaneously reduced dislocation may be more surreptitious. One may only appreciate subtle asymmetry or widening of the joint, or the aforementioned bony injuries (Fig. 23G-5). Repeat radiographs are mandatory to confirm joint reduction if applicable. If there is any concern about the integrity of the popliteal artery, arteriography should be used (Fig. 23G-6). It remains the gold standard for assessment of intimal injury. Some centers have used magnetic resonance angiography (MRA) in the evaluation of possible vascular injury associated with knee dislocation. In addition to being less invasive, it avoids the potential for contrast reactions and arterial punctures. More information is needed before MRA supplants direct arteriography as the first-line vascular study in this patient population. After the acute management of the dislocated knee, magnetic resonance imaging (MRI) of the affected knee may be obtained to identify injury to the menisci, articular cartilage, ligaments, and some neurovascular structures. This study is of vital importance to ascertain the complete

Figure 23G-6   Arteriogram demonstrating intact popliteal artery system after knee dislocation.

nature of the injury. Numerous studies verify the high sensitivity and accuracy of MRI in this particular setting.44 MRI has become a routine imaging modality for preoperative evaluation and planning.

TREATMENT OPTIONS Nonoperative Direct comparison of treatment algorithms regarding the multiligamentous knee injury is nearly impossible. The orthopaedic literature can be difficult to interpret because these injuries can be quite heterogeneous and often result from a variety of mechanisms. The presence or absence of a neurovascular injury will have a direct correlation with patient outcome. Timing of surgery, surgical technique, and postoperative protocols can vastly differ among treating surgeons. All these factors make it difficult to identify clinical differences and statistical significance when comparing one treatment protocol to another. Until recent decades, most knee dislocations were traditionally managed conservatively with immobilization for several months.38,45 Various authors reported reasonable results after nonoperative treatment of this devastating injury.17,19 In 1991, Almekinders and Logan46 reported a retrospective review of their series of knee dislocations treated both operatively and nonoperatively. Nearly all had functional limitations. Both groups developed similar degenerative changes. The group treated operatively, however, had better motion (129 degrees versus 108 degrees of flexion) and superior ligamentous stability.

1754 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 23G-7   Spanning external fixation applied to severely unstable knee after ultralow-energy knee dislocation. Note the delta frame configuration necessary to achieve adequate stability.

Today, few indications for nonoperative treatment of the multiligamentous knee injury persist. Critically ill patients who are unable to tolerate a surgical procedure or comply with a postoperative protocol may be candidates for nonoperative management. Patients with grossly contaminated wounds and significant soft tissue injury about the surgical site should be considered carefully before any attempt at ligament reconstruction. Lastly, elderly patients with low functional demand and high operative risk may be best treated conservatively.11 The most basic method of nonoperative management is a long-leg or cylinder cast. A long-leg brace with the knee locked in extension may also suffice. A brace is useful when access to wounds about the knee is needed and in critically ill patients who may be prone to extreme fluid shifts and peripheral edema. In some instances, a knee-spanning external fixator may be practical (Fig. 23G-7). It may provide more stability than a brace while allowing access to the knee. Regardless of the type of nonsurgical management, frequent radiographs should be obtained to verify continued reduction of the knee.

Operative As stated earlier, there is great variability in the indications and technique used in the management of multiple ligament injured knees. In addition, many published studies are observational or in the form of a single author’s retrospective review. The literature, while flawed to some degree, still represents our best tool to mold a reasonable and effective treatment algorithm. Many early reports of surgical treatment were made based on the technique of formal arthrotomy and direct repair of the ligaments. One small review found similar results in conservative treatment and direct suture repair.34 Another article47 compared early versus late direct repair of torn ligamentous structures in 13 of 17 patients. They determined that patients treated with early repair fared better than those with repairs done in a delayed fashion. This study supported surgical ­ management of the

­ islocated knee and introduced the idea that long-term d benefits exist if ligamentous stability can be achieved in the knee. Since the early 1990s, arthroscopically assisted ACL and PCL reconstruction has gained popularity. Several factors have allowed this technique to thrive: (1) improved availability, procurement, sterilization, and storage of allograft tissue; (2) improved arthroscopic ­instrumentation; (3) better graft fixation methods; (4) improved surgical technique; and (5) improved understanding of the ligamentous anatomy and biomechanics of the knee. The literature is now becoming rife with evidence supporting this surgical approach to multiple ligament–injured knees.15,48-52 Timing of intervention varies greatly based on individual patient considerations. These patients have quite often sustained severe multisystem trauma. Life-threatening injuries obviously take precedence. That being said, open dislocations and dislocations associated with vascular injury are relative surgical emergencies because these represent limb-threatening conditions. These two situations call for provisional stabilization with external fixation combined with either irrigation and débridement in the case of an open knee injury or vascular repair or bypass in the setting of vascular insult. Our preferred surgical tactic is described later in the text.

Weighing the Evidence As stated previously, strategies for the management of knee dislocations have been varied and controversial.53-55 Over time, conservative management of these injuries has been associated with residual instability and poor long-term outcomes.38,56 Most experienced knee surgeons now recommend operative treatment of all compromised ligamentous, capsular, and meniscal structures.1,46,47 Controversy exists regarding operative technique, surgical timing, graft selection, and rehabilitation. The debate over the management of these complex injuries is for the most part due to inconsistent treatment protocols, small and poorly defined patient populations, and a variety of surgical techniques described in the literature.

Nonoperative versus Operative Treatment Multiple comparisons of operative and nonoperative treatment exist in the literature.54,57 A recent retrospective review by Richter and colleagues revealed that operative treatment was superior to conservative management in patients who had sustained traumatic knee dislocations.58 Patients treated operatively had either bicruciate repair or reconstruction using autograft tissue. Rehabilitation protocols varied based on the specific type of cruciate intervention. Patients treated nonoperatively were immobilized for 6 weeks in either a long-leg cast or an external fixator. At an average 8.2-year follow-up, the mean Lysholm and Tegner scores were significantly higher in the surgical group. Patients younger than 40 years of age, who had sports injuries, and who had functional rehabilitation rather than immobilization had better results. In an effort to end the controversy between nonoperative and operative treatment, Dedmond and Almekinders performed a metaanalysis literature review.59 The authors extracted raw data

knee 1755

from several other studies so that 206 knee dislocations were evaluated. Seventy-four of these injuries were treated nonoperatively, whereas 132 were treated operatively. The average Lysholm knee score for the surgical group was 85.2, and it was 66.5 for the nonoperative group. These results were quite similar to Montgomery and coworkers’ earlier review.60

Surgical Timing The optimal timing of operative intervention is another controversial issue in the literature. Acute surgical intervention avoids excessive scarring of the collateral ligaments, making primary repair less technically demanding, yet in some studies has been shown to have a higher incidence of motion loss at final follow-up.55,61 In the absence of an absolute surgical indication, most authors recommend delaying operative treatment for 1 to 3 weeks. This allows for a period of vascular monitoring in addition to minimizing swelling and allowing capsular injuries to heal, which may permit an arthroscopic reconstruction. The literature differentiates between acute and chronic reconstructions at the 3-week time frame. Acute primary repair of the PLC is nearly universally recommended. Medial-sided injuries are more controversial. Shelbourne and associates61 have recommended nonoperative treatment for MCL and lowgrade PCL injuries with delayed ACL reconstruction if patient symptoms and activity level dictate. They reported satisfactory results in nine patients treated in this manner and believed that the arthrofibrosis potentially associated with concurrent ACL reconstruction and MCL repair could be avoided. Their patients had an average extension loss of 3 degrees and flexion loss of 15 degrees. Fanelli and associates62 also showed successful bracing of low-grade MCL injuries followed by delayed central pivot reconstruction. In this series, there was no difference in postoperative Lysholm scores between those patients treated acutely and those treated in a delayed fashion. In contrast to the two previous studies, most series have reported better results in patients treated with acute ligamentous reconstruction. Noyes and Barber-Westin51 reviewed the results in 11 patients who had undergone allograft bicruciate and PLC reconstruction. Seven patients were treated acutely, whereas 4 patients were treated in a chronic setting. The patients who had undergone delayed reconstruction had lower overall ratings and more subjective difficulties during activity than did patients in the acutely treated cohort. Wascher and colleagues56 reported on 13 patients who underwent ACL and PCL ­reconstruction using allograft tissue. Nine patients were treated acutely. Better results were noted in those patients treated within 3 weeks of injury. A recent retrospective review by Harner and coworkers63 confirmed that patients treated in an acute fashion have better postoperative results. Nineteen of 31 patients in their series were treated with acute surgical reconstruction with allograft tissue. All knee rating scores were significantly better in the group treated with early surgery. There was no difference in final range of motion between the acutely and chronically treated cohorts. Laxity tests were consistently improved in both groups, but results were more predictable in the acutely treated patients.

Graft Selection Graft selection remains a matter of surgeon preference. Many authors have reported using allograft tissue in treating multiligamentous knee injuries.15,56,64,65 Reduced donorsite morbidity and shorter operative times are advantages of using allograft tissue. Fanelli and colleagues62 reviewed 20 bicruciate ligament reconstructions, in which 14 PCL reconstructions and 4 ACL reconstructions were performed with allograft tissue. After comparing KT-1000 results, the authors concluded that the autografts and allografts were equivalent. Cole and Harner reported only one graft failure out of 60 allograft cruciate reconstructions in a series of 31 patients.64 Ohkoshi and associates66 reviewed 13 knees that had been treated in a staged fashion with all autograft tissue. All patients underwent acute PCL reconstruction using contralateral hamstring tendons. Three months after surgery, the knees were reassessed, and any residual laxity was treated with delayed reconstruction. Anterior cruciate injuries were addressed using ipsilateral hamstring or patellar tendon autografts. Medial structures were reconstructed using ipsilateral semitendinosus, and lateral-sided injuries were augmented using autogenous biceps tendon. Average range of motion was 0 to 139.5 degrees of flexion. No collateral instability was noted at final follow-up, and average anteroposterior laxity was 2.3 mm.

Posterior Cruciate Ligament Reconstruction Significant refinements in PCL reconstruction have been developed recently as authors have begun to appreciate its importance in knee stability following multiligamentous injury. Controversies surrounding transtibial versus tibial inlay and single- versus double-bundle reconstruction continue. At the present time, no randomized prospective studies exist comparing various PCL reconstruction techniques in this particular patient population. Advocates of single-bundle transtibial PCL reconstruction cite multiple studies justifying its use.56,63,67,68 In both Harner’s and Fanelli’s series, all patients treated with this technique had grade I laxity or better on posterior drawer testing at final follow-up. The tibial inlay technique has been developed as a way to avoid the “killer turn.” Biomechanical studies have shown less graft abrasion and pretension with this method of PCL reconstruction.69,70 Recent studies have shown good clinical outcomes with the tibial inlay technique.71,72 In addition to the transtibial tunnel versus tibial inlay debate, controversy exists regarding single- versus doublebundle reconstruction. Normal PCL function is more accurately reproduced in biomechanical studies with a double-bundle technique.73,74 Single-bundle reconstructions have shown good clinical results for both transtibial62,63,75,76 and tibial inlay techniques.77 Double-bundle techniques are being developed, and early clinical results are promising.72,78-81 Stannard and colleagues72 were the first to report on double-bundle tibial inlay PCL reconstruction in the multiple ligament–injured knee. Allograft Achilles tendon was used in 29 patients with 30 knee dislocations to treat PCL insufficiency. All patients had concomitant PLC injuries, and 29 of 30 knees had ACL tears.

1756 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

At 2-year follow-up, PCL stability was grade 0 in 24 of 30 knees and grade I in the remaining 6. Cooper and Stewart71 performed single-bundle tibial inlay PCL reconstruction in 44 patients with multiligamentous involvement. At final follow-up of 39 months, all patients were improved subjectively. Patients with full-thickness chondral defects had significantly worse outcomes. Posterior stability was reconstituted in all patients.

REHABILITATION Stiffness and recurrent instability are two of the most common complications following operative treatment of multiligamentous knee injuries. Long-term immobilization usually imparts improved stability at the expense of motion. Early motion and aggressive rehabilitation lead to improved function, yet risk failure of ligament reconstructions. No true randomized study exists comparing early functional rehabilitation with prolonged immobilization following these injuries. Obviously, a delicate balance must be achieved between motion and stability to achieve optimal results. Shelbourne and associates11 had noted problems with stiffness following bicruciate ligament reconstruction. They modified their technique to include early PCL reconstruction with PLC repair. ACL reconstruction was performed on a delayed basis if necessary. Early active motion and continuous passive motion were used to ­optimize function. In their initial study, all patients trea­ted in this fashion achieved excellent range of motion. Noyes reported results on 11 bicruciate ligament reconstructions treated with early protected postoperative motion.51 For the first 4 weeks after surgery, patients began active-assisted range of motion from 10 to 90 degrees, 6 to 8 times a day. Between therapy sessions, the knee was

immobilized in full extension in a split cylinder cast. Full weight-bearing was allowed after 3 months. Nearly 5 years after surgery, the average knee range of motion was normal in 9 of 11 patients. One patient had a 5-degree loss of both flexion and extension, whereas a second patient reached only 100 degrees of knee flexion. Failure rates of PCL and ACL reconstruction were 18% and 9%, respectively. Nearly half of the patients in this series required surgical manipulation during their postoperative rehabilitation, and 2 patients were treated for motion loss with arthroscopic débridement of adhesions. Recent literature has focused on attempts to maintain early motion while minimizing shear forces on reconstructed grafts. Fitzpatrick and associates82 performed a cadaveric study in which knee stability was assessed biomechanically using an articulated external fixator. Anterior and posterior tibial translation was significantly decreased at 30 and 90 degrees of flexion, respectively, after application of the fixator to the cruciate-deficient knee. The theoretical advantage of this device would be to allow early range of motion in the immediate postoperative period while limiting posterior shear forces on the freshly reconstructed central pivot grafts. Stannard and colleagues83 performed a functional outcome study in which 40 patients with 43 knee dislocations were evaluated. Group A consisted of 12 patients who underwent multiligamentous knee reconstruction followed by placement of a hinged-knee external fixator. Group B included 27 knees that underwent an identical surgical and rehabilitation protocol with the exception that a brace was used in lieu of the hinged fixator. At an average 2-year follow-up, the ligament failure rates were 7% in group A and 29% in group B. No differences were noted in range of motion between the groups. Further study is needed to determine the optimal rehabilitation protocol in these patients.

Authors’ Preferred Method Fanelli Sports Injury Clinic Experience

Surgical Timing

Our practice is at a tertiary care regional trauma center. There is a 38% incidence of PCL tears in acute knee injuries, with 45% of these PCL-injured knees being combined ACL and PCL tears.75,84 Careful assessment, evaluation, and treatment of vascular injuries is essential in these acute multiple ligament–injured knees. There is an 11% incidence of vascular injury associated with these acute multiple ­ligament–injured knees at our center.19 Our preferred approach to combined ACL and PCL injuries is arthroscopic ACL and PCL reconstruction using the transtibial tunnel technique, with collateral and capsular ligament surgery as indicated.85 Not all cases are amenable to the arthroscopic approach, and the operating surgeon must assess each case individually. Surgical timing is dependent on vascular status, reduction stability, skin condition, systemic injuries, open versus closed knee injury, meniscus and articular surface injuries, other orthopaedic injuries, and the collateral and capsular ligaments involved.

Some ACL, PCL, and MCL injuries can be treated with brace treatment of the MCL followed by arthroscopic combined ACL and PCL reconstruction 4 to 6 weeks after ­healing of the MCL. Other cases may require repair or reconstruction of the medial structures and must be assessed on an individual basis. Combined ACL, PCL, and posterolateral injuries should be surgically addressed as early as is safely possible. ACL, PCL, and posterolateral repair and reconstruction performed between 2 and 3 weeks after injury allows healing of capsular tissues to permit an arthroscopic approach, and still permits primary repair of injured posterolateral structures. Open multiple ligament knee injuries and dislocations may require staged procedures. The collateral and capsular structures are repaired after thorough irrigation and dé­ bridement, and the combined ACL and PCL reconstruction is performed at a later date after wound healing has occurred.

knee 1757

Authors’ Preferred Method—cont’d

Figure 23G-8   Severe soft tissue injury in a patient with bilateral open knee dislocations. Skin issues are the most frequent surgical timing modifiers that cause delay of surgical reconstructions in multiple ligament–injured knees.

Care must be taken in all cases of delayed reconstruction that the tibiofemoral joint reduction is maintained. The surgical timing guidelines outlined previously should be considered in the context of the individual patient. Many patients with multiple ligament injuries of the knee are severely injured multiple-trauma patients with multisystem injuries. Modifiers to the ideal timing protocols outlined earlier include the vascular status of the involved extremity, reduction stability, skin condition, open or closed injury, and other orthopaedic and systemic injuries (Fig. 23G-8). These additional considerations may cause the knee ligament surgery to be performed earlier or later than desired. We have previously reported excellent results with delayed reconstruction in the multiple ligament–injured knee.49,62 Graft Selection

The ideal graft material should be strong, provide secure fixation, be easy to pass, be readily available, and have low donor site morbidity. Our preferred graft for the PCL is an Achilles tendon allograft for the anterolateral bundle and a tibialis anterior allograft for the posteromedial bundle. We prefer Achilles tendon allograft or other allograft for the ACL reconstruction. The preferred graft material for the PLC is allograft tissue.86 Cases requiring MCL and posteromedial corner surgery may have ­ primary repair or ­reconstruction, or a combination of both. Our preferred method for MCL and posteromedial reconstructions is a primary repair and posteromedial capsular ­ advancement with allograft augmentation as needed. Surgical Approach

Our preferred surgical approach is a single-stage arthroscopic combined ACL and PCL reconstruction using the transtibial tunnel technique with collateral and capsular ligament surgery as indicated. The PLC is repaired and then augmented with a split biceps tendon transfer, biceps tendon transfer, semitendinosus free graft, or allograft tissue. Acute medial injuries not amenable to brace treatment ­ undergo

Figure 23G-9   The surgical leg is draped free for multiple ligament reconstruction. A lateral post is used for stability. No leg holder is used. The fully extended operating table supports the well leg.

primary repair, and posteromedial capsular shift and allograft reconstruction as indicated. The operating surgeon must be prepared to convert to a dry arthroscopic procedure or to an open procedure if fluid extravasation becomes a problem. Surgical Technique

The principles of reconstruction in the multiple ligament– injured knee are identification and treatment of all pathology, accurate tunnel placement, anatomic graft insertion sites, use of strong graft material, secure graft fixation, and a deliberate postoperative rehabilitation program. The patient is positioned supine on the operating room table. The surgical leg hangs over the side of the operating table, and the well leg is supported by the fully extended operating table. A lateral post is used for control of the surgical leg. We do not use a leg holder (Fig. 23G-9). The surgery is done under tourniquet control unless prior arterial or venous repair contraindicates the use of a tourniquet. Fluid inflow is by gravity. We do not use an arthroscopic fluid pump. Allograft tissue is prepared, and arthroscopic instruments are placed with the inflow in the superior lateral portal, arthroscope in the inferior lateral patellar portal, and instruments in the inferior medial patellar portal. An accessory extracapsular extra-articular posteromedial safety incision is used to protect the neurovascular structures and to confirm the accuracy of tibial tunnel placement (Fig. 23G-10). The notchplasty is performed first and consists of ACL and PCL stump débridement, bone removal, and ­contouring of the medial wall of the lateral femoral condyle and the intercondylar roof. This allows visualization of the over-­the-top position and prevents ACL graft impingement ­throughout the full range of motion. Specially curved Arthrotek (Biomet Sports Medicine, Warsaw, Ind) PCL instruments are used to elevate the capsule from the posterior aspect of the tibia (Fig. 23G-11). Continued

1758 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

A

B

C

Figure 23G-10   A-C, The 1- to 2-cm extracapsular posterior medial safety incision allows the surgeon’s finger to protect the neurovascular structures and confirm the position of instruments on the posterior aspect of the proximal tibia. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

The PCL tibial and femoral tunnels are created with the help of the Arthrotek Fanelli PCL/ACL drill guide (Fig. 23G-12). The transtibial PCL tunnel is positioned from the anteromedial aspect of the proximal tibia 1 cm below the tibial tubercle to exit in the inferior lateral aspect of the PCL anatomic insertion site. The PCL femoral tunnel originates externally between the medial femoral epicondyle and the medial femoral condylar articular surface to emerge through

Figure 23G-11   Specially curved posterior cruciate ligament (PCL) reconstruction instruments used to elevate the capsule from the posterior aspect of the tibial ridge during PCL reconstruction. Posterior capsular elevation is critical in transtibial tunnel PCL reconstruction because it facilitates accurate PCL tibial tunnel placement and subsequent graft passage. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

the center of the stump of the anterolateral bundle of the PCL. The PCL graft is positioned and anchored on the femoral side, followed by PCL graft tensioning and tibial fixation (Fig. 23G-13). The ACL tunnels are created using the single-incision technique. The tibial tunnel begins externally at a point 1 cm proximal to the tibial tubercle on the anteromedial surface of the proximal tibia to emerge through the center of the stump of the ACL tibial footprint. The femoral tunnel is positioned next to the over-the-top position on the medial wall of the lateral femoral condyle near the ACL anatomic insertion site. The tunnel is created to leave a 1- to 2-mm posterior cortical wall so that interference fixation can be used. The ACL graft is positioned and anchored on the femoral side, followed by ACL graft tensioning and tibial fixation (Fig. 23G-14). Posterolateral reconstruction with the free-graft figureof-eight technique uses semitendinosus autograft or allograft, Achilles tendon allograft, or other soft tissue allograft material. A curvilinear incision is made in the lateral aspect of the knee extending from the lateral femoral epicondyle to the interval between Gerdy’s tubercle and the fibular head. The fibular head is exposed, and a tunnel is created in an anterior-to-posterior direction at the area of maximal fibular diameter. The tunnel is created by passing a guide pin followed by a cannulated drill, usually 7 mm in diameter. The peroneal nerve is protected during tunnel creation and throughout the procedure. The free tendon graft is then

knee 1759

Authors’ Preferred Method—cont’d

Figure 23G-12   The Arthrotek Fanelli posterior cruciate ligament (PCL) and anterior cruciate ligament (ACL) drill guide system is used to precisely create both the PCL femoral and tibial tunnels and the ACL single-incision technique and double-incision technique tunnels. The drill guide is positioned for the PCL tibial tunnel so that a guidewire enters the anteromedial aspect of the proximal tibia about 1 cm below the tibial tubercle, at a point midway between the posteromedial border of the tibia and the tibial crest anteriorly. The guidewire exits in the inferior lateral aspect of the PCL tibial anatomic insertion site. The guide is positioned for the PCL femoral tunnel so that the guidewire enters the medial aspect of the medial femoral condyle midway between the medial femoral condyle articular margin and the medial epicondyle, at least 2 cm proximal to the medial femoral condyle distal articular surface (joint line). The guidewire exits through the center of the stump of the anterolateral bundle of the PCL. The drill guide is positioned for the single-incision endoscopic ACL technique so that the guidewire enters the anteromedial surface of the proximal tibia about 1 cm proximal to the tibial tubercle at a point midway between the posteromedial border of the tibia and the tibial crest anteriorly. The guidewire exits through the center of the stump of the tibial ACL insertion. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

passed through the fibular head drill hole. An incision is then made in the iliotibial band in line with the fibers directly overlying the lateral femoral epicondyle. The graft material is passed medial to the iliotibial band, and the limbs of the graft are crossed to form a figure-eight. A drill hole is made 1 cm anterior to the fibular collateral ligament femoral ­insertion. A longitudinal incision is made in the lateral capsule just posterior to the fibular collateral ligament. The graft material is passed medial to the iliotibial band and secured to the lateral femoral epicondylar region with a screw and spiked ligament washer at the previously mentioned point. The posterolateral capsule that had been previously incised is then shifted and sewn into the strut of figure-eight graft tissue material to eliminate posterolateral capsular redundancy. The anterior and posterior limbs of the figure-eight graft material are sewn to each other to reinforce and ­tighten the construct. The procedures described are intended to eliminate posterolateral and varus rotational instability. Certain cases require reconstruction of the popliteus tendon using an additional graft passed through a transosseous drill hole through the proximal lateral tibia (Fig. 23G-15).

Figure 23G-13   Drawing demonstrates doublebundle double femoral tunnel posterior cruciate ligament reconstruction. Note primary fixation with resorbable interference screws and backup fixation with ligament fixation buttons and screw and spiked ligament washer. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

Posteromedial and medial reconstructions are performed through a medial hockey-stick incision (Fig. 23G-16). Care is taken to maintain adequate skin bridges between incisions. The superficial MCL is exposed, and a longitudinal incision is made just posterior to the posterior border of the MCL. Care is taken not to damage the medial meniscus during the capsular incision. The interval between the posteromedial capsule and medial meniscus is developed. The posteromedial capsule is shifted anterosuperiorly. The medial meniscus

Figure 23G-14   Anterior cruciate ligament reconstruction using the single-incision endoscopic technique. Drawing shows completed combined posterior and anterior cruciate ligament reconstruction. Note primary fixation with resorbable interference screws and backup fixation with ligament fixation buttons and screws and spiked ligament washers. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.) Continued

1760 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

A

A

B

Figure 23G-15   A, Surgical technique for posterolateral and lateral reconstruction using allograft or autograft figureof-eight reconstruction combined with posterolateral capsular shift and primary repair of injured structures, as indicated. This surgical procedure reproduces the function of the popliteofibular ligament and the lateral collateral ligament and eliminates posterolateral capsular redundancy. B, Reconstruction of the popliteus tendon in cases in which that reconstruction is indicated. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

is repaired to the new capsular position, and the shifted capsule is sewn into the MCL. When superficial MCL reconstruction is indicated, this is performed with allograft tissue or semitendinosus autograft. This graft material is attached at the anatomic insertion sites of the superficial MCL on the femur and tibia. The posteromedial capsular advancement is performed and sewn into the newly reconstructed MCL (see Table 23G-1).

B

Figure 23G-16   A and B, Surgical treatment of medial side injuries. Severe medial side injuries are successfully treated with primary repair using suture anchor technique combined with medial collateral ligament (MCL) reconstruction using allograft tissue and the posteromedial capsular shift procedure. The allograft can anatomically reconstruct the superficial MCL. The allograft is secured to the anatomic insertion sites of the superficial MCL using screws and spiked ligament washers. The posteromedial capsule can then be secured to the allograft tissue to eliminate posteromedial capsular laxity. This technique addresses all components of the medial side instability. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

the tibia is internally rotated, slight valgus force is applied to the knee, and final tensioning and fixation of the PLC is achieved. Reconstruction and tensioning of the MCL and posteromedial corner are performed after the ACL, PCL, and PLC reconstructions are completed, and are done in 30 degrees of knee flexion (Table 23G-2). Technical Hints

The posteromedial safety incision protects the neurovascular structures, confirms accurate tibial tunnel placement, and allows the surgical procedure to be done at an accelerated pace. The single-incision ACL reconstruction technique

Graft Tensioning and Fixation

The PCL is reconstructed first, followed by the ACL, followed by the posterolateral complex and medial side (Box 23G-1). The Arthrotek tensioning boot is used for tensioning the ACL and PCL reconstructions (Fig. 23G-17). Tension is placed on the PCL graft distally, and the knee is cycled through a full range of motion to allow pretensioning and settling of the graft. The knee is placed in 70 to 90 degrees of flexion, the Arthrotek tensioning boot is tensioned to 20 pounds to restore the normal tibial step-off, and fixation is achieved on the tibial side of the PCL graft with a screw and spiked ligament washer and an Arthrotek Bio-Core bioabsorbable ­ interference screw. The knee is maintained at 30 degrees of flexion, the Arthrotek tensioning boot is tensioned to 20 pounds with tension on the ACL graft, and final fixation is achieved of the ACL graft with an Arthrotek Bio-Core bioabsorbable interference screw and a Biomet Sports Medicine ligament fixation button or spiked ligament washer back-up fixation. The knee is then placed in 30 degrees of flexion,

Box 23G-1 Order of Anterior ­Cruciate ­Ligament, Posterior Cruciate ­Ligament, ­Posterolateral Corner, and Medial Collateral Ligament Reconstruction 1. Posterior cruciate ligament tibial tunnel 2. Posterior cruciate ligament femoral tunnel 3. Posterior cruciate ligament graft passage and femoral fixation 4. Anterior cruciate ligament tibial tunnel 5. Anterior cruciate ligament femoral tunnel 6. Anterior cruciate ligament graft passage and femoral fixation 7. Posterolateral corner repair and reconstruction 8. Medial-sided repair and reconstruction

knee 1761

Authors’ Preferred Method—cont’d

A

B

Figure 23G-17   The knee ligament-tensioning boot (A) is used to precisely tension posterior cruciate ligament (PCL) and anterior cruciate ligament (ACL) grafts. During PCL reconstruction, the tensioning device is attached to the tibial end of the graft and the torque wrench ratchet set to 20 pounds. This restores the anatomic tibial step-off. The knee is cycled through full flexionextension cycles, and with the knee at about 70 degrees of flexion, final PCL tibial fixation is achieved with an Arthrotek Bio-Core bioabsorbable interference screw and with a screw and spiked ligament washer for backup fixation. The tensioning device is applied to the ACL graft, set to 20 pounds, and the graft is tensioned with the knee in 70 degrees of flexion (B). Final ACL fixation is achieved with Arthrotek Bio-Core bioabsorbable interference screws and with spiked ligament washer or ligament fixation button backup fixation. The mechanical tensioning boot ensures consistent graft tensioning and eliminates graft advancement during interference screw insertion. It also restores the anatomic tibial step-off during PCL graft tensioning and applies a posterior drawer force during ACL graft tensioning. (Courtesy of Biomet Sports Medicine, Warsaw, Ind.)

prevents lateral cortex crowding and eliminates multiple through-and-through drill holes in the distal femur, reducing stress riser effect. It is important to be aware of the two tibial tunnel directions and to have a 1-cm bone bridge ­between the PCL and ACL tibial tunnels. This will reduce the possibility of fracture. We have found it useful to use ­primary and back-up fixation. Primary fixation is with resorbable interference screws, and back-up fixation is performed with a screw and spiked ligament washer or Arthrotek ligament fixation button. Secure fixation is critical to the success of this surgical procedure (see Fig. 23G-14).

   TABLE 23G-2  Order of Tensioning and Final Fixation Structure

Tensioning Position

Posterior cruciate ligament

70 to 90 degrees of knee flexion while restoring normal tibial step-off and neutral rotation 30 degrees of knee flexion while maintaining normal tibial step-off and neutral rotation 30 degrees of knee flexion while internal rotation and anterior translation are applied to the proximal tibia 30 degrees of knee flexion after posterior cruciate ligament, posterolateral corner, and anterior cruciate ligament tensioning and fixation have been performed

Anterior cruciate ligament Posterolateral corner

Medial collateral ligament

Results without the Arthrotek Graft Tensioning Boot

We previously published the results of our arthroscopically assisted combined ACL and PCL, and PCL and posterolateral complex, reconstructions using the reconstructive technique described in this chapter.49,62,67,68,87 Our most recently published 2- to 10-year results of combined ACL and PCL reconstructions without the Arthrotek graft tensioning boot are presented here.67 This study presented the 2- to 10-year (24- to 120-month) results of 35 arthroscopically assisted combined ACL and PCL reconstructions evaluated before and after surgery using Lysholm, Tegner, and Hospital for Special Surgery knee ligament rating scales, KT-1000 arthrometer testing, stress radiography, and physical examination. This study population included 26 males and 9 females with 19 acute and 16 chronic knee injuries. Ligament injuries included 19 ACL, PCL, and posterolateral instabilities, 9 ACL, PCL, and MCL instabilities, 6 ACL, PCL, posterolateral, and MCL instabilities, and 1 ACL and PCL instability. All knees had grade III preoperative ACL and PCL laxity and were assessed before and after surgery with arthrometer testing, three different knee ligament rating scales, stress radiography, and physical examination. Arthroscopically ­assisted combined ACL and PCL reconstructions were performed using the single-incision endoscopic ACL technique and the single femoral tunnel, single-bundle transtibial tunnel PCL technique. PCLs were reconstructed with allograft ­Achilles tendon (26 knees), autograft bone-tendon-bone Continued

1762 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d (BTB; 7 knees), and autograft semitendinosus and gracilis (2 knees). ACLs were reconstructed with autograft BTB (16 knees), allograft BTB (12 knees), Achilles tendon allograft (6 knees), and autograft semitendinosus and gracilis (1 knee). MCL injuries were treated with bracing or open reconstruction. Posterolateral instability was treated with ­biceps femoris tendon transfer, with or without ­primary repair, and posterolateral capsular shift procedures as ­indicated. No Arthrotek graft tensioning boot was used in this series of patients. Postoperative physical examination results revealed normal posterior drawer and tibial step-off in 16 of 35 (46%) knees. Normal Lachman’s test and pivot shift tests in 33 of 35 (94%) knees. Posterolateral stability was restored to normal in 6 of 25 (24%) knees, and tighter than the normal knee in 19 of 25 (76%) knees evaluated with the external rotation thigh-foot angle test. The 30-degree varus stress testing was normal in 22 of 25 (88%) knees, and grade I laxity in 3 of 25 (12%) knees. The 30-degree valgus stress testing was normal in 7 of 7 (100%) surgically treated MCL tears and was normal in 7 of 8 (87.5%) brace-treated knees. Postoperative KT-1000 arthrometer tests of mean side-to-side differences showed 2.7 mm (PCL screen), 2.6 mm (corrected posterior), and 1.0 mm (corrected anterior) measurements, a statistically significant improvement from preoperative status (P = .001). Postoperative stress radiographic side-to-side ­difference measurements at 90 degrees of knee flexion and 32 pounds of posteriorly directed proximal force were 0 to 3 mm in 11 of 21 (52.3%), 4 to 5 mm in 5 of 21 (23.8%), and 6 to 10 mm in 4 of 21 (19%) knees. Postoperative Lysholm, Tegner, and Hospital for Special Surgery knee ligament rating scale mean values were 91.2, 5.3, and 86.8, respectively, demonstrating a statistically significant improvement from preoperative status (P = .001). No Arthrotek graft tensioning boot was used in this series of patients. The conclusions drawn from the study were that combined ACL and PCL instabilities could be successfully treated with arthroscopic reconstruction and the appropriate collateral ligament surgery. Statistically significant improvement was noted from the preoperative condition at 2- to 10-year ­ follow-up using objective parameters of knee ligament rating scales, arthrometer testing, stress radiography, and physical examination. Postoperatively, these knees are not normal, but they are functionally stable. Continuing technical improvements will most likely improve future ­results. Results with the Arthrotek Graft Tensioning Boot

These new data present the 2-year follow-up results of 15 arthroscopically assisted ACL and PCL reconstructions using the Arthrotek graft tensioning boot.76 This study group consists of 11 chronic and 4 acute injuries. These

FUTURE DIRECTIONS We have now converted to performing the double-­bundle, double femoral tunnel PCL reconstruction surgical technique because there is convincing basic science data

i­njury ­ patterns included 6 ACL, PCL, and PLC injuries; 4 ACL, PCL, and MCL injuries; and 5 ACL, PCL, PLC, and MCL injuries. The Arthrotek tensioning boot was used during the procedures as in the surgical technique described previously. All knees had grade III preoperative ACL and PCL laxity and were assessed before and after surgery ­using Lysholm, Tegner, and Hospital for Special Surgery knee ligament rating scales, KT-1000 arthrometer testing, stress radiography, and physical examination. Arthroscopically assisted combined ACL and PCL ­reconstructions were performed using the single-incision endoscopic ACL technique, and the single femoral tunnel, single-bundle transtibial tunnel PCL technique. PCLs were reconstructed with allograft Achilles tendon in all 15 knees. ACLs were reconstructed with Achilles tendon allograft in all 15 knees. MCL injuries were treated ­ surgically ­ using ­primary repair, posteromedial capsular shift, and allograft augmentation as indicated. Posterolateral instability was treated with allograft semitendinosus free graft, with or without primary repair, and posterolateral capsular shift procedures as indicated. The Arthrotek graft tensioning boot was used in this series of patients. Postreconstruction physical examination results revealed normal posterior drawer and tibial-step off in 13 of 15 (86.6%) knees. Normal Lachman’s test in 13 of 15 (86.6%) knees and normal pivot shift tests in 14 of 15 (93.3%) knees. Posterolateral stability was restored to normal in all knees with posterolateral instability when evaluated with the ­external rotation thigh-foot angle test (9 knees equal to the normal knee, and 2 knees tighter than the normal knee). Thirty-degree varus stress testing was restored to normal in all 11 knees with posterolateral lateral instability. Thirty- and zero-degree valgus stress testing was restored to normal in all 9 knees with medial side laxity. Postoperative KT-1000 arthrometer testing mean side-to-side difference measurements were 1.6 mm (range, 3 to 7 mm) for the PCL screen, 1.6 mm (range, 4.5 to 9 mm) for the corrected posterior, and 0.5 mm (range, 2.5 to 6 mm) for the corrected anterior measurements, a significant improvement from preoperative status. Postoperative stress radiographic side-to-side difference measurements measured at 90 degrees of knee flexion and 32 pounds of posteriorly directed proximal force using the Telos stress radiography device were 0 to 3 mm in 10 of 15 knees (66.7%), 4 mm in 4 of 15 knees (26.7%), and 7 mm in 1 of 15 knees (6.67%). Postoperative Lysholm, Tegner, and Hospital of Special Surgery knee ligament rating scale mean values were 86.7 (range, 69 to 95), 4.5 (range, 2 to 7), and 85.3 (range, 65 to 93), respectively, demonstrating a ­significant improvement from preoperative status. The study group demonstrates the efficacy and success of using a mechanical graft-tensioning device (Arthrotek graft tensioning boot) in single-bundle, single femoral tunnel ­arthroscopic PCL reconstruction.

supporting the efficacy of this procedure.73 This doublebundle, double femoral tunnel technique more closely approximates the anatomic insertion site of the native PCL and should theoretically provide improved results (Fig. 23G-18). Our clinical results with 15- to 36-month

knee 1763

Figure 23G-18   Intraoperative photograph demonstrating placement of an Achilles tendon allograft reconstructing the anterolateral bundle of the posterior cruciate ligament (PCL) and a tibialis anterior allograft reconstructing the posteromedial bundle of the PCL. Note how the double-bundle double femoral tunnel PCL reconstruction more closely fills the femoral anatomic insertion site of the posterior cruciate ligament. ACL reconstruction was performed with an Achilles tendon allograft. (Courtesy of Gregory C. Fanelli, MD.)

follow-up of 66 consecutive PCL-based multiple ligament knee reconstructions, however, showed no significant difference between double- and single-bundle PCL reconstruction results evaluated with Telos stress radiography, KT-1000 arthrometer testing, and Tegner, Lysholm, and Hospital for Special Surgery knee ligament rating scales. Both the single- and double-bundle groups had a mean of less than 3 mm side-to-side difference on stress radiography and arthrometer testing. This indicated restoration of normal side-to-side difference measurements for the PCL reconstructions using both the single-bundle and doublebundle surgical techniques. Another area of interest is the incorporation of the Musculoskeletal Transplant Foundation (Edison, NJ) Cascade System autologous platelet rich fibrin matrix into the grafts used in the cruciate and collateral ligament reconstructive procedures (Fig. 23G-19). There are ­several studies indicating favorable effects on the ligament graft tissue and the clinical results.88-90 We have demonstrated favorable initial clinical results with respect to graft incorporation, wound healing, and early stability; however, there is no long-term follow-up as of this writing.

POSTOPERATIVE PRESCRIPTION The postoperative protocol (Box 23G-2) following multiple ligament reconstruction is typically divided into four specific phases. Phase I consists of the first 6 postoperative weeks. The emphasis during phase I is on graft protection. Immediately postoperatively, patients are placed in a long-leg hinged knee brace locked in full extension. Cryotherapy is used to minimize postoperative swelling. The patient is mobilized with physical therapy, with non– weight-­bearing instructions applied to the operative limb. Following discharge from the hospital, outpatient physical therapy is instituted, with emphasis placed on quadriceps strengthening through isometrics and low-intensity

Figure 23G-19   A platelet-rich fibrin matrix clot created by the Musculoskeletal Transplant Foundation Cascade System is incorporated into the lateral posterolateral reconstruction to potentially enhance graft incorporation and wound healing. (Courtesy of Gregory C. Fanelli, MD.)

electric ­ stimulation. During weeks 4 through 6, flexion exercises are allowed from 0 to 90 degrees. Patellar mobilization, gentle hamstring and gastrocnemius stretching, and “foot pump” exercises are also used. Phase II begins during the seventh postoperative week. The brace is unlocked to allow full flexion and may be removed at night. Weight-bearing is initiated at this time as well and is increased in a progressive fashion until the patient is bearing full weight by postoperative week 10. Active and passive range-of-motion exercises are initiated, albeit guardedly to avoid excessive posterior shear forces on the PCL. Once the patient is fully weight-bearing, closed kinetic chain exercises are instituted. These exercises are restricted to a range from 60 degrees to full extension to contain tibiofemoral shearing and patellofemoral compressive forces. Balance and proprioception training activities are also initiated as part of phase II rehabilitation. Open chain flexion and extension exercises are avoided during this time. Between the 10th and 12th postoperative weeks, the hinged knee brace is discontinued. A functional brace may be used at this time to assist with varus and valgus stability. Knee flexion should approach 120 degrees during the third phase of rehabilitation. A lack of the last 10 to 15 degrees of knee flexion is not uncommon but rarely results in any functional limitation. If active flexion of 90 degrees has not been reached by postoperative week 12, manipulation under anesthesia with or without arthroscopic débridement may be considered. During the fourth through sixth postoperative months, closed chain strengthening exercises are advanced. Aerobic conditioning exercises are also initiated at this time. As rehabilitation progresses, open chain exercises and straight-line jogging can commence. At the end of the sixth postoperative month, isokinetic assessments are obtained. Quadriceps and hamstrings deficits of 20% or less are desired before the patient is allowed to begin sport-specific activities. Phase IV of the multiple ligament–injured knee rehabilitation protocol occurs during the 7th through 12th

1764 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 23G-2 Multiple Ligament Reconstruction Rehabilitation Phase I: 0 to 6 weeks Goals

• Graft protection • Maintain patellar mobility • Maintain quadriceps tone • Maintain full passive extension • Control pain and swelling Program

• Non–weight-bearing ambulation with crutches • Brace locked in full extension • Cryotherapy • Quadriceps sets • Patellar mobilization • Ankle pumps and stretching exercises • Active range of motion from 0 to 90 degrees from weeks 4 to 6

Phase II: 6 to 12 weeks Goals

• Initiate weight-bearing • Increase knee flexion • Maintain quadriceps tone • Improve proprioception • Avoid isolated quadriceps and hamstring contraction Program • Progress to full weight-bearing over the next 4 weeks • Open brace to full flexion • Prone hangs • Passive flexion exercises • Closed chain strengthening

postoperative months. Return to sports or heavy labor is predicated on the absence of swelling with minimal to no pain. Strength and functional tests should be within 90% of the contralateral uninjured side. Patients return to clinic 1 year after the initial postoperative date. A complete physical examination is performed along with stress radiographs, arthrometric testing, and completion of ligament rating forms. The functional brace is continued for sports or rigorous activities for an additional 6 months. Patients continue to return on a yearly basis so that long-term effectiveness and patient satisfaction can be assessed.

POTENTIAL COMPLICATIONS Complications involving multiligamentous knee injuries are seen rather frequently, and great care must be employed to minimize their incidence and significance. Accurate diagnosis, appropriate surgical timing, and structured rehabilitation are all essential to avoid complications. In the initial management of these injuries, a detailed neurovascular examination is paramount. Failure to recognize injury to the popliteal artery or vein can lead to loss of limb. In cases in which the vascular status of the leg is in question, further diagnostic tests should be obtained.

• Stationary bicycle • Hip strengthening and continued proprioceptive ­retraining

• Discontinue hinged brace by week 12 Phase III: 4 to 6 months Goals

• Maximize flexion and maintain full extension • Improve quadriceps and hamstring strength • Improve functional skills • Improve cardiovascular endurance Program

• Progressive closed chain strengthening • Single leg proprioception exercises • Isolated quadriceps and hamstring exercises with ­increasing resistance

• Straight-line jogging followed by sprinting • Fit for anterior/posterior cruciate ligament brace • Begin agility drills in brace • Sport-specific drills • Isokinetic testing at end of phase III Phase IV: 7 to 12 months Program

• Assess functional strength via single-leg hop • Return to sports when the following criteria are met

o



o



o



o

 inimal to no pain M Isokinetic and functional tests within 10% of the uninvolved side Successful completion of sport specific drills Anterior/posterior cruciate ligament functional brace

Delay in revascularization must be avoided. Focusing on the potential limb-threatening vascular injuries remains the first step in avoiding serious complications of these injuries. Failure to recognize the severity of the injury complex can also lead to an inadequate treatment plan and worse outcomes. MRI should be used early to supplement physical examination findings in order to identify all compromised ligamentous structures. Specific perioperative complications following multiple ligament knee reconstructions include iatrogenic ­neurovascular injury, compartment syndrome, fluid extravasation, and wound problems. In patients with vascular injury, vascular repair needs to be among the first treatments rendered. During definitive ligament reconstruction, the vascular repair must be protected. Excessive manipulation of the knee may place a vascular anastomosis under tension. It is ideal to have the vascular surgeon that performed the aforementioned revascularization available, in case the vascular repair fails. The highest potential for iatrogenic vascular injury exits during posterior cruciate reconstruction. The posteromedial safety incision is described earlier and is extremely useful in protecting the popliteal artery and tibial nerve during PCL surgery.62 Passage of the guide pin for the

knee 1765

t­ ibial tunnel should be done under direct visualization using the arthroscope or fluoroscopic imaging. Hand drilling of the posterior cortex of the tibial tunnel is also effective in reducing the risk for penetration into the neurovascular structures. The common peroneal and tibial nerves are also at significant risk during these multiligamentous reconstructions. In many cases, some degree of nerve injury exists preoperatively. Great care must be employed to avoid further trauma to the nervous structures. Lateral-sided injuries are often associated with injury to the common peroneal nerve. During open reconstruction of the PLC, the nerve must be identified and protected at all times. It is at significant risk when drilling tunnels in and around the fibular head and when using the biceps tendon as part of the ligamentous reconstruction. The tibial nerve is most at risk for injury during PCL tibial tunnel preparation. It should be protected in a similar manner as that of the popliteal artery. Compartment syndrome of the leg can be a potentially disastrous complication of knee dislocations. It can occur early after the initial injury, or it may manifest in a delayed fashion, especially in cases of revascularization following a period of lower leg ischemia. Additionally, fluid extravasation during arthroscopy can create an iatrogenic compartment syndrome if capsular structures have not healed sufficiently to allow an arthroscopic approach.91 One must be prepared to convert to open central pivot reconstructions in these specific cases. High pump pressures should be avoided. Postoperative infection and wound dehiscence are also potential complications in these cases. Multiple incisions are usually required to perform these procedures. Adequate skin bridges must be maintained to avoid skin necrosis. Open injuries should be thoroughly débrided and irrigated in an operative setting. Undue tension on surgical incisions creates potential for delayed wound healing. Delaying surgery for a short time to allow improvement in the quality of the soft tissue envelope before definitive surgery is ideal.92 Stiffness and recurrent laxity and instability are the two primary long-term complications patients experience following operative treatment of the dislocated knee. A fine balance exits between creating a stiff, stable knee and a supple knee with functional laxity. A review of the literature indicates that stiffness is the primary complication found in patients treated with operative ligament repair and reconstruction.15,34,57,59 However, the few series that have compared operative and nonoperative treatment have recommended surgical treatment despite this known complication.20,51,60,64,93 To minimize postoperative stiffness, some authors have suggested a short delay between the initial injury and definitive ligamentous reconstruction.55,61 Three weeks has been suggested as an optimal time to allow the early post-traumatic inflammatory process to subside before adding a second postsurgical insult to the knee. Benefits of this approach are that capsular injuries are allowed to begin to heal which may allow for arthroscopic intervention. The negative aspect is that primary repair of the PLC structures is more optimal at an earlier time frame. Delay creates secondary scarring, which can force one to proceed with PLC reconstruction.

Recurrent or persistent ligamentous laxity is seen much less frequently in treating the multiple ligament–injured knee because early, aggressive surgical treatment has become more commonplace. However, residual posterior sag of the tibia is not uncommon. Failure to address collateral or corner injuries often leads to functional instability and potential failure of the central pivot reconstructions. Prolonged delay to surgery has been shown to have worse long-term outcomes secondary to increased laxity as well as an increased number of articular cartilage and meniscal injuries. Avoiding complications in the treatment of the multiple ligament–injured knee depends on early recognition and treatment of all potential neurovascular complications. Complete diagnosis of all injured structures needs to be followed by careful preoperative planning and surgical timing. Neurovascular structures must be protected at the time of ligamentous reconstruction. A detailed rehabilitation program must be employed and followed. Stiffness and arthrofibrosis remain a complex issue that should be addressed promptly for optimal results.

CRITERIA FOR RETURN TO PLAY Return to athletic activities or heavy labor is predicated on multiple factors. It is considered at about the seventh postoperative month. Swelling should be absent, and pain should be at a minimum. Isokinetic strength and functional testing should be within 90% of the contralateral, uninjured side. Patients are fitted with a functional combined instability brace. The brace is recommended for strenuous activity for an additional 12 months. It is important to counsel patients that a return to their previous level of function and activity may not be attainable following these potentially devastating injuries.

CONCLUSIONS AND SUMMARY Multiple ligament injuries of the knee are complex injuries requiring a systematic approach to evaluation and treatment. Gentle reduction, along with documentation and treatment of vascular injuries, is a primary concern in the acute dislocated and multiple ligament–injured knee. Arthroscopically assisted combined ACL and PCL reconstruction with appropriate collateral ligament surgery is a reproducible procedure. Knee stability is improved postoperatively when evaluated with knee ligament rating scales, arthrometer testing, and stress radiographic analysis. Acute MCL tears, when combined with ACL and PCL tears, may in certain cases be treated with bracing. PLC injuries, combined with ACL and PCL tears, are best treated with primary repair as indicated, combined with reconstruction using a post of strong autograft or allograft tissue. Surgical timing depends on the ligaments injured, the vascular status of the extremity, reduction stability, additional injuries, and the overall health of the patient. We prefer the use of allograft tissue for reconstruction in these cases because of the strength of these large grafts and the absence of donor site morbidity. Our most recent study group demonstrates the efficacy and success of using a mechanical graft-tensioning device (Arthrotek graft tensioning boot) in single-bundle, single femoral tunnel arthroscopic PCL reconstruction.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1766 DeLee & Drez’s� O

C

r i t i c a l

P

o i n t s

l

 nee dislocations result from violent trauma and are K orthopaedic emergencies. l Vascular injury is common (32%) and must be diagnosed and treated promptly. l Absolute surgical indications include vascular injury, open injuries, and irreducible dislocations. l Surgical timing is dictated by the ligaments injured, the vascular status of the extremity, reduction stability, additional injuries, and the overall health of the patient. l Surgical delay of 2 to 3 weeks helps minimize the risk for postoperative stiffness. l Better surgical results are achieved with anatomic repair and reconstruction of all compromised structures. l Fluid extravasation may preclude an arthroscopic approach to ligament reconstruction. l Specialized instruments and the posteromedial safety incision can help minimize the risk for iatrogenic trauma during PCL reconstruction. l A dedicated rehabilitation program is essential to maximize function and stability. l Stiffness is the most common complication, and manipulation under anesthesia may be needed to maximize results.

S U G G E S T E D

R E A D I N G S

Cole BJ, Harner CD: The multiple ligament injured knee. Clin Sports Med 18: 241-262, 1999. Fanelli GC (ed): Posterior Cruciate Ligament Injuries: A Practical Guide to Management. New York, Springer-Verlag, 2001. Fanelli GC (ed): The Multiple Ligament Injured Knee: A Practical Guide to Management. New York, Springer-Verlag, 2004. Fanelli GC: Rationale and surgical technique for PCL and multiple knee ligament reconstruction. Arthrotek, Inc. Surgical Technique Manual No. Y-BMT979/071506/K, 2006. Fanelli GC, Edson CJ: Arthroscopically assisted combined ACL/PCL reconstruction: 2-10 Year follow-up. Arthroscopy 18:703-714, 2002. Fanelli GC, Edson CJ, Orcutt DR, et al: Treatment of combined anterior ­cruciateposterior cruciate ligament-medial-lateral side knee injuries. J Knee Surg 18: 240-248, 2005. Good L, Johnson RJ: The dislocated knee. J Am Acad Orthop Surg 3:284-292, 1995. Green A, Allen BL: Vascular injuries associated with dislocation of the knee. J Bone Joint Surg [Am] 59:236-239, 1977. Shelbourne KD, Wilckens JH, Mollabashy A, et al: Arthrofibrosis in acute ­anterior cruciate ligament reconstruction: The effect of timing of reconstruction and ­rehabilitation. Am J Sports Med 19:332-336, 1991. Sisto DJ, Warren RF: Complete knee dislocation: A follow-up study of operative treatment. Clin Orthop 198:94-101, 1985.

R efere n ces Please see www.expertconsult.com

S e c t i o n

H

Osteochondritis Dissecans C. Thomas Vangsness Jr. The etiology of osteochondritis dissecans (OCD) remains unclear, and all treatment algorithms are anecdotal in nature without evidence-based medicine.1,2 OCD was originally thought to be an inflammatory problem and initially was named osteochondritis. The disease was formally termed OCD by Koenig3,4 to explain the late problems of loose bodies in the joint. Although not necessarily an inflammatory process, the term OCD has persisted in the literature.5

CLASSIFICATION A uniformly accepted classification system for the anatomic changes seen with OCD has not been established. OCD has been classified by bone maturity: 1. Juvenile lesions have open distal femoral physes. 2. Adolescents have closing distal femoral physes. 3. Adults have fully closed distal femoral physes.

Historically, the classification of OCD has been made from plain radiographs to define the lesion size and location. Most radiologic classification systems attempt to determine the presence and stability of any loose fragments. Whether we use conventional tomography, computed tomography (CT) with or without arthrography, standard arthrography, scintigraphy, or magnetic resonance imaging (MRI) with or without dye, none of these is ideal. Berndt and Harty6 originally described four stages of lesions on plain radiographs of talar OCD, which has since been applied to the knee. Stage I shows a small area of compression of the subchondral bone. Stage II is a partially detached osteochondral fragment. Stage III is a completely detached fragment that remains in the crater of origin, and stage IV shows complete detachment and a loose body. Cahill and Berg7 divided the knee radiograph into 15 distinct zones (Fig. 23H-1). The zones are numbered 1 to 5 from medial to lateral and essentially divided by the notch of the knee. Each compartment is divided in half. In

Knee 1767

   TABLE 23H-2  Magnetic Resonance Imaging

Classification of Knee Osteochondritis Dissecans

5

4

3

2

Blumensaat’s line

1 A B

C

Stage

Description

1 2

Small change of signal without clear margins of fragment Osteochondral fragment with clear margins but without fluid between fragment and underlying bone Fluid is visible partially between fragment and underlying bone. Fluid is completely surrounding the fragment, but the fragment is still in situ. Fragment is completely detached and displaced (loose body).

3 4 5

B

A

Figure 23H-1    Alphanumeric zone classification of knee osteochondritis dissecans lesions. Anteroposterior (A) and lateral (B) views of the knee demonstrate the 15 regions. The five numbered zones on the anteroposterior view are divided centrally by the notch (zone 3). The lettered zones on the lateral view are divided by Blumensaat’s line anteriorly and the posterior cortical line. The half-moon–shaped shaded area in each view of the distal femur represents an old lesion. (Redrawn from Cahill BR, Berg BC: 99m-Technetium phosphate compound joint scintigraphy in the management of juvenile osteochondritis dissecans of the femoral condyles. Am J Sports Med 11:329-335, 1983.)

the lateral radiograph, Blumensaat’s line divides the knee into areas A, B, and C. This classification system applies to adult knees. Milgram8 has documented the radiographic characteristics of OCD around the femoral condyles. Fragmentation and collapse, along with increased separation, was noted with a sclerotic base around the femoral defect and a radiolucent crescentic zone. Barrie9 outlined the components of fragmentation in the pathogenesis of OCD. The classification system for juvenile OCD is based on technetium-99m phosphate scintigraphy findings (Table 23H-1). The scan’s activity is related to the plain radiograph.7,10 Zero is normal, and stage I demonstrates a small change on radiography, but no increased activity on bone scan. Stage II shows an increased uptake in the lesion, but not in the adjacent femoral condyle. Stage III shows an increased uptake in both the lesion and the adjacent femoral condyle. Stage IV demonstrates involvement in both the femoral lesion and the adjacent tibial surface. Stages III and IV are thought to be more systemic OCD    TABLE 23H-1  Bone Scan Classification of Juvenile Osteochondritis Dissecans Stage

Description

0 I

Normal radiographic and scintigraphic appearance The lesion is visible on plain radiographs, but bone scans reveal normal findings. The scan reveals increased uptake in the area of the lesion. In addition, there is increased isotope uptake in the entire femoral condyle. In addition, there is uptake in the tibial plateau opposite the lesion.

II III IV

lesions. Paletta and colleagues11-13 noted that patients with open physeal plates and increased activity on bone scan had improved healing compared with older patients with closed growth plates. MRI has been used to develop a better understanding of these lesions (Table 23H-2). MRI techniques have limitations in detecting articular cartilage changes in OCD. MRIs have been classified with the lesions containing fluid behind them as partially detached.14 High signal intensity on T2-weighted images can help distinguish a break in the articular cartilage surface. Gadolinium as a contrast material has also been used to predict the stability of these OCD lesions.14 Future pulsing sequences with and without contrast will help delineate and explain the OCD process.15 Arthroscopy has improved our understanding of the diagnosis and management of OCD. Chondral lesions that have an intact articular cartilage are called closed lesions. Open lesions can be partial or complete depending on the involvement of the articular cartilage. Cartilage stability can be partial or complete depending on the probing at the time of surgery. Mesgarzadeh and associates16 used conventional radiography, bone scintigraphy, and MRI to assess the mechanical stability of OCD in the femoral condyles. Arthroscopy examination showed lesions to be grossly loose when lesions were large (>0.8 cm3) or associated with broad sclerotic margins (especially >3 mm thick) on routine radiographs. Significant accumulation of bone-seeking radionuclides during the flow, blood-pool, and late phase of radionuclide examination generally indicated a loose fragment. MRI showing fluid at the interface of the fragment and the bone also correlated with loose fragments. The density of the fragment was not a useful indicator of stable or unstable lesions.

EPIDEMIOLOGY The peak prevalence of juvenile OCD is noted during the preteen years, and OCD is thought to be rare among children younger than 10 years. The adolescent OCD is estimated to be between 0.2% and 0.3% based on knee radiographs and 1.2% based on knee arthroscopy studies.6,17 The highest risk occurs in patients aged 10 to 15 years. Male-to-female ratios have been quoted in the literature as between 2:1 and 5:3. Twenty to 25% of the cases are bilateral.18,19 OCD usually involves the lateral portion of the medial condyle in three fourths of the cases (Fig 23H-2).20,21 Posterior portions of the femoral

�rthopaedic ����������� S �ports ������ � Medicine ������� 1768 DeLee & Drez’s� O

A

B

C

D

Figure 23H-3   Osteochondritis dissecans of the patella.

E Figure 23H-2   Location of osteochondritis dissecans of the femoral condyles. Medial condyle: classic, 69% (A); extended classic, 6% (B); inferocentral, 10% (C). Lateral condyle: inferocentral, 13% (D); anterior, 2% (E). (Redrawn from Aichroth P: Osteochondritis dissecans of the knee. J Bone Joint Surg Br 53:440-447, 1971.)

c­ ondyles and the tibial plateau can be affected. Patellar lesions can occur 5% to 10% of the time and are usually located in the inferior medial area (Fig. 23H-3).22-24 The femoral trochlea region for OCD is rare (Fig. 23H-4).25

ETIOLOGY The literature defines three major areas of etiology: constitutional-hereditary, vascular, and traumatic. Repetitive or persistent microtrauma to a vulnerable area appears to be the major etiologic source.1,26 Many predispositions to OCD lesions have been suggested in the literature.27,28 OCD has been found in a variety of inherited conditions including dwarfism, tibia vara, Legg-Calvé-Perthes disease, and Stickler’s syndrome.29 OCD has been suggested to represent a variation of epiphyseal dysplasia.27 Ultimately, genetic inheritance will most likely be proved over time to be multifactorial in nature. Authors have looked at the pathophysiology between osteonecrosis and OCD.19 An analogy has been made to a sequestrum. Studies have shown detached OCD lesions with no histopathophysiologic evidence of osteonecrosis.30 A limited uptake of tetracycline and radionuclide has been noted in OCD lesions.31 Traumatic injury has been reported in up to 40% of patients with OCD providing evidence for the traumatic origin of OCD lesions.32 It is also been thought that the tibial spine eminence may create shearing effects to the lateral aspect in the medial knee condyle, creating increased contact forces.33,34

NATURAL HISTORY Outcomes in the literature have looked at patient maturity as the major predictive determinant of outcome. Younger patients have a high healing potential, although adolescent improvement is still unpredictable.19,33,35,36 About 50% of these lesions tend to heal. In the mature group, the rate of healing is decreased, depending on the location and size of the lesion. Degenerative joint disease can become the long-term final outcome of OCD.37 Specific outcomes depend on the origin of the OCD and the size, location, and specific status of the articular cartilage lesion.38 It must be emphasized that no randomized clinical trials exist to elucidate the natural history of this disease process. A large multicenter review of the European Pediatric Orthopaedic Society (500 knees, 318 juveniles, 191 adults)29 provided several important conclusions: A stable fragment had a better prognosis, and pain and ­ swelling were not

Figure 23H-4   Osteochondritis dissecans of the femur.

Knee 1769 Juvenile (open physis)

Radiographs

Adult (closed physis)

Stable

Stable

Physical examination

Physical examination

Stable

Stable

MRI

Unstable

MRI

Stable

Stable

Not positive

Bone scan

Malalignment

Positive

No

Activity restriction (3 mo)

Yes

Osteotomy

• Impending physeal closure • Clinical signs of instability • Expanding lesion on plain films

Arthroscopy

Stable

Unstable reducible

Unstable with fragmentation or osteolysis

Transchondral drilling

Fixation graft

Fixation and graft

Treat symptomatically

Unstable and chondral damage

Fixation and graft, chondrocyte transplant, or osteochondral graft

Figure 23H-5   Treatment algorithm for osteochondritis dissecans of the knee.

good indicators of loosening. Radiography and CT were not useful in predicting loosening of the fragment. Sclerosis on plain films gave a poor response to drilling treatments. A lesion larger than 2 cm in diameter gave a worse prognosis, and when there was loosening of the fragment, surgical results were better than nonsurgical results. Classically defined anatomic lesions had a better prognosis. An adult onset of symptoms had more abnormal findings after the treatment period, and more than one in five of those with open physeal plates had abnormal knee radiographs an average of 3 years after treatment.

CLINICAL TREATMENT Crawford and Safran39 presented a treatment algorithm for knee OCD (Fig. 23H-5). Early clinical diagnosis and treatment are critical to optimize patient outcomes. The literature recommends plain radiographs of the knee, including anteroposterior, lateral, notch, and skyline radiographs followed by an MRI. Sometimes an MRI with contrast will further clarify the classification of the lesion. Bone scans can also provide bone activity information.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1770 DeLee & Drez’s� O

Nonoperative Management In juvenile and adolescent lesions with open physeal plates, the natural history of stable OCD lesions is generally favorable, and nonoperative treatment is appropriate.40,41 Activity levels must be decreased with consideration for immobilization. The length of this immobilization and duration of limited activity are uncertain. Continuous passive motion for the common OCD can be considered. The side effects of prolonged immobilization must be considered, including ongoing joint stiffness, muscle atrophy, and potential ­cartilage degeneration. Partial weight-bearing issues with this conservative protocol have not been established. Repeat MRIs with intra-articular contrast can be considered, although the best timing of these scans has not been established. Older patients with an OCD lesion have a less certain course, and mechanical symptoms, fragment loosening, and detachment should be monitored with plain radiographs or sequential MRI examinations with or without contrast.

Operative Management Cahill42 reported a 50% success rate of nonoperative management of juvenile and adolescent OCD. Juvenile OCD lesions can undergo an arthroscopic examination and drilling.13 Epiphyseal drilling with or without articular cartilage penetration aims to promote vascular channels for potential healing.2 Retrograde drilling is also an acceptable technique, although technically more difficult. Adult OCD must be observed for potential loose bodies. If adult OCD lesions are unstable or there are loose bodies, surgical intervention is appropriate.43 The maintenance of the articular cartilage surface is important and may necessitate repair of the unstable fragments or the osteochondral defect. Inadequate healing is associated with nonclassic OCD lesions, multiple lesions, or underlying medical ­ conditions such as smoking. Débridement of the fibrous tissue and drilling of the sclerotic subchondral base of the OCD lesion is recommended.44 An autologous bone graft can be placed in the crater before reduction and fixation. Screws have also been used successfully to fix larger fragments.4,28,45-47 The recent use of bioabsorbable screws and pins helps avoid a second surgery to remove hardware.48 Loosening and failure of bioabsorbable screws have been reported in the literature.49,50 If the OCD lesion is too small, has fragmentation, or has inadequate bony backing (less than 2 mm of bone), the surgeon may be forced to simply remove the OCD piece.51,52 To promote filling of the defect, drilling, abrasion arthroplasty, and microfracture methods can be used to recruit pluripotential cells from the marrow elements.53,54 Autologous osteochondral plugs can be obtained from neighboring regions of the knee to fill the defect in the skeletally mature patient.55,56 Autologous chondrocyte implantation (ACI) has been used to treat larger femoral defects. In the skeletally mature patient, a bone graft can be placed into the base of the crater followed by the ACI procedure.13,57,58 Staged and planned defect reconstruction with fresh osteoarticular allografts is difficult in terms of the logistics of obtaining and processing this fresh graft.26,59-61 Good results have been published, although long-term results are not available.43 Regardless of the chosen surgical technique

applied to knee OCD lesions, long-term evidence-based literature does not exist.

CONCLUSION The etiology and natural history of OCD continues to be uncertain. The effect of skeletal maturity has been emphasized. With newer diagnostic techniques and newer MRI pulsing sequences, a better understanding of the biology of these lesions over time will be learned. The long-term problems are pain and arthritis in the knee, and patient education needs to be emphasized, although it difficult because of the relative infrequency with which OCD occurs. Longterm randomized clinical trials need to be done. C

r i t i c a l

P

o i n t s

l OCD is a disease originating in the subchondral bone with late secondary problems involving the articular cartilage surface. l Clinical manifestations are variable and relate to the site of involvement and stage of the disease. l Genetics probably play a role, and knee trauma is a common cause. l Classifications are made from radiographs and MRIs. l Treatment protocols depend on the age, location of disease, and stage of the disease process. l There are no published randomized clinical trials to provide good evidence-based medicine to specifically delineate treatment regimens. l A uniformly accepted classification system has not been established. Classification of OCD has been made from plain radiographs to define the lesion size and location. l The peak prevalence of juvenile OCD is noted during the preteen years, with the highest risk appearing between patients aged 10 to 15 years. Juvenile OCD lesions can undergo an arthroscopic examination and drilling. l Older patients with an OCD lesion have a less certain course, and mechanical symptoms, fragment loosening, and detachment should be monitored with plain radiographs or sequential MRI examinations with or without contrast. l If adult OCD lesions are unstable or there are loose bodies, surgical intervention is appropriate. Maintenance of the articular cartilage surface is important.

S U G G E S T E D

R E A D I N G S

Aichroth P: Osteochondritis dissecans of the knee: A clinical survey. J Bone Joint Surg Br 53:440-447, 1971. Bentley G, Biant LC, Carrington RW, et al: A prospective, randomised comparison of autologous chondrocyte implantation versus mosaicplasty for osteochondral defects in the knee. J Bone Joint Surg Br 85:223-230, 2003. Cepero S, Ullot R, Sastre S: Osteochondritis of the femoral condyles in children and adolescents: Our experience over the last 28 years. J Pediatr Orthop B 14(1): 24-29, 2005. Flynn JM, Kocher MS, Ganley TJ: Osteochondritis dissecans of the knee. J Pediatr Orthop 24(4):434-443, 2004. Garrett JC: Osteochondritis dissecans. Clin Sports Med 10:569-593, 1991. Hefti F, Beguiristain J, Krauspe R: Osteochondritis dissecans: A multicenter study of the European Pediatric Orthopedic Society. J Pediatric Orthop B 8:231-245, 1999. Hui JH, Chen F, Thambyah A, Lee EH: Treatment of chondral lesions in advanced osteochondritis dissecans: A comparative study of the efficacy of chondrocytes, mesenchymal stem cells, periosteal graft, and mosaicplasty (osteochondral autograft) in animal models. J Pediatr Orthop 24(4):427-433, 2004.

Knee 1771 Kocher MS, Tucker R, Ganley TJ, Flynn JM: Management of osteochondritis dissecans of the knee: Current concepts review. Am J Sports Med 34:1181-1191, 2006. Makino A, Muscolo DL, Puigdevall M, et al: Arthroscopic fixation of osteochondritis dissecans of the knee: Clinical, magnetic resonance imaging, and arthroscopic follow-up. Am J Sports Med 33(10):1499-1504, 2005. Murray JR, Chitnavis J, Dixon P, et al: Osteochondritis dissecans of the knee: Longterm clinical outcome following arthroscopic debridement. Knee 14:94-98, 2007. Ronga M, Zappala G, Cherubino M, et al: Osteochondritis dissecans of the entire femoral trochlea. Am J Sports Med 34(9):1508-1511, 2006.

Twyman RS, Desai K, Aichroth PM: Osteochondritis dissecans of the knee: A longterm study. J Bone Joint Surg Br 73:461-464, 1991. Uematsu K, Habata T, Hasegawa Y, et al: Osteochondritis dissecans of the knee: Long-term results of excision of the osteochondral fragment. Knee 12(3): 205-208, 2005.

R efere n ces Please see www.expertconsult.com

S e c t i o n

I

Articular Cartilage Lesion Andreas H. Gomoll, Adam Yanke, and Brian J. Cole Damage to the articular cartilage comprises a spectrum of disease entities ranging from single, focal chondral defects to more advanced osteoarthritis disease, the latter of which is not discussed in this chapter. Left untreated, articular cartilage lesions have no spontaneous repair potential. Therefore, various techniques have evolved to stimulate defect repair or overtly replace these defects. Conventional techniques, such as abrasion arthroplasty, drilling, or microfracture, attempt to fill the defect with a fibrocartilaginous scar produced by marrow-derived pluripotent stem cells. This tissue, however, is of lesser biologic and mechanical quality than hyaline cartilage. More recently developed techniques, such as autologous chondrocyte implantation (ACI) and matrix autologous ­chondrocyte implantation (MACI), achieve a tissue that more closely resembles the original hyaline cartilage but are expensive and involve sophisticated and prolonged rehabilitation. This chapter provides a concise overview of current techniques for cartilage repair, presents new developments in this evolving area, and subsequently discusses the authors’ preferred techniques in more detail.

RELEVANT ANATOMY AND BIOMECHANICS Partial-thickness chondral lesions do not penetrate the subchondral bone and are therefore avascular, do not heal, and may enlarge over time. Full-thickness defects, especially with injury to the underlying vascular bone, have the potential to fill with a fibrocartilaginous scar formed by cells invading from the marrow cavity. The resulting tissue, however, is predominantly composed of type I collagen, resulting in inferior mechanical properties compared with the type II collagen-rich hyaline cartilage. Long implicated in the subsequent development of osteoarthritis, focal chondral defects result from various causes. Patients are about evenly split in reporting a ­traumatic versus an insidious onset of symptoms; athletic

activities are the most common inciting event associated with the diagnosis of a chondral lesions.1 Traumatic events and developmental causes such as osteochondritis dissecans (OCD) predominate in younger age groups. For example, traumatic hemarthroses in young athletes with knee injuries are associated with chondral defects in up to 10% of cases2; patellar dislocation is strongly associated with damage to the articular surface, with chondral defects of the patella see in up to 95% of patients3; the incidence of OCD is estimated at 30 to 60 cases per 100,000 people.4 Several large studies have found high-grade chondral lesions (Outerbridge grades III and IV) in 5% to 11% of younger patients (<40 years) and up to 60% in older age groups.1,5,6 The most common locations for these defects are the medial femoral condyle (up to 32%) and the patella,5,6 and most are detected incidentally during meniscectomy or anterior cruciate ligament (ACL) reconstruction.1,7 Notably, despite this relatively high incidence, many of these defects are incidental in nature and asymptomatic. On careful evaluation, a large percentage of chondral defects are associated with structural abnormalities, such as malalignment, patellar instability, and insufficiency of the ligamentous and meniscal structures. The disappointing early results of cartilage repair have been explained by the failure to diagnose and correct these associated bony and ligamentous abnormalities; for example, in early studies of patellar defects treated with ACI alone, good and excellent results were found in only one third of patients.8 Later studies, however, identified patellar maltracking as an important associated abnormality, and performance of a corrective osteotomy concurrently with cartilage repair led to 71% good or excellent results.9 These reports emphasize the importance of a thorough patient evaluation to correctly identify and treat all associated abnormalities to ensure the long-term success of chondral repair. Varus or valgus malalignment of the lower extremity results in compartment overload and is associated with degeneration of the articular surface. Coventry’s early work

�rthopaedic ����������� S �ports ������ � Medicine ������� 1772 DeLee & Drez’s� O

   TABLE 23I-1  Classification Systems for Cartilage Defects Description Grade of Outerbridge   Lesion Classification

ICRS Classification   (with Subclassifications)

0 1

Normal cartilage A. Softening or fibrillations B. Superficial fissuring Less than half the cartilage depth

2

3

4

Normal cartilage Cartilage with softening and swelling Partial-thickness defect with fissures on the surface that do not reach subchondral bone or exceed 1.5 cm in diameter Fissuring to the level of subchondral bone in an area with a diameter of >1.5 cm Exposed subchondral bone

More than half the cartilage depth, and: A. Not to the calcified layer B. To the calcified layer C. To the subchondral bone D. Blisters Osteochondral lesion violating the subchondral plate A. Superficial B. Deep

CLASSIFICATION

ICRS, International Cartilage Repair Society.

with osteotomies popularized this technique for the treatment of osteoarthritis with comparatively large correction angles.10 The population treated for chondral defects today, however, is predominantly athletic and does not tolerate large degrees of overcorrection. When performed concurrently with cartilage repair, osteotomy around the knee should restore the mechanical axis to at least neutral alignment. Even

Normal

ICRS Grade 1 Nearly Normal

A

in patients with early joint space narrowing, overcorrection of the mechanical axis should be limited to 2 degrees or less. Ligamentous insufficiency, most commonly of the ACL, increases shear forces in the knee joint, predisposes the joint to further injury, and thus contributes to chondral damage. Any patient undergoing cartilage repair should therefore be carefully evaluated for instability, which can be corrected in a staged or concomitant fashion. Meniscal insufficiency, such as after subtotal meniscectomy, increases contact stresses by up to 300% in the respective compartment and is associated with the development of osteoarthritis.11 In carefully selected patients with meniscal insufficiency, meniscal allograft transplantation can provide pain relief and improved function. The ideal candidate for allograft transplantation has a history of prior total or subtotal meniscectomy with refractory, activity-related pain localized to the involved compartment. Following meniscal allograft transplantation, good to excellent results are achieved in nearly 85% of cases, and patients demonstrate a measurable decrease in pain and increase in activity level.12

B

Earlier classification schemes were mainly descriptive in nature and have largely been abandoned. Newer systems have evolved to classify chondral defects based on size and depth to establish a universal language among ­clinicians and researchers, and to ideally provide a correlation of lesion grade with clinical outcome. Currently, the most commonly used classifications are the Outerbridge13 and International Cartilage Repair Society systems (Table 23I-1; Fig. 23I-1).14

ICRS Grade 2 Abnormal

ICRS Grade 3 Severely Abnormal

ICRS Grade 4 Severely Abnormal

A

B

A

C

D

B

Figure 23I-1   International Cartilage Repair Society classification of chondral defects. (See Table 23I-1 for further explanation of grades.)

Knee 1773

Figure 23I-2   Computed tomographic arthrogram demonstrating an osteochondritis dissecans lesion in the medial femoral condyle. The defect was previously treated with arthroscopic fixation using multiple screws.

EVALUATION Clinical Presentation and History Patients often present with a history of knee injury, or prior surgical procedures such as meniscectomy or ACL reconstruction. They report activity-related knee pain and swelling, especially with impact activities such as running. Often, the pain localizes to the affected compartment, and occasionally synovitis develops, resulting in diffuse pain. Larger defects can be associated with mechanical symptoms, profound crepitus, popping, and giving-way.

Physical Examination and Testing Depending on the acuity of symptoms, typical physical ­examination findings include an antalgic gait, soft tissue swelling and joint effusion, quadriceps atrophy, ­tenderness with palpation of the joint line and femoral condyle, and occasionally, mild varus or valgus laxity due to loss of cartilage or meniscal substance. With the exception of very advanced cases, or in large lesions with loose bodies, motion is generally preserved. It is important to evaluate limb alignment and ligamentous stability because any deficiencies should be treated in either staged or concomitant procedures.

Imaging Radiographic work-up should include a standard weightbearing anteroposterior (AP) view in extension, posteroanterior (PA) view in 45 degrees of flexion (Rosenberg view), flexion lateral view, and axial view of the patellofemoral joint (Merchant or skyline view). Double-stance, weightbearing, long-leg radiographs are obtained to quantify lower extremity alignment to determine whether corrective osteotomy is required. Computed tomography (CT) is used infrequently, unless the lesion also affects the subchondral bone, such

Figure 23I-3   Sagittal magnetic resonance image showing a focal chondral defect (arrow) with associated marrow edema of the subchondral bone. Fat-saturated proton density fast-spin echo sequence.

as in osteochondritis dissecans (OCD) or traumatic osteochondral defects. Here, CT, especially when combined with arthrography, can be very helpful to delineate the exact dimensions of the defect more precisely and to assess bone healing (Fig. 23I-2). Another application for CT is in the evaluation of patellofemoral cartilage lesions, where it allows calculation of the tibial tubercle to trochlear groove (TT-TG) distance. This parameter is an alternative to using the quadriceps angle and is important when considering a tibial tubercle osteotomy or anteromedialization. Magnetic resonance imaging (MRI) assessment of the articular surface (Fig. 23I-3) has received increased attention because of newly developed protocols for cartilagespecific high-resolution imaging and contrast enhancement with intravenous and intra-articular gadolinium. Delayed gadolinium-enhanced MRI of cartilage is a new imaging protocol that provides an assessment of the glycosaminoglycan content of cartilage (Fig. 23I-4).15 It represents a useful tool for noninvasive follow-up evaluation after cartilage repair techniques such as ACI. Although arthroscopy remains the gold standard for assessing articular injury, sensitivities and specificities approaching 90% have been reported with MRI protocols using a 1.5-Tesla magnet.16-19 Furthermore, MRI provides additional information on the ligamentous and meniscal structures, which, if compromised, would require staged or concomitant treatment.

TREATMENT OPTIONS Nonoperative Treatment Nonoperative treatment options for cartilage defects can be separated into physical measures and pharmacologic treatment, which is further subdivided into oral and injectable agents. Physical measures include weight loss, exercise, and bracing. Obesity has been established as an independent

�rthopaedic ����������� S �ports ������ � Medicine ������� 1774 DeLee & Drez’s� O 1.46e+003

1.17e+003

879

586

293

0

Figure 23I-4   Color-coded delayed gadolinium-enhanced magnetic resonance imaging of cartilage allows assessment of the glycosaminoglycan content of a defect after cartilage repair, here after autologous chondrocyte implantation. (Courtesy of Dr. Tom Minas.)

risk factor for symptomatic arthritis of the knee, with odds ratios between 3 and 10,20 and weight loss of as little as 10 lb has been shown to decrease symptomatic arthritis by as much as 50%.21 Quadriceps weakness has been implicated in the development of symptomatic knee arthritis,22 and a strengthening program is thought to improve pain and function. Although mild to moderate levels of activity are thought to be beneficial, high- and elite-level activities, especially those of impact and torsion, contribute to symptomatic arthritis of the hip and knee.23 Bracing is frequently prescribed by orthopaedic surgeons to either stabilize a ligament-deficient knee or unload a compartment in unicompartmental arthritis of the knee. Functional bracing of ACL-deficient knees serves to decrease the risk for reinjury.24 Unloader braces have been demonstrated to significantly decrease symptoms and improve quality of life in unicompartmental knee arthritis.25 Pharmacologic treatment of knee arthritis includes oral agents such as acetaminophen, nonsteroidal antiinflammatory drugs (NSAIDs), and neutraceuticals such as chondroitin sulfate and glucosamine. Acetaminophen and NSAIDs have long been the mainstay of treatment for many musculoskeletal disorders. The risks for severe liver damage with acetaminophen overdose and for cardiovascular and gastrointestinal complications with NSAID use should be discussed with the patient. More recently, ­several trials have demonstrated the efficacy of chondroitin sulfate and glucosamine for the treatment of arthritic hip and knee pain with improvements in pain and function of about 50%,26 although there is high variability in the quality of individual preparations. Injectables, including steroids and viscosupplementation, represent a more invasive but viable option if oral treatment fails to improve symptoms. Steroid injection has been proved beneficial in numerous trials, with pain relief of 30% to 50%, which is usually only temporary and is strongest in the first 4 weeks.27 Viscosupplementation with hyaluronic acid provides similar levels of, but longer lasting, pain relief.28

Operative Treatment Conventional Cartilage Repair Techniques Before the development of modern bioengineering techniques, orthopaedists were restricted to procedures that aimed to palliate the effects of chondral lesions or attempted to stimulate a healing response initiated from the subchondral bone resulting in the formation of fibrocartilage to fill the defect. Simple arthroscopic lavage and débridement of arthritic joints has been used since the 1940s29 in an effort to reduce symptoms resulting from loose bodies and cartilage flaps. Although lavage alone has not been found to be effective, in combination with débridement, it can result in adequate pain reduction in slightly more than half of patients.30,31 The goal of débridement of chondral defects is to remove any loose flaps of articular cartilage and to create a defect shouldered by a stable rim of intact cartilage leading to reduced mechanical stresses in the defect bed. Currently, its use is limited to the treatment of small, incidental lesions found during arthroscopy, or for larger and usually more diffuse arthritic lesions in an attempt to delay the need for more invasive procedures such as total joint replacement. Marrow stimulation techniques (MSTs), such as drilling, abrasion arthroplasty, and predominantly, microfracture, attempt to induce a reparative response. This is achieved by perforation of the subchondral bone after radical débridement of damaged cartilage and removal of the tidemark “calcified” zone to enhance the integration of repair tissue. Perforation of the subchondral bone results in the extravasation of blood and marrow elements with formation of a blood clot in the defect. Over time, this blood clot, and the primitive mesenchymal cells contained within, differentiates into fibrocartilaginous repair tissue that fills the defect. Unlike hyaline cartilage, this fibrocartilage largely consists of type I collagen and exhibits inferior wear characteristics. Postoperatively, MSTs require extended periods of relative non–weight-bearing for 6 weeks or longer as well as the use of continuous passive motion (CPM) for up to 6 hours per day to enhance maturation of the repair tissue. Even though MSTs result in reparative tissue with inferior wear characteristics, treatment of smaller defects (<4 cm2) results in good outcomes in 60% to 80% of patients.32 However, especially with larger defects, symptoms tend to worsen again after 18 to 24 months.

Advanced Cartilage Repair Techniques Cartilage Restoration Autologous Chondrocyte Implantation (ACI)

Restorative cartilage repair techniques introduce chondrogenic cells into the defect area, resulting in the formation of a repair tissue that more closely resembles the collagen type II–rich hyaline cartilage. The original technique of ACI was developed more than 15 years ago8 and has been used in the United States to treat more than 10,000 patients since its approval by the U.S. Food and Drug Administration (FDA) in 1997. Second- and third-generation techniques that involve the use of collagen matrices to replace the periosteal patch cover, or as a preseeded carrier, are available in Europe with more than 5-year follow-up

Knee 1775

results. These techniques offer the benefit of a less-invasive surgical approach and have demonstrated excellent early results without the periosteum-related problems seen in conventional ACI, but are not yet approved by the FDA. ACI is indicated for the treatment of medium-sized to large chondral defects with no or shallow associated osseous deficits. It has been FDA approved for application in the femoral condyle (medial, lateral, and trochlea) but has also been used to treat patellar defects. Originally reported in 1994 for the treatment of chondral defects in the knee, it has more recently been applied to other joints such as the shoulder33 and ankle,34 although such use is off-label. ACI in its current form is a two-stage procedure in which a cartilage biopsy of about 200 to 300 mg is harvested during an initial arthroscopic procedure, most commonly from the intercondylar notch. The tissue contains about 200,000 to 300,000 chondrocytes, which are released by enzymatic digestion of the surrounding matrix and expanded in a monolayer culture for several weeks, followed by staged reimplantation through an arthrotomy. Although the ideal cell density for reimplantation is controversial, in current practice, reimplantation of about 12 million cells is attempted for an average-sized lesion of 4 to 6 cm2. The postoperative rehabilitation is similar to that of an MST, using protected weight-bearing for 6 to 8 weeks and CPM for up to 6 weeks. Return to impact and pivoting activities is considered after 9 to 12 months.

Cartilage replacement techniques remove damaged cartilage along with subchondral bone and replace it with osteochondral grafts harvested from the patient or a tissue donor.

damaged or diseased cartilage and bone. Unfortunately, the supply of osteochondral allograft tissue remains limited because of issues related to the donor pool and aseptic processing. However, improved preservation techniques have been developed that allow storage times of up to 4 weeks with no significant compromise in chondrocyte viability with grafts stored at 4° C. Osteochondral allograft transplantation is used predominantly in the treatment of large and deep osteochondral lesions resulting from OCD, osteonecrosis, and traumatic osteochondral fractures but can also be used to treat peripherally uncontained cartilage and bone defects. Furthermore, osteochondral allografting presents a viable salvage option after failure of other cartilage resurfacing procedures. When it is used for the treatment of cartilage or shallow osteochondral lesions, a thin subchondral bone graft (5 to 7 mm) results in the most rapid integration and best chance of success because the mechanism of bulk allograft failure historically has been through creeping substitution and collapse of the transplanted osseous bed and not necessarily failure of the cartilage itself. The main advantages over autograft transplantation are the ability to very closely match the curvature of the articular surface by harvesting the graft from a corresponding location in the donor condyle, the ability to transplant large grafts, and the avoidance of donor site morbidity. The main concern with allograft transplantation is the small risk for disease transmission, which is estimated at 1 in 1.6 million for the transmission of human immunodeficiency virus (HIV).36 Since the advent of strict donor screening criteria in combination with polymerase chain reaction testing for HIV and hepatitis, there has not been an identified case of viral disease transmission.

Osteochondral Autograft Transfer

New Developments

Cartilage Replacement Techniques

Osteochondral autograft transplantation is used in procedures such as OATS (Arthrex, Naples, Fla) and ­mosaicplasty (Smith & Nephew, Andover, Mass) to address mediumsized defects (1 to 4 cm2), often with associated bone loss. In this technique, multiple small cylinders of cartilage and subchondral bone are harvested from lesser–weight-­bearing areas of the knee joint. The chondral defect is prepared with a punch to create a recipient hole that matches the graft cylinders, which are then press-fitted into the defect. Commonly, multiple cylinders have to be transplanted to fill larger defects. Osteochondral autografting is limited by the amount of cartilage that can be harvested without violating the weight-bearing articular surface.35 The main advantage lies in its autogeneity, avoiding the risk for disease transmission, immediate graft availability through harvesting of the patient’s own tissue, and decreased cost of this single-stage procedure.

Osteochondral Allograft Transfer More than 750,000 musculoskeletal allografts were transplanted in 1999, mainly for the treatment of bone defects and for the reconstruction of the ACL. More recently, the treatment of chondral defects with fresh osteochondral allografts has garnered significant attention because of its potential to restore and resurface even extensive areas of

Autologous Chondrocyte Implantation– Associated Matrix Use Matrices were introduced to improve on a number of ­perceived shortcomings of the first-generation ACI (ACI-P, for periosteal cover). These techniques are also performed as staged procedures, with an initial arthroscopic cartilage biopsy followed by cellular expansion and reimplantation. A modification (ACI-C, for ­collagen cover) replaces the periosteal patch with a collagen matrix, commonly a two-layered, type I/III porcine collagen or other synthetic construct. This modification was successful in virtually eliminating patch hypertrophy37 and obviates the need to harvest the periosteal patch, thus decreasing surgical time and morbidity associated with a wider ­exposure. For ­secondgeneration ACI, or MACI, the ­ chondrocytes are seeded onto a type I/III porcine-derived collagen carrier matrix after initial expansion in culture. This matrix is then sized to match the defect and implanted through either a miniopen approach or arthroscopically with fibrin glue fixation. The use of a preseeded matrix also addresses disadvantages of the ACI procedure that are associated with implantation of a cell suspension, such as the risk for cell leakage from the defect and the potentially uneven cell ­ distribution within the defect. European ­ studies ­ investigating MACI have

�rthopaedic ����������� S �ports ������ � Medicine ������� 1776 DeLee & Drez’s� O

shown clinical results comparable to the ACI-P and ACI-C techniques.38 However, similar to ACI, MACI is limited by relatively slow cell growth and differentiation, which precludes early aggressive rehabilitation. MACI is not available in the United States because of FDA ­restrictions.

Other Matrix-Associated Techniques The successful application of biologic matrices as carrier devices for autologous chondrocytes has led several groups to investigate the use of such matrices in conjunction with MSTs. A collagen matrix is placed into a chondral defect following microfracture to stabilize the resultant blood clot and allow cell adherence. In comparison with ACI and MACI, no initial harvest procedure is needed to obtain a cartilage biopsy, and the technique can be performed all-arthroscopic. Most important, the decision to perform this type of chondral repair can be made intraoperatively because the acellular carrier matrix has an extended shelf-life and is not patient specific. Early work with this technique in a sheep model, however, demonstrated results worse than microfracture alone.39 Future research to modify these matrices with growth factors to enhance cell adherence and differentiation holds promise to improve the results of this technology. Another technique currently in clinical trials uses a resorbable scaffold that is seeded with chondrocytes in the operating room, just before implantation (Cartilage Autograft Implantation System, Depuy, Raynham, Mass). A cartilage biopsy is harvested from the patient’s knee and minced to release the chondrocytes; the resultant paste is then reimplanted during the same setting using a synthetic matrix. In vitro studies have demonstrated the viability of chondrocytes and outgrowth of matrix following the mincing process.40 A phase I clinical trial is currently under way.

Synthetic Plugs Synthetic PLA-PGA CaSO4 (OBI TruFit, OsteoBiologics Inc., San Antonio, TX) plugs have been FDA approved to back-fill donor sites and thus decrease morbidity after osteochondral autograft procedures. Studies investigating the use of this plug technology for the primary treatment of chondral defects are being conducted in animal models.41 Currently, this plug technology has not received FDA approval to treat chondral defects.

Tissue-Engineered Cartilage Autologous articular cartilage engineered by pressure perfusion is currently undergoing clinical trials in the United States (NeoCart, Histogenics Inc., Northampton, Mass) for the treatment of small lesions (2 to 3 cm2). The chondrocytes are harvested arthroscopically in an initial staging procedure. The cells are then grown to confluence in 2 to 3 days, seeded on a type I bovine collagen membrane, and then placed in a fluid chamber that pressure-cycles nutrients through the matrix until near-mature tissue is produced. During reimplantation, the tissue is cut to fit the templated defect and secured to the subchondral bone with a collagen-based glue. The patient follows an ­accelerated rehabilitation protocol that includes immediate full weightbearing. Early results are encouraging.

Allogeneic tissue-engineered cartilage that is derived from immature donors younger than 12 years of age has apparent promise for nonimmunogenic incorporation into mature recipients (Neocartilage, ISTO Technologies, St. Louis, in collaboration with Zimmer Holdings Inc.). The results to date are limited to animal models and are unpublished as of this time. Phase I trials are currently under way in the United States.

Gene Therapy and Growth Factors Autologous mesenchymal cells can be obtained either from peripheral blood or through marrow-stimulation techniques and combined with a carrier matrix that provides a mechanical and biologic environment conducive to chondrocyte differentiation. To succeed, several areas have to be addressed, including improved differentiation of mesenchymal cells into chondroblasts, production and maintenance of a hyaline cartilage matrix, and successful integration with the surrounding cartilage. The use of growth factors offers a potential solution to these issues, including the transforming growth factor-β (TGF-β), fibroblast growth factor (FGF), and bone morphogenic protein (BMP) families, which can influence cell differentiation (e.g., TGF-β1 and TGF-β2), proliferation (FGF-2, insulin-like growth factor-I [IGF-I]), and matrix production (IGF-I, BMP-2, and BMP-7).42 The short half-life of growth factors in vivo limits their use when injected or even when bound to a carrier matrix. Gene therapy offers a potential solution to this problem by creating cells that can locally produce and deliver growth factors in higher concentrations for prolonged time. Initial experiments have shown promise, but current techniques remain limited by the only transient expression of growth factors, and further research needs to establish the optimal combination of growth factors and their application.

Stem Cells Mesenchymal stem cells (MSCs) have the ability to differentiate into many diverse cell lineages, including chondrocytes. Initially identified in bone marrow aspirates, MSCs have also been isolated from adipose tissue, skin, umbilical cord, and peripheral blood. MSCs are autogenous, can be obtained through simple in-office biopsy of muscle or fatty tissue, and expanded more than 500-fold, with a potential cell yield in the billions.43 Studies are under way to seed the expanded cells onto a suitable carrier matrix, which are then exposed to a combination of environmental factors, such as hypoxia and hydrostatic pressure, and to biochemical agents, such as growth factors to commit the cells to the chondrogenic pathway44 before surgical implantation.

Weighing the Evidence Microfracture Even though marrow stimulation techniques result in a repair tissue with inferior wear characteristics, treatment of smaller defects (<4 cm2) results in good outcomes in 60% to 80% of patients.32,45

Knee 1777

   TABLE 23I-2  Results of Autologous Chondrocyte Implantation for Chondral Defects in the Knee Study

No. of Subjects

Defect Type

Follow-up

19948

  23

Knee, all locations

Average, 39 mo

Peterson et al, 200053

  94

Knee, all locations

Micheli et al, 200150 Minas, 200154

  50 107

Knee, all locations Knee, all locations

Peterson et al, 200355 Mithöfer et al, 200556

  58 OCD   20 adolescents Knee, all locations

Minas & Bryant, 20059

  45

Brittberg et al,

Patellofemoral defects

Results and Comments

14 of 16 femoral lesions with good or excellent results; 5 of 7 patellar grafts failed Average, 4 yr Good or excellent results in 24 of 25 patients with femoral lesions; 11 of 19 patients with patellar defects treated with concomitant realignment; 16 of 18 OCD lesions; 12 of 16 with concomitant ACL reconstruction; 9 of 15 with multiple lesions >36 mo 94% graft survivorship 36 mo after surgery >12 mo Overall, 87% improvement, 13% failures (defined as lack of improvement or objective graft failure) Average, 5.6 yr 91% good or excellent results; 93% patient satisfaction Average, 47 mo 96% good or excellent results; 60% return to athletic activity levels equal to or greater than before injury Average, 46.4 mo 71% good or excellent results

ACL, anterior cruciate ligament; OCD, osteochondritis dissecans.

Autologous Chondrocyte Implantation

Osteochondral Autograft Transplantation

Several long-term studies have reported good to ­excellent results in 70% to 80% of patients after ACI for the ­treatment of chondral lesions in the knee (Table 23I-2). The results of ACI compare favorably with other forms of treatment, such as débridement,46 microfracture,47 ­ mosaicplasty,48 and osteochondral autograft transfer.49 Patch hypertrophy resulting in mechanical symptoms such as clicking and popping occurs in up to 15% to 20% of patients, typically 7 to 9 months after the procedure,50 and can be addressed with arthroscopic débridement of the hypertrophic tissue.

Patients treated with osteochondral autograft transplantation experienced good to excellent results in about 90% of condylar lesions, 80% of tibial defects, and 70% of trochlear lesions.51 The treatment of patellar defects remains controversial, with some groups reporting almost universal failure in this location.48

Osteochondral Allograft Transplantation Following osteochondral allograft transplantation, good to excellent results are achieved in nearly 85% of cases, and patients demonstrate a measurable decrease in pain and increase in activity level (Table 23I-3).

   TABLE 23I-3  Results of Osteochondral Allograft Transplantation for Chondral Defects in the Knee Study 198957

Meyers et al, Garrett, 199458 Ghazavi, 199759 Chu et al, 199960 Aubin et al, 200161 Shasha et al, 200362

No. of Patients

Mean Age (yr)

Location*

Mean Follow-Up

Results

  39   17 123   55   60   65

38 20 35 35 27 N/A

F, T, P F F, T, P F, T, P F T

3.6 yr 3.5 yr 7.5 yr 75 mo 10 yr 12 yr

78% success, 22% failure 94% success 85% success 76% good or excellent, 16% failure 84% good or excellent, 20% failure Kaplan-Meier survival rate: 5 yr—95%; 10 yr, 80%; 15 yr, 65% 20 yr, 46%

*Defect location: F, femur; T, tibia; P, patella.

Authors’ Preferred Method We have developed a comprehensive treatment a­ lgorithm (Fig. 23I-5) based on defect size and location that takes into consideration the patient’s activity and demand ­ level. 52 Four commonly used techniques that have ­yielded excellent results in our experience are discussed here.

Microfracture

Microfracture is most commonly performed as an ­ all­arthroscopic procedure, and the set-up and patient positioning follow that of routine knee arthroscopy. In very ­posterior defects, the patient should be positioned so that knee ­hyperflexion can be achieved. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1778 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d Lesion Location#

Femoral Condyle

Patellofemoral

Malalignment

Meniscal Deficiency

Rehabilitation

Ligament Insufficiency

Patellofemoral Alignment

Size

Size

< 2-3 cm2

≥ 2-3 cm2

< 2-3 cm2

Microfracture ++ 1˚ OC Autograft ++ ACI OC Allograft

+/+/++ ++

Low Demand Microfracture ++ ACI/AMZ* +/OC Autograft/AMZ* OC Allograft/AMZ*

OC Autograft ++ 2˚ ACI +/OC Allograft

++ ++

ACI/AMZ* ++ OC Autograft/AMZ* + High Demand OC Allograft/AMZ* +

≥ 2-3 cm2

++ +/+ ++ ++

Figure 23I-5   Treatment algorithm for isolated femoral and patellofemoral focal chondral lesions. For condylar lesions, comorbidities of ligament instability, meniscal deficiency, and malalignment must be assessed and corrected if needed. For trochlear and patellar lesions, patellofemoral alignment must be evaluated to select the proper degree of anteromedialization. ACI, autologous chondrocyte implantation; AMZ, anteromedialization tibial tubercle osteotomy; OC, osteochondral; #, assumes minimal bone loss; *, AMZ is often performed for lateral and central patellofemoral lesions and is of questionable benefit for medially located patellofemoral lesions. (Redrawn from Alford JW, Cole BJ: Cartilage restoration, part 2: Techniques, outcomes, and future directions. Am J Sports Med 33[3]:443-460, 2005.)

Approach and defect preparation. After routine ­ iagnostic arthroscopy, the chondral defect is débrided d with a motorized shaver in forward or reverse, and a curet is used to achieve stable vertical shoulders. The débridement includes the calcified cartilage layer and should not violate the subchondral bone. Occasionally, accessory portals have to be created depending on the exact defect size and location. Microfracture. After thorough débridement, ­ multiple holes are created in the subchondral bone with a ­microfracture awl (Fig. 23I-6A). In an effort to remain ­perpendicular to the chondral surface, it may be necessary to rotate the ­articular surface in line with the awl or create accessory ­portals. It is important to preserve the integrity of the subchondral bone, which can be violated if holes are not spaced wide enough and thus connect or become confluent. Ideally, the micr­o­ fracture holes should be spaced about 3 to 4 mm apart, resulting in 3 to 4 holes per cm2. Stability of the transition zone between surrounding cartilage and regenerate ­fibrocartilage

can be improved by placing holes directly adjacent to the defect shoulders. After completion of the microfracture, pump pressure is lowered, and bleeding should be observed from all holes (see Fig. 23I-6B). Closure. Arthroscopy portals are closed with interrupted sutures. The patient should be counseled that occasionally joint aspiration may become necessary because of persistent bleeding from the treated defect. Autologous Chondrocyte Implantation

After arthroscopic cartilage biopsy and culture, a process that usually takes about 6 weeks, the cell suspension is shipped to the surgical facility. The patient is positioned supine on a standard operating table with a thigh tourniquet. Especially in very posterior lesions of the femoral condyle, a leg positioning device is helpful to stabilize the knee in hyperflexion. The lower leg is prepared into the field to just above the ankle to allow harvesting of the periosteal patch.

Knee 1779

Authors’ Preferred Method—cont’d

A

B

Figure 23I-6   A, Arthroscopic image showing a full-thickness chondral defect on the tibial plateau. Multiple pick holes are created with the microfracture awl. B, Bleeding is observed from the pick holes after the tourniquet has been released and the arthroscopic fluid pressure has been lowered.

Approach. For single lesions of the femoral condyles, a ­ limited medial or, less commonly, lateral ­ parapatellar ­arthrotomy is used (Fig. 23I-7A). Adequate exposure is critical, and it may become necessary to mobilize the meniscus by incising the coronary ligament and taking down the ­intermeniscal ligament, with subsequent repair at the end of the procedure. Correct placement of retractors is crucial, ­especially in limited incisions. A bent Hohmann or Z-­retractor placed into the intercondylar notch is helpful to displace the patella to the contralateral side. A Z- or rake ­retractor is ­helpful to control the peripheral soft tissues. For multiple ­defects, a standard medial parapatellar arthrotomy is performed with lateral ­subluxation or dislocation of the ­patella. Defect preparation. Meticulous preparation of the ­lesion is critical for the success of the procedure. The ­defect must be cleaned of all degenerated tissue to achieve a ­stable rim of healthy cartilage with vertical shoulders. This is performed by first outlining the defect with a scalpel incision down to the subchondral plate, taking as much of the surrounding cartilage to remove all unstable or undermined areas. However, if this would transform a contained into an uncontained lesion, it is advisable to leave a small rim of degenerated cartilage to sew to rather than using bone tunnels or suture anchors. The defect is then thoroughly débrided with small ring or conventional curets while maintaining an intact subchondral plate to minimize bleeding, which results in migration of a mixed stem cell population from the marrow cavity into the chondral defect. If the subchondral plate is sclerotic, such as in chronic defects, partially healed OCD lesions, or after prior microfracture, we prefer to carefully thin out the sclerotic bone with a fine bur and cold irrigation. Minor bleeding from the subchondral bone is controlled with thrombin- or epinephrine-soaked sponges or, rarely, a needle-tipped electrocautery device in the cutting mode. After the defect is prepared, it is templated using glove paper, which is oversized by about 2 mm in both length and width because there is shrinkage of the periosteum as it is procured. Periosteal harvest. The most accessible site for procurement of the periosteal patch is the proximal medial tibia.

Either the arthrotomy is extended distally or a second incision is made located centrally over the anteromedial surface of the proximal tibia starting 3 to 5 cm inferior to the pes ­anserine insertion. The subcutaneous fat is incised superficially, and further dissection with Metzenbaum scissors exposes the tibial periosteum. The template is used to outline the periosteum, which is incised using a fresh No. 15 blade and mobilized with a small, sharp periosteal elevator. The patch should be gently removed from its bony bed to avoid tearing; the periosteum is pulled upward with nontoothed microforceps as it is gently removed from the tibia with a gentle pushing motion of the periosteal elevator (see Fig. 23I-7B). After the patch has been harvested, it should be spread out on a moist sponge to avoid desiccation and shrinkage. If a tourniquet has been used, it can be deflated at this point for the remainder of the procedure. Patch fixation. The periosteal patch is retrieved from the back table and placed over the defect, with the ­cambium ­layer facing the defect. The periosteum is gently unfolded and stretched with nontoothed forceps; an obviously ­oversized patch can be trimmed back carefully at this time, preserving a small rim of 1 to 2 mm. Suturing is performed with 6-0 Vicryl on a P-1 cutting needle immersed in mineral oil or glycerin for better handling. The sutures are placed through the periosteum and then the articular cartilage, ­exiting about 3 mm away from the defect edge, everting the periosteal edge slightly to provide a better seal against the defect wall. The knots are tied on the patch side to remain below the level of the adjacent cartilage. Interrupted sutures are initially placed on each side of the patch (3, 6, 9 and 12 o’clock), adjusting the tension of the patch after each suture and trimming the periosteum as needed to obtain a patch that is neither too loose as to sag into the defect, nor so tight that it would cut out of the sutures. Thereafter, additional sutures are placed in-between to circumferentially close the gaps. An opening wide enough to accept an angiocatheter is left in the most superior aspect of the defect to inject the chondrocytes. Water tightness of the suture line is first tested by slowly injecting saline into the covered defect with a tuberculin syringe and plastic 18-gauge ­angiocatheter. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1780 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d Figure 23I-7   A, A large chondral defect has been outlined on the weight-bearing femoral condyle. B, Periosteum is being harvested from the proximal tibia: fine, toothless pickups keep tension on the patch, which is freed from the bone with a sharp periosteal elevator. C, The periosteal patch has been sutured in place and the suture-line waterproofed with fibrin glue.

A

B

Any leakage should be addressed with additional sutures or fibrin glue as needed. Lastly, the saline is reaspirated to prepare the defect for implantation. Chondrocyte implantation. The cells are now resuspended and sterilely aspirated from the transport tubes with a tuberculin syringe through an 18-gauge or larger needle because smaller gauge needles can damage the cells. The needle is then removed and replaced with a flexible, plastic 18-gauge, 2-inch angiocatheter. The angiocatheter is ­introduced into the defect through the residual opening of the periosteal patch. As the angiocatheter is slowly withdrawn, cells are injected until the defect is filled with fluid. One or two additional sutures and fibrin glue are then used to close the injection site (see Fig. 23I-7C). Wound closure. We minimize the use of intra-­articular drains to avoid damage to the periosteal patch. When drains are used, it should be without suction and with care to ­position the tubing away from the defect. The wound is closed in layers, and a soft dressing is applied to the knee. Prophylactic intravenous antibiotics are used for 24 hours after surgery.

C

Osteochondral Autograft Transplantation

Osteochondral autograft transplantation can be performed through either an arthroscopic or open approach (Fig. 23I-8A), based on the exact defect size and location and surgeon’s preference. Frequently, we assess the defect arthroscopically, followed by harvesting of the autograft from the trochlear ridge through a small, 1.5- to 2-cm parapatellar incision. We believe that a mini-open, rather than arthroscopic, approach allows better curvature matching of the donor to the recipient site. This is especially true for the donor site, which is typically difficult to access through an all-arthroscopic approach. Several proprietary systems are available, and the surgeon should follow the guidelines of the respective system used. We describe the general technique here, without reference to individual systems. Approach and defect preparation. The lesion is evaluated through routine knee arthroscopy. Needle localization with an 18-gauge spinal needle aids in the creation of ­accessory portals that allow perpendicular orientation of the

Knee 1781

Authors’ Preferred Method—cont’d

A

C harvesting tube to the articular surface. Occasionally, a central, transpatellar tendon approach is required. In this case, the patellar tendon should be split in line with its fibers and repaired at the end of the procedure. An appropriately sized defect harvesting tube is selected, and dependent on the size of the lesion, one or more recipient holes are created, thus removing the chondral defect along with about 8 to 10 mm of subchondral bone. This is best performed under tourniquet to improve visualization. Graft harvest. Again, an accessory portal is used, or more frequently, we recommend a small, 1.5- to 2-cm incision ­after needle localization. Common donor locations include the ­intercondylar notch and medial and lateral trochlea. We ­prefer to harvest from the medial trochlea or close to the ­sulcus ­ terminalis in the lateral femoral condyle because of lower patellofemoral contact stresses in that region.35 The corresponding, slightly oversized graft harvesting chisel is ­selected and placed perpendicular to the articular surface. With a mallet, it is advanced to a depth of about 8 to 10 mm; the harvester is then twisted or toggled and retrieved with the graft. This process is repeated until the required number of osteochondral plugs has been harvested. The donor site can be

B

Figure 23I-8   A, A trochlear defect from a nail gun injury has been exposed through an arthrotomy (the patient also required autologous chondrocyte implantation to the patella). B, The graft has been advanced slightly out of the harvesting chisel to allow trimming to the correct length. C, An osteochondral cylinder has been transferred from the lateral trochlea to the more centrally located defect. The harvest site has been back-filled with a synthetic plug.

left ­untreated, or filled with synthetic graft, such as the TruFit back-fill plug (OBI, San Antonio, Tex) or similar material. Graft placement. The depth of the recipient site is measured to ensure that the defect is about 1 mm deeper than the length of the graft. The graft is then advanced within the harvesting tube so that the end is just visible (see Fig. 23I-8B). The harvester is introduced into the joint and oriented perpendicularly to the articular surface, and the graft is slowly advanced into the recipient site. Subsequently, the harvester is removed, and the graft is fully seated and made flush by gentle pressure with an oversized tamp. It is preferable to slightly recess the graft rather than leaving it proud (see Fig. 23I-8C). Closure. The arthroscopic portals are closed with interrupted sutures. A mini-arthrotomy should be closed in layers, including the joint capsule and retinaculum. Osteochondral Allograft Transplantation

Similar to ACI, osteochondral allograft transplantation is performed through an arthrotomy sized to be consistent with the location and extent of the lesion. The patient Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1782 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

A

B

C

D

E

F

Figure 23I-9   A, A large chondral defect of the weight-bearing femoral condyle is inspected through a limited, peripatellar approach. B, A guide pin has been introduced perpendicularly into the defect and is being overdrilled with an appropriately sized reamer. C, The prepared defect after reaming to a depth of about 6 to 8 mm. D, Selecting the appropriate harvest site on the fresh allograft hemicondyle. E, Allograft cylinder before trimming to the appropriate depth, which has been marked out for each quadrant. F, The transplant has been introduced into the recipient site. Press-fit fixation is usually sufficient but can be augmented with resorbable pins.

Knee 1783

Authors’ Preferred Method—cont’d is positioned supine on a standard operating room table. A leg positioning device can be helpful for very posterior ­lesions that require hyperflexion of the knee. The most commonly used technique is the press-fit plug or mega-OATS ­technique; several proprietary systems have been developed to facilitate graft sizing and preparation. Approach. Most commonly, an anterior midline incision is made from the proximal pole of the patella to the tibial tubercle, but medial or lateral paramedian incisions can be used as well. The incision is carried down to the capsule; then full-thickness skin flaps are raised to create a mobile window. A medial or lateral peripatellar capsulotomy is performed from the superior pole of the patellar to the tibial tubercle. Recently, more limited incisions such as the subvastus or midvastus approaches have gained popularity, and the authors feel that these approaches allow for accelerated postoperative quadriceps rehabilitation. The patella is retracted with either a Z-retractor or bent Hohmann retractor placed into the notch. We have found it helpful to release the fat pad, and dissect the anterior meniscal horn from the capsule to allow for better exposure, especially with small incisions. Defect preparation. Once the lesion is exposed, the abnormal cartilage is identified (Fig. 23I-9A). It is of utmost importance to reconstruct the normal geometry of the articular surface with the donor graft. A cylindrical sizing guide is placed over the defect to determine the

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Rehabilitation Tables 23I-4 to 23I-6 describe our postoperative rehabilitation protocols after microfracture and osteochondral autograft and allograft transplantation. The rehab protocol after ACI (Table 23I-7) is divided into three phases, based on the slow maturation of the repair tissue, which at the same time has to be protected from overloading and stimulated to encourage tissue ­maturation. The three phases of the healing process are the proliferative (fill) phase, the transitional (integration) phase, and the remodeling (hardening) phase, each of which can accommodate increasing amounts of load. During the initial proliferative phase, protection of the graft is paramount, and the patient is limited to touchdown weight-bearing for 6 weeks. During this phase, patients also use a CPM machine for 6 to 8 hours per day to reduce the likelihood of adhesions and aid in maturation of the transplant. This initial period is followed by the transitional stage in which patients advance to full weight-bearing over the course of several weeks. Additional exercises are prescribed based on the specific

­ ptimal plug diameter, and a guide pin is drilled through o the sizing guide to a depth of 2 to 3 cm. A 6- to 8-mm deep recipient socket is created with a cannulated reamer (see Fig. 23I-9B and C), and the exact depth of the cylindrical defect is measured in all four quadrants. Multiple drill holes are created in the floor of the defect to improve blood supply. Graft preparation. The appropriate donor site is then identified on the allograft condyle (see Fig. 23I-9D), which is secured in the workstation, and a mark is made at the 12-o’clock position to aid in orientation. A bushing of appropriate diameter is selected and set to the angle required to match the contour of the recipient site. The donor harvester is passed through the proximal graft housing and drilled through the entire depth of the donor condyle. After extracting the graft from the donor harvester, the four quadrants of the graft are marked (see Fig. 23I-9E) and trimmed down to match the depths previously recorded from the recipient site. It is helpful to slightly bevel the edges of the graft to facilitate insertion and avoid excessive impaction pressures. Graft insertion. The recipient site is now prepared for insertion with a calibrated dilator, and the graft is press-fit with manual pressure and gentle tapping (see Fig. 23I-9F). It is preferable to recess the graft slightly rather than leaving it proud. Absorbable pins may be used for supplemental fixation.

l­ ocation and type of the defect. ­During the final remodeling phase that begins about 3 months after transplantation, the joint is increasingly loaded with strengthening and impact-loading activities. A full return to high-impact and pivoting activities should be delayed for at least 12 months until near-complete graft maturation has been achieved. Complete maturation is not expected until 12 to 24 months.

Outcomes Measures Commonly used outcomes measures include functional scores, such as the Lysholm score, Tegner activity, and International Knee documentation Committee rating systems, as well as health surveys, including the SF-12 or SF-36. Other measures include subjective parameters such as patient satisfaction, and objective ones such as range of motion, swelling, quadriceps atrophy, radiographic joint space narrowing, and others.

Complications General complications inherent to knee surgery are infection and stiffness, which are more common with open than arthroscopic procedures. Nerve damage after knee surgery usually takes the form of injury to the infrapatellar branch of the saphenous nerve, resulting in a numb area over the

�rthopaedic ����������� S �ports ������ � Medicine ������� 1784 DeLee & Drez’s� O

   TABLE 23I-4  Rehabilitation Protocol for Microfractures Phase

Weight-Bearing

Brace

Range of Motion

Therapeutic Exercise

Microfracture of the Femoral Condyle

I (0-8 wk)

Touchdown weight-bearing (20-30%) for the first 6-8 weeks.

None

II (8-12 wk)

Gradual return to full weight-bearing Full with a normalized gait pattern.

None

CPM: use is for 6-8 hours per day. Set Passive stretching/exercise for the at a rate of 1 cycle/minute, advancing first 6-8 weeks, quad/hamstring 10 degrees daily. Begin at a level of isometrics. flexion that is comfortable for the patient. Advance to full flexion as tolerated. Gain full and pain free. Progressive active strengthening

None

Full and pain free.

III (6-9 mo)

Return to full activities, including cutting, turning, and jumping.

Microfracture of the Patellofemoral Articulation

I (0-8 wk)

Weight-bearing as tolerated

II (8-12 wk)

Full

Locked 0-40 degrees CPM: use is for 6-8 hr per day. Set at of flexion for a rate of 1 cycle/min, ranging from weight-bearing 0-40 degrees None Gain full and pain free

III (≥12 wk)

Full

None

Full and pain free

Passive stretching and exercise for the first 6-8 wk; quad and hamstring isometrics Begin closed chain activities, emphasizing a patellofemoral program Return to full activities, including cutting, turning, and jumping

CPM, continuous passive motion.

   TABLE 23I-5  Rehabilitation Protocol for Osteochondral Autograft Transplantation Phase

Weight-Bearing

Brace

Range of Motion

Therapeutic Exercise

I (0-6 wk)

Non–weight-bearing

0-6 wk: CPM is used for 6-8 hr/day. Begin at 0-40 degrees, increasing 5-10 degrees/day per patient comfort. Patient should gain 100 degrees by week 6.

Passive and active-assisted ROM to tolerance; patella and tibiofibular joint mobilizers (grades I and II); stationary bike for ROM; quad, hamstring, adduction, and gluteal sets; hamstring stretches; hip strengthening; SLRs; ankle pumps

II (6-8 wk)

Progress to full weight-bearing

0-1 wk: Locked in full extension (removed for CPM and exercises) 2-4 wk: Gradually open brace in 20-degree increments as quad control is gained; discontinue use of brace when quads can control SLR without an extension lag None

Gradually increase flexion. Patient should have 130 degrees of flexion.

III (8-12 wk)

Full with a normalized gait pattern

Gait training; scar and patellar mobilizers; quad and hamstring strengthening; begin closed chain activites (wall sits, shuttle, mini squats, toe raises); begin unilateral stance activities Advance phase II activities

None

Full and pain free

CPM, continuous passive motion; ROM, range of motion; SLR, straight leg raises.

   TABLE 23I-6  Rehabilitation Protocol for Osteochondral Allograft Transplantation Phase

Weight-Bearing

Brace

I (0-6 wk)

Non–weight-bearing

II (6-8 wk)

Partial weight-bearing (25%)

0-1 wk: Locked in full extension 0-6 wk: CPM is used for (removed for CPM and exercises) 6-8 hr per day. Begin at 2-4 wk: Gradually open brace in 0-40 degrees, increasing 20-degree increments as quad 5-10 degrees/day per control is gained; discontinue use patient comfort. Patient of brace when quads can control should gain 100 degrees SLR without an extension lag by week 6. None Gradually increase flexion. Patient should have 130 degrees of flexion.

III (8-12 wk) Gradually return to full weight-bearing IV (12 wk to 6 mo)

None

Full with a normalized None gait pattern

CPM, continuous passive motion; ROM, range of motion; SLR, straight leg raise.

Range of Motion

Progress to full and pain free. Full and pain free.

Therapeutic Exercise Passive and active-assisted ROM to tolerance; patella and tibiofibular joint mobilizers (grades I and II); quad, hamstring, and gluteal sets; hamstring stretches; hip strengthening; SLRs Scar and patellar mobilizers; quad and hamstring strengthening; stationary bike for ROM; continue to advance lower extremity strengthening activites Gait training; begin closed chain activites (wall sits, shuttle, mini squats, toe raises); begin unilateral stance activities Advance phase III activities

Knee 1785

   TABLE 23I-7  Rehabilitation Protocol for Implantations Phase

Weight- Bearing

Brace

Range of Motion

Therapeutic Exercise*

0-2 wk: Quad sets, SLR, hamstring isometrics—complete exercises in brace if quad control is inadequate. 2-6 wk: Begin progressive closed chain exercises. †6-10 wk: Progress bilateral closed chain strengthening, begin open chain knee strengthening. 10-12 wk: Progress closed chain exercises using resistance less than patient’s body weight, progress to unilateral closed chain exercises, begin balance activities. Advance bilateral and unilateral closed chain exercises with emphasis on concentric and eccentric control. Continue with biking, Stairmaster, and treadmill; progress balance activities. Advance strength training, initiate light plyometrics and jogging. Start 2-min walk/2-min jog. Emphasize sportspecific training. Continue strength training. Emphasize single-leg loading. Begin a progressive running and agility program. Highimpact activities (e.g., basketball, tennis) may begin at 16 mo if pain free.

Femoral Condyle Autologous Chondrocyte Implantation

I (0-12 wk)

0-2 wk: Non–weightbearing 2-4 wk: Partial weightbearing (30-40 lb) 4-6 wk: Progress to use of one crutch 6-12 wk: Progress to full weight-bearing

0-2 wk: Locked in full extension (removed for CPM and exercise) 2-4 wk: Gradually open brace 20 degrees at a time as quad control is gained. Discontinue use of brace when quads can control SLR without an extension lag.

0-4 wk: CPM: use in 2-hr increments for 6-8 hr/day at 1 cycle/min. Begin at 0-30 degrees, increasing 5-10 degrees/day per patient comfort. Patient should gain at least 90 degrees by week 4 and 120 to 130 degrees by week 6.

II (3-6 mo)

Full with a normalized gait pattern

None

Full active range of motion

III (6- 9 mo)

Full with a normalized gait pattern

None

Full and pain free

IV (9-18 mo)

Full with a normalized gait pattern

None

Full and pain free

Patellofemoral Autologous Chondrocyte Implantation‡

I (0-12 wk)

0-2 wk: Non– weight-bearing 2-4 wk: partial weightbearing (30-40 lb) 4-8 wk: Continue with partial weight-bearing. Progress to use of one crutch. 8-12 wk: Progress to full weight-bearing and discard crutches.

0-2 wk: Locked in full extension, removed only for CPM and exercise 2-4 wk: Locked in full extension with weightbearing 4-6 wk: Begin to open 20 to 30 degrees with ambulation. Discontinue use after 6 weeks.

0-4 wk: CPM: use in 2-hr increments for 6-8 hr/day. Begin at 0-30 degrees—1 cycle/min; after week 3, increase flexion by 5-10 degrees/day. 6-8 wk: Gain 0-90 degrees 8 wk: Gain 0-120 degrees

II (3-6 mo)

Full with a normalized gait pattern

None

Full range of motion

III (6-9 mo)

Full with a normalized gait pattern

None

Full and pain free

IV (9-18 mo)

Full with a normalized gait pattern

None

Full and pain free

1-4 wk: Quad sets, SLR, hamstring isometrics—complete exercises in brace if quad control is inadequate. 4-10 wk: Begin isometric closed chain exercises. At 6-10 wk, may begin weight-shifting activities with involved leg extended if full weight-bearing. At 8 wk, begin balance activities and stationary bike with light resistance. 10-12 wk: Hamstring strengthening, Thera-Band 0-30 degrees resistance, light open chain knee isometrics Begin treadmill walking at a slow to moderate pace, progress balance and proprioceptive activities. Initiate sport cord lateral drills. Advance closed chain strengthening. Initiate unilateral closed chain exercises. Progress to fast walking and backward walking on treadmill (initiate incline at 8-10 mo), initiate light plyometric activity. Continue strength training—emphasize single-leg loading, begin a progressive running and agility program. High-impact activities may begin at 16 mo if pain-free.

*If pain or swelling occurs with any activities, they must be modified to decrease symptoms. †Respect chondrocyte graft site with closed chain activities: If anterior, avoid loading in full extension, if posterior, avoid loading in full flexion > 45 degrees. ‡Most trochlear and patellar defect repairs are performed in combination with a distal realignment. Weight-bearing is restricted for the first 4 to 6 weeks to protect the bony portion of the distal realignment during healing. May consider patellofemoral taping or stabilizing brace if improper patella tracking stresses implantation. Postoperative stiffness in flexion following trochlear or patellar implantation is not uncommon, and patients are encouraged to achieve 90 degrees of flexion at least 3 times a day out of brace after their first postoperative visit (days 7-10). CPM, continuous passive motion; SLR, straight leg raise.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1786 DeLee & Drez’s� O

lateral aspect of the knee. Rarely, nerve damage can result from direct injury to the peroneal nerve, or indirect, tourniquet-related injury to the sciatic and tibial nerves. Specific risks associated with individual procedures are graft detachment or delamination after ACI, and subchondral collapse or nonunion after osteochondral allograft transplantation.

Criteria for Return to Play Most cartilage repair procedures require extensive post­ operative rehabilitation and delayed return to athletic activities. As a general rule, the operated extremity should be relatively pain free, and without residual stiffness, swelling or muscle atrophy before a return to play can be considered. More important than these general guidelines, however, cartilage repair requires a protracted healing period that frequently extends long after the previously mentioned criteria have been met. This is not apparent to outside observation by the patients, who therefore will have to be frequently reminded of their restrictions. Return to play is individualized based on the specific procedure and sport; participation in high-impact and pivoting activities such as basketball and soccer is delayed for 9 to 12 months, whereas impact-free activities such as stationary biking can be considered as early as 8 to 12 weeks after surgery.

C

r i t i c a l

P

o i n t s

l Careful preoperative assessment of number, size, and location of cartilage lesions is critical to successfully plan a cartilage repair procedure. A low threshold for a diagnostic arthroscopy should be maintained if imaging cannot provide a conclusive answer. l Cartilage defects are frequently caused by, or made more symptomatic because of, associated pathology, such as malalignment or meniscal deficiency. This pathology has to be correctly identified and addressed in a staged or concurrent manner. l Bone length radiographs are a necessity to evaluate alignment and to plan for concurrent osteotomy if needed. l A frank preoperative dialogue with the patient and family is needed to discuss the complex nature of the procedure, the long rehabilitation required, and the expected results. Patients often present with unrealistic expectations leading to less than optimal satisfaction rates. l Most cartilage repair procedures are time sensitive; fresh osteochondral allografts and autologous cultured chondrocytes have a limited shelf life of only a few days, and cannot be frozen. A discussion with the operating room staff and receiving is necessary to ensure correct delivery and handling of the grafts. The patient needs to be aware that the surgical date cannot be moved once the process has started.

Special Populations Historically, cartilage repair procedures have been limited to the younger patient. However, our aging population wants to remain active longer and is less willing to accept the limitations of joint replacements. Therefore, chronologic age older than 50 years as a contraindication to cartilage repair is being revisited. Adolescents with open growth plates require special consideration. Osteochondral allograft transplantation has to be considered carefully in order not to injure the physeal plate. Also, the osteotomies performed concurrently to cartilage repair procedures, especially in the patellofemoral joint, are often contraindicated in the growing child. Inflammatory arthritis remains a contraindication to cartilage repair, except in the rare case in which it is burnt out and the lesions are focal and contained. Congenital malformation, such as multiple epiphyseal dysplasia, can occasionally be successfully addressed with osteochondral allograft transplantation.

S U G G E S T E D

R E A D I N G S

Alford JW, Cole BJ: Cartilage restoration, part 1: Basic science, historical perspective, patient evaluation, and treatment options. Am J Sports Med 33(2):295-306, 2005. Alford JW, Cole BJ: Cartilage restoration, part 2: Techniques, outcomes, and future directions. Am J Sports Med 33(3):443-460, 2005. Cole BJ, Schumacher HR: Injectable corticosteroids in modern practice. J Am Acad Orthop Surg 13(1):37-46, 2005. Farr J, Lewis P, Cole B: Articular cartilage: Patient evaluation and surgical decision making. J Knee Surg 17(4):219-228, 2004. Gomoll AH, Minas T, Farr J, Cole BJ: Treatment of chondral defects in the patellofemoral joint. J Knee Surg 19(4):285-295, 2006. Hangody L, Rathonyi GK, Duska Z, et al: Autologous osteochondral mosaicplasty: Surgical technique. J Bone Joint Surg Am 186(Suppl):65-72, 2004. Minas T, Peterson L: Advanced techniques in autologous chondrocyte transplantation. Clin Sports Med 18(1):13-44, 1999. Steadman JR, Briggs KK, Rodrigo JJ, et al: Outcomes of microfracture for ­traumatic chondral defects of the knee: Average 11-year follow-up. Arthroscopy 19(5): 477-484, 2003.

R efere n ces Please see www.expertconsult.com

Knee 1787

S e c t i o n

J

Knee Replacement in Aging Athletes Mark J. Billante and David R. Diduch

A painful arthritic knee in a young patient can be a challenge for today’s orthopaedist. Nonoperative treatment, such as anti-inflammatory medication, physical therapy, injections, lifestyle modification, and bracing, often provides only limited and temporary relief. Operative options include arthroscopic débridement, proximal tibial osteotomy, arthrodesis, and unicompartmental or total knee replacement. Most patients are not willing to accept the functional limitations of a knee arthrodesis. Proximal tibial osteotomies have good short-term results, but do not predictably relieve pain in the long term.1-4 Unicompartmental knee replacement is an option for patients with single compartment osteoarthritis and angulation of the knee, but frequently degenerative changes are not limited to one compartment. Knee replacement is an excellent option for treatment of both osteoarthritis and rheumatoid arthritis in elderly patients. The literature clearly demonstrates both predictable pain relief and improved function.5-16 People in the United States are living longer than ever. As the average age of our population increases, so does the prevalence of arthritic joints. With changing demographics, our outlook regarding total knee replacement is also changing. The senior citizens of today are no longer content to be less active in their “golden years”; they want to maintain their active lifestyle after joint replacement, placing additional stresses on prosthesis and surgeon alike. The excellent results seen with total knee replacement in older patients have helped expand the indications to include arthritis in younger patients. Total knee replacement has been shown to be successful in treating younger patients as well.7,17-19 Yet, concerns about the durability of prostheses with regard to loosening and wear debris generated by more active patients still exists. Currently, many of the recommendations regarding recreational and sporting activity after knee replacement are based on personal and consensual opinion. This chapter reviews the current literature regarding sports participation after knee replacement while highlighting clinical evaluation, surgical techniques, and future directions for treatment in this difficult patient population.

CURRENT LITERATURE REVIEW REGARDING SPORTS AFTER KNEE ARTHROPLASTY Few studies specifically address recommendations regarding athletic activity and sports participation after knee arthroplasty. Those that do are frequently based on ­surgeon

advice and experience rather than concrete data. Prospective randomized studies are not available and would likely be impractical for this patient population. Few physicians could tell their patients they have been randomized into the group that jogs 3 times a week, or conversely tell a tennis player he or she has been randomized to the sedentary group and cannot resume tennis after surgery. Physicians therefore are left somewhere between good advice and hard evidence. The following section reviews the current pertinent literature and provides recommendations based on the best available information to date. Physicians and patients must balance the beneficial and deleterious effects of activity after knee arthroplasty. Inactivity can lead to reduced aerobic fitness, loss of coordination and postural reflexes, loss of muscle mass, and osteoporosis, whereas physical fitness and exercise reduce mortality, anxiety, and depression, and improve muscle coordination, strength, and bone density.20 Ries and colleagues found benefits in cardiovascular fitness after hip and knee arthroplasty. Significant improvements were seen in exercise duration, maximal workload, and peak oxygen consumption after 2 years. They concluded that joint replacements enable patients to increase their activities and improve their physical fitness.21,22 Conversely, other studies have shown increased activity to be associated with increased wear of the prosthesis. Schmalzried and coworkers reported that up to 500,000 submicron particles are released with each step after total knee replacement. The particles can initiate a cascade of processes that eventually lead to periprosthetic osteolysis and loosening.23,24 Schmalzried also showed that wear is a function of use. He found that wear was related more to activity than age of the patient. Lavernia and colleagues found a positive correlation between activity level, length of implantation, and wear rates at an average of 74 months.25 Implant loosening is another important consideration after knee arthroplasty. Athletic activity may increase the stress on implant fixation in compression, tension, rotation, and shear. Mallon and Callaghan found that radiolucent lines occurred in 79% of cemented total knee arthroplasties (TKAs), 45% of uncemented TKAs, and 54% of TKAs when the groups were combined in patients who played golf a minimum of 3 times per week.26,27 They also found that the occurrence of radiolucent lines and the incidence of pain during and after play were higher for patients with TKA than total hip arthroplasty. However, Diduch and coworkers reported on TKA in young, active patients and found that 9% had radiolucent lines that were present immediately postoperatively and did not progress over the course of the study.28 None of the patients in the study underwent revision surgery due

�rthopaedic ����������� S �ports ������ � Medicine ������� 1788 DeLee & Drez’s� O

to femoral or tibial component loosening. To date, there has not been a definitive study demonstrating a causative relationship between activity and implant loosening. Healy and associates surveyed 58 members of the Knee Society regarding their recommendations for athletics and sports participation after knee replacement.29 They recommended low-impact aerobics, stationary bicycling, bowling, croquet, ballroom dancing, jazz dancing, square dancing, golf, horseshoes, shooting, shuffleboard, swimming, and walking. Road bicycling, canoeing, hiking, rowing, cross-country skiing, stationary skiing, speed walking, tennis, and weight machines were recommended if the patient previously participated in these activities. In general, high-impact activities were not recommended. These activities included high-impact aerobics, baseball, softball, basketball, football, gymnastics, handball, hockey, jogging, lacrosse, racquetball, squash, rock climbing, soccer, singles tennis, and volleyball. No conclusions were made regarding fencing, rollerblading, downhill skiing, and weightlifting. Healy also emphasized delaying return to recreational or athletic activity until quadriceps and hamstrings were sufficiently rehabilitated. Healy felt surgeons should educate patients regarding risks associated with higher levels of activity but ultimately let the patient decide what activities to participate in postoperatively. Kuster used six criteria to provide more “scientifically” based guidelines regarding activity after total joint replacement.30 These criteria included wear of total joint replacements, joint load and moments during sports activities, activity and fixation of the prosthesis, recreation versus exercise, and the difference between total hip arthroplasty and TKA. Ultimately, he recommended that patients should remain physically active after joint replacement for general health, prevention of cardiac problems, improvement of bone quality, and enhanced prosthesis fixation. Fitness should be maintained by low-impact aerobic activities such as swimming, cycling, walking, or aqua aerobics. Patients who wished to continue participation in sports with higher joint loads such as skiing, tennis, and hiking should do so only on a recreational basis.31 Kuster noted that knee designs show much smaller stress levels near extension than in flexion for the same load.32 Activities such as hiking or jogging have high joint loads between 40 and 60 degrees of knee flexion, when many knee designs are not conforming.33 This may lead to very high polyethylene inlay stress. Hence, regular jogging or hiking with intense downhill walking produces a large overloaded area with the danger of delamination and polyethylene destruction for many modern knee designs. Mont and associates compared clinical and radiographic outcomes for 50 patients who engaged in high-impact activities such as golf, skiing, tennis, cycling, or jogging at a minimum of 4 years after TKA, to an age-matched cohort of 50 sedentary patients.34 At a mean follow-up of 7 years, there were two revisions and one clinical failure in each group, with no progressive radiolucencies reported. High-impact activities conferred no difference in outcomes in the study. The authors also evaluated 30 total knee replacements in patients younger than 50 years in a separate study. At a mean of 7 years’ follow-up, they reported good or excellent clinical and radiographic outcomes in 29 of 30 patients. Mont and associates also studied patients who played tennis after

TKA. Forty-six knee replacements in 33 patients were reviewed. Two of 46 (4%) knees required revision surgery due to polyethylene wear at 8 and 11 years.35 Diduch and colleagues investigated 103 TKAs in patients younger than 55 years.28 They found that patients had an increase in their Tegner scores from 1.3 preoperatively to 3.5 postoperatively. This improvement reflects a change from sedentary desk-type work with limited walking on even ground to an occupation that involves light labor, such as nursing or truck driving, and some recreational activities, such as cycling, cross-country skiing, or swimming. The Tegner score improved postoperatively in all but two patients who had no change in the score. Thus, no patient had deterioration of functional status after knee replacement. Additionally, 19 patients (24%) had a score of at least 5 points, indicating regular participation in activities such as tennis, downhill skiing, cycling, or strenuous farm or construction work. Despite the patients’ active lifestyles, loosening that necessitated revision was not a problem in their series. The survivorship estimate, with revision of the femoral or tibial component as the end point, was 94% at 18 years. The only aseptic revision involved a 22-year-old patient, who participated in football, baseball, basketball, and softball and was employed as a firefighter. The revision, 7 years postoperatively, revealed wear of a relatively thin, carbon-fiber–reinforced polyethylene spacer without loosening of the tibial tray. Accelerated wear of carbonfiber–reinforced polyethylene has been well documented, and this material is no longer recommended for use in knee replacements.36,37 The literature is particularly sparse regarding sports after unicompartmental knee replacement, with only one recent article directly addressing the issue. Fisher and colleagues reported on 42 patients who had a unicompartmental knee replacement using the Oxford prosthesis at an average follow-up of 18 months.38 They reported that 93% of patients successfully returned to their regular sporting activities following surgery. These sports included swimming, golf, dancing, cycling, hiking, jogging, and squash.

RECOMMENDATIONS REGARDING SPORTS AFTER KNEE ARTHROPLASTY Remaining physically active after joint replacement is beneficial for maintenance of general health, prevention of cardiac problems, and improvement of bone quality. Ideally, regular exercise should be limited to low-impact activities such as swimming, cycling, water aerobics, or walking. Patients who want to continue sports such as skiing, hiking, and tennis can do so on a recreational basis. Patients should also be made aware of joint-load–reducing measures such as using ski poles during hiking, skiing on flatter slopes, and avoiding icy conditions. Diagonal instead of skating techniques can be used while cross-country skiing. Cycling with low loads, higher frequencies, and increased seat height is beneficial. It may be unwise to take up new technically demanding sports after knee replacement such as skiing, hiking, mountain biking, horse riding, or tennis. The joint loads and risk for injury are higher for these activities in unskilled individuals.30 However, if patients

Knee 1789

Box 23J-1 Recommended Sporting Activities after Knee Arthroplasty Recommended/Allowed Low-impact aerobics Stationary bicycling Golf Walking Speed walking Swimming Elliptical machines Ski machines Rowing Weight machines Bowling Dancing Horseback riding Hiking Cross-country skiing Allowed with Previous Experience Road bicycling Singles, doubles tennis Racquetball Squash Downhill skiing Ice skating Not Recommended Football Basketball Baseball Hockey Soccer Lacrosse Rock climbing Volleyball Gymnastics Jogging Handball

participated in these sports before surgery, it is safe to resume them postoperatively (Box 23J-1). It is important to reiterate that there is a lack of a definitive consensus regarding sporting activity after knee arthroplasty in the literature. Healy’s article is the closest paper to a consensus, but it is based on survey results of Knee Society members rather than definitive data.29 We concur with the recommendations seen in most papers that encourage low-impact activities for maintenance of general health and fitness while avoiding high-impact activities that would place dangerous levels of stress on the prosthesis. However, we permit lateral movement activities such as skiing or racquet sports because there has not been a definitive study proving these sports to be detrimental in the survivorship of TKA. There does appear to be consensus in recommending against jogging, running, and jumping sports. As physicians, we can facilitate discussion of appropriate postoperative activities and encourage patients to use common sense when choosing what sports to ­participate in after surgery.

CLINICAL EVALUATION History A complete history and physical examination is necessary in order to avoid missing other potential sources of knee pain. Inconsistent findings at the knee should alert the examiner to evaluate the patient for alternative diagnoses such as referred pain from the hip or back. Most patients report knee pain exacerbated by activity, relieved by rest, and associated with varying degrees of swelling. The swelling may be intermittent or constant in nature. Pain localized to one compartment is common early in the disease ­process, as opposed to diffuse knee pain that suggests multiple compartment involvement and more advanced arthritis. Rest pain is also more common in advanced osteoarthritis or osteonecrosis. Pain with stair climbing, prolonged sitting, or squatting suggests patellofemoral involvement. Mechanical symptoms such as intermittent locking or catching may be related to articular surface irregularity, loose bodies, or meniscal pathology, which is common ­secondary to osteoarthritis. Pain and instability may both be present when arthritis and ligamentous insufficiency coexist. It is important to differentiate instability due to pain, effusion, and quadriceps inhibition from true ligamentous insufficiency, which may or may not be associated with pain. Responses to previous treatments such as nonsteroidal anti-inflammatory drugs (NSAIDs), therapy, injections, bracing, osteotomy, or arthroscopic débridement are helpful to direct further management. A frank discussion regarding patient expectations should also occur. Older patients may be content with limited activity, as long as they are pain free. In contrast, younger or more active patients may find the same activity levels to be severely disabling. The Internet has enabled patients to become educated about their diagnosis and treatment options. Time spent with patients answering questions, discussing treatment options in detail, and differentiating good and bad information can facilitate realistic expectations for surgeon and patient alike. Having pamphlets or websites for patients to review at their own pace is also helpful.

Physical Examination A complete examination of the knee is essential. Special attention should be paid to previous incisions on the knee, joint line tenderness, patellar tenderness and crepitus, varus and valgus deformities, and ligamentous stability. In addition, body habitus, range of motion, flexion contractures, and extension lags should be noted. The back and ipsilateral hip and ankle should be examined for abnormalities, including decreased range of motion. Distal pulses, sensation, and strength should also be evaluated.

Diagnostic Imaging A standard radiography series includes a standing posteroanterior (PA) view of both knees in 45 degrees of flexion, a lateral view, and a sunrise view of the affected knee. The

�rthopaedic ����������� S �ports ������ � Medicine ������� 1790 DeLee & Drez’s� O

that will not worsen their pain is helpful. Exercise that includes high-impact activity such as jumping and running should be avoided. Running on treadmills, stair-climbing machines, and leg extension exercises put added pressure on the patellofemoral joint and can exacerbate symptoms. Low-impact activities such as swimming and bicycling are excellent recommendations. Elliptical machines, stationary bikes, and leg press exercises are good for quadriceps strengthening without excessive loading of the patellofemoral joint. Limiting stair-climbing and squatting activities can also reduce pain in patients with patellofemoral arthritis. If possible, patients also should be encouraged to modify their work responsibilities to those that are physically less demanding. Adaptations at home such as raising the level of a chair or toilet seat can be beneficial for those with chronic symptomatic knee arthritis.

Therapy A

B

Figure 23J-1   A, Medial compartment narrowing demonstrated on a weight-bearing anteroposterior radiograph. B, A posteroanterior view with 45 degrees of flexion weightbearing in the same patient demonstrates advanced arthritic changes.

45-degree PA view can demonstrate subtle loss of joint space, especially in the lateral compartment, which is indicative of early chondrosis. Frequently, the 30- to 60-degree flexion zone has the earliest loss of cartilage and is easily overlooked on full extension radiographs (Fig. 23J-1). Because the 45-degree PA view provides an excellent view of the notch, changes that suggest chronic anterior cruciate ligament (ACL) deficiency can be seen, such as hypertrophy of the tibial spines and narrowing of the notch itself. Changes common after meniscectomy (joint space narrowing, flattening of the femoral condyles, and osteophyte formation along the periphery of the tibia) are more apparent with this view. If joint space narrowing is minimal, magnetic resonance imaging (MRI) of the knee can help identify meniscal pathology or other intra-articular abnormalities. MRI can also identify more rare pathologic processes such as avascular necrosis, spontaneous osteonecrosis of the knee, and neoplastic bone lesions. Degenerative meniscal tears often are present concurrently with osteoarthritis. The surgeon must keep in mind that the arthritis is the primary process involved and treat the patient accordingly. In most cases, if joint space narrowing is present on the 45-degree PA view, MRI is unnecessary.

TREATMENT OPTIONS Lifestyle Modifications Obesity is a known risk factor for osteoarthritis, and losing weight decreases the risk for developing and exacerbating osteoarthritis.39 Educating patients about exercise options

Range of motion exercises reduce or prevent contractures. A flexion contracture results in increased patellofemoral contact stresses during standing and walking, which can exacerbate arthritic symptoms. Periarticular muscle strengthening helps stabilize the knee and can reduce knee symptoms. Cross-training and flexibility are important components of a complete rehabilitation program. Modalities such as heat, ultrasound, hydrotherapy, and ­cryotherapy are thought to work by reflex-mediated pathways involving free nerve endings, vasodilation, and other mechanisms. Duration and frequency of these modalities should be adjusted to optimize results and minimize ­symptoms.

Bracing and Support Devices Even though knee sleeves do not alter joint reaction forces or alignment, they can provide a sense of stability through enhanced proprioception. Patients with unicompartmental arthritis can be fitted with an “unloader” brace. The device provides a three-point bending force, with one force applied at the center of the knee and two opposing forces applied proximal and distal to the knee joint. This reduces joint reactive forces in the affected compartment. Prospective studies with valgus bracing for medial compartment arthritis reported a 50% decrease in the number of patients complaining of pain with activities of daily living after brace wear for an average of 7 hours a day, 5 days a week.40 A recent study demonstrated that “off the shelf” and custom braces were both beneficial with regard to pain relief and stiffness in patients with medial compartment arthritis; however, additional relief was gained with custom braces.41 Bracing also has limitations. Cost ($800 to $1000) and the cumbersome nature of the brace may limit usefulness of this treatment option. The best candidate for brace treatment has unicompartmental arthritis with at least a partially correctable, mild deformity and minimal patellofemoral symptoms. Using a cane in the contralateral hand is an effective way to reduce symptoms by relieving force on the affected knee. Canes should reach the top of the greater trochanter of a patient wearing shoes. Unfortunately, patients are often reluctant to use supportive devices in the long term owing to perceptions of lost independence.

Knee 1791

Patients often view pain relief as the most important goal of treatment. For this reason, acetaminophen is an accepted first-line analgesic agent for treatment of osteoarthritis despite lacking anti-inflammatory action. Furthermore, its favorable side-effect profile makes it an option for patients unable to tolerate traditional NSAIDs because of gastric toxicity. The recommended dosing is 650 mg every 4 to 6 hours as needed, to a maximal dosage of 4000 mg per day. A dose of 1000 mg 3 to 4 times daily is usually sufficient. Use within recommended dose levels is rarely associated with renal toxicity or hepatotoxicity, but routine laboratory work should be done for patients with long-term use.42

­ lacebo-controlled trials suggest that glucosamine is effip cacious in managing osteoarthritis without toxicity or side effects.44 Chondroitin sulfate is a glycosaminoglycan found in articular cartilage that is important for binding collagen fibrils. It has protective effects due to competitive inhibition of the degradative enzymes that lead to cartilage breakdown. Chondroitin sulfate also inhibits thrombus formation, which can occur in periarticular tissues and limit subchondral and synovial blood flow. About 70% of the oral dose is absorbed in the gut. Clinical studies showed effective pain relief and increased function without toxicity or side effects.45 Glucosamine and chondroitin sulfate have synergistic actions when taken together. Concurrent use appears to result in a net increase in the amount of normal cartilage matrix, potentially slowing progression of osteoarthritis.46

Nonsteroidal Anti-inflammatory Medications

Corticosteroid Injections

NSAIDs have been the preferred oral medication for the treatment of the swollen and painful arthritic knee because of their analgesic and anti-inflammatory actions. There are multiple preparations on the market without definitive studies demonstrating superior efficacy of one preparation over another. Generally, they act by reversibly inhibiting the cyclooxygenase (COX) side of arachidonic acid metabolism, thereby blocking production of proinflammatory agents, such as prostaglandins and leukotrienes. Unfortunately, the beneficial effects of prostaglandins are also inhibited, such as the protective effects on the gastric mucosal lining, renal blood flow, and sodium balance. Their most common side effect is dyspepsia. Other potential side effects include gastrointestinal ulcers, hepatotoxicity, renal toxicity, and cardiac failure. These effects are dose related and more severe in older patients with prolonged elimination. Contraindications to NSAIDs include a history of gastrointestinal disease, hepatic disease, renal disease, or concurrent anticoagulation therapy. Patients on prolonged NSAID therapy should be monitored closely for side effects by the orthopaedist or primary care physician. Recent scrutiny of selective COX-2 inhibitors has brought their role in treatment of arthritis into question. These drugs were developed to treat pain and inflammation while reducing the risk for serious gastrointestinal side effects seen with other nonselective NSAIDs. Several questions remain about the safety advantage of COX-2 inhibitors including whether they actually lower the risk for serious gastrointestinal events and whether their potential gastrointestinal advantage is negated by an increased risk for thromboembolic complications. Clinical trials are currently under way to help answer these questions. Until the results of these trials are known, COX-2 inhibitors should be used with caution, especially in patients with established or increased risk for cardiovascular disease.

Corticosteroid injections are helpful in patients who have failed first-line anti-inflammatory therapy or who have contraindications to use of acetaminophen or NSAIDs. An intra-articular injection is a potent anti-inflammatory agent with minimal risk for systemic effects or complications, although diabetic patients should be counseled that their blood glucose level will likely rise after receiving the injection. Crystalline corticosteroids can induce a ­corticosteroid-crystal synovitis or poststeroid flare, but this is rare and usually self-limited. Injections provide variable relief, lasting from a few days to 6 months or longer, particularly in the absence of mechanical symptoms. Steroid injections should be limited to a maximum of three per year because they can cause articular cartilage softening and erosion. Complications include skin pigmentation changes and subcutaneous fat atrophy when delivered in the subcutaneous space. Contraindications include suspected septic joint and recent fracture or trauma.

Medical Management and Injections Acetaminophen

Glucosamine and Chondroitin Sulfate Glucosamine provides the building blocks for the chondral matrix production normally produced by chondrocytes from glucose metabolism. When taken orally as a salt, 87% of the dose is absorbed in the gut and primarily processed through renal excretion, with lesser amounts processed by the liver.43 The results of randomized, double-blinded,

Viscosupplementation Injectable hyaluronic acid, or “viscosupplementation,” is available as a series of three to five weekly injections for treatment of symptomatic osteoarthritis. The injections supplement the reduced concentrations of hyaluronic acid found in the joints of patients with osteoarthritis. Viscosupplementation provides improved “elastoviscosity,” enabling the synovial fluid to be more effective in absorbing joint loads and lubricating articular surfaces. Enhanced endogenous hyaluronic acid synthesis by synovial cells, proteoglycan synthesis by chondrocytes, anti-inflammatory effects, and analgesic effects on nociceptive pain receptors are additional benefits. Any appreciable effusion should be aspirated before injection. Local anesthetic should not be combined with the hyaluronic acid other than infiltration into the skin for local anesthesia. Complications include hypersensitivity to hyaluronic acid preparations and severe postinjection inflammation occurring in 1% of patients. Newer compounds made with recombinant methods appear to reduce this occurrence. Contraindications include patients with a known hypersensitivity to hyaluronic acid preparations and those with skin diseases or infections in the area of injection.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1792 DeLee & Drez’s� O

Cost is a consideration with viscosupplementation. Currently, a series of injections costs between $500 and $1000 for the medication and injection procedure fees.

Operative Treatment Options Arthroscopy Proinflammatory cytokines are released by degenerating articular cartilage and synovium in osteoarthritis. These cytokines cause chondrocytes to release lytic factors that cause breakdown of proteoglycans and type II collagen. One of the benefits of arthroscopic treatment is lavage of these factors out of the knee. Edelson and colleagues found that lavage alone had good or excellent results in 86% of their patients at 1 year and in 81% at 2 years using the Hospital for Special Surgery Scale.47 Jackson and Rouse compared the results of lavage alone versus lavage combined with débridement. 48 Of the 65 patients treated with lavage alone, 80% showed initial improvement, but only 45% maintained that improvement at final follow-up. Of the 137 patients treated with lavage plus débridement, 88% showed initial improvement with 68% maintaining improvement at final followup. Others like Gibson and associates found no statistically significant improvement with either method even in the short term.49 Because of these mixed results, the efficacy of arthroscopic treatment of osteoarthritis is controversial. Patients who get the most benefit from arthroscopic débridement are those who present with a history of mechanical symptoms, symptoms of short duration (<6 months), normal alignment, and only mild to moderate arthritic changes on radiographs. Literature suggests that arthroscopic treatment, when performed for appropriate indications, will provide relief in 50% to 70% of patients lasting from several months to several years. It is important to note that patients may have unrealistic expectations regarding arthroscopic treatment and should be counseled about the limited indications and potential results.

Osteotomy Proximal tibial osteotomy is a viable treatment option for young patients with unicompartmental knee arthritis. Multiple studies have demonstrated excellent pain relief and some improvement in function.50-55 However, these initial results deteriorate over time with progression of arthritis, requiring most patients to proceed to TKA. Matthews and coworkers found useful function was preserved in 86% of patients at 1 year, 64% at 3 years, 50% at 5 years, and 28% at 9 years.56 Insall reported that proximal tibial or distal femoral osteotomy permits unrestricted activity and good short-term results but does not predictably relieve pain over the long term.57-60 Patients had 97% good or excellent results at 2 years, compared with only 58% at 10 years postoperatively.59 One potential advantage of osteotomy over arthroplasty is that unrestricted athletic activity is permitted following osteotomy. This can be helpful for patients interested in continuing running or jumping sports after surgery. However, what activities patients actually perform is another matter. Nagel and associates found that after proximal tibial osteotomy, the patient’s level of activity reaches a plateau that is lower than the preoperative level and then

gradually decreases with time.61 Nagel also emphasized the complications associated with osteotomy, including nonunion, intra-articular fracture, peroneal nerve palsy, infection, vascular injuries, and prolonged recovery. He recommended against preoperative guarantees of increased function. At best, preoperative levels of activity will be maintained. The reported deterioration of the results of proximal tibial osteotomy increases the difficulty of selecting appropriate treatment for young active patients. There have been conflicting reports in the literature regarding outcomes of TKA in patients with previous proximal tibial osteotomy. Several studies claim inferior results,62-64 whereas others do not.65-69 Parvizi and associates found that the overall functional outcomes were inferior in knees that had previous proximal tibial osteotomies.70 They also found a very high rate of radiographic loosening in this patient population after total knee replacement. Risk factors for early failure included male gender, increased weight, young age at the time of TKA, coronal laxity, and preoperative limb malalignment. Currently, a tibial osteotomy is preferred for young, active patients because of its potential to allow unlimited strenuous activity, which is contraindicated after TKA. Therefore, if a patient wishes to continue to participate in sports involving running or jumping or to engage in a manual occupation that involves bending, lifting, or climbing, a proximal tibial osteotomy may be the procedure of choice. It is best suited for patients with true varus and valgus deformities in their 30s or 40s. The survivorship with arthroplasty makes it a better option in patients in their late 40s and 50s. Osteotomy and simultaneous ACL reconstruction is an option in patients with knee instability and varus angulation. Multiple studies have demonstrated good early results. Aqueskirchner and associates reported improvement in Lysholm score from 66 preoperatively to 81, 87, and 93 at 3, 6, and 12 months postoperatively, respectively.71 They found that the combined procedure facilitated early rehabilitation and return to activities of daily living and sports. Bonin and associates studied the combined procedure at 12-year follow-up.72 They found that only 17% of knees had progressed one arthritis grade, 47% returned to intensive sport, and another 37% returned to moderate sports. The mean difference in side-to-side anterior tibial translation was 3 mm. Stein and colleagues reported that patients with chronic instability secondary to ACL deficiency and unicompartmental arthritis will generally benefit from arthroscopic débridement, ACL reconstruction, knee osteotomy, or any combination thereof.73 However, they warned against lofty expectations stating that a return to competitive or high-level sports after these procedures is an unrealistic goal and should be discouraged.

Arthroplasty Unicompartmental Knee Arthroplasty Unicompartmental knee replacement is an alternative to tibial osteotomy or total knee replacement in selected patients. It has a well-established role in treatment of osteoarthritis limited to one compartment (Fig. 23J-2). Osteotomy may be preferred for young, more active, and

Knee 1793

A

B

Figure 23J-2   A, Radiograph demonstrating advanced arthritic changes in the medial compartment without degenerative changes in the lateral compartment. B, Lateral radiograph demonstrating proper sagittal alignment of an implanted unicompartmental knee arthroplasty.

overweight patients. However, because of the increased survivorship of current designs, unicompartmental knee arthroplasty is appropriate for younger patients than previously thought. Patient selection is critical, and the procedure requires technical precision for proper alignment (Fig. 23J-3). Laborers and impact athletes are not good

A

candidates for this procedure unless they are willing to modify their activity levels. Discouraging reports were initially published regarding the results of unicompartmental arthroplasty.74,75 Historically, inappropriate patient selection, suboptimal surgical technique, and inferior prosthetic design contributed

B

Figure 23J-3   A, Unicompartmental knee arthroplasty can be performed through a limited approach. Dotted lines represent the patella, tibial tubercle, joint line, and border of the medial femoral condyle. B, Intraoperative photograph showing final implants through the limited approach.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1794 DeLee & Drez’s� O

to these early failures. Correction of these factors has led to more recent reporting of improved results with studies showing greater than 90% survival at 10 years.76-81 During the 1980s, there was a significant decline in the number of unicompartmental knee arthroplasties performed in this country. In 1996 and 1997, only 2500 unicompartmental knee arthroplasties were performed in the United States, representing about 1% of all knee arthroplasties performed.82 With the aforementioned changes, the procedure has seen a drastic increase in popularity. In 2000 and 2001, 33,900 unicompartmental knee arthroplasties were performed, representing 6% of all knee arthroplasties performed. Unicompartmental knee arthroplasty has advantages over TKA in properly selected patients. Benefits include shorter hospital stays, fewer serious complications, improved walking ability, and lower cost.83,84 Patients also have a more normal gait, better quadriceps function, and better knee flexion than those who are managed by TKA.85,86 Furthermore, when a failed unicompartmental knee arthroplasty is converted to a TKA, results compare favorably with those of primary TKA.87,88 Improved patient selection is a crucial factor for both the increased success and popularity of unicompartmental knee arthroplasty. Currently, indications include noninflammatory unicompartmental arthritis, contained mature osteonecrosis confirmed by MRI, at least 90 degrees of knee flexion, intact ACL and PCL, a flexion contracture of less than 10 degrees, maximal varus or valgus of less than 20 degrees that can be passively corrected to within 5 degrees of normal with the knee in maximal extension, minimal or no pain or tenderness in the opposite compartment or patellofemoral articulation, and refractory symptoms to conservative care. Equally important are the contraindications, which include inflammatory arthritis, synovial chondromatosis, villonodular synovitis, ligamentous instability, tibiofemoral translation or shift more than 5 mm on standing anteroposterior radiographs, recurvatum of more than 5 degrees, hemophilia, proven previous infection, proximal or distal extra-articular deformity of more than 5 degrees, and obesity.89 The absence of pain or symptoms with activity or palpation in the patellofemoral or opposite compartment is probably more important than the presence of radiographic changes. Because of the high correlation of patellofemoral symptoms with obesity, caution should be undertaken when considering unicompartmental knee arthroplasty in these patients. Several recent Journal of Bone and Joint Surgery articles have demonstrated good survivorship at extended followup. Berger and coworkers investigated 49 unicompartmental knee arthroplasties in 38 patients with a minimum follow-up of 10 years.90 Based on the Hospital for Special Surgery (HSS) rating system, 80% had excellent results, 12% had good results, and 8% had a fair result. HSS scores improved from 55 preoperatively to 92 at final follow-up. Two knees with well-fixed components underwent revision to TKA, at 7 and 11 years, because of progression of patellofemoral arthritis. Radiographically, no components were loose, and there was no evidence of periprosthetic osteolysis. Survivorship analysis showed a rate of 98% success at 10 years and 95.7% at 13 years. Argenson and colleagues reported on 166 unicompartmental knee replacements in 147 patients with a mean

follow-up of 66 months.91 The average HSS score improved from 59 preoperatively to 96 at follow-up. Three knees were revised because of progression of arthritis, another two knees for poly exchange. The 10-year survivorship analysis showed a cumulative survivorship rate of 94%. Swienckowski and Pennington reported results for unicompartmental knee arthroplasty in patients 60 years of age or younger.89 Forty-six unicompartmental knee arthroplasties in 41 patients younger than 60 years revealed that HSS scores were excellent in 93% and good in 7% of patients. Three knees had been revised. Two asymptomatic patients had revision of a modular tibial component because of substantial radiographic evidence of poly wear. The third patient was converted to a total knee replacement because of continuing knee pain and progression of a tibial radiolucent line. At a mean follow-up of 11 years, there was a survivorship of 93.3%. As previously mentioned, there is little literature specifically addressing sports participation after unicompartmental knee replacement. Fisher and associates reported that 93% of their patients returned to their regular sporting activities following surgery.38 Despite the lack of specific data, it seems prudent to use a common-sense approach by suggesting these patients observe the same principles recommended for sports after TKA. Participation in lowimpact activities can be encouraged, whereas high-impact activities should be discouraged unless done occasionally on a recreational basis.

Total Knee Arthroplasty TKA is well established as the gold standard for relief of pain, correction of deformity, and restoration of function of the arthritic knee. The excellent patient satisfaction and clinical function achieved with TKA make it one of the most successful procedures currently available. Both posterior cruciate ligament (PCL)-retaining and PCL-­sacrificing/ substituting implants have been used without consensus for one over the other. Rand and Ilstrup demonstrated a 10-year survival rate of 91% for 9200 knee replacements when a cemented, PCL-retaining implant was used.14 Colizza and associates demonstrated an 11-year survival rate of 96% when a cemented, PCL-sacrificing/substituting implant was used.5 The rate of TKA continues to increase in the United States owing to a variety of factors, including increased longevity of the population, the success of the procedure, and the gradual expansion of TKA indications to include younger, more active patients. Our society’s changing philosophy has helped change the indications for total knee replacement. During the 1970s and 1980s, knee replacements were performed mostly for pain, disability, or deformity. The expectations were reduction in pain, limb realignment, and functional improvement. However, pain was the primary reason for knee replacement. Thus far in the 21st century, function has become the main goal of knee replacements. Patients are not satisfied with the reduced function that can accompany a stiff, painful, arthritic knee and want a total knee replacement to restore their function. Increasingly, this desire for improved function includes athletic activity. Secondary arthritis can also occur at a younger age, usually after injury resulting in either varus or valgus malalignment, intra-articular fracture, or ligamentous and meniscal

Knee 1795

   TABLE 23J-1  Clinical Results of Total Knee Arthroplasty in Young Patients Study

No. of Patients

Indications for Surgery

Age of Patients (yr)

Follow-up (yr)

Results (Knee Society Score)

Survivorship

Ranawat et al, 200595 Ranawat et al, 198996 Duffy et al, 199897 Gill et al, 199798 Diduch et al, 199728

54 knees (38 patients)

OA

57 (46-60)

5 (2-11)

95% good or excellent

98.2% at 5 yr

93 knees (62 patients)

OA/RA

48.7

6.1

70% excellent; 27.7% good

96% at 10 yr

74 knees (54 patients) 72 knees (52 patients) 103 knees (88 patients)

OA, RA, PTA OA, RA, AS OA

43 50.7 (30-55) 51 (22-55)

13 (10-17) 9.92 8 (3-18)

88% average 100% good or excellent 94% good or excellent [100% good or excellent Hospital for Special Surgery Score]

99% at 10 yr 96.5% at 10 yr 94% at 10 yr

AS, ankylosing spondylitis; OA, osteoarthritis; PTA, post-traumatic arthritis; RA, rheumatoid arthritis.

deficiency.92 Rangger and coworkers studied 284 patients and found radiographic increases in osteoarthritis after partial arthroscopic medial or lateral meniscectomy (38% and 24%, respectively) at an average follow-up of 53.5 months.93 Maletius and Messner investigated the long-term effects of partial meniscectomies in knees with severe chondral damage and suggested that the progression of arthrosis was exacerbated by meniscectomy.94 Earlier studies suggested that TKA in younger patients predisposed patients to loosening, premature implant wear, and osteolysis. Recent studies refute these previous studies (Table 23J-1). Ranawat and colleagues implanted 54 TKAs in 38 patients younger than 60 years using a cemented all-polyethylene tibial component.95 At an average follow-up of 5 years, there were two failures, one for infection and one for post-traumatic loosening. There was no radiographic evidence of loosening, progressive radiolucent lines, or osteolysis seen in the remaining knees. In a previous study, Ranawat reported results of 93 TKAs in patients younger than 55 years with a mean follow-up of 6.1 years. Good to excellent results were present in 97.7% based on the HSS rating system. The survivorship analysis showed a cumulative survivorship rate of 96% at 10 years.96 Duffy and colleagues investigated 74 TKAs in

Authors’ Preferred Method

for

54 patients younger than 55 years (average age, 43 years) with a minimum follow-up of 10 years (average follow-up, 13 years).97 Two implants were revised, one because of ligamentous laxity and one at 13 years because of aseptic loosening of the tibial component. The estimated survivorship at 10 years was 99%. Gill and coworkers reported on 68 TKAs in 50 patients younger than 55 years at an average follow-up of 9.92 years.98 Two knees in the same patient required revision for loose components. Good or excellent results were found in all knees, and survivorship was 96.5% at 10 years. Diduch and associates investigated 103 TKAs in 80 patients younger than 55 years at an average follow-up of 8 years.28 All knees were rated as good or excellent using the HSS scoring system. There were two revisions for infection, three revisions of patellar components, one polyethylene exchange for instability, and one tibial component exchange for wear of a carbon-fiber–reinforced polyethylene failure at 7 years as described earlier in this chapter. The survivorship estimate, with revision of the femoral or tibial component as the end point, was 94% at 18 years; with revisions of patellar components, the survival rate was 90% at 18 years; and when any operative intervention was included, the survival rate was 87%.

Total Knee Arthroplasty

Many different systems are available in the marketplace today; however, the surgical principles remain the same. Adequate exposure, preparation of bone surfaces, restoration of limb alignment, soft tissue balancing, and correct component position and rotation are crucial to ensure reproducible results. An anterior midline incision is used. If a previous incision is present, it can be incorporated into the incision unless large flaps would be created. We prefer to identify and elevate the bursal layer from the extensor mechanism so that the bursal layer can be closed as a separate layer. We have found this reduces subcutaneous hematoma formation and improves the appearance of wounds postoperatively. A standard medial parapatellar arthrotomy is then performed. The anterior horn of the medial meniscus is transected, and the medial capsule and deep medial collateral ligament (MCL)

are elevated subperiosteally from the proximal 3 to 4 cm of the medial tibia being careful to protect the superficial MCL. The infrapatellar fat pad is excised, and the patella is everted. If the patella cannot easily be everted, we first extend our quadriceps incision superiorly. At last resort, we use a quadriceps snip by continuing the incision from the apex of the arthrotomy into the fibers of the vastus lateralis at an angle of 45 degrees. The knee is then flexed, and the ACL is released from the femur. Once adequate exposure has been obtained, attention is turned to preparation of the femur and tibia. Either the femur or tibia may be cut first, depending on surgeon training and preference. Both intramedullary and extramedullary guides are available. We prefer an intramedullary guide for the femur and an extramedullary guide for the tibia, and to begin with the femur. A pilot hole is made with a punch followed Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1796 DeLee & Drez’s� O

Authors’ Preferred Method

for

Total Knee Arthroplasty—cont’d

Figure 23J-4   After completion of the tibial cut, long rods can be placed through the spacer blocks to verify proper alignment.

by a reamer, slightly medial to the midline of the trochlea just above the PCL origin. The canal is then suctioned to reduce risk for fat embolism when the guide is inserted. The guide is set to produce the desired valgus cut at the distal femur, typically 5 degrees for males and 6 degrees for females. The distal femoral cutting guide is attached to the intramedullary guide and pinned into place. An oscillating saw is then used to make the distal femoral cut. Using the sizing guide, the appropriate anteroposterior cutting guide is selected. The goal is to resect the same thickness of bone from the posterior condyles as will be replaced by our femoral component without notching the anterior femur. If we are between sizes, we select the smaller size as long as notching the anterior femoral cortex can be avoided. The epicondylar axis is then marked to serve as a guide for placement of the anteroposterior guide with regard to rotation. The axis is defined by the center of the sulcus of the medial epicondyle and the high point of the lateral epicondyle. This axis has been shown to be the most reliable landmark for determining accurate femoral rotation.99 Whiteside’s trochlear line is an additional line that can be drawn and should be perpendicular to the epicondylar axis if rotation is correct. Once the guide is correctly positioned, the oscillating saw is used to make the anterior and posterior femoral cuts. Completing the angled chamfer cuts at this point is optional but can make assessment and any needed changes in femoral size difficult. Attention is then turned to the tibial side. The tibial extramedullary guide is positioned to provide a cut perpendicular to the anatomic axis in the coronal plane. The resection height is then determined with the concept of taking at least 10 mm of bone off the “high side” and 2 mm of bone off the “low side.” In younger patients, slightly thicker (about 12 mm) polyethylene inserts may be advantageous to withstand increased loads of an active lifestyle over many years. The resection guide is then pinned into position, retractors are placed to protect the collateral ligaments, and the ­oscillating saw is used to make the tibial cut. After the bony

A

B

Figure 23J-5   Proper cuts and soft tissue balancing produce symmetrical flexion (A) and extension (B) gaps.

cut is made, the remnants of both menisci and cruciate ligaments should be removed, as well as posterior condylar osteophytes using an osteotome. Wide lamina spreaders can help facilitate these steps by providing distraction. Stability and symmetry of the flexion and extension gaps are assessed using spacer blocks. The orientation of the tibial cut can also be assessed using the long alignment rod (Fig. 32J-4). If the appropriate femoral and tibial cuts have been made and the overall alignment is acceptable, the surgeon can proceed to soft tissue balancing of the knee. The goal of soft tissue balancing is to release the tight structures to create a symmetrical rectangular space in both flexion and extension (Fig. 23J-5). Failure to restore soft tissue balance has potential to lead to instability or loss of motion.100 In a varus knee, if the medial structures are tight, an incremental release is performed to correct the imbalance. A blunt osteotome is used to elevate the distal, superficial MCL insertion along the medial border of the tibia (Fig. 23J-6). Additionally, the pes tendons may or may not be released as needed. The lamina spreader can be used to gradually expand the medial space. Next, reinsert the spacer block to evaluate the efficacy of the release. Often, once the extension space symmetry has been restored, the next-thicker spacer block is required. In the valgus knee, tight structures may include the iliotibial (IT) band, lateral collateral ligament (LCL), popliteus, and arcuate ligament–posterolateral capsular complex. If the lateral side is tight after achieving acceptable alignment, the lateral soft tissues are released in a stepwise fashion using an inside-out technique and gradual distraction with the lamina spreader. The surgeon (and patient) must be very aware of stretch injury to the peroneal nerve that can occur even with good technique that restores anatomic alignment. If the knee is tight in extension, multiple “pie crusting” incisions

Knee 1797

Authors’ Preferred Method

for

Total Knee Arthroplasty—cont’d

Medial collateral ligament Pes anserinus

A

B

Figure 23J-6   Release of medial soft tissues in a knee with fixed varus deformity. A, Subperiosteal release of the deep and superficial medial collateral ligament, capsule, and semimembranosus insertion about the proximal tibia. B, Complete release is accomplished by subperiosteal elevation of the distal pes and superficial medial collateral ligament insertions to the mid tibia. (Redrawn from Insall JN, Scott WN: Surgery of the Knee, 3rd ed. Philadelphia, WB Saunders, 2000.)

are made through the IT band and the capsule at the level of the tibial cut and proximal to the joint. In unusual cases, the IT band may have to be totally released from Gerdy’s tubercle. If the knee is primarily tight in flexion, pie crusting to elongate the popliteus and/or the posterolateral joint capsule is performed. If the knee remains tight in flexion and extension, the LCL can be cautiously lengthened through pie crusting. A complete release will necessitate a constrained knee design and should be avoided in high-demand patients because of increased risk for loosening. After the appropriate extension and flexion gap symmetry has been obtained, the femoral finishing cuts can be performed. Chamfer cuts and a box cut for a posteriorly ­stabilized component must be made. The guide for these cuts should be placed so that the component optimizes patellar tracking without creating overhang. Next, attention is turned to the final tibial preparation. The tibial rotation is set so that the center of the component is oriented in line with the medial to middle third of the tibial tubercle. Internal rotation should be avoided because it can cause patellar

FUTURE DIRECTIONS AND TECHNOLOGY Mobile-Bearing Prostheses Mobile-bearing prostheses potentially allow both articular conformity and freedom of rotation beyond what is seen in fixed-bearing prostheses. A mobile-bearing prosthesis

subluxation. After these steps have been completed, femoral and tibial trials are used to ensure that appropriate soft tissue balance has been achieved without flexion contracture or hyperextension. Finally, patellar preparation is undertaken by first measuring the patellar thickness at the high point. The goal is to re-create the patellar thickness or reduce the composite thickness by 1 to 2 mm. The patellar component is positioned slightly medially and superiorly to help prevent maltracking while maximally covering the surface. With all trials in place, a no-thumbs technique is then used to assess patellar tracking. If no technical errors can be identified and maltracking is present, we do not hesitate to perform a lateral release. After trialing to confirm optimal stability and motion with poly thickness, the surfaces are lavaged and dried, and the final components are cemented in place. The joint is then copiously irrigated, excess cement is removed, a deep drain may be placed, and the wound is closed in slight flexion. After skin closure, a light sterile dressing is used, facilitating early postoperative passive motion.

­ ermits rotation between the tibial baseplate and polyp ethylene insert. At the same time, articular conformity is maintained between the femoral component and the superior surface of the polyethylene insert. This construct theoretically maximizes articular conformity, thereby reducing contact stresses and wear on the superior surface of the polyethylene insert. Simultaneously, rotation can still occur on the undersurface of the polyethylene, which could lead to back-side damage. Mobile-bearing designs provide

�rthopaedic ����������� S �ports ������ � Medicine ������� 1798 DeLee & Drez’s� O

contact areas in the range of 1000 mm2, whereas modern condylar prostheses are in the range of 100 to 300 mm2. The increased contact area reduces contact stresses from levels that routinely exceed the yield stress of polyethylene to safer levels.101 Additionally, mobile-bearing designs may offer a solution to undersurface polyethylene wear. Undersurface wear has been recognized as a serious concern in fixed-bearing designs that have modular metal-backed tibial components.101,102 Although many different locking mechanisms exist in today’s market, undersurface motion is ubiquitous in modern modular fixed-bearing designs.103 This motion between the tibial baseplate and the polyethylene insert generates debris that may contribute to osteolysis. Questions remain about design characteristics for mobile-bearing designs, including position of the pivot point, whether stops are necessary to prevent bearing spinout, and whether both rotation and translation should occur at the undersurface. Many modern mobile-bearing designs are currently undergoing clinical investigation to help answer these concerns. To date, the published results have been favorable, with outcomes equal to fixed-­bearing designs.104-106 The theory is promising and may offer advantages for younger, active patients, including reduced polyethylene wear, better kinematics, and improved motion. Prospective studies are necessary to answer these questions.

High-Flexion Total Knee Arthroplasty Patient satisfaction after TKA depends not only on pain relief but also on restoration of function, including maximal flexion and range of motion.107-109 Increasing knee

A

flexion after TKA is an important issue for patients ­ ishing to maintain a high-flexion lifestyle, including w such activities as gardening, sports, or cultural activities seen in patients of Middle Eastern or Asian decent. Activities of daily living can be fulfilled with knee flexion of 105 to 110 degrees with a few exceptions. Climbing up and down stairs and sitting in a chair require 90 to 120 degrees of flexion in normal people. Squatting and sitting crosslegged require knee flexion of 110 to 130 degrees with demand up to 150 degrees for prayer activities seen in Asian cultures. Standard TKA designs result in edge loading beyond 135 degrees (Fig. 23J-7A),107,108,110 and range of motion after TKA using conventional implants has been reported to average about 110 degrees.111-113 These issues have led to a new generation of TKA designs that allow congruous contact in knee flexion beyond 145 degrees (see Fig. 23J-7B). The prominent design difference compared with conventional prostheses is an additional cut of 2 mm of bone from the posterior femoral condyle to increase the articulation curvature during deep flexion (Fig. 23J-8). Additionally, an anterior cut out of the tibial polyethylene insert is made to avoid patellar tendon impingement during deep flexion (Fig. 23J-9). The modified cam and post mechanism has an increased jump distance and avoids dislocation at deep flexion angles. With the specific designs and careful surgical technique, both high flexion angle and good stability can be achieved. Studies have demonstrated good early results of highflexion TKA designs. Bin and Nam found an average maximal knee flexion of 129.8 degrees in the high-flexion group, which was significantly higher than the 124.3 degrees seen in the conventional TKA group.114 No ­ differences

B

Figure 23J-7   A, Lateral radiograph demonstrating edge loading posteriorly at high flexion angles seen with conventional total knee arthroplasty (TKA) designs. B, Lateral radiograph demonstrating congruent contact posteriorly in a high flexion TKA.

Knee 1799

Modified posterior condyles

Modified cam

High flex Conventional

Figure 23J-8   The modified cam, posterior condyles, and increase in articulation curvature help accommodate increased flexion angles while preventing dislocation.

were seen in HSS scores, and there was no evidence of osteolysis or aseptic loosening. Huang and colleagues reported an average of 138 degrees of knee flexion in the high-flexion group, which was significantly higher than 126 degrees in the conventional group at average followup of 28 months.115 Eighty percent of patients were able to squat in the high-flexion group, compared with 32% in the conventional group. It is important to note that current literature is limited by short-term follow-up. Additional studies are necessary to evaluate the potential benefits of high-flexion designs.

Gender-Specific Total Knee Arthroplasty Physicians have known of differential anatomy of the distal femur in female patients for many years. The first to apply the differences in design of total knee implant design was John Insall. He created a customized supplement to his knee series that had an extra 5 mm of anteroposterior height for the same medial width to accommodate women’s knees. New gender-specific designs remove the dilemma of accepting the overhang of the femoral component in a female knee dictated by templating or size measurement,

Conventional articular surface

High flex articular surface Figure 23J-9   An anterior cutout in the polyethylene avoids impingement of the patellar tendon in deep flexion.

as well as removing risks associated with downsizing the femoral component, such as by anterior notching with possible femur fracture or over-resection posteriorly that would compromise knee function. The enhancements with the gender-specific design address three main differences with regard to the female knee: anterior femoral overhang, increased Q angle, and decreased anterior flange thickness. If anterior femoral overhang is observed with a standard implant, the surgeon can replace it with a gender-specific implant that does not compromise the lower part of the femoral component or the tibia, and thus kinematics are not affected. Second, the increased Q angle seen in most women is accommodated by 3 degrees of increased lateral angulation of the trochlear groove, thus enhancing patellar tracking and reducing need for lateral release. Finally, the thinner anterior flange addresses the fact that less anterior bone is resected from a female than a male femur. This decreases the risk for overstuffing the patellofemoral joint in females and may help to reduce ­ anterior femoral pain.

Minimally Invasive Total Knee Arthroplasty Minimally invasive surgery (MIS) for knee arthroplasty began in the late 1990s. Repicci established the MIS approach for unicondylar knee arthroplasty and popularized both limited surgical approaches and partial knee replacement.116,117 Surgeons then began to experiment with smaller incisions for TKA. More recently, MIS techniques have received more attention owing to patient demand, possibility of decreased hospital stay, and development of instrumentation and techniques specific to MIS surgery. Although reductions in hospital time and costs are motivating factors, one should not discount the patientdriven desires, including reduction in postoperative pain and quicker rehabilitation. Standard TKA may require an arduous recovery period for patients. Patients are often advised that it can take 6 months to 1 year for full functional improvement. By using a less invasive approach, surgeons aim to produce less operative morbidity and hence a faster recovery. Several different MIS exposures can be used, such as the mid vastus or quad-sparing approach. Central to all approaches are use of a small skin incision, usually between 8 and 10 cm, and minimization of trauma to the quadriceps tendon and the surrounding musculature. Use of intramedullary guides for the femur and extramedullary guides for the tibia are still employed, but all cuts are made from a medial to lateral direction, making the lateral cuts blind (Fig. 23J-10). Specific instrumentation and retractors are available to facilitate exposure and uniform cuts, but careful attention must be paid to control the blade when penetrating the posterior and lateral cortex to protect surrounding soft tissue structures (Fig. 23J-11). There is limited literature on MIS TKA, although the early studies show encouraging results by Bonutti and colleagues at 5-year follow-up,118-120 and with Tria and Coon at 2-year follow.121 Using MIS techniques, patients who underwent TKA showed early return of motion, less blood loss, and shorter hospital stays when compared

�rthopaedic ����������� S �ports ������ � Medicine ������� 1800 DeLee & Drez’s� O

A

B

Figure 23J-10   A, Intraoperative photography showing the distal femoral cutting guide placed on the medial condyle with a “blind” lateral femoral cut. B, Distal femur after resection.

with historical controls using conventional techniques. Concerns remain regarding the learning curve needed to master the technique as well as consistently reproducing correct alignment seen in conventional TKA.122 Prospective randomized studies with longer follow-up are needed to help define the utility of MIS TKA in today’s market.

Bearing Surfaces in Total Knee Arthroplasty Polyethylene and bearing surface wear continues to be a concern in knee arthroplasty. Better understanding of the factors associated with polyethylene wear has led to changes in manufacturing, postprocessing sterilization, and shelf life. Manufacture of ultrahigh-molecular-weight polyethylene

Figure 23J-11   Intraoperative photograph with tibial resection guide positioned medially. It is important to control the blade when penetrating the posterior and lateral cortex to protect surrounding soft tissue structures.

(UHMWPE) by direct compression molding has consistently demonstrated better wear than ram bar extrusion or compression molding into bars with secondary machining into the desired product. The sterilization process has been implicated as a factor for causing high wear rates and rapid polyethylene failure. Polyethylene components sterilized through radiation in air and stored in an oxygen environment generate oxidized polyethylene that, under repetitive cyclic loading, can excessively wear, delaminate, and crack, leading to polyethylene failure. If polyethylene is sterilized in an environment without oxygen, cross-linking is favored. Cross-linked polyethylene has improved resistance to adhesive and abrasive wear and improves wear rates in simulator data. The disadvantage of cross-linking is that it diminishes the mechanical properties of the implant. Shelf life should be minimized for polyethylene inserts. Irradiation of the polyethylene generates free radicals. Free radicals are known to survive within polyethylene for as long as 2 to 3 years. The higher the radiation dose, the more free radicals produced. Therefore, a long shelf life can adversely affect polyethylene performance by means of shelf oxidation. The worst case scenario is irradiated polyethylene packaged in air and allowed to sit for years on the shelf. Advances in biomaterials that reduce polyethylene wear may prove helpful in the long term. However, the long-term results are still in question. In the past, presumed advancements such as carbon-fiber–reinforced polyethylene and Hylamer-M did not prove beneficial in the clinical setting. Cross-linked polyethylene has been shown to reduce wear rate and subsequent osteolysis in total hip arthroplasty. In vitro studies have shown decreased wear rates in TKA as well. Definitive studies in vivo have not yet been published. Although cross-linked polyethylene may offer better wear characteristics, in the younger, more active patient there is a concern that decreased fatigue strength could result in early polyethylene failure. Oxidized metal implants also show promise for wear reduction in TKA. Wear stimulators have demonstrated decreased wear rates with these implants, but results have not yet been confirmed in vivo, and long-term results are currently unknown.

Knee 1801

C

r i t i c a l

P

o i n t s

l Arthroplasty (total or unicondylar) is the most effective way to improve pain and function in the arthritic knee. l Arthroplasty should not be considered until the patient, at any age, is ready to give up impact sports that involve running or jumping. l Sporting activity after knee arthroplasty should be limited to low-impact activities or lateral movement sports. l No study has demonstrated that recreational level, lateral movement sports (tennis, skiing, golf) are detrimental to the survival of knee arthroplasties. l Patellofemoral pain and symptoms are more important than radiographic changes in deciding between total and unicondylar knee arthroplasty. l Osteotomy is preferred over arthroplasty for younger (<45 years old), heavier, high-demand patients with unicompartmental disease due to deformity. l Arthroscopic débridement is most helpful for younger patients with mechanical symptoms of catching and locking, effusions, minimal deformity, and early degenerative changes. l For patients not ready for surgery, options include unloader bracing, NSAIDs, weight loss, and physical therapy (especially to correct stiffness and flexion-­contractures).

S U G G E S T E D

R E A D I N G S

Berger RA, Meneghini RM, Jacobs JJ, et al: Results of unicompartmental knee arthroplasty at a minimum of ten years follow-up. J Bone Joint Surg Am 87(5): 999-1006, 2005. Cole BJ, Harner CD: Degenerative arthritis of the knee in active patients: Evaluation and management. J Am Acad Orthop Surg 7(6):389-402, 1999. Diduch DR, Insall JN, Scott WL, et al: Total knee replacement in young, active patients: Long-term follow-up and functional outcome. J Bone Joint Surg Am 79(4):575-582, 1997. Duffy GP, Trousdale RT, Stuart MJ: Total knee arthroplasty in patients 55 years old or younger: 10- to 17-year results. Clin Orthop 356:22-27, 1998. Gill GS, Chan KC, Mills DM: Five to 18 year follow-up study of cemented total knee arthroplasty for patients 55 years or older or younger. J Arthroplasty 12: 49-54, 1997. Healy WL, Iorio R, Lemos MJ: Athletic activity after joint replacement. Am J Sports Med 29(3):377-388, 2001. Kuster MS: Exercise recommendations after total joint replacement: A review of the current literature and proposal of scientifically based guidelines. Sports Med 32(7):433-445, 2002. Nagle A, Insall NJ, Scuderi GR: Proximal tibial osteotomy: A subjective outcome study. J Bone Joint Surg Am 78:1353-1358, 1996. Seyler TM, Mont MA, Ragland PS, et al: Sports activity after total hip and knee arthroplasty: Specific recommendations concerning tennis. Sports Med 36(7): 571-583, 2006. Swienckowski JJ, Pennington DW: Unicompartmental knee arthroplasty in patients sixty years of age or younger. J Bone Joint Surg Am 86(Suppl 1, Pt 2):131-142, 2004.

R efere n ces Please see www.expertconsult.com

S e c t i o n

K

High Tibial Osteotomy in the Anterior Cruciate Ligament–Deficient Knee with Varus Angulation* Frank R. Noyes and Sue D. Barber-Westin

High tibial osteotomy (HTO) has gained wide acceptance as a treatment option for patients with unicompartmental knee osteoarthritis and lower limb varus malalignment. The rationale for this procedure stems from the hypothesis that a varus deformity produces abnormal loads on the medial tibiofemoral compartment, leading to articular cartilage damage. In patients with unicompartmental arthrosis, the procedure corrects the mechanical abnormality and redistributes the loads onto the lateral tibiofemoral joint. HTO appears to be most beneficial when it is performed early in the course of the arthrosis process in younger ­individuals. An added complexity in the varus angulated knee with medial compartment arthrosis is anterior cruciate ­ligament (ACL) deficiency. Patients who have these combined *Research for this chapter was funded by Cincinnati Sportsmedicine ­Research and Education Foundation, the Noyes Knee Center, and the ­Deaconess Foundation.

abnormalities often experience pain, swelling, giving way, and functional limitations that result in a truly disabling condition. There may also be an associated deficiency of the posterolateral structures, including the fibular ­collateral ligament (FCL), popliteus muscle–tendon–ligament unit (PMTL), popliteofibular ligament (PFL), and posterolateral capsule, that adds to the varus angulation and clinical symptoms.

BIOMECHANICS AND CLASSIFICATION The terms primary-varus, double-varus, and triple-varus knee were devised to classify varus-aligned knees with associated ligament deficiencies (Table 23K-1).1 This classification system is based on the underlying tibiofemoral osseous alignment and the additional effect of separation

�rthopaedic ����������� S �ports ������ � Medicine ������� 1802 DeLee & Drez’s� O

   TABLE 23K-1  Causes of Varus Angulation in the Anterior Cruciate Ligament–Deficient Knee Tibiofemoral Alignment or Geometry

Knee Motion Limits

Knee Joint Position

Ligament Deficiency

Comments

Physiologic tibiofemoral varus alignment

NA

NA

NA

Narrowing or loss of medial joint cartilage

� Varus or adduction rotation

� Separation of medial tibiofemoral compartment

Pseudo laxity or slackness of medial ligament structures

Medial displacement of weight-bearing tibiofemoral line Effect on varus alignment more pronounced when preexisting physiologic varus alignment present

�� Varus or adduction rotation Often coupled with lateral tibial translation for secondary support, intercondylar eminence against lateral femoral condyle

Separation of lateral tibiofemoral joint on standing Varus thrust on stance phase due to lateral condylar lift-off Lateral tibiofemoral joint space compressive forces are insufficient; tension develops in lateral soft tissues

FCL, lateral capsule, iliotibial band (femorotibial portion) Amount of joint opening depends on slackness of lateral soft tissue restraints Absence of ACL secondary restraint to varus angulation

Weight-bearing tibiofemoral line shifts far enough medially to produce separation of the lateral tibiofemoral joint during walking, sports activities Under states of maximal muscle contraction (quadriceps, biceps femoris), sufficient compressive forces may exist to prevent lateral condylar lift-off.

��� Varus or adduction rotation Varus recurvatum in extension — Increased external tibial rotation — Increased hyperextension — Increased external tibial rotation in flexion

� Separation lateral compartment plus hyperextension produces varus recurvatum on standing, walking Varus recurvatum thrust if quadriceps and ankle plantar flexors do not prevent knee hyperextension Posterior subluxation of lateral plateau with external tibial rotation

Above, plus PMTL, PL capsule Knee hyperextension increases with associated damage to ACL, PCL (partial to complete)

Gait training required to teach patient not to walk with varus recurvatum thrust, maintaining 5 degrees of knee flexion on initial weight-bearing Knee hyperextension with physiologic slackness to ACL and PCL may be present without actual injury to cruciate ligaments allowing varus recurvatum.

Primary Varus

Double Varus

Added FCL or soft tissue deficiency

Triple Varus

Added deficiency of all the posterolateral structures (FCL, PMTL, PL capsule)

ACL, anterior cruciate ligament; FCL, fibular collateral ligament; PCL, posterior cruciate ligament, PL, posterolateral; PMTL, popliteal muscle–tendon–ligament unit. Modified from Noyes FR, Simon R: The role of high tibial osteotomy in the anterior cruciate ligament-deficient knee with varus alignment. In DeLee JC, Drez D (eds): Orthopaedic Sports Medicine Principles and Practice. Philadelphia, WB Saunders, 1994, pp 1401-1443.

of the lateral tibiofemoral compartment (due to deficiency of the posterolateral structures) on the overall varus lower limb alignment. A bilateral physiologic varus tibiofemoral alignment is almost always present in patients with a varus-angulated knee. However, a normal tibiofemoral alignment may convert to a varus malalignment following medial meniscectomy and resultant articular cartilage deterioration. For example, a patient with a physiologic varus alignment of 3 degrees (mechanical axis) who has a loss of 3 mm of the medial articular cartilage may develop an overall 6-degree varus tibiofemoral alignment. The term primary varus refers to both the physiologic tibiofemoral angulation and an additional increase in angulation due to altered geometry (narrowing) of the medial tibiofemoral joint (Fig. 23K-1). The tibiofemoral weightbearing line (WBL) shifts further into the medial tibiofemoral compartment as joint narrowing progresses and the lateral compartment is unloaded.

As the WBL shifts into the medial compartment, increased tensile forces occur in the posterolateral structures. Corresponding separation of the lateral tibiofemoral compartment occurs during standing, walking, and running activities (lateral condylar liftoff).2,3 The term double-varus knee refers to lower limb varus malalignment resulting from two factors: tibiofemoral osseous and geometric alignment, and separation of the lateral tibiofemoral compartment from deficiency of the FCL and posterolateral structures. There are a combination of active and passive restraints that resist separation of the lateral tibiofemoral compartment under dynamic loading conditions.4,5 The ­quadriceps, biceps femoris, and gastrocnemius muscles and iliotibial band act in a dynamic manner to resist adduction moments on the knee joint during gait and, along with weight-­bearing loads, resist lateral tibiofemoral separation. If these muscle forces do not provide a functional restraint due to excessive lateral tensile forces, separation of the ­lateral ­tibiofemoral joint occurs.

Knee 1803

WBL=33% With or without medial compartment narrowing Lateral compartment opening

WBL=20%

WBL=5%

Lateral compartment opening

Hyperextension and external tibial rotation

Primary varus • tibiofemoral geometry

Double varus • tibiofemoral geometry • separation of lateral compartment

Triple varus • tibiofemoral geometry • separation of lateral compartment • varus recurvatum

Figure 23K-1   Schematic illustration of primary, double, and triple varus knee angulation. WBL, weight-bearing line. (Redrawn from Noyes FR, Simon R: The role of high tibial osteotomy in the anterior cruciate ligament-deficient knee with varus alignment. In DeLee JC, Drez D [eds]: Orthopaedic Sports Medicine Principles and Practice. Philadelphia, WB Saunders, 1994, pp. 1401-1443.)

The FCL normally allows a few millimeters of separation of the tibiofemoral compartment, and it is appreciated that pathologic stretching (interstitial injury) may occur to this ligament in chronic varus-angulated knees. Under these circumstances, a transfer of all of the weight-­bearing loads to the medial compartment occurs, which can be especially deleterious if articular cartilage damage is already present. The patient’s symptoms frequently increase, with pain in both the medial compartment and lateral aspect of the knee joint (due to excessive medial compressive and lateral soft tissue tensile forces, respectively). Injury to the posterolateral structures produces a varus recurvatum position of the limb.6 The term triple-varus knee indicates lower limb varus malalignment resulting from three factors: tibiofemoral varus osseous malalignment, increased lateral tibiofemoral compartment separation, and varus recurvatum in extension. The varus recurvatum results from abnormal external tibial rotation and knee hyperextension, reflective of the deficiency of all of the posterolateral structures. Because of the increase in lateral compartment opening, the WBL shifts further medially, as shown in Figure 23K-1.

EVALUATION Clinical Presentation and History Symptoms of pain, swelling, and recurrent giving way related to activity are well-known sequela of the chronic ACL-deficient knee.7 What is unique to the varus­angulated ACL-deficient knee is that there may be two or three different knee subluxations (positions) that produce the giving-way symptoms. In the early stages, the patient may relate a subjective sensation of activity-related, partial ­giving-way episodes. It may be difficult to determine whether the symptoms of instability are due to anterior tibial subluxation, lateral tibiofemoral compartment separation on walking (varus thrust), abnormal knee hyperextension and varus recurvatum, posterior subluxation of the lateral tibial plateau (particularly with knee flexion), or a combination of these abnormalities. Patients may undergo an ACL reconstruction to reduce the anterior tibial subluxation, but after surgery state that the knee joint still feels unstable. At this point, lateral tibiofemoral compartment separation and a varus thrust at the knee joint during

�rthopaedic ����������� S �ports ������ � Medicine ������� 1804 DeLee & Drez’s� O

gait may be ­recognized. The varus thrust may be minimal or barely noticeable, but it is often accentuated during running, turning, or twisting motions. Later, as additional injury or stretching of the posterolateral soft tissue restraints occur, the varus thrust with normal walking may be noticed by the patient’s friends or family; indeed, the patient may not appreciate the abnormality until it is pointed out by others. A varus thrust indicates that medial unicondylar weightbearing is occurring along with separation of the lateral tibiofemoral compartment. Many normal individuals with a mild underlying physiologic tibiofemoral varus alignment will have a slight varus thrust during running. It is the increased magnitude and duration of the varus thrust, particularly during walking and athletic activities, that produces the symptomatic state. Complaints of medial joint line pain may result from the concentration of forces to the medial tibiofemoral compartment and presence of early joint arthrosis. There may also be complaints of pain to the posterolateral aspect of the joint due to excessive lateral ligamentous tensile forces. Hughston and Jacobson reported in 1985 the results of surgical reconstruction for chronic posterolateral rotatory instability in 95 patients.6 Preoperatively, the patients complained of hyperextension-related instability; difficulty with climbing stairs and with twisting, pivoting, and cutting activities; and medial joint line pain. The authors stated that although the initial injury caused only slight symptoms and patients were able to continue competitive athletics, with time the injury became a “severe and disabling chronic instability.” With the onset of a varus recurvatum position on standing, additional gait abnormalities may be noted. By this time, the patient often notices that the knee has a strong tendency to snap back into a hyperextended position during walking and may voluntarily contract the quadriceps muscle during the stance phase to maintain the knee in a few degrees of flexion to prevent hyperextension. Still, the patient frequently states that with any jarring motion to the limb (such as turning or twisting or stepping off a curb), the knee may go into an uncontrollable hyperextended position that is painful and may result in a loss of balance and giving way. The patient may note that in maintaining the partially flexed knee position during walking, it is more comfortable to also maintain the lower leg and foot in an externally rotated position. This position produces a position of posterior subluxation of the lateral tibial plateau in reference to the femoral condyle. If the patient allows internal tibial rotation to occur during walking, a marked increase in the varus thrusting motion may occur, as well as anterior subluxation of the lateral tibial plateau. The patient commonly states that the knee appears to have suffered a complete loss of rotatory stability. This is due to a tendency of both anterior and posterior subluxation of the lateral tibial plateau to occur, as well as an absence of a stable neutral position of the joint. If anterior knee pain or patellofemoral arthrosis is present, the patient may not wish to contract the quadriceps muscle to maintain the few degrees of knee flexion during the stance phase. This results in a hyperextended knee position and a markedly abnormal gait. In this state, the quadriceps muscle undergoes marked atrophy because the patient

is literally not using this muscle during most gait activities. It is exceedingly important to correct this gait abnormality before any consideration is given to surgical reconstruction of the posterolateral structures, as described previously.8 If this gait abnormality persists postoperatively, the likelihood exists that large tensile forces will eventually cause stretching and failure of reconstructed ligamentous structures because the knee will reassume the abnormal varus recurvatum position. It is well known that patient complaints of medial joint line pain may not correlate to the amount of medial compartment arthrosis that is present.7,9,10 In the early stages of medial compartment arthrosis, the patient usually complains of medial pain with sports activities but rarely with activities of daily living. When medial joint line pain occurs with daily activities, there is a strong probability that extensive articular cartilage damage exists, with exposed subchondral bone. Loss of the medial meniscus is a major risk factor for arthritic progression of the medial compartment.11-14 We particularly caution our athletically active patients with physiologic varus-angulated ACL-deficient knees to completely avoid any activities that may result in a giving-way injury that may damage the medial meniscus, and to undergo ACL reconstruction to protect a functional medial meniscus. Further, we advocate repair of complex meniscus tears extending into the avascular zone to avoid removal and, in effect, subtotal meniscectomy.15 If pain from medial compartment arthrosis is of moderate intensity preoperatively, there is a strong likelihood that this symptom will continue after HTO and ACL reconstruction, particularly in patients who engage in athletic activities. Therefore, the patient should not expect that an HTO will completely eliminate pain, although pain may temporarily diminish after the procedure. The symptom of swelling with activity carries, in our experience, a poor prognosis. We previously reported that swelling with any athletic activity in the chronic ACL-deficient knee correlated with time elapsed since the original injury and the presence of radiographic osteoarthritic changes.7 Again, this symptom may be present after any operative procedure and may limit the return to athletic activities.

Physical Examination and Testing A comprehensive physical examination of the knee joint is required (Fig. 23K-2). We pay particular attention to (1) patellofemoral abnormalities, including extensor mechanism malalignment accentuated by increased external tibial rotation and posterolateral tibial subluxation; (2) medial tibiofemoral crepitus on varus loading, indicating early articular cartilage damage before radiographic changes; (3) tenderness and inflammation of the lateral soft tissues; (4) gait abnormalities during walking and jogging; and (5) motion limits and subluxations of the affected knee compared with the opposite knee.7 The diagnostic tests to detect the abnormalities that produce a varus angulation in the frontal plane are shown in Table 23K-2. FCL insufficiency is determined with the varus stress test performed at 0 and 30 degrees of knee flexion. The patient’s thigh rests against the examining table, and the examiner places his finger and thumb on the medial and lateral tibiofemoral compartments. The ­ examiner

Knee 1805 HTO Candidate?

Varus knee with ACL posterolateral injury

Varus knee with medial arthrosis

Varus knee: future meniscus transplant, OAT of Carticel procedure

Examination

Lateral radiograph measure tibial slope

Full-length double stance radiograph

If abnormal, consider correction

Varus thrust with walking

Tibiofemoral varus?

Lateral joint opening stress test (30°) External tibial rotation test (30°, 90°) No

Varus recurvatum with standing or walking

Ligament reconstruction if ACL, posterolateral injury

Yes Possible lateral stress views Lateral joint opening on affected side? Yes

Final coronal wedge angle: subtract 1° for 1 mm

No

Mark WBL 62% width tibial plateau if medial arthrosis 50% width for neutral aligment Draw line from CFH to 62% tibial plateau

Anteromedial osteotomy gap 1/2 the posteromedial osteomy gap so that tibial slope is maintained

Draw line from center of tibio-talar joint to 62% coordinate tibial plateau

Calculate the coronal correction wedge

Cut-out radiograph

Measure actual coronal correction wedge

Repeat No

Preoperative coronal correction height should equal wedge gap at posteromedial aspect of osteotomy

Correction wedge values for 1st and 2nd methods equal?

Measure wedge effect on limb length, patellar height

WBL position check using image, radiograph, and computer navigation

Yes

Yes

Select correct wedge plate height based on location anterior to MCL

Bone graft + ORIF opening wedge site

Rehabilitation and osteotomy healing

Medial opening wedge HTO Lateral closing wedge HTO

HTO surgery with intraoperative WBL position check using image, radiograph, and computer navigation

Figure 23K-2   The algorithm demonstrates a step-wise approach to high tibial osteotomy (HTO). Tibial width is 62% for valgus overcorrection if medial arthrosis is present or 50% for neutral alignment. ACL, anterior cruciate ligament; CFH, center femoral head; MCL, medial collateral ligament; OAT, osteochondral autograft transfer; ORIF, open reduction internal fixation; WBL, weight-bearing line. (Redrawn from Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33:378-387, 2005.)

e­ stimates the millimeters of joint opening from the initial closed contact position of each tibiofemoral compartment to the maximal opened position. The motion involves a constrained varus-valgus rotation, avoiding internal or external tibial rotation, to open the respective tibiofemoral compartment from its initial closed position.16 Note that it is incorrect to measure only the degrees of varus or valgus rotation. For example, the degrees of varus malalignment

may be due to both narrowing of the medial tibiofemoral compartment and opening of the lateral tibiofemoral compartment. Rather, the examination is based on the millimeters of increase of the specific tibiofemoral compartment compared with the opposite normal knee. Frequently, there may be an increase in medial joint opening compared with the opposite knee that represents a pseudo laxity. The increased medial joint opening is due to medial

�rthopaedic ����������� S �ports ������ � Medicine ������� 1806 DeLee & Drez’s� O

The tests for abnormal increases in external tibial rotation are based on the biomechanical data shown in Table 23K-3. Sectioning of the posterolateral structures, ­including the FCL, produced a mean of 13 degrees of increased external tibial rotation at 30 degrees of knee flexion.17 The limits of tibial rotation are also dependent on the flexion angle and the ligaments involved. The tibiofemoral rotation test is used to diagnose posterior tibial subluxations in a qualitative manner.16 The tibia is first positioned at 30 degrees of knee flexion and neutral tibial rotation (Fig. 23K-3). The position of the anterior aspect of the medial and lateral tibial plateau is determined by palpation in reference to the femoral condyles. The tibia is rotated externally to a maximal position, and the positions of the medial and lateral tibial plateaus are again palpated. The examiner determines whether there is any increase in external tibial rotation, paying particular attention to the tibial tubercle compared with the opposite normal knee. The reason for the increase in external tibial rotation is determined to be either a posterior subluxation of the lateral tibial plateau (indicating injury to the FCL and other posterolateral structures) or anterior subluxation of the medial tibial plateau (indicating damage to the medial collateral ligament [MCL] and posteromedial structures). In some knees, both subluxations are present. The test is repeated at 90 degrees of knee flexion. The test may also be repeated beginning in neutral tibial rotation and progressing to internal tibial rotation. Additional information during the tibiofemoral rotation tests may be gained by closely observing the location of the axis for internal and external tibial rotation and comparing this location to the normal knee. With posterior subluxation of the lateral tibial plateau during external tibial rotation, the examiner may detect a shift in the axis of tibial rotation of the medial tibiofemoral compartment. Alternatively, with an anterior subluxation of the medial tibial plateau, the center of tibial rotation frequently shifts to the lateral tibiofemoral compartment as the maximal external tibial rotation position is reached. The advantages of the tibiofemoral

   TABLE 23K-2  Diagnosis of Abnormalities Abnormality

Diagnostic Test

Tibiofemoral alignment

Full-length standing radiograph: supine or double support (closure of lateral tibiofemoral joint required) Change in millimeters from opposite side on weight-bearing or stress radiograph Increase in lateral joint opening at 30 degrees of flexion Increase in lateral joint opening at 30 degrees of flexion; greater than when FCL is only structure deficient Increase in external tibial rotation at 20 degrees of flexion Varus recurvatum in extension Standing radiograph shows increased joint width compared with opposite side Amount of increase on stress radiograph compared with opposite side Defined by degrees of hyperextension and varus angulation Elicited on supine varus recurvatum test Standing tests with patient assuming maximal knee hyperextension position provides greatest subluxation

Narrowing of medial tibiofemoral joint FCL insufficiency FCL, PMTL, PL capsule insufficiency

Lateral tibiofemoral joint separation Varus recurvatum

FCL, fibular collateral ligament; PL, posterolateral; PMTL, popliteal muscle– tendon–ligament unit. Modified from Noyes FR, Simon R: The role of high tibial osteotomy in the anterior cruciate ligament-deficient knee with varus alignment. In DeLee JC, Drez D (eds): Orthopaedic Sports Medicine Principles and Practice. Philadelphia, WB Saunders, 1994, pp 1401-1443.

t­ ibiofemoral joint narrowing and collapse. With a varus stress, the medial joint opening returns the limb to a more normal alignment, and no medial ligamentous damage exists. The primary and secondary restraints that resist lateral joint opening have been previously published.4,17 The amount of varus rotation that occurs depends on the flexion angle at which the test is conducted and the involvement of the secondary restraints (Table 23K-3).

   TABLE 23K-3  Increased Motion Compared with Normal Motion that Occurred When Only the Indicated Structures   Were Sectioned*

Angle of Flexion (degrees) 0

15

30

60

90

0.4 ± 0.5 8.5 ±������ ������� 2.6 10.5 ±������� �������� 4.0

0.2 ±������ ������� 0.5 11.8 ±������ ������� 3.0 14.2 ±������� �������� 3.7

0.2 ±������ ������� 0.6 13.0 ±������ ������� 2.3 18.0 ±������� �������� 3.8

0.4 ±������ ������� 0.9 5.2 ±������ ������� 9.0 21.0 ±������� �������� 3.1

0.6 ±������ ������� 1.2 5.3 ±������ ������� 2.6 20.9 ±������� �������� 2.8

2.5 ±������ ������� 0.4 0.4 ±������ ������� 0.6 6.4 ±������ ������� 2.3 8.1 ±������� �������� 2.5

4.5 ±������ ������� 0.4 0.4 ±������ ������� 0.6 7.9 ±������ ������� 2.0 11.3 ±������� �������� 3.0

5.7 ±������ ������� 0.2 0.4 ±������ ������� 0.6 9.0 ±������ ������� 2.0 14.2 ±������� �������� 3.3

5.5 ±������ ������� 0.6 0.8 ±������ ������� 0.6 8.3 ±������ ������� 3.7 18.9 ±������� �������� 3.4

4.3 ±������ ������� 0.9 1.4 ±������ ������� 0.6 6.8 ±������ ������� 4.5 21.2 ±������� �������� 3.0

External Rotation Limit (degrees)

PCL cut PLS cut All cut Varus Angulation Limit (degrees)

FCL cut PCL cut PLS cut All cut

*All values are given as mean and standard deviation. FCL, fibular collateral ligament; PCL, posterior cruciate ligament; PLS, posterolateral structures (seven specimens). From Grood ES, Stowers SF, Noyes FR: Limits of movement in the human knee: Effect of sectioning the posterior cruciate ligament and posterolateral structures. J Bone Joint Surg Am 70:88-97, 1988.

Knee 1807 Figure 23K-3   The tibiofemoral rotation test. The position of the medial and lateral tibial plateaus is assessed at the starting position (neutral tibial rotation) with the knee flexed to 90 degrees (A) and at the final position with the tibia externally rotated (B). The examiner palpates the position of the medial and lateral tibial plateau, which is compared with the normal knee to assess whether a subluxation (anterior or posterior) of the medial or lateral tibial plateau is present. The axis of tibial rotation is observed in the involved knee and compared with the normal knee to detect a shift in the medial or lateral tibiofemoral compartment during tibial rotation. The test is also performed at 30 degrees of flexion. (Modified from Noyes FR, Stowers SF, Grood ES, et al: Posterior subluxations of the medial and lateral tibiofemoral compartments: An in vivo ligament sectioning study in cadaveric knees. Am J Sports Med 21:407, 1993.)

B

A

rotation test over the traditional posterolateral drawer test are (1) the knee can be positioned at different flexion positions; (2) the tibia is less constrained because the foot is not held fixed to the examining table; and (3) the axis of tibial rotation can be observed as the tibia is rotated externally and internally. We prefer to perform the tibiofemoral rotation test in a supine position and not in a prone position because it is difficult in the prone position to palpate the medial and Intact

lateral tibiofemoral position (anteroposterior) required to diagnose the abnormal compartment subluxations. The mean values in millimeters of posterior tibial subluxation found in selective ligament cutting studies of seven knees are shown in Figure 23K-4. We also perform a varus recurvatum test in both the supine and standing positions18 and the reversed pivot shift test.19 These tests are more qualitative and difficult to measure in objective

PLS/LCL Cut

All Cut

Mean displacements Flexion angle = 30 3.9

–1.8

–7.5 mm

3.4

–6.0

18.2°

–15.5 mm

–3.7

–14.5

31.2°

–25.3 mm

36.2°

Mean displacements Flexion angle = 90 0.3

–5.2

–10.7 mm

0.7

–6.4

–13.4 mm

–12.0

22.8°

17.4°

Starting position

–23.1

–34.2 mm

37.9°

Final position

Figure 23K-4   The mean values of external tibial rotation and the millimeters of anteroposterior translation for medial, central, and lateral tibial reference points are shown for 30 and 90 degrees of knee flexion under the predetermined loads applied to the knee joint. The neutral position (0 degrees of rotation, 0 mm anteroposterior translation) was determined in the intact knee by allowing the tibia to hang vertically by its own weight. PLS, posterolateral structures; LCL, lateral collateral ligament. (Redrawn from Noyes FR, Stowers SF, Grood ES, et al: Posterior subluxations of the medial and lateral tibiofemoral compartments: An in vitro ligament sectioning study in cadaveric knees. Am J Sports Med 21:407-414, 1993.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1808 DeLee & Drez’s� O Abnormal Lateral Joint Opening (30°)

Normal Lateral Joint Opening (30°)

10 mm 8 mm 12 mm

A

B

2 mm 4 mm 6 mm

Varus Load

Figure 23K-5   A, Arthroscopic demonstration of the lateral joint opening “gap test.” The amount of lateral joint opening is measured with the knee at 25 degrees of flexion. B, Knees with insufficiency of the posterolateral structures will demonstrate 12 mm of joint opening at the periphery of the lateral tibiofemoral compartment, 10 mm of opening at the midportion of the compartment, and 8 mm at the innermost medial edge. (From Noyes FR, Barber-Westin SD: Surgical restoration to treat chronic deficiency of the posterolateral complex and cruciate ligaments of the knee joint. Am J Sports Med 24:415-426, 1996.)

terms; however, they provide important data on the magnitude of the overall subluxation of the knee joint when two or more abnormal motion limits are present. In addition to these clinical tests, it is important at arthroscopy to perform the lateral tibiofemoral jointopening gap test and measure the amount of joint opening with a calibrated nerve hook (Fig. 23K-5). Knees that have insufficient posterolateral structures will demonstrate at least 8 mm of joint opening at the intercondylar notch, and 12 mm or more of joint opening at the periphery of the lateral tibiofemoral compartment.

Gait Analysis The presence of varus angulation alone is not sufficient to justify a corrective tibial osteotomy. Rather, the symptoms, medial tibiofemoral arthrosis, and functional limitations are the primary indicators for a valgus-producing HTO. Gait analysis allows the detection of patients with (1) abnormally high knee adduction moments who are at increased risk for progression of medial tibiofemoral arthrosis due to excessive loading in that compartment, (2) predicted (calculated) abnormally high tensile forces in the lateral soft tissue restraints who are at increased risk for stretching out these tissues owing to lateral condylar liftoff with activity, and (3) abnormally high knee adduction moments who will have a less than desirable outcome from an HTO. Many studies have documented that the external moments about the knee joint and the corresponding tibiofemoral compartment loads are markedly influenced by individual gait characteristics and gait adaptations that occur after injury.20-30 Patients with ACL deficiency may show a decrease in the magnitude of the external flexion moment (quadriceps-reduced gait), or an increase in the external extension moment (hamstrings protective muscle force). A high adduction moment may be anticipated

because of the varus angulation; however, it is also known that the moments and loads on the knee joint cannot be reliably predicted from the static alignment of the lower limb measured on radiographs. The alignment of the foot markedly influences the knee adduction moment. Patients with toe-in, or less than normal external axial, rotation of the foot during stance phase tend to have a higher knee adduction moment (Fig. 23K-6). We reported on the results of gait analyses in 32 patients with ACL deficiency and varus angulation.31 A force plate and optoelectronic system were used to measure forces and moments of the lower limb and knee joint. Knee joint loads and ligament tensile forces were calculated using a previously described mathematical model.3 Most patients (20 of 32) had an abnormally high magnitude of the moment, tending to adduct the affected knee (Fig. 23K-7). The calculated medial tibiofemoral loads were excessively high in 21 of 32 patients (P < .01). Fifteen of 32 patients had predicted abnormally high lateral ligament tensile forces (P < .05). The adduction moment showed a statistically significant (P <.05) correlation to predicted high medial tibiofemoral compartment loads and high lateral ligament tensile forces (P < .01). We interpreted these findings as indicating a shift of the center of maximal joint pressure to the medial tibiofemoral compartment, with a corresponding increase in the lateral ligament tensile forces to achieve frontal plane stability. These phenomena are illustrated in Figures 23K-8 and 23K-9. If the muscle forces are not sufficient to maintain lateral tibiofemoral compressive loads, tensile forces develop in the lateral soft tissues. The data indicate, in knees with high lateral ligament tensile forces, the likelihood of separation of the lateral tibiofemoral joint occurring with condylar liftoff during periods of the stance phase. The magnitude of the flexion moment (which is related to quadriceps muscle force) was significantly lower in 15 of 32 patients (P < .05), and the extension moment (related to

Knee 1809 Minimum adduction 23-year-old female

Adduction moment = 1.8

Maximum adduction 31-year-old female

Adduction moment = 4.9

Inversion moment = 0.05

Inversion moment = 1.05 Foot angle = 29°

Ground reaction force

Foot angle = 4° Ground reaction force

Figure 23K-6   Diagram depicting the knee and ankle moments and the foot angle recorded during stance for the subjects with the lowest knee adduction moment (left) and the highest knee adduction moment (right). (Redrawn from Andrews MA, Noyes FR, Hewett TE, Andriacchi TP: Lower limb alignment and foot angle are related to stance phase knee adduction in normal subjects: A critical analysis of the reliability of gait analysis data. J Orthop Res 14:289-295, 1996.)

hamstring muscle force) was significantly higher in 16 of 32 knees (P < .05). These findings indicate a gait adaptation to diminish quadriceps muscle activity and enhance hamstring muscle activity to provide dynamic anteroposterior stability of the knee joint.25 Following ACL ­reconstruction, Normal Adduction Moment 6

cutoff = 3.30

these gait adaptations are lost, with resumption of a normal knee flexion moment. Equally important in the abnormal results was the finding that about one third of the patients had normal or low adduction moments and corresponding normal to low medial tibiofemoral compartment loads. These patients had gait characteristics or adaptations that tended to lower medial tibiofemoral loads despite the varus angulation of the knee joint. Gait analysis, therefore, allowed identification of patients with a potentially better overall prognosis; the adduction moment and medial tibiofemoral loads were not excessively high, and an HTO would result in a substantial lowering of the loads placed on the medial tibiofemoral joint. Markolf and coworkers reported that the absence of lateral tibiofemoral compartment loads, along with lateral condylar liftoff, had a marked effect on reducing joint stability.32 Further, we have encountered varus-angulated knees with insufficient lateral ligamentous structures in which multiple ligamentous reconstructive procedures failed and an HTO was required before further soft tissue reconstructive procedures could be done. We believe that the explanation for this clinical finding is that the knees have decompensated from a gait adaptation standpoint, and excessively high lateral tensile forces are placed across deficient lateral ligament tissues (Fig. 23K-10). Gait analysis predicts knees that have a higher risk for failure of lateral ligament reconstructive procedures, potentially affecting the treatment options.

Imaging and Preoperative Calculations The assessment of lower limb alignment is based on an examination of full standing radiographs. Separation of the lateral tibiofemoral compartment may occur, preventing correct assessment of the true tibiofemoral osseous alignment. We recommend double-stance, full-length anteroposterior radiographs that show both lower extremities from the femoral head to the ankle joint. The knee should be flexed to 5 degrees to avoid the abnormal varus recurvatum position. The use of a single-stance radiograph is not recommended because of the increased lateral joint opening that occurs with deficient lateral ligament restraints. Double-stance full-length radiographs taken while the patient contracts both the hamstrings and quadriceps Figure 23K-7   The distribution of the adduction moments during walking in the anterior cruciate ligament–deficient knees. The cutoff value (3.30) (% body weight × height) represents the control mean minus 1 SD. (Redrawn from Noyes FR, Schipplein OD, Andriacchi TP, et al: The anterior cruciate ligamentdeficient knee with varus alignment: An analysis of gait adaptations and dynamic joint loadings. Am J Sports Med 20(6):707-716, 1992.)

Above Normal Adduction Moment

Number of Knees

5 4 3 2 1 0

1.5

2.0

2.5

3.0 3.5 4.0 4.5 Adduction Moment (%Bw � Ht)

5.0

5.5

6.0

�rthopaedic ����������� S �ports ������ � Medicine ������� 1810 DeLee & Drez’s� O No Pretension Insufficient Muscle Force

No Pretension Sufficient Muscle Force

Pretension + Muscle Force OR

FM1

FM FL

FL

FM2 FL 1

MA

MA

FL = 0 FM < MA

FL = 0 FM1 > FM FM1 < MA

FL = Pretension FM2 FM FM1 2) + (FM2

muscles may help to ensure that both tibiofemoral compartments are in contact. It is imperative that simultaneous medial and lateral tibiofemoral contact occur (or that appropriate calculations are made to subtract the lateral compartment opening) so that the true osseous tibiofemoral alignment can be determined. An overcalculation of the degrees of varus alignment to be corrected can occur if abnormal lateral joint opening is present on the

Adduction

0

4.0

Stance

FM �

Laternal soft tissue

MA

2

Axial force

External adducting moment Figure 23K-9   The external adducting moment is resisted   by the minimal sagittal plane muscle force (Fm) and axial load acting over ℓ. Pretension in the lateral soft tissues would maintain equilibrium if the muscle force were insufficient. (Redrawn from Schipplein OD, Andriacchi TP: Interaction between active and passive knee stabilizers during level walking. J Orthop Res 9:113-119, 1991).

1)

Figure 23K-8   A critical interaction between the dynamic muscle forces and the forces in the passive soft tissues is needed to stabilize the knee joint during walking. The knee joint remains closed laterally if either pretension in the lateral soft tissues or increased muscle force resulting from antagonistic muscle groups is present. Distances ℓ and ℓ1 = 20 mm; ℓ2 = 60 mm. Fi, soft tissue force; Fm, Fm1, Fm2, muscle forces; MA, adducting moment. (Redrawn from Schipplein OD, Andriacchi TP: Interaction between active and passive knee stabilizers during level walking. J Orthop Res 9:113-119, 1991).

> MA

radiographs but not taken into consideration. The varus alignment may be due entirely to separation of the lateral tibiofemoral compartment, contraindicating an HTO. It is also important to look for the “teeter” effect described by Kettelkamp and colleagues,33 in which simultaneous contact of the medial and lateral tibiofemoral joints is impossible because of the obliquity of the medial tibial plateau resulting from arthrosis. This indicates that the overall limb alignment will probably remain in a varus position after HTO and that the goals of osteotomy in reducing pain may not be achieved. The results of our studies indicate that determination of the WBL at the knee joint represents the most precise method of preoperative planning (Fig. 23K-11).34 Fujisawa and associates also expressed postoperative alignment in terms of the position of the WBL by dividing the tibial plateau in halves from 0% to 100% medially and from 0% to 100% laterally, with 0% corresponding to the center of the tibial plateau.35 A WBL crossing the knee lateral to the 75% coordinate has the potential for a liftoff of the medial femoral condyle (Fig. 23K-12).34 Unicondylar weight-bearing resulting from distraction of the medial compartment is undesirable and could result in rapid lateral compartment deterioration, gradual MCL failure, and a possible progressive valgus deformity.27 In many reports, a relatively broad range of postoperative alignment, as defined by the tibiofemoral angle or the mechanical axis (MA) angle, has been considered acceptable. The WBL-tibial intersection depends on two separate variables: the final mechanical axis angle and the femoral and tibial lengths of the patient. An example is given in Figure 23K-12 of the postoperative mechanical axis in which any excess of 186 degrees would result in a WBL position lateral to the 75% coordinate. We use two methods to determine the correction wedge on preoperative radiographs. First, the centers of the femoral head and tibiotalar joint are marked on the full-length radiograph. The selected WBL coordinate of the tibial plateau is identified and marked. This is usually placed at 62% of the tibial width, which allows the WBL to pass through the lateral tibiofemoral compartment with a 3- to 4-degree angular overcorrection. If there is no medial tibiofemoral

Knee 1811 heel on

CFH

off

0.00 Adduction Moment (% BwxHt)

WBL WBL

peak = 3.07

peak = 5.70

6.00

Medial Compartment Load (% Bw)

5.00 involved knee control knee

a

b

peak = 2.68 peak = 2.14

0.00

Lateral Soft Tissue Force (% Bw)

1.00

peak = 0.92 CAM

peak = 0.28

0.00 Time (sec) Figure 23K-10   An example is shown of the increased adduction moment, medial compartment load, and lateral soft tissue force in an involved knee compared with a control knee. (Redrawn from Noyes FR, Schipplein OD, Andriacchi TP, et al: The anterior cruciate ligament-deficient knee with varus alignment: An analysis of gait adaptations and dynamic joint loadings. Am J Sports Med 20(6):707-716, 1992.)

compartment arthrosis and the HTO is indicated for varus angulation and thrust, it is not necessary to obtain a valgus overcorrection. In these cases, the 50% WBL tibial intersection point is calculated for the correction. One line is drawn from the center of the femoral head to the tibial coordinate, and a second line is drawn from the center of the tibiotalar joint to the tibial coordinate (Fig. 23K-13). The angle formed by the two lines intersecting at the tibial coordinate represents the angular correction required to realign the WBL through this coordinate. The second method of determining the correction wedge involves cutting the full-standing radiograph horizontally through the line of the superior osteotomy cut (Fig. 23K-14). A vertical cut of the lower tibial segment is then made to converge with the first cut at the level of the medial cortex. The distal portion of the radiograph is

Figure 23K-11   Calculation of the weight-bearing line (WBL) ratio. The WBL is drawn to connect the center of the femoral head (CFH) with the center of the ankle mortise (CAM). Distance a is measured perpendicular from the WBL to the medial edge of the proximal tibia. Distance b represents the entire width of the proximal tibia. Dividing a by b yields the WBL ratio. A WBL ratio of less than 0.5 indicates varus angulation with the load shifted medially, whereas a value of greater than 0.5 denotes valgus angulation with the load shifted laterally. (Redrawn from Noyes FR, Simon R: The role of high tibial osteotomy in the anterior cruciate ligament-deficient knee with varus alignment. In DeLee JC, Drez D [eds]: Orthopaedic Sports Medicine Principles and Practice. Philadelphia, WB Saunders, 1994, pp 1401-1443).

aligned until the center of the femoral head, the selected WBL coordinate point on the tibial plateau, and the center of the tibiotalar joint are all collinear. With the radiograph taped in this position, the angle of the wedge formed by the overlap of the two radiograph segments is measured and compared with the value obtained using the first method. The mechanical axis is measured to determine the angular correction. If there is a discrepancy between the two correction wedge angles, the procedures should be repeated. The height of the patella is measured on lateral radiographs to determine whether an abnormal patella infera or alta position exists, which may contraindicate an opening or closing wedge osteotomy, respectively, because these procedures would further decrease or elevate the patella position. Lateral radiographs are also examined for abnormal tibial slope, such as an excessive posterior sloping of the tibial surface greater than 8 degrees. Abnormal posterior sloping of the tibia in the sagittal plane is usually seen after tibial

�rthopaedic ����������� S �ports ������ � Medicine ������� 1812 DeLee & Drez’s� O MECHANICAL AXIS 180° 183° 186°189°192° 195°

CFH STEPS 1 Draw line from CFH to 62% coordinate 1

0%

50%

75%

Target Area 40 mm

Angle formed by the two lines equals the angle of correction required to result in weight-bearing line through the 62% coordinate

100%

60 mm 80 mm

TW = 80mm

2 Draw line from CTTJ to 62% coordinate

62%

2

Figure 23K-12   Weight-bearing line positions in a male 46 cm taller than the female in Figure 23K-13. Note that a correction beyond 186 degrees results in a weight-bearing line lateral to the desired postoperative position (target area). TW, tibial width. (Redrawn from Dugdale TW, Noyes FR, Styer D: Pre-operative planning for high tibial osteotomy: Effect of lateral tibiofemoral separation and tibiofemoral length. Clin Orthop 271:105-121, 1991.)

f­ ractures or growth abnormalities and may increase forces on an ACL reconstruction, resulting in failure. On cross section, the proximal anteromedial tibial cortex has an oblique or triangular shape, whereas the lateral tibial cortex is nearly perpendicular to the posterior margin of the tibia. Because of this relationship, a medial opening wedge osteotomy that has an anterior tibial tubercle gap (width) equal to the gap (width) at the posteromedial margin would increase tibial slope, decrease knee extension, and potentially increase ACL tensile loads.9 A lateral closing wedge osteotomy that has an equal anterior-to-posterior gap along the lateral tibial cortex would have a small effect on tibial slope. We conducted a study to mathematically calculate through three-dimensional analysis of the proximal tibia (using fine-cut axial computed tomography) how the angle of the opening wedge along the anteromedial tibial cortex influences the tibial slope (sagittal plane) and valgus correction (coronal plane) when performing a medial opening wedge osteotomy.36 Measurements of the wedge angle and gap angle along the anteromedial tibial cortex were made from a computerized model. Standard algebraic calculations were made using the law of triangles to determine the effect of different degrees of opening wedge osteotomy on coronal (valgus) and sagittal (tibial slope) alignment. The opening wedge angle, along the anteromedial tibial cortex to maintain the tibial slope, was found to be dependent on the angle of coronal valgus correction (HTO coronal angle) and the angle of obliquity of the anteromedial tibial cortex. In Figure 23K-15, the results are shown for the calcu­ lation of the opening wedge angle (along the ­anteromedial

CTTJ Figure 23K-13   Graphic depiction of the method used to calculate the correction angle of a high tibial osteotomy using a full-length anteroposterior radiograph of the lower extremity. The lines from the centers of the femoral head (CFH) and tibiotalar joint (CTTJ) converge in this example at the 62% coordinate. This provides the angle of correction, which will result in a weight-bearing line passing through the 62% coordinate. (Redrawn from Dugdale TW, Noyes FR, Styer D: Pre-operative planning for high tibial osteotomy: Effect of lateral tibiofemoral separation and tibiofemoral length. Clin Orthop 271: 105-121, 1991.)

tibial cortex) for five different osteotomy ­corrections in the coronal valgus plane (2.5 to 12.5 degrees). As an example, a 10-degree coronal valgus correction (assuming a 45-degree obliquity of the anteromedial tibial cortex with respect to the hinge axis) would result in a 7-degree opening wedge angle along the anteromedial tibial cortex. A larger wedge angle would decrease the tibial slope, and a smaller wedge angle would increase the tibial slope. The gap angle is perpendicular to the anteromedial oblique surface of the tibia with a vertex on the hinge axis posterior to the tibia. This is shown in Figure 23K-16 where the gap angle in degrees is shown for five different osteotomy corrections as a function of the obliquity of the anteromedial tibial cortex. The opening wedge angle can be set at surgery by measuring and altering the vertical gap at two points along the osteotomy site, Y1 and Y2 (Fig. 23K-17). This has importance in determining that the correct wedge angle is obtained before internal fixation at the osteotomy site. The site at which the vertical gap measurement is taken is dependent on the ­ coronal distance from the hinge axis, obliquity of the anteromedial tibial cortex,

Knee 1813 14 CFH

CORRECTED WBL

Wedge angle (deg)

CFH

WBL

12.5 deg

12 10 deg

10 8

7.5 deg

6

5 deg

4

2.5 deg

2 0 0

15

62% HINGE

45 60 75 Oblique angle (deg) Anteromedial tibial cortex

90

Figure 23K-15   The anteromedial cortex opening wedge angle depends on the oblique angle of the tibial cortex with respect to the hinge axis. Each line represents the desired calculated degrees of correction for the opening wedge osteotomy in the true coronal (90-degree) plane. (Redrawn from Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33:378-387, 2005.)

UNCUT HINGE

CUT

CTT

J

CTTJ

30

ROTATE

Figure 23K-14   Graphic depiction of an alternative method used to calculate the correction angle of a high tibial osteotomy using a full-length anteroposterior radiograph of the lower extremity. The roentgenograph is cut to allow the center of the femoral head (CFH), the 62% coordinate, and the center of the tibiotalar joint (CTTJ) to become collinear. The angle of the resulting wedge of roentgenograph overlap equals the desired angle of correction. The example is provided for a closing wedge osteotomy. The same technique is used for an opening wedge osteotomy where the medial tibial opening wedge is made to obtain the desired correction. (Redrawn from Dugdale TW, Noyes FR, Styer D: Pre-operative planning for high tibial osteotomy: Effect of lateral tibiofemoral separation and tibiofemoral length. Clin Orthop 271:105-121, 1991.)

opening, the correct width of the buttress plate would be 9.2 mm. A wider buttress plate gap would result in excessive valgus alignment and altered tibial slope. In general, the following rules maintain tibial slope in a medial opening wedge tibial osteotomy: • The most anterior gap of the osteotomy wedge at the tibial tubercle should be one half the posteromedial gap to maintain the tibial slope. • Every millimeter of gap error at the tibial tubercle results in about 2 degrees of change in the tibial slope. • Preoperative radiographic measurement of the tibial slope is recommended. • The buttress plate height at Y2 (placed anterior to the MCL) will be 2 to 3 mm less than the posteromedial 14

12.5 deg

and distance along the osteotomy site on the ­anteromedial surface (Tables 23K-4 and 23K-5). In Table 23K-4, the millimeters of opening at the osteotomy site are based on the width of the tibia and the angle of correction. In Table 23K-5, an average 45-degree oblique angle of the anteromedial cortex and a tibial width of 60 mm is assumed. This allows the surgeon to calculate at the time of surgery, making simple measurements, the desired gap height at two points along the osteotomy to maintain the tibial slope. If a buttress wedge plate is used, the appropriate-sized plate may be selected for the opening wedge depending on the site where the plate is placed. An example is given if the surgeon selects to place the buttress plate just anterior to the superficial MCL on the anteromedial tibial cortex about 20 mm anterior to the most posteromedial point of the osteotomy. In Table 23K-5, with a 10 mm posteromedial opening (Y1), the correct width of the buttress plate would be 7.6 mm (tibial width, 60 mm). For a 12-mm ­posteromedial

Gap angle (deg)

12 10 deg

10

8 7.5 deg 6 5 deg 4

2.5 deg

2 0 0

15

30

45 60 75 Oblique angle (deg) Anteromedial tibial cortex

90

Figure 23K-16   The magnitude of the gap angle changes with the obliquity of the anteromedial tibial cortex angle. Each line represents the calculated degrees of correction for the opening wedge osteotomy in the coronal plane. (Redrawn from Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33:378-387, 2005.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1814 DeLee & Drez’s� O

Y2 Y1 L Figure 23K-17   The opening wedge angle along the anteromedial tibial cortex can be calculated using the three linear measurements along the osteotomy opening wedge. Y2, posterior gap; Y1, gap anterior to Y2; L, length between Y1 and Y2. (From Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33:378-387, 2005.)

height at Y1 (gap) to maintain tibial slope during the coronal valgus correction. • Because only a few millimeters of change at the osteotomy site affects slope and coronal angulations, radiographic confirmation of final alignment at surgery and postoperatively is required.

TREATMENT OPTIONS Operative Indications Most patient candidates for HTO are in their third to fifth decade and wish to remain active and avoid unicompartmental knee replacement. The predominant indication is lower limb osseous malalignment in younger patients who have medial tibiofemoral joint pain and wish to maintain

an active lifestyle. Unfortunately, a prior medial meniscectomy is a major risk factor for progression of arthrosis in these knees. Because any underlying arthrosis is expected to progress, it is advisable to perform HTO while the joint damage is in early stages, before the development of severe articular cartilage deterioration and loss of tibiofemoral joint space.37,38 One advantage of performing HTO in young patients who have medial tibiofemoral arthrosis (following medial meniscectomy) is the opportunity to also perform a meniscus repair or transplantation, or a cartilage restoration procedure for full-thickness articular cartilage defects. The alignment is not overcorrected to valgus in knees in which no damage exists to the articular cartilage in the medial tibiofemoral compartment. The appropriate level of physical activity that can be recommended after HTO remains questionable, and patient education is important so that the limitations to be expected postoperatively are well understood. The goal of the osteotomy is to allow an active pain-free lifestyle that includes low-impact recreational pursuits but avoids high-impact activities such as twisting, turning, jumping, and pivoting. The timing of the HTO and ligament reconstructive procedures in knees with deficient knee ligaments is based on several factors shown in Table 23K-6.39 In double-varus knees, the gap test is used at arthroscopy to determine whether it is safe to proceed with ACL reconstruction concomitantly with the osteotomy, or whether it should be staged to allow adaptive shortening of posterolateral tissues. In knees that do not demonstrate abnormal lateral joint opening or external tibial rotation, the HTO and ACL reconstruction may be performed at the same setting. The tibial fixation of the ACL graft is performed by placing interference screws proximal to the osteotomy site and by adding sutures, which are tied to a suture post for additional fixation. An ACL reconstruction should not be performed if there is excessive abnormal lateral tibiofemoral compartment opening because this would place the graft under undue forces. In triple-varus knees, we stage the ligament reconstructive procedures after the osteotomy. Performing all procedures (HTO, ACL reconstruction, and posterolateral reconstruction) simultaneously results in a lengthy operation with increased risk for complications, prolonged

   TABLE 23K-4  Millimeters of Opening at the Osteotomy Site Based on the Width of the Tibia and the Angle of Correction Degrees of Angular Correction TW

5

6

7

8

9

10

11

12

13

50 55 60 65 70 75 80 85 90 95 100

4.37 4.81 5.25 5.69 6.12 6.56 7.00 7.44 7.87 8.31 8.75

5.25 5.78 6.30 6.83 7.35 7.88 8.40 8.93 9.45 9.98 10.50

6.15 6.77 7.38 8.00 8.61 9.23 9.84 10.46 11.07 11.69 12.30

7.00 7.70 8.40 9.10 9.80 10.50 11.20 11.90 12.60 13.30 14.00

8.00 8.80 9.60 10.40 11.20 12.00 12.80 13.60 14.40 15.20 16.00

8.80 9.68 10.56 11.44 12.32 13.20 14.08 14.96 15.84 16.72 17.60

9.70 10.67 11.64 12.61 13.58 14.55 15.52 16.49 17.46 18.43 19.40

10.85 11.94 13.02 14.11 15.19 16.28 17.36 18.45 19.53 20.62 21.70

11.55 12.71 13.86 15.02 16.17 17.33 18.48 19.64 20.79 21.95 23.10

TW, coronal tibial width at osteotomy site. From Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33:378-387, 2005.

Knee 1815

   TABLE 23K-5  Opening Wedge Height Measurements* Tibial Width at Osteotomy (Y2), mm Opening at Osteotomy (Y1), mm

L, mm

50

55

60

65

70

8

0 20 25 30 35 40 45 0 20 25 30 35 40 45 0 20 25 30 35 40 45

8.0 5.7 5.2 4.6 4.0 3.5 2.9 10.0 7.2 6.5 5.8 5.1 4.3 3.6 12.0 8.6 7.8 6.9 6.1 5.2 4.4

8.0 5.9 5.4 4.9 4.4 3.9 3.4 10.0 7.4 6.8 6.1 5.5 4.9 4.2 12.0 8.9 8.1 7.4 6.6 5.8 5.1

8.0 6.1 5.6 5.2 4.7 4.2 3.8 10.0 7.6 7.1 6.5 5.9 5.3 4.7 12.0 9.2 8.5 7.8 7.1 6.3 5.6

8.0 6.3 5.8 5.4 5.0 4.5 4.1 10.0 7.8 7.3 6.7 6.2 5.6 5.1 12.0 9.4 8.7 8.1 7.4 6.8 6.1

8.0 6.4 6.0 5.6 5.2 4.8 4.4 10.0 8.0 7.5 7.0 6.5 6.0 5.5 12.0 9.6 9.0 8.4 7.8 7.2 6.5

10

12

*By measuring the width of the tibia, the opening wedge height at the most medial point (Y1), and the distance between vertical measurement points (L), the vertical height at the second measurement point (Y2) can be found on the table. Calculations based on 45-degree angle of the anteromedial tibial cortex at osteotomy site. From Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33:378-387, 2005.

r­ ehabilitation, and knee motion problems.40-42 We perform the HTO and then, after adequate healing of the osteotomy, an arthroscopically assisted ACL reconstruction and open FCL and posterolateral reconstruction. The ACL and posterolateral reconstructive procedures must be performed together to allow both structures to function together to resist abnormal lateral tibiofemoral joint opening and varus recurvatum. Correction of the varus

alignment decreases the risks for failure of the ligament reconstructive ­procedures.43-46 When a primary ACL reconstruction is performed with the HTO, our first graft choice is a bone–patellar tendon–bone autograft. For knees undergoing ACL revision reconstruction and osteotomy, we prefer to use a contralateral bone–patellar tendon–bone autograft,44 followed by an ipsilateral quadriceps tendon–bone autograft,47 or

   TABLE 23K-6  Indications and Timing of Ligament Reconstructive Procedures Ligament Deficiency

Surgical Procedure

Anterior cruciate ligament (ACL)

Autograft options: Central-third bone–patellar tendon–bone Quadriceps tendon–patellar bone Semitendinosus-gracilis

Fibular collateral ligament (FCL; posterolateral complex intact, no increased external rotation or varus recurvatum) All posterolateral structures

Usually not required

Anatomic FCL, posterolateral reconstruction with autograft or allograft Posterolateral complex proximal advancement

Timing Related to High Tibial Osteotomy (HTO) At or preferably after HTO

Staged procedure: HTO first, ACL, posterolateral reconstruction

Indications/Comments Any patient who had instability before HTO should not risk a further trial of function and possible reinjury. Consider when secondary ligament restraints are lost (pivot shift 3+ impingement, >10 mm increased anterior displacement involved knee) and associated medial or lateral ligament deficiency is present. Consider when meniscus repair is performed. Athletically active patient desiring best knee possible for return to sports activities. Expect adaptive shortening FCL in most patients after valgus-producing osteotomy. At HTO, avoid disrupting proximal tibiofibular joint, which would allow proximal migration and laxity of posterolateral structures. Patients usually have increased lateral joint opening of 8 mm at the intercondylar notch (12 mm or more at periphery), have increased external rotation of 10 to 15 degrees, and require posterolateral reconstruction at time of ACL reconstruction. Combined posterolateral reconstruction and ACL reconstruction are always performed together to limit hyperextension and varus rotations.

Modified from Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament–deficient knees. Am J Sports Med 28:282-296, 2000.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1816 DeLee & Drez’s� O

a four-strand semitendinosus–gracilis tendon autograft. We avoid allografts for ACL reconstructions in revision knees whenever possible because of the higher failure rates of these grafts compared with autografts.46,48,49 If an allograft is required and the knee demonstrates marked anterior displacement indicating involvement of the secondary ligament restraints, consideration may be given to also performing an iliotibial band extra-articular Loseetype procedure.49

Contraindications In general, HTO is avoided in knees that demonstrate a 15 × 15 mm2 area or more of exposed bone on both the tibial and femoral surfaces. There are knees in which the area of exposed bone may be slightly greater that are considered candidates; however, as a general rule, articular cartilage should be present over the majority of the joint surfaces. Major concavity of the medial tibial plateau with loss of bone stock is a contraindication to HTO. Knees that demonstrate (on standing 45-degree posteroanterior radiographs50) no remaining joint space in the medial compartment are not candidates. An arthroscopic procedure is done just before HTO to assess the amount of remaining articular cartilage and remove symptomatic meniscus fragments and other tissues. An absolute contraindication for a medial opening wedge osteotomy is the use of nicotine products in any form. The complication of a nonunion is not worth the risk, and a minimum of 8 weeks of abstinence before surgery should be insisted. The patient is warned that there may still be an increased risk for healing problems. A relative contraindication is body weight greater than 200 lb (91 kg). Although there may be some patients in whom HTO is indicated who weigh up to 225 lb (102 kg), this operation is avoided in patients with a higher body weight because the beneficial effect of unloading the medial compartment will not be achieved. A relative contraindication is increased medial slope to the affected medial tibial plateau in the coronal plane. This finding indicates that it will not be possible to significantly unload the medial compartment with an HTO, and the knee will remain with all of the weight-bearing confined to the medial compartment. This problem can be tested before surgery on examination of the knee joint with the standard varus-valgus stability tests at 20 to 30 degrees of knee flexion. In these knees with advanced medial arthrosis, there is no neutral point in which there is simultaneous contact of the medial and lateral compartments. The tibia goes into a “teeter-totter” state, with contact alternating between the medial or lateral compartment and obvious separation of the noncontacted compartment because of bone loss. Marked patellofemoral symptoms contraindicate HTO. The finding of asymptomatic articular cartilage changes to the patellofemoral joint is not a contraindication to HTO because clinicians have noted that the end result in terms of longevity of the HTO is dependent on the symptomatic medial tibiofemoral compartment.1,51 An abnormal patella infera or alta position contraindicates an opening or closing wedge osteotomy, respectively, because these procedures would further decrease or elevate the patella position.

WEIGHING THE EVIDENCE The survival rates reported after osteotomies are shown in Table 23K-7. Although conversion to total knee arthroplasty (TKA) is uniformly used as an end point for survival of HTO, some investigators also incorporate a low overall Hospital for Special Surgery (HSS) knee rating score, patient dissatisfaction, or the presence of pain in patients who declined TKA as additional end points. The highest long-term survival rate for closing wedge osteotomy was reported by Koshino and colleagues, who followed 75 knees from 15 to 28 years postoperatively.52 At the final follow-up examination, 93.2% of the patients had not converted to total or unicompartmental knee ­arthroplasty or complained of moderate to severe knee pain. The authors attributed the success of the procedure to the achievement of 10 degrees of anatomic valgus, avoidance of a flexion contracture, and incorporation of a patellofemoral decompressive procedure in patients with preexisting patellofemoral degeneration. Other closing wedge osteotomy studies show more modest survival rates 10 and 15 years after surgery. The 10-year survival rates range from 51% to 78%, with an average rate of 64% when Koshino and colleagues’ series is excluded.51,53-56 The investigations of Naudie and coworkers55 and Billings and associates had the lowest 10-year survival rates of 51% and 53%, respectively. Naudie and coworkers reported that the probability of survival increased in patients who were younger than 50 years of age at the time of the HTO and who had preoperative knee flexion greater than 120 degrees.55 Billings and associates54 failed to find a statistically significant association between survival rates and patient age, amount of valgus correction achieved, or postoperative complications. More recent studies from Sprenger and Doerzbacher56 and Aglietti and associates53 reported more favorable results 10 years after surgery, with survival rates of 74% and 78%, respectively. Fifteen-year survival rates of closing wedge osteotomy, provided to date by only a few authors, range from 39% to 57%.53,55-57 Aglietti and associates reported that male gender and preoperative quadriceps strength correlated with a long-term survival rate of 57%.53 Sprenger and Doerzbacher determined that the survivability of HTO (56% at 15 years) was related to the achievement of 8 to 16 degrees of valgus measured 1 year postoperatively.56 Naudie and associates reported the probability of survival decreased in patients older than 50 years, or in whom a previous arthroscopic débridement, presence of a lateral tibial thrust, preoperative knee flexion less than 120 degrees, or delayed union or nonunion postoperatively was present.55 Huang and coworkers identified only preoperative varus malalignment as predictive of survivability of HTO.57 Knees with 9 degrees or less of varus had a 10-year survival rate of 93%, whereas those with more than 9 degrees of varus had a 10-year survival rate of only 56%. Fewer data are available regarding survival rates following opening wedge osteotomy. Three studies reported 84%,58 89%,59 and 94%60 survival rates 5 years after surgery. After 10 years, survival rates are available from only two investigations: 63% in Weale and colleagues’ series of 73 cases,59 and 85% from Hernigou and Ma’s series of 203 knees.60 Weale and colleagues hypothesized that

Knee 1817

   TABLE 23K-7  Survival Rates after High Tibial Osteotomy

Study Huang,

200557

Koshino, 200452

Sterett, 200458 Aglietti, 200353 Sprenger,

200356

Weale, 200159

Type of ­Osteotomy

N, Patient Age (Range)

End Points for Survival Analysis

Closing wedge Closing wedge

93, 57 yr (38-73) 75, 59 yr (46-73)

Opening wedge Closing wedge Closing wedge

38, 51 yr (34-79) 91, 58 yr (36-69) 76, 69 yr (47-81)

Opening wedge

76, 54 yr (36-70)

1. TKA 2. Patient dissatisfaction 1. TKA or unicompartmental arthroplasty 2. Moderate pain at final follow-up >15 yr postoperatively 1. TKA 2. Revision HTO 1. TKA 2. HSS score < 70 1. TKA 2. HSS score < 70 3. Patient dissatisfaction 1. TKA 2. Waiting for TKA 3. Postoperative sepsis precluded revision TKA

Hernigou, 200160 Opening wedge Billings, 200054 Closing wedge Naudie, 199955 Closing wedge Coventry, 199351 Closing wedge

215, 61 yr (48-72) 64, 49 yr (23-69) 106, 55 yr (16-76) 87, 63 yr (41-79)

Insall, 198461a

95, 60 yr (30-83)

Closing wedge

Survival Rates after Surgery 5 Years

10 Years

15 Years

Correlations with Survival Rate

94.6%

87%

75.2%

97.8%

96.2%

93.2%

Preoperative tibiofemoral alignment (9 degrees varus) None

84%

NA

NA

None

96%

78%

57%

86%

74%

56%

Alignment at healing, muscle strength, male gender Alignment at 1 yr after surgery

88.8%

63%

NA

None

94%

85%

68%

None

TKA

85%

53%

NA

None

TKA

73%

51%

39%

1. TKA 87% 2. Moderate or severe pain in patients who declined TKA Survival rate not calculated, but 23% revised to TKA

66%

NA

Body weight, delayed or nonunion, age, preop flexion Body weight, alignment at 1 yr after surgery HSS excellent to good results: 2 yr—97%, 5 yrs—85%, 9 yrs—37%. Alignment did not correlate with results; passage of time determined result.

HSS, Hospital for Special Surgery rating system; NA, not available; TKA, total knee arthroplasty.

­ rogression of arthrosis in the medial compartment correp lated with HTO failure.59 Hernigou and Ma did not comment on factors that could have affected the survival rates in their investigation. These authors provided the only 15-year survival rate published at the time of writing of opening wedge osteotomy of 68%. The ability of HTO to alleviate pain has been demonstrated in many studies53,61-65; however, the longevity of pain relief correlates with the length of follow-up achieved postoperatively. Unfortunately, several investigators did not separately assess pain when determining clinical outcome but only provided final rating scores from knee rating systems such as the Lysholm, Hospital for Special Surgery, WOMAC, and the American Knee Society.54,58,61,66-68 Aglietti and associates followed 61 patients clinically from 10 to 21 years after surgery and reported that 79% had no or only mild knee pain.53 Koshino and colleagues reported in a small series of 18 patients that all had relief of pain on follow-up evaluations ranging from 38 to 114 months postoperatively.61 Only 9 patients had more than 7 years of follow-up in this investigation. Rinonapoli and coworkers followed 60 knees from 10 to 21 years after surgery and reported that pain at rest was absent in 55%, mild in 18%, moderate in 22%, and severe in 5%.65 Pain on

walking was absent or mild in 55%, moderate in 27%, and severe in 18%. Satisfactory pain relief has been reported in most studies that followed knees that had HTO and ACL reconstruction (either concomitant or staged). Williams and associates followed 25 ACL-deficient varus-angulated knees from 24 to 106 months after surgery.69 Thirteen of these knees were treated with a combined HTO and ACL reconstruction, and 12 had only an HTO. At follow-up, 84% had no pain with vigorous activity. Dejour and associates performed a combined HTO and ACL patellar tendon autograft reconstruction in 44 knees.70 At follow-up, which ranged from 1 to 11 years postoperatively, 66% had no pain or pain only after vigorous activity. Functional limitations with daily activities such as walking and stair climbing are typically reported in most patients before HTO. In addition, few patients are able to participate in even light sports activities without experiencing noteworthy symptoms. After surgery, the ability to walk an unlimited distance or more than 1 km is an important measure of daily function. Koshino and associates reported that 94% of 75 knees could walk more than 1 km without pain at 15 to 28 years after a closing wedge HTO.52 A more modest finding was reported by

�rthopaedic ����������� S �ports ������ � Medicine ������� 1818 DeLee & Drez’s� O

Pfahler and associates, who found that 57% of 62 knees followed 6 to 14 years after surgery could walk more than 1 hour.64 Koshino and associates described a small series of 21 patients who received an opening wedge osteotomy followed 38 to 114 months after surgery. At the most recent follow-up examination, 89% reported the ability to walk

an unlimited distance without limitations.61 The American Knee Society walking score improved from a preoperative mean of 19 ± 5.4 points to 46.7 ± 7.3 points at follow-up (P <.0001). However, a long-term investigation of closing wedge osteotomy from Rinonapoli and colleagues found that only 25% of 102 knees had unlimited walking ability 10 to 21 years after surgery.65

Authors’ Preferred Method The two most frequently used techniques for correcting varus deformity are the opening and closing wedge tibial osteotomies. We recommend opening wedge osteotomy in nearly all cases because the lateral dissection and fibular osteotomy are avoided when a large angular correction greater than 12 degrees is required to avoid excessive tibial shortening, when distal advancement or reconstruction of the MCL is also required, when an extensive FCL and posterolateral reconstruction is required (to avoid a proximal fibular osteotomy because ligament grafts are anchored to the proximal fibula), and when patella alta or leg-length shortening will be positively affected by the millimeters of the opening wedge osteotomy. There are disadvantages of opening wedge osteotomy in that autogenous bone grafting is required for healing of the open defect and achieving stability at the osteotomy site. There are no reported clinical trials to date on the use of bone allografts for opening wedge osteotomy. We believe the potential problem of early varus collapse of the osteotomy due to delayed union represents a serious complication that may be lessened by autogenous iliac bone grafting. In addition, a plate with �������������������������������������� locking ������������������������������ screws is required to achieve rigid fixation. There is also the problem of a greater chance of increasing the posterior tibial slope if the buttress plate is placed in a more anterior position. However, with attention to this problem at surgery, the normal tibial slope in the sagittal plane can be preserved. Another disadvantage of an opening wedge osteotomy is that distal transection of the superficial MCL distal attachment is required. In an opening wedge osteotomy of 5 to 7.5 mm, the MCL fibers may be incompletely transected at several different levels (“pie crust” approach) to maintain the distal attachment, effectively lengthening the ligament. The closing wedge osteotomy has the advantage of faster healing and earlier resumption of weight-bearing because contact of two large cancellous surfaces of the proximal tibia is achieved. The initial internal fixation of the osteotomy is more secure than an opening wedge procedure, and there is less chance for a change in osteotomy position and loss of correction. The closing wedge osteotomy involves more dissection, a meticulous approach, and osteotomy of the proximal fibular neck region while avoiding the peroneal nerve. It is also somewhat more tedious to resect more bone and alter the lower limb correction at surgery, should it be necessary. Opening Wedge Tibial Osteotomy

The entire lower extremity is prepared and draped free. The ipsilateral anterior iliac crest is also prepared. Arthroscopy is first performed to evaluate the articular surfaces of the ­medial

and lateral tibiofemoral compartments and patellofemoral joint. ACL-deficient knees frequently have associated meniscus tears that require repair15 or partial removal. Appropriate débridement of tissues, inflamed synovium, and notch osteophytes limiting knee extension is performed. The arthroscopic examination allows confirmation that there is remaining articular cartilage in the medial tibiofemoral compartment. A sterile tourniquet is inflated before the skin incision. A 4-cm incision is made over the anterior iliac crest and deepened to the periosteum (Fig. 23K-18), which is sharply incised and reflected medially only to the pelvic brim. Laterally, meticulous subperiosteal dissection is carried along the outer table of the pelvis. The graft size is defined on the bone using electrocautery. In most patients, the graft dimension is 40 mm in length, 10 mm in width, and 30 mm in depth. The inner iliac cortex is not dissected, the muscle attachments are not disturbed, and a spacer of the outer table defect is not required. Additional cancellous bone is removed from the inner pelvic cortex. The graft is later fashioned into three triangles; one triangle is placed posterior to the plate to close the gap at the posterior tibial cortex, one triangle is placed in the midportion of the osteotomy deep to the plate, and the smaller triangle is placed in the gap in the anterior tibial cortex. A 5-cm vertical skin incision is made medially midway between the tibial tubercle and posterior tibial cortex, starting 1 cm inferior to the joint line (Fig. 23K-19). The sartorius fascia is incised in line with its fibers, proximal to the gracilis tendon. The pes anserinus is retracted posteriorly, exposing the superficial medial collateral ligament (SMCL) and posterior border of the tibia. Anteriorly, the retropatellar bursa is entered by incising the medial patellar retinaculum, allowing the patellar tendon to be lifted to expose the anterior tibia. A small portion of the patellar tendon fibers attaching medially may be incised at the tibial tubercle to achieve an oblique osteotomy line. A sharp periosteal incision is made at the posteromedial tibial border, just posterior to the SMCL, to allow meticulous posterior tibial subperiosteal dissection by a Cobb elevator. Care is taken to protect the inferior medial geniculate artery that lies just beneath the distal fibers of the SMCL. A malleable retractor is placed in the subperiosteal posterior tibial space. Only sufficient posterior tibial subperiosteal dissection is used to protect the neurovascular structures, and wide dissection is not necessary. Management of superficial medial collateral ligament. There are three surgical approaches for management of the SMCL. In a small opening wedge osteotomy of 5 to 7.5 mm, a pie crust procedure using multiple transverse incisions

Knee 1819

Authors’ Preferred Method—cont’d

A

B

Figure 23K-18   A, A 4-cm incision over the anterior iliac crest is made to harvest the iliac crest bone graft. The graft is composed of the anterior crest and outer iliac cortex; the inner table is not removed. B, The usual iliac crest bone graft dimensions are 40 mm in length, 10 to 12 mm in width, and 30 mm in depth. (From Noyes FR, Mayfield W, Barber-Westin SD, et al: Opening wedge high tibial osteotomy: An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34:1262-1273, 2006.)

Figure 23K-19   Proximal tibial skin incision for opening wedge osteotomy midway between the tibial tubercle and posterior tibia along the anterior border of the superficial medial collateral ligament. The gracilis and semitendinosus tendon insertions are marked. (From Noyes FR, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson DW [ed]: Master Techniques in Orthopaedic Surgery, Reconstructive Knee Surgery. New York, Raven Press, 1995, pp 185-210.)

at different places may effectively lengthen the SMCL. In larger opening wedge osteotomies, it is necessary to transect the distal attachment, usually 6 cm from the joint and to perform a distal elevation of the attachment in a subperiosteal plane. This allows for the SMCL to be reattached distally after the opening wedge osteotomy with the posteromedial portion of the SMCL bridging the osteotomy site. Transecting the SMCL at the most distal attachment site preserves its length, allows excellent exposure and bone grafting, and allows the tibial fixation plate to be placed in a correct midline position. The tibial slope is not altered, and function of the SMCL is retained. A third approach is used when distal advancement of the SMCL is required as a reconstructive procedure due to SMCL insufficiency and abnormal medial joint opening. The SMCL is dissected to the medial joint, which is entered anterior and posterior to the SMCL fibers. The posterior incision is at the junction of the SMCL and posteromedial capsule (short oblique fibers). The anterior incision preserves the attachment of the medial patellofemoral ligament. The medial meniscus position is carefully examined when a distal advancement of the SMCL is performed. It may be necessary to incise the medial meniscus capsular attachments and then resuture the attachment to preserve the correct anatomic location of the meniscus when the SMCL is advanced. The SMCL should appear relatively normal without scar replacement because advancement of only scar tissue would be expected to fail and not provide medial stability. In select cases, a semitendinosus tendon augmentation of the SMCL may be necessary. The tendon is detached proximally, passed through a small drill hole at the femoral attachment just anterior and posterior to the SMCL femoral attachment, and then both anterior and posterior tendon arms are sutured to the SMCL after the osteotomy is completed. The distal attachment of the semitendinosus is not disturbed. In some instances, both Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1820 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

Femur

Tibia Pat tendon

Figure 23K-20   This intraoperative photograph shows the placement of two Keith needles (arrows) at the joint line, which assists the surgeon in obtaining an osteotomy that is perpendicular to the normal tibial slope. In the photograph, the Army-Navy retractor is beneath the patellar tendon. The osteotomy site starts at the anteromedial cortex, 35 mm from the joint line in an oblique manner just proximal to the tibiofemoral joint. This maintains sufficient width of the proximal tibia and limits the risk of a tibial plateau fracture. (From Noyes FR, Mayfield W, Barber-Westin SD, et al: Opening wedge high tibial osteotomy. An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34:1262-1273, 2006.)

the ­ gracilis and semitendinosus tendons may be used. The need to surgically reconstruct the SMCL is infrequent; however, the details as presented should allow for restoration of function when required. At the conclusion of the SMCL reconstruction, plication of the posteromedial capsule anteriorly to the SMCL is performed to remove any abnormal redundancy with the knee at full extension. Overtightening of the SMCL and posteromedial capsules is avoided, the repair allows for full knee extension and a normal range of flexion. Placement of tibial guide pins. The patellar tendon and retropatellar bursa are exposed medially. The superior attachment of the tendon is recessed 5 mm using a scalpel to provide adequate exposure for the osteotomy. Retractors are placed anteriorly and posteriorly to complete the exposure. A Keith needle is placed in the anterior medial joint just above the tibia, and the distance is marked on the desired point of the osteotomy along the anteromedial cortex. A second Keith needle is placed at the posteromedial tibial joint space, and the same millimeters are marked to provide a measurement of the tibial slope (Fig. 23K-20). A commercial guide system (Arthrex Opening Wedge Osteotomy System, Arthrex Inc., Naples, Fla) is used to facilitate guidewire placement (Fig. 23K-21). A 2-mm guide pin is placed at the posteromedial cortex at the marked line and advanced across the tibia at an oblique angle. This pin is usually placed at about 20 degrees of obliquity to the tibial shaft and is verified by intraoperative fluoroscopy. To prevent fracture of the lateral tibial plateau, it is important to retain as much lateral tibial width as possible. The error is to have too much obliquity to the guide pin, compromising postoperative tibial internal fixation. A second pin is placed anterior and parallel to the first pin. At this point, it is imperative to ensure that the medial osteotomy line (from anterior to posterior) is perpendicular to the joint line.

C

A

B

D

Figure 23K-21   A-D: Correct placement for guide pin and subsequent osteotomy with a thin osteotome. The lateral cortex is preserved. (Redrawn from Noyes FR, Barber-Westin SD, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson RW [ed]: Master Techniques in Orthopaedic Surgery, Reconstructive Knee Surgery. Philadelphia, Lippincott Williams & Wilkins, 2003, pp 229-260.)

Knee 1821

Authors’ Preferred Method—cont’d A measurement of the perpendicular cut is accomplished to confirm the distance of each guide pin from the articular surface of the tibia. The distances should be equal in order to maintain the original posterior tibial slope. The length of the posterior pin is measured and used following the law of triangles36 to determine the millimeters of osteotomy opening to obtain the desired angular correction. The osteotomy is performed using an oscillating saw for the outer medial and anterior cortices, followed by a 3⁄4-inch osteotome, placed over the guide pin and verified by fluoroscopy (Fig. 23K-22). A 1⁄2-inch osteotome is used for the posterior cortex, with the edge palpated posterior to the tibia as the osteotome is advanced. The osteotomy is carried to within 10 mm of the lateral cortex. Calibrated opening wedges (Arthrex Opening Wedge Osteotomy System) are then gently inserted into the osteotomy site to achieve the desired angular correction, hinging on the intact posterolateral cortex. This step requires several minutes to prevent fracture of the lateral tibial pillar. Intraoperative fluoroscopy verifies the correct hip-knee-ankle WBL at the tibia. In addition, the surgeon should determine that there is closure of the medial and lateral tibiofemoral compartments using axial weight-bearing against the foot with the knee at 5 to 10 degrees of flexion. The alignment is verified and adjusted if required to achieve the desired angular correction. The anterior gap of the osteotomy site should be one half of the posterior gap, following rules previously described to maintain the tibial slope.36 The width of the tibial buttress plate along the anteromedial cortex is measured and is always less than the millimeters at the posterior medial gap because of the angular inclination of the anteromedial tibial cortex. In select cases, the slope may be purposefully increased in posterior cruciate ligament (PCL)deficient knees or decreased in ACL-deficient knees. The three cancellous bone graft triangular segments are impacted tightly into the osteotomy site to obliterate the

A

space and provide added stability, particularly in the sagittal plane (Fig. 23K-23). An appropriately sized plate with trapezoidal block is selected and secured with 6.5-mm cancellous screws proximally and 4.0-mm cortical screws distally (Fig. 23K-24). We avoid the use of plates with a square buttress block because this geometry increases the posterior tibial slope. The SMCL fibers are sutured distally and secured to either the plate screws or suture anchors to maintain tension. The pes anserine tendons and sartorius fascia are ­reapproximated. Closing Wedge Osteotomy

15

17.5

15

17.5

10

12.5

10

12.5

We report the technique for a closing wedge osteotomy primary for completeness and the select cases that require this procedure. The primary indications are instances in which an opening wedge procedure is contraindicated. For example, patients with preexisting patellar infera who require a large osteotomy may develop, in rare instances, a limblength discrepancy (due to an increase in tibial length) following an opening wedge procedure. In addition, those with medial soft tissue scar formation in whom a lateral approach through normal tissue would be preferred are candidates for a closing wedge osteotomy. An oblique incision is made from the head of the fibula to the anterior crest of the tibia, 2 cm distal to the tibial tubercle (Fig. 23K-25). The subcutaneous tissues are incised down to the fascia of the anterior tibialis muscle. A fascial incision is made from the lateral aspect of the tibial tubercle, sloping up proximally to the distal aspect of Gerdy’s tubercle, and is then extended distally and laterally to the anterior bare area of the fibula. The bare area of the fibula is an important and safe anatomic landmark. The FCL and peroneal nerve are usually safely avoided when one identifies this bare area.

B

Figure 23K-22   A and B, The use of a commercial (Arthrex Opening Wedge Osteotomy System, Arthrex, Inc., Naples, Fla) osteotomy wedge to open the osteotomy site gradually. (Redrawn from Noyes FR, Barber-Westin SD, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson RW [ed]: Master Techniques in Orthopaedic Surgery, Reconstructive Knee Surgery. Philadelphia, Lippincott Williams & Wilkins, 2003, pp 229-260.) Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1822 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

A

B

Figure 23K-23   A, The opening wedge triangular osteotomy opening is shown, along with the harvested and prepared (inset) iliac crest bone grafts. The bone segments are oversized by 1 to 2 mm to achieve impaction into the osteotomy site. One graft is placed along the posterior tibial cortex at the osteotomy site (A), one is wedged beneath the buttress plate (B), one graft is placed anteriorly at the tibial tubercle (which is one-half the width of the posterior segment to maintain tibial slope) (C), and the remaining bone graft is packed keep into the osteotomy site to promote healing (D). B, The final appearance at the osteotomy site. The iliac crest bone grafts have been placed to restore the entire medial cortex from posterior to anterior to provide a buttress at the osteotomy site. The authors now use a Tomofix plate with locking screws to add stability to the osteotomy site. (From Noyes FR, Mayfield W, Barber-Westin SD, et al: Opening wedge high tibial osteotomy: An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34:1262-1273, 2006.)

Subperiosteal dissection on the tibia is begun just ­distally to Gerdy’s tubercle, using a scalpel followed by a Cobb ­elevator. The dissection is continued just lateral to the patellar tendon and should be done cleanly, with a relatively bloodless field and without damage to muscle tissue. The retinaculum adjacent to the patellar tendon is excised and the retropatellar space entered. The patellar tendon is retracted anteriorly. The dissection is continued posteriorly on the tibia. It is important to remain in this subperiosteal plane because of the proximity of neurovascular structures. The dissection is completed across the width of the tibia in one location and is then carefully extended proximally and distally in this safe subperiosteal plane. There are three basic options for the fibula when performing an HTO: proximal slide, proximal fibular osteotomy, and distal fibular osteotomy. A proximal slide (disruption of tibiofibular joint) is strongly contraindicated in ACL-deficient knees because this shortens the lateral and posterolateral ligament complex and may lead to posterolateral instability in an already unstable knee. Our procedure of choice is a proximal fibular osteotomy through the fibular neck region (Fig. 23K-26). Meticulous surgical technique and protection and palpation of the peroneal nerve are essential. In order to protect the peroneal nerve, the lateral and posterior peri­ osteal sleeve is carefully preserved and not retracted under tension. If there is any question as to potential damage to the peroneal nerve, exposure of the nerve is indicated with direct visualization and protection. The fibula bone wedge that is removed is 2 to 3 mm less than the computed size of the

tibial wedge to allow compaction of the fibular osteotomy. Excellent bony apposition is achieved when the osteotomy is closed, and it is not necessary to add internal fixation. The proximal tibial closing wedge osteotomy may be performed using a commercially available calibrated osteotomy guide system (NexGen Osteotomy System, Zimmer, Warsaw, Ind). As an alternative, bone cuts can be determined using a free-hand method (Fig. 23K-27). A smooth guide pin is placed transversely just through but not beyond the medial cortex, 25 mm distal to the joint line with the use of fluoroscopy to ensure proper placement. It is critical to leave 25 mm of proximal tibia to avoid a tibial plateau fracture. The transverse length of the guide pin determines proximal tibial width. The method described by Slocum and colleagues71 involving a series of congruent right triangles is used to determine the entry point of the second guidewire, again confirming proper positioning of the wires with fluoroscopy. These guide pins determine the osteotomy triangular bone wedge that is removed to achieve the desired correction. We prefer to make our osteotomy cuts using a micro­oscillating saw to cut only the outer cortex and then complete the osteotomy with a thin osteotome. An oscillating saw in the cancellous bone can wander and potentially change the correction angle. A malleable retractor is placed in the subperiosteal tissues posteriorly to protect the neurovascular structures, and the knee is flexed 15 degrees. The lateral one half of the wedge is removed as a single piece. This wedge of bone may occasionally need to be replaced if an inadvertent overcorrection is obtained at surgery. The remaining wedge

Knee 1823

Authors’ Preferred Method—cont’d

A

B

of bone is removed under direct visualization. The surgeon is seated, using a headlamp to view the depth of the osteotomy. The osteotomy plane is maintained in a perpendicular manner, noting that at the midpoint of the tibia, the tibial width is one half of that at the lateral cortex.

C

Figure 23K-24   Demonstration of a 15-mm opening wedge osteotomy, with incorporation of a large tibial plate. A, Standing bilateral preoperative radiograph shows the desired opening wedge (gap) to be obtained at the proximal tibia. B, An ankle orthosis (AO) tibial side plate is shown, with cortical iliac crest autograft implanted. ������������ The authors now use a Tomofix plate with locking screws to add stability to the osteotomy site. �C, Immediate postoperative anteroposterior radiograph. (From Noyes FR, Mayfield W, Barber-Westin SD, et al: Opening wedge high tibial osteotomy: An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34:1262-1273, 2006.)

The medial 7 to 10 mm of the wedge adjacent to the medial cortex is not disturbed. The medial cortex is preserved to provide stability and prevent tibial medial or lateral translation or varus recurrence. Two to three perforations of the medial cortex with a guidewire are often required before Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1824 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

Inferior aspect of Gerdy’s tubercle Skin incision

Peroneal nerve Outline of fibular osteotomy

A

B

Figure 23K-25   A, An oblique skin incision is used in high tibial osteotomy (HTO), extending from the bare area of the fibula head to the tibial tubercle. B, The proximal fibula is subperiosteally dissected and the peroneal nerve identified and protected in preparation for the proximal fibular osteotomy. A wedge of bone is osteotomized with the micro-oscillating saw and osteotome and then removed. The posterior periosteum of the fibula is preserved. (Redrawn from Noyes FR, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson DW [ed]: Master Techniques in Orthopaedic Surgery, Reconstructive Knee Surgery. New York, Raven Press, 1995, pp 185-210.)

the osteotomy gap can be closed with a gentle valgus force. ­Apposition of the bony surfaces of the tibia and fibula should be visualized and inspected. It is important that the closing wedge technique (or opening wedge technique when used) does not increase or decrease the normal posterior tibial slope. An increase in the posterior tibial slope would result in a loss of normal knee extension and place higher forces on an ACL reconstruction. A decrease or reverse in the tibial slope (anterior tibia sloping distally) would produce hyperextension of the knee and place high forces on a PCL reconstruction. Using fluoroscopy, an alignment guide rod (rigid 3- to 4-mm rod, 1 m in length) is positioned over the center of the femoral head and the center of the tibiotalar joint to determine the newly corrected WBL intersection at the tibial plateau. A large single staple may be placed across the lateral tibial osteotomy site for provisional fixation. During this procedure, the lower limb is axially loaded to maintain closure of both tibiofemoral compartments. The knee is held at 5 degrees of flexion to avoid hyperextension. The alignment guide rod represents a new WBL, which should agree with preoperative calculations. If necessary, further bone may be removed from the osteotomy to adjust the WBL as required. Internal fixation of the osteotomy is achieved using an L-shaped plate. Two 6.5-mm cancellous screws are placed in the proximal tibia, and two to three cortical screws are placed distal to the osteotomy (Fig. 23K-28). Rarely, a 6.5-mm cancellous screw is placed in a lag fashion across the osteotomy site into the medial aspect of the proximal tibial bone for additional fixation. The final WBL determination is made using the rigid rod as a plumb line, as noted previously. The tourniquet is released and hemostasis obtained. The fascia of the anterior compartment musculature is loosely

reattached to the anterolateral aspect of the tibial border. A Hemovac drain is used for 24 hours. We perform HTO as either an outpatient procedure or with the patient leaving the hospital 23 hours after surgery. Postoperative Rehabilitation after High Tibial Osteotomy and Criteria for Return to Play

The protocol for rehabilitation following HTO is shown in Table 23K-8. The program includes immediate range of motion (0 to 90 degrees), quadriceps isometrics, straight leg raises, patellar mobilization, and electrical muscle stimulation. For the first week after surgery, ice or a commercial cooling system, mild compression, and elevation are used to prevent edema and swelling. Patients are ambulatory for short periods but are instructed to elevate their limb, remain home, and not resume usual activities. The prophylaxis for deep venous thrombosis (DVT) includes intermittent compression foot boots for both extremities, antiembolism stockings, ankle pumps, and aspirin (325 mg twice a day for 10 days). We advocate close observation for any signs of DVT, including abnormal calf and tenderness to palpation, tibial edema, and medial thigh tenderness. Doppler ultrasound is performed if a patient demonstrates any of these symptoms. A long-leg postoperative brace is worn for the first 8 postoperative weeks. Patients are allowed only toe-touch weightbearing for the first 3 weeks to prevent excessive forces at the osteotomy site. Weight-bearing is gradually increased to full by the eighth week if radiographs demonstrate adequate healing and maintenance of the HTO position. The protocol emphasizes quadriceps, hamstring, hip, and gastrocsoleus musculature strengthening. Closed-chain exercises

Knee 1825

Authors’ Preferred Method—cont’d

22–25 mm Proximal guide pin Distal guide pin Fibular osteotomy

Figure 23K-26   The proximal guide pin is placed parallel to the joint and 2.5 cm from it. The width of the proximal tibia is determined by measuring the length of this pin. The preoperative planning notes are consulted to determine the starting point for the distal guide pin with reference to the proximal pin, which has been computed for the desired angle of correction. The distance between the two guide pins is essentially the width of the base of the osteotomy wedge. The pins are inserted under image control either free-hand or using a calibrated osteotomy guide, which is our preferred technique. (Redrawn from Noyes FR, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson DW [ed]: Master Techniques in Orthopaedic Surgery, Reconstructive Knee Surgery. New York, Raven Press, 1995, pp 185-210.)

A

are begun on the fourth postoperative week, and weight machine exercises are begun on the seventh or eighth postoperative week. Excessive use of bicycling and weight machines is not allowed in patients with articular cartilage damage. By postoperative weeks 9 to 12, other aerobic conditioning exercises are begun as appropriate, including swimming, ski machines, and walking. Patients who express the desire to return to strenuous sports activities are advised of the risk for further cartilage deterioration if damage was present at the index procedure and are given medical consent only when muscle strength tests show that the strength of the quadriceps on the involved limb is at least 70% that of the contralateral limb, and no pain or swelling occurs with or after the activities. We strongly recommend, in most patients who undergo HTO, a return to only light recreational sports activities because of preexisting articular cartilage damage. Outcomes Measurement

Closing wedge osteotomy in ACL-deficient varus-angulated knees. We conducted a study that prospectively followed 41 patients (100% follow-up) an average of 58 months (range, 23 to 86 months) after closing wedge osteotomy.1 All patients also had ACL deficiency, of which 30 were treated with a reconstruction. Significant improvements were found for pain, swelling, and giving way. Preoperatively, 41% of the patients had moderate to severe pain with activities of daily living, whereas only 10% had this level of pain at follow-up. The decrease in pain could have occurred as a result of patient counseling on activity modification and the fact that none of the patients returned to strenuous jumping, pivoting, and cutting sports. Before surgery, 22 patients were participating in some form of sports activities; all with pain or functional limitations. At follow-up, 24 patients (59%) had returned to sports with no symptoms; however, most were biking or swimming

B

Figure 23K-27   A, The proximal tibial osteotomy is made by using the two guide pins as an external jig system to guide the osteotome. B, The outer half of the tibial width of the wedge of bone is removed as a single piece. (From Noyes FR, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson DW [ed]: Master Techniques in Orthopaedic Surgery, Reconstructive Knee Surgery. New York, Raven Press, 1995, pp 185-210.) Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1826 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

A

B

Figure 23K-28   Anteroposterior (A) and lateral (B) radiographs show the postoperative appearance after proximal tibial osteotomy, fibular osteotomy, and internal fixation. (From Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28:282-296, 2000.)

only. A statistically significant improvement was found between the preoperative and follow-up overall rating scores (scale, 0 to 100 points, P <.01). The mean increase between evaluations was 14 points (range, −8 to 38 points). A subgroup of 15 patients who had subchondral bone exposed in the medial tibiofemoral compartment was analyzed separately. These patients were followed a mean of 67 months (range, 36 to 89 months) after surgery. A statistically significant improvement was noted in the mean overall rating score at follow-up (P < .01). Nine patients returned to low-impact sports activities; the other 6 did not participate. Radiographic evaluation performed in the early postoperative period showed that 37 of 41 (90%) patients were surgically corrected with a WBL between 50% and 80%. Three patients drifted back into varus, and one settled into valgus. At the final follow-up examination, 25 patients (61%) still demonstrated optimal correction (mean WBL, 60%; range, 46% to 79%). However, 11 patients (27%) were in varus (mean WBL, 37%; range, 25% to 44%), and 5 (12%) had an increased valgus position (mean WBL, 90%; range, 81% to 108%). These findings indicate that in the varus knee, progression of the medial arthrosis continues, and long-term correction will not be achieved in one out of four patients. However, there may still be relief of pain symptoms in knees that return to a varus angulation. An interesting finding was that although 22 knees (54%) had abnormally increased lateral joint opening on varus

stress testing preoperatively, only 5 (12%) had this abnormal joint opening at follow-up. The osteotomy appeared to have unloaded the FCL and other posterolateral tissues, allowing physiologic remodeling and shortening to occur. Therefore, no associated posterolateral reconstruction was required during the subsequent ACL reconstruction. This clinical finding appears to support the recommendation to first perform the HTO in double-varus knees, and then assess knee motion limits and symptoms to determine whether a ligament reconstruction is required. There was no evidence of infection, peroneal nerve palsy, or tibial nonunion. One patient had a gentle manipulation under anesthesia 24 weeks following the HTO; full motion was successfully regained. Three patients required a repeat osteotomy due to complications. One patient settled into a valgus position 16 months postoperatively. One patient required an open reduction and internal fixation 4 weeks after surgery due to a loss of fixation at the osteotomy site. A falling injury 3 weeks postoperative caused a 4-mm collapse at the osteotomy site in another patient who underwent an opening wedge osteotomy. In all three patients, follow-up examinations performed 2 to 3 years after the revision procedures demonstrated WBLs and mechanical axes within the optimal range. One patient had a nonunion at the distal fibula osteotomy site; the tibial osteotomy site healed in a satisfactory manner. The fibula nonunion site was resected 12 months after surgery, and symptoms resolved.

Knee 1827

Authors’ Preferred Method—cont’d

   TABLE 23K-8  Cincinnati Sportsmedicine and Orthopaedic Center Rehabilitation Protocol Summary for High Tibial Osteotomy

Postoperative Weeks

Brace: Long-leg postoperative

1-2

3-4

5-6

7-8

X

X

X

X

Postoperative Months 9-12

4

5

6

Range of Motion Minimal Goals:

0-110° 0-135°

X

X

Weight-bearing:

None to toe touch 1/4 body weight 1/2 to 3/4 body weight Full

X

Patella mobilization

X

X

X

X

X X

X X

X X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X X

X X

X X X X X X

X X X X X X

X X X X X

X X X X

X X X X

X

X

X

X

X X X X X X

X X X X X X

X X X X X X

X X X X X X X X X

X

X

X

Modalities:

Electrical muscle stimulation Pain/edema management (cryotherapy) Stretching:

Hamstring, gastroc-soleus, iliotibial band, quadriceps Strengthening:

Quad isometrics, straight-leg raises, active knee extension Closed-chain: gait retraining, toe raises, wall sits, mini-squats Knee flexion hamstring curls (90°) Knee extension quads (90-30°) Hip abduction-adduction, multi-hip Leg press (70-10°)

X

Balance/Proprioceptive Training:

Weight-shifting, mini-trampoline, BAPS, KAT, plyometrics Conditioning:

UBE Bike (stationary) Aquatic program Swimming (kicking) Walking Stair-climbing machine Ski machine Running: straight Cutting: lateral carioca, figure 8s Full sports

X

X X X

X X X

BAPS, Biomechanical Ankle Platform System (BAPS, Camp, Jackson, Mich); KAT, Kinesthetic Awareness Trainer (Breg, Inc., Vista, Calif); UBE, upper body ergometer. From Noyes FR, Mayfield W, Barber-Westin SD, et al: Opening wedge high tibial osteotomy: An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34:1262-1273, 2006.

Treatment of double- and triple-varus knees: closing wedge osteotomy, ACL reconstruction, and posterolateral reconstruction. We conducted a second prospective investigation that followed 41 knees (23 double, 18 triple varus) a mean of 4.5 years (range, 2 to 12) after closing wedge osteotomy.39 Before referral, 19 ACL reconstructions had been done in 15 patients that had failed. Many of these ACL failures were attributed to an associated posterolateral deficiency that was not corrected at the time of the initial ACL reconstruction. Thirty patients (73%) had a partial or total medial meniscectomy before the HTO. In 17 patients (12 double, 5 triple varus), gait analysis testing was conducted preoperatively and a mean of 2 years following HTO. A control population of 28 age- and sex-matched

normal subjects was used for comparisons. All moments were normalized to the product of body weight ­ multiplied by height and were expressed as a percentage of that product. Thirty knees had abnormal articular cartilage lesions. Twenty-six (63%) had abnormal lesions in the medial compartment. In 21 (91%) of the double-varus knees, the ACL was reconstructed a mean of 9 months after the HTO, and in 2 knees, the HTO and ACL reconstruction were done ­simultaneously. In 13 (72%) of the triple-varus knees, the ACL was reconstructed a mean of 8 months after the HTO. One patient had the HTO and ACL reconstruction done simultaneously. Four patients had the ACL reconstruction done before the HTO in an attempt to avoid HTO. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1828 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

20 15

13

10 5 0

8

6

5

3

1 Severe ADL

Moderate ADL

13 10

10

0

3 0 Severe ADL

None ADL

Figure 23K-29   A statistically significant improvement was found for pain from preoperative to follow-up (P < .01).   Sports is defined as participation in light recreational activities such as swimming, bicycling, golfing. (Redrawn from Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28:282-296, 2000.)

2 Moderate ADL

None ADL

None Sports

13 knees were rated as functional, 4 as partially functional, and one as failed. Preoperatively, all patients with double-varus knees had abnormal increases in lateral joint opening (mean, 4 mm; range, 2 to 10 mm). At follow-up, no patient had more than 2 mm of increase in lateral joint opening, and none had an increase in external tibial rotation. All were rated as functional in lateral joint opening and external tibial rotation. There was no significant difference in the mean preoperative adduction moment between the double-varus and triplevarus knees (4.1% ±���������������������������������������� ����������������������������������������� 0.3% and 4.2% ������������������������� ±������������������������ 0.3%, ­ respectively). Full Giving-way 34

Preoperative Follow-up

35 30 25 20

17

15

11

10 5 0

None Sports

12

Figure 23K-30   A statistically significant improvement was found for swelling from preoperative to follow-up   (P < .01). Sports is defined as participation in light recreational activities such as swimming, bicycling, golfing. (Redrawn from Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28:282-296, 2000.)

31

25

15

15

15

5

Number of Patients

Number of Patients

30

20

40

Preoperative Follow-up

27

Preoperative Follow-up

25

Pain

35

Swelling

30

Number of Patients

All triple-varus knees had a posterolateral reconstruction; 12 (67%) had a proximal advancement of the posterolateral complex with the ACL reconstruction, and 6 had a FCL circle graft procedure. The knees that had the proximal advancement procedure had a definitive although lax FCL of normal width and integrity and intact popliteal muscle-tendon and fibular attachments, previously described as an indication for this procedure. The knees that had allograft reconstruction had extensive damage to posterolateral tissues. At follow-up, statistically significant improvements were found for pain (Fig. 23K-29), swelling (Fig. 23K-30), and giving way (Fig. 23K-31) (P < .001). Before the HTO, 18 patients (44%) had severe to moderate pain with activities of daily living, whereas at follow-up, only 7 (17%) had such pain. Overall, 29 patients (71%) improved their pain score, and 28 (68%) improved their swelling score. Giving way was eliminated in 85%. Twenty-seven patients (66%) were able to return to mostly low-impact athletics without symptoms. One patient rated the overall knee condition as normal; 14, as very good; 14, as good; 10, as fair; and 2, as poor. Statistically significant improvements were found in the mean overall rating score from preoperative to follow-up (63 ± 11 points and 82 ±��������������������������� ���������������������������� 14 points, respectively; P = .0001). The average increase was 20 ±������������������������������ ������������������������������� 10 points (range, 2 to 39). At follow-up, 19 knees (42%) had functional ACL reconstructions, 11 knees (24%) had partial function, and 15 knees (33%) failed. Ten of the 15 knees that failed represented ACL revision cases. A statistically significant difference was found in the failure rate for ACL revision cases compared with primary reconstruction cases (67% and 33%, respectively; P < .05). Preoperatively, all of the triple-varus knees had varus recurvatum, increases in lateral joint opening (mean, 8 mm; range, 3 to 15 mm), and increases in external tibial rotation (mean, 9 degrees; range, 3 to 15 degrees). At follow-up,

10 6

3 0 Severe ADL

1 Moderate ADL

None ADL

None Sports

Figure 23K-31   A statistically significant improvement was found for full giving way from preoperative to followup (P < .01). Sports is defined as participation in light recreational activities such as swimming, bicycling, golfing. (Redrawn from Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28:282-296, 2000.)

Knee 1829

Authors’ Preferred Method—cont’d 6

Normal Adduction Moment

Number of Knees

5

Above Normal Adduction Moment

4 3 2 1 0

2

2.5

3 3.5 4 4.5 5 5.5 6 Adduction Moment (% BW x Ht)

6.5

Figure 23K-32   The distribution of the preoperative adduction moments of the 17 patients tested. The mean adduction moment for the study group was 35% higher than that for the control group (P < .001). (Redrawn from Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28:282-296, 2000.)

The preoperative mean adduction moment of the study group was 35% higher (P < .001) than the control group value; 10 of the 17 patients (59%) had values greater than 1 standard deviation above control values (Fig. 23K-32). The study group also had a 22% higher calculated medial compartment load, and a 40% higher lateral ligament tensile force, compared with the control group (P < .01). Abovenormal medial compartment loads were predicted preoperatively in 71% of the involved knees, and above-normal lateral soft tissue forces during walking were predicted in 43% of the involved knees. Postoperatively, the adduction moment and lateral ligament tensile force decreased to significantly lower than control values. The medial compartment load decreased to values equal to those of controls (Fig. 23K-33). There was no evidence of infection, peroneal nerve palsy, patella infra, or knee motion limitations at follow-up. No patient required additional treatment intervention for losses of knee flexion or extension.

Preoperatively, the mean WBL was 22% (range, 3% to 49%) and the mean mechanical axis was −6.2 degrees (range, −12 degrees to −1 degree). At surgery, all knees were corrected to a WBL of 62%. In two knees, a valgus overcorrection occurred, and these were subsequently revised. At follow-up, 33 knees (80%) were in an acceptable position (mean WBL, 61%; range, 50% to 75%), 7 knees were in varus, and 1 knee was in valgus (WBL, 81%). Two knees required revision of the HTO. In both, the optimal correction in the 62% WBL range was obtained at surgery; however, excessive valgus (WBL, 86%) occurred with full weight-bearing. A revision opening wedge osteotomy was done 2 months after surgery in one knee and 6 months after surgery in the other. At follow-up, one knee had maintained the optimal WBL position, but the other had reassumed a varus position (WBL, 40%). One knee had a loss of internal fixation of the osteotomy on the fourth postoperative week that was corrected. The HTO healed without difficulty. Two patients required resection of a distal fibular painful nonunion osteotomy site 12 and 26 months, respectively, after surgery. Opening wedge osteotomy: an operative technique and rehabilitation program to decrease complications and promote early union and function. We conducted a third prospective study of 59 consecutive patients who had a medial opening wedge proximal tibial osteotomy.72 All but 4 patients were followed for at least 6 months after surgery, the minimal follow-up time period required for this study. Fifty-five patients were followed a mean of 20 months after surgery (range, 6 to 60 months). Independent physicians examined preoperative and postoperative radiographs for tibial slope and patellar height and postoperative radiographs for bony union. Radiographs were taken 4 and 8 weeks after surgery and then as required until bone consolidation was evident. Delayed union was defined as lack of bridging callus and presence of radiolucent areas within the opening wedge defect past a period of 3 months after surgery. In six knees, a concurrent operative procedure was performed and included ACL primary reconstruction in two knees, ACL revision reconstruction in one knee, ACL and MCL reconstructions in one knee, and an osteochondral 1

2

4

Preoperative Postoperative

0

Lateral Soft Tissue Force % BW

0

Medial Compartment Load % BW

Adduction Moment (% Bw x Ht)

1

0

Figure 23K-33   The preoperative and postoperative adduction moment, medial compartment load, and lateral ligament tensile forces for the 17 patients tested. Postoperatively, the adduction moment and lateral ligament tensile force decreased to significantly lower than control values. The medial compartment load decreased to values equal to those of controls. % BW, percent body weight; HT, height. (Redrawn from Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28:282-296, 2000.) Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1830 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

A

B

Figure 23K-34   Radiographs of a 48-year-old male patient treated with opening wedge osteotomy for varus malalignment and symptomatic medial tibiofemoral compartment arthrosis. Anteroposterior (A) and lateral (B) radiographs taken 5 weeks after surgery show the iliac crest bicortical bone graft and a 15-mm buttress plate, respectively. (From Noyes FR, Mayfield W, BarberWestin SD, et al: Opening wedge high tibial osteotomy: An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34:1262-1273, 2006.)

autograft transfer procedure on the medial femoral condyle in two knees. In nine knees, staged procedures were performed an average of 8 months (range, 3 to 19 months) after the osteotomy. These included knee ligament reconstructions of both the ACL and posterolateral complex structures in three knees; combined ACL, posterolateral complex, and PCL reconstruction in one knee; PCL reconstruction in two knees; and medial meniscus allografts in three knees. All nine knees achieved bony union at the osteotomy site and were full weight-bearing before undergoing the staged procedures. No complications occurred in these nine knees related to the HTO. Healing and union at the osteotomy site were radiographically evident an average of 3 months after surgery in 52 patients (95%, Fig. 23K-34). A delay in union (with no loss of fixation or correction) occurred in 3 patients (5%). The size of the opening wedge osteotomy in these 3 patients ranged from 11.0 mm to 16 mm. In 2 of these patients, a

POTENTIAL COMPLICATIONS Bony Instability, Teeter Effect Kettelkamp and colleagues first described the teeter effect as a contraindication to proximal tibial osteotomy.33 Excessive bone loss and concavity on the medial tibial plateau prohibits simultaneous weight-bearing on both plateaus after osteotomy and results in an unstable knee in the coronal

bone stimulator was applied, and union was achieved by 6 to 8 months after surgery. The other patient achieved union without intervention by 10 months. An early postoperative loss of fixation occurred in one patient who admitted to full weight-bearing immediately after surgery. The osteotomy was successfully revised 10 days after surgery and proceeded uneventfully to union. There were no instances of shortening of the patellar tendon related to a patella infra syndrome. There was no significant difference between the mean ­preoperative (9 ± 4 degrees; range, 2 to 16 degrees) and postoperative (10 ± 3 degrees; range, 3 to 21 degrees) tibial slope measurements. One patient with a PCL-deficient knee had an intentional increase in posterior slope. There were no deep infections, loss of knee motion requiring intervention, DVTs, nerve or arterial injury, fracture, or complications related to bone grafting. Full weight-bearing was achieved a mean of 8 weeks (range, 4 to 11 weeks) after surgery.

plane. A teeter effect occurs because tibiofemoral contact shifts, or teeters, from one plateau to another, depending on the relationship of the center of gravity to the center of the knee. We agree with these authors that osteotomy as a single corrective procedure is contraindicated when bone loss makes simultaneous contact of both plateaus impossible. Preoperative radiographic evaluation of the bone loss on the tibial plateaus should be performed, specifically evaluating the slope of the plateaus to determine whether loading of both compartments will occur after HTO.

Knee 1831

It has been suggested, but not experimentally confirmed, that if the combined cartilaginous and bony loss on the medial side is greater than 1 cm, it would be impossible to achieve loading of both the medial and lateral compartments after osteotomy. This situation can be evaluated preoperatively by examining the knee joint under image intensification. At the time of surgery, it is necessary to evaluate with image intensification the WBL with both compartments loaded to confirm that simultaneous medial and lateral tibiofemoral contact is present.

Inadequate or Loss of Axial Correction Inadequate or overcorrection of lower limb alignment has been reported by many authors. Matthews and colleagues reported that 7 of 40 patients (18%) had a partial or complete recurrence of the varus deformity within the followup period.73 Hernigou and colleagues reported that a varus deformity was found immediately after osteotomy in 10 of 93 knees; 10 to 13 years later, changes in alignment in a varus direction had occurred in 71 of 76 knees.9 Magyar and coworkers described a loss of correction after closing wedge osteotomy, with 9 of 16 knees collapsing into valgus by 1 year after surgery despite normal alignment achieved immediately postoperatively.74 Marti and associates followed 32 knees that had an opening wedge osteotomy without bone grafting.75 At follow-up of 24 to 62 months, 31% demonstrated a recurrence of the varus deformity, and 19% were in excessive valgus. Stuart and colleagues evaluated standing tibiofemoral angles, which decreased from a mean of 9.3 degrees of varus to 7.8 degrees of valgus at final follow-up.76 Using the Kaplan-Meier survival analysis method, the authors predicted that varus alignment was likely to occur in 18% of knees, lateral compartment arthrosis was likely to pro­ gress in 60%, and medial and lateral compartment arthrosis was likely to progress in 83% by 9 years after surgery. Loss of the axial alignment obtained at the time of surgery may be attributed to several factors. These include lack of internal fixation or the use of inadequate internal fixation with collapse of the distal fragments settling into the cancellous bone of the plateau. Late drifting into a varus position may be due to a progressive loss of the medial osteochondral cartilage complex, or to stretching of the lateral soft tissue structures. Coventry reported no loss of surgical correction obtained immediately postoperatively.77 He attributed the success of the procedure to stability of the closing wedge, the hinge effect of the medial periosteum and cortex, internal fixation, and avoidance of a shift of the fragments.77 There are several surgeon-controlled factors that can help minimize the risk for inadequate or loss of correction. Careful preoperative planning is required to avoid inadequate correction; it begins with the calculation of the mechanical or anatomic axis using full-length standing radiographs. Stress radiographs may be used to evaluate medial osteocartilaginous complex loss and lateral joint opening, as previously described.34 Calculations to determine the amount of bone to be resected should be performed well before the osteotomy. During surgery, confirmation of adequate correction that was determined by preoperative measurements (WBL,

62%) should be accomplished using image intensification or computer navigation when possible. The knee will be placed in an overall excessive valgus position if the amount of lateral opening is not evaluated and if too much bone is resected. Even though an ideal position may be verified at surgery, a change in alignment may be detected when postoperative standing radiographs are obtained. The alignment should be verified by the fourth postoperative week under partial weight-bearing conditions.

Delayed Union and Nonunion Delayed union or nonunion is possible after HTO. In 1974, Jackson and Waugh reported that 19 of 226 (8%) patients undergoing operation had a delayed union; 14 of these patients were immobilized for more than 12 weeks before union was achieved, and the remaining 5 required bone grafting.78 Warden and colleagues reported a low incidence of delayed union (6.6%) and nonunion (1.8%) in 188 opening wedge osteotomies in which an iliac crest bone autograft was used in the majority of knees.79 There was a trend toward an increased incidence of delayed union or nonunion when a coral wedge was used either alone or in conjunction with the autogenous graft (15%, 5 of 33 knees) compared with when an iliac crest graft was used alone (6%, 8 of 128 knees). No problems with union were reported by Marti and colleagues,67 Pace and Hofmann,63 or Patond and Lokhande80 following opening wedge osteotomy and iliac crest bone autograft procedures. The occurrence of healing difficulties can be reduced drastically by following several technical suggestions for closing wedge osteotomies. First, the osteotomy should be performed proximal to the tibial tubercle to increase the amount of cancellous bone surface contact, which will enhance healing and increase inherent stability. The two surfaces should be cut in a manner that will maximize the amount of surface area that will be in opposition. Internal fixation is also believed to enhance healing, particularly when loss of the medial cortex has occurred. In opening wedge osteotomies, a stable construct created by the use of an iliac crest autogenous bone graft (in the anterior, mid, and posterior portions of the osteotomy) and appropriate plate fixation with protection of the lateral tibial buttress (cortex) will sustain postoperative compressive and torsional loads. Plates with different designs should be available during surgery. Staubli and colleagues described the use of an AO medial tibial plate with locking screws (Tomofix) for opening wedge osteotomy without bone grafting.68 Although the authors reported a small incidence of delayed union (2%) and loss of correction (2%) in their series of 92 cases, 40% required removal of the implant. Autogenous bone grafting is recommended for large osteotomies, such as those larger than 10 mm, to facilitate healing and allow earlier weight-bearing. Allografts may be considered in smaller osteotomies, provided a plate with locking screws is used. Many investigators have demonstrated that external fixators can accomplish adequate union and correction, such as the series reported by Sterett and Steadman in which only 1 of 33 patients had a loss of correction early after surgery.58 However, pin tract infections were reported in 45% by these authors, which was similar to the pin tract

�rthopaedic ����������� S �ports ������ � Medicine ������� 1832 DeLee & Drez’s� O

infection rate reported by Weale and colleagues59 of 38% (28 of 73 knees) and Magyar and associates81 of 51% (157 of 308 cases). We use toe-touch weight-bearing in the initial 3-week postoperative period and then permit progression over the next 5 weeks to full weight-bearing based on radiographic signs of osteotomy healing. A delayed union can be treated, if the overall alignment is acceptable, by electric stimulation.

Tibial Plateau Fracture The incidence of tibial plateau fracture following closing wedge osteotomy has been reported to range from 1% to 20%.9,73 This complication appears uncommon after opening wedge osteotomy, although Amendola and coworkers reported 7 (19%) of 37 patients in their initial series of opening wedge osteotomies had intra-articular fractures that extended into the lateral compartment.66 The fractures were believed to be caused by a combination of a vertical osteotomy site (closer to the lateral tibial plateau joint line than the lateral cortex) and use of thick osteotomes. After adjusting the obliquity of the osteotomy and switching to thin, flexible osteotomes, the authors did not experience any further cases of fracture. Fracture of the tibial plateau can be avoided by meticulously following the previously described operative technique. The proximal osteotomy should remain parallel to the joint surface. It is important to keep the proximal tibial portion 20 to 25 mm thick to avoid fracture and create a thin medial tibial portion after osteotomy. The cortex on the applicable side of the wedge may be penetrated by multiple drill holes, and forceful closure of the wedge should be avoided. If plateau fracture does occur, anatomic reduction and internal fixation of the plateau fragments should be performed and osteotomy of the apical cortex completed.

Arterial Injury Vascular complications are exceedingly rare; there have been reports of single episodes of anterior tibial artery injury during surgery. Bauer and associates reported on 1 case in 60 in which the anterior tibial artery was severed during the resection of the fibular head.82 Recognition of the anatomic relationship of the vessels to the surgical dissection is imperative. The anterior tibial artery is at risk because it pierces the proximal interosseous membrane. Injury to this artery may result in an anterior compartment syndrome, which must be recognized and dealt with promptly. The popliteal artery is at risk during posterior dissection and osteotomy of the posterior tibial cortex. By flexing the knee and gently retracting the popliteal structures, the risk for injury to these structures is significantly reduced.

Peroneal Nerve Injury or Palsy In 1974, Jackson and Waugh reported 27 of 226 (12%) cases of partial or complete injury to the peroneal nerve during osteotomy.78 There was a higher incidence of injury during curved osteotomies located below the tibial tuberosity; the authors concluded that osteotomy in this location was too dangerous. A 7% incidence of peroneal nerve injury was reported by Sundaram, and colleagues,83

and a 6% incidence was described by Harris and Kostuik.84 Slawski and associates reported a 4.3% incidence of peroneal neurapraxias in 225 pediatric tibial osteotomies.85 Flierl and colleagues compared the incidence of neurologic complications after opening wedge osteotomy between a technique using a conventional oscillating saw and a technique that created multiple drill holes and osteoclasis.86 After the conventional method, acute transient peroneal nerve palsy developed in 15.7% of 89 patients, with persistent deficits found in 12.4% 6 months after surgery. In the osteoclasis group, 14% had acute transient events, and 4.7% reported permanent weakness. Other than the report by Flierl and colleagues,86 nerve injury after opening wedge HTO appears to be extremely rare. Peroneal nerve palsy may result from several causes; the most common is a cast or bandage that is applied too tightly after surgery. The use of internal fixation alleviates the need for casting. The nerve may also be directly injured during surgery. Osteotomy of the fibula is performed at different sites by different authors depending on the operative technique. There is a risk for injuring the peroneal nerve if the osteotomy is performed in the proximal third of the fibula.87 Osteotomy in the middle portion of the fibula may injure the peroneal nerve innervation to the extensor hallicis longus. The rate of injury is extremely high (25%) after dome osteotomy below the tibial tubercle78; it is our opinion that this procedure should not be performed for this reason.

Arthrofibrosis and Patella Infera Early HTO studies cited unacceptably high rates of arthrofibrosis and patella infera following postoperative immobilization. Windsor and associates reported on 45 patients who underwent total knee arthroplasty following high tibial osteotomy.88 The authors found that 80% of the knees had a patella infera following HTO and postulated that the postoperative cast immobilization of the knee after the osteotomy allowed the quadriceps muscle and patellar ligament to relax and shorten. Westrich and colleagues reported that 16 of 34 knees (47%) that underwent closing wedge high tibial osteotomy followed by cast immobilization developed patella infera (Insall-Salvati index) postoperatively.89 In the same series, only 3 of 35 knees (8%) that received immediate motion postoperatively also developed patella infera postoperatively. Scuderi and coworkers found that 11 of 66 knees treated with cast immobilization after closing wedge osteotomy had developed patella infera postoperatively.90 No correlation was found between the development of this complication and the necessity for eventual TKA. A decrease in the patellar height ratio as measured by the Blackburn-Peel ratio is expected after opening wedge osteotomy.91 Wright and associates reported that all 28 patients in their series of medial opening wedge osteotomies had a decrease in patellar height, but no significant change in patellar ligament length (as measured by the Insall-Salvati ratio92).93 The authors explained that because the osteotomy increased the distance between the tibial tubercle and tibial articular surface, the patella migrated distally. Similar findings in decreased patellar height following opening wedge osteotomy have been reported by

Knee 1833

­ thers.60,91,94-96 Patella infera correlated with the magnio tude of angular correction after HTO in two series.93,96 For example, Wright and colleagues evaluated patellar height and ligament length in 28 patients and found a relationship between the amount of infera and the degree of angular correction.93 Postoperatively, lateral radiographs are taken to detect any decrease in the patellar vertical height ratio; these are repeated often for any patient who shows early signs of developmental patella infera (inability to perform a strong quadriceps contraction after surgery, decreased patellar mobility, decreased palpable tension in the patellar tendon with failure of the patella to displace proximally on quadriceps contraction, or distal malposition of the involved patella compared with the opposite side). Rigid internal fixation should also help to reduce the occurrence of arthrofibrosis and a resultant patella infera. An immediate knee motion program and exercise protocol of straight-leg raises, multiangle isometrics, and electrical muscle stimulation are advocated to decrease the incidence of quadriceps weakness and knee motion limitation following HTO. In addition, a phased treatment program is begun for limitations of motion early in the postoperative course when restriction of either extension or flexion is noted.97

Iliac Crest Harvest Site Pain Most of the complications described in the literature related to iliac crest bone graft harvest may be avoided by the surgical technique described in this chapter. For example, the dissection is limited to include only 10 mm of the superior iliac crest. A meticulous subperiosteal exposure of the outer iliac crest is performed without violating the muscle plane. The inner iliac cortex is never dissected, and the muscle attachments are kept intact. This minimally invasive harvest technique avoids the frequency of complications reported by larger exposures, such as those used for spine fusions. The standard cortical iliac crest graft harvested is 40 mm in length, 12 mm in width, and 30 mm in depth, although larger osteotomies require a longer graft of about 45 mm. The iliac crest harvest site can be painful, with trunk flexion activities for up to 4 weeks postoperatively, and patients are advised accordingly. Even though there were no complications in the authors’ series, added time is required for the operative procedure, and a complication is always possible, which must be weighed against the benefits of the rapid healing provided by autogenous bone at the osteotomy site.

Deep Venous Thrombosis The incidence of DVT has not been adequately studied following HTO. Turner and coworkers demonstrated a 41% rate of DVT using venography after HTO in 84 patients; only 15% were clinically diagnosed.98 Only calf clots were diagnosed clinically; there were 3 proximal and 12 mixed clots diagnosed only with venogram. Leclerc and associates performed a randomized, prospective trial comparing lowmolecular-weight heparin (LMWH) to placebo after 129 HTO.99 The incidence of DVT was 17% in the LMWH

group, compared with 58% in placebo. Nineteen percent of the placebo group had femoral vein clots, compared with none in the treatment group. Thus, it appears that the incidence of DVT following HTO is similar to that for other large reconstructive knee procedures such as TKA. Patients are treated with prophylaxis against DVT after surgery. Sequential compression foot or limb venous compression devices and TED hose are used immediately following surgery. Aspirin is prescribed and, in high-risk patients, LMWH or warfarin sodium (Coumadin). Initially, rehabilitation emphasizes quadriceps isometrics, range of knee motion, ankle pumping, and ambulation. The surgeon should have a high suspicion for DVT and a low threshold for ordering ultrasound evaluation in the postoperative period.

SPECIAL POPULATIONS Our classification system of the primary-, double-, and triple-varus knee allows for rationale treatment recommendations to be made according to each category. The issues of timing and requirement for ligament reconstructive procedures are summarized in Table 23K-6. We prefer to stage osteotomy and ligament reconstructive procedures for several reasons. First, not all patients with double-varus knees and ACL deficiency require ACL reconstruction. In our experience, patients who undergo ACL reconstruction typically have giving-way symptoms with daily or light sports activities. Patients who modify their sports activities after osteotomy and who do not experience instability may not require this procedure. We profile each patient individually regarding the level of activity he or she wishes to resume after osteotomy. Second, our studies demonstrate that double-varus knees with a posterolateral deficiency do not require a posterolateral reconstructive procedure because the HTO allows adaptive shortening and remodeling of the FCL and other soft tissues. Clinical examination performed a few months after surgery will demonstrate the return of normal lateral joint opening and external tibial rotation knee motion limits. We have also found that in triple-varus knees that require HTO and combined ACL and posterolateral reconstruction, staging these major procedures lessens the risk for postoperative complications. We prefer a bone–patellar tendon–bone anatomic graft replacement of the FCL and plication or advancement of the posterolateral capsule, described in detail elsewhere (Fig. 23K-35).100 In some cases, a second graft reconstruction of the PMTL is also required (Fig. 23K-36). Thus, the requirement of one to two grafts for the posterolateral reconstruction, along with the ACL reconstruction, produces a large operative procedure, which we believe should be performed after the patient has recovered from the osteotomy. We have also noted that there are two distinct groups of young, active patients in whom the goals of HTO may differ. One group has significant medial tibiofemoral arthrosis and pain and swelling with daily activities. The goal of HTO for these patients is to diminish these symptoms, not to return to athletics. These patients are counseled regarding the extent of the disease process and the goal of the HTO, which is to buy time until total joint replacement is required.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1834 DeLee & Drez’s� O Figure 23K-35   Anatomic substitution of the   fibular collateral ligament (FCL) with a bone–patellar   tendon–bone autograft or allograft showing two methods for fibular graft fixation. A, Two small fragment screws are used to fix the bone into a slot created in the proximal fibula. Interference screw fixation is used at the femoral anatomic site of the FCL. B, A fibular tunnel is made, graft is seated, and an interference screw is used for fixation. A backup suture post is shown, which is usually not required but is an option. (Modified from Noyes FR, Barber-Westin SD: Treatment of complex injuries to the posterior cruciate and posterolateral ligaments of the knee. Am J Knee Surg 9:200, 1996.)

A

Figure 23K-36   Anatomic popliteus   muscle–tendon–ligament reconstruction and fibular collateral ligament (FCL) reconstruction with   bone–patellar tendon–bone autograft or allograft. A, A soft tissue interference screw is used for tibial fixation of the popliteus graft. B, Passage of popliteus graft beneath the FCL bone–patellar tendon–bone graft. The posterior exit tunnel is at the posterolateral corner of the tibia. C, Final fixation of the popliteus and FCL graft reconstructions. An interference screw is shown on the fibula for fixation. An alternative procedure is to use two small fragment cortical screws. D, Suture of popliteus graft to posterior margin of the FCL graft at the fibular attachment site to restore the popliteofibular ligament. E, Suture plication of posterolateral capsule to posterior margin of the FCL graft. (Redrawn from Noyes FR, Barber-Westin SD, Roberts CS: High tibial osteotomy in knees with associated chronic ligament deficiencies. In Jackson DW [ed]: Reconstructive Knee Surgery, 2nd ed. Philadelphia: Lippincott Williams & Wilkins, 2003, pp 229-260.)

B

Knee 1835

The second group of patients does not have advanced joint arthrosis; their pain symptoms are of relatively recent onset and occur with sports or manual labor activities. These individuals desire HTO to continue some form of athletics, or they may have a strenuous occupation. These patients also usually require an ACL reconstruction because they have higher activity demands and frequently have experienced a giving-way injury. However, data are not available that accurately predict whether athletics are advisable after successful HTO. Large in vivo joint loading occurs with athletics, and patients who resume these activities do so at risk for further joint damage. Many patients in our experience who require osteotomy also require either a meniscus transplant,101 due to a prior meniscectomy, or an articular cartilage restorative procedure. In these cases, the osteotomy is performed first to ensure that adequate lower limb alignment and successful transfer of weight-bearing loads to the lateral tibiofemoral compartment is achieved. Then, several months later, a meniscus transplant or articular cartilage procedure (osteochondral autograft transfer or autologous chondrocyte implantation) is performed. We prefer this staged approach also to lessen the risk for complications, especially arthrofibrosis. One final distinct group consists of patients with a highly abnormal tibial slope, greater than 1 standard deviation above the mean. In these knees, a biplanar osteotomy is required.

C

r i t i c a l

P

o i n t s

l Classification includes primary-, double-, and triple-varus knee. l Preoperative planning involves diagnosing all ligamentous deficiencies; calculating of the weight-bearing line for desired angular correction; and checking for abnormal patella infera, alta, and tibial slope. l Operative technique includes meticulous, autogenous bone graft and secure internal fixation, with confirmation of angular correction at surgery.

l Observe the patient postoperatively for deep venous thrombosis. l Rehabilitation includes immediate knee motion, bracing for 8 weeks, partial weight-bearing after 3 weeks; and full weight-bearing after 8 weeks. Most patients return to low-impact activities only. l Ensure that correction obtained and maintained postoperatively. l Stage the osteotomy in double- and triple-varus knees. Stage the osteotomy in knees that require meniscus transplantation or articular cartilage restorative procedures.

S U G G E S T E D

R E A D I N G S

Dugdale TW, Noyes FR, Styer D: Preoperative planning for high tibial osteotomy: The effect of lateral tibiofemoral separation and tibiofemoral length. Clin ­Orthop 274:248-264, 1992. Noyes FR, Barber-Westin SD: Posterolateral knee reconstruction with an anatomical bone-patellar tendon-bone reconstruction of the fibular collateral ligament. Am J Sports Med 35:259-273, 2007. Noyes FR, Barber-Westin SD, Albright JC: An analysis of the causes of failure in 57 consecutive posterolateral operative procedures. Am J Sports Med 34(9): 1419-1430, 2006. Noyes FR, Barber-Westin SD, Hewett TE: High tibial osteotomy and ligament reconstruction for varus angulated anterior cruciate ligament-deficient knees. Am J Sports Med 28(3):282-296, 2000. Noyes FR, Barber SD, Simon R: High tibial osteotomy and ligament reconstruction in varus angulated, anterior cruciate ligament-deficient knees: A two- to sevenyear follow-up study. Am J Sports Med 21(1):2-12, 1993. Noyes FR, Dunworth LA, Andriacchi TP, et al: Knee hyperextension gait abnormalities in unstable knees: Recognition and preoperative gait retraining. Am J Sports Med 24(1):35-45, 1996. Noyes FR, Goebel SX, West J: Opening wedge tibial osteotomy: The 3-triangle method to correct axial alignment and tibial slope. Am J Sports Med 33(3):378387, 2005. Noyes FR, Mayfield W, Barber-Westin SD, et al: Opening wedge high tibial osteotomy: An operative technique and rehabilitation program to decrease complications and promote early union and function. Am J Sports Med 34(8):1262-1273, 2006. Noyes FR, Schipplein OD, Andriacchi TP, et al: The anterior cruciate ligamentdeficient knee with varus alignment: An analysis of gait adaptations and dynamic joint loadings. Am J Sports Med 20(6):707-716, 1992. Noyes FR, Stowers SF, Grood ES, et al: Posterior subluxations of the medial and lateral tibiofemoral compartments: An in vitro ligament sectioning study in ­cadaveric knees. Am J Sports Med 21(3):407-414, 1993.

R efere n ces Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 1836 DeLee & Drez’s� O

S e c t i o n

L

Vascular Problems—Popliteal Artery Entrapment Turner C. Lisle and Irving L. Kron

HISTORICAL PERSPECTIVE Popliteal artery entrapment was first described in 1879 by Anderson Stuart, an Edinburgh medical student, who noted an “aberrant course” of the popliteal artery while dissecting an amputated leg.1 In 1925, Chambardel-Dubreuil described a case in which the popliteal artery was separated by an accessory gastrocnemius muscle.2 Some 80 years later, the first clinical case of popliteal artery entrapment syndrome was published by Hamming and Vink, who described a young patient with intermittent claudication due to an anomalous course of the popliteal artery.3 They performed the first operative decompression at Leyden University later that year on the very same patient. There remains debate as to who first coined the term popliteal artery entrapment syndrome (PAES), with Carter and Eban publishing their description in 1964 followed closely by Love and Whelan, of Walter Reed Army General Hospital, with their account in 1965.4,5 After these initial descriptions, numerous isolated cases were published throughout the literature.4-20 To date, more than 350 cases have been published worldwide. With these initial reports, the incidence of popliteal artery entrapment was thought to be quite rare.4,5,21-26 In the late 1970s, an autopsy study by Gibson and colleagues showed an incidence of 3.5%.25 Other case series published in the early 1980s and 1990s indicated an incidence of 0.165%.24,27 Despite these and other series, the true incidence of PAES is largely unknown. In the past, this condition was vastly underdiagnosed. However, with an increasing awareness of the condition, particularly in the young healthy patient with intermittent claudication, more and more cases are being recognized.28

RELEVANT ANATOMY AND BIOMECHANICS Popliteal artery entrapment is the result of a developmental abnormality, a concept that is well accepted among treating physicians.29 In the developing fetus, the upper and lower limb buds initially receive their blood supply from a single axial artery. In the lower extremity, the embryologic popliteal artery is a mere continuation of this axial artery, known as the ischiadic artery. At about 6 to 10 weeks of development, the lower limb bud undergoes medial rotation and extension. As the limb bud extends and internally rotates, the ischiadic artery involutes, giving rise to the newly formed external iliac and superficial femoral arteries. The popliteal artery forms from components of both this

femoral artery as well as islands of remnant axial artery. At the knee, the proximal popliteal artery develops as a continuation of the femoral artery, whereas the midpopliteal artery is derived directly from residual axial artery. The distal popliteal artery arises from the developing midpop­ liteal artery beneath the popliteus muscle before involuting later in development, only to redevelop later, superficial to the popliteus muscle, through the coalescence of anterior and posterior tibial arteries.30-32 As these changes in arterial development are taking place, muscular development is occurring as well. The gastrocnemius muscle arises proximally on the lower limb bud from the posterior aspect of the fibula and lateral tibia. As the popliteal artery develops, the muscle mass that is destined to become the medial head of the gastrocnemius begins a medial migration and eventual insertion on the posterior surface of the femur, just behind the medial femoral condyle. Normally, this migration follows the involution of the distal portion of the popliteal artery. After this medial migration, as noted earlier, the definitive distal popliteal artery re-forms. This complex relationship between the development of lower extremity arterial structures and muscle sets the stage for potential development of popliteal artery entrapment.30-32

CLASSIFICATION Several classification schemes have been described for PAES. Insua and colleagues published the first classification scheme in 1970.33 Their system was based on two main types of abnormalities, with variations of each based on the course of the popliteal artery in relation to the medial head of the gastrocnemius muscle, a trend on which the classification systems that followed are largely based. In 1971, a more straightforward classification system for PAES was introduced by Delaney and colleagues that included four basic types of entrapments (Table 23L-1).15 This rather simplified classification scheme was later revised by Whelan,34 and in 1979, Rich and colleagues added a fifth group to incorporate those entrapments that included the popliteal vein as well as artery.26 In 1980, Ferrero and colleagues introduced a highly complex classification system that included more than 10 separate abnormalities.35 Although more complete in its descriptions, this classification scheme has failed to become clinically useful. In 1985, Rignault and colleagues introduced the term functional entrapment, in which patients experience symptoms of popliteal arterial entrapment but have no identifiable anatomic abnormality.36 In 1998, the Popliteal Vascular Entrapment Forum was established in part to clarify the criteria by which PAES was classified. The

Knee 1837

   TABLE 23L-1  Delaney and Colleagues’ Classification of Popliteal Artery Entrapment Syndrome Type

Anatomic Description

I

Popliteal artery lies medial to the medial head of the gastrocnemius, w������������������������������������������ h����������������������������������������� ich lies in its normal location. This is associated with a marked medial deviation ��������������������������� of the artery in the poplit���������� eal fossa. Popliteal artery descends medially and bene������� ath an abnormally (laterally) pl������������������������ aced medial head of the gastrocnemius (the arterial course is vertical with no exaggerated looping). Popliteal artery is compressed by an accessory slip of muscle or by fibrous or tendinous bands arising from the medial head of the gastrocnemius. Popliteal artery lies deep to and is entrapped by the deeper popliteus muscle or fibrous bands.

II

III IV

From Delaney TA, Gonzalez LL: Occlusion of popliteal artery due to muscular entrapment. Surgery 69(1):97-101, 1971.

result was a simplistic classification scheme, based largely on the work done by Delaney, Whelan, Rich, and colleagues (Table 23L-2).15,26,34,36 This classification scheme, illustrated in Figure 23L-1, is the most widely accepted system among physicians who treat PAES. The first and most common type of entrapment is type I, in which there is marked medial deviation of the popliteal artery around the medial head of the gastrocnemius muscle. If the artery is displaced medially but to a lesser degree compared with type I, and there is a variable attachment of the medial head of the gastrocnemius to either the lateral aspect of the medial femoral condyle, the intercondylar area, or the lower femur above the condyle, type II entrapment is the designation. In type III entrapment, the medial head of the gastrocnemius has additional muscular, tendinous, or fibrous attachments on its lateral side arising from the intercondylar area, which compress the normally located popliteal artery. In type

   TABLE 23L-2  Popliteal Vascular Entrapment Forum

Classification of Popliteal Artery Entrapment Syndrome Type

Anatomic Description

I

Marked medial deviation of the popliteal artery around the medial head of the gastrocnemius Variable attachment of the medial head of the gastrocnemius to the lateral aspect of the medial femoral condyle, intercondylar area, or lower femur above the lateral condyle. Artery is displaced medially but to a lesser degree when compared with type I. Medial head of the gastrocnemius has an additional musculotendinous slip on its lateral side arising from the intercondylar area compressing the normally located popliteal artery. Popliteal artery is compressed as it passes underneath the popliteus muscle or the fibrous bands within the popliteal fossa. Primary venous entrapment Variants not described by types I to V Functional entrapment. Symptomatic individuals show compressive symptoms with stress maneuvers but in whom there is no anatomic abnormality.

II

III

IV V VI F

From di Marzo L, Cavallaro A: Popliteal vascular entrapment. World J Surg 29(Suppl 1):S43-45, 2005.

IV entrapment, the normally located popliteal artery is compressed as it passes beneath the popliteus muscle or associated fibrous bands within the popliteal fossa. If the entrapment includes the popliteal vein in addition to the popliteal artery, the entrapment is classified as type V. Type VI is reserved for variants that may be found on advanced imaging that are not described elsewhere. Finally, type F entrapment, or functional entrapment, is a designation given to patients who have compressive symptoms with activity but have no identifiable anatomic abnormality on imaging or at exploration.

EVALUATION Clinical Presentation and History The classic presentation of popliteal artery entrapment syndrome is one of a young, active, otherwise healthy individual with intermittent claudication in the calf and foot. It has been reported that claudication alone occurs in as many as 69% to 90% of patients with PAES.37-39 Symptom onset is often rapid and experienced with intense lower extremity physical activity, such as marathon running.4,5,40 It is thought that this activity leads to a high degree of muscle development that, in turn, may unmask occult pathology.41 Some case studies have shown claudication resulting from walking rather than running, although this is the exception rather than the rule.7,42 Chronic lower limb ischemia is experienced in less than 10% of patients.24,43 Symptoms of chronic ischemia include paresthesias, discoloration of the foot or toes, poikilothermia, and in the most severe cases, rest pain and tissue necrosis. These patients however, tend to lack the typical diffuse ischemic symptoms seen in a patient with atherosclerosis as the cause of limb ischemia, a key differentiation point in the diagnostic algorithm of PAES. Claudication is unilateral in most patients with PAES. However, it has been shown that that once the diagnosis of PAES has been made, the contralateral limb will show entrapment abnormalities in greater than 57% of patients.44,45 Historically, PAES was thought to occur nearly exclusively in the male population. However, more recent literature suggests a male-to-female ratio of 2:1.45 In children, the presence of PAES is seemingly just as rare, if not more so. Most cases of pediatric PAES appear to occur between the ages of 16 and 17 years, and as of 1995, only 40 cases have been published (Box 23L-1).46

Physical Examination and Testing In most cases, PAES can be suspected with a high degree of certainty from the patient’s history alone. Nevertheless, the physical examination can give the physician valuable information regarding the severity and distribution of the entrapment. Femoral, popliteal, posterior tibial, and dorsalis pedis pulses should be carefully scrutinized for any pulse abnormality, both unilaterally and contralaterally. At rest, patients with PAES frequently have a normal pulse examination. Simple maneuvers, such as having the patient passively dorsiflex and actively plantar flex the ankle can bring out subtle pulse amplitude differences within the popliteal arteries bilaterally. In fact, the presence of pulse

�rthopaedic ����������� S �ports ������ � Medicine ������� 1838 DeLee & Drez’s� O

Figure 23L-1   Normal anatomy of the popliteal fossa and the classification types of popliteal artery entrapment syndrome according to the Popliteal Vascular Entrapment Forum, 1998.

deficits with these maneuvers has been considered pathognomonic for PAES by some authors,47 although similar findings have been shown in normal patients.48 In patients with no palpable differences, auscultation of a systolic bruit within the popliteal artery can be an effective tool in the Box 23L-1 Typical Findings in Popliteal Artery Entrapment

• Young, active, otherwise healthy, individual • Intermittent claudication in the calf and foot • Rapid symptom onset with intense lower

extremity physical activity • Symptoms of chronic lower limb ischemia, such as paresthesias, discoloration of the foot or toes, poikilothermia, rest pain, and tissue necrosis are rare. • Claudication is unilateral, although may have entrapment bilaterally.

physician’s diagnostic arsenal. In more advanced cases of PAES, evidence of superficial arterial collateralization may be observed. Although an unusual physical examination finding, the knee may be warm and show prominence of the genicular arteries over the medial and lateral portions of the joint. The presence of a popliteal artery aneurysm in a young, otherwise healthy patient can indicate substantial entrapment with subsequent poststenotic dilation, putting the patient in danger for serious complications. In such cases, the aneurysm may be a source of distal embolization. It has been shown that 18% to 31% of patients with popliteal artery aneurysm concomitant with PAES have undesirable embolic complications if left untreated, and the incidence of limb loss increases dramatically once these complications have occurred.44,49 In addition to examination of the arterial system, a thorough venous examination should be considered essential to the work-up of a patient presumed to have PAES. The presence of vascular congestion, cyanosis, distended superficial veins over the lower leg, or other signs of venous

Knee 1839

obstruction at the level of the knee should lead one to consider the diagnosis of PAES.29 Finally, a helpful, noninvasive maneuver that may add to the physician’s information pool is to check a resting as well as exercise ankle-brachial index (ABI) on the affected side. The ABI is calculated by taking the pressure at which the respective arteries, posterior tibial and brachial, are occluded by a typical blood pressure cuff. These numbers are then divided, and the ratio is obtained. An ABI greater than 1 is considered normal. If significant stenosis or occlusion is present as the result of PAES, the ABI will likely be less than 1. In the subset of patients with a normal ABI and symptoms suggesting the diagnosis of PAES, exercise treadmill testing can be done, as described by McDonald and colleagues.48 In doing so, the patient exercises according to the protocol until symptoms of claudication are reached or until he or she finishes the end of the test. Immediately following that interval, the patient’s ABI is repeated. In this setting, an ABI of less than 1 is highly indicative of exercise-induced vascular insufficiency consistent with PAES.

Imaging In addition to the physical examination, imaging is one of the primary tools in the physician’s diagnostic armamentarium for the evaluation of PAES. Both noninvasive and invasive testing procedures have an important role in the diagnosis, classification, and evaluation of patients with PAES. Noninvasive imaging methods include ultrasound, computed tomography (CT), computed tomographic angiography (CTA), magnetic resonance imaging (MRI), and magnetic resonance angiography (MRA). Invasive methods of imaging include angiography, historically considered the gold standard by many for diagnosing PAES. DiMarzo and colleagues were the first to show that duplex ultrasonography (DUS) could be used successfully in screening and diagnosing patients with PAES (Fig. 23L-2).24 A decrease in the peak systolic flow rate of 50% or more

A

in the popliteal artery during active plantar flexion of the foot against resistance with the knee flexed to 15 degrees was considered diagnostic in all cases of PAES. In another study by Goh and colleagues, the use of DUS was successful in demonstrating the presence of PAES in all limbs tested.28 Numerous other studies have corroborated these findings.24,28,50-53 In fact, Zeuchner and colleagues reported a sensitivity of 94% and a specificity of 98% for the detection of popliteal artery occlusion.53 This study, however, demonstrated arterial occlusion in patients with peripheral vascular disease and not PAES, although the results are nonetheless intriguing. Others have questioned the utility of DUS in the evaluation of PAES, postulating that the compression of the popliteal artery by active plantar flexion could be a normal physiologic occurrence.54-56 Akkersdijk and colleagues, using peak systolic flow velocities, were able to observe compression of the popliteal artery with active plantar flexion in 72% of healthy volunteers.54 Similarly, Erdoes and colleagues found that active plantar flexion could induce partial popliteal artery occlusion in 53% of healthy volunteers.55 Interestingly, there was no difference when normal individuals were compared with well-trained athletes in terms of likelihood of arterial occlusion. Other authors have quoted similar rates in healthy patients as well.52,56 CT and CTA have also been shown to play an important role in the diagnosis and management of PAES.57-61 Both have been used to determine relationships between the popliteal artery and its surrounding structures within the popliteal fossa (Fig. 23L-3). In addition to demonstrating anatomic relationships with a patent popliteal artery, CT and CTA have the advantage of allowing the physician to identify sites of stenosis, occlusion, thrombosis, or other abnormalities that may be missed by other imaging modalities, such as angiography (Fig. 23L-4). CTA has also been shown to be a valuable tool in the diagnosis of other pathologies that may mimic PAES, such as adventitial cystic disease and popliteal artery aneurysm.58 In one study by

B

Figure 23L-2   A 49-year-old man with right popliteal artery entrapment syndrome. A, Doppler ultrasound of the right popliteal fossa with leg in neutral position shows normal triphasic waveform. B, Doppler ultrasound obtained with plantar flexion of the right foot shows compression of the popliteal artery with absence of flow. Dynamic popliteal artery compression elicited by plantar flexion of the foot is a sonographic finding consistent with, but not diagnostic of, popliteal artery entrapment syndrome. This finding can be seen in healthy, asymptomatic subjects. (From Macedo TA, Johnson CM, Hallett JW Jr, Breen JF: Popliteal artery entrapment syndrome: Role of imaging in the diagnosis. AJR Am J Roentgenol 181(5):1259-1265, 2003.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1840 DeLee & Drez’s� O

Figure 23L-3   This 32-year-old patient with intermittent right lower extremity claudication experienced with exercise was later found to have popliteal artery entrapment syndrome. Computed tomographic angiography of bilateral lower extremities shows an abnormal muscle band (orange arrowhead) arising from the medial head of the gastrocnemius muscle. The contralateral extremity shows normal anatomy.

Beregi and colleagues, CT was able to accurately diagnose 14 isolated, hemodynamically significant (>50% diameter reduction) stenoses, 4 occluded arteries, and 11 aneurysms in patients referred for popliteal arteriography. CTA also demonstrated these 14 stenoses and 4 occlusions, and it showed that 8 of the stenoses were associated with other abnormalities. It is important to remember, however, that as with DUS, active plantar flexion against resistance is an important component of the evaluation and should be used adjunctively during CT and CTA because this often adds clinical information in patients who appear to have a normal popliteal artery at rest. Yet another benefit of CT and CTA is the ability to evaluate the contralateral limb simultaneously, thereby eliminating the suspicion of or concern about bilateral entrapment.59,60,62-64

A

In addition to CT and CTA, MRI and MRA are becoming two of the most widely used imaging modalities for diagnosing PAES (Fig. 23L-5). Like CTA, magnetic resonance has the ability to evaluate the popliteal fossa and the abnormal relationships that occur functionally and morphologically within the space (Figs. 23L-6 to 23L-9). It is noninvasive and able to precisely identify abnormalities other than PAES that cause similar symptoms. Unlike CTA, MRI avoids the use of nephrotoxic contrast agents. These reasons have prompted many authors to consider MRI or MRA the diagnostic tool of choice in young patients with intermittent claudication.28,65-67 Several studies have shown MRI and MRA are superior to DUS and CTA in accurately diagnosing PAES.68,69 Atilla and colleagues demonstrated results equivalent to those of angiography in diagnosing PAES.68 Gorres and colleagues argue that MRI and MRA offer a more detailed morphologic examination of the popliteal fossa and are superior in terms of multiplanar imaging over CT or CTA and DUS for identifying PAES.69 MRI and MRA have been shown to be especially valuable when there is occlusion of the popliteal artery, a situation that limits the value of DUS and angiography.67 Because of these reasons, many centers are now using MRI or MRA as the primary imaging modality in the management of young adults with intermittent claudication.65,68 Before the widespread use of CT and CTA and the recent advances in MRI and MRA, contrast angiography was the classic screening and diagnostic tool for evaluating PAES.70 It remains the most widely used diagnostic modality for PAES, although this is changing (see earlier discussion on CT, CTA, MRI, and MRA).38,45,48 Historically, the diagnosis of PAES was strongly suggested when two or more of the following were observed on neutral, nonstressed images29: (1) medially deviated proximal popliteal artery, (2) segmental occlusion of the midpopliteal artery, and (3) poststenotic dilation of the popliteal artery (Figs. 23L-10 to 23L-13). As with all imaging for PAES, arteriography should be performed in both neutral and active plantar flexion positions in order to bring out subtle entrapments that may

B

Figure 23L-4   This 21-year-old college athlete with right lower extremity claudication experienced during athletics practices (lacrosse) was later found to have popliteal artery entrapment syndrome. Computed tomographic angiography (A, proximal view; B, distal view) of bilateral lower extremities shows abnormal muscle band (orange arrowhead) arising from the medial head of the gastrocnemius muscle and associated nearly complete occlusion of the right popliteal artery (blue arrowhead). The contralateral extremity shows normal anatomy.

Knee 1841

Figure 23L-5   A 44-year-old man with normal anatomy of the right popliteal fossa. Axial T1-weighted magnetic resonance imaging depicts normal anatomic relationship of popliteal vessels and muscle. Only fat should surround the popliteal artery and vein.

Figure 23L-6   This 32-year-old woman with popliteal artery entrapment syndrome of the left lower extremity presented with calf claudication. Axial T1-weighted magnetic resonance imaging reveals abnormal muscle slip (blue arrowhead) between the popliteal artery (orange arrowhead) and vein responsible for the arterial entrapment.

Figure 23L-7   This 47-year-old man with right calf claudication was found to have popliteal artery entrapment syndrome (PAES). Axial T1-weighted magnetic resonance imaging (MRI) reveals the underlying cause of the occlusion to be an abnormal muscle band (orange arrowhead) arising from the medial head of the gastrocnemius muscle. MRI has the advantage of showing both arterial changes and underlying abnormalities found in patients with PAES.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1842 DeLee & Drez’s� O

A

B

C Figure 23L-8   A 44-year-old man with popliteal artery entrapment syndrome (PAES) in the left lower extremity. A, Axial   T1-weighted magnetic resonance imaging (MRI) reveals abnormal muscle slip (orange arrowhead) between popliteal vessels.   B, Time-of-flight magnetic resonance angiography (MRA) of popliteal arteries in neutral position shows normal arterial flow.   C, Time-of-flight MRA of popliteal arteries with active plantar flexion of the feet shows near occlusion (blue arrowhead) of the left popliteal artery and no change in the right leg. MRI and MRA have the potential to show both functional arterial changes and abnormal anatomy responsible for PAES.

appear normal on neutral imaging (Fig. 23L-14).55 Despite the clinical success of angiography in diagnosing patients with PAES, many problems exist with this method of imaging. First, angiography is inherently invasive, and there are several potential complications associated with arterial catheterization, including most importantly arteriovenous fistula and pseudoaneurysm formation. In addition to these infrequent complications, the use of heavy loads of contrast dye in this patient population is an important concern. Although healthy patients are generally considered not at significant risk for contrast-induced nephropathy, exposure of a healthy individual to this potential morbidity, when other means of diagnosing PAES exist, should be taken into consideration. These facts, together with the proven success of CT, CTA, MRI, and MRA, have led many physicians to direct their practice toward less invasive means of diagnosis, classification, and operative planning for patients with PAES.28

TREATMENT OPTIONS

Figure 23L-9   A 37-year-old man with popliteal artery entrapment syndrome. Axial T1-weighted magnetic resonance imaging of the right popliteal fossa reveals prominent abnormal muscle (orange arrowhead) responsible for arterial entrapment.

Surgical intervention is the gold standard for the treatment of PAES, regardless of type or severity. The complexity of operative treatment, however, differs greatly depending on the type and extent of disease within the popliteal artery as the result of the entrapment. The two primary options include simple division of the medial head of the gastrocnemius (myotomy) alone, and myotomy with ­endarterectomy and vein patch or replacement of diseased popliteal artery with saphenous vein graft.

Knee 1843

A

B

C

D

Figure 23L-10   Spectrum of angiographic findings encountered in popliteal artery entrapment syndrome (PAES). A, Angiogram of a 34-year-old woman with PAES showing medial deviation of the popliteal artery (orange arrowhead). B, Angiogram of a 21-year-old man with PAES showing nonocclusive acute thrombus of the popliteal artery (blue arrowhead) with embolization to distal tibial artery. C, Angiogram of a 34-year-old woman with PAES showing occlusion of the popliteal artery (orange arrowhead). D, Angiogram of a   37-year-old man with PAES showing popliteal artery aneurysm (blue arrowhead)

Figure 23L-11   A 26-year-old man with popliteal artery entrapment syndrome in the right extremity. Popliteal artery is deviated medially (orange arrowheads) and occluded (blue arrowhead), with reconstitution by distal collaterals.

Figure 23L-12   This 42-year-old woman with a 3-year history of left lower extremity claudication was later found to have popliteal artery entrapment syndrome. The popliteal artery is noted to be medially deviated and occluded with poststenotic dilation (orange arrowheads). Additionally, there are several prominent genicular collaterals that have formed (small blue arrowheads).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1844 DeLee & Drez’s� O

Figure 23L-13   A 31-year-old man with popliteal artery entrapment syndrome. The image shows partial occlusion (orange arrowheads) and subsequent poststenotic dilation   (blue arrowhead).

If popliteal entrapment is diagnosed early in the course of the disease process, the artery is typically normal from an anatomic and cellular perspective. In this instance, following complete exploration of the popliteal fossa, the medial head of the gastrocnemius muscle is released, along with any abnormal muscle slips or fibrous bands that may be contributing to the entrapment (Fig. 23L-15). Reconstruction of the divided muscle does not appear to be necessary because no measurable strength deficit is observed after myotomy.67 Occasionally, local exploration reveals evidence of arterial disease, including luminal irregularity, marked thickening or nodularity of the arterial wall within the entrapment site, aneurysm formation distal to the entrapment, or thromboembolic complications. In these circumstances, the procedure of choice is muscle release with in situ bypass of the diseased popliteal artery with saphenous vein graft (see Fig. 23L-15D).29 In the small subset of patients in which type F entrapment is suspected, and no other orthopaedic cause for the symptomatology is elucidated, the guidelines for the operation of choice are the same as with normal entrapment types I to V. In the presence of aneurysmal disease within the popliteal artery as the result of entrapment, resection of the aneurysm sac and replacement of the dilated segment with a vein graft should be performed. As with any procedure for PAES, release of the medial head of the gastrocnemius should be undertaken before the arterial repair. There are two main approaches to the popliteal artery: the posterior approach and the medial approach (Table 23L-3; Box 23L-2). The posterior approach is the most commonly used approach for the treatment of PAES. First, it affords the surgeon the best exposure of the involved anatomy, an essential aspect of any operative procedure. Furthermore, it allows easier recognition of subtle

A

B

Figure 23L-14   A 30-year-old woman with popliteal artery entrapment syndrome of the left lower extremity who presented with calf claudication. A, Angiogram of left popliteal artery in neutral position is normal. B, Angiogram of same extremity during active plantar flexion of the foot shows significant narrowing of the left popliteal artery (orange arrowhead).

a­ nomalies and is necessary for straightforward placement of a bypass graft. Finally, the posterior approach leaves the patient with a more pleasing cosmetic result. The posterior approach does have a few disadvantages. Because the patient needs to be in the prone position for the exposure, general endotracheal anesthesia must be used. In addition, when compared with the medial approach, the recovery time is slightly longer, although the significance of this is unknown. At our institution, the posterior approach is the procedure of choice because of its ease, straightforward anatomy, and successful long-term results. The medial calf approach, however, has some important advantages (see Table 23L-3). As noted earlier, a quicker return to activity can be expected with the medial approach. In addition, because the patient can be positioned supine during the procedure, regional anesthesia may be instituted. Some argue that the medial approach is better for long-segment occlusions of the popliteal artery when the likelihood of femoropopliteal bypass is high based on preoperative imaging, although this finding is rather infrequent.67 Despite these important benefits, the disadvantages of the medial approach are significant. In inexperienced hands, the anatomy encountered in the popliteal fossa by the medial approach is difficult to completely appreciate, and in some circumstances, it becomes especially arduous to achieve adequate exposure of the entire popliteal artery. Also, the medial approach lends itself to a higher rate of a missed entrapment when compared

Knee 1845

A

B

C

D Figure 23L-15   Operative approach for popliteal artery entrapment syndrome. A, Standard incision for the posterior approach to the popliteal fossa. B, Operative exposure showing type I entrapment and subsequent sharp division of the medial head of the gastrocnemius. C, Successful release of undamaged popliteal artery (artery will not need to be replaced). D, Successful saphenous vein bypass of injured popliteal artery following myotomy (if indicated at operative exploration).

with the posterior approach.29 Finally, and perhaps most importantly, a higher rate of recurrence is observed with the medial approach. The treatment of PAES with endovascular techniques is still largely in its infancy. There are few reports in the literature describing balloon angioplasty after attempted thrombolysis in patents with PAES.61,71 However, their sample sizes are small and their results are largely ­speculative. One

reason for the lack of success with endovascular treatment of PAES is the inability to fix the underlying pathology for the vascular entrapment, and thus there remains a high degree of reocclusion despite seemingly adequate therapy. Some authors argue that treatment of PAES by endovascular means may be adequate after the appropriate operative entrapment release,24,68,72 but this approach appears far too complex.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1846 DeLee & Drez’s� O

   TABLE 23L-3  Characteristics of Posterior and Medial Approaches to the Popliteal Fossa Posterior Approach

Medial Approach

Pros

Pros

Anatomy easy to appreciate Less prominent scar Essential for placement of bypass graft Subtle anomalies easier to recognize and treat Cons

General anesthesia (patient must be prone) Longer recovery

Regional anesthesia may be used Shorter recovery Ease of saphenous vein harvesting Ease of femoropopliteal bypass if needed

Cons

Anatomy difficult to appreciate in inexperienced hands Incomplete exposure of the popliteal artery Higher recurrence rate Higher rate of missed entrapment

Adapted from Levien LJ: Nonatheromatous causes of popliteal artery disease. In Rutherford RB (ed): Vascular Surgery. Philadelphia, Elsevier Saunders, 2005.

Box 23L-2 Treatment Options in Popliteal Artery Entrapment

• Medial head release of gastrocnemius muscle • With or without interposition saphenous vein graft

bypass

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS In most patients with PAES, gastrocnemius muscle release is all that is needed to successfully relieve the symptoms of claudication. Postoperatively, these patients tend to do very well, and their recovery is based largely on adequate wound healing. A fundamental initial step in successful postoperative recovery is early physical therapy with lower extremity range of motion and light strengthening exercises. It has been well documented that patients prescribed to these regimens have better outcomes than in those without physical therapy.73 Once adequate wound healing has occurred and the patient can ambulate autonomously without pain, discharge from the hospital is appropriate. We recommend initial follow-up within 14 days from the day of discharge, or earlier if problems arise. Potential complications following muscle release are minimal and mainly include problems with wound healing. There is little, if any, functional deficit of the lower extremity and ankle as a result of the myotomy. Patients can expect to achieve their preoperative strength level once wound healing has completed. In contrast, patients who undergo arterial reconstructions as part of their surgical therapy are at risk for substantial and potentially more severe complications. The most important ­ complication

Authors’ Preferred Method A patient presenting to our center with presumed PAES first undergoes a thorough history and physical examination. If the clinical history and physical examination findings fit with the picture of PAES, the patient undergoes CT and CTA of the affected extremity as well as the contralateral limb both at rest and during active plantar flexion. This method of imaging is used because it is easier compared with MRI, has a rather quick turnaround, and is cost-effective. Additionally, CT and CTA provide the essential information needed to appropriately plan our operative procedure, particularly regarding whether bypass grafting will be necessary. If there is a contraindication to CT or CTA because of renal dysfunction or an allergy to contrast dye, MRI or MRA is our imaging modality of choice. We advise against the use of DUS in the work-up of PAES because it adds little information to the clinical picture. Once imaging has confirmed the diagnosis of PAES, the individual is scheduled for operative exploration. We prefer the posterior approach for the treatment of PAES for several reasons. First, it provides the best exposure to the contents that encompass the entrapment and the subtleties that often exist in these cases. Second, it provides the surgeon the best exposure for the placement of a bypass graft. Third, it leaves the patients with a much better cosmetic result because there is a less prominent scar. We have not encountered an increased morbidity with the use of general anesthesia, nor have we observed any difference in terms of recovery time, despite the dogma that the posterior approach exposes the patient to undue anesthetic risk and leads to a longer recovery.

Postoperatively, we encourage early ambulation, even in the setting of saphenous vein bypass. In the patients who undergo myotomy alone, discharge from the hospital may be achieved once unassisted ambulation occurs, there is adequate wound healing with no signs of infection, the patient is tolerating a regular diet, and there are no pulse wave amplitude abnormalities or differences on routine postoperative resting and stressed ABIs and pulse volume recordings (PVRs). The discharge criteria for those patients who undergo saphenous vein grafting are the same. On the day of discharge, we have our patients undergo resting and stressed ABIs and PVRs to document the resolution of the entrapment and serve as a baseline for future clinic visits. We provide the patient several detailed education sessions about their new bypass graft, the signs and symptoms of acute graft thrombosis, and what to do if they experience any problems. We typically see patients back in the clinic 10 to 14 days postoperatively for examination of the wound and evaluation of the ABIs and PVRs, both at rest and with active plantar flexion. Following this, we recommend a 2-week period of light activity, followed by increasing amounts of more vigorous activity until the preoperative baseline level is attained, a process that typically takes 6 to 8 weeks. During this time, we like to see the patient in the clinic every 2 to 3 months until full activity is achieved. In patients who have undergone arterial reconstruction, follow-up should consist of graft imaging by DUS every 6 months for the first year and annually thereafter, in addition to the previously mentioned activity prescriptions.

Knee 1847

that these patients experience is graft thrombosis. The incidence of graft failure in this setting is low, with a 5-year primary patency rate of about 98% at our institution. It is well known that heparin, dextran, warfarin, and the antiplatelet drugs clopidogrel, ticlopidine, aspirin, and dipyridamole appear to be appropriate adjuncts in the high-risk femoropopliteal bypass patient; however, the use of these therapies in the PAES patient has not been studied. Despite this, the potential for graft failure remains a problem throughout the postoperative setting and beyond. The patient must be given specific information about the signs and symptoms of acute graft thrombosis and what to do if he or she experiences any of them. Signs of acute ­ thrombosis include pulselessness, poikilothermia, pallor, paresthesias, paralysis, and pain. Additionally, these patients must undergo close serial follow-up examinations. In this setting, physical examination, ABIs, and DUS to monitor graft patency are appropriate and should be performed every 6 months for the first year and annually thereafter. Evidence of vein graft stenosis at either the proximal or distal sites of the anastomosis should be investigated with the appropriate imaging modality and handled according to the findings.

Criteria for Return to Play Return to a full level of activity depends on several factors, including the complete return of strength and flexibility, adequate wound healing, and arterial reconstruction graft patency. A clearance evaluation by a member of the medical team should include documentation of adequate wound healing as well as restored blood flow under circumstances that caused cessation before operative intervention (i.e., active plantar flexion) by way of ABIs and PVRs. Ruppert and colleagues reported that ABIs alone, with and without forced active plantar flexion, were sufficient for a postoperative follow-up examination to evaluate successful decompression and quality assurance in patients with PAES.74 Repeat CTA, MRA, and angiogram are not needed to evaluate graft patency as long as the clinical signs and symptoms have abated after treatment and adequate ABIs are obtained (see previous discussion). Once these criteria have been met (after about 6 to 8 weeks),

Box 23L-3 Return to Play in Popliteal Artery Entrapment

• Complete return of strength and flexibility • Adequate wound healing • Ankle-brachial index and pulse volume recordings

exhibit restoration of blood flow with stress maneuvers • Repeat computed tomographic angiogram, magnetic resonance angiogram, and angiogram are not needed as long as normal ankle-brachial index and pulse volume recordings are documented • Prescribe 6 to 8 weeks of light ambulation only, followed by a 2-week period of increasing activity, followed thereafter by increasing increments of more strenuous activity until the optimal level is achieved

we recommend a 2-week period of light activity such as walking or swimming, followed by increasing increments of more strenuous activity until the preoperative baseline level is achieved. The patient should be followed during this time with return visits to the physician every 2 to 3 months until full activity is achieved. In those patients who have undergone arterial reconstruction, follow-up should consist of graft imaging every 6 months with CTA or MRA for the first year and annually thereafter, in addition to the previously mentioned activity prescriptions (Box 23L-3). C

r i t i c a l

P

o i n t s

l Popliteal artery entrapment is the result of a developmental abnormality. l The classic presentation is a young, active, otherwise healthy individual with intermittent claudication in the calf and foot. l The presence of pulse deficits with active plantar flexion of the ankle should raise the suspicion for PAES (although their occurrence is not conclusive). l A change in the ABIs of greater than 30% with active plantar flexion should raise the suspicion for PAES (although their occurrence is not conclusive). l If PAES is suspected after physical examination, CT and CTA or MRI and MRA should be performed for definitive diagnosis. l Treatment of PAES is solely operative and includes myotomy alone or myotomy plus saphenous vein bypass graft of affected popliteal artery. l Patients need not undergo anticoagulation after placement of bypass graft. l Return to preoperative baseline can be expected in all patients regardless of operative intervention.

S U G G E S T E D

R E A D I N G S

Darling RC, Buckley CJ, Abbott WM, Raines JK: Intermittent claudication in young athletes: Popliteal artery entrapment syndrome. J Trauma 14(7):543-552, 1974. Goh BK, Tay KH, Tan SG: Diagnosis and surgical management of popliteal artery entrapment syndrome. Aust N Z J Surg 75(10):869-873, 2005. Hamming JJ: Intermittent claudication at an early age, due to an anomalous course of the popliteal artery. Angiology 10:369-371, 1959. Levien LJ: Nonatheromatous causes of popliteal artery disease. In Rutherford RB (ed): Vascular Surgery. Philadelphia, Elsevier Saunders, 2005, pp 1236-1255. Levien LJ, Veller MG: Popliteal artery entrapment syndrome: More common than previously recognized. J Vasc Surg 30(4):587-598, 1999. Macedo TA, Johnson CM, Hallett JW Jr, Breen JF: Popliteal artery entrapment syndrome: Role of imaging in the diagnosis. AJR Am J Roentgenol 181(5): 1259-1265, 2003. McGuinness G, Durham JD, Rutherford RB, et al: Popliteal artery entrapment: Findings at MR imaging. J Vasc Interv Radiol 2(2):241-245, 1991. Rignault DP, Pailler JL, Lunel F: The “functional” popliteal entrapment syndrome. Int Angiol 4(3):341-343, 1985. Steurer J, Hoffmann U, Schneider E, et al: A new therapeutic approach to popliteal artery entrapment syndrome (PAES). Eur J Vasc Endovasc Surg 10(2):243-247, 1995. Stuart TP: Note on a variation in the course of the popliteal artery. J Anat Physiol 13:162-165, 1879.

R efere n ces Please see www.expertconsult.com

C H A P T E R  

 24

Leg S ect i o n

A

Stress Fractures of the Leg Bryce Bederka and Annunziato Amendola

The first report of a stress fracture is credited to ­Breithaupt, a Prussian military physician, who described it in 1855.1 He described foot pain and swelling in new military recruits, and the metatarsal fractures he observed are now commonly called march fractures. Early reports in the literature on stress fractures dealt primarily with the military population because they were commonly seen in the basic training programs. More recently, because of the increased participation in athletics, the recreational and collegiate athletes have become a much studied group.2-4 Stress fractures are an important injury in athletes that can result in time lost from training and competition. ­Iwamoto and Takeda reported that nearly 2% of visits to a dedicated sports medicine clinic over a 10-year period were for stress fracture.5 The incidence of stress fracture has been reported to be as high as 21.1% in competitive track and field athletes6 and is commonly reported to be between 3.7% and 6.8% in the military recruit population.7-9 The tibia is the most common bone to suffer a stress fracture, composing 25% to 46% of reported stress fractures,6,9,10 with the fibula composing an additional 5% to 12% of injuries.6,9 Injuries involving the leg, therefore, make up one third to one half of all reported cases of stress fracture. Stress fractures of the lower extremity most commonly involve the younger and more active population. It has been reported that 80% to 91.6% of patients treated for stress fractures are younger than 29 years.2,11 Treatment of stress fractures is dependent on symptoms and on the risk for progression to nonunion and catastrophic failure. Some may respond to activity modification, supports, and pain medications, whereas others may require surgery to heal and return the athlete to sport. We presented an algorithm12 for the assessment of stress fractures as either high or low risk. Those fractures considered low risk can generally be managed by these nonoperative means. Those fractures deemed high risk, because of either risk or consequences of nonunion or catastrophic failure, are best managed by surgery. This chapter addresses the literature available for dealing with stress fractures and in particular those involving the tibia.

RELEVANT ANATOMY AND BIOMECHANICS The tibia and fibula share in the weight-bearing role of the leg. The fibula bears 6% to 17% of the weight of the lower extremity depending on the position of the foot, with the load increasing with foot dorsiflexion and eversion,13,14 and with the tibia bearing the remainder of the weight. Biomechanically, the anterior surface of the tibia is loaded in tension, whereas the posterior surface is loaded in compression. Stress fractures of the anterior cortex have been reported to have a higher incidence of nonunion with nonoperative management as well as a risk for complications with surgical treatment. Most stress fractures of the tibia occur in the proximal metaphysis of the tibia, with a smaller amount in the distal third. Mid-third diaphyseal fractures of the tibia account for a smaller proportion of stress fractures but make up the majority of operatively treated fractures.3,15-24 The area of the stress fracture can often be predicted by the type of sporting activity. Runners are more prone to fracture of the midshaft or distal third of the tibia, whereas volleyball and basketball players are more prone to proximal tibial stress fractures.5,18,23,24

PREDISPOSING RISK FACTORS FOR FOOT AND ANKLE STRESS FRACTURES Individual Factors Epidemiologic studies done in the United States Army have shown that white individuals have a twofold increased incidence of stress fractures.25,26 Women have a substantially increased risk for developing stress fractures.2,25-27 Moreover, exercise-induced amenorrhea was recognized to double the risk for stress fractures in female distance runners.28 Nutritional studies carried out in ballet dancers29 have suggested that body weight that is less than 75% of ideal and eating disorders (anorexia, bulimia, and ­atypical 1849

�rthopaedic ����������� S �ports ������ � Medicine ������� 1850 DeLee & Drez’s� O

eating disorders) are seen more frequently in patients with stress fractures. Taimela and colleagues30 studied personality traits and performed motor ability testing on 108 new Army conscripts in Finland and suggested that those with stress fractures are more likely to have low scores on achievement, dominance, and exhibition traits. They also noted the predominance of tall stature and previous inactivity in sports in this group of conscripts. Conversely and contrary to widely held beliefs, prospective controlled studies have shown that recreational aerobic physical fitness, in addition to participation level, had no correlation with the incidence of stress fractures in Army recruits. Moreover, Giladi and associates31 showed no statistically significant correlation between the incidence of tibial stress fractures and individual height, weight, tibial torsion, tibial alignment, tibial bone mineral content, or radiographic width of the tibial cortex.

Anatomic Factors Different authors suggested that stress fractures of the lower extremities are more prevalent in patients with varus malalignment of the lower extremity and foot.32-34 Once again, however, prospective controlled studies showed that only tibial bone width31 and external rotation of the hip35 were found to be statistically significant risk factors for stress fractures. The effect of different foot types on stress transmission and shock-absorbing capabilities of the foot cannot be overemphasized. Clement and associates have established the effect of increased foot pronation on higher incidence of stress fractures, especially in the tibia and fibula.32-34 In addition, rigid pes cavus was found to be more common in stress fractures of metatarsals and femur,33 possibly owing to its reduced shock absorption capacity. Even in normal feet, Simkin and coworkers36 showed an increased incidence of metatarsal stress fractures in feet with a low longitudinal arch. Conversely, patients with a normal arch may have decreased shock absorption capacity in the foot, leading to increased femur and tibial stress fractures and a surprisingly low incidence of metatarsal stress fractures.36

Shoewear Studies from the American military have shown an increased incidence of stress fractures of the lower extremity in recruits who exercised in combat boots. The use of semirigid insoles has decreased this incidence but not significantly so.37 A similar result was obtained by Simkin and coworkers,36 who showed that patients with normal low longitudinal arch had a decreased incidence of metatarsal stress fractures with the use of semirigid polyolefin arch supports. It is important to note that viscoelastic insoles have no effect on decreasing loads transmitted to the lower extremity; neither do they decrease the incidence of stress fractures in the lower extremity.37 The design and material selection of running shoes have received wide attention lately, with studies suggesting a strong correlation between the mechanical support or shock-absorbing properties of the midsole and development of stress fracture. A decrease in those properties takes place with increasing duration of use. Other factors

that may predispose to stress fractures are inadequate heel wedging, loose-fitting heel counters, excessive heel wear, and narrow toe boxes.

Surface-Related Factors Along with the previously mentioned factors, training on hard surfaces, road camber, or uneven terrain has been implicated as a possible cause of an increasing incidence of stress fractures.2,25,27,32,33,38-40

Sport-Specific Factors Various sports have characteristic sites of predilection for stress fracture. Runners are usually prone to stress fractures of the tibia, fibula, and metatarsals. Basketball players suffer tarsal navicular stress fractures, whereas football players and competitive figure skaters tend to have more stress fractures of the fifth metatarsal.41 Classical ballet dancers may suffer various stress fractures to the foot, involving the tarsal navicular and sesamoids as well as the proximal shafts of the second and third metatarsal bones.42

PATHOGENESIS OF STRESS FRACTURES Bone is a living organ that responds to change in mechanical loads by effecting changes in its internal architecture and external configuration, a process called remodeling. Wolff’s law states simply that form follows function. In the case of stress fractures, it is the repetitive forces applied to bone that lead to this imbalance or failure of bone remodeling. The basic pathophysiologic mechanism is that of an increased osteoclastic activity that stimulates bone resorption. This in turn shifts the balance from bone formation to bone resorption, leading to miniature cortical defects and subsequent microfractures.32 Studies of the histogenesis of stress fractures43-45 certainly confirm the previously mentioned sequence of events. Hartley in 194346 and Anderson in 199038 described stress (fatigue) fractures as analogous to metal fatigue. ­Stanitski and associates47 and others2,43 refuted this view and stated that bone, unlike metal, is a heterologous, anisotropic material that requires stress for remodeling and that subsequent stress fracture is an expression of failure of adaptation of a living structure. Frankel, in 1978, described four mechanisms as causing stress fractures.45 One is repetitive muscle contracture leading to bony overload, as seen in fibular and calcaneal stress fractures. The second mechanism occurs when bones are exposed to increased stresses following muscle fatigue. The third mechanism is the increased loads caused by running on a hard surface with subsequent increased ground reactive forces. The fourth is a high repetitive stress that is seen mainly with low-impact stresses.45 From a practical standpoint, Taunton and colleagues34 advanced five possible causes for overuse injuries that may very well apply to stress fractures. The first factor is training errors, typically seen with the novice runner or the new military recruit. These errors are usually in the form of ­ persistent high-intensity training without alternate easy days, sudden increases in training mileage or

Leg 1851

intensity, a single severe training or competitive session, and repetitive heel running. The second factor is that of muscular dysfunction and flexibility. Inadequate strength of the tibialis anterior and poor flexibility in calf musculature may lead to muscle fatigue and loss of shock absorption or flexibility, followed by altered gait and resultant abnormal stresses. The third factor is footwear. In running shoes, for example, inadequate heel wedging, soft, loosefitting heel counters, loss of midsole mechanical support, and narrow toe boxes all lead to inadequate motion control and abnormal loads. The fourth factor is lack of the use of orthotic devices—a semirigid arch-supporting orthosis has been shown to increase the shock-absorbing capacity and improve load transmission, leading to decreased stress fractures. The fifth and last factor is a biomechanical factor related to training surfaces. Hard surfaces accentuate the ground reactive force, particularly in runners in whom 1000 to 1200 foot strikes per mile, each at 1.5 to 3 times body weight supported by the foot, add to the injury.

CLASSIFICATION Stress fractures are identified according to their location, and this leads to their classification as high-risk or lowrisk fractures.16,48 The site of the fracture determines the principal mode of loading of the bone, either the tension or compression side of the bone, and also to the consequences of undertreatment. Fractures that are primarily loaded in compression or have less serious consequences of undertreatment are considered low risk. Fractures that are primarily loaded in tension or have serious consequences if undertreated are considered high risk (Box 24A-1). Inadequate or delayed treatment of these fractures can result in (1) progression to complete fracture, requiring open reduction and internal fixation; (2) nonunion; and (3) recurrence or refracture (Table 24A-1).

EVALUATION Clinical Presentation and History High index of suspicion, as well as a detailed history, would lead to a diagnosis in most eases. Classically, the patient presents with a history of exertional pain of insidious onset that is localized to the part in question and is usually worse after exercise, often following a change in activity levels.49-52 Box 24A-1   Sites of High-Risk Stress ­Fractures Superolateral femoral neck Patella Anterior tibial diaphysis Medial malleolus Talus Tarsal navicular Fourth and fifth metatarsals Sesamoids From Kaeding CC, Yu JR, Wright R, et al: Management and return to play of stress fractures. Clin J Sport Med 15(6):442-447, 2005.

  TABLE 24A-1  Classification of Stress Fractures Type of loading Natural history Management

High Risk

Low Risk

Tension Poor Aggressive Complete fracture: surgery Incomplete fracture: strict immobilization and non–weight-bearing or surgery

Compression Good Conservative Symptomatic: activity modification Asymptomatic: ­observation

From Kaeding CC, Yu JR, Wright R, et al: Management and return to play of stress fractures. Clin J Sport Med 15(6):442-447, 2005.

When this pain is ignored, it progressively increases in intensity and duration and can progress to a constant ache that forces the patient to stop training.17,23 Night pain is possible in rare cases.17 Ha and colleagues2 and Hershman and Mailly3 believe that exertional pain is the essential factor for diagnosis of stress fracture, even in the absence of a positive focal uptake on technetium-99m bone scan.53 The second important feature is history of training errors that is usually related to increase in duration or distance of training, training on a hard surface, and changes in footwear.3,16,17,23,54 Matheson and colleagues33 found an incidence of 22.4% of patients with distinct training errors. Although the onset of stress fracture is usually not event related,2,11,27,32,40 Matheson and colleagues33 mentioned a 9.9% incidence of a traumatic event in their patients, with only a 5.6% incidence of trauma in runners and a 20% incidence in other sports. For patients with previous stress fractures, the location of and circumstances resulting in the fracture, in addition to subsequent treatment, should be determined.17,33 As far as onset of pain, it is well documented that the pain of stress fracture is mainly elicited in the first few weeks of a new or altered training program. Milgrom and associates27 noticed the onset of symptoms in 53% of their patients in the first 4 weeks of training, and Greaney and colleagues39 confirmed that 64% of their patients presented with pain in the first 2 weeks after exercise. Both these reports were of prospective studies carried out in military recruits with a definite onset of training. Other studies, mainly retrospective studies in recreational athletes, mentioned a prolonged and protracted onset33 and time to diagnosis, especially in stress fractures of the foot and ankle, between 12 and 16 weeks on average. Certain types of patients are more susceptible to stress ­fractures, including those with eating ­ disorders,17,23,28 those participating in sports in which leanness is emphasized,16,23,28,54 women with menstrual abnormali­ ties,3,16,17,23,28,54 patients with a history of stress fractures,17,52,55 and military recruits.27 Special attention should be paid to the female athlete with a stress fracture. She should be screened for the “female athlete triad,” which includes disordered eating, amenorrhea, and osteoporosis.21,23,54,56 Although most athletes do not meet the classic definition of patients with eating disorders, the possibility must be ruled out during the clinical history.17 Obtaining the height and weight,

�rthopaedic ����������� S �ports ������ � Medicine ������� 1852 DeLee & Drez’s� O

with calculation of the percentage of body fat, also may be helpful in assessing a potential eating disorder. For the athlete with a true eating disorder, such as anorexia or bulimia, appropriate referrals to a dietitian, nutritionist, and psychologist are necessary. A nutritional assessment is also important for athletes without eating disorders because the ingested number of calories often is below that required given the athlete’s activity level, and the type of calories ingested may be insufficient in protein.17 Such a nutritional evaluation is not exclusive to female athletes; it also should be conducted in male athletes when indicated. Because female athletes with abnormal menstrual cycles are at particular risk for stress fractures,3,16,17,23,28,54 a menstrual history, including the age of menarche, cycle duration, frequency, date of last menstrual cycle, and previous episodes of amenorrhea or oligomenorrhea, should be documented.17,54 Evaluation by a gynecologist is recommended to rule out non–eating-related causes of menstrual irregularity.51 Pregnancy should also be ruled out as a possible cause for amenorrhea in female athletes of childbearing years. Criteria exist for use of hormonal replacement therapy and serial bone density scans in patients with the female athlete triad; this is discussed elsewhere in this text.

Physical Examination Stress fractures often present with a paucity of signs. A localized point of tenderness and a moderate amount of swelling are usually present but may also be entirely absent.3,17,23,51 Tests mentioned in the literature include the hop test (pain at the site of the presumed fracture produced by one-legged hop),57 the tuning fork test,2,51 and a percussion tenderness3 elicited at the fracture site when the bone is percussed at a distal site. Pain associated with tibial stress fractures typically is localized to the fracture site and is more proximal than that caused by medial tibial stress syndrome.3,23 Erythema or localized swelling also may be present.17,23,51 In the absence of any associated abnormalities, neurovascular examination findings are normal. Joint range of motion typically is normal, but gait analysis may reveal biomechanical risk factors.17,51

Imaging and Testing Plain radiographs are the first imaging study to be obtained in evaluation of the patient with a suspected stress fracture. Significant findings include a linear transverse disruption of one cortex, an area of focal sclerosis, periosteal reaction, cortical thickening, and narrowing of the endosteal canal.50-52,58 The latter findings are suggestive of a longstanding process. Particular attention should be paid to the mid-anterior cortex because a small lucency, commonly referred to as the “dreaded black line” (Fig. 24A-1), may be indicative of a mid–anterior cortex tibial stress fracture.15,17 Often, in the acute phase, radiographic changes may be absent, in which case the imaging should be repeated after an interval of 2 to 4 weeks, or a more advanced imaging modality may be used.16,17,23,59 At presentation, only 20% to 30% of patients have positive plain radiograph findings,27,39 and in one reported study, only 54% of patients developed plain radiographic findings even after resolution of their stress fractures.39 If radiographic ­examination

Figure 24A-1   The “dreaded black line” on the lateral tibial radiograph. Note the cortical disruption on the anterior cortex.

­ emonstrates evidence of a stress fracture, no further imagd ing is necessary. Triple-phase bone scan has for years been the gold standard for diagnosis of stress fractures in the absence of plain radiographic findings.17,39,51,60 The typical appearance on bone scan is of a focal area of increased signal (Fig. 24A-2).16,48,61 This must be differentiated from the imaging findings of medial tibial stress syndrome, which is characterized by a diffuse area of tracer uptake along the posteromedial border of the tibia. Triple-phase bone scanning has high sensitivity,27 with an essentially 0% false-­negative rate62; however, it lacks significant specificity. More recent studies, however, appear to indicate that magnetic resonance imaging (MRI) is the more accurate study, given that imaging of the fracture area is done in great detail, differentiating between other pathologic processes of bone.63-66 Bone scan is more clinically useful in patients who have contraindications to MRI studies and in those who have a low probability of other active bone diseases.67 MRI is emerging as the imaging modality of choice for evaluation of radiographically negative stress injuries of bone. Recent studies63,65 have reported MRI sensitivity ranging from 88% to 100% and specificity ranging from 86% to 100% for the diagnosis of stress fracture. The studies cite additional benefits, including the noninvasive nature of the procedure, precise visualization of the pathologic process, and ability to differentiate stress fracture from other pathologic process of bone as additional benefits. Further, Aoki and coworkers68 demonstrated the utility of a fat­suppressed coronal image of the tibia for distinguishing the bone edema characteristics that ­ differentiate medial tibial

Leg 1853

Figure 24A-2   Focal increased signal intensity characteristic of a tibial stress fracture.

stress syndrome from stress fracture. As with the bone scan, MRI signal characteristics of stress fractures demonstrate focal signal at the fracture site that with MRI is shown to involve the marrow (Fig. 24A-3). Medial tibial stress syndrome differs in having a linear appearance lining the posteromedial cortex, which may involve the medullary surface of the cortex but does not extend across the marrow. Fredericson proposed an MRI classification system for stress reaction and stress fracture,69 which graded the MRI findings based on the degree of edema seen on various sequences for stress fracture and stress reaction. The grading,

A

and the finding of a discrete fracture line, may be used to guide treatment choices. A follow-up study by the same group in 2004 demonstrated a 43% incidence of tibial stress reaction by their MRI criteria in asymptomatic collegiate distance runners.70 Computed tomography (CT) may be useful to better define identified or suspected fracture lines. However, CT does not have significant sensitivity in comparison with other imaging modalities when trying to rule out the possibility of stress fracture or distinguish it from other ­diagnoses.68

B

Figure 24A-3   Coronal (A) and sagittal (B) T2-weighted magnetic resonance imaging sequences demonstrating increased fluid signal involving the periosteal cortex and the medullary canal of a metaphyseal stress fracture. The dark line of the stress fracture is seen, which was not observed on plain radiographs.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1854 DeLee & Drez’s� O

TREATMENT OPTIONS Nonoperative Treatment Initial treatment of nearly all stress fractures and stress reactions of bone is discontinuation of the inciting activities, rest, and limited weight-bearing. The focus is on pain relief and protection from further injury and should be tried for 2 to 4 weeks. Mild analgesics or nonsteroidal anti-inflammatory drugs (NSAIDs) may be prescribed in conjunction with physical therapy modalities such as ice application.17,19 If pain is not relieved after the initial 2- to 4-week rest period, or in patients with high-risk fractures, strict non–weight-bearing with bracing or casting may be warranted for 3 to 12 weeks.71 Training errors, improper shoewear, and muscle imbalance can contribute to the formation of stress fractures and must be addressed to prevent fracture recurrence.17,23,49-51 Training schedules should be individualized for each patient. Shoes must be examined for signs of wear and inadequate support, with replacement suggested every 300 to 500 miles.17,19 Appropriate orthotics should be used, if necessary.17,72 Finally, treatment plans consisting of dietary counseling or estrogen replacement therapy are recommended for athletes with eating disorders and female athletes with menstrual irregularities, respectively.16,17 Gradual resumption of activity is recommended, with progression according to symptoms. Activity should begin with programs that elicit no pain or only a minimum of discomfort. Activity should not be advanced until each stage can be accomplished without pain. If pain recurs, the specific activity must cease and should not be reattempted until the pain is alleviated. Worsening pain at a previously comfortable activity level should elicit investigation for progression of the fracture.17,61 Patients may expect to resume full training in 8 to 16 weeks, and athletes must be aware that a prolonged recovery period may be necessary for more severe fractures, especially those involving the mid-anterior cortex, which often require a significant period of rehabilitation.17,22,61 In most patients, nonoperative treatment is successful in tibial stress fractures. During the treatment period, cross-training to maintain cardiovascular fitness is encouraged and may be maintained.17,23,51,61 Cycling, swimming, deep-water running, or other non–weight-bearing activities may be employed for lower extremity stress fractures. Upper body strength training does not jeopardize fracture healing and is recommended to preserve muscle mass.23 Improvement in muscular strength and endurance, continuation of cardiovascular fitness, and modification of risk factors also are important in the rehabilitation process and in prevention of future stress injuries.17 Functional bracing with pneumatic or other tibial supports has been used to allow early return from tibial stress fractures. An early case series report by Dickson and ­Kichline73 demonstrated immediate return to play in the pneumatic brace, with complete resolution of symptoms within 1 month in all 13 cases. A later prospective randomized trial by Swenson and coworkers74 found a return to full unrestricted activity in 21 days when treated by pneumatic bracing, with a mean 77 days to return to play in the group

without bracing. More recently, Allen and associates75 published a prospective randomized study that ­ contradicted earlier reports, demonstrating no improvement in return to play or activity and no change in reported pain. Unfortunately, the current body of literature consists of studies with small sample sizes and low power, so no definitive conclusions can be drawn. Bone stimulators also may be used in the treatment of tibial stress fractures. The use of ultrasound to accelerate healing has been documented76,77; however, only two studies have looked at the use of ultrasound for tibial stress fractures. Brand and colleagues78 looked at the use of lowintensity ultrasound in competitive collegiate athletes with tibial stress fractures. All eight athletes were able to continue training and competition while using the device, and the authors believed that it allowed return to activity without restrictions or bracing. A more recent prospective randomized trial of ultrasound by Rue and coworkers79 looked at naval midshipmen with tibial stress fractures and found no difference in healing time or resolution of symptoms. Electromagnetic and electrical fields have demonstrated efficacy with fracture healing in the treatment of osteotomies and nonunion,80-82 but have not been studied in the application to stress fractures. Pharmacologic treatment alternatives are limited at this time, save adjusting for abnormal dietary calcium intake. One study employing bisphosphonates has been performed.83 With a small group of collegiate athletes, training was resumed pain free in as little as 72 hours from the first intravenous infusion. Bisphosphonates and other antiresorptives, calcitonin, and recombinant parathyroid hormone are not well studied in the treatment of stress fractures and at this time cannot be recommended.

Operative Treatment Surgical treatment of stress fractures is generally reserved for patients who fail to respond to a course of conservative treatment, who progress to complete fracture, or who have the high-risk type of stress fracture.12,19,20,22,23 At this time, intramedullary nailing is the surgical treatment of choice of most surgeons for patients with diaphyseal tibial stress fractures.11,19,20,22,23,61,84-88 Other options include drilling or excising the fracture site and, as recently described in the literature, a technique of tension-band plating.89

Weighing the Evidence Intramedullary nailing for the treatment of tibial stress fractures was first reported in the literature by Barrick and Jackson.84 Since then, there have been several case reports and small series (level IV evidence) of outcomes with this technique reported in the literature. Varner and associates88 treated seven athletes with reamed intramedullary nailing, and all were able to return to play. They had one patient with chronic anterior knee pain and one with a stress reaction at the distal nail tip. Chang and ­Harris19 reported five cases, with only two of five returning to unlimited activities. All had radiographic union and at least a significant reduction in their pain complaints. Brukner and colleagues85 reported a single patient with bilateral tibial stress fractures. There were four sites of fracture on one leg, two of which

Leg 1855

resolved after intramedullary nailing. The other leg had a single fracture that was managed nonoperatively. The operative leg became asymptomatic despite the lack of union on radiographs, whereas the nonoperative leg remained symptomatic on an activity-related basis. In a recent case report series, Borens and coworkers89 described their technique and results with anterior tension-band plating used to treat anterior tibial diaphyseal stress fractures in four female high-performance athletes. They thought that their technique directly addressed the mechanical deficiency while avoiding the potential complications of anterior knee pain associated with intramedullary nail placement. In their series of four cases, a return to preinjury activity levels occurred in an average of 10 weeks.

A u t h o rs ’ P r e f e rr e d M e t h o d The surgical technique for intramedullary nailing of tibial shaft stress fractures is the same as would be used for complete tibial fractures. The patient is positioned on a radiolucent table so that the leg can be imaged in full extension, and the knee is bent to 90 degrees to allow for reaming and nail insertion. A 3- to 4-cm anterior skin incision is used, ending distally at the tibial tubercle. Direct anterior transpatellar tendon or medial parapatellar approaches can be used at the preference of the surgeon. The knee joint itself should not be entered, instead traversing under the fat pad to the proximal tibial metaphysis. A starting guide pin is placed under fluoroscopy, with the starting point just proximal and posterior to the insertion of the patellar tendon. An entry awl or reamer is then passed over the guide pin while protecting the soft tissues with the appropriate soft tissue protection sleeve. The ball-tip guidewire is then placed down the tibial shaft, and reaming is begun. If needed, a fracture reduction tool can facilitate passage of the guidewire. Once the nail is passed and its distal position verified, the length of the nail is ­ selected by measuring off the guidewire. It is recommended­ to take the shorter length measured if between sizes to avoid prominence of the nail proximally and impingement on the patellar tendon. Reaming is begun typically with the smallest-sized endcutting reamer, and sequential reaming is carried out to a size 1 to 1.5 cm larger than the diameter of the nail to be used. The nail is then inserted over the ball-tip wire, or over a wire specifically for rod insertion. Initial placement of the nail should go smoothly by hand. As the nail encounters the posterior cortex, a posteriorly directed pressure on the driving jig will help to facilitate passage. The nail should be observed to advance with each blow of the mallet. If the nail does not advance, the bone should be checked fluoroscopically for cortex penetration or iatrogenic complete fracture. When fully seated, the proximal tip of the nail must rest below the anterior cortex of the tibia to avoid impingement on the patellar tendon. One to two proximal and distal interlocking screws are the placed, although some surgeons do not place any distal screws in the setting of incomplete stress fractures.

POSTOPERATIVE PRESCRIPTION AND OUTCOMES MEASUREMENT Postoperatively, the patient is allowed to bear weight as tolerated with crutches as needed and to begin knee range of motion exercises as soon as they can be tolerated. The patient can resume wearing a functional brace if one was worn preoperatively for pain control, and discontinuation of bracing and crutches can be done on a symptomrelated basis. Knee and ankle range of motion activities are initiated in the first postoperative week to avoid stiffness, and strengthening is begun within 2 to 3 weeks or as tolerated by the patient. Radiographic healing lags somewhat behind clinical healing, 3 and 2.7 months respectively, and return to play can be expected within 3 to 5 months.88

potenTIAL COMPLICATIONS Anterior Knee Pain Anterior knee pain is the primary complication of intramedullary fixation of tibial stress fractures. No study has directly addressed this complication in the athlete, but in the trauma literature, the incidence has been reported between 50% and 70%.90-95 In one series,91 56% of patients who had a tibial intramedullary nail placed ­ experienced anterior knee pain. Of these patients, 92% had pain with kneeling, 61% had pain with squatting, and 34% had pain at rest. Although earlier studies had indicated that a transtendinous surgical approach leads to an increase in anterior knee pain,92,96 a more recent prospective randomized trial by Toivanen and colleagues93 compared a transtendinous approach with a peritendinous approach and found no difference in the patient outcome with respect to anterior knee pain. Subsequently, the same group95 performed an ultrasound study that demonstrated no significant differences in the thickness or echogenicity of the patellar tendons whether there is pain or not. The fact that the incidence of anterior knee pain with bone– patellar tendon–bone anterior cruciate ligament reconstruction is much less than that reported with a simple incision to implant the nail would support the evidence that there is no significant difference with either surgical approach, and in our institution, a transtendinous approach for tibial intramedullary fixation is taught by the trauma ­service. The position of the proximal end of the nail has been associated with knee pain when it is prominent.90,96 Therefore, prominence of the nail may be an absolute indication for nail removal, whereas anterior knee pain may be less so. In the series by Court-Brown and colleagues91 following nail removal, 97% of the patients experienced some relief of their pain, with 27% having complete resolution of their anterior knee pain. Keating and associates92 reported less successful results with nail removal, with 20% of patients having no improvement and a further 35% experiencing only partial relief of their pain. In both series, more than half of the patients had some permanent anterior knee pain.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1856 DeLee & Drez’s� O

Recurrence of Stress Fracture

Untrained Athlete and Military Recruit

Recurrence of stress fracture or transformation to complete fracture following fixation is not well documented but is present. A recent case report by Baublitz and Shaffer97 demonstrates this risk. A collegiate basketball player who had intramedullary fixation, without interlocking screws, with curettage of the stress fracture defect, was returning to drills 6 weeks after surgery and sustained a spiral fracture at the site of the previous stress fracture. Therefore, continued observation for healing and patient counseling on the risks for fracture recurrence despite surgical treatment should be part of the discussion with the patient and coaches throughout the treatment process.

Overuse injuries are more common in the previously untrained athletic population, as seen with military recruits and older individuals just beginning an athletic endeavor. These athletes should be counseled in the nature of the diagnosis, the stress remodeling of bone, and a rational progression of physical activity to prevent or rehabilitate from a stress fracture. Visits with an athletic trainer may be beneficial to provide the inexperienced athlete with a rational progression program to avoid or rehabilitate from stress fractures and athletic overuse injuries in general.

CRITERIA FOR RETURN TO PLAY In all cases, a graded progression is followed before allowing an athlete to return to play. Initially, weight-bearing activities are resumed, followed by progressive loading, impact or plyometrics, running, and finally sport-specific activities. Gradual progression to impact or loading activities is achieved based on satisfactory performance at each level without pain. Low-risk stress fractures should be treated with a short period of relative rest and supportive bracing as needed. A graded progression to return to play is begun when the symptoms abate. High-risk stress fractures that are treated surgically should be allowed to return to play when there is radiographic evidence of fracture healing and resolution of pain. There may be instances of complete resolution of symptoms without radiographic signs of healing. In these instances, return to play may be allowed without radiographic consolidation, but the athlete and trainers must be instructed to cease activity and seek care with any return of symptoms because of the risk for completion of the ­fracture. Adjunctive therapies to aid in bone healing—­ultrasound, electrical and magnetic stimulation, and pharmacologic intervention—may be necessary to obtain complete radiographic and functional union and should be employed by practitioners based on their experience and the current ­literature. Finally, in instances of lack of clinical or radiographic healing following appropriate interventions, a discussion should be had with the athlete regarding the risks of continued athletic participation as opposed to alteration or cessation of athletic activities.

SPECIAL POPULATIONS Female Athlete The female athlete triad of altered menses, eating disorders, and low bone mineral density is a significant population in the discussion of stress fractures. A detailed discussion of this disorder is contained in Chapter 10. In any athlete with stress fractures, this diagnosis should be entertained and treated when appropriate.

C

r i t i c a l

P

o i n t s

l The risk category of a stress fracture is determined by its location and should guide management decisions. Highrisk stress fractures require more aggressive management, and possibly earlier surgical intervention, than low-risk stress fractures, which may, to the patient, be just as ­symptomatic. l Eating disorders and abnormal exercise behaviors are not confined to the female athlete. These should be investigated and treated appropriately with a multidisciplinary approach when appropriate. l Radiographic union will often lag behind clinical improvement, and the decision to advance in the rehabilitation program will need to be determined on an individual basis.

S U G G E S T E D

R E A D I N G S

Barrick EF, Jackson CB: Prophylactic intramedullary fixation of the tibia for stress fracture in a professional athlete. J Orthop Trauma 6(2):241-244, 1992. Gaeta M, Minutoli F, Scribano E, et al: CT and MR imaging findings in athletes with early tibial stress injuries: Comparison with bone scintigraphy findings and emphasis on cortical abnormalities. Radiology 235(2):553-561, 2005. Kaeding CC, Yu JR, Wright R, et al: Management and return to play of stress ­fractures. Clin J Sport Med 15(6):442-447, 2005. Taunton JE, McKenzie DC, Clement DB: The role of biomechanics in the epidemiology of injuries. Sports Med 6(2):107-120, 1988. Toivanen JA, Vaisto O, Kannus P, et al: Anterior knee pain after intramedullary nailing of fractures of the tibial shaft: A prospective, randomized study comparing two different nail-insertion techniques. J Bone Joint Surg Am 84(4):580-585, 2002.

R eferences Please see www.expertconsult.com

Leg 1857

S e c t i o n  

B

Leg Pain and Exertional Compartment ­Syndromes Bryce Bederka and Annunziato Amendola

Exercise-induced leg pain is a common problem in both recreational and competitive athletes. About half of sports injuries can be attributed to overuse.1 In endurance athletes, most injuries involve the leg (29%) or foot and ankle (27%).2 The term shin splints is commonly used but lacks a significant medical definition. Exercise-induced leg pain only describes the symptom; it is not a diagnosis. The differential diagnosis of leg pain in the athlete is broad (Box 24B-1). However, several entities account for most diagnoses. Studies have found that medial tibial stress syndrome (MTSS) and chronic exertional compartment syndrome (CECS) are the most common entities, with incidences ranging from 13% to 42% and 27% to 33%, respectively.3,4 Stress fracture and nerve entrapment are also common causes of exercise-induced leg pain.3-8 MTSS is similar to CECS in that it is a training-related condition that is exacerbated by activity and usually precipitated by alterations and increases in training programs. Detmer9 developed a classification system and treatment algorithm for MTSS, treating it as a continuum from stress reaction to stress fracture. MTSS can be differentiated from exertional compartment syndrome by history, physical examination, and imaging findings.3,10 The patient complains of pain that persists beyond the training session and is localized to the posteromedial aspect of the tibia. It tends to be diffuse in nature and may be reported as a dull ache. On examination, the patient has tenderness along the posteromedial border of the tibia, unlike in CECS, in which there generally are no pain findings at rest. Swelling may also be present. Plain imaging studies are generally normal or may demonstrate periosteal thickening in the area of irritation. Bone scan will be positive in a diffuse pattern, unlike the focal nature of stress fractures,11,12 and has been considered the standard in diagnosis of MTSS. Computed tomography (CT)13 and magnetic resonance imaging (MRI)6,14 have also been used in the evaluation of MTSS and its differentiation from other pathologic processes. MRI has demonstrated improved sensitivity and specificity when compared with conventional bone scans.14-16 CECS is a syndrome of pain in the leg when the normal expansion of the muscle from exercise demands results in increased pressures in an unyielding osseofascial compartment. The typical presentation is pain at a certain intensity of exercise or after a certain duration of exercise.8,17-21 The pain increases in intensity with further exercise and is alleviated with rest. Most commonly, it is seen in the leg, but it has also been reported in the thigh, foot, hand, and forearm musculature.

Box 24B-1  Differential Diagnosis of Leg Pain in the Athlete Muscle and Tendon Chronic exertional compartment syndrome Muscle strains Fascial hernias Tendinopathy Bone Medial tibial stress syndrome Stress fracture Vascular Popliteal artery entrapment syndrome Intermittent claudication Venous insufficiency Neurologic Peripheral or central nerve entrapment or impingement Referred pain from proximal joints (hip and knee) Tumor Infection

CECS of the legs is typically associated with younger athletes participating in running sports, but is also seen in jumping, cutting, and skating sports. CECS often presents in bilateral form, with equal incidence in male and female athletes.17,21,22 Of the four lower leg compartments, the anterior compartment is affected more frequently than the lateral, deep, and superficial posterior compartments.17-19,21-23 Activity-induced acute compartment syndromes associated with military training were described early in the literature.24-26 Often, these patients reported prodromal symptoms of exertion-related leg pain in the time before the development of acute compartment syndrome. In 1956, Mavor described the first reported case of an athlete with CECS.27 He described a professional soccer player with a 2-year history of exertional pain in the anterior legs, which was limiting his ability to play. The patient was treated with fasciotomy and fascia lata grafting; he had complete resolution of his symptoms and returned to a high level of play. Kirby, in 1970, described a case of superficial posterior CECS that was treated successfully by surgical ­ fasciotomy,28 and Puranen in 1974 described

�rthopaedic ����������� S �ports ������ � Medicine ������� 1858 DeLee & Drez’s� O

a case series of deep-posterior CECS treated successfully with surgery.29 These represent the earliest described cases of CECS reported in the athletic population and demonstrated early efficacy of surgical intervention.

RELEVANT ANATOMY AND BIOMECHANICS There are four muscle compartments (Fig. 24B-1), each bounded by bone and within its own investing fascia. Some authors have advocated for a separately functioning compartment for the posterior tibialis muscle; however, this has not been shown to be a consistent anatomic finding.30 Each compartment contains one or more muscles and one major neurovascular structure (Box 24B-2). The anterior compartment, most commonly involved, contains the extensor hallucis longus, extensor digitorum longus, peroneus tertius, and anterior tibialis muscles, as well as the deep peroneal nerve. The lateral compartment contains the peroneus longus and brevis, as well as the superficial peroneal nerve. The superficial posterior compartment contains the gastrocnemius and soleus muscles and the sural nerve; however, the fascia of this compartment is very thin, and diagnosis related to this compartment is rare. The deep posterior compartment contains the flexor hallucis longus, flexor digitorum longus, and posterior tibialis muscles, as well as the posterior tibialis nerve, and is the most commonly involved posterior compartment. Only two compartments contain significant vascular structures. The anterior compartment contains the anterior tibial artery and vein, which terminate in the dorsalis pedis vessels. The deep posterior compartment contains the posterior tibial artery and vein and the peroneal artery and vein, which terminate in the medial and lateral plantar arteries. The superficial peroneal nerve exits the fascia 11 cm proximal to the tip of the fibula. At this point, herniations may occur that can produce symptoms of CECS or nerve entrapment. The nerve is most vulnerable at this point, so

Box 24B-2  Anatomic Structures of the Four Compartments of the Leg Anterior Tibialis anterior muscle Extensor hallucis longus muscle Extensor digitorum longus muscle Peroneus tertius muscle Anterior tibial artery and vein Deep peroneal nerve Lateral Peroneus longus and brevis muscles Superficial peroneal nerve Superficial Posterior Gastrocnemius and soleus muscles Sural nerve Deep Posterior Flexor hallucis longus muscle Flexor digitorum longus muscle Posterior tibialis muscle Posterior tibial artery and vein Peroneal artery and vein Posterior tibial nerve

the distal incision is centered here to allow direct visualization and protection of the nerve. The cause of pain in CECS is unclear, and the pathophysiology is obscure. Normal muscle physiology reveals that there is up to a 20% increase of muscle volume during exercise.31,32 This is reflected in an increase in the intracompartmental pressures, even in normal asymptomatic individuals.33 This may be due to increased blood volume from increased blood flow, or it may reflect muscle fiber swelling and fluid retention in the muscle. Regular, relatively intense exercise results in muscular hypertrophy, which also leads to increased muscle volume. Some patients have muscle compartments that are confined by relatively tight, unyielding fascia. When the muscle volume increases during exercise, the pressure increases if the muscle has only a limited amount of potential expansion determined by the fascial boundaries. Symptoms of chronic compartment syndrome are thought to be caused by compromised neurovascular function due to elevated pressure within the involved compartment.34

CLASSIFICATION

Figure 24B-1   Cross-sectional diagram of the leg demonstrating the compartments and the location for medial and lateral fascial incisions: (1) anterior compartment, (2) lateral compartment, (3) superficial posterior compartment, (4) deep posterior compartment.

Compartment syndromes may be classified as either acute or chronic. Acute compartment syndromes typically are associated with trauma or an ischemic event. They have also been described in the setting of exercise, especially with the inception of a training program in otherwise untrained individuals, or with increases or changes in a training program,35 or with athletic injuries.36,37 Acute compartment syndromes are more severe in nature and require emergent wide surgical decompression to avoid serious and longterm sequelae.

Leg 1859

Exertional compartment syndromes, by contrast, are less severe. They are symptomatic only during the inciting exercise stress and resolve rapidly following cessation of exercise. They generally do not require acute surgical decompression unless it happens to convert to a more acute syndrome. No classification system has been developed and is in common use for the diagnosis of CECS. Typically, the specific compartments involved are described. The anterior compartment is most commonly involved (45%) followed by the deep posterior (40%), lateral (10%), and superficial posterior (5%) compartments.18 There is also a high incidence of bilaterality, reported to be from 50% to 70%.

EVALUATION The typical presentation is that of an athlete who complains of exercise-related leg pain that occurs at a reproducible point in the exercise session, increases in intensity with continued exercise, and is alleviated with rest. It is usually bilateral and usually involves the younger population (<30 years of age).38 The pain is often described as cramping or burning associated with the specific anatomic compartments involved. It is progressive with continued exercise and generally dissipates with decreasing the intensity or with rest. The athlete cannot run through the pain unless he or she does so at an intensity level much less than the inciting amount. The pain typically returns with the same intensity and duration at subsequent training sessions.8,17-21 In addition to pain, there is a variable amount of transient dorsiflexion weakness from ­compression of the

tibialis anterior muscle and numbness or burning from irritation of the peroneal nerve. In extreme cases, the pain may persist, and acute compartment syndrome may develop. The condition usually arises at the inception of a new exercise program or training season but can also occur when there is an increase in intensity of the workout or changes in training surface or footwear. At presentation to the physician, the physical examination is generally unremarkable. Therefore, examination following an exercise session that reproduces the ­symptoms is necessary. Tenderness, if present, is generally diffuse over the involved muscle compartments, as compared with focal tenderness for nerve entrapment.17,19,20 Focal muscle herniations may also be observed, typically at the exit of the superficial peroneal nerve from the lateral compartment,9,17 and in severe cases, weakness or neurologic deficits may be observed.9,17 Radiographs, bone scans, and MRI scans may be obtained, but these are generally normal in cases of CECS. Their value is in excluding alternative diagnoses when the clinical picture is unclear. MRI done in the acute setting may demonstrate changes in the signal intensity of the involved muscle compartments (Fig. 24B-2), but their value is limited owing to the need for pre-exercise and postexercise comparison MRI scans.39,40 Verleisdonk had athletes run on a treadmill next to the MRI scanner and completed postexercise T2 imaging within 4 minutes of exercise.40 Bone scans may be used to differentiate medial tibial stress syndrome,11,12 and an MRI angiogram can be performed if popliteal artery compression is suspected. Figure 24B-2   Magnetic resonance imaging findings in CECS involving the anterior compartment musculature: before exercise (top row); immediately 5 minutes after exercise (second row); 10 minutes after exercise (third row); 15 minutes after exercise (fourth row).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1860 DeLee & Drez’s� O

CECS, like acute compartment syndrome, is a clinical diagnosis. When the diagnosis is in question or the examination is equivocal, the confirmatory diagnostic test is intracompartmental pressure testing. Numerous devices are available and many ­ techniques have been described. Pressures may be checked at rest18,19,23,41,42 and during19,41,42 and after exercise,18,19,23,42 although pre-exercise and postexercise pressure levels are generally used now because of the difficulty in obtaining reliable pressure measurements during exercise.17-19,22,23 If CECS is not present, pressures generally return to normal within 3 to 5 minutes after exercise.41 If pressures remain elevated for 5 to 10 ­minutes after exercise, CECS is diagnosed. Many authors assert that pre-exercise pressure measurements greater than 15 mm Hg or a delay in normalization of pressures after exercise is key in confirming CECS.19 Currently, the Pedowitz criteria are the established standard for diagnosis of CECS; these include a pre-exercise resting compartment pressure of 15 mm Hg, 1- and 5-minute postexercise pressures of 30 mm Hg and 20 mm Hg, respectively, and any delay in the return to pre-exercise resting pressure levels.22 If 5-minute postexercise measurements are borderline, 15-minute postexercise compartmental pressure measurements must be obtained. Rorabeck and colleagues23 and Styf and Korner43 described an indwelling catheter so that measurement could be obtained continuously over 20 minutes after exercise and stated that the failure to return to baseline (<15 mm Hg by 15 minutes) was the most important. A diagnosis of CECS may be established with a resting pressure equal to or greater than 15 mm Hg, with a pressure equal to or greater than 30 mm Hg at 1 minute, or with a pressure equal to or greater than 20 mm Hg taken 5 minutes after exercise.18,22,23,43

TECHNIQUE OF INTRACOMPARTMENTAL PRESSURE MEASUREMENT Compartment pressures can be reliably measured using the Stryker Intra-Compartmental Pressure Monitor (STIC; Stryker Orthopaedics, Mawah, New Jersey). The STIC can be used for continuous monitoring but is more ­commonly used for intermittent monitoring with percutaneous needle penetration into the compartment. Initial measurements are taken before exercise and 1 and 5 minutes after exercise, with pressure readings taken in those compartments that are symptomatic. The patient lies supine on a table for the pressure measurements. The location for the measurements to be taken is marked on the skin in indelible ink; the skin is then anesthetized with local anesthetic and prepped in an aseptic fashion. The monitor device should be held horizontally and zeroed before each pressure reading. Anterior and lateral compartments are measured at their mid-­bellies directly through independent skin puncture wounds, whereas the superficial and deep compartments are measured through the same skin puncture wound, with further penetration for the deep compartment. With each pressure measurement, a small amount of fluid is injected, and the reading recorded. The modified Pedowitz criteria

  TABLE 24B-1  Modified Pedowitz Criteria Time

Pressure Measurement

Before exercise 1 min after exercise 5 min after exercise

>15 mm Hg >30 mm Hg >20 mm Hg

From Pedowitz RA, Hargens AR, Mubarak SJ, Gershuni DH: Modified ­criteria for the objective diagnosis of chronic compartment syndrome of the leg. Am J Sports Med 18(1):35-40, 1990.

(Table 24B-1) are used to confirm the ­presence of CECS in the involved compartments.

TREATMENT OPTIONS Initial management is generally nonoperative, consisting of a period of rest and removal of the inciting ­ physical activity. Patient education and activity modification are the mainstays of conservative management. Nonoperative treatments such as physical therapy, anti-­inflammatory agents, and orthotics play an important role in the management of many causes of exercise-induced leg pain; however, they usually are not effective in the management of CECS. Proper pre-exercise warm-up and stretching programs may be beneficial. During symptomatic periods, icing and a short course of anti-inflammatory medication may be used. The initial phase of treatment should include relative rest from the inciting activity, stretching, anti­inflammatory agents, physical therapy, correction of training errors, and corrective foot orthotics. Before return to activity, both extrinsic and intrinsic factors must be addressed and corrected if possible. Extrinsic factors that may be altered include the running surface, shoe design, and the training program. Stretching and strengthening exercises and orthoses may be used to address intrinsic factors inherent to muscle balance or limb alignment. Other biomechanical factors identified may also be addressed before return to activity as a means of altering the forces going through the involved compartment. Because adequate response to conservative measures may be difficult to obtain reliably, operative treatment should be employed if the athlete fails to respond to a rehabilita­ tion program of 3 to 6 months’ duration and wishes to continue with the activity associated with the CECS.9,17-19,42,44 Treatment is with fasciotomy, with or without fasciectomy, and resection of any fascial bands. Addressing areas of muscle herniation is also indicated for a com plete release. Open and percutaneous techniques have been described, and current trends are for limited incision techniques with rapid return to weight-bearing, motion, and resumption of activity. The symptomatic compartments should be addressed by surgical release. Previously, those individuals with anterior symptoms had anterior and lateral compartments released. Schepsis and colleagues45 found that, in patients with complaints isolated to the anterior compartment, isolated release of the anterior compartment produced results equal to a combined anterior and lateral compartment release. We now recommend addressing only those compartments that are symptomatic with positive pressure measurement.

Leg 1861

WEIGHING THE EVIDENCE There are no controlled trials in the literature comparing operative and nonoperative treatment of CECS and no ­studies comparing the different surgical procedures. Most surgical procedures report a high rate of satisfaction and return to unlimited physical activity, with 60% to 100% rates of relief.43,45-54 De Fijter and associates reported a 96% return to unlimited exercise in 118 military personnel following a percutaneous fasciotomy, with an average followup of 62 months.47 Raikin and colleagues reported bilateral simultaneous releases in 16 patients; 16 months after surgery, 13 patients were pain free, whereas 3 had continued mild but improved pain, and all returned to sports an average of 10.7 weeks after surgery.52 Moushine and coworkers reported 18 consecutive athletes treated with a two-incision­ fasciotomy technique, all of whom returned to full sporting activity at the 2-year follow-up, with average return to sporting activity of 25 days.51 Howard and colleagues reported slightly less favorable results of 68% pain relief, but when stratified by compartment, patients with anterior release had 81% relief, whereas posterior release yielded 50% relief.49 Slimmon and associates also had less favorable outcomes using a single-­incision technique.54 They reported 60% good or excellent results in patients undergoing a single

operation, with 58% of patients exercising at a lower level than before the development of symptoms. ­ Hutchinson and coworkers demonstrated incomplete releases with a single-incision technique,55 which may account for the less favorable outcomes. ­Schepsis and colleagues looked at the practice of releasing only the pressure-positive compartments or also releasing adjacent compartments.45 They found that release of only the involved anterior compartment and releasing both the ­anterior and lateral compartments yielded identical results, but that release of both compartments resulted in an average 3 weeks’ longer return to sports (11.4 ­versus 8.1 weeks). In conclusion, minimally invasive, percutaneous, and ­ single- and double-incision techniques are all currently used. There is evidence that single-incision techniques yield inferior results54 compared with double-­incision techniques. In general, there is a high rate of satisfactory outcome and return to sporting activities and a relatively low complication rate with surgical treatment of CECS. With accurate diagnosis of CECS, excellent results can be achieved if the procedure is performed properly. When a diagnosis of CECS has been made, the patient wishes to undergo ­surgical treatment, and modification of activity is unacceptable to the patient, fasciotomy of the affected compartments may be recommended.

A u t h o r s ’ P r e f e r r e d M e t h o d Anterior and Lateral Compartment Release

We prefer to use a double-incision technique similar to that described by Rorabeck and colleagues.53 The anterior intermuscular septum is usually superficially located and centered between the palpable anterior border of the tibia and the lateral border of the fibula. Two longitudinal incisions, 2 to 3 cm long, are centered over the intermuscular septum (IMS) at the junctions of the proximal and middle, and middle and distal, thirds of the leg (Fig. 24B-3). They are

carried down full-thickness to the muscle fascia. The IMS and superficial peroneal nerve can be easily identified. ­Using finger dissection, the plane is developed between the muscle fascia and subcutaneous fatty tissues from the knee to the ankle. One channel is made over each ­compartment to avoid making a large subcutaneous space and to minimize the occurrence of a seroma. The superficial peroneal nerve and any branches are visualized through the distal incision (Fig. 24B-4) and are protected throughout the

  Figure 24B-3   Skin markings for anterior and lateral compartment double-incision technique. Incisions are placed along the intermuscular septum, located about midway between the subcutaneous anterior border of the tibia and the subcutaneous lateral border of the fibula. The distal incision is centered over the exit of the superficial peroneal nerve, about   10 cm from the ankle joint line. The proximal incision is centered 10 cm distal to the proximal fibula.

Figure 24B-4   Superficial peroneal nerve in the distal incision. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 1862 DeLee & Drez’s� O

Authors’ Preferred Method—cont’d

Figure 24B-5   Fascial incisions for anterior and lateral compartment releases. Note the intermuscular raphe between the incisions.

Figure 24B-6   Fasciotomy is performed using long Metzenbaum scissors in a push-cut fashion. The nerve is visualized directly and protected in the distal incision.

procedure. The nerve is released if its exit from the ­fascia is felt to be tight. A small longitudinal incision is then made into the anterior and lateral fascia 1 cm on either side of the IMS at both incisions (Fig. 24B-5). From the proximal incision, the anterior and lateral compartment fasciotomy is carried proximally (Fig. 24B-6). We prefer to use 8- and 12-inch Metzenbaum scissors, but a fasciotome may also be used. The distal incision is then used to carry the fasciotomies distally to the level of the superior extensor retinaculum. Using either proximal or distal incisions, the fasciotomies are connected. The advantages of the doubleincision technique are that it gives easier access to the anterior and lateral compartment fascia adjacent to the IMS and confirmation of a complete fasciotomy. We strongly recommend when using this technique not to proceed with the fasciotomy until you have separated the subcutaneous tissue from the fascia. This decreases the risk for injury to the subcutaneous structures, and it makes passage of the instrument much easier and allows distal inspection to ­confirm complete fasciotomy.

must be complete because it also represents the proximal confluence of the flexor hallucis longus and flexor digitorum longus (FDL) fascia. This releases the deep posterior compartment. A Bristow is then used to release the tibialis posterior muscle off the tibia, completing the release of the tibialis posterior compartment (Fig. 24B-8). We have found this technique effective to release the deep posterior compartments. Remaining on the posterior aspect of the tibia throughout the release ensures safety of the posterior tibial neurovascular bundle, which is posterior to the tibialis posterior and FDL. Verification of an adequate release by digital examination is of the utmost importance. Following the anterior or posterior releases, the tourniquet is released, and hemostasis is obtained. The subcutaneous tissues are closed, and the skin is sutured using a subcuticular stitch. A sterile dressing and a compression bandage are applied to both legs. As mentioned, most patients have bilateral symptoms and hence undergo bilateral procedures. Patients with anterior and posterior symptoms have both compartments released.

Posterior Compartment Release

We use a single-incision technique for the release of the ­superficial and deep posterior compartments. The incision is located 1 cm posterior to the posterior subcutaneous border of the tibia. It is centered at the level of the distal gastrocnemius curve and is 8 to 10 cm long (Fig. 24B-7). The long saphenous nerve and vein are usually in the center of the field and are easily identified on the posteromedial border of the tibia. Proximally, the flexor digitorum longus ­occupies this position. A small vertical incision is made at the osseofascial junction, and then, using Metzenbaum scissors and staying directly on the posterior border of the tibia, the fascia is released to the level of the tibialis posterior tendon. The surgeon’s finger should follow the instrument to ensure a complete release. The release is then taken proximally. The soleus will be encountered in the proximal one third of the tibia at the soleus bridge. Release of this stout structure

Figure 24B-7   Skin marking for posterior compartment releases. A 10-cm incision is located along the posteromedial subcutaneous border of the tibia, centered at the distal insertion of the gastrocnemius muscle.

Leg 1863

A

B

Figure 24B-8   A and B, Fascial incisions for posterior compartment releases. The muscle fascia is taken directly off of the posteromedial border of the tibia.

POSTOPERATIVE PRESCRIPTION, OUTCOMES MEASUREMENT, AND POTENTIAL COMPLICATIONS Weight-bearing is initiated immediately after surgery, with crutches discontinued as tolerated. Early passive and active range of motion exercises are implemented postoperatively to prevent postoperative fascial scarring.8,17,18,20 During the first 2 days, patients follow a RICE (rest, ice, compression, and elevation) protocol as well as anterior and posterior stretching (toe pointing) 3 to 5 times per day. From the third day to the 2-week follow-up visit, patients perform aggressive anterior and posterior compartment stretches 3 times per day and increase the walking distance. Once they are weaned from crutches, nonimpact activities such as ­ hydrotherapy, stationary cycling, and elliptical training are begun. After 2 weeks, the wounds are checked, and a ­ formal ­ physical therapy regimen of stretching and func­tional return to sport-specific activity is begun. When strength and control of the ankle and foot are regained, functional training can begin, usually by 4 to 6 weeks. At that point, running may be implemented, with speed and agility drills added during the eighth week.17,18,20 By 8 to 12 weeks after surgery, athletes typically return to full participation in sports. Complications reported have included hematoma or seroma formation (9%), superficial peroneal nerve injury (2%), anterior ankle pain (5%), and recurrence of symptoms (2%).47 A recently published study in the vascular surgery literature highlighted the function of the muscles and their compartments in the return of fluids in the dependent limb and raised the concern of venous insufficiency after fascial release, but there have not been any documented cases to date with clinically significant findings.56

CRITERIA FOR RETURN TO PLAY Objective criteria for return to play following fascial release for CECS do not exist. The return to play is based on satisfactory completion of the progression outlined in the preceding section. The athlete should be nearly pain free, have demonstrated acceptable strength and endurance, and be able to replicate the demands of practice and play in the therapy sessions. Return to full athletic activities should be accomplished by 8 to 12 weeks after surgical intervention. C

r i t i c a l

P

o i n t s

l History on presentation is the most important part of the

l

l l

clinical examination for sifting through the differential diagnosis. The onset of symptoms will be at a consistently reproducible point in the exercise routine and will be relieved only by rest. Double-incision techniques are favored laterally to ­ allow dissection and visualization of the superficial peroneal nerve to avoid injury. Dissection between the subcutaneous tissues and fascia is performed bluntly using the fingers. For anterior and lateral releases, separate channels are formed for each compartment instead of making one large pocket over the entire anterolateral portion of the leg. The superficial peroneal nerve exits the lateral compartment fascia 11 cm from the distal tip of the fibula. The distal incision should be centered at this point to allow for direct visualization and protection of the nerve. Early range of motion exercises and weight-bearing are encouraged postoperatively to avoid secondary scarring after the fasciotomy. Crutch use is as needed and is generally discontinued within 1 week after surgery.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1864 DeLee & Drez’s� O

S U G G E S T E D

R E A D I N G S

Mavor GE: The anterior tibial syndrome. J Bone Joint Surg Br 38B(2):513-517, 1956. Pedowitz RA, Hargens AR, Mubarak SJ, Gershuni DH: Modified criteria for the objective diagnosis of chronic compartment syndrome of the leg. Am J Sports Med 18(1):35-40, 1990. Rorabeck CH, Bourne RB, Fowler PJ: The surgical treatment of exertional compartment syndrome in athletes. J Bone Joint Surg Am 65(9):1245-1251, 1983.

Styf J, Korner L, Suurkula M: Intramuscular pressure and muscle blood flow during exercise in chronic compartment syndrome. J Bone Joint Surg Br 69(2):301-305, 1987.

R eferences Please see www.expertconsult.com

C H A P T E R  

 25

Foot and Ankle S ect i o n

A

Biomechanics Andrew Haskell and Roger A. Mann

ANKLE JOINT The ankle joint allows sagittal plane motion of 20 degrees of dorsiflexion and 50 degrees of plantar flexion; however, there is a great deal of variability among individuals. The ankle joint is not a simple hinge joint, but rather the trochlear surface of the talus is a section from a cone whose apex is based medially (Fig. 25A-2).4 The talus is stabilized within the ankle mortise by bony and soft tissue restraints. The congruity of the ankle mortise leads to considerable inherent bony stability.5,6 Ligament support includes the deltoid ligament medially7 and three separate ligamentous bands laterally: the anterior and posterior talofibular ligaments and the calcaneofibular ligament.8 The anterior talofibular ligament is taut with the ankle joint in plantar flexion, when it is in line with the fibula; and the calcaneofibular ligament is taut with the ankle joint in dorsiflexion, when the ligament is in line with the fibula 250 Body Weight (%)

This section discusses the biomechanical linkage of the joints of the foot and ankle and their effects on the lower extremity. The ankle joint, subtalar joint, transverse tarsal joint, and metatarsophalangeal joints are uniquely interrelated so that function in one reliably alters the mechanics of the others. The foot and ankle complex initially helps absorb the impact of ground contact and later in stance provides the body with a stable platform from which to function. During walking, the toes are lifted from the floor, but in athletics, forceful push-off facilitates rapid acceleration and deceleration, direction changes, and jumping. Dysfunction of the foot and ankle complex may result in an altered gait pattern, degradation of athletic performance, and compensatory changes in the knee and hip joints. The ankle and subtalar joint complex function as a universal joint, linking pelvic, thigh, and leg rotation to hindfoot motion and longitudinal arch stability. This allows the ankle joints to compensate for some degree of dysfunction in the hindfoot, and vice versa. Athletics, however, requires maximal performance from these systems, and dysfunction of ankle and foot mechanics often leads to pain, injury, and loss of performance. Athletics is distinguished from normal walking by the stresses applied to the joints. The stresses can be repetitive, as in a long distance runner, or impulsive, as occurs in the pushoff foot of a shot-putter. The vertical force involved in running is 2 to 2.5 times body weight compared with 1.2 times body weight in walking,1 and can be higher for many sports with extreme push-off, such as a football lineman engaged in blocking or a basketball player engaged in rapid accelerating and jumping activities (Fig. 25A-1). The nature of these forces depends on the activity and includes vertical force, fore-and-aft shear, side-to-side shear, and torque forces. These forces are measured in a variety of ways, including force plate analysis or thin film pressure transducers placed in a shoe.2 The foot and ankle complex must be supple enough to absorb impact and rigid enough to transmit muscle forces, or injuries such as sprains, strains, stress fractures, and fascial tears may result. Athletic training can help to attenuate these forces and minimize the risk for injury.3

Walking Running

200 150 100 50

0

25

50

75

100

Time (% Stance Phase) Figure 25A-1   Comparison of vertical ground reaction force for walking (blue line) compared with jogging (red line). The horizontal axis is scaled as a percentage of total time in stance phase for walking (0.6 sec) and running (0.24 sec). The vertical axis is shown as a percentage of body weight. (From Mann RA, Haskell A: Biomechanics of the foot and ankle. In Coughlin MJ, Mann RA, Saltzman CL [eds]: Surgery of the Foot and Ankle, 8th ed. Philadelphia, Mosby, 2007.)

1865

�rthopaedic ����������� S �ports ������ � Medicine ������� 1866 DeLee & Drez’s� O

Figure 25A-2   The trochlear surface of the talus is a section from a cone. The apex of the cone is directed medially, and the open end is directed laterally. (From Stiehl JB [ed]: Inman’s Joints of the Ankle, 2nd ed. Baltimore, Williams & Wilkins, 1991.)

(Fig. 25A-3).4 The anterior talofibular ligament is injured most frequently during ankle sprains, in part because the ankle has less intrinsic bony stability in plantar flexion when this ligament is under tension.9 Isolated injuries to the calcaneofibular ligament are less frequent, although they often occur in conjunction with anterior talofibular sprains. Dorsiflexion and plantar flexion occur at the ankle joint during gait. At heel strike, the dorsiflexed ankle rapidly plantar flexes. This ends at foot flat, after which ­progressive

A

B

dorsiflexion occurs. Dorsiflexion reaches a maximum at 40% of the walking cycle, when plantar flexion begins as the heel rises, and continues until toe-off, when dorsiflexion occurs again during the swing phase (Fig. 25A-4). The force applied across the ankle joint during walking has been measured at about 4.5 times body weight.10 This maximal stress occurs just before and just after the onset of plantar flexion of the ankle joint. If this force were extrapolated to running, in which the ground reaction force is over double that of walking, we would see stress across the ankle joint that approaches 10 times body weight. Muscle control of the ankle joint can be divided into the anterior and posterior compartments. The anterior compartment consists of the tibialis anterior, extensor hallucis longus, and extensor digitorum longus. The posterior compartment consists of the gastrocnemius-soleus group, tibialis posterior, flexor digitorum longus, and flexor hallucis longus. The lateral compartment, consisting of the peroneus longus and brevis, functions with the posterior compartment. During normal walking, the anterior compartment muscles become active late in the stance phase and in the swing phase to bring about dorsiflexion of the ankle joint by a concentric (shortening) contraction (see Fig. 25A-4).11 This muscle group remains active after heel strike to control the rapid plantar flexion of the ankle joint that occurs by an eccentric (lengthening) contraction. This eccentric contraction during plantar flexion helps to dissipate the forces on the limb at initial ground contact. The anterior compartment becomes electrically silent by foot flat. The posterior compartment muscles are active after foot flat (15% to 20% of gait cycle), during which time the ankle joint is undergoing dorsiflexion, and they remain active until about halfway through the cycle, at which time about 50% to 60% of ankle joint plantar flexion has occurred.12 This muscle group initially undergoes an eccentric contraction controlling forward movement of the tibia over the foot, then a concentric contraction when plantar flexion begins. The electrical activity of the posterior calf group ceases before full plantar flexion has occurred, ­indicating that the last portion of plantar flexion is a passive phenomenon.

C

Figure 25A-3   Calcaneofibular (a) and anterior talofibular (b) ligaments. A, In plantar flexion, the anterior talofibular ligament is in line with the fibula, thereby providing most of the support to the lateral aspect of the ankle joint. B, When the ankle is in neutral position, both the anterior talofibular and the calcaneofibular ligaments support the joint. The obliquely placed structure depicts the axis of the subtalar joint. It should be noted that the calcaneofibular ligament parallels the axis. C, When the ankle joint is in dorsiflexion, the calcaneofibular ligament is in line with the fibula and supports the lateral aspect of the joint. (A to C, From Stiehl JB [ed]: Inman’s Joints of the Ankle, 2nd ed. Baltimore, Williams & Wilkins, 1991.)

Foot and Ankle 1867

As the speed of gait increases to steady running and sprinting, several changes occur. The ankle joint starts in slight dorsiflexion and, at heel strike, rather than the ankle plantar flexing to a foot flat position, the ankle remains dorsiflexed (see Fig. 25A-4). The tibia moves forward, and the foot flat position is achieved. As running speed increases, the magnitude of the motion increases, the stance phase is reduced significantly, and the period of double limb support gives way to a period lacking limb support. The electrical activity of the anterior compartment still begins late in stance and continues through swing, but it now lasts through about the first third of the stance phase. The posterior calf muscles show a significant change in their activity in that they become active late in swing phase and remain active until about halfway through ankle joint plantar flexion.13 This increased activity in the posterior calf musculature probably results in increased stability of the ankle joint at the time of initial ground contact. 30

SUBTALAR JOINT The subtalar joint is a complex joint that permits inversion and eversion. The axis of the subtalar joint is variable but is about 42 degrees to the horizontal plane and passes from medial to lateral at about 16 degrees (Fig. 25A-5).10,14 Measurement of subtalar joint movement during walking and running is difficult and is based on many assumptions. Although the overall pattern of motion appears to be consistent, the magnitude of motion is variable.15 The range of motion of the subtalar joint in its pure form, inversion and eversion, is 30 degrees of inversion and 15 degrees of eversion, with considerable variability among people. To measure this motion accurately, it is important to start with the calcaneus in line with the tibia. The subtalar joint is less constrained than the ankle, being stabilized primarily by the joint configuration and the interosseous ligament.16,17 This joint is stable when the

PLANTAR FLEXION-DORSIFLEXION RUN

Toe-off

20 dorsiflexion 10

Heel strike

0 10 plantar flexion 20 30

gastrocnemius-soleus anterior tibial

30 20 dorsiflexion 10 degrees

JOG

Toe-off

Heel strike

0

10 plantar flexion 20 30

gastrocnemius-soleus anterior tibial

30

Toe-off

20 dorsiflexion 10

WALK

Heel strike

0 10 plantar flexion 20 30

gastrocnemius-soleus anterior tibial

0

0.1

0.2 0.3

0.4

0.5 0.6 sec.

0.7 0.8

0.9

1.0

Figure 25A-4   Ankle joint range of motion for walking, jogging, and running. The muscle function of the anterior and posterior compartment is noted on the bottom of the graph. (Redrawn from Mann RA: Biomechanics of running. In American Academy of Orthopaedic Surgeons: Symposium on the Foot and Leg in Running Sports. St. Louis, CV Mosby, 1982.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1868 DeLee & Drez’s� O

tala Sub is ax

NT

69°

r

OF OI IS R J X A LA TA UB

x = 23°

S

21°

4°

47°

x = 41°

Horiz. plane

B

A

Figure 25A-5   Variations in the subtalar joint axes. In the horizontal plane (A), the axis approximates 45 degrees and (B) passes about 23 degrees medial to the midline. (A and B, Adapted from Isman RE, Inman VT: Anthropometric studies of the human foot and ankle. Bull Prosthet Res 10:97, 1969.)

long axis of the tibia passes medial to the obliquely placed subtalar joint axis. In the normal foot, subtalar joint eversion ceases with weight-bearing owing to the configuration of the joint surfaces and the interosseous ligament. When the weight-bearing line is lateral to the subtalar joint axis, inversion stability depends on lateral ligament support and active muscle function. The subtalar joint has been likened to an oblique hinge that functions to translate motion between the transverse tarsal joint distally and the ankle joint and leg proximally (Fig. 25A-6).18,19 This linkage is important for energy dissipation at heel strike. At the time of initial ground contact during walking, the slightly inverted subtalar joint

A

B

Figure 25A-6   Mitered hinge effect of subtalar joint. The joint acts as a mitered hinge, converting motion in the calcaneus below into the tibia above and, conversely, from the tibia above into the calcaneus below. (A and B, Redrawn from Mann RA, Haskell A: Biomechanics of the foot and ankle. In Coughlin MJ, Mann RA, Saltzman CL [eds]: Surgery of the Foot and Ankle, 8th ed. Philadelphia, Mosby, 2007.)

­ ndergoes rapid eversion, the tibia undergoes internal u rotation, the transverse tarsal joints become supple, and the medial longitudinal arch flattens (Fig. 25A-7). These are passive energy-absorbing mechanisms. In a person with flatfoot and increased eversion of the subtalar joint (Fig. 25A-8), an increased amount of tibial internal rotation may occur, which can affect the knee, patellofemoral, or hip joint in selected cases. An orthotic device that supports the longitudinal arch with medial heel posting may restrict subtalar joint rotation and decrease the internal rotation of the lower extremity, possibly resolving knee or hip pain. This linkage is also important for efficient energy transfer during heel rise and toe-off. After the initial eversion, the subtalar joint undergoes progressive inversion, which reaches a maximum at toe-off, when eversion begins again. This movement increases the stability of the transverse tarsal joints and medial longitudinal arch, stiffening the foot and allowing it to act as a rigid extension of the leg. Although the initial eversion is passive, the inversion that follows appears to be both passive and active. The inversion results from an external rotation torque from the lower extremity above, which is transmitted across the ankle joint and is translated by the subtalar joint into inversion. The plantar aponeurosis mechanism and the oblique metatarsal break enhance the inversion, as described later. The muscle function around the subtalar joint can be appreciated best by looking at the muscles in relation to the subtalar joint axis (Fig. 25A-9). Muscles medial to the axis are invertors, muscles lateral to it are evertors, and the function of the muscles on the axis is determined by the position of the subtalar joint. The main invertors are the tibialis posterior and the gastrocnemius-soleus complex, and the main evertor is the peroneus brevis and, to a much lesser extent, the peroneus longus, which is mainly a plantar flexor of the first metatarsal. The tibialis anterior lies on the ­ subtalar

Foot and Ankle 1869

A

joint axis and has little influence on the subtalar joint, although it is the only functioning muscle at heel strike and, as such, besides controlling plantar flexion of the ankle joint, may resist eversion at the subtalar joint. The inversion that occurs during the last half of stance is due to the passive mechanisms noted previously, along with the input from the gastrocnemius-soleus complex and the posterior tibialis. The patient who lacks posterior tibial tendon function cannot initiate standing on tiptoe but can maintain the position when it is achieved. It can be concluded that posterior tibial tendon function is necessary to initiate inversion, and the gastrocnemius-soleus complex is necessary to maintain it. As noted in the discussion of ankle function, with running, the posterior calf muscles become active late in swing phase and remain active through most of stance (see Fig. 25A-4). Besides providing stability to the ankle joint, these muscles probably bring about some inversion of the subtalar joint before initial ground contact.

B

TRANSVERSE TARSAL JOINT C

D

Figure 25A-7   Model demonstrating flattening and elevation of the longitudinal arch. A and B, Flattening of the longitudinal arch occurs at the time of heel strike with eversion of the calcaneus and internal rotation of the tibia. C and D, Elevation and stabilization of the longitudinal arch are associated with the outward rotation of the tibia, causing inversion of the calcaneus and locking of the transverse tarsal joint. (Redrawn from Mann RA, Haskell A: Biomechanics of the foot and ankle. In Coughlin MJ, Mann RA, Saltzman CL [eds]: Surgery of the Foot and Ankle, 8th ed. Philadelphia, Mosby, 2007.)

The transverse tarsal joint, consisting of the ­talonavicular and calcaneocuboid joints, lies distal to the subtalar joint and is influenced strongly by it.20 The motion of the transverse tarsal joint is that of adduction and abduction and is measured with the calcaneus in neutral position and the forefoot parallel to the floor. Normal motion is about 20 degrees of adduction and 10 degrees of abduction. The main support of the joint is ligamentous, but its stability is derived from subtalar joint inversion without much direct muscle control. The axes of the transverse tarsal joint are aligned such that when the calcaneus is in an everted position, the axes are parallel, permitting more motion to occur around this joint system. During normal walking, this occurs at heel strike, creating a flexible foot to absorb the energy of impact. When the calcaneus is inverted, the axes of the transverse tarsal joint are nonparallel, creating a stable joint system (Fig. 25A-10).20 This occurs at heel rise and toe-off, creating a rigid foot to effectively lengthen the limb and assist in propulsion during running.

SUBTALAR ROTATION HEEL STRIKE

TOE-OFF

HEEL STRIKE

10

NORMAL FOOT

0 INVERSION 10 EVERSION

FLAT FOOT

0

10

0

10

20

30

40

50

60

70

80

90

100

PERCENT OF WALK CYCLE Figure 25A-8   Graph of subtalar joint motion in the normal individual and in a flatfooted individual. (Redrawn with permission from data in Wright DG, Desai ME, Henderson BS: Action of the subtalar and ankle joint complex during the stance phase of walking. J Bone Joint Surg Am 46:361, 1964.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1870 DeLee & Drez’s� O Dorsiflexors

n

rsio Inve

Invertors

Evertors

s axi lar bta Su

Plantar flexi

Tib. ant.

Ext. hal. longus Ext. dig. longus

on

Dorsiflexion ion

rs Eve

Ankle a

xis

Tib. post F. dig. longus F. hal. longus

Peroneus long Peroneus brevis

T. calcaneus

A

Plantar flexors

B

Figure 25A-9   A, The location and the types of rotation that occur about the ankle and the subtalar axes. B, The relationship of the various extrinsic muscles about the subtalar and ankle joint axes. Ext. dig. longus, extensor digitorum longus; Ext. hal. longus, extensor hallucis longus; F. dig. longus, flexor digitorum longus; F. hal. longus, flexor hallucis longus; T. calcaneus, tibialis calcaneus. (Redrawn from Haskell A, Mann RA: Biomechanics of the foot. In American Academy of Orthopaedic Surgeons: Atlas of Orthoses and Assistive Devices. Philadelphia, Elsevier, 2008.)

EVERSION

INVERSION TN

TN

C

C C

The windlass mechanism describes the function of the plantar aponeurosis during gait. The plantar aponeurosis arises from the tubercle of the calcaneus and inserts into the base of the proximal phalanges (Fig. 25A-11). After heel rise, the metatarsophalangeal joints dorsiflex, tightening the plantar aponeurosis. This depresses the metatarsal heads, elevates and stabilizes the longitudinal arch, and helps to bring the calcaneus into an inverted position (Fig. 25A-12).21 The inverted calcaneus causes the transverse tarsal joint axes to diverge, helping to stabilize the midfoot at toe-off. The oblique metatarsal break is created by the lateral slope formed by the metatarsophalangeal joints two through five (Fig. 25A-13).18 This oblique line creates a cam-like action as the body weight is brought over the metatarsal heads, further enhancing external rotation of the lower extremity and inversion of the calcaneus.

joint motions are enhanced further by the function of the plantar aponeurosis, the transverse metatarsal break, and the muscles of the leg and foot. It is this linked series of movements that enables the athlete to absorb the forces of impact, yet create a rigid platform from which to push off. These joint linkages are essential for high-performance function of the lower extremity. If one of these linkages

C

WINDLASS MECHANISM AND METATARSAL BREAK

LINKAGE OF THE FOOT AND ANKLE The functions of the ankle joint, subtalar joint, transverse tarsal joint, and plantar aponeurosis have been examined and their interdependence described. The linkage between these joints should be emphasized further. The terms pronation and supination describe a coordinated series of movements of the foot and ankle that facilitate its two main functions during gait, namely, energy absorption at impact and energy transfer during stance (Table 25A-1). These

Figure 25A-10   The function of the transverse tarsal joint as described by Elftman. When the calcaneus is in eversion, the resultant axes of the talonavicular (TN) and calcaneocuboid (CC) joints are parallel or congruent. When the subtalar joint is in an inverted position, the axes are incongruent, giving increased stability to the midfoot. (Redrawn from Mann RA, Haskell A: Biomechanics of the foot and ankle. In Coughlin MJ, Mann RA, Saltzman CL [eds]: Surgery of the Foot and Ankle, 8th ed. Philadelphia, Mosby, 2007.)

Foot and Ankle 1871

Flexor tendon

Capsule

Plantar pads

Plantar pad Plantar aponeurosis

C

Lateral

Medial

A D

B Figure 25A-11   Plantar aponeurosis. A, Cross section. B, The plantar aponeurosis originates from the tubercle of the calcaneus and passes forward to insert into the base of the proximal phalanges. The aponeurosis divides, permitting the long flexor tendon to pass distally. C, Components of the plantar pad and its insertion into the base of the proximal phalanx. D, Extension of the toes draws the plantar pad over the metatarsal head, pushing it into plantar flexion. (From Mann RA, Haskell A: Biomechanics of the foot and ankle. In Coughlin MJ, Mann RA, Saltzman CL [eds]: Surgery of the Foot and Ankle, 8th ed. Philadelphia, Mosby, 2007, p 24.)

within the system fails to function properly, stress is placed on the joints proximal and distal to it. Although we speak of ankle joint dorsiflexion and plantar flexion, only about half of this motion comes from the ankle joint; the remainder comes from the movement occurring within the subtalar and transverse tarsal joints.22 If there is diminished motion of the ankle joint, perhaps from an anterior impingement,

ME

TA BR TARS EA AL K

x= 62°

53.5°

72.5°

Figure 25A-12   The function of the plantar aponeurosis. The brown outline shows the medial column with the foot at rest. The red figure shows the medial column with the first ray dorsiflexed. Note that dorsiflexion of the metatarsophalangeal joints tightens the plantar aponeurosis, which results in depression of the metatarsal heads, elevation and shortening of the longitudinal arch, inversion of the calcaneus, and elevation of the calcaneal pitch.

Figure 25A-13   The metatarsal break passes obliquely at an angle of about 62 degrees to the long axis of the foot. (Adapted from Isman RE, Inman VT: Anthropometric studies of the human foot and ankle. Bull Prosthet Res 10:97, 1969.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1872 DeLee & Drez’s� O

  TABLE 25A-1  Comparison of Foot Characteristics Based on Foot Position

Foot Position Characteristic

Pronation

Supination

Joint position

Ankle dorsiflexion Subtalar eversion Transverse tarsal abduction Supple Heel strike Energy absorption

Ankle plantar flexion Subtalar inversion Transverse tarsal adduction Rigid Foot flat to toe-off Energy transfer to ground

Arch stiffness Gait cycle Function

degenerative changes within the joint, or fusion, the subtalar and transverse tarsal joints compensate for the lost motion. If there are degenerative changes within the subtalar or transverse tarsal joints, any loss of ankle joint motion is magnified. This compensatory increase in motion of the neighboring joints often leads to pain, loss of function, and degenerative changes over time.23 Moving distally, if the motion in the subtalar joint is restricted, its ability to translate rotation proximally and distally is impaired, placing increased stress on the ankle and transverse tarsal joints. Talocalcaneal coalition can lead to a spastic peroneal flatfoot or ball-in-socket ankle because of the effect of lack of subtalar motion. The degree of ankle joint dorsiflexion and plantar flexion also is affected, and the ankle can become arthritic from the abnormal stresses.24 Impairment of the transverse tarsal joint impairs subtalar joint motion because for subtalar motion to occur, rotation must occur around the talonavicular joint as well as the calcaneocuboid joint. If an isolated arthrodesis of the talonavicular or calcaneocuboid joint is carried out, most subtalar joint motion is lost.25 When performing an arthrodesis around the hindfoot, sparing the talonavicular joint when appropriate usually leaves the patients with a more functional foot. The metatarsophalangeal joints also are affected by loss of motion. First metatarsophalangeal joint dorsiflexion is lost in hallux rigidus, a degenerative arthritis of the first metatarsophalangeal joint. This leads to a compensatory external rotation of the foot during gait to relieve the stress across the involved area. This compensation, in turn, can affect the alignment of the lower extremity. The theory behind orthotic use for many conditions involving the foot, ankle, knee, hip, and back is the effect it has on this linkage system within the lower extremity. Soft orthoses and compliant shoe material help absorb the impact of initial ground contact. For individuals engaged in repetitive sports, such as long-distance running, a material that helps absorb some of this impact could be beneficial if the athlete is having problems related to impact, such as heel pain, metatarsalgia, or shin splints. However, softer material paradoxically can lead to greater vertical impact when landing from jumps in an attempt to improve balance and stability.26 On a more sophisticated level, the use of a medial heel wedge, whether in the shoe or within an orthotic device, may have some influence on the rotation of the subtalar joint.27 Because at the time of initial ground contact rapid eversion

of the subtalar joint and flattening of the longitudinal arch occur, a buildup of material along the medial arch that prevents some of this rotation from occurring, in theory, would decrease the amount of internal rotation being transmitted to the lower extremity, affecting the ankle, knee, and hip. From a clinical standpoint, some patients appear to benefit from an orthotic device, although the benefit may be in part psychological.28 A runner with chronic knee pain may be helped by an orthotic device that limits eversion of the calcaneus, which, in turn, diminishes internal rotation of the tibia and affects the patellofemoral joint. Reliable data to support this theory are lacking.

C

r i t i c a l

P

o i n t s

At heel strike, the foot and ankle help absorb the force of

l

contact with the ground. l During the heel rise and toe-off phases of gate, the foot becomes more rigid, providing a stable platform for the body. l The subtalar joint links motion of the leg and foot such that eversion of the hindfoot causes internal rotation of the tibia at heel strike, and external rotation of the tibia causes inversion of the hindfoot at heel rise. l Eversion of the hindfoot unlocks the transverse tarsal joints, making the foot supple; inversion of the hindfoot locks the transverse tarsal joints, making the foot more rigid. l The muscles of the leg and foot contract both concentrically and eccentrically during the gait cycle to control the rate of ankle plantar flexion and dorsiflexion during walking and to provide stability during running. l Athletics increases the normal stresses on the foot and ankle and can lead to acute or overuse injuries.

S U G G E S T E D

R E A D I N G S

Elftman H: The transverse tarsal joint and its control. Clin Orthop 16:41, 1960. Gross ML, Davlin LB, Evanski PM: Effectiveness of orthotic shoe inserts in the long-distance runner. Am J Sports Med 19:409, 1991. Hicks JH: The mechanics of the foot: II. The plantar aponeurosis and the arch. J Anat 88:25, 1954. Isman RE, Inman VT: Anthropometric studies of the human foot and ankle. Bull Prosthet Res 10:97, 1969. Mann RA, Beaman DN, Horton GA: Isolated subtalar arthrodesis. Foot Ankle Int 19:511, 1998. Mann RA, Haskell A: Biomechanics of the foot and ankle. In Coughlin MJ, Mann RA, Saltzman CL (eds): Surgery of the Foot and Ankle, 8th ed. Philadelphia, Mosby, 2007. Nilsson J, Thorstensson A, Halbertsma J: Changes in leg movements and muscle activity with speed of locomotion and mode of progression in humans. Acta Physiol Scand 123:457, 1985. Tochigi Y, Amendola A, Rudert MJ, et al: The role of the interosseous talocalcaneal ligament in subtalar joint stability. Foot Ankle Int 25:588, 2004. Tochigi Y, Rudert MJ, Saltzman CL, et al: Contribution of articular geometry to ankle stabilization. J Bone Joint Surg Am 88:2704, 2006. Wülker N, Stukenborg C, Savory KM, et al: Hindfoot motion after isolated and combined arthrodeses: Measurements in anatomic specimens. Foot Ankle Int 21:921, 2000.

R eferences Please see www.expertconsult.com

Foot and Ankle 1873

S ect i o n

B

Sports Shoes and Orthoses Andrew H. Borom and Thomas O. Clanton

There should be little surprise that an entire section of one chapter would be devoted to a discussion of sports shoes. The athletic shoewear industry has grown to such an extent that it has reached one of the pinnacles of achievement in our society—the front cover of Sports Illustrated (Fig. 25B-1).1 With this achievement, both increasing recognition from Wall Street investors and harsh criticism alleging consumer exploitation surfaced.2-4 Sports shoes became “high tech” and rode a wave of advertising to become a status symbol.4,5 Among today’s youth, the “right” shoe may vary from week to week. Athletic shoe sales rose to approach $8 billion in 1998, with fully half of that total brought in by the top two athletic shoe manufacturers. Not surprisingly, the top retailer of athletic shoes spent more on promotions and advertisements (some $163.2 million) than the next nine producers combined.6 Major shoewear manufacturers have paid six-figure salaries to high-profile athletes and coaches to endorse their products.4,7 Financial benefits from shoe contracts have become a major consideration for college athletic programs and their coaches.7 In this distorted environment, it is often difficult to wade through the hype to discover the contributions of merit in shoewear technology. This section attempts to do just that. A foundation of relatively stable information is provided to guide the reader through this subject despite the constant changes fueled by fashion trends and advertising gimmickry as well as scientific research. To understand where we are and where we are headed, some historical perspective is necessary.

A more ancient type of shoe was a hide shoe made by folding the skin or hide of a beast around the foot. This is the forerunner of what we call a moccasin, a term derived from the Algonquin Indians and introduced into English literature in 1612 by John Smith’s “Map of Virginia.”10 Examples of this form of shoe come from excavations in Denmark of early Bronze Age oak-log coffins dating to about 1000 bc.10 Although earlier examples do not appear to have been preserved, one can assume that the early hunters of the Stone and Ice Ages must have been capable of seeing the advantages of covering the foot for protection. What could be more logical than using the hide of their prey to provide a suitable foot covering? Rock carvings have provided evidence that these hide shoes were secured to the foot by lashing them around the instep and arch.10 Cave paintings found in Spain dating to 15,000 bc depict boots made of animal skin and fur.11 More recent descriptions of shoemaking from animal hides are provided in the works of Xenophon, Niebuhr, and

HISTORY The history of sports shoes parallels the history of shoewear itself. According to legend, shoes were originally designed after an Arab chief dismounted from his camel onto a thorn and declared that all the earth would be covered with leather. Seeing the error in this logic, the chief’s main advisor decided to make something that would cover just the feet. Although this makes a good story, it has not been supported by the discovery of shoes in the Fertile Crescent.8 Indeed, the earliest footwear was discovered in south central Oregon in 1932 by anthropologist Luther Cressman—a sandal made from sagebrush bark (Fig. 25B-2).8,9 This find dates back 10,000 years to pre-Columbian times, but design features indicate a much earlier origin. It supports the notion that the shoe’s primary function is to protect the sole from the hazards of the environment.

Figure 25B-1   Sports Illustrated cover indicating the notoriety of sports shoes. (Illustration by Julian Allen. From Sports Illustrated, vol 72, May 14, 1990.)

1874 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25B-2   Earliest existing footwear, dating back some 10,000 years. It is made from sagebrush bark and was found in south central Oregon by anthropologist Luther Cressman in 1932. (Photograph by Steve Bonini. Courtesy of University of Oregon Museum of Natural History and State Museum of Anthropology.)

­ inkerton.12-14 Examples of the bear paw used as a shoe P are seen in Figure 25B-3 from the Musée de l’Homme in Paris.10 This bear paw with claws attached could be considered the first shoe with cleats. A later development in shoewear and the second broad type of shoe is made of two components, an upper and a sole.10 These are joined together at the lower edge of the foot. The appearance of this shoe occurs in Roman times, when a hide shoe reinforced by an extra piece of sole material was used.11 Furthermore, during this time, the insole appears as a layer added for comfort and protection against chafing. These design features were a product of necessity owing to the abuse to which soldiers’ feet were subjected. Similarly, modern design features were generated to protect the athletes’ feet, particularly those involved in distance running. Because early humans were largely dependent on hunting, one can postulate that the earliest footwear was used in running. With civilization’s advancement and socialization, shoes took on symbolic functions.11 Papyrus sandals for religious ceremonies and jeweled sandals for high-fashion gatherings have been discovered in the burial holdings of Egyptian pharaohs.15 Although these have little to do with sports shoes, they do foreshadow the current specialization, trendy colors, and designs incorporated into athletic shoewear construction. Competitors in the early Greek games competed barefoot according to early drawings found on vases of that period (Fig. 25B-4).9 Inasmuch as shoemaking was a well-developed trade by this time, it appears that early athletes eschewed comfort for the presumed benefits of barefoot performance, that is, less weight, better feel for the surface, and improved traction. Robbins and Waked revived interest in barefoot running with their hypothesis that the excessive cushioning found in modern shoewear prevents appropriate sensory feedback and results in a “pseudoneurotrophic” effect.16

Figure 25B-3   Example of the bear paw shoe. (Courtesy of the Musee de l’Homme in Paris.)

Figure 25B-4   Competitors in the early Greek games competed barefoot, according to early drawings found on vases from that period. (Courtesy of the Metropolitan Museum of Art, Rogers Fund, 1914.)

Foot and Ankle 1875

The sensibility of the plantar foot is a key reason that gymnasts and dancers perform with bare or minimally shod feet. While an individual is running, plantar tactile reflexes and the intrinsic shock absorption system of the body complement one another and result in behavior modification to control load magnitude. Specifically, humans dramatically reduce impact force by altering knee and hip flexion at ground contact.16 A series of studies by Robbins and coworkers have proposed that cushioned shoes lead to negligible decreases in load because subjects decrease flexion to accommodate the instability produced by softer surfaces.15,16 A recent study demonstrated that subjects presented with a “deceptive” advertisement of the ability of a surface to cushion impact led individuals to increase the ground reaction force of a barefoot footfall when compared with a “warning” and “neutral” message. This was despite the fact that the surface was covered with an identical thickness of ethyl-vinyl acetate surfacing material.17 Although the notion that “deceptive” advertising can lead to potentially harmful behavior associated with shoewear is intriguing in light of the enormous sums spent on advertisements by shoe companies, noted authorities on running and running shoes have not been impressed with this theory of the importance of sensory feedback.18 They point out that biomechanical abnormalities such as excessive pronation, excessive Q angle at the knee, forefoot varus, and so on are the primary causes of running injuries, not a lack of sensory feedback. Furthermore, it is only with shoewear adaptations that these abnormalities can be corrected, according to these experts.18 Ironically, one of the editors of Runner’s World magazine recorded his shoewear experience over a 20-year period in a shoe diary and noted that a 5-year period of barefoot running was his healthiest period.9 There is even a Web site now that promotes the benefits of shoeless running: www.runningbarefoot.com.19 Although it is clear that Western-style shoes have contributed to many of the foot ills of modern society such as bunions, corns, calluses, and neuromas,20,21 there is circumstantial evidence to suggest that improvements in running shoe construction have reduced the prevalence of Achilles tendinitis and allowed greater numbers of average citizens to participate in the sport of distance running.9 Tracing the history of the running shoe is an enlightening look at the shoe industry itself, at the role of sports in society, and at the international trade competition surrounding sport and its premier athletes. The most thorough sources of information in this area are The Running Shoe Book, written in 1980 by Peter Cavanagh, and The Complete Book of Athletic Footwear, written by Melvyn Cheskin in 1987.9,22 Both trace the evolution of running shoes through the footraces of 16th-century fairs and the pedestrian races of the later 1800s to modern-day track and field competition. Important landmarks in this history can be picked out along the way. The turnshoe construction technique was firmly established by the 12th century. It allowed a shoe to be made with the seams on the outside and the smooth material inside next to the foot. The shoe was turned inside out to produce the finished product.9,22 By the 14th century, shoe construction had incorporated small strips of leather called welts to allow a replaceable outsole to be added to the upper.23 Since the 14th century,

shoemaking has been fairly standardized, with shoes consisting of the following: 1. Two parts: upper and lower 2. Four processes: cutting, fitting, lasting, and bottoming 3. Eight tools: knife, awl, needle, pinchers, last, hammer, lapstone, and stirrup24 With the Industrial Revolution, the craft of shoemaking went from an in-home trade to a model of manufacturing method with the use of machines and mass production. Entire books have discussed the significance of this method to the development of industry in the United States.25,26 Leather was the mainstay of shoemaking throughout this period and continued as such into the 1900s. A change in shoemaking and the origin of the sneaker were presaged by the first patent, granted in 1832, for attaching rubber to the sole of the shoe. Unfortunately, the material was too unstable and lacked durability.11 In 1839, Charles Goodyear’s vulcanization process turned rubber into a usable material, but it was more than 100 years before rubber replaced leather as the most desirable outsole material for running shoes.9 Cavanagh cites the development of the Spencer shoe, a spiked shoe found in England’s Northampton Museum, as the 1865 precursor of modern track shoes (Fig. 25B-5).9 This shoe shows the separation of running shoes into a line separate from street shoes, although a spiked shoe used for cricket was patented in England in 1861. Spiked shoes were used in the short races popular in that day. Longer distances became popular in the latter part of the 19th century. Races around circular tracks for 144 straight hours became a spectator event imbued with international flavor. These pedestrians, as the participants were called, wore high-top leather boots and thick wool socks reminiscent of combat boots used in the military. Although pedestrian races faded in popularity, long-distance racing gained an audience,

Figure 25B-5   The Spencer shoe, a spiked shoe found in England’s Northampton Museum, was the 1865 precursor to modern track shoes. (Courtesy of the Northampton Museum, Northampton, England.)

1876 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

and the Olympics were reborn in Athens, Greece, in 1896. A marathon was included as a race of 40 km to commemorate the legend of Pheidippides. Cavanagh marks this Olympic race as the impetus for development of the distance shoe, or training shoe, we use today.9 The popularity of the marathon prompted the Spalding Company to introduce a long-distance shoe for the general public in 1909.9 Three shoes were advertised, two being high-tops (Fig. 25B-6). They had leather uppers and rubber soles and were priced at $5 to $8. A retired shoemaker named Richings began custom making a shoe for distance runners near the 1930s that predated the custom shoewear used by elite athletes of our day.9 By the early 1900s, production of running shoes was in full swing, and the 1915 Spalding catalog advertised shoes for sprinting, middle-distance running, jumping, and pole-vaulting (Fig. 25B-7).22 Competition entered the scene at about the same time when Sears Roebuck entered the catalog shoe sales market and began the continuing controversy over who makes the best running shoe.22 Endorsements by famous athletes were seen much earlier, but notable shoes were the Kiki Cuyler, Jr., basketball shoe and the Chuck Taylor AllStar shoe. We continue to see society’s ongoing enchantment with famous athletes and their shoewear, as evidenced by the incredible popularity of the “Air Jordan” and “Bo Knows” campaigns of the Nike shoe company in the late

1980s and early 1990s. Despite recent cuts in advertising budgets for most of the major shoe manufacturers,27 the fact that basketball shoe sales alone are responsible for as much as 25% of total athletic shoe revenue28 ensures that certain National Basketball Association stars will expand their talent into marketing. A shoe designed for sports alone did not come into existence until the latter half of the 19th century. Croquet was a popular recreation during the Victorian period, and a croquet sandal appeared during this time.11 Known as the sneaker, it was in use by the 1860s and had a fabric upper, a rubber sole, and laces.9,11,29 Further sports development in the late 1800s spawned the need for durable but lightweight shoes with variable traction requirements depending on the playing surface. Wilcox provides several illustrations of these specialized sports shoes of the late 1800s in his book, The Mode in Footwear (Fig. 25B-8).29 From these developments, we can trace the roots of the multibillion-dollar sports shoe industry and can conclude that the protection of our feet and fashionable design have always been important concerns of humankind. From this foundation, an explosion occurred in sports-specific footwear that has provided us with today’s shoes for basketball, rock climbing, tennis, snowboarding, soccer, gymnastics, fishing, rollerblading, skating, jumping, sprinting, and so forth (Fig. 25B-9).

Figure 25B-6   Long-distance shoes introduced by the Spalding Company for the general public in 1909. (From Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987.)

Figure 25B-7   Page from the 1915 Spalding catalog advertising shoes for sprinting, middle-distance running, jumping, and pole-vaulting. (From Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987.)

Foot and Ankle 1877

ANATOMY OF THE SPORTS SHOE Just as the anatomy of the human body is the basis on which the surgeon’s skill rests, the anatomy of the sports shoe is critical to those who must understand athletes and their injuries. Although the process of shoe manufacturing has evolved into a multibillion-dollar industry, the basic shoe remains the same. This section first looks at the anatomy of the basic sports shoe and then discusses the shoe features unique to particular sports. The actual ­ manufacturing ­process is discussed briefly.

Most of what has been written about athletic footwear has concentrated on shoes designed for the runner. Therefore, the prototype shoe for this section is the running shoe, and shoes for other sports are described in similar terms with specific modifications. Figure 25B-10 illustrates the components of the shoe. For the sake of simplicity, the shoe can be broken down into two basic components: the upper and the bottom. The upper covers the foot, whereas the bottom cushions it and provides the interface between the foot and the surface. These two basic components are then subdivided into

Man’s summer sport shoe, Balmoral style, white canvas with leather 1879

Polo ankle boot (today called Chukkar or Jodhpur) 1850s

Gentleman’s riding boot 1850s Man’s buckled hunting boot, gaiter style, leather and cloth, protruding rubber insertion in heel 1850s

Man’s gymnastic shoe, eyelets halfway, hooks to top, calf or canvas 1890s

Man’s hunting shoe in Blucher style with toebox and tongue of heavy calf 1885

Sports shoe worn hunting, striped fabric and leather 1850s

Football player’s shoe, veal calf and leather thongs 1912

Cyclist’s boot of calf or canvas and leather 1910

Figure 25B-8   Illustrations of specialized sports shoes of the late 1800s. (Adapted from Wilcox RT: The Mode in Footwear. New York, Charles Scribner and Sons, 1948.)

1878 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25B-9   Photograph from typical shoe store with wall of sport-specific shoes in all varieties. (Photograph by Andrew Borom.)

their various parts. The upper is composed of toe box, toe cap, vamp, quarter, saddle or arch bandage, eyelet stay, eyelets, throat, tongue, collar, Achilles tendon protector, heel counter, foxing, forefoot and rearfoot stabilizers, and lining. The bottom consists of sockliner or insole, insole board, midsole, wedge, and outer sole. Although some of these names differ from those used in traditional shoemaking, the actual construction of a sports shoe does not vary remarkably from the traditional shoe manufacturing ­process (see Glossary in Box 25B-1).

To understand the shoe itself, it is necessary to review the steps by which the shoe is made. Integral to this ­process is the last (from the Old English laesk, meaning sole or footprint), which acts as an artificial foot form.22,30 This allows the shoe upper to be created in the proper shape, size, and dimensions. Important measurements with respect to the last are the toe pitch or toe spring, the girth, and the heel height or pitch (Fig. 25B-11).22,31 The variation in toe spring and heel height can affect the movement of the foot by improving or impairing the rocker action of the foot during gait.31 Shape is also a critical consideration in analyzing the form for the last of a shoe. This shape can be either straight-lasted or curve-lasted depending on the amount of inward curve built into the last. Because most feet have a slight inward curve, the curved type of last provides better comfort and fit for most feet. The curved last allows the most shoe flexibility and is particularly well suited to the athlete with a highly arched or cavus foot. When the last is straighter, it translates into better medial support for the foot and is best suited to the flatter foot or to the person with an overpronated foot. Figure 25B-12 depicts the difference between straight and curved lasts. Sports shoe manufacturers use a curve of about 7 degrees in a curved last.22 Variations exist now as slightly curved and semicurved lasts. As we examine the last, the next consideration is the shape of the toe box. The various alternatives are depicted in Figure 25B-13.22 It is clear that this feature has an important bearing on fit and comfort. The best example of this is the need for the athlete with clawing in the toes or even a single hammer toe to have a toe box of sufficient height to prevent chafing. Text continued on p 1886.

Tongue Achilles tendon Collar protector Lining Throat Eyelet Eyelet stay Vamp Toebox

Heel counter (under foxing) Upper

Foxing Sock liner or insole Rearfoot stabilizer Midsole with wedge

Bottom

Outsole Rearfoot stabilizer

A

Quarter Stabilizing straps

Toecap

B Lining Sock liner or insole

Heel counter

C

Wedge

Outsole

Midsole

Insole board

Figure 25B-10   Illustrations of athletic shoes. A, Overview of external appearance. B, Separation of shoe into component parts.   C, Sectional view of interior of shoe.

Foot and Ankle 1879

Box 25B-1 Glossary of Foot and Shoe Terminology* Abduction: To move away from the midline of the body. Achilles notch: A depression cut into the back of the heel collar to provide a secure fit and prevent irritation of the Achilles tendon. Adduct: To move toward the midline of the body. Adduction: Moving a part toward the midline of the body. Adhesive (cement): Substance capable of holding materials together by surface attachment. Air: First introduced in 1979, Nike’s cushioning concept of encapsulated air units in the midsole isn’t actually air, it’s Freon. Depending on the model, the air units may be in the heel or forefoot, or both. Air ball: An air-pressurized ball imbedded in the heel of HiTec Badwater models for additional shock absorption. Anatomical: Pertaining to the structure of the body. Anatomical last: A stabilizing footbed contoured in such a way that the heel sits down in the midsole, rather than resting atop a flat platform. Developed and used extensively by Turntec. Anatomy: The study of the structure of the body and the relationships between its parts. Anterior: Front portion. Anterior heel: Type of metatarsal bar, also known as Denver bar or Denver heel; the apex coincides with the posterior edge under the posterior half of the metatarsal shafts. ARC: Avia’s stabilizing system, which is made of plastic in one of two configurations. Placed in the rearfoot, the “fingers” of the ARC spread out on impact to absorb shock and stabilize the foot. Arch bandage: Reinforcing strips of fabric stitched inside the shoe on the medial and lateral quarters. Arch cushion (cookie): Support pad for the medial arch of the foot. Asymmetric: In shoemaking this applies to lasts and patterns that have uneven shapes, the right side different from the left. ATP (heel horn): Extended padding at the back heel ­collar to protect the Achilles tendon (Achilles Tendon ­Protector). Autoclave: Vessel or oven in which chemical reaction or cooking takes place under pressure such as in the vulcanizing construction method. Axis: A reference line for making measurements. Ground reaction forces are usually evaluated relative to a set of three orthogonal axes: vertical, longitudinal (direction of motion), and transverse (right angle to direction of motion). Backpart (rear foot): Portion of the last extending rearward from the break of the joint to the back of the last. Backpart width: The width of the heel end measured parallel to the heel featherline plane at a specified distance from the heel point. Bal (Balmoral): Front-laced shoe in which the meeting of the quarters and the vamp is stitched or continuous at the distal end of the throat. Bal is the abbreviation of Balmoral, the Scottish castle where this style was first introduced. Ball: Widest part of the sole, at the metatarsal head. Ball girth: Circumference measure around the last encompassing the first to the fifth metatarsal area.

Bar, comma: Comma-shaped bar wedged laterally and posteriorly, also known as Hauser. Bar, Denver: See Anterior heel. Bar, Jones: Metatarsal bar placed between the inner sole and outer sole. Bar, Mayo: Metatarsal bar with the anterior edge curved to approximate the position of the metatarsal heads. Bar, metatarsal: Rubber, leather, or synthetic bar applied transversely across the bottom of the sole, with the apex immediately posterior to the metatarsal heads. Bar, rocker: Sole bar having its apex beneath the metatarsal shafts causing rocking instead of flexing action. Bar, Thomas: Narrow metatarsal bar with abrupt anterior and posterior drop-offs. Bar, transverse: See Bar, metatarsal. Base plane: The plane to which the last in its proper attitude is referenced for the purpose of defining certain terms. Bias cut: Cut away upswept heel. Bilateral: Affecting both right and left sides. Biomechanics: The study of the internal and external ­forces acting on the human body and the effects produced by these forces. Blind eyelet: A metal or plastic eyelet concealed beneath the top surface of the shoe leaving only a small, rimless hole. Blown rubber: The lightest kind of rubber outsole material. As the outsoles are manufactured, air is injected into the rubber to lighten and soften the outsole. Few outsoles today are made with full blown rubber because it lacks durability, but many outsoles have blown rubber in the forefoot and midfoot for lightness and a harder carbon rubber in the high-wear area of the heel. Blucher: Front-laced shoe in which the quarters are not attached distally to the vamp, giving more allowance at the throat and instep in fitting. Opposite of bal style. Front quarters or tabs are stitched over the vamp for a short distance at the throat. Board last: One of three ways shoes are constructed. A fully board-lasted shoe is constructed by gluing the upper to fiberboard before it is attached to the midsole. Board lasting promotes stability and provides a good platform for orthotics but lacks flexibility. Few new models are fully board lasted. (See combination last and slip last for information about the more common types of last.) Boot, high top: High quarter shoe in which the quarters cover the malleoli. Bottom: The sole up to the breast of the heel. On a wedge sole, the term covers the complete sole. Bottom filler: Material that fills the cavity between the outer and inner soles. Bottoming: The operation of attaching the completed sole to the upper. Bottoming out: When the midsole material has worn out and is too soft relative to a runner’s size, it compresses too quickly, which results in compromised shock absorption and support. Box toe: Hardener used to maintain shape of front toe area. Continued

1880 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Break: Flex point or path; creasing formed at the vamp when the shoe is dorsiflexed. Breastline: An arbitrary line defining the forward boundary of the heel seat. Breathability: The ability of a material to absorb and ventilate foot moisture; not to be confused with porous. Bumper: Rubber toe strip attached over front toe area. Calfskin leather: Leather made from the skin of calves. Cantilever: A concave outsole design in which the outer edges flare out on impact to dissipate shock. Used extensively by Avia. Carbon rubber: The most durable kind of rubber outsole material. It’s a solid rubber with a carbon additive that makes the material stronger. If an outsole is not full carbon rubber, it probably has a carbon-rubber heel pad. Celluloid: A thermoplastic material. Cellulose: Natural polymeric. Cement process: Construction stuck-on bottoming method. Center of pressure: An outsole design in which the middle area is bordered by an elevated tread pattern along the outer rim to promote stable footing. This also makes for a lighter shoe by exposing the midsole and eliminating unnecessary outsole material in the center. Used in some Nike, Diadora, New Balance, Hi-Tec, Avia, and Adidas models. Certified pedorthist: One who is certified by the Board for Certification in Pedorthics (BCP). Chainstitch: Sewing method used for stitching uppers to soles. Chukka: Three-quarter Blucher boot with two or three eyelets or a strap with a buckle. Circular vamp: Design vamp extended from toe to heel breast. Coefficient of friction: A number between 0 and 1 indicating the slip resistance of a material such as a shoe sole on a particular surface. The greater the value, the less likely any slipping. Collar: Top line of the shoe quarters. Many are padded. Narrow strip of material stitched around the proximal edge of the quarter. Combination last: Last with wider forepart and narrow heel fitting. Indicates the shoe is board lasted in the rearfoot for stability, but slip lasted in the forefoot to promote flexibility. Remove the sock liner of your shoe; if it is combination lasted, you will find a fiberboard in the rearfoot and stitching in the front. The last deviates from standard proportions, usually to accommodate feet with narrow heels. Compaction: Permanent flattening and deformation of sole material (bottoming out). Composition: Scraps that are pulverized, compressed, and held with a binder to form a sheet material for insoles, midsoles, heel bases, and other components. Compression deflection: The amount of deformation observed in a material after it has been subjected to a compressive or impact load. Compression mold: Shaping materials by heat and ­pressure.

Compression set: The amount of permanent deformation observed in an unloaded material following a single or multiple load application. Conformability: Ability of a material to mold itself to the shape of the foot. Contact face: The surface brought into contact with another surface or object. Contoured midsoles: Similar to anatomic lasts, contoured midsoles are shaped to the foot. This promotes stability. Contour insoles: Foam insoles capable of retaining pressure pattern of the foot. Cookie: Arch pad in shoe; wafer-shaped longitudinal arch support. Copolymer: A natural or synthetic compound. Cordovan: Leather not made from hide or skin but from a ligament-like shell found under the buttocks of animals such as the horse, mule, and zebra. Often referred to as a “shell” and tanned in a burgundy (cordovan) color. Cork: Made from the bark of the cork tree, which may be combined with other materials. Available in various forms such as sheet cork, natural cork, cushion cork. Different names are given to the cork according to the binders used. Corrective shoe: Shoe with special features designed to help correct some type of foot disorder. The term has been displaced by other terms such as a prescription shoe or a modified shoe. Corset, ankle: Reinforcement to preserve the shape of the quarter’s counter. Laced fabric within the shoe intended to retain the hindfoot on the inner sole. Counter, long lateral: Counter extended anteriorly ­beyond the breastline on the lateral side. Counter, long medial: Counter extended anteriorly ­beyond the breastline on the medial side. Counter (pocket): See Heel counter. Crepe rubber: Natural rubber soling material; latex ­rubber compounded for use as soles and heels. Crest, toe: Convex cushion under the plantar phalangealsulcus. Curved last: Refers to the shape of a shoe. Curve-lasted shoes are shaped somewhat like a banana and offer less medial (inner) support but greater foot mobility. Generally, curved-lasted shoes are for biomechanically efficient, faster runners who want a responsive shoe. Cushioning: The ability to absorb shock. Because a runner generates a force of about 3 times the body weight on impact, this is a crucial shoe characteristic. Cushioning is primarily a function of the midsole. Custom shoes: Shoes made to the customer’s specifications. Dellinger web: Embedded into the midsole of some Adidas models, the Dellinger web is a fabric that maintains the durability of the midsole. Designed by University of Oregon coach Bill Dellinger. Denier: Weight of synthetic fibers (measure of fineness). Density: Weight per unit volume of a substance. The measure of the firmness of the midsole material. Many shoes have midsoles of varying densities. For example, a two-density EVA midsole will usually have the firmest material (designated by a darker color) on the medial side to control pronation.

Foot and Ankle 1881

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Derby: Design quarters overlapping vamp and tongue. Design: Pattern or cut of upper. Die cutting: Cutting of upper or sole materials with metal dies. Differential loading: The application of forces of varying magnitudes. Dip construction: Direct injection process. Distal: Part farthest from the central portion of the body. Dorsal: Top of foot, the upper surface of the foot. Dorsiflexion: Moving the toes up toward the distal end of the foot toward the leg. Doubler: Interlining placed between the vamp and vamp lining for additional reinforcement; interfacing between upper material and lining. Drop-off: Anterior vertical edge of a metatarsal. Duo process: Method of upper assembly construction by cementing instead of stitching edge. Durometer scale: A method of determining material hardness on a scale of 0 to 100, with lower readings indicating softness. Dutchman: Lateral sole wedge. DVP: Direct vulcanizing process. Elasticized material: Resilient fabric used for goring in panels and inserts for shoe uppers. Elastomer: Term used for synthetic rubber. Electrodynograph: An instrumentation system consisting of individual sensors to measure pressure at selected locations on the bottom of the foot. Electromyography: The measurement of the electrical activity associated with muscular contractions. Elevation: Material added to the entire sole or heel. Elvalite: A new foam developed by DuPont that’s being used as a midsole material in some Reebok models. ­Elvalite has the cushioned feel of EVA but is more durable. Elvaloy: Resin modifier added to PVC. Embossing: Depressing a specific pattern in leather or ­fabrics. EVA: Ethylene vinyl acetate (EVA) is the most common midsole foam used in running shoes. Compressionmolded EVA is heated and compressed into the shape of the midsole. It is light, resilient, and has good cushioning properties. Nearly every running-shoe company uses EVA in at least some of its midsoles. Eversion: Turning out the plantar aspect of the foot from the midline of the body. Evert: To turn out the plantar aspect of the foot so that it faces away from the midline of the body. Exercise physiology: The study of the effects of exercise on the biochemical function of the body and its parts. Expanded vinyl: Soft, nonbreathable PVC (stretchy base) material (as opposed to nonexpanded vinyl: harder, nonbreathable [rigid base] material). Extended eyestay: A design wherein the eyestay is ­extended to form the toe cap. External: Outer part; lateral. External heel counter: A rigid, plastic collar that wraps around the heel of the shoe for support and to control pronation.

Eyelets: Holes for lacing (blind) with metal reinforcements or eyelet hooks. Eyestay: Reinforcement around lacing holes. Fabric: Woven or nonwoven flexible material. Feather edge: Last bottom profile. Finish: Coating on leather or synthetic material. Finishing: End of manufacturing process. Flanging: The edge where the upper is turned outside for attachment to outsole or midsole. Flare: Widened heel or sole base. Flared heel: Wider flanged heel for landing. Flex: To bend. Flex grooves: Strategically placed ridges in the midsole of the forefoot that make the shoe more flexible at toe-off. Flexibility: A shoe’s ability to bend; the rigidity of the shoe bottom composite usually evaluated in the forefoot region of the shoe. Flex path (break): Girth area at the main metatarsal of foot, which must flex as foot pushes off from ground. Flexion energy: The energy required to bend a shoe or object through a flexion cycle. Flow molding: The construction method of molding PVCcoated materials as an exact replica of original ­uppers. Footbridge: A stability device. As used by Nike in the Air Structure and Air Span II, the footbridge is molded into the midsole across the rearfoot. Reebok uses a footbridge in its Ventilator, but it’s placed under the arch on the medial side. Asics uses a variation of the footbridge in the Gel-MC. Footframe: An extension on top of the midsole or an additional piece that cradles the foot for added support and prevents the foot from rolling over. Force: A pushing or pulling effect that produces motion or deformation of an object or material. Forefoot stability strap: A leather or plastic overlay on both sides of the ball area of the shoe that reinforces the upper and offers stability and support. Forepart: Area of foot from the ball to the toe; portion of the last extending from the ball to the toe. Forepart centerline: The best line of symmetry of the forepart bottom pattern. Frontal plane: The vertical plane that passes through the body dividing into a front and a back half. Functional anatomy: The study of the effects of body structure on performance. Functional shoe: Shoe designed to serve a specific ­purpose. Gait laboratory: A testing lab equipped with specialized equipment for the study of walking and running. Gel: The primary cushioning system used by Asics in all of its performance shoes. It is a pad of silicone in the midsole, which, depending on the model, is found in the heel or the forefoot, or both. Geometric last: Last using a geometric rather than the traditional arithmetic grading system. Intended for better shoe fit and to facilitate automated manufacturing. Girth: Circumferential dimension measured around the last; widest part of the last. Continued

1882 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Goniometer: An instrument used to measure angles. It can be employed to evaluate motions of the joints, particularly the knee and ankle joints. Goodyear welt: Construction method of stitching uppers to sole; shoemaking process in which the joining of the upper, inner sole, and outer sole is accomplished with a welt. Goring: Elastic fabric inserted in the front or sides of an upper, the expansion of which allows a larger opening to insert the foot. Grade increment: The change per size and shoe width of any last dimension. Grade rate: The ratio of the change in girth per size to the change in length per size. Grading: Method used by designers to size original ­patterns. Heel breast: Anterior margin of heel; front face of the heel or anterior portion at the shank. Heel counter: A firm, usually plastic cup that is encased in the upper and surrounds the heel. It controls excessive rearfoot motion. It may be notched to accommodate the Achilles tendon. Heel counter pocket: Rearpart upper material pocket containing heel stiffening material. Heel curve: A side-view profile of the back end of the last from the top of the last to the heel seat or featherline. Heel curve angle: Angle between the heel featherline plane and heel point 2½ inches (63 mm) up from the heel point intersecting the heel curve. Heel elevation: Modification measured in a vertical line at the center of the heel; the vertical distance between the base plane and the heel point is the heel elevation. Heel featherline: A line that defines the heel seat shape. Heel height: Vertical measurement from the plantar surface to the heel seat at the anterior surface of the heel, usually in increments of eighths of an inch. 1. Spring heel, 3⁄8 to 6⁄7 inch; heel base lies under the outer sole eliminating a definite heel breast. 2. Flat heel, 6⁄8 to 10⁄8 inches. 3. Military heel, 10⁄8 to 13⁄8 inches. 4. Cuban heel, 13⁄8 to 14⁄8 inches. 5. Wedge heel, 4⁄8 to 14⁄8 inches; slopes upward from ball to posterior heel. Heel pad: Resilient material to cushion or raise the heel. Heel pitch: Slant at the posterior aspect of the heel; amount of rise at back of last when last is held level. Heel plug: Found in multidensity outsoles, where the most durable rubber is placed in the high-wear area of the heel. Heel point: The rearmost point of the heel featherline. Heel seat: Area of the shoe upon which the foot rests; the bottom surface of the heel end of the last from the breast line back. Heel seat width: The greatest width of the heel seat measured from featherline to featherline perpendicular to the heel centerline. Heel, Thomas: Heel with anteriorly curved medial border. High cut: Over the ankle shoe design.

Horizontal plane: See Transverse plane. Hytrel: A resilient and durable polymer plastic developed by DuPont and used for a variety of shoe components by several companies. It’s most commonly used as forefoot support straps. IMP (injection): Injection molding process form of shoe construction (see also Lasting insole). Inflare: Asymmetric inward swing of last shape; last or shoe whose forepart provides more medial than lateral surface area. Injection molded: Shoe construction whereby a heat­softened plastic is injected into a mold, then compressed against the mating surface of a concentric mold and ­allowed to cool and harden. Inlay: Prefabricated removable material upon which the foot directly rests inside the shoe. In some shoes, the inlay is an integral design component. Inner sole: Material conforming to the size and shape of the last bottom upon which the foot rests (see also Insole). Insert: A type of orthosis, although the term has been used interchangeably in some circles with inlays and insoles to designate an off-the-shelf device placed inside the shoe. The Health Care Financing Agency defines it as a total contact, multiple-density, removable inlay that is directly molded to the patient’s foot or a model of the patient’s foot and that is made of a suitable material with regard to the patient’s condition. Inserts: Metal threaded retainers for spikes or studs. Insole: Integral design component (layer) of the shoe that is the shoe’s structural anchor to which is attached the upper, toe box, heel counter, linings, and/or welting. Instep: Medial inside arch area of the shoe. Instep: Portion of the upper over the midfoot. Instep girth: The dimension around a last passing through the instep point. Interface: The surface forming the common boundary between two bodies or spaces. Internal: Inner part; medial. Inversion: Turning the plantar aspect of the foot in toward the midline of the body. Invert: To turn in the plantar aspect of the foot so that it faces the body’s midline. Ionic cushioning: Pillars of polyurethane in the midsole used by Saucony to add durability. Ionomer resins: A family of thermoplastic resins. Iron: Dimension used for measuring sole thickness, 1⁄48-inch; thus a 6-iron sole is 1⁄8-inch thick. Isoprene: Fundamental rubber molecule. Joint girth: The greatest dimension around the last passing through the break joint. Kinematics: The science of pure or abstract motion. Kinetic energy: Energy associated with motion. Lace locks: Plastic devices on the upper that maintain tension on the laces. Lace stay: Portion of the upper containing eyelets for lacing. Lace-to-toe: Low- or high-quarter shoe laced to the toe; a design in which the eyestay is extended down to the toe box area.

Foot and Ankle 1883

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Last: Three-dimensional facsimile of the foot; model approximating the shape and size of the weight-bearing foot, made of wood or plastic, over which a shoe is formed. The shape of the last determines the shape of the shoe. The straighter the last, the greater the medial support. Generally, faster, lighter runners who need less support prefer curved-lasted (or semicurved) shoes. Runners who need maximal medial support and those who overpronate opt for straight or slightly curved shoes. Last bottom centerline: A line defined by the toe and heel point. Last bottom featherline: A line that defines the bottom shape of the last (last bottom pattern). Last bottom width: The width across the ball area of the last bottom at its widest point. Lasting: Fitting and shaping of the upper to the last. Lasting allowance: Extra material on shoe patterns to fit around and under the bottom edge of the last. Lasting insole: An insole used to attach an upper to an insole before bottoming; the bottom surface of the ­upper. Lasting margin: See Lasting allowance. Last joint break: Point located at the intersection of the shank and the forepart, tangent to heel point and perpendicular to last centerline. Last systems: Methods of sizing last dimensions: Arithmetic, Geometric, Dynametic, Europoint. Lateral: Outer side of the foot or limb; the side away from the midline of the body. Leather: Material created by tanning a hide or skin. Length: Dimension on the center of the last bottom from toe point to heel point. Levy mold: Full-length inlay that conforms to contour of the plantar foot. Lightweight trainer: A training shoe that weighs less than 10 ounces. It can be used for training or racing, but it’s not as durable or supportive as most training shoes. Lining: The inside backing material for uppers. Lockstitch: A method of sewing the upper to the bottom. Long heel girth: The dimension around a last passing through the instep and heel featherline point. Long heel plate: A sheet metal bottom surface extending from the heel to midway of the shank area. Longitudinal arch: Curvature of hind and midfoot. Longitudinal force: The force generated in the direction of motion by a walker or runner during foot contact and related to the slip characteristics of a shoe. Also referred to as the anteroposterior force. Low cut: Below the ankle shoe design. McKay: A shoe construction method that uses tacks and a stitched sole; the upper is tacked, stapled, or cemented, and the sole is attached with chainstitches. MCR: Microcellular rubber. Medial: The side closest to the midline of the body; inside area of the foot. Memory: The speed and extent to which a material recovers to its original shape after load compression. Mesh: Woven or knitted nylon material for uppers.

Metatarsal pad: A soft wedge of material placed under the ball of the foot to add shock absorption and comfort for forefoot strikers. Metatarsals: The long bones of the foot between the toes and ankle. Midsole: The layer of material between the upper and outsole. It’s the most important component of the shoe because it provides most of the cushioning. The midsole is usually made of EVA or polyurethane or some combination of the two. Moccasin: A method of construction whereby the upper is placed under the last and extended up and around to form the quarter and vamp. Modification: Alteration, change, or addition. Mold, mould: That which is shaped, molded, or formed; a cavity used to shape plastic or rubber by pressure and heat. Molded shoe: Shoe made from a model of the foot. Monk strap: Shoe with a wide buckled strap across the ­instep. Motion analysis: The analysis of total or partial body movements for the purpose of better understanding how the body functions. The analysis is usually done in conjunction with high-speed filming and computers. Motion control devices: Materials and designs that control the inward rolling (overpronation) of the foot. Nap: The surface pile or layer of textile fabric. Negative heel: Heel with plantar surface lower than the ball of the shoe. Neoprene: Synthetic, rubber-like material, very ­ durable, used for outsoles, heels, and other components; oil­resistant; an elastomer, polychloroprene. Neutral position: The most efficient functional position for the foot producing the least amount of stress on the joints, ligaments and tendons. Open toe: Shoe design with no front center seam. Orthopaedic shoe: Shoe designed with features to accommodate or reposition foot abnormalities. Orthosis: Corrective device that is used to protect, support, or improve function of parts of the body that move. Outflare: Last or shoe whose forepart provides more lateral than medial surface area. Outsole: Bottom, ground-contacting portion of the shoe; the black material on the bottom of the shoe that strikes the ground. Carbon rubber is the most common outsole material because it is firm and resilient. Blown rubber has a more cushioned feel but is less durable. Oxford: A shoe design with a laced, low-cut shoe; low ­quarter-laced shoe. Pad: A device placed inside a shoe to provide support or relieve pressure from a specific location such as a longitudinal arch pad or a metatarsal pad. They are made of various materials and come in a variety of shapes and sizes. Pattern: The cut-out pieces making up the design of the upper. Pedorthics: Allied health profession concerned with the design, manufacture, fit, and modification of footwear and related appliances. Continued

1884 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Pedorthist: Practitioner of pedorthics. Phylon: A foam similar to EVA and the name of the midsole material that Nike uses in several models. Hi-Tec also uses Phylon in one model. Pivot point: The rotation area on sole under ball of foot. Plantar: The bottom or sole of the foot. Plantar flexion: The downward movement of the toes or distal end of the foot away from the leg. Plastic: Synthetic material or human-made polymeric substance, excluding rubber. Platform: Elevated sole. Polyethylene: A thermoplastic material or ethylene. Polymer: A molecular compound, natural or synthetic. Polypropylene: A tough lightweight plastic. Polystyrene: A transparent thermoplastic. Polyurethane: A synthetic rubber that’s a common midsole material. It is firmer, heavier, and more durable than EVA, but it’s not as cushioned. Polyurethane is often used with EVA in many popular models, such as the Nike Air Pegasus, Avia 2200, and New Balance 997, with the polyurethane in the rearfoot for firmness and durability and the EVA in the forefoot for ­flexibility. Polyurethane resins: A family of resins from which polyurethane is produced. Polyvinyl: A semirigid plastic used for some heel counters. Polyvinyl acetate: A thermoplastic material. Polyvinyl chloride: Thermoplastic material with various applications such as soles and heels and as a coating for uppers and linings. Porous: Material having pores. Posterior: Behind, back. Posting: The use of a firmer material (usually midsole material) to slow foot motion in the rear or middle of the midsole. Posting is usually used to limit overpronation. Potential energy: Energy associated with position. Prefabricated: A sole unit built from more than one layer. Prescription: Legal order, requesting specific treatment, signed by a medical doctor, podiatrist, or osteopath. Prescription shoe: Footwear prescribed by a medical practitioner, either stock or custom-made. Prewelt construction: Shoe construction in which the upper and welt are joined by chainstitches, the insole and upper are cemented, and the outer sole is lockstitched to the welt. Pronation: A complex multijoint action of the foot that is usually estimated from the inward rotation of the heel relative to the leg producing the inward rolling motion that takes place in the foot and ankle joint following footstrike during running; a triplane motion of the foot or part of the foot that consists of simultaneous movements: abduction, dorsiflexion, and eversion; basically, a movement away from the midline of the body, up and out; lowering the medial foot arch; the opposite of supination. Nearly all runners pronate to some degree, or should. If the foot rolls too far inward, however, injuries can result. Overpronation, or the extreme inward roll of the foot, places a strain on tendons and ligaments.

Proportional last: Last with geometric, rather than arithmetic, grading that conforms better to proportional size increments. Prosthetic foot: An imitation foot closely resembling the shape, texture, flexibility, and weight of a human foot. Used in testing procedures. Proximal: Closest to a reference point such as the center of the body. PU: Polyurethane (cellular plastic). Pump: Low-cut shoe not built above the vamp line and usually held onto the foot without fastenings. Push rod: A rod that functions in conjunction with a cam to open or close valves. PVC: Polyvinyl chloride (plastic material). Quarter: Posterior aspect of the upper; the major pattern piece making up the sides of the upper. Rearfoot stability: The ability of the shoe to control foot pronation during the initial 40% to 50% of the support phase. Resiliency: The ability to regain quickly the original shape (return energy rebound). Resin: Solid organic products of natural or synthetic origin. Ridge: A well-defined intersection of the wall and the conical section of the forepart. Rigid shank: Firm, stiff, inflexible area of the shoe between the heel breast and ball. Rocker bar: See Bar, rocker. Rocker bottom: See Bar, rocker. Rubber: An elastomer or natural rubber compound; resilient natural or synthetic material. Running machine: A piece of equipment, used to test shoe characteristics, that simulates the actions of running on a shoe. Sagittal plane: The vertical plane that passes through the body from back to front dividing it into a left and right half. Seam: Sewing that joins together pieces of the upper. Semi cut: A design cut just on or over the ankle. (Also called three-quarter cut.) Shank: The reinforcement under the arch between the heel and the sole; the bottom area of the last between the breastline and the joint break. Shank piece: Rigid reinforcement of the shank. Shank plug: A metal piece inserted in the shank in order to clinch metal shank fasting staple. Shearing force: A force that causes or tends to cause two parts of a body to slide relative to each other. Shoe size: Prewalkers: 3000-4 Big boys: 5½-11 Infants: 1-8 Growing girls: 3������ ½����� -10 Children: 8������ ½����� -12 Men: 6������ ½����� -16 Misses: 12����� ½���� -4 Ladies: 4-13 Youths: 12����� ½���� -4 Boys: 3�������� ½������� -6

Foot and Ankle 1885

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Silicone: A slippery polymeric material used in treating shoes for water repellency. Skive: The thinning down of edges of leather or poromeric material; to cut in thin layers or to a fine edge. Slip last: The most flexible type of shoe construction. With a slip-lasted shoe, the upper is stitched together like a moccasin and glued to the midsole. Slip lasting allows for a better fit. Lasting method whereby a closed upper is formed before being stretched over the last. Sneaker: The American name for vulcanized, canvas rubber shoe. Sockliner: The material (regularly called an insole) inserted between the foot and lasting insole next to the foot; material covering the dorsal surface of the inner sole. Sole: Bottom or ground contact area of footwear. Sole leather: Heavy leather, usually cattle hide that is dryfinished and used for outer soles. Speed lacing: A lacing method that uses D-rings. Splint, Denis-Browne: Rigid bar between both shoes used to abduct the feet. Splint, Friedman-counter: Flexible strip attached to both counters; used to limit internal rotation. Split: The flesh or the underside of the leather hide after the grain side has been removed. Split leather: See Sole leather. Stability: The ability of the shoe to keep the foot moving in a forward direction, rather than allowing for excessive side-to-side movement. Stabilizer: An ingredient used in formulating elastomers and synthetics. Standard deviation: A standard measure of dispersion of a frequency distribution around the average value. A distribution is typically made up of 3 standard deviations on either side of the average value. Stitchdown: A method of sewing the uppers to the bottom. Straight last: A last that is relatively straight on the medial side to add stability. The straighter the last, the greater the medial support. 1. Form for constructing a shoe that can be worn on either foot. 2. Form for constructing a shoe in which the medial border approximates a straight line. Studs: Large knobs protruding from the sole. Suction cups: Indentations on the outsole that provide traction on smooth surfaces. Supination: A triplane motion of the foot or part of the foot that consists of simultaneous movements: adduction, plantar flexion, and inversion; basically, a movement toward the midline of the body, down and in; elevation of the medial foot arch; the opposite of pronation. Oversupination occurs when the foot remains on its outside edge after heel strike instead of pronating. A true oversupinating foot underpronates or does not pronate at all, so it does not absorb shock well. It is a rare condition, occurring in less than 1% of the running population. Symmetrical: In shoemaking, this applies to lasts or patterns that have even sides, the right side the same as the left side. Synthetic: Something resulting from synthesis rather than occurring naturally; a product (as drug or plastic) of chemical synthesis.

Tanning: Process of converting raw hides and skins into leather by a combination of chemical and mechanical means. Tensile strength: The pulling force expressed in measuring leathers or fabrics; the resistance of a material to being pulled apart. Terminal wear condition: A condition in which the outsole of a shoe is worn completely through to the midsole or underlying material. Thermoplastic: Material capable of being repeatedly softened by heat and hardened by cooling; a type of rigid, durable plastic used for most heel counters. Thomas bar: See Bar, Thomas. Throat: Entrance of the shoe where normally the vamp and quarters meet; the topline of the vamp in front of the ­instep. Throat opening: The distance in a straight line from the vamp point to the back seam tuck. Toe box: Reinforcement used to retain the original contour of the toe and guard the foot against trauma or abrasion. Toe cap: An additional protective device on the frontal toe area. Toe recede: The slope of the top surface of the last extending from the toe point to the point of full toe thickness. Toe spring: The vertical distance between the base plane and the toe point of a last having the desired heel elevation; the vertical distance between the ground and the toe point giving the shoe frontal pitch. Tongue: A layer of upper material that protects the top part of the foot from pressure from the laces. Some tongues on Asics shoes are now split to allow the foot to expand. Tongue guide: The tag or slit in the tongue through which the laces are slotted to hold the tongue in place (lace ­keeper). Top line: The open area of the shoe around the ankle. Torque: A force that causes or tends to cause rotation of an object about an axis. The torque (also called moment) is the result of the magnitude of the force, its direction, and distance from the axis of rotation. Torsion: The stress caused by twisting a material. Torsional rigidity: The amount of stiffness in the shank and waist of a shoe. Torsion system: The flagship technology of Adidas. It’s a system designed to allow the forefoot and rearfoot to move independently of each other to encourage freedom of movement. Shoes designed with the Torsion system have a groove cut into the midsole where the foot bends naturally during the running gait. The Torsion Bar is a Kevlar strip embedded lengthwise into the midsole to control excessive twisting of the foot. TPR: Thermoplastic rubber. Traction: The amount of friction or resistance to slip between a shoe outsole and the contact surface. Transverse arch: Curvature of metatarsal heads. Transverse force: The force generated at a right angle to the direction of motion by a walker or runner during foot contact and most closely related to rearfoot stability. Also referred to as the mediolateral force. Continued

1886 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 25B-1 Glossary of Foot and Shoe Terminology—cont’d Transverse plane: The horizontal plane that passes through the body from side to side and back to front dividing it into an upper and lower half. Tread: The soling configuration of the outsole. Treadmill: A rotary machine-driven belt that allows subjects to run in a confined space. Treadpoint: Point of the bottom forepart of last or shoe in contact with the treading surface. Tricot: Knitted fabric commonly used for linings in women’s shoes. Unit sole: The bottom unit with sole and heel portions molded together as a single piece. Universal last: A standard last used by sports shoemakers for all width fittings. Upper: The material making up the “top” part of the shoe. Urethane: Plastic used for uppers, soles, top lifts, and other components; commonly known as polyurethane. Urethane: A resin combination with polymers. U-throat: A lacing eyestay pattern at the front of the shoe. Vamp: Forepart of the upper. The top or front part of the upper over the toe and lacing area. Vamp length or depth: The distance measured along the toe profile from the vamp tack to the toe point. Vamp tack: An arbitrary point on top of a last forepart marked by a tack, measured from the toe. Vegetable tanning: Tanning that uses plant or vegetable materials; uses materials derived from plant life such as oak, chestnut, quebracho, myrobalans, or divi-divi. Velcro: Nylon hook and loop tape fastener that clings on contact. Vertical force: The force perpendicular to a level surface. The dominant force generated by a walker or runner during foot contact and most closely related to the shock absorption characteristics of a shoe. Vinyl: A PVC material that’s available in expanding and nonexpanding types.

Vulcanize: A method of shoemaking in which the rubber sole and/or foxing is cured by heat after attaching to the upper; bonding of the outer sole to the upper from a sole mold in which the soft rubber molds to the shoe, then is allowed to cool and harden. Common in footwear such as sneakers. Waist: Section of the last or shoe between the ball and instep. Waist girth: The smallest dimension around a last between the joint girth and the instep girth. Wall: Straight sides around the periphery of the forepart of certain style lasts. Water repellent: Shoes treated so that they will shed ­water. Waterproof: Shoes treated so that water cannot penetrate. Wear tester: A piece of equipment used to evaluate the resistance of the outsole of a shoe to abrasion. Wedge: Tapered leather, rubber, or other material used to elevate one side of the sole or heel; replaces the heel. Wedge angle: The angle between the heel featherline plane and the base plane, with the last positioned on the base plane. Wedging: Insertion of wedges inside the shoe or on the sole or heel. Weighted shaft: A shaft with a weighted end that is used to impact shoes or material. Welt: A narrow strip around the outside of the sole, stitched between the upper and the sole. Width: Measurement of circumference around the ball of the foot; a coded method for shoe girth sizing. Width sizing: Most running shoes are available in just one width; a few selected models are offered in two. New Balance is the only company that offers its shoes in several widths. For men: from AA to EEEE. For women: AA to EE. Wing tip: The design of the toe cap. ZO2: A Turntec feature; it is a silicone pad that is used for cushioning in the insole.

*This glossary has been compiled from: An A to Z guide to shoe terminology. Runner’s World, 25:48-49, 1990; Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987, pp 163-245; Prescription Footwear Association/Board for Certification in Pedorthics: 1992/93 Desk Reference and Directory. Columbia, Md, Prescription Footwear Association, 1992, pp 65-78; and Janisse D (ed): Introduction to Pedorthics. Columbia, Md, Pedorthic Footwear Association, 1998.

In the evolution of shoewear, the last was originally chiseled out of stone.31 Later models were whittled from wood. A machine used in shaping gunstocks was converted to make lathes and led to the first lastmaking plant in Lynn, Massachusetts, in 1820.31 Today most lasts are made from plastic, a process developed by the Sterling Last Corporation in 1969.31 Metal lasts are used when direct- or ­ injectionmolded soles are attached to the upper because the heat used in this process is poorly tolerated by wood or plastic.22 The dimensions of lasts are based on the average measurements of the segment of the population to whom the shoe will be marketed (e.g., men or women).9,22 In the past, women’s shoes were based on scaled-down versions of lasts derived from the male foot anatomy. Recent investigations have noted several structural differences between the male and the female foot. Specifically, the female foot typically has a narrower Achilles tendon, a narrower heel in relation

to the forefoot, and a foot that is narrower in general than its male counterpart.32 In addition to these dimensional discrepancies, women have proportionately shorter leg length to total body height than do men, necessitating more foot strikes per distance covered. Because of their smaller feet, the heel-to-toe gait cycle is completed more quickly. Consequently, the cumulative ground ­reaction force is increased in the female runner, particularly in elite women runners, who tend to be midfoot strikers.33 The repetitive nature of running causes these factors to be magnified tremendously over the life of a typical athletic or running shoe. Until recently, women’s shoe manufacturers typically scaled down all key internal dimensions of a male athletic shoe in fixed proportion. This practice, termed scaling or grading, persists today. Fortunately, most major athletic shoe companies now have divisions devoted to female ­athletic footwear, and many have developed lasts based on the anatomy

Foot and Ankle 1887

Toe spring

Heel height

Girth

Narrow round

Oval

Pitch Figure 25B-11   Important measurements with respect to the last are toe pitch or toe spring, girth, and heel height or pitch. (Adapted from Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987; and Stacoff A, Luethi SM: Special aspects of shoe construction and foot anatomy. In Nigg BM: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics, 1986; Copyright © 1986 by Benno M. Nigg.)

of the female foot.32 Now, the last is divided and measured in ways that more closely duplicate the average shape of the foot, and a method exists for custom-making a last and producing a shoe to individual specifications. This is frequently done for elite athletes and particularly for athletes in certain sports such as figure skating. It is considered cost ineffective and unnecessary for the general public.22 The divisions and measurements used for the last are shown in Figure 25B-14.22 Individualized lasts are made from an outline of the weight-bearing foot, a weight­bearing impression (to determine pressure distribution), a profile showing the height of the big toe and the instep contour, measurements of overall width and length, and specific girth measurements. Girth measurements are made from (1) the joint—around the metatarsophalangeal joints, (2) the waist—the smallest circumference behind the metatarsophalangeal joints, (3) the instep—the smallest circum­ ference around the arch, (4) the long heel girth—the circumference from the lower edge of the heel around the instep, (5) the short heel girth—the circumference from the lower edge of the heel around the ankle at the lowest crease line, and (6) the ankle—the circumference around the malleoli that is used in high-top shoes and boots.9,22 The process of lastmaking has achieved some level of automation with respect to the upper, for which computer technology has been used more effectively. ­­Computer-aided design

Oxford

Oval round

Natural shape

High/wide

High

High/round

Figure 25B-13   Various alternatives for the shape of the toe box. (From Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987.)

and manufacturing processes produce three-­dimensional designs for the upper and allow direct transference of this information into automated methods of pattern grading and cutting.22 This reduces the work of a formerly labor­intensive operation and foreshadows further innovations in computer-assisted production of sports shoes. Figure 25B-10 illustrates shoe parts.

Specific Shoe Parts: Upper The upper of the shoe is the material that covers the foot. As such, the most important consideration is the specific material used and its relationship to the purposes for which the shoe is purchased. A shoe with a nylon mesh upper is far from ideal for a cold-weather hiking boot, and thick leather is, by the same token, less than ideal for a running shoe in warm climates. These material considerations are discussed in a later section.

Toe Box The front part of the shoe upper is crucial to the health of the toes. Adequate depth is necessary to prevent chafing of the skin over the bony prominences at the interphalangeal joints. Reinforcements of the toe box in the form of stiffeners can vary from being nonexistent in running shoes to being quite stiff in hockey skates or hiking boots. An inserted stiffener protects the toes and prevents the collapse of the upper material onto the toes. Its disadvantages are added weight and stiffness.

Toe Cap Straight

Semicurved

Curved

Figure 25B-12   Illustration of the difference between straight, semicurved, and curved last shoe.

The addition of material called foxing to the front of the shoe protects the toes and increases the durability of the toe box. The toe cap is usually an isolated component in running shoes and is made of suede or rubber stripping.

1888 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Shank

Fore part

Heel seat

Bottom view

Heel curve or rake

Side view Instep Comb Waist

Short heel girth

Back part Long heel girth

Ball/girth Toe width Toe recess

Shank Fore part

Cuboid allowance

Last division and measurements Figure 25B-14   The divisions and measurements used with the last. (From Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987.)

Vamp One of the 16 to 20 pieces of material forming the upper of a shoe is called the vamp. It is the piece (or two pieces) of material forming the front part of the upper and sewn to the eyestays and quarters. Most shoes use a single piece of material for the vamp to eliminate a seam in the toe box area. The vamp is sewn to the quarters at the midfoot level, and these seams are usually hidden by the various trademark stripings of different companies. Split-vamp construction has been popularized as a method allowing better shoe fit for some individuals. It splits the vamp

into two separate pieces with separate lacing systems (Fig. 25B-15).

Quarter This is the other major piece of material composing the upper. Two pieces form the sides of the shoe and conform to the midfoot and arch area of the foot. In shoes designed for side-to-side movement, the vamp and quarter are usually reinforced by extra material (usually leather) called the saddle or arch bandage.

Eyelet Stay and Eyelets

Figure 25B-15   Split vamp shoe with two separate pieces on the upper with separate lacing systems. (Photograph by Thomas O. Clanton.)

The eyelet stay reinforces the holes or eyelets used for lacing. It can be incorporated into the reinforcing material of the saddle and provides additional support for the forefoot and midfoot. Holes are replaced with plastic or metal rings or hooks in some athletic footwear to allow quicker or more forceful lacing. Many athletic shoes have extra eyelets for individualization of a snug but comfortable fit. Widening the reinforcing layer and the eyelets has allowed variable-width lacing to accommodate some of the width sizing difficulties seen in sports shoes. Variable-width or dual-lacing systems can be used with conventional or nontraditional lacing patterns to accommodate variations in foot size, bone spurs, nerve irritation, or other problems. Widely placed eyelets allow the laces to pull the quarter tighter for narrow feet, and narrowly placed eyelets are better suited to a wider foot (Fig. 25B-16).32

Foot and Ankle 1889

control. The size of the heel counter and the quality of the material vary between shoes.

Forefoot and Rearfoot External Stabilizers (Footframe) A

B

C

D

Material is used as a reinforcing component to cup the rearfoot or forefoot of the shoe for greater stability. This is a molded material that may or may not be an integral part of the midsole.

Lining This is the material that acts as the inside backing for the material of the upper. It must be smooth and nonirritating because it is in direct contact with the foot. E

F

G

Figure 25B-16   Examples of lacing patterns for use in certain forefoot conditions. A, Variable-width lacing used   for wide foot. B, Variable-width lacing used for narrow foot.   C, Independent lacing system using two separate laces.   D, Crisscross lacing pattern designed to avoid area of dorsal prominence or pain. E, Lacing pattern useful in highly arched foot so that laces never cross over the top of the foot. F, Lacing pattern designed to pull the toe box up to relieve pressure on the toes. G, Crisscross and loop lacing system used to hold the foot snugly in the heel of the shoe to treat or prevent heel blisters or chafing. (Redrawn from Frey C: Foot health and shoewear for women. Clin Orthop 372:32-44, 2000.)

Specific Shoe Parts: Bottom The bottom, or sole, of the shoe is important in protecting the foot from the environment. Therefore, it requires materials that are both comfortable and durable. In modern athletic footwear, these two features are accomplished by using multiple components.

Sockliner or Insole

The portion of the upper that extends under the laces is called the tongue. It may be padded to reduce irritation to the dorsum of the foot. The tongue is often slit in a way that allows the laces to anchor the tongue and prevent it from sliding laterally.

This material cushions the foot and is the layer between the foot and the bottom of the shoe. Various materials have been used for this layer and are discussed in the next section. Most sports shoes come with removable sockliners to allow them to be replaced when they are worn out or when a corrective orthosis is necessary. The insole reduces friction and provides some degree of shock absorption. It also absorbs perspiration and can provide some canting or control of overpronation. The capability of this material to mold to the shape of the foot can be a source of additional comfort and control.

Collar

Insole Board

The collar forms the uppermost part of the quarters and is the part through which the foot enters the shoe. When excessively stiff or high, the collar can irritate the hindfoot or ankle malleoli. It often has extra padding.

The cellulose fiberboard to which the upper is attached in the conventional lasting process is called the insole board; hence the term board-lasted. This process provides the greatest stability, as opposed to slip lasting, which has no insole board under the sockliner, and leads to improved comfort and flexibility. The intermediate option is combination lasting, in which there is an insole board in the hindfoot with stitching in the forefoot where it has been slip lasted (Fig. 25B-17).

Tongue

Achilles Tendon Protector The extended area on the back of the shoe acts as a pull tab and protection for the Achilles tendon. It should be both molded and padded well to prevent irritation of this area. The high tab design that caused irritation of the Achilles tendon has been replaced by a “bunny-ear” design with a dip in the center.34 The cutaway should be wide enough to prevent friction on the sides of the Achilles tendon.

Heel Counter This reinforcement to the upper of the shoe is located in the heel area. It is a stiffened material of fiberboard or plastic that is molded to the heel and provides greater rearfoot

Midsole Known as the heart of the running shoe, the midsole is sandwiched between the upper and the outsole of the shoe and provides the bulk of shock absorption. With the wedge, this component also produces the desired heel lift, rocker action, and toe spring. Through the use of canting and variable hardness, the midsole can control foot motion. With the use of anatomic contouring, even greater stability and comfort can be achieved. Variations in the materials used add another dimension to what the midsole can do for the foot.

1890 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

C

B

A

Figure 25B-17   Examples of different lasting methods for running shoes. A, Slip-lasted shoe. B, Board-lasted shoe. C, Combination-lasted shoe. (Photographs by Thomas O. Clanton.)

Many of the more significant and recent design advances have occurred through alteration of the midsole.35 These modifications are seen as significant enough by the manufacturers that they are frequently incorporated in the name or advertising campaigns of the various shoe products (Table 25B-1). One particular midsole modification has even contributed to the most recognizable nickname of one of the greatest basketball players of all time—Michael “Air” Jordan.

  TABLE 25B-1  Materials Encapsulated in Midsole Cushioning Designs Shoe Manufacturer

Trade Name

Material and Design

Asics Avia

Gel Arc

Brooks

Hydroflow

Converse

He:01

Etonic

Soft Cell

New Balance

ENCAP

Nike Puma Reebok Reebok

Air

Silicone resin in a pad DuPont Hytrel in polyurethane Silicone fluid in a twochambered plastic bladder Helium gas in polyurethane and nylon Combined gel and ambient air Ethylene vinyl acetate core in polyurethane shell Freon gas in polyurethane Honeycomb pads Honeycomb pads Pads connected by tubing, which allows air flow during stride

Hexalite DMX

Compiled from Heil B: Running shoe design and selection related to lower  limb biomechanics. Physiotherapy 78:406-417, 1992; Frey C: Footwear and stress factures. Clin Sports Med 16:249-257, 1997; and shoe product brochures and the Web sites of the named shoe manufacturers.

Outsole This is the bottom layer of the shoe that makes contact with the ground. The outsole can be constructed with different materials, patterns, colors, and densities. These factors, excluding color variations, can be used to modify the shoe’s stability, flexibility, comfort, and shock absorption. These features are discussed in greater detail later in this chapter.

MATERIALS USED IN SHOES AND SHOE INSERTS The petroleum industry has affected many areas of society, including shoe construction. Traditional materials such as leather, rubber, and cotton (canvas) still have their place, but the search for lighter, more durable, more shock-absorbing materials has led to the development of complex foams with high-tech names such as Elvalite, Hexalite, Hydroflow, Hytrel, Kevlar, Millithane, Phylon, and ZO2.22,36-40 Technologic developments in the field of materials science have progressed at such a rate that it is virtually impossible for the athletic shoe salesperson, the equipment manager, or the sports medicine specialist (much less the average runner) to maintain a current working knowledge in this area. A comprehensive discussion of materials science and the various materials of shoe construction is beyond the scope of this textbook and beyond the attention span of either the author or the reader. Therefore, this section is designed to educate the reader on the fundamentals while providing resources for those interested in further research.

Polymer Science, in Brief The history of polymer science has been summarized succinctly by Cheskin, although he differs with the Encyclopaedia Britannica about some dates.22 Various landmarks are listed in Table 25B-2.

Foot and Ankle 1891

  TABLE 25B-2  History of Polymer Science 1493-1496 1615 1735 1835 1839 1860 1862 1869 1879 1909 1920s 1921 1922 1928 1931 1935 1937 1939 1940s

Christopher Columbus notes Indians playing with balls made from gum of a tree during his second voyage— rubber discovered by Western man Spaniard describes the use of tree milk in Indian footwear and cloaks French geographical team to South America describes “caoutchou” as the condensed juice of the Hevea tree Preparation of vinyl chloride Discovery of vulcanization process for rubber by Charles Goodyear Basic component of rubber discovered and named isoprene First plastic produced by English chemist Alexander Parkes, called Parkesine and later Xylonite. It was a nitrocellulose softened by vegetable oil and camphor. Plasticizing effect of camphor recognized by John Wesley Hyatt; cellulose nitrate patent for celluloid introduced French chemist Bonchardat introduces process of heatcracking rubber. Heat and pressure patent for phenolic resins introduced by Backebrend (Bakelite) Discovery of foam rubber made by confining gaseous bubbles within the rubber First injection molding machine produced and automated by polymerization German chemist Studinger writes that rubber is a chain of isoprene units. This provided theoretical background for polymerization. DuPont begins laboratory study of polymers leading to “superpolyamide” or nylon. Polychloroprene (Neoprene) first marketed commercially Germany produces synthetic rubber: Buna S from styrene butadiene, Buna N from nitrate butadiene Polyurethanes produced British invention of polyethylene Redox process for low-temperature polymerization introduced in Germany to provide more uniform product.

Rubber and Plastic Technology for the Outsole and Midsole Rubber is an organic substance that is a primary component of most athletic footwear. It can be obtained in nature as the milky latex produced by tropical and subtropical trees, or it can be manufactured synthetically.41,42 Natural rubber has the empirical formula C5H8, as assigned by Michael Faraday in 1826.43 The synthetic rubber most commonly used in outsoles is styrene-butadiene rubber.22,43 Styrene’s empirical formula is C8H8, whereas butadiene’s is C4H6.43,44 Different properties of styrene-butadiene rubber can be created by altering the ratio of these two substances or by

varying other elements within the manufacturing process. The addition of carbon black as a filler in the final curing process improves the elasticity and tensile strength of rubber, resulting in improved durability.43 According to the Encyclopaedia Britannica, plastics are “synthetic materials that are capable of being formed into usable products by heating, milling, molding and similar processes. The term is derived from the Greek plastikos, ‘to form.’”45 They have come into increasing use in the shoewear industry in large part as a result of their ability to soften but not melt when heated, which allows for a change in shape without a loss of cohesiveness or mechanical properties and allows for processing into a stable new form on cooling. These synthetics include a vast array of materials that have properties of diverse usefulness. They can be grouped into thermoplastic or thermosetting varieties,46 the basics of which are outlined in Table 25B-3. Regardless of their type, plastics are dependent on the process of polymerization for their existence. This is the process whereby two or more molecules are joined into chains or networks of repeating units (the monomer).45 In an effort to increase the shock absorbency of hard rubber soles, gaseous bubbles were introduced into liquid rubber in a process that produced foam rubber in the 1920s.43 When the structure of the material has openings to environmental air (like a sponge), the material is defined as open-cell foam. Closed-cell foams differ in that their structure is not open to environmental air.47 This cellular construction provides cushioning as a result of the compressibility of the cellular structure as well as the encapsulated gas. Repetitive impact stress can cause a breakdown in the foam material owing to compaction of the foam cell structure. This can occur in as little as 1 hour in some open-cell, lightweight foams. This effect has been studied in running shoes and indicates a loss of 25% of the initial shock absorption after just 50 miles and a loss of more than 40% after 250 to 500 miles, according to machine testing.48 Most experienced runners do not change shoes during this mileage range because it may be reached within 1 month of purchasing the shoe and there may be little external appearance of wear. This type of breakdown has stimulated a search for new and improved polymers for the midsole as well as alternative cushioning systems. Polyurethanes are one of the most versatile groups of synthetic rubber polymers.22,41 They are produced by the polymerization process, in which diisocyanates react with polyols (multiple OH groups) into polyesters or polyethers.45 A liquid catalyst or resin hardener is used to initiate the chemical process. In liquid or semiliquid form, the polyurethane rubber can then be cast, mixed, milled, or vulcanized. Variations in the process can result in materials with

  TABLE 25B-3  Synthetic Material Properties Group

Properties

Examples

Thermoplastic

Soft when heated Hard when cooled Heating and reheating repeatable (heat labile) Shaped by heating Retains shape when cooled Once set, process not repeatable (heat stable)

Ethylene vinyl acetate, polyvinyl chloride, polyethylene Polypropylene, some polyurethanes

Thermosetting

Melamine, phenolic and furan resins, aminoplastics, alkyds, epoxy resins Polyesters, silicones, most polyurethanes

1892 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

a wide range of properties, ranging from hard plastics to soft foams with consequent differences in rigidity and flexibility. Polyurethanes can be either thermoplastic or thermoset resins. Because of their versatility, these compounds have become an integral part of the athletic footwear industry with uses as both midsoles and outsoles. ­ Relatively lightweight and quite durable, polyurethane can be injectionmolded into specific shapes or used as a flat sheet. This allows it to be processed into multistudded athletic shoes and intricately patterned running shoe outsoles.49 Companies continue to support research and development departments to find the ultimate polyurethane for cushioning and durability without sacrificing additional weight.38,39 Ethylene vinyl acetate (EVA) is a copolymer made from ethylene and vinyl acetate in a high-pressure addition polymerization process.22 It is used most commonly as a foam produced by dispersal of gaseous bubbles within the liquid plastic. With a density less than that of polyurethane, it is lighter and less expensive while providing good resiliency and cushioning properties.13,50 It can be adjusted to provide varying hardness, and this has been incorporated into “dualdensity” and “tridensity” midsoles.30,38,39 The harder foam usually has a darker color to denote this difference for the educated consumer. When placed in certain key areas of the midsole, these firmer midsole components can theoretically add control features to the shoe. Because EVA has a tendency to deform with repetitive stress, a process of compression molding is often added, using a combination of heat and pressure to improve the memory and the durability of this foam.30 The competitiveness between shoewear manufacturers and the expanding field of polymer science are irrepressible stimuli to advances in the materials used in athletic shoewear. New forms of lightweight and durable polyurethane have already been introduced and are gradually replacing EVA midsoles in the more expensive running shoe lines.38-40,50 In nonrunning shoes, in which wear and weight are less important, polyurethane midsoles and outsoles are already quite common.51-53 Foams made from new polymers are being introduced with such rapidity that it is hard to keep them all straight. Shoewear companies traditionally attach proprietary names to their technologically based cushioning systems and materials, and it can be difficult to obtain relevant background information. The introduction of air encapsulation in 1979 by Nike signaled a new era in shock absorption technology. Despite early problems with instability, other companies rapidly adopted this encapsulation using other materials (see Table 25B-1). Encapsulation of a cushioning material such as air or gel avoids or delays the compaction seen with traditional midsole foams, thus improving a shoe’s durability.35,54-57 Whether such innovations prevent injuries remains open to speculation, but one can be certain that the athletic shoewear industry will continue to merge science and technology with marketing to give us more sophisticated names and polished promotions.

Heel Counter Materials The next shoe region in which materials play an integral role is the heel counter. The increasing attention placed on rearfoot stability by research resulted in the ­incorporation

of more rigid materials into the heel cup of the shoe. Originally, the heel counter was found only in the running shoes made by the Dassler brothers at Adidas and Puma and consisted of a fiberboard construction.9 The fiberboard lost its stiffness with repetitive wear and constant moisture, leading to the use of more durable plastics. Today, molded plastics take the place of sheet plastic to provide better fit. Heel counters are made from a variety of synthetics, including polyethylene, polyvinyl chloride, and other thermoplastic materials.22,30,39 The stability of the heel counter is enhanced by foxing, external stabilizers, or footframes made of leather or synthetic materials.

Upper Materials Leather remains the most commonly used material in general shoewear construction, particularly for the upper.47,58 It has gradually been replaced in many athletic shoes, in which synthetics provide certain specialty features such as better breathability or reduced weight.22 Leather comes from the skin of an animal and goes through a tanning process to fix the proteins in the skin and eliminate components that would promote degradation. This process can be varied to affect the properties and texture of the finished product. Two layers of the skin are available for use in shoemaking once the skin has been split (Fig. 25B-18).22,47 This process reduces the thickness of the material and changes its properties, the inner split being softer but tending to fray more readily. Leather’s usefulness is apparent from its universal applicability to footwear construction since ancient times. It has the ability to adapt to the shape of the foot and maintain the altered configuration. Leather transmits perspiration (“breathes”) and can be treated to resist or repel water.22,58 Tensile strength is outstanding (up to 4 tons per square inch), flexibility is excellent, and puncture and abrasion resistance is superior.22 The tanning process, along with the finishing and dyeing of the leather, can accentuate one or all of these properties. The disadvantages of leather are its deformability under stress, its tendency to crack when successively moistened and dried, its weight, and cost variations. Nylon-weave uppers are made from polyamide resin fibers woven together into a taffeta and doubled or interfaced with a thin foam or tricot lining.22 In the assembly process, these layers are “flamed,” or heat-bonded. This bonds the laminates together, resulting in a material more flexible and absorbent than if the layers were glued.34 The variables in this composition process are the exact ­material

Grain side or outer split Full thickness of leather

ain it gr Top spl knife Chamois or sp lit s ue Flesh side or inner split de

Figure 25B-18   Splitting of leather into two layers of skin for use in shoemaking. (Redrawn from Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987; and Philps JW: The Functional Foot Orthosis. New York, Churchill Livingstone, 1990.)

Foot and Ankle 1893

used in the thread, the size of the thread, the number of threads in the bundles, and the number and orientation of the bundles going lengthwise and widthwise (Fig. 25B-19).9 The closeness of the weave affects the mechanical properties of the fabric. The nylon weave of a taffeta upper has good durability, softness, and flexibility and is lightweight, making it a superior replacement for leather in many athletic shoes.9,22 Nylon mesh is made from the same nylon threads, but in a knitted rather than a woven process (Fig. 25B-20).9 This knitted process adds space within the strands without compromising strength. Thus, the breathability of the upper is improved. It can be used as a single-, double- or triplemesh knit. Increasing the denier of the thread adds body and strength to the fabric. Whether used in mesh or woven form, the nylon upper is generally combined with a thin foam and a tricot lining for improved fit and ­comfort.22 Other synthetic materials are finding their way into athletic shoe uppers, including Gore-Tex (W. L. Gore and Associates, Elkton, Md).30 Thermoplastic vinyl has uses in golf shoes in coated or laminated applications to leather or fabric. Slush- and dip-molded uppers of thermoplastic materials are used in recreational ice skates and certain waterproof footwear, whereas injection-molded thermoset plastic is the norm for ski boots.22 Depending on the shoe’s intended use, the shoewear manufacturer can provide a full range of breathability, extending from complete breathability up to complete insulation and waterproofing. The upper can be easily or barely deformable. Weight can be varied over a wide range, and the foot can be protected minimally or maximally.

Inlays, Inserts, Insoles, and Orthoses Since ancient times, it has been known that the addition of leaves, moss, or animal skin to the inside of the shoe could provide cushioning for the foot and protection from environmental stresses.10 For soldiers on long marches, this extra protection might mean the difference between life and death. For modern-day runners and athletes, this

Figure 25B-19   Nylon-weave uppers made from fibers woven together into a taffeta. (Redrawn from Cavanagh PR: The Running Shoe Book. Mountain View, Calif, Anderson World, 1980.)

cushioning is intended to protect the most readily identifiable weak link in the kinetic chain. Many athletes would provide testimonial support for the merits of their orthotic devices or inserts in preventing injury or enhancing performance. Although it has been more difficult to document these beneficial effects objectively,31,59,60 scientific evidence of the shock absorption properties of the various materials used in these devices does exist.54,57,61,62 This section provides some acceptable definitions to help make sense of this confusing area and then describes some of the materials used. The following section discusses their ­biomechanical properties. Because of the confusion that surrounds this area, we have chosen to use the definitions of terms accepted by the Pedorthic Footwear Association.63 The insole is the integral design component (layer) of the shoe that is the shoe’s structural anchor to which is attached the upper, toe box, heel counter, linings, and welting. The inlay is a prefabricated removable material upon which the foot directly rests inside the shoe. In some shoes, the inlay is an integral design component. The insert is a type of orthosis, although the term has been used interchangeably with inlays and insoles in some circles to designate an off-the-shelf device placed inside the shoe. For the purpose of this chapter, we use the definition for an insert supplied by the Health Care Financing Agency: a total contact, multiple density, removable inlay that is directly molded to the patient’s foot or a model of the patient’s foot and that is made of a suitable material with regard to the patient’s condition. An orthosis (or orthotic device) is a device that is used to protect, support, or improve function of parts of the body that move. A common error is to use the adjective orthotic as a noun. A pad is a device placed inside a shoe to provide support or relieve pressure from a specific location such as a longitudinal arch pad or a metatarsal pad. Pads are made of various materials and come in a variety of shapes and sizes. Inlays and inserts can be made from a single material or a composite of several materials. The most commonly used materials are leather, cork, foam, felt, and plastic.41 Based on the previous definitions, inlays are “off-the-shelf,” whereas inserts are “custom made.” The latter can be subdivided into

Figure 25B-20   Nylon mesh made from threads with a knitted rather than a woven process. (Redrawn from Cavanagh PR: The Running Shoe Book. Mountain View, Calif, Anderson World, 1980.)

1894 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

those made from a casting of the foot and those made from an impression of the foot (Fig. 25B-21).41,47,64 These can be formulated in a weight-bearing, partial weight-bearing, or non–weight-bearing fashion. A recent trend in orthosis manufacture uses digital foot scanners, obviating the need for a negative mold of the foot. A topographic reading of the plantar foot is obtained using laser scanning, digitized force-plate gait and pressure analysis, or digital analysis of multiple air pegs. These manufacturing devices employ computer-assisted design in the final conversion of data to orthosis.65-68 The devices, which are at present quite expensive, largely rely on the subtalar neutral position as a starting point for data collection. The subtalar neutral position has been variously defined as the position from which there is equal inversion and eversion range of motion, the position from which there is twice as much inversion as eversion, or the position from which the talar head is most fully covered by the tarsal navicular when palpating the foot. It is unclear how much a small variation in this point affects readings and ultimately function. Adequate data in the form of prospective, randomized, controlled studies comparing the use of orthoses manufactured with these systems and inserts made with traditional methods are lacking. There are no data confirming their superiority or justifying their cost.69 Given the uncertainty surrounding the multitude of theories guiding orthotic prescriptions, it would be helpful to have some scientific support for the ability of foot orthoses to accomplish their stated objective. For general classification purposes, inserts can be further divided into accommodative or functional varieties. Accommodative devices are those that are designed with a primary goal of conforming to the individual’s anatomy, whereas functional devices are designed with the primary goal of controlling an individual’s anatomic function, such

as ­ providing support or stability, or assisting ambulation.47,63,70 Different materials can be used for these purposes, leading to an additional subdivision into rigid, semirigid, and soft, in which the material becomes the critical factor. As discussed previously, the materials used in shoes, and now in orthotic devices, can be either natural or synthetic.41 The most commonly used natural materials are leather, rubber, cork, metal, and felt. Synthetics include plastics and foams (both closed and open cell), which can be manufactured with varying qualities of hardness, density, durability, and moldability. The following paragraphs consider the advantages and disadvantages of these materials.

Natural Materials Leather is extremely durable and conforms well to the contours of the foot. It is readily available, is tolerated well by the skin, and combines well with other materials as a composite.41,58 On the negative side, leather is relatively expensive and provides very little shock absorption.47 Rubber has been discussed in the preceding section. It provides good shock absorption with durability but is heavy, and its qualities vary considerably according to the exact process used in its production.41,42 Rubber foams are made by the addition of chemical additives or air. Closedcell foam rubbers are more stable and durable.47 Cork is a lightweight cushioning material made from the outer bark of a tree.71 Although it does not work well when in direct contact with the skin, it is usually used in combination with leather and has a good history of use as an insert. Synthetic forms of cork are now used. Although cork was the material used in the first formal shoe insert in the 18th century,72 it has fallen into disfavor because of its mediocre shock and shear absorption capacities in comparison with some of the newer products.

Synthetic Materials

Figure 25B-21   Foam box used for taking an impression of a foot to make a custom-made orthosis. (Photograph by Andrew Borom.)

Among synthetic materials, there are a wide variety of plastics and foams that have been used in the fashioning of orthoses, inlays, inserts, pads, and insoles. Included in this group are various synthetic rubbers, polyolefins, thermoplastics, thermosetting materials, viscoelastic materials, polyurethane foams, and certain copolymers and composites.41,47,54,64 Because many of these have been described previously, the present discussion is confined to materials in common use that have not been previously described. Styrene-butadiene rubber is a synthetic rubber that has found considerable applicability in orthotic devices both as a basic shell and as a posting material.41 Neoprene is a special rubber made of polychloroprene and is ­ primarily known for its use as an inlay or insole material in the form of a closed-cell foam commonly known as Spenco or as an open-cell foam called Lynco (Table 25B-4).41,54,73 It functions well to reduce friction and attenuate shock but is somewhat hot (particularly in the closed-cell form). The polyolefins consist of polyethylenes, polypropylenes, and their copolymers.41,74 When fabricated as foam, these materials fill a multitude of uses. Because of the modifications available in the manufacturing process, it is possible to vary the mechanical properties of the material over

Foot and Ankle 1895

  TABLE 25B-4  Neoprene Foams

  TABLE 25B-6  Polyurethane Foams

Name

Qualities

Company

Name

Qualities

Company

Spenco

Closed cell, nylon cover Open cell

Spenco Medical Corp Waco, Tex Apex Foot Products, Englewood, NJ

PPT

Open cell, single density Open cell, single density

Langer Biomechanics Group, Deer Park, NY Rogers Corp, East Woodstock, Conn Force 10-Polymer Dynamics, Allentown, Pa UCO International, Prospect Heights, Ill UCO International, Prospect Heights, Ill

Lynco

Poron Axidyne OVA-FLEX

a wide range. Therefore, these foams can be constructed as flexible, semirigid, or rigid orthoses. Table 25B-5 lists some of the trade names for polyethylene foams that are in common use along with their manufacturers.41 Thermoplastic materials are materials that become malleable when heated and hold the set configuration when cooled. They allow reheating for further remodeling.45,46 Polyvinyl chloride (PVC) is one example of such a material. As a plastic, PVC is frequently found in the heel counter of athletic shoes. Polyethylene thermoplastics are one of the most commonly used orthotic materials because they can be fashioned to conform to the patient’s foot and maintain this position under weight-bearing stress.41 The material is relatively stable yet possesses adequate flexibility and strength. Depending on the manufacturing process, the polyethylene thermoplastic can have a low, medium, high, or ultrahigh density. For orthoses, ultrahigh-density polyethylene is most commonly used. Trade names for these products include Ortholen, Subortholen, and Vitrathene.41,47 Thermosetting materials differ from thermoplastics in that they can be molded after heating, but once set, they do not allow repetition of this process. Therefore, they are heat stable.46 The material can be processed into either a plastic or foam and is used in both forms. The most common thermosetting resins are phenolics, furan resins, aminoplastics, alkyds, allyls, epoxy resins, polyurethanes, some polyesters, and silicones.45 The original thermosetting material was a combination of phenol and formaldehyde (Bakelite) formed by high pressure and high temperature.45,46 The physical properties of these plastics can be varied by altering the filler material.46 Polyurethanes can be either a thermosetting resin or a thermoplastic.45 Their principal application is in the form of an open-cell or closed-cell foam. There are a variety of trade names for these products (Table 25B-6). The materials come in various densities and thicknesses, and their mechanical properties are described later in this section.

  TABLE 25B-5  Polyethylene Foams Name

Qualities

Company

Plastazote

Three densities

Pelite

Three densities

Aliplast Berkezote

Four densities Medium density

UCOlite

Medium density

UCOplast 10

Very firm

Apex Foot Products, Englewood, NJ Durr-Fillauer Medical Inc., Chattanooga, Tenn AliMed Corp., Dedham, Mass Foot & Ankle Orthopaedic, Bedford Hills, NY UCO International, Prospect Heights, Ill UCO International, Prospect Heights, Ill

OVA-FIT

Low to medium density Medium density

Viscoelastic materials are materials that combine the mechanical properties of a viscous fluid and an elastic solid. When subjected to a constant deformation or load, their response varies in relationship to the time of application.75 This is an important mechanical property that is also exhibited by tissues of the human body. The viscoelastic nature of a material allows both the storage and dissipation of mechanical energy.59 This would seem to be a desirable quality in an inlay or insert, leading to the claims of manufacturers that it is the perfect material for shock attenuation in running or jumping. Because the time required for stress relaxation to take place is a critical factor in functional shock absorption, however, one cannot assume that these claims are entirely accurate. The trade names for the most commonly used viscoelastic orthoses are given in Table 25B-7. Copolymers are substances formed by a process of co­ polymerization wherein two unlike molecules are united in either a randomly or regularly alternating sequence within a chain.74 An example of this is the nylon acrylic resin Rohadur, which is made from methylmethacrylate and acrylonitrile.41 It is heat moldable and rigid and is primarily used as a functional orthosis. According to Levitz and coauthors, this material was developed in Germany in the 1950s and sold in the form of resin pellets. The different companies that marketed the product for orthopaedic use turned the pellets into sheets of resin.41 When the original material was discovered to be possibly carcinogenic, it was reformulated and is now a clear mahogany-colored plastic. It is widely used in making orthoses and comes in sheets varying from 2 to 5 mm in thickness. It is one of the most rigid materials available.47 Although each of these materials can be found among the various orthotic devices on the market today, each has specific advantages and disadvantages. One foam may have excellent cushioning properties yet be inordinately heavy or hot, and one plastic may be very strong and stable while being difficult to work (Table 25B-8). For this reason, the use of composite orthoses has emerged to combine the properties of two different materials to their best ­advantage.

  TABLE 25B-7  Viscoelastic Materials Name

Company

Sorbothane Viscolas

Sorbothane Inc., Kent, Ohio Chattanooga Corp., Chattanooga, Tenn

1896 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25B-9  Proposed Concept for Inserts and Orthoses

  TABLE 25B-8  Copolymer Foams Trade Name

Qualities

Manufacturer

Polyolefin Foams

Evazote

Closed cell EVA material

Apex Foot Products, Englewood, NJ

Open cell

Apex Foot Products, Englewood, NJ

Polyvinyl Chloride Foam

S.T.S.

The ideal orthosis is moldable to the foot, durable, lightweight, effective in maintaining the proper foot position, and capable of eliminating detrimental stress. Because it appears that no one material can combine all these properties, we can expect an increase in the types and complexities of these composites.

BIOMECHANICAL ASPECTS OF SHOES AND ORTHOSES An understanding of biomechanics as it applies to athletic shoes and the use of orthotic devices requires some understanding of gait and the history of its study, which dates back to the time of Aristotle. Cavanagh has reviewed this history in his book Biomechanics of Distance Running, wherein he points out the influence of such men as Leonardo da Vinci, Isaac Newton, Giovanni Borelli, the Weber brothers, Etienne Jules Marey, Vierordt, Braune and Fischer, Eadweard Muybridge, A. V. Hill, Wallace Fenn, Nicholas Bernstein, Herbert Elftman, O. Boje, and Rodolfo Margaria.76 Contemporary workers in this field include Barry Bates, David Brody, Peter Cavanagh, Tom Clarke, Ed Frederick, John Hagy, Harry Hlavac, Verne Inmann, Stan James, Roger Mann, Benno Nigg, Merton Root, Don Slocum, Thomas Sgarlato, John Weed, and others. It would take an entire chapter to review the contributions of these men—hence the interested reader is referred to their original works, the references from Mann’s section on biomechanics, and the preceding section on the causes of injury to the foot and ankle. Without the contributions of these individuals, there would be no foundation on which we could base further research. Whether one is talking about shoes or orthoses or a combination of these, from a biomechanical standpoint, one is primarily concentrating on (1) the effect of shoes or orthoses on reducing the forces present at foot strike, (2) their ability to improve functional motion within the foot, and (3) their efficacy in preventing or treating pathologic conditions in the lower extremity. Although it is known that certain other biomechanical parameters such as running economy and speed can be altered by shoes and orthoses, these factors are not discussed in any depth in this section. Nigg reviewed the available literature and combined this with nearly 2 decades of his own investigations in the field of shoewear research to formulate a new concept for inserts and orthoses. Although the interested reader is referred to this excellent review for details, the Human Performance

Situation-Dependent Variables

Subject-Dependent Variables

A force signal acts as an input variable on the shoe, based on the chosen movement.

The soft tissue and mechanoreceptors on the plantar surface of the foot act as a third filter. The shoe acts as a first filter for The filtered information is the force input signal. detected by the central nervous system, which provides a subjectspecific dynamic response. The insert or orthosis acts as a The subject performs the second filter for the force input movement task at hand. signal. From Nigg BM, Nurse MA, Stefanyshyn DJ: Shoe inserts and orthotics for sport and physical activities. Med Sci Sports Exerc 31(Suppl 7):S424-S428, 1999.

Laboratory in Calgary proposes that the traditional view of the ability of orthoses or inserts to align the skeleton is not supported in the available literature. Rather, the concept that an orthosis or insert functions most effectively if it minimizes muscle work is advanced by these investigators.77 The basics of this proposed concept are found in Table 25B-9. The situation-dependent variables can be influenced by the shoe, insert, orthosis, or selection of the movement task, whereas the subject-dependent variables, by definition, vary from individual to individual. If this proposal is accepted, the authors suggest that it is possible to conclude the following regarding orthoses: • The skeleton has a preferred path for a given movement task (e.g., running). • If an intervention supports the preferred movement path, muscle activity is decreased. Interference with the preferred path increases muscle activity. • An optimal insert or orthosis decreases muscle activity. • An optimal insert feels comfortable owing to decreased muscle activity with resultant decreased fatigue. • With an optimal insert, performance should increase in association with decreased muscle activity and fatigue. This proposal is clearly innovative and controversial because it is unsupported by adequate experimental evidence. Further research matching subject and insert characteristics to identify the optimal solution for insert and orthosis fitting will, it is hoped, fill this evidence gap.77 As mentioned previously, the functional orthosis is fabricated based on a specific biomechanical theory. It usually uses the “subtalar neutral” position as the starting point. This theoretically aligns the hindfoot with the forefoot and allows the foot to function in its most biomechanically advantageous position. The functional orthosis changes the position of the foot with respect to the weight-bearing surface. The accommodative orthosis, in contrast, brings the surface up to meet the foot in its steady-state position in an effort to improve weight distribution and alleviate symptoms. The accommodative device must be fabricated from a material that will mold easily to the surface of the foot because one of its primary purposes is to accommodate deformities. In contrast, the functional orthosis must be rigid enough to maintain the foot in the position chosen

Foot and Ankle 1897

for maximal function. This makes it apparent why plastic is usually selected for the functional device and a polyethylene or polyurethane foam is more commonly the choice if accommodation is the goal.

Shock Absorption Loading of the athlete’s body during sports activities has been implicated as a significant causal factor in pain and injury.78-81 The study of this relationship is an essential element in the field of sports biomechanics.76,82,83 Numerous works have been published on the biomechanics of walking and running,83-90 and a like number have documented the forces acting on the foot during various activities.79,88-97 Measurement of the pressure under the foot during gait dates back to 1882 with the work of Beely, who used crude manual methods. Since then, measurement of load has progressed to the use of force plates with the aid of piezoelectric transducers or strain gauge technology with computer analysis.97,98 The vertical force component produces skeletal transients beginning at heel strike or foot strike, and these have been theorized to produce injury through their resultant shock and shear waves.99-104 Experimental support for this analysis was provided by the finding that osteoarthritis developed in the joints of sheep housed on concrete.105,106 The association between chronic repetitive trauma, exercise, and arthritis has been the source of considerable controversy.107-110 Although controlled ­ studies have failed to demonstrate a clear relationship between osteoarthritis and the loads generated in sports such as running (in otherwise healthy participants),109-114 this relationship has nevertheless served as an important catalyst for the athletic shoe industry to improve the shock absorption qualities of shoes. Although increasing cushioning in shoewear seems intuitively appealing as a method to diminish shock transmission to the skeleton, Robbins and coworkers have produced a sizeable volume of work outlining the ­potentially detrimental effects of soft materials in shoes.17,115-118 ­Robbins and Hanna proposed that shoewear creates a pseudoneurotrophic condition that eliminates the plantar tactile response from the human system designed to minimize impact loading through alteration of musculoskeletal response.119 It has been demonstrated that although certain interface materials reduce vertical impact from inanimate objects dropped on them, human landing paradoxically increases these forces.120 Gymnasts landing on a 10-cm thick mat demonstrated a 20% increase in vertical impact compared with a rigid surface.121 Behavioral modifications that can either amplify or reduce vertical impact include variation in amplitude of hip and knee flexion.122 Bending at these joints is decreased when landing on soft surfaces. This stiff-legged landing serves to heighten impact, whereas landing on hard surfaces results in increased hip and knee flexion to absorb energy.121 Stability is part of this equation. When soft materials are placed beneath the plantar surface of individuals on a force platform, stability declines as measured by increased sway.123 Human balance improves when placed on thin, stiff surfaces.124,125 To accommodate these factors, humans adopt landing strategies to deal with the landing surface. Decreased hip and knee flexion is used momentarily to increase ­ stability by

compressing the interface material.121 Robbins and Waked examined ground reaction force in 12 men without disability using a 4.5-cm foot fall onto a force platform covered with one of four materials.121 Vertical impact was inversely related to surface stiffness, with the softest ­surface producing the greatest impact. They concluded that balance and vertical impact are closely related and hypothesized that landing on a soft surface is accompanied by an attempt to render it more stable by compressing the material, decreasing its thickness and increasing its stiffness. Although recognizing the impracticality of barefoot activity, they propose that currently available athletic shoes are too soft and thick and recommend redesign.121 In a further study by the same investigators, a “deceptive” advertising message associated with an EVA-covered platform was shown to produce higher vertical impact forces than the identical platform with a preceding “neutral” or “warning” message.17 Although this theory has not been universally accepted, there is evidence that shoe manufacturers are incorporating thinner midsole shoes into their product lines, with the purported advantage of “more speed and stability.”55 Despite this opposing viewpoint, there are scientific data to support the incorporation of elements of cushioning for the reduction of certain overuse injuries.

Experimental Work Available independent data on laboratory testing of various materials used in shoes and orthotic devices are somewhat limited compared with other areas of research in the fields of sports medicine, orthopaedics, and podiatry. In one of the earliest studies of this kind, Brodsky and coworkers studied the effects of repeated compression and the effects of repeated shear and compression on the behavior of five commonly used materials for shoe inserts.54 They also determined the force-attenuation properties of the new and used materials. The materials tested were Plastazote (Apex Foot Products, South Hackensack, NJ), Pelite (Durr­Fillauer Medical Inc., Chattanooga, Tenn), PPT (Langer Biomechanics Group, Inc., Deer Park, NY), Sorbothane (Sorbothane, Inc., Kent, Ohio), and Spenco (Spenco Medical Corp., Waco, Tex). To test compression, the authors used an Instrom testing machine and subjected the materials to cyclic loads. The greatest degree of compression was seen with the soft-grade Plastazote, which went from an original thickness of 6.6 mm to 4.55 mm after 5000 cycles. Lesser amounts of compression were seen with mediumgrade Pelite, Spenco, and Sorbothane, and the least change was found with PPT. In the same study, the authors also looked at the resilience of the different materials. Resilience is a measure of a material’s ability to resume its original shape after having been distorted. This is an important property in an insole or orthotic device. The resilience of the above materials was determined by remeasuring their thickness after a period of rest. Both Plastazote and Pelite showed good rebound in thickness after a 12-hour rest, which allowed a return to 6.0 mm for the Plastazote (a 70% return). Unfortunately, after rebound takes place, accelerated compression occurs when the material is again subjected to the same stress. Some of the results of this study are displayed in Table 25B-10.54

1898 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25B-10  Maximum Loss in Thickness Expressed as Percentage of Original Thickness after 10,000 Cycles Material

Compression (%)

Shear Compression (%)

Plastazote (soft) Pelite (medium) Spenco Sorbothane PPT

55 15 15 3 0

45 16 3.6 10.2 0

From Brodsky JW, Kourosh S, Stills M, et al: Objective evaluation of insert material for diabetic and athletic footwear. Foot Ankle 9:111-116, 1988. © American Orthopaedic Foot and Ankle Society.

Foto and Birke recently investigated resilience for the most commonly prescribed multidensity material combinations used in manufacturing orthotic devices. Dynamic strain, or the material’s percent deformation per cycle, as well as strain loss (or compression set), reflective of the material’s permanent deformation, was measured for each of four combinations of materials. Cyclic loading to 10,000 and 100,000 cycles demonstrated marked differences in the temporary and permanent deformation of the various combinations. Although the materials tested are more commonly used in the treatment of diabetic plantar pressure problems, the authors point out that an ideal pressurerelieving orthosis should demonstrate a dynamic strain of 50% or better at 350 kPa of pressure and that compression set should be minimal. Excessive compression set corresponds to losses in posting, pressure relief, or accommodation of deformity.55 Furthermore, composite material performance is affected by the overall thickness as well as the ratio of the individual materials. The forces on the foot during walking, running, and other sports activities are rarely only compressive in nature. Therefore, a more physiologic test measures how these materials withstand both shear and ­ compression. This ­process was examined in the study by Brodsky using a special jig designed for this purpose.54 The highest degree of change in thickness was again seen in Plastazote with a lesser amount of change in the samples of Pelite and Sorbothane, minimal change in Spenco, and essentially no change in PPT. Rebound was again noted in the ­materials after a period of rest. One interesting finding was that although Plastazote and PPT showed similar results in both the compression and the shear-compression tests, a notable difference was evident between the Spenco and the Sorbothane. Spenco tended to resist shear better than compression, whereas Sorbothane showed a better resistance to compression than shear.54 From the standpoint of the patient and the clinician, the most pertinent information concerning a given insole material is how much reduction in force it affords, that is, how much of the force that is transmitted is actually experienced by the foot? Bench research confirms that all materials (Plastazote, Pelite, PPT, Spenco, UCOlite, ZDEL, and Sorbothane) reduce the force transmitted to the load cell protected by these materials by 10% to 60%.54,57,126-128 Nigg and colleagues, in a review of the literature, examined the available data on shock absorptive insert materials

and concluded that the typical reduction of impact loading is in the 10% to 20% range. They questioned whether these small reductions are capable of reducing injury, and suggest that material alteration of inserts may produce an effect through adjustments in the muscular response of the locomotor system.77 The ability of different materials to accomplish shock absorption can change depending on the amount of force involved and the study methodology, with the exception of Sorbothane, which transmits the highest forces over the entire range.54,57 In contrast, an in vivo study of British Royal Marine recruits evaluating in-shoe pressures found Sorbothane to be significantly better in attenuating peak pressure at heel strike for both marching (23% decrease) and running (27% decrease).62 All materials demonstrate a reduction in shock absorbency after cyclic stress (a phenomenon known as stress relaxation), although this reduction is considerably more apparent in the softer grades of polyethylene foams such as Plastazote, which can lose more than 50% of its shock absorbency after 25,000 cycles.54,126 The actual amount of cushioning contributed by a given material varies according to several factors that have significant implications for their applicability to sports. In general, greater cushioning occurs with increasing thickness of the material, but this also increases weight and affects the stability, comfort, and flexibility of the device or the shoe. Furthermore, it has been shown that softer shoes may allow increased pronation compared with shoes with a firm midsole, for example, and this could produce rather than prevent certain types of injuries.129-131 As stated earlier in this chapter, the correlation between laboratory testing conditions and the physiologic situation in the athlete may be rather poor.120,130-137 Such factors as the size of the missile head used in the impact tests, the shape of the head, and the height from which the missile is dropped can all affect the results of the tests.134 Other factors influencing load magnitude include running velocity, joint kinematics, running strategy, choice of surface, and pattern of foot contact.23,72,50,93,130,132-135,138-140 Therefore, one must be cautious in interpreting test results and include in the equation the data gained from experience and clinical ­trials.

Clinical Work The clinical influence of improved shock absorption provided by shoes and inserts is reviewed to some extent in Chapter 25C. Other studies dealing with the effect of cushioning on injury and pain are discussed here. Stress fractures are an obvious example of an injury for which one would expect to see a reduction in incidence with the use of improved cushioning in shoes or with shockabsorbing insoles. Unfortunately, studies of stress fractures in military recruits have shown inconclusive results. Milgrom and colleagues studied the effect of a semirigid, composite orthotic device on the incidence of stress fractures in Israeli Army recruits.141 There was a reduction in the incidence of femoral stress fractures, but the effect on tibial and metatarsal fractures was insignificant. In fact, the average number of stress fractures per recruit was identical in both groups of recruits, those with the orthotic device and those without. In a subsequent study, Milgrom and coworkers prospectively randomized 390 recruits to

Foot and Ankle 1899 Prospective study n=131 Avg. 30 km/wk Nigg et al., 1995

Relative injury frequency 40 Percent

30 20 10 0

A

F Low 25% 605–1018

Mid 50% 1019–1378

High 25% 1379–2000

(N)

40 Percent

30 20 10 Loading rate

0

B

Low 25% 0.8–47.2

Mid 50% 47.3–79.1

High 25% 79.2–97.4 (N/s)

Figure 25B-22   Relative injury frequency for groups with (A) high-, medium-, and low-impact force peaks, and (B) high-, medium-, and low-maximal loading rate. (Redrawn from Nigg BM, Kahn A, Fisher V, Stefanyshyn D: Effect of shoe insert construction on foot and leg movement. Med Sci Sports Exerc 30:550-555, 1998.)

train in either standard issue military boots or a modified ­basketball shoe. The latter group sustained significantly fewer metatarsal stress fractures and overuse injuries of the foot. Prevention was limited, however, to injuries related to vertical impact loading. In particular, tibial stress fractures were not prevented because they are the result of bending stress. The overall incidence of lower extremity overuse injuries was likewise unaffected.142 Milgrom and associates prospectively examined stress fracture incidence in Israeli army recruits fitted with either a soft or semirigid custom-

made functional orthosis, and compared them with a group who wore no biomechanical orthosis. All recruits trained in a modified infantry boot whose sole design resembled a basketball shoe. Although their data appeared to suggest a protective benefit in the soft orthosis group, more than half of the subjects failed to complete the study. The leading cause of failure to complete the study was dissatisfaction with the orthosis, although the soft orthoses were better tolerated than the semirigid variety.62 In a separate study, Nigg prospectively followed 131 runners for 6 months. Although no difference in injury frequency was seen for subjects with low-, medium- or high-impact force peaks, those runners with a high loading rate sustained roughly 50% fewer injuries than those with a low loading rate (Fig. 25B-22).143 Gardner and coworkers found no reduction in stress fractures with the use of a viscoelastic insole placed in the shoes of military recruits.144 Another element of shock absorption relates to foot pressure and involves proper distribution of load. Locally concentrated forces can cause pathologic lesions such as intractable plantar keratoses, corns, and in the diabetic population, ulceration. Additionally, comfort may be affected by concentration of local pressure. Plantar pressure assessment has played a key role in the management of diabetic patients with neuropathy.145-149 Running shoes, by virtue of their design, have been used to better distribute plantar pressures in this patient population in the hopes of avoiding ulceration.147 Shoewear that reduces peak plantar pressure can do so either by reducing the force or by increasing the area over which the force is distributed. The ability of an insole or insert to distribute peak pressures and cushion the plantar foot, although related to the composition and thickness of the insert material, seems more dependent on the plantar tissue thickness.150 Although the footwear industry is known to evaluate inshoe pressure, this information (like so much else in the athletic shoe industry) is considered proprietary and is not available to the scientific community.146 Although pressure assessment does not seem necessary or practical in the normal population, ­ Figure 25B-23 demonstrates the

Peak pressure = 177 kPa at great toe

No contact on medial longitudinal arch

Peak pressure = 53 kPa at great toe

Minimal contact on medial longitudinal arch Peak pressure =146 kPa at heel

Peak pressure = 181 kPa at heel

A

Contact area = 131 cm2

B

Contact area = 144 cm2

Figure 25B-23   Example of peak pressure measurements comparing different shoes. A, Shoe with limited cushioning properties. B, Shoe with improved cushioning properties. (Redrawn from Mueller MJ: Application of plantar pressure assessment in footwear and insert design. J Orthop Sports Phys Ther 29:747-755, 1999.)

1900 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

ability of a moderately priced running shoe to distribute pressure more effectively over its plantar surface than an inexpensive sports shoe.146 Considering the amassed results of bench and clinical studies, there is considerably less correlation than one would expect for the hypothesis that shock attenuation reduces injury. At best it would appear that improvement in a shoe’s shock-attenuating characteristics can decrease vertical impact-related lower extremity injuries and can change the pattern, but not the overall incidence, of overuse injury.

Alignment and Control As noted in the chapter sections on the causes of injury to the knee, lower leg, and foot and ankle, there is a suspected association between positional abnormalities in the feet and injuries occurring in other areas. As more attention is focused on sports and their attendant injuries, many workers have sought an explanation for the causes of these injuries and a method of treatment that would allow continued participation in sports. The running shoe industry has capitalized on this proposed association between injuries and the feet and has developed entire shoe lines as well as specific shoe features based on the idea of providing improved “control” for the foot. Biomechanical theories have provided the rationale for this development as well as a method of treatment using prescription orthoses. The concept is that there is an ideal functional position for the various articulations of the lower extremity. Theoretically, when the foot is in a properly balanced position (achieved by having the subtalar joint in a neutral position), the foot has the greatest ability to adapt to the stresses placed on it in weight-bearing. In this subtalar neutral position, the maximal amount of inversion and eversion is available and can be used to dissipate the forces of weight-bearing and to transfer load properly to other areas of the lower ­extremity. Alignment has particular implications for problems about the knee because pronation through the subtalar joint results in internal rotation of the tibia. Excessive pronation has therefore been implicated as a causal factor in anterior knee pain,151-157 iliotibial band syndrome,139,151,155-158 pes anserinus bursitis,155 and popliteal tendinitis (see Chapter 25C).151,155,159 The problems attributed to excessive pronation have not been limited to the knee. In the leg, ankle, and foot, excessive pronation has been associated with posterior tibial tendinitis,151,153,156,160-162 overuse syndromes in runners,153,163 medial tibial stress syndrome,162,164 tarsal tunnel syndrome,165,166 cuboid syndrome,167,168 plantar fasciitis,153,156,169 Achilles tendinitis,129,151,153,162 metatarsalgia,162 and stress fractures.170,171 From this listing of virtually every known medical condition affecting the lower leg, it should become obvious to the astute reader that excessive pronation is a seemingly disastrous condition for the athlete. As such, it is natural to expect a plethora of laboratory and clinical studies offering scientific confirmation of the relationship between excessive pronation or rearfoot instability and these pathologic conditions. Unfortunately, the few studies that have addressed this question have failed to provide a conclusive answer.

Although most studies that have been done on the use of rearfoot control features and orthotic devices have focused on the advantages of these features in the treatment of the pronated foot, shoe features and orthotic devices have also been used to treat the opposite foot condition—the cavus or highly arched foot. As discussed in Chapter 25A, the cavus foot type tends to be more rigid and has less available motion to dissipate the forces of weight-bearing. Consequently, it is the cavus foot that has been implicated in the production of plantar fasciitis,151,163,169,172,173 stress fractures,170,171,174 and medial tibial stress syndrome.162 Because the cavus foot is generally more rigid and is commonly associated with a plantar flexed first ray and a varus hindfoot, orthotic support is designed to alleviate these problems. A less rigid orthosis is preferable to provide improved shock absorption and allow some degree of flexibility. Although reports have found that up to 75% of patients who have pronation-related problems benefit from the use of an orthotic device, there has been a considerably less favorable response to the use of orthoses in the cavus foot population.153,175-177 Clinical experience has been the source of most of the evidence supporting the use of orthotic devices for the treatment of a variety of conditions that plague the athlete, particularly the runner. Although much of the scientific and anatomic basis for the use of orthoses can be traced to the work of Manter,178,179 Elftman,178,180-182 Hicks,183,184 Close,185-187 and Inman,43,45,188 it has primarily been orthotists and podiatrists who have experimented with various shapes and materials in an effort to develop a practical approach to the prescription of orthoses. The field of orthotic prescriptions has a pseudoscientific aura. This is created by many factors: erudite yet ambiguous terminology, seemingly contradictory theories, failure to establish what constitutes the normal foot (much less the abnormal), lack of recognition of normal anatomic variation, confusing concepts of what is compensated and what is not, and limited use of the scientific method in establishing the criteria for employment of orthoses and evaluation of their usefulness.31,70,189-195 Faced with this conundrum, it is valuable to reflect on the former foundational work while viewing current shoe and orthotic research with a combination of skepticism and open-mindedness. One can then investigate the available experimental and clinical work, which either supports or refutes the scientific basis for prescribing orthoses or using particular modifications in athletic shoewear.

Experimental Work Studies concerned with control of alignment and maintenance of rearfoot stability must begin by determining an acceptable indication of foot pronation.196 Because pronation is a complicated triplane movement, it is difficult to quantify this movement with currently available techniques. Therefore, it has been generally agreed in the research community to use the degree of calcaneal eversion (valgus) as the indicator of pronation.189,197,198 By using heel eversion alone, the associated abduction and dorsiflexion are discounted.179 Nevertheless, this appears to be the most practical method and the one that has gained acceptance. Measurement is done by observing the subject from a posterior viewpoint using reference markers on the lower

Foot and Ankle 1901

7°

Supinated

–11°

Neutral

Pronated

Shoe with midsole

Shoe without midsole

Bare foot

Figure 25B-24   Position of reference markers on the lower leg to define its axis and a second set of markers on the calcaneus to denote its position in kinematic analysis of gait. (Redrawn from Sport Research Review. Beaverton, Ore, Nike Sport Research Laboratory, Nov/Dec 1989.)

Figure 25B-25   Depiction of the effect of shoewear in increasing pronation from the barefoot condition. (Redrawn from Sport Research Review. Beaverton, Ore, Nike Sport Research Laboratory, Nov/Dec 1989.)

leg to define its axis and a second set of markers on the calcaneus to denote its position (Fig. 25B-24). Gait analysis is then performed using high-speed film cinematography, video cameras, or optoelectronic systems to visualize the markers during each phase of gait.199 The marker positions can then be plotted using anatomic landmarks and sent to a computer for analysis. This is the process of digitalization, from which are derived the specific angles exhibited at specific points in time in the gait cycle. In the cinematographic system originally used, the manual plotting and calculations needed for each frame of film made this an incredibly time-consuming and ­laborious method subject to a certain degree of human error. Kinematic analysis provides information on a number of variables including initial Achilles tendon angle, maximal Achilles tendon angle, initial pronation, total pronation, initial pronation velocity, and so on.95 By employing a video system, one can use a video processor to analyze the film and eliminate part of the tedious process of manual plotting. In the more sophisticated optoelectronic systems, the markers are actually infrared light-emitting diodes and are filmed by infrared-sensing cameras, thereby allowing further automation of the kinematic analysis.197 Regardless of the visualization method used, certain important factors must be taken into consideration. Sampling must be performed at a rate that is at least twice the frequency of the movement being analyzed, requiring a minimal rate of 200 Hz for rearfoot movement.199 The accuracy of the data collection system must be ensured both by the equipment manufacturer and by the on-site testing facility. Calibration, marker-to-marker distance testing, optimization of the collection environment, and meticulous attention to detail are just some of the factors necessary to ensure validity in kinematic testing.199 For example, the markers can be placed using either a relative method, wherein four markers are arbitrarily placed on the posterior foot or shoe and the posterior calf, or an absolute method, which uses standard anatomic landmarks. Furthermore, it should be remembered that testing is done in

a variety of conditions including different test subjects, different speeds from walking to sprinting, different surfaces ranging from treadmill to various over-ground conditions, in shoed and shoeless conditions, in varying types of shoewear, and in shoes with or without orthoses.189 This background is necessary to understand some of the information provided by the studies on rearfoot control. Although it is easy to see how a determination of rearfoot position could be performed in the barefoot runner, an obvious problem exists when the heel is hidden inside a shoe.200 How does one determine the proper position for the markers? This question has been answered by studies that have used a window in the heel of the shoe to allow visualization of the calcaneus position.196 One such study reported by Nigg analyzed measurements using one type of shoe in three test subjects with three trials per subject. A 2- to 3-degree shift was noted between the subject’s heel and the shoe itself.95 Because this shift was systematic, it was not believed to invalidate the test method. In a more recent study, Stacoff placed intracortical bone pins into volunteers to monitor movement coupling between shoe, calcaneus, and tibia. Apart from the difficulty of recruiting volunteers for this type of invasive monitoring, they observed considerable individual differences in coupling between these areas and suggested that we have yet to unravel the details of this complex interaction.201 Shoes and shoe design characteristics can have considerable effect on the kinematics of the foot in running and other sports activities including rearfoot control. Some of this effect is simply the result of displacement of the foot away from the contact surface by the shoe. It has been shown by several investigators that the Achilles tendon angle is decreased in the shoeless condition.189,197,200 The total rearfoot movement and rate of pronation are also reduced in the barefoot runner.95,189,197,200,202 This suggests that wearing shoes increases not only pronation but also other temporally related variables.189,197 This effect of shoewear is demonstrated in Figure 25B-25.

1902 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

The shoe design variables that have been investigated most thoroughly with respect to kinematic effect include sole hardness, heel height, use of a heel flare, width of the midsole, and torsional flexibility.189,197 Sole hardness is measured in terms of durometer, a 25 shore A durometer sole being considerably softer than a 45 shore A durometer sole. Clarke and associates showed that the softer the sole, the greater the degree of maximal pronation and total rearfoot movement.129 A similar study by Nigg found exactly the opposite result, the softer sole having less total pronation.95 A later study by Nigg and colleagues confirmed their previous results, and further elaborated that the total foot eversion was roughly double for hard versus soft inserts.143 The harder inserts allowed for more individual variation of movement and did not force a preset movement pattern to the foot. Despite significant interindividual variance, the subjects with a flexible foot were more likely to have diminished tibial rotation. This study concluded that individual variation must be taken into account when matching feet to inserts. Because other factors such as overall shoe stiffness, shoe construction techniques, and variations in sole geometry can have significant effects on the kinematic function of the foot, it is easy to surmise how different results can be forthcoming from different laboratories when performing similar tests.189 Potential confusion in test results also occurs in the relationship of heel height to rearfoot control variables. Bates and coworkers found that raising the heel height in relation to the forefoot could reduce both maximal pronation and the period of pronation.200 Stacoff and Kaelin analyzed the effect of heel height over the range from 18 to 43 mm and found the same effect in the range from 23 to 33 mm but the opposite effect on pronation at the upper and lower ends of heel height.203 To cap things off, Clarke and associates discovered no significant effect of varying heel heights between 10 and 30 mm.129 It is apparent that there are inconsistencies in the studies generating dissimilar results and adding to an already confusing picture. With the introduction of the New Balance 305 Interval shoe in 1975, a new variable appeared in the shoe-foot control equation.9 The flared heel, seen originally in this shoe, was added primarily to improve the stability of the shoe by widening the base of support when the foot was on the ground. Following the principle that “if a little bit is good, then more will be better,” Nike introduced a shoe in the late 1970s that had a full 1-inch lateral heel flare. Unfortunately, a negative byproduct was quickly perceived in the increasing incidence of lateral knee pain that occurred with this shoe.9 Nigg and Morlock discovered the underlying problem with the flared heel in a scientific study of 14 runners reported in 1987.156 Using three shoes identical except in their degree of lateral heel flare, the authors noted increased initial pronation with wider flare but no difference in total pronation or impact forces at heel strike. Increasing the lateral heel flare increases the lever arm on the axis of motion across the subtalar joint, resulting in earlier initiation of a pronating movement, greater rearfoot angulatory velocity, and an increased initial Achilles tendon (touchdown) angle. A similar study by Nigg and Bahlsen confirmed this result but indicated that heel flare was less important when a softer midsole was used.204 This study supported the earlier work of Cavanagh on a

rounded heel design by demonstrating a reduction of more than 15% in the time of peak force with use of this modification.189,204 Two shoe design characteristics that have been introduced recently are related to midsole width and torsional flexibility. Both have been supported by shoe companies—Nike promoting the wider midsole and Adidas promoting running shoes with greater torsional flexibility.40,196,205 According to the results provided by the Nike Sports Research Laboratory, reductions in maximal pronation and maximal pronation velocity can be obtained by increasing sole width to about 90 mm.197 Torsional flexibility is a new concept related to evaluating the foot in three dimensions rather than by the traditional twoplane analysis of rearfoot inversion and eversion. Work by Stacoff and colleagues has demonstrated that large torsional movement occurs between the forefoot and hindfoot in the barefoot state.202 This three-dimensional linkage between the forefoot and hindfoot is mediated through the tarsometatarsal (Lisfranc’s) and transverse tarsal (Chopart’s) joints and in a shoe can be influenced by shoe sole construction factors. Shoe soles that are stiff in the longitudinal direction restrict the torsional movement normally produced when the forefoot adapts to the ground. Adidas has taken this concept and developed an entire line of running shoes with “a unique construction that controls the natural torsion, or twisting of the foot.”205 They also claim “the foot moves as nature intended it to. The natural twisting of the front part of the foot is controlled so that it no longer strains joints and tendons. Performance greatly improves, whereas injury and muscle strain are substantially diminished.”205 This seems to be an exaggerated claim for a shoe design feature that affects a foot movement that itself is not entirely understood, much less how it is changed by the shoe design and how that relates to other kinematic values. More time and research will be necessary to establish the importance of such factors as shoe width and torsional flexibility because they appear to be mutually exclusive based on examination of the shoes produced with this technology (Fig. 25B-26). The fabrication of orthotic devices to control rearfoot stability, thereby treating various disorders of the lower extremity, has a rather short history. In 1962, Rose presented one of the earliest studies on the ability of a ­positionmodifying device (the Schwartz meniscus) within the shoe to alter the rotation of the lower leg.206 Sheehan, writing in the Preface to The Foot Book by Harry Hlavac, traced the origin of biomechanical therapy to the early 1970s and the development of sports podiatry.207 He theorized that the increasing problem with anterior knee pain in runners was the key to the development of orthotic devices for the foot. The ability of a shoe insert to provide benefit when a surgical solution was not forthcoming led to “the ascendancy of orthopedic medicine.”207 Ascend it did, as orthoses for sports scaled new heights in cost, in numbers, and in varieties of conditions for which they were recommended. At present, many athletes and coaches think that the orthotic device is standard athletic equipment. Athletes come to the clinician with one sort of complaint or another and a self-made diagnosis requesting “orthotics.” In this environment, it is critical that individuals involved in the

Foot and Ankle 1903

Figure 25B-26   Example of wide versus narrow midsole widths used in the Nike (right) and Adidas (left) running shoes. (Photograph by Thomas O. Clanton.)

field of sports medicine understand not only the biomechanical principles behind the use of orthotic devices but also the relevant literature from gait laboratory studies. As in the evaluation of shoewear, kinematic analysis of various parameters is essential to the documentation of

it was Nigg who first determined the effect of shoe inserts on gait using film analysis.31 Nigg’s study concluded that a properly functioning insert should change gait characteristics toward values consistent with those of normal feet. Cavanagh and colleagues presented a paper at the 1978 meeting of the American Orthopaedic Society for Sports Medicine in Lake Placid, New York that showed a ­reduction in maximal pronation and maximal velocity of pronation in runners who used a properly placed felt shoe pad as a medial support.189 Nigg and associates reported in 1986 on the use of various conditions and positions (from no support to anterior to posterior positions) for medial supports within the shoe, demonstrating that placing the pad (elastic cork in this instance) more posterior reduced initial pronation and maximal pronation to a lesser extent (Fig. 25B-27).208 Taunton and coworkers found a similar ­reduction in maximal pronation with the use of an orthosis.209 These studies support the ability of an orthosis to affect kinematic parameters. Further work in this area was provided by Smith and colleagues, who studied 11 well-trained runners using soft or semirigid orthoses while running on a treadmill at a 6- and 7-minute mile pace.210,211 They found that calcaneal eversion was reduced in their subjects, who had an average rearfoot varus posting of 4.2 degrees. This reduced maximal pronation from a mean of 11.3 degrees in controls to 10.5 degrees for the soft orthosis to 10.1 degrees for the semirigid device, only a 1% change. Maximal velocity of pronation was reduced from a mean of 540 degrees per second in controls to 430 degrees per second for the soft orthosis to 464 degrees per second for the semirigid group, an 11% change. Although the reduction in maximal pronation is quite small compared with the expected result for this amount of posting, the reduction in pronation velocity may play a larger role in symptomatic relief provided by orthoses—an effect that can be achieved as adequately by the less expensive soft orthosis as by the semirigid orthosis. Although these studies further augment the role of

Changes in Angles Degrees

15

2

∆βpro = maximal Achilles tendon angle

10

3

Lateral

∆β10 = initial penetration

4

5

5

Medial

∆γ10 = initial change of rearfoot angle

3 4 5 Bare With 2 foot out Anterior Posterior Figure 25B-27   Graphic demonstration that placing an insert more posteriorly reduces initial pronation and maximal pronation. (From Nigg BM: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics, 1986. Copyright © 1986 by Benno M. Nigg.)

orthotic effects on gait. According to Stacoff and Luethi,

1904 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

­ rthoses in adjusting kinematic variables, other authors o have had results that are less supportive. One of the most widely recognized studies of foot orthoses and their effect on gait is that done by Bates and associates in 1979.163 This study failed to confirm a significant reduction in maximal pronation with a custom-molded orthotic device compared with shoes in six symptomatic runners. The study compared its results with control ­values obtained in a previous study of 10 asymptomatic runners.200 Their results also suggested that an increased velocity of pronation occurred with the orthotic device in contrast to the findings of the previously mentioned studies. Rodgers and LeVeau produced a similar study using 29 runners fitted with custom-made semirigid orthoses made from polypropylene.212 Subjects ran in their own shoes on an outdoor track and were filmed with 16-mm film at 120 frames per second. Runners completed three randomly sequenced runs in the following conditions: (1) barefoot, (2) shoe without orthosis, and (3) shoe with orthosis. Values obtained for maximal pronation, angulatory velocity of pronation, and time spent in pronation were not significantly different among the three conditions. There was, however, a trend toward decreased maximal pronation and time in pronation in the shoe with orthosis group. Smith and colleagues pointed out some of the flaws in the various studies related to inadequate control of confounding variables such as type of shoe, type of orthosis, and physical measurements of test subjects.211 Given the number of variables that are being analyzed in these gait studies of running, there should be little surprise that some discrepancies exist. More studies will be necessary to determine which of the variables is important in relation to the clinical situation and the prevention or treatment of injuries. Progress in this field would proceed at a more rapid rate if there could be some universal agreement on terminology, methods of measurement, normal values, and sharing of information without commercial self-interest.

Clinical Work Although there is a widespread belief in sports medicine circles that biomechanical abnormalities are a significant causal factor in lower extremity injuries, clinical studies have had paradoxical results. The military formerly believed that the pronated flat foot was more susceptible to injury, but recent work has disproved this belief. DeVan and Carlton suggested as early as 1954 that stress fractures of the metatarsals were equally common among pronated, normal, and cavus foot types.213 Bensel214 and Gilbert and Johnson215 reached the same conclusion from their research. Further confirmation of this conclusion has come from the Israeli Army study of 295 recruits, which indicated that the low-arched foot might even protect against the development of stress fractures.174 The overall incidence of stress fractures in their study was 39.6% in recruits with a high arch, 31.3% in those with average arch height, and 10% in those with a low arch. It is important to note that in these studies, the criteria for defining arch height were subjective, and determinations were made in a non–weight-bearing position. A continuation of the Israeli Army study was reported in 1989 with a more quantitative determination of the

  TABLE 25B-11  Risk Factors in Running Injuries Characteristics of ­Runners

Characteristics of ­Running

Characteristics of ­Running Environment

Age Gender Structural abnormalities Body build Experience Individual susceptibility Past injury

Distance Speed Stability of pattern Form Stretching, weighttraining, warm-up and cool-down

Terrain Surface Climate Time of day Shoes

From Powell KE, Kohl HW, Caspersen CJ, et al: An epidemiological perspective on the causes of running injuries. Physician Sportmed 14:100-114, 1986.

­longitudinal arch based on radiographic analysis.171 This study reported a higher number of metatarsal stress fractures in recruits with low arches, whereas the incidence of femoral and tibial fractures in these recruits was less than that in those who had a higher arch. This result seemingly contradicts the findings of the previous studies. It should be noted that both of the Israeli Army studies excluded those with marked pes cavus or pes planus from the outset. A subsequent Israeli Army study examined the use of custom soft or semirigid functional orthoses and suggested a benefit in reducing stress fractures; however, fewer than half the recruits completed the study using their assigned orthoses (25% dropped out because of dissatisfaction with the orthosis). Nearly 50% of the arriving recruits were already using orthoses (30% custom-made).62 All of this adds further confounding variables complicating interpretation of the various clinical studies and extrapolation of an association between biomechanical parameters and ­symptomatology. Among runners, it is evident that a number of risk factors have been implicated in the production of injuries (Table 25B-11).216-218 Because so much emphasis is placed on the relationship between overpronation and injury, it is natural to expect confirmation of this in the epidemiologic surveys of running injuries. One of the first studies to focus attention on the role of pronation in runners was that of James and colleagues in 1978.153 They reported a 58% incidence of pronation in 180 patients evaluated for a variety of complaints. It is important to note that one cannot draw the conclusion that most injured runners have pronated feet because there is no way of determining the overall incidence of pronation in the running population. Despite this fact, it is our opinion that this study was instrumental in focusing the attention of sports medicine specialists on the role of alignment in sports injuries. Unfortunately (or perhaps fortunately for the overpronator), no such relationship has been clearly determined in epidemiologically valid studies.217 One of the most comprehensive studies of running injuries is the Ontario cohort study, which included 1680 runners. Anthropometric measurements including femoral neck anteversion, pelvic obliquity, knee and patella alignment, rearfoot valgus, pes cavus and pes planus, somatotype, and running shoewear pattern were made in 1000 of these runners. The study concluded that “none of the anthropometric variables measured was

Foot and Ankle 1905

  TABLE 25B-12  Effectiveness of Orthoses

Diagnosis

No. of Patients (%)

Percentage Improved with Orthosis

Posterior tibial syndrome Pes planovalgum Metatarsalgia Plantar fasciitis Calcaneal spur Iliotibial band tendinitis Cavus foot Leg-length inequality Chondromalacia patellae Achilles tendinitis Miscellaneous Total

55 (27.5) 23 (11.5) 30 (15) 20 (10) 18 (91) 14 (7) 13 (6.5) 10 (5) 6 (3) 4 (2) 7 (3.5) 200 (100)

77 90 87 81 66 25 NA NA NA NA

NA, not available. From D’Ambrosia RD: Orthotic devices in running injuries Clin Sports Med 4:611-618, 1985.

s­ ignificantly related to risk.”218 The most consistent risk factor for a running-related injury is weekly training mileage, and this has been proved in study after study.216-220 When weekly distance reaches 64 km or 40 miles per week, the risk for injury increases by 3 times.216,218 Additionally, no correlation has been shown between shoe characteristics (e.g., varus wedge or waffle sole) or shoe expense and injury reduction in these studies.144,218,221 Given all this ­information, what role does rearfoot stability provided by shoes or orthoses play in the prevention or treatment of athletic injury? Although the scientific approach of systematically establishing a basis for the use of orthoses has failed, there does appear to be some practical basis for prescribing orthoses to injured athletes. Seventy-eight percent of the injured runners in the oft-quoted series of James and colleagues reported some benefit from the use of either rigid or flexible orthoses.153 D’Ambrosia subdivided patients by diagnosis and analyzed the numbers who benefited from the use of an orthosis. The subgroup with pes planovalgus had the most benefit (90% of patients), whereas the subgroup with pes cavus had the least benefit (25% of patients) (Table 25B-12).175 Another study from the Louisiana State University Medical Center Runner’s Clinic reported a 72% improvement in symptoms with the use of orthoses, although these were prescribed for only 10% of their total patient population.222 Blake and Denton surveyed 180 patients and reported that 70% claimed that they were “definitely helped” by treatment with functional foot orthoses.223 Donatelli and coworkers used a similar questionnaire survey of patients to document a 90% relief of pain in patients treated with semirigid orthoses alone.224 A recent multicenter study compared custom-made and prefabricated orthoses in the treatment of plantar fasciitis. Patients were treated with stretching alone or stretching with one of four different orthoses, three of which were prefabricated and one of which was custom-fitted. Stretching alone produced improvement in 72%, whereas those who combined stretching with a silicone heel cup showed the most improvement (95%). Subjects who stretched and used a custom orthosis improved only 68% of the time.

Stretching alone or combined with a prefabricated inlay of any of the types tested was significantly more likely to lead to improvement than use of a custom orthosis with stretching.225 Santilli and Candela found that 100% of 40 athletes complaining of metatarsalgia were relieved of symptoms by a custom-molded polyurethane insert that enabled them to resume sports within 3 weeks.226 Kelly and Winson found that a prefabricated foam inlay with the ability to individualize the placement of the metatarsal support was more effective in relieving metatarsalgia than a prefabricated silicone orthosis, which could not be “semi-customized.”145 A study by Gross and associates reported that 75% of 347 long-distance runners had complete resolution of or great improvement in their symptoms with various types of orthoses (mostly flexible).176 Similar success rates of 70% to 90% have been reported by other clinics using orthotic prescriptions for various athletic injuries.31,176 Based on this evidence that orthoses can improve symptoms and generate documentable alterations in defined kinematic variables, one would assume that a positive attitude exists supporting their continued use. The inconsistency is that “different schools of applying inserts seem to have equal success despite the fact that their inserts look quite different.”31 Support for the value of expensive running shoes (incorporating some of the same principles as orthoses) to treat injuries or symptoms is harder to generate.221 Injury rates for runners have shown no decrease during the past 3 decades of improved shoewear technology, although many factors are at work in this statistic.9,216-218,220,221,227-229 These include the entry of the less physically fit into, and an aging of, the running population. Comparative studies of injury rates in runners have been performed for two separate periods by the University of British Columbia Sports Medicine Clinic.227,228 The results indicate a reduction in foot and ankle injuries in the more recently studied group but an increase in knee injuries. This finding begs the question originally proposed by Cavanagh of whether the increasing technology applied to shoewear has actually contributed to this change.9 The work of the Human Performance Group in Quebec appears to substantiate the idea that increasingly well-cushioned and controlled athletic footwear has tended to produce injuries rather than protect the athlete from them.115-119,229 Perhaps an enterprising shoe company will spend part of its advertising budget on a cooperative study with a runners’ clinic to produce a study that will help answer the question of whether these technologic advancements are helping or hurting. Realistically, the number of variables involved presents an almost insurmountable barrier to epidemiologically valid studies of this nature. Although much has been accomplished in this field in a relatively short time, there are still many unanswered questions and ample opportunities for the inquisitive researcher.

Energy Return Although several shoe companies either state or imply the ability of their products to return energy to the athlete, little support exists for this in the scientific literature.77,230,231 Any shoe that would return energy of a beneficial nature to its user would need to return this energy at the ­correct location, time, and frequency. Unfortunately, ­ materials

1906 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

used to increase shock absorption tend to be poor for energy return.232,233 Additionally, the location of maximal possible energy storage (the rearfoot) is not ideal for use of any returned energy.231,233 Nigg and Segesser have demonstrated that running with running shoes actually requires 3% to 5% more energy than running barefoot.231 They have calculated that the maximal amount of returnable energy from shoewear is less than 1%, which is not even sufficient to overcome the energy requirement of accelerating the shoe itself, or to make up the additional vertical movement added by the shoe.231 Consequently, claims of a shoe’s ability to improve or alter performance through return of energy to its wearer should be viewed with skepticism. It appears that energy considerations may be better met through limitation of energy loss.233 Stefanyshyn and Nigg examined the effect of increased bending stiffness in the midsole, through insertion of a carbon-fiber plate, on jump height and loss of energy through the metatarsophalangeal joint.233 They surmised that stiffening the metatarsophalangeal joint would decrease the amount of energy lost and increase jumping performance. The former was indeed the case for both running and jumping, and vertical jump height was increased by 1.7 cm in the stiffest shoe tested. No comment was made on the potential for increased risk for injury to the Achilles tendon complex, as has been suggested in stiff-soled designs.34

Friction and Torque This aspect of shoewear is thoroughly covered in Chapter 25C and will not be repeated here. Obviously, shoe design features carry major significance in this area. Nigg and Segesser suggested that frictional loads on the human body are of greater importance in the production of sportsrelated pain and injuries than impact loads.234 The interested reader is referred to the previously noted chapter for a discussion of friction and torque related to athletic shoewear.

PROPER FIT AND SHOE PURCHASE DECISIONS Although biomechanical abnormalities have caught the attention of both the athlete and the sports medicine practitioner in recent years, these problems are much less likely

to be the source of day-in and day-out problems compared with the difficulties created by poorly fitting shoes. It is the poorly fitted shoe that creates such commonplace annoyances as blisters, ingrown toenails, certain forms of calluses, metatarsalgia, nerve compression syndromes, “black toes,” corns, and a variety of other unnecessary ills. Most of these conditions are preventable with a working knowledge of shoe-fitting techniques. The most classic case of a footwear-related problem is the bunion. Although there are certainly individuals who have an anatomic predisposition to develop this condition, it has become evident from accumulated scientific research that the improperly fitted shoe is of major significance in the causes of bunions, or hallux valgus. Hoffman’s study of barefooted peoples demonstrated “progressive characteristic deformation and inhibition of function” in ­people who wore shoes compared with those who remained ­shoeless.235 Kato and Watanabe pointed out the relationship between the development of hallux valgus as a clinical entity in Japan and the introduction of the Western-style shoe to replace the traditional geta sandal.21 Other studies have reached similar conclusions, but this has had little effect on the shoe manufacturing industry or the consuming public, who continue to believe that the dainty foot is the most attractive foot. Shoe fit is governed primarily by sizing systems used in the manufacturing process. The English system, originated in 1324, was based on the length of barleycorns (one barleycorn equals 1⁄3 inch).22,74 Today there are numerous systems of sizing in use around the world, making it difficult to fit the foot based on size alone. The most common systems are the English, the American standard, the Continental sizing, and the centimeter systems.22 When determining the proper size to try at the shoe store, it is traditional to use a device such as the Brannock, the Ritz, or the Scholl to size the foot. These devices measure the overall length of the foot, the position of the metatarsal break (related to arch length and toe length) and the girth or width of the foot. The latter is designated by a letter in most circumstances, with AAA being the most narrow, C and D being standard medium widths, and EEE being the widest.22 This form of width sizing is relatively standard and corresponds to various last width standards as seen in Table 25B-13 and Figure 25B-28. In athletic shoewear, it is uncommon to find width sizing available. The New Balance Company has made width sizing a relatively successful marketing strategy in the

  TABLE 25B-13  Last Width Standards (Generally Accepted) American Men’s 8A 8B 8C 8D 8E 8EE 8EEE

9A 9B 9C 9D 9E 9EE

10A 10B 10C 10D 10E

Inches (cm) 8½ 8¾ 9 91⁄8 9¼ 9½ 9¾

(21.6) (22.2) (22.9) (23.2) (22.5) (24.1) (24.8)

Inches (cm) 8 ⁄16 815⁄16 91⁄8 93⁄8 97⁄16 911⁄16 10 11

(22.1) (22.7) (23.2) (23.8) (24.0) (24.6) (25.4)

From Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987.

Widths Available AAAAA AAAA AAA AA

A B C D

E EE EE EEE EEEE

Foot and Ankle 1907 International Size-Scale Comparison Chart 15

16

14

15

11

13

14

10

12

13

9

11

12

8

10

11

9

7

9

10

8

6

8

9

7

8

4

6

7

3

5

6

2

4

5

3

4

13

2

3

12

1

2

11

13

1

10

12

13

9

11

12

8

10

11

7

9

10

6

8

9

5

7

8

4

6

7

3

5

6

2

4

5

1

3

4

0

2

3

1

2

0

1

13 13 12

12

12

11 11 10

11

10

5

7 6 5

9

4

1

3 2

8

1 7

6

5

4

3

13 12

32 31

11

30

10

29

9

28

8 7 6

27 26

5

25

4

24

3

23

2 1 13 12

22 21 20

11

19

10

18

9

17

8 7 6 5

16 15 14

4

13

3

12

2 1 0

0

2

1

49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1

32 31 30 29 28 27 26 25

13½ 12½

24

11½

23

10½

22

10

21 20

9 8 7

19

6

18

5

17

4

16 15

3 2 1

14

2/3

13

1/3

12

1/4

11

1/6

10 9 8 7 6 5 4 3 2 1 Approx Age

Metric Scale

French Sizes

Japanese Sizes

Boston Sizes

American Custom Sizes

American Standard Sizes

English Sizes

Inches

American Ladies Sizes

Figure 25B-28   International size scale comparison chart. (From Cheskin MP: The Complete Handbook of Athletic Footwear. New York, Fairchild Publications, 1987.)

1908 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

F

C A

B

D

E

A — Ball girth B — Waist girth C — Instep girth D — Long heel girth E — Short heel girth F — Heel-toe length Figure 25B-29   Right and left feet with different shapes and shoe fit requirements. (Photograph by Thomas O. Clanton.)

running shoe market. At present, most athletic shoes are made on a standard D last for men and a C last for women. Adjustability for width is often incorporated into the lacing system. A variety of lacing strategies can be used to accommodate a range of temporary or permanent foot shape considerations, and these are demonstrated in Figure 25B-16.32 Additionally, newer, round “laceroni” shoelaces have found their way into shoes. These slide more easily than flat laces when used with the newer loop and web eyelet designs. Pressure is thus more evenly distributed on the dorsal foot than with flat laces, and problems with kinking are minimized. Nevertheless, these lace designs have the slight disadvantage of untying more easily.56 Although women’s shoes are made on a narrower last than men’s, many athletic women have a foot shape more like a man’s and therefore are more properly fitted by the use of a man’s shoe. The manufacturing process has a major impact on how the shoe fits, specifically with regard to the exact last that is used. Different manufacturers use different lasts, which naturally affect the size and fit of the shoe and cause the discrepancies seen in the fit of the same size shoes from different companies. It is for this reason that when an athlete finds one particular make and model of shoe that is satisfactory, he or she will not vary from it except in extraordinary circumstances. This has led to such practices as a famous athlete endorsing a particular company’s shoes while wearing the shoe of a different company, or having his shoes modified by a logo change to resemble the shoes being endorsed.22 Clearly, the proper fitting of the shoe is indispensable to maximal performance in the eyes of many sports participants. Because fit and comfort are so critical, it would seem that the popular shoe surveys would try to determine how to achieve proper fit for the benefit of their readership. Evidently, this is a matter of individual preference and is not quantifiable in the same sense that shock absorption and pronation control can be analyzed. There are too many variables among individuals and even major differences between right and left feet in the same individual (Fig. 25B-29). Comfort and fit are a matter of individual

Figure 25B-30   Measurements used in proper fitting of a shoe include the ball girth, the waist girth, the instep girth, the long heel girth, the short heel girth, and the heel-to-toe length or stick length. (Redrawn from Cavanagh PR: The Running Shoe Book. Mountain View, Calif, Anderson World, 1980.)

preference—some people like a snug fit, whereas others like a more loosely fitting shoe; some like the feel that a soft insole provides, whereas others do not like this sensation; and some like the feel of a higher heel, and others cannot tolerate this. In our own dealings with patients, it has become evident that although we can make ­suggestions based on reasonable empirical considerations, it is impossible for us to predict with accuracy which shoe a specific patient with a specific foot type will select as the most ­comfortable. It should be apparent from the earlier discussion that much of what goes into the fit and comfort of the shoe is derived from the lastmaking process (see the earlier section entitled “Anatomy of the Sports Shoe”). Six measurements are taken into consideration in this process: the ball girth, the waist girth, the instep girth, the long heel girth, the short heel girth, and the heel-to-toe length or stick length (Fig. 25B-30).22 Other important specifications are the toe spring, the heel breast, the heel height, and the heel pitch.22,31 These measurements are then used by the lastmaker to turn a piece of unfinished rock maple or other wood into the finished last. The variability of design produces endless possibilities for fit. Stacoff and Luethi calculated that if 20 different elements in shoewear construction were varied by five systematic steps, more than 95 trillion test shoes would be produced.31 Traditionally, a great deal of craftsmanship has gone into the area of lastmaking. Now, with computer technology, standardization has been introduced that will ideally lead to better fit and greater comfort. In determining the proper fit of a pair of shoes, it is commonly believed that the shoe may be somewhat uncomfortable on the first fitting but can then be “broken in” over time. Conversely, it is often thought that the shoe that feels comfortable in the store the first time will then fit comfortably for the rest of its natural life. Unfortunately, both of these concepts are subject to error. For these reasons, it is important to approach thoughtfully the purchase or ­fitting

Foot and Ankle 1909

Figure 25B-32   Demonstration of the “pinch” test. The individual stands in the shoe bearing weight while the examiner pinches a small amount of material in the upper between the thumb and index finger across the forefoot of the shoe. (Photograph by Thomas O. Clanton.) Figure 25B-31   Demonstration of proper fitting of a shoe for length. Between half width and a full width of the examiner’s thumb can be placed between the end of the longest toe and the end of the shoe. (Photograph by Thomas O. Clanton.)

of a pair of athletic shoes. The decision to purchase a particular shoe should be based on the quality of construction (brand name may or may not be a factor in this). One should avoid buying a shoe simply because it is made by a specific company or endorsed by a certain athlete. Many universities and professional sports teams receive shoes at considerable discounts or even have them donated for the publicity derived by the shoe company. In this situation, it is not uncommon to encounter fitting problems in certain athletes who simply do not have a foot that fits well into the selected shoe. Also, some athletes’ feet require shoes with greater stiffness or other specific characteristics that are not available in the offered shoes. Rather than forcing athletes to adjust to the shoe for the sake of conformity, it is preferable to let them participate in the selection of a shoe that they know will fit well and that will allow them to perform to the best of their ability. Although this is seldom possible in intercollegiate or professional sports, the sports medicine specialist should be sensitive to the relationship between poorly fitting shoes and certain complaints of the athlete as well as the need to switch to a more supportive shoe when appropriate. There are three basic determinations to be made in the fitting of shoes. One must first ascertain that the length is correct. This can be guided by the “rule of thumb” test performed by pressing on the end of the shoe while the wearer is applying full weight. There should be between half and a full width of the examiner’s thumb between the end of the longest toe and the end of the shoe (Fig. 25B-31).

It is essential to note that for many people, the second toe is longer than the great toe. Another important test for length is to have the athlete kick the plantar forefoot into the ground as he would in a sudden stop. If the toes jam uncomfortably into the end of the shoe, the shoe will not last long, the toes will suffer, or the shoe will be shelved. It is wise to do this test before the shoe is purchased. The next step in the fitting process is to determine proper width. The “pinch” test helps with this. The individual stands in the shoe while the examiner tries to pinch a small amount of material in the upper between the thumb and index ­finger across the forefoot of the shoe (Fig. 25B-32). The final test is a determination of the flex point of the shoe in relation to the metatarsal break of the foot. If the shoe does not have the proper degree of flexibility in the appropriate location, one can expect problems (Fig. 25B-33). In the past, the flexibility test was one of the common tests used by Runner’s World in their annual shoe survey.9 The shoe was bent through a 40-degree range, and the force required to do this was measured with a strain gauge. It has been assumed that the less force required, the better, because this is force that must be generated by the runner. There is a fallacy with this assumption, however. It has become evident in shoes designed for artificial surfaces that the overly flexible shoe can predispose the wearer to sprains of the metatarsophalangeal joints, such as turf toe (see Chapter 25C for further details).236 Furthermore, certain athletes may have underlying problems such as hallux rigidus or plantar fasciitis that are aggravated by the overly flexible shoe. Joseph, in a study from the 1930s, found that the average male needs only 30 degrees of flexibility in the first metatarsophalangeal joint for normal walking and that the stiffer soled shoe provided better support for

1910 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25B-14  Ten Points of Proper Shoe Fit A

B C

A

1. Sizes vary among shoe brands and styles. Don’t select shoes by the size marked inside the shoe. Judge the shoe by how it fits your foot. 2. Select a shoe shape that conforms as nearly as possible to the shape of your foot. 3. Have your feet measured regularly. The size of your feet changes as you grow older. 4. Have both feet measured. Most people have one foot larger than the other. Fit to the largest foot. 5. Fit at the end of the day when your feet are largest. 6. Stand during the fitting process and check that there is adequate space (3⁄8 to ½ inch) for your longest toe at the end of each shoe. 7. Make sure that ball of your foot fits comfortably into the widest part (ball pocket) of the shoe. 8. Don’t purchase shoes that feel too tight, expecting them to “stretch” to fit. 9. Your heel should fit comfortably in the shoe with a minimal amount of slippage. 10. Walk in the shoe to make sure it feels right! (Fashionable shoes CAN be comfortable!) From American Orthopaedic Foot and Ankle Society, National Shoe Retailers Association, and Pedorthic Footwear Association: 10 Points of Proper Shoe Fit.

B

1

2

Figure 25B-33   Demonstration of differences in metatarsal break point seen in feet of the same length and how this affects shoe fit. A, Both feet have the same overall length (line A) but different heel-to-ball lengths (lines B and C). B, Improper fit of the heel-to-ball length can create functional problems in the shoe. Point 1 represents the center of rotation of the first metatarsophalangeal joint, and point 2 represents the flexion point of the shoe. (From Prescription Footwear Association: Reference Guide and Directory to Pedorthic Practice. Columbia, Md, 1988.)

the foot.237 The findings documented in the earlier section on “Energy Return” also support increased stiffness in the shoe’s sole. One can quickly realize that all the answers are not available on this aspect of comfort and fit. This is one of the many areas in which a great deal of individual variability exists in both objective and subjective factors. A number of other considerations should be taken into account in the shoe selection process and are mentioned only briefly here. As noted earlier, the shape of the shoe is determined by the last, and this shape can be divided into either straight or curved forms (see Fig. 25B-12). The straight last provides greater support along the medial aspect of the foot and is better suited to the athlete who has a lower arch or who tends to overpronate. Cheskin also recommends the straight-lasted shoe for the athlete who participates in activities demanding slower and more controlled movements, whereas the curved last is better for faster movements.22 The curved last is generally better for the individual who has an adducted foot or a cavus foot. In purchasing shoes, one should try to mimic the conditions under which they will be worn as much as possible. Because feet tend to swell at the end of the day or after

vigorous activity, this should be taken into consideration. In most individuals, one foot is larger than the other or has certain anatomic features that mandate greater emphasis in the fitting process. One generally uses a specific sock in specific shoes, and it is important to remember this in the fitting process. Also, one’s shoe size does not remain static over the years. A recent survey by the Women’s Footwear Council of the American Orthopaedic Foot and Ankle Society found that women’s feet increase in size in a relatively consistent fashion from the second to the sixth decade of life.20 Even in the same manufacturer, there can be variations within the same labeled size for different shoes. Despite the fact that in the United States alone, there are an estimated 1.2 billion pairs of shoes sold annually by some 200,000 salespeople employed by more than 50,000 shoe stores or departments, fewer than one fourth of Americans can recall the last time they had their feet sized. It is estimated that nearly three fourths of Americans wear shoes that fit improperly, with the leading offenders being shoes that are too narrow or too short.238 To help combat the seemingly bewildering task of finding a pair of shoes that fit well, the American Orthopaedic Foot and Ankle Society, in combination with the National Shoe Retailers Association and the Pedorthic Footwear ­ Association, has summarized many of the points contained in the preceding paragraphs and compiled a brochure outlining the “10 Points of Proper Shoe Fit” (Table 25B-14). The quality of production is not always the same for every shoe even within the same product line. The purchaser should pay close attention to the construction of the shoe in the fitting process. There may be a bad seam or an improperly applied layer that will affect fit and comfort. The type of material used in the construction of the shoe upper is also of interest because it affects the conformability of the shoe as well as its breathability. The temperature

Foot and Ankle 1911

and humidity perceived by the foot are related to this latter property and are a factor in the shoe’s comfort. One should appreciate the knowledge required of the athletic equipment manager or skilled salesperson in selecting the correct shoe for individuals who have markedly varying feet and participate in numerous athletic endeavors. For this reason, it is helpful if the manager or salesperson has personal experience with athletic shoes as well as some book knowledge and apprenticeship training along with a desire for a satisfied athlete-customer. It is quite common for those involved in the care of athletes, particularly runners, to be asked for suggestions about the best running shoe, tennis shoe, skating boot, and so on. From the foregoing discussion, it should be obvious that there is no one shoe that can fulfill all the criteria necessary to be the ideal shoe for all individuals. We have seen shoes used in sports progress from a rather simplistic design to a design that involves more computers and researchers than many fields of medicine. By the same token, the price has achieved similar emphasis, and we see teenagers getting summer jobs in order to purchase $150 basketball shoes, only to keep them in the closet for fear of having them stolen.60 There is no question that a decision about which shoe to wear or to recommend is very ­important to an industry that pays six-figure salaries to college coaches and offers multimillion-dollar contracts to high-profile professional athletes. Nevertheless, it is this climate of commercialism and competitiveness that has produced dramatic achievements in the athletic shoewear industry during the past 20 years. No doubt even the least expensive shoes are a far cry from the technologically unsophisticated shoes worn by our forefathers. It is unnecessary to reach the conclusion that one must wear the shoe with the latest gimmick or the highest price tag to achieve one’s maximal performance or to avoid serious injury. For the most part, all the well-known companies in the athletic shoe industry make a high-quality product and have high- and low-end options. In selecting a shoe, one must remember that price can be used only as a very general guide to the quality and functional usefulness of a shoe. Once a running shoe, for example, reaches the $50 level, it usually incorporates all the principal components that are critical to satisfactory performance in such a shoe. Also, the price of the shoe is

directly proportional to the amount of advertising used to market that shoe. Interestingly, there is usually another shoe of virtually identical quality available at a lower price from the same manufacturer. The consumer should always consider that shoe manufacturers might provide misleading information to the consumer. For example, a 1991 survey by Running magazine asked manufacturers to provide information regarding the type of runner for whom their shoes were designed. Of 171 models of shoes, including spikes as well as running and general athletic shoes, 26% were purported to be for both pronators and supinators, 36% claimed an advantage for both the high-arched and flat-footed individual, and amazingly, two shoes in the survey claimed benefit for every distance, foot type, ­­ heel-strike pattern, and surface mentioned.34

SUMMARY Knowledge of athletic shoes, pads, inlays, inserts, and orthoses has become important in the field of sports medicine for many reasons. This knowledge is essential not only from the standpoint of treating and preventing injuries but also to halt the propagation of misleading information and avoid unnecessary expense. Athletic shoewear is essential for the protection of the athlete’s foot, but this protection must extend to the athlete as a whole. With a brief historical perspective and knowledge of the construction of shoes and orthoses, individuals can better understand the factors involved in this protection. These factors include shock absorption, stability, friction and torque, and proper fit. Their specific contributions to athletic performance, comfort, and injury risk remain incompletely elucidated despite numerous laboratory and clinical studies. It is hoped that this chapter not only has stimulated increased awareness of the role of athletic shoes and orthoses in the field of sports medicine but also has pointed out the need for ­ further research.

R eferences Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 1912 DeLee & Drez’s� O

S e c t i o n

C

Ligament Injuries 1. Ligament Injuries of the Foot and Ankle in Adult Athletes Melissa D. Koenig

The foot and ankle serve as a constant interface with our environment. This unique collection of tissues, each with a variety of specialized functions, allows efficient, upright stance and locomotion. Injury to the foot and ankle ligaments results in varying degrees of impairment and associated disability. Athletic ability, regardless of competitive level, is dependent on foot and ankle function. Diagnosis and treatment of foot and ankle ligament injuries are dependent on a complete understanding of foot and ankle anatomy and biomechanics (Fig. 25C1-1). An excellent review of foot and ankle biomechanics is presented within the first section of this chapter (see Chapter 25A, Biomechanics). The diagnosis and treatment of common ligament injuries to the ankle and foot in the adult athlete are reviewed in the following sections.

INJURY TO THE ANKLE LIGAMENTS In the following section, ankle ligament injuries are arbitrarily divided into lateral ankle sprain, medial ankle sprain, ankle syndesmosis sprain, and dislocation of the ankle without fracture. The division is artificial in that many ankle sprains represent a combination of ligament injuries. Ankle ligament injury associated with malleolar fracture is not discussed as a separate topic. An anatomic division of the ankle ligaments is presented in Table 25C1-1 for the purpose of completeness and discussion.

Lateral Ankle Sprain Lateral ankle sprain represents one of the most common injuries in the athletic population.1 Inversion of the plantar flexed ankle is the accepted mechanism of injury for lateral ankle sprain (Fig. 25C1-2). Stretching or tearing of the lateral ankle ligaments leads to inversion instability, a condition that may present in an acute or chronic setting. The highest incidence of injury is localized to the lateral ankle ligaments but may also include the subtalar joint. Among the lateral ankle ligaments, the most commonly injured structure is the anterior talofibular ligament (ATFL). Among cadets at West Point, one third sustained one or more inversion injuries during their 4-year placement.1 Among high school students, ankle injuries are estimated to occur in 1 of every 17 participants per season; 85% of these injuries are ankle sprains.2

Gerber and associates prospectively evaluated cadets at West Point over a 2-month period.3 Football, soccer, jogging, and basketball were the activities most often associated with ankle sprain. Among high school athletes, ankles sprains are most common among men’s and women’s basketball players, followed by participants in football and women’s cross-country.2 A varsity high school basketball survey revealed that 70% of players had a history of ankle sprain.4 Ankle sprain is the most common soccer injury (17% of injuries in senior division men’s soccer), with a cumulative incidence of 27% over an 8-year period.5 Lateral ankle injuries were the injury most commonly associated with disruption of training among runners in one study.6 Commonly cited risk factors associated with lateral ankle injury include generalized ligamentous laxity, inappropriate shoewear, irregular playing surface, and cutting activity. Glick and colleagues reported that preexisting laxity of the lateral ankle ligaments, in the form of increased talar tilt on stress radiograph, is a significant risk factor.7 Furthermore, Thacker and coworkers completed a review of the literature and determined that a history of previous lateral ankle sprain is the most commonly cited risk factor for ankle sprain.8 Although hypermobility, generalized joint laxity, and previous ligament injury should intuitively qualify as significant risk factors for lateral ankle injury, Baumhauer and associates published a contrary conclusion.9 They completed a prospective study of joint laxity, foot and ankle alignment, ankle ligament stability, and isokinetic strength as risk factors for inversion ankle injuries. Among 145 ­ college-aged athletes, 15 injuries were reported during a single intercollegiate season (lacrosse, soccer, and field hockey). No significant differences were found between the injured and uninjured groups with regard to the stated risk factors.

Relevant Anatomy Normal Anatomy The talus articulates with the tibia and fibula to form the ankle joint (talocrural joint). The clinical range of motion is variable but usually ranges from 0 to 10 degrees of dorsiflexion and 40 to 50 degrees of plantar flexion. The empirical axis of the joint is somewhat oblique such that plantar flexion and dorsiflexion produce concomitant internal and external rotation of the foot relative to the leg. The rotational movements are translated through the subtalar joint

Foot and Ankle 1913

Inferior extensor retinaculum

Posterior talofibular ligament

Anterior inferior tibiofibular ligament Anterior talofibular ligament Lateral talocalcaneal ligament Interosseous talocalcaneal ligament Cervical ligament

Calcaneofibular ligament

Bifurcate ligament

A

Superficial Deltoid Ligament

Deep Portion Deltoid Ligament

Tibiocalcaneal ligament Tibionavicular ligament Superficial tibiotalar ligament

B

Anterior inferior tibiofibular ligament Anterior talofibular ligament Calcaneofibular ligament

Posterior tibiotalar ligament

Spring ligament

Posterior inferior tibiofibular ligament Deltoid Deltoid ligament ligament

Posterior talofibular ligament Calcaneofibular ligament

Cervical ligament

C

D

Figure 25C1-1   Compendium of the foot and ankle ligaments. A, Lateral view of the foot and ankle demonstrating the anterior talofibular ligament, calcaneofibular ligament, posterior talofibular ligament, anterior-inferior tibiofibular ligament, lateral talocalcaneal ligament, inferior extensor retinaculum, interosseous talocalcaneal ligament, cervical ligament, and bifurcate ligament. B, Medial view of the foot and ankle demonstrating the superficial deltoid ligament, including the tibionavicular, spring ligament, tibiocalcaneal, and superficial tibiotalar components. C, Anterior view of the ankle and hindfoot demonstrating the deltoid ligament with its superficial and deep components, the anterior-inferior tibiofibular ligament, the cervical ligament, the anterior talofibular ligament, and the calcaneofibular ligament. D, Posterior view of the ankle and hindfoot demonstrating the deltoid ligament with its superficial and deep components, the posterior-inferior tibiofibular ligament, the posterior talofibular ligament, and the calcaneofibular ligament.

and the remainder of the foot to produce supination and pronation during the gait cycle. Inman noted that the anterior margin of the talar dome is wider than the posterior margin by an average of 2.4 mm.10 The implication of this differential width is the stability imparted to the ankle joint during ankle ­ dorsiflexion,

along with the relative instability associated with ankle plantar flexion. The functional stability of the ankle is the product of its soft tissue support. The ankle capsule is reinforced by several groups of ligamentous structures. The lateral ligamentous complex includes the ATFL, the calcaneofibular

�rthopaedic ����������� S �ports ������ � Medicine ������� 1914 DeLee & Drez’s� O

  TABLE 25C1-1  Ankle Ligament Groups Lateral ankle ligaments Medial ankle ligaments

Ankle syndesmosis

Anterior talofibular ligament (ATFL) Calcaneofibular ligament (CFL) Posterior talofibular ligament (PTFL) Superficial deltoid (tibionavicular ligament, tibiospring ligament, and superficial tibiotalar ligament) Deep deltoid (deep anterior tibiotalar ligament and deep posterior tibiotalar ligament) Anterior-inferior tibiofibular ligament (AITFL), posterior-inferior tibiofibular ligament (PITFL), distal interosseous ligament (IOL)

ligament (CFL), and the posterior talofibular ligament (PTFL) (Fig. 25C1-3). The ATFL originates from the anterior aspect of the distal fibula and inserts onto the talar body just anterior to the articular facet. It measures 5 mm in width and 12 mm in length.11 Its fibers blend with the anterior lateral capsule of the ankle. The CFL originates from the anterior border of the distal lateral malleolus and courses medially, posteriorly, and inferiorly to its insertion on the calcaneus. Its fibers blend with the peroneal tendon sheath. Typically a cordlike structure 4 to 6 mm in diameter and 2 to 3 cm long, the CFL is directed 10 to 45 degrees posterior to the line of the longitudinal axis of the fibula.11 The PTFL originates from the posterior border of the fibula and inserts at the posterior lateral aspect of the talus; it is 6 mm in diameter and 9 mm in length.11 The PTFL blends with the posterior ankle capsule. The position of the talus relative to the long axis of the leg is important for determination of the function of the lateral ankle ligaments (Fig. 25C1-4).10 At a position of neutral dorsiflexion, the ATFL is perpendicular to the axis

Anterior talofibular ligament

of the tibia, and the CFL is oriented parallel to the tibia.12 In this position, the CFL provides resistance to inversion stress or varus tilt of the talus. If, however, the talus is plantar flexed (the most common position for lateral ankle inversion injuries), the ATFL is parallel and the CFL is perpendicular to the axis of the tibia. Therefore, with the foot in the most common position for lateral ankle ligament injury, the ATFL is placed in the precarious situation of providing resistance to inversion stress.13 Colville and colleagues used 10 cadaveric ankles to measure strain in the lateral ankle ligaments with the ankle moving from dorsiflexion to plantar flexion.14 The ATFL strain increased with increasing degrees of plantar flexion, internal rotation, and inversion. Conversely, the CFL strain increased with increasing degrees of dorsiflexion and internal rotation. The authors concluded that the ATFL and the CFL work in tandem to provide lateral ankle stability throughout the entire range of ankle motion.

Anatomy and Biomechanics The biomechanical characteristics of the ankle ligaments are such that failure (rupture) is due to increasing load as opposed to twisting or shearing.15 Isolated testing of the individual ankle ligaments demonstrates that the ATFL is the first to fail and the deep deltoid ligament is the last to fail.15 The ATFL is considered the weakest lateral ankle ligament.16-18 Lateral ankle ligament failure is typically midsubstance rupture or a talar avulsion.15 Isolated division of the three lateral ankle ligaments yields predictable results.12,19-24 Division of the ATFL allows increased talar tilt with the ankle in a plantar flexed position, but not in a neutral dorsiflexion position. Division of the CFL allows increased subtalar motion in a neutral dorsiflexion position, but the ankle remains stable in a plantar flexed position. Division of both the ATFL and the CFL results in an unstable ankle. Division of the PTFL allows increased ankle dorsiflexion but does not impart lateral ankle instability. The stabilizing effect of each of the lateral ankle ligaments is dependent on the position of the talus at the time of the applied stress. Stormont and coworkers further advanced the method of ligament testing through the addition of physiologic loads with fixed axial rotation to the sequential ligament

Posterior talofibular ligament

Anterior talofibular ligament

Calcaneofibular ligament

Figure 25C1-2   Inversion of the plantar flexed ankle is the mechanism of injury for lateral ankle sprain associated with a tear of the anterior talofibular ligament.

Figure 25C1-3   The lateral ligamentous complex of the ankle consists of the anterior talofibular ligament, calcaneofibular ligament, and posterior talofibular ligament.

Foot and Ankle 1915

ATFL ATFL

CFL

CFL

A

B

Figure 25C1-4   A, At a position of neutral dorsiflexion, the anterior talofibular ligament (ATFL) is perpendicular to the axis of the tibia, and the calcaneofibular ligament (CFL) is oriented parallel to the tibia. In this position, the CFL provides resistance to inversion stress or varus tilt of the talus. B, If, however, the talus is plantar flexed (the most common position for lateral ankle inversion injuries), the ATFL is parallel and the CFL is perpendicular to the axis of the tibia, and the ATFL provides resistance to inversion stress or varus tilt of the talus.

sectioning protocol.25 Inversion stability was provided by the CFL and the ATFL in the unloaded ankle and entirely by the articular surface in the loaded ankle. Eversion stability was provided by the deltoid in the unloaded ankle and entirely by the articular surface in the loaded ankle. The stabilizing effect of the articular cartilage as described earlier (physiologic load and fixed axial rotation) does not represent the in vivo situation. Remember, normal gait and foot and ankle motion are associated with internal and external rotation of the tibia.10 Kinematic experiments by Cass and Settles placed axial loads on the ankle while allowing axial rotation.26 Computed tomography (CT) was used to evaluate the ankle subjected to an inversion stress. Interestingly, the talar tilt did not change with isolated section of the ATFL or the CFL. Division of both ATFL and the CFL produced an average of 20.6 degrees of talar tilt. The authors concluded that ankle joint stability is provided by the lateral ankle ligaments and not by the ankle articular surfaces. Broström described the ligamentous lesions found during the surgical exploration of 105 recent ankle sprains.27 The ATFL was the most commonly injured structure. The ATFL was completely torn as an isolated injury in 65 cases and as an associated injury in an additional 25 patients. The CFL was the second most commonly injured ligament; it was completely or partially torn as an associated injury in 23 patients. The most common combination of ligamentous injuries was a complete tear of the ATFL and a partial or complete tear of the CFL as described in 20 patients.

Complete ligament tears were noted to occur with concomitant rupture of the adjacent joint capsule. Broström observed that with the talus in a reduced position, the torn ends of the capsule and ligament remained well apposed in most cases. Broström further noted that a tear of the CFL was always associated with a tear of the adjacent peroneal tendon sheath, an observation that is key to ankle arthrography (Table 25C1-2). The advent of magnetic resonance imaging (MRI) has allowed a more detailed evaluation of ligamentous injuries, beyond that provided and limited by surgical exposure. ­Tochigi and colleagues performed MRI on 24 patients with acute inversion injury of the ankle.28 They detected 23 ATFL, 15 CFL, 11 PTFL, 8 deltoid ligament, 13 interosseous talocalcaneal ligament, and 12 cervical ligament injuries. In addition to the loss of stability imparted to the ankle by ligament rupture, an interruption of normal neural ­processes has been documented. The disruption of capsular mechanoreceptors29 and the subsequent loss of afferent nerve function and ankle motor coordination may further contribute to the development of chronic instability.30 This loss of function is addressed directly by proprioceptive and coordinated motion rehabilitation. The most significant problem associated with acute lateral ankle sprain is the predisposition to development of chronic instability. The natural history of untreated chronic lateral ankle instability is loss of function and progressive osteoarthritic changes around the ankle.31-33

  TABLE 25C1-2  Distribution of Ligament Injury after Inversion Injury Study 196540

Broström, Brunner & Gaechter, 199134 Povacz et al, 199868

Diagnostic Method

No.

ATFL (%)

ATFL + CFL (%)

Arthrography Surgery Surgery

239 180 73

152 (64) 52 (29) 29 (40)

40 (17) 101 (56) 42 (58)

ATFL + CFL + PTFL (%)

2(3)

CFL (%)

AITFL (%)

Deltoid (%)

0 (0) 27 (15)

25 (10)

6 (2.5)

AITFL, anterior-inferior tibiofibular ligament; ATFL, anterior talofibular ligament; CFL, calcaneofibular ligament; PTFL, posterior talofibular ligament.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1916 DeLee & Drez’s� O

Clinical Evaluation

History

Although rupture of the lateral ankle ligaments is appropriately considered when a plantar flexion–inversion injury is evaluated, it is imperative that the examination not be limited to these structures. An inversion foot or ankle injury is approached as a constellation of possible injuries, including ATFL sprain, CFL sprain, syndesmosis sprain, deltoid sprain, subtalar sprain (chronic insufficiency of the lateral hindfoot associated with subtalar instability isolated or combined with lateral ankle ligament instability has been shown to occur in up to two thirds of patients34), subtalar coalition, bifurcate ligament sprain, peroneal tendon instability, peroneal tendon tear, lateral malleolus fracture, talar dome osteochondral injury, anterior process of the calcaneus fracture, and fracture of the base of the fifth metatarsal. Although classification systems abound, no single system is routinely used in the literature. Sprains can be considered from the perspective of graded ligament injury, as suggested by the American Medical Association35 and O’Donoghue.36 Injuries are graded based on stretch, partial tear, or complete rupture of the ligament, as is noted in Table 25C1-3. Additional information with regard to associated ligamentous injuries is noted. Grading lateral ankle injuries is much more of a gestalt process than a scientific endeavor. It is not important for the physician to differentiate a grade I from a grade II injury, but the physician should be able to discern a grade I from a grade III injury, or an isolated ATFL injury from an ATFL injury associated with rupture of the syndesmosis. Jackson and associates have established a functional classification system that hinges on the ability of the patient to walk with no limp (mild), walk with a limp (moderate), or not walk (severe).1 The degree of injury was related to a return to full activity in 8, 15, and 19 days for mild, moderate, and severe sprains, respectively.

Relevant historical information includes previous ankle injury, the mechanism of injury, the ability of the patient to continue to play or bear weight, and current symptoms. Severe lateral ankle sprains are associated with a history of inversion injury with a characteristic “pop.” Acute pain and swelling develop quickly, and frequently the athlete is unable to continue playing.

Physical Examination Examination of the patient includes evaluation of the entire extremity. Inspection of the leg, ankle, and foot may reveal swelling, ecchymosis, blister formation, or gross deformity. A vascular and sensory assessment is always performed. The region is palpated systematically with attention to pain over ligamentous, bony, or tendinous structures. Tenderness at the ATFL and the CFL is particularly important to note. Muscle testing is assessed for strength and pain during activation. Motion around the foot and ankle is always assessed with the patient seated and relaxed. The knees are flexed and the feet allowed to fall into an equinus position. The leg is gently grasped while the heel is held in a neutral position and the ankle brought to a right angle or neutral dorsiflexion. From this position, maximal dorsiflexion and plantar flexion are observed in both passive and active modes and compared with the uninjured side. Stress testing is a useful clinical tool that provides a portion of the diagnostic data needed for grading ankle sprains. Stress testing alone is not adequate for reproducible diagnosis of lateral ankle ligament injuries.37,38 Ankle stability is evaluated by several stress tests. The anterior drawer test is used to demonstrate the integrity of the ATFL (Fig. 25C1-5). The patient is seated, and

  TABLE 25C1-3  Grading System for Ankle Ligament Injury Acute Grade

Anatomic Injury

Historical Findings

Examination Findings

I

Stretching of the ATFL

II

Partial tearing of the ATFL

Mild swelling, mild ATFL tenderness, stable ankle Moderate swelling, moderate ATFL tenderness, stable ankle

III

Complete rupture of the ATFL

Inversion injury, subacute pain and swelling, continuous athletic activity Inversion injury, acute pain and swelling, inability to continue athletic activity, painful gait Inversion injury with associated “pop,” acute severe pain and swelling, inability to walk

Severe swelling, severe ATFL tenderness, unstable ankle

Subclassification CFL injury

ATFL and CFL injury

Chronic instability

Persistent laxity at lateral ankle ligaments

Medial ankle injury

Deltoid ligament—complete or partial disruption AITFL, PITFL, and IOL injury

Syndesmosis injury Subtalar injury

CFL, interosseous talocalcaneal ligament, cervical ligament

Mechanism related to ankle dorsiflexion Recurrent ankle sprains, “giving way,”30,75 apprehension and anxiety related to the ankle Abduction or eversion mechanism, pain over the medial ankle External rotation mechanism, pain over the ankle syndesmosis Frequent “ankle sprains,” sinus tarsi pain, difficulty on uneven ground

Additional tenderness at CFL, increased varus tilt of the talar dome Swelling and tenderness at lateral ankle ligaments associated with recurrent injury; ankle instability despite grade of injury Swelling and tenderness over deltoid, valgus instability Swelling and tenderness over syndesmosis, pain at syndesmosis with squeeze test177 or forced external rotation176 Increased subtalar range of motion, sinus tarsi tenderness

AITFL, anterior-inferior tibiofibular ligament; ATFL, anterior talofibular ligament; CFL, calcaneofibular ligament; IOL, distal interosseous ligament; PITFL, posteriorinferior tibiofibular ligament.

Foot and Ankle 1917

A

B

Figure 25C1-5   A and B, The anterior drawer test of the ankle. Note the skin dimple consistent with a positive test.

the flexed leg hangs off of the table. The examiner stabilizes the distal tibia with one hand while the other hand grasps the heel behind and pulls the foot forward to produce forward translation. The test is performed with the ankle in both neutral and plantar flexion positions. The results are compared with those of the contralateral ankle, and the test is repeated as required. A few millimeters of translation is normal. With a complete ATFL tear, the talus subluxates anteriorly and a dimple appears over the anterolateral joint due to suction. A portable ankle ligament arthrometer may be used in conjunction with manual testing to improve the accuracy and reliability of the test.39 Testing is occasionally uncomfortable, particularly in the acute setting. False-negative results may be caused by involuntary guarding or pain response. Local anesthesia

A

increases the accuracy of the anterior drawer test (when performed with a mechanical testing device).37 Broström noted that clinical instability was almost never present among arthrographically proven, acute, nonanesthetized ankle ligament ruptures.40 After spinal anesthesia was established, all ankles with a proven ATFL rupture demonstrated a positive anterior drawer sign. Broström went on to report in a separate publication that the anterior drawer test without any form of anesthesia was useful in diagnosing persistent (chronic) tears of the ATFL.41 The results of the anterior drawer test are also influenced by the thickness of the fat pad at the posterior calcaneal tuberosity and by ligamentous laxity.42 The talar tilt test is performed with the patient seated (Fig. 25C1-6). The leg is secured with the examiner’s open hand,

B

Figure 25C1-6   A and B, The talar tilt (inversion stress) test of the ankle.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1918 DeLee & Drez’s� O

the heel is grasped from behind with the opposite hand, and a varus or inversion force is placed in an effort to produce talar tilt. The results are compared with those of the contralateral ankle, and the test is repeated as required. The test is performed with the ankle in both neutral and plantar flexion positions. Stressing the ankle in neutral dorsiflexion differentially tests the function of the CFL, whereas stressing a plantar flexed talus tests the ATFL.12 Increased inversion of the calcaneus may represent ankle or subtalar instability.34 Varus tilt to a limited degree is probably normal.43 The high incidence of bifurcate ligament sprain warrants a brief discussion with regard to its presentation. Broström noted clinical evidence of bifurcate ligament injury in 18.6% of patients with acute ankle sprains and 3.7% of patients with confirmed lateral ankle ligament ruptures.40 Bifurcate ligament injury is characterized by diffuse lateral hindfoot and midfoot swelling with associated ecchymosis. Tenderness tends to localize to the course of the bifurcate ligament, which is distinct from the course of the ATFL. The ankle and midfoot remain stable. Pain is easily reproduced with forced inversion of the plantar flexed foot. Broström noted that the differentiation between lateral ankle ligament injury and bifurcate ligament injury was best achieved by eliciting indirect tenderness.40 He suggested manipulation of the heel to produce lateral ankle pain and stabilization of the heel with simultaneous forced forefoot motion to produce bifurcate pain.

Imaging Radiographs Radiographs in the anteroposterior, mortise, and lateral projections are required for ankle evaluation. Weight-bearing radiographs better reproduce physiologic loading, but they are not always obtainable in the acute phase owing to pain. The radiographs are evaluated with regard to malleolar fracture, physeal fracture, osteochondral fracture, and avulsion fracture. Alignment and translational abnormalities are also noted, particularly at the syndesmosis and the medial ankle joint space. (See “Ankle Syndesmosis Sprain” and “Medial Ankle Sprain,” later.) During surgical exploration of 60 chronic ankle sprains, Broström noted 5 cases (8%) with anterior lateral talar osteochondritis dissecans.44 Anderson and Lecocq reported a 22% incidence of osteochondral lesions in a mixed series of 27 cases of single and recurrent lateral ankle injuries.13 The lesions were located at the lateral talus in 5 patients and at the medial talus in 1 patient. The presence of a subfibular ossicle may be indicative of acute or chronic injury to the ATFL,45 or it may be a normal variant (os subfibulare). Broström noted that avulsion of bone fragments is an uncommon pathologic finding after an acute ankle sprain.40 He further noted that patients who sustained an avulsion fracture were more likely to be older and female.

acute lateral ankle ligament injury.1,34 Accuracy is compromised by pain response, peroneal spasm,46 variable stress technique, and lack of control data in the case of a previously injured contralateral ankle. The use of a local anesthetic injection may eliminate some of these variables and result in more accurate stress radiographs. The talar tilt stress radiograph is an anteroposterior view of the ankle taken while an inversion force is applied. The stress can be performed manually or with commercially available devices that provide standardized and quantitative applied stress. The degree of tilt is determined by measuring the angular divergence between the distal tibial articular surface and the talar dome (Fig. 25C1-7). Stress radiographs have been described with the foot in dorsiflexion, neutral, and plantar flexion with the knee flexed or straight. The literature offers no consensus with regard to normal and pathologic findings in radiographic stress tests. ­Bonnin reported radiographic data demonstrating 4 degrees of varus tilt in 10% to 15% of noninjured ankles.47 Hughes evaluated varus stress radiographs of both ankles in 90 injured and 90 noninjured patients and concluded that 6 degrees of increased talar tilt represents the transition from “normal to abnormal” talar tilt.48 Rubin and Whitten published data analyzing the range of talar tilt present in stress radiographs of 152 normal ankles.38 About 56% of the noninjured ankles in their study had talar tilts of 3 to 23 degrees. However, only 2 ankles measured more than 20 degrees, and only 6 had more than 15 degrees of talar tilt. Seligson and associates used mechanical devices to obtain controlled stress radiographs of 25 functionally normal ankles.49 The talar tilt in these asymptomatic ankles varied from 0 to 18 degrees. The anterior translation as seen on the lateral stress radiograph never exceeded 3 mm. Cox and Hewes performed stress radiographs on 404 ankles of patients with no history of previous ankle injury.50

Stress Radiographs The bilateral stress radiograph is used to quantify anterior talar translation and varus tilt of the talus. Stress radiographs are not routinely necessary for the evaluation of

Figure 25C1-7   The talar tilt (inversion) stress radiograph. The talar tilt angle refers to the angle between two lines drawn to the tibial plafond and the talar dome.

Foot and Ankle 1919

Manual stress was applied to the plantar flexed ankle for an anteroposterior radiograph. No talar tilt was detected in 365 (90.3%) of the ankles, 1 to 5 degrees of talar tilt was detected in 32 ankles (7.9%), and greater than 5 degrees of talar tilt was detected in only 7 ankles (1.7%). Glasgow and colleagues suggested the importance of a lateral stress radiograph.21 They cited the prevalence of ATFL ruptures and the associated anterior instability. The anterior drawer stress radiograph is a lateral radiograph taken while an anterior displacement stress is applied to the ankle with the foot in gentle plantar flexion. The degree of translation is determined by measuring the shortest distance between the talar dome and the posterior margin of the tibial articular surface (Fig. 25C1-8).51 Numerous studies have found a range of translation between 2 and 9 mm, with most less than 4 mm. With more than 5 mm of anterior translation, most consider this a positive test consistent with ATFL rupture.

A

C

Other Imaging Arthrography The unpredictable effect of patient guarding during stress radiography is overcome by the use of ankle arthrography.52,53 This method is relatively simple. Contrast material is injected into the acutely injured ankle, preferably with fluoroscopic guidance, and radiographs are obtained in various projections with attention to areas of extravasation. A false-negative result from an early capsular seal is possible. The procedure has been used much less since the increased availability of MRI. Broström and colleagues performed arthrography of 321 fresh ankle sprains.54 Extra-articular leakage occurred in 239 cases (74%). Surgical exploration was performed in 99 of these cases and in an additional 6 cases that did not demonstrate leakage. The authors concluded that

B

Figure 25C1-8   A, The anterior drawer stress radiograph. Anterior talar displacement (in millimeters) is recorded by measuring the shortest distance from the most posterior articular surface of the tibia to the talar dome. B, The anterior drawer stress radiograph with no anterior talar displacement, negative test. C, The anterior drawer stress radiograph with increased anterior talar displacement, positive test.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1920 DeLee & Drez’s� O

a­ rthrography was useful within the first 7 days of an acute ankle sprain. Leakage of contrast into the peroneal tendon sheath correlated with tear of the CFL. Leakage in front of the lateral malleolus correlated with ATFL rupture. Leakage in front of the syndesmosis correlated with complete rupture of the syndesmosis. Leakage at the medial malleolus correlated with partial deltoid rupture. The presence of extra-articular contrast in the flexor hallucis longus (FHL) and flexor digitorum longus (FDL) tendon sheaths, as well as the posterior facet of the subtalar joint, was not ­diagnostic. Peroneal sheath injection (tenography), as described by Black and associates, is useful in discerning injury of the CFL.53 Contrast leakage from the sheath or passage into the ankle joint suggests CFL disruption.

Magnetic Resonance Imaging MRI is a useful method for evaluation of acute, subacute, and chronic lateral ankle ligament injuries (Fig. 25C1-9).55 Associated injuries to the talar dome, subchondral bone, and peroneal tendons, as well as the interosseous and cervical ligaments of the subtalar joint, are visualized.28 MRI also reveals tarsal coalition. Rijke and colleagues used a dedicated knee coil and axial images of the neutral and plantar flexed ankle to describe various ligament injuries.55 Complete disruption, partial disruption, and laxity were all visualized. Additionally, hemorrhage and soft tissue swelling were indicative of an acute injury. The high sensitivity associated with MRI mandates that images be carefully correlated with clinical findings. Its accuracy, lack of ionizing radiation, noninvasiveness, decreasing cost, and increasing availability suggest that MRI is the imaging modality of choice when a definitive diagnosis with objective documentation is important.

Therapeutic Options Nonsurgical treatment is the primary choice of management for most lateral ankle sprains, no matter how severe. All acute injuries are treated with the RICE (Rest, Ice, Compression, Elevation) method and protected weightbearing. The amount of rest required after an acute ankle sprain is determined by several factors. Many studies have reviewed the effect of cold application to the injured extremity. Cold therapy is an effective, inexpensive, and easy-to-use modality for the treatment of acute musculoskeletal injury. Appropriately applied cold therapy decreases both pain perception and the biochemical reactions that produce inflammation, and produces vasoconstriction with a concomitant reduction in soft tissue swelling and bleeding. Hocutt and coworkers demonstrated the importance of using cold therapy (ice whirlpool or ice pack for 15 minutes 1 to 3 times a day) in the treatment of acute ankle sprains.56 Furthermore, the group demonstrated a faster recovery associated with early cold therapy (within the initial 36 hours after an ankle sprain) compared with delayed cold therapy or early heat therapy. Compression is typically provided in the form of an elastic bandage but may involve casting, splinting, pneumatic orthosis, or mechanical compression devices.57

Elevation of the ankle helps to reduce swelling and pain. Elevating the extremity above the level of the heart should be emphasized. Any decrease in height will reduce the pressure that edema must overcome before being mobilized out of the acutely swollen tissues surrounding the ankle.

Grade I and Grade II Lateral Ankle Ligament Injuries Treatment of mild and moderate ankle sprains is symptomatic, with an emphasis on recovery of range of motion, strength, and coordination. The detailed rehabilitation program is outlined here. A nonrigid functional ankle brace, such as a lace-up brace, or a semirigid pneumatic ankle brace is used during all phases of recovery.

Grade III Lateral Ankle Ligament Injuries Several reliable treatment methods are available for complete lateral ankle ligament injuries. Treatment methods include early mobilization, cast immobilization, and surgical repair. Injection of anesthetics or corticosteroids into the acutely sprained ankle is never indicated.1,58 Aspiration of the joint hematoma and injection of hyaluronidase did not produce improved outcomes in one study.59 The ankle is supported by the use of a variety of methods, including a nonwalking cast, a walking cast, a removable cast boot, a semirigid pneumatic ankle brace, a nonrigid functional ankle brace, and various ankle taping methods. Various immobilization methods are ­summarized in Table 25C1-4.

Cast Immobilization Practically speaking, casting has a useful role when limited to the initial 2 to 3 weeks after acute injury. Application of a short leg cast allows a rapid return to work activities and early discontinuation of crutch walking. Published reports discuss casting periods as long as 6 weeks. Prolonged immobilization is not recommended. In fact, Jackson and colleagues suspect that the use of a cast for moderate and severe ankle sprains simply delays recovery for the number of days that the cast is used.1 Complete immobilization is provided with a short leg cast. The assumption is that casting allows the ruptured ends of the lateral ankle ligaments to heal in a near­anatomic position. Various cast positions have been advocated in an effort to reapproximate the torn ligament ends. Smith and Reischl suggested 5 to 15 degrees of dorsiflexion based on fresh cadaveric models.4,60 Slight eversion is also a common suggestion.61 As soon as the swelling and pain remit, a functional rehabilitation program is instituted with a semirigid pneumatic ankle brace. Drez and coworkers used a more protracted immobilization method to treat 39 patients with first-time, combined ATFL and CFL injuries.61 All cases were evaluated by stress radiography before and after treatment. The protocol included 7 to 10 days in an everted splint and 6 weeks in an everted walking cast, followed by ankle rehabilitation and a 1-month delay in return to athletic activity. A 79.5% success rate was obtained as determined by a

Foot and Ankle 1921

ATFL

ITCL talus

talus

talus fibula

calcaneus

PTFL

tibia

CL

tibia

talus talus

DTL fibula

talus

calcaneus calcaneus

calcaneus

CFL

A

talus talus

talus ATFL

fibula

CFL calcaneus

PTFL

fibula

tibia CL ITCL

talus

DTL talus

calcaneus

talus

calcaneus

calcaneus

B Figure 25C1-9   A, Magnetic resonance imaging key for the normal ankle. B, Magnetic resonance imaging key for the injured ankle. ATFL, anterior talofibular ligament; CFL, calcaneofibular ligament. (Redrawn from Tochigi Y, Yoshinaga K, Wada Y, Moriya H: Acute inversion injury of the ankle: Magnetic resonance imaging and clinical outcomes. Foot Ankle Int 19:730-734, 1998.)

repeat talar tilt stress radiograph with 5 degrees or less angulation compared with the uninjured ankle. Eiff and associates used a prospective study to compare early mobilization and immobilization for the treatment of first-time lateral ankle sprains.62 The early mobilization group was treated with an elastic wrap for 48 hours followed by application of a semirigid pneumatic brace and a common rehabilitation program. The ­ immobilization

group was treated with a non–weight-bearing plaster splint for 10 days followed by a common rehabilitation program. Both methods produced excellent results at 1-year follow-up with low rates for residual symptoms (5%) and reinjury (8%). A semirigid pneumatic ankle brace (Air-Stirrup, Aircast, Inc., Summit, NJ) has shown proven results with advantages that include cost-effective treatment, the capacity for

�rthopaedic ����������� S �ports ������ � Medicine ������� 1922 DeLee & Drez’s� O

  TABLE 25C1-4  Immobilization Methods for Foot and Ankle Injuries Immobilization Method

Common Application

Advantage

Disadvantage

Short-leg nonwalking cast

Initial treatment for severe ankle and midfoot sprains; definitive treatment of stable syndesmosis and Lisfranc injuries Initial treatment for severe ankle and moderate foot sprains

Excellent protection for all foot injuries and most ankle injuries; effective edema management

Poor rotational control for ankle syndesmosis; rapid deconditioning; inconvenient for dressing, showering, and sleeping Poor rotational control for the ankle syndesmosis; rapid deconditioning; inconvenient for dressing, showering, and sleeping

Short-leg walking cast

Removable cast boot (3D Walker, Bledsoe Boot [Bledsoe Brace Systems, Grand Prairie, Tex], CAM boot)

Initial treatment for moderate ankle and foot sprains

Semirigid pneumatic ankle brace (Air-Stirrup, Aircast, Summit, NJ)

Functional treatment for hindfoot and ankle injuries at various recovery phases, including acute ankle sprains,51,58,63,66,72 prevention of recurrent ankle sprains159,162-164

Nonrigid functional ankle brace (lace-up or Velcro closures)

Functional treatment for hindfoot and ankle injuries at various recovery phases, including chronic injuries

Ankle and foot taping

Functional treatment for foot and ankle injuries at various recovery phases, including chronic injuries

independent rehabilitation, an early return to function, and predictable results (Fig. 25C1-10).63 The semirigid orthosis is lined by opposing dual air cells; the system produces a milking action that actively reduces ankle edema. Several experimental studies have demonstrated the brace’s effectiveness in reducing ankle inversion.23,64,65

A

Excellent protection for most foot and ankle injuries; improved ability to bear weight; continuous rehabilitation provided; self-applied device Removable protection for ankle and foot injuries; improved ability to bear weight; continuous rehabilitation provided; self-applied device Rigid support for hindfoot and ankle injuries—allows ankle range of motion, allows continued athletic participation, facilitates resolution of edema (air cell systems)58; self-applied device; device used within shoe; low cost Nonrigid support for hindfoot and ankle injuries—allows ankle range of motion, allows continued athletic participation; self-applied device; device used within shoe Custom-applied support for foot and ankle injuries; provides resistance to inversion,64 provides biofeedback,70 allows continued athletic participation

No rotational control through ankle; poor edema control; athletic participation restricted No rotational control through ankle; bulky within shoe

No rotational control through ankle

Rapid loosening with time-limited effectiveness159; requires trained personnel for application; high cost over the course of a season158

Early Mobilization Early mobilization, or functional treatment, is the current favored treatment method. Treatment and rehabilitation are directed without the use of rigid cast immobilization. After the acute pain and swelling remit (after about

B

Figure 25C1-10   A, Semirigid pneumatic ankle brace. B, Lace-up ankle brace.

Foot and Ankle 1923

48 hours), weight-bearing to tolerance is encouraged, and a rehabilitation program is instituted. This method specifically avoids immobilization, which many believe simply prolongs the recovery period. A semirigid pneumatic ankle brace, walking boot, or taping is used throughout the rehabilitation period. Konradsen and colleagues used a prospective, randomized study to evaluate the effectiveness of early mobilization compared with total immobilization for complete lateral ankle ligament injuries.51 Early mobilization allowed an earlier return to work and resumption of athletic activity. The 1-year follow-up was similar for both groups. Klein and coworkers used a randomized study to evaluate the same methods.66 Patients treated with a pneumatic ankle brace fared better than those treated with a short leg cast for 6 weeks. Stress radiographs did not significantly differ after treatment. Sommer and Schreiber used a prospective, randomized study to compare plaster cast immobilization (for 3 weeks followed by a pneumatic ankle splint), early ­mobilization with a pneumatic ankle brace, and early ­mobilization with an Unna boot for the treatment of stress radiograph-­documented lateral ankle ligament ruptures.67 They reported lower direct cost and improved stability in the mobilization groups. A recent study to evaluate the effectiveness of early mobilization was published by Povacz and associates in 1998.68 The group performed a randomized prospective study of 146 adults with acute lateral ankle sprains diagnosed by clinical findings and stress radiographs. The treatment was either immediate operative repair followed by 6-week immobilization in a short leg cast or placement of an ankle orthosis for 6 weeks. Nonoperative treatment produced subjective and objective results comparable to those associated with operative treatment. The nonoperative treatment was also associated with a significantly shorter recovery period.

Surgical Repair Surgical management of the acute unstable lateral ankle sprain remains controversial. Several authors have recommended primary surgical repair for severe injury or injury in the high-demand athlete (Fig. 25C1-11).41,69-72 ­Kaikkonen

and colleagues reported excellent and good results at a 6- to 8-year follow-up after primary repair of acute lateral ligament rupture.73 No operative complications occurred. Others have reported favorable, reproducible results with primary repair of acute ruptures.11,41,74 Most physicians currently agree that the role of surgery in the acute setting is very small. Rare circumstances, such as open injuries, large avulsion fractures, and frank dislocation may warrant acute lateral ligament repair. Some authors have reported on adverse outcomes with acute surgical treatment. Freeman compared strapping and early mobilization, immobilization with plaster for 6 weeks, and ligament repair with immobilization for the treatment of complete lateral ankle ligament ruptures.75 The periods of disability were 12, 22, and 26 weeks for patients in the early mobilization, immobilization, and ligament repair groups, respectively. Ligament repair produced the greatest number of complaints at 1-year follow-up, including the only cases with residual loss of motion. Freeman suggested early mobilization as the treatment of choice for lateral ankle ligament ruptures. Broström used a prospective study to compare the use of primary surgical repair followed by 3 weeks of plaster immobilization (95 cases), 3 weeks of plaster immobilization alone (82 cases), and strapping with early mobilization (104 cases).41 Although surgical repair provided excellent results, including a low (3%) residual symptomatic instability rate, Broström went on to suggest that primary surgical repair should not be the routine treatment for acute ruptures. He cited the protracted postoperative recovery, the risk for infection, the risk for painful scar formation, and the success of late lateral ankle ligament ­reconstruction. Evans and coworkers recognized the discrepancy in immobilization periods and performed a prospective, randomized trial with 3 weeks of immobilization in a plaster cast with or without surgical repair of the acute lateral ankle ligament rupture.76 An independent 2-year follow-up concluded that surgical repair yielded similar radiographic results (stress radiographs) and slightly worse functional results. Twice as many patients in the operative group were forced to give up athletic activity. Loss of subtalar inversion and surgical complications were also cited as disadvantages of primary repair of acute injury.

Superficial peroneal nerve

Oblique incision Sural nerve

A

B

C

D

Figure 25C1-11   Surgical technique for primary lateral ankle ligament reconstruction. A, A short oblique incision is made at the anterior margin of the distal fibula. B, Blunt subcutaneous dissection is performed, with care taken to protect the superficial peroneal nerve and the sural nerve. C, The anterior lateral capsule is exposed, and the anterior talofibular ligament, calcaneofibular ligament, and capsular tears identified. D, The tears are approximated and repaired with absorbable suture.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1924 DeLee & Drez’s� O

Repair of acute injury produces results similar to those associated with delayed reconstruction. Because most patients respond to nonsurgical management, primary repair offers few advantages. Today, operative repair of the complete, acute lateral ankle ligament injury is very infrequent.58 Most authors favor nonsurgical treatment for acute ruptures.41,75,77,78

Technique: Primary Lateral Ankle Ligament Repair 1. The procedure is performed with the patient under general anesthesia. The patient is supine with a wellpadded proximal tourniquet and a soft bump placed beneath the ipsilateral hemipelvis. 2. An image intensifier is used to obtain bilateral varus tilt and anterior drawer stress views. Brodén’s stress views are also obtained, if indicated. 3. Arthroscopy of the ankle is performed if clinical presentation or imaging suggests intra-articular disease. 4. The extremity is exsanguinated, and the tourniquet is inflated. A bump is placed under the leg to prevent anterior translation of the talus. 5. A short oblique incision is made at the anterior margin of the distal fibula. This incision incorporates the lateral arthroscopy portal. 6. Blunt subcutaneous dissection is performed, with care taken to protect the superficial peroneal nerve and the sural nerve. 7. The anterior lateral capsule is exposed, and the ATFL, the CFL, and capsular tears are identified. The ankle joint is inspected for chondral or osteochondral fragments. The joint is copiously irrigated. The peroneal sheath is evaluated for tears. 8. The tears are approximated and repaired with absorbable suture. All sutures are placed and then tied from posterior to anterior with the ankle held in neutral dorsiflexion and slight eversion. Avulsion fractures are repaired directly to the fibula or talus. 9. The wound is closed in layers, and a short leg splint is applied. At 10 days, the incision is inspected, and a short leg weight-bearing cast or walking boot is applied. Four to 6 weeks after surgery, the patient is transitioned to an ankle brace. A guided rehabilitation program is instituted and monitored. Patients may actively dorsiflex and evert the foot with inversion limited to neutral. Progressive stretching, strengthening, and proprioception exercises are instituted 6 to 8 weeks after surgery. A semirigid pneumatic ankle brace is used for 6 months after surgery.

Chronic Lateral Ankle Ligament Instability The chronic or recurrent lateral ankle sprain is associated with apprehension, discomfort, swelling, muscular weakness, tenderness, and loss of coordination.30,44 Instability may be overt or subtle and has been described as a “giving way” of the ankle.30 Significant disability is noted, particularly if the patient runs on uneven or loose surfaces. Objective instability of the ankle joint is defined by patient symptoms and positive stress radiographic findings. The condition develops after acute injury in up to 20% of patients.41,77,79

A related condition is functional lateral ankle instability, defined by frequent sprains, difficulty running on uneven surfaces, and difficulty jumping and cutting. Freeman reported this sensation in 21 of 42 patients 1 year after initial lateral ankle ligament rupture.80 Increased varus tilt on stress radiograph, was present in only 6 of these cases. Hansen and coworkers noted a similar lack of agreement between clinical symptoms and persistent talar tilt.81 Brand and colleagues reported a 10% prevalence of functional lateral ankle instability among 1300 Naval Academy freshmen.82 Functional instability may be related to previous ankle sprain, chronic lateral ankle instability, or peroneal weakness.

Rehabilitation without Surgery Successful resolution of chronic lateral ankle instability is possible without surgical stabilization.44,83,84 Patients with chronic, recurrent injuries are treated with a rehabilitation program, and their activity levels are reduced. Functional ankle bracing is continued throughout the treatment period. The key points of rehabilitation are motor strength (particularly of the peroneal muscles), proprioception, and coordination.

Surgical Treatment Patients with chronic injuries that remain symptomatic after a supervised rehabilitation program are candidates for surgical management. Radiographic criteria include talar tilt greater than 15 degrees (or a side-to-side difference of more than 10 degrees) and anterior drawer translation greater than 5 mm (or a side-to-side difference of more than 3 mm). A multitude of procedures have been used for lateral ankle ligament reconstruction, many with reasonable success. In 1932, Nilsonne described a procedure for stabilization of the lateral ankle.85 He repaired a chronic CFL rupture and reinforced the repair with a peroneus brevis tenodesis. Evans86-88 and Watson-Jones84,89 modified the tenodesis. A further modification in the form of an augmentation procedure was proposed by Elmslie90 and refined by Chrisman and Snook.91 The Evans procedure is a transposition of the entire peroneus brevis tendon through the distal fibula (Fig. 25C1-12). Evans recognized the tenodesis resulted in the loss of subtalar inversion after this procedure.87 The Watson-Jones procedure reconstructs the lateral ankle ligaments with a peroneus brevis tenodesis configured to replace the course of the ATFL (Fig. 25C1-13).84,89 The classic method uses the entire peroneus brevis tendon.92 The procedure does not reproduce the anatomic orientation of the CFL. Because the tenodesis of the peroneus brevis lies at a right angle to the subtalar joint empirical axis, loss of subtalar motion is to be expected.10 The Watson-Jones procedure has been modified to spare the function of the peroneus brevis tendon with the use of a split peroneus longus tendon graft,33,93 a complete peroneus longus tendon graft,94 a split peroneus brevis tendon graft,33,95 a plantaris tenodesis,96 a free plantaris tendon graft,34,97 and a split Achilles tenodesis.98,99 Using the split peroneus longus modification of the Watson-Jones procedure, Barbari and associates reported

Foot and Ankle 1925

Peroneus longus

Peroneus brevis Figure 25C1-12   The Evans procedure.

excellent or improved outcomes in 39 of 42 ankles (Fig. 25C1-14).93 Problematic issues include loss of ankle dorsiflexion and subtalar inversion. Brunner and Gaechter completed a retrospective comparison of the split peroneus brevis modification with the free plantaris tendon graft modification of the WatsonJones procedure.34 This study demonstrated slightly more favorable results with the plantaris tendon graft. Increased patient satisfaction and fewer reoperations were noted. The free plantaris graft modification is a more anatomic approach to addressing lateral ligament failure, particularly of the CFL. This procedure preserves the function of the peroneus brevis while sacrificing the vestigial100 plantaris. In 1934, Elmslie described the use of a fascia lata graft passed through drill holes to re-create and augment the ATFL and the CFL.90 The idea of re-creating the ­anatomic

Figure 25C1-13   The Watson-Jones procedure.

Figure 25C1-14   The Barbari modification of the WatsonJones procedure. (Redrawn from Barbari SG, Brevig K, Egge T: Reconstruction of the lateral ligamentous structures of the ankle with a modified Watson-Jones procedure. Foot Ankle 7:362-368, 1987.)

configuration of the ATFL and CFL was adopted by Windfeld.101 He used the entire peroneus brevis to perform a modified Watson-Jones repair. The peroneus brevis tendon was passed through the fibula from front to back and sewn to the remnants of the ATFL at the talus and the CFL at the calcaneus. The procedure limited the graft to local tissue and attempted to reconstruct both lateral ­ligaments. In 1969, Chrisman and Snook modified the Elmslie repair by using a split peroneus brevis tendon, instead of a strip of fascia lata, to reconstruct the ATFL and the CFL (Fig. 25C1-15).91 The 2-year follow-up of seven patients confirmed a successful return to strenuous athletic activity in all participants. Although moderate loss of subtalar motion occurred, it was not problematic. In 1985, the same authors published a long-term follow-up (average followup, 10 years).102 Of the 48 ankles evaluated, 43 had attained excellent and good outcomes. Of the 3 patients with fair and

Figure 25C1-15   The Chrisman-Snook procedure.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1926 DeLee & Drez’s� O

poor results, all had sustained subsequent severe trauma, and one had suffered from generalized ligamentous laxity. Other complications included sural nerve injuries and an asymptomatic loss of inversion. Riegler reported the 2-year follow-up of 11 young athletes after a Chrisman and Snook reconstruction.103 Ten patients returned to their primary athletic activity. All patients lost inversion of the subtalar joint compared with the uninvolved side. Savastano and Lowe published similar results.104 Among 10 ankles treated with a Chrisman and Snook reconstruction, 9 obtained satisfactory results despite limited subtalar inversion. Colville and coworkers performed a biomechanical comparison of the Evans, Watson-Jones, and Chrisman and Snook reconstructions.105 It was determined that each procedure stabilized the ankle but also produced limited subtalar inversion. Colville and Grondel described an anatomic split peroneus brevis lateral ankle reconstruction.106 This procedure reproduces the anatomic orientation of the ATFL and the CFL. The peroneus tendon function appears to remain intact; furthermore, the anatomic approach preserves subtalar function (80% of nonoperated hindfoot inversion). Mechanical stability was verified with stress radiographs in all 17 patients. Minor difficulties with aching and swelling persisted at the 42-month average follow-up. The authors advocate the technique for patients with inadequate local tissue for anatomic reconstruction, generalized hypermobility, and previously failed reconstructive techniques. Paterson and associates described an anatomic reconstruction of the ATFL with a semitendinosus autograft.107 The average 24-month follow-up suggests that the reconstruction is useful for instability not associated with the subtalar joint. Kin-Com dynamometer testing revealed no significant loss of knee flexion strength. Anderson and Lecocq reported a successful delayed repair of the ATFL and CFL with simple plication.13 Broström advocated a similar approach (Fig. 25C1-16).44 His report of successful delayed lateral ankle ligament repair remains significant. Sixty patients with chronic symptoms after lateral ankle sprain were treated with end-to-end ligament repair or advancement of the torn end of the ATFL into the anterior border of the fibula. In 3 of these cases, the repair was fortified with a flap of the lateral talocalcaneal ligament. Average follow-up was 2.9 years. Forty-three patients were completely asymptomatic. Only 3 patients (5%) complained of severe or moderately severe symptoms. Forty-six of the 56 patients who were re-examined were “normal.” Abnormal findings included positive anterior drawer sign (4), impaired mobility (2), tenderness (5), and soft tissue induration (4). Broström’s technique has several advantages over tenodesis and augmentation methods. The technique accomplishes anatomic restoration and maintains joint mobility, it is relatively simple, and it is less morbid with regard to the use of peroneal tendon grafts and length of incision. Other authors have obtained similar good results.108 Gould and associates recommended modification of the Broström technique.109 They removed the intervening scar from the ATFL, repaired the ligament, and supplemented the repair with both a flap from the lateral talocalcaneal ligament and an advancement of the inferior extensor retinaculum (Fig. 25C1-17). Late repair was

A

B Figure 25C1-16   The repair of a chronic lateral ankle ligament rupture as described by Broström. A, Chronic anterior talofibular ligament (ATFL) rupture and direct repair. B, Chronic ATFL rupture with insufficient tissue for simple direct repair and reconstruction using ATFL repair with advancement of the flap of the lateral talocalcaneal ligament into the fibula. (Redrawn from Broström L: Sprained ankles. VI. Surgical treatment of “chronic” ligament ruptures. Acta Chir Scand 132:551-565, 1966.)

­ erformed on 50 patients. The anterior translation of the p talus was reduced to 2 mm or less, and talar tilt to less than 12 degrees. All patients returned to athletic activity. A rating scale based on activity, stability, mobility, swelling, and overall satisfaction revealed that all patients scored between 8 and 10 of 10 possible points. The inferior extensor retinaculum is composed of two layers, superficial and deep to the extensor tendons. Harper concluded that the superficial layer of the inferior extensor retinaculum remains a constant, substantial tissue, suitable for lateral ankle and subtalar joint reconstructions, as proposed by Gould and associates.109,110 Biomechanical testing with cadaveric ankle has proved the efficacy of the Broström repair in restoring ankle stability while maintaining ankle and subtalar range of motion.111 Others have used the Gould modification of the Broström repair with consistent reproducible stabilization and return to athletic activity.112 Harper further modified the Gould modification by using a flap fashioned from the superficial layer of the inferior extensor retinaculum (Fig. 25C1-18).113 When a Broström repair is performed, identification of the torn ligament ends is often difficult, if not impossible. To bypass this difficulty, Karlsson and colleagues used division and imbrication of the attenuated ATFL and CFL.114 They reported good to excellent results in 132 of 153 ankles available for follow-up (mean, 6 years). Reconstruction of both the ATFL and the CFL produced better results than did isolated reconstruction of the ATFL. Most patients with unsatisfactory results were noted to have ­generalized hypermobility, longstanding lateral ankle instability, or a history of previous ankle reconstruction.

Foot and Ankle 1927

Anterior talofibular lig.

Calcaneofibular lig.

A

Lateral malleolus

Lateral malleolus

Lateral talocalcaneal lig. sutured to lat. malleolus

Extensor retinaculum reinforcement

B

Figure 25C1-17   Gould modification of the Broström technique. After repair of the anterior talofibular or calcaneofibular ligament, reinforcements with the lateral talocalcaneal ligament (A) and extensor retinaculum (B) are made.

Karlsson and coworkers performed 60 lateral ankle ligament reconstructions with a modification of the Broström repair.115 These authors advanced the ATFL and the CFL into a 4 × 4-mm bone trough at the anterior aspect of the distal fibula. The remaining periosteal flap was repaired over the ATFL. Good or excellent results were obtained in 53 (88%) of the patients. Unsatisfactory results were associated with generalized ligamentous laxity or longstanding instability. Despite the many surgical options, ankle stabilization is effective and produces good or excellent results in more than 91% of chronic lateral ankle ligament instability cases.116,117 Radiographic stability does not always correlate with clinical outcome. Patients with early signs of osteoarthritis may experience progressive arthrosis despite stabilization. Persistent instability leads to osteoarthritis of the ankle joint.31-33 Ankle arthroscopy at the time of lateral ankle ligament reconstruction provides the surgeon with an opportunity for a more thorough evaluation of the ankle joint. Komenda and Ferkel suggest the use of ankle arthroscopy before lateral ankle ligament stabilization for treatment of loose bodies, osteochondral lesions of the talus, and ankle Anterior talofibular ligament

lnferior extensor retinaculum

A

pain unrelated to instability.118 They also suggest the procedure for evaluation of ATFL integrity and suitability of local tissue for reconstruction. Patients should be carefully evaluated for severe ankle or hindfoot cavovarus. A calcaneal osteotomy may need to be performed in conjunction with lateral ligament repair to prevent failure of the procedure. Patients with failed lateral ankle reconstruction due to re-rupture are placed into rehabilitation and fitted with an orthosis. Repair of the acute injury is a reasonable alternative to functional treatment. Persistent instability can be treated with a repeat Broström repair with the Gould modification, or an anatomic reconstruction with local tissue or a free tendon graft.119 Technique: Lateral Ankle Ligament Reconstruction with a Gould Modification of the Broström Repair

1. The procedure is performed with the patient under general anesthesia. The patient is supine with a wellpadded proximal tourniquet and a soft bump placed beneath the ipsilateral hemipelvis.

Anterior talofibular ligament Calcaneofibular ligament

lnferior extensor retinaculum

Calcaneofibular ligament

B

Figure 25C1-18   Harper modification of the Gould modification of the Broström technique. A flap from the inferior extensor retinaculum is mobilized and sutured to the lateral fibula. (Redrawn from Harper MC: Modification of the Gould modification of the Broström ankle repair. Foot Ankle Int 19:788, 1998.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1928 DeLee & Drez’s� O

2. An image intensifier is used to obtain bilateral varus tilt and anterior drawer stress views. Brodén’s stress views are also obtained, if indicated. 3. Arthroscopy of the ankle is performed if clinical presentation or imaging suggests intra-articular disease. 4. The extremity is exsanguinated, and the tourniquet is inflated. A bump is placed under the leg distally to prevent anterior translation of the talus. 5. A short oblique incision is made at the anterior margin of the distal fibula. The incision incorporates the lateral arthroscopy portal if used. 6. Blunt subcutaneous dissection is performed, with care taken to protect the superficial peroneal nerve and the sural nerve. The extensor retinaculum is identified and tagged. 7. The anterior lateral capsule is exposed, and the ATFL and the CFL are identified. The capsule is incised just distal to the origin of the ATFL and the CFL, and the ankle joint is inspected. 8. The anterior margin of the fibula is exposed by subperiosteal elevation of the proximal capsule. The anterior fibula is decorticated, and three to five tunnels are created with a small K-wire. The capsule with the ATFL and the CFL is secured with multiple nonabsorbable braided sutures. The sutures are passed through the tunnels, and the capsule is advanced onto the decorticated margin of the fibula with the ankle held in dorsiflexion and eversion. The lateral periosteal sleeve and the proximal capsule are repaired over the advanced ligament. This repair can also be performed with suture anchors in the fibula or by soft tissue imbrication. 9. The superficial layer of the extensor retinaculum is secured with suture and advanced to the anterior fibula. This provides significant subtalar stability. 10. The wound is closed in layers, and a short leg splint is applied in slight eversion. At 10 days, the incision is inspected, and a short leg weight-bearing cast is applied. Four weeks after surgery, a rehabilitation program is instituted and monitored. A removable cast boot or a semirigid pneumatic ankle brace is used for an additional 4 to 6 weeks. The repair is further protected during subsequent athletic activity with an ankle brace for 6 months after surgery.

addition to orthotic devices, an elastic sock is available for a­ dditional mobilization of edema (Fig. 25C1-19). In the acute phase, the athlete’s pain and inflammation are addressed with rest, cold therapy, and whirlpool. A trial of electrical stimulation may be considered. Ankle and subtalar joint passive and active range of motion are re-established with inversion limited to neutral for 6 weeks. Isometrics around the ankle and subtalar joints is initiated as pain allows. Weight-bearing to tolerance is ­encouraged. Once the acute pain subsides, flexibility is addressed in all planes. An inclined board is a useful adjunct to gastrocnemiussoleus and Achilles stretching (Fig. 25C1-20). Strengthening is initiated with towel scrunches (Fig. 25C1-21), toe pickup activities, manual resistive inversion and eversion, elastic bands (Fig. 25C1-22), seated toe and ankle dorsiflexion with progression to standing, and seated supination-pronation with progression to standing. Closed chain activities are gradually introduced, including one-leg balance and sport-specific activities on a trampoline, as well as use of the biomechanical ankle platform system (BAPS) (Fig. 25C1-23). Aerobic fitness is maintained with cross-training activities such as water running (Fig. 25C1-24) and cycling. Heat therapy, such as the application of warm packs, is a useful modality before the therapy session. It reduces pain and spasms and thus facilitates increased range of motion. Cold therapy, compression, and elevation are used after each therapy session to reduce inflammation. As the athlete returns to sport, protective bracing, range of motion, and strength activities are continued from the subacute phase. Walking and running activities are allowed to progress within the limits of a pain-free schedule. Once running activity is mastered, a monitored plyometric

Rehabilitation after Surgery Tissue injury initiates a predictable and sequential series of events known as the healing response. The response is typically divided into three phases with arbitrary and overlapping time lines.120 The initial phase is the inflammatory phase, which includes the first through the third day following injury. The second phase is a proliferative phase of tissue repair that extends from day 3 to day 20. The final phase is a remodeling phase that proceeds after day 9. To a certain degree, rehabilitation follows the phases of the healing response in an effort to reduce the undesirable effects of inflammation (i.e., pain, swelling, loss of function) while simultaneously promoting tissue repair and functional recovery. For rehabilitation of the ankle, emphasis is placed throughout the protocol on ankle and subtalar flexibility, motor function, and coordination.30 The ankle is supported by a semirigid pneumatic ankle brace. In

Figure 25C1-19   Elastic sock used for foot and ankle edema mobilization and control.

Foot and Ankle 1929

Osteochondral Lesion

Figure 25C1-20   Gastrocnemius-soleus and Achilles stretching is facilitated with an inclined board.

(­ cutting) program is introduced with progressive difficulty. Schedules are carefully controlled to avoid reinjury during these activities.

Associated Injuries If a lateral ankle ligament injury is suspected either by the reported mechanism of injury (inversion mechanism) or by the initial physical findings (lateral ankle or hindfoot swelling), a multitude of diagnoses must be considered. These may present as isolated findings or as concomitant injuries. Evaluation begins with a detailed history and physical examination with particular attention to the ankle and hindfoot.

Anderson and Lecocq reported a 22% incidence of osteochondral lesions in a mixed series of 27 cases of single and recurrent lateral ankle injuries.13 These lesions were located at the lateral talus in five patients and at the medial talus in one patient. Meyer and associates reported the successful use of CT in the evaluation of the chronically painful ankle after ankle sprain.121 These scans demonstrated intra-articular or juxta-articular bony fragments in 13 of 31 patients. The fragments were located in the anterolateral or lateral aspect of the ankle or subtalar joints in 12 of the 13 patients. Recurrent lateral ankle instability is associated with repetitive shear and compression forces across the ankle articular surface. Taga and colleagues used ankle arthroscopy to evaluate chondral lesions before lateral ankle ligament reconstruction in 31 patients.122 Articular cartilage damage was seen in 89% of the acute ankle injuries and 95% of the chronic ankle injuries. They determined that cartilage damage most frequently was located at the anteromedial tibial articular surface. Furthermore, lesions tended to worsen with regard to depth of injury as the period of ankle instability lengthened. The location and severity of the cartilage damage as seen with the arthroscope correlated with clinical findings. The significance of the treated and untreated chondral lesions associated with a previously unstable ankle remains unknown. Komenda and Ferkel also used ankle arthroscopy to evaluate ankles before lateral ankle ligament reconstruction for chronic instability in 54 patients.118 Intra-articular disease, such as loose bodies, synovitis, and osteophytes, was seen in 93% of the ankles. Osteochondral lesions of the talus or chondromalacia, or both, were found in 25% of patients. Options for treating osteochondral injuries include conservative and surgical approaches. Nonsurgical treatment includes rehabilitation, bracing, oral anti-inflammatory medications, and intra-articular steroid injection. When these treatments fail, surgery is considered. Various techniques, including arthroscopy with débridement, drilling, microfracture, internal fixation, and cartilage transfer, are used. Rehabilitation depends on the procedure performed.

Bone Bruise

Figure 25C1-21   Towel scrunches. The towel is gathered beneath the foot with active toe motion. The activity begins in a seated position and progresses to standing.

With the advent of MRI, subtle injuries to the subchondral bone are easily imaged. Mink and Deutsch described a bone bruise as a traumatic, nonlinear marrow lesion localized to the subchondral bone, typically represented by a T1-weighted signal loss and a T2-weighted signal intensity.123 The natural history of bone bruise is not completely understood and remains controversial. Isolated bone bruise subsequent to knee injury has a predictable and benign course.124 Lahm and coworkers reported no arthroscopic changes (and no clinical sequelae) related to bone bruise associated with various knee injuries.125 Conversely, ­ Johnson and associates identified arthroscopic (e.g., softening, fissuring, and fracture) and histologic (e.g., chondrocyte and osteocyte necrosis) changes that suggested significant damage to the cartilage overlying bone bruise associated with anterior cruciate ligament ­disruption.126

�rthopaedic ����������� S �ports ������ � Medicine ������� 1930 DeLee & Drez’s� O

A

B

Figure 25C1-22   Elastic band exercises. The Thera-Band is posted on a table leg and is used to provide resistance as the foot is exercised. A, Elastic band exercises for resisted inversion. B, Elastic band exercises for resisted eversion.

A

C

B

Figure 25C1-23   Closed chain activities. A, One-leg   balance on a trampoline. B, Sport-specific activity   (throwing) on a trampoline. C, Biomechanical ankle   platform system.

Foot and Ankle 1931

Ant. inferior tibiofibular lig. Distal Fascicle

Figure 25C1-24   Water running for cross-training after foot or ankle injury.

Nishimura and colleagues and Labovitz and Schweitzer suggested that the bone bruise is an indicator of concomitant ligamentous injury to the ankle and that the pattern and location of the lesion correlate with a specific mechanism of injury.127,128 Alanen and coworkers used a prospective study to establish a 27% incidence of bone bruise (i.e., microtrabecular fracture) after ankle inversion injury.129 Ninety-five patients with otherwise normal radiographs were imaged. No clinical significance was related to the occurrence of the bone bruise.

Anterior Lateral Ankle Impingement Chronic anterior lateral ankle pain after an inversion ankle injury is a well-recognized entity. Bassett and associates described a distal fascicle of the anterior-inferior tibiofibular ligament; the structure was found in 10 of 11 cadaveric specimens130 (Fig. 25C1-25). Impingement of the fascicle against the talar dome occurs with ankle dorsiflexion between 9 and 17 degrees. The clinical component of the study identified abrasion of the talar articular cartilage in five of the seven patients. Resection of the fascicle was curative and did not increase ankle instability. A meniscoid lesion of the anterior lateral ankle has also been described (Fig. 25C1-26).131 Hamilton reported finding entrapment of the capsule between the talus and the lateral malleolus during the surgical management of two of three acute, high-grade, lateral ankle sprains.72 He speculated that the capsular interposition might provide the substrate for the classic meniscoid lesion. Chronic anterior lateral ankle pain is treated with an aggressive 6-week course of physical therapy and bracing to eliminate subtle instability. Oral anti-inflammatory and cortisone injection therapy may also be used. Patients who fail conservative management are treated with ankle arthroscopy and débridement of the anterolateral lesion (meniscoid lesion), if present.

Peroneal Tendon Instability Peroneal tendon instability is an entity that may be associated with lateral ankle instability. The condition may be secondary to an inversion ankle injury.13 It may produce

Calcaneofibular lig.

Ant. talofibular lig.

Figure 25C1-25   The distal fascicle of the anterior-inferior tibiofibular ligament is normally parallel and distal to the main ligament and separated from it by a fibrofatty septum. Inset,: After an inversion sprain of the ankle, the distal fascicle may impinge on the anterolateral aspect of the talus. (Redrawn from Bassett FH, Gates HS, Billys JB, et al: Talar impingement by the anteroinferior tibiofibular ligament: A cause of chronic pain in the ankle after inversion sprain. J Bone Joint Surg Am 72:55-59, 1990.)

functional ankle instability caused by the subluxation of the peroneal tendons. Chronic subluxation or frank dislocation of the peroneal tendons may also produce degenerative tears of the peroneus brevis tendon. The peroneus brevis and longus muscles form individual tendons that pass behind the lateral malleolus to turn anteriorly toward their respective insertions at the base of the fifth metatarsal and the base of the first metatarsal. At the level of the lateral malleolus, the tendon of the peroneus brevis remains anterior to the peroneus longus. The tendons are retained within the peroneal groove by the superior peroneal retinaculum (SPR).132 The SPR originates from the periosteum on the posterolateral ridge of the fibula.133 The peroneal groove is a shallow bony groove134 deepened by a fibrocartilaginous ridge. The mechanism of injury for an acute dislocation is related to a sudden, forceful, passive dorsiflexion of the inverted foot combined with reflex contraction of the peroneal tendons.135-137 As has been stated previously, chronic subluxation may also be related to chronic lateral ankle instability.138 The injury produces a variety of pathologic features, which include elevation of the SPR off the lateral border of the fibula with concomitant dissection of the tendons beneath the lateral fibular periosteum, tear of the SPR, or fracture of the posterolateral margin of the fibula. The anatomic classification of peroneal tendon instability, as described by Oden, is based on the location of SPR disruption.139 Chronic subluxation of the peroneus brevis tendon onto the posterolateral border of the fibula has been implicated in the development of longitudinal tears of the peroneus brevis tendon.140,141 Longitudinal tear of the peroneal tendons has also been described after acute and chronic lateral ankle inversion injury.142,143 Acute peroneal tendon dislocation produces pain over the course of the peroneal tendons as well as along the lateral border of the fibula. The patient may recall a pop at the time of injury. Often, the patient is

�rthopaedic ����������� S �ports ������ � Medicine ������� 1932 DeLee & Drez’s� O

A

B

Figure 25C1-26   The “meniscoid lesion” associated with chronic anterior lateral ankle pain after lateral ankle ligament injury.   A, Arthroscopic image of an anterior-lateral meniscoid lesion as seen from the medial portal. The lateral talofibular and talotibial joint space is occupied by the lesion. B, Arthroscopic image of the ankle after complete resection of the meniscoid lesion.

capable of providing a vivid description of dislocation. The tendon may or may not spontaneously reduce. A careful examination of the acute injury confirms swelling and tenderness posterior to the lateral malleolus.135,136,144-146 Active dorsiflexion of the foot from a plantar flexed and everted position may produce apprehension, subluxation, or dislocation (Fig. 25C1-27). Dislocation is not always actively elicited. Lateral ankle ligament stability is also assessed as part of a comprehensive examination. As with all lateral ankle injuries, routine radiographs are obtained. Fracture of the posterolateral margin of the fibula is a rare finding but indicates SPR disruption. MRI is the best imaging modality to evaluate peroneal disease, including tenosynovitis, partial or complete peroneal tendon rupture, peroneal tendon subluxation or dislocation, integrity of the SPR, and the competency of lateral ankle ligaments and internal derangement of the ankle and subtalar joints. A single, acute dislocation of the peroneal tendons is treated with an initial course of immobilization. A short leg cast is applied, and the patient is allowed to bear weight as tolerated. At 6 weeks, the stability of the peroneal tendons

A

is verified. An ankle rehabilitation program is instituted with an emphasis on peroneal strength and proprioception. Casting is aborted if the patient detects peroneal instability within the cast. Patients with recurrent or chronic dislocation do not respond to nonsurgical treatment methods.136,144,146-152 For these patients, surgical reconstruction is required. Surgical treatment includes exploration, tendon repair or tenodesis, peroneal groove deepening, and superior peroneal retinaculum reconstruction, as dictated by surgical findings. Technique: Deepening of the Peroneal Groove and Imbrication of the Superior Peroneal Retinaculum

1. The procedure is performed with the patient under general anesthesia. The patient is supine with a wellpadded proximal tourniquet and a soft bump placed beneath the ipsilateral hemipelvis. 2. A curvilinear incision is made over the course of the peroneal tendons. The initial exposure can be limited to 4 cm, with most of the incision proximal to the tip of the lateral malleolus. Care is taken to protect the sural nerve at the distal aspect of the incision.

B

Figure 25C1-27   A, Reduced peroneal tendons. B, Dislocated peroneal tendons as foot is dorsiflexed and everted.

Foot and Ankle 1933

3. The SPR is exposed and instability of the peroneal tendons verified by manipulation of the foot and ankle. The SPR, identified as a thickening in the sheath, and synovial sheath are sharply divided in line with the posterior border of the lateral malleolus. Exposure is completed by systematic synovectomy and tenolysis of the peroneal tendons. The peroneus brevis muscle that lies within the peroneal groove is resected off of the tendon. 4. Partial longitudinal tears involving less than 50% of the peroneal tendons are débrided and repaired with running sutures. 5. A shallow peroneal groove is deepened through a threesided, medially based osteoperiosteal flap. After a portion of the underlying cancellous bone is removed, the flap is replaced. Alternatively, the flap is resected and the underlying cancellous bone smoothed with bone wax. 6. The SPR is repaired and advanced onto the posterior lateral aspect of the fibula. This is typically accomplished through multiple drill holes using an absorbable suture. 7. The wound is closed in layers and a short leg splint applied in slight plantar flexion and eversion. At 10 days, the incision is inspected and a short leg weight-bearing cast applied for 4 weeks. Patients are then transitioned to a removable boot for 6 more weeks, and a rehabilitation program is instituted. An ankle brace is used for an additional 4 to 6 months. Activities are gradually resumed as the athlete regains full range of motion and strength.

Nerve Palsy The superficial peroneal nerve is susceptible to tension injury after inversion ankle injury. Patients complain of numbness at the dorsal foot or pain of the fascial hiatus radiating distally. Careful physical examination confirms loss of sensation on the dorsal foot. Pain may be reproduced with percussion of the superficial peroneal nerve, particularly at the crural fascia hiatus, or with passive plantar flexion and inversion of the foot. Differential injection of local anesthetic is also used to establish the diagnosis. Treatment is symptomatic but must include conservative management for the ankle instability. Johnston and Howell reported seven cases of superficial peroneal nerve neuropathy associated with inversion ankle injury.153 Five of the patients were also diagnosed with reflex sympathetic dystrophy. Surgical exploration revealed several abnormalities, including a distal exit from the crural fascia, scar, anomalous nerve, and a vessel leash. Nitz and colleagues performed electrodiagnostic testing on 60 consecutive patients with severe ankle sprains.154 The group of patients with medial and lateral ligament injuries (30) included 5 patients with peroneal and 3 patients with posterior tibial injuries. The group of patients with medial, lateral, and syndesmosis ligament injuries (36) included 31 patients with peroneal and 30 patients with posterior tibial injuries. Traction injury to the peroneal and tibial nerves at the bifurcation of the sciatic nerve is one possible mechanism of injury. Electrodiagnostic studies are used for documentation at presentation and follow-up. Sensory deficits do not require intervention. Motor deficits may require bracing with an ankle foot orthosis. Spontaneous recovery is typical.

Subtalar Sprain Meyer and coworkers performed ankle stress radiographs and subtalar joint arthrography on 40 patients with acute inversion sprains.155 They classified the injuries based on the extent of lateral ankle ligament and subtalar ligament injury. Thirty-two (80%) of the patients sustained injury to both the lateral ankle and the subtalar joint. Six patients with negative stress ankle radiographs had positive subtalar arthrograms. Subtalar joint sprain and subtalar instability are discussed in detail in the following foot section.

Subtalar Coalition Recurrent ankle sprain is not an uncommon presentation for the 10- to 14-year-old athlete with tarsal coalition. Lateral hindfoot pain, recurrent sprains, flatfoot, and decreased subtalar motion all point to talocalcaneal or calcaneonavicular coalition. The entity is diagnosed with oblique radiographs, CT, or MRI. Treatment is symptomatic initially with orthoses. Persistent symptoms require excision or occasionally hindfoot arthrodesis.

Bifurcate Ligament Sprain Injury to the bifurcate ligament or the anterior process of the calcaneus is associated with a plantar flexion–inversion mechanism (see Figs. 25C1-1, 25C1-48, and 25C1-49). The injury is often associated with, or mistaken for, a concomitant lateral ankle sprain. Diagnosis is clinical, but MRI confirmation may be obtained. Treatment is symptomatic with foot and ankle rehabilitation and is occasionally protracted owing to persistent symptoms. This injury is also reviewed in great detail in the foot section.

Tibiofibular Synostosis Heterotopic ossification of the interosseous ligament occasionally is noted after disruption of the syndesmosis (Fig. 25C1-28). Whiteside and associates identified six patients, all professional athletes, with tibiofibular synostosis and a history of inversion and internal rotation injuries.156 The authors’ experiences suggest excision for cases of persistent pain and limited dorsiflexion. Recurrence occurred in two of three cases. In my experience, the presence of ossification of the interosseous ligament or complete synostosis does not appear to be associated with a poor outcome (among patients with syndesmosis injury, not isolated lateral ankle injury). Certainly, a poorly reduced syndesmosis with an associated synostosis presents a need for excision of the synostosis and anatomic reduction.

Fifth Metatarsal Base Fracture The base of the fifth metatarsal is also susceptible to injury after a plantar flexion–inversion mechanism. The associated lateral hindfoot swelling and tenderness are distal and inferior to the ATFL. The diagnosis is confirmed with

�rthopaedic ����������� S �ports ������ � Medicine ������� 1934 DeLee & Drez’s� O

Chronic lateral ankle instability produces a variable amount of impairment. The treating physician should ­emphasize a comprehensive and well-supervised rehabilitation program to prevent recurrent instability. Until ankle stability is achieved, a nonrigid or semirigid ankle brace is used to support the ankle. If significant improvement is not demonstrated within 6 weeks, I use a Broström procedure as modified by Gould to reconstruct the ATFL and the CFL. Intraoperative imaging is used to obtain bilateral stress views of the ankles to document instability. Arthroscopy of the ankle is used before stabilization, if indicated by clinical examination (e.g., anterolateral tenderness, ­talar dome tenderness) or MRI findings (e.g., osteochondral ­lesion of the talus, loose body).

Return-to-Play Criteria

Figure 25C1-28   Ossification of the syndesmosis in a former collegiate football player.

routine foot radiographs. Treatment is generally supportive with immobilization in a cast, walking boot, or stiffsoled shoe for 4 to 6 weeks.

Author’s Preferred Method Injury to the lateral ankle requires an individualized a­ pproach. Treatment must balance the needs of the athlete with the available professional support and facilities. High school, collegiate, and professional athletes are treated ­according to an accelerated schedule. These athletes are fortunate enough to have professional trainers for daily monitoring and treatment. Trainers and their staff are also able to provide ongoing ankle taping for practice and ­competition. Acute grade I and grade II sprains are treated with a ­ supervised physical therapy program. Medical anti­inflammatory therapy and cold therapy are used to reduce swelling and pain. Cast immobilization is avoided. A semirigid pneumatic ankle brace is used throughout the rehabilitation period and well into the return-to-sports phase. High-grade (grade III) lateral ankle sprains are treated with a brief 3- to 7-day period of weight-bearing to tolerance in a removable boot or a cast. The most comprehensive approach avoids casting and focuses on a supervised physical therapy program, including modalities (e.g., cold therapy, ­ compression, electrical stimulation) for swelling and pain control. A semirigid pneumatic ankle brace is used throughout the rehabilitation period and well into the return-to-sports phase, typically 4 to 6 months.

Recovery of function after a lateral ankle sprain follows a logical sequence of events. Once the initial pain and swelling subside, coordination and strengthening activities are emphasized. Gradually, the patient is able to return to walking, running, and cutting programs. I return patients to sport once they master sport-specific drills with minimal discomfort. A semirigid pneumatic ankle brace or taping is used to prevent recurrent injury. Prevention of recurrence and new injuries has received much attention in the literature. Key components of prevention include strength conditioning, coordination, proprioception, stretching, and external support. Taping of the ankle is perhaps the most widely used prophylactic method.157 Rovere and colleagues published a retrospective analysis of taping, and the laced ankle stabilizer that confirmed that taping was much less effective in the prevention of new and recurrent ankle injuries among collegiate football players.158 Taping appears to be limited by time-related loosening.159,160 Glick and coworkers studied the effect of tape on six ankles with significant (>5 degrees) talar tilt.7 Each subject performed 20 minutes of exercise. Only one of the six ankles remained firmly supported by the ankle tape. Walsh and Blackburn acknowledge the time-related loosening associated with ankle taping.161 They contend that the method continues to play a supportive role, perhaps limited to restricting the extremes of motion. They recommend taping against the skin and secondarily around the shoe (football). They emphasize that tape, in any amount or configuration, is not a substitute for rehabilitation. Although taping is inexpensive on a per-use basis, significant cost is associated with long-term application by trained personnel.158 Hamill and colleagues suggested the use of a semirigid pneumatic orthosis for the prevention of recurrent lateral ankle sprains; they cite a reduction in mediolateral excursion force and velocity in an experimental setting.162 Decreased inversion, based on angular displacement, has also been demonstrated.163 Soccer players with and without a history of lateral ankle sprains were randomly assigned by Surve and associates to use a semirigid pneumatic ankle brace (Sport Stirrup, Aircast, Inc., Summit, NJ).164 The orthosis reduced

Foot and Ankle 1935

the ­incidence of lateral ankle sprains in the group with a history of previous ankle injury but not in the group without such a history. Thacker and colleagues completed a review of 113 studies and concluded that supervised rehabilitation must be completed before resumption of running or practice.8 Furthermore, these authors recommended the use of an orthosis for 6 months after a severe lateral ankle sprain. A sobering study completed by Gerber and associates clearly illustrates the long-term disability associated with ankle sprain.3 This prospective observational study of 104 West Point cadets confirmed that 95% of the cadets had returned to athletic activity by the 6-week follow-up. At the time of the 6-month follow-up, all cadets had returned to athletic activity, but 40% remained symptomatic. The persistent dysfunction was related neither to the grade of sprain nor the presence of joint laxity. In summary, athletes return to sport after recovery of pain-free ankle motion, strength, and protective reflexes. The athlete must complete sport-specific drills with a high degree of confidence and comfort. Most important, the ankle is braced until functional and anatomic stability are achieved.

Medial Ankle Sprain Isolated injury to the deltoid ligament is rare. Staples reviewed 110 cases of deltoid injury.165 Deltoid rupture without associated ankle fracture was noted in only 10 cases. Of these 10 cases, 5 were isolated to the deltoid, 3 were associated with syndesmosis injury, and 2 were ­associated with anterior capsule injury.

Relevant Anatomy The deltoid ligament consists of superficial and deep components (Fig. 25C1-29).166,167 The superficial deltoid ligament originates from the anterior portion of the medial malleolus and spreads out to insert on the navicular, talus, and calcaneus. The superficial deltoid includes four parts—the tibionavicular ligament, the tibiospring ligament, the tibiocalcaneal ligament, and the superficial tibiotalar ligament. The deep deltoid ligament includes two parts—the deep anterior tibiotalar ligament and the deep posterior tibiotalar ligament. It is a short, thick ligament that traverses from the intercollicular groove onto the medial talus and blends with the medial capsule of the ankle joint. The biomechanical characteristics of the ankle ligaments are such that failure (rupture) is due to increasing load as opposed to twisting or shearing.15 Isolated testing of the individual ankle ligaments demonstrates that the ATFL is the first to fail and the deep deltoid ligament is the last to fail.15 The deltoid ligament functions to limit abduction. It is a strong structure that requires significant force to cause rupture. Siegler and colleagues tested 20 fresh cadaveric ankles.18 Based on increasing ultimate load, components of the deltoid were ordered from weakest to strongest as follows: the tibiocalcaneal ligament, the tibionavicular ligament, the tibiospring ligament, and the posterior tibiotalar ligament.

Superficial portion deltoid ligament

Deep portion deltoid ligament

Spring ligament Sustentaculum tali Figure 25C1-29   The superficial and deep layers of the deltoid ligament. (Redrawn from Close JR: Some applications of the functional anatomy of the ankle joint. J Bone Joint Surg Am 38:761781, 1956.)

Close has demonstrated the importance of the medial ligaments in maintaining a normal medial clear space, that is, a normal intermalleolar distance.166 He provided a detailed anatomic description but no data to support his conclusions. Earll and coworkers conducted a cadaveric study to assess the importance of the deltoid ligament relative to talocrural contact and pressure.168 Division of the tibiocalcaneal fibers of the superficial deltoid resulted in significant decreased contact area (maximum, 43%) and an associated increase in contact pressure (maximum, 30%). Division of the other components of the deltoid resulted in insignificant changes in joint contact. Kjærsgaard-Andersen and associates used a cadaveric model to study the effect of isolated division of the tibiocalcaneal ligament.169 They reported a maximal median increase in tibiotalocalcaneal abduction of 6.1 degrees and a corresponding maximal median increase in talocalcaneal abduction of 3.6 degrees. The authors concluded that the tibiocalcaneal ligament is an important stabilizer of the medial hindfoot. Rasmussen and colleagues used a cadaveric model to study the effect of isolated division of the deltoid ligament with a device that allowed recording of rotatory movements in two planes.170 The tibiocalcaneal ligament (superficial deltoid) and the intermediate tibiotalar ligament (deep deltoid) provided resistance to abduction of the talus. The ­deltoid also provided significant resistance to external ­rotation of the talus.

Clinical Evaluation Isolated deltoid rupture is a rare injury, usually associated with a traumatic mechanism of injury. Most deltoid ruptures are associated with ankle fractures. Ankle fractures associated with a pure deltoid rupture typically involve the posterior deep tibiotalar ligament.171 Evaluation of the medial ankle sprain is designed to elicit information that allows classification of the injury. Various classifications have been described, but graded ligament injury, as suggested by the American Medical Association35 and by O’Donoghue,36 is sufficient. Injuries

�rthopaedic ����������� S �ports ������ � Medicine ������� 1936 DeLee & Drez’s� O

are graded based on stretch (grade I), partial tear (grade II), or complete rupture of the ligament (grade III). Additional information with regard to associated ligamentous injuries is noted.

History Information relevant to previous ankle injury, the mechanism of injury, the ability of the patient to continue to play or walk, and current complaints represent the salient historical points. The patient frequently reports feeling a pop in the medial ankle with associated pain and swelling.

Incisura fibularis

1 cm

Physical Examination It is imperative that the examination not be limited to the medial ankle ligaments. Inspection of the leg, ankle, and foot may reveal swelling, ecchymosis, blister formation, or gross deformity. A vascular and sensory assessment is performed, followed by palpation of the entire leg, ankle, and foot. Tenderness at the deltoid and surrounding area is particularly important to note. Attention is paid to symptoms in other areas that could indicate ankle fracture, lateral ligament sprain, syndesmotic injury, combined proximal fibular fracture (Maisonneuve pattern), or proximal tibiofibular joint injury. Motion about the foot and ankle is determined with the patient seated and relaxed. The knees are flexed and the feet allowed to fall into an equinus position. Ankle and subtalar range of motion is documented and motor function graded. Stress testing is a useful clinical and radiographic tool that provides a portion of the diagnostic data required for grading ankle sprains. The anterior drawer test and the varus talar tilt test are used to demonstrate the integrity of the ATFL and the CFL. A valgus talar tilt test is used to evaluate the integrity of the deltoid. The test is performed in both ankle neutral and plantar flexion positions. Valgus stress is applied to the talus through the hindfoot, and a comparison is made between injured and noninjured ankles. Testing is occasionally uncomfortable, particularly in the acute setting. False-negative results may be caused by involuntary guarding or pain response. The ankle can be injected with local anesthetic to facilitate a better ­examination. The chronic or recurrent medial ankle sprain is associated with functional or mechanical medial instability, apprehension, discomfort, swelling, and tenderness over the deltoid. This clinically manifests as a valgus and pronation deformity that many patients can correct by contracting the posterior tibialis muscle.

Imaging Radiographs Radiographs in the anteroposterior, mortise, and lateral projections are required for ankle evaluation. Weightbearing radiographs are used in an effort to reproduce physiologic loading.

0.5 cm Medial joint space width

Syndesmosis width Displacement of distal fibula

Figure 25C1-30   Techniques of measuring the lateral displacement of the lateral malleolus (mortise view) and the width of the syndesmosis (mortise view) and medial joint space (anteroposterior view). (Redrawn from Harper MC: The deltoid ligament: An evaluation of need for surgical repair. Clin Orthop 226:156-168, 1988.)

Radiographs are evaluated with regard to malleolar fracture, physeal fracture, osteochondral fracture, avulsion fracture, and deltoid ossification. Alignment and translation deformity are also inspected, particularly at the syndesmosis and the medial ankle joint space (Figs. 25C1-30 and 25C1-31). An increased medial clear space suggests complete rupture of both components of the deltoid.166 Radiographs are frequently normal with partial ligament injuries.

Stress Radiographs Valgus talocrural instability may occur after deltoid disruption.172 The bilateral stress radiograph is used to quantify valgus tilt of the talus. The valgus tilt stress radiograph is similar to the clinical test but is performed during an anteroposterior radiograph. The degree of tilt is determined by measuring the angular divergence between the distal tibial articular surface and the talar dome. Leith and associates performed valgus stress radiography on 32 previously uninjured patients.173 This examination was performed on both ankles in a neutral position with and without valgus stress. The authors demonstrated that 91% of the ankles had less than 2 degrees and that the remaining 9% of the ankles had between 2 and 3 degrees of valgus tilt. They suggest that an ankle with a valgus tilt greater than 2 degrees has a high probability of deltoid injury. The external rotation stress radiograph is also used to evaluate associated syndesmosis injury. This test is described in the section on “Ankle Syndesmosis Sprain.”

Foot and Ankle 1937 Anteroposterior View

A

B

B

C

A = Lateral border of posterior tibial malleolus B = Medial border of fibula C = Lateral border anterior tibial tubercle

B

A B

C

B

A Syndesmosis A (<6 mm)

Syndesmosis B (>6 mm or 42% of fibular width)

Figure 25C1-31   Syndesmotic radiographic criteria. A, The syndesmosis clear space as depicted on the anteroposterior view and by coronal section. The tibiofibular clear space is the distance between the lateral border of the posterior tibial malleolus (point A)   and the medial border of the fibula (point B) on the anteroposterior radiograph. This space is normally less than 6 mm.187 B, The syndesmosis overlap as seen on the anteroposterior view and by coronal section. The tibiofibular overlap is the distance between the medial border of the fibula (point B) and the lateral border of the anterior tibial prominence (point C) on the anteroposterior radiograph. This space is normally greater than 6 mm, or 42% of the fibular width.187 (Redrawn from Stiehl JB: Complex ankle fracture dislocations with syndesmosis diastasis. Orthop Rev 14:499-507, 1990.)

Arthrography The unpredictable effect of patient guarding during stress radiography is overcome by the use of ankle arthrography.52,53 This method is relatively simple and can provide objective evidence of deltoid disruption. Contrast material is injected into the acutely injured ankle, preferably with fluoroscopic guidance, and radiographs are obtained in various projections with attention to areas of extravasation. A false-negative result from an early capsular seal is possible. The procedure has been largely replaced by the use of MRI.

Magnetic Resonance Imaging MRI is a useful method for evaluation of acute, subacute, and chronic ankle ligament injuries.55 Deltoid disruption, partial or complete, is demonstrated by fiber disruption, edema, and associated injuries to the surrounding soft tissue and bone (see Fig. 25C1-9). Visualization of the deltoid ligament requires careful attention to the position of the foot during the test. On coronal images, the tibionavicular and anterior tibiotalar components are best seen with the foot in plantar flexion, whereas the tibiocalcaneal and posterior tibiotalar portions are visualized with the foot in dorsiflexion.174

Therapeutic Options The treatment of this injury depends on the severity of disruption and associated injuries. All acute injuries are treated with the RICE method followed by gentle range of motion and protected weight-bearing. Many studies have reviewed the effect of cold application to the injured extremity. Cold therapy is an effective, inexpensive, and easy-to-use modality for the treatment of

acute musculoskeletal injury. Appropriately applied cold therapy decreases pain perception, decreases the biochemical reactions that produce inflammation, and produces vasoconstriction with a concomitant reduction in soft tissue swelling and bleeding. Compression is typically provided in the form of an elastic bandage but may also include casting, splinting, pneumatic orthosis, or mechanical compression devices.57 Elevation of the ankle helps to reduce swelling and pain. The ankle is supported by a variety of methods, including the use of a nonwalking cast, a walking cast, a removable cast boot, a semirigid pneumatic ankle brace, a nonrigid functional ankle brace, or various ankle taping methods. Various immobilization methods are summarized in Table 25C1-4.

Grade I Medial Ankle Sprains Treatment is symptomatic with an emphasis on recovery of range of motion, strength, and coordination. A structured ankle rehabilitation program remains an important part of treatment. A nonrigid functional ankle brace, such as a lace-up brace, or a semirigid pneumatic ankle brace is used during all phases of the recovery. Return to competition is usually delayed when compared with lateral ankle sprains.

Grade II and Grade III Medial Ankle Sprain Complete immobilization is provided with a short leg walking cast or boot. Weight-bearing to tolerance is encouraged. The assumption is that immobilization allows the ruptured ends of the medial ankle ligaments to heal in a near-anatomic position. Immobilization is ­continued for 6 to 8 weeks depending on resolution of swelling,

�rthopaedic ����������� S �ports ������ � Medicine ������� 1938 DeLee & Drez’s� O

t­ enderness, and instability. A comprehensive ankle rehabilitation program is used along with a semirigid pneumatic orthosis for up to 6 months from the date of injury. Surgical treatment is rarely necessary and is reserved for patients with a persistent valgus tilt or in whom the medial clear space is not reduced.165,175 Technique: Repair of Medial Ankle Ligaments

Author’s Preferred Method of Treatment Completely isolated deltoid injuries are rare. ­Occasionally, an athlete presents with a high-grade lesion with ­predominantly medial symptoms. In this instance, I prefer to use a short leg cast or walking boot for 4 to 6 weeks in an ­ effort to promote anatomic healing of the deltoid fibers. A ­ comprehensive, supervised physical therapy ­program then follows. An ankle brace is used throughout the ­rehabilitation period and well into the return to sports phase. Medical anti-inflammatory therapy is provided to reduce swelling and pain. Chronic medial ankle instability is rare. The clinical and radiographic work-up must be meticulously done before a medial reconstruction is undertaken. Partial tears and ­ossicles associated with the anterior colliculus can be a common source of dysfunction. Isolated removal and local repair can provide relief.

1. A longitudinal incision is created over the medial malleolus and extended to the talonavicular joint. 2. Both the superficial and deep components of the deltoid ligament are inspected as are the adjacent posterior tibial and flexor digitorum tendons. 3. The ankle joint is inspected for articular surface lesions and loose bodies. This can be done arthroscopically before the open procedure if desired. 4. Anatomic repair of the deep fibers is completed before repair of the superficial fibers. Nonabsorbable sutures are used. In cases of complete avulsion from the medial malleolus, a suture anchor can be used. 5. The incision is closed in a layered fashion, and a splint is applied. At 10 days, the patient is placed into a short leg cast. Patients are not allowed to bear weight for 4 weeks after surgery. At that point, the patient is transitioned into a walking boot and allowed to progress to full weight-bearing over 4 weeks. An ankle rehabilitation program with emphasis on range of motion, strengthening, and proprioception is started. The athlete continues to wear an ankle brace during sports for 6 months after surgery.

Recovery follows a logical sequence of events. Once the initial pain and swelling subside, coordination and strengthening activities are emphasized. Gradually, the patient is able to return to walking, running, and cutting programs. Patients are returned to sport once they master sport­specific drills. A semirigid pneumatic ankle brace or taping accelerates the schedule.

Chronic Medial Ankle Sprain and Instability

Ankle Syndesmosis Sprain

Occasionally, a medial ankle sprain produces a chronic deltoid insufficiency or medial ankle pain. Difficulty with push-off may also be noted. Careful review of the initial treatment program may suggest the need for an aggressive and well-supervised rehabilitation program along with a semirigid pneumatic orthosis. If it is believed that the rehabilitation was adequate, further imaging with stress radiographs and MRI is obtained to better delineate the condition. An ossicle within the anterior deltoid may produce medial symptoms without clinical instability. Symptomatic anterior deltoid ossicles are surgically débrided, followed by postdébridement valgus stress radiographs. Sympto­ matic medial ligament insufficiency is most easily treated with local repair and imbrication of the deltoid. Advancement of the deltoid to the medial malleolus may be facilitated through drill holes or with suture anchors. If the local tissues are inadequate for a direct repair, a reconstruction can be performed. Autograft or allograft material is placed from the tibia to the talus or navicular. This technique can be combined with spring ligament repair or reconstruction if that ligament is attenuated as well.

Injury to the ankle syndesmosis, or a high ankle sprain, is most common in collision sports.176 Injury to the syndesmosis results in more impairment to the athlete when compared with lateral ankle sprains. Boytim and colleagues reported 98 ankle injuries among the players of a professional football team over a 6-year period.176 Twenty-eight significant lateral ankle sprains and 15 syndesmosis sprains were reported. The players with syndesmosis sprains missed more games and more practices and used more treatments than players with lateral ankle sprains. Hopkinson and coworkers retrospectively reviewed 1344 ankle sprains that occurred over a 41-month period at the U.S. Military Academy.177 Fifteen of these patients (1.1%) were diagnosed with syndesmosis sprain. A subsequent prospective study at the same institution revealed that syndesmosis sprains accounted for 17% of ankle sprains over a 2-month period.3

Rehabilitation A detailed rehabilitation program is outlined in the section on lateral ankle ligament injury. Ankle and subtalar flexibility, motor function, and coordination30 are emphasized throughout the protocol.

Return-to-Play Criteria

Relevant Anatomy The interosseus membrane connects the tibia to the fibula. At the level of the ankle, three defined ligaments are present: the anterior-inferior tibiofibular ligament (AITFL), the posterior-inferior tibiofibular ligament (PITFL), and the interosseous ligament (IOL) (Fig. 25C1-32). The AITFL courses from the anterodistal fibula to the anterolateral (Tillaux-Chaput) tubercle of the tibia. The AITFL is the most commonly injured ligament in syndesmotic sprains and can result in symptomatic impingement in some cases.

Foot and Ankle 1939

IOL

IOL

Syndesmotic ligaments AITFL ATFL

PITFL

DL

PTFL DL

A

Syndesmotic ligaments

CFL

B

Figure 25C1-32   A, The tibiofibular syndesmosis from the front—anterior-inferior tibiofibular ligament (AITFL), anterior talofibular ligament (ATFL), and deltoid ligament (DL). B, The tibiofibular syndesmosis from the back—posterior inferior tibiofibular ligament (PITFL), distal interosseous ligament (IOL), posterior talofibular ligament (PTFL), calcaneofibular ligament (CFL), deltoid ligament (DL).

The PITFL is composed of the deep transverse t­ ibiofibular ligament and a superficial portion. The two components form a strong yet elastic structure, which ­usually fails last in syndesmotic injury. The IOL connects the tibia to the fibula about 0.5 to 2.0 cm above the plafond. Proximally, it continues as the interosseus membrane, which provides little additional strength to the syndesmotic ligaments. The syndesmosis, along with the deltoid ligament, maintains the critical anatomic relationship between the tibia and the talus. Ramsey and Hamilton used a carbon black transference technique to clearly demonstrate that lateral displacement of the talus results in an incremental decrease in contact area with each millimeter of translation.178 The first millimeter of lateral translation produced an average 42% reduction in contact area between the tibia and the talus. Failure to reduce the ankle syndesmosis and the associated lateral talar translation greatly increases the risk for post-traumatic ankle arthritis (Fig. 25C1-33). Anatomic dissection by Close revealed that division of the syndesmosis and of the interosseous ligament produces minimal widening of the intermalleolar distance.166 Only after the deltoid ligaments are divided does the syndesmosis separate. His conclusion is that significant trauma to the ankle must occur for the ankle mortise to appear wide. Broström described the ligamentous lesions found during the surgical exploration of 105 recent ankle sprains.27 The ATFL was the most commonly injured structure. The ATFL was completely torn as an isolated injury in 65 cases and as an associated injury in an additional 25 patients. The CFL was the second most commonly injured ligament. It was completely or partially torn as an associated injury in 23 patients. The AITFL was completely torn in 6 cases. No incomplete ruptures were noted. Five of these injuries occurred at midsubstance, and 1 involved an avulsion fracture off the anterior tibia. The torn ends of the ligament remained well apposed. External rotation of the talus produced up to 5 mm of diastasis in 5 cases and 1 to 2 mm in the sixth case. The mechanism of injury is thought to be external rotation, although researchers have been unable to ­ reliably

reproduce syndesmosis sprain without fracture. An ­external rotation force first disrupts the AITFL, followed by the IOL, but usually spares the PITFL. Increased force can lead to spiral fractures of the proximal fibula (Maisonneuve fracture). In 1968, Lovell described the case of a 13-year-old tobogganner who sustained a forced external rotation injury to the ankle.179 The patient presented with a fixed external rotation foot deformity that was associated with a posterior dislocation of the fibula perched behind the lateral tibia. Closed reduction and casting produced an excellent result. Boytim and associates suggested two specific mechanisms of injury in the professional football player.176 The first is direct force applied to the posterior leg of a downed player whose foot is in an externally rotated position. The

Figure 25C1-33   Post-traumatic ankle arthritis subsequent to an incompletely reduced ankle syndesmosis and lateral talar translation. Note the wide syndesmosis and medial joint space width, the narrow superior joint space, and the lateral subchondral cyst formation.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1940 DeLee & Drez’s� O

second is an external rotation force at the knee while the foot is firmly planted. Fritschy evaluated 10 world-class slalom skiers with syndesmosis injuries.180 He speculated that a common mechanism of forced external rotation of the talus against the fibula produced all of their injuries. A retrospective review by Hopkinson and colleagues failed to establish a consistent mechanism of injury among athletes of different sports with syndesmosis sprain.177

Clinical Evaluation Sprains can be considered from the simplistic perspective of graded ligament injury, as suggested by the American Medical Association35 and by O’Donoghue.36 Injuries are graded based on stretch (grade I), partial tear (grade II), or complete rupture (grade III) of the AITFL. Additional information with regard to associated ligamentous or bony injuries is noted. Edwards and DeLee described a classification system for ankle diastasis without fracture (grade III sprain) based on the presence of radiographic diastasis with and without stress.181 A latent syndesmosis injury appeared normal on an unstressed radiograph and abnormal or widened on external rotation stress mortise radiograph. A frank syndesmotic injury was seen as a widened syndesmosis on unstressed radiographs. The frank injuries were further divided into four types: type I, lateral fibular subluxation without plastic deformity of the fibula; type II, lateral fibular subluxation with plastic deformity of the fibula; type III, posterior subluxation of the fibula behind the lateral tibia; and type IV, superior dislocation of the talus between the tibia and fibula with diastasis and no fibula fracture.

to a former chief athletic trainer at the U.S. Military Academy. The test is performed by compression of the midleg from posterior lateral to anterior medial. Pain produced at the AITFL suggests injury to the same, as long as fracture, contusion, and compartment syndrome are not ­ present. The authors retrospectively reviewed eight patients with syndesmosis sprains; all were noted to have a positive squeeze test at initial evaluation. A separate biomechanical analysis of the squeeze test demonstrated that the squeeze test produced reproducible separation of the fibula and tibia.183 Boytim and associates described an external rotation stress test (Fig. 25C1-35).176 The patient is seated and relaxed with the hip and knee flexed and the foot and ankle held in a neutral position. The knee is maintained in a forward-facing position while a gentle but firm external rotation force is applied to the foot. Pain reproduced at the anterior syndesmosis is diagnostic of a syndesmosis injury. A secondary test is the direct eversion maneuver (Fig. 25C1-36).184 The maneuver is accomplished with the patient in a seated and relaxed position. The examiner gently secures the leg and foot as a direct eversion or abduction force is applied across the ankle. Increased translation compared with the contralateral ankle is a positive result.

Imaging Radiographs The radiographic examination includes weight-bearing views of the ankle, orthogonal views of the leg, and, when indicated, a computed tomographic scan of the syndesmosis (see Figs. 25C1-30 and 25C1-31).185,186

History Information relevant to previous ankle injury, the mechanism of injury, the ability of the patient to continue to play or walk, and current complaints represent the salient historical points. Syndesmosis injury is suggested by a mechanism of forced external rotation of the foot.

Physical Examination The examination is systematic and includes careful palpation along the entire interosseous ligament and the fibula. Fracture of the fibula at all levels must be considered. Although dislocation of the proximal tibiofibular joint is rare,182 it must also be considered when proximal leg symptoms are present. Local tenderness at the AITFL or along the interosseous ligament suggests a syndesmosis sprain. There may be tenderness over the deltoid ligament, usually with an associated abduction force at the time of injury. Ankle range of motion is carefully assessed. Also, lateral ankle stability is determined with performance of the anterior drawer and talar tilt tests. One must never forget that pain out of proportion to the injury is a finding consistent with acute compartment syndrome. Hopkinson and coworkers described a squeeze test used to identify syndesmosis ruptures at the time of initial presentation (Fig. 25C1-34).177 The squeeze test was ­attributed

Figure 25C1-34   The squeeze test.177 Syndesmosis injury is suspected when compression of the midleg produces pain at the ankle syndesmosis.

Foot and Ankle 1941

B

A

Figure 25C1-35   The external rotation stress test of the syndesmosis. (A, Redrawn from Boytim MJ, Fischer DA, Neumann L: Syndesmotic ankle sprains. Am J Sports Med 19:294-298, 1991.)

Harper and Keller used 12 cadaveric legs to establish radiographic criteria for a normal syndesmosis.187 Plastic spacers with 1-mm increments were used to produce the diastases. With the use of standard anteroposterior and mortise radiographs, it was determined that the “clear space” on either view is normally less than 6 mm. This was in fact the most reliable parameter for evaluating the integrity of the syndesmosis. The normal overlap between the fibula and the anterior process of the tibia

A

is greater than 6 mm, or 42% of the width of the fibula on the anteroposterior view, or greater than 1 mm on the ­mortise view. Ostrum and colleagues used a more specific approach to the question of radiographic diastasis.188 The authors included 40 female and 40 male normal volunteers to establish normal values. Their first finding was a distinct difference between the sexes. To bypass the sex variation, they suggested sexspecific absolute values or non–sex-specific ratios.

B

Figure 25C1-36   A, The direct eversion maneuver is accomplished with the patient in a seated and relaxed position. The examiner gently secures the leg and foot as a direct eversion or abduction force is applied across the ankle. Increased translation compared with the contralateral ankle is a positive result. B, A positive stress radiograph showing increased translation. (A, Redrawn from Stiehl JB: Complex ankle fracture dislocations with syndesmotic diastasis. Orthop Rev 19:499-507, 1990.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 1942 DeLee & Drez’s� O

Late radiographs after a syndesmosis rupture may reveal varying degrees of interosseous ligament calcification. Hopkinson and associates retrospectively reviewed radiographs of 10 patients with syndesmosis sprains, as diagnosed by a positive squeeze test at initial evaluation.177 Radiographs for nine of the patients demonstrated heterotopic calcification. At an average of 20 months after injury, all ankles were asymptomatic. Taylor and colleagues reported finding heterotopic ossification on 11 of 22 follow-up radiographs after syndesmosis sprain sustained during football (diagnosed by tenderness at the syndesmosis).189 No player developed frank synostosis. The lower incidence of heterotopic ossification, as compared with the Hopkinson report, may be related to the method of diagnosis (squeeze test versus local tenderness). Patients with heterotopic ossification experienced a delayed recovery (i.e., 32 days without heterotopic ossification, 43 days with heterotopic ossification). The authors suggested a correlation between severity of injury and the formation of heterotopic ossification. Patients with heterotopic ossification also were more likely to experience recurrent inversion ankle sprains.

Stress Radiographs When routine radiographs are normal and there is a concern for a latent syndesmotic injury, stress radiographs are obtained. Radiographs are taken with application of an external rotation and abduction force through the ankle. Xenos and coworkers performed an experimental study with 25 cadaveric ankles.190 Each ankle was tested under a constant external rotation torque. After sequential division of the syndesmosis and subsequent repair, the authors concluded that the lateral external rotation stress radiograph is superior to the mortise stress radiograph for the detection of syndesmosis injury.

acute syndesmosis injury. Among 27 patients, the imaging was 100% sensitive and 71% specific.191

Computed Tomographic Scan Computed tomographic scans are able to obtain images in the axial, sagittal, and coronal planes. These images can be reconstructed to create a detailed three-dimensional representation of the relationship of the tibia to the fibula. Ebraheim and coworkers demonstrated the superiority of CT over plan radiographs in detecting syndesmotic disruption.192 They used a cadaveric model and plastic spacers to demonstrate the inability of routine radiographs and CT to identify 1-mm diastases at the syndesmosis. Computed tomographic scans identified 2- and 3-mm diastases. Routine radiographs failed to identify 2-mm and 50% of the 3-mm diastases.

Magnetic Resonance Imaging MRI is a useful method for evaluation of syndesmosis injuries. Its accuracy, lack of ionizing radiation, noninvasiveness, decreasing cost, and increasing availability suggest that MRI is the imaging modality of choice for the ankle joint. Ligament discontinuity, wavy ligament contour, and the inability to image the ligament are all findings consistent with a syndesmosis injury. The high sensitivity associated with MRI mandates that images be carefully correlated with clinical findings (Fig. 25C1-37A).

Therapeutic Options All acute injuries are treated with the RICE method and non–weight-bearing until definitive diagnosis is established.

Isolated Grade I and Grade II Syndesmosis Sprains

Before the development of MRI, arthrography was commonly used to document injury to the syndesmosis. After injection of contrast material into the ankle, a positive test showed dye leakage more than 1 cm superior to the plafond. Broström and associates performed arthrography of 321 fresh ankle sprains.54 Extra-articular leakage occurred in 239 cases (74%). Surgical exploration was performed in 99 of these cases and in an additional 6 cases that did not demonstrate leakage. The authors concluded that arthrography was useful within the first 7 days of an acute ankle sprain. Leakage of contrast into the peroneal tendon sheath correlated with tear of the CFL. Leakage in front of the lateral malleolus correlated with ATFL rupture. Leakage in front of the syndesmosis correlated with complete rupture of the syndesmosis. Leakage at the medial malleolus correlated with partial deltoid rupture.

After the acute pain and swelling remit (within 72 hours), weight-bearing to tolerance is encouraged, and a rehabilitation program is instituted. Treatment is symptomatic with an emphasis on recovery of range of motion, strength, and coordination. A semirigid pneumatic ankle brace or taping is used throughout the rehabilitation period. Taping is applied with an effort to restrict or reduce external rotation. While facing the athlete, the trainer applies the tape from lateral to medial. Gerber and coworkers prospectively evaluated 96 West Point cadets with acute ankle sprains.3 Sixteen of these injuries were primarily syndesmosis sprains treated with early mobilization. Regardless of grade of injury, patients with syndesmosis sprains were most likely to experience an unacceptable outcome at 6 months’ follow-up. The authors speculated that lack of accurate evaluation, underestimation of extent of injury, and incomplete rehabilitation may account for the poor results among patients with syndesmosis sprains.

Nuclear Imaging

Isolated Grade III Syndesmosis Injuries

Radionuclide imaging of the acute syndesmosis injury without fracture is not a common practice. Marymont and colleagues used radionuclide imaging 1 to 2 weeks after

Treatment of the complete syndesmosis disruption is based on displacement and stability. Latent injuries as described by Edwards and DeLee181 may be treated with a

Arthrography

Foot and Ankle 1943

A

B

Figure 25C1-37   A, Magnetic resonance image demonstrating complete tear of AITFL and partial tear of PITFL. B, Open treatment of syndesmotic rupture.

protracted course of non–weight-bearing in a short leg cast or a removable cast boot. Non–weight-bearing status is maintained for 6 to 10 weeks depending on the resolution of local tenderness and pain with provocative maneuvers. Nonoperative treatment is recommended only for isolated, stable injuries. If the injury is associated with medial bony or ligamentous injury, internal fixation is suggested.

A

B

Frank injuries with a displaced or widened syndesmosis require closed reduction and internal fixation or open reduction and internal fixation (Fig. 25C1-38). The adequacy of the reduction is based on radiographs of the noninjured ankle. Based on the evaluation of 34 ankle fractures, Leeds and Ehrlich concluded that the poorly reduced ­tibiofibular diastasis predisposed the patient to a poor ­ outcome and

C

Figure 25C1-38   Syndesmosis rupture. A, Anteroposterior injury radiograph. B, Internal fixation with fully threaded 4.5-mm screw. C, Anteroposterior radiograph after screw removal.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1944 DeLee & Drez’s� O

osteoarthrosis.193 Reduction of the syndesmosis to within 2 mm of the contralateral side correlated with good subjective and objective results at an average follow-up of 4 years. Published recommendations for syndesmosis screw removal vary from as early as 6 weeks194 to nonremoval. Most authors maintain screws for a minimum of 12 weeks after surgery. Harper reported syndesmosis failure following removal of syndesmosis screw fixation 6 and 8 weeks after insertion.195 The syndesmotic screw eventually loosens and does allow at least some motion at the distal tibiofibular joint.195-197 Grath noted in his comprehensive treatise that the removal or nonremoval of a syndesmotic screw does not produce detrimental effects.198 Thordarson and associates acknowledged the disadvantages of permanent screw fixation to include prominent painful hardware, disruption of normal biomechanical relationships at the syndesmosis, screw fracture, need for a second operative procedure if removal is selected, stress shielding of bone, and interference with MRI and CT.199 To obviate the need for permanent hardware, the group tested 4.5-mm polylactide (PLA) screws against 4.5-mm stainless steel screws in a cadaveric model. They concluded that the PLA screws were of sufficient strength to maintain fixation and allow healing of the syndesmosis.

to prevent over-reduction of the mortise and subsequent loss of ankle dorsiflexion (see Fig. 25C1-38B). 7. The wound is closed in layers and a short leg splint applied. At 10 days, the incision is inspected and the patient is transitioned to a short leg cast or removable boot. Patients are prevented from bearing weight for a minimum of 4 weeks. After 4 weeks, a rehabilitation program is instituted and monitored. Partial ­weightbearing is allowed after 4 to 6 weeks and progressed to full by 8 to 12 weeks. The fixation is removed 3 months after surgery, before resumption of ­athletic activity is permitted. The patient continues for 6 months with an ankle brace during rehabilitation and sports activities.

Unusual Syndesmosis Injuries

Surgical Repair

Olerud reported a single case of posterior dislocation of the fibula (Edwards and DeLee type III) and subluxated talus associated with a violent supination and external rotation mechanism of injury.201 This patient was successfully treated with primary open reduction and syndesmotic screw placement. Edwards and DeLee used a fibular osteotomy to reduce lateral displacement of the fibula associated with plastic deformity of the fibula (Edwards and DeLee type II).181 The procedure was successfully used on two patients with this unusual injury.

Technique: Closed or Open Reduction and Internal Fixation of the Ankle Syndesmosis (see Fig 25C1-37B)

Chronic Syndesmosis Sprains

1. The procedure is performed with the patient under general anesthesia. The patient is supine with a wellpadded proximal tourniquet and a soft bump placed beneath the ipsilateral hemipelvis. 2. An image intensifier is used to attempt closed reduction of the syndesmosis. If reduction is not satisfactory, an open reduction is performed. 3. The extremity is exsanguinated and the tourniquet inflated. A linear incision is placed over the tibiofibular space about 2 to 3 cm above the plafond. Care is taken to protect the lateral branch of the superficial peroneal nerve as it courses over the syndesmosis. 4. The syndesmosis is exposed, and tearing of the AITFL is noted. The tear is approximated and repaired with absorbable suture. Avulsion fractures are repaired directly to the fibula or tibia. Open reduction is accomplished by débridement of the distal tibiofibular articulation. Reduction is performed and held with a large forceps on the tibia and fibula. 5. If reduction of the medial joint space remains incomplete, a medial ankle arthrotomy and a deltoid ligament repair are performed. Sutures are placed but are not tied until the syndesmosis is satisfactorily reduced and stabilized. 6. The syndesmosis repair is then protected by placement of one or two 3.5- or 4.5-mm screws. These screws are placed 2 cm200 above the ankle joint line with the ankle in a neutral dorsiflexion position. The fixation is directed from the relatively posterior fibula into the more anterior tibia. Three or four cortices are captured with the screw. A nonlag technique is used in an effort

Little literature exists on how to optimally treat chronic syndesmosis injuries. Reconstruction requires scar débridement from the syndesmosis and ankle joint. Reduction is held with large clamps while a tendon graft is placed through drill holes in the fibula and tibia to recreate the AITFL. Trans-syndesmotic fixation is placed to protect the reconstruction, and the medial deltoid ligament is repaired or reconstructed. Some authors have advocated arthroscopic treatment of chronic syndesmotic injuries.202 Patients who underwent resection of the torn portion of the interosseus ligament and chondroplasty reported improvement in postoperative pain, swelling, stiffness, and activity level. The authors also reported normalization of the external rotation stress test. Although the authors did not perform transsyndesmotic fixation, it can be done in conjunction with this procedure to stabilize the syndesmosis. Katznelson and coworkers used a distal tibiofibular arthrodesis to treat chronic syndesmosis ruptures.203 All five ruptures were the sequelae of injuries initially diagnosed as lateral ankle ligament sprains. All five patients obtained excellent results at the time of final follow-up. The report did not discuss return to athletic activity, nor was the time to follow-up noted.

Rehabilitation Treatment of patients with low-grade syndesmosis sprains is ­symptomatic, with bracing and a rehabilitation program as outlined previously. Rehabilitation for ­ postoperative syndesmosis injuries is effectively delayed by 4 to 6 weeks.

Foot and Ankle 1945

Non–weight-bearing status is maintained for 6 to 8 weeks, followed by ­ progression to full weight-bearing by 12 weeks. Ankle and subtalar flexibility, motor function, and coordination30 are emphasized throughout the protocol. The ankle is supported with an ankle brace during later rehabilitation and return to sports for 6 months postoperatively. In addition to the ankle brace, an elastic sock is available for mobilization of edema (see Fig. 25C1-19).

Author’s Preferred Method of Treatment I treat acute syndesmosis injury not associated with fracture with the RICE method and non–weight-bearing until the definitive diagnosis is established. Isolated grade I or II syndesmosis injuries are allowed to begin weight-­bearing to tolerance after the acute pain and swelling remit. Treatment is symptomatic with an emphasis on recovery of range of motion, strength, and coordination. Patients frequently benefit from the support of a walking boot shortly after injury. They are later transitioned into a supportive ankle brace. I use the Edwards and DeLee classification system for grade III sprains.181 This system is based on the presence of radiographic diastasis with and without stress. A latent syndesmosis injury appears normal on an unstressed radiograph and abnormal or widened on external rotation stress mortise radiograph. A frank injury is seen as a widened syndesmosis on unstressed radiographs. Patients with latent injuries do not need surgery if the reduction of the fibula is anatomic on CT or MRI. Patients are not allowed to bear weight for at least 4 to 6 weeks. Sequential radiographs are used to monitor alignment. Patients are transitioned to a walking boot and are allowed to progressively start bearing weight after 6 weeks. Frank injuries require closed or open reduction and screw fixation. Rehabilitation is similar to that described for stable latent injuries. Screw fixation is removed 3 to 4 months postoperatively before resumption of full athletic activity is permitted.

Return-to-Play Criteria Recovery follows a logical sequence of events. Once the initial pain and swelling subside, coordination and strengthening activities are emphasized. Gradually, the patient is able to return to walking, running, and cutting programs. Patients are returned to sport once they master sport-specific drills. A semirigid pneumatic ankle brace is maintained during this time. Fritschy evaluated 10 world-class slalom skiers with syndesmosis injuries.180 The unexpected finding was that all skiers returned to their original level of competition, but the time to return ranged between 18 months and 12 years. Gerber and colleagues completed a prospective observational study of 96 West Point cadets with ankle sprains, including 16 syndesmosis sprains.3 All patients were treated with a functional rehabilitation program. At the time of the 6-week and 6-month follow-up examinations, grade I syndesmosis sprains were associated with worse outcomes

compared with all ankle sprains, including grade II and III syndesmosis sprains.

Ankle Dislocation without Fracture Dislocation of the ankle joint is typically associated with major or minor bony injuries. Dislocation without associated fracture is rare.

Relevant Anatomy The ankle joint ligamentous support is described in detail earlier. The stability of the ankle joint is such that ankle dislocation without associated malleolar fracture is ­uncommon. Wilson and coworkers reviewed the literature and found 14 cases of ankle dislocation without fracture.204 Most of these injuries were associated with falls or direct trauma. The authors further described two cases. The first was a posterior dislocation of the ankle and the second an upward dislocation of the ankle associated with a wide diastasis of the distal tibiofibular joint.

Clinical Evaluation History Dislocation of the ankle joint with or without fracture produces dramatic pain and deformity. The patient may report spontaneous reduction, or reduction on the field may be noted by the patient or a trainer. Reduction produces significant relief of pain. Information relevant to previous ankle and subtalar injury, the mechanism of injury, and current complaints represent the salient historical points.

Physical Examination Examination of the patient includes evaluation of the entire extremity. Inspection of the leg, ankle, and foot may reveal swelling, ecchymosis, blister formation, or gross deformity. A vascular and sensory assessment is performed, followed by a palpation of the entire leg, ankle, and foot.

Imaging Radiographs Radiographs of a dislocated ankle are typically dramatic (Fig. 25C1-39). Three standard views are reviewed, and a lack of talocrural continuity is diagnostic for an ankle dislocation. Radiographs taken after a formal reduction or after spontaneous reduction are less dramatic but nonetheless must be obtained and carefully scrutinized for malleolar fractures, talar fractures, and osteochondral fractures (Fig. 25C1-40). Subsequent radiographs are obtained to ensure anatomic reduction and to monitor heterotopic ­ossification.

Magnetic Resonance Imaging MRI is useful for delineation of ligamentous injury after ankle dislocation. The ligament remnants are occasionally imaged and may suggest continued nonoperative management. Osteochondral injury is also assessed.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1946 DeLee & Drez’s� O

with primary open reduction and syndesmotic screw placement.

Rehabilitation The rehabilitation phase begins as soon as the ankle is deemed stable. The program focuses on reducing swelling and re-establishing range of motion. After completing rehabilitation, the athlete is gradually returned to progressive activity. An ankle brace is used to provide additional support and biofeedback.

Author’s Preferred Method of Treatment Figure 25C1-39   Ankle dislocation without fracture.

Therapeutic Options This injury is usually identified on the field. Reduction brings about tremendous pain relief and reduces ongoing swelling and damage to nerves, vessels, and articular cartilage. Once the patient presents to the hospital, a complete examination is performed before reduction. General anesthesia is recommended, but early reduction with intravenous sedation may be possible.204 If anatomic reduction is not possible, open reduction is required. If at the time of reduction, the ankle remains unstable, internal or external fixation is required. A tibiofibular diastasis associated with a syndesmosis rupture must also be reduced and stabilized (see previous section). Olerud reported a single case of a posterior dislocation of the fibula and subluxated talus in association with a violent supination and external rotation mechanism of injury.201 This patient was successfully treated

A

After radiographic documentation is obtained, a closed reduction is performed with the patient under conscious sedation or general anesthesia. The stability of the reduction is ideally simultaneously determined. Immobilization and complete radiographic studies, including CT or MRI, are obtained after reduction. The length of immobilization is determined by postreduction stability. Ideally, a non–weight-bearing cast is used for 4 weeks. This is followed by application of a removable cast boot for an ­additional 4 to 6 weeks of weight-bearing to tolerance and ­initial rehabilitation. A semirigid pneumatic orthosis is used for up to 6 months after the injury. Emphasis is placed on ­ reducing swelling and improving range of ­ motion, strength, and proprioception.

Return-to-Play Criteria The patient must be prepared for an unpredictable return to competitive activity. Many pitfalls await the patient. The most significant is persistent stiffness. Return to play is allowed only after range of motion has been re-­established and local tenderness and swelling have resolved.

B

Figure 25C1-40   A, Ankle dislocation with fracture at the tip of the lateral malleolus. B, Ankle dislocation with fracture at the medial tuberosity of the talus (different case).

Foot and Ankle 1947

c

d

b

a

e

Talus c

d b

a

ATFL

CFL

Calcaneus

A

CL

IOL

Figure 25C1-41   Anterolateral ligamentous structures of the subtalar joint—anterior talofibular ligament (ATFL), calcaneofibular ligament (CFL), cervical ligament (CL), and interosseous talocalcaneal ligament (IOL). (Redrawn from Meyer JM, Garcia J, Hoffmeyer P, Fritschy D: The subtalar sprain: A roentgenographic study. Clin Orthop 226:169-173, 1988.)

INJURY TO THE FOOT LIGAMENTS In the following section, subtalar sprain, subtalar dislocation, bifurcate sprain, and Lisfranc sprain are reviewed. First ­metatarsophalangeal joint sprain, also called turf toe, is reviewed in a separate section within this chapter (see Chapter 25H). The incidence of injury to the foot ligaments is certainly lower than that to the ankle ligaments. The cause of injury is reviewed in great detail in a separate section within this chapter (see Chapter 25J).

Subtalar Sprain Recently, more attention has been directed at the subtalar joint as a source of pathology. Subtalar joint injury varies from a mild subtalar sprain to a complete subtalar and talonavicular dislocation without fracture. Subtalar joint sprain is most commonly associated with a lateral ankle sprain. The injury less frequently presents as an isolated entity that produces persistent pain and instability after inversion injury to the foot and ankle.205

Relevant Anatomy and Biomechanics A review of hindfoot bony anatomy and function facilitates understanding of the ligamentous anatomy in this region. The talus articulates with the calcaneus through the subtalar joint, which is composed of anterior, middle, and posterior facets. The most important of these articulations is the posterior facet.206 The oblique empirical joint axis of the posterior facet converts rotatory movement from the leg to the foot. Internal rotation of the leg produces (by means of the oblique empirical axis) eversion of the ­ calcaneus,

B

Figure 25C1-42   A, Ligaments of the sinus tarsi: a, lateral retinacular root; b, intermediate retinacular root; c, medial retinacular root; d, cervical ligament; e, interosseous ligament. B, Calcaneal attachments of the ligaments of the sinus tarsi. (Redrawn from Harper MC: Lateral ligamentous support of the subtalar joint. Foot Ankle 11:354-358, 1991.)

which, in turn, places the foot in a supple configuration (pronation). External rotation of the leg produces the opposite effect of calcaneal inversion and increased foot rigidity (supination). Failure of the hindfoot to move through this natural motion reduces the effectiveness of the foot as a mechanical shock absorber and as a rigid lever for ­propulsion. Subtalar motion is a complex three-plane motion that includes motion in the sagittal, frontal, and axial planes.207-209 The clinical range of motion as described by Sarrafian includes 25 to 30 degrees of inversion and 5 to 10 degrees of eversion.210 The determination of clinical range of motion is imperfect at best owing to the obliquity of the joint, its association with other joints, and the soft tissues surrounding the joint. Pearce and Buckley found a threefold overestimation of the clinical range of motion compared with the motion determined by CT.211 Stability of the subtalar joint is accomplished through bony configuration and ligamentous orientation. Bony stability is greatest with the calcaneus everted, a position that allows the greatest degree of posterior facet contact and congruity.210 Lateral subtalar joint stability is provided by a group of lateral ligamentous structures. Harper reviewed the ­anatomic literature and compiled data from 10 cadaveric ­dissections.110 Based on Harper’s work, the lateral ligaments of the subtalar joint are divided into three layers— ­superficial, intermediate, and deep (Figs. 25C1-41 and 25C1-42; Table 25C1-5). The inferior extensor retinaculum is composed of two layers—superficial and deep to the extensor tendons. Harper concluded that the ­ superficial layer of the inferior extensor retinaculum remains a constant, substantial tissue, suitable for lateral ankle and subtalar joint reconstructions, as proposed by Gould and ­associates.109 Magnusson highlighted the importance of the talocalcaneal interosseous ligament, the ankle ligaments, and the

�rthopaedic ����������� S �ports ������ � Medicine ������� 1948 DeLee & Drez’s� O

 TABLE 25C1-5  Lateral Ligamentous Support of the Sinus Tarsi

Superficial layer Intermediate layer Deep layer

Lateral root of the inferior extensor retinaculum Lateral talocalcaneal ligament Calcaneofibular ligament Intermediate root of the inferior extensor retinaculum Cervical ligament Medial root of the inferior extensor retinaculum Interosseous talocalcaneal ligament

5 mm

From Harper MC: Lateral ligamentous support of the subtalar joint. Foot Ankle 11:354-358, 1991.

A medial and lateral malleoli for the restriction of extreme supination and pronation.212 Smith, after empirically determining the axis of the subtalar joint, demonstrated that the cervical ligament limits inversion and the interosseous ligament limits eversion of the calcaneus across the subtalar joint.213 In 1968, Laurin and colleagues completed anatomic studies that demonstrated the importance of the CFL in maintaining subtalar stability.23 This group sequentially divided the ATFL and the CFL (along with the lateral talocalcaneal ligament) before taking stress radiographs. Isolated division of the ATFL produced primarily talocrural instability, and isolated division of the CFL produced primarily talocalcaneal instability (lateral joint opening). The group further speculated that the mechanism of subtalar joint sprain (isolated CFL injury) is forced inversion of the foot below a dorsiflexed ankle. Chrisman and Snook also used a cadaveric model to demonstrate that subtalar instability occurs after section of both the CFL and part of the lateral talocalcaneal ­complex.91 Kjærsgaard-Andersen and coworkers, also using a cadaveric model, obtained measurements continuously through the ankle range of motion with a constant adduction force across the tibiotalocalcaneal joint complex.214 Isolated division of the CFL resulted in increased hindfoot adduction through the talocalcaneal joint, as opposed to the talocrural joint. This incremental difference was maximal at 5 degrees of ankle dorsiflexion. The clinical application, as suggested by the authors, is to place the ankle in slight dorsiflexion during stress testing of the ankle and subtalar joints. A second cadaveric study by Kjærsgaard-Andersen and associates revealed that the CFL provides significant rotatory stability to the talocalcaneal joint.215 External rotation after isolated division of the CFL increased up to 5.4 degrees at the tibiotalocalcaneal complex and up to 2.9 degrees at the talocalcaneal complex. The authors concluded that the CFL is the primary restraint to hindfoot external rotation. Heilman and colleagues used 10 fresh cadaveric ankles to demonstrate that the CFL tightens with ­supination and dorsiflexion.216 Selective division of the CFL produced 5 degrees of lateral opening across the posterior facet of the subtalar joint. Division of the lateral subtalar capsule added no further instability to

B

C

Figure 25C1-43   Stability of the subtalar joint after serial ligament sectioning. A, Brodén’s view of the subtalar joint, intact ligaments. B, Brodén’s view of the subtalar joint, after sectioning of the calcaneofibular ligament (CFL). C, Brodén’s view of the subtalar joint, after sectioning of the CFL, capsule, and interosseous ligament. (Redrawn from Heilmen AE, Braly G, Bishop JO, et al: An anatomic study of subtalar instability. Foot Ankle 10:224-228, 1990.)

the joint. Finally, division of the interosseous ligament completely destabilized the joint, leading to dislocation (Fig. 25C1-43). A third experimental study by Kjærsgaard-Andersen and coworkers demonstrated that isolated division of the cervi­ cal or interosseous ligament resulted in relatively minor increases in three-plane joint motion.208 These authors concluded that the resulting instability was significant despite the small angular changes. Furthermore, injury to either ligament may be related to the sinus tarsi syndrome or talocalcaneal instability. Knudson and associates used a cadaveric model to specifically study the effect of interosseous ligament division.209 Measurements taken before and after interosseous ligament sectioning suggested that the interosseous ligament provides significant subtalar joint support, particularly in supination. The exact cause of subtalar instability remains unsolved. Several factors, including local bony and ligamentous anatomy, contribute to this entity. Treating this condition requires an understanding of the anatomy and biomechanics of this region.

Clinical Evaluation As has been stated previously, sprains are frequently evaluated by graded ligament injury, as suggested by the American Medical Association35 and by O’Donoghue.36 Acute injuries are graded based on stretch (grade I), partial tear (grade II), or complete rupture (grade III) of the subtalar capsule or supporting ligaments, including the CFL, the interosseus ligament, and the cervical ligament. A grade III tear is suggested by a clinical history of severe deformity or swelling, examination consistent with gross subtalar instability, or MRI demonstrating ligamentous disruption. The clinical evaluation provides additional information with regard to associated ligamentous injuries, especially of the lateral ankle.

Foot and Ankle 1949

History History alone is usually insufficient to distinguish between subtalar and lateral ankle instability. Information relevant to previous ankle and subtalar injury, the mechanism of injury, the ability of the patient to continue to play or walk, and current complaints are important to obtain. Severe subtalar sprains are associated with a history of inversion injury with a characteristic pop; acute pain, swelling, or deformity; and the inability to continue activity. After the acute event, patients may report recurrent instability with walking, running, and sports. They frequently report difficulty on uneven surfaces. Lateral pain in the region of the sinus tarsi may be present.

135° 45°

45°

A

B

Figure 25C1-44   Clinical examination of the subtalar joint is facilitated by placement of the empirical subtalar joint axis horizontal with the ground. A, Place the foot in 45 degrees of equinus. B, Examine the patient prone with the knee flexed 135 degrees. (Redrawn from Inman VT: The Joints of the Ankle. Baltimore, Williams & Wilkins, 1976, p 108.)

Physical Examination The differential diagnosis of an inversion foot or ankle injury includes ATFL sprain, CFL sprain, syndesmosis sprain, deltoid sprain, subtalar sprain, subtalar coalition, bifurcate ligament sprain, peroneal tendon instability, peroneal tendon tear, lateral malleolus fracture, talar dome osteochondral injury, anterior calcaneus process fracture, and fracture of the base of the fifth metatarsal. Chronic insufficiency of the lateral hindfoot associated with subtalar instability isolated or combined with lateral ankle ligament instability has been shown to occur in up to two thirds of patients.34 Examination of the patient includes evaluation of the entire extremity. Inspection of the leg, ankle, and foot may reveal swelling, ecchymosis, blister formation, or gross deformity. A vascular and sensory assessment is documented, followed by a palpation of the entire leg, ankle, and foot. Tenderness at the sinus tarsi, the ATFL, and the CFL is particularly important to note. Clinical examination of the subtalar joint is difficult to complete owing to the complex nature of subtalar motion and its association with leg, ankle, and foot motion.217 Evaluation of motion is determined by gently grasping the leg while the ankle is held at a right angle or in a neutral dorsiflexion position. With the widest part of the talus engaged into the ankle mortise, adduction of the heel is more likely to represent motion through the talocalcaneal joint rather than the talocrural joint. From a neutral position (heel vertical), maximal passive inversion and eversion are measured. An alternative method is placement of the empirical subtalar joint axis horizontal with the ground by allowing the foot to fall into 45 degrees of equinus, or by examining the patient prone with the knee flexed 135 degrees (Fig. 25C1-44).10 Gross clinical instability is consistent with a grade III sprain.

Imaging Radiographs Radiographic demonstration of the subtalar joint is difficult.218 Brodén described an oblique view of the foot designed to produce tangential images of the posterior facet of the subtalar joint (Fig. 25C1-45).219 The view is obtained with the foot in 45 degrees of internal rotation and the beam centered on the sinus tarsi and angled

posteriorly (by 10, 20, 30, or 40 degrees). The images are carefully inspected with attention to small fractures, nonconcentric joint alignment, loose body, and arthrosis.

Stress Radiographs Several methods for stress radiography of the subtalar joint have been described. Stress radiography of the ankle joint to quantify the extent of concomitant ankle instability is recommended and described previously. The ankle talar tilt test is dependent on the contralateral ankle for control measurements. Varus tilt of the ankle, to a limited degree, is probably normal. Increased inversion of the calcaneus may represent talocrural (ankle) or talocalcaneal (subtalar) instability.34 Radiographic data suggest that 4 degrees of varus tilt occurs in 10% to 15% of noninjured ankles.47 Varus stress radiographs of both ankles in 90 injured and 90 normal ankles revealed that 6 degrees of increased tilt represents the transition from “normal to abnormal” tilt.48 The subtalar varus tilt test is performed with the ankle in a neutral position.214 A Brodén view of the posterior facet

BRODÉN I PROJECTION FOOT POSITION limb internally rotated 45° ankle dorsiflexed 90°

CENTRAL RAY centered 2-3 cm distal and anterior to lateral malleolus four pictures, 10°, 20°, 30°, and 40° off the perpendicular Figure 25C1-45   Brodén’s oblique view of the foot designed to produce tangential images of the posterior facet of the subtalar joint. The view is obtained with the foot in 45 degrees of internal rotation and the beam centered on the sinus tarsi and angled posteriorly (10, 20, 30, or 40 degrees).

�rthopaedic ����������� S �ports ������ � Medicine ������� 1950 DeLee & Drez’s� O

of the subtalar joint is obtained as a varus stress is applied to the subtalar joint. Angular divergence or lateral opening at the posterior facet is compared with the contralateral foot for quantification of instability. A relatively new method is the forced manual ­dorsiflexion-supination stress lateral radiograph described by Ishii and colleagues.220 The true lateral ankle radiograph is used to establish the position of the lateral talar process relative to the posterior facet of the calcaneus. Clinical and cadaveric experiments indicate that this method is useful for detecting subtalar instability. Stress radiographs can be performed in the office with a mini C-arm or in the operating room. A 40-degree Brodén view is obtained with inversion stress applied to both the calcaneus and the fifth metatarsal head laterally while the medial distal tibia is stabilized. Heilman and associates described a 5-mm separation between the talus and the calcaneus as indicative of subtalar instability.221 The reliability of the stress Brodén view has been challenged by ­ other authors. Harper completed a review of ankle and subtalar stress radiographs in 14 injured extremities, the contralateral noninjured extremities, and 18 additional asymptomatic extremities.222 Lateral opening of the subtalar joint, as seen on the Brodén stress view, did not significantly differ between the injured and the uninjured contralateral extremity. Furthermore, results from the asymptomatic extremities revealed subtalar articular divergence between 0 and 20 degrees, with an average of 9 degrees. Similar findings using fluoroscopic methods have also been reported.223

Arthrography Subtalar arthrography for evaluation of acute injury to the CFL or the interosseous ligament remains a useful technique. Meyer and coworkers performed ankle stress radiographs and subtalar joint arthrography in 40 patients with acute inversion sprains.155 They classified the injuries based on the extent of lateral ankle ligament and subtalar ligament injury. Thirty-two (80%) of the patients sustained injury to both the lateral ankle and the subtalar joint. Six patients with negative stress ankle radiographs had positive subtalar arthrograms. Sugimoto and associates recommend using an image intensifier to obtain anteroposterior, lateral, and 45-degree oblique views of the subtalar joint.224 Extravasation of contrast into the peroneal sheath or ankle joint suggests CFL rupture; leakage into the sinus tarsi suggests interosseous ligament or anterior capsule of the posterior facet joint rupture. The authors also noted that this method is not limited by the lack of availability and the cost associated with MRI.

Computed Tomography The inability of the stress Brodén view to provide useful screening for subtalar instability was confirmed by van Hellemondt and colleagues.225 Using helical CT, the authors demonstrated no significant difference in subtalar tilt between a group of patients with suspected subtalar instability and the contralateral asymptomatic extremities. The authors suspect that translation produces the tilt seen on routine stress radiographs, as opposed to true divergence of the talus and calcaneus. It is interesting to note

that CT did isolate four cases of fibrous middle facet coalition and a large calcaneal cyst. Finely cut CT is also useful in the evaluation of high-grade sprains to rule out associated fractures.

Magnetic Resonance Imaging MRI is useful for the diagnosis of acute and chronic ankle and subtalar ligamentous injury28 and their sequelae (see Fig. 25C1-9). With proper imaging, MRI can identify injuries to the cervical and interosseus ligaments. High sensitivity mandates that images be carefully correlated with clinical findings. MRI is also useful for identification of tarsal coalition.

Therapeutic Options All acute injuries are treated with the RICE method, as detailed earlier. Patients with high-grade injuries are immobilized and protected from bearing weight until the acute symptoms resolve.

Grade I and Grade II Subtalar Sprains Low-grade sprains are treated with early mobilization. Patients are allowed to bear weight to tolerance, and a rehabilitation program is instituted. Treatment is symptomatic, with an emphasis on recovery of foot and ankle range of motion, strength, and proprioception. A removable fracture boot is used during the initial recovery period, followed by use of a semirigid pneumatic ankle brace.

Grade III Subtalar Sprains High-grade injuries, as determined by clinical findings, are treated with 3 weeks of immobilization in a short leg cast. Patients may bear weight as tolerated. Once clinical stability is achieved, a comprehensive foot and ankle rehabilitation program is instituted. A removable fracture boot is used during the second 3-week period, followed by the use of a semirigid pneumatic ankle brace.

Chronic Subtalar Instability The Broström anatomic ligament reconstruction with the Gould modification reliably addresses lateral ankle and subtalar instability. The lateral capsular imbrication or advancement must extend to include the CFL as well as the ATFL. Imbrication of the extensor retinaculum to the lateral fibula effectively limits excessive subtalar motion. Historical procedures using tendon graft or transfer to reconstruct the lateral ligaments also stabilize the subtalar joint. Elmslie’s original technique using a piece of fascia lata to reconstruct the lateral ankle also constrained the subtalar joint. Chrisman and Snook’s technique split the peroneal brevis and rerouted it to stabilize the ankle and subtalar joints. This nonanatomic reconstruction can overly constrain the subtalar joint. Newer techniques attempt to reconstruct the joints in a more anatomic manner. Kato demonstrated that conservative management designed to prevent anterior translation of the calcaneus relative to the talus was effective for most patients.226

Foot and Ankle 1951

B 3 2 A

1

Figure 25C1-46   Interosseous ligament reconstruction as described by Schon has the advantage of an anatomic reconstruction. (Redrawn from Schon LC, Clanton TO, Baxter DE: Reconstruction of subtalar instability: A review. Foot Ankle 11:324, 1991.)

Fourteen patients underwent reconstruction of the interosseus ligament with an Achilles tendon graft augmented by cervical and lateral ligament reconstruction. Because the reconstruction was performed near the center of rotation, subtalar motion was not limited (the publication did not provide data to substantiate this finding). All patients reported excellent outcomes. Schon and coworkers reviewed several tendon transfers used for lateral ankle stabilization that also have utility for subtalar stabilization.227 Stabilization with a tendon transfer is suggested in cases of severe injury, generalized ligamentous laxity, and previous failed reconstruction. Interosseous ligament reconstruction, as described by Schon, has the advantage of an anatomic reconstruction (Fig. 25C1-46). The anatomic configuration ensures preservation of ankle and subtalar motion. The authors recommend its use for mild subtalar instability.

Rehabilitation With nonoperative treatment, tissue injury initiates a predictable and sequential series of events known as the healing response. This response is typically divided into three phases with arbitrary and overlapping time lines.120 The initial phase is the inflammatory phase and includes the first through third days after injury. The second phase is a proliferative phase of tissue repair that extends from day 3 to day 20. The final phase is a remodeling phase that proceeds after day 9. To a certain degree, rehabilitation follows the phases of the healing response in an effort to reduce the undesirable effects of inflammation (e.g., pain, swelling, loss of function) while simultaneously promoting tissue repair and functional recovery. The emphasis of rehabilitation is placed on ankle, subtalar, midfoot, and forefoot flexibility, motor function, and coordination.30 The hindfoot is supported by a functional ankle brace or various taping methods. An elastic sock is available for additional mobilization of edema (see Fig. 25C1-19).

The pain and inflammation associated with the first few days following subtalar joint sprain are addressed with rest, cold therapy, and whirlpool. A trial of electrical stimulation may be considered for nonbony injuries. Foot and ankle passive and active range of motion are re-established. Isometrics may be initiated as pain allows. Once the acute pain subsides, flexibility is addressed in all planes. An inclined board facilitates stretching the gastrocnemius-soleus complex (see Fig. 25C1-20). Strengthening is initiated with towel scrunches (see Fig. 25C1-21), toe pick-up activities, manual resistive inversion and eversion, elastic bands (see Fig. 25C1-22), seated toe and ankle dorsiflexion with progression to standing, and seated ­supination-pronation with progression to standing. Closed chain activities are gradually introduced, including one-leg balance, sports-specific activities on a trampoline, and use of the BAPS (see Fig. 25C1-23). Aerobic fitness is maintained with cross-training activities such as water running (see Fig. 25C1-24) and cycling. Heat therapy, such as the application of warm packs, is a useful modality before the therapy session. It reduces pain and spasms and thus facilitates increased range of motion. Cold therapy, compression, and elevation are used after each therapy session to reduce inflammation. Patients are allowed to progress to walking and running activities within the limits of a pain-free schedule. Once running activity is mastered, a monitored plyometric program is introduced with progression of difficulty. Schedules are carefully controlled to avoid reinjury.

Associated Injury Sinus Tarsi Syndrome O’Connor described a sinus tarsi syndrome as persistent pain at the sinus tarsi that follows a lateral ankle sprain.228 Others have noted the syndrome to follow as a sequela of subtalar instability.208 Taillard and associates noted that, in addition to pain over the sinus tarsi, patients with sinus tarsi syndrome complain of lateral ankle instability (Table 25C1-6).229 The cause of the condition is poorly understood and few diagnostic criteria exist. O’Connor speculated that the condition was related to fat pad scarring.228 The diagnosis was made in 45 patients, 14 of whom were treated with sinus tarsi débridement, a procedure that included resection of the fat pad and the superficial ligaments at the floor of the sinus tarsi. Complete relief was noted in 9 patients and partial relief in the remaining 5. Brown performed the O’Connor operation on 11 patients with sinus tarsi syndrome.230 Ten patients reported significant relief. A single patient with early signs of talar osteoarthritis did not improve. Meyer and Lagier presented four cases of sinus tarsi syndrome, all successfully treated by curettage of the sinus tarsi contents.231 Histologic findings from the small series indicated fibrous scar formation consistent with a traumatic cause. Two of the patients were treated with simultaneous subtalar arthrodesis, and one was treated with repair of a ruptured CFL. Parisien and Vangsness described the use of subtalar joint arthroscopy for evaluation of the posterior facet and its anterior recess.232 Parisien used this technique to rule

�rthopaedic ����������� S �ports ������ � Medicine ������� 1952 DeLee & Drez’s� O

  TABLE 25C1-6  Clinical Signs of Sinus Tarsi Syndrome History of foot or ankle inversion sprain Subjective instability Tenderness Ankle stability Anesthetic injection to the sinus tarsi Radiographic findings Arthrography (subtalar) Magnetic resonance imaging

Laboratory testing (rule out systemic inflammatory process)

Yes Yes Localized to the sinus tarsi Yes/no Temporary symptomatic improvement Normal Loss of sinus tarsi filling Rupture of interosseous ligament or cervical ligament; bone edema, soft tissue swelling, or fibrosis at sinus tarsi

Data from references 155, 228, and 229.

out intra-articular disease, to perform biopsy of synovium, and to release periarticular adhesions in three patients with post-traumatic hindfoot pain.233 A diagnostic approach to chronic sinus tarsi pain includes routine and stress radiographs to rule out ankle and subtalar instability as well as MRI to rule out contiguous arthrosis and ganglion formation. Local injection of anesthetic and corticosteroid is most important for diagnosis and initial empirical treatment. Ankle bracing and foot and ankle rehabilitation are used as adjunctive treatments. Occasionally, sinus tarsi exploration and débridement are performed either by open method or arthroscopically. When indicated, open débridement begins with an incision over the sinus tarsi. Care is taken to avoid the lateral branch of the superficial peroneal nerve. The inferior extensor retinaculum is reflected distally along with the extensor digitorum brevis muscle. The capsule of the subtalar joint is incised. The joint is inspected for chondral injury and osteophytes, which are addressed as needed. Fibrofatty tissue is resected from the sinus tarsi, but the ligaments are kept intact. The wound is closed in a layered fashion with attention to reattachment of the extensor digitorum brevis. Patients are placed in a splint postoperatively. After 10 days, the patient is placed into a weight-bearing short-leg cast for 4 more weeks.

Author’s Preferred Method of Treatment Low-grade subtalar sprains are treated with a semirigid pneumatic orthosis, weight-bearing to tolerance, and early rehabilitation of the foot. Most of these injuries are associated with a lateral ankle sprain, and treatment is dictated by the ankle injury. High-grade injuries, as suggested by a clinical history of severe deformity or swelling, examination consistent with gross subtalar instability, or MRI demonstrating ligamentous disruption, are placed in non–weight-bearing short leg casts for 3 weeks. This is followed by application of a removable cast boot for an additional 3 weeks of bearing weight to tolerance. A comprehensive rehabilitation program follows, with an emphasis placed on reducing swelling and improving range of motion, strength, and proprioception.

Return-to-Play Criteria Recovery follows a logical sequence of events. Once the initial pain and swelling subside, coordination and strengthening activities are emphasized. Gradually, the patient is able to return to walking, running, and cutting programs. A semirigid pneumatic brace (see Fig. 25C1-10) or taping accelerates the schedule. Patients are returned to sport once they have mastered sport-specific drills.

Subtalar Dislocation Subtalar joint dislocation varies from isolated subtalar and talonavicular dislocation to dislocation associated with talar, calcaneal, or navicular fractures. The injury is unusual during athletic participation; it most commonly is associated with high-energy mechanisms. Grantham reported five case of medial subtalar dislocation, all with an inversion mechanism of injury.234 Four of these patients sustained the injury while playing basketball. Dendrinos and coworkers reported a single case of subtalar dislocation without fracture in a professional basketball player.235 DeLee and Curtis identified 17 subtalar dislocations in a 7-year period.236 The direction of the dislocation was anterior, lateral, and medial, for 1, 4, and 12 patients, respectively. Four of the medial dislocations were associated with an inversion mechanism of injury; each patient attained full subtalar range of motion and was asympto­matic at the last follow-up. Avascular necrosis did not occur. Christiensen and associates noted arthrosis in each of 17 patients with a fracture-dislocation of the subtalar joint and 6 of 13 patients with isolated dislocation of the subtalar joint.237

Relevant Anatomy A review of hindfoot bony and ligamentous anatomy is presented in the preceding section (see “Subtalar Sprain”). The subtalar joint is stable in an everted position. This suggests that lateral dislocations are more likely to be associated with higher degrees of trauma, fractures, and less favorable outcomes.217,237,238

Clinical Evaluation History Patients with subtalar joint dislocation present with a history of acute and severe pain associated with obvious hindfoot deformity.

Physical Examination After trauma protocols are followed, the extremity is carefully examined (Fig. 25C1-47). Open injuries are documented along with the neurologic and vascular status of the extremity.

Imaging Radiographs Routine radiographs of the foot and ankle in orthogonal views confirm the preliminary diagnosis. Repeat

Foot and Ankle 1953

Author’s Preferred Method of Treatment

Figure 25C1-47   Subtalar dislocation.

Subtalar dislocations, both related and unrelated to ­athletic activity, warrant significant attention at the outset. Historical information to determine the mechanism of injury is paramount. Trauma protocols are followed as indicated. After radiographic documentation is obtained, a closed reduction is performed with the patient under conscious ­sedation or general anesthesia. The stability of the reduction is assessed at the time of reduction. Complete radiographic studies, including CT, are obtained after reduction. The length of immobilization is determined by postreduction stability. A stable reduction is immobilized initially in a non–weight-bearing cast for 2 weeks. This is followed by application of another short leg cast or removable fracture boot for an additional 4 to 6 weeks. Patients are encouraged to bear weight as tolerated. Initial rehabilitation emphasizes swelling reduction and improved range of motion, strength, and proprioception.

r­ adiographs with anteroposterior, oblique, Brodén,219 and lateral projections, as well as CT, are required before definitive and complete diagnosis is determined.

Therapeutic Options Subtalar dislocation without fracture is treated expediently with closed reduction. Reduction is required for decompression of neurovascular structures and should not be delayed without appropriate reason. General anesthesia is probably more predictable and is less likely to produce associated injury to the foot. The knee is flexed in an effort to relax the gastrocnemius muscle.217 Relocation is performed by accentuation and then reversal of the deformity along with traction applied through the foot. As has been noted earlier, repeat and complete radiographs and CT are required for verification of subtalar joint reduction and assessment of associated fractures. After reduction of isolated subtalar joint dislocations is performed, the joint is usually stable. With a stable joint, immobilization for 4 to 8 weeks is indicated. DeLee and Curtis suggest 3 weeks of immobilization, immediate toe range of motion, and early subtalar range of motion. Recurrent dislocation did not occur in any of the 17 subtalar dislocations in the series.236 Dendrinos and colleagues reported a single case of subtalar dislocation without fracture in a professional basketball player.235 The patient was treated with closed reduction. After 5 years of continued professional play, the same foot suffered an almost identical subtalar dislocation. The recurrence was attributed to coincidence.

Rehabilitation After a period of immobilization, a comprehensive foot and ankle rehabilitation program (as described earlier in the section, “Subtalar Sprain”) is instituted for maximal recovery of hindfoot function. Continued subtalar support is provided by a semirigid pneumatic orthosis.

Return-to-Play Criteria After completing short-term rehabilitation, the athlete is gradually returned to progressive activity. An ankle brace is used to provide additional support and biofeedback. Return to play is allowed only after local tenderness and swelling have resolved.

Bifurcate Sprain Injuries to the bifurcate ligament typically result from forceful inversion and plantar flexion of the foot. This results in either bifurcate ligament sprain or an avulsion fracture of the anterior process of the calcaneus. Broström noted clinical evidence of bifurcate ligament injury in 18.6% of patients with acute ankle sprains and 3.7% of patients with confirmed lateral ankle ligament ruptures.40 Backmann and Johnson considered bifurcate ligament rupture to be a common injury.239 Søndergaard reported a 24% incidence of bifurcate or talonavicular sprain among patients presenting with an acute ankle or foot inversion sprain.240 An additional 9% of patients were noted to have a combination of bifurcate and talonavicular ligament and lateral talocrural ligament injuries.

Relevant Anatomy The bifurcate ligament is a short, stout ligament that originates from the anterior process of the calcaneus, divides into two arms, and inserts onto the navicular and cuboid (Fig. 25C1-48; see Fig. 25C1-1).206 The origin is contiguous with the superior aspect of the calcaneal facet of the calcaneocuboid joint. The origin of the ligament is routinely visualized during the lateral approach to the sinus tarsi for triple arthrodesis. The ligament is distal and anterior to the inferior tip of the fibula, and superior and proximal to the base of the fifth metatarsal.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1954 DeLee & Drez’s� O

Physical Examination Anterior inferior tibiofibular ligament Anterior talofibular ligament

Bifurcate ligament Figure 25C1-48   Schematic drawing demonstrating the bifurcate ligament.

Physical examination confirms diffuse lateral hindfoot and midfoot swelling with associated ecchymosis. Tenderness tends to localize to the course of the bifurcate ligament, an area that is distinct from the course of the ATFL. The ankle and midfoot remain stable. Pain is easily reproduced with forced inversion of the plantar flexed foot. Broström noted that the differentiation between lateral ankle ligament injury and bifurcate ligament injury was best achieved by eliciting indirect tenderness.40 He suggested manipulation of the heel to produce lateral ankle pain and stabilization of the heel with simultaneous forced forefoot motion to produce bifurcate pain.

Imaging Radiographs

Clinical Evaluation Sprains can be classified according to the perspective of graded ligament injury, as suggested by the American Medical Association35 and by O’Donoghue.36 Acute injuries are graded based on stretch (grade I), partial tear (grade II), or complete rupture (grade III) of the bifurcate ligament. Additional information with regard to associated ligamentous injuries, especially at the lateral ankle, is noted.

History A bifurcate ligament sprain or an avulsion fracture of the anterior process of the calcaneus must be considered when a patient presents with a suspected ankle sprain. The plantar flexion inversion mechanism associated with injury to the ATFL is the same mechanism as that responsible for the bifurcate ligament sprain. The patient often recalls a pop or snap followed by swelling and ecchymosis. The patient’s ability to continue play after the acute injury is variable.

A

Routine radiographs, including anteroposterior, lateral, and oblique views of the foot and ankle, are obtained. A pure bifurcate ligament sprain is not associated with bony injury; however, an avulsion fracture of the anterior process of the calcaneus is confirmed with the lateral radiograph (Fig. 25C1-49). The size of the fragment may vary from a fine calcified body to a significant portion of the anterior process and the contiguous calcaneocuboid facet.

Computed Tomography Computed tomography is the preferred method for assessing avulsion fractures, but it is not particularly useful for evaluating an isolated bifurcate ligament sprain.

Magnetic Resonance Imaging Isolated bifurcate ligament injury is not routinely imaged by MRI. A sprain is confirmed by edema within or adjacent to the ligament as well as by increased marrow signal at the anterior process of the calcaneus.

B

Figure 25C1-49   A, Avulsion fracture of the anterior process of the calcaneus. The triangular fragment is either intra-articular or extra-articular and varies in size. B, Magnetic resonance image demonstrating increased signal within the bifurcate ligament consistent with sprain.

Foot and Ankle 1955

Therapeutic Options Acute bifurcate sprains are treated with the RICE method followed by gentle range of motion and protected weightbearing. The hindfoot is supported by the use of a variety of methods, including a splint, a walking cast, a removable boot, a functional ankle brace, or various taping methods. Chronic bifurcate sprains are treated with a range of motion program and reduced activity levels. Intralesional and intra-articular (calcaneocuboid) steroids are placed, under fluoroscopic guidance, with the patient either in the sports medicine practitioner’s office or in a radiology suite. Operative treatment is rare for bifurcate ligament injuries. Large, displaced, intra-articular anterior process fractures are treated with open reduction and internal fixation through a sinus tarsi approach. The extensor digitorum brevis muscle is elevated and the fragment reduced and provisionally pinned. Fixation is accomplished with a small or mini-fragment screw. For symptomatic nonunions, the same approach is used, with the addition of local bone graft. Unfortunately, most bony nonunions are not amenable to open reduction with internal fixation owing to their small size. Excision is a reasonable option, but it does not yield the immediate relief that both patient and surgeon expect. Therefore, excision of a symptomatic nonunion is considered only after 6 to 12 months of rehabilitation.

Rehabilitation Emphasis throughout the protocol is placed on rehabilitation of the foot and ankle and on subtalar flexibility, motor function, and coordination.30 The foot is supported by a functional ankle brace or various taping methods. An elastic sock is available for additional mobilization of edema (see Fig. 25C1-19). In the acute phase, the athlete’s pain and inflammation are addressed with rest, cold therapy, and whirlpool. A trial of electrical stimulation may be considered for nonbony injuries. Foot and ankle passive and active range of motion are re-established. Isometrics may be initiated as pain allows. Once the acute pain subsides, flexibility is addressed in all planes. An inclined board is a useful adjunct to ­gastrocnemius-soleus and Achilles stretching (see Fig. 25C1-20). Strengthening is initiated with towel scrunches (see Fig. 25C1-21), toe pick-up activities, manual resistive inversion and eversion, elastic bands (see Fig. 25C1-22), seated toe and ankle dorsiflexion with progression to standing, and seated supination-pronation with progression to standing. Closed chain activities are gradually introduced (see Fig. 25C1-23), including one-leg balance, sport-specific activities on a trampoline, and use of the BAPS. Aerobic fitness is maintained with cross-training activities such as water running (see Fig. 25C1-24) and cycling. Heat therapy, such as the application of warm packs, is a useful modality before the therapy session. It reduces pain and spasms and thus facilitates increased range of motion. Cold therapy, compression, and elevation are used after each therapy session to reduce inflammation. As patients prepare to return to sports, walking and running activities are progressed within the limits of a pain-free

schedule. Once running activity is mastered, a monitored, plyometric and cutting program is introduced. Schedules are carefully controlled to avoid reinjury. Søndergaard and coworkers reviewed the results of treatment for 162 midtarsal sprains (bifurcate ligament or talonavicular ligament, or both) and 161 talocrural sprains and concluded that the two injuries produce similar outcomes.240 Both groups returned to preinjury athletic activity at an average of 21 days.

Author’s Preferred Method Bifurcate ligament sprains are initially treated with a short removable fracture boot. Patients are allowed to bear weight as tolerated. Early rehabilitation of the foot and ankle follows. Occasionally, I place a less athletic patient in a short leg walking cast in an effort to improve pain control and allow increased levels of independent activity. Injuries associated with small anterior process fractures are treated as severe sprains. Delayed union or nonunion is unlikely and is even less likely to remain symptomatic. Large, intra-articular fractures are treated with a short leg cast and weight-bearing to tolerance. Displaced fractures are treated surgically with internal fixation. Chronic pain related to nonunion or malunion is treated conservatively for a minimum of 6 to 12 months. If a small fragment remains symptomatic, it is excised. Attempted union of larger fragments requires open treatment with internal fixation and bone grafting.

Return-to-Play Criteria Recovery follows a logical sequence of events. Once the initial pain and swelling subside, coordination and strengthening activities are emphasized. Gradually, the patient is able to return to walking, running, and cutting programs. A protective brace or taping accelerates the schedule. Patients are returned to sport once they master sport-­specific drills.

Lisfranc Sprain Much of the literature discussing injury to the Lisfranc joint complex is in association with severe high-energy fracture-dislocations. The poor outcome among these patients is well known by orthopaedic surgeons. Subtle Lisfranc injuries can occur during athletic activity and can be challenging to diagnose. Curtis and associates described 19 such injuries associated with athletic activity.241 The most common activity was basketball, followed by running. Meyer noted 24 midfoot sprains among university football players between 1987 and 1991.242 The incidence was calculated at 4% of football players per year. Shapiro and colleagues identified nine injuries to the Lisfranc ligament associated with collegiate gymnastics (four), collegiate football (three), collegiate pole vault (one), and recreational tennis (one).243

�rthopaedic ����������� S �ports ������ � Medicine ������� 1956 DeLee & Drez’s� O

Relevant Anatomy The midfoot is a stable configuration of five bones (navicular, cuboid, medial cuneiform, middle cuneiform, and lateral cuneiform) joined together in a complex system of multifaceted, relatively immobile joints (Fig. 25C1-50). The tarsometatarsal articulation between the midfoot and the metatarsals is known as the Lisfranc joint. The second metatarsal cuneiform joint is the most stable of the entire complex. Two factors contributing to the second metatarsal cuneiform joint stability include a recessed bony configuration (keystone) and a strong plantar ligament connecting the base of the second metatarsal to the medial cuneiform (Lisfranc’s ligament). A significant amount of force is required to produce fracture-dislocation. The mechanism of injury is either direct crushing or indirect loading of the fixed forefoot.244 Wiley reviewed 20 cases of Lisfranc injury and identified direct and indirect mechanisms of injury.245 The indirect mechanisms of injury were acute abduction of the forefoot (most common) and forced plantar flexion of the forefoot. Forced dorsiflexion of the forefoot may occur as the result of landing from a jump, continued forward motion on a planted forefoot, a fall from a height, or a brake pedal injury. Curtis and coworkers described the mechanism of injury to include plantar flexion and rotation with or without abduction of the forefoot.241 Meyer and associates determined a mechanism of injury from 16 football players with midfoot sprains.242 Eight players reported an indirect twisting mechanism; six players reported contact to the heel of a plantar flexed forefoot; two players reported a crush injury to the dorsum of the foot. Shapiro and colleagues identified a consistent mechanism in nine athletes who sustained a Lisfranc injury.243 Each athlete placed full weight onto the first ray with the foot in an externally rotated and pronated position. The resulting injury includes tearing of a relatively weak dorsal capsular structure, tearing of the strong plantar ligament between the medial cuneiform and the base of the second metatarsal, and to a varying degree, fracture of chondral and bony structures on both sides of the joint. Subsequent to the displacement that occurs at the time of

injury, the joint complex either returns to a nondisplaced state or remains displaced owing to the interposition of the capsule and osteochondral fragments.

Clinical Evaluation Sprains can be classified according to ligament injury, as suggested by the American Medical Association35 and by O’Donoghue.36 Acute injuries are graded based on stretch (grade I), partial tear (grade II), or complete rupture (grade III) of the Lisfranc capsule and supporting ligaments, including the Lisfranc ligament. Stable injuries, including grade I and II sprains, are not associated with displacement or deformity. Unstable injuries, grade III sprains, vary between nondisplaced injuries and frank fracture-dislocations. Nunley and Vertullo have proposed a classification system for athletic midfoot injuries. Stage I represents a ligament sprain with no diastasis or loss of arch height on a weight-bearing radiograph. Stage II injuries have diastasis of 1 to 5 mm between the first and second metatarsals without arch height loss. Stage III injuries are associated with diastasis and loss of arch height as defined by a decrease or reversal in the distance between the plantar medial cuneiform base and the fifth metatarsal base on a weight-bearing lateral foot radiograph.246

History Typically, the athlete can recall a specific mechanism of injury. The injury is associated with a pop or snap followed by pain, swelling, and ecchymosis localized to the midfoot. Some patients are able to bear weight with pain, but are unable to run, jump, or continue to play after the acute injury. After a severe injury, weight-bearing is very painful and unlikely. Pain localized to the midfoot should raise suspicion for a subtle Lisfranc injury.

Physical Examination Physical examination confirms diffuse midfoot swelling with associated ecchymosis. Sensory and vascular examination with particular attention to the deep peroneal nerve and the dorsalis pedis artery is documented. Stability is tested in the sagittal plane (dorsiflexion and plantar flexion) by securing the midfoot with one hand and grasping the first metatarsal with the other hand. A dorsiflexion force is applied; when compared with the opposite midfoot, pain and increased mobility are abnormal findings. Frontal plane stability is demonstrated by applying an adduction or abduction force across the Lisfranc joint. Myerson and colleagues described a passive pronationabduction test to evaluate the stability of the joint complex (Fig. 25C1-51).247-249 This maneuver elicits pain and reproduces the patient’s symptoms.

Imaging Figure 25C1-50   The Lisfranc articulation with its ligamentous attachments. Note the recessed second tarsometatarsal joint and the Lisfranc ligament in place of the first-second intermetatarsal ligament.

Radiographs The bilateral standing anteroposterior radiograph provides critical information used to help classify the injury. Additionally, weight-bearing lateral and oblique views of

Foot and Ankle 1957

with a poor outcome.250 The authors also studied 20 normal volunteers. With the use of the standing lateral radiograph, the distance between the plantar aspect of the medial cuneiform was related to the plantar aspect of the fifth metatarsal base. In all normal subjects, the medial cuneiform was higher than the fifth ­metatarsal base.

Stress Radiographs

Figure 25C1-51   Frontal plane stability at the Lisfranc joint is demonstrated by application of a passive pronation-abduction. (Redrawn from Komenda GA, Meyerson MS, Biddinger KR: Results of arthrodesis of the tarsometatarsal joints after traumatic injury. J Bone Joint Surg Am 78:1668, 1996.)

the foot are obtained. A diastasis between the bases of the first and second metatarsals suggests an unstable injury. A small fragment of bone, called the fleck sign, represents an avulsion fracture of the Lisfranc ligament from the base of the second metatarsal (Fig. 25C1-52). The medial border of the second metatarsal should be parallel to the medial edge of the middle cuneiform on the anteroposterior view. The medial border of the fourth metatarsal should align with the medial border of the cuboid on the oblique radiograph. Dorsal displacement of the metatarsal bases is best assessed on the weight-bearing lateral projection. Faciszewski and associates reviewed 15 cases of subtle injury of the Lisfranc joint and determined that ­flattening of the longitudinal arch correlated with a poor outcome, whereas persistent diastasis up to 5 mm did not ­correlate

A

In cases with strong clinical signs and normal radiographs, stress films can help elicit subtle widening.251 The procedure can be performed in the office after injection of local anesthetic into the region or in the operating room with the patient under general anesthesia. Both pronation-abduction and supinationadduction forces are applied to the forefoot, and widening or instability of the tarsometatarsal articulation is noted.

Computed Tomography Further imaging with CT allows for a more detailed analysis. Alignment, displacement, and subtle osseous injury are best evaluated with fine-cut CT (see Fig. 25C1-52).

Magnetic Resonance Imaging MRI is occasionally used to identify the hemorrhage and edema associated with acute ligamentous injury or to differentiate complete and partial tears.252

Therapeutic Options All acute injuries are treated with the RICE method, as described earlier. Patients are immobilized and prevented from bearing weight until a definitive diagnosis is

B

Figure 25C1-52   A diastasis and occasionally a bone fragment between the first and second metatarsal bases suggest injury to the Lisfranc joint. A, Lisfranc injury as seen on routine standing radiograph. B, Lisfranc injury as seen on computed tomography.

�rthopaedic ����������� S �ports ������ � Medicine ������� 1958 DeLee & Drez’s� O

e­ stablished. Subsequent treatment is predicated on the stability of the Lisfranc complex.

Grade I Lisfranc Sprain These injuries are immobilized in a short leg cast and kept from bearing weight for 4 to 6 weeks. On verification of stability by clinical and radiographic examination, patients are transitioned to a walking boot, and weight-bearing to tolerance is encouraged. A rehabilitation program is instituted, with an emphasis on recovery of foot and ankle range of motion, strength, and coordination. As healing progresses, the boot is discontinued, and a steel or fiber carbon shoe insert is used to reduce bending forces at the midfoot. Supportive tape or custom-molded orthotics may decrease the time before return to athletic activity. If pain persists or the patient is unable to return to competition, operative management is considered.

Grade II Lisfranc Sprain As previously described, this group includes injuries with 2 to 5 mm of displacement. There is some disagreement about whether these injuries can be addressed through closed reduction and internal fixation instead of open treatment. Those that favor this technique use large reduction forceps and fluoroscopy to guide reduction before screw placement. Other authors advocate open treatment described next.

Grade III Lisfranc Sprain Unstable nondisplaced injuries, as verified by radiographs and CT, are treated with open reduction and internal fixation. Weight-bearing is delayed for 8 to 12 weeks after surgery. Routine screw removal remains controversial, but generally does not occur until 12 to 16 weeks after ­surgery. Meyer and colleagues reported no untoward effect associated with small, persistent diastasis between the first and second metatarsal bases after a midfoot sprain in a collegiate football player.242 Aitken and Poulson244 and Brunet and Wiley253 found that persistent subluxation or malalignment caused little functional disability. Others contend that anatomic reduction and stable fixation remain paramount.251,254-258 Nondisplaced injury or anatomic reduction does not prevent poor outcome. Resch and Stenström completed a review of 45 consecutive tarsometatarsal injuries, with an average 5-year follow-up.257 They noted that 3 of the 11 patients with nondisplaced tarsometatarsal injuries (<2 mm displacement on initial radiographs) reported persistent pain and poor functional outcome. Displaced grade III injuries require open reduction and internal fixation with the use of screws or plates. K-wire fixation does not provide sufficient fixation for these injuries. The injury pattern dictates the hardware configuration. Fully threaded 4-mm screws or larger are preferred, but partially threaded screws can be used as long as they are not placed in a lag manner. Weight-bearing is delayed for 8 to 10 weeks. The hardware is removed 3 to 4 months after surgery, before the patient’s return to full athletic activity.

Technique: Lisfranc Open Reduction with Internal Fixation (Fig. 25C1-53)

1. The procedure is performed with the patient under general anesthesia. The patient is supine with a well-padded bump placed beneath the ipsilateral ­hemipelvis. 2. The procedure can be performed with or without ­tourniquet. 3. A 5- to 6-cm longitudinal incision is created starting 2 cm proximal to Lisfranc’s joint and carried distally over the first intermetatarsal space. 4. Blunt subcutaneous dissection is performed. The extensor retinaculum is identified and divided. The extensor tendons are carefully retracted laterally. The deep peroneal nerve and dorsalis pedis artery are mobilized and protected. 5. The capsule and the periosteum over the second metatarsal base are inspected and divided as required for exposure of the second tarsometatarsal joint. The interspace between the first and second metatarsal bases is also inspected. Hematoma, capsule, cartilage, and bone fragments are débrided. An anatomic reduction is obtained with manual pressure and held with bone reduction forceps. 6. The anatomically reduced midfoot is temporarily fixed with K-wires. Permanent fixation is carried out with 4- or 4.5-mm screws. The second metatarsal base must be anatomically and rigidly fixed to the medial and middle cuneiforms. If the first metatarsal–medial cuneiform or medial–middle cuneiform articulations are disrupted, they must be stabilized as well. 7. The periosteum and the capsule are repaired with absorbable sutures. 8. The wound is closed in layers and a short leg splint applied. At 10 days, the incision is inspected and a short leg non–weight-bearing cast applied. Alternatively, a removable fracture boot can be used, and a rehabilitation program is instituted once the soft tissues have healed. Patients are not allowed to bear weight for 6 to 8 weeks after surgery. After that, patients are allowed to bear partial weight with crutches. Over 4 to 6 weeks, the patients progress to bear full weight. 9. A comprehensive foot rehabilitation program is continued. The timing and necessity of screw removal remains controversial but generally occurs between 12 and 16 weeks after surgery. Screw retention does not prevent the athlete from training or competing. The foot is protected with a stiff-soled shoe and rigid orthosis as the athlete returns to competition. Delayed-onset arthrosis after missed injuries or even appropriately treated injuries may occur. For advanced arthritic conditions, arthrodesis (fusion) of the midfoot may be required.

Rehabilitation Ankle, hindfoot, and forefoot flexibility, motor function, and coordination30 are emphasized throughout the protocol. The foot is supported by tape, a steel or fiber carbon shoe insert, or a custom-molded orthotic device to reduce

Foot and Ankle 1959

A

C

B

bending forces at the midfoot. An elastic sock is available for additional mobilization of edema (see Fig. 25C1-19). Early in rehabilitation, the athlete’s pain and inflammation are addressed with rest, cold therapy, contrast bath, and whirlpool. A trial of electrical stimulation may be considered. Foot and ankle passive and active range of motion are re-established. Isometrics may be initiated as pain allows. Weight-bearing is protected for 6 to 8 weeks in grade I and II injuries. Weight-bearing after grade III injury is delayed for at least 8 weeks postoperatively. Once the acute pain subsides, flexibility is addressed in all planes. Joint mobilization is used at the midfoot as well as in other affected joints. An inclined board is a useful adjunct to gastroc-soleus and Achilles stretching (see Fig. 25C1-20). Strengthening is initiated with towel scrunches (see Fig. 25C1-21), toe pick-up activities, manual resistive inversion and eversion, elastic bands (see Fig. 25C1-22), seated toe and ankle dorsiflexion with progression to

Figure 25C1-53   A, Percutaneous Lisfranc reduction. B, Open reduction of a Lisfranc dislocation. C. Screw fixation of a Lisfranc dislocation.

s­ tanding, and seated supination-pronation with progression to standing. Closed chain activities are gradually introduced, including one-leg balance, sport-specific activities on a trampoline, and use of the BAPS (see Fig. 25C1-23). Aerobic fitness is maintained with cross-training activities such as water running (see Fig. 25C1-24) and cycling. Heat therapy, such as the application of warm packs, is a useful modality before the therapy session. It reduces pain and spasms and thus facilitates increased range of motion. Cold therapy, compression, and elevation are used after each therapy session to reduce inflammation. After patients are able to bear full weight, walking and running activities are allowed to progress within the limits of a pain-free schedule. Once running activity is mastered, a progressively difficult monitored plyometric and cutting program is introduced. Schedules are carefully controlled to avoid reinjury.

1960 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Author’s Preferred Method I prefer to evaluate all acute-grade Lisfranc sprains with weight-bearing radiographs. If the physical examination or radiograph suggests instability (grade III sprain), a computed tomographic scan is obtained. Occasionally, MRI is obtained to assess a partial injury to the Lisfranc ligament. Stable injuries are treated with a comprehensive rehabilitation program, with emphasis on recovery of foot and ankle range of motion, strength, and coordination. A non–weight-bearing short leg cast is used during the initial recovery period, followed by a walking boot and, ­finally, a carbon fiber shoe insert or a custom orthotic. Unstable nondisplaced injuries, as verified by radiographs and CT, are treated with either closed or open reduction and screw fixation. Weight-bearing is delayed for 8 to 10 weeks. Displaced or angulated injuries are treated surgically with open reduction and screw fixation. Weight-bearing is delayed for 8 to 10 weeks, and the hardware is removed 4 months after surgery.

Return-to-Play Criteria Recovery follows a logical sequence of events. Once the initial pain and swelling subside, coordination and strengthening activities are emphasized. Gradually, the patient is able to return to walking, running, and cutting programs. Patients are returned to sport once they master sport­specific drills. A custom orthotic device or taping supports the midfoot during the transition back to competition. Prolonged convalescence is the rule for the Lisfranc sprain. Persistent stiffness at the midfoot, forefoot, and hindfoot is common after surgery. Shapiro and coworkers identified nine Lisfranc injuries that occurred during

athletic activity.243 Seven players were treated with casting and non–weight-bearing for 4 to 6 weeks. Each player returned to athletic activity, but average time for return to competitive activity was 4 months. Depending on the severity of the injury, return to previous performance level is guarded. Curtis and associates noted that 3 of 19 patients who sustained a Lisfranc injury during athletic activity were unable to return to sport.241 Two of the patients were forced to modify their athletic activity. The average time to return to athletic activity was 4.1 months. C

r i t i c a l

P

o i n t s

l Lateral ankle sprains occur frequently in athletes. Remember to evaluate patients for other sources of lateral ankle pain. l Subtalar instability can occur independently or together with lateral ankle instability. l In patients with recurrent instability, the modified Broström reconstruction works well to address lateral ankle and subtalar instability. l Medial ankle and syndesmotic sprains have a longer rehabilitation and return-to-play time. l With suspected injury to Lisfranc’s joint, a weight-bearing anteroposterior radiograph of both feet should be obtained. l Prognosis for return to competition after Lisfranc injury requiring surgical management is guarded compared with other injuries.

R eferences Please see www.expertconsult.com

S e c t i o n

C

Ligament Injuries 2. Ligament Injuries of the Foot and Ankle in the Pediatric Athlete J. Andy Sullivan

Most of the injuries that occur in the ankle and foot of the pediatric athlete are not unique to athletic participation but occur normally during childhood. Some, however, occur with greater frequency in the athlete. The conditions covered in this chapter occur in childhood and may present in the athlete, raising the question of whether the athlete should be allowed to participate in sports activities.

VARIATIONS OF NORMAL ANATOMY Tarsal Coalition Tarsal coalition is a bony or fibrous or cartilaginous connection of two or more of the tarsal bones. The cause is unknown, but it has been established that the condition

Foot and Ankle 1961

Figure 25C2-1   This patient had a tarsal coalition in the left foot. Note that the foot is held in an everted position. Attempted inversion caused pain and resistance.

results from failure of differentiation and segmentation of the primitive mesenchyme.1 The overall incidence is 1% to 3%.1-3 The most common coalitions are the calcaneonavicular and talocalcaneal types. The first is bilateral in about 60% of cases, and the second is bilateral in about 50%.2-4 More than one type of coalition can exist in one foot. In a review of 60 cases of tarsal coalition Clarke found that 6 of 30 patients had multiple coalitions in the same foot.5 The exact mode of inheritance is unknown, but it is postulated to be autosomal dominant with variable penetrance.2 Most patients seek medical care during early adolescence, at a time when the coalition is ossifying. The pain is vague in nature and insidious in onset. There may be a history of precipitating trauma. Sports participation or running over uneven ground may accentuate the pain. The pain is thought to be due to microfractures in the coalition.6 Physical findings include pain on palpation over the subtalar joint, limited subtalar motion, and at times pes planus and ankle valgus. The peroneal muscles may be tight and

A

resist inversion, but true muscle spasm occurs rarely (Fig. 25C2-1). Any condition that injures the subtalar joint can produce similar symptoms. The clinical diagnosis can be confirmed by radiographic imaging. Plain radiographs, especially the 45-degree oblique view (Fig. 25C2-2), usually demonstrate the calcaneonavicular coalition and other less common coalitions, such as the calcaneocuboid. The talocalcaneal coalition is difficult to visualize on plain radiographs, but secondary changes, which may suggest the need for other studies (Fig. 25C2-3), include beaking and shortening of the talar neck, a middle subtalar facet that cannot be seen, elongation of the lateral process of the calcaneus, and ball-and-socket ankle joint (see Fig. 25C2-3). Computed tomography (CT), which is now the method of choice for the diagnosis of tarsal coalitions not identified on plain films, is comparable in cost and radiation dosage to the other studies. It is technically simple to perform, noninvasive, and accurate. CT provides precise delineation of the anatomy if surgical resection is contemplated. Three-dimensional reconstruction may aid in clearly defining the coalition and planning surgery. CT may also demonstrate the presence of more than one coalition, which is important for treatment planning (see Fig. 25C2-2). El Rassi and colleagues7 reviewed 19 patients with symptoms and signs consistent with tarsal coalition and normal imaging studies. They used technetium-99m scintigraphy and found slightly increased uptake in the middle facet. They obtained good or fair results in all patients after resection of hypervascular capsule and synovium, which had produced arthrofibrosis. Magnetic resonance imaging (MRI), which also can be used, provides the advantage of demonstrating fibrous coalitions.8 The initial treatment of these conditions should include conservative measures aimed at relieving the pain. These measures are empirical and include casting and the use of various shoe inserts and orthotics. The main indication for surgical resection is persistent pain. For calcaneonavicular coalition, resection of the bar with interposition of the extensor digitorum brevis is usually associated with

B

Figure 25C2-2   Plain radiography (A) and computed tomographic scan (B) showing calcaneonavicular and talocalcaneal coalitions, which occurred bilaterally in this patient. The patient presented with a painful, rigid foot and was having difficulty playing tennis.

1962 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25C2-3   Lateral radiograph of the foot. Note the beaking of the talus and the widening of the talonavicular joint. The subtalar joint is narrowed.

good results. Studies by Cowell2 and by Jayakumar and Cowell4 indicated that 23 of 26 feet treated in this manner became symptom free. Talocalcaneal coalition (TCC) is more difficult to recognize, and its surgical management is less certain.9 Before the advent of CT, the diagnosis was often confirmed at the time of surgery. Jayakumar and Cowell4 reported that up to one third of their patients responded to conservative treatment, and they believed that there were few indications for resection. This conclusion was based on evidence showing that family studies indicated that many adults with tarsal coalition were asymptomatic. The surgical alternatives include resection of the bar with interposition of fat or tendon, calcaneal osteotomy, and triple arthrodesis. Scranton10 reviewed 14 patients with 23 sympto­ matic talocalcaneal coalitions. Five feet (3 patients) were treated successfully with casts. Four feet were treated with triple arthrodesis. Eight patients with 13 coalitions that had been resected had a good result. The review was done at a mean of 3.9 years after surgery. In Scranton’s series, about half of the joint surface was removed in some patients. In the series of Swiontkowski and associates,11 10 patients were treated for TCC—four by resection of the bar, and the remainder by some type of arthrodesis. This article stressed that the talar beak is not a true degenerative sign and therefore is not a contraindication to resection of the bar. Olney and Asher12 evaluated nine patients with persistent pain from 10 middle facet TCCs who were treated by resection of the bar and autogenous fat graft. At an average follow-up of 42 months, the results in five were rated excellent, three good, one fair, and one poor. In one patient who underwent reoperation, the fat graft was replaced by fibrous ­tissue. Luhmann and Schoenecker used CT to evaluate TCC in 25 feet. They quantified heel valgus and the size of the coalition relative to the posterior facet.13 The ratio of mean TCC cross-sectional area to the posterior facet was 53.4%. Mean hindfoot valgus was 17.8 degrees. Statistical analysis determined a significant association between TCC greater than 50% the size of the posterior facet and poor outcome (P = .014). Heel valgus greater than 21 degrees was also associated with poor outcome (P = .014). There were good

results, however, in some patients with TCC greater than 50% and in those with heel valgus greater than 21 degrees. These authors advocate using the CT information for preoperative discussion with patients and families. Comfort and Johnson reviewed resection of 20 TCCs at an average of 29 months of follow-up.14 They found good or excellent results with resections involving less than one third of the total joint surface. They did not find increasing age to be a contradiction to the procedure. The management of these coalitions is still controversial and awaits the results of larger series with longer follow-up. Patients with persistent symptoms who do not have degenerative findings have the option of continued conservative care, resection of the coalition, or arthrodesis. Talar beaking is not necessarily a degenerative sign. Severe malalignment of the foot is a contraindication to resection alone. A sliding osteotomy or medial closing wedge as described by Cain and Hyman15 can be used to realign the foot. The size of the coalition that can be resected is unknown, and the question of whether interpositional material is beneficial remains unanswered.

Adolescent Bunion The etiology of adolescent bunion is unknown. Fifty to sixty percent of patients have a positive family history.16 Patients with this condition have an increased intermetatarsal angle (the angle between the first and second metatarsals, normally 10 degrees) and an increased first metatarsal–phalangeal angle (normally 20 degrees). Many also have a relaxed flatfoot and a long first metatarsal ray. None of these conditions is known to be the cause of adolescent bunion. Shoewear has been implicated, but because bunions occur in cultures in which shoes are not worn, this theory seems unlikely. These patients complain of pain, prominence, and difficulty associated with shoewear. On examination, lateral deviation of the toe is found, along with a medial prominence and a wide forefoot. The bursa that is a prominent part of the adult deformity may be present but is usually less impressive. Arthritis and decreased range of motion are also less common. The patient should be evaluated with anteroposterior and lateral weight-bearing radiographs. The joint space is usually maintained. The sesamoids may be laterally displaced in advanced cases. The medial eminence of the metatarsal head is prominent, and a sagittal groove may be present medially. Children should be treated nonsurgically whenever possible. Alteration or stretching of shoewear may alleviate symptoms. Although some series have claimed a success rate of 80% to 95%, this high rate of success has not been the universal experience. Factors implicated in these complications included failure to correct the abnormal deviation of the first metatarsal, failure to correct the soft tissues, too early weight-bearing, inadequate immobilization, and osteotomy performed distal to the open physis. Patients with a hypermobile flatfoot or a long first ray also seemed more prone to recurrence. Indications for surgery include pain that is not responsive to conservative measures and severe deformity. The goals of surgery should include realignment of the first ray and of the metatarsophalangeal joint, cosmesis, and

Foot and Ankle 1963

Cavus Foot Cavus is defined as an increase in the height of the longitudinal arch of the foot. A variety of other modifiers, such as cavovarus and calcaneocavus, are used to further describe the position of the heel. Often, the patient has claw toes or hammer toes and metatarsal head calluses. The presenting complaint can be pain or abnormal wear of the shoe. A cavus foot is usually the result of muscle imbalance that is caused by an underlying neurologic disorder. The patient should undergo a meticulous neurologic examination so that evidence of disorders such as Charcot-Marie-Tooth disease, spinal dysraphism, or a spinal tumor can be detected. Initial radiographic evaluation should include weight-bearing views of the feet and at least an anteroposterior view of the entire spine in search of an occult spinal anomaly.

Miscellaneous Foot Abnormalities

Figure 25C2-4   Large bilateral cornua of prominent accessory naviculars. These are joined by a synchondrosis to the navicular.

prevention of arthritis. Discussion of the types of ­surgical procedures used is beyond the scope of this chapter. In general, the most common procedures are distal soft tissue ­realignment, a distal osteotomy such as Mitchell’s or chevron, or a proximal realignment osteotomy combined with soft tissue procedures. Arthrodesis and resection arthroplasty have no place in surgery for neuromuscularly normal children.

Accessory Navicular Numerous accessory ossicles can occur in the foot, and one needs to be aware of these to avoid confusion with acute fracture. The most common are the os trigonum posterior to the talus and the os vesalianum at the base of the fifth metatarsal. The accessory navicular is a separate ossification center of the navicular. It may be completely separate or joined by a synchondrosis. It may also present as a large or cornuate prominence on the medial side of the navicular (Fig. 25C2-4). We now know that only a small slip of the tendon inserts into this ossicle and that these patients are no more likely to have flatfoot than are those with a normal navicular.17 Many of these patients are asymptomatic. Symptomatic patients experience pain directly over the prominence, usually from shoewear over the prominence. If the shoes are stretched or altered over this prominence, symptoms may be relieved. Other patients experience pain when the posterior tibial tendon is stretched or placed under tension. In those with persistent pain, simple excision of the ossicle without rerouting of the tendon is usually successful.17,18

A variety of other bony variations may produce pain. “Pump bumps,” which are soft tissue bursae associated with prominent calcaneal tuberosities, are irritated by the heel counters of some shoes. Treatment consists of altering the shoewear or padding the heel. Bunionettes are similar bursae that occur over the lateral aspect of the fifth metatarsal head. Stretching the shoe is usually curative. Only rarely is removal of the bony prominence or osteotomy of the fifth ray necessary. A dorsal bunion can occur over the first metatarsal head. In most instances, it is a result of weakness or loss of function in the peroneus longus, which is responsible for depressing the first metatarsal. Treatment must be individualized to balance the overpull of the dorsiflexors of the first ray and may require tendon transfer, osteotomy, or arthrodesis.

SOFT TISSUE INJURIES The ligaments of the ankle insert on the epiphyses distal to the physeal line (Fig. 25C2-5). Because the physis is the weakest link in this bone-tendon-bone interface, it is usually the part that gives way when significant force is applied to the ankle. Serious ankle sprains are unusual in the skeletally immature athlete. Physeal fractures that do occur are discussed in the next section. Minor ankle sprains do occur and are diagnosed by a history of inversion or eversion strain with findings of tenderness over the anterior talofibular or deltoid ligament. Treatment is accomplished by the usual conservative means of rest, ice, compression, elevation, and immobilization. Formal rehabilitation is rarely necessary but may be beneficial for the competitive athlete. Continued pain or disability should provoke a search for other, more serious injury. Recurrent subluxation of peroneal tendons can occur in the adolescent athlete. Usually, a history of injury is followed by recurrent episodes of a snapping sensation and pain. The subluxation can be provoked by forceful dorsiflexion with the foot everted. In patients whose symptoms are sufficiently severe, surgical correction may be indicated. Surgical alternatives include deepening the groove on the fibula, creating a bony block, and reconstructing the superior peroneal retinaculum. The first two are rarely useful in treatment of the pediatric athlete because the physis is still open. Poll and Duijfjes reviewed nine patients

1964 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Deltoid

Medial

Fibulocalcaneal Deltoid Anterior talofibular

Posterior talofibular

Posterior Lateral Figure 25C2-5   The ligaments of the ankle in posterior, medial, and lateral views.

aged 15 to 45 years (average age, 25 years) who underwent reconstruction with the posterior calcaneofibular ligament attached to a bone block.19 Results were said to be good. Contusions on the foot are treated in the same way as those on any other area. Blisters are a frequent problem and require alleviation of the stress, which is usually provided by a new shoe, and protection until healing occurs. Tinea pedis (athlete’s foot) usually responds to a regimen of antifungal medication, along with education about the need to change socks frequently and to use antifungal ­powders.

FRACTURES AROUND THE ANKLE Multiple systems have been proposed to classify ankle injuries. All these classification systems take into account the position of the foot at the time of injury and the force applied. The Salter-Harris classification20 is based on the mechanism of injury and the pathoanatomy of the fracture pattern through the physis, as is interpreted on plain radiographs. The authors described types I through V. Type V, a crush injury to the physis, is difficult to recognize on plain radiographs and is not discussed in this chapter (Fig. 25C2-6). In the skeletally immature patient, the tibiotalar joint surface is rarely disturbed. The injury pattern changes in adolescence as the physis begins to close. The outcome of the injury depends on the type of physeal injury and its management. Tension injuries usually produce SalterHarris I and II injuries of the physis. Compression forces can produce Salter-Harris III and IV injuries. CT and MRI are valuable techniques in assessing and classifying some fractures because they allow more accurate evaluation of the fragments. Carey and associates performed a study of plain films and MRI in 14 patients with acute injuries and reached the following ­ conclusions.21

Direct visualization of cartilage afforded by MRI improved evaluation of growth plate injury in each case. MRI changed Salter-Harris classification or staging in 2 of 9 patients with fractures visualized on conventional radiographs, allowed the detection of radiographically occult fractures in 5 of 14 cases, and resulted in a physical change in management for 5 of the 14 patients studied. MRI has an important role in the evaluation of acute pediatric growth plate injury, particularly when diagnostic uncertainty persists after the evaluation of conventional radiographs. MRI allows detection of occult fractures, may alter Salter-Harris staging, and may lead to a change in patient management. Rohmiller and colleagues have recently used the LaugeHansen system to look at the mechanism of injury in SalterHarris II distal tibial fractures.22 This is discussed later.

Clinical Evaluation The mechanism of injury and the time elapsed since the accident should be noted. The neurovascular status of the foot should be carefully documented. The amount of swelling and the status of the skin are important. Gentle examination should be carried out to seek areas of point tenderness, especially over the physis. This examination may be more useful than radiographs in the diagnosis of Salter-Harris I injuries of the fibula. Radiographs should always include three views of the ankle. Only in this manner will the physician be able to see some fractures and determine whether there is disruption of the ankle mortise. Plain tomography and CT may be indicated in the juvenile fracture of Tillaux and in triplane fractures, which are discussed later. Because treatment and prognosis depend on the Salter-Harris classification, the fractures are discussed according to fracture type.

Foot and Ankle 1965

Type I

Type II lateral

Type II AP

Type III

Type IV

Figure 25C2-6   Fracture patterns of the distal tibial and fibular physes classified by the Salter-Harris system.

Salter-Harris Type I Rohmiller and colleagues combined Salter-Harris I and II fractures of the distal tibia and found an incidence of premature physeal closure (PPC) of 38%.22 No distinction was made between type I and II. Salter-Harris I fractures of the fibula are common and may be missed entirely or misdiagnosed as a sprain. The characteristic history of an external rotation force in a patient who presents with localized tenderness and swelling directly over the distal fibular physis is diagnostic. The radiograph shows only localized swelling and widening of the fibular physis. Many of these injuries probably go unrecognized and untreated.

Salter-Harris Type II This injury is uncommon or is infrequently recognized in the fibula. Salter-Harris II injury of the distal tibia combined with a fracture of the distal fibula is one of the most common injuries of the ankle, accounting for 47.3% of cases in the series compiled by Peterson and Cass.23 Rohmiller and colleagues studied 91 Salter I and II fractures.22 Treatment options include no reduction and cast, closed reduction and cast, closed reduction and percutaneous pins and cast, and open reduction with internal fixation. They found a 39.6% incidence of PPC. Using the Lauge-Hansen classification system, they found a significant increased incidence of PPC in pronation-abduction injuries (54%) compared with supination–external rotation injuries (35%). The most important determinant of PPC was the amount of fracture displacement following reduction. In some cases, periosteum is trapped in the fracture site medially, blocking reduction. They thought that operative treatment may decrease the incidence of PPC.

They recommended less than 2 mm of displacement in a child with 2 years of growth remaining to decrease the risk for PPC. Traditional treatment consists of a long leg bent-knee cast for 2 to 3 weeks followed by a short leg walking cast for 4 weeks. Dugan and coworkers reviewed 56 patients with this injury who were treated with a long leg weightbearing cast for 4 weeks.24 There were no nonunions and no angular deformities. There was one case of clinically insignificant premature closure of the growth plate. This appears to be the treatment of choice because it allows early healing, low morbidity, and rapid rehabilitation.

Salter-Harris Type III In adolescents, this injury is also known as the juvenile fracture of Tillaux. The distal tibial physis closes first in the central region, and then from the medial side toward the fibula. An external rotation force applied to the partially closed physis applies traction on the physis through the anterior talofibular ligament. This avulses a fragment of the lateral physis, which remains attached to the ligament (Fig. 25C2-7). Closed reduction under anesthesia should be attempted. The injury can be treated closed if the fragment is not displaced more than 2 mm, or if it can be reduced closed and percutaneously fixed. Most of these injuries require open reduction and fixation of the fragment with a pin or cancellous screw. Fractures of the medial malleolus can be either type III or IV injuries. If displaced less than 2 mm, they may be treated closed. This treatment should consist initially of a long leg non–weight-bearing cast for 3 weeks followed by a short leg walking cast for 3 weeks. These injuries are the most unpredictable of ankle epiphyseal injuries. Near-anatomic reduction must be obtained.25

1966 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 25C2-7   A and B, The triplane fracture can consist of two or more fragments. C, The juvenile fracture of Tillaux.

Salter-Harris Type IV This group includes some of the medial malleolar fractures and the triplane fractures. The triplane fracture, first described by Marmor, is so named because the fracture lines extend from the physis into the transverse, sagittal, and coronal planes (see Fig. 25C2-7).26-29 This type of fracture may be mistaken for a Salter-Harris II injury if the radiographs are not carefully scrutinized. Many authors have described this fracture and have argued about the number of fragments involved.26,28,29 Most of these studies were based on plain ­radiography. Figure 25C2-7 illustrates the possibilities. In the two-part fracture, the main fragment is the tibial shaft, including the medial malleolus and a portion of the medial epiphysis. The second fragment is the remaining epiphysis, which is attached to the fibula. In the three-part injury, the third fragment is usually an anterior free epiphyseal fragment. Brown and colleagues studied 51 children with tibial triplane fractures.30 By evaluating them with CT with multiplanar reconstructions, these authors have used the best radiographic evaluation possible to define the number of fragments. The classic two-fragment type with medial epiphyseal extension was most frequent (33 of 51). All three-fragment types (8 of 51) had a separate anterolateral fragment. Extension to the medial malleolus was common (12 of 51). None of the four reported fracture types involving anteromedial extension was seen. Karrholm reviewed the literature on this injury.31 Triplane fracture made up 7% of physeal injuries in girls and 15% in boys. Of the injuries, 35% were treated closed without manipulation, 30% by manipulation and casting, and 35% by open reduction and internal fixation. If this injury can be reduced to within 2 mm, it may be treated closed. In the series of Cooperman and associates, 13 of 15 fractures were treated closed,26 and in the series of Dias and Giergerich, 5 of 8 were treated in this way.27 In the series by Ertl and colleagues, residual displacement of more than 2 mm was associated with a high incidence of late symptoms.28 Obtaining a reduction of less than 2 mm by either closed or open means did not ensure an excellent result. Poor results may be related to damage done to the articular surface, or to the amount of displacement. Fractures outside the weight-bearing area did not show this tendency toward poor results.

Evaluation of the adequacy of reduction in this injury is difficult, and because most authors recommend manipulation under general anesthesia, the only radiographic means of diagnosis available is plain radiography. The author’s preferred method is manipulation by internal rotation of the foot under sedation, which usually occurs in the emergency room. If there is any question about the adequacy of reduction on plain radiographs, CT or plain tomography is used to evaluate the articular surface and the reduction (Fig. 25C2-8); if displacement is greater than 2 mm, open reduction with internal fixation is carried out. This may require two incisions. The first is an anterolateral incision, which allows identification of the anterolateral fragment. Usually, it is first necessary to reduce and fix the posterior fragment. If this cannot be done closed, a second posteromedial approach is used to reduce the fragment under direct vision. Fixation is by cannulated screws, cancellous screws, or pins. These injuries require 6 weeks in a cast.

Prediction of Outcome The prognosis for an ankle fracture in a skeletally immature patient depends on the following factors: 1. Mechanism of injury 2. Salter-Harris classification 3. Quality of reduction 4. State of skeletal maturity 5. Amount of displacement 6. Miscellaneous modifiers (open fracture, vascular injury, infection, systemic illness, interposed periosteum) Spiegel and colleagues retrospectively studied a series of closed distal tibial physeal injuries.29 One hundred eightyfour patients (of 237) were followed for an average of 28 months. The authors looked specifically at the complications of angular deformity of greater than 5 degrees and shortening of more than 1 cm, joint incongruity, or asymmetric closure of the physis. These complications appeared to correlate with the Salter-Harris type, the amount of displacement or comminution, and the adequacy of reduction. The patients were divided into the groups shown in Table 25C2-1. The overall complication rate was 14.1% for 184 patients. Salter-Harris II injuries of the tibia appeared to be the least predictable because the incidence of complications remained about the same, regardless of the amount

Foot and Ankle 1967

A

B

C

D

Figure 25C2-8   A and B, The position attained after manipulation of a severely displaced triplane fracture. The position was not acceptable, so further imaging was not necessary. C, The position achieved by open reduction. D, The position that is retained after hardware removal.

  TABLE 25C2-1  Complications of Ankle Fractures Complication Rate (%)

Salter-Harris Group and Bone Involved

Low risk (89 patients)

6.7

High risk (28 patients) Unpredictable (66 patients)

32

Type I and II of fibula, I of tibia, III and IV with displacement of less than 2 mm, epiphyseal avulsion injuries Types III, IV, and V of tibia Tillaux and triplane Type II of the tibia

Group

16.7

of displacement. Displacement is not always mentioned as one of the factors involved in prediction of outcome, but it is intuitive that greater displacement implies greater force, with more likely damage to the articular cartilage, the circulation, and the soft tissues important in healing. Karrholm thought the good results were based on the adequacy of the reduction.31 Near-anatomic reduction of type II injuries of the tibia is desirable. Gruber and associates have show in an animal model that interposed periosteum in an intact physis produces a spectrum of changes at the tissue level and a small but statistically significant leg-length discrepancy

1968 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25C2-9   The Freer elevator is on the metaphyseal periosteum that is inverted into the physeal fracture site. The black mark is on the epiphyseal fragment.

c­ ompared with fracture alone.32 Because this is a spectrum, it may explain the unpredictable nature of these injuries and support the need to remove interposed periosteum (Fig. 25C2-9). Residual displacement after attempted closed reduction may be secondary to this interposition.

The patient in Figure 25C2-10A was treated by closed manipulation and casting. He had residual medial displacement (see Fig. 25C2-10B and C). Follow-up radiographs show the development of a defect in the medial cortex. He subsequently developed a valgus ankle deformity (see Fig. 25C2-10D), which would indicate the medial physis and suspected interposed periosteum grew more than the lateral side. He required a varus tibial osteotomy and physeal closure. Although I have seen patients who developed normal growth despite the medial cortical defect, the unpredictability and high frequency in the Rohmiller article support a near-anatomic reduction with less than 2 mm of displacement in a child with more than 2 years of growth remaining.22 These injuries must be followed until the patient attains skeletal maturity or a normal growth pattern is ensured because some will go on to premature closure and angular deformity.25 The juvenile fracture of Tillaux and the triplane fracture result from incomplete closure of the physis. Because growth is nearing an end, angular deformity and shortening are uncommon. In these patients, the tibiotalar joint surface is disturbed and must be restored to as near normal as possible to prevent incongruity and subsequent traumatic arthritis. In the series by Cooperman and colleagues,26 triplane fractures were reduced with the patient under general anesthesia by internally rotating the foot. The ­ adequacy

Figure 25C2-10   A, Anteroposterior (AP) radiograph on the day of injury of this SalterHarris II tibial fracture. B, Postreduction AP radiograph. C, AP radiograph at 4 months.   D, AP radiograph at 2 years.

A

B

C

D

Foot and Ankle 1969

of reduction was determined by plain tomography. Dias and Giergerich had nine Tillaux and triplane fractures that were followed for an average of 18 months, and all did well.27 Peterson and Cass reviewed all Salter-Harris IV distal tibial injuries seen at the Mayo Clinic, paying particular attention to injuries of the medial malleolus.23 Nine of 18 of these injuries went on to premature physeal closure ­sufficient to require additional surgery for physeal bar resection, angular deformity, or leg-length discrepancy. Thirteen of these patients received their care at the Mayo Clinic, and of these, 11 were closed injuries. Six patients were treated by closed reduction and a short leg cast. Five had open reduction and internal fixation. Five additional patients in the study had been referred to the clinic because of complications of a closed injury that had been treated closed. They concluded that oblique radiographs are necessary to ensure an accurate diagnosis and to confirm the adequacy of reduction. Some injuries that resemble type III injuries are actually type IV. The authors also found that partial arrest that results in angular deformity was more common than complete arrest. They concluded that there are three patterns of medial malleolar injury and that type IV injuries constitute the most common and most dangerous pattern because they usually occur in a patient who has remaining growth potential (Fig. 25C2-11). They also concluded that the medial malleolus requires anatomic reduction, which often necessitates open reduction and internal fixation. In any patient with an open physis, it is preferable to avoid crossing the physis with a fixation device. This goal can usually be achieved by placing smooth pins from metaphysis to metaphysis, or from epiphysis to epiphysis. At times, crossing the physis cannot be avoided. Smooth

pins can be used, and care should be taken that they do not cross within the physis. Patients need to be followed to skeletal maturity, or until one is certain that a normal growth pattern is occurring. An asymmetric Harris growth arrest line may be the earliest clue to an abnormal growth pattern (see Fig. 25C2-11).

FRACTURES IN THE FOOT Fractures of the foot resulting from sports are unusual in children. Fractures of the metatarsals can result from direct trauma (Fig. 25C2-12). These can be treated by immobilization in a short leg walking cast. The most controversial fracture in the foot may be an avulsion injury at the base of the fifth metatarsal. Fractures of the fifth metatarsal in children can be divided into distal physeal fractures, fractures of the proximal diaphysis, and avulsion fractures of the apophysis. The fifth metatarsal has its epiphysis distally and an apophysis proximally. The tendon of the peroneus brevis is inserted into the apophysis. With inversion stress, the apophysis can be avulsed. Findings include tenderness at the base of the fifth metatarsal and radiographic confirmation of widening of the apophysis. Treatment should be symptomatic with compression and partial weight-bearing until the pain subsides. Crutches and an elastic bandage may be sufficient. Two to three weeks in a short leg cast also yields good results. Fracture of the proximal diaphysis of the fifth metatarsal (Jones’ fracture) is less common in skeletally immature patients and usually occurs in the 15- to 20-year age range.33 When such fractures do occur, a trial of immobilization in a short leg walking cast is indicated because

Figure 25C2-11   This patient sustained a Salter-Harris IV medial malleolar fracture, which was treated closed. The patient was referred 6 months after injury, at which time she had trouble remembering which ankle had been injured. These radiographs were taken 18 months after the injury. Resection of a bony bridge and interposition were required. She resumed growth, and the fibular angular deformity had been corrected. Note the irregular Harris growth arrest lines.

1970 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

many acute fractures do heal. Even fractures with delayed union may heal if they are treated conservatively.34 Early operative intervention in highly competitive athletes has been advocated by some, but others have shown that each patient needs to be treated individually because some of these fractures will heal if treated conservatively, allowing early return to athletics.33-35 Early operative intervention in the pediatric athlete is rarely if ever indicated. Patients with established nonunion require operative treatment that includes reopening of the medullary canal, bone grafting, and internal fixation (Fig. 25C2-13). Fractures of the toes are unusual in sports. Most phalangeal fractures can be treated by buddy taping them to the adjacent toe, wearing appropriate shoes for a few weeks, and avoiding sports until the toe is asymptomatic. Articular fractures are even rarer. The only one that may merit consideration of operative management is an intra-articular physeal fracture of the great toe. These should be reduced

Figure 25C2-12   A, This patient sustained fractures of the lateral four metatarsal necks when he caught his foot on a base while sliding. B, This patient fractured his first metatarsal when he was stepped on during a football game.

A

B

to as near-anatomic alignment as possible by whatever means necessary (Fig. 25C2-14). Stress fractures are less common among children than in adults but cannot be entirely dismissed. Some children participate in marathons and other sporting events that can result in stress fractures. Basketball, soccer, and other team sports have tournaments that may require considerable running. The stress fracture shown in Figure 25C2-15 resulted from a tournament and was thought to be a sprain. Yngve found 131 pediatric stress fractures in 23 references in the literature.36 Two reports of the 131 documented metatarsal fractures—two of the tarsal navicular, and one of the medial sesamoid. The primary training error was too much too soon. Other factors that should be considered are a change in training surface or equipment (shoes) and a sudden change in intensity of training ­(tournaments). Diagnosis depends on an appropriate history, a high index of suspicion, and the presence of localized

Foot and Ankle 1971

A

B

Figure 25C2-13   A, This patient sustained a fracture of the diaphysis of the fifth metatarsal. This is the radiograph 2 months after treatment with a cast. The metatarsal was tender to palpation, and the patient walked with a limp. B, This radiograph was taken after treatment with internal fixation and local bone graft from the calcaneus.

tenderness. The differential diagnosis includes contusion, tendinitis, and sprains. The initial radiograph may be normal but should be diagnostic in half of cases. One should look for cortical thickening or a translucent fracture line. A bone scan may be diagnostic at this stage and may be particularly helpful if the diagnosis is in question and one wishes to avoid immobilization. A bone scan may be indicated when the diagnosis is in doubt and the athlete wishes to return to play. MRI has also been used to identify occult stress fractures. Conversely, immobilization for 2 weeks in a cast is usually diagnostic in that pain is relieved; repeat radiographs are then positive, making a bone scan unnecessary. Although some of these fractures heal without a cast, the athlete should be

Figure 25C2-14   This intra-articular fracture was treated by closed reduction and percutaneous pinning.

immobilized for protection from himself or herself as well as from well-meaning parents and coaches.

Osteochondral Lesions of the Talus The term osteochondritis dissecans (OCD) has been used to describe lesions in the dome of the talus. These lesions have been attributed to a vascular insult, but trauma has been implicated, especially in lateral lesions.37-39 ­ Berndt and Harty developed a classification system based on the amount of damage and the degree of displacement involved.37 This system, modified slightly, is still in use. Stage I: Localized trabecular compression Stage II: Incompletely separated fragment Stage IIA: Formation of a subchondral cyst Stage III: Undetached, undisplaced fragment Stage IV: Displaced or inverted fragment Most series consist predominantly of adults; however, 21 of 29 patients studied by Canale and Belding experienced onset of symptoms during the second decade.38 CT and MRI can reveal lesions not seen on plain films and further make staging more accurate. These techniques may reveal that this disorder is more common among adolescents than was previously suspected. Proper treatment depends on identification of the lesion and accurate staging. Canale and Belding recommended that nonoperative treatment by immobilization of all stage I, stage II, and medial stage III (Berndt and Harty classification) lesions would result in a high percentage of good clinical results and delayed development of arthrosis.38 Persistent symptoms after conservative treatment were an indication for operative treatment by excision and curettage. They further recommended that all stage III lateral lesions and all stage IV lesions be treated by immediate excision and curettage of the lesion. Anderson and associates recommended immobilization for 6 weeks for patients with stage I and II fractures, but they cautioned that these patients need to be followed for a prolonged time so that delayed development of arthrosis can be detected.38a Operative treatment was recommended for stage IIA, III, and IV lesions.

1972 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 25C2-15   This patient presented with localized tenderness just above the ankle. After being injured in a basketball tournament, she continued to play with a presumptive diagnosis   of a sprained ankle. The initial radiographs   (A and B) showed periosteal elevation. Follow-up radiographs (C and D) taken after 2 weeks of short leg walking cast treatment illustrate new bone formation.

B

A

C

Letts and colleagues reviewed 24 children treated since 1983 for OCD of either the medial or lateral dome of the talus (2 were bilateral).39 Average age at presentation was 13 years, 4 months. Nonoperative treatment included activity restriction, physiotherapy, and immobilization. Surgical intervention was required in 15 (58%) ankles. Surgical treatment included arthroscopy (1), arthrotomy and drilling of the defect (9), drilling with excision of the lesion (3), excision of the lesion (2), and pinning of the fragment (1). Most recent follow-up revealed resolution or decreased symptoms in 25 (96%) and no change in 1. MRI was useful in preoperative assessment in 6 cases. In this series, there was a slight female preponderance (58%). Higuera and coworkers reported good to excellent results in 94.8% of children.39a

D

Osteochondroses in the Foot The osteochondroses are a group of conditions of unknown origin. Suggested causes have included endocrinopathies, vascular phenomena, infection, and trauma.40 Many of these conditions are now known to represent radiographic variations of normal ossification of the epiphysis. Most are named for the person or persons who originally described them. They all include a pattern of clinical symptoms coupled with a radiograph that suggests that the epiphysis or apophysis is undergoing necrosis. In the foot, the most commonly described conditions are Kohler’s syndrome and Freiberg’s infarction. Kohler’s syndrome is a clinical syndrome consisting of pain in the midfoot coupled with a finding of localized

Foot and Ankle 1973 Figure 25C2-16   This patient presented with undisplaced fractures of the metatarsals. The condensed, narrowed appearance of the navicular is the same as that seen in patients with Kohler’s syndrome, but it was an incidental finding in this patient.

tenderness over the navicular. Radiographs demonstrate increased density and narrowing of the tarsal navicular. Irregular ossification in this bone may be the rule rather than the exception, so the existence of this condition is in question (Fig. 25C2-16). Williams and Cowell reviewed a series of patients with the following findings.41 Thirty percent of males and 20% of females demonstrated irregular ossification in the tarsal navicular. Most patients appeared to respond to 6 weeks of immobilization in a cast. All 23 patients eventually became asymptomatic, and the navicular became normal. The authors believed that patients treated in a cast became asymptomatic sooner than did those treated with shoe inserts. Regardless of treatment, no long-term problems were associated with this condition, again raising the question of whether it is a distinct pathologic condition. Freiberg’s infarction is a condition of condensation and collapse of the metatarsal head and articular surface. It commonly occurs during the second decade of life while the epiphysis is still present.42 It is of unknown cause and is more common among females. Many causes have been proposed, but repetitive trauma probably plays a role. The lesion occurs most commonly in the second or third metatarsal (Fig. 25C2-17).40 These are the longest and least mobile of the metatarsals. The patient presents with pain on weight-bearing and has localized tenderness over the metatarsal head. Radiographs reveal collapse of the articular surface (see Fig. 25C2-17). Conservative treatment with a cast or orthotic device that minimizes weight-­bearing over the involved head is often successful in relieving the pain. Surgical treatment consisting of removal of loose bodies or bone grafting has been reported for persistent symptoms. A dorsiflexion osteotomy to relieve weightbearing has also been reported to work well. Removal of the metatarsal head should be avoided because this results in transfer of weight-bearing to the adjacent metatarsal heads. ­Prosthetic replacement has also been tried but is not indicated in ­children. In most instances, the disease runs its course, and the head reossifies within 2 to 3 years. Sever’s disease is a term used to refer to a nonarticular osteochondrosis or a traction apophysitis (Fig. 25C2-18).

The real question is whether a distinct syndrome exists and, if it does, whether the apophysis has anything to do with it. The calcaneal apophysis appears and develops in the 5- to 12-year age range and is typically irregular. Often, a child with heel pain and an irregular apophysis has the same radiographic finding in the opposite asymptomatic heel. These children are usually in the 9- to 12-year age range and are active in sports. They may have a tight heel cord. The calcaneus serves as the insertion of the powerful gastrocnemius-soleus muscle and the origin of the plantar fascia. Traction or overuse can strain these structures, producing pain. Stretching may be beneficial. Symptomatic treatment by avoidance of the offending exercise is usually curative. Shock-absorbing inserts or a heel cup may be advantageous. A heel lift to relieve some of the pull of the gastrocnemius-soleus, or at times an arch support for a child with a high arch, may give ­symptomatic

Figure 25C2-17   This anteroposterior radiograph of the foot shows irregularity and collapse of the second metatarsal head.

1974 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25C2-18   Lateral radiograph of the calcaneus. Note the irregularity, especially in the superior part of the apophysis. Fragmentation and increased density are common occurrences in the calcaneal apophysis.

relief as well. Heel cord stretching exercises may be tried. One must carefully search for the point of maximal ­tenderness and seek its cause rather than implicating an irregular apophysis, which is probably not the source of the pain. The exact time frame for resolution of symptoms in children with heel pain is unknown and at times can be vexing. If I feel that the child has followed the previous conservative measures and is no better after 2 to 3 months, I proceed with further work-up, such as a bone scan and other studies, to seek more occult sources of the pain.

SYSTEMIC ILLNESS Systemic illness can present with foot pain and must be remembered in the athlete as well as in other children. Rheumatoid arthritis or hemophilia can involve the subtalar joint. Osteomyelitis can involve the foot but is unusual unless there is a history of a puncture wound. Acute lymphocytic leukemia is a great masquerader and can infiltrate the bones of the foot. Although they are rare, these sorts of diseases must be considered. One cannot develop tunnel vision and believe that all pain in an athlete is of traumatic origin.

SHOES AND ORTHOTICS The athletic shoe business is a lucrative one, as is shown by the intense marketing, endorsement, and competition for the introduction of new technology and an edge in the marketplace. Little scientific evidence supports the hype associated with shoe sales. Most often, the advertisements depict current sports heroes wearing shoes from the high end of the price scale, and they tell us little about the shoes themselves. Athletic shoes should fit adequately in both width and length. The models and range of widths available are more limited for children. The material should be reasonably soft. Too often, children’s athletic shoes are made of stiff, unyielding, synthetic material and are poorly padded around the heel counter. Multisport shoes with small-diameter, evenly spaced cleats that distribute weight-bearing more evenly are preferable to cleated or

studded shoes. Padding over the heel counter and ankle may increase comfort. Most shoes now come with a builtin arch support that has little scientific basis but may give some support to children with a well-developed arch. Those with a flatter foot may actually find it necessary to remove the pad. Orthotics is another area of controversy. An asymptomatic flexible flatfoot should be left alone. There is no evidence to support the idea that an orthotic will bring about any structural change in such a foot. A painful flatfoot should prompt a thorough search for its cause, such as a tarsal coalition. Orthotics may be tried in a patient with aching feet or shins and a flexible flatfoot. Heel cups may be beneficial in the symptomatic treatment of heel pain. There is little if any scientific information available about the use of sports orthotics in children.

C

r i t i c a l

P

o i n t s

l Patients presenting with cavus deformity of the foot must have a thorough neurologic examination and workup. l Serious ankle sprains are rare in the skeletally immature patient. The physes are the weak link and usually give way before the ligaments. l One must have knowledge of the factors involved in predictions of outcome in ankle fractures in skeletally immature patients in order to choose treatment modalities and obtain informed consent. l Salter Harris II fractures of the tibia associated with an abduction mechanism of injury must be reduced to near anatomic position. l Patients with Salter Harris III and IV injuries require restoration of normal joint and physeal alignment and congruity.

S U G G E S T E D

R E A D I N G S

Bennet GL, Weiner DS, Leighley B: Surgical treatment of symptomatic accessory navicular. J Pediatr Orthop 10:445-449, 1990. Brown SD, Kasser JR, Zurakowski D, Jaramillo D: Analysis of 51 tibial triplane fractures using CT with multiplanar reconstruction. AJR Am J Roentgenol 5: 1489-1495, 2004. Carey J, Spence L, Blickman H, Eustace S: MRI of pediatric growth plate injury: Correlation with plain film radiographs and clinical outcome. Skelet Radiol 27(5):250-255, 1998. El Rassi G, Riddle EC, Kumar SJ: Arthrofibrosis involving the middle facet of the talocalcaneal joint in children and adolescents. J Bone Joint Surg Am 87(10): 2227-2231, 2005. Karrholm J: The triplane fracture: Four years of follow-up of 21 cases and review of the literature. J Pediatr Orthop B 6(2):91-102, 1997. Letts M, Davidson D, Ahmer A: Osteochondritis dissecans of the talus in children. J Pediatr Orthop 23(5):617-625, 2003. Luhmann SJ, Schoenecker PL: Symptomatic talocalcaneal coalition resection: Indications and results. J Pediatr Orthop 18(6):748-754, 1998. Peterson HA, Cass JR: Salter-Harris IV injuries of the distal tibial epiphysis. J Bone Joint Surg Am 65:1059, 1983. Rohmiller MT, Gaynor TP, Pawelek J, Mubarak SJ: Salter-Harris I and II fractures of the distal tibia: Does mechanism of injury relate to premature physeal closure? J Pediatr Orthop 26(3):322-328, 2006. Vincent KA: Tarsal coalition and painful flatfoot. J Am Acad Orthop Surg 6(5): 274-281, 1998.

R eferences Please see www.expertconsult.com

Foot and Ankle 1975

S e c t i o n

D

Tendon Injuries of the Foot and Ankle Geoffrey S. Baer and James S. Keene

Although the Achilles tendon has been the focus of most reports on tendon injuries of the foot and ankle, athletes sustain many disabling injuries that involve other tendons (e.g., anterior and posterior tibial, flexor hallucis longus [FHL]) in this anatomic region. The frequency of tendon injuries around the foot and ankle has been recorded in only a limited number of studies.1-5 In 1959, Anzel and colleagues reviewed 1014 cases of muscle and tendon disruptions treated at the Mayo Clinic and found that 143 (14%) of the injuries occurred in the lower extremity.1 Of these 143 injuries, 34 (24%) involved the Achilles tendon or triceps surae, 10 (7%) the anterior tibial, 3 (2%) the posterior tibial, 3 (2%) the FHL, and 2 (1%) the peroneus longus or brevis tendon. Although earlier studies6-8 did not provide specific numbers or percentages, they did state that muscles and tendons are torn in the following order of frequency: triceps surae, quadriceps, biceps, triceps, and rectus abdominis. Most of these studies emphasize the rarity of injuries to the tendons of the foot and ankle, and several suggest that these injuries, if untreated, cause little functional impairment.9-13 Other studies emphasize that disruptions of the tendons of the foot and ankle, particularly those that occur in the fibro-osseous tunnels, are subject to the same complications characterizing tendon injuries in other anatomic regions.2,3,5,14-20 These latter studies conclude that the same principles of treatment well established in injuries of the wrist and hand should be applied to tendon injuries of the foot and ankle if maximal function is to be regained. In this chapter, the issues related to operative versus nonoperative treatment of injuries to the Achilles, anterior and posterior tibial, FHL, and peroneal tendons are addressed, and pertinent demographic data (age, sex, sport, and so forth) relative to an athlete’s predisposition to injure the aforementioned structures are delineated.

INJURIES DEFINED Before beginning a discussion of specific tendon injuries, it seems germane to define the terms used to describe and classify these conditions. The dictionary definition of tendinitis (also spelled tendonitis) is “inflammation of tendon” and that of tenosynovitis (or tenovaginitis) is “inflammation of a tendon and its sheath.” It is apparent that these definitions are too generic when one considers that many large tendons, such as the Achilles, do not have a true synovial sheath and are composed of dense fibrous connective tissue that has little vascularity.21 Thus, these tendons lack

the extrinsic vascularity that would predispose them to an inflammatory process. They are, however, encased by a peritendineum—a vascular, hypercellular, peritendinous connective tissue that has visceral and parietal layers, and septa that run between the fibers of the tendon. It is this peritendinous tissue that has the potential for an inflammatory response, and thus inflammation of tendons that lack a synovial sheath (e.g., Achilles) is more accurately described as either peritendinitis22-24 or paratendinitis.12,25 The latter term appears to be less descriptive because studies have shown that in cases of “pure peritendinitis” the underlying tendon is entirely normal.23,24 It has been documented, however, that tendon ruptures often are preceded by recurrent episodes of peritendinitis.22,24,26 In these cases, an inflammatory process occurs in the peritendinous tissues and degenerative changes also occur in the tendon itself. Puddu and colleagues defined this situation as a “peritendinitis with a tendinosis.”24 Tendon ruptures also occur with no previous history of peritendinitis.24,26,27 Surgical specimens from these cases demonstrate microscopic evidence of degeneration of the tendon but no associated peritendinitis.24,27 Thus, we agree with others who suggest that the term tendinosis best describes these latter pathologic changes in the tendon.22-24

ANTERIOR TIBIAL TENDON INJURIES Disruptions of the anterior tibial tendon are uncommon and most often are the result of lacerations or open injuries.1-3,5,28-30 Anagnostakos and colleagues31 identified 111 previous reports of anterior tibial tendon disruptions in their review of the literature and report of 3 additional cases. Anzel and colleagues reviewed 1014 cases of muscle and tendon injuries treated at the Mayo Clinic and found that only 10 (9 of which were open injuries) involved the anterior tibial tendon.1 In contrast, Hovelius and Palmgren29 and Simonet and Sim30 reported on a total of seven boot-top tendon lacerations in ice hockey players. In Hovelius and Palmgren’s series of two cases, the anterior tibial, extensor hallucis, and extensor digitorum tendons were all cut, as were the deep portion of the peroneal nerve, the anterior tibial artery, and the saphenous vein.29 Similar structures were found lacerated in Simonet and Sim’s series of five cases.30 In their series, the diagnosis of the tendon lacerations was missed by the initial treating physician in two of the five cases. They concluded that the lack of the apparent seriousness of the skin ­laceration

1976 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

c­ ontributes to this error. They recommended that all boottop lacerations be thoroughly explored.30 Both reports stressed that this injury could be prevented if players wore their skate tongues in the up position, laced the skate to the top, and put the tongue under the shin guard.29,30 In both series, none of the injured hockey players were wearing the tongue of the skate in the up position. Closed disruptions are uncommon, with only a limited number reported to date in the literature,5,9,13,14,18-20,28,31-39 and only six of these injuries occurred in athletes: three during cross-country skiing,33,38,40 one during alpine skiing,41 two during fencing.31 The alpine injury occurred in a 45-year-old man who fell while downhill skiing,41 and the others occurred in a 72-year-old retired physician who caught the tip of his ski while cross-country skiing,40 a 40-year-old lawyer who was attempting the skating technique for the first time while cross-country skiing,33 and a 68-year-old man who was attempting cross-country skiing for the first time and fell, which led to forced plantar flexion of his foot.38 Both fencing injuries occurred during a lunge maneuver with both athletes (67 and 53 years old) feeling acute sharp pain over the anterior ankle.31 Additionally, Hamilton and Ford42 reported a case of a longitudinal tear within the substance of the tibialis anterior tendon in a 60-year-old woman who had a painful mass and swelling over her anterior ankle. These reports indicate that diagnosis of a closed disruption often is delayed because the inability of the patient to dorsiflex the foot forcefully is interpreted as an acute neurologic problem.5,34 Attention to the details of the history of onset and a thorough physical examination of the foot and ankle, however, will delineate the problem.

Pertinent Anatomy The main extensor muscles of the foot and ankle are the tibialis anterior, extensor hallucis longus, extensor digitorum longus, and peroneus tertius. At the level of the ankle joint, the tendons of the tibialis anterior, extensor hallucis longus, and extensor digitorum longus are immediately adjacent to the distal tibia and pass under the superior extensor retinaculum (transverse crural ligament) and the inferior extensor retinaculum (cruciate crural ligament) in tight, fibro-osseous canals. Each of these tendons not only has its own compartment but also has a separate synovial sheath. The tibialis anterior tendon, a subject of this chapter, passes behind the superior retinaculum and lies within the most medial compartment of the inferior extensor retinaculum. After crossing the front of the ankle joint, it inserts into the medial and plantar aspects of the medial cuneiform bone and the adjacent base of the first metatarsal. At least seven insertion types for the tibialis anterior tendon have been identified. Musial43 described four types of insertion, all with two slips, with a wide insertion on the medial cuneiform and a narrow insertion onto the first metatarsal (type 1, 56.6%), two slips of equal width inserting on the medial cuneiform and first metatarsal (type 2, 37.7%), principal insertion onto medial cuneiform with accessory slips to the first metatarsal (type 3, 4.1%), and a wide insertion onto the first metatarsal and a narrow one onto the medial cuneiform (type 4, 1.6%). Arthornthurasook and Gaew-Im44 identified an additional configuration where there was only a single slip inserting onto the medial cuneiform (15.9%).

Brenner45 found ­ similar insertion sites with differing frequencies but also found two cases of tendons inserting only onto the first metatarsal. Luchansky and Paz46 identified an additional variant with the primary tendon inserting on both the medial cuneiform and first metatarsal with a second tendon originating from the primary tendon near the cuneiform and inserting onto the first metatarsal. Anagnostakos and colleagues identified similar insertions for the tibialis anterior tendon, with 68% of tendons inserting into both the medial cuneiform and first metatarsal.31 The tibialis anterior (the largest anterior muscle of the leg) is the primary dorsiflexor of the foot, and because of the tendon’s medial site of insertion, it also adducts and inverts (supinates) the foot. The tendon, however, is not a very powerful adductor owing to its proximal insertion on the foot. The blood supply of the tibialis anterior tendon may be a causal factor in spontaneous rupture of the tendon. One study, which only used injection techniques, found that the anterior tibial tendon had a complete blood supply without any evidence of an avascular zone.47 A more recent study that employed histochemical methods and injection techniques demonstrated an avascular zone at the usual site of rupture, which is 5 to 30 mm from the site of insertion of the tendon.48 Specifically, they found that the blood supply of the proximal part of the tendon arises from the anterior tibial artery. Distally, the tendon is supplied by branches of the medial tarsal arteries. The branches of both arteries reach the tendon through vincula from the posterior side. Distribution of the blood vessels within the tendon, however, is not homogeneous. The avascular zone is located at the level of the superior and inferior retinaculum. This is the site where the tendon changes its direction of traction and where closed disruptions usually occur.48

Clinical Evaluation In general, closed disruptions of the anterior tibial tendon occur in individuals older than 45 years of age, and surgical specimens have documented that degeneration of the tendon often precedes the disruption.5,13,19,20,33,34 The rupture is usually the result of minor trauma (such as stumbling over a door jamb or a stone), resulting in forceful plantar flexion of the foot. Degenerative tears after corticosteroid injections also have been reported.31,33,49,50 Other reported causes of spontaneous rupture include systemic lupus erythematosus,51 hyperparathyroidism,52 chronic acidosis of lead nephropathy,53 diabetes mellitus,54-56 rheumatoid arthritis,18,57,58 psoriasis,59 and gouty tophi.60 With this injury, an individual usually experiences a transient twinge of pain in the anterior aspect of the ankle but is able to continue his or her usual activities. The injury usually does not produce discoloration or swelling, and thus it ultimately is manifested only by a painless, flatfooted (slapping) gait. Because of the lack of awareness of this rare condition, late diagnosis of this injury occurs in most cases.20,32,33 The differential diagnosis for this condition often includes peroneal nerve palsy, compartment syndrome, lumbar disk herniation, tendinitis, insertional tendinopathy, and neoplasm.31 In one series, time to diagnosis averaged 21⁄2 months (range, 0 to 8 months).33 Most individuals seek medical evaluation only when they or others notice the abnormal gait pattern, often many weeks after the injury occurred.

Foot and Ankle 1977

On physical examination, it will be noted that the patient stands with the foot everted and that there is a loss of dorsiflexion when the patient attempts to walk on the heels. The degree of footdrop is variable and depends on the amount of time that has elapsed since the injury. Within weeks of the rupture, the proximal end of the tendon will become reattached to the surrounding structures and will provide variable power for dorsiflexion of the ankle. On examination of the ankle, however, one will find a visible and palpable defect in the tendon just distal to the transverse retinacular (cruciate crural) ligament, or in the case of tenosynovitis or a longitudinal tear, a mass may be palpated.42,61 On active dorsiflexion of the foot, one may feel the tendon curl up and produce a mass in the anterior leg. This mass disappears with relaxation of the muscle. If there is a question about the cause of the footdrop, electrodiagnostic evaluation will rule out other causes (e.g., peroneal nerve palsy) for this abnormality. Standard anteroposterior and lateral radiographs of the ankle should be obtained to rule out an avulsion fracture at the tendon’s insertion on the navicular bone.19,62 Although diagnosis can usually be established by physical examination, sonographic studies will demonstrate discontinuity of the tendon fibers separated by a hypoechoic defect.49 Magnetic resonance imaging (MRI) has been advocated for diagnosis of both partial and complete anterior tibial tendon ruptures when clinical history and physical findings are confusing.40,49,63

Treatment Options Treatment of open injuries and lacerations is surgical and consists of end-to-end repair with nonabsorbable sutures.2,3,5,28-30 Closed ruptures have been treated successfully with both operative2,5,14,18,19,28,31,33,35,36 and nonoperative regimens.9,64 Moskowitz reported on two patients treated with orthotics (two men, 69 and 70 years of age) and found that 1 year after injury they both walked with an almost normal gait, experiencing only slight difficulty in walking on their heels.13 He concluded that surgical repair “hardly seemed indicated” in elderly patients. Markarian and coworkers18 found no difference in outcome in a comparison of operative and nonoperative treatment in a group of 16 patients but recommended surgical repair in acute cases and in the delayed management of young, highdemand patients. Many authors, however, have concluded that surgical repair is the best treatment for acute (≤4 weeks) and subacute (≤3 months) injuries because it prevents the complications of nonoperative treatment, which include a flatfooted gait, decreased ankle motion, heel cord contractures, and a flat, pronated foot.2,5,9,14,18-20,28,33,35,36 These studies document that restoration of normal function is more rapid with surgical repair of the tendon. Acute injuries are best treated with primary, end-to-end repair of the tendon when possible, whereas tendon-lengthening procedures may be required for a larger defect.18,31,33,65 Wong66 treated a chronic traumatic tear in which a 7-cm defect was bridged by harvesting a 7-cm free tendon graft from the proximal stump of the tibialis anterior and then attaching in an end-to-end fashion to bridge the defect. This same technique has been used to treat smaller defects and has the advantage of no donor site morbidity.67,68 After ­ surgery,

the ankle is immobilized (in dorsiflexion) in a short leg cast for 6 weeks; return to strenuous activity (including mountain climbing) ranges from 14 to 24 weeks.5,9,19 Appropriate treatment of avulsion fractures from the navicular bone is predicated on the amount of displacement. Scheller and colleagues, who stated that they have observed this injury in athletes, recommend closed treatment if displacement is less than 5 mm and open reduction and internal fixation if displacement is greater than 5 mm.62 Longitudinal split tears within the tibialis anterior tendon are rare injuries, and when conservative management has failed to provide relief, surgical intervention has been shown to be effective. Hamilton and Ford42 reported on the surgical management of a 60-year-old woman with four to five longitudinal tears within the tendon just proximal to its insertion. The tears were treated with débridement and tubularization of the tendon with nonabsorbable 2-0 suture. The patient was treated in a splint for 2 weeks, returned to regular shoewear at 6 weeks, and by 3 months had painless, full range of motion and strength compared with the contralateral foot.

Authors’ Preferred Method Our preferred method of treatment is predicated on the patient’s age, level of activity, and the time elapsed from injury to evaluation. If the patient is not active and is in the sixth or seventh decade of life, the results of operative and nonoperative treatment are presented to the patient, and nonoperative treatment is recommended. In active individuals (regardless of age), operative treatment is recommended. We perform an end-to-end repair with a Krakow-type stitch of 2-0 nonabsorbable suture. The periphery of the tendon subsequently is sutured with a running absorbable 3-0 suture. If the tendon ends cannot be brought together, reconstruction of the tendon should be performed. Several reconstructive techniques have been advocated.18,20,28,36,50 The technique using the tendon of the extensor hallucis longus as a graft, as described by Moberg, appears attractive.19 To date, however, we have not had a case for which this or other reconstructive techniques were needed. After end-to-end repair, the ankle is immobilized in a neutral position with either a cast or a removable plastic orthosis for 3 weeks. At that time, the orthosis is removed several times each day for active range of motion exercises. Six weeks after surgery, all immobilization is discontinued, and stretching exercises are initiated. Return of full function usually is achieved within 12 to 14 weeks of surgery.

POSTERIOR TIBIAL TENDON INJURIES Injuries of the posterior tibial tendon include closed disruptions, anterior dislocations, and stenosing tenosynovitis. Isolated anterior dislocations are rare injuries and have been reported infrequently in the literature with mechanisms often including twisting the ankle while running or

1978 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

inverting the ankle during dismounts or ballet maneuvers when the foot becomes inverted and either dorsiflexed or plantar flexed with contraction of the posterior tibial tendon.69-72 In a large systematic review, Lohrer and Nauck73 identified 61 cases of posterior tibial tendon dislocation in the English, French, and German literature.72-107 They found that more than 75% of the dislocations were caused by trauma, and when specified, 58% had occurred with sporting activities. The injury was more prevalent in males, and the average age was 33 years. At least 50% of the patients were misdiagnosed initially with ankle sprains or posterior tibial tendon dysfunction. Patients often complain of persistent pain, swelling, and tenderness over the medial malleolus following an inversion injury. In 58% of the patients, a cord-like structure could be felt passing over the medial malleolus. The patients have often undergone conservative care, including periods of immobilization and physical therapy for suspected ankle sprains, and may have imaging studies in which tendon subluxation or dislocation was not present or was not recognized. Lohrer and Nauck identified 10 cases75,79-81 in which the patients were treated conservatively, whereas results were reported for only four patients, with three excellent results and one fair result. Surgery was required in the other cases, with retinacular reconstruction in 42.9%, primary repair of the retinaculum in 32.7%, and groove deepening in 18.4% of patients. Patients were typically immobilized in either a cast or walking boot for an average of 4.8 ± 1.9 weeks, and full weight-bearing was achieved an average of 3.6 ± 2.4 weeks after surgery. Excellent results were identified in 80%, good in 12%, and fair in 8% of the patients at final follow-up. The frequency of closed disruptions has only recently been appreciated.16,17,108 In 1983, Johnson reported the results of 11 surgical repairs of ruptured posterior tibial tendons that were performed over a 3-year period.16 In 1985, Mann and Thompson109 reported on 17 cases that were treated over a 4-year period and noted that the results of only 50 operatively treated disruptions had been previously reported.11,15,16,110-113 In 1991, Woods and Leach reported on six athletic people, 20 to 50 years of age, who required surgery for partial (three cases) or complete ruptures of the posterior tibial tendon.108 Holmes and Mann reported on 67 patients diagnosed with rupture of the posterior tibial tendon.114 They divided their 67 patients into older (>50 years of age) and younger (<50 years of age) groups. The profile of the older group was significantly different in that 60% of the cases were associated with hypertension, diabetes, and obesity. Younger patients had no significant association with the aforementioned problems, but 5 of these 14 patients had significant trauma to the foot.114 All the aforementioned studies stressed that closed disruptions often are not diagnosed because the signs and symptoms, although pathognomonic, are not recognized. In Mann and Thompson’s series of patients, all had seen one or more physicians (in only two was the proper diagnosis made), and the average delay in diagnosis was 43 months (range, 1 month to 12 years). Average time to diagnosis in Johnson’s series of patients was 19 months; he noted that 10 of his 11 patients had had previous medical evaluations, but in no instance was disruption of the posterior

tibial tendon recognized.16 Woods and Leach noted that the delay in diagnosis in five of their six athletes (range, 6 to 18 months) adversely affected the results of operative treatment.108 They concluded that the key to successful treatment, particularly in athletic individuals, is to make the diagnosis early, before deformity occurs.

Pertinent Anatomy The posterior tibial tendon is formed in the distal third of the leg by the convergence of the large muscle units that arise more proximally from the posterior surfaces of the tibia and fibula and the adjacent interosseous membrane. At the level of the medial malleolus, the posterior tibial tendon is the most anterior structure, located immediately adjacent to the posterior surface of the malleolus. The other structures, in order from anterior to posterior, are the flexor digitorum, the posterior tibial artery, vein, and nerve, and the FHL (one mnemonic for remembering the order of these structures is “Tom, Dick, and A Very Nervous Harry”). In this region the structure of the posterior tibial tendon changes to a more fibrocartilaginous structure with a change in cellularity and extracellular matrix components.115 The ability of this altered structure to resist repetitive microtrauma may make this area more vulnerable to spontaneous rupture.115 Histologic analysis of spontaneously ruptured tendons has not shown normal tendon structure; instead, degenerative tendinosis, mucoid degeneration, tendolipomatosis, and calcifying tendinopathy have been identified.116,117,184 The posterior tibial tendon has numerous insertion sites on the plantar aspect of the foot. These include the inferior aspect of the navicular bone, all three cuneiform bones, and the bases of the second, third, and fourth metacarpals.118,119 The incidence of an accessory navicular or os tibial externum, a sesamoid bone invested in the posterior tibial tendon near its insertion, is 10% to 20%.118-120 If present, it may be a small accessory bone within the posterior tibial tendon without attachment to the navicular (type I); an accessory navicular with a synchondrosis (type II); or a cornuate bony, navicular tuberosity (type III).118 The tibialis posterior is the main dynamic stabilizer of the hindfoot against valgus (eversion) forces, owing to its multiple insertions, which are variously taut at different positions of foot function, and to the size and strength of its muscle. Disruption of the tendon, therefore, results in elongation of the ligaments of the hind and midfoot and a painful flatfoot deformity. Variations in the vascular anatomy of the flexor tendons also may play a role in the frequency and location of injury. In a recent study, Frey and colleagues used injection techniques to examine the intrinsic vascularity of the posterior tibial and flexor digitorum longus tendons in 28 cadaveric limbs.121 In all 28 specimens, they found that the posterior tibial tendon had a zone of hypovascularity in its midportion. The zone started about 4 cm from the tendon’s insertion on the medial tubercle of the navicular bone and ran proximally for an average of 14 mm. Their dissections confirmed that there was no mesotenon present in this region and that the visceral layer of synovial tissue was present but hypovascular. Although this zone of hypovascularity may predispose the tendon to

Foot and Ankle 1979

­ egenerative processes, it is proximal to the most comd mon site of rupture of this tendon (1.0 to 1.5 cm from its insertion into the navicular) reported in the literature.15-17 In contrast, the midportion of the flexor digitorum longus tendon has ample vascularity throughout its length, owing to a longitudinal system of vessels arising from the proximal and distal arteries.121

Clinical Evaluation The hallmark of a posterior tibial tendon disruption is the progressive deformity of the foot. Patients usually present with a painful flatfoot and note that they cannot walk “normally,” have difficulty going up and down stairs, and have lost control of their foot. Disruptions most commonly occur in women older than 40 years of age, and many patients cannot recall any significant previous acute trauma.16 Most patients do, however, have a history of antecedent tenosynovitis108 and state that the problem began after twisting the foot, slipping off a curb, or stepping in a hole.17 Most do not seek emergent care because their pain, located medially between the tip of the malleolus and the navicular bone, is not incapacitating.16,17,108 Salient physical findings include valgus of the hindfoot, abduction of the forefoot, a positive single-heel rise test, a too-many-toes sign, and loss of function on manual testing of the tendon. The concurrent development of abnormal hindfoot valgus and forefoot abduction produces the flatfoot deformity. These deformities are best evaluated by observing and comparing the posterior aspects of both feet with the patient standing. Patients with tendon ruptures have either a unilateral flatfoot or, in those who had previous flat feet, a relatively flatter foot on the involved side. Excessive forefoot abduction also can be suspected from posterior observation when more toes are visible lateral to the patient’s heel on the involved side. This finding is called the too-many-toes sign (Fig. 25D-1).

Figure 25D-1   The too-many-toes sign. More toes are visible lateral to the heel on the patient’s right side.

The best method for evaluating loss of posterior tibial function is the single-heel rise test.16 In this test, the patient is instructed to stand on one leg with the knee extended and to rise onto the ball of the foot and the tips of the toes. The test is positive if the hindfoot (heel) on the involved side fails to invert and assume a stable (varus) position (Fig. 25D-2). The patient also will have great difficulty in raising the heel off the floor because gastrocnemius muscle function is compromised when there is no posterior tibialis to bring the heel into a locked, stable varus position.16 Manual testing of the strength of the posterior tibial tendon can be misleading because of the inversion strength of the tibialis anterior. This latter muscle and its tendon effectively resist eversion of the foot when the foot is in full inversion. Thus, loss of posterior tibial function is best assessed by everting the heel and placing the forefoot in full abduction. The patient is then asked to invert the foot. The presence of posterior tibial weakness will be apparent with this maneuver, particularly if one compares it with the strength of the opposite foot. Standard weight-bearing anteroposterior and lateral radiographs show variable changes depending on the time elapsed since the injury. Although several studies state that standard radiographs are not useful in making a diagnosis of posterior tibial tendon disruption,11,17,122 Johnson concluded that standard films usually show pronounced changes.16 Specifically, an increase in the talocalcaneal angle and inferior subluxation of the talus at the talonavicular articulation are evident on the lateral view. On the anteroposterior view, subluxation of the forefoot at the talonavicular articulation and an increased talocalcaneal angle are evident.16 In Mann’s study, the lateral talar-first metatarsal angle averaged 22 degrees.17 This angle is 0 degrees in normal feet, 1 to 15 degrees in mild flatfoot deformities, and more than 15 degrees in severe flatfoot deformities.123 Ancillary studies such as computed ­tomography (CT) and MRI are rarely indicated. CT scans should be obtained in feet with rigid deformities to rule out tarsal coalitions. Although MRI demonstrates the site or sites of tendon disease, it is not required to establish the diagnosis of posterior tibial tendon disruption.

Figure 25D-2   Single-heel rise test. The right heel does not assume a stable (varus) position (as seen on the left) when the patient attempts to stand on the toes.

1980 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Treatment Options Nonoperative treatment of posterior tibial tendon disruption is appropriate in patients who have low activity levels, are in their 60s or 70s, have life-limiting medical problems, or have minimal discomfort from their acquired deformity. If the injury is diagnosed acutely, immobilization of 4 weeks’ duration in a short leg cast that is well molded along the longitudinal arch is recommended.11,12,15,16 In subacute cases, the use of an orthosis (a flexible leather arch support with a 3⁄16-inch heel wedge) is appropriate if the patient’s only complaint is that he or she is wearing out the medial counter of his or her shoe because the results of later surgery will not be compromised if the foot is flexible.15,17 In most patients, however, the prognosis for an unrepaired posterior tibial tendon is poor because of the unopposed pull of the peroneus brevus tendon and loss of hindfoot inversion and the inevitable development of a painful flatfoot.15,124 Goldner and colleagues15 reported on three patients with untreated posterior tibial tendon lacerations, all of whom reported progressive loss of the arch and weakness with increased physical activity. Thus, early surgical repair or reconstruction is the best treatment, particularly in athletic individuals.108,125,126 Appropriate surgical treatment of a posterior tibial tendon disruption is predicated on the site and extent of the tear, the flexibility (duration) of the foot deformity, and the expectations of the patient. Based on these factors, there are four distinct surgical procedures: (1) end-to-end suture, (2) reattachment or advancement of the tendon to its primary site of insertion on the navicular bone, (3) reconstruction of the tendon, and (4) a triple or limited arthrodesis. Primary end-to-end repair often is not possible owing to retraction of the proximal segment of the tendon.11,17,113 In Johnson’s series of 11 patients, primary repairs were performed successfully in three cases,16 but in other series of patients (Mann and Thompson [17 patients],17 Wren [12 patients],113 Kettelkamp and Alexander [3 patients]11), end-to-end suture was not possible, even when surgery was performed within 7 days of the injury.11 Advancement of the tendon and tendon transfers are indicated when the posterior tibial tendon is found to be within its sheath but is partially torn and lengthened, or is ruptured within 3 cm of its distal insertion. In the latter situation, advancement of the posterior tibial tendon and anchoring the tendon through a vertical drill hole in the navicular bone may be indicated. One must ascertain, however, that the proximal part of the tendon moves freely within the tendon sheath, particularly at the medial ­malleolus, and that there is good elasticity of the muscle­tendon unit. Tendon transfers are often indicated even when the proximal end of the tendon is retrievable because in most cases there are more proximal segments of degenerated tendon.15-17,110,113 In these cases, resection of the degenerated segments usually does not allow an end-to-end repair of the tendon. Some studies have found that tendon transfers are not necessary for the aforementioned problem and have bridged the gap and reconstituted the function of the posterior tibial tendon with Z-plasty lengthening or turndown techniques.11,127 Because of the short excursion of the

posterior tibial tendon, however, a Z-plasty lengthening of more than 1 to 2 cm will cause a functional deficit.116 The flexor digitorum longus is most often used in ­tendon transfers because of its size, strength, and ­location.15-17,108,110,113 When the flexor digitorum longus tendon is used, it is transected just distal to the medial malleolus, and the proximal segment of the tendon is placed within the posterior tibial tendon sheath and tenodesed to the distal end of the posterior tibial tendon.17 The distal stump of the flexor digitorum longus is tenodesed to the adjacent FHL. Tenodesis of the more proximal segment of the flexor digitorum longus to the musculotendinous junction of the posterior tibial tendon is indicated only when the posterior tibial muscle has normal color and good elasticity.15-17 When scarring within the sheath of the posterior tibial tendon is too extensive to permit free motion of the transferred flexor digitorum longus, the latter tendon is left in its own sheath and is passed dorsally through a vertical drill hole in the navicular and sutured back on itself. The flexor digitorum longus tendon should be sutured under maximal tension while the foot is held in maximal inversion and plantar flexion. Most studies recommend the use of the flexor digitorum longus as the graft because although it is only one third the diameter of the posterior tibial tendon, it is only slightly smaller than the diameter of the peroneus brevis. The peroneus brevis is the strongest everter of the foot, and in patients with posterior tibial tendon ruptures, its unopposed function results in the flatfoot deformity.16,17,113 In contrast, Goldner and colleagues recommend transferring the FHL when advancement or end-to-end suture of the posterior tibial tendon is not advisable.15 They note that the key to success of this transfer is to section the FHL tendon proximal to its attachment to the flexor digitorum so that the flexor digitorum will provide flexion of the big toe and prevent hyperextension of the distal phalanx. Other authors do not use the FHL because they believe it plays an important role in push-off and in stabilization of the longitudinal arch of the foot.16,17,113 Tendon transfers are not appropriate when there are severe fixed deformities of the foot and secondary degenerative changes in the talonavicular or talocalcaneal joints. In these cases, tendon transfers will not correct the deformity or alleviate the patient’s pain, and a limited or triple arthrodesis is indicated. A triple arthrodesis should be performed when there are degenerative changes in the talonavicular and calcaneocuboid joints and fixed medial-inferior rotation of the head and neck of the talus. A single talocalcaneal arthrodesis is appropriate when the talonavicular and calcaneocuboid joints do not have secondary degenerative changes.16,17 If the primary deformity is abduction of the forefoot, and the foot is flexible and can be placed in its normal position, a talonavicular and calcaneocuboid fusion (double arthro­ desis) can be performed and will produce the same result as a triple arthrodesis.16 In Johnson’s series of 11 patients, two patients with fixed deformities were treated by single (subtalar) arthrodeses,16 and in Mann and Thompson’s 17 patients, one had a double (calcaneocuboid and talonavicular) arthrodesis after an unsuccessful flexor digitorum longus tendon transfer.17

Foot and Ankle 1981

Authors’ Preferred Method To date, we have not found it necessary to perform tendon transfers in the limited number of patients we have treated with posterior tibial tendon disruptions. In these few cases, posterior tibial function has been restored either by end-toend suture or by advancing and reattaching the tendon to the navicular bone as described earlier. One should ­always be prepared, however, to perform tendon transfers when one operates to restore posterior tibial tendon function because, as noted previously, the surgical findings indicate which procedure is most appropriate.

POSTERIOR TIBIAL TENDINITIS Based on our own experience and a review of the literature, it is apparent that tenosynovitis of the posterior tibial tendon is far more common than a total disruption of the tendon.62,112,128-131 Patients with this problem typically are women (in their fourth to sixth decades of life) who have occupations that require long periods of standing. In general, onset of symptoms is gradual, and the pain is aggravated by prolonged walking and standing. Posterior tibial tendinitis is often an initiating event in the development of adult-acquired flatfoot deformity. There is a paucity of literature that addresses this problem in younger individuals, however. Goldner had three patients younger than 20 years of age with tibialis posterior dysfunction in his series. In all three patients, trauma was the cause of their dysfunction.132 In nonathletic individuals, the pain from their posterior tibial tendinitis is not disabling and has been present for several months before the patient comes in for evaluation. The problem is less common in younger patients. The incidence has been reported as 2.3% to 3.6% in runners presenting to sports medicine clinics.133,134 In athletic people, the pain is debilitating because they are unable to run properly or perform any athletic activity requiring a strong push-off action.108,128 The pain, which is usually localized to the posterior tibial tendon in the vicinity of the medial malleolus, increases with prolonged and strenuous ­activity.

Clinical Evaluation Physical examination reveals medial ankle swelling and, in later stages, flattening of the arch of the involved foot. Palpation along the course of the posterior tibial tendon causes pain, and the site of maximal tenderness usually is located behind the medial malleolus. The integrity of the posterior tibial tendon can be confirmed by the functional tests outlined earlier, but there will be relative weakness of the tendon compared with the contralateral side owing to pain. Specifically, lack of heel inversion when the patient attempts a single-leg toe raise indicates rupture of the posterior tibialis tendon. In contrast, patients with tendinitis demonstrate heel inversion with this test but are unable to perform the single-leg toe raise 10 or more times owing to posterior tibialis pain. In addition to tenderness, swelling is usually found along the lower part of the tendon, and there may be palpable crepitus over the tendon on active

­ ovement. Active inversion and passive eversion of the m foot usually reproduce the patient’s pain. Radiographic evaluation of these patients should include standard, weight-bearing anteroposterior, and lateral views of the foot, as well as anteroposterior, lateral, and mortise views of the ankle. The relationship between the bones of the foot, including the longitudinal arch (a 0 to 10 degrees lateral talometatarsal angle), is normal in posterior tibial tendinitis. MRI,135-138 ultrasonography,139-141 and scintigraphy142 have been advocated for evaluation of posterior tibial tendinitis. Several studies have documented that MRI is more effective for determining abnormalities within the posterior tibial tendon in cases of tendinitis than ultrasound or tenography.135,136,138 Other studies found that sonographic evaluations had greater accuracy (94% versus 66%) and sensitivity (100% versus 23%) than MRI for demonstrating posterior tibial tendon abnormalities,139-141 and may demonstrate tendon ruptures (two cases) not diagnosed by MRI.140 The exact role of these modalities will be defined only through prospective studies that demonstrate that the information gained from sonographic or MRI evaluation altered the patients’ final outcome. Posterior tibial tendinitis is usually evident by clinical evaluation, but in our practice we obtain an MRI to rule out osteochondral fractures of the talar dome, navicular stress fractures, and accessory navicular injuries. MRI is also useful to evaluate the posterior tibial tendon for skip areas and to determine the extent of internal tendinous degeneration when operative intervention is necessary for treatment of this problem. Tendon degeneration found at surgery is often much greater than that expected based on clinical evaluation. In the evaluation of posterior tibial tendinitis, it is also important to evaluate for the development of adult-acquired flatfoot deformity. Adult-acquired flatfoot deformity has been described as occurring in four stages.143 In stage I, the patient may or may not have flatfoot deformity but will present with tenosynovitis or tendinosis.16,109,143,144 Flatfoot deformity then develops as the posterior tibial tendon fails, and subsequent ligament failure (often involving the spring ligament supporting the talonavicular joint) occurs. Once the talonavicular joint fails and begins to subluxate, the interosseous ligament becomes involved, and the subtalar joint begins to subluxate. The talar head continues to migrate medially and plantar, leading to further ligamentous failure and progression of deformities at the naviculocuneiform and tarsometatarsal joints and failure along the entire medial arch. In stage II disease, the flatfoot deformity can be passively corrected with inversion of the talonavicular joint and correction of heel alignment. Radiographs of stage II disease will show uncovering of the talar head and forefoot abduction. In stage III disease, the flatfoot deformity is no longer correctable. Lateral pain may develop as subtalar subluxation advances and the hindfoot progresses into further valgus deformity.144,145 With stage IV disease, lateral talar tilt is noted at the tibial-talar joint.146

Treatment Options Initial treatment of this condition is immobilization for 2 to 6 weeks in a short leg cast or rigid plastic boot, along with anti-inflammatories, which in most cases relieve the

1982 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

pain.62,147,148 In several earlier studies, local steroid injections into the synovial sheath were administered to patients who did not respond to nonoperative (supportive casts and anti-inflammatory medications) treatment.130,131,149 These injections, however, have been associated with tendon rupture.110,112,114 In Trevino and associates’ series of eight operative cases, three of the five tendons that received local steroid injections before surgery were partially or completely ruptured.112 Based on our current knowledge of the deleterious effects of steroids on the strength of tendons, we feel that steroid injections are never indicated for treatment of this problem. Operative treatment of this condition was required in 12 of the 52 patients reported by Williams131 and also in other smaller series of patients. In most cases, the operation performed included a release of the tendon sheath, excision129-131 of the scar tissue, and a partial synovectomy. At surgery, the tendon had a normal appearance and was not opened. The reported results of these procedures were good in 19 of 20 patients.108,129,130,150 Crates and Richardson reported their results on seven patients who had failed at least 6 weeks of cast immobilization and were available for follow-up at 10.9 months after tenosynovectomy and débridement of partial tendon tears (3 patients).150a Three and one-half months after surgery, 6 of 7 patients were completely pain free and could perform a single-heel rise test. The single failure had had persistent symptoms for

12 months before surgery and significant intrasubstance degeneration of the tendon at the time of surgery and later required lateral column lengthening and tendon transfer for progressive disease. Some authors noted the frequent presence of a bulbous distal enlargement of the tendon causing triggering and recommended resection and thinning of the tendon down to the approximate diameter of the noninvolved proximal and distal portions.112,130,131 We have not observed this phenomenon but have found that patients with posterior tibial tendinitis often have an intact, normal-appearing tendon, even when areas of intrinsic degeneration are present (Fig. 25D-3A). Recently, tendoscopy has been advocated for treatment of posterior tibial tendon disorders.151-153 In 1997, van Dijk and colleagues reported the results of posterior tibial tendon sheath endoscopy performed in 16 consecutive patients.151 All patients had pain on palpation over the posterior tibial tendon, a positive tibial tendon resistance test, and local swelling. Indications for arthroscopy were diagnostic in 11, chronic tenosynovitis in 2, screw removal in 1, and posterior ankle arthrotomy for a loose body in 2. The results of surgery, performed through a two-­portal technique, were good in 3 of 4 patients with adhesions resected, and in the 2 patients who had a tenosynovectomy and tendon sheath release performed. All of these patients were free of symptoms at 1-year follow-up.151 Chow and colleagues153 reported on 6 patients with stage I ­

A

B

C

D

Figure 25D-3   Posterior tibial tendinitis. A, The surface of the tendon appears normal, but intrinsic degeneration is apparent when the tendon is opened. B, The degenerated central segment of tendon was excised, but the peripheral fibers were intact. C, The periphery of the tendon was closed with a running absorbable 2-0 suture. D, The retinaculum was closed at the medial malleolus to prevent subsequent subluxation of the tendon.

Foot and Ankle 1983

posterior tibial tendon dysfunction treated with tendoscopic débridement. No complications were reported, none of the patients progressed to stage II dysfunction, and the patients had smaller scars, less wound pain, and short hospital stays compared with patients with traditional open débridement.

Authors’ Preferred Method The posterior tibial tendon is exposed through a curved medial incision from its origin to its musculotendinous junction. The tendon sheath is then divided longitudinally, and hypertrophic synovium, when present, is excised. The tendon is then explored throughout its length to detect unanticipated, partial, or complete ruptures. If disruption of the tendon is encountered, end-to-end suture, advancement, and reinsertion into the navicular, or tendon transfers (as outlined earlier), are performed as indicated. When the tendon is intact, we recommend opening and exploring areas of nodularity, areas where pain was localized ­preoperatively, and areas where the MRI has demonstrated abnormalities because, as noted earlier (see Fig. 25D-3A), intrinsic degeneration may not be apparent on the surface. Subsequently, all degenerative tissue is removed with a ­curet or knife (see Fig. 25D-3B), and after débridement, the tendon is closed with a running absorbable 2-0 suture (see Fig. 25D-3C). Although several studies have concluded that releasing and not repairing the tendon sheath is appropriate,129-131 subluxation of the posterior tibial tendon has been reported as a complication of tarsal tunnel decompressions.154 Thus, the posterior tibial tendon sheath is routinely preserved, restored, or reconstructed at the medial malleolus to prevent this complication (see Fig. 25D-3D). The wound is then closed with a running, subcuticular, nonabsorbable 3-0 Prolene suture, and the ankle is immobilized for 2 weeks to allow the incision to heal. At that juncture, immobilization is discontinued, the suture is removed, and the patient is instructed and supervised in gaining full range of motion, full strength, and full function. Relief of symptoms and return to full activity can be anticipated within 12 to 14 weeks of surgery.

FLEXOR HALLUCIS LONGUS INJURIES There are a limited number of published reports about complete disruptions of the FHL tendon.1,10,125,155-161 Only eight cases of closed complete disruptions of the FHL tendon have been reported to date.155-162 Six of the eight individuals were injured during the following sports activities: diving from a board, marathon running, walking; and playing soccer.155-158,160,161 The average age of the eight individuals was 40 years (range, 27 to 54 years), and the sites of tendon injury were in the groove of the posterior process of the talus (one), under the sustentaculum tali (two), just distal to the knot of Henry (two), at the metatarsal head (one), and 0.5 cm proximal to the tendon’s ­insertion on the great toe (two). All eight FHL tendon

r­ uptures were repaired because of functional disability due to loss of great toe push-off. Six were repaired acutely, and two late, at 4 months and at 2 years. Although only three of the eight patients regained active interphalangeal (IP) joint flexion, all of the six athletes regained their preinjury level of performance. 155-158,160,161 Postoperative stiffness of the IP joint did not cause any functional deficit. Boruta and Beauperthuy reported on three patients with longitudinal split tears within the FHL tendon just distal to the knot of Henry.163 The patients were treated with release of the knot of Henry, débridement and repair of the longitudinal tears, and excision of the interconnections between the FHL and flexor digitorum longus tendons. All three patients obtained good long-term results. The frequency and causes of FHL tendon lacerations are well chronicled in two studies.10,125 In 1977, Frenette and Jackson reported on 10 young athletes (median age, 11 years) with FHL tendon lacerations.10 In 8 of their 10 patients, the laceration was caused when the athlete was running barefooted and stepped on a sharp object, usually broken glass. To find that number of cases, they surveyed 100 orthopaedic surgeons and did a 5-year review of the hospital records of four large hospitals. They found that three of the six primary repairs of this tendon resulted in no active plantar flexion of the IP joint of the great toe and concluded that repair of the FHL tendon was not essential for good push-off or balance in running sports. In the nonoperatively treated group, one athlete was a former Olympic high jumper and noted no change in jumping ability, whereas a second athlete went on to become the city 100-yard dash champion. However, one of the patients did develop a cock-up deformity that required an IP joint fusion and extensor hallucis longus transfer.10 In contrast, Floyd and associates reported on 13 cases of open FHL tendon lacerations and found that of 12 tendons repaired (10 sutured primarily, 2 delayed), 9 retained active motion of the IP joint of the great toe.125 Most of the reports on open or closed disruptions of the FHL tendon have concluded that an intact tendon is essential for good push-off and balance in running sports.130,155-161,164 Tendinitis of the FHL tendon is more common than complete disruption. Although cases of flexor hallucis tendinitis and tenosynovitis have been reported in tennis players and long-distance runners,165-167 it is much more prevalent, and is the most common site of lower extremity tendinitis, in classical ballet dancers.168-175 Washington’s survey study of musculoskeletal injuries in dancers puts the incidence of this problem in perspective.176 Of 414 injuries reported by individual dancers, 55 involved the ankle joint, and 3 of these 55 (5.5%) were diagnosed as tendinitis. Hamilton and colleagues noted that tendinitis around the ankle is common in classical ballet dancers and that the tendon involved is almost always the FHL.171 They stated that, in ballet, the FHL is the Achilles tendon of the foot when a dancer is en pointe.171 FHL tendinitis is more often seen in the left foot than the right because choreography more often calls for the dancer to turn to the right (clockwise). This requires the dancer to be en pointe on the left foot. Michelson and Dunn recently reported on 81 patients with FHL tenosynovitis.178 The patient population consisted of 55 women and 26 men with an average age of 38.3 years and with 33% involved in athletic activities. All

1984 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the patients had tenderness over the FHL, and 82% of the patients who had an MRI had evidence of synovitis. Nonoperative management consisted of a stretching program and short-term immobilization. Sixty-four percent of the patients treated nonoperatively had successful outcomes, whereas all 23 of the patients who underwent decompression and synovectomy had successful outcomes.

Pertinent Anatomy The FHL arises from the lower two thirds of the posterior surface of the shaft of the fibula, and the FHL tendon inserts into the base of the distal phalanx of the great toe. In the sole of the foot, the FHL tendon is connected by a fibrous slip to the flexor digitorum longus tendon. This slip (see earlier) tethers the FHL tendon and prevents excessive retraction of the proximal segment after the FHL tendon is severed. In the great toe, the FHL tendon lies superficial to and between the two heads of the flexor hallucis brevis. Thus, it is relatively easy to find the proximal end of the FHL tendon when it is severed in this location.10 At the level of the ankle joint, the FHL, posterior tibial, and flexor digitorum longus tendons pass under the flexor retinaculum (lancinate ligament), and septa from this strong fibrous retinaculum convert a series of bony grooves into fibro-osseous tunnels. The tendons also are enclosed in separate synovial sheaths, which are 8 cm long and extend proximal and distal to the aforementioned tunnels. The blood supply to the FHL tendon arises from the posterior tibial and medial plantar arteries.179 Injection and immunohistochemical studies revealed two avascular zones that correlated with the sites of frequent tendon degeneration and rupture: (1) as the tendon passes behind the talus and (2) around the first metatarsal head.179 The constrictive nature of the fibro-osseous tunnels can cause “triggering” of a flexor tendon when partial tearing and healing of the tendon produce exuberant scar tissue. The most common tendon involved is the FHL, and the most common precipitating activity is classical ballet dancing.168-175 The FHL tendon is further constrained by bony grooves in the posterior surface of the tibia, the talus, and the sustentaculum tali. Stenosing tenosynovitis has also been reported at the level of the sesamoids, and surgical tenolysis at this level has provided successful relief of symptoms in patients who did not respond to conservative management.180,181

Clinical Evaluation In evaluating patients with lacerations to the FHL tendon, it is important to evaluate for associated injuries as well. In two studies, injury to the digital nerve was found in 46% and 80% of the patients with open laceration of the long flexor.10,125 The injury to the digital nerve may be identified with a thorough physical examination for sensory loss as well as with wound exploration. The function of the flexor hallucis brevis tendon should be evaluated (ability to flex the metatarsophalangeal joint of the great toe) and documented. Additionally, the wound should be explored when there is concern for an open joint as well. Dancers with FHL tendinitis often note the insidious onset of pain at the posterior medial aspect of both ankles

behind the malleolus and often seek evaluation only when triggering of the FHL tendon produces so much pain that they are unable to dance en pointe.168,171,173-175 Sammarco and Miller reported on four ballet dancers who had triggering of the FHL; in two of these dancers, the hallux became locked in a flexed position.174 They noted that the triggering becomes more severe over a period of several months, but pain is not an outstanding characteristic of this condition.174 In Kolettis and associates’ series of 13 female ballet dancers that had operative release because of isolated stenosing tendosynovitis, all had pain and tenderness over the medial aspect of the subtalar joint.172 Their symptoms were exacerbated by jumping and attempts to perform en pointe work, and all 13 had lost the ability to stand en pointe. Crepitus was present in 6 patients, but triggering was present in only 3. None of their dancers had pain or tenderness in the posterolateral aspect of the ankle with forced passive plantar flexion, which would have suggested involvement of an os trigonum. In contrast, Hamilton and associates reported on 37 dancers who had 41 operations for posterior ankle pain. In their series, only 9 operations were performed for isolated tendinitis. Of the remaining 32 ankles, 26 were operated on for tendinitis and posterior impingement and 6 for isolated posterior impingement.171 Physical examination of patients with triggering of the FHL tendon reveals that the hallux can be flexed with ease when the foot is in a neutral position. When the foot is brought into plantar flexion, however, the patient is unable to flex the hallux. On forcible active contraction of the FHL, however, a snap or pop is noted in the posterior medial region of the ankle, and the patient is then unable to extend the IP or metatarsophalangeal joints of the great toe. Subsequent passive extension of the IP joint produces a painless snap or pop posterior to the medial malleolus with subsequent freeing of motion in the great toe. In dancers who have tendinitis without triggering, localized tenderness and swelling are present, occasionally with crepitus over the FHL tendon just posterior to the medial malleolus.169 Hamilton stated that FHL tendinitis can be distinguished from tendinitis of other adjacent tendons by the following maneuver. The foot is placed in the pointe position, and the patient is asked to flex the great toe against mild resistance and then to flex toes two through five against resistance. Sequential palpation of the FHL and the flexor digitorum longus will reveal which tendon is locally tender and has the tendinitis.170,171 Differential diagnosis of FHL tendinitis includes peroneal tendinitis, posterior impingement of the os trigonum, acute fracture of the lateral tubercle of the posterior process of the talus, Achilles tendinitis, bone spurs, and arthritic or osteochondrotic lesions of the talus.170,171,182 Posterior impingement can be tested for by forcibly plantar flexing the ankle. When impingement is present, this maneuver usually produces pain in a posterolateral location.171,182,183 Achilles tendinitis can be differentiated from FHL tendinitis by carefully ascertaining the location of tenderness. In Achilles tendinitis, the pain is over the sheath of the Achilles tendon, not over the fibrous tunnel of the FHL. Osteochondritis and arthritic changes of the talus and bone spurs emanating from the tibia or fibula can be excluded by obtaining standard radiographs of the ankle. Hamilton

Foot and Ankle 1985

suggests that an additional lateral view of the ankle in the full pointe position should be obtained to evaluate the significance of bony exostoses.170

Treatment Options Although Frenette and Jackson found that repair of FHL tendon lacerations was not essential for restoring good push-off in running sports, they concluded that, in young athletes, all FHL tendon lacerations should be explored, and they recommended the following protocol based on the operative findings.10 If the tendon ends are easily identifiable, the tendon should be repaired using the same meticulous technique that one would use for flexor tendon lacerations in the hand. Repair is recommended because they found that adhesion formation was a problem, particularly when the laceration is at the level of the first metatarsal head. Patients treated in this manner were subsequently immobilized in a plaster cast and kept non–weight-bearing for 5 to 6 weeks. Second, if there is concomitant injury to the flexor hallucis brevis and the proximal end of the longus cannot be located, the brevis should be repaired by suturing the proximal end of the distal segment of the FHL tendon to the brevis muscle to prevent the possibility of a hyperextension deformity. Preoperatively, one may be able to establish the integrity of the flexor hallucis brevis by asking the patient to flex the big toe. If active flexion of the metatarsophalangeal joint is present without active flexion of the IP joint, the flexor hallucis brevis is intact. They also found that the first common digital nerve was divided when lacerations were located distal to the origin of the flexor brevis. In their series of patients, primary neurorrhaphy was attempted in four patients, but restoration of normal sensation was achieved only in one case. They concluded that nerve injuries often were associated with a hypersensitive scar but that the sensory deficit, although annoying, had little effect on total function. They recommended repair, however, of nerves during primary treatment if feasible or, if repair was not possible, resection of the proximal end of the nerve to prevent its entrapment in the scar and the formation of a neuroma in a weight-bearing area. Scaduto and Cracchiolo recommend surgical exploration and repair of the FHL tendon following a laceration, although the patient should be informed that results of repair of the tendon led to normal active flexion of the IP joint in only 61% of patients.164 The tendon should be repaired in an end-to-end fashion with a Kessler or ­Bunnell suture technique with 2-0 or 3-0 nonabsorbable suture. A short leg, non–weight-bearing cast, with the foot in mild equinus, extended to block great toe extension should be worn for 4 weeks. A neutral position weight-bearing cast should then be worn for an additional 2 weeks. Active flexion and extension of the hallux is permitted at 6 weeks, with gradual return to unlimited activity over the following 3 to 4 weeks. In patients with closed FHL rupture, surgical repair or reconstruction has been recommended to alleviate pain and not necessarily restore IP joint flexion. With closed ruptures, it is recommended that an MRI scan or ultrasound be obtained to localize the disruption and to evaluate the amount of distraction.164 Multiple surgical techniques

have been described in the literature, and primary end-toend repair is recommended when feasible.162,164 In cases in which significant distraction exists, techniques including a side-to-side repair of the proximal and distal stumps of the FHL tendon to the flexor digitorum longus tendon,157 grafting of the defect with a tensor fasciae latae graft,161 and a distal tenodesis of the distal FHL tendon to the flexor digitorum longus162 have been used to restore some active IP joint flexion. Treatment of FHL tendinitis is predicated on the stage at which it is diagnosed. In the acute phase, treatment includes anti-inflammatory medications and avoidance of activities that stress the FHL. Specifically, the individual may be allowed to continue dance workouts but should be instructed to avoid dancing en pointe. In most cases of tendinitis without triggering of the tendon, the pain will resolve within a matter of days to weeks.169-173 Prevention of recurrences should focus on instructing the dancer to reduce the amount of turnout at the hips so that they are working directly over the foot, avoiding hard floors (a major contributing factor), and using firm, well-fitting toe shoes so that the foot is well supported and no additional strain is placed on the tendon.170,171 When the tendinitis has been present for several months, a prolonged recovery period of up to 3 months should be anticipated. In these cases, a short period of immobilization (2 to 3 weeks) may be indicated. Chronic tendinitis is most often seen in relatively “tight” dancers, especially those with stiff feet and an incorrect pointe position.170,171 Operative treatment should be considered when persistent synovitis or triggering of the FHL tendon prevents dancing en pointe. The primary procedure performed is release of the tendon sheath. Hamilton stated that this operation is rarely if ever indicated and observed that dancers often say that the presence of scar tissue from surgery is almost as incapacitating as the tendinitis before ­surgery.170,171 In more recent reports, however, several authors have recommended exploration of the tendon and release of the tendon sheath in recalcitrant cases of flexor hallucis ten­ dinitis.158,168,169 Garth explored the FHL in both ankles of a 21-year-old ballet dancer who was unable to return to dancing after 2 years of treatment including intermittent rest and anti-inflammatory medications.169 At surgery, he found that both tendons were eroded beneath the retinaculum so that they were only 25% and 50% of their normal diameter. He used double strips of plantaris tendon to reinforce the FHL and released the flexor retinaculum. At follow-up 7 months after surgery, the patient had no pain and good FHL function but was not dancing en pointe. Cowell and colleagues explored the FHL tendon of an 18-year-old who had a 1-year history of tendinitis without triggering.168 At surgery, he found calcified nodules within the tendon that did not allow the tendon to glide freely in its sheath. He released the tendon sheath and excised the calcific nodules. The patient was able to return to dancing en pointe 5 months after surgery. Several studies have concluded that surgical treatment is indicated earlier in patients who have triggering of the FHL tendon compared with those who have pure tendinitis.172,174,175 Sammarco and Miller reported on two ballet dancers with bilateral triggering and locking of the FHL

1986 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

tendon.174 All four tendons in these dancers were explored, and in every case, there was a fusiform enlargement of the tendon just distal to the flexor retinaculum at the medial malleolus. There were ruptures of the central fibers in the enlarged area of the tendon. These fibers, which were thickened and contracted, were encompassed by peripheral, intact fibers of the tendon. In 1984, Tudisco and Puddu reported similar surgical findings in a 25-year-old professional ballet dancer who had a 2-year history of bilateral triggering of his FHL tendon.175 At surgery, the authors were able to demonstrate entrapment of the tendon within the flexor retinaculum by passively flexing the great toe. In both of the aforementioned reports, release of the flexor retinaculum relieved the symptoms and allowed the tendon to move smoothly in its tract. Sammarco and Miller concluded that the active dancer requires at least 3 months of slow, progressive rehabilitation of the ankle before he or she will be able to return to dancing en pointe.174 In contrast, Tudisco and Puddu’s patient, who began active range of motion at 3 days and walked unaided at 14 days, returned to dancing en pointe 40 days after surgery without problems.175 Operative treatment of dancers with tendinitis and posterior impingement or isolated posterior impingement requires individualizing the surgical approach.171,182 ­Hamilton stressed that a lateral approach to the ankle should only be used to treat isolated posterior impingement.171 He advocated a medial approach for treatment of both FHL tendinitis and posterior impingement because he

believed that the fibro-osseous tunnel of the FHL tendon cannot be released safely from the lateral side.171 Marotta and Micheli recommended a similar approach.182 They reported the results of 12 ballet dancers (15 ankles) that had excision of the os trigonum, through a lateral approach, for treatment of posterior impingement. There were two minor complications (one superficial wound infection and one transient tibial nerve neurapraxia). Both complications resolved without sequelae. At follow-up, 2 years after surgery, 8 (67%) still had occasional discomfort, but all 12 dancers returned to unrestricted dance activity.182 Recently, tendoscopy has been proposed as a possible treatment for FHL decompression.187,188 Keeling and ­ Guyton performed a cadaveric study to evaluate the utility of FHL decompression at the posterolateral talar process.189 They found that in 3 of 8 ankles, the tendon was injured during the release and that in no case was the sheath completely released. Additionally, the sural nerve and the lateral calcaneal branch of the sural nerve, as well as the first branch of the lateral plantar nerve, were closely associated to the tendon, and the portal locations placed these nerves at risk during endoscopic release. They thought that open release had a better reliability and less potential morbidity associated with it.189 For more distal disease, Lui and Chow have shown that tendoscopy can be effective in the management of toe flexor tenosynovitis in a 2-year follow-up study of 3 patients with tenosynovitis and metatarsalgia.188 No other reports in the literature could be identified to expand on these findings.

Authors’ Preferred Method Our preferred method of treatment of FHL tendinitis is based on the length of time the tendinitis has been present and whether there is triggering of the tendon. In the one professional ballet dancer treated by the senior author (JSK) for this problem, tendinitis of the FHL ­ tendons of

To date, the authors have not treated an athlete with a l­aceration of the FHL tendon; however, the operative protocol of Frenette and Jackson10 and Scaduto and Cracchiolo164 presented earlier appears sound and would be our approach to this injury.

A

B

Figure 25D-4   Flexor hallucis tendinitis. At surgery, a longitudinal tear and a bulbous enlargement of the flexor hallucis longus tendon were found just distal to the flexor retinaculum (A). With passive flexion of the toe, the enlarged portion of tendon became entrapped within the retinaculum. The degenerated fibers and the central scar tissue were excised, and the tendon was repaired (B).

Foot and Ankle 1987

Authors’ Preferred Method—cont’d both ankles was present for several months before the onset of triggering of the tendons. In the left ankle, the triggering subsided after 3 weeks of rest—specifically, ­ avoiding dancing en pointe—and the use of anti-inflammatory medications. In the right ankle, however, the triggering persisted and became so painful that the dancer was unable to rise onto or push off her right great toe. Thus, her right FHL was explored through a 5-cm incision that was located 2 cm distal to the tip of the medial malleolus (Fig. 25D-4). At surgery, it was evident that the peripheral fibers were intact and that there was a longitudinal tear and a bulbous enlargement of the tendon at the distal edge of the flexor retinaculum (see Fig. 25D-4A). After release of the tendon sheath, the central degenerated fibers and scar tissue

PERONEAL TENDON INJURIES Peroneal tendon injuries include ruptures, tendinitis, and acute or chronic subluxations. Subcutaneous ruptures of the peroneus brevis tendon (PBT) occur more often than peroneus longus tendon (PLT) tears, but both are uncommon injuries. A limited number of cases of complete traumatic peroneus longus tendon ruptures have been reported in the literature.35,190-199 Two involved soccer players,192,193 two occurred in runners,194,200 one occurred in a collegiate football player,197 and one in a walker.195 All six of these individuals twisted the ankle and felt a distinct “pop” at the lateral aspect of the injured ankle. Most of the cases occurred in older, less active individuals. One was attributed to overgrowth of the fibula after osteomyelitis of the tibia and a varus deformity of the foot,191 and two occurred in middle-aged women (ages 48 and 58 years) who had twisted their ankles.35,190 Disruption of the PLT due to a fracture of the os peroneum also has been reported.201-206 Most cases are the result of either repetitive inversion injuries or forced eversion of the ankle against resistance. Although PLT rupture with an associated os peroneum fracture is an uncommon occurrence, it should be included in the differential diagnosis of any patient with lateral ankle pain and instability and either of the aforementioned mechanisms of injury. Although longitudinal and interstitial tears of the PBT have been well described in the literature,195,196,207-214 complete ruptures of the PBT are less common, with a limited number of cases reported in the literature.195-197,215-217 Most of the PBT tears occurred in combination with partial or complete PLT tears.196,217-219 One occurred in a young, healthy collegiate athlete,197 two were discovered at the time of surgical treatment of the PLT tears,195,216 and another was found when the PBT was to be used for correction of chronic lateral ankle instability.215

Pertinent Anatomy The PLT and PBT cross the ankle joint within a common fibro-osseous tunnel and synovial sheath. The tunnel is created laterally by the superior peroneal retinaculum; medially by the posterior talofibular, calcaneal fibular, and

were excised, and the tendon was repaired with a running 2-0 absorbable suture (see Fig. 25D-4B). The sheath was left open, and the tendon glided smoothly throughout a full range of motion of the great toe. The subcutaneous tissues and skin were closed with interrupted 2-0 absorbable sutures and a running 3-0 nonabsorbable Prolene ­ suture, respectively. The ankle was immobilized for 2 weeks to ­facilitate wound healing, and floor exercises were initiated 1 week later. The patient was allowed to do a complete workout with the exception of going en pointe 8 weeks after surgery. Three months after surgery, she was able to resume en pointe positioning without fatigue or pain. At ­follow-up 1 year later, the patient was dancing en pointe without difficulty.

posterior inferior tibiofibular ligaments; and anteriorly by the posterior surface of the lateral malleolus. The superior peroneal retinaculum (SPR) is a strong fibrous band that blends with the periosteum of the lateral surface of the lateral malleolus and is rarely torn when subluxations or dislocations of the peroneal tendons occur.220 Eckert and Davis, who described the retinaculum origin from the periosteum of the posterior ridge of the fibula, found no actual tears of the SPR in their 73 repairs for subluxating peroneal tendons.220 Davis and colleagues delineated the different patterns of insertions and the relationship of the superior retinaculum to the peroneal tendons and the lateral ankle ligaments.221 They observed that the SPR has at least one insertional band that parallels and inserts just lateral to the calcaneofibular ligament and that these two structures are at maximal length in ankle dorsiflexion. They concluded that inversion ankle sprains that cause calcaneofibular ligament ruptures may often produce concomitant SPR injuries.220 Numerous studies have documented the relationship between lateral ankle instability and peroneal tendon disease.195,201,204,206,208,209,211-213,218 The inferior peroneal retinaculum is continuous (anteriorly) with the fibers of the inferior extensor retinaculum and is attached posteriorly to the lateral surface of the calcaneus. Some retinacular fibers attach to the peroneal tubercle of the os calcis, forming a septum that creates separate compartments for the peroneus longus and brevis tendons. A thin vincula-like structure that runs between the PLT and PBT and is dorsally attached to the dorsolateral aspect of the fibula has also been described.222 The peroneal tendons have a common synovial sheath until they separate under the inferior retinaculum. At that juncture, each tendon is enveloped by a distal progression of the common sheath. Proximal to the lateral malleolus, the PBT lies deep (medial) to that of the peroneus longus. Distal to the lateral malleolus, the PBT lies anterior to the peroneus longus. Knowledge of the changing spatial relationships of these tendons is important if one uses the PBT for lateral ankle reconstructions. The PBT inserts into the tuberosity of the lateral aspect of the base of the fifth metatarsal. The PLT exits

1988 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the ­ posterior compartment of the inferior retinaculum, crosses the lateral surface of the cuboid bone, and traverses the plantar surface of the cuboid in a tunnel created by a groove in the bone and the lateral plantar ligament. The tendon inserts into the lateral side of the base of the first metatarsal and the medial cuneiform. There is an abrupt change in direction of the PLT at two points: first at the tip of the lateral malleolus and second at the distal (lateral) edge of the cuboid bone. In both of these locations, the tendon is thickened, and at the point of contact with the smooth surface of the edge of the cuboid, there usually is a sesamoid within the tendon composed of fibrocartilage or bone. The blood supply of the PLT and PBT comes from the peroneal artery, and microvascular studies have documented that there are no critical zones of hypovascularity within these tendons that may be attributed to ­rupture.223,224

Relevant Biomechanics Several studies have noted the critical role that subluxation of the tendons over the posterolateral edge of the fibula plays in the development of peroneal tendon tears.208,209,211,213 Sobel and associates, in their study of 15 fresh-frozen specimens, demonstrated that PBT splits were uniformly located at the sharp posterior lateral edge of the fibula.213 They concluded that PBT tears were the result of either acute or repetitive mechanical trauma caused by subluxation of the tendon over the posterior corner of the lateral malleolus. Subsequent studies have supported this proposed mechanism of injury.208,211,212

Clinical Evaluation Diagnosis of complete peroneal tendon disruptions is difficult and often is delayed.35,190,192-194,197,200 In one soccer player, the diagnosis was not suspected and was discovered 10 hours after injury when an emergency fasciotomy was performed for an acute peroneal compartment syndrome.192 The PLT was found to be avulsed from the muscle belly, and subsequently the entire muscle became necrotic and was excised. Seven months after the injury, function of the peroneus brevis and the anterior compartment muscles was normal, and the patient had returned to playing soccer. In the other case involving a soccer player, the individual sought treatment 8 months after his injury because of recurrent, painful swelling at the outer aspect of the heel.193 At that time, he had full range of motion and full strength except for eversion, which was painful and weaker than normal. Ultimately, the tendon was explored because of his persistent tenosynovitis. At surgery, a bulbous enlargement and transverse tear of the tendon were discovered just below and behind the lateral malleolus. The enlarged portion of the tendon was excised, and the two ends of the peroneus longus were sutured to the adjacent tendon of the peroneus brevis. Two months after surgery, the patient had returned to full activity. The two runners with this injury were initially treated for lateral ankle sprains. The diagnosis of a complete PLT tear was made 4 weeks after injury when both sought a second opinion because of persistent pain and difficulty walking.194,200 One collegiate football player was found to

have complete ruptures of the PLT and PBT tendons after being treated for 2 years with taping and bracing for a lateral ankle sprain.197 In the two middle-aged women with this injury, the diagnosis of a PLT disruption was made 1 or more years after the injury. One was treated successfully with 3 weeks of immobilization,35 and the other was treated with endto-end repair of both the peroneus longus and brevis ­tendons.190 It is apparent from the reports cited earlier that rupture of the PLT is an uncommon injury and that diagnosis is delayed because the physical findings (pain and swelling about the lateral malleolus) usually imply a lateral ankle ligament injury. In addition, swelling often makes determination of point tenderness more difficult. Thus, one should look for increased hindfoot varus and test for pain with active eversion in all individuals with lateral ankle pain and tenderness proximally along the peroneal tendons. Although fractures of the os peroneum without disruption of the tendon occur186,225 (and are the subject of a subsequent chapter), the association of this sesamoid fracture with PLT disruption warrants further emphasis. The os peroneum, which is present in 5% to 26% of individuals, is found within the PLT at the level of the plantar and lateral aspects of the cuboid.206 One mechanism of injury, sudden forceful dorsiflexion of the foot and a violent reflex contraction of the peroneus muscles, is the same mechanism that produces a dislocation of the peroneal tendons.202,205 The other mechanism proposed (repetitive inversion injuries) is more subtle,201,204,205 and thus recognition of this injury requires knowledge of its existence, careful examination of the lateral aspect of the foot, and standard radiographs that include anteroposterior, lateral, and calcaneal views. The differential diagnosis for this combined injury includes lateral ankle ligament sprains, avulsion fractures at the insertion of the PBT, peroneal tendon subluxation, and tenosynovitis of the peroneal tendons. With a fracture of the os peroneum and disruption of the tendon, there is point-specific tenderness over the area of the sesamoid, and either wide separation of the fracture fragments or proximal migration of the sesamoid is evident on the standard radiographs.200,203-206

PERONEAL TENDINITIS Far more common than complete disruption of one or both of the peroneal tendons are interstitial tears and chronic tendinitis. The evaluation and management of chronic tendinitis and tenosynovitis of the peroneal tendons have been addressed in only a limited number of studies.62,112,190,191,210,217,226 Scheller and colleagues postulated that peroneal tendinitis is due to the pulley action and abrupt change in direction of these tendons at the lateral malleolus. 62 They suggested that the oblique stress placed on the tendon at the lateral malleolus produces an area of decreased vascularity that predisposes the tendon to injury. Burman concluded that tenosynovitis of the peroneal tendons occurs in three specific locations: posterior to the lateral malleolus (peroneal sulcus), at the peroneal trochlea (a tunnel created by the attachment of the calcaneal fibular ligament to the peroneal tubercle of the os calcis), and at the plantar surface of the cuboid.210 Burman also ­concluded

Foot and Ankle 1989

that stenosing tenosynovitis of the peroneus longus and brevis occurs mainly in individuals with well-developed peroneal tubercles, and noted that anatomic studies have documented that 37% to 44% of humans have prominent peroneal tubercles. He operated on 6 of 25 patients who had chronic peroneal tenosynovitis at the level of the peroneal tubercle. In all six cases, the tendons were found to be intact, but the fibrous sheath and fibrous septum of the retinaculum were thickened. After resection of the stenotic sheath, the patients experienced relief from their peroneal pain.210 Trevino and coworkers reported on four cases of peroneal tenosynovitis that were treated surgically.112 They noted that the site of involvement was either the peroneal sulcus (two cases) or the peroneal trochlea. Roggatz and Urban described three cases of peroneal tendinitis that were located distal to the peroneal trochlea at the lateral border or plantar surface of the cuboid bone.226 Peroneal tendinitis also can be caused by congenital anomalies of the peroneal tendons.177,227-229 Of the four cases reported to date, three had an anomalous bifurcation or trifurcation of the PBT,177,227,229 and one had an accessory peroneal muscle and tendon.228 In all cases, resection of the anomalous tendon and muscle produced good results. The evaluation and management of interstitial tears (splits) of the peroneus longus and brevis tendons have been addressed in an increasing number of studies.195,207-213 Interstitial tears of the PLT can be classified as acute or chronic. Acute tears occur after a single event, after which symptoms develop. With chronic tears, there is no acute traumatic episode, and symptoms are insidious in onset and develop over a long period.228 Acute tears may occur as primary injuries or may occur in conjunction with other ankle injuries. Bassett and Speer reported on eight athletes who sustained longitudinal ruptures of PLT (five patients) or PLB (three patients) as a result of a plantar flexion-inversion ankle injury.208 Sammarco and Brainard chronicled the evaluation and management of 14 cases of PLT tears, of which 8 had acute onset of symptoms.228 In these two series of patients, delay in diagnosis was common, ranging from 7 days to 6 months, and from 7 to 48 months, respectively.208,228 In Bassett and Speer’s study, all eight athletes noted persistent lateral ankle swelling, popping, and retrofibular pain after their initial sprain.208 They all complained of a subjective feeling of ankle instability but were able to compete in their sport by taping the ankle. On physical examination, all ankles were stable, and all had retrofibular tenderness, synovial sheath fullness, and palpable retrofibular popping with active foot rotation. None had evidence of clinical subluxation of the peroneal tendons. They and others209,230 have concluded that splits in the PLT and PBT that go unrecognized may be the cause of residual pain after ligamentous repairs of lateral ankle injuries. In the past decade, longitudinal tears (split lesions) of the PBT have been reported with greater frequency; however, despite the recent focus on this problem, fewer than 50 cases have been reported in the literature.208,209,211-213 These studies stress that chronic PBT tears are frequently overlooked or misdiagnosed and are often found in cases of chronic ankle laxity. A great majority (90%) of patients with PBT tears have swelling and tenderness over the

­ eroneal tendons. Tenderness with resisted eversion of the p foot also is a common finding,211,213 and re-creation of the retromalleolar pain during an anterior drawer test usually indicates a tear.209 Sobel and colleagues coined the term peroneal tunnel compression test to describe the maneuver by which the superior peroneal retinaculum is compressed against the posterior ridge of the fibula while the patient forcefully dorsiflexes and everts the foot against resistance.213 The test is positive if this maneuver re-­creates the patient’s pain. Imaging studies for evaluation of peroneal tendon tears have limited value. Standard radiographs of the ankle (including anteroposterior, lateral, and oblique views) usually are normal.208 However, MRI has been found useful in differentiating longitudinal split tears in the peroneal tendons from other lateral ankle disorders.230 Several studies have described the features and characteristics of peroneal tendon tears on MRI,231-233 and some have attempted to define the accuracy of MRI.211,233,234 Rosenberg and associates did a retrospective review of ankle MRI done on 27 patients who had clinical “evidence” of longitudinal PBT splits.233 Seven (26%) of the 27 patients had surgery, and the tears were confirmed in all 7 cases. Yao and associates reported on 4 patients who had MRI performed before surgical exploration of the “peroneal tunnels.”234 Surgery and MRI revealed two PBT and three PLT splits. They concluded that axial MRI through the ankle and hindfoot can help distinguish tendinitis from longitudinal tendon splits.234 Other studies, however, have noted a 27% falsenegative rate for detection of peroneal tendon splits.211 Thus, it appears that appropriate treatment of clinically diagnosed peroneal tendon splits should not be predicated on the findings on MRI until prospective studies have established its accuracy and better diagnostic criteria. Ultrasound also has been recommended for the diagnosis of peroneal tendon tears. Grant and associates evaluated 60 peroneal tendons that had preoperative ultrasound examinations and found that the sensitivity, specificity, and accuracy of ultrasonography for detecting the 25 tears found at surgery were 100%, 85%, and 90%, respectively.235

Treatment Options Successful treatment of peroneal tendon disruptions with associated fractures of the os peroneum has been achieved with 6 weeks of cast immobilization,203 internal (suture) fixation of the sesamoid fracture,205 and excision of the sesamoid fragments and repair of the tendon.200,201,204,206 Peroneal tendon disruptions without an associated fracture of the os peroneum have been variably treated by immobilization,226,228 tenodesis of the proximal and distal tendon ends to the peroneus brevis,191,193,204,236 suture to the cuboid,200 and primary repair of the tendon.190,206 Although operative treatment with primary repair190,206 and tenodesis191 193,204,236 have produced uniformly good long-term results, the outcome of nonoperative treatment, reported to be successful in one case,203 has not been well documented.35,186 Initial treatment of peroneal tendinitis should include decreased activity, immobilization, anti-inflammatory medications, and protected (crutch) ambulation. Some authors state that they have never found it necessary to operate

1990 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

on a patient with peroneal tendinitis for relief of symptoms.62,226 Most published reports, however, have found that operative treatment that includes release of the peroneal tendon sheath, along with débridement and repair of the tendon, often is necessary to alleviate the patient’s symptoms.112,208-212,237 All the methods of surgical treatment reported include débridement of the tendon and either primary repair of the tendon, tenodesis of the tendon to the PLT, or a tendon graft.41,112,127,174,209,238,239 Krause and Brodsky based the method of surgical treatment on the cross-sectional area of tendon involvement.211 If 50% or more of the tendon remained after débridement of the damaged portion (grade I tear), they repaired the tendon. If less than 50% remained (grade II), proximal and distal tenodesis of the PBT to the longus tendon was performed.211 All of the aforementioned studies emphasize that these procedures must be augmented by a peroneal tendon stabilization procedure when there is concomitant, underlying subluxation of the peroneal tendons. One study concluded that the reduction of pressure on the tendon that occurs with deepening of the peroneal groove may be advantageous for treatment of recalcitrant peroneal tendinitis.240 The results of surgery from either single or combined procedures have been uniformly good or ­excellent.112,201,209-212,237 Recently, tendoscopy for treatment of peroneal tendon disorders has been advocated.237 Van Dijk and Kort reported the results of peroneal tendoscopy on nine consecutive patients.237 The indications for the arthroscopy were diagnostic (five), snapping sensation (two), removal of exostosis (one), and partial tendon rupture (one). After surgery performed through a two-portal technique, three of four patients with adhesions resected and one patient with symptomatic prominent peroneal tubercle removed were symptom free, and one patient had a longitudinal rupture of the PBT successfully repaired.237 Scholten and Van Dijk reported similar results in 23 patients who underwent peroneal tendoscopy.241

Authors’ Preferred Method Our preferred method of treatment is based on the duration, site, and cause of the peroneal tendinitis. It has been our experience that most cases of tendinitis are secondary to either subluxation of the peroneal tendon or chronic lateral ankle instability. We have treated operatively four patients with longitudinal tears in their peroneal tendons. They were managed as has been outlined for the treatment of posterior tibial tendinitis. The central degenerated fibers were excised, and the peripheral intact fibers were ­repaired. It was not necessary to perform a tenodesis in any of these cases. However, we agree with those who note that pure peroneal tendinitis can usually be treated successfully with a nonoperative regimen of decreased ­ activity, immobilization (when necessary), anti-­inflammatory medications, protected ambulation, and a stretching and strengthening program.

SUBLUXATION OF PERONEAL TENDONS Traumatic dislocation of the peroneal tendons, either acute or chronic, is an uncommon injury. When the condition is recurrent, however, it causes disabling pain and usually prevents participation in sports. Although Monteggia originally described this injury in a ballet dancer in 1803, only sporadic reports were published on this topic in North America before 1960.242-245 Subsequently, there has been an abundance of publications on this topic that have elucidated the incidence, mechanism of injury, and difficulties of establishing the diagnosis.62,185,207,220,238,240,246-269 The injury most commonly occurs in skiing,246,247,258,270,271 and it is estimated that complete peroneal dislocations occur in 0.5%35 and sprains of the peroneal retinaculum in 2.5% of all skiing injuries.246 Moritz reported that four to five complete peroneal dislocations occur each year at Sun Valley.271 Although skiing accidents are the primary cause of this injury, soccer, basketball, football, ice skating, and a variety of other activities have precipitated this injury as well.62,207,246,251,253,255,259,260,272 Peroneal tendon dislocations can be classified as acute or chronic. Several studies have noted that acute injuries are often misdiagnosed as lateral ankle sprains and that owing to the lack of early diagnosis and proper treatment, recurrent dislocations invariably occur.40,207,248,253,254,263,272 The aforementioned reports indicate that this injury occurs in all age groups, the youngest and oldest patients being 13 and 54 years, respectively.

Pertinent Anatomy Several authorities have stated that patients sustaining peroneal dislocations usually have predisposing anatomic abnormalities, which include a shallow peroneal groove and an incompetent or absent peroneal retinaculum.220,253,262,263 Anatomic texts state that the posterior surface of the lateral malleolus usually has a definitive sulcus to accommodate the peroneal tendons. Several studies have shown, however, that the contour of the posterior surface of the distal fibular is quite variable. Edwards found that 11% of cadaveric fibulas had a flat posterior surface with no lateral edge, and 7% had a convex posterior surface.242 In the remaining 82%, a groove was present but was never more than 2 to 3 mm deep. She concluded that the lateral ridge of the fibula, when present, was not of sufficient proportion to maintain the peroneal tendons in their groove. She also noted that the length of the sulcus was not measurable because of its gradual inception and that the width of the sulcus ranged from 5 to 10 mm.242 Szczukowski and associates performed CT on eight normal ankles and on two patients with peroneal dislocations.262 They found that the eight normal ankles had well-defined, concave fibular grooves and that the two patients with dislocations had convex peroneal grooves. Other authors have found convex or shallow grooves in 40% to 100% of patients operated on for recurrent dislocations.220,253,259,263 Eckert and Davis examined the distal fibulas of 25 amputated legs and found that the posterior surface was essentially flat; when a groove was present, it was so variable in depth, length, and orientation that it was unrealistic to

Foot and Ankle 1991

attempt a description of its “usual” anatomy.220 They also documented that there is a discrete ridge of dense collagen and elastin fibers that extends along the posterior lip of the lateral malleolus.220 This ridge, which is most pronounced near the tip of the fibula, tapers proximally and is usually 3 to 4 cm in length. The superior retinaculum lacked strong connections to this dense strip of collagen and attached mainly to periosteum of the lateral aspect of the malleolus. Based on 73 cases of acute dislocations that were surgically explored, they concluded that rupture of the retinaculum occurs rarely, but instead the retinaculum strips the periosteum from the lateral malleolus or avulses a thin cortical shell. These authors based their conclusions on the aforementioned series of patients in whom they found no tears of the retinaculum, and they described three patterns of injury (grades I, II, and III). In grade I injuries, the retinaculum and periosteum were elevated from the lateral malleolus, and the tendons lay between the periosteum and the bone. This lesion, which occurred in 51% of their patients, also has been described by Das De and Balasubramaniam.246 In grade II injuries (33% of their patients), the distal 1 to 2 cm of the fibrous ridge was elevated along with the retinaculum, and in grade III injuries (16% of their patients), a thin cortical rim of bone was avulsed along with the retinaculum, the fibrous lip, and the periosteum.220 The peroneal tendons lay between the two exposed cancellous surfaces of the bone. In grade I injuries, the peroneal tendons, once reduced, were unstable only when placed under tension; in grade II and III injuries, the tendons were very unstable after they were reduced.220 Oden described a similar classification scheme based on 100 cases, but defined a type II injury as a tear rather than an elevation of the retinaculum from the fibula, and added a type IV injury, an avulsion of bone from the posterior rather than anterior fibular insertion site.258 Anatomic variations in the superior peroneal retinaculum also have been cited as a factor contributing to dislocation of the peroneal tendons. Congenital absence of this structure has been reported, and acquired laxity may occur in individuals subject to habitual pronation of the foot such as jockeys.243,245,271 Stover and Bryan concluded that primary sprains of the retinaculum may result in permanent laxity, predisposing the tendons to dislocation.245 They also concluded, however, that although the absence or convexity of the posterior fibular groove or the laxity of the retinaculum contributes to dislocation of the tendons, the presence of these factors is not a prerequisite for the occurrence of a dislocation.

Relevant Biomechanics The most commonly described mechanism of injury is sudden, forceful, passive dorsiflexion of the ankle with the foot in slight eversion.62,185,220,238,245,253,260,261 This results in a sudden, strong reflex contraction of the peronei, which are the dynamic lateral stabilizers of the ankle. This sequence of events produces the various grades of injuries summarized previously. Eversion of the ankle tends to tense the tendons against the retinaculum, making dislocation easier if there is preexisting retinacular laxity. Eversion of the ankle also tenses the calcaneofibular ligament, which decreases the width of the fibro-osseous tunnel and forces the tendons against the retinaculum.245,263 The position of

dorsiflexion causes a maximal change of direction of the tendons at the lateral malleolus and, in combination with eversion, also thrusts the tendons against the superior retinaculum. The key component of the injury, however, is the forceful reflex contraction of the peroneal muscles.

Clinical Evaluation In acute injuries, the athlete usually notes a popping or snapping sound or sensation at the posterolateral aspect of the ankle, accompanied by intense localized pain behind the fibula and above the joint line.220 There is usually diffuse lateral swelling and ecchymosis. The pain usually subsides rapidly and becomes relatively mild. If the individual is examined within hours of injury, one usually finds swelling localized over and just posterior to the lateral malleolus. If the swelling is not extensive, the peroneal tendons often can be palpated over the lateral malleolus, or their displacement can be appreciated when the ankle is dorsiflexed (Fig. 25D-5). When there is an associated fracture of the lateral malleolus, the bone fragment may be palpable in the deep tissues. In most cases, there are no signs of injury to the anterior talofibular, fibular calcaneal, or anterior tibiofibular ligaments as is commonly seen with lateral ligament injuries. If the injury is evaluated 4 or more hours after the injury, the lateral aspect of the ankle often is markedly swollen, and palpation reveals only tenderness at the posterior lip of the lateral malleolus. At this juncture, the swelling may obscure frankly dislocated tendons and makes the diagnosis difficult. One must therefore attempt to stress the retinaculum by eliciting active eversion of the foot with the ankle held in dorsiflexion. In patients with an acute subluxation, this test may produce apprehension or severe pain, or may show obvious dislocation, all of which are diagnostic of the problem. Safran and associates advocate that this provocative testing be done with the patient prone and the knee flexed to 90 degrees.260 They concluded that this position is a more comfortable position for the examiner and that dynamic instability is better visualized. Because of the rapid resolution of symptoms and an initial diagnosis of an ankle sprain, most patients seek medical evaluation when chronic subluxation is established. Patients with recurrent subluxation complain of lateral ankle instability and pain. They also note a popping or snapping sensation in the ankle just before the ankle giving way.62,185,238,250,251,254,256,260,261,263 The snapping or popping sensation is noted particularly during such activities as jogging or walking on uneven ground. Chronic subluxation may tend to occur with increasing frequency owing to progressive stretching of the residual lateral soft tissue structures.207 Many authors state that they can reproduce the tendon subluxation in all their patients by having the patient contract the peroneal muscles with the foot in eversion.62,248,253,259-261,263 Marti, however, stressed that one must examine the other ankle to identify those individuals with bilateral congenital subluxation of the peroneal tendon.253 In addition, DeHaven reported that one can reproduce the dislocation in only about 50% of patients with chronic dislocation.274 The differential diagnosis of this injury includes longitudinal tears of the peroneal tendons, sprains of the lateral

1992 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 25D-5   Subluxation of the peroneal tendons. With dorsiflexion and eversion of the right foot (A), the peroneal tendons dislocate anteriorly over the lateral malleolus. The dislocated tendons reduce when the foot is inverted (B).

ligaments of the ankle, and in children, Salter I fractures of the lateral malleolus. Often a longitudinal tear of one of the tendons will result in internal snapping of one tendon over another, mimicking a subluxation.250,255 Patients with chronic dislocation of the peroneal tendons do not have instability of the ankle or subtalar joints. Specifically, they usually have a negative anterior drawer test and no increased lateral laxity in comparison with the opposite ankle. Salter I fractures can be documented by obtaining radiographs of the opposite ankle and noting the location of the pain. In patients with acute and chronic peroneal dislocations, the pain is located along the posterior lip of the lateral malleolus and not over the anterior tibial-talar area. Radiographic evaluation of this injury should include anteroposterior, lateral, and mortise views of the ankle. Stress views, to assess increased talar tilt and lateral ankle instability, may be performed as indicated. Radiographs will be diagnostic only in patients with grade III injuries, those with an associated rim avulsion fracture of the lateral malleolus.245,270-272 The fragment, which is 1 to 1.5 cm in length, is best seen on the mortise view. This fracture occurs in 15% to 50% of individuals sustaining peroneal tendon dislocations,272 and when this finding is present, it is pathognomonic of a dislocation of the peroneal tendons.185,245,248,253,259,260,263,270-272,275 MRI will give further information regarding interstitial tendinopathy,232-234 and dynamic ultrasound imaging will demonstrate peroneal tendon subluxation.276 However, these imaging modalities are not routinely necessary in cases of isolated peroneal instability.

Treatment Options Nonoperative Treatment Although most authorities agree that there is no place for­ nonoperative treatment of chronic dislocations, the treatment of acute injuries remains ­ controversial.62,220,245,248,265,277

Stover and Bryan245 stated that all acute cases should be treated conservatively with a well-molded cast and non–weight-bearing ambulation for 5 to 6 weeks. They reported on 5 patients treated in this manner, and all five had excellent results; however, 2 other patients in their series were immobilized for a shorter interval, and 6 of 10 patients treated by taping of the ankle in a neutral position for 4 to 6 weeks had recurrent dislocations. Thus, of a total of 17 acute peroneal dislocations treated nonoperatively, 7 ultimately needed surgical repair.245 In contrast, Scheller and colleagues treated seven patients with taping and a lateral felt pad over the fibula, and all seven had recurrent dislocations.62 Similarly, Escalas and associates248 reported that 28 of their 38 patients with acute dislocations, who were treated by immobilization and a compressive bandage for several weeks, had recurrent dislocations. Eckert and Davis220 noted that only 1 of their 7 patients who were treated with either adhesive strapping or casts for 4 weeks had a stable pain-free ankle, whereas four had recurrent dislocations. McLennan reported on 16 athletes with subluxations or dislocations of the peroneal tendons.277 Of the 6 patients with acute injuries (diagnosis within 1 week), 1 had immediate surgery, and 5 had 3 weeks of taping of the ankle with a crescent-shaped, laterally placed pad. Three (60%) of these 5 athletes later had operative correction because of recurrent dislocations. Ten athletes with subacute and chronic injuries (diagnosis at 2 weeks to 2 years) also were treated with the aforementioned regimen of taping. Four (40%) of these athletes ultimately had surgery.277 ­McLennan concluded that although nonoperative treatment gives satisfactory results in more than 50% of the cases, operative treatment of acute injuries is indicated in athletes seriously involved in sports, particularly those with rim fractures.277 All the aforementioned authors who have experienced failures with nonoperative treatment still recommend treating acute cases initially with a short leg, non–weight-­bearing cast for a minimum of

Foot and Ankle 1993

Peroneus brevis tendon Peroneus longus tendon Wedge of bone driven backward Lateral malleolus

A

B

Figure 25D-6   Kelly’s bone block procedures. The original technique (A) is shown on the left, and the modified procedure (B) is shown on the right.

4 weeks.62,248,277 They reserve ­ surgical treatment for patients who have ­subsequent ­dislocations.

Operative Treatment The results of operative treatment of acute dislocations have been reported in only a limited number of studies.220,253,272 In Eckert and Davis’s series of 73 patients, grade I (retinaculum and periosteum elevated from lateral malleolus) and grade II (fibrous ridge and periosteum elevated) lesions were repaired by suturing the retinaculum to the fibrous ridge or the ridge and retinaculum to the malleolus, respectively.220 Grade III injuries (rim fractures) were treated with open reduction and K-wire fixation. At follow-up 6 or more months after surgery, the tendons had redislocated in 3 patients, all with grade II injuries. Twelve additional patients experienced mild pain after vigorous activity, but none thought their symptoms warranted further treatment.220 Results of acute surgical repairs also have been reported by Marti253 and Murr.272 All three of Murr’s patients had rim fractures of the lateral malleolus and were treated by open reduction and internal (suture) fixation. Within 1 year of surgery, all three patients had returned to skiing and were symptom free.272 Marti reported on five patients with acute injuries, one of whom had a rim fracture, and concluded that to achieve the best results, acute injuries should be treated operatively with primary repair of the torn peroneal retinaculum.253 At an average follow-up of 31⁄2 years, all five patients were symptom free and had normal ankle motion. In all the aforementioned series of acute injuries, the patients were immobilized in a short leg cast for 6 weeks after surgery. Although more than 18 surgical procedures have been proposed for treatment of recurrent dislocations, there are only five basic types of procedures. These are bone block procedures,244,253,254,275,277,278 rerouting ­procedures,185,250,259 reattachment of the retinaculum and reinforcement with local tissue,62,246,251,260,266,268,275,279 reconstruction of the superior peroneal retinaculum with tendon slings,243,248,249,255,256,261 and groove-deepening ­procedures.238,262,263,265,267,269

Figure 25D-7   The Du Vries modification of Kelly’s bone block procedure.

The procedures that alter the osseous structure of the fibula are designed to contain the peroneal tendons either by deepening the peroneal groove or by augmenting the lateral osseous ridge of the fibula. Augmentation of the lateral border of the fibula usually is achieved with a bone block procedure. In the original procedure described by Kelly in 1920,244 a “veneer-like graft, almost wholly composed of compact bone,” was created by two saw cuts (one sagittal and one horizontal) in the distal 2 to 3 cm of the fibula. The graft was then rotated posteriorly so that it overlapped the posterior cortex of the fibula by 5 to 6 mm and was fixed in its new position with two countersunk screws (Fig. 25D-6A). Kelly subsequently described a modification that eliminated screw fixation by dovetailing the graft so that it was wider anteriorly and was held tightly in its bed when it was displaced posteriorly (see Fig. 25D-6B). DuVries modified Kelly’s technique by using a wedgeshaped section of distal fibula that was 2 cm wide and half the depth of the lateral malleolus.275 The wedge was displaced posteriorly 0.5 cm and held in place with a small screw or autogenous bone peg (Fig. 25D-7). Patients treated in this manner were immobilized in a short leg cast for 5 to 8 weeks. The largest series of patients (11 to 12 cases) treated by the modified Kelly procedure were reported by Marti in 1977,253 Micheli and colleagues in 1989,278 and Mason and Henderson in 1996.254 None of Marti’s had redislocations 2 or more years after surgery, but two patients had crepitation of the tendons. He concluded that the crepitation was caused by insufficient posterior displacement of the bone block, which resulted in a shallow groove with a sharp inferior edge. In the more recent series of sliding,278 or rotational,254 bone block procedures, similar results have been reported. In patients with a shallow, flat, or convex peroneal groove, the groove-deepening procedure described by Zoellner and Clancy appears to be an effective method of correcting the basic deformity.263 With this technique,

1994 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25D-8   Deepening of the peroneal groove through an osteoperiosteal flap. Figure 25D-9   The Jones (tendon-sling) procedure.

a 1 × 3-cm ­osteoperiosteal flap is elevated from the posterior aspect of the distal part of the fibula and lateral malleolus. The posterior medial border of the cortical flap is left intact to act as a hinge. The flap is elevated and swung posteriorly, and the cancellous bone from the posterior aspect of the fibula is removed to a depth of 6 to 9 mm. The osteoperiosteal flap is then tapped back into place, and the peroneal groove is deepened over a length of 3 to 4 cm (Fig. 25D-8). Zoellner and Clancy used this technique in 10 patients and stated that the results were excellent at an average follow-up of 2 years (range, 7 months to 3 years and 10 months).263 In 6 of the 10 patients, however, the peroneal retinaculum was so tenuous that an additional (1 × 1 cm) periosteal flap was fashioned from the lateral surface of the malleolus and sutured to the medial part of the peroneal retinaculum. After surgery, the patients were immobilized in a short leg cast for 3 weeks and then in a short leg cast with an ankle hinge for an additional 3 weeks. Zoellner and Clancy recommended this technique because it (1) is technically easy to perform, (2) does not require metallic fixation, (3) corrects the basic deformity of a shallow peroneal groove, and (4) allows early motion because prolonged immobilization for union of bone or tendon is not needed.263 Three subsequent studies have also reported excellent results with this technique in 17 additional patients,207,238,262 and one study reported excellent results with a similar but indirect groove-deepening procedure.265 Results of the various soft tissue procedures (e.g., rerouting, periosteal flaps, and tendon slings) generally are very good. To date, however, there has not been a controlled study that compares the results of the different procedures to determine whether one is better than another. Rerouting of the peroneal tendons has been advocated by Sarmiento and Wolf185 and by Poll and Duijfjes.259 Sarmiento and Wolf reported one case in which the tendons were divided, repositioned under the calcaneofibular ligament, and then repaired with a Bunnell stitch. At 3-year follow-up, the patient participated in athletic activities and had had no further dislocations.185

In 1984, Poll and Duijfjes reported on 10 patients in whom the insertion of the calcaneofibular ligament was mobilized and lifted with a cancellous bone block from the calcaneus. The peroneal tendons were then brought under the ligament, and the bone block was replaced and fixed with a small cancellous screw. After surgery, the ankle was immobilized for 6 weeks in a short leg plaster cast, but weight-bearing was allowed after 2 weeks. Poll and ­Duijfjes reported excellent results in all 10 patients and recommended the procedure because it precluded scarring and adhesions of the peroneal tendons to the surrounding structures.259 They also described and critiqued similar procedures proposed by Platzgummer and by Leitz. Platzgummer divided the calcaneofibular ligament near the fibula, and Leitz osteotomized the lateral malleolus and refixed it with Kirschner wires.259 Poll and Duijfjes stated that the disadvantage of the first technique was that the integrity of the ligament was disturbed, and a disadvantage of the second technique was that the osteotomy was near the articular surface of the fibula. Thus, these procedures increased the risk for adhesions forming between the tendons and the ligament or bone. Recently, Harper reported a good result in one patient who had transfer of both tendons under the calcaneofibular ligament after detachment of the ligament from the fibula.250 The use of a tendon sling to reconstruct the peroneal retinaculum was first described by Jones in 1932 (Fig. 25D-9).243 He reconstructed the retinaculum in a 22-yearold football player with a strip of tendon that was fashioned from the Achilles tendon at its calcaneal insertion. The tendon slip, which was 21⁄2 inches in length and about 1⁄4 inch in width, was passed through a transverse drill hole in the fibula, 1 inch above the tip of the lateral malleolus, looped posteriorly, and sutured to the periosteum of the fibula and to the tendon slip itself. Jones stressed that the tendon slip should be anchored with the foot held in full dorsiflexion and supination. After surgery, the athlete was placed in a short leg cast for 6 weeks. At 6 weeks, he was permitted to

Foot and Ankle 1995

return to full activity and subsequently returned to football without symptoms 3 months later. The largest series of patients treated with the Jones procedure was reported by Escalas and colleagues in 1980.248 They performed the procedure in 28 patients, 15 of whom were followed for an average of 6.8 years (range, 3 to 11 years). Fourteen of the 15 patients had excellent results and returned to sports activities after an average of 4.2 months. One patient reported instability of the ankle, but no instability could be demonstrated on physical examination. Three of the 15 patients had a moderate decrease in inversion of the hindfoot, and four lost up to 7 degrees of dorsiflexion. The authors noted that minor loss of motion occurred despite the fact that the tendon slip was sutured in all patients with the foot held in maximal dorsiflexion.248 Tendon slings have also been created from a portion of the PBT261 and by transposition of the peroneus quartus tendon.256 Stein used a free graft consisting of 50% of the diameter of the tendon of the peroneus brevis and anchored it through two drill holes to the fibula at a point 1.5 cm proximal to the inferior tip of the lateral malleolus.261 The graft was passed through the distal drill hole in an anteroposterior direction, looped over the peroneal tendons, and passed through the proximal medially placed hole in a posterior-to-anterior direction. The free tendon was then sutured to itself with sufficient tension to restrain anterolateral subluxation of the tendons. Stein described the use of this technique in a 17-year-old quarterback who was placed in a short leg cast for 6 weeks after the operation. The athlete was able to resume full sports-related activities 3 months after the operation and was asymptomatic when examined 3 years later. Stein believed that his procedure was superior to that of Jones243 because the latter procedure had two disadvantages: (1) it weakened the Achilles tendon and (2) restoring full ankle motion was a problem because the slip of tendon was left attached to the calcaneus. Mick and Lynch256 reported a case in which the peroneus quartus was used to reconstruct the superior and inferior peroneal retinaculum. The peroneus quartus, which is estimated to occur in 13% of the population, originates on the posterior surface of the fibula and inserts on the calcaneus. They rerouted this tendon anteriorly over the peroneus longus and brevis tendons and placed it in a 5-mm deep oblique groove created in the anterior aspect of the lateral malleolus. The periosteum and soft tissue on either side of the groove were sutured over the transferred peroneus quartus tendon. At follow-up 1 year later, their patient was performing all required duties as a firefighter and was asymptomatic.256 Reconstruction of the peroneal retinaculum with an osteoperiosteal flap was conceived and described by ­WatsonJones in 1955.279 With this technique, a flap of periosteum (1.0 × 1.35 cm) is elevated from the distal fibula. The posterior periosteum and soft tissue structures are left intact. The flap is reflected posteriorly and sutured to the remaining portion of the superior retinaculum and the fascia of the flexor muscles of the great toe. The tendon sheath of the peroneal muscles is not opened, and the groove is not deepened. After surgery, the patient is placed in a short leg cast for 4 to 6 weeks and then rehabilitated until he or she has gained full range of motion and full strength.

The results of this procedure have been reported in only a limited number of studies.63,246,268 Scheller and colleagues performed this procedure in seven patients and subsequently described their follow-up at a minimum of 15 years.62 All seven patients had excellent results and returned to their preinjury sporting activities. In contrast, Das De and Balasubramaniam noted that two of three patients treated by them in this manner had recurrent dislocations and poor results.246 They concluded that the failures occurred because the procedure did not address the primary disease involved. They operated on seven subsequent patients with recurrent dislocations and noted that in all cases the periosteum was stripped from the lateral malleolus but remained in continuity with the superior retinaculum. The detached periosteum extended forward to the anterior margin of the lateral malleolus and down to its tip, creating a false pouch into which the peroneal tendons easily dislocated. They suggested that this injury was similar to that found in Bankart’s lesion in recurrent dislocations of the shoulder. They repaired the avulsed periosteum by scarifying the outer surface of the lateral malleolus and then suturing the periosteal flap to the lateral malleolus through the drill holes in the lateral edge of the peroneal groove. Their results in all seven cases were excellent at a minimum follow-up of 2 years.246 In 1998, they published the results of 21 patients treated as outlined previously. They reported that 18 had good results and had returned to their preinjury level of activity. The three fair results were due to painful scars or neuromas.251 Maffulli and colleagues performed anatomic repairs of the superior peroneal retinaculum on 14 patients with chronic, recurrent peroneal tendon subluxations. At follow-up of 38 months, none had experienced recurrent subluxations, and all had returned to their normal activities.266 Subluxation of the peroneal tendons within the peroneal sheath has been described by McConkey and Favero,255 Harper,250 and Raikin and collegues.264 McConkey and Favero reported on a 28-year-old runner who, 5 years after a plantar flexion inversion injury, complained of intermittent pain and clicking in the area of the peroneal tendons during jumping and pivoting activities. There was no evidence of anterior subluxation of the peroneal tendons, but the patient was able to reproduce the painful clicking of the tendons with dorsiflexion and eversion of the foot. Direct compression over the tendons eliminated the clicking. When nonoperative treatment was unsuccessful, the peroneal tendons were surgically explored under a local anesthetic. At surgery, it was evident that dorsiflexion and eversion of the foot caused the peroneus brevis to rise out from under the peroneus longus and to displace posterolaterally. This maneuver produced the click that the patient had been experiencing. They corrected this problem by taking strips of the retinaculum and passing them under the peroneus longus but over the peroneus brevis. At ­follow-up 4 years later, the patient had returned to athletic participation, including basketball, without recurrence of the symptoms.255 Harper described two cases of subluxation of the tendons within the peroneal groove. One occurred insidiously in a 13-year-old girl, and the other developed in a 26-yearold man after several inversion injuries to the ankle.250 Both were treated successfully, one with tenodesis of the

1996 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

brevis tendon to the longus after failure of a tendon-sling procedure, and the other with rerouting of the tendons.250 Raikin and colleagues described 14 patients in whom preoperative ultrasound demonstrated intrasheath peroneal tendon subluxation during dorsiflexion and eversion of the ankle.264 At surgery, two types of intrasheath subluxation were found. In 10 patients, the subluxation occurred

when the intact tendons switched their normal anatomic positions, and the peroneus longus came to lie deep to the peroneus brevis tendon. In the other 4 patients, there was a split in the peroneus brevis tendon through which the longus tendon subluxated. All 14 patients were treated with a peroneal groove–deepening procedure with a retinacular reefing and had excellent results.264

Authors’ Preferred Method The authors’ experience parallels that of Eckert and ­Davis and of Murr, who concluded that the results of closed methods of treatment are disappointing and doomed to failure.220,272 Thus, our preferred method of treatment for acute dislocations is to reduce the tendons and tape the ankle with a felt pad over the perineal groove to allow the athlete to attempt to complete the current season. At the end of the season, a reconstruction of the retinaculum and deepening of the groove (as detailed later) are performed. One author has treated one athlete who presented with dislocated peroneal tendons that could not be reduced by closed methods. The injury had occurred 1 month before my evaluation of the athlete. During that interval, the individual had been examined by two physicians who, despite the mother’s insistence that the tendon was dislocated, failed to recognize and diagnose the injury. This individual sought another opinion because he was unable to jog or participate in sports owing to pain, weakness, and lateral ankle instability. This athlete was surgically treated with the method outlined later for chronic dislocations. The authors’ surgical treatment of chronic dislocations includes deepening the retromalleolar groove and reconstructing the superior peroneal retinaculum. Deepening of the peroneal groove is achieved using the method described by Zoellner and Clancy.263 In this technique, a 5- to 7-mm J-shaped incision is made over the posterior aspect of the lateral malleolus along the course of the peroneal tendons. The superior and inferior retinaculi are incised, and the tendons are freed from their sheath and retracted anteriorly over the lateral malleolus. A small drill bit is then used to define the borders of an osteoperiosteal flap (4 cm in length and 1 cm in width) in the peroneal sulcus of the distal fibula and lateral malleolus. A small osteotome or an oscillating saw with a 3⁄8-inch blade is used to connect the drill holes along the superior, lateral, and inferior margins of the flap. The ­periosteum of the posterior medial border of the flap is left intact to act as a hinge. The flap is then raised and swung posteriorly to allow access to the cancellous bone beneath it. The cancellous bone from the posterior aspect of the fibula is removed to a depth of 8 to 10 mm. One should also remove the cancellous bone that remains on the posterior aspect of the flap and the cancellous bone under the intact medial edge of the sulcus to prevent the flap from springing open when it is tapped back into place. The flap will usually

hold firmly in place when it is tapped back into position. If the flap does not stay in place, it can be held in its new bed with sutures that are passed through drill holes in the flap and the anterolateral cortex of the fibula. The peroneal tendons are replaced in the new groove, and the ankle is put through a full range of motion. If the tendons do not remain well seated and show a tendency to subluxate, the groove should be deepened further by removing more cancellous bone. We also routinely reconstruct the superior retinaculum with a periosteal flap (1 cm2) fashioned from the lateral surface of the malleolus, hinged on its posterolateral side, and sutured to the medial part of the peroneal retinaculum. Although we have not found it necessary to perform additional procedures, we would not hesitate to augment the lateral edge of the fibula and further deepen the groove by using the modified bone block procedure described by Kelly.244 The subcutaneous tissues are closed with a running 2-0 absorbable suture, and the skin is closed with a running subcuticular 3-0 nonabsorbable suture. After surgery, the patient is placed in a short leg boot for 2 weeks. At that juncture, the running subcuticular suture is removed, and the patient is allowed to bear full weight as wound healing and the patient’s pain dictate. Three weeks after surgery, the patient is instructed to take off the boot and begin dorsiflexion and plantar flexion exercises 3 to 4 times per day. Immobilization is discontinued 6 weeks after surgery, and the patient begins further range of motion and strengthening exercises. The criteria for return to sports participation are the achievement of full range of motion and full strength, and completion of a running program. The running program is initiated when the patient has a full range of motion and strength equal to 80% of that of the contralateral extremity. The athlete begins by jogging one quarter of a mile and then, if there is no pain or limp, progressing to 1 mile. At that time the athlete is instructed to perform six 40-yard “sprints” sequentially at half, threequarters, and full speed. Athletes are allowed to increase their speed only when they have had no pain or limp at slower speeds. They subsequently repeat the same routine, alternately “cutting off” the injured and uninjured ankle every 10 yards. Most athletes are able to obtain a full range of motion and normal strength and to complete the running ­program within 12 to 16 weeks of surgery.

Foot and Ankle 1997

ACHILLES TENDON INJURIES Injuries of the Achilles tendon include peritendinitis, tendinosis, and partial or complete ruptures. Several authors report that peritendinitis of the Achilles is the most common overuse syndrome seen in sports medicine clinics.25,280-284 Although the true incidence of peritendinitis and tendinosis in athletes is not known, Achilles tendon pain accounted for 6.5% to 11% of the complaints of runners who were examined for lower extremity problems,280,281 and it has been reported in a variety of other sports such as soccer, football, tennis, volleyball, basketball, badminton, and handball.23,26,283 Similarly, although there have been several studies of partial ruptures of the Achilles tendon, the true incidence of this phenomenon also has not been documented.285-289 Reported estimates of the prevalence of Achilles tendinitis, however, are 11% in runners,242 9% in dancers,290 5% in gymnasts,291 2% in tennis players,293 and less than 1% in football players.292 In the various published reports to date, the mean age of patients with peritendinitis and tendinosis ranged from 24 to 30 years, and the youngest and oldest were 16 and 52 years, respectively.280,283,285,290-293 Basically, Achilles peritendinitis and tendinosis are overuse phenomena and are the result of accumulative impact loading and repetitive microtrauma to the tendon.22,23,26,280,281,284 In Clement and colleagues’ series of 109 runners with Achilles tendinitis, the average distance run per week for males was 34 miles (range 12 to 70 miles) and for females was 24 miles (range, 10 to 50 miles).280 There are, however, both intrinsic and extrinsic factors that predispose an athlete to these injuries. Intrinsic factors include areas of decreased vascularity,21,294 aging and degeneration of the tendon,22,24,27 and anatomic deviations such as heel-leg or heel-forefoot malalignment,21,281,295 and poor gastrocnemius-soleus flexibility.280,282,284 Clement and colleagues proposed that a varus position of the heel or supination of the forefoot produced functional overpronation of the foot during running. He and others have concluded that this type of mechanical deformation may result in a whipping action in the tendon and increased friction between the tendon and the ­peritendinous ­tissues.22,280 Extrinsic factors that predispose an athlete to tendinitis include a sudden increase in training intensity, interval training, change of surface (soft to hard), fluoroquinolone antibiotics (e.g., levofloxacin), and inappropriate or worn-out footwear.280,283,295-298 McCrory and coworkers performed a discriminant analysis of significant variables to determine causal factors associated with Achilles tendinitis.299 Their study revealed that plantar flexion peak torque, touch-down angle, and years running were the strongest discriminators between runners afflicted with Achilles tendinitis and runners who had no history of overuse injury.300 Although complete ruptures most often occur in middle-aged persons after a specific precipitating event, partial ruptures occur in younger individuals (20 to 30 years of age) who have reached their highest level of performance.26,285

Anatomic Considerations The tendocalcaneus (Achilles tendon), which is the thickest and strongest tendon in the body, is formed about 15 cm above the heel at the confluence of the soleus and the

­ astrocnemius muscles. Although the tendon begins where g the muscle belly of the gastrocnemius ends, it continues to receive muscle fibers on its anterior surface from the soleus almost to the malleolar level. The soleus and gastrocnemius components of the tendon can be separated and identified almost to the tendon’s insertion at the calcaneus, which is about 1.5 cm distal to the tip of the superior tuberosity. Between its origin and insertion, the tendon twists laterally (about 90 degrees) so that the tendinous fibers from the gastrocnemius insert into the posterolateral and those from the soleus insert into the posteromedial aspect of the calcaneus. Proximal to the site of insertion of the tendon, the retrocalcaneal bursa is interposed between the tendon and the upper part of the bony surface of the calcaneus. The narrowest part of the tendon is 4 cm proximal to its insertion, and throughout its length, the tendon is separated from the deep muscles by areolar and adipose tissue. The small saphenous vein and sural nerve are located along its lateral side. ­Lagergren demonstrated that the Achilles tendon has an avascular zone 2 to 6 cm above its insertion into the calcaneus.21 Stein and colleagues confirmed these findings using a new method with injection of radioisotopes.301 They found that the intravascular volume of the middle part of the tendon (3 to 6 cm above the calcaneal insertion) was significantly lower than the proximal or distal parts of the tendon. This area of avascularity is the most common site of peritendinitis, tendinosis, and rupture of the tendon.26,109,280,283,284,302-308 Hastad and colleagues demonstrated through isotope clearance studies that there is deterioration in the nutrition of the tendon with advancing age.294 The pathologic changes in Achilles tendon injuries span a continuum of abnormalities but generally can be defined by three stages. In stage I, the tendon is normal, but there are inflammatory changes in the peritendinous tissue that include thickening and adherence of the sheath to the tendon and on occasion an exudative fluid around the tendon. This stage is most aptly described as peritendinitis.23,24 In two series of patients with Achilles tendinitis (a total of 188 tendons) in which the tendon was surgically explored, isolated peritendinitis was observed in 50% to 59% of the cases.283,284 The terms tendinosis and peritendinitis with tendinosis best describe the second stage, which is characterized either by degenerative and inflammatory changes within the tendon or by degenerative changes within the tendon and associated inflammation of the peritendinous tissue, respectively. Macroscopic examination of the tendon may reveal nodular thickening, areas of metaplastic calcification, or loss of the normal luster of the tendon at several locations.24,26,27,283,286,308 Microscopic examination reveals areas of mucoid degeneration, fibrinoid necrosis, and tearing of the tendon fibers.24,27,283,284 In general, tendinosis does not affect the whole tendon but may affect nonadjacent areas of the tendon. In the series of Nelen and associates283 and Schepsis and Leach,284 tendinosis (intrinsic degeneration) without macroscopic ruptures was observed in 20% and 33% of the operated tendons, ­respectively. In the final stage of injury, macroscopic, visible disruption of the tendon occurs. These macroscopic tears occur in both the peripheral and central areas of the tendon. In the studies cited earlier, macroscopic tears and partial disruption of the tendon were found in 21% of the cases.283,284

1998 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Clinical Evaluation The predominant symptom of Achilles tendinitis is pain, which is localized to the area of the tendon 2 to 6 cm above its insertion.22,16,280,283-285,309-311 In Nelen and associates’ series of 91 patients who were operated on for chronic Achilles tendinitis, the extrinsic factors that contributed to the onset of symptoms were sudden changes of training in 31% of the patients, inappropriate training surface in 15%, inappropriate shoes in 7%, and direct trauma in 10%.283 Clement and colleagues studied 109 runners and concluded that the three most prevalent causal factors were overtraining (75%), functional overpronation (56%), and poor gastrocnemius-soleus flexibility (38%).280 In the early stages of tendinitis, the athlete experiences pain only with prolonged running. The pain usually subsides rapidly with rest but may be exacerbated by climbing stairs. In the subacute stage, the pain is present at the start of a run and becomes worse with sprinting, sometimes forcing the athlete to stop or to cut down on sports activity. In the advanced stages, when there is tendinosis or a partial rupture, the athlete is unable to run and experiences pain at rest. The athlete may also complain of weakness and intermittent swelling.289 Several authors think that the presence and severity of morning stiffness is a good standard by which to evaluate the seriousness of the condition.282,283 Nelen and associates found the following incidence of symptoms in the 91 patients whom they explored surgically for chronic Achilles tendinitis.283 Thirty-five percent had pain only during sports activities, 65% had pain with daily activities, 86% had morning stiffness, 10% had pain at rest, and 35% had acute sharp pain, felt during a sprint or acceleration while running.283 The aforementioned constellation of symptoms caused 54% of these athletes to decrease their sports activity and 46% to discontinue it.283 The physical findings in patients with Achilles peritendinitis and tendinosis include soft tissue swelling, local tenderness, and crepitus.24,282,283,295,311-313 In the early stages, there is focal swelling and tenderness limited to a small area, usually no larger than the breadth of the palpating fingertips. The area of tenderness can be defined by squeezing the diseased tendon segment between the thumb and index finger. Crepitus, which is the result of an exudation around the tendon, is more commonly found in the acute stage and is accentuated by active dorsiflexion and plantar flexion of the foot. In all stages of tendinitis and tendinosis, including partial ruptures, Thompson’s test is negative. In chronic tendinitis, the area of tenderness is somewhat larger, and often the area is thickened and nodular. When one detects nodularity in the tendon, one should be suspicious of tendinosis or a partial rupture of the tendon.283 In Denstad and Roass’s series of 58 partial ruptures that were treated surgically, there was localized tenderness in 56 (97%), focal swelling in 53 (91%), but a palpable defect in the tendon in only 2 (3%).285 In contrast, Skeoch examined 11 patients (16 tendons) with partial ruptures and concluded that palpation of the tendon generally reveals a partial defect.289 Differential diagnosis of patients with posterior heel pain includes retrocalcaneal bursitis and superficial tendoAchilles bursitis or “pump bumps.”22,282 The superficial bursa of the Achilles tendon lies between the tendon and

the skin and becomes inflamed as a result of friction from the heel counter of a shoe. The retrocalcaneal bursa lies anterior to the tendon, and inflammation of this structure usually is associated with prominence of the posterosuperior tuberosity of the os calcis. Standard anteroposterior and lateral radiographs of the ankle rarely show calcification of the soft tissues around the tendon or in the tendon itself; however, a prominent superior tuberosity of the os calcis will be evident on the lateral radiograph.282 Lateral radiographs taken with a soft tissue technique may show localized swelling, particularly in the area of Kager’s triangle,25 which includes the retrocalcaneal bursa.25,285 The incidence of calcification within the Achilles tendon is not known because bone formation itself does not cause symptoms.286 As noted in subsequent sections of this chapter, however, rupture of the Achilles tendon may occur through the ossified area308 or in the tendon adjacent to the site of ossification.286 Studies on sonographic evaluation of athletes with chronic Achilles tendinosis have found that this modality accurately demonstrates both tendinitis and tendinosis (and associated microtears) of the tendon.314-317 Ultrasound also has shown that the increase in neovascularization seen after an eccentric training program correlates highly with the amount of improvement in outcome/VISA scores.316

Treatment Options Nonoperative Treatment Most cases of Achilles peritendinitis and tendinosis are successfully managed nonoperatively. The basic modalities of nonoperative treatment are rest or a decrease in the runner’s weekly mileage, use of a 1⁄4- to 3⁄8-inch heel lift, oral nonsteroidal anti-inflammatory medications, use of an orthotic to correct excessive pronation, ultrasound, and stretching exercises.280,282-284 Total rest usually is not required, but the athlete should be instructed to avoid hill work and interval training.282,284 Clement and colleagues suggest a form of modified rest in which the athlete is allowed to participate in swimming and cycling activities but is not allowed to resume running for 7 to 10 days after the symptoms have subsided.280 Although several authorities advise the use of a 1⁄4- to 3 ⁄8-inch heel lift to decrease tension on the Achilles tendon, studies have differed on the effectiveness of these viscoelastic pads.273,318 In a blind-observer, random, prospective study of 33 subjects with Achilles tendinitis, Lowdon and associates found that subjects treated with ultrasound and exercises showed more profound improvement at both the 10-day and 2-month assessments than the two patient groups that received heel pads and exercises.273 In a recent randomized, controlled clinical trial, Petersen and associates found that an AirHeel brace was as effective as eccentric training for treatment of chronic Achilles tendinopathy.318 The positive effects of ultrasound therapy on the repair of Achilles tendon injuries have been documented (in rats) by Jackson and coworkers.287 They found that ultrasound increased the rate of collagen synthesis and the breaking strength of the Achilles tendon. The breaking strengths of treated tendons were significantly greater than those of untreated tendons 5, 9, 15, and 21 days after injury.

Foot and Ankle 1999

Stretching exercises have been proposed by several authorities as a key to nonoperative treatment of Achilles tendinitis.280,282-284 Leach and associates stressed that the common finding in patients with posterior heel pain due to Achilles tendinitis is loss of passive dorsiflexion.282 There are two recommended methods of stretching, but with either method, the stretching should be slow and should be sustained for 20 to 30 seconds. With one method, the individual leans against a wall with the knees straight and the heel flat on the floor. With the other method, the involved foot is placed forward and both the knee and the ankle are flexed while the heel is held flat on the floor. It has been our experience that the former method of stretching affects the more proximal tendon and musculotendinous junction, whereas the latter method has a more profound effect on the insertion and distal aspect of the Achilles tendon. Immobilization of the ankle for a period of 7 to 10 days may be indicated in individuals with severe acute symptoms.282 However, a recent study found that continued sports activity during rehabilitation for the tendinopathy using a pain-monitoring model produced the same results as active rest.319 Steroid injections, however, should be limited to the area of the retrocalcaneal bursa and should be employed only in patients with recalcitrant retrocalcaneal bursitis.282,284 There is a growing amount of evidence that steroid injections in and around the Achilles tendon may increase the risk for tendon rupture.32,282,320,321 ­Kleinman and Gross reported on three cases of Achilles tendon ruptures that occurred 2 to 6 weeks after the injection of steroids into the tendon.321 In all cases, the rupture was the result of minor trauma, and the tears were transverse. They concluded that the tears were directly attributable to the steroid injection because rupture of the Achilles tendon usually is a sudden traumatic event, and in such patients, surgical exploration reveals that the tendon ends are shredded and interdigitated. In the three patients of Kleinman and Gross, however, the tears were transverse, and the tendon ends were rounded, with obvious preexisting degenerative changes. Astrom and Rausing analyzed 342 cases of partial ruptures in 298 patients who were operated on for chronic painful Achilles tendinopathy.312 A logistic regression analysis of age, gender, physical activity, and preoperative steroid injections found that only preoperative steroid injections and male gender predicted a partial rupture.312 A study by Balasubramaniam and Prathap320 supports the conclusions reached by Kleinman and Gross321 and Astrom and Rausing.312 They created central tears in the Achilles tendon of rabbits and then injected hydrocortisone acetate into one side and compared the effects of the injection with the rabbit’s contralateral tendon. They subsequently sacrificed the animals at various intervals and observed the following changes on the injected side. Within 45 minutes there was separation of collagen bundles, and by 24 to 72 hours, areas of necrosis within the collagen were evident. At both 1 and 8 weeks after injection, they found that there was no evidence of repair of the Achilles tendon lesions, and in some animals, there was evidence of dystrophic calcification.320 Although most authors recommend nonoperative treatment of Achilles tendinitis, only Clement and colleagues have reported the results of a large series of patients

(109 runners) treated in this manner.280 Of these 109 athletes, 86 had complete follow-up, and 85 had good or excellent results (e.g., they had achieved preinjury training levels and either were symptom free or had minor symptoms on occasion). Clement and colleagues and other authorities, however, stress that athletes will not remain symptom free unless they understand the extrinsic factors that caused the injury.280,282,284,309 They should also be instructed in preventive measures, which include warming the Achilles tendon before running, applying ice for 10 to 15 minutes after running, and monitoring their shoewear, particularly the posterolateral aspect of the shoe. The most important preventive measure is stretching the posterior structures to prevent contractures and loss of passive ­dorsiflexion.280,282,284 Over the past decade, many new nonoperative methods of treatment have been advocated. These include extracorporeal shock-wave therapy (ESWT)322-325 heavy-load eccentric training,316,318,326 topical medications,327,328 and intratendinous injections of proteinase inhibitors,329,330 sclerosing agents,331 and dextrose.331a Alfredson and colleagues prospectively studied the effect of heavy-load eccentric training in 15 athletes who had chronic Achilles tendinosis.309 All 15 experienced fast recovery in eccentric and concentric calf muscle and returned to previous running activity symptom free. They concluded that there was little place for surgery in the treatment of chronic Achilles tendinosis located at the 2- to 6-cm level in the tendon.309 However, Woodley and associates reviewed all of the controlled trials using eccentric exercise (EE) for the treatment of Achilles and other tendinopathies and concluded that there was a dearth of high-quality research in support of the clinical effectiveness of EE over other methods of treatment.333 The effectiveness of ESWT has been evaluated in several recent randomized, double-blind, placebo-controlled studies.322,325 Rassmussen and colleagues reported on the 4, 8, and 12-week results of 48 patients with chronic Achilles tendinopathy that were treated with either active ESWT or sham ESWT.322 Significantly better results were seen in the intervention group at both 8 and 12 weeks. Rompe and colleagues reported on 50 patients with chronic, insertional Achilles tendinopathy. Twenty-five patients received eccentric training, and 25 patients received repetitive lowenergy ESWT. At the 4-month primary follow-up, all of the outcome measures (e.g., VISA-A scores, pain ratings) were significantly better in the group that received ESWT. They concluded that eccentric training as applied in their study gave inferior results to ESWT for the treatment of chronic, insertional Achilles tendinopathy.325

Operative Treatment Operative treatment should be considered when a comprehensive nonoperative treatment program of several months’ duration has failed and the athlete is not willing to alter or abandon the precipitating sports activity. Surgical treatment is required in about 25% of athletes with Achilles tendon overuse injuries.334 Although the surgical procedure performed should depend entirely on the disease found at the time of the operation, there are basically four distinct surgical procedures. These include release of the

2000 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Achilles tendon sheath; excision of degenerated segments of tendon and side-to-side repair; excision of degenerated tissue and reconstruction of the tendon with fascial flaps from the gastrocnemius; and excision of the retrocalcaneal bursa and ostectomy of the superior tuberosity of the os calcis.23,282-286,310-314 In cases of pure peritendinitis, it is recommended that the sheath of the Achilles tendon be released from the musculotendinous junction to the insertion of the tendon.23,282-284,295 Adhesions between the tendon and the sheath should be released but only on the dorsal, medial, and lateral sides of the tendon. Circumferential dissection of the tendon will damage the anterior vascular supply of the tendon and cause excessive scarring.280,282-284 Several authors also make a number of vertical slits in the tendon to “vent” the tendon and encourage ingrowth of new blood vessels.22-24 Endoscopic and ultrasound-guided percutaneous tenot­ omies also have been advocated for the treatment of chronic Achilles tendinopathy.317,335 Testa and ­ associates reviewed the 3-year results of 63 patients who had ­ ultrasoundguided percutaneous longitudinal tenotomy of the ­Achilles tendon.317 At that juncture, there were 35 excellent, 12 good, 9 fair, and 7 poor results. Nine of the 16 patients with fair or poor results underwent an open exploration of the Achilles tendon 7 to 12 months after the percutaneous procedure. They concluded that percutaneous ­ultrasoundguided tenotomy should be considered in patients with recalcitrant Achilles tendinopathy. However, they noted that in patients with diffuse or multinodular tendinopathy, an open exploration with stripping of the paratenon and multiple longitudinal tenotomies may be preferable.317 Maquirriain and coworkers reported results of seven patients with chronic Achilles tendinopathy who had endoscopic peritenon release, débridement, and longitudinal tenotomies performed.335 According to their scoring system, all seven patients improved after surgery from an average score of 39 points preoperatively to 88 points postoperatively. The only complications were a minor hematoma and edema that resolved spontaneously. Endoscopic Achilles tenodesis also has been described for the treatment of chronic insertional Achilles tendinopathy.336 In cases with focal tendinosis and partial rupture of the tendon, the diseased area of the tendon should be excised and the tendon repaired by side-to-side suture of the remaining normal tendinous fibers.283-285 One must carefully inspect and palpate the tendon throughout its length because nonadjacent areas of tendinosis are common. The tendon should be opened in zones that have lost their normal luster and in areas of nodularity. However, a recent study has documented that normal-appearing areas adjacent to Achilles tendon lesions also will have histologic and biochemical degenerative changes.337 In addition, a recent study has shown that patients with focal intratendinous lesions have poorer surgical results than those with only peritendinous adhesions.338 In cases in which excision of the degenerated tissue disrupts the continuity of the tendon, one should reconstruct and reinforce the tendon with a turn-down flap of fascia from the gastrocnemius. In these cases, a modified Lindholm339 repair, using only one flap of gastrocnemius fascia, will cover the defect. An FHL transfer also can be used for

repair of large defects caused by the excision of degenerated segments of the tendon.340 When it is evident preoperatively that the patient has a retrocalcaneal bursitis, the bursa should be excised, and, if prominent, the posterosuperior tuberosity of the os calcis should be removed.283,284 Ostectomy of the superior angle of the os calcis should begin just superior to the insertion of the Achilles fibers at an angle of 45 degrees to the long axis of the tendon. After excision of the fragment, any rough edges should be removed with a rasp or rongeur.282,284 The foot should then be put through a full range of motion to make sure that all areas of bony impingement have been excised. Leitze and associates have reported that endoscopic decompression of the retrocalcaneal space, including bursectomy and excision of a Haglund spur, produces results equal to or better than open techniques.341 Overall results of the aforementioned operative procedures were rated as good or excellent in 84% to 100% of the patients in the various series published to date.282285,289,295,311-313 In several studies, the results of specific procedures were reported.283,284,311,313 In Nelen and associates’ series of 143 tendons that were surgically explored, release of the tendon sheath alone produced good or excellent results in 89% of the 93 tendons with peritendinitis. Excision of the degenerated tissues and side-to-side repair produced good or excellent results in 73% of the 26 tendons with tendinosis, and turned-down gastrocnemius flaps produced good or excellent results in 87% of the 24 tendons with more extensive tendinosis.283 Complications from the aforementioned procedures included six cases of skin edge necrosis, two superficial wound infections, and one case of phlebitis. Schepsis and Leach reported the results of 45 patients who underwent surgical exploration.284 In the 28 patients who had release of the sheath (15 cases), excision of areas of calcification (4 cases), or débridement of partial tears and repair of the tendon (9 cases), the results were good or excellent in 89%. In the 11 patients who had excision of the retrocalcaneal bursa and a partial ostectomy of the os calcis, the results were good or excellent in 71% of the cases. In the 6 remaining cases, a release of the tendon sheath, a retrocalcaneal bursectomy, and partial ostectomy of the os calcis were performed. All 6 patients had good or excellent results. They concluded that the higher percentage of unsatisfactory results in those treated for retrocalcaneal bursitis alone was due to technical errors. Specifically, either an inadequate amount of bone had been removed or concomitant adjacent Achilles tendinitis had not been appreciated at the time of the initial operation. In a subsequent long-term follow-up study that included an additional 21 patients (66 total), Schepsis and coworkers reported satisfactory results in 75% and 86% of the patients with retrocalcaneal bursitis and insertional tendinitis, respectively.313 They also reported that the highest percentage of satisfactory results (89%) were obtained in the paratendinitis group, and the lowest (67%) in the ­tendinosis group.313 In a recent study and the largest to date, Paavola and colleagues reported the results and complications of surgical treatment of Achilles tendon overuse injuries in 432 consecutive patients.311 There was a total of 46 (11%) complications. Sixteen complications (35%) were classified

Foot and Ankle 2001

as major (14 skin edge necroses, 1 new partial rupture, 1 deep vein thrombosis), and 30 (65%) were considered minor (11 superficial wound infections, 5 seromas, 5 hematomas, 5 extensive scar or fibrotic reactions, and 4 sural nerve irritations). Fourteen (30%) of the 46 patients with complications had reoperations. They noted that every 10th patient treated surgically suffered a postoperative complication that clearly delayed recovery.311 They suggested that these complications could be prevented through the meticulous adaptations in surgical techniques they describe.

Postoperative Management and Rehabilitation In cases of peritendinitis in which only the sheath is divided, the postoperative protocols in the aforementioned series range from no immobilization22,295,311 to cast immobilization for 2 to 6 weeks.282-284 A recent study found no

advantages in recovery of muscle strength after surgery with a short immobilization time (2 weeks) compared with a longer (6 weeks) period.310 In individuals with tendinosis and side-to-side repairs of the tendon, a non–weightbearing splint is worn for 2 weeks, and a short leg walking cast is worn for an additional 2 weeks.283,311 Some authors, however, recommend no immobilization even after sideto-side repairs of the tendon.285 When a flap of gastrocnemius has been used to reconstruct the Achilles tendon, most authorities recommend immobilization for a period of 5 to 7 weeks.282-284,313 Stretching and strengthening of the Achilles tendon are initiated after immobilization has been discontinued. Patients then progressively increase their activity from walking and swimming to bicycling, jogging, and running. After a tendon reconstruction, they usually are able to jog within 8 to 12 weeks of surgery, but full recovery takes 5 to 6 months.282-284,311,312

Authors’ Preferred Method Initial treatment of Achilles peritendinitis and tendinosis always is conservative and focuses on control of pain and inflammation, correction of functional malalignment, and rehabilitation of the gastrocnemius-soleus muscle-tendon complex. In athletes who present within 1 to 2 weeks of the onset of symptoms, a short course (7 to 10 days) of oral, nonsteroidal anti-inflammatory medications and 2 weeks of rest usually will allow them to return to running, symptom free. In addition, they are counseled about the extrinsic factors (e.g., errors in training) that may have caused their problem and about prophylactic measures (proper shoes, stretching) that can prevent Achilles tendinitis. In most cases, however, the athlete seeks treatment only ­after he or she has attempted to “run through” the pain and has had to curtail or stop sports activities. In these cases, successful nonoperative treatment requires a more comprehensive program. First, the athlete must stop the precipitating ­activity (e.g., interval training, sprints, and so on) but is ­allowed to maintain aerobic fitness with alternative activities (such as swimming or cycling) if such activities can be performed without symptoms. This program of modified rest should be continued for 7 to 10 days after the ­Achilles tendon pain has subsided. Second, the athlete is started on a daily program of gastrocnemius-soleus stretching and strengthening exercises. Stretching is done in the manner outlined previously, and strengthening, initiated when pain has subsided, is achieved by performing toe raises with the heel hanging over the edge of a stair and moving the ankle through a full range of motion. Eccentric strengthening is accomplished with the toe raises by progressively increasing the speed of the heel drop. Third, functional components of malalignment are corrected with orthotics. Off-the-shelf flexible leather longitudinal arch supports (with or without a medial heel wedge) often are all that are needed to ­correct overpronation of the foot. Custom-made, rigid orthoses are obtained for patients with more severe or complex foot ­deformities.

Ten days to 2 weeks after symptoms subside, the athlete begins a gradual return to the preinjury level of activity with a running program staged to include a progressive increase in intensity and duration. The rate at which an athlete is able to return to preinjury training levels varies according to the severity and duration of each individual’s symptoms. I use steroid injections only in athletes who have retrocalcaneal bursitis. In those individuals, the bursa (not the tendon) is injected with 1 mL of lidocaine (Xylocaine), 0.25% bu­ pivacaine (Marcaine), and 40 mg of dexamethasone. After the injection, the athlete is instructed to refrain from all physical activity for a minimum of 3 days. At that juncture, he or she begins the rehabilitation program outlined earlier for nonoperative treatment of Achilles tendinitis. Operative treatment is offered to athletes who have Achilles pain after completing a 2- to 3-month program that includes “modified rest,” oral anti-inflammatory medications, use of orthoses (arch supports and heel lifts), and physical therapy (stretching and ultrasound). The operation is performed with the patient prone on the operating table, and the extent of the surgical procedure is dictated by the operative findings. In all patients, the tendon sheath is released throughout the length of the tendon, and the tendon is examined from its musculotendinous junction to its insertion on the calcaneus. In patients with peritendinitis and no macroscopic evidence of tendinosis, adhesions between the tendon and the tendon sheath are released, and vertical slits are made in the tendon in the area or areas that correspond with the patient’s preoperative pain. If these vertical incisions reveal intrinsic degeneration of the tendon, the degenerated areas are excised, and a side-to-side repair of the tendon is performed. In patients with partial ruptures or macroscopic evidence of tendinosis (nodularity, calcification, or loss of normal luster of the tendon), each area is opened through a longitudinal incision, and the abnormal segments of the tendon are excised. The method used to repair the tendon is predicated on the amount of diseased tissue removed. A side-to-side Continued

2002 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d repair is performed when 20% or less of the width of the tendon is excised. When wider segments are removed, an end-to-end repair and a turned-down flap of gastrocnemius fascia are used to reconstitute the integrity of the tendon. Details of the surgical technique and postoperative protocol for this procedure are provided in the following section on repair of Achilles tendon ruptures.

ACHILLES TENDON RUPTURES Spontaneous ruptures of the Achilles tendon generally occur in healthy, vigorous, “young” adults with no previous history of calf or heel pain.116,303,333,343,344 Although the exact incidence of this injury is not known, two studies determined the incidence in two larger populations of people.344,345 Nistor reported that over a 4-year period, 107 people (0.02%) from a population of 500,000 were treated at his hospital (the only one in the area) for this injury.344 Cretnik and Frank found that during a 5-year period, 116 (0.04%) of 273,609 inhabitants of their region sustained this injury.345 In published series on this injury, the average age of the patients ranged from 37 to 43.5 years,26,239,303,305,306,332,343,344-346 and 75% were between the ages of 20 and 49 years.26,239,304,306 In most series, most individuals with this injury were men in their third to fifth decades of life who were participating in recreational sports activities.239,257,303,304,307,332,343,344,346 The left Achilles is ruptured more frequently than the right,261,347 possibly because of the higher prevalence of individuals who are right-hand dominant and thus push off with their left foot. The reported risk for subsequently rupturing the contralateral tendon ranges from 6% to 26%.243,345 In many series, basketball and racket sports accounted for more than half of the injuries,304,307,332,344 and in one large study, 58 (52%) of 111 patients were playing badminton at the time of their Achilles tendon rupture.348 One report noted a high (14.3%) incidence of gout in their patients compared with the normal prevalence in the general population of 0.2% to 0.3%.332 Although ruptures of the gastrocnemius musculotendinous junction (so-called tennis leg) occur, they are extremely rare injuries,64 and the Achilles tendon most commonly ruptures 3 to 4 cm proximal to its insertion on the calcaneus, within the area of decreased vascularity.21,301 The pathogenesis of Achilles tendon rupture has been attributed to both degeneration of the tendon and excessive mechanical forces. Arner and Lindholm reported that all 92 ruptured tendons that they examined histologically had degenerative changes.349 Kannus and Józsa found that 864 (97%) of the 891 spontaneously ruptured tendons they examined histologically had degenerative changes.116 If tendon ischemia and secondary degeneration, however, were major factors, patients older than 30 years of age would be expected to have a higher rate of rupture, which is not the case.26,239,304-306,332 In addition, pathologic specimens removed from acute ruptures

Patients who undergo a release of the tendon sheath or a side-to-side repair of the tendon are put in a short leg, ­removable plastic boot for 2 weeks to facilitate wound healing. During that interval, they are allowed to bear weight on the extremity as tolerated. When the wound has healed, they begin the rehabilitation program outlined under the section on nonoperative treatment of Achilles tendinitis.

often reveal the ­ presence of hemorrhage and inflammation and no associated peritendinitis or tendinosis.24,304,332 Thus, several authors have concluded that ruptures are due to a sudden overloading of the musculotendinous unit in a poorly conditioned individual rather than to underlying pathologic processes in the tendon.303,304 In contrast, ­Tallon and associates have shown that ruptured Achilles tendons histologically have a significantly greater degree of degeneration than tendons with tendinopathy, and with both conditions, the tendons are significantly more ­degenerated than ­normal tendons.350 The common precipitating event in 90% to 100% of individuals that sustain this injury is active forceful, sometimes unexpected, plantar flexion of the foot.304,332 Arner and Lindholm classified the trauma that resulted in rupture in 92 patients into three catagories.349 The mechanism in the first category was pushing off with the weight-­bearing foot while extending the knee. This type of movement, seen with sprint starts and in jumping sports, accounted for 53% of the ruptures in their series. The mechanism in the second category was sudden, unexpected dorsiflexion of the ankle. This mechanism, seen when an individual steps in a hole, accounted for 17% of the ruptures. The third category was violent dorsiflexion of a plantar flexed foot as occurs after a fall from a height. This mechanism was reported by 10% of the patients.349 The major current controversy regarding this injury is whether surgical repair or cast immobilization is the most appropriate method of treatment.

Clinical Evaluation Most patients who sustain spontaneous ruptures of the Achilles tendon note a sudden snap in the heel region at the time of injury and subsequent pain with flexion of the foot.26,116,304,307,332,343,351 Kannus and Józsa reported that only 297 (33%) of the 891 patients in their study had symptoms before rupture of their Achilles tendon.116 Many patients do not seek immediate treatment because they still are able to plantar flex their ankle, and about 70% of patients complain of pain in the ankle or heel only at the time of their initial medical evaluation. This presenting complaint, along with a moderate limp and weakness in the ankle, may suggest a mild sprain rather than a heel cord rupture. Thus, the correct diagnosis is missed at the time of initial evaluation in 25% of cases.257 283,303,304 In acute cases, physical examination will reveal a palpable depression over the area of the tendon rupture,

Foot and Ankle 2003

weakness of plantar flexion, and positive results on the calf squeeze or Thompson’s test.26,127,303,304,332 Thompson’s test is performed with the patient prone on a table with the feet extending over the end of the table. The calf muscles are then squeezed between the examiner’s thumb and forefingers in the middle third (the musculotendinous junction) below the place of widest girth. A “normal reaction” is shown by passive plantar movement of the foot. A “positive reaction” occurs when there is no plantar ­movement of the foot, indicating a rupture of the heel cord.127 The test also can be performed with the patient kneeling on a chair. The accuracy of Thompson’s test for the diagnosis of fresh ruptures has been reported to be between 96% and 100%.127,304,352 O’Brien, however, concluded that Thompson’s test can be falsely positive.352 He noted that with Thompson’s test, tension in the Achilles tendon is produced by lifting and functionally shortening the ­gastrocnemius-soleus complex. When the gastrocnemius aponeurosis is torn and is no longer connected to that of the soleus, Thompson’s test will indicate a complete rupture when in fact the Achilles tendon is intact.352 Thus, O’Brien described and recommended a needle test for diagnosis of complete ruptures of the tendon. The needle test is performed by inserting a 25-gauge needle into the calf just medial to the midline at a point 10 cm proximal to the superior border of the calcaneus. The foot is then passively dorsiflexed and plantar flexed, and the movement of the hub of the needle is noted. Movement of the needle in the direction opposite the direction of the foot indicates that the tendon is intact throughout its distal 10 cm. O’Brien performed this test on 10 patients with suspected rupture of the Achilles tendon and found that two patients had a positive Thompson’s test but a negative needle test. At surgery, both of these patients were found to have a partial rupture of the musculotendinous junction of the gastrocnemius, but the Achilles tendon was intact.352 He further commented that false-positive results on Thompson’s test may cause unwitting errors

A

when the results of operative and nonoperative treatment of ­“complete” ­ruptures are compared. A recent prospective study of 174 patients found that palpation of the gap was the least sensitive clinical test for an Achilles tendon rupture.351 Although both the calf squeeze and O’Brien tests had a high positive predictive value, the calf squeeze test was significantly more sensitive (0.96 versus 0.8) for diagnosis of an Achilles tendon tear.351 In chronic cases, defects in the tendon may not be evident or palpable owing to hematoma formation at the site of rupture and generalized swelling about the heel and ankle. Most patients do not have ecchymosis, swelling, or point tenderness as are usually found with acute ruptures,109 and in one series, Thompson’s test was positive in only 80% of the chronic cases.332 Standard radiographs usually are not diagnostic of an Achilles tendon rupture. Although Arner and colleagues reported that deformation of the contour of the distal stump was pathognomonic of an Achilles tendon rupture,353 most authorities have concluded that standard radiographs are helpful only in confirming the diagnosis of rupture in patients with calcification of the tendon (Fig. 25D-10).109,332 With the advent of MRI, studies have been published that assess the diagnostic capabilities of this technique. One study found that MRI was extremely accurate for assessing the condition (e.g., shredded, uniform) and ­orientation (e.g., antegrade, retrograde) of the torn fibers and the width of the diastasis (with and without ankle ­flexion) between the ends of the tendon.354 A second study identified four types of Achilles tendon lesions: type I, inflammatory reaction; type II, degenerative change; type III, incomplete rupture; and type IV, ­ complete ­rupture.355 Ultrasound also has been promoted as an excellent imaging method for confirming the clinical diagnosis of an Achilles tendon rupture.356 Margetic and associates reported that ultrasound imaging correctly identified complete Achilles tendon ruptures in 86 (98%) of 88 patients

B

Figure 25D-10   Lateral radiographs of a 56-year-old patient with calcification of the Achilles tendon. The radiograph on the left (A) was obtained when the patient was evaluated for plantar fasciitis. Three years later (B), the Achilles tendon was avulsed from the calcaneus.

2004 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

treated operatively.356 The other patients had partial ruptures. However, the authors do not believe that MRI or ultrasound imaging is indicated for evaluation of Achilles tendon injuries because clinical tests are extremely reliable for diagnosis of complete ruptures, and treatment of lesser injuries should be predicated on the results of ­nonoperative treatment.

Treatment Options Controversy about Treatment Method The best method of treatment for complete ruptures of the Achilles tendon remains controversial. A review of the literature reveals that there are comparable numbers of studies that advocate operative239,302-305,307,332,342,345,348,357-363 and nonoperative299,302,306,344,347,364-366 methods of treatment. The advantages of nonoperative treatment are absence of wound complications, decreased patient cost, and lack of a scar. The disadvantages of nonoperative treatment are a re-rupture rate as high as 39% (Table 25D-1), a higher percentage of dissatisfied patients, and a significant loss of power, strength, and endurance compared with surgically treated patients.303-305,332 In addition, the results of surgical repair of re-rupture after nonsurgical treatment are not as good as those of primary repair.342 The advantages of surgical repair are a much lower rate of re-rupture, which ranges from 0% to 5% (Table 25D-2), a higher percentage of patients returning to sports, and a greater recovery of strength, power, and endurance.304,305,332,357,359,360,363,366 The disadvantages of surgical repair are the cost of hospitalization and the major and minor complications of surgery (Table 25D-3). Complications such as deep venous thrombosis and pulmonary embolism have been reported with both methods of treatment.347,359,360,365 The limited number of studies (Table 25D-4) comparing operative and nonoperative methods of treatment have reached contradictory conclusions.302,303,305,344,366,367

  TABLE 25D-1  Comparison of the Rate of Re-rupture in

Reported Series of Nonoperatively Treated Achilles Tendon Ruptures Study Nistor, 1981344 Lea & Smith, 1972306 Lildholdt & Munch-Jorgensen, 1976364 Stein & Luekens, 1976347 Gillies & Chalmers, 1970302 Jacobs et al, 1978305 Edna, 1980357 Persson & Wredmark, 1979360 Saleh et al, 1992368 Inglis et al, 1976303 Cetti et al, 1993348 Ingvar et al, 2005365 Van der Linden-van der Zwaag et al, 2004366 Moller et al, 2002363 Totals

No. of Cases

No. of Percentage Re-ruptures of Cases

60 66 14

5 7 2

 8 11 12

8 7 32 10 20 40 23 55 196 80

1 1 7 3 7 2 9 7 14 4

13 14 22 30 35  5 39 13  7  5

53 664

11 80

21 12.1

  TABLE 25D-2  Comparison of the Rate of Re-rupture

in Reported Series of Surgical Repairs of Achilles Tendon Ruptures No. of Cases

Study Jacobs et al, 1978305 Inglis & Sculco, 1981304 Lennox et al, 1980307 Inglis et al, 1976303 Percy & Conochie, 1978359 Shields et al, 1978391 Gillies & Chalmers, 1970302 Ma & Griffith, 1977239 Beskin et al, 1987332 Jessing & Hansen, 1975376 Arner & Lindholm, 1959349 Nistor, 1981344 Cetti et al, 1993348 Soldatis et al, 1997361 Mortensen et al, 1999371 Speck & Klaue, 1998373 Aoki et al, 1998370 Sölveborn & Moberg, 1994372 Totals

No. of Re-ruptures

26 159 20 44 74

0 0 0 0 0

32 6 18 42 108 92 44 56 23 71 20 22 17 874

0 0 0 0 2 4 2 3 0 0 0 0 0 11

Percentage of Cases 0 0 0 0 0 0 0 0 0 2 4 5 5 0 0 0 0 0 1.1%

­ istor performed a prospective randomized study and N found only minor differences in the results of the two methods of treatment.344 He concluded that the treatment of choice is nonsurgical because in patients treated conservatively, there were fewer complications, fewer complaints, and no hospitalization. Gillies and Chalmers also

  TABLE 25D-3  Comparison of the Rate of Postoperative Complications in Reported Series of Surgical Repairs of Achilles Tendon Ruptures

Study Inglis & Sculco, 1981304 Jessing & Hansen, 1975376 Arner & Lindholm, 1959349 Percy & Conochie, 1978359 Inglis et al, 1976303 Nistor, 1981344 Beskin et al, 1987332 Shields et al, 1978391 Jacobs et al, 1978305 Lennox et al, 1980307 Ma & Griffith, 1977239 Gillies & Chalmers, 1970302 Soldatis et al, 1997361 Cetti et al, 1993348 Totals

No. of Cases

No.of Major* No. of Minor† Complications (%) Complications (%)

159

20 (13)

—

108

7 (7)

18 (17)

92

22 (24)

49 (53)

74

16 (22)

7 (10)

44 44 42 32 26 20 18

2 (5) 4 (9) 3 (7) 1 (3) 5 (19) 4 (40) 0 (0)

— 29 (64) — — — 1 (5) 2 (11)

6

1 (17)

—

23 56 744

0 (0) 2 (4) 91 (12.2)

2 (9) 15 (27) 123 (16.5)

*Major complications include wound infection, delayed wound healing, skin slough, sinus tract formation, and re-rupture. †Minor complications include adhesions of the tendon to the skin and sural nerve injury. (In several series, the number of minor complications was not reported.)

Foot and Ankle 2005

  TABLE 25D-4  Summary of Studies Comparing the Re-rupture Rate and Strength Achieved after Operative and Nonoperative Treatment of Achilles Tendon Ruptures

Operative Study Nistor, 1981344

No. of Patients

No. of Re-ruptures

Percentage of Cases

Nonoperative Strength* Achieved (%)

No. of Patients

No. of Re-ruptures

Percentage of Cases

Strength* Achieved (%)

44

2

5

83

60

5

 8

79

Gillies & Chalmers, 1970302

6

0

0

84

7

1

14

80

Jacobs et al, 1978305

26

0

0

75

32

7

22

65

Inglis et al, 1976303

44

0

0

88

23

9

39

62

212

10

4.7

N/A

80

4

 5

N/A

80

53

11

20.8

80

255

37

14.5

Van der Linden-van der Zwaag et al, 2004366 Moller et al, 2002363 Totals

59

1

1.7

391

13

3.3

*Strength of the normal leg was used to calculate the percentage of strength achieved after surgery.

concluded that owing to the complications of surgery, the results of ­operative repair of fresh ruptures were not significantly superior to those achieved with nonoperative management.302 Moller and associates also concluded that if re-rupture is avoided, both surgical and nonsurgical treatment produce good functional outcomes.363 In all three of these studies, the authors found no significant difference in the functional strength achieved by surgically and nonsurgically treated patients (see Table 25D-4). In contrast, studies by Inglis and colleagues,303 Jacobs and associates,305 and Cetti and coworkers348 not only documented a much higher rate of re-rupture with nonoperative treatment but also found that surgical repair resulted in significantly better restoration of strength, power, and endurance (see Table 25D-4). Both studies concluded that surgical treatment is the treatment of choice, particularly for active individuals. Two other relevant studies that examined the postoperative strength of patients found that surgical repair of re-rupture after failed conservative treatment and early (<1 month of injury) surgical repair of acute ruptures produced better functional (strength) results than a second (8-week) period of casting and late surgical repair, respectively.304,360

Nonoperative Treatment The largest series of patients treated nonoperatively were reported by Nistor,344 Lea and Smith,306 and Ingvar.365 Nonoperative treatment was popularized by Lea and Smith, who reported on 66 patients in 1972.306 They immobilized their patients in a short leg walking cast with the foot in a gravity equinus position for 8 weeks and allowed weightbearing as soon as the cast was dry. The length (8 weeks) of immobilization was important because they documented a higher incidence of re-rupture with shorter periods of casting. After the cast was removed, the patient began active gastrocnemius strengthening exercises and used a 2.5-cm heel lift for 4 weeks. In their series of patients, the average time to immobilization was 2.6 days (range, 1 to 42 days), and the majority (56%) of patients were placed in a cast within 24 hours

of injury. At an average follow-up of 26 months (range, 5 months to 6.5 years), re-ruptures had occurred in 7 (11%) of these 66 patients. All re-ruptures occurred within 1 to 4 weeks after the 8-week period of immobilization was completed. When a re-rupture occurred, the authors recommended a second full 8-week period of immobilization to ensure a good result.306 Nistor344 modified this treatment protocol by using a below-knee gravity equinus cast for 4 weeks, a similar cast with less equinus for an additional 4 weeks, and a 2.5-cm heel lift for 4 subsequent weeks. In his series of 60 patients, there were five (8%) re-ruptures. All re-ruptures, which occurred between 9 and 16 weeks after the initial rupture, were treated with plaster casts for 7 to 9 weeks. In 2005, Ingvar and associates reported the results of nonoperative treatment in 196 consecutive patients.365 The re-rupture rate was 7%, 7 patients suffered other complications (7 had deep vein thrombosis, 1 had pulmonary embolism), and 62% of the patients reported a full recovery. The authors concluded that the low ­re-rupture rate challenged the claim of other studies that acute Achilles tendon ruptures should be treated ­operatively.365 More recent studies have found that patients treated nonoperatively with splinting that allows controlled early mobilization are able to return to normal activities sooner.300,368,369 In 1997, McComis and colleagues reported on 15 patients who had nonoperative treatment of their Achilles tendon rupture with a functional bracing protocol.300 The brace allowed immediate weight-bearing and active plantar flexion of the ankle, but limited dorsiflexion of the ankle. They graded their results (100-point scoring system) as excellent in 3 patients, good in 9, fair in 2, and poor in 1. Five patients had a positive result on Thompson’s squeeze test at their 2-year follow-up.300 In 2007, Twaddle and Poon found that the common denominator between operative and nonoperative treatment was early motion.369 They found no differences in the outcomes of operative and nonoperative treatment in patients who were treated with early motion controlled in a removable orthosis.

2006 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25D-5  Comparison of the Frequency of Various Procedures Performed for Open Repair of Acute Ruptures of the Achilles Tendon Study 1981304

Inglis & Sculco, Beskin et al, 1987332 Lennox et al, 1980307 Jacobs et al, 1978305 Shields et al, 1978391 Jessing & Hansen, 1975376 Totals

No. of Ruptures

Primary* Suture (%)

Local† Graft (%)

Three-Tissue Bundle (%)

Other‡ (%)

162 42 20 25 32 102 383

79 (49) 19 (45) 16 (80) 16 (64) — 54 (53) 184 (48)

81 (50) 18 (43) 4 (20) 9 (36) 26 (81) 48 (47) 186 (49)

— 5 (12) — — — — 4 (1)

2 (1) — — — 6 (19) — 8 (2)

*In the primary suture group, direct repair with either a Bunnell or an amplified Kessler suture often was secured with a pullout wire. †Local grafts included flaps of gastrocnemius fascia or augmentation with the plantaris or peroneus brevis tendon. ‡Includes reinsertion of the Achilles tendon into the os calcis via drill holes and os calcis bolt fixation.

Operative Treatment Surgical treatment of Achilles tendon ruptures initially recommended by Abrahamson in 1923 and by Queru and Stainovitch in 1929 has become increasingly popular since Arner and Lindholm published their results in 1959.349 Open surgical procedures proposed for acute disruptions (<4 weeks after injury) include end-to-end repair with a Bunnell, modified Kessler, or pullout wire suture304,305,307, 332,359,370-375; end-to-end repair with the three-tissue bundle technique332; and direct repair and augmentation with tendon grafts,332,359 fascial flaps and grafts,304,307,332,376 or synthetic (e.g., Dacron, Gore-Tex, and the like) grafts.358 Closed surgical methods of repair include percutaneous suture239,362,377-382 and external fixation.258 In general, repairs are performed with the aid of a tourniquet and with the patient in the prone position. Although local anesthesia has been used successfully for repair of Achilles tendon ruptures,370,371,383,384 a general or regional anesthetic usually is administered to the patient for open repairs.304,332,359,373,376 Many studies stress the importance of delaying surgery if there is substantial soft tissue swelling. They note that there is no difference in the functional result if the tendon is repaired within 30 days of rupture, and therefore optimizing local skin conditions is preferred over hasty surgical intervention.304,307,332,346 For open repairs, an incision medial to the medial border of the Achilles tendon is recommended to minimize sural nerve injury. Inglis and colleagues recommend a darted incision, with darts created 1 inch apart at a 30degree angle, to distribute skin pressure more evenly along the line of the incision and to increase the operating area of the incision,304 while Lansdaal advocated a minimally invasive procedure.375 The subcutaneous tissues are then opened to expose the deep (crural) fascia of the leg. The Achilles tendon is not a subcutaneous structure but lies beneath the deep crural fascia. The fascial layer is sharply dissected so that it can be closed carefully at the completion of the tendon repair. Closure of the deep fascia not only reduces tension on the skin closure but also acts as an interface between the subcutaneous tissue and the tendon repair and helps to prevent adherence of the tendon to the subcutaneous tissues and skin. Although numerous surgical procedures have been proposed for repairing ruptures of the Achilles tendon, there

is no single, uniformly superior technique. Analysis of the efficacy of one type of repair has been poorly documented because in most published series, several techniques were employed (Table 25D-5). It is evident from Table 25D-2, however, that all the techniques used in the different studies have been uniformly successful. In one study, Jessing and Hansen compared the results of end-to-end suture (54 patients) with those of direct repair augmented with a turndown graft of gastrocnemius fascia (48 patients).376 There was one re-rupture in each group, and they concluded that there were no significant differences in the functional results of these two operative procedures. They did note, however, that although the rate of re-rupture with either technique was very low, the risk for sustaining a rupture of the contralateral tendon was quite high. Eight (26%) of the 31 patients in their series who resumed full sporting activities ruptured the other Achilles tendon within 2 to 14 years of their ipsilateral rupture. Aroen and colleagues reported that 6% of their 168 patients with Achilles tendon repairs subsequently ruptured their contralateral tendon.385 The basic concepts and principles of specific types of repairs are summarized in the following paragraphs. Although some authors recommend excision of the frayed tendon ends,349 most remove only the intervening hematoma and repair the tendon in its frayed state.304,332,376 End-to-end repairs are accomplished with either a Bunnell suture, a Krackow locking loop stitch, or a modified Kessler suture.305,307,359,370,371-373,386 A recent study tested the Krackow locking loop technique and found that increasing the number of locking loops had minimal effect on the strength of the repair.386 However, using a second interlocking suture placed at 90 degrees to the first significantly increased the load to failure of the repair. Two other studies found that the overall strength of the repair can be further enhanced by the edition of an epitenon suture.387,388 However, load to failure was significantly improved only when the epitenon was sutured with a cross-stitch rather than a running stitch technique.387,388 The three-tissue bundle technique, used exclusively for repair of acute ruptures, was first described in the American literature in 1987.332 With this technique, the disorganized tendon fibers are gathered into three bundles by means of three heavy nonabsorbable sutures (Fig. 25D-11A and B). The three bundles are sutured together in a functional

Foot and Ankle 2007

A

B

C

Figure 25D-11   The three-tissue bundle technique. The torn tendon fibers (A) are gathered into three bundles with three nonabsorbable (Bunnell) sutures (B). The three bundles are then interdigitated and sewn to each other with the suture ends and additional nonabsorbable sutures (C). (Redrawn from Beskin JL, Sanders RA, Hunter SC, Hughston JC: Surgical repair of Achilles tendon ruptures. Am J Sports Med 15:1-8, 1987.)

position (see Fig. 25D-11C), and then the tendon sheath is closed over the repair. A recent cadaveric study using eight pairs of fresh-frozen Achilles tendons compared the tensile strength of the triple-bundle technique with the Krackow locking loop technique.389 Biomechanical testing revealed that the triple-bundle technique was almost 3 times stronger (average load to failure, 387 N versus 161 N). After surgery, the foot and ankle are immobilized in below-the-knee plaster splints with the foot in a relaxed equinus position for 5 to 7 days. The patient is instructed to keep the leg elevated with the knee flexed 45 degrees but is allowed to move around on crutches without bearing weight to perform necessary daily functions. The splints are removed 1 week after surgery, and active dorsiflexion is initiated. When 0 or more degrees of dorsiflexion are achieved (usually in 1 to 3 days), a short leg walking cast is applied with the ankle in a neutral position and is worn for 6 to 8 weeks. After the cast is removed, the patient is placed on crutches and returns gradually to full weight-bearing over 4 to 6 weeks. Active range of motion is encouraged as soon as the cast is removed, and no further immobilization or heel lifts are employed. The authors found that range of motion at the time of cast removal was significantly better in patients treated with the three-bundle technique than with other surgical procedures. Although this difference was not present at long-term follow-up, the greater range of motion after cast removal did imply an earlier return to normal activities. Their long-term follow-up (average, 4.7 years) documented that most patients returned to their preinjury level of activity and that most patients did not seek further follow-up more than 6 to 12 months after their surgery.332 Augmentation of end-to-end repairs has been recommended by several authorities.332,339,359,376,390 Lindholm described a method of augmentation that reinforces the repair and prevents adhesion of the repaired tendon to the overlying skin.339 Through a standard medial incision, direct repair of the tendon is performed as described previously. Subsequently, two fascial flaps, 1 cm wide and 7 to 8 cm long, are fashioned from the proximal tendon

and aponeurosis of the gastrocnemius muscle. The flaps, which are left attached at a point 3 cm proximal to the site of the repair, are rotated 180 degrees and twisted so that the smooth external surface faces the subcutaneous tissues. The two flaps are then sutured to the distal stump of the Achilles tendon and to each other so that they completely cover the site of the repair. Lindholm stresses that during wound closure, one must be careful to approximate the sheath over the site of the repair.339 Jessing and Hansen described a modification of the Lindholm technique using only one flap of gastrocnemius fascia.376 The strip of fascia, about 4 cm in width, is rotated on its distal pedicle and turned distally to cover the rupture site (Fig. 25D-12). The smooth outer surface of the gastrocnemius fascia is thus interposed between the site of repair and the crural fascia. Lynn described a method of augmenting end-to-end repairs in which the plantaris tendon is fanned out to make a membrane an inch or more wide for reinforcing the tendon.390 This method can only be applied to acute injuries, however, because after 2 weeks, the plantaris often becomes incorporated in the scar and cannot be identified. With this technique, a standard medial incision 12 to 18 cm long is made parallel to the medial border of the tendon. The ends of the tendon are sewn together, and the plantaris tendon is divided at its insertion on the calcaneus. The tendon is then fanned out to form a membrane, and the membrane is placed over the repair and sutured in place with interrupted sutures. When possible, the Achilles tendon is covered for 1 inch both proximal and distal to the repair. Closure of the wound and postoperative treatment are the same as those described for other primary repairs. After the aforementioned surgical procedures, the foot is placed in 20 to 30 degrees of equinus, and a long leg cast is applied.304,307,359,391 Excessive equinus is not desirable because it produces local subcutaneous tissue and skin ischemia.304 Three weeks after surgery, a short leg cast is applied and is worn for 3 to 4 additional weeks. At that juncture, immobilization is discontinued, and the patient is given heel lifts, 2- to 2.5-cm in height, for both

2008 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

C

Figure 25D-12   Augmentation of a direct repair of the Achilles tendon with one strip of gastrocnemius fascia. After repair of the tendon (A), a strip of gastrocnemius fascia about 4 cm wide and 12 cm long is twisted 180 degrees upon itself and is then turned down over the repair so that its smooth surface underlies the tendon sheath and subcutaneous tissues (B). The defect in the gastrocnemius fascia is closed, and the flap is secured to the Achilles tendon (C).

heels. In general, weight-bearing is not permitted until immobilization is discontinued.304,307,359,391 The elevated heels are worn for 3 to 4 weeks, and during that time the patient is allowed to swim and exercise on a stationary bicycle with instructions to increase force gradually as power returns to the gastrocnemius-soleus complex. Light jogging and athletics are not initiated until about 3 months after ­surgery. In recent years, several studies have advocated immediate motion and early full weight-bearing after surgical repair of acute Achilles tendon ruptures.369,370-373,392,393 Sölveborn and Moberg studied 17 patients who were allowed free ankle motion in a patellar tendon–bearing plaster cast with an outrigger frame under the foot that made weight-­bearing possible immediately after surgery.372 They reported that no re-ruptures or other complications occurred, and range of motion, strength, and recovery time were better than results of traditional ankle immobilization. Several subsequent studies have reported similar results.370,371,373 In 1998, Speck and Klaue prospectively evaluated the clinical outcomes of 20 patients who had 6 weeks of early full weight-bearing in a removable anklefoot orthosis after an open repair of a torn Achilles tendon.373 All 20 patients reached their preoperative level of sports activity and had no significant side-to-side difference in ankle mobility and isokinetic strength. There were no re-ruptures. In a larger prospective study, Mortensen and colleagues randomly assigned 71 patients who had repairs of acute Achilles tendon ruptures to either conventional postoperative management (a cast for 8 weeks) or early restricted motion of the ankle in a below-the-knee brace for 6 weeks.371 They found that early motion patients had a smaller initial loss in range of motion, and returned to work and sports activities sooner than those managed with a cast. There were no re-ruptures in either group. As noted previously, Twaddle and Poon found that outcomes of operative and nonoperative treatment were the same when controlled early motion was used.369 Kangas and

colleagues found that tendon elongation after ­ operative repair was less in patients treated with early motion and that patients who had less elongation had better isokinetic calf muscle strength and significantly better clinical outcomes.392 Closed techniques for repair of acute ruptures have been described in several reports.239,257,362,377-382,394 In 1977, Ma and Griffith reported on 18 tendons that were repaired using a percutaneous suturing technique.239 Through stab wounds, sutures were passed through the proximal and distal stumps of the tendon with straight and curved needles. Through tension on the two ends of the suture, the ends of the tendon were approximated while the ankle was positioned in maximal equinus. The suture was then cut short, tied with a surgeon’s knot, and pushed subcutaneously. The six small puncture wounds (three medial and three lateral) were left open and were covered with dry sterile dressings. Aftercare included a short leg, non–weight-bearing cast for 4 weeks, and then a low-heeled, weight-bearing, short leg cast. When the cast was removed, the patient started toe-heel raising for 4 weeks and then performed heel cord stretching exercises for an additional 4 weeks. Ma and Griffith stated that the procedure could be performed under a local, regional, or general anesthetic. Results of the percutaneous suture were no ruptures in the 18 patients and only two minor complications. One was a skin retraction dimple at the operative site, and the other was a tender nodule at the site of the surgical knot. Small, nontender nodules generally were present in most of the 18 patients at the site of the surgical knot, but the sutures were removed only in the two patients who were symptomatic.239 In the past decade, there have been additional studies to assess long-term results,377 present new percutaneous techniques,362,378,395 and compare the results of percutaneous and open surgical repair.394 In 1990, Bradley and Tibone compared the results of 15 patients treated by open repair augmented with a gastroc-soleus fascial graft, and 12 patients

Foot and Ankle 2009

treated by percutaneous (Ma and Griffith type) repair.394 They found that percutaneous repair resulted in strength levels similar to open repair but had a re-rupture rate of 12%, significantly higher than the results of open repairs. They concluded that open repair should be performed in all high-caliber athletes who cannot afford any chance of re-rupture.394 Recently, new percutaneous surgical techniques have been developed to increase the strength of repairs395 and decrease complications such as sural nerve injury.362,379-382 Webb and Bannister reported results of 27 patients who had repairs through three midline stab incisions rather than Ma and Griffith’s six (three lateral, three medial) skin incisions.362 They had no sural nerve injuries, compared with the 13% injury rate reported with the Ma and Griffith technique. Amlang and associates described percutaneous repairs using the Dresden instrument in 61 patients and reported that very good and good results were achieved in 78% and 20% of the patients, respectively.379 Cretnik and coworkers compared the results of 105 patients who had open repairs with those of 132 patients who had percutaneous repairs.380 There were slightly more re-ruptures (3.7% versus 2.8%) and sural nerve injuries (4.5% versus 2.8%) in the group of percutaneous repairs, but there were significantly more major complications (12.4% versus 4.5%, P = .03) in the group of open repairs. More recent reports have advocated endoscopically assisted percutaneous ­Achilles tendon repairs.381,382 In 2003, Halasi and colleagues compared the results of 89 patients with percutaneous repairs with those of 57 patients who had endoscopically assisted percutaneous repairs.381 They reported excellent to good results in 88% and 89% of the percutaneous and ­endoscopic-­percutaneous groups, respectively. They noted that the re-rupture rate was lower (2 total, 3 partial versus 1 partial) in the endoscopic group. In 1985, Nada described a method of external fixation for treatment of acute ruptures that he used in 33 consecutive patients.257 With this technique, Kirschner wires (0.08 and 12 cm long) are inserted through stab wounds into the calcaneus and the proximal stump of the tendon. The ankle is put into an equinus position, and the proximal wire is pulled distally (1 to 1.5 cm) to approximate the ends of the tendon. The two K-wires are then maintained in their approximated position with an external fixator. The operation was performed under local anesthesia in 26 of the 33 patients. Four weeks after the operation, the clamps and wires were removed, and the patients were placed in an equinus short leg walking cast for an additional 4 to 5 weeks. After 8 weeks, immobilization was discontinued, and the patients were given shoes with the heels built up 2.5 cm. Results at a mean follow-up of 2.4 years (range, 9 months to 4 years) were good or excellent in 30 of the 33 patients. Two patients had a fair result owing to sural nerve injuries,

and one had a poor result owing to a Sudeck’s atrophy. All except the patient with the poor result returned to their original level of activity, and there were no re-ruptures.257 In patients with neglected Achilles tendon ruptures, end-to-end repair often is not possible after excision of the intervening scar tissue. In these cases, grafts of fascia lata, local tendons (plantaris and peroneus brevis), and strips of Achilles tendon or free autologous gracilis tendon grafts have been used to bridge the gap.345,396-399 Mann and colleagues opposed the use of free fascia and turn-down fascial grafts because they are avascular structures that must be revascularized to be incorporated into the repair.17 They also noted that transfer of the peroneus brevis carries the risk for changing the balance between the everters and inverters of the foot. Thus, they recommended transfer of the flexor digitorum longus and reported good and excellent results in six of the seven patients in which this procedure was performed.17 In contrast, Sebastian and associates recommended the use of either peroneus brevis or flexor hallucis longus for reconstruction of chronic Achilles tendon ruptures.398 Abraham and Pankovich also found that results of repair of neglected rupture with strips of Achilles tendon and fascia lata were unsatisfactory.400 They concluded that only an end-to-end repair of the chronic lesion would promote optimal functional recovery and restore muscle strength. Thus, they described a technique by which an end-toend anastomosis was made possible by a proximal V-Y tendinous flap. With this technique, the tendon ends are appropriately trimmed, and the length of the defect in the tendon is measured with the knee in 30 degrees of flexion and the ankle in 20 degrees of plantar flexion. An inverted V incision, with arms that are 11⁄2 times the length of the defect, is made over the central aponeurosis of the gastrocnemius. The incisions extend through the fascia and underlying muscle tissue of the gastrocnemius. The flap is then pulled distally and sutured to the distal stump of the ruptured tendon. Maffuli and Leadbetter reported the use of free autologous gracilis tendon grafts in 21 patients and concluded that the procedure is safe but technically demanding and affords good results even in neglected ruptures of 9 months duration.399 Postoperatively, the patient is placed in a long leg cast for 6 to 8 weeks and then a short leg, weight-bearing cast for an additional 4 to 6 weeks. The patient is subsequently placed in shoes with 3- to 5-cm heel lifts, which are worn for 1 month. Progressive resistive exercises are initiated immediately after the second cast is removed. In their series of four patients, there was only one surgical complication (a neuroma of the sural nerve), and three of the four patients regained full strength of the triceps surae muscle. The other patient had some residual weakness but was happy with the result.400

Authors’ Preferred Method Operative repair of Achilles tendon ruptures is our treatment of choice because most of our patients are intercollegiate or recreational athletes. Nonoperative treatment is always ­offered and discussed but usually is not recommended owing

to the reported rate of re-rupture of 7% to 39% (see Table 25D-1) and the loss of strength after this method of treatment (see Table 25D-4). We have, however, treated acute ruptures with cast immobilization or a removable ­orthosis in Continued

2010 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� 400

Authors’ Preferred Method—cont’d

A

B

Figure 25D-13   The authors’ method for repair of Achilles tendon ruptures. The frayed ends of the tendon (A) are not routinely débrided. The ends of the tendon are reapproximated with a No. 5 nonabsorbable suture (B), and the peripheral fibers are aligned and repaired with a running 2-0 absorbable suture.

patients who declined to have surgery, in elderly and chronically ill patients, and in middle-aged, sedentary, executivetype individuals who feel they cannot afford the time away from work required for operative treatment. Our preferred method of repair is shown in Figure 25D-13. The procedure is performed on an outpatient basis, under a general or regional anesthetic. The patient is placed prone on the operating table, and the ruptured tendon is approached through a medial incision that parallels the medial border of the Achilles tendon. The skin and subcutaneous tissue are gently retracted with skin hooks and narrow retractors to ­ expose the crural fascia. The fascia is incised along the ­medial border of the tendon so that its closure will be offset from that of the skin incision. The ends of the tendon (see Fig. 25D-13A), which are not routinely débrided, are reapproximated with a Krackow locking loop stitch of 5-0 nonabsorbable suture. The peripheral fibers of each stump subsequently are sewn together with a 2-0 absorbable suture using a cross-stitch technique (see Fig. 25D-13B). The repair is then reinforced with a strip of gastrocnemius fascia that is twisted 180 degrees on its distal pedicle and rotated distally to cover and extend 2 to 3 cm beyond the site of repair (see Fig. 25D-13C). A strip that is 7 to 8 cm long and 1.2 to 1.4 cm wide usually is sufficient because the fascia can easily be stretched to a width of 2 to 3 cm with tissue forceps to encompass the width and length of the sutures used for the primary repair. The fascial strip is secured to the distal stump and to the medial and lateral

edges of the tendon with a running 2-0 absorbable suture (see Fig. 25D-13C). The ankle is then placed in a neutral position to assess the degree of tension on the repair. The defect in the gastrocnemius fascia is closed with a running 0-0 absorbable suture, and the sheath of the Achilles tendon is closed with interrupted and running 2-0 absorbable suture (see Fig. 25D-13D). The subcutaneous tissues are closed with buried 2-0 absorbable sutures, and the skin is closed with a running subcuticular, 3-0 nonabsorbable suture and 3 to 4 escape stitches. The leg is placed in a short leg plastic boot with the ankle in 20 degrees of plantar flexion. Two weeks after surgery, the subcuticular suture is removed, the angle of flexion of the ankle hinge on the boot is ­decreased to 10 degrees, and the patient is allowed full weight-bearing with the ankle locked in 10 degrees of plantar flexion. Three weeks after surgery, the ankle is brought up to a neutral position, and the patient is allowed to continue full weight-bearing on the extremity. Immobilization is discontinued 4 weeks after the repair, and the patient begins active and active-assisted range of motion exercises, swimming, and stationary bicycle activities and is allowed to ambulate in a shoe that has a 1-cm heel lift. When the patient has achieved a full range of motion (usually in 2 to 4 weeks), strengthening (e.g., one-leg heel ­raises) of the extremity is initiated. Patients begin the previously described running program when isokinetic strength testing reveals that their strength is at least 70% of that of the ­nonoperated extremity. They are allowed to return

Foot and Ankle 2011

Author’s Preferred Method—cont’d

C

D

Figure 25D-13, cont’d   C, The repair is reinforced with a turned-down flap of gastrocnemius fascia. The defect in the gastrocnemius fascia and the Achilles tendon sheath are repaired (D) before closure of the wound.

to athletic activities when they have full strength and full endurance and have completed the aforementioned running program. We also encourage athletes to complete an ankle joint proprioception training program. Bressel and colleagues have shown that patients with a previous history

C

r i t i c a l

P

of an Achilles tendon rupture have proprioception deficits in both limbs, which adversely affect their function.401 In most cases, the athletes have achieved full strength and completed the running and proprioceptive programs ­within 6 months of ­surgery.

o i n t s

l Early recognition and management of tendon injuries of the foot and ankle can optimize outcome. l A thorough history of the injury and physical examination to evaluate potential tendon injuries are usually adequate for diagnosis; however, MRI or ultrasound examination are useful for diagnosis of partial injuries, evaluation of associated structures, and determination of the location of the injury and tendon ends. l Most closed disruptions of tendons of the foot and ankle are preceded by chronic degeneration of the tendon. l Early surgical intervention in young active patients with tendon disruptions to the foot and ankle may lead to im­proved outcomes compared to nonoperative management. l After a rupture of the Achilles tendon, an MRI of the Achilles tendon is needed only if nonoperative treatment is selected to determine if the ends of the tendon are apposed when the ankle is in plantar flexion. l After Achilles tendon repairs, early motion and weight bearing produce the best results.

S uggested

R eadings

Bell W, Schon L: Tendon lacerations in the toes and foot. Foot Ankle Clin 1:355-372, 1996. Hamilton WG, Geppert MJ, Thompson FM: Pain in the posterior aspect of the ankle in dancers: Differential diagnosis and operative treatment. J Bone Joint Surg Am 78:1491-1500, 1996. Myerson MS: Adult acquired flatfoot deformity: Treatment of dysfunction of the posterior tibial tendon. Instr Course Lect 46:393-405, 2005. Redfern D, Myerson M: The management of concomitant tears of the peroneus longus and brevis tendons. Foot Ankle Int 25(10):695-707, 2004. Scaduto AA, Cracchiolo A III: Lacerations and ruptures of the flexor or extensor hallucis longus tendons. Foot Ankle Clin 5:725-736, 2000. Squires N, Myerson MS, Gamba C: Surgical treatment of peroneal tendon tears. Foot Ankle Clin 12(4):675-695, vii, 2007. Thompson TG, Doherty JH: Spontaneous rupture of tendon of Achilles: A new clinical diagnostic test. J Trauma 12:126-129, 1963. Thoradson DB, Shean CJ: Nerve and tendon lacerations about the foot and ankle. J Am Acad Orthop Surg 13:186-196, 2005.

R eferences Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 2012 DeLee & Drez’s� O

S e c t i o n

E

Stress Fractures of the Foot and Ankle James W. Brodsky and Nathan Bruck

Repetitive stress or “overuse” injuries of the bones of the foot and ankle cause numerous problems and disturbances in training and recreation. Beginning even before President John F. Kennedy’s public awareness program in the early 1960s, fitness and recreation have gained increasing attention and importance in modern American society. As sports and exercise have become the focus of personal health programs, as well as preventive medicine projects that promote healthier lifestyles, stress-related injuries have become more and more commonplace. They are now estimated to account for up to 10% of a sports medicine practice.1 The first description of a stress fracture was recorded in 1885 by the Prussian military physician, Breithaupt, who described soldiers with swollen, painful feet. After the discovery of roentgenographs in the late 19th century, this condition was attributed to a metatarsal shaft fracture, and the term march fracture was coined. Through the ensuing 100 years, numerous reports have described fatigue-type fractures; most of the early literature identified these injuries in a military setting.2-10 The literature continues to abound with reports of many different kinds of stress fractures that affect all manner of athletes, sportspersons, and individuals engaged in health-conscious fitness programs. A wide range of the population is affected as well, including both sexes and all ages from adolescence to senescence, as well as athletes participating in virtually every kind of sporting endeavor.11-36 Recent literature has placed a new emphasis on the nature and occurrence of stress injuries in elite athletes as well as on nutritional and hormonal contributions to disease.1,12,13,17,19,28,30,37 A stress fracture can be defined as a partial or complete fracture of bone that results from repeated application of a stress lower than that required to fracture the bone in a

   TABLE 25E-1  Characteristics of Stress Fractures Bone

Location

Activity

Hallucal sesamoids

Medial > lateral

Metatarsal shaft, neck

2nd, 3rd, 4th

Metatarsal base Metatarsal shaft base (Jones’ fracture) Navicular

2nd 5th

Running, football, golf, gymnastics Military recruits, running, athletes Ballet dancers Basketball, football, soccer Track and field, basketball, football, soccer Running, jumping, basketball Distance runners

Dorsal, middle

Medial malleolus Tibia

Shaft or distal metaphysis

single loading situation.15 Stress fractures can be divided into two types—fatigue fractures and insufficiency fractures. Fatigue fractures result when histologically normal bone is subjected to repetitive stress below the monotonic fracture threshold, leading to incomplete repair and eventual mechanical failure. These tend to occur in specific anatomic sites based on particular loading patterns associated with specific activities (Table 25E-1). Insufficiency fractures, in contrast, occur when mechanical failure of the bone results from relatively normal stresses applied to a histologically abnormal bone, which is deficient in mineral or abnormally inelastic, or both.38 Insufficiency fractures typically occur in postmenopausal osteoporotic women.25 They are also seen in patients with underlying conditions such as Paget’s disease, diabetes mellitus, hyperparathyroidism,39 rheumatoid arthritis,38,40 and postirradiation osteopathy.1

ANATOMY Depending on what one includes, the foot and ankle is composed of about 30 bones (i.e., tibia, fibula, talus, calcaneus, navicular, cuboid, three cuneiforms, five metatarsals, fourteen phalanges, and two hallux sesamoids) and associated accessory ossicles and variable sesamoids joined by myriad ligamentous structures. Extrinsic and intrinsic musculotendinous units power the articulations that allow for both motion and dissipation of force. The normal walking cycle consists of a stance phase and a swing phase, with the stance phase consuming about 62% of the walking cycle. The stance phase is divided into three intervals: heel strike, foot flat, and push-off. During the first interval, the foot is being loaded with the weight of the body while the ankle undergoes rapid plantar flexion under control of the anterior musculature. The foot is in pronation that originates at the subtalar joint, allowing absorption and dissipation of the forces generated by ground contact. The second interval begins with foot flat and forward transfer of the center of gravity over the foot. The ankle initially dorsiflexes, and the subtalar joint demonstrates progressive inversion brought on by multiple factors. The progressive inversion helps transform the flexible hindfoot and midfoot into a rigid lever. In the third interval, rapid plantar flexion of the ankle by concentric contraction of the posterior musculature occurs with further inversion of the subtalar joint, reaching a maximum at toe-off. When the heel is elevated at the time of lift-off, the weight of the body is normally shared by all of the metatarsal heads. Because of the oblique nature of the metatarsophalangeal break, the foot must supinate to distribute the body weight evenly among the metatarsal heads.

Foot and Ankle 2013

Alterations and variations in anatomy and range of motion may affect the loading of the foot during the stance phase of gait, which can lead to localized overloading and subsequent stress fracture. Patients with rigid foot deformities such as cavovarus foot or tarsal coalition are poorly able to absorb the forces during the first interval of the stance phase.41-43 Stress is subsequently transferred to the midfoot (navicular, base of metatarsals), laterally (fifth metatarsal), or distally to the forefoot. Inability to invert (supinate) the hindfoot during the second and third intervals leads to failure to distribute the body weight evenly among the metatarsals. Kaufman and colleagues showed that pes planus increases the risk for stress fracture42 and does not protect the foot, as other studies have suggested.41 Clinically, overloading of the medial column of the foot occurs with sesamoid symptoms. Many authors accept that an abnormally short first metatarsal (congenital, traumatic, iatrogenic) or an abnormally long second or third metatarsal (Morton’s foot) contributes to a stress fracture because of loss of the normal metatarsophalangeal break, or abnormal pattern of pressure distribution within the forefoot, which results in isolated overloading of specific metatarsals.44 Other authors showed that the lengths of the first and second metatarsals and first and second toe lengths had no effect on metatarsal stress fracture incidence.45 Conversely, some authors demonstrated that the relative length of the second metatarsal is a risk factor regarding stress fracture at the base of the second metatarsal.46 No correlation was found between first ray hypermobility and second and third metatarsal stress fractures.47 Kaufman was unable to demonstrate a relationship between loss of ankle or hindfoot motion and stress fracture.42 There is one report of a stress fracture of the anterior process of the calcaneus, in essence a portion of the coalition, in the presence of a calcaneonavicular coalition that caused restricted subtalar motion.48 In an inverse scenario, there is also a report of talar stress fracture developing after resection of a talocalcaneal coalition.49

BIOMECHANICS Bone is a dynamic organ that is capable of repair and reconfiguration. Wolff’s law states that stresses applied to a bone stimulate remodeling of the bony architecture for optimal withstanding of the forces being applied. Rapid disuse osteopenia results when normal forces are withdrawn from a healthy bone. This phenomenon is clearly demonstrated in experimental models as well as in humans experiencing weightlessness.50 During steady state, daily stresses applied to bone stimulate development of a bony architecture that is optimally fit in terms of strength and density, according to Wolff’s law. When activities and therefore stresses are increased, remodeling of the bone takes place. The exact mechanism that initiates this process is unknown. Several authors state that it is likely due to the creation of microfractures.1 Microfractures stimulate osteoclastic resorption in initiation of the repair process. As osteoclasts resorb bone, osteoblasts fill in the voids with new osteons. If high levels of activity continue, microfractures and osteoclastic resorption exceed osteoblastic new bone formation, resulting in metabolic imbalance. If stresses continue to accumulate at a rate beyond that occurring in the repair process,

fatigue fracture may result. Although microfractures of the bone are a physiologic process that responds to Wolff’s law, disease is apparent when reparative processes cannot keep pace with the fracture process. Stresses to bone result from both tension and compression. Tensile forces occur on the convex sides of long bones; compressive forces occur on the concave side. Muscular activity and muscle fatigue can attenuate and increase, respectively, these forces applied to bones. In normal, nonmaximal activity, muscular contractions tend to dampen the forces applied with weight-bearing. As activity level and duration increase, muscles fatigue, and the dampening effect of muscle diminishes and the stresses at the bones increase.51,52 Microfractures begin to accumulate, thus shifting the balance in favor of bone damage. Continued impact forces from weight-bearing lead to damage that exceeds repair capabilities and eventually to fatigue fracture. Fatigue fracture is a function of the number of loading cycles and the amount of force applied. Because the metabolic balance in normal healthy bones relies on the ability of bone to heal, osseous vascularity plays an important role in this balance. As such, “watershed areas” of bone with relative hypovascularity (proximal metaphyseal-­diaphyseal junction of the fifth metatarsal; central third of tarsal navicular; metatarsal necks; anterior tibia) are known to be predisposed to fatigue fractures.

RISK FACTORS Various risk factors for development of a lower extremity stress fracture have been identified. Clearly, running is responsible for the greatest number of stress fractures in the nonmilitary population.1 Running is the athletic activity that produces the greatest number of repetitions of the same (high) impact-loading activity. The force of each stride is a magnification of body weight, as has been shown in the gait studies of Mann and associates.53 Thus, a 70-kg individual dissipates 120 tons of force per foot per mile walked, and this increases to 210 tons of force per foot per mile in running.53 It is not hard to see why runners and joggers develop stress injury after running multiple miles, several times a week, on concrete streets. In this context, it is hard to understand why the incidence of stress fractures is not higher still. Other predisposing factors may include poor training techniques, such as abruptly increasing the intensity, duration, or type of training. Alterations in warm-ups, running surface or grade, shoewear, and orthoses may be predisposing factors, although Milgrom and colleagues showed no difference in stress fracture rate in military recruits when using insoles as a prophylactic ­measure.54 Other impact activities such as jumping, sprinting, and hurdles can independently lead to stress fractures; they also increase the risk for stress injury in runners. Anatomic factors play a role in stress-related injuries. The height of the arch has been shown by Simkin and colleagues43 to correlate with an increased risk for stress fracture. A cavus foot deformity tends to be rigid and unable to absorb forces imparted during physical activities. Leglength discrepancies overload the longer limb, resulting in the application of higher stresses. Hallux valgus deformities and long second metatarsals (Morton’s foot) lead to second ray overload, with development of stress fractures

�rthopaedic ����������� S �ports ������ � Medicine ������� 2014 DeLee & Drez’s� O

as well as to the rare stress fracture of the first proximal phalanx.55 Excessive external rotation at the hip has been shown to increase the risk for stress fracture.4 Female gender as a risk factor has received increased attention in the recent literature on stress fractures. In their study of stress fractures in dancers, Kadel and coworkers24 showed that excessive training and prolonged duration of amenorrhea contribute independently to the risk for fracture. Bone mineral density has been shown to be significantly decreased in amenorrheic athletes compared with eumenorrheic athletes and controls.28,56 Girls with anorexia nervosa are at higher risk for stress fractures, and the bone appearance on magnetic resonance imaging (MRI) is altered and thus may interfere with stress fracture diagnosis.57 Although estrogen supplementation has been shown to be clearly beneficial in preventing postmenopausal bone loss, as well as to increase bone mineral density in amenorrheic woman,58 the literature is contradictory as to whether estrogen supplementation decreases the incidence of stress fracture among amenorrheic athletes.56,58,59 Prophylactic treatment with risedronate did not appear to decrease the risk for stress fractures in military recruits.60 Previous surgical procedures to the lower extremities may expose the bones to new or varied forces. Metatarsal osteotomies, either for hallux valgus or for lesser ray problems, may lead to overloading of adjacent rays and increased risk for subsequent development of stress fractures.61 Ankle and hindfoot procedures to fuse or realign the joint can lead to increased loading, especially if there is a malunion. Sammarco and Idusuyi62 reported a case of a third metatarsal stress fracture following endoscopic plantar fascia release. They postulated that excess force was created because of the loss of the stress-relieving function of the plantar fascia. Stress fractures following total joint replacement have been reported, presumably owing to newly increased stresses in a poorly conditioned, previously underused extremity.63 Age has been shown to be a risk factor for stress fracture, likely related to osteoporosis. Myburgh and associates28 showed that athletes with stress fractures had lower bone mineral density in both the appendicular and the axial skeleton. They also showed that these same athletes had lower dietary calcium intake than did control subjects. White race has also been shown to be a risk factor.

DIAGNOSIS History Although history and physical examination have classically been the standard for diagnosis of stress injuries, imaging studies have evolved to aid significantly in the diagnosis. Patients sustaining stress fractures of the foot and ankle generally present with a history of insidious onset of a vague, aching pain. The patient usually does not remember a specific traumatic event but may note a recent change in activity level or a change in type or duration of training. For the casual athlete, a new recreational sport or a diet-related exercise program may be the only change. For more highly trained athletes, the change may be as simple as new athletic shoes, a different surface for running, or

minor increases in speed or distance. Routine daily activities without a history of strenuous activity may produce insufficiency stress fractures in patients with osteoporosis or underlying diseased bone.25 Dull, aching pain is characteristically present midway through or near the end of the activity; it usually diminishes with cessation of the activity. The patient can usually localize the pain to a specific area. This pain progresses and eventually may develop into a sharp, severe pain that results from all weight-bearing activities as the stress fracture becomes complete. Although a history of menstrual irregularities certainly aids in the diagnosis of stress fracture, a high index of suspicion is necessary early in the course, especially if a history of increased activity level is not forthcoming.

Physical Findings The most reliable physical finding is a discrete, localized area of tenderness and swelling over the affected bone. Sixty-six percent of the athletes with stress fractures in Matheson’s study had localized tenderness, and 25% had soft tissue swelling.64 Erythema and warmth may be ­present but are much less likely. Patients may present with a limp but rarely with muscular atrophy. Joint range of motion is usually not affected. Percussion and vibratory and ultrasonic stimulation have been reported to aid in the diagnosis but are usually unnecessary.65 Physical examination should also include assessment of ankle and hindfoot position. Equinus contracture or hindfoot varus can predispose an athlete to increased levels of forefoot or lateral foot stress, respectively. Alignment and range of motion of the knee and hip should be assessed as should leg lengths.

Imaging Studies A patient’s history of temporally related altered activity level, along with findings of localized tenderness, is often sufficient for the diagnosis of stress fracture to be made and treatment initiated. Plain radiographs, however, should still be obtained. Standard three-view studies of the ankle and of the foot are initially performed (anteroposterior, lateral, and oblique views of the foot; anteroposterior, lateral, and mortise views of the ankle). Special oblique studies can be obtained as required by the particular case. Not uncommonly, in fact, more often than not, the initial radiographs are negative, with radiographic changes lagging behind clinical symptoms by at least 2 to 3 weeks.8,25 Cortical hypertrophy may be present, suggesting a condition of longstanding overload. Linear radiolucencies from osteoclastic resorption may eventually appear. More commonly, increased radiodensity as the body attempts a healing response can be present. Prather and colleagues8 found radiographic assessment alone to be diagnostic only 64% of the time. In many patients with nonarticular involvement, diagnosis of stress fracture by history and physical findings, despite negative plain radiographs, is sufficient (and reasonable) for treatment to be initiated. With higher risk lesions or in high-caliber or elite athletes, further diagnostic studies are appropriate. Bone scintigraphy with radionuclide technetium-99m diphosphonate has an advantage because

Foot and Ankle 2015

of its ability to demonstrate early physiologic abnormality, thus increasing sensitivity.66 Bone scans are routinely positive as early as 2 days after injury and may be so as early as 24 hours.67 The isotope is incorporated into the bone by the activated osteoblasts. Even in subradiographic stress fractures, scintigraphic uptake is usually intense (Fig. 25E-1). The possibility of a false-positive study should be considered because it has been shown that in cases of os trigonum, no correlation was found between technetium-99m diphosphonate uptake and the clinical picture.68 Some fractures have a characteristic appearance on bone scan, and certain fractures (e.g., navicular, sesamoid, talar processes) often require further anatomic definition with computed tomography (CT). MRI is the most commonly used adjunct in the diagnosis of stress fractures (Fig. 25E-2). Stress fractures have a characteristic appearance on MRI, that is, low signal intensity on T1-weighted images and high signal intensity on T2-weighted images.69 T1 and short T1 (tau) inversion recovery (STIR) sequence images demonstrate the lesion best when it is out of the acute phase.70 MRI has the advantage of better accuracy in diagnosis than bone scan in terms of definition of the anatomy of the fracture site.69 The benefits may be offset by the higher cost, and often the MRI diagnosis does not change the treatment decisions. Thus, this approach should not be used indiscriminately in all patients with suspected stress fracture. With the more frequent use of MRI and considering its high sensitivity, more subtle findings may be addressed as stress fractures. A more rigorous definition of the term stress fracture would be reserved for cases in which a fracture line can be identified either on T1, T2, or STIR sequences. Other findings, such as bone edema without

A

a definite fracture line, could be addressed as stress reaction, although there is obvious ambiguity between the two terms, presumably because the physiologic process is on a continuum.

TREATMENT PRINCIPLES The treatment of stress fractures of the foot and ankle may vary somewhat in duration, timing, and the election of aggressive intervention in comparison with the treatment of those fractures caused by acute, discrete traumatic incidents (Table 25E-2). These variations are discussed later according to the specific fractures. The basic principles of fracture care must still be followed, however, to achieve predictable healing. The two most basic of these principles are rest and immobilization of the injured limb. Rest takes on an additional (and negative) nuance of meaning for many athletes in sports-induced stress fractures because it is most often the highly repetitive motions and impacts of the sport itself that induce the injury. Rest, then, does not simply suggest the sensible alteration of everyday activity that comes to mind in the nonathlete; rather, it means that the athlete must severely modify or even suspend the previously intense pursuit of the sport. This can, and often is, met with howls of protest and anger, given the passion or payment associated with participation in amateur or professional sports, respectively. The physician or surgeon is, in turn, subjected to pressure from patient, parent, coach, trainer, or agent (sometimes subtle and often overt) to violate these basic principles so that the athlete may return to the sport as quickly as possible. Sometimes the doctor feels that the demand is not just for a prognosis, nor even treatment, but for something just short of magic, as if the

B

Figure 25E-1   A, Minimal radiographic findings with no cortical involvement. B, Diagnosis of tibial stress fracture confirmed by technetium-99m bone scan. (Courtesy of James W. Brodsky, MD.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2016 DeLee & Drez’s� O

A

B

Figure 25E-2   A and B, Tibial stress fracture: incomplete medial to lateral, but cortical portion facilitates diagnosis. C, Magnetic resonance imaging also demonstrates the lesion with high sensitivity. (Courtesy of James W. Brodsky, MD.)

C

   TABLE 25E-2  Authors’ Preferred Method of Treatment and Time to Return to Sport Stress Fracture

Treatment

Return to Sport (mo)*

Tibia Medial malleolus Distal fibula Calcaneus Talus Navicular Metatarsals 2-4 Proximal metaphyseal-diaphyseal junction of the 5th metatarsal Medial sesamoid

Conservative—NWB Open bone grafting and plate fixation Conservative—WBAT in a CAM walker Conservative—NWB Conservative—NWB Open bone grafting and fixation with two lag screws Conservative—WBAT in a CAM walker Early (no sclerosis)—closed fixation with an intramedullary screw Late (sclerosis)—open bone graft and intramedullary screw fixation Conservative—off-loading orthosis; if displaced fracture, sesamoidectomy and FHB reconstruction or bone grafting

4-5 3-4 2-3 2-3 3-6 4-6 1.5-2 3-4

*Approximate time from treatment. FHB, flexor hallucis brevis; NWB, non–weight-bearing; WBAT, weight-bearing as tolerated.

2-4

Foot and Ankle 2017

f­ racture will be healed if the doctor would only acquiesce to a quick return to the sport. In addition to the science discussed elsewhere in this chapter, successful treatment usually requires the skillful application of the art of medicine, encompassing that unique blend of rational decision making, sympathetic coercion, and thoughtful ­compromise. Rest, then, signifies decrease in, or cessation of, the sport, together with the judicious use of cross-training, so generalized deconditioning is minimized while the specific injury to the limb is protected from further microtrauma and repetitive stress. Traditionally, immobilization has signified cast application. The advent of prefabricated braces, orthoses, and walking boots, as well as rigid-soled shoes, has made it easier to comply with immobilization. Although the use of casts has been supplanted in many cases, there are still many instances in which the superior immobilization provided by a cast is required in treatment of the fracture. The advantage of the removable devices is that even in fractures that require strict non–weight-bearing, the benefit of being able to remove the device for bathing is exceeded by the benefit of the patient being able to perform active range of motion exercises to maintain some muscle tone and bulk and to reduce joint stiffness and the period of rehabilitation once the fracture is healed. Most stress fractures of the foot and ankle heal by the application of these two principal treatments alone. Patient compliance with instructed use should never be forgotten as possible cause of treatment failure when using removable orthotic devices and other forms of treatment with a discretionary component. Dietary change and metabolic treatment are important adjuncts to treatment of stress fracture in the patient with insufficiency fracture, less for the healing of the specific stress fracture than for the prevention of future stress fractures because stabilization or reversal of the disease process is slow. Fractures in osteoporotic bone are not slower to heal than those in normal bone; however, they heal with the same porotic bone. Dietary and metabolic interventions are frequently supervised by an internal medicine or endocrine specialist. These include adequate intake of calcium with vitamin D in a diet that has sufficient calories and protein to ensure that the patient is not in a catabolic state. These interventions also frequently include hormone replacement therapy in postmenopausal women (unless there is a medical contraindication such as a history of estrogen-sensitive breast cancer) and often the use of bisphosphonates such as alendronate (Fosamax), risedronate (Actonel), and ibandronate (Boniva), or the parathyroid hormone–like drug teriparatide (Fortéo). The use of osteoblast-inhibiting medications as a preventive measure has not shown to reduce the risk for stress fracture.60 Although surgery is required in the minority of stress fractures of the foot and ankle, the need for surgery follows the same basic indications as for other fractures: nonunion, malunion, or displacement that would lead to either of the two. In practical terms, there are several stress fractures of the foot and ankle for which surgery is more the rule than the exception, although the opposite is true for the rest. Those fractures that frequently require surgery are vertical stress fractures, distal medial tibia fractures, stress fractures of the tarsal navicular, and Jones’ fifth metatarsal fracture

at the junction of the proximal metaphysis and diaphysis of the fifth metatarsal. Fractures that not infrequently require surgery are stress fractures of the base of the second metatarsal, disruption of the synchondrosis between the accessory and primary navicular bones, and stress fractures of the sesamoids of the hallux. Specific recommendations are discussed in the following sections. It should be emphasized to all patients with fractures, injuries, or surgeries of the foot or ankle that it is typical to have a great deal of swelling, and for the swelling to be persistent, often lasting 6 months or longer.

SPECIFIC FRACTURES AND THEIR TREATMENT Ankle Fractures Stress fractures of the supramalleolar area are relatively uncommon. Fractures of the distal tibial metaphysis are most often associated with overuse, usually in the absence of deformity. Factors that increase the risk for a distal tibial stress fracture include previous prolonged immobilization, change in the modulus of elasticity of the bone caused by adjacent internal fixation, osteoporosis, and sudden increase in physical activity. Radiographic appearance often does not include a cortical component of the fracture. The initially negative radiographs later demonstrate condensation of new bone within the diaphysis (see Fig. 25E-2A and B), which seldom progresses to a complete fracture or to any discernible displacement (Fig. 25E-3). Treatment consists of immobilization and patience because these tend to be slow to heal. An initial period of non–weight-bearing of 6 to 8 weeks is followed by weight-bearing immobilization until symptoms subside entirely. Weaning out of the removable walking brace is followed by participation in a rehabilitation program. Vertical fractures of the distal medial tibia, ­including and extending superiorly from the medial malleolus, require

Figure 25E-3   Less typical finding of cystic resorption at tibial stress fracture site. (Courtesy of James W. Brodsky, MD.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2018 DeLee & Drez’s� O Figure 25E-4   A, An intra-articular, vertical stress fracture of the medial malleolus in a high school quarterback. B, Healed fracture, one year after percutaneous planting of the fracture.

A

special consideration of mechanical factors. Vertical forces involved are apparent relative to orientation of the fracture line. Predisposing factors include varus at the ankle or of the hindfoot that causes wedge-like force from the medial corner of the talus against the medial ankle mortise, although this is not always present. Open reduction with internal fixation requires an antiglide buttress plate to counteract the vertical forces. Two screws in the medial malleolus are usually inadequate and likely to fail (see Fig. 25E-6). Figure 25E-4 shows a medial malleolus fracture before displacement in high school quarterback. The plate was placed percutaneously as an outpatient procedure, and the patient returned to play after 3 months. The symptoms subsided completely. In contrast to distal tibial fractures, stress fractures of the distal fibular diaphysis and metaphysis are not infrequently associated with deformity. Typically, this would be a valgus hindfoot that increases biomechanical stress on the distal fibula (Fig. 25E-5). Otherwise, the same predisposing factors apply as those enumerated in the discussion of stress fractures of the distal tibia. As early as 1940, Burrows subclassified stress fractures of the distal fibula into two groups—young male athletes with fractures 5 to 6 cm proximal to the distal tip of the lateral malleolus, and middle-aged women with fractures 3 to 4 cm proximal to the malleolar tip (Fig. 25E-6).25,26,71 Treatment in the two groups is substantially the same, that is, immobilization with or without activity modification. In the second subgroup, consideration should be given to metabolic status and appropriate diagnostic tests, or referral should be considered. As with all stress fractures, clinical healing may precede completion of radiographic fracture healing. Surgery is seldom required. Fractures of the medial malleolus, which are even less common than supramalleolar fractures of either the tibia or fibula, tend to be fatigue-type fractures due to overexertion in athletes or patients initiating a new exercise program. Although general treatment principles should be followed, treatment is more likely to be surgical because of

B

a greater proclivity for nonunion. Nonsurgical treatment with immobilization can be used initially in most patients. Exceptions include fractures that are fully intra-articular, displaced fractures, and fractures in selected professional or other serious athletes who require the fastest possible return to sport. Depending on the facture pattern, percutaneous screw fixation may not be sufficient mechanical stabilization, especially if the fracture has displaced or has developed a delayed union or nonunion. Medial plating and bone grafting is usually required in cases of nonunion or delayed union. Illustrated in Figure 25E-7 is a professional basketball player with recurrent stress fractures of both medial malleoli. Neither pedobarographic analysis nor gait laboratory evaluation revealed an underlying biomechanical abnormality to explain the occurrence of the fractures. Treatment in this individual was variably successful, with failures and recurrences both after periods of immobilization and after internal fixation with and without bone grafting.

Fractures of the Hindfoot and Midfoot Stress fractures of the talus are relatively uncommon and are seen most often in runners and military recruits but can occur in dancers and jumping athletes such as gymnasts. Fractures classically occur in the talar neck, but several reports describe talar process (i.e., lateral, posteromedial, posterolateral) stress fractures as well as talar body stress fractures in gymnasts.72 Treatment is usually ­nonoperative, but the challenge is to make the diagnosis promptly. Plain radiographs are typically unrevealing, and an imaging study such as MRI or technetium scanning is required in most cases. Misdiagnosing talar osteochondral lesion as a stress fracture should be avoided. Making the diagnosis can be difficult owing to the misleading, deep, but vague localization of the patient’s pain. Healing time varies, averaging between 3 and 4 months, but occasionally longer. Fractures of the posterolateral process of the talus and disruption of the synchondrosis between an os trigonum

Foot and Ankle 2019

A

B

Figure 25E-5   This patient with valgus hindfoot (A) developed a fibular stress fracture (B) after initiation of a walking program. (Courtesy of James W. Brodsky, MD.)

and the posterior body of the talus are also uncommon injuries. Scintigraphy has been shown to be nondiagnostic in these instances because 63% of recruits who had ­technetium-99m diphosphonate uptake at the os trigonum had no pain over that area.68 They are most typically associated with classical ballet because of the hyperplantar flexion of the tibiotalar joint when the female dancer is dancing on toe (en pointe) (Fig. 25E-8).73,74 An important variation of stress fracture is the posterior impingement of the os trigonum against the posterior tibia, which causes direct mechanical pain (Fig. 25E-8). This often requires excision of the os trigonum through a posterior or posterolateral approach, usually with excellent results.73-75 Stress injuries of the lateral process of the talus are unusual. Cases of acute fracture, although uncommon, are much more common than is stress injury. An example of an MRI of a stress injury of the lateral process is demonstrated in Figure 25E-9. The treatment of these fractures should be tailored according to the duration of symptoms, the displacement, and whether there are the radiographic characteristics of nonunion. Cases in which a sclerotic line is found on CT require surgical intervention with excision of the fragment unless it is a large intra-articular fragment, for which bone grafting and internal fixation is the treatment of choice. Although calcaneal pain is one of the most common if not the single most common affliction of the foot in both athletes and nonathletes, few of these cases are true stress fractures. Most heel pain in both groups is attributable to soft tissue conditions that are frequently manifested at the

interface between tendon or ligament and bone, but are not stress fractures, nor even true osseous lesions. These diagnoses typically include plantar fasciitis, insertional Achilles tendinitis, plantar heel pain syndrome, and nerve entrapments. The symptoms may correspond to the anatomic location of these structures, most commonly pain on the plantar surface of the heel or on the posterior aspect of the heel. Owing to their extreme rarity in relation to other common heel pain diagnoses, stress fractures of the calcaneus are frequently overlooked because the initial radiographs are nearly always negative, and the pain is frequently quite diffuse and nonspecific. Although imaging studies such as MRI or technetium bone scan may be required to make the diagnosis, repeat radiographs after an interval of several weeks eventually show the typical, albeit often subtle, radiographic appearance. Calcaneal stress fractures are typically incomplete and manifest as a vertical condensation or radiodensity within the cancellous bone of the calcaneal tubercle. This usually occurs in the dorsal two thirds, as can be seen on the lateral standing radiograph of the foot, and does not extend down to the plantar cortex (Fig. 25E-10). Rarely, the fracture might be posterior in the tubercle (Fig. 25E-11). Recently, a review of 44 cases of calcaneal stress fractures using MRI found a certain percentage of stress fractures to appear in the middle and anterior part of the calcaneus62 as well as a case report of a stress fracture of a calcaneonavicular coalition in a runner.48 Changes on the imaging studies correspond to the same anatomic distribution as that described for radiographs. Technetium bone scans that show increased uptake on the plantar surface of

�rthopaedic ����������� S �ports ������ � Medicine ������� 2020 DeLee & Drez’s� O

B

A

Figure 25E-6   Finding in a 50-year-old physician with ankle pain after inception of a new jogging regimen. A and B, Initial radiographs. C, After 7 weeks of immobilization, note cystic resorption and new peripheral bone formation. (Courtesy of James W. Brodsky, MD.)

C

the calcaneus do not signify stress fracture, but rather are typical of plantar fasciitis. The inflammatory process at the insertion of the plantar fascia involves the periosteum and Sharpey’s fibers, which produce the positive uptake on the scan. On close physical examination, the distinguishing characteristic of calcaneal stress fracture is that the pain is dorsal and anterior on the calcaneal tubercle, and the

A

patient will point to the medial and lateral sides of the heel anterior to the Achilles tendon as the center of the pain. Treatment consists of diminished activity with variable immobilization until symptoms subside, usually between 2 and 3 months. Stress fractures of the tarsal navicular are the most difficult stress fractures of the foot and ankle to treat because of the typically slow and recalcitrant healing. Presentation is

B

Figure 25E-7   Professional basketball player with bilateral recurrent stress fractures of the medial malleolus. A, Nonunion after open reduction with internal fixation. B, Healing after revision of fixation and bone grafting. This degenerated again to a recurrent stress fracture after resumption of play but rehealed with conservative therapy. (Courtesy of James W. Brodsky, MD.)

Foot and Ankle 2021

B

A

Figure 25E-8   Professional ballerina with posterior ankle pain when dancing en pointe. Comparison of lateral radiographs in standing (A) and “on-toe” (B) positions reveals not only compression of the os trigonum but also displacement of the os trigonum away from the talus as well as anterior subluxation of the tibiotalar joint. (Courtesy of James W. Brodsky, MD.)

one of vague, poorly localized pain that is gradual in onset. On careful questioning, the patient will point to the dorsum of the hindfoot, but the examiner can easily be misled into thinking that the pain emanates from the adjacent tibiotalar joint. These relatively uncommon fractures occur primarily in running athletes. The largest series to date is that undertaken by Torg and associates.34 They showed that all 21 of the fractures in their study were in the sagittal plane, in the central one third of the navicular. Strict enforcement of non–weight-bearing has a markedly positive effect on healing of the fracture.

A

The routine, three-view foot radiographic evaluation is frequently unrevealing because the navicular is underpenetrated and the fracture is in the sagittal plane. Cone-down views assist in diagnosis. Telltale radiographic signs include sclerosis in the subchondral bone adjacent to the talonavicular joint. Although bone scan or MRI adequately screens for the presence of a navicular stress fracture, the study of choice is CT. Sections in two planes are recommended. Images in the coronal plane should be parallel to the talonavicular joint and roughly parallel to the plantar surface of the foot. The typical appearance of the fracture on CT is a vertical fracture

B

Figure 25E-9   This athlete underwent magnetic resonance imaging for persistent lateral hindfoot pain. T1-weighted (A) and   T2-weighted (B) sequences demonstrate the marrow edema indicative of a nondisplaced stress fracture of the lateral process. (Courtesy of James W. Brodsky, MD.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2022 DeLee & Drez’s� O Figure 25E-10   Calcaneal stress fracture in a runner. The propagation of the fracture from the superior cortex of the tubercle is the most common pattern. (Courtesy of James W. Brodsky, MD.)

line (in the sagittal plane) that begins on the dorsum of the navicular, proceeds plantarward, and is almost always incomplete (Fig. 25E-12A to C). Although Torg and associates34 described these as occurring in the central third, the location is usually more lateral than medial, and often at the junction of the medial two thirds and the lateral third. ­Sclerosis often is noted on both sides of the fracture, depending on the age of fracture, which in the horizontal plane extends to and through the proximal subchondral bone of the navicular at the talonavicular joint. The location of the fracture line has been demonstrated to correspond to a zone of avascularity in the tarsal navicular. Microangiographic studies showed the central one third of the navicular to be relatively avascular.34 The intraosseous blood supply has been shown to enter from medial and lateral poles of the bone, and to diminish in the zone where the fracture occurs. Tarsal navicular stress fractures frequently require surgical intervention because diagnosis is commonly delayed. If treatment is initiated early, an attempt at non–weightbearing immobilization may be tried. Surgical treatment consists of bone grafting and internal fixation; the surgical procedure is technically challenging. The incision should be on the dorsolateral aspect of the hindfoot, corresponding to the central one third of the navicular bone. The fracture site is exposed and carefully débrided. The major pitfall is to lose the three-dimensional orientation of the navicular bone, so one must remember that the medial and lateral parts of the articular surface (and cartilage) are more proximal, and the articular surface is farther distal in the midportion of the bone. Thus, it is easy to penetrate the central portion of the proximal articular surface, damaging the talonavicular joint. Further risks to the joint are posed by the placement of the internal fixation screws, again attributable to the sharply curved proximal articular surface of the navicular. The screws are best and most easily placed from lateral to medial because the medial fragment is larger, thus affording a greater area for placement of the screw threads. For damage to the talonavicular joint to be avoided, thoughtful screw placement requires a more distal starting point on the lateral aspect of the navicular (see Fig.

25E-11D). It is advisable to place the screws under fluoroscopic control; the process may be facilitated by the use of 3.5- or 4.0-mm cannulated screws, placed over guidewires. It is necessary to warn the patient about the typically prolonged healing time of 3 to 5 months for this fracture. Postoperatively, non–weight-bearing immobilization in a cast is recommended for at least 8 weeks, followed by a walking cast for at least 1 month, and a removable walking boot thereafter, with modification based on the rate of fracture healing, as judged by radiographs, and if necessary, repeat computed tomographic scan. Disruption of the synchondrosis between the accessory navicular and the medial tubercle of the navicular is uncommon but must be mentioned among the stress

Figure 25E-11   Persistent heel pain in this 29-year-old woman was initially attributed to plantar fasciitis at another institution, where her radiographs were negative. After an interval of 4 weeks, the stress fracture of the calcaneal tubercle became apparent. (Courtesy of James W. Brodsky, MD.)

Foot and Ankle 2023

A

B

C

D

Figure 25E-12   A, Defensive lineman with vague, persistent hindfoot pain. Subtle linear lucency in the navicular is seen on the radiograph. B and C, Computed tomography confirmed the diagnosis. D, The patient was treated with internal fixation and bone grafting. (Courtesy of James W. Brodsky, MD.)

i­njuries of the hindfoot. Although this may occur as a result of chronic loading in a running athlete, the examiner’s index of suspicion should be raised, especially in jumping athletes such as hurdlers or long-jumpers. Because this is the site of attachment of the tibialis posterior tendon, this injury also occurs as a result of acute forced eversion. The injury also may be subacute or superimposed on chronic trauma. Physical examination reveals tenderness at the medial hindfoot, overlying the most distal portion of the tibialis posterior tendon, right at its insertion. This tenderness is exacerbated by passive eversion and active inversion. There may be increased medial prominence and swelling. Radiographic diagnosis may be delayed because of difficulty in visualizing the accessory navicular on routine foot radiographs. A reverse oblique (medial oblique) plain radiograph is the view of choice. Even excellent radiographs can be inadequate in assessing whether the attachment of the accessory navicular has been disrupted. The plane between the accessory navicular and the tuberosity of the navicular is oblique in all three planes, making it difficult to assess if it has retracted proximally. Technetium-99m bone scan is the screening test of choice, if making this determination is not possible on radiograph and physical examination. Highly collimated views with appropriately adjusted gain are necessary for localization of the increased scintigraphic uptake to the synchondrosis (Fig. 25E-13). Surgery consists of advancement of the tibialis posterior tendon to prevent incompetence and an acquired pes planus. This may be accomplished by excising the accessory navicular, then reattaching the tendon directly to the

medial navicular tubercle, or by arthrodesis of the accessory navicular to the medial tubercle. The latter is possible only if the accessory bone is large (type II). The result is enhanced if the medial tubercle of the navicular is reduced in size, which serves to tighten the tendon by advancing the insertion distally and laterally, while reducing the external prominence on the medial border of the hindfoot. Either way it is essential to advance the tibialis posterior tendon to ensure it is reconstructed under sufficient tension, that it will be subsequently functional. It is wise to err on the side of greater than lesser tension in the reconstructed tendon. Stress fractures of the cuneiforms are reported but rare, and they require conservative management as outlined previously. Stress fractures of the middle and lateral cuneiforms have been reported more frequently than those of the medial cuneiform, presumably because of the propagation of forces from the bases of the second and third metatarsals.9,32 Stress fractures of the cuboid are somewhat less rare but nonetheless very uncommon. One should look for predisposing mechanical factors that cause overload of the lateral column of the foot, such as subtalar ankylosis or a varus position of the hindfoot or ankle, or as a complication after plantar fascia disruption.76 As with cuneiform stress fractures, the ­cancellous nature of the cuboid makes healing very likely. Once healing has resulted from conservative measures, consideration should be given to the use of a soft-soled but supportive shoe, in addition to a custom-molded, cushioning, accommodative insole to compensate for the stiffness or varus.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2024 DeLee & Drez’s� O

A

C

A rare stress fracture is a fracture of the os peroneum that is embedded in the peroneus longus tendon as it passes under the cuboid. Only a few cases of stress fracture of the os peroneum have been reported in the literature.77,78 Some have been associated with a peroneus longus tear, whereas others were shown to be associated with an intact tendon.78

Metatarsal Fractures Stress fractures of the metatarsals are the most common stress fractures of the lower extremity.15 These occur commonly in military recruits, dancers, and other athletes. The most common metatarsal stress fractures occur in the shaft (diaphysis) or neck of the bone. Although insufficiency fractures are known to occur in the metatarsals, most metatarsal stress fractures seen in clinical practice are of the fatigue-type. The most common history involves a precipitous increase in the duration and intensity of weight-bearing sports or exercise activity, such as a sudden doubling of running distance, or the institution of a new and vigorous exercise program in previously sedentary individuals. As with other stress fractures, the initial radiographs are negative more often than not. The diagnosis can usually be made on the basis of the history described earlier in the paragraph, a high index of suspicion, tenderness to palpation directly over the metatarsal (more so than the intermetatarsal web space), and one specific physical finding, that is, the presence

B

Figure 25E-13   Images of a 15-year-old female basketball and softball player with persistent medial pain over the hindfoot. Technetium-99m bone scans (A and B) and comparison view of the contralateral foot (C) confirmed disruption of the synchondrosis between the accessory navicular and the navicular. (Courtesy of James W. Brodsky, MD.)

of a well-­circumscribed swelling only over the dorsum of the forefoot that does not extend to the medial or lateral border of the foot (Fig. 25E-14). Pain is almost always described as dorsal in location. Treatment, besides attention to predisposing conditions, is symptomatic and usually consists of a rigid-bottom surgical shoe or a stiff-soled shoe and cessation of inciting activities (e.g., dancing, running, marching). Healing time is variable, but on average, resumption of the use of normal shoewear begins between 6 and 8 weeks after onset. Advancement to unrestricted shoewear may require several months, depending on the severity of swelling and the rate at which it recedes. The primary complication of metatarsal stress fractures is displacement with subsequent malunion, and its primary result is transfer metatarsalgia. In occasional cases, surgery might be warranted to reconstruct such a forefoot situation, although shoe modification with a cushioning orthosis would usually be tried first. The reconstruction could include plantar flexion osteotomy and bone grafting of the malunion, with or without elevating osteotomy of the symptomatic adjacent metatarsal. Fractures of the proximal fifth metatarsal are of two types, and improper use of the eponyms by some authors has led to unnecessary confusion of these distinct entities. The most common fracture of the foot and ankle is fracture of the base of the fifth metatarsal through the cancellous bone of the proximal metaphysis (properly known as a dancer’s fracture) (Fig. 25E-15). These are acute fractures

Foot and Ankle 2025

A

C

(not stress fractures) that are sustained by a sudden inversion mechanism, which is similar to the twisting injury that produces damage to the lateral collateral ligaments of the ankle. Most of these injuries usually do not extend into the metatarsal-cuboid joint. Most are mildly or minimally displaced and need only conservative treatment, which consists of immobilization. Treatment of choice is usually a removable below-knee walking boot for 6 to 8 weeks. It is important to point out to the patient that there is always a lag between clinical and radiographic healing. Persistent lucencies at the fracture site are common, presumably owing to the displacement of the fracture and the time needed to fill in. Nonunion cannot be diagnosed before 6 months after the initial fracture. Complete evidence of

B

Figure 25E-14   Classic metatarsal stress fracture. A, The initial film shows a minor crack in the diaphyseal cortex.   B, Subsequent radiograph shows the exuberant callus formation. C, Note the foot swelling, especially on the dorsum. (Courtesy of James W. Brodsky, MD.)

radiographic healing is usually not present until after 3 to 4 months. Surgical intervention is required if there is severe displacement, especially if the proximal fragment is highly rotated, but this is uncommon. Most cases are treated nonoperatively. The second type of fracture is properly referred to by the eponym of Jones’ fracture, in reference to its original report by Sir Robert Jones,79 in which he described the injury as it occurred in his own foot. It is a fracture of the metaphyseal-diaphyseal junction. Although many of these fractures have an acute component as in the original fracture, many, if not most, are superimposed on chronic stress injury to the bone. This injury is an incomplete stress

�rthopaedic ����������� S �ports ������ � Medicine ������� 2026 DeLee & Drez’s� O

A

B

Figure 25E-15   A “dancer’s fracture.” This is the most common foot fracture, that is, avulsion fracture of the base of the fifth metatarsal associated with acute inversion injury. (Courtesy of James W. Brodsky, MD.)

fracture, which explains the intense bony sclerosis, hypertrophic bone formation, and incomplete nature of the fracture that is seen on the radiograph even immediately after the acute twisting injury. The fracture is based plantar and lateral, with the dorsomedial cortex often intact (Fig. 25E-16). Nonoperative treatment is appropriate for injuries with an acute component in patients who do not have an established pseudarthrosis or chronic bony sclerosis at the margins of the fracture on radiographic evaluation. Nonoperative treatment consists of immobilization in a cast or below-knee boot with strict non–weight-bearing for at least 6 to 8 weeks. A significant number of these fractures heal without surgery, but radiographic consolidation takes at least 3 months on average. Fractures with less chronic stress change (as judged by less sclerosis and less hypertrophic spurring) are more likely to heal nonoperatively; surgical treatment is more often necessary in fractures with evidence of chronic stress injury. Surgical treatment is indicated in cases in which nonoperative treatment has failed, as determined by persistent, unresolving pain or established pseudarthrosis, or in which there is a need to return a high-performance athlete to his or her Figure 25E-16   Acute Jones’ fracture at the diaphyseal-metaphyseal junction of the fifth metatarsal. (Courtesy of James W. Brodsky, MD.)

sport as quickly as possible. In many athletes, it is almost routine to treat these fractures with early surgery because it provides speedier and more reliable healing, especially in the middle of an athletic season. The surgery typically consists of internal fixation with an intramedullary screw (usually between 4.5 and 6.5 mm diameter) placed from the proximal tubercle in a distal direction past the fracture site (Fig. 25E-17). Although several authors have recommended the use of cannulated 4.5-mm screws80,81 in some cases, cannulated screws will bend and break either in cases of nonunion or a recurrent stress fracture. This may be augmented by débridement and bone grafting of the fracture site, especially in cases with evidence of a hypertrophic nonunion, or chronic stress injury. Six weeks of cast immobilization and non–weight-bearing is recommended after surgery, followed by appropriate progression, based on radiographic findings, to weight-bearing in a cast followed by a below-the-knee boot. In high-performance athletes, it is recommended that the screw be retained after healing, as long as the player continues to compete, because of the risk for recurrent fracture. Once the fracture is healed, strong consideration should be given to the use of a supportive athletic shoe and a custom-molded cushioning orthosis.

Foot and Ankle 2027

A

B

Figure 25E-17   Images of an 18-year-old collegiate soccer player with recurrent pain and nonunion of Jones’ fracture after previous surgery. A, Note the hypertrophic beaking and sclerosis at the fracture site. B, Healing was achieved 10 weeks after revision intramedullary fixation and bone grafting. (Courtesy of James W. Brodsky, MD.)

A good example is a dual-density device made of heatmolded polyethylene foam (e.g., medium-grade Plastazote or Pelite) combined with non-deforming open-cell foam (e.g., Poron). To the plantar surface of the orthosis should be added a lateral wedge to counter the varus thrust of the hindfoot and to facilitate the transfer of weight to the medial side of the midfoot. Risk factors for Jones’ fracture include a high level of activity in a running or jumping athlete. This fracture is typically seen in soccer and basketball players, although its occurrence is not exclusive of other sports and activities. Athletes with hindfoot varus of any cause have a predilection for this fracture because of lateral column ­overloading. Basilar metatarsal stress fractures are less common among the general population but are the most common stress fracture seen in ballet dancers.19,30 The typical location is at the proximal metaphyseal-diaphyseal junction, with possible extension into the tarsometatarsal joint.

Figure 25E-18   A 66-year-old woman with large joint osteoarthritis developed this stress fracture of the proximal second metatarsal while participating in a walking program for knee rehabilitation. Note the chronic hypertrophy of the metatarsal. (Courtesy of James W. Brodsky, MD.)

Multiple theories exist as to the mechanism for a basilar metatarsal fracture; all these theories center around the en pointe position. Fractures of the bases of the metatarsals are less common than are those of the diaphysis and neck, but they are not rare.30 Most typically, this fracture occurs at the base of the second metatarsal (Fig. 25E-18). By the time it is diagnosed, radiographs demonstrate sclerosis without displacement. The patient gives a history of chronic aching in the midfoot region, which must be distinguished from tarsometatarsal arthritis. These fractures are chronically symptomatic to varying degrees, although they seldom heal with immobilization alone. Most require surgical intervention with bone grafting, with or without internal fixation. If the fracture is very close to the second tarsometatarsal joint, it may be technically impossible to achieve fixation to the narrow proximal fragment. In these cases, it may be necessary to do an arthrodesis of the adjacent joint in order to gain stabilization and fixation (Fig. 25E-19). An unusual case of a stress fracture of the base of the first metatarsal in a skeletally immature boy that involved the physis and the first tarsometatarsal joint (Salter-Harris III) was reported. This fracture healed with conservative measures,82 but stress fractures of the first metatarsus are rare. Stress injury to the hallux sesamoids is relatively common in active populations. The hallux sesamoids are the points of weight transfer along the medial column of the forefoot, and they are covered with little soft tissue padding. Injuries to the sesamoids vary from acute fracture and stress fracture to sesamoiditis. Sesamoid stress fractures can occur in either the medial or the lateral sesamoid. The medial sesamoid is usually slightly larger than the lateral, and fracture of the medial sesamoid is somewhat more common.35 Radiographic details of the sesamoids are often difficult to visualize on standard radiographs because of the obscuring shadow of the first metatarsal head (Fig. 25E-20). Patients with sesamoid stress fractures are generally active, although most do not necessarily give a history of a recent change in activity level. They present with insidious

�rthopaedic ����������� S �ports ������ � Medicine ������� 2028 DeLee & Drez’s� O

A

B

Figure 25E-19   A, Image of a 27-year-old woman with persistent dull midfoot pain. The correct diagnosis of stress fracture of the base of the second metatarsal was made after 1 year of multiple consultations. B, Operative treatment required bone grafting as well as arthrodesis of the metatarsal-cuneiform joint. (Courtesy of James W. Brodsky, MD.)

onset of medial forefoot pain, primarily with weight-­bearing activities; this pain is exacerbated by running, jumping, and participating in toe-off activities. Physical examination reveals tenderness localized to the plantar first metatarsophalangeal joint (tenderness may localize to a specific sesamoid), which is worsened with passive dorsiflexion and plantar flexion of the hallux. Ecchymosis and swelling are unlikely. Radiographs may demonstrate a transverse fracture of the sesamoid, but differentiation from a bipartite sesamoid becomes problematic (Fig. 25E-21). If ­ physical examination does not absolutely localize the pain to a specific sesamoid, anteroposterior and sesamoid views with a metallic marker are helpful. In most patients, treatment can be initiated without additional studies. If an absolute diagnosis is necessary, bone scan (from the plantar surface of the foot) may be the modality of choice83 because even MRI sometimes fails to diagnose these stress fractures.84

Figure 25E-20   Stress fracture of the medial sesamoid. Note the irregular edges, the elongation compared with the lateral sesamoid, and the difference in density of the proximal and distal portions. (Courtesy of James W. Brodsky, MD.)

Treatment is largely conservative, with activity modification and the use of off-loading orthoses being the mainstay. Protected weight-bearing in a cast or boot is used in recalcitrant cases and in those with a definite acute fracture line on plain films. Surgical treatment is reserved for cases that are recalcitrant to nonoperative treatment and symptomatic cases with radiographically documented distraction between the fragments. Sesamoid excision and reconstruction of the flexor hallucis brevis tendon and intersesamoid ligament is the appropriate surgical treatment. This approach has led to good results in the authors’ hands, with little evidence of late hallux deformity. However, the soft tissue reconstruction is critical to obtaining a good result because the position of the remaining sesamoids must be maintained and late deviation of the great toe must be prevented. (The risk is that hallux valgus deformity may develop after medial sesamoid excision, and hallux varus after excision of the lateral sesamoid.) Other authors have suggested bone grafting of nondisplaced sesamoid nonunions, with satisfactory results. This approach has been reported in a small series of young, primarily collegiate, athletes.85 In Figure 25E-22, a medial sesamoid stress fracture in a teenage athlete is illustrated, which was treated with this method of bone grafting. Although partial excision of one of the poles of the fractured sesamoid has been suggested, this seldom is feasible unless one piece is much smaller than the other. Leaving half-sesamoids usually results in a painful condition because of the poor tracking of the fragment under the metatarsal condyles, but a recent study reported six cases of proximal partial sesamoidectomy, with five out of the six cases achieving complete pain relief.84 The authors recommend partial excision only when the fragment is small, composing 25% to 30% of bone at most. Stress fracture of the first proximal phalanx is unusual. Only about 25 cases have been reported in the literature. About 90% were associated with hallux valgus.55,86 These fractures occurred at the medial base of the proximal phalanx. Most healed with conservative treatment, whereas a few of them were treated with open reduction and bone grafting.55

Foot and Ankle 2029

A

B

C

Figure 25E-21   Sesamoid fracture equivalent. There is disruption of the synchondrosis of this bipartite sesamoid. Note the separation between the two poles. (Courtesy of James W. Brodsky, MD.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2030 DeLee & Drez’s� O

A

B

Figure 25E-22    A, Teenage athlete with a medial sesamoid stress fracture treated with local bone grafting. B, The result of a ­completely healed sesamoid. (Courtesy of James W. Brodsky, MD.)

C

r i t i c a l

P

o i n t s

l Repetitive microtrauma in a muscle-fatigued athlete is the cause of stress fracture. l Clinical presentation with radiographic imaging is still the mainstay of diagnosis. l MRI is the most common adjunct to diagnosis. l Most stress fractures heal with cast or boot immobilization. l Fractures that benefit from early surgical intervention, especially in a high-performance athlete, are fractures of the medial malleolus, navicular, and base of fifth metatarsal (Jones’ fracture). l Return to sport should occur only after the fracture has healed and patient is pain free, depending on the specific case.

S U G G E S T E D

R E A D I N G S

Anderson MW, Greenspan A: Stress fractures. Radiology 199:1-12, 1996. Boden BP, Osbahr DC: High risk stress fractures: Evaluation and treatment. J Am Acad Orthop Surg 8:344-353, 2000. Drez D, Young JC, Johnston RD, et al: Metatarsal stress fractures. Am J Sports Med 8:123-125, 1980. Eisele SA, Sammarco GJ: Fatigue fractures of the foot and ankle in the athlete. J Bone Joint Surg Am 75:290, 1993. Lee JK, Yao L: Stress fractures: MR imaging. Radiology 169:217-220, 1998. Milgrom C, Giladi M, Stein M, et al: Stress fracture in military recruits: A prospective study showing an unusually high incidence. J Bone Joint Surg Br 67:732, 1985. Orava S, Karpakka J, Teimela S, et al: Stress fracture of the medial malleolus. J Bone Joint Surg Am 77:362, 1995. Sormaala MJ, Niva MH, Mattila VM, et al: Stress injuries of the calcaneus detected with magnetic resonance imaging in military recruits. J Bone Joint Surg Am 88:2237-2242, 2006. Torg JS, Pavlov H, Cooley JH, et al: Stress fractures of the tarsal navicular. J Bone Joint Surg Am 64:700, 1982. Weinstein SB, Haddad SL, Myerson MS: Metatarsal stress fractures. Clin Sport Med 2:319, 1997.

R eferences Please see www.expertconsult.com

S e c t i o n

F

Heel Pain Keith L. Wapner and Selene G. Parekh

RETROCALCANEAL BURSITIS (HAGLUND’S DISEASE, ENLARGEMENT OF THE SUPERIOR TUBEROSITY OF THE OS CALCIS) Pain in the posterior-superior portion of the calcaneus may be multifactorial, caused by retrocalcaneal bursitis, enlargement of the superior bursal prominence of the calcaneus,

insertional Achilles tendinosis, or inflammation of an adventitious bursa between the Achilles tendon and the skin (Fig. 25F-1).1-12 Each of these entities may exist as an isolated condition or may be part of a symptom complex. Careful analysis of the patient’s subjective complaints and objective findings are required to arrive at the correct ­diagnosis. Retrocalcaneal bursitis may occur as an isolated entity but is more commonly associated with the prominent ­posterior-superior bursal portion of the calcaneus, or Haglund’s deformity. When Achilles tendinosis occurs

Foot and Ankle 2031

Achilles tendon Adventitial bursa Retrocalcaneal bursa Haglund’s deformity

Figure 25F-1   Illustration of Haglund’s deformity with a retrocalcaneal bursa between the Achilles tendon and the superior bursal prominence and an adventitious bursa between the Achilles tendon and the skin.

concomitantly with this condition, it is generally located in the area of the Achilles tendon just at or above the insertion of the Achilles at the posterior portion of the os calcis and not more proximally as occurs with deterioration in the area of decreased vascularity 2 to 6 cm above the posterior tuberosity of the os calcis. Calcification may occur within the Achilles tendon at this area and probably represents calcification in a degenerative area of the tendon.3 The adventitious bursa, which occurs between the Achilles tendon and the overlying skin, is usually caused by pressure of the counter of the shoe against the prominent area. It is more common in women and less common in athletes. Mann noted that Haglund called attention to the possible relationship between the shape of the calcaneus and the appearance of “pump bumps,” Achilles tendinitis and retrocalcaneal bursitis, and small spurs that are attached to the Achilles tendon.8 Variations in the shape of the superior tuberosity of the os calcis may also play a role in the development of this retrocalcaneal symptom complex (Fig. 25F-2).8

Pertinent Anatomy The Achilles tendon, covered by a thin paratenon, inserts into the middle of the posterior part of the posterior surface of the calcaneus,13 inserting on average between 10 and 13 mm from the superior aspect of the calcaneal tuberosity, extending 19 mm inferiorly, and extending further medially than laterally.14,15 A retrocalcaneal bursa located between the Achilles tendon and the superior tuberosity of the calcaneus is a constant finding.13,16,17 Dorsiflexion of the foot and ankle produces increased pressure in the retrocalcaneal bursa, whereas plantar flexion decreases the pressure in the retrocalcaneal bursa.10 Anatomically, the retrocalcaneal bursa has an anterior bursal wall composed of fibrocartilage laid over the calcaneus, whereas the posterior wall is indistinguishable from the thin epitenon of the Achilles tendon.16 It is a diskshaped structure lying over the posterior-superior aspect of the calcaneus, fitting like a cap over the calcaneus and having a concave aspect anteriorly (Fig. 25F-3).16 The superior tuberosity of the os calcis may be hyperconvex, normal, or hypoconvex.8 The radiographic anatomy of the os calcis has been described by Heneghan and Pavlov4,9 in terms of the following anatomic landmarks on

Figure 25F-2   Variations in the shape of the superior tuberosity of os calcis. Left, hyperconvexity; middle, normal; right, hypoconvexity. (Redrawn from DuVries HL: Miscellaneous afflictions of the foot. In Mann RA [ed]: Surgery of the Foot, 5th ed. St. Louis, CV Mosby, 1986, p 248.)

the lateral projection (Fig. 25F-4). The superior aspect of the talar articulation marks the most proximal portion of the posterior facet. The bursal projection is the area of the superior tuberosity of the os calcis. The tuberosity of the posterior surface marks the site of the Achilles insertion. The medial tubercle is the site of insertion of the central portion of the plantar aponeurosis.

Relevant Biomechanics The retrocalcaneal bursa is a constant, horseshoe-shaped bursa found between the Achilles tendon and the rounded superior bursal prominence of the os calcis. It maintains the relatively constant distance between the axis of the ankle joint and the insertion of the Achilles tendon.10 If the posterior prominence were not present, during dorsiflexion there would be shortening of the distance of the ankle joint axis and the insertion of the Achilles tendon. As this lever arm shortens, the ability of the gastrocnemius-soleus muscle to function is affected. This mechanism allows the tension of the gastrocnemius-soleus muscle group through the Achilles tendon to remain constant with dorsiflexion and plantar flexion.10 Canoso and colleagues showed by magnetic resonance imaging (MRI) the dynamic aspects of the retromalleolar fat pad.18 They said that the tongue-like extension of the fat pad may be viewed as a freely moving spacer with a sliding motion between cartilage and tendons and is facilitated by hyaluronic acid, presumably secreted by its own lining. Because the Achilles tendon inserts in the middle third rather than in the more proximal upper portion of the posterior calcaneal surface, the plantar flexion lever is increased. The retrocalcaneal bursa, by accepting the fat pad, allows the necessary separation of tendon and bone that occurs in plantar flexion without creating excessive tissue tension. Retrocalcaneal pain syndrome is commonly associated with the high-arched cavus foot and the varus heel.2,10 The combination of these factors tends to produce a foot that does not dorsiflex as readily as a normal foot. There is prominence of the heel, which is more susceptible to increased pressure from the tendons and the counter of the shoe. Ruch stated that retrocalcaneal bursitis generally occurs in the circumstances of compensated rearfoot varus, compensated forefoot valgus, and a plantar flexed first ray because of the abnormal motion of the subtalar joint and the frontal and sagittal plane relationships.10

�rthopaedic ����������� S �ports ������ � Medicine ������� 2032 DeLee & Drez’s� O

Fat pad

Retrocalcaneal bursa

Talus

8mm 22mm 4mm

Achilles tendon

Calcaneus

4mm

Figure 25F-3   Demonstration of the disk-shaped retrocalcaneal bursa. (From Frey C, Rosenberg Z, Shereff MJ: The retrocalcaneal bursa: Anatomy and bursography. Presented at the American Orthopaedic Foot and Ankle Society Specialty Day Meeting, Las Vegas, February, 1989.)

Clinical Evaluation The history is generally that of slow onset of dull, aching pain in the retrocalcaneal area aggravated by activity and certain shoewear. A common complaint is start-up pain after sitting or when arising out of bed in the morning. At times there may a history of acute onset of pain, sometimes associated with a traumatic incident. When this occurs, one must think of a tear or calcification of the Achilles tendon as perhaps the initiating factor. Physical examination reveals swelling in the area of the retrocalcaneal bursa between the Achilles tendon and the calcaneus.1 There is generally a prominence in the area of the superior portion of the heel. The swelling in the retrocalcaneal bursa will be found just anterior to the Achilles tendon. By palpating medially and laterally at the same time and with the aid of ballottement, one can sometimes feel fluid within the bursa (Fig. 25F-5). With careful and discrete palpation, one can generally differentiate between swelling in the Achilles tendon and swelling in the retrocalcaneal bursa. The swelling of the Achilles tendon ­associated with retrocalcaneal bursitis is usually at the level of the tendon

Figure 25F-4   Important radiographic landmarks of the os calcis: T, superior aspect of talar articulation; BP, bursal projection; P, posterior tuberosity indicating attachment of Achilles tendon; M, medial tuberosity attachment of plantar aponeurosis; A, anterior tubercle. (From Pavlov H, Heneghan MA, Hersh A, et al: The Haglund syndrome: Initial and differential diagnosis. Radiology 144[1]:83-88, 1982.)

at or just proximal to the insertion. Dorsiflexion of the foot usually increases the pain in the area. A great deal of swelling and inflammation on examination may indicate involvement of the retrocalcaneal bursa and involvement of the Achilles tendon. There may be redness and swelling between the Achilles tendon and the skin, usually due to an adventitious bursitis produced by pressure of the shoe counter against the Achilles tendon. There may be an area of periostitis, which is a discrete localized area of tenderness of the os calcis, usually on the lateral side of the posterior portion of the os calcis and produced by pressure of the shoe counter.

Diagnostic Studies Radiographic Studies A lateral view of the foot is taken with the patient standing. This allows biomechanical evaluation of the foot as well as evaluation of the specific points of the os calcis. The points of the os calcis are identified as the posterior margin of the posterior facet, the superior bursal projection, the tuberosity indicating the site of the Achilles tendon insertion, the medial tubercle, and the anterior tubercle.9 The shape and appearance of the superior bursal prominence are noted. Evaluation of the lateral radiograph may be performed

Figure 25F-5   Illustration demonstrating area of the swelling; retrocalcaneal bursitis with swelling is anterior to the Achilles tendon.

Foot and Ankle 2033

~75º ~50º Figure 25F-6   Measurement of the Fowler and Philip angle. The normal angle is shown on the left and abnormal on the right. Upper level of normal is considered to be 69 degrees. Drawing at right indicates an abnormal angle of 75 degrees.

using the method of Fowler and Philip, which measures the posterior calcaneal angle (Fig. 25F-6).19 Fowler and Philip consider the bursal projection prominent if the angle is greater than 75 degrees. Some authors have concluded that a combination of the Fowler angle and the angle of calcaneal inclination is more effective in correlating the radiographic appearance with symptoms than the Fowler and Philip angle alone,10,20 the combined angle being greater than 90 degrees in patients with symptomatic Haglund’s disease. Parallel pitch lines have been used by Heneghan and Pavlov4 to determine the prominence of the bursal projection (Fig. 25F-7). The base line is constructed by placing a line along the medial tuberosity and the anterior tubercle and a parallel line from the posterior lip of the talar articular facet. The bursal prominence is considered abnormal if it extends above this line. Pavlov and associates noted that the recessions seen on the lateral radiograph extend 2 mm inferior to the bursal projection, and with retrocalcaneal bursitis, there is loss of the sharp interspace with the tendoachilles.9 They also pointed

Figure 25F-7   Parallel pitch lines used to determine the prominence of the bursal projection. A line is drawn along the medial tuberosity and the anterior tuberosity. A parallel is constructed from the superior prominence of the posterior facet. If the bursal projection is above the superior line, the projection is considered abnormally large. (From Pavlov H, Heneghan MA, Hersh A, et al: The Haglund syndrome: Initial and differential diagnosis. Radiology 144[1]:83-88, 1982.)

out that the pre-Achilles fat pad outlines the anterior surface of the Achilles tendon, which normally measures 9 mm from anterior to posterior, 2 mm above the bursal projection. Frey and colleagues described the retrocalcaneal bursa in a study of 12 fresh cadavers, 15 patients with signs and symptoms of retrocalcaneal bursitis, and 8 normal patients.21 They found that in symptomatic patients, the amount of contrast material accepted averaged 0.92 mL, and the outline of the bursa was irregular in 100% of the patients in this group. The asymptomatic group accepted an average of 1.22 mL of contrast material, and 71% were noted to have a smooth bursal outline. The average area of the bursa on the lateral radiograph was 0.77 cm2 in the ­normal patient and 1.18 cm2 in the abnormal patient. Eighty-three percent of the patients noted significant (greater than 80%) improvement of their symptoms when 1% lidocaine was injected into the retrocalcaneal bursa. Burhenne and Connell used xeroradiography to assess soft tissue and calcaneal detail in patients suffering from painful swelling localized in the heel.22 They noted that neither a posterior calcaneal angle (of Philip and Fowler) of more than 75 degrees (see Fig. 25F-6) nor the parallel pitch line (see Fig. 25F-7) proved to be a reliable index. They evaluated 4 patients with heel pain and swelling out of 100 control patients and found that the radiographic triad of retrocalcaneal bursitis, superficial tendoachilles bursitis, and Achilles tendon thickening, in the presence of an intact posterior-superior calcaneal margin, were readily evaluated with xeroradiography. Canoso and coworkers reported finding bursal fluid in cadavers without rheumatic disease and aspirating the retrocalcaneal bursa in four patients, three with Reiter’s syndrome and one with pseudogout.23 They reported that the findings in bursal and synovial joint fluids were similar. In patients with retrocalcaneal bursitis, one should always be aware of the possibility that the bursitis may be a manifestation of systemic arthritis or gout. This should be addressed by the history, general physical examination, and laboratory studies to rule out these disorders. Gerster and Piccinin reported on the painful heel syndrome with plantar fasciitis or Achilles tendinitis in 33 of 150 patients suffering from a seronegative spondyloarthritis.24,25

�rthopaedic ����������� S �ports ������ � Medicine ������� 2034 DeLee & Drez’s� O

By contrast, Achilles tendinitis was not encountered in 220 patients with rheumatoid arthritis. Gerster and colleagues also reviewed 30 patients with severe heel pain, of whom 24 were diagnosed as having seronegative spondyloarthritides in which talalgia is frequently the first symptom of the disease.26 Talalgia is defined as heel pain located either posteriorly (along the Achilles tendon or its insertion on the calcaneus) or at the attachment of the superficial aponeurosis on the plantar surface of the calcaneus. Of these 24 patients, 5 had involvement of the Achilles tendon with peritendinitis, and 6 had plantar fasciitis plus Achilles peritendinitis. Gerster’s group, unlike others, thought that rheumatoid arthritis was not a common cause of severe talalgia. MRI has provided clearer insight into the anatomic abnormalities associated with posterior heel pain.27-30 The imaging allows visualization of the Achilles tendon and the bursa as well as demonstrating any bony abnormalities in the posterior-superior calcaneus. In patients refractory to nonoperative treatment, preoperative MRI defines which anatomic structures need to be addressed (Fig. 25F-8) The degree of tendinosis present in the Achilles tendon is easily visualized and distinguished from isolated bursitis.

A

Treatment Options Puddu and associates31 proposed three stages of inflammation occurring at the insertion of the Achilles. Stage 1, peritendinitis, involves inflammation of the paratenon only. Peritendinitis with tendinosis, stage 2, is characterized by macroscopic thickening, nodularity, and microscopic focal degeneration of the Achilles tendon in addition to inflammation of the paratenon. Stage 3 is characterized by degenerative lesions of the substance of the tendon itself without associated peritendinitis. Clain and Baxter,32 divided Achilles tendon pathology into insertional and noninsertional dysfunction. Insertional tendinosis occurs within and around the Achilles tendon at its insertion and may be associated with Haglund’s deformity or spur formation within the tendon itself. In the presence of associated tendinosis, they advocated transfer of either the flexor digitorum longus, as advocated by Mann and colleagues,33 or the flexor hallucis longus (FHL), as advocated by Wapner and coworkers.34,35 Schepsis and colleagues36 concurred that tendon transfer should be considered because it may enhance the blood supply and reinforce the Achilles tendon.

B

C

D

E

Figure 25F-8   A, Axial magnetic resonance imaging (MRI) of normal Achilles tendon showing normal shape of the Achilles. B, Sagittal MRI of normal Achilles tendon showing normal shape of the Achilles. C, Sagittal MRI showing increased signal in the insertion of the Achilles tendon consistent with tendinosis. There is increased fluid demonstrating an inflamed bursa surrounding Haglund’s deformity. D and E, Axial and sagittal MRI demonstrating chronic tendinosis of the Achilles with marked fusiform swelling of the tendon.

Foot and Ankle 2035

Nonoperative Therapy Initial goals in the treatment of retrocalcaneal bursitis are to control pain and attempt to allow the patient to return to normal function and activity. Rest, particularly soon after the onset of symptoms, can be helpful. The duration of rest may be prolonged, depending on increasing symptom durations.37,38 Cross-training with low-impact exercises, such as an elliptical machine, swimming,16,37 or biking,39 may prevent deconditioning in the athlete. In the athlete who is unwilling to cross-train, modified regimens may be suggested. In patients with exquisite tenderness, immobilization in a CAM boot or short leg walking cast may be helpful.16,40 In the athlete, immobilization should be used cautiously because patients can have resultant tendon and muscle atrophy, degeneration, and decreased blood ­supply.39,41 Nonsteroidal anti-inflammatory drugs (NSAIDs), orally or locally delivered as a patch, may decrease local inflammation.41-43 Cryotherapy and ice can decrease pain, swelling, and inflammation as well.41-43 Corticosteroids, orally, injected locally, or applied topically, have been used.44 Injections must be used cautiously because they can lead to a higher incidence of tendon ruptures and tendinopathy.45,46 Animal model studies with intratendinous injections have been shown to result in localized tendon necrosis and decreased mechanical strength. If used, the steroid injection must be placed anterior to the tendon, in the area of the retrocalcaneal bursa.41,42 Physical therapy is a useful modality in the treatment of Achilles tendinitis. Eccentric loading of the tendon has been shown to have beneficial effects on the microcirculation of the tendon.47,48 Stretching of the contractures of the gastrocnemius and soleus muscles may limit Achilles tendon strain. Traditionally, night splints or dynamic joint stretching devices have been thought to decrease the Achilles contracture.40 However, a recent study by de Vos and associates demonstrated that a night splint, in addition to eccentric exercises, is not beneficial in the treatment of chronic midportion Achilles tendinopathy.49 Orthoses may help these patients by providing a heel lift function, correcting hyperpronation, or minimizing leg-length discrepancies.41,42 Care must be exercised when correcting hindfoot pronation deformities because overcorrection can result in inflexibility of the hindfoot with decreased shock absorption.42 Heel cups can decrease the strain on an Achilles tendon and elevate a prominent superior calcaneal tuberosity away from the tendon. Highheels, clogs, open-backed shoes, gel braces,50 and horseshoe pads40 can also provide symptomatic relief for the patient. Brisement therapy is a technique whereby saline is forcibly injected between the paratenon and the tendon. The principle of this technique is to lyse adhesions between these two structures.40,42 Successful reports have been published in the literature using glycosaminoglycan polysulfate41 and deproteinized hemodialysate injections.51 In addition, sclerosing therapy in insertional tendinopathy has shown promising results in a pilot study.52 Finally, mixed results have been published on shock-wave therapy for the treatment of chronic Achilles tendinopathy. Rompe and coworkers, in a randomized, controlled trial of low-energy shock-wave therapy and eccentric loading, demonstrated inferior results in the low-energy shock-wave treated cohort.53 No difference was found by Costa and colleagues in

a double-blind randomized placebo controlled trial comparing shock wave to placebo.54 Furia published successful treatments with high-energy shock-wave therapy.55 These studies suggest that high-energy, rather than low-energy, shock-wave therapy may be beneficial for chronic Achilles tendinopathy.

Operative Therapy Fiamengo and colleagues reviewed the charts of patients in whom a diagnosis of Haglund’s disease, retrocalcaneal bursitis, or pump bumps had been made.3 They found 19 patients who met the criteria, and radiographs of 12 of these patients were available for review. They also reviewed 104 control cases in which calcaneal spurs and Achilles tendon calcification with a posterior calcaneal step were present. This step is a horizontal ledge in the middle of the posterior portion of the os calcis corresponding to the level at which the Achilles tendon inserts into the calcaneus. The Fowler and Philip angle was measured, and no difference was found between the two groups. However, the symptomatic heels had a significantly longer horizontal calcaneal length, measured by a horizontal line between the most anterior and most posterior portions of the os calcis. The incidence of Achilles tendon calcification and a posterior calcaneal step was higher in patients who had chronic posterior heel pain compared with the control population. Fiamengo and colleagues recommended that in cases of chronic posterior heel pain, resection of the posterior-superior aspect of the calcaneus, as well as excision of the degenerative and calcific soft tissue in and about the distal Achilles, should be performed.3 Fox and colleagues reported on 32 patients with Achilles tendon disease who were operated on between 1968 and 1973.56 They divided the patients into two groups, those with an acute rupture without antecedent complaints and those with a history of chronic pain, weakness, and functional loss. They did not describe distal retrocalcaneal bursitis in these patients. They stated that degenerative disease of the Achilles tendon should be recognized and treated not as a simple injury but as a pathologic lesion. Heneghan and Pavlov noted that in their experimental model, the osseous projection on the plantar surface of the calcaneus produced a more prominent bursal projection, and heel elevation decreased the pitch angle, allowing the prominent bursal projection and the foot to slip forward, thus displacing the posterior calcaneus away from the shoe counter (Fig. 25F-9).4

Figure 25F-9   Demonstration of calcaneal angle or pitch angle, which is an angle drawn between the horizontal baseline and the inferior portion of the os calcis.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2036 DeLee & Drez’s� O

Ippolito and Ricciardi-Pollini described three patients with invasive retrocalcaneal bursitis in whom there were a large bursa and invasion of the os calcis.57 Pathologic examination revealed lymphoplasmacellular infiltrates containing proportionately more plasma cells than lymphocytes. Removal of the bursa provided clinical relief, and no systemic rheumatic disease developed in later years. Keck and Kelly reported on 13 patients with 20 sympto­ matic heels that were treated surgically.6 In 17 heels, the superior bursa was excised, and in 3 heels, a dorsally based closing wedge osteotomy was performed. Good results were reported in 15 of the heels treated. The initial results were good in all but 2 patients whose pain recurred as a manifestation of generalized rheumatoid arthritis. Osteotomy was used to reduce the posterior prominence, and results were rated good in 2, fair in 1, and poor in 2 heels. The authors thought that too few osteotomies were performed in this series to evaluate this method. However, the osteotomy had the disadvantage of requiring a longer convalescence.6 Zadek reported in 1939 on a closing wedge osteotomy of the superior part of the os calcis in three patients for treatment of adventitious bursitis.58 These patients were relieved of their symptoms. Kennedy and Willis reported on the effects of local steroid injections in tendons.59 They found that the most significant effect was actual collagen necrosis and that a return to normal failing strength in the injected tendons occurred by 14 days after the injection. The conclusion was that local steroid placed directly in a normal tendon weakened it significantly for up to 14 days. Therefore, in a patient with posterior heel pain, any injection should be made into the retrocalcaneal area, not into the tendon. However, even then there is some contact with the Achilles tendon.59 Lagergren and Lindholm reported on the vascular supply of the Achilles tendon.60 They found that ruptures of the Achilles tendon are usually limited to the segment of the tendon that lies between 2 and 6 cm proximal to its insertion in the os calcis and that this was an area of decreased vascularity and nutrition. This is an important finding relative to the retrocalcaneal bursal syndrome because this classic type of Achilles tendinosis is proximal to the area usually associated with the retrocalcaneal bursal syndrome. This may suggest that insertional tendinosis is brought on by impingement on the tendon rather than decreased circulation. A soleus muscle anomaly associated with symptoms simulating retrocalcaneal bursitis has been reported.61 In this entity, a soft bulge due to a large mass of anomalous soleus muscle in Kager’s triangle was noted. The condition responded satisfactorily to excision. Pavlov and colleagues reported the use of the parallel pitch line measurement in 10 symptomatic feet and 78 control feet.9 They thought that the symptoms correlated statistically with a positive posterior pitch line but not with an abnormal posterior calcaneal angle. They concluded that radiographically, the syndrome is characterized by (1) retrocalcaneal bursitis (loss of the lucent retrocalcaneal recess between the Achilles tendon and the bursal projection), (2) Achilles tendinitis (an Achilles tendon measuring more than 9 mm located 2 cm above the bursal projection), (3) superficial tendoAchilles bursitis (a convexity of the soft tissues posterior to the Achilles tendon insertion), and (4) a cortically intact but prominent bursal projection with a positive parallel pitch line.

In a series of 65 patients with Haglund’s disease reported by Ruch,10 17 patients were operated on by resection of the posterior-superior portion of the os calcis, resecting the posterior-superior aspect both medially and laterally and removing sufficient bone to render the previous palpable prominence entirely absent. They were evaluated 6 months to 5 years postoperatively. Fifteen demonstrated good to excellent results with elimination of symptoms. Three of the patients required a second procedure to obtain the desired result. Vega and colleagues reported 20 cases of Haglund’s deformity.12 They noted that the combination of the Fowler and Philip angle and the calcaneal angle, when greater than 90 degrees, correlated with the manifestation of symptoms. This was similar to the findings of Ruch.10 Vega and associates reported conservative treatment with pads, shoes, braces, and injections and resorted to surgical treatment only if conservative treatment was not successful. The surgical incision recommended was the lateral para-Achilles tendon approach. Clancy stated that an articular-like surface lines the calcaneus at its superior surface where it comes into contact with the Achilles tendon.2 He found that a bursa may form from constant overuse, enlargement of the bony prominence, or external pressure. He recommended steroid injection behind but not through or into the tendon. In patients who did not obtain relief with conservative measures, ostectomy proved successful. Clancy noted that most of those who required surgery had significant cavus deformities. Schepsis and Leach11 reported that most athletes, particularly runners, who presented with acute or chronic posterior heel pain were successfully managed nonoperatively using a combination of (1) a decrease in or cessation of the usual weekly mileage, (2) temporary termination of interval training and workouts on hills, (3) change from a harder bank surface to a softer surface, (4) a 1⁄4- to 1⁄2-inch lift inside the shoe or added to the shoe, and (5) a program designed to stretch and strengthen the gastrocnemiussoleus complex. These measures were combined with oral anti-inflammatory medications and an occasional injection of corticosteroid into the retrocalcaneal bursa. Postural abnormalities were treated with orthotics. Schepsis and Leach studied retrospectively 45 cases of chronic posterior heel pain treated surgically in 37 patients.11 All but 2 of these patients were competitive long-distance runners who ran an average of 40 to 120 miles per week before the onset of symptoms. Their ages ranged from 19 to 56 years. The surgical approach used by Schepsis and Leach was a longitudinal incision 1 cm medial to the Achilles tendon that was continued transversely to form a J-shaped incision if necessary (Fig. 25F-10). The patient was placed in a cast for 2 to 3 weeks with weight-bearing permitted after 1 week. When there was pathology within the tendon requiring excision and repair, immobilization was continued for 1 to 2 weeks longer. Range of motion exercises were emphasized. A graduated program of swimming and stationary bicycling, combined with isometric, isotonic, and isokinetic strengthening of the calf muscles, was prescribed. Jogging was permitted after 8 to 12 weeks, rarely sooner. Full return to a competitive level of sports activity usually required 5 to 6 months. The patients were divided into three groups: those with Achilles tenosynovitis-tendinitis, those with retrocalcaneal

Foot and Ankle 2037

A

B

Figure 25F-10   Illustration of surgical incisions for retrocalcaneal bursitis. A, Medial approach with J extension as described by Schepsis and Leach11; B, medial and lateral approach incisions as described by Jones and James.5

bursitis, and those with a combination of both. In a group of 24 patients with Achilles tenosynovitis-tendinitis, there were 15 (63%) excellent results, 7 (29%) good results, 1 (4%) fair result, and 1 (4%) poor result. In the 14 patients with retrocalcaneal bursitis, there were 7 (50%) excellent, 3 (21%) good, and 4 (29%) fair results. In the group with a combination of both, there were 5 (71%) excellent and 2 (29%) good results. It was noted that 4 of the 6 unsatisfactory results occurred in the group with retrocalcaneal bursitis. Jones and James5 reported on 10 patients who underwent partial calcaneal exostosectomies for retrocalcaneal bursitis. They suggested that conservative measures should be attempted before considering surgical intervention. These included a decrease in the usual weekly mileage, elevation of the heel, instruction in Achilles tendon strengthening, removal of external pressure from the heel, use of oral anti-inflammatory medications, and evaluation and treatment of postural foot deformities. They also suggested immobilization of the leg in a short leg walking cast for a brief period, allowing the athlete to continue cardiovascular maintenance on an exercise bicycle. Steroid injection was used as a last resort before surgery. During an 8-year period, Jones and James operated on 10 patients with retrocalcaneal bursitis.5 Six patients were competitive long-distance runners, and 4 were avid recreational runners. Their symptoms consisted of pain and tenderness in the retrocalcaneal area that developed either immediately or after running several miles. The patients ranged in age from 21 to 42 years. Surgery was performed through a longitudinal incision on both sides of the Achilles

Authors’ Preferred Method

of

tendon (see Fig. 25F-10) and included exostosectomy and excision of the bursa. Jones and James emphasized that the ridge of bone at the insertion site must be carefully removed with a small curet and rongeur so that no prominence of bone is left beneath the Achilles tendon posteriorly. A short leg walking cast was used for 8 weeks, with partial weightbearing allowed for the first 2 weeks and then full weightbearing. After casting, a 1-inch heel elevation was used until the foot assumed the neutral position easily. General muscle conditioning was carried out until Cybex testing revealed symmetrical muscle strength. All the patients went back to their desired level of activity within 6 months. Endoscopic techniques for the débridement of the retrocalcaneal space and the posterior-superior tuberosity of the calcaneus have been reported in the literature. Ortmann and McBryde reported on 28 patients and 30 heels treated with endoscopic surgery; 86.67% of patients reported excellent results, with one major complication of an Achilles tendon rupture.62 Although the medial column of the Achilles tendon, the sural nerve, and the plantaris tendons are at risk during this technique,63 the purported benefits of lower morbidity and recovery time make this a desirable option for some surgeons. Sullivan,20 addressing the problem of recurrent pain in the pediatric athlete, stated that heel pain may be due to osteochondrosis of the apophysis of the calcaneus (Sever’s disease) or to Achilles tendinitis, which is characterized by pain on palpation of the tendon just above its insertion. In severe cases, there may crepitation of the tendon. He recommended treatment consisting of rest and aspirin or other mild anti-inflammatory agents. In differentiating these two entities, it may be helpful to note that the pain of osteochondrosis occurs on the inferior portion of the os calcis and the pain associated with Achilles tendinitis is felt proximal to the insertion of the Achilles tendon on the os calcis. In summary, retrocalcaneal bursitis is a condition characterized by inflammation of the retrocalcaneal bursa, the Achilles tendon just above its insertion, and at times the tissue between the Achilles tendon and the skin. It is generally managed by conservative measures consisting of anti-inflammatory medication, decreased activity, padding to prevent pressure on the affected area, orthoses or heel lifts, and strengthening and stretching exercises. If it does not respond to these modalities, surgical intervention may be considered. Surgery generally consists of excision of the exostosis and the retrocalcaneal bursa and at times the adventitious bursa, if it is present, and correction of the Achilles tendon pathology with tendon transfer if necessary. Although most series do report good results after surgery, in the athlete, this condition may present a serious threat to continued full activity even after surgical ­intervention.

Treatment

The patient is first evaluated to ascertain the exact reason for the pathology. Adventitious bursitis is usually seen in women and does not appear to be a prominent problem in athletes. It is generally treated conservatively by softening the heel counter, using a small U-shaped pad (Fig. 25F-11) to relieve the pressure of the shoe or counter against the inflamed area, and anti-­inflammatory

medications. If the pain is refractory to these modalities, an injection of steroid directly into the inflamed area of the bursa, with care to avoid the tendon, may be tried once. Because of the risk for Achilles rupture after injections, the foot is immobilized in a removable cast walker for 2 weeks. Surgical intervention performed solely for adventitious bursitis is unusual. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 2038 DeLee & Drez’s� O

Authors’ Preferred Method

of

Treatment—cont’d

Figure 25F-11   Demonstration of use of U-shaped pad to remove pressure on the bony prominence of the heel.

Nonoperative management of insertional Achilles tendinitis is determined by the degree of inflammation and tendinosis present. When mild, use of a U-shaped heel pad, home stretching of the gastrocnemius-soleus muscle, avoidance of activities such as running, cross-training with bike riding, and use of a night splint to keep the foot in neutral is generally successful. The addition of NSAIDs can be considered. In moderate to severe cases of tendinosis, a period of immobilization with a molded ankle-foot orthosis can be used.

A

This allows decreased load across the tendon but does not completely immobilize the tendon. The brace should have a relief molded into it to avoid direct contact on the ­posteriorsuperior aspect of the calcaneus. The patient is allowed to continue to ambulate with full weight-bearing and may do activities such as stationary bike riding with the brace. The brace is continued until the swelling and tenderness are resolved. Aggressive stretching of the gastrocnemius-soleus complex and strengthening are then started, and the patient is weaned out of the brace. Shoe modifications with a U-shaped heel pad and soft heel counter are used. If nonoperative management fails, an MRI is obtained before surgery to assess the degree of tendinosis within the insertion of the Achilles tendon. This allows for preoperative discussion with the patient regarding the need for possible tendon transfer to reinforce the Achilles tendon. Surgical procedures are usually performed for retrocalcaneal bursitis associated with the superior bony prominence. The retrocalcaneal bursa and the superior bursal prominence are excised. The adventitious bursa is excised if it is prominent. If the adventitious bursa is excised, the surgeon must take care to excise it carefully and meticulously to prevent damage to or an adverse effect on the blood supply or the skin overlying this area because a skin slough would be a serious complication. A medial, lateral, or combination of incisions 1.5 cm anterior to the Achilles tendon is made as determined by the location and width of the bony prominence (see Fig. 25F-10). The incision is carried directly to the level of the paratenon, and dissection is performed at this level to maintain fullthickness skin flaps to avoid skin loss. Attention should be paid to the calcaneal branch of the sural nerve on the lateral side and the medial calcaneal nerve on the medial side. The Achilles tendon is inspected to confirm the degree of tendinosis. The retrocalcaneal bursa is excised. An exostosectomy is performed, removing the bone from the area of insertion of the Achilles tendon to the superior portion of the posterior facet of the os calcis (Fig. 25F-12). Adequate bone is removed, and the edges are smoothed with a rasp.

B

Figure 25F-12   A, Haglund’s deformity with prominence of posterior superior portion of os calcis. B, Appearance of os calcis after surgical resection of posterior superior prominence for symptomatic Haglund’s deformity.

Foot and Ankle 2039

Authors’ Preferred Method

of

Treatment—cont’d

B

A

D

C

FHL

E

F

Figure 25F-13   A, Diagram of incisions used for flexor hallucis longus (FHL) reconstruction of Achilles tendon. B, Clinical demonstration of incisions. C, Abductor and flexor hallucis brevis are reflected plantar and the FHL tendon identified. D, The FHL has been tagged and divided at the level of the proximal third of the first metatarsal. E, The flexor digitorum longus (FDL) is identified and tagged. F, Diagram of FHL and FDL anastomosis. Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 2040 DeLee & Drez’s� O

Authors’ Preferred Method

of

Treatment—cont’d

G

H

I

J

K

Figure 25F-13, cont’d   G, Posterior incision is made, and the posterior fascia of the leg is opened to allow transfer of the FHL into the wound. H, Two drill holes are made, one superior and the other medial, to intersect in the posterior body of the calcaneus to create a tunnel for tendon transfer. I, The FHL tendon is passed through the superior hole and out the medial side of the tunnel. J, The FHL is woven through the Achilles tendon using a tendon weaver. K, Diagram of completed weaving of FHL through the Achilles tendon demonstrating the orientation of the tunnel through the posterior calcaneus.

Foot and Ankle 2041

Authors’ Preferred Method

of

Treatment—cont’d

If significant tendinosis is present, the patient is positioned supine. A medial incision is used to expose the Achilles tendon. Dissection is carried down to the level of the paratenon, and the paratenon is opened and the tendon débrided. If significant calcification is present within the tendon, it should be removed. At times, this necessitates partial release of the insertional fibers of the Achilles tendon. Once the tendon is adequately débrided, the decision must be made about whether to add a tendon transfer. If more than 30% of the tendon is involved, I prefer to add a transfer of the FHL (Fig. 25F-13). Attention is first directed to the medial border of the foot, where the FHL tendon is harvested. A longitudinal incision is made along the medial border of the midfoot, just above the level of the abductor muscle from the navicular to the head of the first metatarsal. The skin and subcutaneous tissues are sharply divided down to the level of the abductor hallucis fascia. The abductor is then reflected plantarward, and a small Weitlaner retractor is placed in the wound. The flexor hallucis brevis is then reflected plantarward, exposing the deep midfoot anatomy. In some instances it is necessary to release the origin of the short flexors to assist visualization. The FHL and flexor digitorum longus tendons are identified within the midfoot. They are generally covered by a layer of fatty tissue. Identification of the tendons is assisted by placing a finger over the lateral wall of the short flexor and manually plantar flexing and dorsiflexing the first toe proximal interphalangel (PIP) joint. The motion of the tendon can be felt, and dissection can be carried down to identify the tendons of the FHL medially and the flexor digitorum longus laterally. The FHL is divided as far distally as possible, but one must allow an adequate distal stump to be transferred to the flexor digitorum longus. The proximal portion is tagged with a suture. The distal limb of the FHL is then sewn into the flexor digitorum longus with all five toes in a neutral posture, providing flexion to all five toes through the flexor digitorum longus. Attention is again turned to the posterior medial incision. The fascia overlying the posterior compartment of the leg is then incised longitudinally directly over the muscle belly of the FHL. The tendon is then retracted from the midfoot into the posterior incision.

Criteria for Return to Sports Participation The time needed for return to sports participation will depend on the severity and perniciousness of the condition, the relationship of the pain component to the patient’s sporting activity, and the extent of involvement of the Achilles tendon. Should the situation be severe enough to force the patient to stop the desired athletic activity, the following plan is used. If retrocalcaneal bursitis is present with a normal Achilles tendon, conservative therapy is tried until the patient has been asymptomatic for 4 to 6 weeks. The patient may then return to sports participation starting with limited activity and working up to full activity within 4 to 12 weeks, assuming that the patient has

A transverse drill hole is placed just distal to the insertion of the Achilles tendon halfway through the bone from medial to lateral. A second, vertical, drill hole is made just anterior to the level of resection of the posterior-superior prominence to meet the first hole. A large towel clip is used to augment the tunnel created. A suture passer is placed through the tunnel from distal to proximal. The suture is then pulled through the tunnel, thus drawing the FHL tendon through the drill hole. If the Achilles insertion has been detached, suture anchors can be used to reattach the tendon before securing the FHL transfer. The FHL is then woven from distal to proximal through the Achilles tendon using a tendon weaver. The tendon weaver is passed through the Achilles, creating a “tunnel” in the tendon. The tag suture on the flexor hallucis is then grasped and pulled back through the tunnel, bringing the flexor tendon through the Achilles. This ­process is repeated to use the full length of tendon harvested. The tendon is secured with multiple sutures of No. 1 Cottony Dacron. After completion of the reconstruction, the paratenon is repaired. The subcutaneous tissue and skin are closed. Compressive dressings and plaster splints are applied to maintain 15 degrees of ankle plantar flexion. The patient is placed in a short leg non–weight-bearing cast at 15 degrees of equinus for 4 weeks. When the patient returns at 4 weeks, the dressing is removed, and the forefoot is placed on a foot rest with the patient seated on an examining table and with the hip flexed and allowed to stay in this position until the foot reaches neutral. The foot is then placed into a short leg walking cast or removable cast walker with the ankle at neutral for an additional 4 weeks, and weight-bearing is begun. A rehabilitation program for strengthening and range of motion is begun 8 weeks before surgery. The patient is maintained in a removable cast walker for community ambulation until 10 degrees of dorsiflexion is obtained and grade 4/5 strength is demonstrated. In-home ambulation is allowed with a 7⁄16-inch heel lift during this time. The patient is then advanced to regular shoewear and continued on a home strengthening program with Thera-Band. Athletic activity is restricted for 6 months after surgery.

recovered full strength and mobility without pain. If retrocalcaneal bursitis is associated with degeneration of the Achilles tendon, nonoperative treatment should be used until the patient is ­asymptomatic. A gradual increase in activity is then allowed over a 6- to 12-week period. If surgery has been performed on the tendon in combination with excision of the retrocalcaneal bursa and exostosectomy, immobilization is continued for 8 weeks, but active range of motion is begun at 3 weeks. Strengthening and stretching exercises are started at 6 to 9 weeks. Increased activity according to tolerance can be started at 12 weeks, and return to strenuous activity is allowed at 4 to 6 months if local symptoms have resolved. When tendon transfer is used, the protocol is as described in the previous section.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2042 DeLee & Drez’s� O

PLANTAR FASCIITIS ASSOCIATED WITH PAIN IN MEDIAL TUBEROSITY (HEEL SPUR SUBCALCANEAL PAIN SYNDROME) Pain in the region of the medial tuberosity of the heel associated with increased pain with activity and sometimes related to a spur of the os calcis has been described for many years. Initially, this condition was thought to be associated with gonococcus and was described as “gonorrheal spurs.”64,65 Later it was thought that this entity was due to pull of the plantar fascia and musculature.66-71 More recently, Przylucki and Jones72 and Baxter and Thigpen64 have attributed this pain to entrapment of the nerve to the abductor digiti quinti arising from the lateral branch of the plantar nerve. Freeman73 and others66,74 have attributed this pain to irritation of the medial calcaneal nerve. Other authors have attributed the pain to herniation or to compression of fat nodules.75,76 Bordelon66,67 described a clinical syndrome characterized by pain beneath the heel that is aggravated by ambulation and is not associated with any trauma. He proposed that this condition be considered in light of the structures that are present in the area. Specific treatment should be directed toward the structures that are inflamed, the theory being that inflammation in one structure may produce inflammation in other structures.66,67 In reviewing the literature, it is apparent that many different theories exist about the etiology of subcalcaneal pain, and hence many different methods of treatment have been suggested for it.64-69,71-102 It has been said that although this condition is familiar to all orthopaedic surgeons, it is probably fully understood by none.67 Snook and Chrisman100 noted that there is conflicting literature on this subject on two salient points. The first is that there is no accepted explanation of the etiology of the condition. Second, there is no generally approved method of treatment. They thought that perhaps the basic cause lay in the subcalcaneal pad, which in some unknown manner lost its compressibility, either by local loss of fat with thinning or by rupture of the fibrous tissue septa. Ali103 stated that “the painful heel is due to a fibrotic response, similar to plantar fibromatosis and not to the spur of bone which is the end result of recurrent strain on the plantar fascia.” Tanz101 thought that “inferior heel pain is often due to irritation of a branch of the medial calcaneal nerve.” Similarly, Baxter and Thigpen64 and Przylucki and Jones72 advanced the thought that the heel pain is due to an entrapment neuropathy that involves the branch of the lateral plantar nerve to the abductor digiti quinti. It has been noted that this branch passes more proximally than is shown in most anatomic studies and is in the area of the heel spur. Mann71 wrote that “in the early stages, fibrositis of low chronicity, with or without pain, anterior to the calcaneal tuberosity represents the pathological change. Continuation of the process leads to osteophytic changes and bone deposits on the sulcus just anterior to the tuberosity.” Kopell and Thompson104 stated that calcaneodynia, or painful heel, “is usually ascribed to an inflammatory reaction based on mechanical stress at the common muscular and fascial origins on the anterior inferior surfaces of the calcaneus and in many cases, the inflamed structures are

the calcaneal nerves which innervate the region of the common origin.” Leach and co-workers69,70 also thought that the cause of plantar fasciitis appears to be repetitive trauma, which produces microtrauma in the plantar fascia near its attachments and leads to attempted repair and chronic inflammation. Katoh and colleagues105 reported on objective analysis of foot function during gait using vertical impulse distribution along the sole of the foot during the load-bearing period of gait. This was demonstrated to be reliable in distinguishing between patients with painful heel pads and those with plantar fasciitis. Although there has been much discussion about the relationship of the calcaneal heel spur to subcalcaneal pain (Fig. 25F-14), the relationship has not been definitely established. Tanz101 stated that a heel spur is located in the origin of the short toe flexors and not in the plantar fascia. He noted that 15% of normal asymptomatic adult feet have subcalcaneal plantar spurs, whereas about 50% of adult feet with plantar heel pain have spurs. He thought that heel spurs contributed to the plantar heel pain, although many patients with plantar heel pain did not have spurs. Snook and Chrisman100 agreed with this. Their report on 27 patients with subcalcaneal pain noted that 13 had a calcaneal spur and 11 did not. Mann71 stated that “over a long period of time, proliferative bony changes at the origin of the fascia may lead to the formation of a spur.” Shmokler and colleagues106 reviewed 1000 patients at random with radiographs of the foot. There was a 13.2% incidence of heel spurs. Only 39% of those with heel spurs (5.2% of the total sample) reported any history of subcalcaneal heel pain. Shmokler and coworkers believed that these statistics tended to support the premise that the presence of a heel spur did not mandate pain.

Figure 25F-14   Lateral view of a foot demonstrating a spur in the region of the medial tuberosity. The spur is in the region of the origin of the short flexors. Half of symptomatic patients have heel spurs.

Foot and Ankle 2043

Leach and colleagues69,70 stated that the spur is located in the short toe flexor origins as opposed to the plantar aponeurosis, casting doubt on the concept that the heel spur contributes to the pain in the plantar fascia. In evaluating 45 patients with 52 painful heels, Williams and associates107 found that 75% of those with painful heels had a heel spur compared with 63% of the opposite nonpainful heels. Comparing 63 heels in 59 age- and sex-matched controls, the incidence of heel spur was 7.9%. Warren and Jones108,109 attempted to predict which factors would be associated with plantar fasciitis but found that a set of predictable variables was not present. Heel spurs may be present or absent and may or may not be the primary pathologic entity in heel pain. However, they have to be considered in the context of the entire syndrome.

Pertinent Anatomy The plantar aponeurosis arises from the os calcis and is composed of three segments (Fig. 25F-15).110-113 The central segment is the largest and arises from the plantar aspect of the posteromedial calcaneal tuberosity. It inserts into the toes. The lateral portion arises from the lateral process of the tuberosity of the os calcis and inserts into the base of the fifth metatarsal. The medial portion is thin and covers the undersurface of the abductor hallucis. Clinically, when considering the plantar aponeurosis, one is generally referring to the central portion, which extends from the medial tuberosity of the os calcis to the

toes. It originates from the os calcis and passes to the proximal phalanges of the lesser toes through the longitudinal septa, to the big toe through the sesamoids, and into the skin of the ball of the foot through the vertical fibers.114 Hyperextension of the toes and the metatarsophalangeal joints tenses the plantar aponeurosis, raises the longitudinal arch of the foot, inverts the hindfoot, and externally rotates the leg. This mechanism is passive and depends entirely on bony and ligamentous instability. This mechanism, whereby the arch is raised and supported with dorsiflexion of the toe, providing more stability to the foot, has been termed the windlass mechanism by Hicks.115 The posterior tibial nerve is located on the medial side of the foot behind the medial malleolus and beneath the flexor retinaculum (Fig. 25F-16). The medial calcaneal nerve arises at the level of the medial malleolus or below and passes superficially to innervate the skin of the heel. It may consist of one or two branches. The important anatomic point is that this nerve passes in the subcutaneous tissue between the plantar fascia and the skin. The next nerve, which is the nerve to the abductor digiti quinti and which branches off the lateral plantar nerve, passes deeper, beneath the plantar ligament and underneath the spur, if present, to innervate the abductor digiti quinti (Fig. 25F-17). It is important to differentiate the medial calcaneal nerve from the nerve to the abductor digiti quinti.66-68 They can be readily differentiated because the medial calcaneal nerve passes in the subcutaneous tissue and is superficial to the plantar aponeurosis, whereas the nerve to the

5

8 9

8 4

11 7 2

6 3

9 8 2

10 5 1

4

1 7

6

3

Figure 25F-15   Plantar aponeurosis composed of three parts: (1) central component of plantar aponeurosis, (2) medial component of plantar aponeurosis, (3) lateral component of plantar aponeurosis. From the clinical standpoint, the central portion is considered to be in the plantar aponeurosis. (Redrawn from Rondhuis JJ, Hudson A: The first branch of the lateral plantar nerve and heel pain. Acta Morphol Neerland-Scand 24[4]:269-279, 1986.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2044 DeLee & Drez’s� O

1 2 3 1

4 2

5

3

6 7

Figure 25F-16   Relationship of structures commonly associated with heel pain: (1) long plantar ligament, (2) plantar fascia, (3) skin, (4) medial plantar nerve, (5) lateral plantar nerve, (6) nerve to abductor digiti quinti, and (7) medial calcaneal nerve. Note that the medial calcaneal nerve supplying sensation to the heel passes superficial to the plantar fascia. The nerve to the abductor digiti quinti passes deep to the plantar fascia and beneath the spur.

abductor digiti quinti is deeper and passes beneath the plantar fascia. Rondhuis and Hudson116 noted that entrapment of the nerve to the abductor digiti quinti occurs between the abductor hallucis and the medial margin of the medial head of the quadratus plantae muscle. They did not find a perforation of the fascia as described by Baxter and Thigpen,64 nor did they find any bursa in the origin of the plantar aponeurosis. They concluded that there are fibers that innervate the perichondrium and that sensory fibers are present and form free endings, producing pain sensation. They found motor branches to the flexor digitorum brevis and abductor digiti quinti muscles (Fig. 25F-18). The medial and lateral plantar nerves continue to the foot and pass through prospective foramina of the abductor muscles. When considering entrapment of the posterior tibial nerve, it is important to note that this nerve may be entrapped either beneath the flexor retinaculum at the level of the medial malleolus or at the point where the medial and lateral plantar nerves exit through separate foramina in the abductor muscles (Fig. 25F-19).

Relevant Biomechanics The foot must be evaluated to ascertain which specific type of foot and which biomechanical components are involved.68 A supple foot with a tendency toward a flatfoot deformity will place increased strain on the origin of the plantar fascia on the calcaneus because the windlass mechanism will be under increased strain in maintaining a stable arch during the propulsive phase of gait. Biomechanically, in an effort to avoid this strain, one could consider an orthotic device to correct the biomechanical deformity and to increase the support of the foot during the stance phase. Strapping with tape may also be used to hold the forefoot in adduction and the heel in varus to relieve the

4 5

Figure 25F-17   Sagittal section of os calcis showing relationship of the nerves to the abductor digiti quinti and the medial calcaneal nerve as described by Baxter and Thigpen64 and Przylucki and Jones.72 (1) Long plantar ligament, (2) plantar fascia, (3) skin, (4) nerve to abductor digiti quinti, (5) medial calcaneal nerve.

pressure on the origin of the plantar aponeurosis during propulsion. When a cavus foot is present, there may be excessive strain on the heel area because the foot lacks the ability to evert to absorb the shock and adapt itself to the ground. With a cavus foot, a soft cushioning material may be used to decrease the shock component and increase the area of contact. The goals of these orthotic devices are to reduce the stress on the medial tuberosity and the plantar fascia (Fig. 25F-20).

Clinical Evaluation History The history usually reveals a slow but gradual onset of pain along the inside of the heel.66-68 Occasionally, the pain may be associated with a twisting injury of the foot, producing an abrupt onset of pain.112 However, the clinical course is generally similar regardless of the onset. The location of the pain is generally described as along the medial side of the foot at the bottom of the heel. The pain is worse upon first arising in the morning and then decreases with increased activity. However, it may increase after prolonged activity. Periods of inactivity are generally followed by an increase in pain as activity is started again. Numbness of the foot is not present. When severe pain is present, the patient is unable to bear weight on the heel and will bear weight on the forepart of the foot.

Physical Examination Physical examination consists of evaluating the foot to determine what type of foot is present.68 As with any examination of the foot, examination must include the entire lower extremity as well. Specific examination of the foot reveals acute tenderness along the medial tuberosity of the os calcis. This tenderness may be at the origin of the central slip of the plantar fascia, or it may be deep, in which case it probably represents a deep inflammation, perhaps with involvement of the nerve to the abductor digiti quinti. The medial calcaneal nerve is palpated and tapped to search for ­paresthesias

Foot and Ankle 2045

1

3

2

B

1

2

3

A Figure 25F-18   A, Illustration of the nerve to the abductor digiti quinti. (1) Branch running to the medial process of the calcaneal tuberosity, bifurcating into a branch covering the perichondrium of the medial process and into another one running to a more lateral part of the calcaneal perichondrium. (2) Branch to the flexor digitorum muscle. (3) Branches to the abductor digiti minimi muscle.   B, Drawing of the nerve to the abductor digiti quinti as dissected in an adult foot showing parts of the abductor hallucis (1) and flexor digitorum (2) muscles, which have been removed; the nerve runs across the quadratus plantae muscle (3) in a plantar direction, then turns into a horizontal plane and proceeds laterally. (Redrawn from Rondhuis JJ, Hudson A: The first branch of the lateral plantar nerve and heel pain. Acta Morphol Neerland-Scand 24[4]:269-279, 1986.)

of the medial calcaneal nerve in the subcutaneous tissue; if found, this condition indicates inflammation and entrapment of the medial calcaneal nerve in this area. The plantar fascia is palpated to determine whether the plantar fascia is tender just at its origin or throughout its course. The plantar fascia is also palpated for nodules, the presence of

which suggests plantar fibromatosis. The plantar fascia is palpated both with the toes flexed so that it is supple and with the toes extended, which places tension on the plantar fascia. The tarsal tunnel is palpated and percussed to elicit any tenderness, inflammation, or Tinel’s sign of the posterior tibial, medial or lateral plantar, or medial calcaneal

Post. tibial n. Laciniate ligament (Flexor retinaculum) Lateral plantar n. Medial calcaneal branches

Medial planter n.

Laciniate ligament

Fascial leash

Nerve to abductor digiti quinti m.

Abductor hallucis m. Figure 25F-19   Site of entrapment of the posterior tibial nerve and its branches, demonstrating possible entrapment beneath the laciniate ligament and at the point at which the nerve passes through the fascia of the abductor hallucis muscle. (Redrawn from Baxter DE, Thigpen CM: Heel pain—operative results. Foot Ankle 5[1]:16-25, 1984. ©American Orthopaedic Foot and Ankle Society, 1984.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2046 DeLee & Drez’s� O

A

C nerves. Sensation of the foot is evaluated by light touch and pinprick to ascertain the status of the sensory nerves. The subtalar joint and ankle are examined for motion and mobility, both actively and passively to rule out referred pain from these areas. Active motor power of the muscles that cross or affect this area, such as the posterior tibial, anterior tibial, peroneus longus and brevis, and toe flexors and extensors, are checked to determine their motor power and also to see whether any pain is produced with active motor function. Neurologic examination of the remainder of the lower extremities and back is performed as indicated.

Diagnostic and Radiographic Studies Standing, full-weight bearing radiographs of the heel including the foot in the anteroposterior and lateral standing projections are taken. Radiographs taken in this manner provide information about the osseous structures of the foot as well as specific details of the os calcis. ­Standing radiographs also provide information about the biomechanical status of the foot because they are taken in the base and angle of gait and with the foot loaded. They thus represent the position and relationships of the bones of the foot during the stance phase of gait. Such radiographs also help to classify the foot as to type: normal, flat, or cavus. The presence of a spur or calcification along the medial tuberosity are also shown. Axial non–weight-bearing views of the os calcis may be taken to provide information about the os calcis in a second plane. Graham86 reported on radiographs of the heel taken at a 45-degree angle that showed bony condensation on the medial side (Fig. 25F-21). This was thought to represent a fatigue fracture. Ninety-eight

B

Figure 25F-20   A and B, Orthotic devices for flatfoot deformity due to increased strain on plantar fascia. A, Orthotic with forefoot post to prevent collapse of foot and thus strain on plantar fascia. B, Orthotic device constructed of polypropylene designed to prevent eversion and abduction of the foot to prevent strain on plantar fascia. C, Soft orthotic device constructed from polyethylene closed-cell foam thermoplastic material to provide cushioning and increased contact of arch.

percent of his cases were associated with a positive bone scan on the side of the heel that had pain. A technetium99m bone scan may be performed to provide objective evidence of an inflammatory abnormality in this area (Fig. 25F-22). Sewell and associates117 interpreted these bone scans in five patients with heel pain to show that the uptake is increased at the site of insertion of the plantar fascia into the calcaneus in patients with clinical signs of plantar fasciitis and that such signs may occur in the absence of any radiologic change. Changes in intensity of tracer uptake reflected symptomatic improvement. They suggested that radiologic imaging allowed quantitative assessment of inflammation of the “enthesis” and enabled therapy to be evaluated. Vasavada and colleagues118 obtained early blood pool images to detect soft tissue inflammation (plantar fasciitis) when delayed images were normal. Williams and colleagues107 evaluated a total of 45 patients with painful heel syndrome without evidence of associated inflammatory arthritis. These patients were studied using technetium-99m isotope bone scans and lateral and 45-degree medial oblique radiographs of both feet. Of the 52 painful heels in these 45 patients, 59.6% showed increased uptake at the calcaneus. These authors found that patients with scans showing increased uptake tended to have more severe heel pain and responded more frequently to a local hydrocortisone injection. Their findings suggested that patients with normal scans had some noninflammatory lesion such as simple trauma or entrapment of the medial calcaneal nerve. Their findings showed no evidence of stress fractures in contrast to Graham’s findings.86 Intenzo and Wapner119 studied 15 patients complaining of chronic heel pain that underwent three-phase

Foot and Ankle 2047

M

EA

B AY X-R

45

°

90°

90°

Figure 25F-21   Position for roentgenogram of the heel to demonstrate fatigue fracture on the medial side of the os calcis as described by Graham. The x-ray beam is angled downward from the heel, making a 90-degree angle to the toe-heel plane of the foot and a 45-degree angle to the horizontal plane of the cassette. (Redrawn from Graham CE: Painful heel syndrome: Rationale of diagnosis and treatment. Foot Ankle 3[5]:261-267, 1983. ©American Orthopaedic Foot and Ankle Society, 1983.)

t­ echnetium-99m MDP bone scintigraphy. Ten patients demonstrated abnormal scan findings consistent with plantar fasciitis with uptake only in the early soft tissue phase and had responded to conventional therapy. Two patients were found to have calcaneal stress fractures, and one patient demonstrated a calcaneal spur that required no treatment. The remaining two patients had normal scans and did not appear clinically to have plantar fasciitis. They found the three-phase bone scan useful in diagnosing plantar fasciitis and in distinguishing it from other causes of the painful heel syndrome.

Figure 25F-22   Bone scan showing increase in uptake in the calcaneal area associated with subcalcaneal pain syndrome.

Laboratory studies in most patients with subcalcaneal pain syndrome are normal. When the subcalcaneal pain syndrome is present, and especially if it is persistent and severe, consideration must be given to the diagnosis of a systemic disorder such as seronegative arthropathy. It has been reported that in patients with the subcalcaneal pain syndrome, there may be an incidence as high as 16% of subsequent development of a systemic arthritic disorder.72 Eastmond and associates120 described 26 patients with seronegative pauciarticular arthritis and positive HLA-B27. They noted that low back and buttock pain, Achilles tendinitis, and dactylitis of the toes were more frequent in HLAB27–positive patients. They suggested that arthritis and an increased frequency of HLA-B27 may occur in adults who do not have clinical or radiologic signs of any other sero­negative arthritis. The authors believe that this test should be considered part of the systemic work-up for a patient with chronic, recurrent, and incapacitating heel pain. Gerster121 found the painful heel syndrome with plantar fasciitis or Achilles tendinitis in 33 of 150 patients suffering from a seronegative spondyloarthritis. An HLA-B27 antigen was found in 91% of the patients. He contrasted this study with a study of 220 cases of rheumatoid arthritis. In these patients, he found that Achilles tendinitis was not encountered and that plantar fasciitis was exceptional. Gerster and Piccinin114 also described severe heel pain in 4 of 18 cases of juvenile-onset seronegative spondyloarthropathy. The plantar fascia was affected in each case. Four other patients among these 18 had mild heel pain. Gerster and colleagues122 also reported on 30 cases of severe heel pain in patients with seronegative spondyloarthropathy. Four patients underwent surgical intervention with either disinsertion of the plantar aponeurosis or rasping of the calcaneal spur. This surgery failed, and Gerster and colleagues stated that surgery is contraindicated in patients with severe heel pain associated with seronegative spondyloarthritis. Proximal neurologic causes of heel pain should also be considered. Tarsal tunnel syndrome may be present with referred pain to the heel and sole of the foot. A positive Tinel’s sign may suggest this diagnosis. Electromyographic and nerve conduction studies should be used to rule out this condition.68 Heel pain may also be referred from the lumbar spine. If a spinal or proximal cause appears to be a possibility, appropriate laboratory and radiographic studies should be performed as indicated. MRI has shed light on the anatomic structures involved in subcalcaneal heel pain.123 Heel pain can be caused by disorders of the plantar fascia, calcaneus, tendons, or adjacent nerves. Because these conditions can lead to pain located in a small area of the heel, a precise clinical diagnosis may be difficult. This article describes some of these various causes of heel pain and how MRI helps to characterize them. Grasel and colleagues124 evaluated various MRI signs of plantar fasciitis and to determine whether a difference in these findings exists between clinically typical and atypical patients with chronic symptoms resistant to conservative treatment. They found signs on MRI that included occult marrow edema and fascial tears. Patients with these manifestations appeared to respond to treatment in a manner similar to that of patients in whom MRI revealed more benign findings.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2048 DeLee & Drez’s� O

Treatment Options Nonoperative Therapy Management of subcalcaneal heel pain should initially begin with nonoperative treatment.125 Plantar fasciitis is a common cause of heel pain, which frustrates patients and practitioners alike because of its resistance to treatment. Although normally managed with conservative treatment, plantar fasciitis is frequently resistant to the wide variety of treatments commonly used, such as NSAIDs, rest, pads, cups, splints, orthotics, corticosteroid injections, casts, physical therapy, ice, and heat. Although there is no consensus on the efficacy of any particular conservative treatment regimen, there is agreement that nonsurgical treatment is ultimately effective in about 90% of patients. Because the natural history of plantar fasciitis has not been established, it is unclear how much of symptom resolution is in fact due to the wide variety of commonly used treatments. Gill and Kiebzak126 studied 411 patients with a clinical diagnosis of plantar fasciitis who were assessed for predisposing factors. Each patient completed an outcomes assessment survey instrument that ranked effectiveness of various nonsurgical treatment modalities. Listed in descending order of effectiveness, the treatment modalities assessed were short leg walking cast, steroid injection, rest, ice, runner’s shoe, crepe-soled shoe, aspirin or NSAID, heel cushion, low-profile plastic heel cup, heat, and Tuli heel cup. Treatment with a cast ranked the best. The Tuli heel cup ranked the poorest. Most of the treatments were found to be unpredictable or minimally effective. However, in their study, stretching and night splints were not included. A randomized trial by Landorf and associates in 2006 found that foot orthoses produce only small short-term benefits in function and pain in people with plantar fasciitis.127 When compared with a sham device, these orthoses do not have any long-term benefits. However, another study performed by Roos and colleagues demonstrated that foot orthoses and anterior night splints are equally effective in both short- and long-term times.128 Further studies will be necessary to elucidate the role of foot orthoses in the treatment of plantar fasciitis. O’Brien and Martin94 studied 58 painful heels in 41 patients who received conservative treatment. Seventy percent were classified as having excellent results (no remaining symptoms), 26% were classified as having good results (50% or less of the symptoms remained), and only 3.5% were symptomatic and classified as having poor results. Injection therapy was most successful when the preceding symptoms had been present for an average of 2.6 months. Orthotic therapy was most successful when the duration of the preceding symptoms was 21⁄2 years. Shikoff and coworkers98 reported a retrospective study of 195 patients with heel pain. The typical patient was middle-aged and overweight; 91% of the respondents were classified as having above normal weight for their sex and height. About half of the patients continued to wear heel padding or to take oral medication, or both, for months after the initial visit. Thirty percent experienced only marginal relief from pain or had an unsatisfactory result. The efficacy of oral NSAIDs was evaluated by Donley and colleagues.129 In this randomized, prospective,

placebo-controlled study, 29 patients were treated with a conservative regimen. In addition, patients were randomly assigned to placebo or NSAIDs. The study concluded that NSAIDs, particularly when taken for 2 months, increased pain relief and decreased disability when used in conjunction with a conservative treatment regimen. Callison80 reviewed 400 consecutive patients with heel pain who were seen in his office during a 40-month period from October 1985 to February 1989. Radiographs of all were obtained. Heel spurs were present in 45% and absent in 53%. The series consisted of 65% women and 35% men; 30% of the women were obese, but only 10% of the men were obese. Seven percent were involved in active sports. Patients were treated with steroid injections, orthoses, calfstretching exercises, and NSAIDs. Plaster immobilization was occasionally used. Results showed that 73% improved significantly within 6 months, 20% failed to improve, and 7% did not return and were lost to follow-up. Davis and coworkers130 studied 105 patients (70% female and 30% male; average age, 48 years) with 132 symptomatic heels who were treated according to a standard nonoperative protocol and then reviewed at an average follow-up of 29 months. The treatment protocol consisted of NSAIDs, relative rest, viscoelastic polymer heel cushions, Achilles tendon stretching exercises, and, occasionally, injections. Obesity, lifestyle (athletic versus sedentary), sex, and presence or size of heel spur did not influence the treatment outcome. Ninety-four patients (89.5%) had resolution of heel pain within 10.9 months. Six patients (5.7%) continued to have significant pain but did not elect to have operative treatment, and 5 patients (4.8%) elected to have surgical intervention. They concluded that despite attention to the outcome of surgical treatment for heel pain in the current literature, initial treatment for heel pain is nonoperative. The treatment protocol used in this study was successful for 89.5% of the patients. Wolgin and associates131 evaluated the long-term results of patients treated conservatively for plantar heel pain. After eliminating those patients with worker’s ­ compensationrelated complaints and those with documented inflammatory arthritides, data on 100 patients (58 females and 42 males) were available for review. The average patient age was 48 years (range, 20 to 85 years). The average follow-up was 47 months (24 to 132 months). Clinical results were classified as good (resolution of symptoms) for 82 patients, fair (continued symptoms but no limitation of activity or work) for 15 patients, and poor (continued symptoms limiting activity or changing work status) in 3 patients. The average duration of symptoms before medical attention was sought was 6.1, 18.9, and 10 months for the three groups, respectively. They concluded that although the treatment of heel pain can be frustrating because of its indolent course, a given patient with plantar fasciitis has a good chance of complete resolution of ­symptoms. The use of steroid injection in the treatment of painful heel syndrome has been advocated but is controversial.131-133 Few reports on the long-term efficacy of injection are available. Miller and colleagues134 evaluated the results of a single injection of corticosteroids in patients with painful heel syndrome. Twenty-seven heels in 24 patients were injected with a combination of 1 mL of lidocaine and 1 mL of betamethasone (6 mg). These patients had never previously received

Foot and Ankle 2049

an injection to their heels and had continued symptoms of pain after a trial of other nonoperative treatment modalities. After the injection, patients were seen and surveyed periodically for a period of 5 to 8 months. The amount of pain relief that they obtained, the length of time this lasted, and the amount of heel pain present at the final follow-up were recorded. At final follow-up, the pain had returned to its preinjection level in 13 feet. Based on the results of our study, they believed that a steroid injection is a reasonable adjunct in the treatment of painful heel syndrome, but that it is unlikely to provide permanent pain relief. Complications from steroid injection have been reported and can be severe. Rupture of the plantar fascia and irreversible fat pad atrophy135-138 have been reported. Loss of the plantar fad pad decreases the physiologic protection in the subcalcaneal region and increases the risk for the return of intractable symptoms. Injection should be used with caution in patients with a competent fat pad and not repeated more than once in any 3-month interval. If fat pad atrophy occurs, repeat injection should be avoided. In one study published in 2007, Lee and Ahmad evaluated the results of patients injected with intralesional autologous blood injection compared with corticosteroid injections.139 Sixty-four patients were randomized in this prospective, controlled trial. Patients were followed for 6 months. As compared with corticosteroid treatment, intralesional autologous blood injection is not as expeditious or effective in treating chronic plantar fasciitis. Iontophoresis has been used as an adjunct to conservative care by physical therapists. Acetic acid or dexamethasone can be used. A recent study by Osborne and Allison demonstrated that six treatments of acetic acid iontophoresis combined with taping provided greater relief from stiffness and equivalent pain relief when compared with dexamethasone iontophoresis with taping.140 Pfeffer and associates141 reviewed a 15-center prospective randomized trial to compare several nonoperative treatments for proximal plantar fasciitis (heel pain syndrome). Included were 236 patients with duration of symptoms of 6 months or less. Patients with systemic disease, significant musculoskeletal complaints, sciatica, or local nerve entrapment were excluded. Patients were randomized prospectively into five different treatment groups. All groups performed Achilles tendon and plantar fascia stretching in a similar manner. One group was treated with stretching only. The other four groups stretched and used one of four different shoe inserts, including a silicone heel pad, a felt pad, a rubber heel cup, or a custom-made polypropylene orthotic device. Patients were re-evaluated after 8 weeks of treatment. Combining all the patients who used a prefabricated insert, they found that their improvement rates were higher than those assigned to stretching only (P = .022) and those who stretched and used a custom orthosis (P = .0074). We conclude that when used in conjunction with a stretching program, a prefabricated shoe insert is more likely to produce improvement in symptoms as part of the initial treatment of proximal plantar fasciitis than a custom polypropylene orthotic device. Wynne and associates reported on the effect of counterstrain treatment in patients with plantar fasciitis.142 This ­ single-blind, randomized controlled study of crossover design evaluated 20 patients treated with either

c­ ounterstrain or placebo. This group found a significant relief of symptoms that were most pronounced in the first 48 hours after treatment. However, a 2-week stretching program has been shown to not statistically benefit firststep pain, foot pain, foot function, or general foot health when compared with patients not stretching.143 Other studies have demonstrated a tissue-specific plantar fasciastretching protocol as a key component in the treatment of chronic plantar fasciitis.144,145 Long-term use of this therapy produces a marked decrease in pain and functional limitations and a high rate of satisfaction. The use of night splints to prevent plantar flexion of the ankle, as an adjunct to the treatment of subcalcaneal pain, has received considerable interest in the past few years. Wapner and Sharkey146 originally reported the results of the use of molded ankle-foot orthosis night splints for the treatment of recalcitrant plantar fasciitis in 14 patients with a total of 18 symptomatic feet. All patients had symptoms for greater than 1 year and had previously undergone treatment with NSAIDs, cortisone injections, shoe modifications, and physical therapy without resolution. All patients were provided with custom-molded polypropylene anklefoot orthoses in 5 degrees of dorsiflexion because no commercially manufactured splints were yet available. With continued use of NSAID medication, Tuli heel cups, Spenco liners, and general stretching exercises, successful resolution occurred in 11 patients in less than 4 months. There were three failures. It is thought that night splints provide a useful, cost-effective adjunct to current therapeutic regimens of plantar fasciitis by assisting in maintaining the flexibility gained by stretching exercises and relieving morning start-up pain, and thus reducing the time to resolution of symptoms. Multiple subsequent studies have demonstrated the usefulness of these devices, and they are now readily commercially available.79,147-150 Martin and colleagues151 reported on an outcome study of 400 patients with chronic plantar fasciitis treated nonoperatively and concluded that patients could expect a good outcome. Compliance with their protocols did not have a correlation with outcome, with one exception. Patients with chronic conditions who were compliant with the use of the night splint had a better outcome. Patients’ subjective perceptions were that stretching, night splints, and heel pads were of equal importance in their treatment. This study suggests that within the first 12 months of onset, early, aggressive, nonsurgical treatment offers the best chance of a good outcome. Hyland and associates performed a randomized controlled trial comparing calcaneal taping, sham taping, and plantar fascia stretching for the short-term management of plantar heel pain.152 They found calcaneal taping to be more effective for the relief of plantar heel pain than stretching, sham, taping, or no treatment. Alvarez77,153 was one of the first to report on the OssaTron as another alternative for management of heel pain syndrome after failure of nonoperative management and before surgical management. His study evaluated primarily the safety and early preliminary efficacy of the OssaTron in the treatment of patients with plantar fasciitis unresponsive to nonoperative management. Twenty heels of 20 patients were treated with 1000 extracorporeal shock waves from the OssaTron to the affected heel after administration

�rthopaedic ����������� S �ports ������ � Medicine ������� 2050 DeLee & Drez’s� O

of a heel block. Each patient was evaluated by radiography, Kin Com, range of motion, and physical examination, including evaluation of point tenderness by means of a palpometer and according to a 10-point visual analog scale. The control was the contralateral heel. Patients also performed self-evaluation by means of patient activities of daily living questionnaire and pain reported by a 10-point visual analog scale. There were no complications or adverse effects attributed to the procedure of orthotripsy. Of the 20 patients treated, 17 were improved or pain free. Eighteen of the 20 subjects treated stated that they would undergo the procedure again instead of surgery. Based on these results, Alvarez concluded that orthotripsy is a safe and effective method of treating heel pain syndrome that has been unresponsive to nonoperative management. Zingas and associates154 studied the safety and efficacy of musculoskeletal shock-wave therapy in the treatment of chronic plantar fasciitis in 29 patients with chronic plantar fasciitis who were enrolled in a randomized, 1:1 allocated, placebo-controlled, prospective, double-blind clinical study with two groups: one receiving extracorporeal shockwave therapy with the Dornier Epos Ultra and the other receiving sham treatment. The authors hypothesized that shock-wave therapy will be useful in the treatment of chronic plantar fasciitis that has failed conventional conservative methods. Preliminary results indicate that shockwave therapy is a safe and efficacious treatment for chronic plantar fasciitis with only minor, transient adverse effects. The U.S. Food and Drug Administration has approved this treatment for chronic plantar fasciitis. Since that time, numerous studies have been published evaluating the efficacy of shock-wave therapy for the treatment of chronic proximal plantar fasciitis. One of the most compelling studies is a double-blind, multicenter, randomized controlled study performed by Malay and associates.155 The cohort in this study was followed for 1 year after one treatment of shock-wave therapy. The authors found that patients treated with shock-wave therapy had objectively and subjectively less pain than those treated with sham intervention. Similar results have been published by other groups in the literature.156-160 Although some of these studies use one treatment of high-energy shock-wave therapy, whereas others use three weekly treatments of low-energy shock-wave therapy, it is unclear what constituted “high-energy” treatments. One study evaluating moderate-dose shock-wave therapy has demonstrated no effect on outcome measures, 6 months after treatment when compared with sham therapy.161 There is no correlation between the presence or absence of a heel spur and the eventual treatment outcome.162 Location of the treatment probe for shock-wave therapy has been found to be most effective when placed, with patient aid, at the most tender site.163 As extracorporeal shock-wave therapy continues to be studied, its effectiveness and efficacy in the treatment of chronic plantar fasciitis will continue to be elucidated.

Operative Therapy Surgical management of chronic plantar fasciitis remains controversial. The American Orthopaedic Foot and Ankle Society (AOFAS) issued a position statement on the timing of surgical intervention (Box 25F-1).

Review of the literature provides no clear consensus on the role of surgery. Snook and Chrisman,100 reporting on 22 patients with 25 painful heels, found that 16 patients with 18 painful heels obtained relief from pain with a variety of conservative measures, including a plastic heel cup. Seven of the patients (8 painful heels) required surgical therapy consisting of excision of the medial and inferior tuberosities of the os calcis. All obtained pain relief. The follow-up period in these patients ranged from 2 to 7 years. Goulet85 reported on the use of soft orthoses for treating plantar fasciitis. Ali103 reported that steroid injections provided relief in 13% of the patients treated in his series. Plantar fasciotomy provided permanent relief in 75%, and plantar fasciotomy with excision of the calcaneal spur produced a cure in 85%. However, Mann71 stated that surgical treatment of calcaneal spurs gives only 50% to 60% satisfactory results. Ward and Clippinger102 used a curved oblique plantar incision on the proximal aspect of the medial longitudinal arch to release the plantar fascia in eight feet with recalcitrant plantar fasciitis (Fig. 25F-23). Seven feet became pain free, and the eighth was 75% improved. Normal sensation was preserved in all cases. There were no painful scars or neuromas. Shmokler and colleagues106 used fluoroscopy to aid in the removal of heel spurs. They used this modality in 130 patients between January 1, 1984 and September 5, 1987 but did not provide a statistical analysis. Contompasis81 operated on 129 patients through a medial horizontal incision, performing exostosectomy and Box 25F-1 American Orthopaedic Foot and Ankle Society Position Statement: Endoscopic and Open Heel Surgery   1. Nonsurgical treatment is recommended for a ­minimum of 6 months and preferably 12 months.   2. More than 90% of patients respond to nonsurgical treatment within 6 to10 months.   3. When surgery is considered in the remaining patients, a medical evaluation should be considered before ­surgery.   4. Patients should be advised of complications and risks if an endoscopic or open procedure is not indicated.   5. If nerve compression is coexistent with fascial or bone pain, an endoscopic or open procedure should not be attempted.   6. The AOFAS does not recommend surgical procedures before nonoperative methods have been used.   7. The AOFAS does support responsible, carefully planned surgical intervention when nonsurgical ­treatment fails and work-up is complete.   8. The AOFAS supports cost constraints in the treatment of heel pain when the outcome is not altered.   9. The AOFAS recommends heel padding, medications, and stretching before prescribing custom orthoses, and extended physical therapy. 10. This position statement is intended as a guide to the orthopaedist and is not intended to dictate a treatment plan.

Foot and Ankle 2051

fasciotomy, fasciotomy only, or exostosectomy only. Most heels underwent more exostosectomies (115 cases) than fasciotomies. Overall, there was complete improvement in 43%, some improvement in 38%, no improvement in 17%, but no worsening of symptoms in any patient. Du­Vries84 in 1957 treated 37 patients with medial incision and a fasciotomy and rasping of the spur; he reported good results without recurrence of symptoms. Hassab and El-Sherif  87 drilled the os calcis to obtain relief of recalcitrant heel pain. They performed 68 operations in 60 patients and reported excellent results in 62, good in 2, and poor in 4. Jay and associates89 performed calcaneal decompression for chronic heel pain through a lateral approach in four patients and achieved pain relief in three of the four patients. One patient had immediate relief. Two experienced relief from symptoms several weeks postoperatively. The fourth patient did not become pain free. Kenzora164 considered the painful heel syndrome an entrapment neuropathy. He described exploration of the nerve to the abductor digiti quinti and release of the nerve through a midline plantar incision of the heel. He submitted a preliminary report on six patients, none of whom were dissatisfied but had only a short follow-up. Two patients had temporary numbness along the medial aspect of the incision. He considered the procedure experimental. Satku and colleagues165 described heel pain produced by an unusual cause—that of an osteocartilaginous nodule within the heel. This was a glistening, circumscribed, ­lobulated, ovoid mass measuring 1.6 × 2.5 × 1.4 cm. The patient was relieved of pain with excision of the nodule. Shaw and associates166 called attention to the fact that heel pain can occur in patients with sarcoidosis. They reported seven cases in which sarcoid-related heel pain was a major characteristic of the sarcoid disease process and noted that bilateral heel pain can be a presenting symptom of sarcoidosis and can accompany or precede sarcoid arthritis. Hoffman and Thul167 reported two cases of fracture of the os calcis following surgical procedures for excision of the calcaneal exostosis and plantar fasciotomy. Lester and Buchanan91 reported on 10 patients who underwent stripping of the plantar fascia and superficial plantar muscles

4

2 1

3

Figure 25F-23   Incisions used in subcalcaneal pain surgery: (1) medial incision, (2) oblique incision, (3) incision of Ward and Clippinger, (4) incision for complete release and exploration.

from the calcaneus. These patients had been treated for an average of 12.4 months before surgery. They were followed for 24 months after surgery. Complete symptomatic relief was obtained in all patients, although hypoesthesia of the heel was present in five feet after the operation. Three patients were receiving workmen’s compensation and returned to work within 16 weeks of surgery. Endoscopic plantar fascia release has received significant attention in the past several years. Barrett and Day78 described this technique as having the advantage of less tissue damage than open treatment. In a follow-up multicenter study of 652 procedures, they reported 62 complications in 53 patients but thought that it afforded satisfactory results. O’Malley and colleagues168 reviewed the surgical results after endoscopic plantar fasciotomy in 16 patients (20 feet) with an average preoperative duration of symptoms of 4 years. Of the 20 feet, 9 had complete relief of pain, whereas symptoms were improved in 9 feet. One patient with bilateral symptoms had no relief in either foot. The average AOFAS hindfoot score improved from 62 to 80, a statistically significant difference. Unilateral patients did better than bilateral, with no bilateral patients reporting complete resolution of symptoms. There were no iatrogenic nerve injuries. On the basis of their review, they recommend endoscopic plantar fasciotomy as an alternative to open plantar fascial release for those patients with recalcitrant heel pain who fail conservative treatment. Other authors have reported various complications from endoscopic plantar fascial release, including stress fractures,169 pseudoanuerysm formation,170 and recurrence of pain.171 Because of the high incidence of lateral foot pain, it is recommended that only the medial two thirds of the plantar fascia be released. Kitaoka and colleagues172-175 in a series of articles has shed light on the biomechanical risk to the foot from release of the plantar fascia. They have demonstrated that changes in displacement were more pronounced in unstable or destabilized feet. Their data suggest that operations involving fasciotomy affect arch stability and should not be performed in patients with evidence of concomitant pes planus deformity, because of the likelihood of further deformation. Flattening of the longitudinal arch occurred in their clinical series. Dynamic force-plate studies showed differences in peak vertical, fore-aft, and lateral-medial forces between patients and matched controls. More rapid progression of weight-bearing along the longitudinal axis of the foot during stance phase in patients indicated avoidance of heel loading. Yu and colleagues176 studied 17 patients (15 women, 2 men; age range 22 to 59 years; mean age, 40 years) with foot pain with a mean duration of 22 months after undergoing a fasciotomy. Each patient was instructed to localize the pain to a region of the foot; classify the pain as new onset, persistent, or recurrent; and characterize it as to the action that produced the greatest pain. MRIs were evaluated for abnormalities of the plantar fascia, perifascial soft tissues, tendons, and osseous structures. The plantar fascia appeared thick in all ankles (mean, 8.0 mm; range, 6 to 12 mm). A total of 25 symptomatic sites were assessed. An acute plantar fascia rupture explained plantar symptoms in 2 feet. In another 16 feet (12 with plantar heel pain and 4 with nonspecific heel pain), 6 had documentation of acute

�rthopaedic ����������� S �ports ������ � Medicine ������� 2052 DeLee & Drez’s� O

plantar fasciitis, and 9 demonstrated perifascial edema. Of the latter 9 feet, 5 demonstrated abnormalities of the posterior tibialis, peroneus longus, and peroneus brevis tendons. The pain localized to the medial arch in 6 feet; 5 feet had abnormalities of the posterior tibialis tendon; and 1 foot demonstrated edema in the flexor digitorum brevis muscle. The pain localized to the lateral midfoot in 1 foot, which had a cuboid stress fracture. The cause of foot pain in patients who had a plantar fasciotomy appeared to be multifactorial. Three likely causes of pain were identified: persistent or recurrent acute plantar fasciitis, pathology related to arch instability, and structural failure from overload.

Treatment of Athletes with Plantar Fasciitis In 1986, Lutter92 outlined the decision-making process in athletes with subcalcaneal pain. He described 182 patients with heel complaints related to sports injuries; most of the patients were runners (76%). About 20% of these patients required 3 to 4 months of conservative treatment before returning to sports activity. Five percent had chronic heel pain and did not recover within 9 to 12 months. For these, a surgical approach was considered. Lutter stated that the decision to operate on an athlete should be based on six specific tenets: (1) a correct diagnosis has been made, (2) about 12 months of conservative treatment have been tried, (3) electromyography and appropriate nerve blocks have been performed for diagnosis, (4) the surgeon has a thorough knowledge of the anatomy or performs a complete review, (5) the patient understands that even successful surgery may not allow him or her to return to high-­performance athletics, and (6) correct and appropriately directed surgery is chosen. The procedure used in these patients depended on the preoperative diagnosis and varied from release of the nerves to release of the fascia to complete exploration of the posterior tibial nerve and its branches and release of the plantar fascia. Cycling or swimming was begun 2 weeks postoperatively. Gentle walk-dash-run training and a gradual escalation up to running was allowed about 6 weeks after surgery. Patients were asked to refrain from walking until they were pain free and had no tenderness. If pain occurred with increasing activity, the work-up was cut by 50% until the patient could tolerate the work-up without pain. Shock-wave therapy has been examined in the treatment of chronic plantar fasciitis in the running athlete. Rompe and colleagues demonstrated that after 6 months of three weekly treatments with shock-wave therapy, the treatment group experienced greater relief than did the sham treatment group.177 Baxter and Thigpen64 performed 34 operative procedures in 26 patients with recalcitrant heel pain. The procedure consisted of isolated neurolysis of the nerves supplying the abductor digiti quinti muscle as it passed beneath the abductor with release of the deep fascia of the abductor hallucis longus and removal of the heel spur if it impinged on or produced entrapment of the nerve. Among the 34 heels operated on, there were 32 good results and 2 poor results. Clancy2 treated patients with a medial heel wedge and flexible leather support, heel cord stretching, and rest

for 6 to 12 weeks with a gradual return to running wearing the orthotic and the medial heel wedge for 10 weeks. In patients who failed to respond, surgery consisting of release of the plantar fascia and the fascia over the abductor ­hallucis longus was recommended. The 15 patients in whom surgery was performed returned to running within 8 to 10 weeks. D’Ambrosia and colleagues82,83 had success using anti-inflammatory medication, physical therapy, orthotic devices, and shoe modifications. Orthotic devices seemed to be the most useful part of the treatment. These orthotic devices were made of Vitrathene, Plastazote II, or Plastazote III. Henricson and Westlin88 described 11 heels in 10 athletes with chronic heel pain that was unrelieved by conservative therapy. The pain was due to compression of the calcaneal branches of the tibial nerve. There was entrapment of the anterior calcaneal branch where the nerve passed between the tight and rigid edges of the deep fascia of the abductor hallucis and the medial edge of the os calcis. Surgery consisted of identifying and releasing the tibial nerve and both calcaneal branches and releasing the deep fascia of the abductor hallucis. Follow-up for 58 months after surgery revealed that 10 of the 11 heels were asymptomatic. The patients had resumed athletic participation after an average of 5 weeks. It seems to us that both the nerve to the abductor digiti quinti and the medial calcaneal nerve were released. Jørgensen178 described three athletes with unusually soft and fat heel pads who were heel strikers. These athletes were treated successfully with external heel shock absorption pads. Jørgensen reported a diagnostic method for evaluating the shock-absorbing ability of the heel pad by using a visual compressible index calculated on the basis of radiographic films of the heel, loaded and unloaded by body weight. According to Kwong and colleagues,90 fasciitis is produced by an excessive amount or a prolonged duration of pronation. Temporary relief was obtained in his patients by the use of anti-inflammatory drugs and therapy. Longterm relief was obtained by achieving adequate control of pronation through the use of semirigid custom-molded orthotics that reduced plantar fascial strain by supporting the first metatarsal bone and controlling calcaneal position. These devices were used in conjunction with a firm posterior counter shoe. Leach and coworkers137 stated that most patients respond well to conservative therapy consisting of decreased activity, stretching, heel cups, and occasional local steroids. They described 15 competitive athletes in whom 16 operations were performed. Surgery consisted of release of the plantar fascia at the insertion of the os calcis, making the incision along the medial aspect of the heel. In one instance, the medial calcaneal nerve was involved in the inflammatory process. One patient returned to running at 6 weeks; most returned to running 9 weeks after surgery. Most patients continued to improve up to 6 months after the surgical procedure. Of the 15 operations, 14 were entirely successful in that the athletes returned to their previous level of activity. One failure occurred in a marathon runner who improved but was unable to train at the level he desired. There were no complications.

Foot and Ankle 2053

In 1984, McBryde93 reported that in his running clinic, plantar fasciitis composed 9% of the total running disorders seen. The conservative (nonoperative) approach of McBryde and his group consisted of (1) ice massage for 2 minutes 4 to 6 times daily, including before and after runs; (2) heel cord stretching for 3 to 5 minutes 3 to 4 times daily; (3) posterior tibial and peroneal strengthening; (4) heel cushioning and control; and (5) anti-inflammatory medication. This regimen was usually successful in treating runners with plantar fasciitis who were seen within the first 8 weeks. In runners with symptoms lasting longer than 6 weeks, a period of absolute rest with casting was usually required. Orthoses were used. Five percent of the patients in the series underwent surgery, consisting of plantar fascial release through a short 1-inch longitudinal incision in the medial arch. All re-embarked on a successful running program 6 to 12 weeks after surgery. Overall, among the 100 patients with plantar fasciitis, 82 recovered with a conservative approach, 11 stopped running, 5 underwent surgery (all of whom returned to running), and 2 refused surgery and continued to be symptomatic. Rask95 reported a medial plantar neurapraxia that he termed jogger’s foot. Three cases were reported in which there was probable entrapment of the medial plantar nerve behind the navicular tuberosity in the fibromuscular tunnel formed by the abductor hallucis; the inciting factor was eversion of the foot. All three patients were treated successfully with conservative measures, including change in running posture of the foot, anti-inflammatory medication, and proper footwear. Sammarco97 divided heel pain into three clinical classifications and provided a treatment algorithm for these conditions. First, calcaneodynia, produced by a stress fracture,

A

was treated with a foot orthosis and decrease in running. Second, plantar fasciitis was treated with a foot orthosis, anti-inflammatory medication, and a flexibility program, plus occasional injections of corticosteroids. Recalcitrant plantar fasciitis (defined as symptoms lasting for more than 1 year) was considered for release of the plantar fascia and bone spur if one was present. Finally, calcaneodynia involving entrapment of the medial calcaneal nerve or the nerve to the abductor digiti quinti was diagnosed by Tinel’s sign and treated with an orthosis or release of the nerve. Snider and colleagues99 reported 11 operations for plantar fascial release for chronic fasciitis in nine distance runners who had had symptoms for an average of 20 months and had not responded to nonsurgical treatment. The results of the operations were excellent in 10 feet and good in one foot, with an average follow-up time of 25 months. Eight of nine patients returned to their desired level of full training at an average time of 4.5 months. In 2004, Saxena179 reported on 16 athletes with intractable heel pain who had failed conservative care. Most of these athletes were runners. These patients were treated surgically with a uniportal endoscopic plantar fasciotomy. Saxena found that runners were able to return to athletic activity on average 2.6 months after surgery. Five poor results were found in patients with a body mass index of greater than 27.

Treatment of Children Calcaneal apophysitis (Sever’s disease) is a common cause of heel pain in the young child. Micheli and Ireland180 reported calcaneal apophysitis in 137 heels in 85 children, both heels being affected in 61% of the patients. Soft Plastazote orthotics or heel cups were used in 98% of the patients

B

Figure 25F-24   A, Over-the-counter heel cups designed to compress the fat pad and cushion the heel. B, Custom orthoses used for treatment of subcalcaneal pain syndrome. These may be made of many types of material. Consistent features are support of the arch, presence of cushioning material, recess for area of pain beneath the heel, and slight medial elevation.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2054 DeLee & Drez’s� O

in combination with proper athletic footwear. Physical therapy consisted of lower extremity stretching and ankle dorsiflexion strengthening. All patients improved and were able to return to the sport of their choice 2 months after the diagnosis. Two patients had recurrent difficulty. Sullivan181 reported that radiographic irregularity of the calcaneal apophysis is the rule rather than the exception. He found no evidence that treatment altered the radiographic picture. Because there are a variety of conditions that cause pain in the heel, he believes that Sever’s disease might not even be a true entity. This author notes that differentiation can be made on the basis of location of the pain, the pain of Achilles tendinitis being proximal to the medial bursa

Authors’ Preferred Method

of

and the pain of calcaneal apophysitis being along the distal portion of the posterior part of the os calcis. In summary, conservative treatment consisting of antiinflammatory medications, orthoses, heel cups (Fig. 25F-24), injections, physical therapy, and decreased activity level is effective in 95% of cases of subcalcaneal pain syndrome. In patients who do not respond to an adequate trial of these conservative measures, surgery consisting of release of the plantar fascia or release of the nerve to the abductor digiti quinti or the sensory medial calcaneal nerve can be performed with an expectation of good results. Heel spur, if present, is removed. However, problems with decreased sensation of the heel and persistent pain have been reported.

Treatment

Plantar fasciitis arising at the medial tuberosity probably represents a traction periostitis with degeneration and tears of the plantar fascia and subsequent secondary involvement of the adjacent structures such as the medial calcaneal nerve and the nerve to the abductor digiti quinti. Occasionally, primary entrapment of the nerve of the abductor digiti ­quinti and the sensory branch of the medial calcaneal nerve may occur. A calcaneal spur is present in 50% of the cases and may be part of the inflammatory process. In longstanding cases, the chronic inflammation and traction on the insertion of the plantar fascia may result in a stress fracture of the os calcis, although this is more common in the older patient. An accurate diagnosis of the etiology of the pain and any concomitant conditions such as chronic heel fat pad atrophy, cavus foot deformity, contraction of the Achilles tendon, or nerve entrapment must be determined because they will affect both the prognosis and treatment modalities employed. At the first visit, we recommend a discussion with the patient to explain the nature of this condition. We tell them the good news is that it almost always gets better without the need for surgery. The bad news is that it usually does not get better quickly. In our initial consultation, we explain that we expect it to take at least 6 to 9 months to resolve completely and sometimes up to year. It is important for both the physician and the patient to set this time frame early on to decrease the frustration often associated with this diagnosis. The patient with subcalcaneal pain syndrome and associated proximal plantar fasciitis is evaluated to determine which foot type is present and whether there is any associated abnormality of the lower extremity or body. If there is none, I begin treatment consisting of anti-inflammatory medication and an orthotic device designed to cushion the heel and relieve the pressure on the tender area of the heel. There are many types of over-the-counter heel cups and cushioning devices that may be used, but we generally prescribe viscoelastic silicone heel pads and ask the patient to use well-cushioned shoewear such as a running shoe. Activities are restricted according to the patient’s symptomatic tolerance. The mainstay of treatment is to stretch both the Achilles and plantar fascia. Physical therapy consisting of stretching, strengthening, ultrasound, and at times, iontophoresis and phonophoresis may be used. We stress to the patient that following up with a daily home program of stretching is essential. If the condition is not responsive to this regimen within

the first 6 weeks or has been present for more than 3 months, a night splint with the foot maintained in dorsiflexion is prescribed. We explain to the patient that we are treating pain on the bottom of their foot while we are attempting to allow them to keep ambulating on the foot. This is similar to trying to fill up a bath tub with the plug out of the drain. The only way it works is if more water is going in than is coming out. Therefore we use multiple modalities at once to try to diminish the inflammation more rapidly than we are creating it. Cross-training activity is important. The patient should avoid repetitive impact activities such as running or treadmill and cross-train with a bike or elliptical trainer. These assist in maintaining conditioning and increasing flexibility while avoiding cyclic loading. If this does not suffice after a period of 3 months and the patient’s symptoms are severe enough, complete rest in a short leg weight-bearing cast for 4 weeks is considered. We explain this is like plugging the drain in the tub analogy. Once the cast is removed, the protocol is restarted. The exception to this is the patient with severe pain to medial-lateral compression of the calcaneus at the level of the insertion of the plantar fascia that is greater than pain to palpation at the insertion on the calcaneus. This is more common in the elderly patient and often indicates a stress fracture. In these patients, a bone scan is ordered before starting therapy, and if it is positive for stress fracture, casting will be started. Once the patient has become asymptomatic without tenderness and has maintained this status for 4 to 6 weeks, a gradual increase in activity may be allowed. The orthotic is continued for several months. After several months, the orthotic is discontinued unless the patient has a biomechanical abnormality of the foot such as a flatfoot or cavus deformity. If the patient has a flatfoot deformity, a device designed to correct the biomechanical abnormality, support the foot, and prevent the abnormal biomechanical stresses along the plantar fascia and the medial side of the heel is used. If the patient has a cavus deformity, a soft orthotic designed to decrease the shock and increase the weight-bearing area may be used indefinitely depending on the patient’s symptoms. When a patient does not have a normal foot and has had a significant episode of subcalcaneal pain requiring treatment, I suggest that the orthotic be continued permanently. ­Although the over-the-counter type of heel cup can be used

Foot and Ankle 2055

Authors’ Preferred Method

of

Treatment—cont’d

Incision

B

A

Superficial fascia of abductor hallucis Deep fascia of abductor hallucis Plantar fascia Abductor hallucis m.

C

D

E Figure 25F-25   Plantar fascia and nerve release. A, Incision. B, Release of the abductor hallucis muscle. C, Abductor hallucis muscle is reflected proximally. D, Abductor hallucis is retracted distally. E, Resection of small medial portion of the plantar fascia.

initially to try to provide some symptomatic relief, in a patient with a true biomechanical abnormality, a specific orthosis should be used. In our practice, we generally avoid steroid injection to prevent any subsequent loss of fat pad. Although reasonable results have been obtained with injection, we believe that fat pad atrophy is an irreversible complication with long-term negative consequences. We use cast immobilization and NSAIDs instead. If the patient does not respond to this conservative regimen, has an injury or impairment sufficient to prevent the performance of the desired activity, and demonstrates no other abnormality or systemic cause of the pain, I consider surgical intervention after 9 to 12 months of therapy. It has been our experience that surgery is rarely required unless there is an associated nerve lesion. In the athlete, we attempt to perform the least amount of surgery commensurate with a probably good result. In the athlete with a recalcitrant subcalcaneal pain syndrome who desires to continue

a­ thletic ­activity, the exact site of the abnormality is evaluated carefully by means of differential blocks using a long-­acting anesthetic such as bupivacaine hydrochloride. This allows more precise localization of the exact area of pathology. We attempt to rule out tarsal tunnel syndrome and to define whether the abnormality lies along the medial tuberosity or is produced by the sensory medial calcaneal nerve or the nerve to the abductor digiti quinti. Before surgery, nerve conduction and electromyographic studies are considered in patients in whom a tarsal tunnel syndrome is possible. Laboratory studies are done to exclude systemic arthritis or spondyloarthropathies. Bone scans with technetium-99m may be considered if one suspects a fatigue fracture or if the exact location of the pain is not clear. We approach the area through a medial oblique incision along the heel as described by Schon (Fig. 25F-25).182 Loupe magnification may be used. The sensory branch of the medial calcaneal nerve is located, inspected, and preserved. If there is entrapment of the medial calcaneal nerve as it comes through Continued

�rthopaedic ����������� S �ports ������ � Medicine ������� 2056 DeLee & Drez’s� O

Authors’ Preferred Method

of

Treatment—cont’d

the fascia, this is explored and released. The central slip of the plantar fascia is released at its origin. The spur is removed using a small rongeur after stripping the muscles. The deep fascia of the abductor hallucis and the fascia of the quadratus plantae are divided to release the nerve to the abductor digiti quinti as it passes laterally. The wound is irrigated and closed with several plain ­subcutaneous and ­interrupted mattress sutures. The patient is placed in a short leg cast and kept non–weight-bearing for 4 to 7 days. Weight-bearing in a cast or an over-the-counter brace is maintained for 14 more days. At 3 weeks, weight-bearing with a shoe is permitted. Running is started at 6 to 12 weeks, and activity following this is allowed as tolerated. If the patient has a biomechanical foot abnormality, an orthotic ­device is used postoperatively. In the patient who has symptoms that may involve the posterior tibial nerve as well as pathology of the medial tuberosity, or had undergone recurrent or failed surgery, a

Criteria for Return to Sports Participation Because plantar fasciitis associated with pain in the medial tuberosity in the athlete may be associated with a variety of presentations and degrees of severity and does not produce any significant structural effect other than alteration in form, activity may be allowed within the patient’s symptomatic tolerance, although this varies greatly according to the pain component and the athlete’s restriction. The exception to this is if the patient is developing a stress fracture. Once the pain has subsided with conservative treatment and no tenderness is present, gradually increasing activity is allowed using an orthotic within the shoe. If symptoms do not recur with the specific activity desired, full activity is allowed. After surgical intervention, the patient is allowed increased activity when the symptoms of pain with activity and acute tenderness have resolved. This usually occurs between 6 and 12 weeks after surgery.

C

r i t i c a l

P

o i n t s

RETROCALCANEAL BURSITIS • Retrocalcaneal pain syndrome is commonly associated with the high-arched cavus foot and the varus heel. • It is a condition characterized by inflammation of the retrocalcaneal bursa, the Achilles tendon just above its insertion, and at times the tissue between the Achilles tendon and the skin. • Steroid injections should be made into the retrocalcaneal area, not into the tendon. • It is generally managed by conservative measures consisting of anti-inflammatory medication, decreased activity, padding to prevent pressure on the affected area, orthoses or heel lifts, and strengthening and stretching exercises. • Surgery generally consists of excision of the exostosis and the retrocalcaneal bursa and at times the adventitious bursa, if it is present, and correction of the Achilles tendon pathology with tendon transfer if necessary.

more extensive operation is considered on the understanding that whereas it may relieve the pain, the operation may produce more morbidity with a decreased chance of recovery. The operation for complex, resistant, or recurrent pain along the medial side of the heel consists of exploration of the posterior tibial nerve from the medial side of the ankle to the point at which it exits through the foramina of the abductor muscles and exploration and release of the medial calcaneal nerve and the nerve to the abductor digiti quinti along with release of the central portion of the plantar fascia and excision of the heel spur, if present (see Fig. 25F-25). The patient is kept non–weight-bearing for 2 weeks and can then bear weight in a short leg cast for 2 more weeks; increased activity is started at 12 weeks. This operation is used only for patients with recalcitrant conditions. It carries with it the expectation that the patient will probably, but not certainly, be able to return to his or her preinjury status.

PLANTAR FASCIITIS • Conservative treatment consisting of anti-inflammatory medications, orthoses, heel cups, injections, physical therapy, and decreased activity level is effective in 95% of cases of subcalcaneal pain syndrome. • Surgery consists of release of the plantar fascia or release of the nerve to the abductor digiti quinti or the sensory medial calcaneal nerve.

S uggested

R eading

deVos RJ, Weir A, Visser RJA, et al: The additional value of a night splint to ­eccentric exercises in chronic midportion Achilles tendinopathy: A randomized controlled trial. Br J Sports Med 21:e5, 2007. DiGiovanni BF, Nawoczenski DA, Malay DP, et al: Plantar fascia-specific stretching exercise improves outcomes in patients with chronic plantar fasciitis. J Bone Joint Surg Am 88:1775-1781, 2006. Gerken AP, McGarvey WC, Baxter DF: Insertional Achilles tendinitis. Foot Ankle Clin 1:237-248, 1996. Heneghan JA, Pavlov H: The Haglund painful heel syndrome: Experimental investigation of cause and therapeutic implications. Clin Orthop 187:228-234, 1984. Leach RE, James S, Wasilewski S: Achilles tendinitis. Am J Sports Med 9:93-98, 1981. Schepsis AA, Jones H, Haas AL: Achilles tendon disorders in athletes. Am J Sports Med 30:287-305, 2002. Schepsis AA, Leach RE: Surgical management of Achilles tendinitis. Am J Sports Med 15(4):308-315, 1987. Wapner KL, Pavlock GS, Hecht PJ, et al: Repair of chronic Achilles tendon rupture with flexor hallucis longus tendon transfer. Foot Ankle 14(8):443-449, 1993. Wapner KL, Sharkey PF: The use of night splints for treatment of recalcitrant plantar fasciitis. Foot Ankle 12(3):135-137, 1991.

R eferences Please see www.expertconsult.com

Foot and Ankle 2057

S e c t i o n

G

Entrapment Neuropathies of the Foot Gregory P. Guyton, Lorenzo Gamez, and Roger A. Mann

An entrapment neuropathy results from localized damage or inflammation of a peripheral nerve, which may be due to an injury such as direct trauma or a mechanical irritation such as impingement against an anatomic structure. This entrapment may result in direct axonal compression or vascular changes. Generally, the common factor in an entrapment neuropathy is a nerve passing through a rigid compartment, through a fibrous or fibro-osseous canal in which the nerve changes direction, or over a bony ­prominence. The main clinical complaint of patients with entrapment neuropathy is pain or dysesthesia. Depending on the nerve involved, the pain may be worse at night or at rest, or at other times pain may be activity related. Retrograde pain, known as the Valleix phenomenon, may be a prominent feature of the patient’s complaint. When a purely sensory nerve is involved, a rather clear distribution of pain can be elicited from the patient, whereas compression of a motor nerve gives rise to a more diffuse type of pain. Occasionally, atrophy of the innervated muscles may occur. In all cases of entrapment neuropathies, it is imperative to elicit a careful history from the patient, carry out a physical examination, and obtain appropriate electrodiagnostic studies.

TARSAL TUNNEL SYNDROME The tarsal tunnel syndrome in its basic form is an entrapment of the posterior tibial nerve within the tarsal canal. It was first described by Keck and Lam in 1962. ­Interestingly, some contemporaries of Lam believed ­ tarsal tunnel Flexor digitorum longus

s­ yndrome to occur exclusively in jockeys. Currently there is no associated occupational risk ­attributed to the development of tarsal tunnel syndrome.1,2 It may involve only one of the terminal branches distal to the tarsal canal. The tarsal canal is located behind the medial malleolus and becomes the tarsal tunnel as the flexor retinaculum passes over the structures, creating a closed compartment (Fig. 25G-1). Distally, at about the level of the medial malleolus, the posterior tibial nerve divides into its terminal branches, giving rise to the medial ­plantar nerve, the lateral plantar nerve, and the medial calcaneal branches. Diagnosis of the tarsal tunnel syndrome is a specific one that is made by correlating the patient’s history, the physical findings, and electrodiagnostic studies.3 If the findings on all three of these studies are not positive, the possibility of an entity other than the tarsal tunnel syndrome should be considered carefully.

Clinical Symptoms Most patients with tarsal tunnel syndrome complain of a poorly defined burning, tingling, or numb feeling on the plantar aspect of the foot (Fig. 25G-2). At times, this ­pattern of pain may be localized to one of the three terminal branches of the posterior tibial nerve rather than the entire nerve. Generally, the pain is aggravated by activity and relieved by rest, but sometimes patients note that the symptoms are most bothersome in bed at night; this type of pain can be relieved by getting up, moving around, and massaging the foot. In our experience, younger patients who present with this problem complain of activity-related symptoms.

Tarsal tunnel Posterior tibial art. Tibialis posterior Posterior tibial n. Calcaneal branches Nerve to abductor digiti quinti m.

Figure 25G-1   Tarsal tunnel.

Figure 25G-2   Topographic areas of pain based on diagnosis.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2058 DeLee & Drez’s� O

Etiology The cause of tarsal tunnel syndrome is idiopathic in about 50% of cases, and in the other 50%, a specific cause can be identified. The most commonly identified cause is a spaceoccupying lesion such as a synovial cyst,4 ganglion protruding from a tendon sheath,5,6 lipoma,3 neurilemoma,7 venous varicosities,3,8 tenosynovitis,9 severe pronation or valgus hindfoot deformity,5,10 trauma resulting in a fracture of the distal tibia or calcaneus,3 or occasionally a severe ankle sprain.3

Diagnosis Physical Examination Examination of the patient with a suspected tarsal tunnel syndrome is important. Initially, the patient is examined in the standing position to evaluate the overall posture of the foot and the presence of edema, venous varicosities, or thickening around the medial aspect of the ankle or heel. The range of motion of the ankle and of the subtalar and transverse tarsal joints is evaluated. The physician then gently taps along the course of the posterior tibial nerve starting above the malleolus and passing distally below the malleolus along each terminal branch of the nerve. As this area is percussed, an attempt is made to elicit tingling along the course of the posterior tibial nerve or its terminal branches to identify the site of possible pathology. Careful palpation along the tendon sheaths is important because at times a cyst or ganglion may arise, placing pressure on the posterior tibial nerve or one of its terminal branches. A sensory examination on the plantar aspect of the foot is made, and the motor function to the toes is evaluated.

Electrodiagnostic Studies In the patient in whom tarsal tunnel syndrome is suspected, electrodiagnostic studies should be carried out.3,11 The studies should include conduction velocities of the peroneal nerve to rule out the possibility of a peripheral neuropathy. The terminal latencies of the medial plantar nerve to the abductor hallucis and the lateral plantar nerve to the abductor digiti quinti are determined. As a general rule, the terminal latency of the medial plantar nerve to the abductor hallucis should be less than 6.2 msec, and that of the lateral plantar nerve to the abductor digiti quinti should be less than 7 msec. The presence of fibrillation in the abductor hallucis and the abductor digiti quinti is a late sign, and its absence does not necessarily rule out tarsal tunnel syndrome. Variation tends to occur between electromyographers, and the terminal latencies should be compared with the standard of the electromyographer carrying out the studies.12,13 Evaluation of the sensory components of the posterior tibial nerve likewise may be of benefit.14,15 There are conflicting reports among electromyographers about precisely which is the most important examination in determining the presence of an abnormality.

Differential Diagnosis Box 25G-1 lists remote, intraneural, and extraneural causes.

Box 25G-1 Differential Diagnosis of Tarsal ­Tunnel Syndrome Remote Causes

• Interdigital neuroma • Intervertebral disc degeneration • Plantar fasciitis Intraneural Causes

• Peripheral neuritis • Peripheral vascular disease • Diabetic neuropathy • Leprosy • Neurilemoma Extraneural Causes

• Ganglion • Nerve tethering • Blunt trauma • Valgus of the hindfoot • Rheumatoid arthritis with tenosynovitis • Venous varicosities • Hypertrophy of the abductor hallucis muscle origin • Lipoma • Bony impingement secondary to a previous fracture Treatment Conservative Management Conservative management may be of benefit, particularly if the cause is thought to be postural, in which case an orthotic device may be used. If there is significant swelling of the extremity, this should be brought under control. The use of nonsteroidal anti-inflammatory medication or, occasionally, an injection of a steroid preparation into the area of the tarsal tunnel may be beneficial. At times, immobilization in a short leg cast and the use of a polypropylene ankle-foot orthosis may bring relief. If the problem is thought to be due to mechanical pressure on the nerve, usually conservative measures are not beneficial.

Surgical Management Decompression of the posterior tibial nerve and its terminal branches should be undertaken through a longitudinal incision, which starts about 20 cm proximal to the medial malleolus and 3 to 4 cm behind the posterior border of the tibia and extends distally behind the medial malleolus and along the medial aspect of the foot. The direction of the incision below the malleolus depends on the area that needs to be explored. If most of the symptom complex involves the medial plantar nerve, the incision is brought along the dorsal margin of the abductor hallucis muscle, or if the symptoms appear to involve mainly the lateral ­plantar nerve, the incision is carried slightly more plantarward along the origin of the abductor hallucis muscle. The incision is developed down to the deep fascia of the abductor. Care is taken to identify and cauterize all bleeders. The investing retinaculum is opened behind the posterior tibial tendon sheath, and usually the posterior tibial nerve lies just posterior to the flexor digitorum longus

Foot and Ankle 2059

out, if any ganglion or cystlike structure appears to be arising from a tendon sheath, it must be identified and excised. This structure could be the main cause of the patient’s symptoms and should be looked for carefully, particularly if there is a localized Tinel’s sign preoperatively over one of the terminal branches below the level of the malleolus. After the dissection has been completed, the tourniquet is released, and all bleeding is brought under control. The wound is closed in a routine manner, and a compression dressing is applied. Postoperatively, the patient is kept non– weight-bearing for 3 weeks, after which progressive weightbearing is begun as tolerated. Swelling should be controlled with an elastic stocking. Impact-type ­ activities are not allowed for at least 2 months and then only as ­tolerated.

Clinical Results Figure 25G-3   Lateral plantar nerve.

t­ endon. Occasionally, the flexor digitorum longus tendon is in its own tendon sheath; then one must open up the next tendon sheath posteriorly to identify the posterior tibial nerve. The nerve always is identified proximally, which makes the dissection easier as one proceeds distally toward the medial malleolus. As one reaches the area of the medial malleolus, the vascular leash, which accompanies the posterior tibial nerve, crosses over the nerve, making the dissection more difficult. At times, it is almost easier to identify the medial plantar nerve distal to the malleolus and work in a retrograde fashion, allowing identification of the medial plantar nerve along its entire course. One follows the medial plantar nerve distally past the talonavicular joint, after which it passes through a fibroosseous tunnel onto the plantar aspect of the foot. It is important that a clamp be passed through this tunnel to ensure that no entrapment is present. If entrapment does appear to be present, the tunnel should be opened. Returning proximally, the lateral plantar nerve carefully is identified posterior to the medial plantar nerve (Fig. 25G-3). Generally, the medial calcaneal branches come off the posterior aspect of the lateral plantar nerve at varying heights, usually starting at about the level of the medial malleolus, but at times they may branch off more proximally. Because of this, the dissection should be carried out along the anterior aspect of the lateral plantar nerve to avoid injury to the medial calcaneal branches. The lateral plantar nerve is traced distally behind the vascular leash, after which it passes down beneath the origin of the abductor hallucis. At times, it is difficult to dissect out the lateral plantar nerve, but in patients in whom there is a positive Tinel’s sign over the area of the nerve, it is imperative to take down the abductor origin sufficiently to allow the entire course of the lateral plantar nerve to be identified. Behind the abductor hallucis, the lateral plantar nerve passes through several firm fibro-osseous ­ tunnels confluent with the deep fascia of the abductor hallucis muscle, each of which needs to be identified by blunt ­dissection. If there is any evidence of entrapment, the tunnel needs to be released. The medial calcaneal branch should be identified as it passes along the posterior aspect of the lateral plantar nerve, but it does not need to be dissected out unless the symptoms point to entrapment. As this dissection is carried

The results after release of the tarsal tunnel are variable. The best results occur in patients who have a mass-­occupying lesion. In this group, a satisfaction rate of about 90% can be achieved. The remainder have about a 70% satisfaction rate. Many patients note that the pain or dysesthesias still are present but at a significantly reduced level. The success of revision tarsal tunnel surgery in the case of recurrent symptoms largely depends on the adequacy of the original release. Skalley and colleagues reviewed 12 patients (13 feet) who had failed initial tarsal tunnel release.6 Three groups of patients emerged from their analysis. First, patients who failed the first procedure because of entrapment of the tibial nerve in scar but had an adequate release did not do well after their revision surgery. Individuals with scar entrapment and an inadequate release had mixed results. Finally, patients whose symptoms were purely due to an inadequate release did well after revision surgery. The overall picture for revision tarsal tunnel surgery is bleak, with only a small percentage of patients finding any real relief of their symptoms.6

ENTRAPMENT OF THE MOTOR BRANCH TO THE ABDUCTOR DIGITI QUINTI Entrapment of the nerve to the abductor digiti quinti has been identified as a source of heel pain.10,16-19 Although it is not a common cause of heel pain, this nerve entrapment is something that should be sought carefully, particularly in the patient who presents with refractory pain around the heel.

Clinical Symptoms Patients with this entrapment neuropathy complain of a vague burning pain around the heel pad area. The pain usually is poorly localized and rarely radiates out to the forepart of the foot or proximally in the heel. Generally, the symptoms are aggravated by activity and relieved by rest.

Etiology The cause is believed to be the result of entrapment of the nerve to the abductor digiti quinti muscle (Fig. 25G-4), which usually arises as the first branch of the lateral plantar nerve. The entrapment occurs as this nerve passes beneath

�rthopaedic ����������� S �ports ������ � Medicine ������� 2060 DeLee & Drez’s� O

the plantar aspect of the deep investing fascia of the abductor hallucis muscle as it courses laterally across the foot. The nerve, after passing the abductor hallucis, passes along the calcaneus, lying between the long plantar ligament, which is dorsal to it, and the plantar fascia, which is plantar to it. It lies adjacent to the plantar tuberosity of the calcaneus. Continuing laterally, the nerve terminates in the muscle mass of the abductor digiti quinti. The entrapment of the nerve usually occurs beneath the abductor fascia or, owing to its proximity to the calcaneal spur, secondary to inflammation, trauma, or healing of a stress fracture.

Nerve to abductor digiti quinti m. Abductor hallucis m.

Abductor hallucis deep fascia

Diagnosis Abductor digiti quinti m.

Physical Examination Pressure over the medial and slightly plantar aspect of the abductor hallucis muscle reproduces the patient’s pain. Occasionally, a positive Tinel’s sign can be elicited in the area. The patient with this condition usually needs to be evaluated on several occasions to be sure that the ­symptom complex is reproducible. Generally, electrodiagnostic studies are not useful in defining this problem.

Differential Diagnosis The differential diagnosis in patients with this entrapment includes: • Plantar fasciitis • Fasciitis or tendinitis of the origin of the abductor hallucis muscle • Periostitis • Stress fracture of the calcaneus • Tarsal tunnel syndrome • Systemic arthritides • Mechanical foot problem

Treatment Conservative Management Conservative management consists of the use of nonsteroidal anti-inflammatory medications, local steroid injection, an orthotic device to attempt to keep the stress off the involved area, and occasionally, immobilization of the foot in a short leg cast.

Surgical Treatment In patients who fail to respond to conservative management, operative release can be considered.16 The release is carried out through a 5-cm incision made along the upper border of the heel pad, with care taken to preserve the sensory branches to the calcaneus. The investing fascia over the abductor hallucis muscle is split, and the muscle is retracted dorsally. The deep layer of the ­investing ­fascia is released to observe the lateral plantar nerve. The branch to the abductor digiti quinti can be identified as it passes laterally. The nerve is traced laterally, and a hemostat is placed along the course of the nerve to ensure that an entrapment is not present as it passes adjacent to the calcaneal ­tuberosity. If

Figure 25G-4   Nerve to the abductor digiti quinti.

there appears to be any evidence of pressure against the nerve from the medial half of the plantar fascia or the calcaneal tuberosity, the former is released, and the latter is excised. Postoperatively, a compression dressing is used, and weight-bearing is begun when it can be tolerated.

Clinical Results Baxter and Thigpen16 reported on 34 procedures in 26 patients, of which good results were obtained in 32 cases. Poor results were reported in 2 cases. Although satisfactory results have been obtained with this procedure, it is imperative to make a precise diagnosis; otherwise, satisfactory resolution of the problem cannot occur. There are many causes of heel pain, and entrapment of the nerve to the abductor digiti quinti is a cause of heel pain in only a few isolated instances.

SURAL NERVE INJURY An entrapment of the sural nerve, which is a continuation of the tibial nerve formed by the communicating branch of the common peroneal nerve and the medial sural cutaneous nerve, is not common. The sural nerve is accompanied by the lesser saphenous vein behind the lateral malleolus, where the nerve usually divides into anterior and posterior branches. The nerve supplies sensation to the back of the leg, the lateral aspect of the heel, the lateral border of the foot, the fifth toe, and sometimes the lateral portion of the fourth toe (Fig. 25G-5).20 A primary entrapment of this nerve is unusual because it is surrounded by adequate fatty soft tissue throughout its course and does not pass through any structures that result in constriction. It is damaged most frequently secondary to extrinsic factors, such as trauma or surgery.21-24

Clinical Symptoms Patients usually complain of a burning pain or numbness along the distribution of the sural nerve or its terminal branches. At times, the symptoms are brought about by wearing a shoe or boot that places pressure against the involved area.

Foot and Ankle 2061

Sural nerve

Deep peroneal n. Extensor retinaculum

Figure 25G-5   Anatomy of the sural nerve.

Etiology Although external pressure might be applied to the nerve, symptoms are unusual because even if a mass were present, the nerve is surrounded by so much soft tissue that sufficient pressure rarely can be applied to the nerve to bring about an entrapment. After a crush injury, however, the nerve can become entrapped in fibrous tissue. After ­surgical procedures about the malleolus or hindfoot area, the sural nerve can be entrapped either proximal to the malleolus or distally, after it divides into two branches.

Diagnosis Physical Examination The physical findings consist of tenderness along the course of the sural nerve along with tingling when the nerve is percussed in the involved area. One should start by examining the nerve proximal to the malleolus, then proceed down behind the malleolus and along its two terminal branches. Numbness can be detected distal to the area of the entrapment, or occasionally dysesthesias may be present if there is a neuroma in situ. Electrodiagnostic studies are not necessary to confirm the diagnosis of a sural nerve entrapment.

Treatment Conservative Management Conservative management consists in attempting to relieve the pressure on the involved area by padding or ­changing the type of boot or shoe that is worn. ­ Occasionally, a ­steroid injection into the area of maximal tenderness may be beneficial.

Surgical Treatment If a specific neuroma is identified, the area can be explored surgically, resecting the nerve back into an area of adequate soft tissue and attempting to bury it beneath the structures so that when the traumatic neuroma forms again, it will be, ideally, in an area that will not be subjected to pressure. Sometimes, however, even after successful excision of the neuroma, it may reform and cause the patient difficulty. When the neuroma occurs above the level of the malleolus, the nerve should be traced proximally, resected, and buried in the anterior aspect of the gastrocnemius-soleus muscle so that it is not subjected to pressure from boots.

Figure 25G-6   Deep peroneal nerve entrapment beneath the retinaculum.

Postoperatively, the foot is placed in a compression dressing until the soft tissue heals and then weight-­bearing is permitted as tolerated. Swelling should be controlled with an elastic stocking.

Clinical Results Generally, resection of the neuroma results in satisfactory resolution of the problem. If a recurrent neuroma develops, a second operative procedure may be necessary.

DEEP PERONEAL NERVE ENTRAPMENT (ANTERIOR TARSAL TUNNEL SYNDROME) The deep peroneal nerve may become entrapped at several sites as it passes distally (Fig. 25G-6). As it passes over the ankle joint beneath the inferior extensor retinaculum just lateral to the extensor hallucis longus tendon, it divides into medial and lateral branches. The lateral branch innervates the extensor digitorum brevis muscle and the surrounding joints. The medial branch passes with the dorsalis pedis artery and distally supplies the web space between the first and second toes. Compression against the nerve may occur at the ankle, as it passes over the talonavicular joint, and at the tarsometatarsal joints.

Clinical Symptoms The patient usually complains of a vague pain over the dorsomedial aspect of the foot with occasional radiation into the first web space. At times, the main complaint is numbness within this area. The symptom complex often is aggravated by activities but in some cases is more bothersome when the patient is in bed at night. Examination may reveal sensory changes in the first web space as well as weakness of the extensor digitorum ­brevis

�rthopaedic ����������� S �ports ������ � Medicine ������� 2062 DeLee & Drez’s� O

muscle. In about 22% of limbs, the extensor digitorum brevis muscle is innervated by the accessory deep peroneal nerve, not the deep peroneal nerve.25,26 Percussion along the course of the nerve is important, and usually tingling is noted at the level of the entrapment. If a bone ridge is encountered and is thought to be the cause of the patient’s problem, radiographic evaluation is essential to define the size of the exostosis.

Diagnosis Electromyographic Studies Electrodiagnostic studies can be useful. These studies show an increase in the distal motor latency of the deep peroneal nerve with normal conduction velocities proximal to the ankle.27,28 There are, however, no specific criteria for nerve latencies established for this location. There may be signs of chronic denervation in the extensor digitorum brevis muscle.

Differential Diagnosis An L5 radiculopathy, as well as a neuropathy of the common or superficial peroneal nerve, may mimic entrapment of the deep peroneal nerve. The electrodiagnostic studies should differentiate these entities.

Treatment Conservative Management If the entrapment is over a bony prominence at the level of the tarsometatarsal joints or at the level of the talonavicular joint, padding the area or wearing a looser shoe may permit resolution of the symptoms. Occasionally, a steroid injection may be of some benefit. If the impingement is mainly mechanical in nature, surgery is usually necessary.

SUPERFICIAL PERONEAL NERVE ENTRAPMENT Entrapment of the superficial branch of the common peroneal nerve may occur where it exits through the sharp ­fascial edges in the distal anterolateral aspect of the leg about 10 to 15 cm proximal to the lateral malleolus (Figs. 25G-7 and 25G-8).3,29-31 After exiting through the fascia, the nerve divides into its two terminal branches, forming the medial dorsal cutaneous and the intermediate dorsal cutaneous nerves. These nerves provide sensation to the ­dorsum of the foot, with the exception of the first web space, which is innervated by the superficial portion of the deep peroneal nerve.

Clinical Symptoms The patient usually complains of numbness and tingling over the dorsal aspect of the foot and ankle, but the first web space is not involved. The symptoms may be ­stimulated by sporting activities, but the condition occurs in the nonathlete as well. If the entrapment is due to a fascial defect with resultant muscle herniation, the patient may note a fullness or swelling over the anterolateral aspect of the leg where the nerve exits through the fascia.

Etiology The cause frequently is undetermined, but sometimes the condition is caused by a fascial defect with secondary ­muscle herniation, which results in tenting of the superficial peroneal nerve.32 At times, compression from an ­external source, such as a lipoma or occasionally a bony callus resulting from a tibial fracture, may place pressure on the nerve.29,31,32

Superficial peroneal n.

Surgical Management Surgical release of the nerve depends on the site of the entrapment. If the symptoms are located at the level of the extensor retinaculum, this area can be released ­easily through a small anterior incision. If the ­ entrapment is secondary to bony spurs over the anterior aspect of the ankle or the talonavicular joint or at the level of the ­ tarsometatarsal joints, the nerve should be exposed ­carefully over the involved area; the exostosis is then exposed and excised carefully. Postoperatively, the patient is placed in a compression dressing with weight-bearing as tolerated and is permitted to return to activities as ­tolerated.

Clinical Results Generally, release of an entrapment of the deep peroneal nerve is successful. Caution must be exercised, however, because the nerve can be damaged while trying to release it. Occasionally, the exostosis may re-form, resulting in further difficulty for the patient.

Figure 25G-7   Entrapment of the superficial peroneal nerve as it exits from the deep fascia of the leg.

Foot and Ankle 2063

Several studies have demonstrated nerve conduction abnormalities of the distal peroneal nerve, superficial peroneal nerve, and common peroneal nerve after inversion injury of the ankle.33 A recent biomechanical study showed the magnitude of shear force generated across the superficial peroneal nerve during an inversion injury to be within the range necessary to structurally alter the nerve.34

Diagnosis Physical Findings The physical examination usually brings out the cause of the problem. Tenderness or irritability of the nerve to percussion is evident where it exits through the fascial defect. A muscle herniation or an underlying lipoma or callus usually can be palpated. Frequently, there is some sensory loss over the distribution of the nerve around the anterior aspect of the foot, sparing the first web space. Resisted dorsiflexion of the ankle joint may show the muscle herniation through the fascial defect.

Electromyographic Studies Electrodiagnostic studies are rarely of any benefit in the diagnosis of superficial peroneal nerve injury or ­entrapment.

Differential Diagnosis Superficial peroneal nerve entrapment may mimic a common peroneal nerve entrapment or possibly an L5 radiculopathy. The symptoms of a chronic ankle sprain or instability can mimic this condition, but with a careful history and examination, these usually can be differentiated.

Treatment Conservative Management Occasionally, a steroid injection or a diminished level of activity may be beneficial. If symptoms persist, however, and particularly if a muscle herniation or mass is present beneath the nerve, surgical release is indicated.

Surgical Treatment The site of the entrapment is identified through a longitudinal incision, and the fascia over the anterior compartment is released. It is important to carry out an adequate release so that a muscle herniation does not result from the surgical decompression. If a mass is present, such as a lipoma or callus, this should be excised. Postoperatively, a compression dressing is used until the soft tissues heal, at which time activities can be resumed as tolerated.

Clinical Results Generally, surgical decompression of the nerve results in satisfactory relief for the patient. One must be cautious, however, not to damage the nerve inadvertently when the decompression is carried out.

C

r i t i c a l

P

o i n t s

l The most common error in surgical release of the tarsal tunnel is to make too short an incision and inadequately release the nerves distally as they pass underneath the abductor fascia. l Surgical decompression is dramatically more reliable for tarsal tunnel syndrome due to mass-occupying lesions in the hindfoot than that due to idiopathic causes. l Entrapment of the nerve to the abductor digiti quinti is an important component of the differential diagnosis of heel pain and may coexist with plantar fasciitis. l Sural nerve injury due to extrinsic causes or iatrogenic surgical exposure is among the most common nerve problems encountered in sports medicine practice. Refractory cases are best dealt with by proximal resection and burial rather than neurolysis. l Traction neuritis of the superficial peroneal nerve is an underappreciated source of additional pain in cases of acute or chronic ankle instability. Surgical release can be performed, but the neuritis almost always resolves following surgical stabilization of the ankle.

S U G G E S T E D

R E A D I N G S

Baxter DE, Thigpen CM: Heel pain: Operative results. Foot Ankle 5:16, 1984. Gould N, Alvarez R: Bilateral tarsal tunnel syndrome caused by varicosities. Foot Ankle 3:290-292, 1983. Jones JR, Klenerman L: A study of the communicating branch between the medial and lateral plantar nerves. Foot Ankle 4:313, 1984. Kaplan PE, Kernahan WT: Tarsal tunnel syndrome: An electrodiagnostic and surgical correlation. J Bone Joint Surg Am 63:96-99, 1981. Keck C: The tarsal tunnel syndrome. J Bone Joint Surg Am 44:180-182, 1962. Mann RA, Reynolds JC: Interdigital neuroma: A critical clinical analysis. Foot Ankle 3:238-243, 1983. O’Neill PJ, Parks BG, Walsh R, Simmons LM, Miller SD. Excursion and strain of the superficial peroneal nerve during inversion ankle sprain. J Bone Joint Surg Am. 2007 May;89(5):979-86.

Figure 25G-8   Cadaveric specimen: superficial peroneal nerve exiting hiatus.

R eferences Please see www.expertconsult.com

�rthopaedic ����������� S �ports ������ � Medicine ������� 2064 DeLee & Drez’s� O

S e c t i o n

H

Conditions of the Forefoot Sandra E. Klein

HALLUX VALGUS Hallux valgus is a static subluxation of the first metatar­ sophalangeal (MTP) joint characterized by lateral or val­ gus deviation of the great toe and medial or varus deviation of the first metatarsal.1 Occasionally, pronation or rotation of the hallux occurs with severe deformity. A hallux valgus deformity can also occur with lateral ­deviation of the distal articular surface of the first metatarsal. This anatomic vari­ ation is seen most commonly in juvenile hallux valgus.2-4 Hallux valgus is found almost exclusively in shoe­wearing societies, although it has been noted in primitive societies that do not wear shoes.5,6 A higher incidence in females has also been reported in the literature.7-9 Many factors other than shoewear have been investigated for a causal or contributing relationship to hallux valgus defor­ mity, including genetic predisposition,2,9-11 pes planus,9,12 first metatarsocuneiform hypermobility, ligamentous ­laxity,5,13-15 and Achilles contracture.5,13,16,17 In the athletic population, hyperextension injuries or acute dislocation of the first MTP joint may predispose to the development of a hallux valgus deformity. A symptomatic hallux valgus deformity in a highperformance athlete presents a special problem for the treat­ ing physician. The increased stress manifested by running or jogging can affect the function of the foot significantly.18

The increased force generated in the forefoot can approach 250% of body weight with running compared with 80% with walking. Lillich and Baxter18 noted an increased range of motion of the joints of the lower extremity and an altera­ tion in phasic activity of the muscles of the lower extrem­ ity with running activity. These factors must be considered when evaluating a hallux valgus deformity in an athlete.19

Anatomy and Biomechanics The MTP joint of the hallux is differentiated from the lesser toes by the sesamoid mechanism as well as by ­specific intrinsic muscles that stabilize the great toe and provide motor strength.20 The first metatarsal head has a rounded cartilage-covered surface that articulates with the concave base of the proximal phalanx. There is significant ­variation in the joint surfaces of the first MTP articulation. A rounded MTP articulation is more prone to the develop­ ment of hallux valgus; a flattened or chevron-shaped MTP articulation is a stable configuration and is less prone to progressive development of hallux valgus (Fig. 25H-1).1 The first MTP joint is stabilized on the plantar medial and plantar lateral aspects by the collateral ligaments and sesamoid ligaments (Fig. 25H-2A). The fan-shaped ­collateral ligaments originate from the medial and lat­ eral epicondyles and course in a distal-plantar direction,

Figure 25H-1   A, A rounded metatarsal head is at risk for subluxation of the metatarsophalangeal joint. B, A flat metatarsophalangeal articulation is more resistant to subluxation. (© M. J. Coughlin. Used by permission.)

A

B

Foot and Ankle 2065

Sesamoid ligament

Collateral ligament

A

Extensor hallucis longus

Extensor hallucis brevis Hood ligament Collateral ligament

B

Sesamoid ligament

Figure 25H-2   A, The collateral ligaments and the sesamoid ligaments of the first metatarsophalangeal joint. B, The extensor hallucis longus inserts into the hood ligament.

inserting on the medial and lateral aspect of the base of the proximal phalanx. The collateral ligaments interdigi­ tate with the sesamoidal ligaments, which originate in the epicondylar region of the metatarsal head and insert into the margins of each sesamoid as well as the plantar plate. Although the plantar medial and plantar lateral aspects of the MTP joint are stabilized by a thick capsular structure, the dorsal aspect is supported by the relatively thin exten­ sor hood (see Fig. 25H-2B). With pronation of the foot

or with ­ progressive subluxation of the first MTP joint, increased pressure is placed on the weaker dorsal medial capsule, which becomes stretched and attenuated with time (Fig. 25H-3). The intrinsic musculature of the first ray is composed of several distinct tendinous structures. The two tendons of the flexor hallucis brevis, located directly plantarward, insert into the medial and lateral sesamoids. The sesamoids articulate on their dorsal surfaces with medial and lateral facets on the plantar aspect of the first metatarsal head (Fig. 25H-4A). The intersesamoidal ridge (the crista) (see Fig. 25H-4B) separates these facets and affords stability to the metatarsal-sesamoid articulation. Distally, the sesamoids are attached to the fibrous plantar plate, which inserts onto the plantar base of the proximal phalanx. The abductor hal­ lucis, located on the plantar medial aspect, inserts onto the plantar medial base of the proximal phalanx. The adductor hallucis, located on the plantar lateral aspect, inserts onto the plantar lateral base of the proximal phalanx. There are no tendinous insertions onto the ­metatarsal head; the control of the metatarsal is maintained by the capsuloliga­ mentous complex and the intrinsic muscles that insert onto the plantar half of the proximal phalanx. As a hallux valgus deformity progresses, the ligaments and intrinsic muscula­ ture of the first ray play a significant role in the orientation of the MTP joint. The extensor hallucis longus, located on the dorsal aspect of the hallux, is stabilized medially and laterally by the extensor hood (see Fig. 25H-2B); it inserts onto the dorsal base of the distal phalanx, whereas the extensor hal­ lucis brevis inserts onto the base of the proximal phalanx. The tendon of the flexor hallucis longus, contained within a fibrous tendon sheath on the plantar aspect of the sesa­ moid complex, inserts onto the plantar base of the distal phalanx.

Extensor hallucis longus

Extensor hallucis longus

Medial hood ligament Adductor hallucis Lateral sesamoid

A

Attenuation of medial capsule

C

Abductor hallucis

Adductor hallucis

Medial sesamoid Flexor hallucis longus

B

Abductor hallucis Lateral Medial sesamoid sesamoid Flexor hallucis longus

Figure 25H-3   A-C, With subluxation of the metatarsophalangeal joint, the dorsal capsule and the hood ligament become attenuated.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2066 DeLee & Drez’s� O

Crista

Adductor hallucis

Abductor hallucis

Adductor hallucis

Sesamoids Abductor hallucis

A

Flexor hallucis brevls

B

Intersesamoidal ligament

Flexor hallucis brevis

Figure 25H-4   A, The sesamoids are contained within the double tendon of the flexor hallucis brevis. B, Cross section through the metatarsophalangeal joint shows relationship of sesamoids and extrinsic muscles.

Specific angular measurements on a weight-­bearing anteroposterior radiograph quantify the characteristics of a hallux valgus deformity: the hallux valgus angle, 1-2 intermetatarsal angle, hallux interphalangeal angle, and distal metatarsal articular angle (Table 25H-1; Fig. 25H-5).4,10,16,21-23 The first metatarsocuneiform joint plays an integral role in the stability of the first metatarsal. Increased obliquity of the first metatarsocuneiform joint increases the magnitude of the metatarsus primus varus (Fig. 25H-6), although rou­ tine radiographic imaging of this joint is often imprecise. A horizontal orientation of the metatarsocuneiform joint appears to be a stable configuration and resists deformity, whereas a curved metatarsocuneiform joint appears to be more flexible, allowing medial metatarsal deviation as the hallux valgus deformity increases.1 Congruency of the first MTP joint is also assessed. In a congruent joint (Fig. 25H-7A), the center of the metatarsal

articular surface corresponds to the center of the articu­ lar surface of the proximal phalanx. In contrast, with an incongruent articulation, there is subluxation of the MTP joint, that is, the base of the proximal phalanx is deviated ­laterally in relation to the articular surface of the first metatarsal, resulting in an uncovering of the metatarsal head (see Fig. 25H-7B).3 Piggott,24 in a study of congru­ ous and incongruous first MTP joints with hallux valgus, found that a ­congruous joint (with or without an increased distal ­metatarsal articular angle) was a stable articulation. A hallux valgus deformity in this situation does not increase with time. With an incongruous MTP joint, there is a ­significant risk for further progression of the deformity.

Valgus <15°

  TABLE 25H-1  Angular Measurements in Hallux Valgus Deformity

Measured Angle

Radiographic ­Parameters

Normal Values

1,2 Intermetatarsal angle10,16,21

Angle between the axes of the first and second metatarsals (see Fig. 25H-5) Angle between the axes of the first metatarsal and the proximal phalanx (see Fig. 25H-5) Angle between the axis of the proximal phalanx and an axis drawn through the center of the base of the distal phalanx and the tip of the distal phalanx Angle between the axis of the first metatarsal and a perpendicular to a line defining the lateral slope of the distal metatarsal articular surface

Less than 9 degrees

Hallux valgus angle10,16,21 Hallux interphalangeal angle22

Distal metatarsal articular angle4,23

Less than 15 degrees Less than 10 degrees

Less than 6 degrees of lateral inclination

1–2 intermetatarsal <9° Figure 25H-5   The 1-2 intermetatarsal angle is measured by lines bisecting the axis of the first and second metatarsals. The hallux valgus angle is the angle subtended by the axes of the proximal phalanx and the first metatarsal.

Foot and Ankle 2067

A

Straight metatarsal cuneiform joint

B

Curved metatarsal cuneiform joint

C

Obilque metatarsal cuneiform joint

Figure 25H-6   Metatarsocuneiform orientation. A, A straight orientation of the metatarsocuneiform joint is stable. B, A curved metatarsocuneiform articulation is more flexible and may be at risk for deviation with an increased intermetatarsal angle. C, An oblique orientation of the metatarsocuneiform joint is often resistant to correction. An osteotomy may be necessary in this situation.

The dynamic development of a hallux valgus deformity is characterized by changes at the MTP, metatarsocunei­ form, and metatarsosesamoid articulations. As hallux valgus develops, the proximal phalanx is displaced in a valgus or lat­ eral direction and often is pronated on the head of the first metatarsal.25 The insertion of the adductor hallucis plays an integral role in the subsequent deformity. The adduc­ tor hallucis insertion onto the lateral sesamoid anchors the sesamoid complex to the lesser metatarsals so that they can­ not drift medially with the first metatarsal (Fig. 25H-8). As the hallux is forced in a lateral direction by extrinsic forces, the base of the proximal phalanx pushes the metatarsal head

medially and attenuates the medial capsule. The insertion of the adductor hallucis onto the plantar lateral base of the proximal phalanx and lateral sesamoid exerts not only a lat­ eral force on these structures but also a rotational force on the proximal phalanx. As the metatarsal drifts medially, the proximal phalanx rotates along the longitudinal axis of the adductor hallucis, the pivot point being the point of inser­ tion of the adductor. As the great toe displaces laterally, the intrinsic plantar cuff (consisting of the abductor hallucis, the flexor hallucis brevis, and the adductor hallucis) (Fig. 25H-9) rotates in relation to the metatarsal head, exposing the thin dorsal medial capsule to further deforming forces. On the plantar aspect, the sesamoid complex displaces in relation to the first metatarsal head. The sesamoid mecha­ nism, owing to the adductor hallucis insertion, retains its relationship to the lesser metatarsals, whereas the first meta­ tarsal literally slides off the sesamoid complex. Although the term sesamoid subluxation is used to describe this ­progressive

Lateral sesamoid

A

B

Figure 25H-7   A, The metatarsal articular orientation. The distal metatarsal articular angle is deviated in a lateral direction. B, A subluxated metatarsophalangeal joint. In this noncongruent joint, the articular surface of the phalanx subluxates laterally on the metatarsal articular surface.

Conjoined adductor tendon

Figure 25H-8   The adductor hallucis anchors the lateral sesamoid and the proximal phalanx. When subluxation of the metatarsophalangeal joint occurs, the toe may rotate or pronate owing to insertion of the conjoined adductor tendon.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2068 DeLee & Drez’s� O EHB

EHB

Abductor hallucls

FHB FHB

Adductor hallucis

Abductor hallucis

Adductor hallucis

A

FHB FHB

B

Figure 25H-9   A, Schematic representation of the intrinsic muscles surrounding the first metatarsal head. Normal articulation.   B, With valgus deviation, the intrinsic muscles rotate. The abductor hallucis assumes a more plantar position, and the adductor hallucis assumes a more lateral orientation. EHB, extensor hallucis brevis; FHB, flexor hallucis brevis.

deformity, it is the first metatarsal that is displaced in rela­ tion to the sesamoids and lesser metatarsals. As this dis­ placement occurs, the intersesamoidal ridge is smoothed out gradually (Fig. 25H-10) until it affords no resistance to further deformity. In severe hallux valgus deformities, as the first metatarsal displaces medially off the sesamoid mechanism, the extensor hallucis longus displaces into the first intermetatarsal space. As a result, contraction of the extensor hallucis longus not only extends the toe but also adducts the toe. The abductor hallucis, having assumed a more plantar position in relation to the first metatarsal, loses its splinting effect on the first metatarsal head. The lateral sesamoid may come to lie on the lateral aspect of the first metatarsal head and on occasion lies vertically above the medial sesamoid. The medial sesamoid may articulate with the lateral facet of the first metatarsal. Running places higher demands on the foot than walk­ ing.18,26,27 Various athletic activities make varying demands on the foot and ankle. A sprinter may require extreme range of motion in dorsiflexion and plantar flexion at the first MTP joint, in contrast to a middle-distance or long-­distance run­ ner.18 Analysis of the particular sports avocation, training techniques, strength requirements of the first ray, and range of motion needed by the first MTP joint plays an important role in the assessment of the functional biomechanics of the forefoot. This careful evaluation helps the practitioner to develop a plan for the symptomatic ­athlete. Although much attention has been paid to the bun­ ion or medial eminence, it is most often the increased 1-2 intermetatarsal angle coupled with lateral deviation of

the great toe that creates a prominent and painful medial border of the first metatarsal head. At times, the medial eminence may become hypertrophied and symptomatic; however, this is uncommon. With progressive subluxation, a groove or sagittal sulcus (Fig. 25H-11) develops on the first metatarsal head, delineating the articular surface from the medial eminence. Although this anatomic landmark delineates the border of the metatarsal articular surface, its location is variable and depends on the severity of the deformity. Sometimes the sagittal sulcus is located in the center of the metatarsal head. It is important not to use the sagittal sulcus as a landmark for the medial eminence resec­ tion because in severe deformities an excessive amount of metatarsal head is removed. In mild and moderate cases of hallux valgus, attention must be directed to the location of the sagittal sulcus on the anteroposterior radiograph to determine whether resection of the medial eminence should be performed through the center of the sagittal sul­ cus or medial or lateral to the center of the ­sagittal sulcus.

Classification Hallux valgus deformities are typically classified as mild, moderate, or severe.1 Classification is used as a general guideline in the decision-making process when treating a hallux valgus deformity (Box 25H-1).

Sagittal groove

Sesamoid subluxation

Proximal phalanx

Medial eminence

A

Sesamoid ridge

B

Ridge C obliterated

Figure 25H-10   A, Schematic representation shows sesamoid view with normal crista. B, Schematic representation of a patient with hallux valgus shows moderate obliteration of crista. C, Complete obliteration of crista with sesamoid subluxation.

Figure 25H-11   The sagittal sulcus is created at the medial margin of the articular surface and is an osteophyte delineating the border of the medial eminence.

Foot and Ankle 2069

Box 25H-1 C  lassification of Hallux Valgus Deformities Mild Deformity

Box 25H-2 T  ypical Findings in Hallux Valgus Deformities History

• Hallux valgus angle < 20 degrees • Hallux valgus interphalangeus

• Medial eminence pain • Metatarsophalangeal joint pain • Bursal or skin irritation • Callus formation

Moderate Deformity • Hallux valgus angle 20-40 degrees • 1-2 intermetatarsal angle 11-16 degrees • Subluxation of first metatarsophalangeal joint (unless distal metatarsal articular angle is abnormal) • Hallux may be pronated

• Medial deviation of the first metatarsal • Lateral deviation of the hallux • Pronation of the hallux

may contribute to ­ eformity d • First metatarsophalangeal joint often congruent • 1-2 intermetatarsal angle < 11 degrees

Severe Deformity

• Hallux valgus angle > 40 degrees • 1-2 intermetatarsal angle greater than 16 degrees • Subluxation of metatarsophalangeal joint • Hallux typically pronated

Evaluation Clinical Presentation and History The major subjective complaint of a patient with hallux valgus is pain over the medial eminence. At times, espe­ cially with repetitive athletic activities such as running, blistering of the skin or development of an inflamed bursa overlying the medial eminence may occur. Pressure from a shoe may cause compression of the dorsal medial ­sensory nerve to the hallux, causing either neuritic pain or numb­ ness. With a severe deformity, diminished weight-bearing capacity of the first ray may lead to lateral metatarsal­ gia with ­ development of an intractable plantar keratosis beneath the lesser metatarsal heads. Runners with hallux valgus may develop a callus on the medial border of either the great toe or the medial eminence (Box 25H-2).

Physical Examination Evaluation of the athlete with hallux valgus begins with an examination of the patient while standing. Often, defor­ mities are accentuated with weight-bearing. Assessment of postural abnormalities such as pes planus, pes cavus, and a contracted Achilles tendon; neuromuscular abnormali­ ties; and hyperelasticity associated with a collagen defi­ ciency syndrome is important. The hallux is evaluated for ­lateral deviation and pronation. Medial deviation of the first metatarsal may also be noted on physical examina­ tion. A ­ vascular evaluation should assess the presence or absence of pulses. Doppler evaluation may assist if vascular integrity is in doubt. Neurologic evaluation is important for ­ assessment of sensation, motor strength and control, and gait abnormalities. Assessment of active and passive range of motion is important in an athlete.28 Often bun­ ion surgery may be associated with diminished first MTP joint range of motion postoperatively, and quantitation of

Physical Examination

Radiographic Examination 1-2 intermetatarsal, hallux valgus, or hallux interphalangeal angles • Increased distal metatarsal articular angle • Congruent versus incongruent first metatarsophalan­ geal joint

• Increased

­ reoperative range of motion may help in developing a plan p of treatment. The mobility of the first metatarsocuneiform joint is evaluated.29,30 Skin changes such as inflammation of the medial eminence, callus formation, or intractable plantar keratoses are noted. The magnitude and rigidity of a pes planus deformity may influence preoperative or postoperative orthotic management (see Box 25H-2).

Imaging Standard radiographic views include standing anteroposte­ rior, oblique, lateral, and sesamoid views. The magnitudes of the 1-2 intermetatarsal angle and the hallux valgus angle help to quantitate the severity of the deformity. Analysis of the MTP joint for congruity, joint subluxation, and the presence of degenerative arthritis or hallux rigidus is indicated. The metatarsocuneiform joint articulation, although often difficult to assess radiographically, should be analyzed as well. The radiographs should be inspected for other lesser toe abnormalities that may be associated with the hallux valgus deformity (see Box 25H-2).

Treatment Options Nonoperative For the athletically active person, conservative care is most important. Often pain and blistering can be reduced by diminishing friction over the medial eminence. Evaluation of the athlete’s footwear is important. Modification of shoes by stretching constricting areas or relieving pressure areas may relieve an athlete’s symptoms completely. Wider foot­ wear with a roomy toe box may afford significant reduction in discomfort related to static forefoot abnormalities. The use of pads, arch supports, and various insoles may assist in maintaining an athlete’s activity. The overall philosophy of treatment of an athlete with a hallux valgus deformity should be nonoperative until pain or discomfort forces the athlete to make significant

�rthopaedic ����������� S �ports ������ � Medicine ������� 2070 DeLee & Drez’s� O

modifications in athletic activity. As Lutter stated,19 “The stiffness potential occurring after an operative procedure is potentially more harmful to a running career than pain.” If, despite conservative care and modifications of athletic activity, significant discomfort still exists, operative inter­ vention may be considered. Operations that lead to sig­ nificant alterations in the weight-bearing pattern or that are associated with significant restriction of motion post­ operatively should be avoided whenever possible. A patient should be counseled about the risks, complications, and expected outcomes. Unreasonable patient expectations about postoperative athletic activities are discussed before surgery. As Lutter noted,19 “Surgery is not contraindi­ cated, but a runner must be willing to trade relief of pain for a lower ability to run.”

Operative Many surgical repairs have been devised to correct the hal­ lux valgus deformity. The huge number of operative pro­ cedures available makes it clear that no one operation can suffice for every deformity. Various anatomic abnormali­ ties, different pathologic entities, and varying degrees of deformity make it imperative that a surgeon have several methods of hallux valgus repair within his or her surgical repertoire. The surgical procedure chosen must address the various anatomic abnormalities that may be present (Box 25H-3). The preoperative assessment of the patient is important to establish which significant pathologic com­ ponents are present and to select the most appropriate pro­ cedure for a particular bunion deformity (Fig. 25H-12). The objective of this discussion is to describe pertinent operative procedures that may be considered for the ath­ lete. Just as various operations address mild, moderate, and severe hallux valgus deformities, there are a wide variety of requirements and abilities in athletes (from walkers and joggers to high-performance sprinters, dancers, and professional athletes). A careful decision-making process is important to select the appropriate procedure for the symptomatic athlete (Box 25H-4; see Fig. 25H-12).

Akin Procedure The Akin31-34 procedure involves a medial eminence resection and medial capsular reefing combined with a medial closing wedge osteotomy of the proximal phalanx (Box 25H-5; Fig. 25H-13). A correction of an increased 1-2 intermetatarsal angle cannot be achieved with this Box 25H-3 A  natomic Factors Addressed in Surgical Treatment of Hallux Valgus Prominent medial eminence Valgus angulation of proximal phalanx Increased 1-2 intermetatarsal angle Sesamoid subluxation Pronation of great toe Increased distal metatarsal articular angle (congruent first metatarsophalangeal joint)

­ rocedure. Indications for its use include hallux valgus p interphalangeus, mild hallux valgus without significant metatarsus primus varus, and recurrent hallux valgus of a mild nature (Fig. 25H-14). In the presence of a congruent MTP joint, a proximal phalangeal osteotomy can be com­ bined with a metatarsal osteotomy without significantly altering MTP joint congruity.

Chevron Procedure The chevron osteotomy, as described by Johnson and col­ leagues35 and others,20,36 is indicated for a mild-to-moderate hallux valgus deformity with a hallux valgus angle of less than 30 degrees, an intermetatarsal angle of less than 15 degrees, and no pronation of the hallux. Because the chevron osteotomy achieves an extra-articular correction, it may be used to correct a congruous deformity as well as an incongruous or subluxated MTP joint (Box 25H-6; Fig. 25H-15). Neither the lateral capsule nor the conjoined ten­ don is released. Although release of these structures may allow a greater correction,36 Mann37 and others38,39 have reported several cases of avascular necrosis of the metatar­ sal head after a chevron procedure (Fig. 25H-16). Although several authors have not found an increased incidence of avascular necrosis after a lateral MTP release,35,36,40,41 an extensive lateral release may place a patient at greater risk for avascular necrosis of the metatarsal head.42,43

Distal Soft Tissue Realignment The distal soft tissue procedure has been used to correct mildto-moderate hallux valgus deformities.25,44,45 The description of this technique as a distal soft tissue realignment is meant to clarify the fact that this is an intra-articular repair that does not use an osteotomy to achieve realignment. Although an osteotomy may be combined with this procedure to achieve an adequate repair, a distal soft tissue repair, when indicated, may be used by itself to correct a mild-to-moderate defor­ mity (Box 25H-7; Figs. 25H-17 and 25H-18).

Proximal First Metatarsal Osteotomy A proximal metatarsal osteotomy that corrects an increased 1-2 intermetatarsal angle can be combined with a distal soft tissue reconstruction that realigns the first MTP joint. As a rule, a proximal osteotomy is indicated if the 1-2 inter­ metatarsal angle is greater than or equal to 15 degrees, if the hallux valgus angle is greater than or equal to 30 to 40 degrees, or if after a distal soft tissue realignment an ade­ quate correction of the 1-2 intermetatarsal angle cannot be obtained. Sometimes a more severe hallux valgus deformity (>40 degrees) with an increased 1-2 intermetatarsal angle can be realigned adequately without an osteotomy; some­ times a moderate hallux valgus deformity (20 to 40 degrees) cannot be corrected adequately without a first metatarsal osteotomy. Although an opening or closing wedge osteot­ omy can be performed, significant lengthening or shorten­ ing of the first ray usually is not desirable. For this reason, a proximal crescentic osteotomy frequently is used.1,46,47 This osteotomy is performed in the proximal first metatar­ sal metaphysis, an area that provides a broad, stable cancel­ lous surface that allows fairly rapid healing. A distal soft

Foot and Ankle 2071

A

Hallux valgus ‘‘mild’’

Incongruent joint (subluxation)

Congruent joint

Chevron osteotomy

Distal soft tissue realignment

Akin osteotomy

B

Chevron osteotomy

Hallux valgus ‘‘moderate’’

C

Congruent joint

Incongruent joint

Chevron osteotomy with Akin

Distal soft tissue realignment with proximal osteotomy (crescentic first metatarsal osteotomy, opening wedge first cuneiform osteotomy)

Hallux valgus ‘‘severe’’

Congruent joint

Double osteotomy 1. Akin and chevron 2. Akin and 1st metatarsal 3. Akin and 1st cuneiform opening wedge Triple osteotomy 1. Akin, distal, and proximal 1st metatarsal 2. Akin, distal metatarsal, and cuneiform

Hypermobile 1st metatarsocuneiform joint

Incongruent joint

Distal soft tissue realignment with proximal osteotomy • 1st metatarsal crescentic osteotomy • 1st metatarsal opening/ closing wedge osteotomy • 1st cuneiform opening wedge osteotomy

Regenerative arthritis

Distal soft tissue realigment and fusion 1st metatarsocuneiform joint

Keller procedure

Metatarsophalangeal fusion

Figure 25H-12   A, Algorithm for treatment of mild hallux valgus deformity. B, Algorithm for treatment of moderate hallux valgus deformity. C, Algorithm for treatment of severe hallux valgus deformity.

tissue ­realignment is typically performed first in conjunc­ tion with the proximal first metatarsal osteotomy. Intra­ operative radiographs are taken to evaluate the correction of the 1-2 intermetatarsal angle and the internal fixation. Internal fixation can be removed easily under local anesthe­ sia 6 weeks after surgery (Box 25H-8; Fig. 25H-19).

Combined Multiple First Ray Osteotomies In the presence of a congruent first MTP joint associated with a hallux valgus deformity, a distal soft tissue realign­ ment is contraindicated because it would create an incon­ gruent MTP joint.1-3 An incongruent joint is at risk for

�rthopaedic ����������� S �ports ������ � Medicine ������� 2072 DeLee & Drez’s� O

Box 25H-4 C  ommon Surgical Procedures to Correct Hallux Valgus Deformity Akin procedure • Proximal phalangeal osteotomy, medial eminence ­resection, medial soft tissue reefing Distal Soft Tissue Realignment (modified McBride ­procedure) • Medial eminence resection, lateral capsule and adductor hallucis release, medial soft tissue reefing Distal metatarsal osteotomy (chevron procedure)

• Distal metatarsal chevron osteotomy, medial eminence resection, medial soft tissue reefing

Proximal metatarsal osteotomy with distal soft tissue realignment • Medial eminence resection, lateral capsule and adductor hallucis release, medial soft tissue reefing, proximal first metatarsal osteotomy Combined multiple first ray osteotomies osteotomy, distal and proximal first ­metatarsal osteotomies, cuneiform osteotomy

• Phalangeal

Salvage procedures (not indicated as primary procedures in the athlete) • First metatarsophalangeal joint arthrodesis, Keller ­resection arthroplasty

later degenerative arthritis or recurrence of deformity. When a hallux valgus deformity occurs in a patient with a congruent first MTP joint, an extra-articular MTP joint correction with periarticular osteotomies (Akin, distal metatarsal, cuneiform) is indicated. The magnitude of the distal metatarsal articular angle determines the necessity of multiple first ray osteotomies. An Akin phalangeal osteotomy1,2,31,48 decreases phalan­ geal angulation associated with an increased proximal pha­ langeal articular angle; a first ray osteotomy (proximal first metatarsal osteotomy)12,49 or cuneiform osteotomy2,50 may correct an increased intermetatarsal angle; occasionally, an increased distal metatarsal articular angle may necessitate a medial closing wedge distal first metatarsal osteotomy.51-53 When a proximal first metatarsal osteotomy is performed in conjunction with a distal first metatarsal osteotomy, care must be taken to avoid excessive soft tissue stripping, which may devascularize the first metatarsal. Alternatively, an opening wedge cuneiform osteotomy may be performed (Fig. 25H-20). Mitchell and Baxter54 recommended a com­ bined chevron osteotomy and phalangeal osteotomy. Richardson and coworkers23 stated that the average dis­ tal metatarsal articular angle in normal feet was 6 degrees. As this angle increases, the magnitude of a congruent hal­ lux valgus deformity increases.2 Piggott24 noted that 9% of adults with a hallux valgus deformity had a congruous MTP joint. Coughlin found this to occur in 46% of juve­ niles with hallux valgus and 37% of adult men with hallux valgus.55 Coughlin and Carlson3 reported a 2% incidence of congruent hallux valgus deformities in a large series of adults requiring hallux valgus surgery.

Box 25H-5 Akin Procedure Technique

• Make

a medial longitudinal incision over the medial e­ minence. • Protect the dorsal and plantar digital nerves within the skin flap. • Create an L-shaped distally based capsular flap. • Resect the medial eminence with an oscillating saw starting toward the medial aspect of the sagittal sulcus. • Make the phalangeal osteotomy in the proximal meta­ physeal region (see Fig. 25H-13). • Cut and remove a small medially based wedge of bone. • Score the lateral cortex with the saw and close the ­osteotomy. • Stabilize the osteotomy with one or two Kirschner wires placed obliquely. • Repair the medial capsule to the surrounding metatar­ sophalangeal joint capsule and periosteum. • If necessary, place a drill hole in the medial metaphyseal cortex to suture the capsule flap to bone.

Salvage Procedures Excisional arthroplasty,56 placement of MTP joint implants, and MTP arthrodesis all are techniques that must be considered salvage procedures. Although Cleveland and Winant8 and Jordan and Brodsky57 found that the Keller procedure produced acceptable results, most authors agree that the procedure is associated with multiple complica­ tions. The decreased weight-bearing capacity of the first metatarsal,58 the increased incidence of metatarsalgia, the cock-up deformity in the hallux, and the decreased strength and loss of stability created by disruption of the plantar aponeurosis and intrinsic musculature of the first ray make it unlikely that the Keller procedure would result in useful improvement for the symptomatic athlete with hallux valgus. As a salvage procedure in older patients, this

A

B

Figure 25H-13   A, Location of osteotomy of proximal phalanx. B, After closure of osteotomy site.

Foot and Ankle 2073 Figure 25H-14   A, Preoperative radiograph shows hallux valgus interphalangeus deformity. B, After Akin phalangeal osteotomy, adequate alignment is maintained.

A

B

Box 25H-6 Chevron Procedure Technique

A

Chevron

• Make

a medial longitudinal incision over medial emi­ nence from the midportion of the proximal phalanx ­extending proximally 5 cm (see Fig. 25H-15A). • Protect the dorsal and plantar digital nerves within the skin flap. • Create an L-shaped distally based capsular flap. • Resect the medial eminence with an oscillating saw starting toward the medial aspect of the sagittal sulcus cutting in line with the medial border of the foot. • Place a horizontal drill hole in the center of the metatar­ sal head to mark the apex of the osteotomy. • Make a horizontal osteotomy from medial to lateral at an angle of approximately 60 degrees (Fig. 25H-15B). • Avoid dissection of the lateral aspect of the metatar­ sophalangeal joint to reduce the risk for avascular necrosis of the first metatarsal head (Fig. 25H-16). • Displace the osteotomy one third of the metatarsal width (Fig. 25H-15C, D). • Stabilize the osteotomy with one 0.062 Kirschner wire placed obliquely. • Resect the remaining metaphyseal flare. • Repair the medial capsule to the surrounding metatar­ sophalangeal capsule and periosteum. • If necessary, place a drill hole in the medial metaphyseal cortex to suture the capsule flap to bone.

B

C

D Chevron osteotomy

Figure 25H-15   A, Proposed chevron osteotomy of first metatarsal. B, A V-shaped osteotomy at about a 60-degree angle based proximally is centered in the metaphyseal region.   C and D, The osteotomy is displaced laterally, and the remaining medial eminence is resected.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2074 DeLee & Drez’s� O

A

B

Figure 25H-16   A, Postoperative radiograph shows avascular necrosis after a distal metatarsal osteotomy. B, Computed tomographic scan shows cystic degeneration consistent with avascular necrosis of the first metatarsal head.

Box 25H-7 D  istal Soft Tissue Realignment Technique Lateral Release

• Make a dorsal longitudinal incision in the first intermetatarsal web space about 3 cm in length. • Retract the first and second metatarsals to expose the tendon of the adductor hallucis. • Insert the scalpel blade into the interval between the fibular sesamoid plantarly and the metatarsal head dorsally. • Direct the blade distally along the adductor tendon until it strikes the base of the proximal phalanx. • Turn the blade laterally against the adductor tendon and release it from the base of the proximal phalanx. • Bring the knife proximally to dissect the tendon from the lateral sesamoid. • Reposition the retractors to expose the transverse intermetatarsal ligament. • Transect the intermetatarsal ligament, taking care not to injure the common digital nerve lying directly below

the l­igament. • Perforate the lateral capsule with several puncture incisions. • Stretch the hallux into varus to tear the remaining lateral capsule. • Approximate the first and second MTP capsules with three interrupted sutures incorporating the adductor tendon into this repair. • Tie these sutures after the medial plication is complete. • Compress the transverse metatarsal arch when tying the sutures to approximate the first and second metatarsal heads. Medial Eminence Resection and Medial Plication

• Make a medial longitudinal incision directly over the medial eminence (see Fig. 25H-17A). • Protect the dorsal and plantar digital nerves within the skin flap. • Make a vertical capsular incision with a No. 11 blade about 2 to 3 mm proximal to the base of the proximal phalanx. • Make a second parallel incision more proximal to the first (4 to 8 mm depending on the severity of the deformity). • Connect the two capsular incisions dorsally by an inverted-V 5-10 mm medial to the extensor hallucis longus tendon. • Connect the two capsular incisions plantarly by a V, keeping the blade inside the joint to prevent damage to the plantar medial cutaneous nerve.

• Incise the capsule along its dorsomedial aspect to create a proximally and plantarly based flap. • Resect the medial eminence in line with the medial diaphyseal cortex of the first metatarsal shaft starting about 2 mm ­medial to the sagittal sulcus (see Fig. 25H-17B).

• Reef the medial capsule with interrupted sutures holding the hallux in appropriate alignment (neutral varus-valgus and neutral rotation).

• Excise additional capsule if needed to improve the correction.

Foot and Ankle 2075

Sagittal sulcus

Capsular flap

A

B

Figure 25H-17   A, A proximally based, L-shaped capsular flap is used to expose the medial eminence. B, The medial eminence is resected along a line parallel with the long axis of the medial cortex of the first metatarsal.

­ rocedure has merit, but in young and middle-aged ath­ p letes, strength and weight-bearing capacity of the first ray are functions that should be preserved if possible. The use of a silicone hemiarthroplasty or double-stem total joint replacement is contraindicated in an athlete because the decreased weight-bearing capacity of the first ray after joint implantation may lead to lateral ­ transfer metatarsalgia. The increased stress placed on the first

A

MTP joint by a jogger or runner creates the potential for early failure of the implant. Arthrodesis of the first MTP joint for severe hallux val­ gus and hallux rigidus is probably the best alternative of all these salvage procedures in an athlete. Arthrodesis leaves the athlete with increased rigidity of the forefoot, which leads to early lift-off59 with walking and running and to potentially decreased function.

B

Figure 25H-18   A, Preoperative radiograph. B, Postoperative radiograph after distal soft tissue realignment. (© M. J. Coughlin. Used by permission.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2076 DeLee & Drez’s� O

Box 25H-8 P  roximal First Metatarsal ­Osteotomy Technique

• Make a dorsal 3-cm longitudinal incision over the prox­

imal first metatarsal (see Fig. 25H-19A). down to the subperiosteal level to expose the proximal metatarsal shaft (along the medial aspect of the extensor hallucis longus tendon). • Identify the metatarsocuneiform joint. • Perform the crescentic osteotomy with the concave as­ pect directed proximally 1 cm distal to the metatarsocu­ neiform joint (see Fig. 25H-19B). • Orient the plane of the osteotomy halfway between the perpendicular plane to the first metatarsal shaft and the floor. • Displace the osteotomy 2 mm laterally. • Fix the osteotomy with a single 0.062 Kirschner wire and a single compression screw (see Fig. 25H-19C, D, and E). • Bend the Kirschner wire, keeping it subcutaneous after closure.

• Dissect

Although these salvage procedures should be part of the surgical armamentarium of an orthopaedic foot surgeon, their applicability in an athletically active individual is lim­ ited. The primary procedures discussed earlier have lower postoperative morbidity than these salvage procedures and are preferable when surgery is indicated for an athlete with a symptomatic hallux valgus deformity.

Weighing the Evidence Akin Procedure A phalangeal osteotomy as described by Akin31 produces slight, if any, correction of the 1-2 intermetatarsal angle.60,61 Plattner and Van Manen61 reported on a series of 22 patients who had undergone an Akin procedure. An initial correction of the hallux valgus angle of 13 degrees decreased to a cor­ rection of only 6 degrees at an average 4.5 years of follow-up. Seelenfreund and associates32 reported a 16% recurrence (8 of 50 patients), and Goldberg and colleagues60 reported a 21% recurrence (75 of 351 patients) (Fig. 25H-21). Inter­ nal fixation to stabilize a phalangeal osteotomy site is pre­ ferred, and nonunion is uncommon.32 Although Colloff and Weitz62 used a lateral MTP capsular release, this technique may devascularize the proximal phalangeal fragment and is not recommended. Other complications reported after the Akin procedure include a poor cosmetic appearance60 and a high level of subjective postoperative patient dissatisfac­ tion.60,61 Goldberg and associates60 concluded that a pha­ langeal osteotomy as an isolated procedure in treatment of a hallux valgus deformity “does not have a sound biome­ chanical basis and should not be performed as an isolated procedure.” Plattner and Van Manen61 recommended this procedure for hallux valgus interphalangeus and not for a subluxated hallux valgus deformity (see Fig. 25H-14). Mitchell and Baxter54 and Colloff and Weitz62 suggested that a phalangeal osteotomy may be used in combination with a proximal repair to gain increased correction.

Chevron Procedure Some authors have recommended the chevron procedure for mild-to-moderate hallux valgus deformities,20,35,36 and Lillich and Baxter18 reported the use of the chevron proce­ dure in two elite female middle-distance and marathon run­ ners. The rationale for using this procedure was that toe-off power could be maintained, and range of motion would not be altered significantly in this extra-articular type of repair. Lillich and Baxter18 further stated that the stable nature of this osteotomy would make displacement unlikely, and the possibility of a transfer keratotic lesion is avoided. Johnson and colleagues35 and others36,41,42,63-65 noted a high level of excellent results with the chevron procedure, with a reported average correction of the hallux valgus angle of 12 to 13 degrees and an average correction of the 1-2 intermetatar­ sal angle of 4 to 5 degrees. Because of the limited correction of the hallux valgus angle offered by the chevron procedure, it should be reserved for mild and low-moderate deformities. Extension of the indications for this procedure to more severe deformities appears to increase the risk for recurrence, patient dissatisfaction, and complications. Meier and Kenzora39 reported on 50 patients (72 feet) after a distal metatarsal oste­ otomy and noted a 74% satisfaction rate when the preopera­ tive 1-2 intermetatarsal angle was greater than 12 degrees and a 94% satisfaction rate when the 1-2 intermetatarsal angle was 12 degrees or less. The most frequent complication associated with the chevron procedure is undercorrection or recurrence, which varies from 10% to 14%.36,63,65 Recurrent hallux valgus may develop when the indications for the chevron procedure are expanded to more severe deformities. Loss of correction can occur because of inadequate fixation or slippage at the osteotomy site. Shortening may occur as a result of excessive bone loss.36,37,39,63 Klosok and coworkers40 reported postop­ erative transfer lesions in 12% of cases.

Distal Soft Tissue Realignment Although Silver66 recommended resection of the medial eminence and a lateral soft tissue release, he did not report on the results of this procedure. Later, Kitaoka and asso­ ciates,67 in reporting on the Mayo Clinic experience after simple bunionectomy and medial capsulorrhaphy with or without lateral capsulotomy, noted at an average 4.8-year follow-up that the hallux valgus angle had increased 4.8 degrees from the preoperative deformity, and the 1-2 inter­ metatarsal angle had increased almost 2 degrees. Of the feet that had undergone a bunionectomy without a lateral cap­ sulotomy, 29% underwent reoperation at 5 years. A failure rate of 24% for the entire group was reported. Bonney and Macnab7 reported generally poor results after simple exos­ tectomy. Of their patients, 37% underwent additional treat­ ment, and the authors concluded that the only indication for a simple bunionectomy is a large medial eminence that is the sole cause of symptoms in a patient whose general medi­ cal condition contraindicates an extensive procedure and to whom the postoperative appearance is unimportant. Meyer and colleagues68 reported the results of the modi­ fied McBride procedure in 21 women joggers who had a ­symptomatic hallux valgus deformity. These authors reported a successful correction if the preoperative 1-2 intermetatar­ sal angle was less than or equal to 14 degrees and the hallux

Foot and Ankle 2077

A

C

E

B

D

Figure 25H-19   A, For a proximal first metatarsal osteotomy, a dorsal incision is made over the first metatarsal. B, A curved saw blade is used to create a crescentic osteotomy. C, Preoperative radiograph. D, Postoperative radiograph after a first metatarsal osteotomy. E, Lateral postoperative radiograph shows internal fixation within the first metatarsal and not crossing the metatarsocuneiform joint. (© M. J. Coughlin. Used by permission.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2078 DeLee & Drez’s� O

1

valgus angle was less than 50 degrees. They reported an aver­ age overall hallux valgus correction of 20 degrees and an aver­ age correction of the 1-2 intermetatarsal angle of 4.2 degrees. Of 21 patients, 19 (90%) reported significant improvement postoperatively. Two thirds of the patients resumed jogging 3 months after surgery, and 19 of 21 patients were involved in athletic activities 6 months after surgery. Mann and Coughlin,13 in a review of the results of 100 McBride procedures, reported an average correction of the hallux valgus angle of 14.8 degrees and an average correction of the 1-2 intermetatarsal angle of 5.2 degrees. Mann and Coughlin13 recommended that if more than 20 degrees of correction of the hallux valgus angle was indicated, the proce­ dure should be combined with a first metatarsal osteotomy. The limitations of a McBride or distal soft tissue recon­ struction are substantial. Mann and Pfeffinger69 observed that a severe hallux valgus deformity was not corrected adequately by a distal soft tissue reconstruction in half of the cases. The indication for this procedure is a subluxated hallux valgus deformity of less than 30 degrees with a 1-2 intermetatarsal angle of less than 15 degrees. One of the most significant complications of the distal soft tissue procedure is a postoperative hallux varus defor­ mity (Fig. 25H-22). Mann and Coughlin13 reported hal­ lux varus to occur in 11% of patients, although a severe deformity occurred in only 4%. Mann and Pfeffinger69 reported a higher incidence of hallux varus deformities after attempted correction of severe deformities.

1

2 2

3 3 4 4

A B Medial exostectomy Bone resection

1

2

C

Bone graft

3

Proximal First Metatarsal Osteotomy

4

Figure 25H-20   Schematic diagram of double and triple first ray osteotomies. (A and B are anteroposterior views; C is a lateral view.) 1, Phalangeal osteotomy (closing wedge); 2, distal metatarsal osteotomies (closing wedge); 3, crescentic proximal first metatarsal osteotomy; 4, cuneiform osteotomy (opening wedge). (From Coughlin M, Carlson R: Treatment of hallux valgus with an increased distal metatarsal articular angle: Evaluation of double and triple first ray osteotomies. Foot Ankle Int 20:765, 1999.)

A

B

A high satisfaction rate has been reported with the combined procedure of distal soft tissue reconstruction and proximal first metatarsal osteotomy (78% to 93%).70-75 The aver­ age correction of the hallux valgus angle has been reported consistently to be 23 to 24 degrees.2,6,71,75 The magnitude of improvement achieved is directly proportional to the severity of the preoperative hallux valgus deformity. A crescentic oste­ otomy is preferred1,47 because it results in minimal ­shortening

C

Figure 25H-21   A, Immediate postoperative radiograph shows correction with an Akin phalangeal osteotomy and medial eminence resection. B, One year after surgery, a recurrent deformity has developed with an increase in the hallux valgus angle. C, Ten years after surgery, further progression of deformity is noted.

Foot and Ankle 2079

­ alunion, avascular necrosis, and degenerative arthritis of m either the interphalangeal or MTP joint. These techniques are difficult and should be reserved for the occasional case of a hallux valgus deformity associated with a congruent joint with an increased distal metatarsal articular angle of greater than 15 degrees (Fig. 25H-23).

Author’s Preferred Method

Figure 25H-22   Hallux valgus deformity (overcorrection) caused by excessive medial eminence resection.

of the first metatarsal. Lengthening of the first metatarsal with an opening wedge osteotomy may lead to instability at the osteotomy site, and a closing wedge osteotomy may lead to lateral metatarsalgia as a result of first ray shortening. Complications associated with this procedure include overcorrection or hallux varus, undercorrection or recur­ rence, lateral metatarsalgia, and delayed union or malunion. Overcorrection often has been associated with lateral sesa­ moidectomy, and Simmonds74 and Mann and Coughlin13 have recommended that a lateral sesamoidectomy should be avoided. Retention of the lateral sesamoid has decreased the prevalence of hallux varus in one series from 11%13 to 8%.69 Often hallux varus deformities less than 10 degrees are asymptomatic and are associated with a satisfactory result.

Combined Multiple First Ray Osteotomies Funk and Wells51 and others12,48,52,53 reported success with dis­ tal metatarsal osteotomies and double osteotomies. Funk and Wells51 reported an average correction of the 1-2 intermeta­ tarsal angle of 7.2 degrees. Coughlin and Carlson3 reported on the use of double and triple osteotomies in the treatment of adolescent hallux valgus deformities in 18 patients (21 feet) with a hallux valgus deformity characterized by an increased distal metatarsal articular angle, who underwent either double or triple first ray periarticular osteotomies. The average hal­ lux valgus correction measured 23 degrees, and the average 1-2 intermetatarsal angle correction was 9 degrees. The distal metatarsal articular angle averaged 23 degrees preoperatively and was corrected to an average of 9 degrees postoperatively. Peterson and Newman49 reported a similar correction of the hallux valgus angle and the 1-2 intermetatarsal angle. Complications associated with multiple metatarsal osteotomies include loss of correction, loss of fixation,

In general, it is preferable to treat most high-level athletes symptomatically until their desire for high-level compe­ tition wanes. If the hallux valgus deformity continues to be symptomatic, surgery can be performed at that time. When surgery is contemplated, careful analysis of the com­ ponents of the hallux valgus deformity is essential in the decision-making process to ensure the appropriate choice of hallux valgus repair (Box 25H-9). For a mild hallux deformity (0 to 20 degrees), usually con­ servative care suffices. An Akin procedure or a chevron repair is used when indicated. Because these two repairs are extraarticular, they can be carried out in a patient with a congru­ ent (increased distal metatarsal articular angle) or subluxated MTP joint. For a low-moderate deformity (a hallux valgus angle of 20 to 30 degrees), a chevron procedure or a distal soft tissue realignment is the procedure of choice. The chev­ ron procedure may be used to correct a congruent or sublux­ ated MTP joint, but distal soft tissue realignment can be used only in a patient with a noncongruent or subluxated joint. For a high-moderate deformity (a hallux valgus angle of 30 to 40 degrees), distal soft tissue repair with or without a proximal first metatarsal osteotomy appears to offer the best chance for successful hallux valgus repair in the pres­ ence of a subluxated first MTP joint. For a congruent hal­ lux valgus deformity, periarticular osteotomies are indicat­ ed in the presence of an increased distal metatarsal articular angle. An intra-articular correction of a congruent MTP joint with hallux valgus places the repair at high risk for recurrence, degenerative joint disease, or at least restricted range of motion. Adding an osteotomy increases the post­ operative morbidity and recovery time in an athlete. When a more severe hallux valgus deformity is present, the magni­ tude of the surgical procedure increases, and postoperative swelling, reduced MTP joint range of motion, and reduced strength become more likely. For these reasons, patient counseling is especially advocated in this group. A change in athletic activity, substituting activities such as bicycling and other nonimpact activities for running, may reduce symptoms significantly in an athlete with hallux valgus.

Box 25H-9 P  rimary Goals of Hallux Valgus Correction Surgery

• Maintain

a flexible metatarsophalangeal joint with as normal range of motion as possible. • Restore normal weight-bearing pattern in the forefoot. • Correct the deformity without producing residual ­disability. • Allow a reasonable route of salvage if a complication ­develops.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2080 DeLee & Drez’s� O

A

B

C

D Figure 25H-23   A, Preoperative radiograph shows hallux valgus angle of 33 degrees, intermetatarsal angle of 18 degrees, and distal metatarsal articular angle (DMAA) of 25 degrees. B, Postoperative radiograph shows proximal crescentic first metatarsal osteotomy, distal first metatarsal closing wedge osteotomy, and phalangeal closing wedge osteotomy. C, Lateral radiograph shows periarticular osteotomies. It is likely that the screw and pin did cross the metatarsocuneiform joint. Hardware typically is removed 6 weeks after surgery, as shown in D. D, Radiograph at 3 years’ follow-up shows hallux valgus angle of 16 degrees, intermetatarsal angle of 13 degrees, and DMAA of 11 degrees. (From Coughlin M, Carlson R: Treatment of hallux valgus with an increased distal metatarsal articular angle: Evaluation of double and triple first ray osteotomies. Foot Ankle Int 20:767, 1999.)

Postoperative Prescription, Outcomes Measurement, and Potential Complications A soft compression dressing is applied at surgery, holding the toe in appropriate alignment, and is changed weekly in the first 6 to 8 weeks for most hallux valgus correction procedures. A postoperative shoe is appropriate for most

realignment procedures for the first 6 weeks. When mul­ tiple first ray osteotomies are performed, a short leg cast is recommended. A cast may also be used in the case of a proximal first metatarsal osteotomy, but is not required. Weight-bearing is allowed mostly on the heel in the first 6 weeks after surgery. Non–weight-bearing in the cast is preferred for multiple first ray osteotomies. Kirschner wires used to fix Akin or chevron osteotomies are removed 4 to 6 weeks after surgery.

Foot and Ankle 2081

Box 25H-10 C  omplications After Hallux ­Valgus Surgery

Box 25H-11 C  riteria for Return to Play After Hallux Valgus Surgery

• Undercorrection or recurrent deformity • Hallux varus • Malunion or nonunion of the osteotomy site • Avascular necrosis of the metatarsal head • Lateral metatarsalgia or transfer lesions • Degenerative arthritis in the hallux metatarsophalan­

• Full healing of osteotomies on radiographs • Adequate range of motion, strength and endurance • Running 7 weeks postoperatively in a loose-fitting

At 6 weeks, the patient is transitioned to a stiff-soled sandal and then gradually into a supportive athletic shoe as pain and swelling allow. Range of motion exercises are started between 3 and 6 weeks. Aggressive walking activity can be started 6 weeks after surgery if the patient is pro­ gressing well. A comfortable, supportive athletic shoe with a roomy toe box and a spacer between the first and second toes protects the athlete as he or she returns to activities. Activity is increased as tolerated, with expected return to jogging at 2 to 3 months, running at 3 to 4 months, and full return to sport at 3 to 6 months. Distal soft tissue realign­ ment requires a longer return to full activity to maintain the alignment of the hallux. Full healing of the proximal first metatarsal osteotomy or multiple first ray osteotomies occurs at about 6 to 8 weeks. Advancing range of motion and more aggressive walking should be delayed until oste­ otomy sites are completely healed. The most common complication following hallux valgus surgery is undercor­ rection or recurrence of the deformity (Box 25H-10).

for excessive dorsiflexion and push-off than sprinters. It is wise to treat most high-level athletes symptomatically until their desire for high-level competition wanes; if the hallux valgus deformity continues to be symptomatic, surgery can be performed at that time.

geal joint or interphalangeal joint

shoe with a toe spacer placed between the hallux and second toe • Full return to sport 3 to 6 months postoperatively

TURF TOE

With any hallux valgus correction, return to play occurs when full healing of osteotomies is noted on radiographs and the athlete regains adequate range of motion, strength, and endurance to return to sport. Lillich and Baxter18 stressed the importance of a precise operative technique with minimal joint dissection. They allowed running 7 weeks after surgery in a loose-fitting shoe with a toe spacer placed between the hallux and second toe. Running speed and duration were increased gradually, with a return to full activity achieved at 12 weeks after surgery. In many cases, full return to sport may not occur until 6 months after sur­ gery (Box 25H-11).

MTP joint injuries are commonly the result of a hyper­ extension injury and have become an increasing source of forefoot pain and dysfunction in professional athletes.76-80 Occasionally, these injuries are characterized by a signifi­ cant delay in return to sporting activities.76 Bowers and Martin81 initially described a syndrome in which “a sprain of the plantar capsuloligamentous complex of the great toe of the MTP joint” occurred. Bowers and Martin81 reported on 27 such injuries of the foot in football play­ ers and coined the term turf toe because of the frequency with which the injury occurred on artificial playing sur­ faces as opposed to natural grass. Clanton and associates76 reported on 62 MTP joint injuries in 53 collegiate football players, all of whom sustained injuries on artificial playing surfaces. Severe trauma to the MTP joint of the hallux is relatively uncommon. Jahss,82 Giannikas and coworkers,83 DeLee,84 and Rodeo and colleagues85 reported on fracture-disloca­ tions and dislocation of the first MTP joint after hyperex­ tension injuries. Usually these injuries are associated with a significant dissipation of energy, and sesamoid fractures may occur. Hyperextension injuries77,81,86,87 to the hallux and lesser MTP joints have been linked to flexible shoe­ wear88 and artificial playing surfaces77,81,86,87 and in certain cases may have long-term sequelae for the symptomatic athlete.

Special Populations

Anatomy and Biomechanics

A symptomatic hallux valgus deformity presents a special problem to the treating physician. Although, in general, conservative methods should be used in all athletes, a high-performance athlete requires more deliberate treat­ ment. Hamilton (quoted by Lillich and Baxter18) stated that bunionectomies should be avoided in sprinters and ballet dancers. Restricted dorsiflexion and plantar flexion may interfere with a dancer’s functional ability. Likewise, the need for dorsiflexion as well as strength in push-off makes sprinters poor candidates for a hallux valgus repair. ­Middle- and long-distance runners appear to have less need

The first MTP joint is a relatively shallow articulation between the convex metatarsal head and concave base of the proximal phalanx.89 Joint stability is primarily provided by the capsuloligamentous structures and enhanced by musculotendinous and bony structures that surround the joint (Box 25H-12). The sesamoids lie within the flexor hallucis brevis tendon (Fig. 25H-4A) and articulate with the medial and lateral plantar facets of the first metatar­ sal head (see Fig. 25H-4B). The tendons of the abductor hallucis and adductor hallucis contribute to the joint cap­ sule medially and laterally and insert on medial and lateral

Criteria for Return to Play

�rthopaedic ����������� S �ports ������ � Medicine ������� 2082 DeLee & Drez’s� O

Box 25H-12 S  tabilizing Structures of the First Metatarsophalangeal Joint Joint Capsule Ligaments Plantar plate Medial and lateral collateral ligaments Sesamoid ligaments Tendons Flexor hallucis brevis Adductor hallucis Abductor hallucis Bone Medial and lateral sesamoids

borders of the sesamoids as well as onto the base of the proximal phalanx. These structures provide plantar stabil­ ity along with the plantar plate, a thick fibrous structure extending from a loose attachment to the metatarsal neck to a firm attachment to the base of the proximal phalanx. The medial and lateral collateral ligaments interdigitate with the sesamoid ligaments (Fig. 25H-2A) and contribute to the medial and lateral stability of the MTP joint. Dor­ sally, the extensor hallucis longus, extensor hallucis bre­ vis, and hood ligaments (see Fig. 25H-2B) of the extensor expansion form the major capsular stabilizing structures. However, the dorsal structures are weak relative to the plantar supporting structures.

Motion at the MTP joint occurs by means of a slid­ ing action along the joint surface. Joseph90 reported that normal active extension of the MTP joint approximates 80 degrees (Fig. 25H-24). Clanton and coworkers76,91,92 and others81 reported that an average of 60 degrees of dorsiflexion can be expected with normal gait. As the pha­ lanx approaches the upper range of dorsiflexion excur­ sion, this sliding action is replaced by axial compression forces on the articular surfaces of both the metatarsal head and base of the proximal phalanx, which can lead to joint injury. Stiff sole shoes diminish MTP dorsiflexion by up to 30 degrees without causing any noticeable impairment of gait (Fig. 25H-25). The primary mechanism of injury is forced hyper­ extension of the hallux MTP joint,76,79,81,91 often with an axial load to the heel.77 The resulting injury to the plantar structures of the joint ranges from a mild sprain to frank dislocation of the joint. The plantar plate fre­ quently tears at the weakest point, the neck of the first metatarsal. However, the plantar structures can also be disrupted distal to the sesamoids, through a bipartite sesamoid or with sesamoid fracture. With further dor­ siflexion of the joint, an impaction injury to the joint surface can occur. A valgus force can be associated with the injury causing disruption of the plantar medial cap­ suloligamentous structures or the tibial sesamoid. Less frequently, a first MTP joint sprain is the result of a hyperflexion injury.77

Classification Rodeo and coworkers85 proposed a classification of first MTP joint injury (Box 25H-13). Clanton76,91,92 ­proposed a classification for more acute injuries (Table 25H-2).

80°

25°

Figure 25H-24   Normal active extension of the first metatarsophalangeal joint averages 80 degrees. Active plantar flexion averages 20 to 25 degrees. This excursion may decrease with advancing age.

Figure 25H-25   A shoe with a flexible insole affords little protection for a hyperextension injury. (From Clanton TO, Butler JE, Eggert A: Injuries to the metatarsophalangeal joints in athletes. Foot Ankle 7:162-176, 1986. © American Orthopaedic Foot and Ankle Society, 1986.)

Foot and Ankle 2083

Evaluation Clinical Presentation and History Both clinical history and physical examination are directed toward identifying the severity and location of the injury. The history elicited from the athlete gives some indication of the mechanism of the injury. Players may complain of localized pain, swelling, and pain with range of motion and ambulation. The patient history should also include the type of footwear used and the playing surface involved.

Physical Examination On physical examination, periarticular swelling and ecchy­ mosis are typically present. Mild injuries present with plantar or plantar medial tenderness. Dorsal tenderness is elicited in more severe capsular disruption. In the most

Box 25H-13 C  lassification of First Metatarsophalangeal Joint Injuries Grade 1: Acute sprain of the first metatarsophalangeal (MTP) joint plantar capsule • Localized tenderness, swelling, and pain with ­dorsiflexion • Normal radiographs • Conservative treatment Grade 2: Acute sprain of the first MTP joint with signifi­ cant plantar capsule disruption • Ecchymosis, painful dorsiflexion, and loss of motion. • Diastasis of a partite sesamoid or joint instability on radiographs • No degenerative first MTP joint changes • Conservative or surgical treatment Grade 3: Chronic symptoms involving the first MTP joint resulting from previous injury • Loss of motion • Degenerative joint disease, hallux rigidus, or malalign­ ment on radiograph • Treatment is often surgical

severe injuries, fracture-dislocation of the joint can occur with obvious joint deformity. A palpable plantar defect may be noted, indicating a partial or complete disruption of the plantar plate–sesamoid mechanism.93 The athlete may present with an antalgic gait, attempt­ ing to walk with the limb in external rotation or the foot everted to minimize MTP joint excursion. Push-off is impaired with running, and it may be difficult to crouch with the MTP joint extended.81 Range of motion and maneuvers to load the joint will also elicit discomfort on physical examination.81 Decreases in range of motion can be observed in both dorsiflexion and plantar flexion. Weak­ ness of the MTP joint in plantar flexion may be observed (Box 25H-14).

Imaging Weight-bearing anteroposterior and lateral radiographs, as well as oblique views, are helpful in evaluating first MTP dor­ siflexion injuries. Bilateral weight-bearing ­anteroposterior views are helpful to identify proximal migration of the

Box 25H-14 T  ypical Findings in Turf Toe Injuries Physical Examination

• First metatarsophalangeal (MTP) joint swelling • Ecchymosis adjacent to the area of capsular injury • Plantar tenderness at the MTP joint • Pain with passive MTP joint dorsiflexion • Pain with joint loading: walking, push-off, crouching with the MTP joint extended

• Decreased dorsiflexion of the MTP joint Radiographic Examination

• Soft tissue swelling • Small periarticular bony avulsions • Intra-articular loose bodies • Diastasis of bipartite sesamoid • Sesamoid fracture • Migration of sesamoids

  TABLE 25H-2  Acute First Metatarsophalangeal Joint Injury Grade

Objective Findings

Activity Level

Pathology

Treatment

1

Localized plantar or medial tenderness Minimal swelling No ecchymosis More diffuse and intense tenderness Mild to moderate swelling Mild to moderate ecchymosis Painful and restricted range of motion Severe and diffuse tenderness Marked swelling Moderate to severe ecchymosis Range of motion painful and limited

Continued athletic participation

Stretching of the capsuloligamentous complex

Symptomatic

Loss of playing time for 3-14 days

Partial tear of the capsuloligamentous complex

Walking boot and crutches as needed

Loss of playing time for 2-6 weeks

Tear of the capsuloligamentous complex Articular cartilage and subchondral bone injury Possibility of sesamoid fracture or separation of bipartite sesamoid Possibility of dislocated first metatarsophalangeal joint with spontaneous reduction

Long-term immobilization in boot or cast versus surgical repair

2

3

�rthopaedic ����������� S �ports ������ � Medicine ������� 2084 DeLee & Drez’s� O

A

B

C

Figure 25H-26   A, A fleck of bone may indicate ligamentous or capsular disruption. B, Intra-articular loose bodies (arrow) may occur after turf toe injury. C, A bone scan may show increased uptake even when radiographs still appear normal. (From Clanton TO, Butler JE, Eggert A: Injuries to the metatarsophalangeal joint in athletes. Foot Ankle 7:162-176, 1986. © American Orthopaedic Foot and Ankle Society, 1986.)

sesamoids. A forced dorsiflexion lateral view assists in the diagnosis of distal sesamoid migration and diastasis of a bipartite or fractured sesamoid. In general, initial radio­ graphs typically reveal only soft tissue swelling.76,81,92 However, radiographs should be reviewed thoroughly for more subtle findings associated with these injuries (see Box 25H-14; Fig. 25H-26A). When a compression frac­ ture of the metatarsal head has occurred, intra-articular loose bodies may be apparent (see Fig. 25H-26B). Chon­ drolysis of the joint space can occur with time, and Clan­ ton and colleagues76 have also reported cystic changes in the metatarsal head denoting progressive degenerative changes. Bone scan can be helpful in distinguishing a bipar­ tite sesamoid from an acute fracture (Fig. 25H-26C). ­Magnetic resonance imaging (MRI) can be used to ­evaluate the presence and extent of capsular or plantar plate disruption. In addition, MRI will reveal osseous or articular damage in the presence of normal radiographs (Fig. 25H-27).94

(Fig. 25H-28). Typically, refraining from athletic activity for 1 to 2 weeks allows significant improvement. For grade 3 injuries, immobilization and a period of restricted weight-bearing is usually required in addition to ice, elevation, and compression. Weight-bearing is initi­ ated within the limits of pain. Mobilizing the joint as the patient is able to tolerate assists in achieving more normal range of motion. The athlete may progress weight-­bearing when gait becomes less painful and gradually increase speed and activity level. Cutting maneuvers are initiated last within the limits of pain. Gradual return to sporting activities depends on the resolution of symptoms.76,91,92 Intra-articular injections of steroids disguise symptoms or give temporary relief only and are generally discour­ aged.76,77,91,92 When joint dislocation is present (Fig. 25H-29), prompt reduction should be performed. Reduction is performed by hyperextending the joint with concurrent longitudinal traction and plantar pressure on the base of the proximal

Treatment Options Nonoperative For grade 1 injuries, ice, compression, and nonsteroidal anti-inflammatory drugs may be used.76,81,91,92 The patient can continue to participate in athletics if pain is mini­ mal. Taping of the toe helps compress the joint and limit motion. Inserting a rigid forefoot insole76,81,91,92 to stiffen the shoe may decrease discomfort and help to reduce the incidence of recurrent injury. Similar nonoperative methods are employed for grade 2 injuries. However, activity should be restricted in athletes with more severe injuries and significant discomfort. Addi­ tion of a firm insole prevents dorsiflexion strain to the toe

Figure 25H-27   Magnetic resonance image of the first metatarsophalangeal joint may reveal osseous or articular cartilage abnormalities.

Foot and Ankle 2085

associated with restricted range of motion, which may limit running ­activity significantly.

Weighing the Evidence

Figure 25H-28   Use of a firm insole that restricts forefoot motion may prevent injury and may diminish symptoms after injury. (From Clanton TO, Butler JE, Eggert A: Injuries to the metatarsophalangeal joints in athletes. Foot Ankle 7:162-176, 1986. © American Orthopaedic Foot and Ankle Society, 1986.)

phalanx. If the joint is irreducible, operative intervention is required (Box 25H-15).84

Operative Operative treatment is considered in more severe ­injuries with joint instability, diastasis of bipartite sesamoid or ses­ amoid fracture, retraction of sesamoids, traumatic ­ hallux valgus, or presence of a loose body or chondral injury. Operative intervention is rarely necessary in acute injuries; management is usually centered on conservative methods. After defining a capsular injury, an open repair occasion­ ally is performed, especially in the presence of a partial or complete plantar plate injury.77 Coker and associates77 recommended operative repair for these injuries. When a sesamoid injury (fracture, osteonecrosis) occurred and did not heal successfully, Coughlin93 believed that surgi­ cal excision may be warranted. Open débridement of the joint may be required when intra-articular injury occurs. If hallux rigidus develops, cheilectomy or arthrodesis is ­considered (see Box 25H-15). With a progressive hallux valgus deformity, it is likely that disruption of the medial or plantar medial capsule has occurred. Conservative treatment in the competitive ­athlete is warranted initially; however, with progression of pain and deformity, surgical repair is often necessary. Roomy footwear and the use of orthotic devices are pre­ ferred to surgical intervention if the progression of the deformity is insidious and asymptomatic. Operative treat­ ment of ­ hallux valgus in the competitive athlete can be

Although there are few documented studies of athletes with MTP joint injuries, Rodeo and colleagues79 evaluated professional football players. Of players surveyed, 57% (38 players) reported symptoms of turf toe. Although this incidence was comparable to that seen in players from other teams whose home field was a natural grass surface, 84% of the players in this series reported that their initial injury occurred on artificial turf. Other factors that predispose an individual to injury include increasing age of the player, number of years in professional football,78 increased range of ankle dorsiflexion,78 pes planus, and decreased preinjury MTP range of motion.76 The classic mechanism of injury is a forced hyperex­ tension injury to the first MTP joint.76,79,81,91 Coker and associates77 described this as an axial loading of the foot with the MTP joint in hyperextension. Dorsiflexion in excess of a normal range of motion can lead to varying degrees of soft tissue capsular disruption, plantar plate rupture, or injury to the articular cartilage and subchon­ dral bone. Only 35 of 51 patients in Clanton’s series could recall the mechanism of injury76; however, 32 of the 35 patients (92%) could recall and describe a hyperextension injury. Rodeo and associates79 reported that 85% of the football players in their series sustained a hyperextension injury. Coker and associates77 reported that the mecha­ nism of injury could be hyperextension, hyperflexion (which is rare), or valgus, depending on the applied stress. Massari and coworkers95 described a rare case of varus injury of the first MTP joint. Clanton and colleagues76 hypothesized that more severe MTP joint injuries occur in players who have preexisting restricted motion of the first MTP joint. With continued forced dorsiflexion, cap­ sular, ligamentous, or tendinous injury and axial compres­ sion of the joint surfaces occur sooner than they would in a joint with more excursion. Although Garrick88 noted an increased injury rate in players playing on artificial surfaces, Bowers and Mar­ tin81 found a correlation between flexible shoes and a relatively hard artificial playing surface. The use of relatively lightweight, flexible shoes on artificial turf apparently predisposes the forefoot to injury.77,86,87 In contrast, on grass surfaces, the use of a conventional cleated shoe, which is much stiffer (a steel plate is incor­ porated into the sole for attachment of the cleat), limits forefoot excursion.76 Clanton and colleagues76 observed that the rate of injury on artificial turf could be reduced markedly by inserting a steel forefoot insole inside the players’ shoes. Depending on the magnitude of force, the position of the MTP joint, and the direction of force, various injuries to the MTP joint can occur. Although most often a hyper­ extension injury leads to symptoms of turf toe, Coker and associates77 and Clanton and colleagues76 noted that occa­ sionally a ­plantar flexion injury is associated with a turf toe injury. Valgus stress can occur as well, either alone or in combination with other forces, leading to different injury patterns.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2086 DeLee & Drez’s� O Figure 25H-29   A-D, Fracture-dislocation of the metatarsophalangeal-sesamoid joint (grade 3 injury).

A

B

C D

Box 25H-15 T  reatment Options for First Metatarsophalangeal Joint Injuries Nonoperative • Immobilization • Restricted weight-bearing • Taping of the toe • Rigid forefoot insole Operative

• Plantar plate and capsular repair • Partial or full sesamoid excision • Joint débridement • Correction of traumatic hallux valgus • Cheilectomy • First metatarsophalangeal arthrodesis

Author’s Preferred Method Preseason physical examination helps identify which play­ ers are at high risk for turf toe. Players with less than 60 degrees of dorsiflexion at the MTP joint or with in­ creased dorsiflexion at the ankle joint should be protected with a stiff-soled shoe or a stiff insole. These modalities may prevent MTP joint injury or reduce the magnitude of injury should it occur. An MRI scan should be obtained when there is suspicion of significant injury. In the high-­performance athlete, open repair of partial and complete capsular, ­plantar plate, or sesamoid disruption may be ­indicated. The use of appropriate footwear provides a great opportu­ nity to reduce the incidence of first MTP joint injury in the athlete. Although turf toe can be a disabling injury that can

Foot and Ankle 2087

bring about the premature conclusion of an athletic career, prophylactic shoewear modification and counseling athletes at risk can reduce the frequency and magnitude of this injury. In the athlete with an acute grade 1 turf toe injury, ice, compression, nonsteroidal anti-inflammatory drugs (NSAIDs), and taping of the toe make continued sports participation possible. For more severe injuries, more ag­ gressive physical therapy is necessary. In patients with grade 2 injuries, a firm forefoot insole is helpful to prevent further injury to the MTP joint of the great toe. Usually it is neces­ sary to refrain from athletic activities for 1 to 2 weeks. With more severe injuries, such as grade 3 turf toe, an ­ athlete may be unable to return to sports for the remainder of the season. Surgical intervention is considered in the presence of a partial or complete plantar plate disruption. This in­ jury can occur with a dislocation or subluxation of the MTP joint or a severe articular or nonarticular injury.

Postoperative Prescription, Outcomes Measurement, and Potential Complications Postoperative management must balance the soft tissue heal­ ing with early range of motion of the MTP joint.96 Immobi­ lization in a removable splint with restricted weight-bearing continues for 4 weeks. Seven to 10 days after surgery, passive range of motion is started. At 4 weeks, weight-bearing and active range of motion are initiated. Modified shoewear is allowed at 8 weeks with return to full sport at 3 to 4 months. After return to sport, the athlete continues shoe modifica­ tions to protect from excessive dorsiflexion. After a turf toe injury, athletes can experience persis­ tent pain with toe-off and stiffness of the first MTP joint. Degenerative joint disease can cause pain and restricted range of motion. Progressive hallux valgus deformity develops in cases of medial capsule injury. The develop­ ment of chronic pain, restricted motion, and discomfort with running may herald the end of an athletic career.

Criteria for Return to Play Return to sport in first MTP joint injuries is gradual and dependent on the resolution of symptoms. Shoe modifica­ tions with a stiff forefoot insole will decrease discomfort with return to activity. Recovery from an acute injury typi­ cally ranges from 2 to 6 weeks depending on the severity of the injury. In the case of an established turf toe, sport­ ing activity can exacerbate the injury. Physical therapy, NSAIDs, and the use of a stiff insole may allow return to sports and continued athletic activity (Box 25H-16).

Special Populations The preference of athletes for lightweight, flexible shoes that afford good traction but little structural support pre­ sents an increasing risk factor for competitive athletes. First MTP joint sprains were first noted by Bowers and Martin81 in 1976 and associated with artificial playing surfaces and flexible footwear. Rodeo and colleagues79

Box 25H-16 C  riteria for Return to Play After First Metatarsophalangeal Joint Injury Nonoperative Treatment

• Return to sport is gradual. • Recovery from an acute

injury ranges from 2 to 6 weeks. • A stiff forefoot insole minimizes discomfort with return to sport. Operative Treatment

• Full return to sport is typically 3 to 4 months after ­surgery.

Box 25H-17 Sesamoid Function

• Protect the tendon of the flexor hallucis longus. • Absorb most of the weight-bearing on the medial aspect of the forefoot.

• Increase the mechanical advantage of intrinsic muscula­ ture of the hallux.

evaluated ­professional football players, and 45% (36 of 80 players surveyed) reported symptoms of a turf toe. When the players could recall the initial injury, 83% of players reported their injury occurred on artificial turf. Players with restricted motion at the MTP joint may sustain an injury with relatively less excursion of the hallux. Protec­ tion of the forefoot with an orthotic device can avoid more serious injury.76,91,92

SESAMOID DYSFUNCTION The first MTP joint is characterized by the two sesamoids, which play a significant role in the function of the great toe (Box 25H-17).97 Sesamoid dysfunction is uncommon; however, it can occur with arthritis,97-102 trauma,97,100,103-114 osteochondritis,99,105,107,115-118 infection,119-125 or sesa­ moiditis.98,107,115-117,126,127

Anatomy and Biomechanics The sesamoids, contained within the double tendon of the flexor hallucis brevis (see Fig. 25H-4A) on the plantar aspect, articulate dorsally with plantar facets on the first metatarsal head. These concave facets are separated by a crista or intersesamoidal ridge, which divides the medial and lateral metatarsal facets. This crista affords intrinsic stability to the sesamoid complex (Fig. 25H-30) and may be eroded or become obliterated when severe hallux valgus causes an insidious dislocation of the sesamoid complex. The sesa­ moids insert on the base of the proximal phalanx through the plantar plate, an extension of the tendons of the flexor hallucis brevis. Suspended from a sling-like mechanism composed of the sesamoid ligaments and the MTP collat­ eral ligaments (Fig. 25H-2) on the medial and lateral aspects of the MTP joint, the sesamoids are stabilized superiorly by

�rthopaedic ����������� S �ports ������ � Medicine ������� 2088 DeLee & Drez’s� O

Box 25H-18 Vascular Supply of the Sesamoids

• Type

A: arterial circulation derived from the medial plantar artery and the plantar arch (52%) • Type B: circulation predominantly from the plantar arch (24%) • Type C: circulation from the medial plantar artery (24%)

Box 25H-19 Common Causes of Sesamoid Pain

Figure 25H-30   The intersesamoidal ridge or crista is well seen on an axial radiograph. (© M. J. Coughlin. Used by permission.)

the plantar capsule and the plantar aponeurosis. The flexor hallucis brevis provides an active plantar flexion force at the MTP joint; through its insertion into the sesamoid mecha­ nism, an increased mechanical advantage in plantar flexion is maintained.97 The tendon of the flexor hallucis longus, which is protected in its tendon sheath by the medial and lateral sesamoids, provides a plantar flexion force to the interphalangeal joint of the great toe (Fig. 25H-4B). The abductor hallucis inserts onto the plantar medial base of the proximal phalanx and the medial sesamoid, pro­ viding a medial stabilizing force on the sesamoid mecha­ nism. Similarly, the adductor hallucis tendon inserts onto the plantar lateral base of the proximal phalanx and the lat­ eral sesamoid, providing a lateral stabilizing force. Kewenter100 reported that the medial sesamoid is located slightly more distal and is slightly larger than the lateral sesamoid. Orr128 measured the tibial sesamoid to be 9 to 11 mm in width and 12 to 15 mm in length on average. The fibular sesamoid has an average width of 7 to 9 mm and an average length of 9 to 10 mm. Kewenter100 reported that ossification of the hallucal sesamoids occurs between the ages of 7 and 10 years. Frequently, ossification of the sesamoids occurs from multiple centers; this may be the reason for the development of multipartite sesamoids. Several authors have described the vascular supply of the sesamoids.129,130 Sobel and coworkers129 determined that the primary vascular supply to the sesamoids enters proxi­ mally and plantarly. The distal blood supply is much more tenuous, contributing to delayed healing and nonunion of sesamoid fractures. Pretterklieber and Wanivenhaus130 described three different types of arterial circulation (Box 25H-18). These authors concluded that the pattern of arte­ rial supply of the sesamoids plays a role in the development of avascular necrosis. With standing, the sesamoids are located slightly poste­ rior to the metatarsal head. As the MTP joint dorsiflexes, the sesamoids are pulled distally, protecting the ­ plantar surface of the first metatarsal head and absorbing the weight-bearing forces on the medial aspect of the forefoot (see Box 25H-17).

• Sesamoid fracture (acute trauma or stress fracture) • Bipartite sesamoid • Nerve compression • Osteochondritis and avascular necrosis • Arthritis • Sesamoiditis • Intractable plantar keratosis • Infection Classification A classification system for sesamoid dysfunction has not been described. However, there are several different causes associated with sesamoid pain (Box 25H-19).

Bipartite Sesamoid and Fracture The incidence of bipartite sesamoids as well as their cause is a subject of significant controversy in the literature.114 Dobas and Silvers131 found a 19% incidence of bipartite sesamoids, whereas Kewenter100 found a 31% incidence. Rowe123 reported a 6% to 8% frequency of bipartite sesa­ moids and noted that 90% of these were bilateral. Dobas and Silvers131 noted that 87% of the bipartite sesamoids in their series involved the tibial or medial sesamoid. Dobas and Silvers also noted that about 25% of the divided tibial sesamoids had an identical bipartite tibial sesamoid, and the remaining sesamoids were asymmetric in regard to division. Jahss99 reported that the medial sesamoid has a 10 times greater incidence of bipartitism. Although Giannestras132 stated that bipartite sesamoids are ­symmetrical, this ­finding has not been shown in other studies. The tibial sesamoid often is divided into two or more parts, whereas a lateral sesamoid rarely is divided into more than two fragments. Inge and Ferguson97 reported a 10.7% incidence of bipar­ tite sesamoids and noted that only 25% of these were bilateral. Of bipartite sesamoids that were bilateral, 85% showed asymmetric division. Fracture of a sesamoid is usually due to either sudden load­ ing of the forefoot, a fall onto the forefoot, or a crush injury.

Nerve Compression The plantar lateral digital nerve and the plantar medial digital nerve are located adjacent to the lateral and medial sesamoids (Fig. 25H-31). Impingement of either one of

Foot and Ankle 2089

Figure 25H-31   The medial and lateral common digital nerves are found on the medial and lateral aspects of the respective sesamoids. They should be protected in any surgical excision. (© M. J. Coughlin. Used by permission.)

Figure 25H-32   An intractable plantar keratosis may develop beneath the tibial sesamoid. (© M. J. Coughlin. Used by permission.)

these branches may be a source of pain in the area of the sesamoids. Helfet133 noted compression of the plantar lat­ eral cutaneous branch to the hallux. The plantar medial digital nerve may be compressed by the medial sesamoid in a similar fashion. Often this type of pain is difficult to dif­ ferentiate from pain localized to a particular sesamoid.

Sesamoiditis

Osteochondritis and Avascular Necrosis Osteochondritis of the sesamoid is characterized by tender­ ness on palpation, pain localized to the area of the involved sesamoid, and fragmentation or mottling of the involved sesamoid on radiographic evaluation.116,117 Although Renander127 first described this condition in 1924, the cause of osteochondritis of the sesamoid is unknown. Osteochondritis appears to be infrequent,116 and trauma frequently has been thought to be the cause of fragmen­ tation.98 Jahss99 compared osteochondritis with osteone­ crosis. Fleischli and Cheleuitte134 and Julsrud135 reported cases of osteonecrosis of the sesamoids. Kliman and asso­ ciates117 suggested that the cause of osteochondritis is a stress ­ fracture and that the subsequent healing process is the cause of the fragmentation.

Arthritis Osteoarthritis of the sesamoids has been reported97-102 in association with hallux valgus, hallux rigidus, and rheuma­ toid arthritis and as an isolated occurrence. Osteoarthritis of the sesamoids may be a natural progression of early ses­ amoiditis, chondromalacia, or localized sesamoid trauma. Scranton and Rutkowski102 described erosion of the articu­ lar cartilage in what they termed progressive sesamoid chondromalacia that eventually led to surgical resection.

The diagnosis of sesamoiditis is one of exclusion. Usually it is associated with trauma or repetitive stress and often occurs in young athletes. Apley115 found remarkable simi­ larity between the articular cartilage degeneration seen in sesamoiditis and that seen in chondromalacia of the patella, and he termed this condition sesamoid chondromalacia.

Intractable Plantar Keratosis A keratotic lesion may develop beneath either sesa­ moid.136,137 When associated with a high-arched or cavus type of foot, a plantar flexed first ray may be the cause of the callus formation. A large diffuse callus beneath the meta­ tarsal head is usually associated with a cavus foot. When a sesamoid is involved, there is usually a more localized or concentric lesion (Fig. 25H-32).

Evaluation Clinical Presentation and History In the history of a patient with a symptomatic sesamoid, the most frequent subjective symptoms are pain and dis­ comfort in the toe-off phase of gait. Sesamoid dysfunction that restricts the range of motion of the MTP joint may lead to a pathologic gait pattern. An athlete may tend to evert the forefoot to decrease dorsiflexion excursion in the latter part of gait or may tend to toe-off prematurely to minimize dorsiflexion excursion of the hallux. In the case of sesamoid fracture, an athlete often walks on the lat­ eral aspect of the foot to decrease weight-bearing on the ­sesamoid complex.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2090 DeLee & Drez’s� O

Box 25H-20 T  ypical Findings in Sesamoid ­Dysfunction Physical Examination

• Pain on direct palpation of sesamoids • Restricted range of motion at the first ­metatarsophalangeal (MTP) joint

• Pain with motion of the first MTP joint • Swelling in the first MTP joint • Diminished strength in plantar flexion or dorsiflexion • Neuritic symptoms in cases of nerve compression Radiographic Examination • Radiographs frequently normal • Sesamoid fracture (trauma) or fragmentation ­(osteochondritis) • Increased uptake on bone scan

Physical Examination Pain, warmth, swelling, decreased range of motion, and diminished strength are common findings on examina­ tion of patients with sesamoid dysfunction (Box 25H-20). ­Localized tenderness to palpation of the involved sesamoids with inflammation on the plantar aspect of the sesamoid mechanism can be present. Often, swelling of the MTP joint or synovitis is noted. In addition to restricted range of motion of the MTP joint, the patient may experience pain with forced dorsiflexion of the MTP joint or pain on localized palpation. A plantar keratosis beneath either the tibial of fibular sesamoid occasionally may accompany a symptomatic sesamoid. Deviation of the hallux either medially (hallux varus) or laterally (hallux valgus) can be associated with sesamoid dysfunction. Traumatic disruption of a sesamoid may cause progressive, insidious deviation of the great toe. A hallux valgus or hallux varus deformity can also occur after previ­ ous sesamoid resection. A careful sensory examination may reveal neuritic symp­ toms or numbness secondary to digital nerve compression by either the tibial or fibular sesamoid. In this setting, a positive Tinel’s sign may be elicited along the borders of the sesamoids. Patients may or may not note decreased sensation distal to the nerve compression.

Imaging Routine anteroposterior and lateral radiographs may not provide sufficient information with which to evaluate the sesamoids (Fig. 25H-33A and B). The tibial sesamoid is seen best on an oblique radiograph, with the MTP joint extended about 50 degrees (see Fig. 25H-33C). The x-ray beam is directed 15 degrees cephalad from a lateral position and is centered over the first metatarsal head. The fibular sesamoid is seen best on a lateral oblique radiograph (see Fig. 25H-33D), where it is shown in the first interspace between the first and second metatarsal heads. Often the most useful radiograph is the axial sesamoid view (see Fig. 25H-33E). Fragmentation of a sesamoid in osteochondri­ tis may be seen on the axial radiograph (Fig. 25H-34).

Radiographs are frequently normal despite subjective symptoms (sesamoiditis). In these cases, a bone scan may aid the diagnosis of sesamoid dysfunction.99,117 Increased uptake within the sesamoids may occur on bone scan before the development of any significant radiographic changes (Fig. 25H-35). MRI138 is useful in visualizing bone and soft tissue abnormalities of the sesamoid mechanism (see Box 25H-20).

Treatment Options Nonoperative Most sesamoid problems in the athlete can be treated effec­ tively without surgery (Box 25H-21). Decreasing activity such as running and implementing other aerobic activity may diminish symptoms. Decreasing the heel height of shoes will diminish pressure over the sesamoids. A molded insole or a metatarsal arch support is helpful in decreas­ ing plantar pressure in the sesamoid region for treatment of a fracture, sesamoiditis, or osteonecrosis. This type of support may also relieve the pressure of an intractable plantar keratosis (IPK). Patients can be instructed in shav­ ing the callus, which may alleviate symptoms as well. Oral NSAIDs may give relief of pain associated with sesamoid­ itis, osteonecrosis, or arthritis. A steroid injection occasion­ ally will help when there is localized inflammation or when sesamoiditis has occurred, but injections should be used judiciously. In the presence of a fracture or osteonecrosis, intra-articular injection is contraindicated. In patients with sesamoid arthritis, a stiff insole, a rocker-bottom sole, and metatarsal padding reduce metatarsal joint motion and can relieve pain with ambulation. In the case of sesamoid fractures, a walking cast or taping of the hallux into a neutral position (Fig. 25H-36) with a stiff-soled shoe may be used. Typically, 6 weeks of immobilization and activity restriction will allow resolu­ tion of symptoms and fracture healing. After 6 weeks, the athlete may gradually return to activities with repeat radiographs at 3 to 4 months after injury to evaluate healing.

Operative If conservative treatment has been unsuccessful, several surgical options are available depending on the pathology (see Box 25H-21). Although sesamoid abnormalities are infrequent, injury or degeneration of a sesamoid may lead to significant dysfunction of the MTP joint in athletes. Scranton and Rutkowski102 and Speed139 described excision of both sesamoids. Scranton102 advised a reap­ proximation of the surgical defect to minimize postopera­ tive disability. Helal98 and Inge and Ferguson97 cautioned against combined medial and lateral sesamoid excision because postoperative clawing of the hallux may occur with dual sesamoid excision. Jahss99 recommended repair of the defect in the flexor hallucis brevis after dual sesamoid exci­ sion on the theory that it was comparable to repairing the quadriceps mechanism after patellectomy. Occasionally, dual sesamoidectomy may be necessary (i.e., with infec­ tion); elective dual sesamoidectomy should be avoided because a claw-toe deformity may occur.

Foot and Ankle 2091

B

A

C

D

E

Figure 25H-33   A, Anteroposterior radiograph shows the sesamoids. B, Lateral radiograph shows sesamoids, although there is significant overlap that may conceal disease. C, An oblique radiograph shows the tibial sesamoid. D, A lateral oblique radiograph shows the fibular sesamoid. E, Axial view shows fracture of the lateral sesamoid. (A, From Coughlin MJ: Sesamoids and accessory bones of the foot. In Mann RA, Coughlin MJ [eds]: Surgery of the Foot and Ankle, 6th ed. St. Louis, Mosby, 1993; B-E, © M. J. Coughlin. Used by permission.)

For isolated arthritis of either the medial or the lateral sesamoid, surgical excision may be performed. When both sesamoids are involved, however, a combined medial and lateral sesamoid excision is likely to destroy the insertion of the flexor digitorum brevis and may lead to eventual claw­ ing of the hallux.97,99 Arthrodesis of the MTP joint may alleviate arthritic symptoms and provide a stable medial buttress to the first ray. Although pain may be relieved considerably with operative resection of an involved sesa­ moid, MTP joint motion may not improve when signifi­ cant arthritis is present. Depending on which sesamoid is to be resected, vari­ ous surgical approaches are available. The lateral or fibular sesamoid is approached through either a dorsal or a plantar

approach. Mann and Coughlin136 advocated a dorsal lateral approach to resect the fibular sesamoid in patients with hal­ lux valgus; they recommended this approach for isolated fibular sesamoid excision also.137 The fibular ­ sesamoid is resected through a dorsal first interspace incision (Box 25H-22; Fig. 25H-37).140,141 Inge and Ferguson97 and oth­ ers111,117 recommended a dorsal approach to avoid a pain­ ful plantar scar as well as injury to the plantar sensory nerve in the first interspace. If a plantar approach is used to resect the fibular sesa­ moid, an intermetatarsal incision is recommended (Box 25H-23; Fig. 25H-38). Van Hal and colleagues,113 Helal,98 and Jahss99 recommended a longitudinal plantar incision adjacent to the fibular sesamoid. Jahss99 found a dorsal

�rthopaedic ����������� S �ports ������ � Medicine ������� 2092 DeLee & Drez’s� O

A

B

Figure 25H-34   A, Radiograph shows fragmentation or osteochondritis. B, Anteroposterior radiograph shows osteochondritis of lateral sesamoid. (© M. J. Coughlin. Used by permission.)

approach for a sesamoidectomy difficult. If one wishes to approximate the surgical defect, a plantar approach is ­necessary. The medial sesamoid may be resected by either a plan­ tar medial incision (Fig. 25H-39A)98,113 or a direct medial incision (Box 25H-24).97,99,117,126 Kliman and associates117 advocated a direct medial approach to avoid the plantar digital nerves, which may be in close proximity to the sur­ gical dissection through a medial plantar approach. Mann and coworkers137 discouraged a plantar approach because of the possibility of developing a painful postoperative plantar scar. Although sesamoidectomy is occasionally necessary for IPKs, surgical shaving of the involved sesamoid may pre­ serve function in the athlete and allow more rapid recovery Box 25H-21 T  reatment Options for Sesamoid Dysfunction Nonoperative

• Nonsteroidal anti-inflammatory drugs • Decreased shoe heel height • Molded insole or metatarsal arch support • Shoe modifications (stiff insole, rocker sole) • Shaving of a painful callus Operative

Figure 25H-35   The arrow marks increased uptake in the medial sesamoid consistent with osteochondritis.

• Sesamoidectomy (tibial or fibular) • Sesamoid shaving • Screw fixation of sesamoid fracture • Autologous bone grafting

Foot and Ankle 2093

Figure 25H-36   Taping of the hallux minimizes dorsiflexion and may relieve symptoms. (© M. J. Coughlin. Used by permission.)

(Box 25H-25; Fig. 25H-40).142,143 Resection of the plantar half of the medial sesamoid may reduce a prominent tibial sesamoid and allow resolution of a plantar keratotic lesion. Sesamoid shaving or resection in the presence of a plantar flexed first ray is associated with a high rate of recurrence of the keratotic lesion.142,143 In this situation, a closing wedge dorsiflexion first metatarsal osteotomy is preferable. Anderson and McBryde reported on a series of sesamoid fracture nonunions treated with autologous bone grafting with good results in a series of 21 patients. 144 Autologous bone grafting may offer an alternative to sesamoid excision in some individuals. Other authors have described screw fixation of sesamoid fractures.145

Weighing the Evidence Surgical excision of the chronically painful sesamoid has been advocated by multiple authors.97-99,107,108,113-117,119-127,139,146-149 However, controversy exists about whether an accept­ able result can be achieved after sesamoidectomy. Lee and colleagues150 describe tibial sesamoidectomy as “safe and effective,” emphasizing meticulous surgical technique.

Box 25H-22 F  ibular Sesamoidectomy, Dorsal Approach

• Make a dorsal incision over the first interspace. • Retract the first and second metatarsals with a

Figure 25H-37   The fibular sesamoid is exposed through a dorsal incision. (© M. J. Coughlin. Used by permission.)

Ilfeld and Rosen116 reported excellent postoperative pain relief after surgical excision of the involved sesamoid in cases of sesamoiditis. Although Van Hal and colleagues113 and others111,126,151 reported acceptable results after hallu­ cal sesamoidectomy, Inge and Ferguson97 reported signifi­ cant dysfunction of the MTP joint after sesamoidectomy. The incidence of hallux varus deformity after fibular sesamoidectomy was about 8% according to Mann and Coughlin in their review of the McBride procedure. After isolated fibular sesamoidectomy, Coughlin146 and others137 reported a 5% varus drift of the hallux. Often residual pain exists after the removal of a diseased sesamoid. Zinman and colleagues114 and others113,126 noted complete resolution of pain and resumption of normal activities. Inge and Ferguson97 found that only 41% of their patients experienced complete relief of pain. ­Coughlin146 and Mann and coworkers137 reported complete relief of pain in only 50% of patients postoperatively. Box 25H-23 F  ibular Sesamoidectomy, Plantar Approach

• Make a longitudinal plantar incision in the first inter­ metatarsal space.

Weit­

• Identify the plantar digital nerve adjacent to the lateral

• Reflect the adductor hallucis off the lateral sesamoid. • Protect the continuity of the adductor tendon (to pre­

• Protect the plantar digital nerve. • Detach the capsular fibers of the flexor hallucis brevis. • Release the sesamoid proximally, laterally, and distally. • Incise the intersesamoidal ligament. • Remove the fibular sesamoid. • Inspect the tendon of the flexor hallucis longus for

laner retractor to sublux the lateral sesamoid. vent postoperative varus).

• Detach the intersesamoidal ligament. • Excise the fibular sesamoid from

the surrounding t­ issues. • Confirm that the adductor hallucis and the flexor hal­ lucis longus are still intact.

aspect of the hallux and the lateral sesamoid.

­continuity.

• Close the defect created by the sesamoid excision.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2094 DeLee & Drez’s� O

Box 25H-24 Tibial Sesamoidectomy

• Make either a plantar medial or direct medial incision. • Identify the medial plantar cutaneous nerve on the ­medial aspect of the sesamoid (see Fig. 25H-39A). the nerve with the plantar skin flap to avoid postoperative neuroma. • Incise the medial capsule to expose the articular surface of the sesamoid. • Dissect the sesamoid free of surrounding tissues. • Incise the intersesamoidal ligament (see Fig. 25H-39B). • Inspect the tendon of the flexor hallucis longus for ­continuity. • Close the defect created by the sesamoid excision.

• Retract

Box 25H-25 S  having a Prominent Tibial ­Sesamoid

• Make either a plantar medial or direct medial incision. • Identify the medial plantar cutaneous nerve on the

Figure 25H-38  A plantar incision to excise the lateral sesamoid should be located on the lateral aspect of the lateral sesamoid.(© M. J. Coughlin. Used by permission.)

Stiffness is a common complaint after sesamoid­ ectomy.99,117,120,125,126,137,146 Coughlin146 and Mann and coworkers137 noted that one third of patients had decreased MTP motion at long-term follow-up. Torg­ erson and ­ Hammond125 and Colwill120 noted decreased range of motion preoperatively but did not mention restricted motion after surgery. Jahss99 reported that surgical excision could lead to stiffness. Inge and Fergu­ son97 reported that 58% had restricted range of motion. Both Mann and coworkers137 and Coughlin146 reported a 60% incidence of plantar flexion weakness. Diminu­ tion of strength of the flexor hallucis brevis was hypoth­ esized by Van Hal and associates.113 Inge and Ferguson97 reported a 17% incidence of clawing in the hallux, but many of these patients had had dual sesamoid resection.

A

­medial aspect of the sesamoid (see Fig. 25H-40A). the nerve with the plantar skin flap to avoid postoperative neuroma • Retract the plantar fat pad and protect the flexor ­hallucis longus • Resect the plantar surface of the tibial sesamoid with a sagittal saw (see Fig. 25H-40B). • Bevel any remaining prominent edges of the sesamoid.

• Retract

Postoperative migration of the sesamoid has been noted to occur in 10% of cases.137,146 Nayfa and Sorto151 quan­ titated this drift to average 6.2 degrees in either varus or valgus postoperatively. Results of tibial sesamoid shaving routinely have been good. Aquino and colleagues142 reported on 26 feet in which tibial sesamoid shaving was performed for an IPK. Satisfactory results were reported in 89%. This high

B

Figure 25H-39  A, Excision of the medial sesamoid through a plantar medial approach. B, Location of the medial plantar nerve.  (© M. J. Coughlin. Used by permission.)

Foot and Ankle 2095

A

C

s­ uccess rate is probably due to the fact that the intrinsic musculature was not disrupted with this dissection. Mini­ mal weakness was observed at the first MTP joint. Mann and Wapner,143 reporting on their results of 16 tibial sesamoid shavings, noted no functional limita­ tions and good or excellent results in 94%. In one case, a

B

Figure 25H-40  A, A plantar medial incision is used to expose the medial sesamoid for shaving. B, Half of the tibial sesamoid is shaved to treat an intractable plantar keratosis. Care is taken to protect the flexor hallucis longus tendon. C, Radiograph after tibial shaving.  (© M. J. Coughlin. Used by permission.)

­recurrent callus developed. Mann and Wapner143 observed that there is a possibility of fracture if an excessive resec­ tion is performed, although this did not occur in the series. A tibial sesamoid shaving is contraindicated in the presence of plantar flexion of the first ray greater than 8 degrees.

Author’s Preferred Method Conservative treatment with shoe modifications and taping of the toe is preferred in most athletes with sesamoid dysfunc­ tion. For a diseased tibial or fibular sesamoid that is refractory to conservative care, surgical excision can be considered. The preferred method for a tibial sesamoid resection is a plantar medial approach. Repair of the surgical defect should be at­ tempted. A fibular sesamoid may be excised through either a dorsal or plantar approach. In the presence of a hallux valgus deformity, the dorsal approach provides excellent visualiza­ tion of the fibular sesamoid. The plantar approach is more direct in cases of normal alignment. With a plantar approach, a painful plantar scar can develop, and a patient should be in­ formed of this before surgery. Care must be taken not to dis­ rupt the lateral capsule of the first MTP joint or to ­disrupt the adductor hallucis brevis in order to maintain function

and stability of the first MTP joint. Although resection of a diseased sesamoid often achieves acceptable results, the athlete can experience continued symptoms, restricted mo­ tion, or diminution of strength. An attempt should be made to avoid a dual sesamoid resection unless absolutely ­necessary. Despite careful dissection and postoperative care, it is common for 50% of patients to have continued symptoms after surgical resection. About 50% of patients may note diminution of plantar flexion strength, and restricted range of motion may be experienced. Postoperative physical ther­ apy exercises to rehabilitate plantar flexion strength and to achieve greater range of motion are helpful in the recovery of an athlete.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2096 DeLee & Drez’s� O

Postoperative Prescription, Outcomes Measurement, and Potential Complications After sesamoid excision, a soft compression dressing is used, and the patient is allowed to ambulate in a postop­ erative shoe. The toe is taped into appropriate alignment for 6 to 8 weeks after surgery. After successful healing of the surgical incision, range-of-motion activities may be started in the athlete. Although postoperative mobiliza­ tion of the MTP joint may be initiated within 2 to 3 weeks after surgery, it is still important to tape the toe in correct alignment to minimize postoperative drift. By 6 to 8 weeks, aggressive walking activity can be initiated, and by 8 to 10 weeks, jogging and running activity can be instituted. Postoperative dressings and later taping are important to protect the hallux from either medial migration of the hallux (hallux varus) after a fibular sesamoid resection or development of hallux valgus after a medial sesamoid resection. Nayfa and Sorto151 reported a 42% incidence of hallux valgus after medial sesamoidectomy. Mann and colleagues137 reported a 5% incidence of valgus drift after tibial sesamoid resection. Inge and Ferguson97 found that impaired motion, defor­ mity, and pain occurred in many patients after sesamoid­ ectomy. The development of postoperative IPKs beneath a remaining sesamoid,136,137 valgus deviation after tibial sesamoidectomy,99,137,151,152 varus deviation after fibular sesamoidectomy,99,117,137 development of restricted motion in the MTP joint,97 and weakening of plantar flexion strength of the MTP joint97 all have been reported. Claw­ ing of the hallux after a medial and lateral sesamoidectomy has been noted.97,102 The development of a postoperative neuroma after surgery may be more symptomatic than the patient’s presurgical complaint. Taping of the toe is usually not necessary after shaving of the tibial sesamoid because the MTP joint is not entered. After tibial sesamoid shaving, a soft dressing is used, and the patient is allowed to ambulate in a postoperative shoe for 3 weeks. Range of motion exercises are started as soon as wound healing is complete. Usually the patient can start aggressive walking the fourth week. Jogging and running are initiated 6 weeks after surgery. The athlete may gradu­ ally return to full activity as pain and swelling allow.

Criteria for Return to Play Return to play after surgical intervention for sesamoid dys­ function is typically between 2 and 4 months after surgery. When shaving of the tibial sesamoid is performed, the

Box 25H-26 C  riteria for Return to Play After Sesamoid Surgery

• Typically expected 2 to 4 months postoperatively • Return to play earlier (2 months) with sesamoid ­shaving

• Athlete

must be running with resolution of pain and swelling.

athlete is allowed to resume running about 6 weeks after ­surgery. With excision of a sesamoid, running is delayed until 8 weeks after surgery, when taping of the toe is dis­ continued. Once the athlete is running without pain or swelling, gradual return to play is allowed (Box 25H-26).

Special Populations Sesamoid dysfunction has been reported in both danc­ ers and gymnasts.153,154 Any sport that requires repeti­ tive impact loading of the forefoot may place an athlete at increased risk.

INGROWN TOENAILS Onychocryptosis, or ingrown toenail, occurs with great fre­ quency in young athletes. Trauma such as stubbing the toe can lead to inflammation or infection of the toenail edge.155 Cutting the toenails in a curved fashion and tearing them often leads to ingrowth with subsequent infection. As the toenail grows out, the corner of the advancing nail plate impinges on the lateral nail groove.156 Tight-fitting shoes can cause increased pressure between the nail and the nail fold, leading to ingrowth of the toenail.156,157

Anatomy and Biomechanics The toenail, or the nail plate, is composed of several layers of dense overlapping keratinized cells. The nail plate consists of three layers, each originating from a different area of the nail unit. The thin dorsal layer is relatively stiff or brittle and covers the thicker middle layer. The deep layer is believed to be derived from the nail bed.158 The nail is supported by the nail unit, an area of epithelial tissue that is divided into four components (Box 25H-27; Fig. 25H-41).159 The nail plate lies on the nail bed, a roughened epithe­ lial surface that consists of longitudinal grooves that inter­ digitate with corresponding grooves on the undersurface of the toenail. This interdigitation creates a firm bonding of the nail plate to the nail bed. The nail bed is believed to contain some germinative cells that produce a portion of the nail plate.158,160-162 At the distal end of the nail bed, as the nail bed and nail plate separate, there is a smooth bor­ der of skin (the hyponychium), which forms a seal between the distal end of the nail and the nail bed. On the tibial and fibular borders of the toenail, the nail plate is surrounded by epidermal skin folds called lateral

Box 25H-27 Components of the Nail Unit

• Nail bed: a roughened epithelial surface that consists of

longitudinal grooves that interdigitate with correspond­ ing grooves on the undersurface of the nail • Hyponychium: smooth border of skin at the distal end of the nail bed, which forms a seal between the distal end of the toenail and the nail bed. • Proximal nail fold: a complex structure that participates in the germination of the nail plate • Nail matrix: main germinal area of the nail

Foot and Ankle 2097 Nail plate Lunula Lateral nail groove

Cuticle

Matrix

Proximal nail fold

A Cuticle

Lateral nail groove Nail bed

Proximal nail fold

Lunula

Distal phalanx

Nail plate

Eponychium

Distal groove

B

Hyponychium

Matrix

Figure 25H-41   A, Germinative layer of nail matrix. B, Cross section of toe shows nail bed, nail matrix, toenail, hyponychium,   and eponychium.

nail folds. The base of the nail is covered by the proximal nail fold, a complex structure that participates in the ger­ mination of the nail plate.162 The dorsal surface of the nail fold is composed of the skin on the dorsal surface of the toe. On the plantar surface of this fold, the eponychium forms a thin surface that attaches to the nail plate. The distal surface of the two components of the proximal nail fold forms the cuticle. The nail matrix is the main germinal area for the toenail. The matrix extends from a point just distal to the lunula, as far laterally as the entire width of the nail plate155 and 8 mm proximally to the edge of the cuticle. The matrix closely borders the insertion of the long extensor tendon.158 Nail matrix is seen beneath the nail plate as the lunula, the opaque crescent-shaped area at the base of the toenail. Dis­ tally, the nail matrix is contiguous with the nail bed. The germinal matrix is covered by a small epidermal surface and does not have the epidermal ridges characteristic of the nail bed. At the proximal margin of the matrix, there appears to be a small area on the plantar surface of the proximal nail fold that contributes to nail plate growth.158,163,164 The major area of matrix germination occurs between the apex of the matrix and the distal border of the lunula. The area covered by the proximal nail fold forms the thin dorsal layer of the nail plate.158 The distal area of the matrix (the lunula) produces the thicker and softer area of the mid­ dle portion of the nail plate.158 Microscopic toenail matrix contained within the nail folds and the distal nail bed160,161 produces a thin layer of the ventral toenail plate, a factor that occasionally causes postoperative ­recurrence.155,158,163 Biomechanical imbalance may be a cause of ingrown toenails in athletes. Tight-fitting shoes causing pressure at either the medial or lateral nail folds can result in infection of the toenail margins of the great toe. Pronation ­deformity of the hindfoot can also cause pressure on the medial nail

fold. Trauma to the great toe with formation of a subun­ gual hematoma often leads to the loss of a toenail.

Classification Heifetz156 proposed a classification of ingrown toenails into three distinct stages (Box 25H-28).

Evaluation Clinical Presentation and History A careful history should be taken to elicit any familial traits or genetic factors that predispose an athlete to congeni­ tal nail deformities. Prior toenail problems may indicate extrinsic factors such as ill-fitting footwear, a problem related to the type of athletic activity, or other causes of toenail abnormalities. Recent stubbing or trauma to the toe may indicate the cause of toenail pathology. Changes in footwear or athletic shoes can also contribute to the development of toenail problems.

Box 25H-28 Classification of Ingrown Toenails Stage I: Swelling and erythema occur along the lateral fold. The edge of the nail plate may be embedded in an irritated nail fold. Stage II: Acute or active infection is characterized by increased pain. Drainage is present. Stage III: With chronic infection, granulation tissue may develop in the lateral nail fold. Hypertrophy of the ­surrounding soft tissue is present.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2098 DeLee & Drez’s� O

Physical Examination In evaluating a toenail, physical examination of the entire foot and lower extremity is important. Palpation of arte­ rial pulses and evaluation of the skin are required because vascular insufficiency may lead to toenail abnormalities. Examination of the skin also provides information about fungal infections, ulcerations, blisters, or other sources of skin lesions. The patient’s gait should be observed to look for gait patterns that subject the toenails to undue pressure. The postural position of the foot may show pes planus, pes cavus, or other anatomic abnormalities. Swelling and erythema may be present at the nail fold. In acute infection, purulent drainage is present, often with significant tenderness along the nail fold. Granulation tis­ sue and hypertrophy of the surrounding soft tissue occur in chronic infections (Box 25H-29).

Imaging Radiographic evaluation of the involved toes is warranted in patients with recurrent toenail infections or toenail abnormalities. Evaluation of the distal phalanx by radio­ graph helps to rule out osteomyelitis in a chronically infected hallux. A subungual exostosis is best seen on a lateral radiograph and can be an explanation for chronic toenail elevation, ulceration, or infection.

Treatment Options Nonoperative Stage I lesions of the nail frequently respond to conserva­ tive care. Appropriate footwear is essential in preventing recurrence. Effective treatment requires elevation of the lateral toenail plate from the inflamed nail fold. A wisp of cotton is inserted carefully beneath the edge of the nail plate, taking care not to fracture the nail plate (Fig. 25H42A). Frequently a digital anesthetic block is required to reduce discomfort when the nail edge is elevated. When an adequate length of the nail plate has regrown, the patient is instructed to cut the nail in a transverse fashion. Tearing or picking a nail or cutting the nail in a curved fashion can result in recurrent infection. Diagonal trimming of a nail helps diminish the inflammation of an acute infection, but it is merely a temporary treatment and is often associated with recurrent infection as the nail plate grows (see Fig. 25H-42B). A subungual hematoma can result from a single trau­ matic episode or repetitive stress of the toe striking the end of the toe box in tennis, jogging, or other athletic activities.

Box 25H-29 T  ypical Findings in Ingrown ­Toenails

• Swelling • Erythema • Tenderness to palpation at the nail fold • Purulent drainage in acute infection

A

B

Figure 25H-42   A, Elevation of toenail with cotton placed under edge. B, Diagonal trimming of toenail edge relieves pressure on adjacent soft tissue. (From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 56.)

If acute pain develops, decompression of the hematoma by penetrating the nail plate with either a small nail drill or a hot paper clip alleviates discomfort significantly. As this nail detaches, a new nail grows beneath the old one. As the new nail grows distally, attention must be directed to the re-establishment of a new lateral nail fold so that an ingrown toenail does not develop. Painless discoloration underneath the toenail with no history of trauma may be an asymptomatic melanoma, and the toenail should be reinspected at a later date. In this situation, a biopsy is necessary to further evaluate the possibility of a subungual melanoma.

Operative To be effective, treatment of ingrown toenails must be tai­ lored to the individual patient. Heifetz156 stated that only during stage I can conservative means be used to obtain a cure. When conservative care is unsuccessful or when acute (stage II) or chronic (stage III) infection occurs, more aggressive care is necessary (Box 25H-30). A digital anes­ thetic block is performed routinely before any toenail pro­ cedure (Fig. 25H-43). A small Penrose drain may be used as a tourniquet if needed.

Box 25H-30 S  urgical Treatment of Ingrown Toenails

• Partial nail plate avulsion • Complete nail plate avulsion • Plastic nail wall reduction • Partial onychectomy • Complete onychectomy • Syme amputation of the toe • Phenol and alcohol matrixectomy

Foot and Ankle 2099

Partial Onychectomy (Winograd or Heifetz Procedure)

A C

D

B Figure 25H-43   Digital block of great toe. A, Injection of medial aspect of toe. B, Vertical injection techniques. C, The needle is turned in a horizontal direction. D, A final injection is used to anesthetize the lateral aspect of the toe.

Partial Nail Plate Avulsion After elevating the outer edge of the nail plate proximally to the level of the cuticle, a scissors or small bone cutter is used to section the nail longitudinally. Only the outer edge of the nail is resected, removing as narrow a section of nail as possible. The toe should be examined closely to ensure that no spike of the nail plate remains in the lateral nail fold because this would continue to incite a foreign body reac­ tion. After removal of the nail edge, the acute or chronic infection often subsides (Fig. 25H-44).

Complete Toenail Avulsion When more extensive infection is present, a complete toe­ nail avulsion may be necessary. After the nail plate is ele­ vated proximally, the toenail is avulsed. Re-­epithelialization of the nail bed occurs in 2 to 3 weeks.155 As the nail plate regrows, the advancing edges must be elevated, or an ingrown toenail can develop (Fig. 25H-45).165

Plastic Nail Wall Reduction A plastic nail wall reduction is performed only after an acute infection has resolved following either suc­ cessful conservative care or partial or complete toe­ nail avulsion. This technique is often used in younger patients.155,166,167 A wedge of tissue from the lateral toenail pulp, including skin and subcutaneous tissue, is excised from the medial or lateral border or from the tip of the toenail. The skin is approximated loosely with interrupted nylon sutures. Often this procedure reduces the prominent lateral toenail fold, removing the source of impingement of the lateral toenail edge (Fig. 25H-46).

Partial onychectomy is performed only after an acute infection has resolved, usually following partial nail plate avulsion.156,168,169 A vertical incision is made along the edge of the toenail (usually where the previous nail plate edge was avulsed). The resection is carried distally to the terminal extent of the nail matrix (the Heifetz proce­ dure)156 or the terminal extent of the nail bed (the Wino­ grad procedure)168 and proximally to the proximal edge of the nail matrix. An oblique incision is made at the apex of the nail bed, and the proximal nail matrix and edge of the cuticle are removed. Care must be taken not to injure the extensor tendon insertion and not to violate the interphalangeal joint. The germinal matrix is character­ ized by a pearly white color and smooth texture.155 The resection is carried into the nail fold laterally. The cortex of the distal phalanx is stripped of remaining matrix, and the skin is approximated with interrupted nylon sutures (Fig. 25H-47).

Complete Onychectomy (Zadik Procedure) Occasionally, a patient desires a complete and permanent removal of a toenail. The procedure usually is performed after an initial toenail avulsion when previous infection has been present.170 It is advantageous to delay surgery until the infection and inflammation have subsided completely. An oblique incision is made at the medial and lateral api­ ces of the nail bed. The toenail is avulsed. The cuticle, eponychium, and proximal nail bed are excised completely, and the matrix is excised proximal to the cuticle and distally as far as the lunula. The resection includes each nail fold as well. The skin is approximated with interrupted nylon sutures loosely closing the proximal edge of the remaining nail bed (Fig. 25H-48). Excess tension on the suture line may lead to sloughing of the skin.

Syme Amputation (Thompson-Terwilliger Procedure) For symptomatic recurrence or when the patient requests a more reliable ablation, Thompson and Terwilliger171 described a Syme amputation of the distal phalanx. An ellip­ tical incision is used to resect the entire nail bed, matrix, lateral and proximal nail folds, cuticle, and proximal border of skin. About half of the distal phalanx is removed, and the remaining edges of bone are beveled. Skin edges are approx­ imated with interrupted nylon sutures (Fig. 25H-49).

Phenol and Alcohol Matrixectomy A phenol and alcohol matrixectomy may be used rather than surgical resection.155,172,173 The advantage of this technique is that it can be performed in the presence of concurrent infection. The nail plate edge is avulsed as pre­ viously described. The toenail groove and nail matrix are curetted. A cotton-tipped applicator is used for the proce­ dure, but most of the cotton is removed from the applica­ tor so that it can fit along the area of the lateral nail fold where the toenail has been avulsed and into the matrix area below the eponychium. The cotton-tipped applicator is moistened with 88% fresh carbolic acid (phenol), and the

�rthopaedic ����������� S �ports ������ � Medicine ������� 2100 DeLee & Drez’s� O

A

B

C

D

E

excess is blotted on a gauze pad. The applicator is inserted into the nail groove and matrix area for 1 minute. The applicator is removed, and the area is flushed with alcohol. After this application, two subsequent 30-second phenol applications are performed. After each application, alcohol is used to flush the nail groove and matrix region. Any area of skin that should be protected is covered with petroleum jelly to prevent injury to the tissue (Fig. 25H-50).

Weighing the Evidence The most frequent complication after attempted toe­ nail ablation is recurrence. After partial nail avulsion, the recurrence rate with subsequent infection is 47% to 77% when further definitive care is deferred.165,174 ­ Murray175

Figure 25H-44   Avulsion of edge of toenail. A, Stage III infection of medial edge of toenail with granulation tissue and purulence. B, Elevation of edge of toenail. C, Longitudinal sectioning of toenail. D, Avulsion of toenail edge. E, Typical postoperative compression dressing. (From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, pp 57, 61.)

estimated that recurrent infection occurred in 64% of cases after the initial nail avulsion and in 86% of cases after a sec­ ond toenail avulsion in the absence of definitive treatment. Although a toenail avulsion gives dramatic relief not only of an infection but also of symptoms, it has a poor longterm cure rate, and usually a second procedure is needed. Dixon155 noted a much higher recurrence rate in patients in whom multiple avulsions of a single toenail have been performed. Lloyd-Davies and Brill165 reported that within 6 months after nail avulsion, 31% of patients required further treatment. Palmer and Jones176 reported a 70% recurrence rate after a nail plate avulsion. Plastic nail wall reduction has a reported recurrence rate of 25%.174 After partial onychectomy with either a Winograd or a Heifetz procedure, recurrence rates of 6% to 33% have

Foot and Ankle 2101

A

B

C

excision is necessary. The cause of this recurrent nail plate growth may be germinal cells within the nail bed. Pettine and associates178 found that the Heifetz proce­ dure had a 90% rate of patient satisfaction and a recurrence rate of 6%. Nail excision with chemical matrixectomy had an 80% rate of patient satisfaction and a 20% recurrence rate. Fewer patients underwent complete nail ablation and a Syme procedure; for both techniques, a satisfaction rate of about 80% to 90% was reported. Shaath and associ­ ates182 compared recurrence after complete ablation with the Zadik procedure with chemical ablation. They found a recurrence rate of 60.5% in the patients treated surgically compared with 15.6% after chemical ablation. Murray and Bedi177 concluded that because of the recurrent nail plate regrowth after either a Winograd or partial or complete nail plate ablation a terminal Syme procedure was the pro­ cedure of choice.

Author’s Preferred Method Figure 25H-45   Complete toenail avulsion for infection. A, Nail is elevated with sharp dissection. B, Nail is avulsed. C, Remaining nail bed and matrix after avulsion.

been reported.168,174,176-178 Murray and Bedi177 noted a 27% recurrence rate after an initial Winograd procedure and a 50% recurrence rate if a double Winograd proce­ dure was performed. Gabriel and colleagues179 reported a 1.7% recurrence rate with a 79% satisfaction rate follow­ ing this procedure. Wadhams and coworkers180 reported the development of 10 epidermal inclusion cysts after 147 partial matrixectomies (7%) at an average of 5.5 months after surgery. Complete toenail ablation procedures are associated with a recurrence rate of 16% to 50%.176,177,181 All patients should be informed that minor recurrence of toenail tissue can emanate from the remaining toenail bed. Although this regrowth is not a significant problem, occasionally a repeat

The goal in treatment of toenail infections is to control the infection initially. A partial toenail avulsion is preferable in the presence of acute or chronic infection. Once the infec­ tion has resolved, definitive surgery can be contemplated, but surgery can be deferred depending on the patient’s de­ sire to participate in competitive sports. A plastic nail wall reduction is useful for younger pa­ tients with mild disease. Excision of the toenail edge and matrix (the Winograd procedure) is also preferable because of its high rate of predictability. Complete nail ablation is complicated by a high rate of recurrence as a result of the presence of microscopic nail matrix in the nail bed. This procedure is preferable for a chronic toenail abnormality or for recurrent toenail infections that have not been ame­ nable to more conservative methods. A Syme amputation is associated with a low rate of recurrence but is a more radi­ cal procedure and probably is less cosmetically acceptable to patients. Syme amputation should be reserved for more chronic toenail problems in older patients.

Postoperative Prescription, Outcomes Measurement, and Potential Complications A

B

C

D

Figure 25H-46   Soft tissue plastic repair to decompress lateral nail fold. A, Cross section of plastic toenail repair. B, Lateral view shows elliptical skin incision and soft tissue excision. C, Cross section after closure of skin and soft tissue ellipse. D, Lateral view after closure of ellipse.

Postoperative care after a toenail avulsion requires initial hemostasis, a compressive dressing, and efforts to reduce inflammation and infection. Usually, removal of the toe­ nail edge allows rapid resolution of inflammation. Prophy­ lactic antibiotics are used in more severe infections, but for the most part, antibiotics are not necessary. Soaks in tepid salt solution should be initiated within 24 to 48 hours after toenail avulsion. Aftercare is important after a partial toenail avulsion. As the toenail regrows, the advancing edge is at risk for recurrent infection. A cotton wisp must be placed beneath the nail plate to elevate it as the edge grows distally. A digital block may be necessary to replace the initial pack­ ing. The patient is instructed in the packing procedure,

�rthopaedic ����������� S �ports ������ � Medicine ������� 2102 DeLee & Drez’s� O

A

B

C Figure 25H-47   Winograd procedure. A, Longitudinal incision of toenail matrix with oblique incision at apex. B, After excision of toenail matrix. C, Postoperative photograph shows bilateral Winograd procedure. (From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 63.)

and ­continues to perform the nail edge packing until the advancing nail edge has grown past the distal edge of the nail groove. With more definitive procedures such as complete or partial toenail ablation or a Syme amputation, a compres­ sion dressing is applied and changed 24 hours postopera­ tively. Repeat dressings are changed intermittently until there is no further drainage. Then a bandage is used to protect the toe. Sutures are removed 2 to 3 weeks after surgery. A sterile dressing is applied and changed daily after phe­ nol and alcohol matrixectomy. The patient is instructed to

soak the foot daily in a tepid salt solution. Daily soaking continues until the drainage has subsided, usually between 2 and 8 weeks.

Criteria for Return to Play In athletes, a partial or complete toenail avulsion is used to resolve an infectious process quickly, allowing the patient to return to competitive athletics within a short time. A patient is allowed to return to athletic partici­ pation after the inflammation has subsided and drainage is minimal. Aggressive running, jumping, and cutting

Foot and Ankle 2103

A

B

C

D

E

Figure 25H-48   Toenail ablation. A, Incisions are made at the apex of the lateral nail grooves, and the cuticle is excised.   B, The toenail is avulsed. C, Toenail matrix and lateral nail fold are excised. D, Skin edges are loosely approximated. E, Occasionally, recurrence of toenail tissue may develop after an attempt at complete toenail ablation. (From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991.)

and contact sports are possible after the inflammation ­subsides. Definitive surgery, such as partial or complete toe­ nail ablation or a Syme amputation of the toe, should be deferred until adequate time for healing is available. These

procedures are reserved best for the off-season or when an athlete can afford several weeks of down time. When definitive surgery is performed, the patient must be willing to avoid significant running activities until adequate heal­ ing has occurred (3 to 6 weeks) (Box 25H-31).

�rthopaedic ����������� S �ports ������ � Medicine ������� 2104 DeLee & Drez’s� O

B

A

C

D

E Figure 25H-49   A, Chronic incurved toenail. B, Proposed excision. C, Cross section shows area of resection. D, After Syme amputation of the great toe. E, Development of inclusion cyst with recurrence after attempted Syme amputation. (A and E, From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991; B to D, © M. J. Coughlin. Used by permission.)

Foot and Ankle 2105

Box 25H-31 C  riteria for Return to Play After Treatment of Ingrown Toenail

A

• Resolution of pain and inflammation after conservative

care or partial or complete toenail avulsion wound healing after definitive nail ablation procedures (usually 3-6 weeks)

• Complete B

often causes discomfort,195 and athletic activity tends to exacerbate the symptoms. The association of trauma with the later development of subungual exostosis, as suggested by Miller-Breslow and Dorfman183 and Bendel,190 may indicate a high risk for this abnormality in athletes. Many patients describe a history of pain that is aggravated by walking or running. This pain is due to the pressure of an expanding lesion beneath the nail plate that is traumatized by extrinsic pressure from either the end or the medial aspect of the toe box. Figure 25H-50   Phenol ablation of toenail edge.   A, Curettage of nail bed and matrix. B, Phenol application with cotton-tipped applicator along nail edge.

SUBUNGUAL EXOSTOSIS A subungual exostosis is a benign tumor of the distal phalanx most commonly occurring in the great toe.183-186 An osteo­ chondroma has a similar clinical presentation; however, the two entities are characterized by differences in histology,187 age of occurrence,183,185,188,189 and ­etiology.183,188,190-193

Anatomy and Biomechanics The histologic differentiation of the cartilage cap is the primary distinguishing factor between a subungual exos­ tosis and a subungual osteochondroma.187 A subungual exostosis is composed of a fibrocartilaginous cap with reac­ tive fibrous growth and cartilage metaplasia.183,187 Jahss194 suggested that the histology is characteristic of chronic irritation. A trabecular bone pattern typically underlies the fibrocartilaginous cap and connects with the distal pha­ lanx.183 In a subungual osteochondroma, hyaline cartilage composes the cartilaginous cap; however, a trabeculated bone pattern often underlies the cap. Ippolito and colleagues189 reported a high incidence of subungual exostosis in individuals participating in sports. In 21 patients with subungual exostosis, 12 were noted to be dancers, gymnasts, cyclists, or football players. Pressure of the toe box against the dorsal aspect of the nail plate

Classification A specific classification system for subungual exostoses has not been described. However, several key factors help dis­ tinguish a subungual exostosis and subungual osteochon­ droma (Table 25H-3).183,185,187-193,196

Evaluation Clinical Presentation and History The patient often gives a history of discomfort with exces­ sive walking, jogging, or running. An athlete may give a history of traumatic injury to the toe with later develop­ ment of pain (Box 25H-32).

Physical Examination On physical examination, direct pressure on the dorsal aspect of the toenail causes discomfort. A subungual hem­ orrhage can occur and is often mistaken for a chronic sub­ ungual hematoma, when in fact the nail has been elevated by an underlying exostosis.190 Pressure of an underlying subungual exostosis often separates the toenail from the toenail bed.195 Occasionally, the mass will penetrate the nail plate and can be mistaken for a malignant tumor (see Box 25H-32).186 A subungual exostosis is frequently misdiagnosed or confused with other toenail abnormalities.189 Elevation of the nail bed can resemble a subungual hematoma or chronic onychomycosis (Fig. 25H-51). Differential ­ diagnosis

  TABLE 25H-3  Subungual Exostosis Versus Subungual Osteochondroma Age (yr)

Gender

Etiology

Histology

Radiographic Features

Subungual exostosis

20-40

Female

Trauma

Fibrocartilage cap

Subungual osteochondroma

10-25

Male

Trabeculated bone from dorsomedial tuft Trabeculated bone from juxtaepiphyseal region

Hyaline cartilage

�rthopaedic ����������� S �ports ������ � Medicine ������� 2106 DeLee & Drez’s� O

Box 25H-32 T  ypical Findings in Patients with Subungual Exostosis

deformed toenail in a younger athlete may be an indication of a subungual exostosis. Radiographic examination is the key to accurate diagnosis.

• Pain occurs with activities such as walking or running. • Discomfort is elicited on examination with direct

Treatment Options

­pressure on the dorsal nail.

Nonoperative

actual size due to the fibrocartilaginous cap.

An asymptomatic subungual exostosis seen on radiographs does not require intervention. Once the exostosis becomes symptomatic, surgical treatment is preferred.

• Lateral or oblique radiographs best show the lesion. • The bony portion of the lesion may be smaller than the includes subungual verruca, pyogenic granuloma, glomus tumor, melanotic whitlow, keratoacanthoma, carcinoma of the nail bed, subungual nevus, subungual melanoma, and myositis ossificans.183,184,188,189,193,197

Imaging The key to the correct diagnosis is radiographic demon­ stration of the exostosis.198 Lateral or oblique radiographs are most helpful when the anteroposterior radiograph fails to show the lesion. The exostosis arises from the dorsal medial aspect of the distal phalanx (see Fig. 25H-56). It is often oval in shape and irregular in density.195 The trabec­ ular pattern is usually contiguous with the distal phalanx. Often, if the cartilage cap is of significant size, radiographs show a bony lesion much smaller than the actual size of the growth (see Box 25H-32). The major reason for delay in diagnosis of a subungual exostosis is the misdiagnosis of the condition as a toenail deformity or as onychomycosis. The presence of a painful

Operative Surgical resection of a subungual exostosis is the most fre­ quent treatment. A partial or complete toenail avulsion is performed depending on the size and location of the exos­ tosis (Fig. 25H-52). A longitudinal incision is made in the nail bed, and the nail bed is reflected from the exostosis. Care should be taken to avoid damage to the toenail matrix. The exostosis is resected with an osteotome or bone cut­ ter, and the base is curetted. The nail bed is relocated and sutured when possible. The wound is covered with a com­ pression dressing to encourage hemostasis.

Weighing the Evidence DePalma and colleagues199 reported successful resec­ tion in 11 cases. Recurrence of the exostosis, although not a common occurrence, results if incomplete resection occurs188,189 or if a continuing source of irritation is present. Miller-Breslow and Dorfman183 noted a 53% incidence of Deformity of toenail

A

B

C

Figure 25H-51   A, Subungual exostosis may elevate the toenail and may be misdiagnosed as a chronic fungal infection.   B, Anteroposterior radiograph shows subungual exostosis (same patient). More commonly, a lateral or an oblique radiograph may be used to show the exostosis. C, Often, the toenail is deformed. The exostosis often deviates in a medial direction. (B and C, From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991.)

Foot and Ankle 2107 Figure 25H-52   Resection of subungual exostosis.   A, The toenail is avulsed. B and C, The toenail matrix is incised longitudinally, then peeled off the exostosis to allow adequate exposure. D, The exostosis is resected at its base, and the base is curetted (see line of resection).

A

C

B D recurrence when subtotal excisional biopsy was performed. With wide local excision and curettage of the base, a 5% to 6% rate of recurrence can be expected.192 Fickry and coworkers200 reported a recurrence in 4 of 28 cases without evidence of malignant transformation. Postoperative toenail deformity can occur if the surgical dissection damages the nail matrix. Ippolito and associates189 reported a 66% rate of nail deformity or absence of nail growth after resection of a subungual exostosis. Although postoperative infection is a possibility with surgical resec­ tion of the exostosis, the infrequent occurrence of infection is most likely due to the fact that these lesions are not asso­ ciated routinely with active infection or ingrown toenails.

Author’s Preferred Method Partial or complete toenail avulsion followed by a wide surgical excision and curettage of the base is the preferred treatment. A patient should be informed that surgical excision may be complicated by recurrence or by postop­ erative toenail deformity caused by toenail matrix injury.

Postoperative Prescription, Outcomes Measurement, and Potential Complications Weight-bearing in a postoperative shoe begins immedi­ ately after the procedure. The surgical dressing is changed at the first postoperative visit preferably 1 to 3 days after surgery. Dressing changes are continued until the wound is

clean and dry, and the nail bed is protected until ­tenderness resolves. Sutures are removed 2 weeks after surgery. Ath­ letic shoes are permitted once the nail bed is nontender and the postoperative wound is healed.

Criteria for Return to Play Gradual return to sport occurs when the wound is healed and all pain, swelling, and drainage resolve. This usually occurs 4 to 6 weeks after surgery.

INTRACTABLE PLANTAR KERATOSES An IPK is a localized callosity occurring on the plantar aspect of the foot.201 An isolated keratotic lesion typically develops beneath a bony prominence as a direct result of increased pressure or friction. Establishing the correct diagnosis is essential to treatment. Care must be taken to distinguish an IPK from other lesions of the plantar skin (Box 25H-33; Figs. 25H-53 to 25H-55).202,203 Box 25H-33 D  ifferential Diagnosis of ­Intractable Plantar Keratosis

• Verrucae plantaris, plantar wart (see Fig. 25H-53) • Seed corn (small, well-circumscribed) • Diffuse callous formation (see Fig. 25H-54) • Discrete well-localized callous formation (see Fig. 25H-55) • Epidermal cysts • Blistering or ulceration due to vascular insufficiency

�rthopaedic ����������� S �ports ������ � Medicine ������� 2108 DeLee & Drez’s� O

A

B

Figure 25H-53   A, Verruca plantaris, or wart. Warts usually are located in areas other than beneath the metatarsal head (e.g., heel region). B, Wart beneath the third metatarsal head. (A, © M. J. Coughlin. Used by permission; B, from Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 86.)

Generalized callus can develop in the forefoot of an athlete as a result of increased pressure. It is normal for a moderate amount of callus to form. In contrast, an IPK is a well-localized keratotic lesion in an area of mechanical irritation such as friction or pressure.201 Lesions may be either discrete or diffuse.

Anatomy and Biomechanics Anatomic abnormalities often lead to callus formation. A plantarflexed metatarsal (Fig. 25H-56),204,205 an elongated metatarsal in relation to adjacent metatarsals,201,204,206 or a malunion of a metatarsal after fracture204 can cause increased pressure beneath a metatarsal and subsequent callus forma­ tion. Positional deformities204 (Box 25H-34; Fig. 25H-57) often lead to malalignment manifested by increased pres­ sure beneath one or more metatarsal heads. Jahss204 noted that minor physiologic variations in the athlete can lead to development of callosities as a result of the repeated stresses

Figure 25H-54   Diffuse plantar keratosis. (© M. J. Coughlin. Used by permission.)

that are involved with particular sporting activities. The repetitive stresses of athletic activity such as running, jog­ ging, or walking increase mechanical irritation to vulner­ able areas, frequently leading to increased symptoms. The first, fourth, and fifth metatarsal-tarsal articula­ tions allow considerable flexibility compared with the more rigid second and third metatarsal-tarsal articula­ tions.204 Reduced flexibility in the middle rays can increase susceptibility to pressure buildup beneath the metatarsal heads. The length of individual metatarsals is variable, but in some situations, an abnormally long second metatarsal (Fig. 25H-58) or second and third metatarsals subject a forefoot to increased weight-bearing, leading to symptoms of increased pressure.207-209 Midfoot flexibility and alignment can have a consider­ able effect on the distal metatarsals. A pes planus configu­

Figure 25H-55   A well-circumscribed intractable plantar keratosis resulting from a prominent fibular condyle. (From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 86.)

Foot and Ankle 2109

Box 25H-34 P  ositional Deformities ­Associated with Intractable Plantar ­Keratosis Formation

• Ankle equinus • Forefoot equinus • Cavus midfoot coupled

with a rigid forefoot (see Fig. 25H-62) • Rotatory deformity of the midfoot • Varus or valgus malalignment of the forefoot

Classification Figure 25H-56   Plantar flexed first metatarsal.   (© M. J. Coughlin. Used by permission.)

ration presents a flexible forefoot that rarely is associated with plantar keratoses.204 However, a cavus midfoot is characterized by decreased flexibility and is associated with a higher incidence of plantar keratoses. A rotary mid­ foot deformity (Fig. 25H-59) can result in varus or valgus malalignment. A varus abnormality often causes increased pressure on the fifth ray, whereas a valgus malalignment is characterized by increased pressure beneath the first ray. Forefoot equinus can occur without a significant hindfoot deformity, presenting as a rigid forefoot that is less resis­ tant to pressure beneath the metatarsal heads. In differentiating a plantar wart from an IPK, there are significant differences not only in the cutaneous ­presentation but also in location. Typically, an IPK is located under the weight-bearing part of the forefoot beneath the metatarsal heads.203 A wart is rarely located under a metatarsal head and is more likely to be seen under other areas on the plantar aspect of the foot. When an IPK is trimmed, a small, pearlgray avascular core is noted. This keratotic core is a buildup of the plantar callosity that invaginates owing to localized pres­ sure beneath the metatarsal head. A wart is a well-localized lesion with sharp margins. When shaved, it bleeds vigorously because of the end arterioles that are present in the lesion.

A

Keratotic lesions have not been specifically classified. Lesions are usually characterized as discrete or diffuse (Box 25H-35).

Evaluation Clinical Presentation and History Clinical evaluation includes a thorough history of the aggravating and relieving factors related to the lesion. Ath­ letes often report increased symptoms with specific activi­ ties or shoewear. Pain is usually directly under the callus (Box 25H-36).

Physical Examination Physical examination first includes careful evaluation of the alignment of the foot when standing. Position of the hind­ foot, the arch, and any great toe or lesser toe deformity should be noted. The plantar aspect of the foot should be carefully evaluated, noting the characteristics and location of any lesions. Trimming of the lesion usually reveals a well-circumscribed keratotic lesion. Keratotic lesions occur directly under a bony prominence, whereas plantar warts are less likely to occur underneath the metatarsal heads.203 Warts are quite vascular, so bleeding will be encountered with trimming a wart, unlike keratotic lesions, which are avascular (see Box 25H-36).

B

Figure 25H-57   A and B, A cavus deformity may lead to multiple areas of callus formation. (© M. J. Coughlin. Used by permission.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2110 DeLee & Drez’s� O

Box 25H-35 C  lassification of Intractable Plantar Keratosis

• Small core (or seed corn) with keratotic buildup around the periphery

• Well-circumscribed lesion (<1 cm in size) • Broad diffuse lesion (>1 cm in size) Treatment Options Nonoperative

Figure 25H-58   A long second metatarsal may lead to increased callus formation. (© M. J. Coughlin. Used by permission.)

Imaging Evaluation includes weight-bearing views of the foot. Sesamoid views are included depending on location of the lesion. Careful attention is given to bony abnormalities that may be responsible for areas of increased pressure (see Box 25H-36).

The initial treatment of an IPK involves trimming the lesion to reduce the keratotic buildup. The keratotic cen­ ter of a callus is delineated as it is shaved. This keratosis is a typical response to increased pressure; with time, the lesion invaginates slowly, and the keratosis increases in depth. As it deepens, it typically becomes more sympto­ matic. Frequently, it is not possible to trim a keratotic lesion completely in a single office visit; it may be necessary for a patient to return for subsequent visits to reduce the callus further.201 As the keratotic core becomes more superficial, it typically becomes less symptomatic. A seed corn has a well-differentiated keratotic core, usually 1 to 2 mm in size, and responds well to trimming or curettage.201 Trim­ ming a callus also helps in differentiating it from a wart. When the callus has been trimmed, a soft metatarsal pad (Fig. 25H-60) is placed proximal to the keratosis to redistribute the pressure more uniformly. Jahss204 recom­ mended relieving the pressure on areas of excess weightbearing and increasing pressure on areas of too little weight-bearing. The use of a soft insole (Fig. 25H-61) can alleviate the pressure further in athletes.210 Athletic shoe­ wear should provide a wide toe box and a soft sole to lessen impact when running. When varus or valgus malalignment of the forefoot or hindfoot results in subsequent development of an IPK, appropriate orthotic devices are selected to compensate for the malalignment and to redistribute weight-bearing forces beneath the metatarsal head. A Plastizote orthosis of medium or high density can be fabricated to relieve pressure beneath the IPK and provide correction for a postural deformity.

Box 25H-36 Typical Findings in Patients with Intractable Plantar Keratoses

• Patients report pain directly under the callus. • Patients may have hindfoot, midfoot, or forefoot

­deformity that leads to areas of increased pressure. of the callus reveals a well-circumscribed, keratotic lesion occurring directly under a bony ­prominence. • Keratotic lesions are avascular, so significant bleeding with trimming may indicate the presence of a plantar wart. • Radiographs may reveal a bony lesion adjacent to the callus.

• Trimming

Figure 25H-59   A rotatory midfoot deformity may subject the lateral border of the foot to callus formation. A varus deformity of the forefoot with a fixed hindfoot deformity is shown.   (© M. J. Coughlin. Used by permission.)

Foot and Ankle 2111

Operative

Figure 25H-60   A small metatarsal pad may be placed   just proximal to the keratosis to redistribute the pressure.   (© M. J. Coughlin. Used by permission.)

If a keratotic lesion continues to be symptomatic and significantly impairs athletic function, surgical interven­ tion is considered. Because of the lengthy postoperative recovery time, the possibility of restricted MTP motion, and the possibility of recurrence of a lesion or development of a transfer lesion, a rigorous trial of trimming, padding, and orthotic management should be carried out before surgery is performed.

IPKs beneath the second, third, and fourth metatarsal heads usually are classified as diffuse or discrete keratoses. Differentiating between these two lesions is important to institute proper treatment. A large, diffuse IPK often is associated with a long or plantar flexed metatarsal. Although the second and third metatarsals are the most common sites of occurrence,203 the other metatarsals are occasionally involved. ­Typically, these lesions are 1 to 2 cm in size and do not have an invaginated keratotic core. Multiple osteotomies have been described for treatment of metatarsalgia and plantar keratoses (Box 25H-37).201-203,211-214 Giannestras203 described a step-cut proximal metatarsal osteotomy that is used to decrease the length of a symptomatic metatarsal. Mann and Mann201,202 suggested the use of a long oblique longitudinal osteotomy rather than a step-cut osteotomy to shorten the elongated metatarsal (Fig. 25H-62). Giannestras203 reported develop­ ment of transfer lesions in 10% of patients postoperatively. Mann206 reported a 5% rate of transfer lesions. Delayed union occasionally occurs; however, with time, most oste­ otomies go on to successful healing. Mann and Du Vries215 proposed that a small, discrete, intractable plantar keratosis is caused by a prominent fibu­ lar condyle on the plantar aspect of the metatarsal head. A discrete callus can develop after a metatarsal head frac­ ture with a plantar flexion deformity of the metatarsal and a hyperextended MTP joint, leading to buckling of a toe. Discrete lesions are also associated with an idiopathic plantar flexed metatarsal. DuVries216 described a plantar condylectomy to correct this deformity. Coughlin recom­ mends removal of 20% to 30% of the condyle through a dorsal incision.217 This procedure was later modified by Mann and DuVries,215 who performed an MTP arthro­ plasty, removing about 2 mm of the articular surface along with the plantar condylectomy (Fig. 25H-63). After MTP joint arthroplasty, MTP joint motion is diminished by 25% to 50%. Although stiffness does not always affect the ­ function of a sedentary person, a competitive athlete typically requires more normal motion, and if so, an MTP arthroplasty is contraindicated. Mann and DuVries evaluated 100 patients with dis­ crete IPKs and noted a recurrence rate of 17.6% after MTP arthroplasty.215 Of these, 5% recurred under the ­symptomatic metatarsal. A transfer lesion developed in 13% of cases. Despite these results, 93% of the patients were satisfied with their outcome. There was a 5% rate of complications, which included fracture of the ­ metatarsal Box 25H-37 Metatarsal Osteotomies for ­Intractable Plantar Keratoses

Figure 25H-61   A soft insole may be used to decrease the pressure in the forefoot region. (© M. J. Coughlin. Used by permission.)

• Step cut osteotomy • Long oblique metatarsal osteotomy • Distal oblique (Helal) osteotomy • Capital oblique (Weil) osteotomy • Dorsal closing wedge osteotomy • Vertical chevron osteotomy • Segmental metatarsal osteotomy

�rthopaedic ����������� S �ports ������ � Medicine ������� 2112 DeLee & Drez’s� O 4 mm

Figure 25H-62   A, An oblique osteotomy of the proximal metatarsal may be used to   achieve shortening. B, Preoperative radiograph of a patient with intractable plantar keratosis beneath the second metatarsal head. C and D, After longitudinal step-cut osteotomy with internal   fixation, successful healing is shown   (C, anteroposterior view; D, lateral view).   E, Failure of fixation with fracture of Kirschner wire after osteotomy. (B to E, © M. J. Coughlin. Used by permission.)

A

B

4 mm

A

B

A A B

B

A

B

C

D

E

head, avascular necrosis, drift of the involved toe, and cock-up of the involved toe. When a competitive athlete requires a more normal motion, a distal metatarsal osteotomy is considered. A ver­ tical chevron procedure (Fig. 25H-64),205,214,218 a distal oblique osteotomy (Fig. 25H-65),212,213,219-222 or capital oblique osteotomy223 (Fig. 25H-66) is commonly used. A distal oblique osteotomy allows dorsal displacement and shortening of the osteotomy. It is believed that elevation of about 3 mm is necessary for adequate decrease of pressure beneath the symptomatic metatarsal head.224

For discrete lesions under the tibial sesamoid, tibial sesamoid shaving has been advocated. A broad, diffuse cal­ lus under the first metatarsal is usually associated with a ­plantar flexed first metatarsal, whereas a discrete lesion is associated most commonly with a prominent sesamoid. When conservative treatment is unsuccessful, a dorsalbased ­closing wedge osteotomy of the proximal first meta­ tarsal is used to correct the plantarflexed first ray. Sesamoid shaving or, more infrequently, sesamoid excision is consid­ ered for discrete lesions (see earlier section on sesamoid ­dysfunction).

Foot and Ankle 2113

A

D

B

C

E

F

Figure 25H-63   A, A hockey-stick–shaped incision from the middle of the metatarsal shaft into the adjoining web space. The extensor tendon and the skin are retracted, and the capsule is excised longitudinally. B, The capsule is released on the medial and lateral aspects. C, The toe is plantar flexed, exposing the metatarsal head. D, Two millimeters of the distal metatarsal articular surface are resected. E, The plantar condyle is removed. F, The articular surfaces are smoothed to create a congruous surface. Often, there is a certain amount of restricted motion after this procedure.

A

A

B Figure 25H-64   Technique of vertical chevron osteotomy.   A, Preoperatively. B, After osteotomy.

B Figure 25H-65   Technique of oblique lesser metatarsal osteotomy showing dorsal displacement. A, Preoperatively.   B, After osteotomy.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2114 DeLee & Drez’s� O

A

B C Figure 25H-66   A, Preoperative diagram shows dislocated lesser metatarsophalangeal joint. B, Technique of capital oblique osteotomy allows shortening but not plantar flexion. C, After internal fixation and resection of the dorsal flare, the lesser metatarsophalangeal joint has been reduced.

Weighing the Evidence Helal,212,213 in describing the oblique distal metatarsal osteotomy, noted that the purpose was to elevate the meta­ tarsal head and to reduce length. Although in his initial report, Helal212 did not note specific complications, in a retrospective review of 310 patients (508 feet), 84% were noted to have no pain, and an 8% recurrence of plantar callosities was noted.213 Winson and coworkers222 reported on 94 patients (124 feet) who underwent a similar procedure. In their report, 53% of the patients had significant postoperative symp­ toms, including transfer lesions in 32%, nonunion in 13%, and an overall recurrence of an IPK in 50% of patients. In 66 of the 124 feet, major complaints were noted ­postoperatively. The authors stressed that in either a cavus or a rigid foot, a distal oblique sliding osteotomy was con­ traindicated. These results agreed with those of Giannes­ tras,203 indicating that the presence of a contracture at the MTP joint was a further contraindication to a metatarsal osteotomy. Pedowitz,221 in a report on 69 distal oblique osteoto­ mies in 49 patients, reported good results in 83% of cases. There was a 27% incidence of either residual callosity or transfer lesion and two nonunions in his series. ­Pedowitz221 stressed that this procedure was contraindicated in patients who had a fixed MTP joint deformity. Idusuyi and asso­ ciates,220 in a report on 20 patients (23 feet) who had single osteotomies of the second, third, or fourth meta­ tarsals without internal fixation, noted a 20% reoperation rate for recurrent plantar callosities. Of these patients, 65% were limited with footwear choices or required a shoe insert. Of 23 feet, 13 (56%) were rated as poor or fair, and the authors had significant reservations about recommend­ ing this procedure.

Trnka and colleagues,223 in a comparison of the results of an intra-articular capital oblique osteotomy (Weil type) and a distal oblique osteotomy (Helal type), reported a high level of satisfaction and a lower inci­ dence of recurrent metatarsalgia in transfer lesions with the capital oblique osteotomy. No transfer lesions were noted with the capital oblique osteotomy, whereas 41 feet with a Helal-type osteotomy developed transfer lesions. Five malunions and three pseudarthroses occurred in the 15 cases that underwent the Helal-type osteotomy com­ pared with no malunions or pseudarthroses in the Weiltype group. Trnka and colleagues223 concluded that the Weil-type osteotomy was a satisfactory method for treat­ ing metatarsalgia, but because of the high complication rate, the Helal-type osteotomy was not an acceptable procedure. Metatarsal head resection should be avoided because it tends to concentrate pressure beneath the remaining meta­ tarsal heads. Several authors have investigated the mechanism and magnitude of pressure relief after capital oblique osteot­ omy of the metatarsal.225-228 Grimes and Coughlin225 rec­ ommended using a 2-mm saw blade for a capital oblique osteotomy in order to offset plantar displacement of the metatarsal head. Khalafi and colleagues,226 Snyder and coworkers,227 and Vandeputte and colleagues228 found that overall plantar pressure decreased under the metatar­ sal head with capital oblique osteotomy. However, when compared with a vertical chevron osteotomy,227 both pro­ cedures resulted in decreased pressure, but only the chev­ ron osteotomy demonstrated a corresponding decrease in load. Hatcher and colleagues219 advocated multiple metatar­ sal osteotomies as a method of reducing the occurrence of transfer lesions and the recurrence of plantar keratoses. Mann206 and others222 favored an isolated osteotomy as a more reliable procedure. If two adjacent metatarsals are significantly longer, an osteotomy of these two metatar­ sals is considered.201 Internal fixation allows the surgeon to alter either the length or the inclination of the metatarsal, then to stabilize it in the desired position. A higher rate of displacement, angulation, and shortening can occur if the metatarsal osteotomy is not internally fixed.222 Rigid inter­ nal fixation also tends to reduce the chance of malunion or delayed union. Although delayed union or nonunion is possible with the use of a proximal dorsal closing wedge osteotomy for an IPK, the location of this osteotomy in the metaphysis of the involved metatarsal usually allows fairly rapid heal­ ing. A transfer lesion can develop if an excess amount of bone is resected, and an IPK can recur if too little bone has been resected. Mann206 reserved this osteotomy for recurrent lesions after failure of a longitudinal oblique osteotomy. Occasionally, an IPK develops beneath a hallucal sesa­ moid.229,230 Mann and Wapner229 reported on the results in 12 patients and noted that 58% had excellent results with no recurrence and 33% had good results with slight recurrence of a plantar callosity. Van Enoo and Cane230 reported on 17 tibial sesamoid shavings with an average 4-year follow-up. In two of the feet, a mild recurrent ­callosity developed postoperatively.

Foot and Ankle 2115

Author’s Preferred Method Localized trimming, instruction of the athlete in methods of reducing the callus by self-care, redistribution of pres­ sure with metatarsal pads or orthotic devices, and use of proper footwear are the preferred methods of treatment. Surgery should be reserved for individuals who are limited significantly by an IPK and in whom other conservative methods have been unsuccessful. The significant risk for reduced range of motion, recurrence of plantar keratosis, and development of transfer lesions makes a strong case for the use of conservative care in competitive athletes.231 Surgical intervention should be considered only after a lengthy period of conservative care. A capital oblique osteotomy is useful in the setting of either a diffuse or discrete IPK. In patients with a small, discrete IPK secondary to a prominent fibular condyle, a capital oblique osteotomy (Weil type) is preferable either to an MTP joint arthroplasty (because of the possibility of restricted postoperative range of motion) or to a vertical chevron osteotomy. However, the metatarsal head must be raised or at the least not depressed, or symptoms can be made worse. When a contracture of the MTP joint has occurred, a plantar condylectomy or a distal metatarsal os­ teotomy should be avoided because the buckling of the toe at the MTP joint leads to plantar keratoses.232

Postoperative Prescription, Outcomes Measurement, and Potential Complications A compressive dressing is placed at the time of surgery, and the patient is allowed to bear weight in a postoperative shoe. If necessary, a cast can be used for increased protec­ tion. Kirschner wires are removed usually 3 to 4 weeks after surgery. After a metatarsal osteotomy, adequate time must be allowed for healing. Premature athletic activity may lead to failure of fixation, displacement of an osteotomy, or nonunion. In general, proximal metatarsal osteotomies require 6 to 12 weeks for osseous union to occur. A verti­ cal chevron osteotomy, if performed with a thin oscillat­ ing saw blade, heals in about 6 weeks. After an MTP joint arthroplasty and condylectomy, about 6 weeks is necessary for adequate healing to occur at the MTP joint. Once pins are removed and adequate healing has occurred (4 to 6 weeks), gentle range of motion is initi­ ated at the MTP joint to diminish postoperative stiffness. When radiographs demonstrate bony union, aggressive walking activity is initiated. Taping of the forefoot or the use of soft metatarsal pads helps to alleviate symptoms with athletic activity during surgical recovery. About 4 weeks after the initiation of walking, jogging is initiated, followed by running activities as pain permits. Decreased range of motion is a significant risk after MTP arthroplasty as well as after distal metatarsal oste­ otomy. Recurrence of plantar keratoses or development of a transfer lesion occurs in 10% to 50% of patients after metatarsal osteotomy.

Box 25H-38 Criteria for Return to Play ­After Osteotomy for Intractable Plantar Keratosis

• Bony healing of osteotomies on radiographs and clinical examination

• Resolution of pain and swelling • Full return to activities expected months

between 3 and 6

Criteria for Return to Play Full return to athletic activity is expected when bony heal­ ing occurs. Usually between 6 and 12 weeks after surgery, aggressive walking is initiated. Walking is advanced to jog­ ging after 4 weeks as pain and swelling allow. The athlete then increases running activities as tolerated, gradually returning to sport between 3 and 6 months postoperatively (Box 25H-38).

Special Populations All athletes are at risk for plantar keratoses. The risk for developing an IPK is higher in athletes with mild posi­ tional deformity or a plantarflexed or elongated metatarsal. Risk may also be increased in athletes involved in repetitive running activities.

LESSER TOE ABNORMALITIES The most common pathologic entities involving the lesser toes are hammer toes, mallet toes, and claw toes.233 Other pathologic entities of the lesser toes include lateral fifth toe corns, interdigital corns, and bunionettes. The severity and frequency of occurrence of these deformities increases with the age of the patient, and they have been reported as more frequent in women.234-236 Although there appears to be a correlation between the use of high-fashion footwear and the development of lesser toe deformities, intrinsic predis­ posing factors, such as a wide forefoot, an abnormally long ray, inflammatory arthritis, isolated or repetitive trauma, or neuromuscular disease may predispose the lesser toes to deformity.233 The development of callosities on the lateral aspect of the fifth toe or between the lesser toes often causes problems with walking or running. Keratoses in either of these areas frequently necessitate modification of athletic activities. Although the onset of these problems usually is insidious,233 they can cause significant discomfort in older athletes, specifically athletes in their 40s through 60s.

Anatomy and Biomechanics The position of the lesser toes relies on a balance between intrinsic and extrinsic muscle forces. On the dorsal aspect of the lesser toes, the extensor digitorum longus dorsiflexes the proximal phalanx by its insertion into the extensor expansion.235 When the proximal phalanx is in a neutral position, the extensor digitorum longus extends the proxi­ mal interphalangeal joint.235,237 When the MTP joint is

�rthopaedic ����������� S �ports ������ � Medicine ������� 2116 DeLee & Drez’s� O Flexor digitorum brevis

A

EDL

Flexor digitorum longus Extensor tendon

Extensor hood

Interossei

Intrinsics

Flexor tendon

B Interossei

Lumbrical Extensor tendon

Extensor hood

Flexor tendon

C

Intrinsics

Lumbrical

Figure 25H-67   A, Plantar view of lesser toe. The tendon of the flexor digitorum longus inserts onto the base of the distal phalanx. B and C, Side views show the insertion of the intrinsic muscles, which allows them to plantar flex the metatarsophalangeal joint and extend the interphalangeal joints.

hyperextended, the dorsiflexion force at the interphalan­ geal joint decreases significantly.235,237 The flexor digito­ rum longus inserts into the distal phalanx on the plantar aspect of the toes and flexes the distal interphalangeal joint (Fig. 25H-67A). The flexor digitorum brevis inserts into the middle phalanx and plantar flexes the proximal inter­ phalangeal joint. Neither flexor has a significant influence on MTP joint plantar flexion because there is no inser­ tion into the plantar aspect of the proximal phalanx.237 The interossei and lumbricals pass plantar to the axis of motion of the MTP joint and dorsal to the axis of the interphalan­ geal joints. These muscles provide a plantar flexion force at the MTP joint and an extension force at the proximal interphalangeal and distal interphalangeal joints (see Fig. 25H-67B and C).237,238 The plantar aponeurosis and plantar capsule are static structures that provide a stabilizing force on the plan­ tar aspect of the MTP joint.235,237 Aided by the dynamic power of the interossei and lumbricals, plantar flexion of the proximal phalanx occurs. At the MTP joint, the long and short extensor muscles are opposed by the plantar aponeurosis, plantar capsule, and intrinsic musculature. The reactive force as the foot strikes the ground helps to achieve passive extension at the MTP joint; however, the long and short extensors are significantly stronger than the intrinsics. At the interpha­ langeal joints, the flexor digitorum longus and flexor digi­ torum brevis are opposed by the interossei and lumbricals. At both the MTP joint and the interphalangeal joints, a mismatch occurs (Fig. 25H-68) in which the more pow­ erful extrinsic muscles overpower the weaker ­ intrinsic

Flexors

Figure 25H-68   At the various joints of the lesser toe, intrinsic and extrinsic muscles oppose each other. An obvious mismatch occurs between the larger intrinsic and the smaller intrinsic muscles, leaving two deformities. EDL, extensor digitorum longus.

­ uscles.237 The position of the lesser toes is crucial in m whether this mismatch leads to development of pathol­ ogy. When the MTP joint is held in a neutral position, the extensor digitorum longus assists in extending the proxi­ mal interphalangeal joint and the flexor digitorum longus flexes the MTP joint. When the MTP joint is hyperex­ tended, these functions are diminished. The long and short flexors come under increased tension, and the interossei and lumbricals are overpowered by the stronger extrinsic muscles.235,237 As a result, a chronic hyperextension defor­ mity develops at the MTP joint and the flexion deformity at the interphalangeal joints increases. Subluxation or dislocation of the MTP joint com­ monly occurs as a result of the imbalance between stabi­ lizing plantar structures and intrinsic and extrinsic muscle forces.239-243 The plantar aponeurosis or plantar capsule becomes incompetent as a result of repetitive trauma, a constricting toe box, or excessive length of the metatar­ sal.237 This diminishes the stabilizing force of the plantar structures over the MTP joint.235,237,244 Concurrent con­ tractures of the dorsal capsule and extensor tendons add to the deformity. A hallux valgus deformity destabilizes the toe by exerting extrinsic pressure on the second toe.237 With time, subluxation can progress to frank dislocation. Although the most common deformity of the second MTP joint is dorsal dislocation, occasionally the second toe deviates medially.237,240,242 A gap or space occurs between the second and third toes (Fig. 25H-69).240-242 This defor­ mity is often associated with a hallux valgus deformity. Medial deviation of the toe can also be seen after trauma,240 as a result of degenerative or rheumatoid arthritis,237,240 nonspecific synovitis,245 synovial cyst or ganglion forma­ tion,240 or erosion of the fibular collateral ligament.240 A corn occurs when a keratosis develops over the lateral aspect of the fifth toe. Pressure from constricting footwear over the area of the lateral condyle of the interphalan­ geal joint can lead to the development of keratoses (Fig. 25H-70).246 Occasionally, skin breakdown occurs, although callus formation is a more common abnormality. An interdigital corn is a hypertrophic keratotic lesion that occurs between the lesser toes either along the shaft or in the web space as a result of pressure between two prominent areas on adjacent toes (Fig. 25H-71).233,246 This lesion often is mistaken for a fungal infection when there is maceration between the lesser toes.233,246

Foot and Ankle 2117

Classification

Figure 25H-69   Crossover second toe shows space between the second and third toes. (© M. J. Coughlin. Used by permission.)

Lesser toe deformities include hammer toes, claw toes, mallet toes, and curly toes (Table 25H-4). These clinical entities are characterized by the location of the deformity (MTP, proximal interphalangeal, or distal interphalangeal joints) and the direction of the deformity (hyperextension or flexion). The deformity is further identified as passively correctable, flexible, or rigid. A hammer toe is a plantar flexion contracture of the proximal interphalangeal joint237,247 and frequently is asso­ ciated with a hyperextension deformity of the MTP joint (Fig. 25H-72).235 Hammer toes can occur in multiple toes, although most commonly they occur in the second toe.247 A hammer toe is believed to be an acquired deformity. Neuromuscular disease, degenerative disk disease, and connective tissue disorders may be associated with ham­ mer toe development234,237; however, the long-term use of fashionable footwear probably is the most common cause that leads to the development of progressive contractures of the lesser toes.234,235,248

Callus formation

A

B

Bony resection

C

D

Figure 25H-70   A and B, Lateral fifth toe corn. C and D, Resection of lateral condyle. Capsular repair helps prevent postoperative deformity. (A and C, From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 50; B and D, © M. J. Coughlin. Used by permission.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2118 DeLee & Drez’s� O

A

B

Bony resection

C

D

Figure 25H-71   A, Interdigital corn in web space. B, Radiograph shows area of bone impingement between distal aspect of proximal phalanx of fifth toe and proximal aspect of proximal phalanx of fourth toe. C, Padding may be placed in a symptomatic web space to relieve symptoms. D, Resection of condyle. (A-C, From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 51.)

A claw toe is characterized by dorsiflexion of the MTP joint and plantar flexion of the proximal interphalangeal joint.233,237 Multiple toes often are involved. Neurologic conditions, such as poliomyelitis, cerebral palsy, spinal cerebellar degeneration, muscular dystrophy, menin­ gomyelocele, Friedreich’s ataxia, diastematomyelia, and ­Charcot-Marie-Tooth syndrome, are common causes, but many of these deformities have no clear cause.233,237

One distinction between a hammer toe and a claw toe is that with a claw toe many toes are involved, and there is always a hyperextension deformity of the MTP joint.237 Both deformities typically have a fixed flexion deformity of the proximal interphalangeal joint. Sometimes there is no clear distinction, and often the treatments are similar. A mallet toe is characterized by a contracture or defor­ mity of the distal interphalangeal joint (Fig. 25H-73).237,249

  TABLE 25H-4  Characteristics of Lesser Toe Deformities Metatarsophalangeal Joint

Proximal Interphalangeal Joint

Distal Interphalangeal Joint

Hammer toe

Hyperextension

Flexion

None

Claw toe Mallet toe

Hyperextension None

Flexion None

± Flexion deformity Flexion

Curly toe

None

Flexion

Flexion

Acquired deformity Common in second toe Typically single toe involvement Multiple toe involvement Trauma, impingement in shoes Second toe most commonly involved Congenital

Foot and Ankle 2119 Callus from pressure of shoe

Callus Figure 25H-74   Dorsal aspect of the proximal interphalangeal joint strikes the toe box, leading to callus formation. An intractable plantar keratosis may develop beneath the metatarsal head.

deformities typically involve all of the lesser toes. Patients with a mallet toe develop a callus at the tip of the toe from repetitive injury as the toe strikes the ground. Occasion­ ally a callus can develop over the dorsal aspect of the distal interphalangeal joint. Instability, subluxation, or dislocation at the MTP joint occurs with a lesser toe deformity or in isolation. Patients frequently complain of pain caused by the development of a callus beneath the metatarsal head or pain on the ­plantar aspect of the MTP joint at the insertion of the plantar cap­ sule.237 With the development of a deviated second toe, a patient complains of vague pain in the second intermetatarsal space.240 This condition is often difficult to distinguish from an interdigital neuroma except that typically paresthesias or neuritic symptoms are not present in the toes (Box 25H-39). When a keratosis develops over the lateral aspect of the fifth toe, patients complain of pain with pressure from shoewear directly over the area of callus formation (see Fig. 25H-70A and B).246 Occasionally, skin breakdown occurs, although callus formation is a more common abnormality. In the case of an interdigital corn, the patient may complain of symptoms of recurrent infection, and in such cases a sinus tract is found extending into a mass of ­subcutaneous scar tissue. Occasionally, patients have been treated for a long time with antibiotics with no resolution of the ­problem.

Figure 25H-72   Hammer toe deformity. (© M. J. Coughlin. Used by permission.)

The abnormality is believed to result from pressure of the toe box of a shoe against a long second toe.234 In younger patients, contracture of the flexor digitorum longus is often the cause of this typically flexible deformity. In older patients, this is frequently a fixed contracture.237

Evaluation Clinical Presentation and History The major complaint of the patient with a lesser toe deformity is discomfort related to pressure on the toe with shoewear or callus formation. In a hammer toe deformity, discomfort is located over the dorsal aspect of the proximal interphalangeal joint where a callosity develops as the toe buckles and strikes the top of the toe box (Fig. 25H-74).248 On the plantar aspect of the foot, a callus develops beneath the metatarsal head sec­ ondary to subluxation or dislocation at the MTP joint. Claw toe deformities can present with the same pattern of pain and callus formation as a hammer toe. These ­clinical entities can be difficult to distinguish; however, claw toe

A

B

Figure 25H-73   A and B, Mallet toe deformity. (From Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991, p 48.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2120 DeLee & Drez’s� O

Box 25H-39 Typical Findings Associated with Lesser Toe Deformities Hammer Toe formation under metatarsal head and over ­proximal interphalangeal joint • Metatarsophalangeal joint subluxation or dislocation • Flexion deformity of adjacent toes (contracture of flexor digitorum longus) • Adjacent toe deformity such as hallux valgus

• Callus

A

Claw Toe

• Callus

formation under metatarsal head and over ­ roximal interphalangeal joint p • Metatarsophalangeal joint subluxation or dislocation • Multiple toe involvement Mallet Toe

• Callus formation over the distal interphalangeal joint or on the tip of the toe

Metatarsophalangeal Joint Instability

• Plantar capsule pain • Pain with drawer testing of the joint • Dorsal, medial, or lateral deviation of the toe • Pain in the adjacent interspace Physical Examination The patient with a lesser toe deformity should be examined in a standing as well as a sitting position.233,237,248 Each joint is evaluated for deformity, and, when present, deformities are identified as either flexible or rigid. It is important to evalu­ ate for subluxation or dislocation of the MTP joint, most commonly of the second toe. Subluxation and dislocation (Fig. 25H-75) occur in a dorsal medial or lateral plane.240,242 Pain can occur with ambulation or is elicited with palpation

A

B

B Figure 25H-76   Drawer test for metatarsophalangeal (MTP) joint instability. A, The toe is grasped between the thumb and the second finger. B, With a dorsal force, an attempt is made to subluxate the MTP joint. With instability of the MTP joint, pain is elicited with stress on the plantar structures. (From Coughlin MJ, Mann RA: Lesser toe deformities. In Coughlin MJ, Mann RA [eds]: Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999, p 354.)

or manipulation of the MTP joint. A dorsal plantar drawer test (Fig. 25H-76) is administered by thrusting the toe in a dorsal plantar direction. With capsulitis or MTP joint insta­ bility, pain is elicited. When a patient complains of presum­ able MTP pain, but no deformity is present, eliciting pain with a drawer test assists in making the correct diagnosis.236 In the evaluation of lesser toe deformities, it is ­important to note the amount of space present between adjacent toes. Lesser toes are inspected for deformity, callus ­formation, and interdigital corns. Occasionally, deformity diminishes the space available for a lesser toe correction. This diminished

C

Figure 25H-75   A-C, Gradual subluxation of the metatarsophalangeal joint may occur. This tennis player developed subluxation and dislocation during a 5-year period. (A, From Coughlin MJ: Lesser toe deformities. Orthopaedics 10:65, 1987; B, from Coughlin MJ: Lesser toe deformities. In Mann RA, Coughlin MJ [eds]: Surgery of the Foot and Ankle, 6th ed. St. Louis, Mosby, 1993.)

Foot and Ankle 2121

Treatment Options Nonoperative

Figure 25H-77   Contracture of all the lesser toes may indicate a tightness of the flexor digitorum longus. (© M. J. Coughlin. Used by permission.)

space requires correction of an asymptomatic adjacent toe (i.e., hallux valgus or an adjacent lesser toe deformity).237 With single hammer toe deformity, the adjacent toes are also evaluated in a standing position. Contracture of the flexor digitorum longus can result in a flexion deformity of an adjacent toe (Fig. 25H-77). In this case, in addition to carrying out a hammer toe repair, a flexor tenotomy is ben­ eficial in the treatment of the hammer toe deformity.237 On physical examination, deviation of the second toe either medially or dorsally is common. Pain is noted with ambulation and elicited with palpation of the second inter­ metatarsal space (see Box 25H-39).240

Imaging Radiographs of the forefoot are important in analyzing lesser toe deformities. The presence of subluxation or dislocation at the MTP joint is best seen on an AP radiograph, whereas eval­ uation of a hammer toe is carried out best with a lateral radio­ graph (see Box 25H-39). Attention is given to bony abnormality or deformity underlying interdigital or lateral corns.

A

In the early stages of a lesser toe deformity, a roomy shoe with a low heel and an adequate toe box is appropriate for most flexible deformities.233,237,248 Range of motion exer­ cise at each joint on a regular basis can help maintain flex­ ibility of the toe.237 A deeper toe box eliminates dorsal pressure on the ham­ mer toe or claw toe at the level of the proximal interpha­ langeal joint.237 When a callus has developed, use of a soft insole or liner can decrease symptoms.237 Foam or visco­ elastic padding over a callosity or padding at the tip of a toe where calluses have occurred (Fig. 25H-78) often relieves the pressure. A toe cap or foam rubber pad can protect the end of the second toe in a mallet toe deformity. Padding the bony prominence over the lateral aspect of the fifth toe and shaving the keratotic lesion are recommended for symp­ tomatic relief.233 Padding placed in the interspace between the toes relieves discomfort of interdigital corns (see Fig. 25H-71C). If an interdigital corn shows signs of chronic infection, the infection is treated by performing a culture and sensitivity studies and then instituting appropriate antibiotic therapy. When the acute infection has resolved, surgical treatment is considered to prevent recurrence. Metatarsal pads placed just proximal to the MTP joint alleviate plantar pressure and can decrease the ­hyperextension deformity at the MTP joint.235 Taping the toe in a corrected position helps to stabilize the sublux­ ated joint (Fig. 25H-79)240; with dislocation, however, tap­ ing rarely is successful in alleviating symptoms.237 When a painful deformity does not respond to conservative care, surgical intervention is considered.

Operative When a rigid deformity of the proximal (hammer toe or claw toe) (Fig. 25H-80) or distal (mallet toe) (Fig. 25H-81) interphalangeal joint is present, an interphalangeal joint

B

Figure 25H-78   A, Tube gauze may alleviate pain over a hammer toe. B, A toe cap pads a callus at the tip of the toe. (A, From Coughlin MJ: Lesser toe deformities. In Mann RA, Coughlin MJ [eds]: Surgery of the Foot and Ankle, 6th ed. St. Louis, CV Mosby, 1993; B, from Mann RA, Coughlin MJ: Video Textbook of Foot and Ankle Surgery. St. Louis, Medical Video Production, 1991.)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2122 DeLee & Drez’s� O Figure 25H-79   A and B, A toe may be taped into a stable position. (From Coughlin MJ: Crossover second toe deformity. Foot Ankle 8:29-39, 1987. © American Orthopaedic Foot and Ankle Society, 1987.)

B

A

Claw toe

Hammer toe

Flexible

Fixed

Flexible

Flexor tendon transfer

Condylectomy, proximal phalanx

Flexor tendon transfer

Fixed

Metatarsophalangeal soft tissue release • Dorsal capsular release • Medial and lateral capsular release • Extensor tendon lengthening

Hyperextended proximal phalanx

Fixed hammer toe repair Kirschner wire fixation

Metatarsophalangeal soft tissue release • Dorsal capsular release • Medial or lateral capsular release • Extensor tendon lengthening

B

Realigned toe

Kirschner wire fixation

A

Realigned toe

Figure 25H-80   A, Algorithm for treatment of hammer toe deformity. B, Algorithm for treatment of claw toe deformity. (A and B, From Coughlin MJ, Mann RA: Lesser toe deformities. In Coughlin MJ, Mann RA [eds]: Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999, pp 331, 348.)

Foot and Ankle 2123 Mallet toe

A Flexible

Fixed

Flexor digitorum longus tenotomy

Condylectomy, middle phalanx

Flexor digitorum longus tenotomy

B

C

D Kirschner wire fixation Realigned toe Figure 25H-81   Algorithm for mallet toe repair. (From Coughlin MJ, Mann RA: Lesser toe deformities. In Coughlin MJ, Mann RA [eds]: Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999, p 343.)

Box 25H-40 I nterphalangeal Joint ­Arthroplasty Technique

• Center an elliptical skin incision over the dorsal aspect

of the proximal (hammer toe and claw toe) or distal (mallet toe) interphalangeal joint. • Excise the dorsal callus, the extensor tendon, and the joint capsule.233,237 • Release the plantar capsule and the collateral ligaments carefully to deliver the head of the phalanx. • Resect the condyles of the proximal (hammer toe and claw toe) or middle (mallet toe) phalanx (the amount of bone removed depends on the severity of the contracture). • Perform a flexor tenotomy in a hammer toe repair if there is significant tightness of the long flexor tendon or an adjacent lesser toe appears to be contracted.237 In a mallet toe deformity, the flexor digitorum longus is released through the same incision. • Remove the articular cartilage of the base of the middle (hammer toe and claw toe) or distal (mallet toe) phalanx. • Fix the toe with a 0.045 intramedullary Kirschner wire.237,248 • Close the skin with vertical mattress sutures and apply a small compression dressing (see Figs. 25H-87 and 25H-88).

E Figure 25H-82   A, Technique of fixed hammer toe repair. A dorsal elliptical incision excises skin, extensor tendon, and dorsal capsule. B, The collateral ligaments are released. C, The condyles of the proximal phalanx are delivered. D, The condyles are resected with a bone cutter. E, An intramedullary Kirschner wire is used for fixation.

arthroplasty is carried out (Box 25H-40; Figs. 25H-82 and 25H-83). The technique is similar at either the proximal interphalangeal joint for a hammer toe or claw toe or the distal interphalangeal for a mallet toe. Whether an arthrod­ esis or an arthrofibrosis is obtained is not as important as the correction of the deformity and attaining stiffness at the interphalangeal joint.244 With a fibrous proximal inter­ phalangeal joint arthroplasty, about 15 degrees of motion is expected.237,248 When the deformity of the proximal interphalan­ geal joint in a hammer toe or claw toe is flexible, a flexor ­tendon transfer is performed (Box 25H-41; Fig. 25H-84). A Kirschner wire is placed from the tip of the toe in a prox­ imal direction. Even when the wire fails to penetrate the metatarsal head, it still acts as a splint for the MTP joint. The Kirschner wire is used only to stabilize the repair. If the toe is not corrected completely, the deformity can recur when the wire is removed. Alternate fixation devices such as bioabsorbable pins have also been described.250 An MTP joint abnormality is assessed in regard to the severity and rigidity of the deformity. Surgical correction is performed at this level to prevent recurrence of defor­ mity.233 A soft tissue release at the MTP joint (Box 25H-42)

�rthopaedic ����������� S �ports ������ � Medicine ������� 2124 DeLee & Drez’s� O

Box 25H-41 Flexible Hammer Toe Correction Technique

• Center a dorsal longitudinal incision over the proximal A

B

C

D

E

Figure 25H-83   Operative repair of mallet toe deformity.   A, Elliptical skin incision. B, Extensor tendon and dorsal capsule are excised. C, The collateral ligaments are severed, exposing the condyles in the middle phalanx. D, The condyles of the middle phalanx are excised in the supracondylar region with a bone-cutting rongeur. E, The articular surface of the distal phalanx is excised. (From Coughlin MJ, Mann RA: Lesser toe deformities. In Coughlin MJ, Mann RA [eds]: Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999, p 344.)

is used for mild (Fig. 25H-85A) or moderate (see Fig. 25H-85B) subluxation when the toe can be reduced easily. By achieving arthrofibrosis at the MTP joint, the stability of the second MTP joint is increased; however, this occurs at the expense of motion. With more severe instability and deformity (Fig. 25H- 86), further surgery is necessary to realign the second MTP joint. In this case, a complete soft tissue release of the MTP joint capsule and ligaments does not achieve a stable reduction. A flexor tendon trans­ fer (Fig. 25H-87) can enhance the correction and achieve adequate MTP joint stability. Each limb of the flexor ten­ don can be sutured in place (see Box 25H-41), and the stability of the second toe is evaluated by dorsiflexion and plantar flexion of the ankle. When instability is still present or if the toe cannot be reduced easily, decompression of the MTP joint must be performed.237 Several procedures, including partial proximal phalangectomy (Fig. 25H-88), metatarsal head arthroplasty (Fig. 25H-89), and shorten­ ing osteotomy of the metatarsal, have been described to decompress the MTP joint in more severe deformities. A partial proximal phalangectomy251 (see Fig. 25H-88) ­frequently requires a syndactylization between the ­second

phalanx.237,269-272,279 • Dissect superficial to the extensor tendon and ­adjacent to the extensor hood on each side of the toe. This ­dissection is superficial to the extensor hood and deep to the neurovascular bundle. • Make a transverse incision at the plantar flexion crease of the metatarsophalangeal joint. • Identify the tendon sheath of the flexor tendons and ­incise longitudinally. The tendon of the flexor ­digitorum longus is characterized by a median raphe and is the largest and deepest of the three tendons in the sheath (see Fig. 25H-89).248 • Apply tension to the tendon through this proximal plantar incision and percutaneously detach the tendon distally at the level of the distal interphalangeal joint. • Pull the tendon into the proximal plantar incision and split longitudinally along the median raphe. • Pass each half of the tendon ion either side of the ­proximal phalanx in a dorsal direction. • With the ankle held in a neutral position and the toe held in about 20 degrees of plantar flexion, suture the tendon to the extensor expansion or to the other limb of the tendon. • Tighten either limb until adequate alignment is achieved. • Stabilize the repair with a 0.045 Kirschner wire (the toe should be realigned adequately before placement of the Kirschner wire).

and third toes to stabilize the floppy toe. Although a par­ tial proximal phalangectomy decompresses the MTP joint, a more mechanically stable articulation is achieved by ­preserving the base of the proximal phalanx and decom­ pressing the joint instead through a metatarsal osteotomy (Box 25H-43) or joint arthroplasty.237,248,252,253 Lateral fifth toe corns are treated with removal of a bony prominence at the interphalangeal joint.237,248 If only the lateral condyle is prominent, it is excised (see Fig. 25H-70C and D). When the base of the middle phalanx is prominent, it is also shaved with a rongeur. When a con­ tracture of the interphalangeal joint is present, resection of the condyles of the proximal phalanx achieves a more longlasting repair.233 A technique similar to that used for a fixed hammer toe deformity is employed (see Box 25H-40), stabilizing the toe with an intramedullary Kirschner wire. With recurrence of a lateral fifth toe corn after a condylec­ tomy, a flexor tenotomy reduces the fifth toe contracture and diminishes symptoms. Interdigital corns with an underlying exostosis are treated with removal of the exostosis (see Fig. 25H-71D).237,248,254 The incision should be kept out of the web space because these wounds are slow to heal and can become macer­ ated or secondarily infected. If possible, the capsule of the involved joint should be repaired. When prominent exos­ toses are adjacent to each other, both are resected.246

Foot and Ankle 2125

EDL

A

FDB

FDL EDL

FDL

D EDL

Figure 25H-84   Technique of flexor tendon transfer. A, Lateral view shows flexor digitorum longus (FDL), flexor digitorum brevis (FDB), and extensor digitorum longus (EDL). B, The flexor digitorum longus is detached through a distal puncture wound and is delivered through a transverse incision at the plantar metatarsophalangeal joint flexion crease. C, The tendon is split longitudinally, and each half is delivered on either side of the proximal phalanx and is sutured into either the extensor expansion or the corresponding limb of the flexor tendon. D, Dorsal view shows transferred flexor digitorum longus tendon. E, Cross-sectional view shows the characteristic position of the flexor digitorum longus tendon. It is deep to the flexor digitorum brevis and is characterized by a midline raphe.

B

FDL

C E

Box 25H-42 Metatarsophalangeal Joint ­Release Technique

• Center a curvilinear incision over the metatarsophalan­ geal joint.

• Lengthen or release the extensor tendon.240 • Release the dorsal, medial, and lateral capsules

and

r­ esect any hypertrophic synovium. • Release or tighten additional structures if needed to ­reduce the metatarsophalangeal joint: • Release the first lumbrical if it is acting as a ­deforming force.237 • Release plantar adhesions.237 • Tighten elongated medial or lateral collateral ­ligaments to correct varus or valgus deformity. • Reduce the subluxated metatarsophalangeal joint after the soft tissue release and test for stability by passive dorsiflexion and plantar flexion of the ankle joint. • Stabilize the toe with an intramedullary 0.045 Kirschner wire, if the toe is stable and does not redislocate.

FDB

Often hindfoot or midfoot deformity is the cause of discomfort in patients with lesser toe deformity, especially claw toes.237 It is important to address the major deformity when treating the patient.

Weighing the Evidence Ohm and colleagues255 reported on 25 patients (62 ham­ mer toe repairs) in whom an interphalangeal fusion was performed. A 100% fusion rate was achieved. An equal number of corrections were performed in the second, third, and fourth toes. Although many authors advocate attempted proximal interphalangeal arthrodesis,256-258 adequate resection with realignment that achieves a stable alignment of the toe is considered a successful result.259-261 A fusion of the proximal interphalangeal joint or an arthro­ fibrosis succeeds by converting the flexor digitorum longus to a flexor of the entire digit. Coughlin and associates247 reported on 63 patients (118 toes) with a fixed hammer toe deformity. Involvement of the second toe was noted in 35%, the third toe in 21%, the

�rthopaedic ����������� S �ports ������ � Medicine ������� 2126 DeLee & Drez’s� O Mild MTP subluxation

Subluxation moderate

Soft tissue release • Mediolateral capsular release • Dorsal capsular release • Extensor tendon lengthening

Soft tissue release • Medial or lateral capsular release • Dorsal capsular release • Extensor tendon lengthening • Lumbrical release

Remaining dorsal subluxation

Medial or lateral MTP malalignment

Plantar capsular release

Medial or lateral capsular reefing

Remaining dorsal subluxation

Medial or lateral MTP malalignment

Flexor tendon transfer

Flexor tendon transfer

MTP joint realigned Fixed hammer toe deformity?

A

Hammer toe repair

MTP joint realigned Fixed hammer toe deformity?

Hammer toe repair

Kirschner wire fixation (optional)

Kirschner wire fixation

Realigned toe

Realigned toe

B

Figure 25H-85   A, Algorithm for treatment of mild subluxation of lesser toe metatarsophalangeal (MTP) joint. B, Algorithm for treatment of moderate MTP joint subluxation. (From Coughlin MJ, Mann RA: Lesser toe deformities. In Coughlin MJ, Mann RA [eds]: Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999, pp 360, 362.)

fourth toe in 24%, and the fifth toe in 20%. After a resec­ tion arthroplasty technique with intramedullary Kirschner wire fixation, fusion of the proximal interphalangeal joint occurred in 81% of the involved toes. Subjective acceptable alignment was achieved in 86% of cases. Pain was relieved in 92%, and subjective satisfaction was noted by 84% of patients. Malalignment and numbness were the major fac­ tors reported to be associated with an unsuccessful result.

The rate of pseudarthrosis in some series approaches 50%, although a higher fusion rate has been achieved with peg-and-dowel type of technique.256,262,263 Lehman and Smith262 reported a 50% patient satisfaction rate with this technique, however. Major reasons for postoperative dissat­ isfaction included angulation of the lesser toe and incomplete relief of pain. McConnell264,265 reported on a large series of patients treated with diaphysectomy. This ­ technique

Foot and Ankle 2127 Dislocation of MTP joint

Soft tissue release • Medial, lateral, and dorsal capsular release • Extensor tendon lengthening

Acute dislocation (reducible)

Chronic dislocation (irreducible)

Reducible but unstable

MTP joint arthroplasty

Hammer toe repair

Kirschner wire fixation

Fixed hammer toe deformity?

Hammer toe repair Kirschner wire fixation

Partial proximal phalangectomy

Flexor tendon transfer

Flexor tendon transfer Fixed hammer toe deformity?

Metatarsal osteotomy

Fixed hammer toe deformity?

Hammer toe repair Kirschner wire fixation

Realigned toe Figure 25H-86   Algorithm for treatment of subluxation/dislocation of metatarsophalangeal (MTP) joint. (From Coughlin MJ, Mann RA: Lesser toe deformities. In Coughlin MJ, Mann RA [eds]: Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999, p 366.)

was useful in treating a hammer toe deformity as well as shortening a substantially long lesser toe. The actual post­ operative alignment of the lesser toe and complication rate were not reported in either of McConnell’s series. Daly and Johnson266 reported on a large series of hammer toes treated with partial proximal phalangectomy. Although 75% patient satisfaction was noted, 43% of patients noted ­ moderate footwear restrictions, 27% reported residual pain, 28% noted cosmetic dissatisfaction, and 18% reported recurrent cock-up deformity. Cahill and Connor267 reported on 78 patients (84 feet). They noted 50% poor results after partial proximal phalangectomy. Conklin and Smith268 noted 29% postoperative dissatisfaction after this procedure.

Taylor,269,270 Pyper,271 and others266,272 reported satis­ factory results ranging from 51% to 89% after ­correction of a flexible hammer toe deformity. Thompson and Deland273 reported excellent pain relief; however, only 54% of patients examined were noted to have complete cor­ rection of deformity. Kuwada274 and Barbari and Brevig275 reported greater than 90% satisfaction in patients after flexor tendon transfer. Boyer and DeOrio276 used a modifi­ cation of the flexor tendon transfer with 72% satisfaction. Trnka and colleagues253 reported on 25 osteotomies of the second, third, and fourth metatarsals. Of 15 patients, 12 were satisfied with their results. Reasons for dissatis­ faction included pain associated with a prominent plantar

�rthopaedic ����������� S �ports ������ � Medicine ������� 2128 DeLee & Drez’s� O Figure 25H-87   A, Preoperative photograph of crossover second toe deformity. B, Preoperative radiograph shows second toe deformity. C, Three years after soft tissue release and flexor tendon transfer. D, Radiograph 3 years after surgical repair. (From Coughlin MJ: Cross over second toe deformity. Foot Ankle 8:29-39, 1987. © American Orthopaedic Foot and Ankle Society, 1987.)

A

B

C

D

screw. Three patients developed an asymptomatic callus beneath the involved lesser metatarsal head. All osteoto­ mies healed. All 25 dislocated MTP joints were relocated successfully with an average metatarsal shortening of 4.4 mm. No cases of avascular necrosis were described, and the authors concluded that this technique enabled an accurate metatarsal shortening because the distal meta­ tarsal fragment could be positioned exactly and stabilized with internal fixation. Coughlin240 reported on the surgical correction of 15 toes (11 patients) with a variety of methods depending on the severity of the deformity. Extensor tendon length­ ening or tenotomy, MTP joint release, flexor tendon transfer, Kirschner wire fixation, and occasionally metatar­ sal articular resurfacing were used in the treatment of these acute and chronic MTP joint subluxations. ­ Satisfactory

results were reported in 93%. Of these patients, 79% were women with an average age of 60 years. In a follow-up study, Coughlin242 reported on a younger group of patients (9 patients [11 toes]) with an average age of 50 years. A positive result on drawer testing was noted in all cases and was pathognomonic of early second MTP joint instabil­ ity. Using a similar soft tissue realignment technique, good and excellent results were reported in 71%. After a mallet toe repair, generally satisfactory results have been reported. Coughlin249 obtained successful fusion in 72% of cases; 86% of patients were satisfied with their results. Of patients with a fibrous union, 75% were satis­ fied, although slightly less so than patients with a success­ ful distal interphalangeal arthrodesis. Pain relief was noted by 97% and correction of deformity by 91% of patients. Although not performed in all cases, a flexor tenotomy was

Foot and Ankle 2129

Author’s Preferred Method

Figure 25H-88   A partial proximal phalangectomy   may be used to decompress the second metatarsophalangeal joint. (© M. J. Coughlin. Used by permission.)

associated with a slightly higher rate of satisfaction and maintenance of correction. Barbari and Brevig275 performed a flexor tendon trans­ fer for correction of a flexible claw toe deformity and noted satisfactory results in 90% of cases, although one third of patients reported postoperative metatarsalgia. Postopera­ tively, 12 patients noted no interphalangeal joint motion, 15 had reduced motion, and 12 reported the same amount of motion at the interphalangeal joint. Continuing com­ plaints were noted by patients who had a fixed claw toe deformity that would have been treated more appropri­ ately with resection arthroplasty. With refinement of this procedure and the use of a flexor digitorum longus trans­ fer, Thompson and Deland273 reported improved results, although complete realignment of MTP subluxation was achieved in only 54%. After a simple condylectomy to repair a lateral fifth toe corn, in about 2% of patients, a keratotic lesion persists despite resection of the bony prominence.237 In this situation, soft tissue trimming with a scalpel usually results in reduction of the keratotic lesion to a minimally symptomatic state. An excision of an excessive portion of the middle phalanx should be avoided because this can destabilize the fifth toe. Zeringue and Harkless277 reported their results after the treatment of 30 patients with an interdigital corn excision. Of the affected feet, 94% were noted to have rotation of the fifth toe. These patients were treated with a proximal interphalangeal joint arthroplasty of the fifth toe combined with lateral-based condylectomy of the fourth toe proximal phalanx. A high level of satisfaction was found at an average follow-up of 3 years. Recurrence of a soft corn is the most common postoperative com­ plication. When recurrence develops, a more extensive resection is considered. A complete condylectomy may be necessary in cases of recurrence in the presence of a severe deformity.246

Deformities of the lesser toes can cause significant mor­ bidity in athletes. Pressure from ill-fitting footwear leads to blistering or callus formation over bony prominences. Padding of areas at risk, coupled with the use of well­fitting shoes, alleviates many symptoms pertaining to the lesser toes. Because most surgery of the lesser toes involves some form of joint arthroplasty, reduced range of motion of the involved joint is a common postoperative occurrence. An athlete should be counseled that ­although surgery typically reduces discomfort significantly, de­ creased range of motion can impair athletic ­ activity or performance. It usually is preferable to perform surgery in athletes when the athlete already has modified activ­ ity because of discomfort and conservative treatment has been ­unsuccessful. A fixed hammer toe or claw toe deformity is corrected by performing a condylectomy of the proximal phalanx. A flexor tenotomy is used when contractures of adjacent toes are present. Release of contracted dorsal medial and dorsal lateral capsular structures is performed when a hyperex­ tension deformity of the MTP joint is present. If severe subluxation or dislocation has occurred, decompression of the MTP joint often is necessary. A distal metatarsal shortening osteotomy is preferable because it affords more stability to the MTP joint and preserves the articular car­ tilage. A flexor tendon transfer often is necessary to give a stabilizing plantar flexion force to the proximal phalanx. Care must be taken not to perform excessive surgery on the second toe that leads to vascular compromise. A flexible hammer toe or claw toe is repaired by per­ forming a flexor tendon transfer. During the preoperative physical examination, it is important to determine that the toe is passively correctable. If a flexor tendon transfer is performed on a toe with a fixed deformity, a less than com­ plete repair results. A flexible mallet toe in a younger person is corrected with a flexor tenotomy. When a fixed contracture devel­ ops, a condylectomy of the middle phalanx, coupled with a flexor tenotomy, allows realignment of the toe. A lateral fifth toe corn is treated by removing the promi­ nent exostosis of the lateral condyle of the proximal phalanx. A complete capsular repair is important to realign the toe. When a more severe deformity is present or when a flexion deformity is present at the interphalangeal joint, a condylec­ tomy of the proximal phalanx is performed, and Kirschner wire stabilization is used. An interdigital corn repair is ad­ dressed by removing either one or both of the correspond­ ing exostoses.246 Through a dorsal approach, the exostosis is identified and resected. If possible, a capsular repair is performed. The most important aspect of surgery on the lesser toes is the need to repair all components of the deformity so that adequate alignment is achieved. Care must be taken to ­protect the vascular supply to the lesser toes. Sometimes an intramedullary Kirschner wire must be removed when postoperative vascular compromise occurs. It is better to perform a two-stage lesser toe repair than to risk severe ischemia or necrosis of a lesser toe postoperatively.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2130 DeLee & Drez’s� O A B

B

A

B

C

Figure 25H-89   A, A metatarsophalangeal arthroplasty requires removal of 2 to 3 mm of articular surface and beveling of the metatarsal head on the dorsal and plantar aspects. B, Radiograph after metatarsophalangeal arthroplasty. C, Five-year follow-up after metatarsophalangeal arthroplasty. Joint motion is reduced about 50%. (A, © M. J. Coughlin. Used by permission; B and C, from Mann RA, Coughlin MJ [eds]: Surgery of the Foot and Ankle, 6th ed. St. Louis, Mosby, 1993, p 390.)

Postoperative Prescription, Outcomes Measurement, and Potential Complications When surgical intervention is used to repair hammer toes, mallet toes, claw toes, and hard and soft corns, postopera­ tive management should concentrate on the goals of main­ taining physical fitness, while protecting the repaired lesser toes until adequate healing has occurred. The patient is allowed to ambulate in a postoperative shoe immediately.

While wound healing is occurring, upper extremity activi­ ties and exercising on a stationary bicycle can be used to maintain fitness. Sutures are removed 2 to 3 weeks after surgery. Pins are removed 4 weeks after surgery. After pin removal, the involved lesser toe is taped to maintain adequate alignment for another 6 weeks. In the case of a flexible hammer toe repair or release of the MTP joint, the toe is taped in slight plantar flexion (see Fig. 25H-79), or, in cases of medial deviation or a crossover second toe, the toe is taped in a lateral direction. If correction is ­performed

Foot and Ankle 2131

Box 25H-43 Capital Oblique Osteotomy ­Technique

Box 25H-44 Expectations After Correction of Lesser Toe Deformity

• Center

• Swelling frequently lasting 6 months or more • Stiffness at the metatarsophalangeal or interphalangeal

a 3-cm longitudinal incision over the lesser ­ etatarsophalangeal joint. m • Release the dorsal, medial, and lateral capsule. • Reduce the metatarsal head and plantar flex the phalanx to expose the metatarsal head. • Perform a longitudinal oblique distal metatarsal ­osteotomy, holding the saw blade parallel to the plantar surface of the foot • Penetrate the distal superior metatarsal articular surface 2 to 3 mm inferior to the dorsal metatarsal surface. • Create the osteotomy in the plane parallel to the plantar surface of the foot and proceed in a proximal ­direction until the saw blade has penetrated the ­ proximal ­metatarsal cortex. • Translate the distal fragment proximally by the desired amount (2 to 6 mm). • Stabilize the osteotomy with one or two minifragment lag screws. (Excessively long screw can cause plantar pain.) • Resect the excess dorsal metatarsal surface that extends beyond the articular surface with a rongeur (see Fig. 25H-66).

at the fifth toe interphalangeal joint, the fifth toe is taped to the fourth toe after the pin is removed. It is important during the 6-week period of taping after pin removal to protect the toe from stress that increases the risk for recur­ rence of the deformity. Walking activities are allowed about 6 weeks after surgery; the involved toe is protected in a shoe with a stiff sole and a roomy toe box. Running should be avoided until 9 to 12 weeks after surgery to allow adequate healing to occur. Lesser toe deformity correction frequently results in postoperative stiffness at either the interphalangeal joint or the MTP joint (Box 25H-44). Passive manipulation helps to alleviate these symptoms. Patients should be counseled that an interphalangeal joint arthroplasty will leave resid­ ual stiffness. After a tendon transfer, the ability to curl the toes is sacrificed. When correction of a flexible deformity is planned, a patient should be counseled preoperatively that dynamic function of the lesser toe is absent after flexor tendon transfer. Although this is not a cause of disability, a patient must be counseled about the tradeoff of function of the flexor digitorum longus for stability or realignment of the lesser toe. After distal metatarsal osteotomy, patients experience limitation of plantar flexion at the MTP joint as well as reduced range of motion. After an operative procedure on any of the lesser toes, the most frequent postoperative finding is swelling. Although swelling often persists for 1 to 6 months, it invari­ ably subsides with time.249,255,262 Postoperatively, the toe often assumes the shape of adjacent toes, which is known as molding. Molding of a lesser toe because of intrinsic pres­ sure from an adjacent toe is a common occurrence. It is often unavoidable,233 and it is best to counsel the patient

joints

• Decreased function after flexor to extensor transfer • “Molding” of the toe

preoperatively that this molding may occur. Usually, the main anxiety of the patient is the cosmetic appearance of the toes. Preoperative counseling alerts the patient to this possibility. Recurrence of deformity occasionally is noted (Box 25H-45). Adequate decompression of the deformity at the joint involved is necessary to prevent recurrence. Excessive resection can lead to an unstable or floppy toe, and inade­ quate resection can lead to recurrence. If the toe is not cor­ rected completely at the time of surgery, an intramedullary Kirschner wire should not be used to achieve correction only for stabilization. In this situation, when the Kirsch­ ner wire has been removed, the deformity often recurs. In the face of recurrence of a hammer toe deformity, a flexor tenotomy can aid in correction. Occasionally, recur­ rence of a mallet toe deformity is noted because the flexor digitorum longus tendon is not released. Postoperative malalignment is also a source of patient dissatisfaction.247 Arthrodesis or arthrofibrosis gives an inherent stability to the digit and helps to resist the deforming forces of lesser toes. Complications associated with Kirschner wire fixa­ tion are uncommon, but migration or breakage or pin tract infection occasionally occur.278 If patients develop pain at the site of the pseudarthrosis, an injection of corticosteroid usually gives lasting relief. The most significant complication after lesser toe sur­ gery is vascular compromise. If extensive surgical correction is required, it is preferable to do a two-stage repair rather than to incur a vascular injury to a lesser toe. If the circu­ latory status of the toe is impaired, constrictive dressings are corrected first. Occasionally, a Kirschner wire must be removed to improve circulation. When the Kirschner wire fixation is removed, the surgical correction is maintained with postoperative dressings.

Box 25H-45 C  omplications Associated with Correction of Lesser Toe ­Deformity

• Recurrent deformity • Postoperative malalignment • Vascular compromise • Localized numbness • Kirschner wire complications: migration, breakage, ­infection

�rthopaedic ����������� S �ports ������ � Medicine ������� 2132 DeLee & Drez’s� O

Occasionally, a Kirschner wire fatigues and breaks. Usually, this break occurs proximal to the metatarsal articular surface, and if the proximal portion of the pin is retained completely within the metatarsal, it does not need to be removed. If the pin compromises joint func­ tion, it can be removed through an arthrotomy of the involved joint. Injury to an adjacent digital nerve results in an area of numbness, which is associated with patient dissatisfaction. Preoperative toenail deformity typically does not resolve after correction of a mallet toe, and the patient should be counseled appropriately.

Criteria for Return to Play Walking is initiated 6 to 8 weeks after surgery in a stiffsoled shoe with a roomy toe box. Taping of the toe begins at 4 weeks when pins are removed and continues for a total of 6 weeks. Occasionally, swelling prohibits appropriate shoewear, and initiation of walking is delayed. The ath­ lete progresses to running at 9 to 12 weeks postoperatively. Osteotomies of the metatarsal should be healed both clini­ cally and radiographically before advancing to running activities. Athletic activities are advanced after 12 weeks as pain and swelling allow. Full return to sport is expected 4 to 6 months after surgery (Box 25H-46).

BUNIONETTES A bunionette, or tailor’s bunion, is characterized by a prominence of the lateral eminence of the fifth metatarsal head. This deformity originally was described by Davies,280 who observed that pressure over the lateral aspect of the fifth metatarsal head could lead to chronic irritation of the overlying bursa. Sitting in a cross-legged position gave rise to the term tailor’s bunion. Pressure from constrict­ ing footwear leads to the development of thickened kera­ toses over the lateral281-283 or plantar lateral284 aspect of the metatarsal head. Typically, the fifth toe deviates in a medial ­direction at the MTP joint, whereas the fifth meta­ tarsal head deviates in a lateral direction with respect to the fourth metatarsal.

Anatomy and Biomechanics Kelikian285 stated that a bunionette could be considered analogous to the medial eminence of the first metatarsal in hallux valgus. The cause and anatomic variations that are present with a bunionette appear to be much more

Box 25H-46 C  riteria for Return to Play After Correction of Lesser Toe Deformity

complex, however, than those originally described by Kelikian285 and Davies.280 Several anatomic factors have been attributed to the development or presence of a bun­ ionette deformity (Box 25H-47). An enlarged fifth metatarsal head can lead to a bun­ ionette deformity (Fig. 25H-90A).281,286-288 Although hypertrophy of the lateral condyle of the fifth metatarsal head occurs,281,288 Throckmorton and Bradlee287 and later Fallat and Buckholz286 reported that with pronation of the forefoot, the lateral plantar tubercle of the fifth metatar­ sal head rotates to a more lateral position, creating the radiographic impression of fifth metatarsal head enlarge­ ment. These investigators noted an average increase of the 4-5 intermetatarsal angle of 3 degrees with pes planus. Whether there is true hypertrophy of the fifth metatarsal head or a prominence of the fifth metatarsal head owing to pronation of the foot, the lateral condyle of the metatarsal head may become symptomatic without divergence of the fifth metatarsal. The angular measurements that define a bunionette deformity are the 4-5 intermetatarsal angle and the MTP-5 angle (see Fig. 25H-90B). The MTP-5 angle is the mag­ nitude of medial deviation of the fifth toe in relation to the fifth metatarsal shaft. The 4-5 intermetatarsal angle cal­ culates the divergence of the fourth and fifth metatarsals and is measured by the intersection of lines bisecting the base and neck of the fourth and fifth metatarsals.289 A 4-5 intermetatarsal angle of greater than 8 degrees is consid­ ered abnormal (see Fig. 25H-90C).288-291 Lateral bowing of the fifth metatarsal diaphysis results in prominence of the lateral metatarsal condyle (see Fig. 25H-90D).281,286,291-294 Nestor and colleagues,293 in reporting on the anatomic variations in patients with symptomatic bunionettes, found that an increased 4-5 intermetatarsal angle frequently was associated with a bunionette deformity, whereas fifth metatarsal bowing and an enlarged fifth metatarsal head were seen much less frequently. In general, a bunionette deformity is a static deformity that is exacerbated by excess pressure of footwear against a prominent fifth metatarsal head. Anatomic variations and repetitive activity can lead to symptoms in the ath­ letic population. A thickened or inflamed bursa with an ­associated hyperkeratotic lesion overlying the deformity is seen in athletes participating in repetitive activities such as ­running.

Box 25H-47 Anatomic Factors Contributing to Bunionette Deformity

• Enlarged or prominent fifth metatarsal head (width > 13 mm)

• Clinical and radiographic evidence of bony healing after

• Hypertrophy of the lateral condyle of the fifth metatar­

• Decrease in swelling that allows a stiff sole shoe with a

• Increased 4,5 intermetatarsal angle (>8 degrees) • Fifth metatarsal lateral bowing • Pes planus

metatarsal osteotomy roomy toe box

• Return to sport expected 4 to 6 months after surgery

sal head

Foot and Ankle 2133 Figure 25H-90   A, Enlarged metatarsal head. B, The 4-5   intermetatarsal angle and metatarsophalangeal 5 angle.   C, Increased 4-5 intermetatarsal   angle. D, Lateral angulation of the   distal fifth metatarsal. (A, C, and D, © M. J. Coughlin. Used by permission.)

Metatarsal phalangeal-5 angle

4–5 intermetatarsal angle

A

C

B

D

Classification

Evaluation

Three types of deformity have been described (Box 25H-48).286 Coughlin295 further reported the frequency of occurrence of each type of deformity. Coughlin noted a type 1 deformity in 27% of cases, type II in 23% of cases, and type III in 50% of cases.

Clinical Presentation and History The major subjective complaints of an athlete are pain and irritation caused by friction between the underlying bony abnormality and restricting footwear. Patients complain

�rthopaedic ����������� S �ports ������ � Medicine ������� 2134 DeLee & Drez’s� O

Box 25H-48 Classification of Bunionette ­Deformities

Box 25H-50 Treatment Options for Bunionette

Type I: Enlargement of the metatarsal head or lateral exostosis Type II: Abnormal lateral bend of the fifth metatarsal with a normal 4,5 intermetatarsal angle (23%) Type III: Increased 4,5 intermetatarsal angle (>8 degrees [50%])

• Shoe modifications • Callus shaving • Orthotics

of swelling, pain with shoes, and callus formation over the deformity (Box 25H-49).

Nonoperative

Operative

• Lateral condylectomy • Metatarsal head resection • Distal metatarsal osteotomy • Diaphyseal osteotomy • Proximal fifth metatarsal osteotomy • Fifth ray resection

Physical Examination On clinical evaluation, the examiner inspects the lat­ eral eminence and lateral border of the foot for an inflamed bursa,284,287,296 a plantar keratosis,284,297,298 a lateral keratosis,299 or a combined plantar lateral kera­ tosis.282,299 Diebold and Bejjani299 noted that two thirds of the patients in their series had significant pes planus. Diebold and Bejjani299 also noted that one third of the patients had a plantar lesion, and half had a lateral kera­ totic lesion. Force plate studies or imprints that evaluate the pressure concentration on the plantar aspect of the foot are helpful in the analysis of a plantar keratosis (see Box 25H-49). A bunionette can develop in combination with a hallux valgus deformity. An increased 1-2 intermetatarsal angle combined with an increased 4-5 intermetatarsal angle results in a wide or splay foot abnormality.280,284,290,300,301

Imaging Radiographic evaluation includes standing anteroposterior and lateral radiographs. The 4-5 intermetatarsal angle and the MTP-5 angle are measured on standing anteroposte­ rior radiographs (see Box 25H-49).

Treatment Options Nonoperative Treatment Most bunionette deformities respond well to nonopera­ tive measures. Constricting footwear is often a signifi­ cant cause of symptoms in the athlete. Pain, swelling, and

Box 25H-49 Typical Findings in Bunionette Deformities

• Pain

and irritation over the lateral aspect of the fifth metatarsal head • Plantar or lateral keratosis formation • Elevated 4,5 intermetatarsal angle or metatarsophalan­ geal 5 angle

chronic irritation over the lateral bursa of the fifth meta­ tarsal head are reduced significantly by the use of properly fitted shoes.280,283,285,294,296,302 Padding of the prominent metatarsal head283,294 and shaving of the hypertrophic cal­ lus will relieve symptoms. An orthotic device can control pronation and secondarily can reduce discomfort over the prominent fifth metatarsal head (Box 25H-50).

Operative Treatment Surgical treatment of a bunionette is indicated for defor­ mities that continue to cause symptoms despite appropriate nonoperative measures. Numerous operative techniques have been proposed for surgical correction of a sympto­ matic bunionette deformity (see Box 25H-50). Lateral condylectomy is considered when an isolated enlargement of the fifth metatarsal head of lateral condyle occurs (Box 25H-51; Figs. 25H-91 and 25H-92). Failure of

Box 25H-51 Lateral Condylectomy

• Center a longitudinal skin incision over the lateral con­ dyle of the fifth metatarsal.

• Protect the dorsal cutaneous nerve of the fifth toe. • Create an inverted L-type capsular incision by

de­ taching the dorsal and proximal capsular attachments, allowing exposure of the fifth metatarsal head (see Fig. 25H-91). • Distract the fifth metatarsophalangeal (MTP) joint and release the medial capsule. • Resect the lateral eminence with an osteotome or ­sagittal saw (see Fig. 25H-92A). • Close the MTP capsule by suturing it to the dorsal peri­ osteum and to the abductor digiti quinti proximally (see Fig. 25H-92B). • If necessary, place a suture through a drill hole in the fifth metatarsal metaphysic dorsal and lateral to ensure a stable capsular closure and prevent recurrence or lateral subluxation of the MTP joint.

Foot and Ankle 2135 L-shaped capsule incision

Box 25H-52 Distal Metatarsal Osteotomy Types

• Distal chevron osteotomy • Distal oblique osteotomy • Crescentic osteotomy • Transverse distal osteotomy • Distal closing wedge osteotomy Abductor digiti quinti

Capsule

Figure 25H-91   An L-shaped capsular incision is used to expose the fifth metatarsal head.

lateral condylectomy as a treatment of the bunionette defor­ mity has led to the development of more extensive resec­ tion procedures. Excision of the fifth metatarsal head,303 resection of the distal half of the metatarsal,298 and fifth ray resection304 all have been used to treat a bunionette defor­ mity but are not appropriate in the initial treatment of the symptomatic athlete. These procedures are used as salvage procedures for infection, ulceration, or severe deformity. Less radical operative procedures that preserve foot or toe function are preferred in athletes. Many different distal fifth metatarsal osteotomy tech­ niques have been described for treating the sympto­ matic bunionette (Box 25H-52). Hohmann305 originally described a transverse osteotomy of the metatarsal neck, although the lack of stability of this procedure increases the risk for a transfer lesion or a malunion. Kaplan and associates282 used a distal closing wedge osteotomy inter­ nally fixed with 2-mm Kirschner wire. These authors282 suggested the use of internal fixation because they believed that a distal osteotomy is unstable and can rotate postop­ eratively with loss of correction. Most frequently, a distal oblique, crescentic, or chev­ ron osteotomy is considered. A distal oblique osteo­ tomy300,301,306 is performed from distal lateral to proximal

medial (Box 25H-53; Figs. 25H-93 to 25H-95). Haber and Kraft284 used a distal crescentic osteotomy. These authors did not use internal fixation and reported delayed healing and excessive callus formation at the osteotomy site. An alternative procedure is resection of the prominent lateral condyle in combination with a distal chevron osteotomy. Throckmorton and Bradlee287 performed a transverse chevron-type osteotomy without fixation, relying on this stable osteotomy shape to hold the postoperative position (Box 25H-54; Fig. 25H-96). Boyer and DeOrio307 describe fixation of the chevron osteotomy with a bioabsorbable pin. Recently, several minimally invasive techniques have been reported in the literature with good results.308,309 The indications for a diaphyseal fifth metatarsal oste­ otomy are a bunionette deformity associated with either an increased 4-5 intermetatarsal angle or lateral bowing of the distal metatarsal. MTP joint realignment with a lateral eminence resection is performed simultaneously if neces­ sary (Box 25H-55; Fig. 25H-97). A diaphyseal metatarsal osteotomy has been used to correct a bun­ionette defor­ mity.295,310 Voutey288 carried out a transverse osteotomy in the diaphysis but described problems with rotation, angulation, and pseudarthrosis. Yancey292 used a double ­transverse closing wedge osteotomy in the diaphyseal region to correct a bunionette deformity characterized by lateral angulation of the fifth metatarsal. Gerbert and colleagues311 recommended use of a biplane osteotomy

Medial capsular release

A

B

Figure 25H-92   A, The lateral eminence is resected, and a medial capsulotomy is performed. B, Lateral capsular reefing may be reinforced by a drill hole through the lateral metaphysis of the fifth metatarsal.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2136 DeLee & Drez’s� O

eminence.

cephalad direction to create an elevating effect on the dis­ tal fragment, then the fragment is rotated (Fig. 25H-98). Proximal osteotomies are associated with a higher inci­ dence of nonunion secondary to potential injury to the blood supply to the fifth metatarsal.299,312

capsular incision (see Fig. 25H-91)

Weighing the Evidence

Box 25H-53 Distal Oblique Osteotomy

• Make a midlateral longitudinal incision over the lateral • Release the proximal and dorsal capsule using an L-type • Release the abductor digiti quinti and resect the lateral

eminence with an osteotome or sagittal saw (see Fig. 25H-93A) • Create the oblique osteotomy of the metaphyseal neck using either a saw or osteotome (see Fig. 25H-93A). The osteotomy is oriented in a proximal lateral to ­distal medial direction. • Displace the distal fragment medially on the metatarsal and impact the bone on the proximal fragment (see Fig. 25H-93B and C). (The osteotomy can be fixed with a Kirschner wire [see Fig. 25H-94B]).

for a combined plantar lateral keratotic lesion to displace the distal fragment in a medial direction. Mann294 and Coughlin295,310 used an oblique diaphyseal fifth ­metatarsal osteotomy to treat diffuse keratotic lesions on either the plantar or plantar lateral aspects of the fifth metatarsal. The oblique orientation of the metatarsal osteotomy per­ mits a dorsal medial translation of the metatarsal as the dis­ tal fragment is rotated. Internal fixation was recommended with either a small fragment screw or wire loop or a Kirsch­ ner wire. Mann294 did not realign the fifth MTP joint with this procedure, and no results were reported, although one case of nonunion was noted. Coughlin310 modified Mann’s oblique diaphyseal osteotomy by performing a fifth MTP joint realignment and lateral eminence resection (see Box 25H-55). The fifth MTP medial capsular structures are not released because release might impair the circulation to the fifth metatarsal head. If a combination plantar lateral keratosis is present, the oblique osteotomy is oriented in a

A

B

C

Figure 25H-93   A, Distal oblique osteotomy coupled with lateral eminence resection. B, Resection of fifth metatarsal metaphysis with distal oblique osteotomy. C, Impaction of osteotomy site.

Kitaoka and Holiday313 reported on 21 feet that had undergone a lateral condylar resection. These authors313 concluded that a minimal correction was achieved with the procedure, although it did relieve symptoms. As Kelikian285 noted, “at best a lateral condylectomy is a temporizing measure like simple exostectomy on the medial side of the foot; in time, deformity will recur.” The only indication for a lateral condylectomy is an enlarged lateral condyle. Symptomatic relief often follows a condylectomy alone; however, a distal metatarsal osteotomy achieves greater correction of the deformity. Kitaoka and Leventen314 reported an average of 5 degrees of correction of the 4-5 intermetatarsal angle and a diminished forefoot width of 4 mm with 87% patient satisfaction after distal oblique osteotomy. Sponsel,301 who advocated an oblique distal osteotomy, noted an 11% delayed union rate, and Keating and coworkers300 reported 75% of patients to have transfer lesions with a 12% recur­ rence rate. Pontious and colleagues315 reported a much

A

B

Figure 25H-94   A, Preoperative radiograph shows severe bunionette deformity. B, After distal oblique osteotomy with internal fixation. (Courtesy of H. Zollinger-Kies, Zurich, Switzerland.)

Foot and Ankle 2137

A

B

Figure 25H-95   A and B, Preoperative and postoperative radiographs of distal oblique osteotomy. (A and B, © M. J. Coughlin. Used by permission.)

higher rate of success in oblique osteotomies that were internally fixed (see Fig. 25H-94). Throckmorton and Bradlee287 and others283,316,317 reported high levels of good and excellent results with the chevron osteotomy. Kitaoka and associates317 reported

Box 25H-54 Distal Chevron Osteotomy

• Make a midlateral longitudinal incision over the lateral eminence.

• Release the proximal and dorsal capsule using an L-type capsular incision (see Fig. 25H-91).

• Avoid soft tissue stripping to avoid vascular insult to the

distal metatarsal fragment. about 2 mm of the lateral eminence with an osteotome or a sagittal saw. • Mark the apex of the osteotomy with a drill hole in the midportion of the metatarsal. • Create a horizontal chevron osteotomy with a sagittal saw. The osteotomy is based proximally with an angle of 60 degrees (see Fig. 25H-96A) and oriented in a ­lateralto-medial direction. • Displace the distal fragment about 2 to 3 mm in a ­medial direction and impacted onto the proximal phalanx (see Fig. 25H-96B). • Use Kirschner wire fixation when necessary. • Remove any remaining prominent bone in the metaphy­ seal region of the fifth metatarsal with a sagittal saw. • Reef the lateral capsule to the abductor digiti quinti or the dorsal periosteum of the fifth metatarsal. If ­necessary, reattach the capsule through drill holes on the ­dorsal aspect of the metaphysis.

• Remove

the 4-5 intermetatarsal angle was reduced an average of 2.6 degrees, and the MTP-5 angle was reduced an average of 8 degrees with a chevron osteotomy. Moran and Claridge318 stressed that there was a low margin of error with this oste­ otomy and that there was a high risk for either recurrence or overcorrection. As a result, these authors encouraged the use of Kirschner wire stabilization of the osteotomy site. Coughlin295 reported on 30 feet that had undergone a midshaft diaphyseal metatarsal osteotomy. All went on to successful union. The average 4-5 intermetatarsal angle was reduced 10 degrees, and the MTP-5 angle was reduced 16 degrees. No transfer lesions developed, and a 93% patient satisfaction rate was reported. The average foot width was reduced 6 mm. Midshaft osteotomies do not appear to have an increased nonunion rate. Vienne and colleagues319 reported good or excellent results in 97% of their series of patients with diaphyseal osteotomies for bunionette correction. In this study, the 4-5 intermetatarsal angle was reduced from an average of 10 degrees preopera­ tively to 1 degree postoperatively. Shereff and associates312 noted that more proximally positioned osteotomies are at increased risk for delayed healing as a result of interruption of the interosseous and extraosseous blood supply to the proximal fifth metatarsal (Fig. 25H-99). Salvage procedures are less desirable in the athletic popu­ lation. Although McKeever298 reported a high level of suc­ cess in his series of 60 cases of metatarsal head resection, no criteria for postoperative evaluation were included. Kitaoka and Holiday320 reported on a small series and noted that 82% had fair or poor results, including the complications of severe shortening, transfer metatarsalgia, stiffness, and continuing symptoms. Dorris and Mandel321 reported malalignment of the fifth toe in 59% of patients, and Addante and col­ leagues322 reported malalignment and wound problems after metatarsal head resection and silicone implant arthroplasty.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2138 DeLee & Drez’s� O Figure 25H-96   A, Lateral view of a fifth metatarsal chevron osteotomy. B, Anteroposterior view after chevron osteotomy.

A

B

Author’s Preferred Method

Box 25H-55 Oblique Diaphyseal Osteotomy

• Make a midlateral longitudinal incision is from the base

of the fifth metatarsal to the middle of the proximal phalanx (see Fig. 25H-97A). • Carry the dissection down to the fifth metatarsal shaft. • Protect the dorsal cutaneous nerve and retract the ­abductor digiti quinti in a plantar direction to expose the diaphysis of the fifth metatarsal. • Create an L-type capsular incision (see Fig. 25H-91) to expose the lateral eminence. • Remove the lateral eminence with a sagittal saw or ­osteotome. • Perform the resection in a line parallel with the ­metatarsal shaft. • Make a direct horizontal osteotomy in a dorsal ­proximal– to–plantar distal plane (see Fig. 25H-97B). • Drill the fixation holes before final displacement of the osteotomy site. • Drill a gliding hole in the dorsal distal fragment and create a tapped fixation hole in the proximal plantar fragment. • Complete the osteotomy and rotate the distal fragment medially (see Fig. 25H-97C). • Fix the osteotomy with either a small fragment compres­ sion screw or two minifragment compression screws. • Repair the fifth MTP joint capsule and bring the fifth toe into proper alignment (see Fig. 25H-97D and E). • Approximate the abductor digiti quinti and the MTP capsule. If necessary, reattach the capsule through drill holes on the dorsal aspect of the metaphysis.

Conservative management of a symptomatic bunionette in­ cludes the use of padding, shaving of keratotic lesions, and roomy footwear. In many cases, an athlete can continue sports activities with the use of orthotic devices or pads. Development of chronic bursal thickening, blistering, and symptomatic keratoses leads to operative treatment in certain patients. Because of the risk for transfer lesions, recurrence of deformity and malunion, delayed union, or nonunion of osteotomy sites, surgical intervention should be delayed until a patient experiences significant difficulty in sports activities. Surgical versatility in the treatment of a bunionette de­ formity is important. Attention to the underlying pathol­ ogy helps to determine whether a condylectomy with a distal soft tissue repair, a distal metatarsal osteotomy, or a diaphyseal biplane osteotomy offers the best treatment for the symptomatic bunionette deformity in the athlete. Analysis of the physical findings including examination of the plantar aspect of the foot for the presence of keratotic lesions helps to differentiate the type of bunionette present and the appropriate treatment. Evaluation of radiographs is necessary to analyze the nature of the deformity. When an enlarged fifth metatarsal head or medial emi­ nence is present (with or without a pronated foot or fifth ray), lateral condylectomy with MTP joint realignment or distal metatarsal osteotomy is the treatment of choice. The presence of a pure lateral keratotic lesion makes a chevron osteotomy preferable because of the stability of this osteotomy. Kirschner wire fixation often helps to stabilize the osteotomy site. When a plantar lateral keratotic lesion is present (with or without an increased 4-5 intermetatarsal angle or lateral deviation), a distal oblique osteotomy as described by Kitaoka and Leventen314 or a diaphyseal biplane osteotomy is considered. When there is an abnormally wide 4-5 intermetatarsal angle or when lateral deviation of the distal fifth metatarsal is present, a diaphyseal biplane osteotomy affords an excellent means of correction. Although there is some disagreement about the need for internal fixation after a fifth metatarsal osteotomy,315 the development of delayed union, malunion, nonunion, or transfer lesions in patients in whom floating osteotomies have been performed indicates a need for internal fixation.

Foot and Ankle 2139

A

C B

D

E

Figure 25H-97   A, Incision for a fifth metatarsal diaphyseal osteotomy. B, The horizontal osteotomy is performed from a proximal-dorsal to a plantar distal site. C, The osteotomy is rotated. D, Preoperative radiograph. E, Postoperative radiograph after oblique osteotomy. (D and E, © M. J. Coughlin. Used by permission.)

Postoperative Prescription, Outcomes Measurement, and Potential Complications Although recovery from bunionette surgery usually is rel­ atively rapid, conservative methods often are used either to alleviate symptoms or to help postpone surgery until

the off-season. Postoperatively, the foot is wrapped in a soft gauze and tape dressing, and the patient ambulates in a postoperative shoe. Sutures are removed 2 weeks after surgery. Pins are removed at 6 weeks, and the toe is taped in the proper alignment for 4 more weeks. A below-knee cast can be used if the surgeon is concerned about fixa­ tion or the reliability of the patient. Weight-bearing in

�rthopaedic ����������� S �ports ������ � Medicine ������� 2140 DeLee & Drez’s� O

A

90°

Saw blade 90°

20° Saw blade

12mm

B

C

Figure 25H-98   A, To achieve elevation of the fifth metatarsal head, the saw is oriented in a cephalad direction, and the osteotomy site is rotated. B, Schema illustrating effect of horizontal osteotomy. With the saw blade oriented in a lateral-to-medial direction, the osteotomy site is rotated and does not elevate the distal metatarsal. C, Schema illustrating effect of oblique osteotomy. With the saw blade oriented in a medial-to-lateral but also superior direction, as the osteotomy site is rotated, the distal fragment is elevated. (B and C, Adapted from Lutter L: Atlas of Adult Foot and Ankle Surgery. St. Louis, Mosby, 1997, pp 110-111.)

a postoperative shoe usually is carried out by having the patient bear more weight on the inner aspect of the foot at first. By 3 weeks, a plantigrade stance and gait pattern are ­encouraged. Recurrence of deformity is the most common com­ plication after lateral condylectomy (Fig. 25H-100).

A

B

­ ccasionally, fifth MTP joint realignment is complicated O by joint subluxation or dislocation (Fig. 25H-101). An adequate capsular joint repair helps to avoid this complica­ tion. Any dissection in this area can result in injury to the ­lateral ­cutaneous nerve to the fifth toe (a branch of the sural nerve), leading to numbness or formation of neuroma.

C

Figure 25H-99   A, Preoperative radiograph shows moderate bunionette deformity. B, After proximal fifth metatarsal osteotomy. C, Symptomatic nonunion after proximal osteotomy. This osteotomy took about 12 months to heal and prevented the patient from participating in high school athletics for that season. (From Mann RA, Coughlin MJ [eds]: Surgery of the Foot and Ankle, 6th ed. St. Louis, Mosby, 1993.)

Foot and Ankle 2141

A

B

C

Figure 25H-100   A, Radiograph shows type 1 bunionette deformity with a large metatarsal head. B, After lateral condylectomy.   C, Recurrence 3 years after lateral condylectomy.

Dorsal angulation with development of an IPK or trans­ fer lesion beneath the fourth metatarsal head is a reported complication of distal metatarsal osteotomy. Delayed union or nonunion has been observed as well. The use of internal fixation reduces the incidence of malunion and nonunion after fifth metatarsal osteotomies. An oblique diaphyseal osteotomy can also be complicated by malunion, delayed union, nonunion, or transfer metatarsalgia.

Criteria for Return to Play Return to athletic activities is expected earlier after a lat­ eral condylectomy or MTP joint realignment than after a fifth metatarsal osteotomy. After a lateral condylec­ tomy, usually aggressive walking can be initiated 4 weeks

after surgery, with running after 6 weeks. After a distal ­osteotomy, aggressive walking is initiated at 8 weeks, and if no complications are encountered, jogging and running can be started progressively between 10 and 12 weeks postoperatively. Diaphyseal osteotomies require slightly longer healing time. In these cases, aggressive walking is initiated between 8 and 10 weeks and progressive run­ ning after 12 weeks. If there is clinical concern of incom­ plete healing of the osteotomy, progression of activity should be delayed. Roomy footwear with an adequate toe box is more comfortable during initiation of athletic activities.

C

Figure 25H-101   Dislocated metatarsophalangeal joint after simple condylectomy.

r i t i c a l

P

o i n t s

 ost conditions of the forefoot are successfully treated l M nonoperatively in athletes. l When athletes with forefoot deformity (hallux valgus or lesser toe deformity) have pain associated with shoewear, modification of shoes by stretching constricting areas or relieving pressure areas can relieve the athlete’s symp­ toms completely. l Choosing appropriate orthotics or shoes for an athlete depends on careful examination and recognition of subtle malalignment that may be contributing to symp­ toms. l When choosing operative treatment, the surgical procedure(s) must address all anatomic abnormalities present. l Athletes should expect up to 3 to 6 months until full return to sport after most forefoot surgeries, espe­ cially when deformity correction through osteotomy is required.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2142 DeLee & Drez’s� O

S U G G E S T E D

R E A D I N G S

Anderson RB: Turf toe injuries of the hallux metatarsophalangeal joint. Tech Foot Ankle Surg 1(2):102-111, 2002. Baxter DE: Treatment of bunion deformity in the athlete. Orthop Clin North Am 25(1):33-39, 1994. Boyer ML, DeOrio JK: Transfer of the flexor digitorum longus for the correction of lesser-toe deformities. Foot Ankle Int 28(4):422-430, 2007. Campbell D: Chevron osteotomy for bunionette deformity. Foot Ankle Int 2:355-356, 1982. Coughlin MJ: Hallux valgus in the athlete. Sports Med Arthrosc Rev 2:326-340, 1994. Coughlin MJ: Operative repair of the mallet toe deformity [erratum appears in Foot Ankle Int 16(4):241, 1995]. Foot Ankle Int 16(3):109-116, 1995.

Coughlin MJ: Second metatarsophalangeal joint instability in the athlete. Foot An­ kle 14(6):309-319, 1993. Coughlin MJ: Sesamoid pain: Causes and surgical treatment. Instr Course Lect 39:23-35, 1990. Coughlin MJ, Dorris J, Polk E: Operative repair of the fixed hammertoe deformity. Foot Ankle Int 21(2):94-104, 2000. Mann RA, Mann JA: Keratotic disorders of the plantar skin. J Bone Joint Surg Am 85(5):938-955, 2003.

R eferences Please see www.expertconsult.com

S e c t i o n

I

Osteochondroses and Related Problems of the Foot and Ankle S. Terry Canale and David R. Richardson

OVERVIEW OF OSTEOCHONDROSIS The term osteochondrosis has been applied to more than 50 eponymic entities to describe a variety of conditions characterized by abnormal endochondral ossification of physeal growth. Osteochondrosis is the singular term and is used to describe a noninfectious disease process involving the growth or ossification centers in children that begins as a degeneration or osteonecrosis followed by regenera­ tion or recalcification. Because normal endochondral ossi­ fication does not always present a uniform radiographic ­pattern, the differentiation of osteochondrosis from nor­ mal growth often is difficult. The etiology of osteochondrosis is complex and has been described as traumatic, constitutional, idiopathic, and hered­ itary. Most authors now believe that multiple factors are responsible for these changes. For example, excessive physi­ cal demands during athletic activity may incite osteochondral

changes in growing bone made vulnerable by constitutional factors. Once the process has begun, repetitive trauma or pressure may prolong recovery or contribute to deformity. All osteochondroses heal, but treatment may be required to relieve pain or prevent residual deformity, especially in osteochondroses around the foot and ankle in athletes. The osteochondroses have been classified according to etiology, anatomic location, and type of growth center, but none has had much practical application. In a clinically ori­ ented classification, Siffert1 divided the osteochondroses into three basic groups: articular, nonarticular, and phy­ seal (Table 25I-1). Mintz and colleagues2 described a clas­ sification based on the magnetic resonance imaging (MRI) appearance of the lesion (Table 25I-2). Osteochondroses of specific locations also have been further classified, and these specialized classification systems are discussed with the specific osteochondroses. According to the Siffert classification, osteochondro­ ses involving the foot and ankle generally are articular or

  TABLE 25I-1  Clinical Classification of Osteochondrosis Type

Involvement (Example)

Characteristics

Treatment

Articular

Primary: articular and epiphyseal cartilage and subadjacent endochondral ossification (Freiberg disease) Secondary: articular and epiphyseal cartilage as consequence of osteonecrosis of adjacent bone (Legg-Calvé-Perthes disease, Köhler disease) Tendon attachment (Osgood-Schlatter disease) Ligament attachment (vertebral ring, epicondyle) Impact sites (Sever disease, Iselin disease) Long bones (tibia vara) Vertebrae (Scheuermann disease)

Degenerative arthritis, pain, limitation of motion

Minimize epiphyseal deformity Encourage joint congruity

Local pain with activity, local tenderness, adolescents, self-limited

Individualized Allow rapid, safe return to activity while minimizing sequelae

Nonarticular Physeal

Modified from Siffert RS: Classification of the osteochondroses. Clin Orthop 158:10-18, 1981

Foot and Ankle 2143

  TABLE 25I-2  Magnetic Resonance Imaging Classification of Osteochondrosis Grade

Magnetic Resonance Imaging Appearance

0 1 2 3 4 5

Normal cartilage Abnormal signal but intact Fibrillation or fissures not extending to bone Flap present or bone exposed Loose undisplaced fragment Displaced fragment

Metatarsal head (Freiberg)

Data from Mintz DN, Tashjian GS, Connell DA, et al: Osteochondral lesions of the talus: A new magnetic resonance grading system with arthroscopic correlation. Arthroscopy 19:353-359, 2003

nonarticular. Articular osteochondroses, such as Freiberg and Köhler diseases, may result in degenerative arthritis, pain, and limitation of motion, and treatment should be aimed at minimizing epiphyseal deformity and maximizing joint con­ gruity. Nonarticular osteochondroses, such as Sever disease of the calcaneus and Iselin disease of the base of the fifth metatarsal, cause local pain with activity and local tender­ ness, usually occur in adolescents, and are self-­limited con­ ditions. Treatment of any osteochondritic condition must be individualized to allow the athlete a rapid, safe return to activity and to minimize sequelae of the condition.

OSTEOCHONDROSES OF THE FOOT AND ANKLE Although a number of osteochondritic conditions in the foot and ankle have been described, most are rare and ­ usually asymptomatic. The most common osteochondroses in the foot and ankle that cause symptoms and require treatment affect the talus, calcaneus, navicular, cuneiforms, metatar­ sals, and sesamoids (Fig. 25I-1). Because an osteochondro­ sis may have its onset in childhood or adolescence and not become evident until adulthood, it is difficult to divide these conditions into clear-cut adult and pediatric categories.

Osteochondral Lesions of the Talus Konig,3 in 1888, first used the term osteochondritis dissecans to describe loose bodies in the knee joint, theorizing that they were caused by spontaneous necrosis of bone. In 1922, Kappis4 noted the similarity of lesions of the ankle to those in the knee and referred to osteochondritis dis­ secans of the ankle. In 1932, Rendu5 suggested that osteo­ chondritic lesions represented traumatic intra-articular fractures, and the primarily traumatic etiology of osteo­ chondral lesions of the talus (OLT) has been supported by numerous authors.6-12

Relevant Anatomy and Biomechanics The talus is a uniquely shaped bone divided into three ana­ tomic regions: dome, neck, and head. The talar dome artic­ ulates with the tibia and fibula on its superior, medial, and lateral surfaces to form the ankle joint. Inferiorly, it articu­ lates with the posterior facet of the calcaneus and, along with the inferior surfaces of the head and neck of the talus,

Cuneiforms (Buschke)

Navicular (Köhler)

Fifth metatarsal base (Iselin)

Talus (Koenig, Kappus)

Calcaneus (Sever) Figure 25I-1   Common sites of osteochondrosis in the foot. The specific form of disease is mentioned in parentheses.

forms the subtalar joint. It plays a key role in ankle motion and in supporting the axial load during weight-bearing. Because it lacks muscular or tendinous insertions, indirect perfusion of the talar dome is limited; injury to the artery of the tarsal canal, a branch of the posterior tibial artery, disrupts the main intraosseous blood supply to the central two thirds of the talar dome. In addition, about 60% of the talar dome is covered with hyaline cartilage, further reduc­ ing its vascular supply and reparative capacity.13 Boyd and Knight14 showed that the tibiotalar articulation is subjected to more load per unit area than any other joint in the body, and Millington and coworkers15 demonstrated that more force than previously thought is placed at the talar shoulders where osteochondral lesions typically occur. Several authors16-20 have hypothesized that even minimal talar displacement can result in medial stress concentration in the tibiotalar joint and lead to cartilage damage. Berndt and Harty21 proposed two possible mechanisms for osteochondral fractures of the talus: (1) compressive injury to a dorsiflexed and inverted ankle (direct tibiotalar impact) that crushes the subchondral bone of the lateral talar dome, with or without overlying cartilage damage; and (2) inversion and external rotation forces on a plantar flexed ankle that can produce osteochondral injuries to the medial surface of the talus. O’Farrell and Costello22 also described a combination of inversion and plantar flexion as a mechanism of injury for medial talar lesions. Other mechanisms have been described, including impaction of the talar dome against

2144 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

the lateral malleolus (causing lateral dome lesions) or against the posterior tibial lip (causing medial defects).21,23,24 Ankle sprains have been identified as the most common injuries leading to OLT.25-28 In ankle sprains that involve a plantar flexed foot forcibly supinating and injuring the anterior talofibular ligament, the plantar flexed foot and talus subluxate or dislocate, and the posteromedial corner of the talus strikes the tibial plafond, causing a chondral bruise, chondral cracks, shearing off of the corner of the talus with intact chondral attachments, or a complete frac­ ture of the corner of the talus. Lateral lesions usually are located anterior or central on the talar dome. Also, they are shallower and more wafer shaped than medial lesions, possibly because of a more tangential force vector that results in shearing-type forces. Medial lesions are central and posterior as well as deeper and cup shaped because the combination of inversion, plantar flexion, and external rotation forces causing the posteromedial talar dome to impact the tibial articular sur­ face has a relatively more perpendicular force vector.21,29

Classification The classification of Berndt and Harty, based on radio­ graphic findings, remains a useful system for describing osteochondral lesions of the talus. However, more recent classification systems reflect advances in technology, such as computed tomography (CT),30 MRI,31 and ankle arthroscopy,32,33 (Table 25I-3) and may contribute to a more accurate prognosis. Raikin and colleagues34 divided the talar dome articular surface into nine zones in a grid configuration: zone 1 was the most anterior and medial, zone 3 was anterior and lat­ eral, zone 7 was the most posterior and medial, and zone 9 was the most posterior and lateral (Fig. 25I-2). From MRI examinations of 424 ankles with reported osteochondral talar lesions, they recorded the frequency of involvement and size of lesion for each zone. The medial talar dome was more fre­ quently involved (62%) than the lateral talar dome (34%); in the anteroposterior direction, the midtalar dome was much more frequently involved (80%) than the anterior (6%) or posterior (14) thirds; zone 4 (medial and mid) was most fre­ quently involved (53%), with zone 6 second (26%). Lesions in the medial third of the talar dome were significantly larger and deeper than those in the lateral talar dome.

Evaluation Clinical Presentation and History In patients with acute injuries, the ankle and foot usually are swollen and painful, which can limit the specificity of physi­ cal examination. Patients with chronic injuries generally complain of mechanical symptoms, such as locking or giving way, or a feeling of instability of the ankle joint, in addition to pain and persistent swelling. Pain may occur only with certain ankle movements during sport or strenuous activity. An OLT should be considered in any patient who pre­ sents with a history of a “persistent ankle sprain.” Eighty percent of patients with traumatic OLT have a history of a seemingly benign ankle sprain. Taga and associates35 found cartilage lesions in 89% of acutely injured ankles and in

95% of ankles with chronic injuries. They concluded that the longer the time from initial injury, the more severe the associated cartilage lesions. They suggested that these car­ tilage lesions are the primary cause of persistent pain in patients with chronic ankle instability. Studies have found cartilage damage in up to 66% of ankles with lateral liga­ ment injuries and 98% of ankles with deltoid ligament injuries.36,37 No correlation has been found between the amount of cartilage damage and the severity of lateral liga­ ment injury.36 In contrast, Komenda and Ferkel26 found chondral injuries in only 25% of 55 unstable ankles.

Physical Examination and Testing The ankle and foot, especially, should be palpated to iden­ tify locations of tenderness; the medial and lateral cor­ ners of the talar dome should be palpated with the ankle ­maximally plantar flexed. A careful neurovascular examina­ tion is essential. Range of motion in the involved foot and ankle should be compared with that of the contralateral extremity. Stability of the ankle should be evaluated with an anterior drawer test with the ankle plantar flexed and dorsi­ flexed and with inversion and eversion stress testing. Other soft tissue or bony causes should be ruled out (Box 25I-1).

Imaging Oblique and plantar flexed radiographic views that avoid tibial overlap generally show the lesion more clearly than plain films. If radiographs are suggestive but not diagnostic of OLT, technetium-99m bone scanning can help identify localized bony pathology. If an OLT is suggested by either radiograph or bone scan, CT or MRI, or both, can provide more definitive information. Axial and sagittal CT cuts can help determine the location of the lesion (anterior, medial, or posterior) as well as its depth and size (Fig. 25I-3). MRI is useful for both preoperative evaluation and postopera­ tive follow-up. Anderson and colleagues38 demonstrated that low signal intensity in T1-weighted images is an early and definitive sign of even stage I lesions. A high signal rim between the osteochondral fragment and the talar bed is considered indicative of instability of the fragment, with the presence of joint fluid or fibrous granulation tissue as a result of the mobility of these fragments. It has been noted that the diameter of the lesion measured on MRI was significantly larger than indicated on radiographs,39 an important factor in preoperative planning. We recommend MRI evaluation to detect changes that provide information about detachment and viability of the fragment and help make the decision to preserve or to excise the fragment (Fig. 25I-4). MRI also may allow more appropriate treat­ ment because it delineates the lesion more accurately than either radiography or CT.40 OLTs that have a high signal rim on T2-weighted images are most likely unstable.41 In a study of 22 ankles with OLT, Higashiyama and associates42 found that the low and high signal rims present before sur­ gery disappeared in 100% and 77% of ankles, respectively. A decrease in or disappearance of the signal rim correlated well with clinical results: no patient in whom the signal rim persisted had a good result. It has been suggested that heli­ cal CT, MRI, and diagnostic arthroscopy are significantly better than history, physical examination, and standard

Foot and Ankle 2145

  TABLE 25I-3  Classification of Osteochondral Lesions of the Talus Radiographic Classification*

Stage

Radiographic Finding

I IIa IIb III IV V

Small area of compression of subchondral bone Partially detached osteochondral fragment Subchondral cyst on magnetic resonance imaging38 Completely detached osteochondral fragment remaining in the crater Displaced osteochondral fragment Radiolucent lesions on computed tomography scan43

Computed Tomography Classification�

Stage

Computed Tomography Appearance

I IIa IIb III IV

Cystic lesion of talar dome with intact roof Cystic lesion with communication to talar dome surface Open articular surface lesion with overlying nondisplaced fragment Nondisplaced lesion with lucency Displaced osteochondral fragment

Magnetic Resonance Imaging and Arthroscopy Classification� Cartilage

Grade

Magnetic Resonance and Arthroscopy Appearance

A B

Viable and intact Breached and nonviable

Bone

Stage

Magnetic Resonance and Arthroscopy Appearance

1 2 3 4

Subchondral compression or bone bruise that appears as high signal on T2-weighted images Subchondral cysts not seen acutely, develop from stage 1 lesions Partially separated or detached fragments in situ Displaced fragments

Arthroscopic Classification§

Grade

Arthroscopic Appearance

I II III

Intact, firm, shiny articular cartilage Intact but soft articular cartilage Frayed articular cartilage

Arthroscopic Classification¶

Grade

Arthroscopic Finding

A B C D E F

Articular cartilage is smooth and intact but may be soft or ballottable Articular cartilage has a rough surface Articular cartilage has fibrillations or fissures Articular cartilage with a flap or exposed bone Loose, nondisplaced osteochondral fragment Displaced osteochondral fragment

*Data from Berndt AL, Harty M: Transchondral fracture of the talus. J Bone Joint Surg Am 41:988-1029, 1959. †Data from Ferkel RD, Sgaglione NA: Arthroscopic treatment of osteochondral lesions of the talus: Long-term results. Orthop Trans 17:1011, 1993. ‡Data from Taranow WS, Bisignani GA, Towers JD, et al: Retrograde drilling of osteochondral fragments of the talar dome. Foot Ankle Int 20:474-480, 1999 §Data from Pritsch M, Horoshovski H, Farine I: Arthroscopic treatment of osteochondral lesions of the talus. J Bone Joint Surg Am 68:862-865, 1986. ¶Data from Cheng MS, Ferkel RD, Applegate GR: Osteochondral lesions of the talus: A radiologic and surgical comparison. In Ferkel RD, Whipple TL, Burst SE (eds): Arthroscopic Surgery: The Foot and Ankle. Philadelphia, Lippincott-Raven, 1996.

radiography for detecting or excluding OLT. Diagnostic arthroscopy does not perform better than helical CT and MRI.40 In general, arthroscopy should not be used as the initial method for diagnosing OLT.

Treatment Options Nonoperative Nonoperative treatment may be attempted for CT stage I or II lesions and for stage III lesions in skeletally imma­ ture patients. Nonoperative treatment of acute OLT

­ enerally involves an initial period of non–weight-­bearing g with cast immobilization, followed by progressive weightbearing and mobilization to full ambulation by 12 to 16 weeks. The recommended duration of nonoperative treatment is varied, with some authors recommending 6 months7 and others6 up to 12 months before opera­ tive treatment is chosen. Based on the results of nonop­ erative treatment of 35 chronic cystic OLT, Shearer and colleagues43 concluded that (1) nonoperative manage­ ment of chronic cystic OLT is a viable option with little or no risk for developing significant osteoarthritis; (2) most lesions remain ­ radiographically stable; (3) there is

2146 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Box 25I-1 D  ifferential Diagnosis of Osteochondral Lesions of the Talus

• Syndesmosis injury • Ankle soft tissue impingement lesions • Complex regional pain syndrome type I • Degenerative arthrosis • Occult talar fractures • Lateral ankle instability • Tarsal coalition • Peroneal tendon subluxation or tendinitis • Subtalar dysfunction Lavage, Débridement, and Excision

Small, chronic, symptomatic lesions may benefit from arthroscopic lavage and débridement by removing catabolic cytokines and loose bodies from the ankle, which can be the source of mechanical symptoms. However, ­adding curettage and drilling has been associated with better results.12,48 Figure 25I-2   Anatomic nine-zone grid scheme on the talar dome: nine equal surface area zones, with zones 1, 4, and 7 positioned on the medial talus and zones 1, 2, and 3 positioned anteriorly. (From Raikin SM, Elias I, Zoga AC, et al: Osteochondral lesions of the talus: Localization and morphologic data from 424 patients using a novel anatomical grid scheme. Foot Ankle Int 28:154-161, 2007.)

poor ­correlation between changes in lesion size and clini­ cal outcome, although the few patients with lesions that decrease significantly in size tend to do well, and those with lesions that increase significantly in size tend to do poorly; (4) the development of mild radiographic changes of ­osteoarthritis does not correlate with clinical outcome; (5) the general course of chronic cystic OLT is benign, with more than half of patients improving to good or excellent results with nonoperative management; (6) lat­ eral lesions tend to do better than medial ones; and (7) adult-onset lesions tend to do better than juvenile-onset lesions. Alexander and Lichtman44 suggested that a delay in treatment does not affect outcome; however, more recent studies have questioned this.45-47 Lesions present­ ing more than 1 year after injury or the onset of symptoms may have a poorer prognosis.45 Also, radiographic results are improved when the interval between injury and opera­ tive treatment is reduced.45,47 This indicates that early diagnosis and treatment are advisable.

Operative Options for open or arthroscopic treatment of OLT gen­ erally are based on one of three specific goals: (1) stimulat­ ing the bone marrow by débridement or drilling, with or without loose body removal; (2) securing the lesion to the talar dome so that it will heal in place; or (3) stimulating the development of hyaline cartilage. Techniques include excision, drilling, and curettage, alone or in combination; internal fixation with screws, Kirschner wires, or bioab­ sorbable devices; cancellous bone grafting; osteochondral autograft or allograft procedures; and autologous chondro­ cyte implantation or transplantation.

Curettage, Drilling, and Microfracture

Marrow-inducing reparative techniques, such as abrasion, drilling, and microfracture, aim to stimulate chondropro­ genitor cells within the underlying marrow. These stem cells populate the fibrin clot in the talar defect and produce a fibrocartilaginous matrix composed of chondroblasts, chondrocytes, fibrocytes, and an unorganized matrix that protects the surface from excessive loading. The dis­ advantage of these reparative techniques is the weaker ­mechanical properties of the fibrocartilage matrix, which lacks the normal biomechanical and viscoelastic character­ istics of normal tissue. Arthroscopic results appear superior to open procedures.48 Internal Fixation

Fixation devices include permanent or bioabsorbable low-profile pins, nails, or headless screws. Acute OLTs do markedly better after fixation than do chronic lesions. Lesions need to be larger than 8 mm to allow secure ­internal ­fixation. Restoration of Articular Hyaline Cartilage

Restorative techniques usually are recommended for defects larger than 2 cm2. These techniques can include autologous chondrocyte implantation (ACI), matrix or membrane ACI (MACI), collagen-covered autologous chondrocyte implantation (CACI), arthroscopic allograft or autograft (AAP) with platelet-rich plasma (PRP) implan­ tation, osteochondral autograft, osteochondral autologous transfer system (OATS) and mosaicplasty, fresh osteochon­ dral graft, stem cell–mediated implants, and scaffolds. Osteochondral autografting procedures such as OATS and mosaicplasty require the harvest of grafts from a donor site, such as the lateral supracondylar ridge or intercondy­ lar notch of the femur, for insertion into the OLT. These techniques generally are used for moderate to large grade III or IV lesions. Concerns about donor site morbidity have prompted graft harvest from sites other than the dis­ tal femur, including the anterior talar dome, and the use of allografts.

Foot and Ankle 2147

A

B

C

Figure 25I-3   A, Posteromedial osteochondral lesion of the talus (arrow). B, Coronal plane CT image. C, Axial plane CT image. (From Richardson DR: Ankle injuries. In Canale ST, Beaty JH [eds]): Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Fresh-frozen allografts (mega-OATS procedure) have been used for large osteochondral lesions in the knee, but seldom in the talus. Gross and colleagues49 listed as their indication for this procedure a lesion at least 1 cm in diameter and 5 mm deep that could not be internally fixed. Chondrocyte viability is a primary concern, and it is essential that the graft be harvested within 24 hours of the donor’s death and be stored at 4° C for less than 5 days. A benefit of using an ipsilateral talar allograft is the ability to harvest from a similar area as the defect and thus have a closely matched graft. Autologous Chondrocyte Implantation or ­Transplantation

ACI involves harvesting 200 to 300 mg of autologous chon­ drocytes from the distal femur, growing the cells in vitro for 2 to 5 weeks, then implanting them into the defect. An autologous periosteal flap is harvested and sewn over the implanted cells and sealed with fibrin glue. A “sandwich” procedure has been described for lesions with concomitant bone loss.50 The bony defect is grafted and covered with a periosteal patch with its cambium side facing the cartilage. A second periosteal patch with its cambium side facing the bone is sewn over the first patch to create a space for the

Figure 25I-4   Magnetic resonance imaging appearance of stage IV osteochondral lesion of the talus.

cells. ACI is best suited for large and well-contained stage III or IV defects, large lesions with extensive subchondral cystic changes, and lesions in which previous operative treatment has failed. According to Ferkel and Hommen,51 the ideal patient for ACI is between 15 and 55 years old and has no malalignment, degenerative joint disease, or instability of the joint. The procedure is contraindicated in bipolar (kissing) lesions that involve both the tibia and the talus.51,52 Because of concerns about donor-site mor­ bidity after harvest from the distal femur,53,54 other donor sites have been suggested. Giannini and associates55 used detached osteochondral fragments as a source of cells.

Weighing the Evidence Most of the literature about OLT consists of case series (level IV evidence) or case reports (level V evidence). For some of the newer techniques, numbers are too small and follow-up too short to make definitive recommendations. Shearer and colleagues43 reviewed the results of ­nonsurgical management of 35 OLTs and concluded that nonsurgical management of chronic cystic (stage 5) OLT is a viable option with little or no risk for development of osteoarthritis. Their clinical results were good or excellent in 54%, fair in 17%, and poor in 29%. A meta-analysis of 14 studies with a total of 201 patients showed only a 45% success rate of nonsurgical treatment of grades I and II and medial grade II OLTs, and nonoperative treatment of chronic lesions had a success rate of 56%.48 The highest success rate was obtained with excision, curettage, and drill­ ing (85%), followed by excision and curettage (78%) and curettage alone (38%). Shelton and Pedowitz56 reported just 25% satisfactory results after nonoperative treatment of grade II and III lesions. Gobbi and coworkers,57 in a randomized trial ­comparing chondroplasty, microfracture, and OATS in 33 patients, found no significant differences in clinical results among the three methods. However, each treatment group con­ tained a small number of patients, and three different sur­ geons were involved in the surgeries. Individual studies of various treatment methods have reported good results.

2148 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Lavage, Débridement, and Excision

In the meta-analysis by Verhagen and coworkers,48 exci­ sion alone had an overall success rate of 38%, with a range of 30% to 100% in individual studies. Excision and curettage had a success rate of 76% (range, 53% to 100%). Arthroscopic procedures had a higher success rate (84%) than did open procedures (63%). Savva and associ­ ates58 described repeat arthroscopic débridement in 12 of 215 patients who had arthroscopic treatment of OLT; at an average 6-year follow-up, results were good in all 12, and 8 had returned to their preinjury levels of sports. Curettage and Drilling or Microfracture, Bone Grafting

Good to excellent results after drilling have been reported in 28% to 93% of patients. Ferkel and colleagues51 reported 72% excellent or good results in 64 patients, Taranow and coworkers59 reported an 81% success rate in 16 patients with retrograde drilling, and Becher and Thermann60 reported 83% excellent and good results and 17% satisfactory results in 30 patients at an average follow-up of 2 years after micro­ fracture. To determine whether the presence of a subchon­ dral cyst affected the results of arthroscopic microfracture or abrasion arthroplasty, Han and colleagues61 compared the results in 20 defects with cysts to those in 18 defects without cysts and found no differences in the clinical results. They concluded that small cystic lesions can be successfully treated by arthroscopic microfracture or abrasion arthroplasty. Autogenous Cancellous Bone Grafting

Saxena and Eakin62 compared the results of microfracture procedures in 26 patients to those after bone grafting in 20 patients. Overall, 96% of patients had excellent or good results, and there was no difference between the groups in the percentages of those who returned to sports. Bone grafting, however, required a longer time to return to activity than did microfracture in high-demand patients, but the two groups had similar postoperative American Orthopaedic Foot and Ankle Society (AOFAS) scores. Regardless of treatment type, patients with anterolateral lesions had the fastest returns to activity and the highest AOFAS scores. Draper and Fallat63 compared the results of 14 patients treated with bone graft­ ing with those of 17 patients treated with curettage and drill­ ing. At almost 5-year follow-up, those with bone grafting had better range of motion and less pain. Kolker and colleagues,64 however, reported that 6 of 13 patients required further surgery after open antegrade autologous bone grafting and concluded that autologous bone grafting alone should not be used as primary treatment for patients with symptomatic advanced OLT and deficient or absent overlying cartilage.

Osteochondral Autografts (Osteochondral ­Autologous Transfer System, Mosaicplasty)

Scranton and associates65 reported 90% good to excel­ lent results in 50 patients with type V OLT at an average 3-year follow-up after osteochondral autograft transplan­ tation using a single, arthroscopically harvested graft from the distal femur. Thirty-two of their 50 patients (64%) had at least one previous operation that failed to relieve symp­ toms. Hangody and collegues66 described good to excellent results in 34 of 36 patients 2 to 7 years after mosaicplasty. Kreuz and coworkers67 used mosaicplasty procedures for the treatment of 35 OLTs after failure of arthroscopic excision, curettage, and drilling. The osteochondral graft was harvested from the ipsilateral talar facet, and a mal­ leolar or tibial wedge osteotomy was used to access central or posterior lesions. Although there were no nonunions of the osteotomies, patients with small osteochondral lesions accessible through an anterior approach without additional osteotomy had the best results. Osteochondral Allografts

Although several studies have reported good results with this technique in the knee, there are few reports of its use in the ankle. Gross and associates49 reported that six of nine allografts remained in situ with a mean sur­ vival rate of 11 years; three patients required arthrodesis because of graft resorption and fragmentation. Kim and colleagues68 used tibiotalar osteochondral shell autografts in seven patients; at 10-year average follow-up, only four had excellent or good results. Complications included graft fragmentation, poor graft fit, graft subluxation, and nonunion. Autogenous Chondrocyte Implantation or Transplantation

Koulalis and colleagues69 reported excellent to good results at 17 months’ follow-up in all 8 of their patients treated with ACI, and Whittaker and associates70 described ACI in 10 patients, 9 of whom were “pleased” or “extremely pleased” with their results at 4-year follow-up; however, 1 year after surgery, Lysholm knee scores had returned to preoperative levels in only 3 patients, suggesting donorsite morbidity in the other 7. Baums and coworkers52 reported 12 patients with ACI of the talus for defects that averaged 2.3 cm2. At about 5 years’ follow-up, 7 had excellent results, 4 had good results, and 1 had a satisfac­ tory result. The AOFAS mean score improved from 43.5 before surgery to 85.5 after surgery. Patients who had been involved in competitive sports were able to return to their full ­activity levels.

Author’s Preferred Method For a stage I or II OLT, non–weight-bearing in a cast or boot is first tried for 6 to 10 weeks, depending on the size of the lesion. If this fails to relieve symptoms, arthroscopic excision, curettage, and microfracture or drilling is done. In a skeletally immature patient with a stage III OLT, a trial of conservative treatment is warranted before surgical treat­ ment. For stage III or IV OLT in skeletally mature patients,

arthroscopic microfracture or drilling is the first choice and has obtained good results in about 90% of our patients (Fig. 25I-5). The use of a noninvasive ankle distractor (Fig. 25I-6) will help with visualization of posterior lesions. If this ­option fails to relieve symptoms, an osteochondral autograft ­(lesion < 1.5 cm2) (Table 25I-4) or allograft (lesion >1.5 cm2) is used.

Foot and Ankle 2149

Authors’ Preferred Method—cont’d

A

B

C

D

Figure 25I-5   A, Stage IV osteochondral lesion of the talus. B, Arthroscopic view of displaced osteochondral fragment.   C, Arthroscopic excision and drilling. D, Note vascular channels created in defect.

Technique of Arthroscopic Drilling, ­ xcision, or Pinning of OLT E

• View anterolateral lesions through an anteromedial por­

Figure 25I-6   Noninvasive ankle distraction for ankle arthroscopy. (From Richardson DR: Ankle injuries. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

tal, with instrumentation for drilling, excision, or pinning inserted through an anterolateral portal, changing portals as necessary for optimal viewing and fixation. • Posteromedial lesions can be more difficult to view and treat. With noninvasive distraction and use of a small, 2.7-mm scope in the anterolateral portal, most postero­ medial lesions can be treated through anteromedial and posterolateral portals. • Use a small, curved curet or curved microfracture awl to make perforations in the subchondral bone. • If needed, make a small bony trough on the anteromedial tibia to improve access to posterior lesions. • If the lesion still is not accessible, use a guide to place a Kirschner wire through the medial malleolus for drilling of the lesion (Fig. 25I-7). • A malleolar osteotomy may be required for pinning of larger lesions. • Other helpful instruments are an open-end curet, a small 2.7-mm full-radius resector, and a small 2.7-mm bur. Continued

2150 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d   TABLE 25I-4  Authors’ Preferred Treatment of OLT MRI Stage

Conservative Treatment

Primary Operative Treatment

After Failed Primary Procedure

I or II

Cast or boot: non–weight-bearing for 6-10 weeks depending on size; ankle brace for 3 months

Arthroscopic excision, curettage, microfracture/drilling

III

Cast or boot: Acute injury (< 10 weeks) skeletally immature: non-weight-bearing for 6-10 weeks Skeletally mature or chronic injury: proceed to operative treatment No role for conservative treatment

Minimally damaged surface: arthroscopic transtalar drilling Damaged surface: microfracture/drilling

Damaged surface < 1.5 cm2, osteochondral autograft transport; > 1.5 cm2, osteochondral allograft As above

IV

V

Cast or boot: weight bearing as tolerated for 6-10 weeks; ankle brace for 6 months

Minimally damaged surface: If < 1 cm2, arthroscopic transtalar drilling; If > 1 cm2, internal fixation Damaged surface: microfracture or drilling Minimally damaged surface: arthroscopic transtalar drilling + bone graft Damaged surface: microfracture or drilling

Technique of Osteochondral Autograft or Allograft Transplantation

• With

the patient under general anesthesia, prepare the affected lower extremity from the ankle to the knee. ­Examine the ankle arthroscopically to further delineate the chondral lesion. • Harvesters are made for lesions 5 to 11 mm (larger sizes also are available). • Approach lateral lesions through an anterior sagittal incision and perform a medial malleolar osteotomy for ­medial lesions. Rarely, a lateral malleolar osteotomy will be needed to access posterolateral lesions. • Use a commercially available recipient sizer and harvester to create a recipient hole for the donor osteochondral plug. Extract the plug to a depth of 10 mm (Fig. 25I-8A and B). Place the harvester perpendicular for dome ­lesions (see Fig. 25I-8C) and at 45 degrees for talar shoulder ­lesions.

As above

Cyst < 1.5 cm: as above + bone graft Cyst > 1.5 cm: bulk allograft

• Drill

multiple holes into the subchondral bone of the r­ ecipient hole (see Fig. 25I-8D). • Obtain a graft from the ipsilateral knee, ­arthroscopically from the medial femoral condyle, or from the lateral femoral condyle through a small incision (see Fig. 25I-8E and F). For talar shoulder lesions, obtain a graft from the lateral trochlea. • Use the specially designed donor harvester to obtain ­osteochondral grafts that measure 5 to 11 mm in diam­ eter and 10 to 12 mm in depth (slightly deeper than the recipient hole). • Insert the cylindrical grafts carefully into the recipient hole using the designed extruder or collared pin through the donor harvester (see Fig. 25I-8G and H). • Do not remove the OATS harvester before completion of full graft extrusion. Do not allow the harvester to deviate from the insertion angle. Either of these may cause frac­ ture of the donor core.

Anteromedial OLT

Transmalleolar portal

Kirschner wire

Anterolateral portal

A

Anteromedial port

B

Anteromedial OLT

Figure 25I-7   A, Transmalleolar drilling of osteochondral lesion using a guide. The scope is in the anterolateral portal, and inflow is through the posterolateral portal. B, Holes are drilled through the medial malleolus into the talus down to areas of bleeding bone. (From Ferkel RD: Arthroscopy of ankle and foot. In Mann RA, Coughlin MJ [eds]: Surgery of the Foot and Ankle, 8th ed. Philadelphia, Elsevier, 2006.)

Foot and Ankle 2151

Authors’ Preferred Method—cont’d

A

D

B

E

C

F

Figure 25I-8   Osteochondral autograft and allograft transplantation. A, Trial sizer for harvester. B, Recipient harvester. C, Plug 10 to 12 mm deep is removed from recipient site. D, Multiple holes are drilled at the base of the lesion. E, Autograft is obtained from the femoral condyle with a donor harvester (for talar shoulder lesions, graft is obtained from corner of trochlea). F, Donor graft in harvester.

• Use the sizer-tamp to gently tamp the core flush with the

surrounding cartilage. • Test range of motion of the ankle to ensure that the graft is well seated and secured. • Close the incision and secure the osteotomy in the usual fashion (see Fig. 25I-8I). Place one drain in the knee and

apply a compressive dressing to the ankle. Apply a poste­ rior splint with strips. For very large lesions, allografts can be harvested from an ipsilateral donor talus (see Fig. 25I-8J).

Continued

2152 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

I G

H

J

Figure 25I-8—cont’d   G and H, Graft is placed in recipient hole. I, Malleolar osteotomy is secured with two partially threaded cancellous screws (holes are predrilled before osteotomy). J, For large defects, allografts can be taken from donor talus. (From Richardson DR: Ankle injuries. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008; courtesy of Dr. Robert Anderson, Charlotte, NC.)

Postoperative Prescription, Outcomes Measurement, and Potential Complications Postoperative Prescription After arthroscopic excision, curettage, and drilling, the patient is non–weight-bearing in a boot for 4 weeks, then progresses to weight-bearing in the boot in physical therapy. Active motion is begun at 12 days after surgery. After inter­ nal fixation or OATS, the patient is non–weight-bearing in a cast for 8 weeks, then progresses to weight-bearing in a boot for 4 weeks in physical therapy. A brace is then worn during a gradual return to activities as symptoms dictate.

Outcomes Measures Clinical outcomes measures include pain relief, ankle sta­ bility, and ankle motion.

Potential Complications The most common complication is continued pain. Repeat MRI evaluation or second-look arthroscopy (Fig. 25I-9) is reasonable if pain persists after 4 months. We have had several patients with continued pain after osteochondral grafting of an OLT; all had complete or significant relief of symptoms after arthroscopic débridement of the graft. Other potential complications are wound infection and neural injury, most often to the superficial peroneal nerve.

Criteria for Return to Play • Minimal or no symptoms, minimal swelling • Participating in physical therapy out of boot • After internal fixation, OATS, or cartilage replacement, a brace must be worn while participating in sports for 6 months after the procedure.

Foot and Ankle 2153

26 with ­microfracture and 20 with autogenous bone graft­ ing. Results were excellent or good in 44 (96%) of the 46 lesions, and the average time to return to sports activity was 17 weeks. The return to activity was significantly lon­ ger in the bone graft group (20 weeks) than in the micro­ fracture group (15 weeks); return to sports was faster after arthroscopic treatment (16 weeks) than after arthrotomy (17.5 weeks), but there was no difference in postoperative AOFAS scores. Patients with anterolateral lesions had the fastest return to sports and the highest AOFAS scores.

Osteochondral Lesions of the Distal Tibia

Figure 25I-9   Second-look arthroscopy 15 months after osteochondral autologous transfer system (OATS) procedure shows good incorporation of the osseous portion of the graft, but incomplete incorporation of the cartilage. Repeat arthroscopy was done because of impingement symptoms, which resolved after surgery.

Special Populations Skeletally Immature Patients The literature concerning the treatment of OLT in skel­ etally immature patients is scarce, and treatment recom­ mendations usually are based on the Berndt and Harty classification: nonoperative treatment for stages I and II and medial stage III lesions and operative treatment for lateral stage III and stage IV lesions. Higuera and associ­ ates71 reported excellent or good results in 18 of 19 patients, and Kumai and colleagues45 reported good results in 10 of 11 with nonoperative treatment. In the series of Letts and associates,72 13 of 24 children initially treated non­ operatively required operative treatment. More recently, Perumal and associates73 concluded that few OLTs in skeletally immature patients heal with 6 months of nonop­ erative treatment. In their 31 patients (mean age 12 years), after 6 months of nonoperative treatment only 5 (16%) had complete clinical and radiographic healing, 24 (77%) had persistent lesions on radiographs, and 2 had severe pain. Of the 13 who subsequently had operative treatment, 11 healed clinically and radiographically within 12 months; the other 2 had persistent lesions on radiographs but no clinical symptoms. In their compilation of the literature reporting operative treatment of OLT in 48 children, Letts and colleagues72 found excellent or good results in 34 (71%), fair results in 12 (25%), and poor results in 2 (4%). In the only comparison of outcomes of surgery for OLT in adults and adolescents, Bruns and Rosenbach74 found that adolescents had better long-term outcomes than adults, regardless of the severity of the lesion.

High-Level Athletes Although a number of treatment methods have been shown to be successful in treating OLT in young, active patients, the ability to return to high-level athletic activ­ ity has not been well-documented. Saxena and Eakin,62 in a series of 44 athletic patients with 46 OLTs, treated

Osteochondral lesions of the distal tibia are much less com­ mon than those of the talus, and there is little informa­ tion in the literature about their etiology, natural history, or treatment. It appears that they, like talar lesions, are ­primarily caused by trauma. “Mirror image” or “kissing” lesions of the talus and distal tibia have been described.75 In one of the largest series of osteochondral lesions of the distal tibia,76 11 of 17 patients recalled an inversion injury to the ankle. Symptoms may include pain, stiffness, swelling, locking, and instability. Radiographs usually are not help­ ful, but MRI and CT can identify the lesion (Fig. 25I-10). Treatment is similar to that of osteochondral lesions of the talus: débridement and curettage of the lesion, abrasion of the defect to subchondral bone, and drilling or micro­ fracture of the subchondral bone. Mologne and Ferkel76 reported excellent or good results in 14 of 17 patients an average of 44 months after débridement, curettage, abra­ sion, and drilling or microfracture. If this is unsuccessful, a retrograde osteochondral autograft or allograft trans­ fer procedure can be done77,78; instrumentation has been developed to make this easier.

OTHER LESIONS THAT CAN MIMIC ANKLE SPRAINS Injuries of the ligaments around the ankle joint are com­ mon in athletic individuals, accounting for the highest percentage of injuries in epidemiologic studies of sports injuries,79-83 regardless of the sport, level of participation, or age or sex of the participants. Usually, these injuries are diagnosed promptly and treated appropriately; however, continued ankle pain or instability should raise suspicion of some other entity. Conditions that can be misdiagnosed as ankle sprains include fractures, neoplasms, impingement syndromes, and coalitions of the tarsal bones. Delayed or incorrect diagnosis can result in prolonged disability.

Fractures of the Talus and Calcaneus Fractures of the talar and calcaneal processes can be mis­ taken for ankle sprains (Table 25I-5). Misdiagnosis or delayed diagnosis of these injuries can lead to nonunion, which can cause ankle pain that limits athletic activity. Missed lateral talar process fractures were found in about 1% of 1500 patients initially diagnosed with lateral ankle sprains84; in a series of 25 anterior calcaneal process frac­ tures, 7 (28%) were initially diagnosed as anterior talo­ fibular ligament sprains.85 In 20 patients with avulsion

2154 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

C

B

Figure 25I-10   Osteochondral lesion of the distal tibia in a female collegiate basketball player. A, Coronal fat-suppressed magnetic resonance imaging. B, Axial computed tomography (CT) shows posterior-central lesion and small subchondral cysts. C, Sagittal CT shows depth of the defect. (From Mologne TS, Ferkel RD: Arthroscopic treatment of osteochondral lesions of the distal tibia. Foot Ankle Int 28:865-872, 2007.)

fractures of the posterior talus that were all initially diag­ nosed as ankle sprains, the average number of physician visits before correct diagnosis was about 6, and 1 patient was seen 17 times.86 Clark and coworkers87 reviewed ankle radiographs of 1153 patients with acute ankle trauma and determined that an ankle effusion of 13 mm or more had a positive predictive value of 82% for occult fracture.

Relevant Anatomy and Biomechanics The lateral talar process is an osseous protuberance that articulates superolaterally with the fibula and helps to stabilize the ankle mortise. Inferomedially, it articulates with the calcaneus to form the lateral portion of the sub­ talar joint. The posterior process of the talus is made up of the lateral and medial tubercles. The lateral tubercle is the larger of the two and serves as the attachment of the posterior talocalcaneal and posterior talofibular ligaments. The posterior third of the deltoid ligament attaches on the medial tubercle. The undersurface of both tubercles

forms the posterior fourth of the subtalar joint. An acces­ sory bone, the os trigonum, often is present posterior to the lateral tubercle and can be confused with a fracture of the lateral tubercle. Fractures of the lateral tubercle of the posterior process can be caused by hyper–plantar flexion or inversion, whereas those of the medial tubercle usually are caused by dorsiflexion and pronation injuries because the medial tubercle is avulsed by the deltoid ligament.

Evaluation and Classification Most fractures of the talar and calcaneal processes result in pain, tenderness, or swelling in specific locations (see Table 25I-5) that help distinguish them from ankle sprains and from each other. Patients with fractures of the anterior calcaneal process usually report a sudden twist of the ankle with immedi­ ate pain on the outer aspect of the midportion of the foot and discomfort with weight-bearing. Pain and tenderness are located in the region of the sinus tarsi, with maximal

  TABLE 25I-5  Talar and Calcaneal Process Fractures That Can Mimic Ankle Sprains Fracture

Mechanism

Physical examination

Radiograph

Lateral talar process

Inversion + dorsiflexion

Mortise view Lateral view

Posterior talar process (lateral tubercle)

Hyper–plantar flexion (compression fracture) or inversion (avulsion fracture)

Posterior talar process (medial tubercle)

Dorsiflexion, pronation

Anterior calcaneal process

Inversion + plantar flexion (avulsion fracture); dorsiflexion (compression fracture)

Point tenderness anterior and inferior to lateral malleolus (over lateral process); pain with plantar flexion, dorsiflexion, subtalar joint motion Tenderness to deep palpation anterior to Achilles tendon, over posterolateral talus; plantar flexion may produce pain; swelling in posterolateral ankle Tenderness to deep palpation between medial malleolus and Achilles tendon; swelling posterior to medial malleolus and anterior to Achilles tendon Point tenderness over calcaneocuboid joint (about 1 cm inferior and 3 to 4 cm anterior to lateral malleolus)

Lateral view

Oblique view (foot in 40 degrees of external rotation) Lateral view Lateral oblique view

Foot and Ankle 2155

  TABLE 25I-6  Classification of Calcaneal Anterior and Lateral Talar Process Fractures

Classification of Calcaneal Anterior Process Fractures

Type I Type II Type III

Nondisplaced tip avulsion Displaced avulsion fracture not involving the calcaneocuboid articulation Displaced, larger fragments involving the calcaneocuboid joint

Classification of Lateral Talar Process Fractures

Type A Type B Type C

Small, minimally displaced, extra-articular avulsion Medium-sized fracture involving only the talocalcaneal articular surface Larger fracture involving both talocalcaneal and talofibular articulations

tenderness 2 cm anterior and 1 cm inferior to the anterior talofibular ligament, which helps distinguish this lesion from a lateral ankle sprain. Avulsion fractures of the ante­ rior calcaneal process are best seen on a lateral oblique projection, whereas compression fractures are best seen on a lateral view of the hindfoot. Anterior calcaneal process fractures often are associated with other ankle pathology, including tarsal coalitions, ankle sprains, and bifurcate lig­ ament abnormalities. In a review of 1479 foot and ankle MRI studies, Petrover and associates88 found 15 fractures of the anterior process (1%), only 2 of which had no associ­ ated abnormality. Fractures of the anterior calcaneal proc­ ess generally are classified according to displacement and involvement of the calcaneocuboid joint (Table 25I-6).85 Fractures of the posterior process of the talus most often involve the lateral tubercle (Fig. 25I-11). Lateral tubercle factures can be caused by hyper–plantar flexion (compres­ sion) or inversion (avulsion) and have been associated with football and rugby kicking, which places the ankle in a forced plantar flexed position. Pain may be exacerbated by plantar flexion or, because of the proximity of the flexor hal­ lucis longus tendon, by dorsiflexion of the hallux. In almost

50% of ankles, the os trigonum (fused or separate) is just posterior to the lateral tubercle of the posterior talar proc­ ess and may be mistaken for a fracture. Differentiation of a fracture of the lateral tubercle from a nonunited second­ ary ossification center is best made on a lateral radiograph. An acute fracture is suggested by a rough, irregular corti­ cal surface along the line of separation, whereas a normal os trigonum has a smooth and rounded cortical surface. These fractures also can be classified according to size, dis­ placement, and joint involvement (see Table 25I-6). Fractures of the medial tubercle generally are caused by dorsiflexion-pronation injuries that cause avulsion of the medial tubercle by the deltoid ligament. They usually result in a tender, firm mass posterior to the medial mal­ leolus, with no ankle instability or limitation of motion. These fractures are difficult to see on routine radiographs, and CT or MRI may be necessary to confirm the diagnosis. Ebraheim and colleagues,89 in a cadaver study, determined that the 30-degree external rotation view of the ankle is most likely to show this injury. Fractures of the entire ­posterior process (both tubercles) have been described but are rare. Fractures of the lateral talar process, the ­“snowboarder’s fracture,” are relatively infrequent but should be suspected in patients who complain of lateral ankle pain after an inversion or dorsiflexion injury. From 33% to 41% of these injuries are missed on initial examination.84,90-92 Point tenderness over the lateral talar process should prompt CT evaluation. Von Knoch and associates93 found that 14 of 16 displaced or unstable lateral process fractures were associated with severe concomitant hindfoot injuries. A morphologic classification of these fractures has been described94: type I, chip fracture; type II, large fragment; type III, ­comminuted.

Treatment Nonoperative treatment generally is sufficient for acute process and tubercle fractures with small (<1 cm), mini­ mally displaced (<2 mm) fragments.95 Nonoperative treat­ ment also may be appropriate for larger fragments that are nondisplaced or minimally displaced.96 Usually 6 weeks of immobilization in a non–weight-bearing cast is followed by transition to a removable walking boot and progressive weight-bearing with crutches. Operative treatment should be strongly considered for large (>1 cm), displaced (>2 mm) fragments with significant articular involvement.90,91 Valderrabano and colleagues97 developed a treatment algorithm based on the fracture type (McCrory-Bladin94 classification) (Fig. 25I-12). Operative treatment may con­ sist of open reduction and internal fixation of large frag­ ments, primary excision of severely comminuted fractures, or delayed excision of chronic nonunions.

Weighing the Evidence

Figure 25I-11   Lateral radiograph shows fracture of the lateral tubercle of the posterior talar process (arrow). (From Judd DB, Kim DH: Foot fractures frequently misdiagnosed as ankle sprains. Am Fam Physician 66:785-794, 2002.)

Because of the relatively infrequent occurrence of these process and tubercle fractures, there is little evidence in the literature on which to base treatment recommendations. Most studies include only a small number of patients or are isolated case reports. In one of the larger series of ante­ rior calcaneal process fractures (25 fractures), all 18 treated

2156 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Fracture Type McCrory-Bladin Classification

Fracture Displacement

Type I Chip

Type II Large Fragment

Type III Comminuted

Primary Therapy

Secondary Therapy (if necessary)

Nonoperative Nondisplaced ORIF

Débridement

Displaced Débridement

Figure 25I-12   Treatment algorithm for fractures of the lateral talar process. (Redrawn from Valderrabano V, Perren T, Ryf C, et al: Snowboarder’s talus fracture: Treatment outcome of 20 cases after 3.5 years. Am J Sports Med 33:871-880, 2005.)

nonoperatively had good results, whereas 5 of 7 treated with excision had good results.85 The worst outcomes were in patients with the longest delays in diagnosis and ­treatment. Medial talar process fractures generally are reported to do well with nonoperative treatment, if diagnosed and treated promptly.98,99 Although Giuffrida and colleagues100 reported that five of six medial process fractures required arthrodesis after nonoperative treatment, the fractures in their series were all associated with high-energy medial subtalar dislocations. Lateral talar process fractures appear to do less well with nonoperative treatment. In a series of 23 lateral talar process fractures in snowboarders,93 7 minimally displaced fractures were treated nonoperatively, and 16 displaced or unstable fractures were treated with open reduction and internal fixation. Fifteen of the patients (10 treated opera­ tively and 5 treated nonoperatively) regained their prein­ jury levels of athletic activity. The 6 operatively treated patients who failed to regain preinjury participation levels all had severe associated injuries of the ankle or hindfoot. At 3.5-year follow-up of 20 lateral talar process fractures,97 the AOFAS scores were higher in the 14 treated with open reduction and internal fixation (ORIF) (97 points) than in the 6 treated non­operatively (85 points); all 14 patients treated with ORIF regained their preinjury levels of sport, whereas only 2 of the 6 treated nonoperatively did so.

Authors’ Preferred Method Most fractures of the calcaneus and talus that mimic an­ kle sprains heal with cast immobilization, and this is our preferred treatment unless displacement or comminution is severe or the fracture fragment is large. ORIF is indicat­ ed for noncomminuted, displaced fractures because large, displaced, articular fragments, if unreduced, have a high risk for nonunion. For displaced intra-articular process and ­tubercle fractures, especially those of the lateral talar process, that are too comminuted for internal fixation, pri­ mary excision allows early mobilization without the risk for painful nonunion.

Postoperative Prescription, Outcomes Measurements, and Potential Complications A short leg walking cast is applied over a compression dressing after ORIF and worn for 3 weeks, after which the dressing is removed, and a weight-bearing cast or brace is worn for another 3 weeks. After primary excision, immobi­ lization usually consists of 2 to 3 weeks in a weight-­bearing cast or removable boot. Full weight-bearing is allowed when radiographic healing of the fracture is evident. Phys­ ical therapy is instituted for muscular strengthening, pro­ prioception, balance, and sport-specific functions. The primary clinical outcome measures are pain relief, joint motion, and return to activity. The AOFAS hindfoot score can be used for functional evaluation. Radiographs or CT scans evaluate healing. The most frequent complications after process and tubercle fractures of the hindfoot are chronic pain and the development of arthritis. Symptomatic nonunion is likely if these fractures are not diagnosed and treated promptly. Late excision of the un-united fragment can improve symp­ toms,96,101-103 but results appear to deteriorate the longer the interval between nonunion and excision.85,101

Criteria for Return to Play Return to play depends on both the fracture and the ath­ lete. Motivated athletes generally can return to sports when fracture healing is documented and normal strength and motion have been regained. A range of motion equal to the uninjured ankle and 85% of contralateral strength should be obtained before returning to sport.

Impingement Syndromes The term ankle impingement encompasses a wide variety of soft tissue and bony conditions around the ankle that can cause chronic pain and decreased motion and limit athletic activity. Various forms of mechanical impingement can be caused by synovial proliferation, bone spur formation, or ligamentous scarring and hypertrophy. Impingement gen­ erally is described as anterolateral, anterior, or posterior. Combined anterior and posterior impingement also has been described.104

Foot and Ankle 2157

A

B

Figure 25I-13   Anterior impingement syndrome. A, Magnetic resonance image shows osteophyte on distal tibia. B, Radiograph after excision of osteophyte. (From Richardson DR: Ankle injuries. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Evaluation Anterior impingement probably is the most common of the impingement syndromes and occurs most often in bal­ let dancers and football, basketball, and soccer ­ players. The first symptom is pain that begins as a vague discomfort and becomes sharper and more localized to the front of the ankle and usually is exacerbated by cutting or pivoting maneuvers. Physical examination finds tenderness between the anterior tibial tendon and the medial malleolus, which is exacerbated with dorsiflexion and relieved with plantar flexion. With the ankle plantar flexed, exostoses can be pal­ pated on the superior surface of the talar neck. Imaging studies show a beak-like prominence at the anterior rim of the tibial plafond, usually associated with a corresponding area of the opposed margin of the talus proximal to the talar neck, within the anterior ankle joint capsule (Fig. 25I-13). These osteophytes can impinge on each other, and soft tissues can become entrapped between them. Anterior impingement has been classified into four grades to indi­ cate severity (Fig. 25I-14).105 Two distinctive impingement lesions in the anterior talus have been described: a localized “divot” that accepts the growing tibial spur during dorsi­ flexion, prevents the formation of an osteophyte on the anterior talar neck, and allows unimpeded dorsiflexion106; and a “tram track” lesion, formed when a prominent osteo­ phyte carves a longitudinal trough in the articular surface of the talar dome.107 Anterolateral impingement is believed to be caused by relatively minor inversion injuries of the ankle and to occur after about 3% of ankle sprains.108 Tearing of the anterolateral soft tissues and ligaments, without substan­ tial associated mechanical instability, followed by repeated microtrauma can result in hypertrophy of the synovial tissue and fibrosis in the anterolateral ankle gutter, causing pain and mechanical impingement. Several studies109-113 have suggested that a contributing factor may be hypertrophy of an accessory fascicle of the anterior tibiofibular ligament (Bassett ligament) (Fig. 25I-15). Most patients complain of chronic vague pain over the anterolateral ankle, exacerbated by cutting or pivoting maneuvers, and physical examina­ tion identifies tenderness along the lateral gutter and the anterior tibiofibular ligament. Maximal dorsiflexion and deep palpation to the anterolateral corner of the ankle joint

reproduces the pain. The “charger” stance, with the patient bearing weight, flexing the knee, and keeping the heel flat on the ground (Fig. 25I-16), also produces pain in this cor­ ner of the ankle.114 Impaired proprioception also has been identified in patients with anterolateral impingement. Anteromedial impingement is an uncommon cause of chronic ankle pain that also is associated with an inversion mechanism that injures the lateral and medial ankle liga­ ments. A meniscoid lesion or a thickened anterior tibio­ talar portion of the deltoid ligament may impinge on the

A

B

C

D

Figure 25I-14   Classification of anterior impingement. A, Grade I: synovial impingement, spur ≤3 mm. B, Grade II: osteochondral reaction exostosis, spur >3 mm. C, Grade III: severe exostosis with or without fragmentation, secondary spur on talus. D, Grade IV: pantalocrural osteoarthrotic destruction. (Redrawn from Ferkel RD, Scranton PE: Current concepts review: Arthroscopy of the ankle and foot. J Bone Joint Surg Am 75: 1233-1242, 1993.)

2158 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Anterior inferior tibiofibular ligament

Distal fascicle

Calcaneofibular ligament Anterior talofibular ligament

A

B

Figure 25I-15   A, Distal fascicle of the anterior inferior tibiofibular ligament is parallel and distal to the anterior tibiofibular ligament proper and is separated from it by a fibrofatty septum. B, With dorsiflexion of the ankle, the distal fascicle may impinge on the anterolateral aspect of the talus. (From Bassett FH 3rd, Gates HS 3rd, Billys JB, et al: Talar impingement by the anteroinferior tibiofibular ligament. J Bone Joint Surg Am 72:55-59, 1990.)

anteromedial corner of the talus during dorsiflexion of the ankle, resulting in osteophyte formation or a chondral lesion, or both. Posterior impingement has been described as os trigo­ num syndrome, talar compression syndrome, and posterior block of the ankle and has been extensively described in ballet dancers as well as in gymnasts, soccer players, and down-hill runners.110,111,115,116 The mechanism of injury has been compared with a nutcracker because the posterior talus and surrounding soft tissues are compressed between the tibia and the calcaneus during plantar flexion of the foot.117 Bony causes of posterior impingement include the os trigonum (an accessory ossicle of the lateral talar tuber­ cle that persists in as many as 7% of adults), the Stieda process (an elongated lateral talar tubercle), a prominent posterior calcaneal process, and loose bodies (Table 25I-7). Soft tissue factors can include synovitis of the flexor hallucis longus tendon sheath, the posterior synovial recess of the subtalar and tibiotalar joints, and the posterior intermal­ leolar ligament. Pain is worse with forced plantar ­flexion and with push-off maneuvers. Posteromedial impingement occurs after severe ankle inversion injury that damages the deep posterior fibers of the medial deltoid ligament. Chronic inflammation and hypertrophy of the ligament result in fibrotic scar tissue that can be trapped between the medial wall of the talus and the posterior margin of the medial malleolus. Figure 25I-16   “Charger” stance causes pain in the anterolateral corner of the ankle in patients with anterolateral impingement. (From Hyer CF, Buchanan MM, Philbin TM, et al: Ankle arthroscopy. In ElAttrache NS, Harner CD, Mirzayan R, Sekiya JK [eds]: Surgical Techniques in Sports Medicine. Philadelphia, Lippincott Williams & Wilkins, 2007.)

Treatment Options: Weighing the Evidence Initial treatment of any impingement lesion should be nonoperative, including restriction of activity, orthoses, nonsteroidal anti-inflammatory drugs (NSAIDs), and

Foot and Ankle 2159

  TABLE 25I-7  Etiology of Posterior Ankle Impingement Syndrome Pathology

Example

Trigonal process Synchondrosis injury True compression Flexor hallucis longus dysfunction Tibiotalar pathology

Fracture (acute or chronic)

Osteochondritis Fracture Subtalar pathology Arthritis Other Prominent calcaneus posterior process Combined

Tenosynovitis Posterior capsuloligamentous injury Osteochondritis Calcified inflammatory tissue Flexor hallucis longus tenosynovitis and synchondrosis injury

From Maquirriain J: Posterior ankle impingement syndrome. J Am Acad Orthop Surg 13:365-371, 2005.

possibly cortisone injection. If nonoperative methods are unsuccessful at relieving pain and mechanical symptoms, arthroscopic débridement of soft tissue lesions or excision of bony lesions, or both, is successful in most patients. In two of the larger series of anterolateral lesions,118,119 includ­ ing 64 lesions, 58 (91%) had excellent or good results, 4 had fair results, and 2 had poor results after arthroscopic treatment. At a mean follow-up of 6 years after arthros­ copy, 40 (70%) of 57 patients with anterior lesions had resumed sports, 21 of them playing soccer.120 Results were good in all patients who had no preoperative osteoarthritis (OA), in 73% of those with grade I OA, and in only 29% of ankles with grade II OA. Posterior arthroscopy also can be used for excision of an os trigonum, tenolysis, loose body removal, spur excision, and débridement of OLT.121 Two studies each reported that 14 of 15 patients returned to their preinjury levels of sports after posterior arthros­ copy.121,122 ­Henderson and La Valette104 described anterior arthroscopic and open posterior treatment for com­ bined anterior and posterior impingement in 62 patients; 47 (81%) had excellent or good outcomes, 9 (15.5%) had fair outcomes, and 2 (3.5%) had poor outcomes.

Authors’ Preferred Method If nonoperative measures do not relieve symptoms, arthros­ copy is indicated for excision, débridement, decompression, or synovectomy. Large exostoses may require open arthrot­ omy for excision. Arthroscopic treatment of ankle impingement begins with a careful inspection of all structures that may be con­ tributing factors (Table 25I-8). Standard anteromedial and anterolateral ankle portals are used (Fig. 25I-17). Technique of Arthroscopic Treatment of Ankle Impingement

• With the patient under general anesthesia, apply and in­ flate a thigh tourniquet. • Insert a needle just medial to the anterior tibial tendon and distend the ankle joint with 15 to 20 mL of saline.

  TABLE 25I-8  Ankle Impingement Syndromes: Arthroscopic Examination Medial Portal

Lateral Portal

Medial gutter Medial malleolus Deep fibers of deltoid ligament

Lateral gutter Lateral malleolus Anteroinferior talofibular ligament Anterior joint line, anterior tibia Talar dome Medial malleolus Medial gutter

Anterior joint line, anterior tibia Talar dome Tibiofibular joint, ligaments Lateral gutter

From Hyer CF, Buchanan MM, Philbin TM, et al: Ankle arthroscopy. In ElAttrache NS, Harner CD, Mirzayan R, Sekiya JK (eds): Surgical Techniques in Sports Medicine. Philadelphia, Lippincott Williams & Wilkins, 2007.

• Make a small longitudinal incision to allow insertion of a

2.7-mm or 4.0-mm, 30-degree angle arthroscope through an anteromedial portal just medial to the anterior tibial tendon. Take care to pass the arthroscope across the an­ terior aspect of the joint and not across the dome of the talus. • Make a separate anterolateral portal just lateral to the peroneus tertius tendon to allow inflow and outflow of saline. Be aware of the superficial peroneal nerve in this area. Instruments and the arthroscope can be switched to either portal as necessary. • Fully examine the ankle with the use of a noninvasive ­ankle distraction device as necessary (see Fig. 25I-6). ­Distraction may need to be removed to identify and gain access to large anterior osteophytes, especially on the talus, because distraction may cause the anterior capsule to tighten. • Use a pressure irrigation system with a 3.5-mm full-­radius resector to clear the anterior synovium and define the an­ terior tibial and superior talar bony spurs. • Use a 3-mm bur to remove the spurs, resecting them back to the level of normal cartilage. • Smooth off the tibial surface with a 3.5-mm full-radius resector. • Carry out a similar procedure on the superior neck of the talus. • Again, examine the whole ankle by passing the arthroscope gently over the dome of the talus. This can be accom­ plished with the use of manual distraction in mid–plantar flexion or with a commercially available noninvasive ankle distraction device. • After irrigation, place 20 mL of 0.25% bupivacaine into the joint, suture the incision, and apply a compression dressing. Continued

2160 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Authors’ Preferred Method—cont’d

A

B

Figure 25I-17   Ankle arthroscopy portals. A, Anteromedial portal site. B, Anterolateral portal site. (From Phillips BB: Arthroscopy of the lower extremity. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

If synovitis is found, typically along the anterior joint line and in the medial and lateral ankle gutters, an arthro­ scopic resector is used to remove the hypertrophic synovitis. ­Débridement begins in the medial gutter, then the anterior joint line, and finally the lateral gutter. Switching portals is necessary for complete débridement. Hypertrophy of the anterior tibiofibular ligament, which appears as a whitish, thick, scar-like lesion, is resected and débrided. For anterior bony impingement, the anterior ankle cap­ sule is carefully reflected and elevated superiorly off the ­anterior tibia and inferiorly off the talar neck. An arthro­ scopic bur is used to remove the exostoses on the tibia and talus. It is important to make sure the bur is facing bone to avoid damage to the soft tissues, especially the anterior neu­ rovascular bundle. A trough can be made with a 3-mm bur about 1 mm proximal and parallel to the anterior edge of the tibia. This trough is taken down to subchondral bone to the level of surrounding normal cartilage, and an arthroscopic bone biter is used to remove the bony spur. This allows more control of the bur with less risk for damage to the articular surface than may occur with an arthroscopic shaver.123 If the exostoses are large, open excision may be preferable through an anteromedial incision and arthrotomy. Tibial osteophytes typically are located laterally and the talar osteophytes medi­ ally. They can be easily removed with an osteotome. Posterior synovectomy can be done arthroscopically through a posterolateral portal (Fig. 25I-18). For excision of posterior bony impingement, we usually prefer an open, pos­ terolateral approach. The interval between the Achilles ten­ don and peroneal tendons is developed, with care to protect the sural nerve. The os trigonum is removed with subperio­ steal dissection. The lateral tubercle of the posterior process

of the talus (Stieda process) is removed with a curved oste­ otome or rongeur to ensure that the posterior talar surface is flush with the posterior tibial surface. If decompression of the flexor hallucis longus is indicated, a posteromedial approach can be developed along the tarsal tunnel.

Figure 25I-18   Posterior portal site for ankle arthroscopy. (From Phillips BB: Arthroscopy of the lower extremity. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Foot and Ankle 2161

Figure 25I-19   CT shows medial facet tarsal coalition of talus and calcaneus.

Postoperative Prescription, Potential Complications, and Criteria for Return to Play After ankle arthroscopy, weight-bearing to tolerance, with or without crutches, usually is allowed the first week after sur­ gery. Then a progressive program of strengthening, range of motion, and functional agility is begun. Generally, return to competitive sports is possible by about 6 weeks after surgery when the patient is pain free and range of motion and strength are comparable to the unaffected ankle. The most frequent complication of ankle arthroscopy is temporary numbness or tenderness at a portal site caused by local nerve dam­ age. Osteophytes may recur, especially in athletes who have recurrent supination trauma or repeated forceful dorsiflexion of the ankle, but these are not always symptomatic.120

Tarsal Coalition Particularly in children and adolescents, there is an asso­ ciation between frequent ankle sprains and tarsal coalitions, including fibrous (syndesmosis), cartilaginous (synchondro­ sis), and bony (synostosis) coalitions. Although the ­frequency of tarsal coalition has long been estimated at 1%, more recent information indicates a frequency of 11%, possibly because of

A

Figure 25I-21   Pigmented villonodular synovitis eroding into superior portion of the talar neck caused symptoms of ankle sprain in young athlete.

more frequent identification by MRI.124 Patients may com­ plain of mild deep pain and limited range of motion, usually after a traumatic ankle sprain that “just never seems to get better.” Often symptoms are relieved by rest and aggravated by prolonged or heavy activity, suggestive of ankle ligament injury.125 Radiographs or CT scans can confirm the ­diagnosis (Fig. 25I-19). If nonoperative measures, such as bracing or casting and NSAIDs, fail to relieve symptoms, open or arthroscopic excision of the coalition is indicated.126 After coalition excision, a removable boot usually is worn for about 6 weeks, and strengthening and flexibility exercises are done.

Neoplasms Neoplasms such as osteoid osteoma, eosinophilic granu­ loma (Fig. 25I-20), and pigmented villonodular synovitis (Fig. 25I-21) can mimic ankle sprain. Neoplasms of the ankle can mimic a number of ankle conditions, ­including chronic sprains, anterior impingement, stress fracture, osteomyelitis, and osteonecrosis. Appropriate treatment of these lesions often is delayed because of misdiagnosis; one

B

Figure 25I-20   Eosinophilic granuloma may cause symptoms of ankle sprain. A, Lesion in head and neck of talus. B, After excision and bone grafting.

2162 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

study reported a 2.5-year delay in the diagnosis of osteoid osteomas in five patients,127 and another reported a delay of more than 3 years in one patient.128 A typical feature of osteoid osteoma is pain that worsens at night and is relieved by aspirin, but this may not be present in all patients with osteoid osteomas. Treatment of osteoid osteomas usually is en bloc excision or curettage of the nidus. Successful arthroscopic treatment of osteoid osteomas of the ankle has been described,128-131 which allows an earlier return to activ­ ity (2 to 3 months after surgery) than open procedures.

OTHER APOPHYSITIS, OSTEOCHONDRITIS, AND DEVELOPMENTAL ANOMALIES OF THE FOOT THAT CAN CAUSE DISABILITY IN ATHLETES Calcaneal Apophysitis (Sever Disease) First described by J. W. Sever132 in 1912, calcaneal apophy­ sitis (Sever disease) is a common cause of heel pain, espe­ cially during running, that most often occurs in children who are engaged in activities such as basketball and soccer. The calcaneal apophysis appears as an independent center of ossification in boys aged 9 to10 years and ossifies around the age of 17 years; in girls, this occurs earlier. The average age at presentation is 11 years in boys and 8 years in girls. Sever disease is 2 to 3 times more common in boys, and 60% of patients have bilateral involvement. Aptly described by Sever as an “inflammation of the calcaneal apophysis result­ ing in the clinical symptoms of pain at the posterior heel, mild swelling, and difficulty with walking,” the condition often causes a child to walk with an antalgic gait. The heel pad and posterior aspect of the heel are tender, but swelling usually is minimal. Pain is located at the most distal portion of the heel pad and along the posterior aspect of the heel up to the most distal portion of the Achilles tendon. The heel cord itself is not especially tight. Ogden and associates133 suggested that this entity should be called Sever injury rather than Sever disease because MRI evidence indicated that the true pathogenesis is a stress microfracture related to chronic repetitive microtrauma. The stress microfractures were found in the metaphysis of the body of the calcaneus adjacent to, but not directly involving, the apophysis. Radiographs are not essential to the diagnosis but can rule out other conditions, such as fracture, infection, or neoplasm. They may show a sclerotic and fragmented calcaneal apophysitis (Fig. 25I-22), but the radiographic appearance usually is normal. The necessity of curtailing the activity that incites pain is controversial. Because they consider this condition a chronic, repetitive injury to metaphyseal bone, some have suggested that temporary discontinuation of the activity is essential. Most patients are able to return to normal sports activity after 2 months of rest; in only the most severe cases is immobili­ zation necessary.134,135 Conversely, Weiner and associates136 reviewed the records of 227 patients with Sever disease and concluded that activity restriction is unnecessary; they recom­ mended an in-shoe orthosis, no limits on physical activity, with 4 to 6 weeks of casting for persistent symptoms.

Figure 25I-22   Increased sclerosis and fragmentation of calcaneal apophysis (Sever disease).

Osteochondrosis of the Tarsal Navicular (Köhler Disease) First described by Köhler137 in 1908, osteochondrosis of the tarsal navicular is a relatively uncommon cause of pain and limp in children. Like Sever disease, boys are more commonly affected than girls; however, the age of onset is younger, 5 years in boys and 4 years in girls. Thirty-three percent of patients have bilateral involvement. Tenderness in the area of the tarsal navicular is the most consistent phys­ ical finding. The radiographic findings are variable degrees of sclerosis, flattening, and irregular rarefaction of the navic­ ular (Alka-Seltzer-on-end appearance) (Fig. 25I-23). Loss of the trabecular pattern and fragmentation also are ­common. Other bones of the foot usually are normal. Köhler dis­ ease is self-limiting, and the final outcome is not affected by treatment. Short-leg walking cast immobilization for 6 to 8 weeks has been reported to result in faster resolution of symptoms than conservative treatment.138

Accessory Navicular The accessory navicular is a separate ossification center for the tuberosity of the navicular that is present in about 14% of all feet, but most of these are asymptomatic and even­ tually fuse with the navicular.139,140 In some individuals, however, the accessory navicular can be a source of medial foot pain. The accessory navicular may exist as a separate ossicle within the posterior tibial tendon (type I), may form a synchondrosis with the navicular (type II), or may fuse with the navicular and create a cornuate navicular (type III) (Fig. 25I-24). Often the posterior tibial tendon inserts onto the accessory navicular instead of onto the tuberosity of the navicular, which has been suggested to alter the pull of the tendon and cause a flatfoot deformity. Symptoms are most common when the accessory navicular forms a synchon­ drosis with the navicular (type II). Symptoms usually begin

Foot and Ankle 2163

B

Figure 25I-23   A and B, Increased sclerosis and narrowing   (Alka-Seltzer-on-end appearance) of the navicular (Köhler disease) in a young athlete with complaints of pain in the medial midfoot area.

A

A

D

B

C

Figure 25I-24   Accessory navicular. A and B, Type I. C and D, Type II. (Redrawn from MacNicol MJ, Voutsinas S: Surgical treatment of the symptomatic accessory navicular. J Bone Joint Surg Br 66:218-226, 1984.)

2164 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

B

A

C

D

Figure 25I-25   A and B, Bilateral accessory navicular. Note “opening up” of metatarsal cuboid-cuneiform articulation, suggesting flattening of the longitudinal arch. C and D, Note sag at talonavicular joint in left foot (C) compared with right foot (D). (From Murphy GA: Pes planus. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

in adolescence and are made worse by activity and weightbearing. In adults, the onset of pain usually is acute after an eversion injury or other foot trauma. Pain is localized over the prominence of the accessory navicular and at the medial arch of the foot. A painful bursa may be present over the navicular prominence. Standard anteroposterior and lat­ eral views of the foot (Fig. 25I-25), along with a 45-degree eversion oblique view, usually are sufficient for diagnosis, but bone scanning and MRI can be helpful if the diag­ nosis is in question.141 Initial treatment of a symptomatic accessory navicular is nonoperative and directed at reliev­ ing pressure at the painful medial prominence and reduc­ ing inflammation. The classic operative treatment for an accessory navicular that does not respond to nonoperative measures is removal of the ossicle (Kidner procedure142), which generally gives good results.143,144 The benefits of

t­ ransposition of the posterior tibial tendon to the plantar aspect of the navicular are questionable.140,145-147 Percuta­ neous drilling of the accessory navicular to induce or accel­ erate fusion was reported to obtain excellent or good results in 30 of 31 feet.148 Percutaneous placement of one or two 3.5-mm or 4.0-mm cannulated screws across the synchon­ drosis has been recommended in athletes 14 to 30 years of age who have symptoms consistent with synchondrotic stress fracture.149 Malicky and associates150 described open fusion between the ossicle and navicular bone: after excision of the synchondrosis and the adjacent subchondral bone, one or two 2.7-mm or 3.5-mm lag screws are inserted to “arthodese” the bones together. They recommended this procedure in adults with large, painful accessory navicular bones because it leaves the posterior tibial tendon attach­ ment intact and provides a more reliable healing surface.

Authors’ Preferred Method Initial treatment is nonoperative and can include physi­ cal therapy aimed at controlling coexisting tendinitis and strengthening the posterior tibial tendon, molded orthoses, shoe modifications, and cast immobilization. If symptoms

persist, we prefer simple excision of the accessory navicu­ lar. It is important to excise enough of the prominence to create a surface that is flush with the medial border of the midfoot.

Foot and Ankle 2165

Authors’ Preferred Method—cont’d

A

C

B Figure 25I-26   Excision of accessory navicular. A, Incision. B, Exposure of the posterior tibial tendon. C, Removal of accessory navicular. (From Murphy GA: Pes planus. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Technique of Excision of Accessory ­ avicular N

• Beginning 1 to 1.5 cm inferior and distal to the tip of the medial malleolus, curve the skin incision slightly dorsal­ ward, peaking at the medial prominence of the accessory navicular and sloping distally to the base of the first meta­ tarsal (Fig. 25I-26A). • Ligate the plantar communicating branches of the saphe­ nous system and identify the posterior tibial tendon as it approaches the accessory navicular (see Fig. 25I-26B). • Identify the dorsal and plantar margins of the tendon 2 cm proximal to the accessory navicular, and expose

the ­ tendon distally, ending at the bone. This exposes the ­entire tendon without disturbing the part extending plantarward toward its multiple insertions. • Use sharp dissection to shell the accessory navicular from the posterior tibial tendon (see Fig. 25I-26C). Return to Sport

After operative treatment, a short-leg cast is worn for 6 weeks. Range of motion exercises are begun after cast removal, and return to competition usually can be accomplished within 8 to 10 weeks.

2166 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Osteochondrosis of the Cuneiform Osteochondrosis of the cuneiform (Buschke disease) is rare, with few reports in the literature. Repetitive trauma with insult to the developing osseous tissue is believed to be responsible for this osteochondrosis, especially in the pronated foot in which continuous pressure is placed on the medial side of the foot and the medial cuneiform. Of the 19 patients described in the English literature, 18 were boys; 12 were 5 or 6 years old, and the youngest was 2 years old.151,152 Pain and limping usually are pres­ ent, but the lesion can be asymptomatic. Pain occasionally can be elicited by gentle pressure over the medial cunei­ form, and mild swelling rarely is present. Radiographs usu­ ally show a sclerotic, fragmented medial cuneiform (Fig. 25I-27). Osteochondrosis of the medial cuneiform has been associated with other tarsal bone lesions in most patients and is almost always bilateral. Treatment is nonoperative and includes control of the foot pronation and reduction of pain with rest and medial arch supports.

Osteochondrosis of the Metatarsal Head (Freiberg Disease) First described by Freiberg153 in 1914, osteochondrosis or osteonecrosis of the metatarsal head is most common in adolescent athletes who perform on their toes in either sprinting or jumping activities. It is more common in girls than in boys, making it the only osteochondrosis with a female predilection. The second, followed by the third, metatarsal head is most commonly involved. Fewer than 10% of patients have bilateral involvement. The patho­ genesis is unknown, but Gauthier and Elbaz154 found that patients with longer second metatarsals and excessive plantar pressure under the second metatarsal head had no significantly higher risk. These findings would appear to call into question mechanical stress being the sole or even the primary cause of Freiberg disease. The primary complaint often is a vague forefoot pain that is worsened by activity and weight-bearing and relieved with rest. The

Figure 25I-27   Increased sclerosis and fragmentation of the first cuneiform (Buschke disease) in a child who complained of pain in the medial midfoot area.

pain usually is worse at extremes of motion, with pain under and around the involved metatarsophalangeal joint. Palpation may identify swelling and slightly increased temperature. The initial process of pain and synovitis is followed by radiographic findings of sclerosis, resorption of the subchondral plate, fracture, collapse, and fragmen­ tation. Secondary degenerative changes and remodeling then occur in the flattened metatarsal head (Fig. 25I-28). Several staging systems have been developed to corre­ late physical and radiographic findings with treatment (Fig. 25I-29 and Table 25I-9), but their reliability is still unproved.

Treatment Options Initial treatment is nonoperative, including metatarsal relief pads, restriction of running and jumping activities, and occasionally a short-leg walking cast worn for 6 to 12 weeks until acute symptoms are resolved. A number of operative procedures have been described for persistent symptoms, including débridement, synovectomy, drilling, osteotomy, interpositional arthroplasty, and joint replacement. Dor­ sal wedge or dorsiflexion osteotomy (Fig. 25I-30) has been used successfully for all stages of the disease,154-157 although the range of motion of the MTP joint is decreased. More recent innovations include the use of absorbable pins for fixation of the osteotomy, which obviates the need for a second operation for implant removal; arthroscopic tech­ niques for synovectomy and drilling158,159; and osteochon­ dral plug transplantation.158,160 Carro and coworkers160 recommended an age-based approach to the treatment of Freiberg disease, beginning with arthroscopic synovectomy and débridement, followed by open or arthroscopic osteo­ chondral transplantation in late adolescence and adulthood, and an arthroscopic Keller procedure (with or without interpositional arthroplasty) for more severe involvement in late adulthood. However, in our experience, resection of the base of the proximal phalanx or the metatarsal head should be avoided to avoid transfer metatarsalgia or hallux valgus caused by instability of the second digit.

Foot and Ankle 2167

A

B

A

B

C

D

C Figure 25I-28   Freiberg infarction. A and B, Note flattening of second metatarsal head. C, Note marrow edema of second metatarsal.

Authors’ Preferred Method Initial treatment is nonoperative, including restriction of sports activities and short-term cast-with-toe-plate ­immobilization. After acute symptoms resolve, metatarsal pads inserted proximal to the MTP joint are used during running and sports. Operative treatment is indicated for persistent symptoms. For adolescents and young adults, joint débridement and removal of loose bodies usually are sufficient, and return to sports generally is possible in 6 to 8 weeks. More extensive surgery (dorsiflexion or metatar­ sal shortening osteotomy) is reserved for older patients with late-stage involvement.

E Figure 25I-29   Levels of progression of Freiberg disease.   A, Early fracture of subchondral epiphysis. B, Early collapse of dorsal central portion of metatarsal with flattening of articular surface. C, Further flattening of metatarsal head with continued collapse of central portion of articular surface with medial and lateral projections; plantar articular cartilage remains intact. D, Loose bodies form from fractures of lateral projections and separation of central articular fragment. E, End-stage degenerative arthrosis with marked flattening of the metatarsal head and joint space narrowing. (Redrawn from Katcherian DA: Treatment of Freiberg’s disease. Orthop Clin North Am 25:69-81, 1994.)

Iselin Disease Iselin disease is a traction apophysitis of the base of the fifth metatarsal that occurs in late childhood or early adolescence at the time of the appearance of the proximal apophysis of the tuberosity of the fifth metatarsal.161 This secondary ossification center is a small, shell-shaped fleck of bone ori­ ented slightly obliquely with respect to the metatarsal shaft and located on the lateral plantar aspect of the tuberosity (Fig. 25I-31). It usually is not visible on anteroposterior or lateral radiographs but can be seen on an oblique view. Symptoms include tenderness over a prominent proximal fifth metatarsal and pain over the lateral aspect of the foot with weight-bearing. ­ Participation in sports that require

2168 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25I-9  Staging and Classification Systems for Freiberg Disease Katcherian

Smillie

Gauthier/Elbaz

Thompson/Hamilton

Description

Radiographs

Level A

Stage I

Stage 0 + 1

Type 1

Level B

Stage II

Stage 2

Type 2

Radiographs normal; bone scan may detect Radiographs may show slight widening of joint, sclerosis of epiphysis, collapse and flattening of articular surface

Level C

Stage III

Stage 2

Type 2

Earliest form; fissure in epiphysis Progression of subchondral fracture with bone resorption; collapse of dorsal central portion of metatarsal head and alteration of articular surface Further deformation and collapse of central portion of head

Level D

Stage IV

Stage 3

Type 3

Level E

Stage V

Stage 4

Type 3

Fracture and separation of central portion and peripheral projections of involved metatarsal head resulting in loose bodies Advanced arthrosis secondary to progressive flattening and deformity of metatarsal head

Progressive flattening of metatarsal head with osteolysis and collapse; zone of rarefaction around sclerotic bone as healing and revascularization take place; premature physeal closure may be present Fragmentation of epiphysis, early joint narrowing, multiple loose bodies

Joint space narrowing, hypertrophy of metatarsal head, irregularity of base of proximal phalanx with osteophyte formation

MT, metatarsal. Modified from Katcherian DA: Treatment of Freiberg’s disease. Orthop Clin North Am 25:69-81, 1994.

r­ unning, jumping, and cutting, which cause inversion stresses on the forefoot, is a common factor.162,163 The affected area over the tuberosity is larger than that on the noninvolved side, with soft tissue edema and local erythema. The area is tender to palpation at the insertion of the peroneus brevis, and resisted eversion and extreme plantar flexion and dor­ siflexion of the foot elicit pain. Oblique radiographs show enlargement of the apophysis and often ­ fragmentation of

this secondary ossification center (Fig. 25I-32) and widen­ ing of the cartilaginous-osseous junction. Technetium-99m bone scanning shows increased uptake over the apophysis. Failure of the apophysis to fuse with the metatarsal (Fig. 25I-33) can cause symptoms into adulthood. Os vesalia­ num, a sesamoid in the peroneus brevis (Fig. 25I-34), must be distinguished from Iselin disease, and the nonunited apophysis should not be mistaken for a fracture.

Figure 25I-30   Dorsal closing wedge osteotomy and crosspinning for Freiberg infarction. (Redrawn from Chao KH, Lee CH, Lin LC: Surgery for symptomatic Freiberg’s disease: Extraarticular dorsal closing wedge osteotomy in 13 patients followed for 2 to 4 years. Acta Orthop Scand 70:483-486, 1999.)

Figure 25I-31   Fragmentation of secondary ossification center (Iselin disease) in a young athlete who complained of pain and swelling over the base of the fifth metatarsal.

Foot and Ankle 2169

Os vesalianum

Iselin disease

Figure 25I-32   Enlargement and fragment of epiphysis in Iselin disease. (From Canale ST: Osteochondrosis or epiphysitis and other miscellaneous affections. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Authors’ Preferred Method Treatment of Iselin disease is always nonoperative and in­ cludes limitation of activity, ice, and NSAIDs. If symptoms persist or are severe, a brief period of cast immobilization can be helpful. After resolution of acute symptoms, stretch­ ing exercises and range of motion exercises for the ankle and subtalar joints on a wobble board are initiated. Return to competition is allowed when discomfort is tolerable after activity. Plantar arch strapping or an arch support system placed in the shoe to elevate and relieve stress on the fifth metatarsal may allow athletic activity without pain.

Figure 25I-33   Nonunion of fifth metatarsal as a result of Iselin disease. (From Canale ST: Osteochondrosis or epiphysitis and other miscellaneous affections. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Figure 25I-34   Os vesalianum must be distinguished from Iselin disease. (From Canale ST: Osteochondrosis or epiphysitis and other miscellaneous affections. In Canale ST, Beaty JH [eds]: Campbell’s Operative Orthopaedics, 11th ed. Philadelphia, Elsevier, 2008.)

Osteochondritis of the Sesamoids Although small, seemingly insignificant bones of the foot, the sesamoids can cause disabling pain in an athlete. Galen, in 180 ad, named these small bones the sesamoids because of their resemblance to flat, oval sesame seeds. During the seventh or eighth week of embryonic develop­ ment, both sesamoid bones of the hallux appear as islands of undifferentiated connective tissue between the first metatarsal heads. The fibular (lateral) sesamoid is present slightly sooner than the tibial (medial) sesamoid. Ossifica­ tion begins at about 8 or 9 years of age in girls and 10 or 11 years of age in boys. There may be two or more centers of ossification, with the tibial sesamoid being bipartite at skeletal maturity in about 10% to 15% of the population; the fibular sesamoid rarely is bipartite.164-166 The two hal­ lucal sesamoids are embedded in the tendons of the short flexor of the hallux and are held together by the intersesa­ moid ligament and the plantar plate, which inserts on the base of the proximal phalanx of the hallux (Fig. 25I-35). The sesamoids function to absorb weight-bearing pressure, reduce friction, and protect tendons. They are important to the dynamic function of the hallux and act as a fulcrum to increase the mechanical force of the flexor hallucis bre­ vis tendon. The sesamoid complex normally transmits as much as 50% of body weight and during push-off may transmit loads of more than 300% of body weight.167 Osteochondritis of the hallucal sesamoids is most com­ mon in young women but has been reported in nearly all age and gender groups from adolescence through adult­ hood.164 Osteochondritis may occur as a primary patho­ logic entity or may be a late stage of repetitive stress injury involving osteonecrosis (Fig. 25I-36). Trauma is believed

2170 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Flexor brevis Abductor Anterior sesamoidal ligament

Medial sesamoid

Flexor hallucis longus

Figure 25I-35   Anatomy of the sesamoids. (Redrawn from Leventen EO: Sesamoid disorders and treatment: An update. Clin Orthop 269:236-240, 1991.)

to the most frequent cause, but osteonecrosis with sub­ sequent regeneration and excessive calcification may be present. The diagnosis of sesamoid injury constitutes a spectrum of abnormality rather than an isolated injury.165 Depending on the stage at which the patient is first seen, the diagnosis may range from bursitis over the tibial sesa­ moid to the vague entity sesamoiditis, which encompasses both osteochondrosis and osteochondritis. In later stages, the diagnosis may include stress fracture, traumatic or degenerative arthritis, and chondromalacia. Hematoge­ nous osteomyelitis of the sesamoid also should be included in the differential diagnosis, and sesamoid periostitis may occur in athletes with one of the rheumatoid variants, such as psoriasis, Reiter syndrome, and ankylosing spondylitis. Typically, patients have pain and tenderness to palpation over the involved sesamoid, without swelling or erythema that would indicate infection or bursitis. An axial radio­ graph or CT may show an enlarged or deformed sesamoid with irregular areas of increased bone density, mottling, and fragmentation.168 Comparison views of the opposite foot are helpful to distinguish a bipartite sesamoid from a fracture.

A

Authors’ Preferred Method Initial treatment is nonoperative, including NSAIDs, ­activity modification, and full-length shoe orthoses with a metatarsal pad and a relief beneath the first metatarsal pad. If no relief is obtained, a period of cast immobilization is tried. Persistent symptoms are an indication for sesamoid excision. Only the symptomatic sesamoid should be excised because of the likelihood of creating an intrinsic minus cock-up (hallux extensus) deformity with excision of both sesamoids. Even excision of a single sesamoid may result in hallux varus deformity, exacerbation of hallux valgus, or overload of the metatarsal head. Removal of both sesa­ moids results in about a 30% decrease in flexion strength of the hallux.169,170 Return to athletic activity usually is possible at 6 weeks after excision of one of the sesamoids; competitive sports are not allowed for 12 weeks after exci­ sion of both sesamoids.

B

Figure 25I-36   A, Osteochondrosis of the tibial (medial) sesamoid in a runner. B, Note fragmentation of sesamoid.

Foot and Ankle 2171

C

r i t i c a l

P

S U G G E S T E D

o i n t s

l Important differences exist related to the mechanism of injury for medial and lateral talar dome lesions, and these must be recognized and considered in choosing treatment methods. l Most OLT, including posterior ones, can be treated arthroscopically, but accessory portals often are required. l The shorter the interval between the diagnosis of MRI stage II and IV OLT and operative treatment, the better the results. l OLT and fractures of the talar and calcaneal processes can be mistaken for ankle sprains and should be suspected when there is an inability to bear weight on the foot and ankle, severe swelling, or continued pain 3 to 4 weeks after “ankle sprain.” l Small, minimally displaced fractures of the talar and cal­ caneal processes usually can be successfully treated non­ operatively, but larger fragments (>1 cm) or displaced fragments (>2 mm) generally require open reduction and internal fixation. l Ankle impingement syndromes usually occur after an inver­ sion ankle injury or repetitive ankle flexion to extremes (e.g., sprinters, ballet dancers) and can be caused by bony lesions (i.e., osteophytes, os trigonum, Stieda process) or soft-tissue lesions (i.e., thickened ATFL [Basset ligament], deltoid ligament injury, tenosynovitis, meniscoid lesion). l Pain with palpation and dorsiflexion suggests anterior impingement (most common), whereas pain with forced plantar flexion is indicative of posterior impingement. l Initial treatment of ankle impingement is nonoperative; if symptoms persist, arthroscopic treatment (débridement, synovectomy, excision, decompression) is successful in 90% to 95% of patients. l Sever, Köhler, Buschke, Freiberg, and Iselin diseases; accessory navicular; and osteochondritis of the sesamoids are all relatively rare conditions that may be confused with traumatic injury; initial treatment is always nonoperative, which is successful in most patients.

R E A D I N G S

Baums MH, Heidrich G, Schultz W, et al: Autologous chondrocyte transplanta­ tion for treating cartilage defects of the talus. J Bone Joint Surg Am 88:303-308, 2006. Dedmond BT, Cory JW, McBryde A Jr: The hallucal sesamoid complex. J Am Acad Orthop Surg 14:745-753, 2006. Gobbi A, Francisco RA, Lubowitz JH, et al: Osteochondral lesions of the talus: Randomized controlled trial comparing chondroplasty, microfracture, and osteo­ chondral autograft transplantation. Arthroscopy 22:1085-1092, 2006. Hassan AH: Treatment of anterolateral impingements of the ankle joint by arthros­ copy. Knee Surg Sports Traumatol Arthrosc 15:1150-1154, 2007. Kopp FJ, Marcus RE: Clinical outcome of surgical treatment of the symptomatic accessory navicular. Foot Ankle Int 25:27-30, 2004. Kreuz PC, Steinwachs M, Erggelet C, et al: Mosaicplasty with autogenous talar autograft for osteochondral lesions of the talus after failed primary arthroscopic management: A prospective study with a 4-year follow-up. Am J Sports Med 34:55-63, 2006. Perumal V, Wall E, Babekir N: Juvenile osteochondritis of the talus. J Pediatr ­Orthop 27:821-825, 2007. Verhagen RA, Struijs PA, Bossuvt PM, van Dijk CN: Systematic review of treat­ ment strategies for osteochondral defects of the talar dome. Foot Ankle Clin 8:233-242, 2003. Watson AD: Ankle instability and impingement. Foot Ankle Clin 12:177-195, 2007.

R eferences Please see www.expertconsult.com

S e c t i o n

J

Etiology of Injury to the Foot and Ankle Christopher B. Hirose, Thomas O. Clanton, and Robert M. Wood The etiology of injury to the foot and ankle is of utmost importance in sports medicine. The capacity to recognize the injury-causing conditions and the ability to control them may help prevent these same injuries. Research has shown that when etiologic factors are addressed, it is pos­ sible to decrease injury rates significantly.1-3 This chapter reviews predisposing conditions for injury of the foot and ankle and the variables that may have a ­protective value. The primary areas of importance are

(1) joint flexibility, (2) shoewear, and (3) the quality of the playing surface. Injury variables can be divided into intrinsic and extrin­ sic factors.4-6 Intrinsic factors include both an athlete’s indi­ vidual physical and personality characteristics. The physical characteristics include age, sex, genotype, somatotype, strength, speed, agility, coordination, fitness, ­ flexibility, malalignment, muscle composition, previous injury, and residual structural inadequacies.4,5,7-35

2172 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Personality characteristics such as extroversion, anxiety level, conscientiousness, self-confidence, determination, responsiveness to coaching, discipline, dependency, sensi­ tivity, and others may play a role in whether an individual is injury prone.4,17,26,36-41 The contribution of these variables constitutes the individual athlete’s intrinsic risk for injury. Extrinsic factors include the kind of sports activity, the training methods, the knowledge and skill of the coach, the level of competition, the environmental conditions, the type of shoewear used, the playing surfaces, and the equip­ ment.4,5,8-10,18,22,26-28,32,42-58 There is, and always has been, an inherent risk for injury associated with athletic endeavors. The kind of athletic activity can further be separated into high, medium, or low risk for foot and ankle injuries (Box 25J-1).59-61 In addition, each sport poses a different risk to a particular body part. For example, running carries a low risk for shoulder injuries but a high risk for foot or knee injuries. Conversely, swim­ ming may pose a high risk for shoulder injuries but a low risk for foot and ankle injuries. Certain other variables play a role in the risk for injury, including playing time, level of competition, the movements required, equipment used, individual versus team play, the level of fitness, and the enforcement of the rules and regulations.60,62,63 Collins and colleagues examined athletes at 100 U.S. high schools and found that nationally 6.4% of all high school sports–related injuries were related to breaking of the rules of the sport.64 Practice methods also have a definite impact on injury statistics.10,45,65,66 In competitive running, training tech­ niques such as high-mileage workouts, excessive use of plyometrics, and interval work have been implicated in the incidence of overuse problems.6,67-69 Certain factors fre­ quently lead to overstress: (1) when unprepared athletes are asked to cross-train with long-distance running for such sports as swimming, basketball, or volleyball; (2) when ath­ letes are out of shape because it is early in the season; and (3) when it is the transition period between seasons when a player shifts from one sport to another without recognizing the different stresses introduced by the variation in perfor­ mance requirements, shoes, and playing surfaces involved. Running programs that involve excessive mileage or, more commonly, “too much too soon” are responsible for stress fractures or other overuse syndromes.8,28 Taunton and coworkers found that being active for less than 8.5 years was positively associated with injury in both sexes for tibial stress syndrome; and women with a body mass index of less than 21 kg/m2 were at a significantly higher risk for tibial stress fractures.70 A lack of stress adaptation frequently results in overload problems and injury. Other training techniques may increase an athlete’s risk for injury. Plyometric training is promoted to improve jumping ability and speed71; however, such exercises as bench jumping and bounding can overload the system.72 This is particularly true when stress accommodation is not allowed and when shoes and surfaces are inappropriate for these activities. Stress tolerance varies widely from individual to indi­ vidual. This must be recognized in the design of preseason, in-season, and off-season training programs. Unnecessary exposure to injury can be attributed to coaching meth­ ods. Uncontrolled scrimmages and excessive contact work increase injury rates in football.10 Gymnasts who perform complex maneuvers without adequate spotting

Box 25J-1 High-, Medium-, and Low-Risk Sports for Foot and Ankle Injury High Risk

• Ballet • Basketball • Dance • Ice skating • Mountaineering • Running • Skateboarding • Snowboarding • Soccer Medium Risk

• Aerobics • Baseball • Football • Gymnastics • Ice hockey • Lacrosse • Racquetball, squash • Roller skating • Rugby • Tennis • Volleyball • Water-skiing Low Risk

• Archery • Boating • Bowing • Cycling • Equestrian • Fishing • Golf • Parachuting • Rodeo • Skiing • Weight training • Wrestling Adapted from Sports Injuries. Accident Facts. Report of the National Safety Council, 1990, p 88; and from Table 26-1 in Clanton TO: Athletic injuries to the soft tissues of the foot and ankle. In Coughlin MJ, Mann RA (eds): Surgery of the Foot and Ankle, 7th ed. St. Louis, Mosby, 1999.

or ­ preparation have a higher risk for injury.22 Baseball and softball players who practice sliding into fixed bases are unnecessarily exposed to foot and ankle trauma (Fig. 25J-1).47,48 Finally, the “no pain, no gain” coaching philos­ ophy can pressure an athlete to return to play after an injury without adequate rehabilitation and increase the potential for injury: an ankle syndesmosis injury treated functionally as a typical lateral ankle sprain can worsen considerably. In these times of astronomical salaries for professional ath­ letes, there is increasing emphasis on players returning to sport after injury in the shortest amount of time possible. The environment may affect injury risk: weather con­ ditions can alter the playing surface, the shoes worn, the interface characteristics between shoe and surface, and the attitudes of players and coaches. The complex interplay

Foot and Ankle 2173

A

C

B

Figure 25J-1   A, Potential for injury created by sliding into a fixed base. B, A 15-year-old boy who injured his ankle sliding into second base. C, Distal fibula fracture of 15-year-old male baseball player.   (A, Photograph courtesy of Rice University Sports Information Office; B, photograph by Christopher B. Hirose.)

between extrinsic and intrinsic elements makes statistical analysis difficult from a precise epidemiologic perspective. As scientists, we must recognize that injury statistics are only as good as the integrity of the investigator, the statisti­ cal methods, and the research design.73-76 The use of performance-enhancing drugs and nutri­ tional supplements has been rearing its ugly head for more than a century.77,78 In 1886, Arthur Linton, a 24-year-old Welsh cyclist, died during a race from Bordeaux to Paris. He was believed to have taken the stimulant trimethyl. Shortly thereafter, Charles Edouard Brown-Séquard, in

1889, extracted testosterone in dogs and injected himself with the extract, claiming that it made him feel younger. From 1940 to 1945, the Germans tested anabolic steroids in prisoners, and Hitler himself may have used them. Some believe his behavior late in life exhibited the characteris­ tics of steroid use: aggressiveness and violent behavior with bouts of depression and suicidal ideation. In 1954, the Soviet weightlifting team dominated the sport, and their team doctors revealed the use of injected testosterone. At the 1976 Montreal Olympics, the East German women’s swim team dominated with 11 of 13 individual gold ­medals,

2174 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

setting 8 world records in the process. Much later, the German team coaches admitted to steroid use. Based on the lawsuits filed in German court, it is believed that up to 2000 of the East German athletes who have used anabolic steroids are now suffering from serious health problems, including liver tumors, heart disease, testicular and breast cancer, infertility, eating disorders, depression, and birth defects. The list of world-famous athletes who used, or are accused of using, performance enhancing drugs continues to grow: Canadian sprinter Ben Johnson, NFL defensive end Lyle Alzado, baseball’s 1996 National League MVP winner Ken Caminiti (who died of a heart attack at age 41 years), Major Leaguers Barry Bonds and Jose Canseco, Tour de France Winner Floyd Landis, Sydney Olym­ pic Champion Marion Jones—who had her five medals stripped by the International Olympic committee, and most recently Roger “The Rocket” Clemens.78 Unfortu­ nately, with the desire to win, sometimes at all costs, this list of famous athletes may continue to grow. Quite clearly, there are negative effects associated with the use of per­ formance-enhancing drugs. They are illegal, constitute an unfair advantage, and place the athlete at an unnecessary and serious risk.77,79,80 Injury risk is a natural and accepted part of sports par­ ticipation. Different sports with different performance factors naturally provide varying degrees of exposure to injury.26,60 The broad categories of analysis in the injury equation are intrinsic and extrinsic factors. For the foot and ankle, the primary areas of importance are joint flexibility, shoewear, and the quality of the playing surface. These can be related to a classification of sports activities based on six basic motions: (1) stance, (2) walking, (3) running, (4) jumping, (5) throwing, and (6) kicking.22,81 Running, jumping, and kicking seemingly pose the greatest risk for injury to the foot and ankle. By using these types of ­categorization and classification, combined with scien­ tific ­analysis of biomechanical stresses, it may be possible to analyze injury potential more accurately and to design effective prevention strategies.

INCIDENCE Injuries to the lower extremities account for most sports injuries.82 Between 55% and 90% of all sports injuries occur from the hip region down to the toe.83 Running, jump­ ing, and kicking sports are associated with injuries to the foot and ankle. When combined with cutting and ­sliding maneuvers, the injury statistics increase dramatically. It has been suggested that the sprained ankle is the single most common injury in sports, and many ankle sprains go ­unreported and untreated (Fig. 25J-2).84-89 Nevertheless, epidemiologic analysis of sports injuries occurring around the world indicates an overwhelming preponderance of injuries to the lower extremities, and most of these are sprains, strains, and contusions.46,84,86,87,89-96 Studies of sports-related injuries in the running and jumping sports have suggested an incidence of injury of 10% to 15% for the ankle and 3% to 15% for the foot.86,90,92,93,96 In their study of more than 12,000 injuries occurring in 19 different sports in a university sports clinic, Garrick and Requa found that 25% of these injuries involved the foot and ankle.86 Patients participating in running, tennis, and

Figure 25J-2   Ankle ligamentous examination at the University of Texas. (Photograph by Thomas O. Clanton.)

dance were seen most frequently in this clinic setting. Kan­ nus and colleagues reported on the injuries seen in their European sports medicine clinic, where the most common sports were soccer, running, and orienteering.92 Ankle injuries accounted for 9% of the visits to the clinic during a 6-month period. Achilles tendon problems and heel pain accounted for another 7%. DeHaven and Lintner, from the University of Rochester Section of Sports Medicine, sur­ veyed injuries seen over a 7-year period, including patients from unorganized sports all the way through the profes­ sional level.90 Of the 3431 cases studied, 12% involved the ankle, and 3% involved the foot. The three sports most commonly producing injury in this clinical setting were football, basketball, and soccer. In 1980, Zaricznyj and associates reported a ­ thorough analysis of the causes and severity of sports injuries in a population of school chil­ dren from elementary through high school.96 This study documented all injuries in a population of 25,512 school children. Of these, 11.4% occurred to the ankle. The foot and toes were injured in 6%. A thorough analysis of the incidence of injury in ­different sports reveals the magnitude of the problem. Soccer claims anywhere from 40 to 120 million participants worldwide and about 13 to 16 million in the United States.97,98 It has an injury rate of 22% to the ankle and 8% to the foot.82 Football involves an estimated 17,400,000 participants in the United States.98 The overall likelihood of player injury in football has been estimated to be anywhere from 10% to 80% of participants. Combining these numbers with the estimated injury rate to the foot and ankle of 15% to 25%, the total injuries to the foot and ankle in football in this country range from a low of about 261,000 to a high of 3,480,000. 9,99-103 The estimated injury rates for the foot and ankle in other sports are listed in Table 25J-1. When combined with esti­ mates of participation numbers in those sports,98 the magni­ tude of the athletic injury problem to the foot and ankle and the need for better prevention become evident. Whether the sport is a team or individual sport, organized or ­unorganized, professional or nonprofessional does not appear to make a major difference in the types of foot and ankle injuries that occur and has only a minor influence on the overall injury rate. This suggests that the intrinsic movements and extrinsic load characteristics of the foot and ankle required by sports participation involve a certain basic risk for injury.

Foot and Ankle 2175

  TABLE 25J-1  Injury Rates Calculated for the Foot and Ankle in Various Sports from a Review of the Literature Sport/Study

Skill Level

Ankle Injury (%)

Foot Injury (%)

Recreational Recreational

12 11

5 18

Professional and nonprofessional Review of literature

17

22

14

15

Professional

19

4

Professional Professional High school

19 18 31

4 6 8

N/A Club

F&A F&A

8 14

Various levels Student

17 22

15 15

Top class Top class

F&A F&A

13 6

Aerobics

Rothenberger Garrick

Sohl

Basketball

Zelisko Henry Moretz

Dance (general)

Washington Rovere Equestrian

Bermhang Bixby-Hammett

Meyers

Blyth Culpepper DeLee Zemper

High school High school High school College

Canale

Ferkel Perlik

Cottlieb Walter Temple Marti Brown Smith

Johnson Blitzer Dowling Pino

Professional Amateur

2 3

3 2

Squash/racquetball

Club High school

21 10

3 8

Tennis

Amateur High school College Professional Junior

0 0 7 0 4

0 0 10 0 1

College College

15 14

4 4

N/A N/A

41 40

8 35

Military

7

0.3

Parachuting

Petras

N/A 2

College/club

8

2

Recreational Recreational N/A N/A

19 15 26 30

11 16 26 10

National males Age 14-19 yr

8 29

8 8

Various Youth

9 F&A

N/A 8

USSA

8

N/A

Recreational

26

3

Swedish senior male division Various Amateur leagues (France)

17

12

36 20

8 N/A

Recreational N/A

21 20

2 7

Elite

11

9

National amateur

18

6

N/A

4

15

Elite, Olympic

2

0

College Olympic College High school

10 10 4 3.8/100 wrestlers

3 0 0

Soccer

Ekstrand Nilson Bareger-Vachon

Berson Soderstrom Winge Volleyball

Schafle Water skiing

Hummel Weight training

Mountaineering

McLennan Tomczak

10 8

Freestyle

College

Lacrosse

Muller Nelson

College N/A

Skiing Downhill

2 4 2 4 AE = 0.25 2

Ice hockey

Park

1

Running

15 11 18 16 AE = 1 11

Gymnastics

Sutherland

6

Rugby

Golf

Caine Garrick

College

Snowboarding

Football

McCarroll McCarroll

Foot Injury (%)

Ice skating

Cycling

Davis Kiburz

Ankle Injury (%)

Rodeo

Micheli

Baseball

Garfinkel

Skill Level

Roller skating

Ballet

Garricki

Sport/Study

Kulund Wrestling

Roy Lok Snook Requa

AE, number of injuries per athletic exposure; F&A, foot and ankle; N/A, not available; USSA, United States Ski Association. From Clanton TO: Athletic injuries to the soft tissues of the foot and ankle. In Coughlin MJ, Mann RA (eds): Surgery of the Foot and Ankle, 7th ed. St Louis, Mosby, 1999, pp 1093-1094.

2176 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

FLEXIBILITY AND STIFFNESS Flexibility is one of the components of physical fitness that has been judged to be critical for injury-free perfor­ mance.104,105 Its effect on injury to the foot and ankle has been discussed primarily in relation to (1) ankle joint stiff­ ness affecting the incidence of ankle sprains and (2) flatfeet or hyperpronated feet as a source of injury or pain.

Definitions Flexibility is the range of motion commonly present in a joint or group of joints that allows normal and unimpaired function.106,107 It can be subdivided into static and dynamic flexibility. Static flexibility is the maximal range of motion that a joint can achieve with an externally applied force, such as gravity. Dynamic flexibility is the range of motion that an athlete can produce and the speed at which he or she can produce it. Dynamic flexibility is important in high-­velocity movement sports such as throwing, sprint­ ing, or jumping.104,107 Having excellent static flexibility does not mean that one possesses excellent dynamic flex­ ibility. Beighton and coworkers noted that articular range of motion is a spectrum, where one end of the spectrum includes considerable joint laxity.108 When flexibility exceeds the normal range of movement in multiple joints, an individual is considered loose-jointed or hypermobile. Hypermobility has been shown to have multiple inheri­ tance patterns and can occur in both benign and patho­ logic forms.109-113 Box 25J-2 shows characteristics of joint mobility patterns and changes with age and gender and ethnicity. Stiffness is the physical measurement of a reduced range of motion of a joint or a group of joints. The loss of flex­ ibility during rapid growth in children causes both mus­ cle tendon imbalance and increased apophyseal traction. Resultant overuse injuries include traction apophysitis and tendinitis. Sever’s disease is a common cause of heel pain in adolescents. Little discussion in the medical literature centered on hypermobility until it was proposed as a heritable cause of joint problems by Finkelstein in 1916.114 Connective tissue diseases such as Ehlers-Danlos syndrome, Marfan syndrome, Larsen’s syndrome, Down syndrome, hyperly­ sinemia, homocystinuria, and osteogenesis imperfecta are

Box 25J-2 Flexibility Generalizations

• Inherited characteristic • Individual variability • Joint specific • Females more flexible then males • Decreases with age • Can be acquired through training • Strength training does not necessarily reduce flexibility • Little relationship with body proportion or limb length • Little relationship with injury rate Data from references 13, 107, 108, 116, 118, 121, 126, and 127.

known to be associated with joint hypermobility and resul­ tant subluxation and dislocation of joints, although this rarely affects the foot and ankle joints.111,115-117 Hypermobility is a trademark of certain sports such as diving, gymnastics, and ballet. Ballet dancers are par­ ticularly noted for excessive motion in their spine, hips, and ankles.118 To a certain degree, individuals with these traits of flexibility are selected through the intense train­ ing that is required in competitive dancing. It seems that inherited hypermobility gives these individuals an advan­ tage during the training years.118 There is also a nega­ tive side: ­ studies have shown that hypermobile dancers have a higher incidence of injuries than those who are not ­hypermobile.65,119 Although Nicholas cited hypermobility as a causative factor in ligamentous injuries in athletes in 1970,120 this relationship was not confirmed in follow-up studies.38,121-123 The connection between hypermobility, particularly when it is familial, and dislocation of one or more joints is rela­ tively well established, but it has not included the other joints of the foot and ankle.116,121 The occurrence of joint effusions in the knees and ankles in the absence of trauma or other known inciting causes was attributed to hypermo­ bility by Sutro in 1947.124 Additionally, there have been some reports suggesting an association between hypermo­ bility and osteoarthritis.116,118,121,125 Although the presence of flatfeet is frequently included among the characteristic findings of individuals who are hypermobile, it is quite clear that not all individuals with flatfeet are hypermo­ bile.126,127 There is considerably less evidence implicating hypermobility with other pathologic conditions of the foot and ankle. Criteria for the diagnosis of hypermobility were first proposed by Carter and Wilkinson128 and later ­modified by Beighton and colleagues.108 The tests com­ monly used include (1) passive thumb apposition to touch the forearm, (2) passive little finger hyperextension of more than 90 degrees, (3) elbow hyperextension of more than 10 degrees, (4) knee hyperextension of more than 10 degrees, and (5) forward flexion of the trunk with the knees straight and the palms of the hands resting flat on the floor (Fig. 25J-3). Hypermobility is diagnosed in ­individuals who can perform three or more of these tests.117 The distinction between hypermobility, or laxity, and instability is an important one. Hypermobility is normal movement carried beyond the range found in most indi­ viduals. This hypermobility is primarily a function of the stiffness within the muscles, ligaments, and tendons cou­ pled with the bony configuration of the joint. Instability is a symptom-producing phenomenon that is related to the ligamentous and bony integrity of joint as well as compres­ sive joint forces and neuromuscular control mechanisms and their opposition to the forces of shear, distraction, and angulation. Instability may be further subdivided into functional and mechanical instability. The idea of functional instability as proposed by Freeman is “… the occurrence of recurrent joint insta­ bility and the sensation of joint instability due to the contributions of any neuromuscular deficits.”1,129 Such deficits would primarily be related to injury to the joint ­mechanoreceptors and afferent nerves resulting in combi­ nations of impaired balance, reduced joint position sense,

Foot and Ankle 2177

A

B

C

Figure 25J-3   Tests for hypermobility. A, Passive thumb apposition to touch forearm. B, Passive little finger hyperextension.   C, Forward flexion of the trunk so that palms rest on the floor. (Photographs by Christopher B. Hirose.)

and slower firing of the peroneal muscles in response to inversion stress. Mechanical stability, in contrast, is defined as “laxity of a joint due to structural damage to ligamentous tis­ sues which support the joint.”129 Structural damage also includes damage to the bony support of the joint. This type of instability is not always evident on physical exami­ nation. The varus stress test applied to the ankle to pro­ duce a talar tilt demonstrating lateral ligamentous injury at the ankle can be difficult to interpret in patients who have increased subtalar motion. Adding to the difficulty of such an ­assessment, as of this writing, there are no scientifically based anatomic and biomechanical definitions that are uni­ versally accepted.

Historical Perspective on Flexibility Hypermobility is often seen in athletes and may be more frequently symptomatic than in the nonathletic population. This is likely due to the added stress of sports participa­ tion. According to Grahame, Hippocrates first mentioned hypermobility as a source of difficulty for athletes as early as the 4th century bc.116 The importance of flexibility in athletic performance and the prevention of injury are of rather recent ori­ gin according to Corbin.107 The immobility caused by wartime injuries, together with the epidemic of polio­ myelitis, instigated research efforts in this field in the past century. Cureton emphasized flexibility as an important component of physical fitness as early as 1941 after his work with swimmers during the 1932 ­Olympic games.130,131 Kraus’ work led to the formation of the President’s Council on Physical Fitness and Sports and fostered much of the subsequent research on flexibil­ ity.107 In a classic work, DeVries proved the value of passive stretching in improving flexibility.132,133 Further­ more, the work of Salter and associates in Toronto has shed new light on the importance of maintaining flex­ ibility and motion postoperatively in patients who have undergone musculoskeletal procedures. The benefits of continuous passive motion to the joints and to the

s­ urrounding musculotendinous and ligamentous struc­ tures are well established.134-136 Although the natural inclination and the accepted teaching for many years in the fields of sports medicine and exercise physiology has been that stretching is a preventive measure for athletic injury, there is little conclusive epi­ demiologic evidence to support this idea, and studies have been contradictory and inconclusive.45,137,138 Two ­ welldesigned studies of running-related injuries failed to show a significant relationship between stretching or its absence and the frequency of injury.32,51 Conversely, a study of mil­ itary basic trainees showed a reduction in overuse injures with an effective hamstring flexibility program.139 Research from Duke University has provided scientific ground­ work on the preventive aspects of stretching and warm-up periods by showing that greater tension is required to rup­ ture a muscle that has been stretched.140,141 There is the possibility that a degree of tightness protects against injury when joints are stressed. This implies that stretching beyond a certain point may reduce the load-sharing ability of the musculotendinous units or the capsuloligamentous com­ plex, which are responsible for joint stability.137 Ingraham believes that the current evidence suggests that stretching and increasing the range of motion beyond function are not beneficial and may actually cause injury and decrease performance.142 If this is indeed the case, excessive stretch­ ing or hypermobility could result in increased stress on the ligaments, bone, and cartilage at the joint, leading to injury or arthritis.141,143,144 This may be the situation when ballet dancers force ankle plantar flexion to such an extent that it creates posterior impingement symptoms and reactive bone formation (Fig. 25J-4).65,119,145-147 The balance between adequate and excessive stretching is further demonstrated in studies of runners. Research has shown that during level running a great deal of energy is stored in the muscle-tendon unit.13,148 Several laborato­ ries have shown that less flexible individuals use less oxy­ gen to cover the same distance while running at the same speed than more flexible individuals.149 This interesting fact may explain why runners as a group are relatively inflexible unless specific stretching exercises are pursued.

2178 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 25J-4   A, Computed tomographic scan demonstrating bony posterior impingement. B, Excised bony fragment. (Photographs by Christopher B. Hirose.)

The importance of muscle stiffness in athletic injury remains an area for continued study because there is no scientifically based program for flexibility training with statistically reproducible results in lowering injury rates or improving performance.13 In summary, there appears to be an optimal range of motion for each individual and for each of his or her joints that may be sport specific. Warming up and stretching to obtain or maintain this range may or may not prevent

Sagittal plane Frontal plane

Horizontal plane

injury, but stretching beyond this range is potentially harm­ ful. Certain sports require a particular range of motion and flexibility. Athletic activity and the competitive process tend to select naturally those who can attain the movement criteria with the most efficiency and without a subsequent increase in injury rate.150

Joint Motion We define joint motion based on the position of the human body in relation to the three cardinal planes: the sagittal, frontal, and horizontal planes (Fig. 25J-5 and Table 25J-2).151 This system considers pronation and supination as a triplane motion. Pronation consists of eversion, abduction, and dorsiflexion, and supination consists of inversion, adduc­ tion, and plantar flexion.151,152 It is apparent that motion in almost all joints and specifically in the ankle, subtalar, and transverse tarsal joints is actually a triplane motion to one degree or another.152-156 As an example, dorsiflexion of the ankle produces some degree of foot abduction as well as external rotation. Similarly, plantar flexion produces some adduction and internal rotation. There are different norms for range of motion of the ankle, subtalar, and first metatarsophalangeal joints (Tables 25J-3 to 25J-5). This is due to the degree of ana­ tomic ­constraints around these joints for each individual. A review of the literature shows that the average ankle range of motion is calculated as 53 degrees, with average dorsiflexion equaling 18 degrees and average plantar ­flexion

  TABLE 25J-2  Terminology Used for Motion, Instability, and Deformity Plane

Motion

Position

Deformity

Sagittal

Dorsiflexion Plantar flexion Inversion Eversion Adduction Abduction Pronation  Supination

Dorsiflexed Plantar flexed Inverted Everted Adducted Abducted Pronated Supinated

Calcaneus Equinus Varus Valgus Adductus Abductus Pronatus Supinatus

Frontal Figure 25J-5   The cardinal planes of motion. (Redrawn with permission from Women in Sports. Sport Research Review. Beaverton, Ore, Nike Sport Research Laboratory, Mar/Apr, 1990; and Oatis C: Biomechanics of the foot and ankle under static conditions. Phys Ther 68:1815-1821, 1988.)

Transverse Triplane description

Foot and Ankle 2179

  TABLE 25J-5  First Metatarsophalangeal Joint Motion

  TABLE 25J-3  Ankle Joint Motion*

Study

Method

Dorsiflexion (Extension) (degrees)

AAOS159 Bonnin406 Boone and Azen158 Sammarco175 WB NWB Weseley169

NS NS A

20 10 to 20 12.6 ± 4.4

50 70 25 to 35 35 to 55 56.2 ± 6.1 66.8 ± 5.5

P P P

21 ± 7.21 23 ± 7.5 0 to 10 (max 23)

23 ± 8 23 ± 9 26 to 35 (min 10) (max 51)

44 ± 7.6 46 ± 8.25 26 to 45 (min 51) (max 84)

13 to 33

23 to 56

36 to 89

24.9 ± 3.0 20

28.5 ± 7.5 53.4 ± 5.25 40 60

Lundberg155 Review/ NS summary Personal study NS AMA407 A or P

Plantar Flexion (degrees)

Study Total (degrees)

*When a range is given, it signifies the range of greater concentration. A, active; NS, not specified; NWB, non-weight-bearing; P, passive; WB, weight-bearing.

equaling 35 degrees.155-159 Variability in ankle motion mea­ surements is introduced depending on the methodology of measurement: radiographic or clinical using flexometry, goniometry, or electrogoniometry; selection of land­ marks; and measurement of specific tibiotalar movement or combined ankle-foot motion.155 Timing with regard to a warm-up period and geographic consideration may also be important.160-162 As one moves distally, range of motion of the joints of the foot and ankle becomes increasingly difficult to measure objectively. This is particularly true of subtalar motion, for which numerous methods of measurement have been described. No consensus for normal motion has been reached (Fig. 25J-6; see Table 25J-4). This absence of a consensus creates considerable difficulty when trying to determine whether or not subtalar instability exists.163 It is clear that there is intersubject variability in measur­ ing subtalar motion. At the current state of knowledge, subtalar motion is generally described as movement in the frontal plane of 10 to 59 degrees, with an average of 24 degrees.164,165

AAOS157 AMA407 Sammarco175 Joseph176

Standing

16 Clanton173

Dorsiflexion (degrees)

Plantar Flexion (degrees)

70 50 90

45 30 30

Active

Total

51 (40 to 100) 72.3 ± 12.4

74

23 (3 to 43) 35.0 ± 10.2

Transverse tarsal joint motion plays an important role in lower extremity kinematics. Its instability has been described as medial swivel syndrome.166,167 There has even been one report of surgical treatment of patients with trans­ verse tarsal joint instability.168 Average motion in the trans­ verse tarsal joint depends on the position of the hindfoot, or subtalar joint. As the subtalar joint undergoes eversion early in the gait cycle, the axes of the transverse tarsal joint are aligned so that they become parallel in this everted position of the heel, and this permits increased motion. This is about 22 degrees.153 Conversely, when the heel is inverted as occurs near the end of stance phase, the axes of the transverse tarsal joint become more divergent and less mobile, averaging 8 degrees of motion.153 This fact is important when assessing range of motion about the tib­ iotalar joint. If range of motion of the tibiotalar joint is tested with the hindfoot in an everted position, one gets a falsely increased range of motion due to a combination of motion through the tibiotalar joint and transverse tarsal joint. With the heel inverted, a truer measure of tibiotalar

  TABLE 25J-4  Subtalar Joint Motion Study AAOS157

Inman154

Sammarco175 DeLee408 McMaster409 Brantigan410 James251 Milgrom411

Inversion (degrees)

Eversion (degrees)

Arc (degrees)

5 — 20 — 25 — 23 ± 6

5 — 5 — 5 — 8±4

10 44 ± 7 25 35 30 38 ± 6 31 ± 7

Method

R

L

ASOS Cailiet James

2.0 ± 7.4 21.4 ± 5.4 18.4 ± 5.2

31.2 ± 7.7 3.9 ± 4.1 21.6 ± 5.5 3.4 ± 3.1 18.4 ± 5.2 6.4 ± 3.7

R

L

4.0 ± 4.2 3.2 ± 3.3 6.7 ± 4.2

25–30°

A

0°

B

5–10°

C

Figure 25J-6   Method of measuring subtalar motion with patient prone and knee flexed. A, Neutral position at   0 degrees on goniometer. B, Inversion. C, Eversion. (Redrawn with permission from American Academy of Orthopaedic Surgeons: Joint Motion-Method of Measuring and Recording. Chicago, American Academy of Orthopaedic Surgeons, 1965; and Oatis C: Biomechanics of the foot and ankle under static conditions. Phys Ther 68:1815-1821, 1988.)

2180 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

joint motion can be obtained because a minimal amount of motion is taking place through the transverse tarsal joint. Large variations in motion exist in this region, and this makes the job of defining separate motions for the talo­ navicular and calcaneocuboid joints difficult.155 The lit­ erature indicates that 13 to 15 degrees of dorsiflexion or plantar flexion movement occurs in the midfoot.169 Motion in the tarsal-metatarsal joints is also a triplane motion, but it occurs primarily in a single plane.153,156 This is most obvious in the first and fifth ray, where we describe dorsiflexion and plantar flexion movement. With dorsiflex­ ion of the first ray, however, some degree of abduction and external rotation occurs, and with plantar flexion, there is adduction and internal rotation.170 Some believe that the first ray, like the transverse tarsal joint, has a locking mechanism. Perez and colleagues have demonstrated that the frontal plane position of the first ray affects the sagit­ tal plane motion.55 An everted position, compared with an inverted position, has the least mobility in the sagittal plane. Average dorsiflexion and plantar flexion in the first through fifth ­ tarsal-metatarsal joints are 3.5, 0.6, 1.6, 9.6, and 10.2 degrees, respectively.171 The triplane motion of supi­ nation and pronation was also described as 1.5, 1.2, 2.6, 11.1, and 9.0 degrees, respectively. This motion is used to advantage in the Lapidus procedure for treating hallux val­ gus and metatarsus primus varus in the hypermobile foot. Translational movements are abnormal in this area, and excessive forces are a cause of pathologic conditions rang­ ing from stress fractures at the base of the second metatarsal to mild diastasis between the first and second metatarsals

(Fig. 25J-7) and on to the more severe forms of Lisfranc’s fracture-­dislocation (Fig. 25J-8). The normal range of motion in the first metatarsopha­ langeal joint is quite variable (Fig. 25J-9). According to Joseph, the average range of motion is 51 degrees of active dorsiflexion and 74 degrees of active plus passive dorsi­ flexion using the axis of the first metatarsal as the neutral line.172-174 The natural position for this joint in the standing posture is 16 degrees of dorsiflexion. Active plantar flexion varies between 23 degrees and 45 degrees.172-174 In per­ forming these measurements, some variability occurs from the positioning of the ankle and subtalar joints. Motion is reduced when the ankle is in dorsiflexion and the subtalar joint is inverted. Motion is also reduced with advancing age.172 When maximal dorsiflexion is reached in the first metatarsophalangeal joint, the normal gliding motion of the proximal phalanx on the metatarsal head ceases, and impingement can occur between the proximal phalangeal base and the first metatarsal head.175 This may be a source of some of the problems seen in turf toe injury as well as hallux rigidus. First metatarsophalangeal joint dorsiflexion is important because of its relationship to gait and to stabil­ ity of the skin of the metatarsal pad.176 Motion in the lesser metatarsophalangeal joints and interphalangeal joints has been studied less thoroughly. Joseph’s study included motion at the interphalangeal joint of the great toe. Average motion was 31 degrees of total extension and 46 degrees of active plantar flexion. Recorded norms for lesser metatarsophalangeal joint motion has a wide range from 40 degrees to 90 degrees of dorsiflexion

Figure 25J-7   A, Non–weight-bearing radiograph with small fleck sign near the base of the second metatarsal. B, Weight-bearing radiograph with subtle diastasis. (Photographs by Christopher B. Hirose.)

A

B

Foot and Ankle 2181 Figure 25J-8   A, Photograph of a swollen midfoot and forefoot in a patient with a Lisfranc injury. B, Radiograph of diastasis and fracture of the second metatarsal. (Photographs by Christopher B. Hirose.)

A

B

and from 35 degrees to 50 degrees of plantar flexion.157,175 Excessive movement or stress to the lesser metatarsopha­ langeal joints can be a rare source of disease. Isolated syno­ vitis and instability of the lesser metatarsophalangeal joints, however, are becoming increasingly recognized as a source of forefoot pain in patients, including athletes.177,178 Inter­ phalangeal joint motion in the lesser toes is even less well

defined, with 0 degrees being the standard for extension and 35 to 40 degrees for plantar flexion.157 Loss of motion in these interphalangeal joints rarely creates symptoms unless there is a fixed flexion deformity, such as a hammer toe or mallet toe (Fig. 25J-10). Knowledge of the normal range of motion allows one to determine the etiology of specific injuries to athletes, while Figure 25J-9   Variability of first metatarsal phalangeal joint dorsiflexion. (Photographs by Christopher B. Hirose.)

A

B

2182 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Figure 25J-10   Proximal interphalangeal joint flexion. (Photograph by Christopher B. Hirose.)

being mindful that a considerable degree of variation exists that obscures a precise cause-and-effect relationship.

Etiologic Role of Flexibility With a more scientific understanding of joint motion and flexibility, we will examine foot and ankle injuries related to flexibility or its absence. Turf toe injuries provide one of the best examples of disease related to loss of flexibility. It was initially believed that turf toe injuries were related to lack of flexibility in the first metatarsophalangeal joint.179 We might expect a greater frequency of hyperexten­ sion injuries in athletes who have less natural dorsiflexion motion in the first metatarsophalangeal joint. All research to date, however, has failed to confirm an etiologic relation­ ship between the loss of dorsiflexion of the first metatarsal phalangeal joint and turf toe injuries.173,174,180,181 An initial first metatarsal phalangeal joint injury can cause long-term morbidity, and hallux rigidus and hallux valgus are two specific long-term sequelae.173,182 In hallux rigidus, a loss of motion occurs in the first metatarsophalangeal joint that produces pain (Fig. 25J-11).183 Lack of flexibility in the lesser metatarsophalangeal joints or interphalangeal joints is rarely linked with inju­ ries or pathologic conditions in athletes. The loss of inter­ phalangeal joint motion is a key component in mallet toes, hammer toes, and claw toes, and flexibility exercises

are often advocated in the treatment of these conditions. A lack of flexibility in such toes can result in painful calluses at the tips of the toes and painful corns over the dorsum of the interphalangeal joints. A lack of dorsiflexion in a lesser metatarsophalangeal joint can be a source of pain and disability, particularly when this involves the second toe. There is usually an iatrogenic or post-traumatic cause. Conversely, increased motion appears to be a factor in conditions involving the second metatarsophalangeal joint, including transient synovitis, crossover second toe, and subluxation and dislo­ cation of the second toe.177,178,184 Flexibility studies on the relation of these conditions to injury of the lesser metatar­ sophalangeal joints are sparse. Hypermobility of the first ray has been implicated as a source of problems in the foot such as stress fractures, but confirmatory studies using accepted statistical techniques are lacking.185,186 Simkin and colleagues have suggested an asso­ ciation between low arches and metatarsal stress fractures.187 Gross and Bunch measured stresses in 21 distance runners and found that the greatest force occurred under the first and second metatarsal heads.188 Using the model of the meta­ tarsals as proximally attached rigid cantilevers, the authors showed that the greatest bending strain and shear force occurred in the second metatarsal, where stress fractures are most common. These two studies provide circumstantial evidence that increased first ray flexibility is an anatomic fac­ tor in stress fractures of the second metatarsal shaft. Studies of the length of the metatarsals in dancers and risk for second metatarsal stress fractures note contradictory and inconclusive findings.189-192 Some believe that a longer second metatarsal predisposes dancers to an increase risk for second metatarsal stress fracture193-195 because a long second metatarsal is believed to be the recipient of added stress, especially in the en pointe position.196 ­Others note no association between the relatively long second metatar­ sal and increased risk for stress fracture.191,197 We may expect that the best source of a direct link between the lack of flexibility and injury would come from the ankle because of the high frequency of ankle injuries in sport. The tight Achilles tendon has been blamed for numer­ ous conditions, including bunions,198 turf toe,181 midfoot strain or plantar fasciitis,199-201 ankle sprains,202 Achilles tendinitis,68 calf strains,203,204 and ­ hyperpronation.205-207

Figure 25J-11   Hallux rigidus. (Photograph by Christopher B. Hirose.)

Foot and Ankle 2183

Although there are numerous studies, evidence remains scarce for a causal relationship. Walsh and Blackburn sug­ gested heel cord stretching as a preventive maneuver to decrease the incidence of ankle sprains, but they reported no results from this program.202 Mahieu and colleagues determined that increased ankle dorsiflexion ­excursion and reduced plantar flexion strength may be linked with Achil­ les tendon overuse injuries.208 The strength of the plantar flexors and amount of dorsiflexion excursion were identi­ fied as significant predictors of an Achilles tendon overuse injury. A plantar flexor strength lower than 50 Nm and dorsiflexion range of motion higher than 9.0 degrees were possible thresholds for developing an Achilles tendon over­ use injury.208 It seems somewhat paradoxical that on the one hand, the athletic trainer works hard to improve the athlete’s flexibility at the ankle, and on the other hand, the trainer tapes or braces the ankle before practice or games to restrict motion. Hyperpronation has been blamed for many problems known to runners.205-207 Many runners come to their local orthopaedist, podiatrist, or running shoe salesperson with the self-made diagnosis of “pronated feet” and ask for a shoe or an orthotic device to cure their shin splints, knee pain, or arch pain. Running shoes have been made and marketed spe­ cifically for pronated feet or for the supinated, rigid, cavus foot.209-211 The complex movements in the foot and ankle associated with pronation produce a more flexible foot at the time of weight transfer. This can produce problems in one of two ways. First, the normal foot212,213 goes through 6 to 10 degrees of subtalar eversion in the frontal plane during gait, and the flatfoot or pronated foot may have 12 to 15 degrees of motion.214-217 This increased motion will produce a cor­ responding increase in transverse plane motion.218 Second, the speed with which the rearfoot angle changes seems to be important.219,220 The pronated foot goes through the avail­ able pronation range more quickly, and this force results in increased load transmission.214,221 Both of these movements are important components of load absorption in running. In addition, the duration of foot pronation may also have a protective effect on tibial stress fractures. In the cavus foot, there is more rigidity in the joints of the foot.219,222 This means that loads are not dampened as effectively, and higher stress is applied at each level, which can become symptomatic. So far, treatment has centered on the use of orthotic devices, but their success in the ath­ lete with a cavus foot is limited.207,222-224 As with all areas of science, studies of the etiology of foot and ankle injuries produce more and more ­questions. Because of the interrelationship of many factors and the extent of individual variability, it has been difficult to impose the proper degree of control in open systems and in vivo studies to come to solid conclusions. Current studies have not provided a reliable prediction of the influence of the pronated or cavus foot on the risk for injury.28,187,215,225-229 A 1999 study from the Mayo Clinic looked at the effect of foot structure and range of motion on musculoskeletal overuse injuries.230 The study group was a well-defined cohort of 449 naval trainees. They were tracked prospec­ tively for injuries throughout training. The risk factors that predisposed trainees to overuse injuries in this study were dynamic pes planus, pes cavus, restricted ankle dorsi­ flexion, and increased hindfoot inversion.

SHOEWEAR-RELATED INJURY Although the role of flexibility in sports injuries of the foot and ankle remains somewhat unclear, more evidence exists implicating shoewear and playing surfaces. The intimate relationship between these two makes separation into indi­ vidual components difficult. Many studies of sports inju­ ries look at these in a combined manner as the shoe-surface interface.231-238 The shoe can be a factor in athletic inju­ ries in other ways, such as improper fit, lack of cushion­ ing, inadequate support, and abnormal force generation. For this reason, shoewear and playing surfaces have been included as separate etiologic factors here, but the reader should maintain an awareness of their interdependency.

History Barefoot participation in sports was the norm in ancient times. Perhaps the first recorded injury in shod feet was noted by the Greeks.61 Today, shoes can cause problems for athletes in so many ways, one has to wonder whether it might be better for athletes to participate without shoes. An advantage to barefoot play can be seen in young chil­ dren who remove their shoes to race. More recently, Zola Budd and Abebe Bikila achieved Olympic fame without shoes (Fig. 25J-12). The symptom-free nature of peoples who trod without shoewear (regardless of their degree of pes planus or pes cavus) also makes one question the true value of modern shoewear.239-242 Numerous studies have demonstrated that the least amount of pronation occurs during barefoot run­ ning.214,220,243,244 This finding stimulates an inquiry into the

Figure 25J-12   Zola Budd running in 1984 Olympic   3000-meter race against Mary Decker. (© 1991 David Madison.)

2184 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

role of pronation in injury and whether a shoe designed for overpronators is protective. Robbins and colleagues cham­ pioned the role of sensory feedback from the plantar ­surface of the foot in modifying load and protecting the runner from stress-related injury.245-247 They suggested that this important system is impaired by modern footwear, creating a “pseudo-­neuropathic” condition. In support of this, they note that studies have shown no trend toward a reduction in injury from the use of modern athletic footwear. DeWit and colleagues looked at the biomechanics of the stance phase during barefoot and shod running.248 Barefoot running is characterized by a significantly larger external loading rate and a flatter foot placement at touchdown. This flatter foot placement correlates with lower peak heel pressures. It was postulated that runners adopt this altered foot ­ placement position in order to limit the local pressure beneath the heel. Divert and colleagues studied the biomechanics of barefoot compared with shod runners.209 Barefoot runners showed lower contact and flight time and lower passive peak than shod. They concluded that barefoot running leads to a reduction of impact peak in order to reduce the high mechanical stress occurring during repetitive steps. They then studied why athletes have a higher oxygen consump­ tion and lower net efficiency when running shod compared with running barefoot. Many believe that this effect is due to the additional mass of the shoe, and their results show that there is a significant mass effect and increasing oxygen consumption. However, stride frequency, vertical stiffness, leg stiffness, and mechanical work were significantly higher in barefoot condition. The lower net efficiency reported in shod running may also be due to the impact-dampen­ ing properties of the shoe as the foot strikes the ground.249 Even though these studies suggest that shoes have no pro­ tective effect, most athletes still use shoes for competition and daily life. We explore a critical analysis of shoes in this chapter and discover how they fall short in their role as pro­ tective equipment for the sports participant.

factors.251 Nineteen percent of the 180 runners in their series were treated by a change or modification in shoewear. Lysholm and Wiklander studied 60 runners with 55 injuries during a 1-year period.252 Surface or shoe problems were the primary sources of injury in 3 cases and one of multiple factors in 10 others. Inferior footwear was the etiologic fac­ tor blamed in 34 of 318 injuries in soccer players. Football has provided the best perspective on shoewear and its relationship to knee and ankle injuries. Torg and Quedenfeld published an extensive study in 1971 relating the incidence and severity of knee and ankle injuries to shoe type and cleat length in Philadelphia high school foot­ ball players.253 Rates of injury to the ankle were reduced from 0.08 per team per game to 0.01 per team per game by switching from conventional cleated football shoes to soccer-style shoes with multiple shorter cleats. This result was substantiated by the work of Mueller and Blyth, who noted a reduction in knee and ankle injuries by resurfacing the playing field (30.5% reduction), changing from regular cleats to soccer shoes (22.3% reduction), or making both changes (46% reduction).254 Although these studies do not consider such shoe-related problems as cleating-induced contusions or lacerations or turf toe, they do provide a framework for studying the role of shoes in sports injury.

Mechanical Factors Shoe Fit

The exact incidence of injury attributable to athletes’ shoes is unknown, but several studies have included shoes as a sep­ arate factor in injury rates.250 James and colleagues included shoes and surfaces as one of three categories under cause of injury in runners in addition to training errors and anatomic

The most obvious problem with shoes that plagues the shod people of the world regardless of whether they are athletes is the proper fit. Even the smallest problem with fit can prevent athletes from performing to the best of their capabilities. For example, a web corn can become so pain­ ful that each step is agony. Metatarsalgia can result from a shoe that is too tight across the forefoot. Similarly, a narrow shoe often aggravates a Morton’s neuroma (Fig. 25J-13). The black toe of long-distance runners can be a sequela of improper shoe fit. Toe deformities such as ham­ mer toes, claw toes, and overlapping fifth toes may become symptomatic in athletes whose shoes have toe boxes that are either too narrow or of insufficient depth. Calluses and blis­ ters are inherent in most sports participation (Fig. 25J-14). Improperly fitted shoes that rub the skin excessively or allow the foot to move or slide disproportionately in the

Figure 25J-13   Morton’s neuroma excised from the third web space. (Photograph by Christopher B. Hirose.)

Figure 25J-14   Calluses in a long-distance runner. (Photograph by Christopher B. Hirose.)

Incidence

Foot and Ankle 2185

shoe enhance the frequency of calluses and blisters. Other irritants include the insole edge, penetrating cleats, promi­ nent seams, and orthotic device edges.

Cushioning The importance that cushioning protects athletes from injury has been promoted largely by the running shoe industry and those whose research it supports.255 The run­ ning literature has seen a proliferation of articles address­ ing impact forces, shoe cushioning, shock absorption, and the effects of various alterations in shoe construction. The logical assumptions have been that load on the human body is directly attributable to impact forces at foot strike and that these forces are naturally altered by the cushioning properties of the shoe.243,255-258 These assumptions fos­ ter the belief that changing the shoe’s material properties (e.g., midsole thickness) can influence impact load, thereby changing rates of injury. Impact forces are a critical feature in the etiology of sports-related pain and injury whether acute or chronic.27 The bionegative effects of impact loads are evident from the damage produced to articular cartilage by high-impact loads.259,260 Radin and coworkers showed that deleterious changes occurred in the biochemical and biomechani­ cal properties of articular cartilage of sheep walking con­ stantly on concrete as opposed to those walking on wooden chips.259,260 Impact force and shock wave transmission play a role in the etiology of experimentally induced osteoar­ thritis.261,262 Shock-absorbing insoles in the boots of South African military recruits produced a 9% reduction in their incidence of overuse injuries.263 In addition, the German armed forces studied the properties of cushioned insoles. The aim was to assess metatarsal head loading in combat boots with respect to the prevention of metatarsal stress fractures comparing cushioned with standard conventional insoles. The cushioned insoles were superior to the con­ ventional insoles with respect to the plantar pressure dis­ tribution.264 These studies support the belief that impact loads are an etiologic factor in certain injuries, but care­ ful analysis leaves the impression that this load plays a less than consequential role. Ground reaction forces are the primary external force acting on the human body during running.27,265-268 These

forces increase as running velocity increases. Higher ­gro­und reaction forces are seen in the progression from walking to jogging to sprinting to jumping. Estimates vary from 1.2 times body weight (BW) for walking to 2.5 times BW for jogging. Sprinting increases load by 3 to 6 times BW, whereas jumping multiplies this force by 6 to 8 times BW27 (Table 25J-6). Impact force amplitude is reduced when soft materials are used in the shoe or running surface.269 This reduction is achieved by increasing the deceleration dis­ tance of the foot. Calculations of impact peaks can be made from the following equation: Fmax = Fxi = v fm,

where Fmax is the maximal force occurring in the vertical direction (Fxi) and is proportional to velocity (v), the mass of the body (m), and the spring constant for that body (f). From this equation it is apparent that impact velocity has the greatest influence on vertical load. It is also reduced by reducing the mass (e.g., body weight) or by reducing the spring constant (e.g., knee flexion angle).246,270-272 As an etiologic factor in injury, shock absorption has not proved to be the critical factor that advertising would lead us to believe. Although changing from a new shoe with good cushioning properties to an old shoe lacking these properties can produce injury, an injury may come from changing to a newer shoe as well. It may be that change itself is the critical factor in an injury that occurs during a repetitive activity. Reinschmidt and Nigg have noted that for running shoes, at least, pronation control and cushioning are still considered the key concepts for injury prevention despite the fact that conclusive clinical and epidemiologic evidence is missing for these design strategies.273 In addi­ tion, recent running shoe research has suggested that cush­ ioning may not be related to injuries and that cushioning during the impact phase of running may be more related to aspects like comfort and muscle fatigue. A more recent study refuted the shock-absorbing qualities of insoles in injury prevention. This study was a randomized con­ trolled trial of 1205 Royal Air Force recruits to assess the differences, if any, in the efficacy of two commonly avail­ able shock-absorbing insoles. Similar rates of lower limb injuries were observed for all insoles (shock-absorbing and non–shock-absorbing) in the trial.274

  TABLE 25J-6  Variations in Ground Reaction Force from Walking to Jogging to Sprinting to Jumping Study

Movement

Velocity (m/sec)

Footwear

Fmax (N)

Fmax (BW)

Cavanagh, 1981 Cavanagh, 1981 Clarke, 1982 Frederick, 1981 Frederick, 1981 Cavanagh, 1980 Frederick, 1981 Nigg, 1981 Nigg, 1978 Nigg, 1981

Walking heel-toe Walking heel-toe Running heel-toe Running heel-toe Running heel-toe Running heel-toe Running heel-toe Running heel-toe Running jump Running jump

1.3 1.3 2.7 3.4 3.8 4.5 4.5 5.5 6.0 8.0

Barefoot Casual shoes Running shoe Barefoot Barefoot Running shoe Barefoot Running shoe Spikes Spikes

— — — 1365 1590 — 1963 2350 4000 5500

0.6 0.3 2.8 2.0 2.3 2.2 2.9 3.6 5.3 7.9

BW, body weight; N, Newtons. From Nigg BM: Biomechanical aspects of running. In Nigg BM (ed): Biomechanics of Running Shoes. Champaign. III, Human Kinetics, 1986, p 21. Copyright 1986 by Benno M. Nigg.

2186 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Energy Absorbed N-m

14

A

Machine Testing Runner A Runner B Test System

12 10 8 6 4 10

Energy Absorbed N-m

One case example for the role of poor cushioning is that of a 26-year-old runner who sustained bilateral stress frac­ tures of the fibulae after he changed from his usual pair to an older pair of shoes for a 14-km race.275 Instrom testing of the original running shoe and the ipsilateral older shoe demonstrated that the original running shoe had twice as much energy absorption and 5 times the deformation as the older pair. Although lack of shock absorption is the pos­ tulated cause of injury in the aforementioned case, other factors might have played a role as well. These include the change itself, the altered muscle activity causing increased bending moments on the fibula, increased muscle activity resulting in muscle fatigue and loss of the protective func­ tion of the muscle, and variations in foot geometry leading to increased pronation.275-283 Cook and coworkers have demonstrated by mechani­ cal testing and in vivo experimentation that different shoe models vary by as much as 33% in shock-absorbing char­ acteristics, and all show significant decreases with mile­ age.284,285 Initial shock absorption values were reduced by 60% or more after 250 to 500 miles of machine-tested running. Other variables that reduced shock absorption were environmental conditions, such as moisture con­ tent of the shoe (e.g., perspiration or rain) and hardness of the testing surface (e.g., asphalt versus grass). Some run­ ning authorities have suggested that running shoes may have time-related life spans. The longevity of the shoe is related to the mileage run and usually falls in the 500- to 1000-mile range (Fig. 25J-15).243 One study examined heel pad stresses during heel strike with simulated wear of the ethylene vinyl acetate foam commonly found in modern sport shoes. Heel pad stresses were consistently increased when the ethylene vinyl acetate thickness was decreased, suggesting that the age of the shoe affects the viscoelastic properties of the shoe.286 Whittle looked at transient impulses beneath the foot and how they are generated and attenuated.287 He found that there is a “shock-wave” that passes up the limb that may lead to degenerative joint disease. Limb position­ ing, materials used in shoe construction, and the use of orthoses can alter these forces. In addition, the intrinsic shock absorption system of the body produces behavior modification to control load magnitude. Impact forces are dramatically reduced by increasing knee and hip flexion at ground contact.288 Several studies have proposed that cushioned shoes lead to negligible decreases in load because ­subjects decrease flexion to accommodate the instability produced by softer surfaces.12,288-290 In reality, the human body accommodates load using complex strategies, and the material used for cushioning in shoes is most likely a very small element in this shock absorption system. Milani and colleagues’ study supports this hypothesis.291 The percep­ tual ratings of eight identical running shoes with a rela­ tively close range of midsole stiffness was examined. Their findings suggest that the body’s sensory systems differenti­ ate well between impacts of different frequency content, and that subjects adapt their running style to avoid high heel impacts.291 Experimental results do not universally confirm that a lack of shock absorption is the major cause of injury.225,292,293 Softer materials of inadequate thickness have a tendency to reach maximal compressibility and then transmit greater

8

6

4 0

B

In vivo Testing Runner A Runner B Test System

100

200

300

400

500

Miles

Figure 25J-15   Life span of running shoes related to mileage as demonstrated by reduction in retention of initial shock absorption with increasing mileage run in the shoes. A, Machine testing. B, In vivo testing. (Redrawn with permission from Cook SD, Kester MA, Brunet ME: Shock absorption characteristics of running shoes. Am J Sports Med 13:248-253, 1985.)

amounts of load.220 To reduce impact forces, material thickness must be of sufficient height to prevent maximal compressibility. In one study, however, a change in mid­ sole hardness from soft to hard did not alter the impact force peaks measured in 14 test subjects at four different running velocities.294 These results appear to be related to the variation in the point of application on the foot of the ground reaction force. Harder materials impose a larger lever arm in the force equation by moving the effective point of application farther lateral from the subtalar joint axis. This produces an increase in both initial pronation and initial pronation velocity, corresponding to an increase in deceleration distance over time and a resulting decrease in the impact force.269,294 The foot-shoe-surface interac­ tion is so complex that investigators are still debating the proper methodology for testing it. There are dramatic differences between machine testing and subject testing, between internal forces and external forces, and between subjects as well as in the same subject under varying test conditions.220,225,292,295-298 Bates and coworkers have even shown that the same test in the same runner can produce

Foot and Ankle 2187

different results between trials.225,295 With all these con­ founding data, it is hardly surprising that investigators in this field frequently disagree and have occasionally pro­ duced contradictory results. In the final analysis, the question remains whether the cushioning provided by the shoe is a critical factor in the etiology of injury in the athlete. The sport shoe industry and its related research have concentrated heavily on improving the shock-absorbing characteristics of its shoes. We have seen a plethora of innovative cushioning designs such as air soles, gel shoes, variable density and encapsulated soles, and soles that will automatically vary cushioning according to the runner’s weight or speed. As remarkable as these designs are, there is still no universally accepted scientific study that proves better cushioning in the shoe lowers the incidence of injury in runners. Robbins and Waked noted that deceptive advertising creates a false sense of security with users of expensive athletic shoes, inducing attenua­ tion of impact-moderating behavior, increased impact, and injury.299 In conclusion, it can be stated that a lack of cushioning may be a factor in producing injury but prob­ ably not to the degree to which some researchers and shoe manufacturers might lead us to believe.

Shoe Control For the running shoe, control or support is primarily inter­ preted in the context of rearfoot control. This is defined as the shoe’s ability to limit the amount or rate of pronation occurring through the subtalar joint at heel strike.214 A less well-defined component of shoe support is its control of take-off supination, which occurs at the lateral forefoot during toe-off.220 Foot support also consists of the rela­ tive flexibility of the shoe, most often measured as forefoot flexibility, which allows the shoe to bend at the metatarsal phalangeal joints.243 Greater flexibility has been consid­ ered desirable, but shoe flexibility can be a major factor in turf toe injuries.179,234 Conversely, stiffening the sole in the area of the metatarsophalangeal joint may decrease energy loss and improve performance.300 Control of movement in the sagittal plane by the shoe remains a subject for further investigation. Support for lateral motion and medial and lateral shear is a concern in court shoes: tennis, basketball, and aerobic shoes.301-305 Lateral stability, torsional flexibility, cushion­ ing, and traction control appear to be important design strategies to decrease the risk for injury.273 The support provided for the ankle by high-top shoes has been an important consideration in preventing injury, particularly in basketball and football (Fig. 25J-16).305-310 Brizuela and associates examined the performance of basketball shoes with increased ankle support compared with a shoe with no ankle support.311 The high support shoes resulted in higher forefoot impact forces and lower shock transmission to the tibia. The use of high support shoes also resulted in lower ranges of eversion and higher ranges of inversion for the ankle on landing. In the motor performance tests, the high support shoes reduced the height jumped and increased the time to complete the running course compared with low support shoes. This study underlines the importance of designing shoes that will maximize performance with­ out compromising safety. Johnson and associates tested

Figure 25J-16   Examples of a current high-top and low-top athletic shoe. (Photographs by Christopher B. Hirose.)

torsional stiffness of high-top football shoes using a special chair and measurement apparatus and found that high-top shoes were 50% stiffer than low-cut models.307 Theoreti­ cally, the high-top shoe should stimulate the propriocep­ tive feedback mechanism, resulting in greater sensitization of the peroneal muscles and improved stability for the ankle.312 Potential negatives induced by the high-top shoe include the reduction in load carried by the collateral liga­ ments of the ankle and the limitation in subtalar motion restricting the foot’s ability to adapt to surface irregularities. Handoll and colleagues believe that the protective effect of high-top shoes still remains to be established.313 They stud­ ied various interventions, including external ankle supports in the form of a semirigid orthosis, air-cast braces, high-top shoes, ankle disk training, taping, muscle stretching, boot inserts, and controlled rehabilitation. The main finding was a significant reduction in the number of ankle sprains in people allocated external ankle support and no statisti­ cal difference in those athletes wearing high-top shoes.313 Although some evidence supports the ability of high-top shoes to lower the rate of injury to the ankle,280,314,315 other studies have found no difference in ankle sprain incidence related to shoe type.316-318 In related work, the effect of ankle bracing and taping on athletic performance and injury prevention remains unclear despite numerous stud­ ies.315,319-325 Control characteristics are essential elements in modern sports shoes designed for prevention of injury, but not all features are beneficial. Rearfoot stability provided by the athletic shoe has become a critical ingredient in managing the overuse prob­ lems attributed to overpronation (Fig. 25J-17).326,327 Run­ ning shoes are designed to accomplish this control. Included are such creative concepts as cantilever soles with imbed­ ded plastic stability devices, plastic torsion bars, composite plates, gel pods, air chambers, progressive-rate polymer columns, variable-density-foam midsole platforms, metal springs buried inside the heel, and even magnetic impactsensing systems, materials of variable hardness in areas of the midsole, thermoplastic heel counters, heel flares, external stabilizers, and combination lasts. The impor­ tance attributed to rearfoot control is derived from the

2188 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine ������� Figure 25J-17   Rear foot stability provided by athletic shoewear. Illustration of two runners switching between their commonly used shoes without control features and shoes with special control features provided in the laboratory. (Redrawn with permission from Nigg BM, Bahlsen AH, Denoth J, et al: Factors influencing kinetic and kinematic variables in running. In Nigg BM [ed]: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics Publishers, 1986, p 157. © 1986 by Benno M. Nigg.)

Subject A Right leg

21°

Without

assumption that overpronation produces injuries in run­ ners. This assumption has been extrapolated from studies of runners beginning with the classic study by James and colleagues.251 Fifty-eight percent of those injured were classified as pronators on biomechanical examination.251,328 The ability of the shoe or an orthosis to control excessive pronation is supported by the work of Bates and associates, Cavanagh, Nigg and Morlock, and Smith and coworkers (Fig. 25J-18).243,329-331 Take-off supination has been blamed for Achilles ten­ don strain and reduced performance.220 Take-off supina­ tion corresponds to the take-off angle, which should be close to 180 degrees. Angles between 160 and 170 degrees are indicative of increased take-off supination. Interest­ ingly, take-off supination is seldom seen in people engaged in barefoot running and is primarily a product of shoewear. It is unaffected by midsole hardness and running velocity but is heightened by additional medial support, especially when such support is located more posterior in the shoe. Nigg and coworkers have shown that geometric solutions are possible by adding lateral forefoot support or midsole

Subject B Right leg

12°

12°

31°

With Without Running heel-toe. 4 m/s

With

grooves.220 Little clinical information is available to sup­ port or refute this concept in shoe control.332 The next category to be assessed in the etiology of ath­ letic injury is shoe flexibility. This quality can be assessed by the amount of force necessary to flex the shoe’s forefoot 40 to 50 degrees.333,334 According to Ryan, the running shoe should be “moderately flexible” to allow proper foot mechanics.335 A shoe with greater flexibility is typically preferred by runners and is even advocated for those with rigid cavus feet. More convincing clinical and experimental evidence exists to implicate excessive flexibility of shoes in causing football injuries. In 1964, artificial grass was introduced, and a flexible sole soccer shoe replaced the traditional stiff, cleated football shoe.179,234 A new symptom complex appeared that was attributed to the use of these more flex­ ible shoes on the harder surface, which was nicknamed turf toe.234 Two cases of turf toe were found in track ath­ letes whose sprains occurred while wearing flexible racing spikes.179 Further characterization of this injury has made turf toe a well-defined clinical entity and has incriminated

Foot and Ankle 2189

Change of the rearfoot angle of the calcaneus

��10

8 Medial Degrees

6 Lateral 4

2

2 3 4 5

Figure 25J-18   Shoe control of excessive pronation as demonstrated by change in rearfoot angle (Δγ10) resulting from systematic alterations in location of medial support (positions 2 to 5) in the running shoe compared with support while barefoot and in shoe without support. (Redrawn with permission from Nigg BM, Bahlsen AH, Denoth J, et al: Factors influencing kinetic and kinematic variables in running. In Nigg BM [ed]: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics Publishers, 1986, p 152. © 1986 by Benno M. Nigg.)

0 1 2 3 4 5 Barefoot Shoe Anterior . . . . . . . . . . Posterior without support Shoe condition and position of medial support

overly flexible shoewear conclusively in its etiology (Fig. 25J-19).173,179-181 To their credit, shoewear manufacturers have responded by either stiffening the forefoot or provid­ ing shoe inserts to stiffen the forefoot.336 These examples support the view that the shoe flexibility plays a critical role in the etiology of injuries to the forefoot.

Orthotics Is the runner with a hyperpronating foot truly more sus­ ceptible to injury? Does a rigid orthosis holding the foot in a subtalar neutral position reduce the likelihood of injury? What role do orthoses have in injury reduc­ tion or production? Further evidence corroborating the effect of pronation control is provided by the results of treatment of injury by the use of orthoses, which led to

i­mprovement and resumption of training in 70% to 90% of those treated.152,337-339 Nigg believes that there is some evidence that orthotics reduce movement-related inju­ ries, but that the orthotic functions as a second filter of force input, the first being the shoe and the third being the athlete’s foot.340 An orthotic can optimally function by simply reducing muscle activity, increasing comfort, and ideally increasing performance. Nigg and associates looked at the effect of shoe insert construction on foot and leg movement.341 They found that the changes resulting from the use of all inserts in total shoe eversion, total foot eversion, and total internal tibial rotation were smaller than 1 degree when compared with the no-insert condi­ tion. Also, they found that the soft insert construction was more restrictive than the harder inserts. They concluded that it is important to match specific feet and shoe inserts

Figure 25J-19   Variability of the forefoot stiffness of two different running shoes. (Photographs by Christopher B. Hirose.)

A

B

2190 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

optimally. Kulcu and colleagues studied over-the-counter insoles and gait patterns with those who have a flexible flatfoot.342 They concluded that over-the-counter insoles have no beneficial effect in normalizing forces acting on the foot and on the entire lower extremity. Crosbie and Burns studied orthotics in those with pes cavus deformi­ ties and found that the mechanisms by which orthotic intervention is effective in improving pain and function in painful, idiopathic pes cavus remain unclear and equivo­ cal.343 In addition, Krivickas studied overuse injuries and bony alignment of the extremities and noted that although malalignment of the lower extremities is frequently cited as predisposing to knee extensor mechanism overuse injuries and foot injuries, orthotics do not have any effect on knee alignment, and although they can alter subtalar joint align­ ment, the clinical benefit of this remains unclear.20 Despite the improvements in rearfoot control offered by current athletic shoewear, it has been difficult to relate this to any evidence for reduction in incidence of running injuries.

Cleats and Sole Modifications Traction is a crucial ingredient in efficient performance in all sports and is related to the design features of ath­ letic shoewear. It can also be influenced by the playing ­surface and by the weight of the participant. Factors to be considered in the analysis of traction related to shoewear include the outsole materials used, the sole pattern, and the presence of cleats as well as their size and configuration. Load related to traction is studied primarily with regard to torque and friction.138,231,238,333,344 Outsoles are the bottom-most layer of the shoe that has direct contact with the ground. The most frequently used materials are carbon rubber, styrene butadiene rubber, ethylene vinyl acetate, and polyurethane.345,346 In general, if the material is harder, its durability is greater. Traction considerations must encompass the interaction of these materials with a variety of playing surfaces and conditions,

including the artificial surfaces used indoors and outdoors as well as the amount of moisture or dust on the field or court.347 Major differences occur in the coefficients of static and sliding friction for various shoe-surface ­combinations (Table 25J-7).348 Torque is the component of traction that is measured as the tendency of a force to rotate an object around an axis. It correlates with the static coefficient of friction. Laboratory experiments have established large differences in torque between different shoe-surface combinations, as illustrated in Figure 25J-20.348 Rheinstein and colleagues performed similar experiments supporting this difference in torque as related to outsole material, outsole hardeners, player weight, and playing surface.347 They found greater maximal torques with the softer outsoles combined with artificial and clean hardwood flooring in heavier players. Rubber-soled shoes demonstrated much more sensitivity to dust than the polyurethane soles when analyzed for loss of traction. This research has considerable implications in sports medicine because higher torque means higher load transmission to the body and an increased potential for injury. Alternatively, lack of traction implies sliding, slipping, falling, and poor performance. Where is the proper balance? This question may be unanswerable but deserves attention. Outersole tread design has developed from the flat rub­ ber of the traditional canvas tennis shoe of yesteryear to the high-technology multiple-patterned sole we see in running shoes and court shoes today (Fig. 25J-21).345,349 These tread designs theoretically alter the mechanical properties of the shoe by enhancing performance or pre­ venting injury by determining the flexion path for the shoe, the proper break point, and the pivot point, as well as by affecting traction and shock absorption.346 The tread design has been used in marketing certain running shoes, including the original Nike waffle sole (Fig. 25J-22). The tread design can clearly alter the traction characteristics of the shoe, but its relationship to foot function and injury prevention remains unproved.

  TABLE 25J-7  Friction Coefficients for Several Floor Combinations Tested under Laboratory Conditions Surface Shoe

Carpet

Synthetic Granular

PVC

Sand

Asphalt

1.05-1.15

0.95-1.05

1.00-1.20

0.40-0.60

0.70-0.80

0.95-1.05

0.80-0.95

0.80-0.90

0.30-0.55

0.60-0.75

0.50-0.60

0.75-0.90

0.40-0.50

0.30-0.50

0.65-0.75

1.15-1.25

1.05-1.15

1.00-1.10

0.50-0.60

0.70-0.80

105-1.15

0.95-1.05

0.80-0.90

0.40-0.60

0.70-0.80

0.60-0.70

0.80-0.90

0.40-0.50

0.40-0.50

0.75-0.85

Sliding Friction Coefficients

All-around shoe Little profile All-around shoe Treaded Profile Tennis shoe Indoor No profile Static Friction Coefficients

All-around shoe Little profile Jogging shoe Treaded Profile Tennis shoe Indoor No profile PVC, pol������������������ yvinyl c���������� hloride.

Maximum Torque (Nm)

Foot and Ankle 2191

40 Mean

Standard deviation

30

la y C

t al As ph

ra ss G

r tif gr icia as l s Te nn is

oo In d

Ar

O

ut do

or

20

Surface Figure 25J-20   Mean values and standard deviations for maximal torque for 12 subjects on seven surfaces in eight different shoes. (Redrawn with permission from Nigg BM, Denoth J, Keir B, et al: Load sport shoes and playing surfaces. In Frederick EC [ed]: Sport Shoes and Playing Surfaces. Champaign, Ill, Human Kinetics Publishers, 1984, p 12. © Nike, Inc.)

Efficiency of movement is an important factor in ath­ letic performance, and proper traction ensures that internal forces generated by the body’s muscles are efficiently con­ verted into movement. A natural byproduct of this relation­ ship was the introduction of cleats to the outsole of the shoe to improve traction. Introduced about 100 years ago for sports, they have been a source of controversy ever since. Rule changes in sports banned pointed cleats and placed size

l­imits on cleats. The polemics continued with the concept that foot fixation was a leading cause of injury of the ankle and knee in sports.346,350,351 Dr. Daniel ­Hanley of Bowdoin College championed this attack on rigid cleating, particu­ larly in the heel area, after he observed the incidence of sig­ nificant noncontact injuries to the knee at Bowdoin.345,351,352 Cleat modifications followed, including plastic heel disks,237 lower profile oval cleats,345 soccer cleats,353 and cleats attached to a rotating turntable (Fig. 25J-23),245,350 as well as cleats with a circular design for artificial grass, the Tanel 360 (Fig. 25J-24).354,355 This design concept theoretically allows pivoting with adequate traction but without the problem of foot fixation. Queen and colleagues examined the effect of different cleat plate configurations on plantar pressure. 356 They noted significant differences in forefoot loading pat­ terns among cleat types, but no definitive conclusions were drawn in regard to injuries. Research has documented the relationship between cleats and sports injuries. Rowe studied the effect of ­different shoes and cleats on knee and ankle injuries in the New York State Public High School Athletic Association ­during the 1967 and 1968 seasons.353 He found a reduction in ­ injuries with the use of a low-cut disk heel shoe com­ pared with low- and high-top shoes with heel cleats and an even more substantial reduction with short soccer-type cleats used by athletes playing on natural grass. Although Rowe’s findings were not subjected to statistical verifica­ tion, ankle injuries were reduced from a high of 77 per 100,000 hours of participation to 34 per 100,000 by ­changing from the low-cut conventional heel to the soccer shoe. Rates of injury with the soccer shoe were lower than those seen with either type of high-top shoe. Torg and Queden­ feld found a similar reduction in ankle injuries from 0.45 per team per game with conventional cleats to 0.23 per team per game with a soccer-type shoe.253 Cameron and Davis found

A

B

C

D

Figure 25J-21   Different tread designs: A, Track shoe. B, Football cleats. C, Baseball spikes. D, Running shoe. (Photograph by Christopher B. Hirose.)

2192 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

A

B

Figure 25J-22   The original Nike Waffle shoe. (© Nike, Inc.: www.nike.com.)

a progressive reduction in ankle injuries from 8.46% with a cleated shoe to 7.69% with a heel plate to 5.64% with a soc­ cer shoe to 3.00% with their swivel shoe design.350 An exhaustive study of injuries in North Carolina high school football players by Mueller and Blyth emphasized the critical role of the playing surface in evaluating the role of cleats.254 They reported a reduction in knee and ankle injuries from 14.8% to 11.5% by changing from tra­ ditional cleats to soccer-type shoes on properly maintained fields. Bonstingl and colleagues concluded that there was a positive relationship between torque and injury.357,358 The highest torque occurred with a conventional foot­ ball shoe, which contained seven ¾-inch high cleats. A substantial reduction in torque occurs in circular pat­ tern outsole design (e.g., the original Tanel 360 shoe). Other studies have confirmed these differences in friction and torque between various outersoles, cleats, and playing

Figure 25J-23   Dr. Bruce Cameron’s swivel shoe alternative cleat design. (Photograph by Thomas O. Clanton.)

surfaces.231,233,344,357,359 This is one of the most obvi­ ous areas in which basic science and clinical research in the field of sports medicine have had an impact on sports equipment designed to prevent injury.

PLAYING SURFACES AND INJURY Of all the etiologic factors involved in foot and ankle inju­ ries, the playing surface may be the most important and, at the same time, the least understood. If it is true that load on the human body is the common ground for discover­ ing clues to the causes of athletic injury, the influence of forces and moments intrinsic to the surface of play must be critically analyzed.348,360 Traditional sports surfaces have been composed of natural materials: wooden basketball courts, clay and grass tennis courts, cinder tracks, grass and dirt baseball diamonds, and natural grass football and soc­ cer fields. With advancing technology, there was a move to replace these with more durable low-maintenance syn­ thetic surfaces. Although such surfaces had some attractive advantages, they were not universally accepted and were subjected to quick criticism that followed their short-lived popularity. Sports surfaces should accomplish three functions: (1) protection, (2) performance, and (3) maintenance.293 The first is concerned with the protection of the ­ athlete from excessive forces and injury. The sports perfor­ mance ­function relates to the optimization of the athletic ­experience through the qualities of the surface. Mainte­ nance refers to the durability and conservation of the sur­ face and preserving the former two qualities. The task of reviewing all sports surfaces is overwhelming, considering that wrestling mats, gymnastic beams, ice-skating rinks, and snow-packed slalom courses could all be included. To

Foot and Ankle 2193

Figure 25J-24   Alternative cleating patterns in the Tanel 360 football shoe. (Photograph courtesy of Tanel Corporation.)

gain some understanding of this broad topic, we divide surfaces into indoor and outdoor surfaces and into natural and synthetic surfaces. There is some overlap between sur­ faces in sports, and multiple uses are the rule in most sports facilities. As the characteristics of a particular surface used in one sport are delineated, the reader is reminded that the surface characteristics are applicable in other sports, but that protective, technical, and performance characteristics may vary among sports, types of athletes, shoewear, and environmental situations.

History Since ancient times, humans have competed athleti­ cally. The surface chosen was that which occurred natu­ rally, most commonly grazed fields and dirt areas freed of rocks and obstacles. As culture progressed, so did the playing fields, and stadiums were created for competition in ancient Greece. Improvements in these outdoor facili­ ties were accomplished by maintaining the fields, using developments in soil and grass technology, crowning to improve drainage, and limiting play to allow the surface to recover. It was not until 1964 that synthetic grass, pro­ duced by Monsanto, was introduced and installed on the playing field at Moses Brown School in Providence, Rhode Island.51,361,362 Developed as a substitute for grass in a place where its natural growth was difficult, this artificial surface created little impact until the 1960s. On April 9, 1965, the “eighth wonder of the world,” the Astrodome, was com­ pleted amid much hype in a city known for space-age tech­ nology: Houston, Texas.363 A special grass was developed to grow inside the Astrodome: Tifway 419 Bermuda.363 Even the foundation soil required a unique design. Unfor­ tunately, when the clear Lucite roof panels were darkened to eliminate glare and lost fly balls, the grass withered, and management scrambled to find a suitable substitute. The

following year, the natural grass was replaced by Monsan­ to’s synthetic grass, and this became known as AstroTurf. Now, grass playing fields in natural and custom-installed varieties compete with synthetic surfaces.364 Entire publi­ cations advise facility managers in the most intricate detail of how to maintain playing fields.32 The debate about which is the better surface has involved both the public and the scientific community, and more than 50 articles have appeared without a summary conclusion.51,57,160,360,365-371 Perhaps the key deficiency in these studies is their poor control of other key factors in the turf equation such as the shoe, field maintenance, and weather. Synthetic grass manufacturers have been forced to make improvements in their fibers, mats, underpads, and drainage systems, and advocates of natural grass have used modern methods to introduce advances of their own. As long as prevention of injury is a prime concern in these advances, both the indi­ vidual athlete and society as a whole will benefit. This same revolutionary process has occurred in other outdoor surface sports, such as tennis and track. Ten­ nis originated from a French handball game called jeu de paume at about the 13th century and was played in court­ yards over a fringed net.372 Major Walter Wingfield of North Wales invented a game in 1873 from which modern outdoor tennis has evolved. Its popularity spread so rapidly that the All England Croquet Club added the name Lawn Tennis to its title and sponsored the first championship in 1877. As the name indicates, grass was the original surface used for play. As the popularity of tennis increased, other playing surfaces were needed to allow winter play and to bypass the problems inherent in growing a playing field. In 1909, Claude Brown introduced the clay court. Although these surfaces remain popular, new court surfaces have flourished to such a degree that modern tennis has the widest choice of playing surface of any major sport (Table 25J-8). Only recently have the surface characteristics of

2194 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

Glare

Initial Cost per Court Including Base* (1992) prices

Maintenance

Avg. Time before Resurfacing

Resurfacing Cost (1992) prices)

Surface Hardness

Fast dry

No

No

14,000-18,000†

Daily and yearly

Annual

1000-3000

Soft

With subirrigation Clay

No

Generally

24,000-26,000 9,000-11,000

Daily and yearly

5 yr

Top dressing 1000-1500

Soft

Grass

No

no

15,000-17,000

Daily and yearly

Indefinite

Varies

Soft

Sand-filled synthetic turf‡

Yes

no

25,000-30,000

Daily and yearly

Indefinite

N/A

Soft

Porous concrete

Yes

yes

27,000-31,000

Minor

Indefinite

N/A

Hard

Court Type

Repairs May Be Costly

  TABLE 25J-8  Chart Comparing Various Tennis Court Surfaces

Porous

Nonporous Noncushioned

Concrete post-tensioned Concrete reinforced

Yes

No (if colored)

22,000-25,000

Very minor

5 yr (if colored)

3000-3500

Hard

Yes

No (if colored)

19,000-21,000

Very minor

5 yr (if colored)

3000-3500

Hard

Asphalt plant mix (colored)

No

No

16,000-18,000

Very minor

5 yr

2500-3000

Hard

Emulsified asphalt mix

No

No

19,000-21,000

Very minor

5 yr

3000-3500

Hard

Asphalt penetration macadam

No

No

15,000-17,000

Very minor

5 yr

3000-3500

Hard

Nonporous Cushioned

Asphalt bound system (colored) Liquid applied synthetic

No

No

19,000-23,000

Very minor

5 yr

2500-3500

Soft

Yes

Possible

30,000-40,000

Very minor

5-10 yr

2500-3500

Soft

Textile§ Modular§ Removable§

No No No

No No No

25,000-28,000 22,000-26,000 25,00-30,000

Very minor Very minor Very minor

Varies Varies Varies

Varies Varies Varies

Soft Soft Soft

*Prices vary regionally, do not include site preparation or fencing, and will be somewhat reduced when building or resurfacing batteries of courts. †Including sprinkler system. ‡Damaged areas may be readily repaired. §Including base construction and structurally sound surface. Reprinted from Tennis Courts 1992-1993 with the permission of the United States Tennis Association, 707 Alexander Road, Princeton, NJ 08540.

Cushioned Surface

Durable

Court Speed Adjustable

Lines Affect Ball Bounce

Yes, slightly

Yes

Yes

Yes

Yes if tapes

No

No‡

Yes

Short if damp court

Yes

Varies

Slow

Yes

Yes

Yes

Yes

No

No‡

Yes

Moderately long

Yes

Green

Slow

Yes

Out only

Yes

Yes

Yes if tapes No

Yes

No

Short

Yes

Green, red

Fast

No

Yes

Yes

Yes

No

Yes

Yes

Short

Yes

Concrete

Fast

No

Out only

No

No

No

No

Hard objects can damage Hard objects can damage Yes

Controllable

Yes

Variety

Fast

No

Yes

No

No

No

No

Yes

Yes

Variety

Fast

No

Yes

No

No

No

No

Yes

Yes

Variety

Fast

No

Yes

No

No

No

No

Yes

Yes

Variety

Fast

No

Yes

No

No

No

No

Yes

Yes

Variety

Fast

No

Yes

No

No

No

No

Yes

Yes

Yes

Variety

Fast

No

Yes

No

No

No

Yes

Yes

Yes

No if glossy finish, yes if gritty finish Yes Yes Yes

Variety

Fast

No

Yes

No

No

No

Yes

Yes

Yes

Variety Green, red Variety

Fast Fast Fast

No No No

In only Yes In only

N/A Yes N/A

No Yes No

No No No

Yes Yes Minor

Yes Yes Yes

No Yes No

Long if glossy finish, medium if gritty finish Long if glossy finish, medium if gritty finish

Long if glossy finish, short if gritty finish Varies shortest to longest Short Medium to short Varies shortest to longest

No if glossy finish

Slide Surface

Fast

Surface OK In & Out

Green, red

Ball Discolored

Drying Time After Rain

Yes

Ball Spin Effective

Short if damp court

Ball Skid Length

Colors

Surface Cool on Hot Day

Foot and Ankle 2195

No

2196 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

these courts been subjected to laboratory and subject tests in an attempt to establish their safety. Track and field is unquestionably the oldest form of organized sports activity, dating back 3500 years to ancient Greece. The religious festival at Olympia featur­ ing this activity became an event of such proportions that the Greeks dated events with reference to the year of the Olympiad.373,374 Records dating back to 776 bc indicate that a race covering 192 m took place on a track about 32 m wide.374 These early contests were held on grass, and later races took place on public thoroughfares. Improvements in track design led to the use of cinders and a mixture of cinders and clay by the late 1800s. Syn­ thetic surfaces began to appear in the 1960s with the pro­ duction of a prototype track by the 3M Corporation for Macalester College in 1965.361 The synthetic track gained wide appeal after its use in the 1968 Olympics in Mexico City, where numerous records were set. Although ath­ letes have long known the difference between “fast tracks” and “slow tracks,” it remained for McMahon and Greene from Harvard to produce a track engineered to optimize running speed.375-378 This was an indoor track that theo­ retically and practically enhanced speed by about 2% (or 5 seconds per mile), and the physics have since been incor­ porated into outdoor track surfacing technology. Although the cinder track has been virtually eliminated from elite competition, its value in reducing injuries remains clear, and some schools have preserved cinder tracks for use in training. Tracks for elite competition are generally made from rubber or synthetic materials such as polyurethane. Corporate research and development departments work­ ing in conjunction with bioengineers at universities in locations as far apart as Calgary, Zurich, Rome, and Boston design these surfaces to maximize performance and reduce injury potential.379 The origin of the movement of sports events from out­ doors to inside can also be traced to ancient Greece. The name gymnasium is derived from the Greek word meaning “school for naked exercise.”380 Exercise was an integral part of Greek society, but the gymnasium also served as a public facility for the training of male athletes to par­ ticipate in the aforementioned public games. From its beginning as a room that served as a gathering place for exercise, the gymnasium grew in proportion to accom­ modate baths, dressing quarters, rooms for specialized purposes, and larger areas for contests and spectators.380 The first modern gymnasium opened in Copenhagen in 1799.381 Pestalozzi, Ling, and Spiess stimulated further development, and groups such as the Young Men’s Chris­ tian Association and the Young Men’s Hebrew Associa­ tion included physical exercise in their activities in the mid-1800s. The international Young Men’s Christian Association training school in Springfield, Massachusetts was the site where James Naismith introduced the new game of basketball to the world in 1891.382 Wooden floors became the traditional gymnasium surface and continued to predominate for basketball courts as well as for other indoor courts and dance studios. Contemporary surfaces have been said to have advantages ranging from improved durability to noise reduction to better use of space, but seldom have their supporters argued for improved safety. In the study of sports surfaces, whether indoor or out­

door, it is evident that many factors affect the safety value of the surface. These include not only the visible surface and the top finish but also, and equally important, the undersurface.

Injury Incidence Although there is consensus that sports surfaces play a crit­ ical role in causing athletic injuries, it has been quite dif­ ficult to establish the incidence of injury in a sport that can be attributed solely to the playing surface. Even in the area of traditional grass football fields, there have been stud­ ies that have suggested differences in incidence of injury from field to field, although the surface material is the same. The 1992 study by Powell and Schootman is one of the best-controlled studies demonstrating that football players were at significantly higher risk for knee ligament injury when playing on AstroTurf as compared with natu­ ral grass.383 Unfortunately, their study does not address injuries outside the knee. A study of North Carolina foot­ ball players conducted by Mueller and Blyth254 showed that a reduction in the injury rate resulted simply from resurfacing and maintaining the game and practice fields. Injury rates plummeted from 29.3% on unresurfaced fields in 1969 to 14.8% on resurfaced fields in 1972, a 30% ­reduction. Poor field conditions were considered a factor in 8 of 34 soccer injuries in the paper by Sullivan,95 in 14 of 18 outdoor soccer injuries in Hoff and Martin’s series,46 and in 62 of 318 injuries in the Swedish study of Ekstrand and Gillquist.91 These studies draw attention to the play­ ing field and its importance and the need for further work on proper ­ playing surfaces and their maintenance to reduce injury rates in sports. FieldTurf was developed to ­duplicate the playing characteristics of natural grass. Mey­ ers and associates studied the differences in injuries of high school football players between the two surfaces and found that the types of injuries were different between the two surfaces. The natural grass surface produced more head and neural traumas and ligamentous injuries. The Field­ Turf produced more injuries during higher temperatures, muscle-related trauma, and epidermal injuries.384 Studies of injuries and injury rates in sports ranging from dance189,385 to ice hockey386,387 to tennis388 have mentioned the sport’s surface as a factor. For example, Pasanen and colleagues studied the injury risk in pivoting indoor sports between artificial floors and wooden floors.389 They found that the risk for a traumatic injury in pivoting indoors sports is two-fold higher on artificial floors when compared with wooden floors, largely due to a higher shoe-surface friction level. Although the surface is frequently named as a source of problems in runners, there have been no studies that have unequivocally confirmed this.28 The Ontario cohort study found no association between running surface and injury, whereas the companion study from South Caro­ lina showed a statistically meaningful relationship only for females running on concrete.228 In baseball, if the base is included as a part of the play­ ing field surface, several studies can be cited indicating that this is the primary factor in ankle injuries in baseball and softball. Janda and coworkers found that 71% of the recreational softball injuries in their study were related to sliding into bases.47 A follow-up to this study showed the

Foot and Ankle 2197

MECHANICAL FACTORS Regardless of the surface, the underlying question to be answered is how the surface affects load in the individual participant.298,348 For this purpose, attention has been focused on certain material properties of the sports sur­ face such as hardness and friction together with perfor­ mance properties such as energy loss or resilience. Nigg has reviewed the methods by which a playing surface may be characterized and specified some of the tests that are important in determining these properties.394

Hardness One of the most obvious differences between surfaces detected by casual observation as well as by sophisticated testing is the relative hardness or softness. Hardness is related to the ground reaction force. In conformity with Newton’s first law (for every action there is an equal and opposite reaction), the vertical reaction force responds to the vertical force component applied by the individual at foot strike. Its amplitude is affected by the shock-absorbing qualities of the surface to which the force is applied. The time needed for force absorption and reaction is the key to the amplitude of the reaction force and relates to the com­ pliance of the surface material. Hard materials deform less than soft materials during identical impact testing condi­ tions: a well-maintained grass lawn is softer than a concrete sidewalk. When impact occurs between an object such as a leg and a surface such as grass or concrete, it is the surface hardness that limits the time of impact while increasing the amplitude of the reaction force (Fig. 25J-25). Kerdok and colleagues built experimental platforms with adjustable stiffness to examine the leg stiffness and metabolic cost of the athlete.395 The 12.5-fold decrease in surface stiffness resulted in a 12% decrease in the runner’s metabolic rate

Surface A (Hard)

Reaction force N

implications of an injury-prevention method such as the use of a breakaway base on these surface-related injuries.48 Janda and associates estimated that breakaway bases could reduce the incidence of serious injury in softball by 96% and result in a $2 billion per year savings in acute medical care costs.390 In the area of children’s playground injuries, statis­ tics suggest that falls to the ground account for 60% of playground equipment injuries, yet barely half of day care playground equipment is installed on impact-absorbing surfaces.246,391 Impact studies of five types of loose-fill playground surfaces at a variety of drop heights, mate­ rial depths, and conditions suggested that shredded rub­ ber was the best performer, and there was little difference between sand, wood fibers, and wood chips; and pea gravel had the worst performance, making it a poor choice for playground surfacing.392 A separate study supported these findings: children sustained significantly more inju­ ries in playgrounds with concrete surfaces than in those with bark or rubberized surfaces. Playgrounds with rub­ ber ­surfaces had the lowest rate of injury, with a risk half that of bark and a fifth of that of concrete. Rubberized impact-­absorbing surfaces are safer than bark.393 Certainly the playing surface is an important consideration in the prevention of injury.

Surface B (Soft)

Time of impact msec Figure 25J-25   Example of relationship between time of impact and type of surface. Hard surface (A) has a short time of impact but a high reaction force, whereas soft surface (B) has a longer time of impact and a lower reaction force.

and a 29% increase in their leg stiffness. They concluded that an increased energy rebound from the ­ compliant ­surfaces studied contributes to the enhanced running economy. It takes little imagination to determine that sports participation on a soft surface such as a gymnastics mat carries less risk for certain impact-related injuries than a similar activity performed on a wooden floor. By the same token, athletes running on shock-absorbing gymnastics mats compared with a tuned track would set few records.378 From these mundane examples, we can conclude that there are some opposing factors confronting us in this analysis. Although a harder surface may provide a better surface for performance needs in certain sports, it may create simulta­ neously an increased exposure to injury. This fact appears to be borne out most obviously by synthetic tracks. Hardness is the resistance generated by a material during deformation in response to an externally applied force.348,396 In the study of various materials, their behav­ ior is described by means of a stress-strain curve or stressdeformation diagram. The plastic or elastic behavior of a material is determined by the remaining deformation once the acting force is removed. When the deformation persists, the material is described as plastic, and when the material returns to its original shape, it is elastic. A sports surface may be further characterized as being either area elastic or point elastic.397 Owing to their high bending strength, area elastic surfaces distribute forces over a wide area. Point elastic surfaces have low bending strength and therefore deform only in a very confined area (Fig. 25J-26). This fact has implications for both performance and health. From the clinical standpoint, there is a widespread belief that harder surfaces are associated with a higher incidence of injury for any given activity. This conclusion seems obvious, but it is difficult to support scientifically. Such conditions as shin splints, stress fractures, tibial stress syndrome, turf toe, bursitis, arthritis, and even acute frac­ tures have been associated with the higher loads imparted by surfaces with limited compliance. Bowers and Martin

2198 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

producing initial decelerations as high as 80 G force (G) is a primary etiologic factor.400 The relationship between surface hardness and stress fractures has been mentioned in several studies.72,159,188,280-282,401 With this as background, it is apparent that surface hardness is important in the study of athletic injury, although its relative contribution among the interrelated factors is imprecise.

Friction

Point Elasticity

Area Elasticity

Figure 25J-26   Example of point elastic surface and area elastic surface. (Redrawn with permission from Denoth J: Indoor athletic playing surfaces—floor vs. shoe. In Segesser B, Pförringer W [eds]: The Shoe in Sport. Chicago, Year Book Medical Publishers, 1989, pp 65-69.)

demonstrated the existence of reduced shock-absorbing characteristics in 5-year-old AstroTurf compared with new AstroTurf, describing this as “clearly detrimental to player safety.”398 Unfortunately, they did not show a clear relationship between this lack of impact absorption and an increase in either acute or chronic injuries. Larson and Osternig399 reported one of the few clinical studies implicating surface hardness as a source of specific athletic injury in 1974. In a survey of Pacific-8 Conference ath­ letic trainers after the 1973 football season, they showed that the incidence of prepatellar and olecranon bursitis was increased on artificial grass compared with natural grass and attributed this increase to the hard underlying subbase. Anecdotally, considerable evidence of problems with harder playing surfaces exists because players commonly complain of aching feet and legs after standing and practic­ ing on older synthetic fields. The injury most frequently associated with sports par­ ticipation on artificial grass is turf toe.179,181,234 This injury is a sprain of the first metatarsophalangeal joint that has been inextricably linked to the artificial playing surface. Turf toe is a distinct clinical entity related to the combina­ tion of a relatively flexible shoe and a hard artificial sur­ face.234 Despite the weight of clinical evidence pointing to the relationship between turf toe and the artificial surface, little statistical support exists implicating surface hardness as a major factor. The injury does indeed occur on natural grass and probably has more to do with the flexibility of the shoe and frictional characteristics of the surface than its hardness. Impact forces, skeletal transient forces, and excess load produced by hard surfaces have been emphasized as etiologic factors in many other conditions ranging from osteoarthritis to shin splints to stress fractures. Statistical verification of this relationship, however, remains absent. An increased incidence of tibial periostitis, Achilles tendi­ nitis, Achilles tendon rupture, and muscle rupture are pos­ tulated to be the result of hard surface synthetic tracks.53 Haberl and Prokop have related these conditions to what they call Tartan syndrome and propose that surface hardness

Without friction, human locomotion would be impossible. The frictional properties of the surface are the second crit­ ical mechanical factor of sport surfaces related to sports performance and injury. Whereas hardness is defined by a high vertical stiffness, friction relates to horizontal stiff­ ness. The static coefficient of friction (μ) is the inherent property of the two contacting materials. It can be calcu­ lated from the equation μ = F/W, where W is the weight of the object being moved over a surface and F is the force required to move the object.344,402 This equation applies to smooth, uniform surfaces, but not to the shoe-turf inter­ face. Therefore, a similar term has been described as the release coefficient (r). It is expressed as r = F/W, where W is the weight in the shoe and F is the force necessary to release the shoe-turf interface when engaged.344 The difficulties imposed by reduced friction are well known to the novice ice skater. High friction between the shoe and the surface translates into good traction for the athlete and improved performance. Unfortunately, this factor also means greater load on the body, which may exceed physi­ ologic limits. Therefore, a trade-off exists between perfor­ mance-enhancing qualities and safety considerations. Friction can be viewed in several ways that are impor­ tant to sports biomechanics. There are two types of fric­ tion to be considered: static and kinetic.403 Static friction is the resistance to movement between two objects that are not moving relative to each other. It is a surface property of the contact surfaces. Kinetic friction occurs when two objects are moving relative to each other, rub together, and typically slow down one or both of the objects. Friction can also be viewed in terms of horizontal or rotational fric­ tion.348,393,403 Functionally, the former corresponds to the force resisting the foot sliding or moving sideways, whereas rotational friction relates to torque generated in activities such as turning. Torque is considered one of the primary etiologic factors in injuries to the knee and ankle.231,357

Energy Loss A separate but equally important property apart from hardness is the energy loss that occurs when a material is loaded. This property varies over a wide range from sur­ face to surface and carries major implications in the field of sports biomechanics. As depicted in Figure 25J-27, materials can exhibit similar elastic behaviors on the stress and strain diagram and yet have very different responses to the effects of loading and unloading. In this figure, material A shows no loss of mechanical energy, whereas material B depicts a loss of mechanical energy equivalent to the shaded area. This energy loss is not a true mate­ rial property because it depends on the loading rate and other variables. When a material has the quality of ­variable

Foot and Ankle 2199

Material 2

Stress

Stress

Material 1

Deformation

Figure 25J-27   Differences in response of materials to loading and unloading despite similar elastic behavior. (Redrawn with permission from Denoth J: Load on the locomotor system and modelling. In Nigg BM [ed]: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics Publishers, 1986, p 96. © 1986 by Benno M. Nigg.)

deformation dependent on velocity of deformation, it is described as viscoelastic. If energy applied to a surface is lost in the surface deformation, performance may suffer.378 Surfaces that deform to greater degrees are called compliant and result in increased contact time. This is the means by which cushioning occurs. By increasing the time of colli­ sion, the force between the colliding bodies is decreased. The final property of importance in athletic performance on a given surface is its resilience. High resilience indicates that the energy stored in the surface owing to its stiffness is returned to the athlete. This has implications for enhanc­ ing performance as well as lessening fatigue.378,397

Experimental Work Because of a presumed association with injuries, fric­ tional properties of various sport surfaces and athletic shoes have been the subject of studies by several authors. Torg and Quedenfeld published pioneering work in the field of sports medicine in 1971 aimed at reducing injury rates by targeting the shoe-surface interface.253 The study was done on grass and concentrated on the relationship between the number and size of shoe cleats and the inci­ dence and severity of knee injuries in high school football players. This investigation showed a reduction in ankle injuries from 72 to 36 and in ankle fractures from 13 to 7 by changing from a traditional seven-cleated football shoe to a multicleated soccer shoe. Continuing this study, Torg and colleagues performed laboratory studies to determine the torque necessary to release an engaged shoe-surface interface.236,237 Twelve shoes and nine surface condi­ tions were tested for a total of 108 release coefficients. These are shown in Tables 25J-9 and 25J-10. Coefficient differences of 0.05 were determined to be significant. The release coefficients ranged from a high of 0.55 ± 0.06, with a conventional seven-cleated football shoe on dry grass, to a low of 0.20 ± 0.02, with a conventional shoe that had an uncleated disk heel (Bowdoin modification) on dry Polyturf. The use of wet versus dry conditions was based on the study by Bramwell and colleagues in 1972, which

s­ uggested that fewer injuries were sustained on wet syn­ thetic fields than dry synthetic fields.404 From their study, Torg and coworkers concluded that the release coefficient varies with (1) the number, length, and diameter of the cleats; (2) the type of surface—natural or artificial; (3) the condition of the surface—wet or dry; and (4) the outsole material of the shoe—polyurethane or soft rubber. They classified shoes as safe for a particular surface when the release coefficient was 0.31 or below. Cawley and colleagues recently studied nine shoes by three manufacturers, which were characterized as turf, court, molded cleat, or traditional cleat and tested on both natural grass and synthetic turf.405 They found that the cleated shoes (both traditional and molded) generated the highest frictional and torsional resistance on the grass surface when compared with the other categories of shoes. Grass generated higher peak moments than turf for the cleated shoes. These results demonstrate the considerable differences between laboratory and physiologic conditions and that the increase in frictional resistance is nonlinear with increasing loads. Stanitski and associates determined the static coefficient of 16 different shoe-surface combinations.402 They used a drag test of a size 13 shoe with a 25-pound load pulled both with and against the grain, across various sections of football fields. Their results are shown in Table 25J-11. The coefficient of friction ranged from a high of l.54 for a Riddell At-31 (standard last, leather upper, molded plas­ tic sole, 203⁄8-inch conical cleats) on Polyturf to a low of 0.92 for the Puma 1430 (“soft last,” molded rubber sole, 233⁄8-inch cylindrical cleats with central indentations) on Tartan Turf and grass fields. These investigators found no grain effect and “essentially no change” when the surface was wet. Bowers and Martin continued the study of cleat-surface friction, adding the new parameter of a new versus a worn synthetic surface.233 The cleats from two different shoes were studied. One cleat was evaluated in both a slightly worn and a very worn state. Three similar shoes were mounted in a triangular pattern on a platform weighted from 2 to

2200 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25J-9  Results of Torque Testing in 12 Shoes for Nine Surface Conditions Release Coefficients Shoe

Grass

Grass Wet

AstroTurf

Astroturf Wet

Tartan Turf

Tartan Turf Wet

Poly Turf

Group I 0.55 ± 0.06 0.35 ± 0.03 0.34 ± 0.03 0.29 ± 0.3 Group II 0.44 ± 0.04 0.31 ± 0.03 0.32 ± 0.02 0.26 ± 0.02 Group III 0.37 ± 0.04 0.26 ± 0.02 0.26 ± 0.03 0.20 ± 0.02 Group IV 0.36 ± 0.03 0.32 ± 0.04 0.41 ± 0.03 0.26 ± 0.03 0.34 ± 0.03 0.23 ± 0.02 0.33 ± 0.02 Group V 0.28 ± 0.03 0.27 ± 0.03 0.29 ± 0.03 0.29 ± 0.03 0.27 ± 0.03 0.24 ± 0.02 0.26 ± 0.02 Group VI 0.40 ± 0.01 0.36 ± 01 0.41 ± 0.02 Group VII 0.45 ± 0.04 0.37 ± 0.02 0.45 ± 0.02 Shoes in groups I-V have plastic or polyurethane soles. Shoes in groups VI-VII have rubber soles. Group I: Conventional 7-posted football shoe, ¾-inch cleat length, 3⁄8-inch tip diameter, plastic sole Group II: Conventional 7-posted football shoe, ½-inch cleat length, 3⁄8-inch tip diameter, polyurethane sole Group III: Conventional shoe with five ¾-inch cleats, 3⁄8-inch tip diameter, Bowdoin heel, polyurethane sole Group IV: Soccer style, 15 cleats with 3⁄8-inch tip diameter, polyurethane sole Group V: Soccer style, 15 cleats with ½-inch tip diameter, polyurethane sole Group VI: Soccer style, 12 cleats (ten 3⁄8-inch length and two ½-inch length), ½-inch tip diameter, rubber sole Group VII: Soccer style, 49 to 121 cleats (3⁄8-inch or 5⁄16 -inch length), ½-inch tip diameter to pointed tips, rubber sole

Poly Turf Wet

0.23 ± 0.02 0.23 ± 0.02

From Torg JS, Quedenfeld TC, Landau S: The shoe-surface interface and its relationship to football knee injuries. J Sports Med 2:261-269, 1974.

14 pounds that was pulled across the new or worn turf. This study showed 16% more friction against the grain of a 5-year-old AstroTurf field when wet and 22% more fric­ tion when dry (Fig. 25J-28). From this study, the authors provided a formula for calculating coefficients of friction for individual shoes and surfaces that increased linearly with the number of cleats. They suggested that changes in friction altered player performance (e.g., smaller slip angles on wet turf) and that increasing friction could pro­ duce “foot lock” and result in increased injuries. Bonstingl and coworkers expanded the study of shoes and surfaces by looking at dynamic torque for 11 shoe types on four dry turf samples at two different player weights.357 They used the swivel shoe of Cameron’s design, five styles of multicleated soccer shoes, four styles of noncleated basket­ ball shoes, and a conventional football shoe for grass (seven ¾-inch plastic screw-on cleats with metal tips). The four surfaces were AstroTurf, Tartan Turf, Polyturf, and grass. The two player weights used for the normal force were 170 pounds (77 kg) and 200 pounds (91 kg). All combinations were tested for both toe stance and foot stance positions. They found that all shoes except the swivel shoe developed about 70% more torque in foot stance than in toe stance

and that the higher player weight resulted in more torque. The conventional shoe tested on grass had among the high­ est torque. Although noncleated shoes generally had less torque for all playing surfaces, this was not absolutely true for all surfaces. This study proved that torque applied to an athlete’s leg depends on (1) the type and outsole design of the shoe, (2) the playing surface, (3) the player weight being supported, and (4) the foot stance assumed. Culpepper and Niemann continued the study of the shoe-turf interface in 1983 by looking at the release coef­ ficient for torque in several shoe-surface combinations.344 They tested five different soccer-style shoes of variable cleat number and configurations on old and new Polyturf and new AstroTurf under wet and dry conditions. Loads rang­ ing from 10 to 90 pounds were transferred down a metal shaft to a prosthetic foot on which the different shoes were mounted (Fig. 25J-29). The release coefficient of torque was calculated for 30 different conditions (Table 25J-12). The authors found that although a specific shoe did vary in its release coefficient ranking on different surfaces, a shoe that had a low coefficient on one surface under one condition was generally lower on all three surfaces whether wet or dry, whereas a shoe that ranked higher did so for all conditions.

  TABLE 25J-10  Relationship of Shoe-Turf Interface

Release Coefficient to Incidence of Football Knee Injuries Release Coefficient

0.60— 0.50— 0.40— 0.30— 0.20— 0.10—

Not safe Probably safe Probably safe Safe

  TABLE 25J-11  Friction Coefficients for Various Surfaces and Shoes

0.49

0.31

From Torg JS, Quedenfeld TC, Landau S: The shoe-surface interface and its relationship to football knee injuries. J Sports Med 2:261-269, 1974.

Coefficient of Sliding Friction Surface Shoe

Poly Turf

Astro Turf

Tartan Turf

Grass

A B C D

1.49 1.54 1.33 1.38

1.34 1.31 1.23 1.16

1.42 1.16 1.13 0.92

1.23 1.21 1.07 0.92

From Stanitski CL, McMaster JH, Ferguson RJ: Synthetic turf and grass: A comparative study. J Sports Med 2:22-26, 1974.

Foot and Ankle 2201

A

B

C

D

Figure 25J-28   AstroTurf evolution. A, Original AstroTurf (new) in its 1966 design with long nylon ribbon. B, Original AstroTurf showing its grain effect. C, Newer AstroTurf 8. D, Weave pattern of AstroTurf designed to prevent a grain effect. (Photographs by Thomas O. Clanton.)

Another study of both static friction and torque appeared in 1983, when Van Gheluwe and colleagues tested nine different shoes on three varieties of artificial turf in both the toe stance and foot stance positions.238 In their analy­ sis, the authors demonstrated higher values for friction and torque for AstroTurf compared with the other surfaces (AstroTurf scored highest in 22 of 36, or 61%, of the test conditions). They attributed this result to the nylon fiber of AstroTurf compared with the polypropylene fiber used in the other surfaces. This confirmed the work of Bonst­ ingl and colleagues; however, the results violate the law of physics for dry friction, which predicts the linear relation­ ship for friction between two surfaces. This contradiction is significant when interpreting the results of other studies that have assumed this linear relationship. Andreasson and coworkers in Sweden published a more detailed study in 1986 evaluating torque and friction in a dynamic mode on an artificial surface.231 This was the first study to specifically add dynamic torque by using a rotat­ ing disk on which the surface was applied to simulate speed from walking to running. Twenty-five different shoes were tested, including running shoes, tennis shoes, soccer shoes for artificial turf, and soccer shoes for natural turf. Torque in toe stance varied from a low of less than 10 Nm for one of the running shoes and one of the ­ multicleated soccer shoes to a high of more than 50 Nm for another multi­ cleated soccer shoe placed against the grain of the Poligrass

Artificial leg

Surface being tested

Rotation

Force plate Figure 25J-29   Example of device for testing torque. (Redrawn with permission from Physical Tests. Sport Research Review. Beaverton, Ore, Nike Sport Research Laboratory, Jan/Feb, 1990; and Van Gheluwe B, Deporte E, Hebbelinck M: Frictional forces and torques of soccer shoes on artificial turf. In Nigg BM, Kerr BA [eds]: Biomechanical Aspects of Sport Shoes and Playing Surfaces. Calgary, University of Calgary Press, 1983, pp 161-168.)

2202 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������

  TABLE 25J-12  Release Coefficients of 30 Shoe-Surface Combinations New Poly Turf Dry

Wet

0.27 ± 0.01 0.35 ± 0.03 0.21 ± 0.01 0.32 ± 0.03 0.28 ± 0.01

0.21 ± 0.01 0.29 ± 0.01 0.23 ± 0.01 0.24 ± 0.01 0.29 ± 0.01

Shoe A B C D E

Old Poly Turf Dry

New Astro Turf Wet

0.24 ± 0.02 0.33 ± 0.05 0.29 ± 0.05 0.32 ± 0.02 0.30 ± 0.01

Dry

0.24 ± 0.02 0.31 ± 0.01 0.00 ± 0.01 0.31 ± 0.01 0.25 ± 0.02

Wet

0.19 ± 0.01 0.38 ± 0.03 0.19 ± 0.01 0.34 ± 0.03 0.26 ± 0.01

0.24 ± 0.01 0.35 ± 0.02 0.22 ± 0.02 0.26 ± 0.01 0.28 ± 0.01

Date from Culpepper MI, Niemann KMW: An investigation of the shoe-turf interface using different types of shoes on Poly-turf and Astro-turf: Torque and release coefficients. Alabama J Med Sci 20:387-390, 1983.

test surface. The authors showed that the frictional force is independent of speed between 1 and 5 m/second, but this is below the 7 to 10 m/second speeds that occur in modern football and soccer. Torque was generally lower for shoes with polypropylene outsoles compared with polyurethane and rubber-like soles. Nigg and Segesser reported in 1988 on the variation in friction related to a change in load.360 They found when studying the six different surfaces considered for the Toronto SkyDome playing surface that the static coef­ ficient of friction changed from lows of 1.13 for 280 N normal loads to highs of 3.48 for 769 N loads (Table 25J13).360 The tested surfaces ranked differently for the two load conditions. This result indicates the complexity of using material tests for choosing a playing surface because the individuals playing vary by a factor of 2 or more in weight and generate forces that may be well beyond those studied to date in laboratory tests.400 Just as there are differences between natural and artifi­ cial grass that affect the frictional properties of the play­ ing surface, similar differences have been known for other sport surfaces. Rheinstein and colleagues looked at static drag and dynamic torque characteristics of different shoes on two basketball surfaces (Fig. 25J-30).347 The three playing surfaces tested were clean hardwood, dusty hard­ wood, and clean artificial flooring (Tartan indoor surface manufactured by the 3M Corporation). Polyurethane soles produce less torque than the elastomer outsoles for all weight, floor, and surface condition parameters. Torque also decreased with increasing sole hardness for the elasto­ mer outsoles on the clean hardwood and artificial surface. As expected, greater player weight increased torque, and dust on the hardwood flooring cut torque almost in half. The polyurethane soles were considerably less affected by dust than the elastomer outsoles. One would expect that high friction and torque could overload the athlete and

result in injury. These findings have obvious implications for performance as well as prevention of injury in sports medicine. Nigg and associates measured the coefficients of static and sliding friction in 1980 for five different surfaces and three styles of shoewear.348 The dynamic, or sliding, ­coefficient of friction ranged from a low of 0.3 on sand (clay) with an indoor tennis shoe or a treaded jogging shoe to a high of 1.20 on a polyvinylchloride floor with an allaround shoe (see Table 25J-7). Nigg expanded on this work in 1986 and has used these studies to emphasize the dominant role of the playing surface on the translational friction coefficient, although he acknowledged the open question of whether the structure or material, or both, are most responsible for the result.218 In a separate study, Nigg and colleagues measured torque for 12 subjects rotating 180 degrees on one leg on seven different sport surfaces and with eight different types of shoes.303 Mean values for the different surfaces are shown in Figure 25J-20. The range varied from 20 Nm to 38 Nm, the highest torque being found on artificial grass. In an expanded study using average torque for five tested surfaces (10 Nm to 20 Nm) and average torque for 10 tested shoes (13 Nm to 18 Nm), Nigg proposed that torque was shoe and surface codependent to a greater degree than translational friction.218 In attempting to ­correlate the material tests for translational friction with the subject tests for rotational friction, Yeadon and Nigg found no relationship.397 This study demonstrates the dif­ ficulty of combining material and subject tests, as occurs also in tests of impact load, in which the subject’s response to a test condition may alter the result.394 Synthetic playing surfaces with rubber or sand infill are now used on many athletic fields such as soccer, football and rugby. Although these surfaces may come closer to the mechanical characteristics of a true grass playing surface

  TABLE 25J-13  Variation in Static Coefficient of Friction for Translation with Variation in Vertical Load Using Football Shoes on Six Different Playing Surfaces (A through F)

Static Coefficient of Friction (Translational) Load (Vertical) (N)

A

B

C

D

E

F

280 769

1.13 3.15

1.42 2.90

1.30 2.57

1.30 2.57

1.56 3.48

1.51 3.15

From Nigg BM, Segesser B: The influence of playing surfaces on the load on the locomotor system and on football and tennis injuries. Sports Med 5:375-385, 1988.

Foot and Ankle 2203 a Ø (3.00 ± 0.10) mm b Ø (1.25 ± 0.15) mm 0.04 c 2.50 + – 0.01 mm d Ø (0.79 ± 0.01) mm r Ø (0.10 ± 0.01) mm f Ø (16.0 ± 2.50) mm

(

Shore A a b

c

)

Shore D a b

30° ± 1°

c

35° ± 0.25°

r d f

f

Figure 25J-30   Methods for measuring shore hardness. (Redrawn with permission from Denoth J: Load on the locomotor system and modelling. In Nigg BM [ed]: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics Publishers, 1986, p 97. © 1986 by Benno M. Nigg.)

than the older turf designs, their potential effects on lower extremity biomechanics and related injury rates necessitate further study. With the continued introduction of different playing surfaces, the relentless study the shoe-surface interaction can only help improve athlete safety. It does, however, raise some interesting questions. Does this put schools that can­ not afford multiple types of player surfaces at a measurable disadvantage? Does it put athletes at greater risk for injury when they are not afforded the best playing surface for a given game condition? Does the use of a shoe on surfaces for which it is not designed carry liability exposure for the school? Does there exist a shoe or surface that is “the best” for a particular sport?

Clinical Relevance In 1980, Nigg and associates reported the results of a ret­ rospective study of tennis injuries using a questionnaire.348 Using one tennis player during one 6-month season as a sin­ gle case, they analyzed 2481 cases to determine the relation­ ship between injury and playing surface. The foot and ankle were the most commonly affected area when ankle joint, Achilles tendon, heel, and sole cases were combined (Fig. 25J-31). The authors factored in the variation and frequency of play for the different surfaces by determining the relative frequency of pain per hour per week (Table 25J-14).360 Their data showed that the lowest frequency of injury occurred on clay and synthetic clay surfaces, and a lower frequency

Number of Cases with Pain

400

300

200

100

0

Back

Groin

Hip

Knee

Calf

Ankle Achilles Heel Joint Tendon Area of Involvement

Sole

Other

Figure 25J-31   Number of cases of tennis players with back and lower extremity pain by area of involvement. (Redrawn with permission from Nigg BM, Segesser B: The influence of playing surfaces on the load on the locomotor system and on football and tennis injuries. Sports Med 5:375-385, 1988.)

2204 DeLee & Drez’s� O �rthopaedic ����������� S �ports ������ � Medicine �������   TABLE 25J-14  Frequency and Relative Frequency of Pain in Tennis Players Related to Six Different Playing Surfaces Surface

Frequency of Pain (%)

Relative Frequency (%/hr/wk)

Clay Synthetic sand Synthetic Surface Asphalt Felt carpet Synthetic grill

2.2 3.0 10.7 14.5 14.8 18.0

0.5 1.6 3.0 3.9 4.8 3.8

From Nigg BM, Segesser B: The influence of playing surfaces on the load on the locomotor system and on football and tennis injuries. Sports Med 5:75-385, 1988.

of pain occurred on asphalt or concrete than on carpet or synthetic grill. Using this information, Nigg and Segesser speculated that the compliance of a surface is less important in tennis injuries than its frictional properties.360 Studies of the frictional properties of track surfaces have also yielded interesting insight on performance and injury. Stucke and colleagues pointed out the importance of static friction coefficients for starting efficiency in running events.403 When the static coefficient of friction is small, shorter steps and less body lean are used. The authors mea­ sured the static coefficient of friction for starting, stopping, and turning during 100 trials using five subjects wearing the same shoe type. They found that the cinder track had an intermediate value between 0.65 and l.72, compared with the synthetic outdoor surfaces’ values of 0.8 to 2.22 and the synthetic indoor surface values of 0.54 to 1.47 (Fig. 25J-32). This study emphasizes the variability between surfaces by pointing out that an artificial surface does not automatically indicate a surface with greater frictional stresses. Automatic changes in movement technique are influenced by the variation in surface friction properties. The authors speculated that the use of surfaces of ­varying frictional properties during training and competition is disadvantageous because of the time necessary to perfect a repertoire of movement skills to meet the requirements

1.0

Stopping

0.6

µSTAT

0.4

0.2

0.2 A

2.5

Starting

B Cinder Ground Artificial Surface Surface

0

Turning

�STAT

2.0

0.6

0.4

0

This discussion of the etiologic factors involved in the foot and ankle injuries should introduce the concepts nec­ essary to understand the injuries discussed in the other sections in this chapter. It is only after one understands the underlying causes of a problem that solutions are ­forthcoming. In the foot and ankle, as in no other area of the body, there is a direct interaction between anatomy and the environment—between the foot and ankle and

0.8

µSTAT

µ*DYN

SUMMARY

�[cm]

µ

0.8

µSTAT

1.0

of different sport surfaces. This conclusion begs the ques­ tion of whether it is preferable to train on one surface (e.g., natural grass) to reduce injury exposure when contests will be held on a different surface (e.g., artificial grass), or to train on the same surface on which the contest will be ­conducted with scheduled training in a sequence of gradu­ ated stress to allow proper adaptation by the body. When frictional properties of a surface are too low, slip­ ping can occur, and injuries may result.360 When the fric­ tional resistance is too high, the load transference to the body may exceed its range of tolerance, resulting in injury. An optimal range between excessive frictional overload and lack of traction exists to prevent injury. Nigg has suggested optimal ranges for the coefficient of translational friction for various sports.360 He based his recommendations on both objective and subjective assessments, and the range was always between 0.5 and 0.7. Stussi and colleagues calculated the coefficients of static friction to be 0.6 and sliding friction to be 0.5 for clay; values on fabric courts approached 1.0.305 Ankle strain is reduced, according to Stussi and colleagues, during braking maneuvers on sandy courts. They acknowl­ edged that greater performance demands might require increasing coefficients of friction and suggested that more stable shoewear (e.g., with improved ankle support) might allow the player to tolerate the greater strain of the higher friction surfaces. Owing to the epidemiologic flaws in the study of tennis injuries by Nigg and coworkers,360 the final answer on the relationship between the frictional properties of playing surfaces and athletic injuries has not been deter­ mined, but the stage is certainly set for such a study.

1.0

A

B Cinder Ground Artificial Surface Surface

0

A B Cinder Ground Artificial Surface Surface

Figure 25J-32   Variation in coefficient of friction between three different track surfaces for starting, stopping, and turning.   μ, coefficient of friction for translation; η, coefficient of friction for rotation; DYN, dynamic; STAT, static. (Redrawn with permission from Stucke H, Baudzus W, Baumann W: On friction characteristics of playing surfaces. In Frederick EC [ed]: Sport Shoes and Playing Surfaces. Champaign, Ill, Human Kinetics Publishers, 1984, pp 91-96. © Nike, Inc.)

Foot and Ankle 2205

s­ hoewear, and between shoewear and the playing surface. As the reader investigates the specific injuries and patho­ logic conditions that beset athletes in sports, he or she should keep in mind the individual nature of these injuries and their potential risk factors. When causes are discov­ ered, prevention is only a step behind. C

r i t i c a l

P

o i n t s

l It is our job as physicians to act as educators for coaches, therapists, and athletic trainers so that they understand the benefit of returning an athlete to sport participation at the appropriate time after injury. Likewise, we should point out the risks involved when criteria for return to competition are ignored. l Warming up and stretching to obtain or maintain this range may or may not prevent injury, but stretching beyond this range is potentially harmful. l The role of flexibility in foot and ankle injuries is unclear. l A lack of cushioning may be a factor in producing injury but probably not to the degree to which some researchers and shoe manufacturers might lead us to believe. l There is more evidence supporting the link between shoe­ wear control and injury incidence than that of shoewear cushioning. l The compliance of a surface may be less important in injuries than the frictional properties of a surface. l The optimal ranges for the coefficient of translational fric­ tion for a playing surfaces is likely between 0.5 and 0.7.

S U G G E S T E D

R E A D I N G S

Botrè F, Pavan A: Enhancement drugs and the athlete. Neurol Clin 26:149-167, 2008. Caplan A, Carlson B, Faulkner J, et al: Skeletal muscle. In Woo SL-Y, Buckwalter JA (eds): Injury and Repair of the Musculoskeletal Soft Tissues. Park Ridge, Ill, American Academy of Orthopaedic Surgeons, 1988, pp 213-291. Clanton TO: Athletic injuries to the soft tissues of the foot and ankle. In Coughlin MJ, Mann RA (eds): Surgery of the Foot and Ankle, 7th ed. St Louis, Mosby, 1999. Divert C, Mornieux G, Baur H, et al: Mechanical comparison of barefoot and shod running. Int J Sports Med 26:593-598, 2005. Hamilton WG: Foot and ankle injuries in dancers. Clin Sports Med 7:143-173, 1988. Inman VT: The Joints of the Ankle, 2nd ed. Baltimore, Williams & Wilkins, 1991. Mann RA, Baxter DE, Lutter LD: Running symposium. Foot Ankle 1:190-224, 1981. Nigg BM: Biomechanics of Running Shoes. Champaign, Ill, Human Kinetics Books, 1986. Schwartz RP, Heath AL, Misiek W: The influence of the shoe on gait. J Bone Joint Surg 17:406-418, 1935. Torg JS: Athletic footwear and orthotic appliances. Clin Sports Med 1:157-175, 1982.

R eferences Please see www.expertconsult.com

A p p e n d i x

Sports Medicine Terminology Dean C. Taylor, Robert A. Arciero, Donald T. Kirkendall, and William E. Garrett, Jr.

The purpose of this appendix is to define commonly used sports medicine terms and to establish a basis for standardized orthopaedic sports medicine terminology. The language must be understandable for patients, health care professionals, the media, orthopaedic surgeons, and sports medicine specialists. If we want to communicate our ideas to others, we need to do so in a comprehensible way. If the language we use is ambiguous, confusing, contradictory, inconsistent, or filled with nonwords (as it often is), our communication is ineffective, misleading, and potentially problematic. The bases for the presentation of the material in this chapter are not our own opinions or pet peeves. We have solicited suggestions from experts (editors of major orthopaedic journals and established medical editors) to assist in improving the language of sports medicine. We have attempted to synthesize the collective wisdom of sports medicine organizations to help develop consensus when possible. We hope that this chapter can serve as a starting point for standardizing sports medicine terminology and lead to further refinement of our unique language. The importance of using consistent and proper terms helps research findings to be more accessible and understandable to the audience. But more importantly, misuse of the language can affect patient care. Physicians in the same group using different definitions of terms could compromise patient care when interacting on a common patient. This could also extend to health care support such as nurses, nurse practitioners, physician assistants, physical therapists, athletic trainers, and more. This chapter is divided into two segments: commonly accepted sports medicine terminology and some special topics along with sports medicine classification systems. We point out common misuses of terminology. In situations in which there are good arguments for consensus, we argue for the acceptance of the terminology or the classifications. If there is conflicting usage, we point out the strengths, the weaknesses, or the limitations.

SPORTS MEDICINE TERMINOLOGY Sports medicine terminology is a highly descriptive mix of athletic, lay, and medical language. It is filled with athletic terms, such as jumper’s knee, tennis elbow, skier’s thumb, and footballer’s ankle, and common terms, such as shoulder separation and hip pointer. The language is colorful and has developed over time as prominent athletes, media personnel, trainers, and physicians have added their own terms, or

misuse of terms, to the mix. This rich and vivid language can also be confusing because many terms are used improperly or have developed different meanings over time. For example, the media routinely confuses strain and sprain. Or how does a dislocation differ from a frank dislocation? What is a nonfrank dislocation and what is the “line” to be crossed when identifying a dislocation as frank, and do all physicians understand where this line is? How about stomach versus abdomen? Does a patient place his or her hand on the stomach or the abdomen? Unless there is an incision to reach into, it’s probably the abdomen. Where is the line separating the lower abdomen from the groin? These may seem to be nonsense examples, but such misuse of terms is what leads to confusion and miscommunication. Medical professionals are aware that humans have upper and lower extremities, not arms and legs. The upper extremity contains the arm and the forearm. The lower extremity contains the thigh and leg. Table APP-1 lists some definitions of commonly used sports medicine terms, some of which can be misleading. It may be difficult to eliminate these terms from our sports medicine language, but their elimination is necessary if we are to improve our ability to communicate consistently. For example, how can one explain to a medical student that a jumping athlete with tenderness of the proximal patellar tendon has a localized degenerative process, not an inflammatory process, when the terminology we use to describe the condition is “patellar tendinitis”? Maffulli and associates1 have discussed this common incorrect usage of the term tendinitis. To try to develop an international consensus on describing tendon problems, the Magellan Orthopaedic Society, the alumni society for the international sports medicine traveling fellowships, considered this terminology problem. The Society decided that most overuse tendon problems are associated with noninflammatory degenerative-type changes in the tendon, which histologically should be called tendinosis. Inflammation around a tendon is more commonly tenosynovitis, such as in de Quervain’s or Achilles tenosynovitis. Because tendinitis, tenosynovitis, and tendinosis are all histologic diagnoses, however, these terms should not be used for the clinical diagnosis of an overuse tendon problem. Rather, the term tendinopathy is a better descriptor for the clinical diagnosis (see Table APP-1). This distinction may seem insignificant to some people, but when terminology interferes with accurately teaching students, residents, and patients, it is a real problem because our language loses clarity. Another problem is the use of terms or phrases in the wrong context. Frequently, strain—an injury to a ­muscle— is incorrectly used to describe a sprain—an injury to a 2207

�rthopaedic ����������� S �ports ������ � Medicine ������� 2208 DeLee & Drez’s� O

   TABLE APP-1  Definitions of Commonly Used Sports Medicine Terms Abrasion—a worn-away area of skin Arm—the part of the upper extremity between the shoulder and the elbow, sometimes used incorrectly in place of “forearm” Arthritis—inflammation of a joint (commonly used to describe arthrosis) Arthrosis—degeneration of a joint Bursitis—inflammation of a bursa Cartilage—the tissue covering the articular surface of bones Chondromalacia—softening of articular cartilage; frequently used incorrectly for patellofemoral pain before operative inspection of the patellar articular cartilage Cramp—a painful spasmodic muscle contraction Dislocation—displacement of the bones of a joint from their normal position; usually implies loss of articulation of the joint surfaces that are normally in apposition Forearm—the part of the upper extremity between the elbow and the wrist Instability—a condition of increased joint motion due to ligament injury; a symptom Laceration—an open wound, commonly referred to as a “cut” Laxity—looseness or slackness, usually when describing the character of a ligament; may be used when describing a normal or an abnormal ligament; a sign Leg—the part of the lower extremity between the knee and the ankle, sometimes used incorrectly in place of “thigh” Lower extremity—thigh + leg + ankle + foot Meniscus—within the knee, a crescent or crescent-shaped tissue of fibrous cartilage Operation—an act performed by a surgeon; a surgical procedure Radiograph (or roentgenogram)—the image produced by passing x-rays through the body onto specially sensitized film Sprain—injury to a ligament secondary to excessive load (sometimes used incorrectly to describe a muscle-tendon unit injury) Strain—injury to a muscle-tendon unit secondary to excessive contractile or stretching load (sometimes used incorrectly to describe a ligament injury) Subluxation—a partial dislocation Surgery—the branch of medicine that treats injuries, diseases, and deformities by manual or operative methods, or conceptually, the work performed by a surgeon (often used incorrectly in place of “operation”) Tendinitis—inflammation of a tendon Tendinopathy—disease process of a tendon Tendinosis—degeneration of a tendon Tenosynovitis—inflammation of a tendon sheath Thigh—the part of the lower extremity between the hip and knee Upper extremity—the arm + forearm + wrist + hand X-ray (or roentgen ray)—the actual electromagnetic radiation used to make radiographs (commonly used as a synonym to radiograph)

­ligament—and vice versa. Incidence and prevalence are other terms that are sometimes incorrectly interchanged. Two examples of the meanings of inappropriately used words being accepted and ingrained in our use are “arthritis” and “x-ray.” Arthritis is often used incorrectly to describe a degenerative joint; the term arthrosis may be the accurate term. X-ray is commonly used instead of the accurate term radiograph to refer to the image made by radiography. We use these terms so frequently on a day-to-day basis that the incorrect meanings have become accepted. This acceptance then becomes a problem when we are communicating with newcomers to the sports medicine language and in formal writing and presentations. The correct terminology is defined in Table APP-1. Surgery is another word that is frequently used incorrectly in place of “operation,” as in “The knee surgery we performed on the patient yesterday was a success.” Surgery is defined as a field of medicine or the concept of the work performed by a surgeon, not an actual surgical procedure. An operation is the act performed by the surgeon. “The knee operation we performed…” is the correct usage. Many times, commonly used terminology has evolved because a certain term may be easier to use than more accurate language. These terms are not incorrect but can be misleading. For example, it may be difficult to explain to a patient how someone who does not play tennis can develop tennis elbow. In these cases, it is important to know the synonyms. The common usage is unlikely to

­ isappear, but it is necessary to know the accurate termid nology to improve understanding and for formal writing or presentation. Some examples of common terms and their associated precise synonyms and definitions are listed in Table APP-2.    TABLE APP-2  Commonly Used Sports Medicine Terms Common Term

Precise Term

Break Bruise Burner/stinger Dead arm

Fracture Contusion/ecchymosis Brachial plexus traction injury Condition of transient episodes of upper extremity loss of function, associated with recurrent, transient anterior subluxation34 Syndesmosis ankle sprain Iliac crest contusion or abdominal muscle strain at iliac crest insertion Patellar tendinopathy Muscle strain Nonspecific term for leg pain; usually implies a condition of overuse; specific conditions (stress reaction, stress fracture, tendinopathy, exertional compartment syndrome, and the like) should be used as diagnosis Acromioclavicular sprain Ulnar collateral ligament sprain of thumb metacarpophalangeal joint Lateral epicondylar tendinopathy

High ankle sprain Hip pointer Jumper’s knee Muscle pull Shin splints

Shoulder separation Skier’s/gamekeeper’s thumb Tennis elbow

Appendix 2209

   TABLE APP-3  Nonwords Commonly Used in Sports Medicine

   TABLE APP-4  Commonly Misused Plural Forms   of Singular Words

Nonword

Correct Word

Singular

Plural

Allograph Crepitence Patulent Sublux

Allograft Crepitus/crepitation Patulous Subluxate or subluxation

Bacterium Basis Criterion Curriculum Datum Fungus Maximum Medium Patella Septum Sequela

Bacteria Bases Criteria Curricula Data Fungi Maxima Media Patellae Septa Sequelae

Other words that are used in a questionable context include the -logy words, such as pathology, morphology, and symptomatology. The -logy suffix comes from the Greek logos, meaning work or reason, and refers to the science or study of the subject designated by the stem to which it is affixed.2 Thus, the traditional meaning of the word pathology is the science or study of disease processes. Morphology refers to the science of the forms and structure of organisms2 and symptomatology to the science of disease symptoms. Common usage has led to other definitions for these -logy words. A secondary meaning of pathology has become “the structural and functional manifestations of disease”2; for morphology, “the form and structure of a particular organism, organ or part”2; and for symptomatology, “the combined symptoms of a disease.”2 For example, orthopaedists frequently refer to intra-articular pathology when describing abnormalities in a joint or the appearance of a meniscal tear. These forms of common usage have become accepted in presentations and verbal communication but are not accepted by some journal editors. In addition, our communication will be easier if we can keep our language as simple as possible. Therefore, we would recommend using the -logy words in their more traditional meanings. The common form of pathology could be replaced with words such as injury, lesion, finding, or damage. Similarly, shape, appearance, structure, or form can replace morphology, and symptoms can be used instead of symptomatology. Methods should be used consistently, not methodology, when referring to “the methods we used in our study.” Suffer is a word that is commonly used inappropriately, as in “a patient suffered a tibia fracture.” This is more common in the lay media but is still present in medical writing and presentations. Inasmuch as the extent of suffering associated with injuries is often unknown and difficult to quantify, it is better to use the word sustain to denote an injury, as in “a patient sustained a tibia fracture.” Some nouns in sports medicine are now being used as verbs. Examples include scope or arthroscope (“we arthroscoped his knee”), biopsy (“we biopsied the lesion”), and radiograph or x-ray as described previously (“we x-rayed his leg”). These usages are common in sports medicine, but not acceptable. The use of nouns as new verbs should be avoided in formal presentations or writing. We can offer numerous other similar instances. Consider gender versus sex. Inanimate objects have gender in languages like Spanish, French, and others, whereas sex is biologic. The proper term for males and females is sex, but most editors will ask that the word gender be used. We also use many nonwords in sports medicine. At worst, this can make us look unintelligent and can lead to

poor communication, especially with those for whom English may be a second language. At best, these terms sometimes become so ingrained in usage that they are adopted as accepted words. Some of the nonwords are listed in Table APP-3. Several common terms are incorrectly used in the plural form. For example, the phrase, “The data is shown in ­Figure 1,” should be stated as, “The data are shown in ­Figure 1.” Table APP-4 provides some commonly misused plural forms of singular words. Some terms remain controversial, and no consensus on usage exists. For example, the patellar ligament or the ligamentum patellae fits the strict definition of a ligament in that it connects two bones, the patella and the tibia; however, as an extension of the quadriceps muscle group, it functionally acts as a tendon, the connection between a muscle and a bone, and can also be called the patellar tendon. Both usages are accepted and understood, but there is no consensus among orthopaedic journals. We use “patellar tendon” because the patella is a large sesamoid, and the collagen that extends distally from the patella acts as a tendon to extend the knee when the quadriceps muscle group contracts. Similarly, the flexor hallucis longus tendon extends distally from the plantar sesamoid bones to the head of the first metatarsal.

SPORTS MEDICINE CLASSIFICATION SYSTEMS We create classification systems to improve our ability to communicate both clinically and in research. We also use the classifications to help decide courses of action when presented with a clinical problem. A few classification systems are widely accepted, make sense, and contribute to the understanding of a process. Rockwood’s classification3 of acromioclavicular joint sprains is an example of a classification that is well accepted in North America. Seddon’s classification4 of nerve injuries (neurapraxia, axonotmesis, neurotmesis) is also well accepted because the terms describe the extent of injury and are easy to understand and remember. Most of our classification systems are plagued by the existence of other competing classification systems, by lack of acceptance, or by varied interpretation of the grading. These problems create a significant communication

�rthopaedic ����������� S �ports ������ � Medicine ������� 2210 DeLee & Drez’s� O

dilemma. When multiple classification systems are in place, using them is equivalent to a situation in which everyone is speaking different languages and no one knows which languages the others are speaking. For example, using the Hughston classification,5 a grade II medial collateral ligament knee sprain has no medial joint space opening with valgus stress at 30 degrees of flexion, but using Fetto and Marshall’s classification6 a grade II injury is “unstable… with a firm end point if chronic and soft if acute.” If different classifications are used, understanding is impaired. Even if there is only one system that is not widely accepted, it can be used to communicate only with those who understand and use the system. For example, if an ­English-speaking physician is trying to describe a LaugeHanson supination-eversion type IV ankle fracture to another English-speaking physician who does not know the Lauge-Hanson classification, communication will be as effective as if one of them were speaking a foreign language. If we are trying to expand education and ­understanding, it is not effective to have a language with limited acceptance. Perhaps the worst situation is when a classification system exists that has different interpretations so that one individual may have a completely different idea of what another is intending to communicate. This situation is common in communications about articular cartilage lesions. These lesions are usually graded I to IV based on Outerbridge’s classification of chondromalacia patellae7; however, there have been several modifications of the Outerbridge classification, and these modified grading schemes are now applied to articular cartilage lesions throughout the knee and in other joints. As a result, in a lecture on cartilage injury, a grade II lesion may mean many different things to an audience, rendering the use of all the classifications ineffective. The fact that there is confusion regarding classification systems is well recognized. In a 1997 questionnaire sent to the Herodicus Society members, 97% believed that there was confusion in grading of knee ligament injuries (Richard J. Hawkins, MD, personal communication). Even though the Society’s members recognized that there was no agreement regarding grading injuries, 82% still used some type of classification. Although we need classification systems to communicate, many of the ones we have now are inadequate.

General Lack of Consensus Regarding Grading and Measurement Most classification systems in orthopaedic surgery and sports medicine have three grades: I/II/III, 1/2/3, or 1+/2+/3+. (What the “+” means in grading is unclear, but it has become a common part of the sports medicine language. One might think “1+” is greater than “1” but not quite “2.” If so, what is “3+”?) The three-level method of grading is simple and easily understood. It generally provides information about the magnitude of injury or measurement. In medicine, these classifications are widely recognized to represent adjectives such as mild/moderate/severe, small/medium/large, or little/moderate/big.

Therefore, there is usually understanding, though inexact, when general grading scales are used. Communication could be greatly improved, however, if descriptive words instead of numbers were used to classify. In the story of Goldilocks, it is much better to talk about a papa bear, a momma bear, and a baby bear instead of a type III, type II, and type I bear. Similarly, it is easier to describe a nerve injury as neurapraxia, axonotmesis, or neurotmesis instead of type I, II, or III. In efforts to make the classification systems more exact, authors have applied quantitative values to the gradations. Differences of opinion have led to different classification systems based on differing magnitudes of the measurements used to define the classification grades. As a result, we find several different classifications in place and are unable to compare one study with another. Because of the many differing classification systems, it might be reasonable to eliminate the grading systems and instead to quantify measurements precisely. The only problem with this concept is that our measurements are so imprecise that comparisons between different examiners have little agreement. The members of the International Knee Documentation Committee clearly illustrated this point. When examining patients with different knee conditions, the International Knee Documentation Committee members differed appreciably in their measurements of translation on Lachman’s test and grading of the pivot shift and reverse pivot shift tests.8 Additionally, there was a wide variability among examiners in how instability tests were performed.9,10 Using different methods for making the same measurement affects the results, thus contributing to the general lack of agreement regarding the measurement values. We have a significant problem in sports medicine if even the experts, such as the International Knee Documentation Committee members, cannot agree on what is measured, how to measure it, or what the results of the measurement are. Partly for this reason, when the Anterior Cruciate Ligament (ACL) Study Group examined the problem of knee classifications in 2000, the members agreed that when examining a knee, a general description of the findings was better than exact millimeter measurements. The group decided that better agreement could be reached in regard to the assessment of whether an ACL was torn and whether a collateral ligament injury was mild, moderate, or severe (grade I, II, or III) than on the millimeter measurements of displacement during a physical examination (Richard J. Hawkins, MD, personal communication). We agree that physical examinations can determine whether there is an injury and generally approximate the extent of injury. Whenever possible, injuries, findings, and examination results for individual cases should be described in common language, and if measurements are reasonably accurate, they can be included in this description. Going back to the example of the three bears, it is much clearer to say that we were chased by an 8-foot (2.4-meter) male bear than by a type III bear, even if the bear in question was only 7 feet 6 inches (2.25 meters tall). More to the point, an articular cartilage lesion is better described as a 2 cm × 2 cm halfthickness lesion than as a type II or III (depending on the classification system used) lesion of the femoral condyle.

Appendix 2211

A

B

C

D

Figure APP-1   Classification of superior labrum injuries. A, Type I, fraying and degeneration of the superior labrum. B, Type II, detachment of the superior labrum and the biceps anchor from the superior glenoid. C, Type III, bucket handle tear of the superior labrum, with the peripheral labrum and the biceps anchor remaining intact. D, Type IV, bucket handle tear of the superior labrum with extension of the tear into the biceps tendon. (From Snyder SJ, Karzel RP, Del Pizzo W, et al: SLAP lesions of the shoulder. Arthroscopy 6:274-279, 1990.)

To provide some understanding of the language, the following subsections outline and clarify the qualities and deficiencies of the classification systems that have been created to (1) describe anatomic changes associated with pathologic conditions, (2) define physical examination findings, and (3) grade the extent of an injury.

Classifications Anatomic Changes Associated with Pathologic Conditions Anatomic changes due to injury or disease that are seen repetitively intraoperatively or on imaging studies are often classified for research purposes or to assist in treatment. The following classifications are examples of these grading schemes.

Superior Labral Injuries With the development of shoulder arthroscopy has come the realization that a severe injury to the superior glenoid labrum can be a source of symptoms. Snyder and colleagues11 coined the term SLAP, which has become associated with I

II

superior labral injuries. SLAP stands for superior labrum anterior to posterior. This term is so firmly ingrained in shoulder terminology that it is often used without clarifying the severity of the injury, the precise diagnosis, and a description of loss of function. Hence, the acronym may be confusing for those just learning about shoulder injuries. Snyder and colleagues11 also described the classification of superior labral injuries (Fig. APP-1).

Acromion Shape Bigliani and coworkers12 described various shapes of the acromion that are associated with impingement syndrome of the shoulder (Fig. APP-2). These shapes are usually defined on a suprascapular outlet view radiograph. Obtaining a good outlet view can be difficult, and the shape of the acromion can vary depending on the angle of the x-ray beam. Therefore, as in any classification system, there can be variability in the measurements.

Radiographic Changes of the Knee In 1948, Fairbank13 outlined radiographic changes associated with previous meniscectomy (Table APP-5). These radiographic findings have come to be known as Fairbank’s changes and are telltale signs of arthrosis that can follow

III

   TABLE APP-5  Fairbank’s Changes Marginal ridge Femoral condyle flattening

Figure APP-2   Classification of acromion shapes. I, Flat acromion from posterior to anterior. II, Anterior acromion has an inferior curve relative to the remainder of the acromion slope. III, Anterior acromion has an acute angle, or “hooked” shape, relative to the remainder of the acromion slope. (From O’Brien SS, Allen AA, Fealy S, et al: Developmental anatomy of the shoulder and anatomy of the glenohumeral joint. In Rockwood CA, Matsen FA [eds]: The Shoulder, 2nd ed. Philadelphia, WB Saunders, 1998, p 45.)

Joint space narrowing

Ridge or osteophyte formation at the margin of the outer aspect of the femoral condyle Flattening of the femoral condyle’s normally concave curvature as visualized on an anteroposterior radiograph Narrowing of the space between the femoral condyle and the tibial plateau; usually best defined on an anteroposterior radiograph with the patient bearing weight. An anteroposterior radiograph taken with the patient bearing weight and the knee flexed 45 degrees is often helpful in demonstrating joint space narrowing and has come to be known as the Rosenberg view33

�rthopaedic ����������� S �ports ������ � Medicine ������� 2212 DeLee & Drez’s� O

   TABLE APP-6  Modified Outerbridge Classification   of Articular Cartilage Lesions Grade I Grade II Grade III Grade IV

Cartilage softening and swelling Fragmentation and fissuring, ≤1-cm diameter area Fragmentation and fissuring, >1-cm diameter area Erosion of cartilage to bone

partial or complete meniscectomy. Fairbank’s changes are now commonly used to describe any radiographic evidence of arthrosis of the knee.

Articular Cartilage Lesions In 1961, Outerbridge7 published his classification of articular cartilage lesions associated with chondromalacia patellae. This classification has subsequently been used to classify articular cartilage lesions in general throughout the knee and in other joints. Furthermore, the Outerbridge classification has been modified based on arthroscopic findings. The grading criteria of the Outerbridge classification have been modified and misquoted in various articles on cartilage injury. Others have attempted to use grading scales based on lesion depth,14 shape,15 and a combination of measures.16,17 Noyes and Stabler17 have reviewed many of the classification systems in their article describing their system. Unless there is a clear definition of the classification system being used to describe a lesion, there will be confusion. We propose that for simplicity’s sake and for general understanding, the appearance of individual cartilage lesions be described by size, depth, and location; for example, a 1.5 cm × 2.0 cm partial-thickness lesion of the medial femoral condyle. When classifying a series of lesions in a research study, it is important that the classification be clearly defined. We suggest the Outerbridge classification for general research use when grading cartilage lesions because it is the original scale for cartilage injury, is widely accepted, and remains the best-known grading scale. The only modification that we make is changing the cutoff between grades II and III from 1⁄2 inch to 1 cm, owing to the conversion to the metric system after the publication of the original article (Table APP-6). In research, if the Outerbridge classification is not used, the classification that is used should be well described and defined. Recently, the International Cartilage Repair Society has developed a rating scale to assess articular cartilage lesions. Although use of this scale is increasing, inter-rater and intrarater reliability data are lacking.

Classification of Physical Examination Findings Physical examination findings are frequently classified into different gradations, especially in the evaluation of ligaments and the translational or rotational changes that may result from injury. Noyes and colleagues18 have provided clear definitions for much of the terminology that is used to document examination findings. Quantitative physical examination measurements should be described as translations or rotations, and not as the amount of “laxity” in

   TABLE APP-7  General Grading of Translation/Joint Opening for Ligament Testing 1-5 mm >5 mm and <10 mm 10 mm or greater

1+ or grade I 2+ or grade II 3+ or grade II

a particular ligament. Laxity implies the general slackness present in a structure and should not be used when discussing measurements. Instability refers to the presence of abnormal displacements between the two opposing bones of a joint as a result of traumatic injury. Like the term laxity, instability should be used in a general sense. It is better to state that there is anterior translation on Lachman’s examination than to say that there is 2+ laxity, or 2+ instability of the ACL. Additionally, the history of traumatic injury to a joint and symptomatic excessive translations differentiates instability from laxity. In this sense, the term multidirectional instability of the shoulder sometimes is used incorrectly to imply generalized shoulder laxity in the absence of trauma. In 1968, the Committee on the Medical Aspects of Sports of the American Medical Association (AMA) published Standard Nomenclature of Athletic Injuries.19 Many physical examination grading scales and injury classifications used today follow the AMA’s guidelines. The grading for ligament testing is shown in Table APP-7. Using this system, an 8-mm posterior drawer test would be graded as 2+, or grade II. Often the “+” is used to clarify that one is referring to a measurement grade and not a grade of injury, which is discussed in the next section. This grading system for evaluation of ligaments can also be used to quantify joint space opening to rotational stress, such as valgus stress to assess a medial collateral ligament (MCL) injury of the knee or the elbow. In this case, 7 mm would be considered a 2+, or grade II, opening to valgus stress. In the following sections, we outline some commonly used physical examination grading scales.

Shoulder Instability Examination (Translation) Scapulothoracic, acromioclavicular, and glenohumeral motion combined provide the shoulder with tremendous mobility in positioning the hand, both in daily activities and in throwing sports. Any attempt to quantify glenohumeral translation when evaluating a shoulder for instability is difficult because of these multiple articulations. The load and shift test20 and the drawer test21 are common methods that attempt to quantify the amount of glenohumeral translation. The presence of different classifications for the load and shift test has resulted in confusion and difficulty in comparing studies in the shoulder literature. Additionally, different examiners conduct the test differently. Despite these problems, grading of the load and shift or drawer test is useful to provide a general idea of the anterior and posterior translation of the humerus relative to the glenoid. For uniformity, we recommend the use of the American Shoulder and Elbow Surgeons (ASES) grading scale for glenohumeral translation (Table APP-8).22 The

Appendix 2213

   TABLE APP-8  American Shoulder and Elbow Surgeons Grading for the Load and Shift Test

   TABLE APP-9  Recommended Grading Scale for Anterior and Posterior Shoulder Translations

Grade

Humeral Head Translation

Degree of Translation

Grade

Humeral Head Translation

0 I

No translation Slightly up the glenoid face

1

II

Translation to the glenoid rim, but not over Translation over rim

None Mild (0- to 1-cm translation) Moderate (1- to 2-cm translation) Severe (>2-cm translation)

Humeral head translation up to, but not over, the glenoid rim Humeral head translation over the glenoid rim with spontaneous reduction Humeral head translation over the glenoid rim without spontaneous reduction (“locked out”)

III

ASES grading scale can also be applied to inferior translation of the humerus when evaluating the sulcus sign. The grading results are reported for each shoulder in isolation and not in comparison with the contralateral shoulder. In 1990, Hawkins and Bokor23 proposed a four-level grading system for glenohumeral translation (Fig. APP-3). In 1998, they subsequently modified this grading scale (Table APP-9),22 and it was then used as part of the ­American GRADE

GLENOHUMERAL TRANSLATION

CLINICAL

Trace Small amount of humeral head translation

I

Humeral head rides up the glenoid slope but not over the rim

II

Humeral head rides up and over the glenoid rim Reduces when stress removed

III

Humeral head rides up and over the glenoid rim Remains dislocated on removal of stress

Figure APP-3   Diagram of grades of translation of the humeral head in the glenoid fossa during the load and shift test. (From Hawkins RJ, Bokor DJ: Clinical evaluation of shoulder problems. In Rockwood CA Jr, FA Matsen 3rd [eds]: The Shoulder, vol 1, 1st ed. Philadelphia, WB Saunders, 1990. pp 149-177.)

2 3

Shoulder and Elbow Surgeons evaluation. Subsequently, McFarland and colleagues24 and Bach and associates25 have described great inter-rater and intrarater variability when using the ASES grading scale. One of the reasons for this variability is that grading by tactile appreciation of translation short of subluxation (as in the original grading scale) is easier to identify than millimeters of translation (as used in the ASES scale).24 In fact, there is a 0.3 greater reliability if the grading of translation is based on whether the humeral head translates over either the anterior or posterior glenoid rim.25 McFarland and coworkers also suggested that the lowest grades (trace and 1 in the first edition of the book or 0 and 1 in the second edition) be combined.24 When grades 0 (or trace) and 1 are combined, the intra-and inter-rater reliability is improved. Therefore, we would recommend using the classification system of Hawkins and Bokor, as modified by McFarland and coworkers,24 for reporting ­glenohumeral translation (see Table APP-9). Inferior translation has been measured most commonly using an inferior traction force applied to the upper extremity, as described in the sulcus sign by Neer and Foster.26 Bahk and colleagues27 pointed out that the sulcus sign usually is graded as I (<1 cm translation), II (1 to 2 cm), and III (>2 cm). An alternative for describing the sulcus sign when measuring the amount of inferior translation is to simply use centimeters, which may be a more accurate method to communicate.

Anterior Cruciate Ligament The diagnosis of an ACL injury is largely based on two physical examination techniques: Lachman’s test and the pivot shift test. Lachman’s test can be based on manual examination, an instrumented examination, or radiographic analysis. The measurement in millimeters is usually the difference between the measurements in the injured knee and in the normal contralateral knee. Different grades have been described. We recommend the use of a three-grade scale based on grading described by the AMA (see Table APP-7).19 Additionally, the end point is graded as A-firm end point, or B-soft, or not a well-defined end point. Compared with the normal knee, a knee with 8 mm of increased anterior translation with a soft end point would be a grade IIB Lachman. The pivot shift has had numerous descriptions and different grading schemes. Owing to varying degrees of the pivot shift phenomenon, we have used a grading system (Table APP-10) based on the classification of Bach and associates.25 The one deficiency of this grading scheme is that a 2+ pivot shift encompasses a wide range of findings;

�rthopaedic ����������� S �ports ������ � Medicine ������� 2214 DeLee & Drez’s� O

   TABLE APP-10  Pivot Shift Test Grade

Description

I

Mild, also known as pivot glide, slight subluxation with test, but no “jump” or “shift” Moderate, obvious jump or clunk with reduction of tibia Severe, marked clunk; tibia may remain subluxated unless reduction maneuver is employed

II III

   TABLE APP-11  American Medical Association Ligament Injury Classification Grade

Description

I

Mild, minor tearing of ligament fibers; no demonstrable increase in translation on examination Moderate, partial tear of ligament without complete disruption; slight to moderate increased translation on examination Severe, complete tear of the ligament; marked increase in translation on examination

II III

however, this system does allow the examiner to assign different grades based on the magnitude of abnormal examination findings.

Posterior Cruciate Ligament The primary test for evaluating the posterior cruciate ligament is the posterior drawer test. Traditionally, the posterior drawer test has been graded similarly to other ligament tests (see Table APP-7). Posterior tibial translation so that the anterior tibia is flush with the femoral condyles has generally been considered to be 10 mm of posterior translation, or the dividing point between a 2+ or 3+ posterior drawer findings. Obviously, patient size varies so that the flush position does not always equate to 10 mm of posterior translation. To improve the understanding of posterior drawer measurement, the “thumb sign” has been introduced. The examiner performs the thumb sign test by sliding his or her thumbs off the femoral condyles onto the tibial plateau. The test is graded as anterior, flush, or posterior depending on the position of the anterior tibial plateau relative to the femoral condyles. The consensus of the ACL Study Group was that the thumb sign was a better assessment than millimeters of posterior tibial translation in documenting the posterior drawer findings.

Collateral Ligaments of the Knee The integrity of the knee collateral ligaments is assessed by applying a valgus (MCL) or varus (fibular collateral ligament) rotational stress to the knee positioned in 30 degrees of flexion. The amount of opening in millimeters of the medial or the lateral joint space is measured and is compared with the normal contralateral knee or is considered an isolated examination. The amount of opening is classified using the AMA grading scale (see Table APP-7) or by documenting the measurement in millimeters. Based on the ACL Study Group consensus, using the 1+, 2+, and 3+ grades may be the best way to communicate, especially because measuring joint space opening is not reproducible to within 1 or 2 mm.

CLASSIFICATION SYSTEMS USED TO DESCRIBE SPORTS INJURIES General Ligament

Ligament injuries are generally graded on three levels based on the AMA guidelines (Table APP-11).19 This classification can be confusing because there are also three

grades for measuring translation on the physical examination, as discussed in the previous section. Some authors use these translational changes to classify the extent of injury. This usage results in conflicting classifications and confusion because they are not the same. For example, a grade I MCL injury of the knee will have no increased medial knee opening to valgus stress, but a grade II injury may have 1+ or 2+ medial joint opening to valgus stress. Additionally, some authors use their own grading scales for ligament injuries. The lack of uniformity is the biggest communication problem in grading the severity of ­ligament injuries.

Muscle Strains Currently, there is no good classification for muscle strain injuries. Unlike ligament injuries, partial disruption of a muscle-tendon unit is difficult to assess with mechanical testing; using a scale similar to the ligament injury classification, therefore, is not exact. The physical examination findings that can be assessed in a strain are tenderness, muscle strength, swelling, ecchymosis, and the presence of a defect in the muscle-tendon unit. Based on these findings, muscle strain injuries can be classified as interstitial strains, intramuscular strains, partial ruptures, or complete ruptures (Table APP-12). Interstitial injury without disruption of blood vessels or muscle fibers is the mildest form of strain. There may be mild to moderate tenderness and mild strength loss, but no swelling, ecchymosis, or defect with an interstitial strain. Intramuscular injuries are the next highest level of severity in muscle strains. In intramuscular strains, there is enough tensile force to cause limited muscle fiber and capillary disruption. The injury is usually localized adjacent to a muscle-tendon junction, where, because of the fiber disruption, there is swelling and possibly ecchymosis. Tenderness and weakness may be moderate to severe, but there is no palpable defect at the injury site.

   TABLE APP-12  Classification of Muscle Strain Injuries Injury Type

Swelling/Ecchymosis

Defect

Interstitial strain Intramuscular strain Partial rupture Complete rupture

Absent Present Present Present

Absent Absent Present, incomplete Present, complete loss of continuity

Appendix 2215

Figure APP-4   The five degrees of nerve injury based on the Sunderland classification of peripheral nerve injuries. 1, Neurapraxia equivalent with conduction block. 2, Axonotmesis equivalent with intact endoneurium resulting in wallerian degeneration. 3, Loss of nerve fiber continuity; perineurium and epineurium intact. 4, Loss of nerve fascicle continuity; only epineurium intact. 5, Complete nerve transection. (From Sunderland S: Nerve Injuries and Their Repair: A Critical Appraisal. Edinburgh, Churchill Livingstone, 1991, p 222.)

In partial muscle ruptures, there is a palpable defect, and tenderness, weakness, and swelling may be more severe than in intramuscular strains. In complete muscle ruptures, there is a disruption of all muscle fibers and total loss of function of the muscle injured. Complete ruptures are uncommon injuries.

Nerve Injuries Based on his extensive experience treating peripheral nerve injuries during World War II, Seddon4 classified nerve injuries as neurapraxia, axonotmesis, or neurotmesis. Neurapraxia is defined as a “failure of conduction in a nerve in the absence of structural changes, due to blunt injury, compression or ischemia; return of function normally ensues.”2 Axonotmesis is a “nerve injury characterized by disruption of the axon and myelin sheath but with preservation of the connective tissue fragments, resulting in degeneration of the axon distal to the injury site; regeneration of the axon is spontaneous and of good quality.”2 Neurotmesis is a “partial or complete severance of a nerve, with disruption of the axon and its myelin sheath

and the connective tissue elements; regeneration does not occur.”2 Sunderland28 classified nerve injuries into five types (Fig. APP-4). Types I and II are neurapraxia and axonotmesis, respectively. Types III to V represent increasing severity of nerve injury to complete transection. Both classifications are used and well known. Seddon’s classification probably has the greatest acceptance among all orthopaedists because of its descriptive nature and because it is simple and easy to understand. Sunderland’s classification may be better accepted among hand surgeons and neurosurgeons because it provides additional clarification that can be helpful in deciding between treatment alternatives.

Specific Injury Classifications Acromioclavicular Joint Injuries Rockwood3 has presented a classification for acromioclavicular joint injuries that is well accepted and commonly used. This classification is shown in Table APP-13.

�rthopaedic ����������� S �ports ������ � Medicine ������� 2216 DeLee & Drez’s� O

   TABLE APP-13  Classification of Acromioclavicular Joint Sprains Type I Type II

Type III Type IV

Type V

Type VI

Acromioclavicular (AC) ligament sprain only; interstitial without evidence of injury to the AC joint on radiographs AC ligament disruption without injury to the coracoclavicular (CC) ligaments; widening of the AC space on radiographs with minimal increase in the CC distance Disruption of the AC and CC ligaments; increase in CC distance on radiographs Ligaments injured as in type III with posterior translation of the lateral clavicle relative to the acromion with displacement of the clavicle into or through the trapezius muscle; posterior translation of the clavicle is seen on the axillary lateral radiograph Ligaments injured as in type III with detachment of the deltoid and the trapezius from the lateral clavicle; marked increase in the coracoclavicular separation, CC space 100% to 300% of normal Ligaments injured as in type III with inferior translation of the lateral clavicle; clavicle is inferior to the acromion and coracoid on radiographs

Data from Rockwood CA: Fractures and dislocations of the shoulder. Part II: Subluxations and dislocations about the shoulder. In Rockwood CA, Green DP (eds): Fractures in Adults, 2nd ed. Philadelphia, JB Lippincott, 1984, pp 722-985.

Medial Collateral Ligament Several grading schemes exist for MCL injuries, and as a result, there is confusion when trying to compare studies on these injuries. Many use the AMA’s ­nomenclature for damage to the ligament as a basis for injury classification. Others use the instability classification of 1+, 2+, and 3+ for grades I, II, and III. Hughston5 clearly outlined how study results can differ because of differences in classification systems used. He compared several studies with his own work and found that each study had a different definition for a grade III MCL sprain. Hughston’s classification is similar to the AMA’s; ­ however, his interpretation is that only grade III injuries can have medial opening to valgus stress. Hughston further subdivides grade III injuries by the extent of medial opening: 1+, 2+, or 3+. Others have interpreted that grade II may have increased opening of the medial joint space consistent with a partial tear (1+ or 2+ opening). As Hughston5 demonstrated, our understanding of the best way to treat MCL injuries is impaired by the use of various classifications. We recommend the use of the AMA system owing to its simplicity and the overall awareness of the classification. Based on physical examination findings, grade I (interstitial) injuries have no increased medial opening to valgus stress, grade II (partial injuries) have 1+ or 2+ opening, and grade III (complete tears) have 3+ opening. We recommend that authors clearly define their classifications and that readers understand that many classifications exist so that the data are properly presented and interpreted.

Fibular Collateral Ligament The grading of fibular collateral ligament injuries usually follows the standard I, II, and III scheme for ligament injuries. Instability is usually graded based on amount of

l­ateral joint-line opening to varus stress with the knee in 30 degrees of flexion. Again, some authors use the instability grading as injury grading, which results in confusion and inability to compare studies. We recommend the use of the AMA guidelines for ligament injury, as described for MCL injuries.

Anterior and Posterior Cruciate Ligaments It is unusual to classify ACL or posterior cruciate ligament injuries using the AMA’s nomenclature because it is difficult to identify grade I injuries. ACL and posterior cruciate ligament injuries therefore are usually considered to be either partial or complete. Additionally, the method of treatment of cruciate ligament injuries is dependent on the presence or the absence of injuries to other knee ligaments. ACL and posterior cruciate ligament injuries therefore are usually classified as “isolated” injuries, with no other ligamentous injury evident on examination, or as “combined” injuries, if there is evidence of injury to the ACL and one of the collateral ligaments.

Ankle Sprains Ankle sprains are usually separated into medial, lateral, and syndesmosis types. Grading of ankle sprains using the AMA classification can be difficult because many times, more than one ligament is involved. As a result, a “mild, moderate, or severe” grading scheme is usually employed to document the severity of injury. The West Point Ankle Grading System29 (Table APP-14) provides guidelines for grade I, II, and III injuries.

Concussive Injury Although brain injury is not an orthopaedic issue, recognizing and reporting these injuries is an important aspect of caring for athletes involved in contact and collision sports. There are few injuries as polarizing or more confusing than a concussion injury. The earlier comments on the grading of injury are magnified here because of the lack of agreement on the definition of injury and the lack of standardization of terms. There are two primary concerns regarding a concussive head injury. First is recognizing that an injury has occurred. Second is when to allow a player to return to play (beyond the scope of this chapter). Recognition of the injury is a function of the definition of the injury. The Concussion in Sport group offers this lengthy definition: “(1) Concussion may be caused by a direct blow to the head, face, neck, or elsewhere on the body with an ‘‘impulsive’’ force transmitted to the head. (2) Concussion typically results in the rapid onset of short-lived impairment of neurological function that resolves spontaneously. (3) Concussion may result in neuropathological changes, but the acute clinical symptoms largely reflect a functional disturbance rather than structural injury. (4) Concussion results in a graded set of clinical syndromes that may or may not involve loss of consciousness. Resolution of the clinical and cognitive symptoms typically follows a sequential course. (5) Concussion is typically associated with grossly normal ­ structural neuroimaging studies.”30

Appendix 2217

   TABLE APP-14  West Point Ankle Grading System Criteria

Grade I

Grade II

Grade III

Evidence of instability Reaction to manual ligament stress testing Localization of tenderness Lateral sprains

None Mild to moderate discomfort

None or slight Moderate to intense discomfort

Definite No pain or intense discomfort

Mild to moderate over ATaFL only Deltoid only Distal syndesmosis only

Intense over ATaFL + CFL + PTaFL Deltoid Distal syndesmosis and proximally > 4 cm Complete tear of ligaments involved Impossible

Well-localized Slight Positive squeeze or external rotation stress test

Moderate to intense over ATaFL + CFL Deltoid Distal syndesmosis and proximally ≤ 4 cm Partial tear, partial macroscopic disruption to ligaments involved Difficult or impossible without supportive device (i.e., brace, tape, cane) ± Localized Significant Positive squeeze and external rotation stress tests

Radiograph—no mortise widening Minimal edema superior and anterior to lateral malleolus

Radiograph—no mortise widening Moderate edema superior and anterior to lateral malleolus

Medial sprains Syndesmosis sprains Suspected disorder Weight-bearing capability Edema, ecchymosis Syndesmosis sprains Special tests

Stretch to ligaments involved without macroscopic disruption Full or partial without significant pain

Diffuse Significant Signs and symptoms of grade II sprain but will have mortise widening radiographically

ATaFL, anterior talofibular ligament; CFL, calcaneal fibular ligament; PTaFL, posterior talofibular ligament. Modified from Gerber JP, Williams GN, Scoville CR, et al: Persistent disability associated with ankle sprains: A prospective examination of an athletic population. Foot Ankle Int 19:655, 1998.

After recognizing that an injury has occurred, many health care professionals will attempt to grade the severity of injury. Table APP-15 shows three popular grading schemes (out of well over a dozen recognized scales). With most grading scales, each higher grade suggests a more serious injury to a specific location, such as in grading ligament injuries. In general, the higher the grade of injury, the more challenging the prognosis. As the previous definition states, this is a functional, not a structural, injury, and the presence or length of loss of consciousness is not prognostic of outcome. Thus, using loss of consciousness or memory loss or other features as criteria for injury severity is problematic. Many physicians who deal with concussive injury prefer not to use a grading system. They prefer simply to describe the injury as “concussion with <1 minute loss of consciousness” or “concussion with retrograde amnesia” because the qualifier is suggestive of where the functional deficit is focused in the brain. There are many grading scales and no    Table APP-15  Concussion Grading Scales Grade

Cantu

Colorado

AAN*

1 (Mild)

No LOC PTA < 30 min

2 (Moderate)

LOC < 5 min, or PTA > 30 min

3 (Severe)

LOC > 5 min, or PTA > 24 hr

No LOC Confusion No amnesia No LOC Confusion Amnesia LOC

No LOC Symptoms < 15 min No LOC Symptoms > 15 min Any LOC Brief vs. prolonged

*American Academy of Neurology. LOC, loss of consciousness; PTA, post-traumatic amnesia.

clear consensus on which scale is best. Assigning a grade to a concussion injury can be very confusing to patients and health care colleagues alike. Until the definitive grading scale is defined and validated, grading of concussion injury probably should be avoided.

BASIC STATISTICAL TERMS The language of statistics can be as confusing to orthopaedic surgeons as the language of medicine is to the lay public. In a profession whose advances are a result of research, however, proper understanding of some statistical terms is important to making decisions about the quality of research and whether results are meaningful to any particular physician’s practice. A statistics course is far beyond the scope of this chapter, but some basic statistical terms that are used on a regular basis require an understanding for communication between professionals and patients alike.

Reliability and Validity The fundamental definition of reliability is the degree of consistency with which an instrument or rater measures a variable.31 Within this definition are multiple levels of reliability. For example, one of the most common types of reliability is test-retest reliability (or intrarater reliability). A test is said to be reliable if measurements taken today and tomorrow by the same rater are similar. Another aspect of reliability is objectivity. Do two (or more) raters get the same results when administering a test? For example, do two different surgeons rate a manual muscle or range of motion or other clinical test the same? This is sometimes called inter-rater reliability. Other aspects of reliability are

�rthopaedic ����������� S �ports ������ � Medicine ������� 2218 DeLee & Drez’s� O

internal consistency and split-half reliability that are critical for questionnaire development. The definition of validity is multifaceted. The most commonly applied definition is the degree to which an instrument measures what it is supposed to measure. In research, studies are designed to allow reasonable interpretations of the data based on controls (interval validity), definitions (construct validity), analysis (statistical validity), and generalizations (external validity).31 Researchers need the company of a statistician to help with the design and analysis of a project to ensure all the nuances of the research process are satisfied.

Samples and Populations Research (especially clinical research) attempts to make some generalization about a population. Thus, the population needs to be defined at the outset of a project. A population is the entire set of cases or units the study attempts to generalize, and a sample is a subset of the population under study.31 Selecting the sample from a defined population requires special procedures to ensure a representative group of cases.

Dependent and Independent Variables In almost all research projects, there is a variable of interest and a variable modified to affect the variable of interest. The dependent variable is assumed to depend on or be caused by another variable. The independent variable is the variable presumed to cause or determine a dependent variable, or the variable manipulated or controlled by the researcher.31 For example, a paper is about outcomes after patellar ligament versus hamstring grafts for ACL rupture. The independent variable is graft (hamstrings versus patellar ligament), and the dependent variable is outcome.

Power Power is the ability of a statistical procedure to find a significant difference that really exists.31 In planning a project, the researchers need to consider what difference is important, the inherent variability in the measurements, and the planned level of significance. When all this is known, the appropriate number of cases to detect the desired difference can be studied. This is probably the most commonly requested service of a statistician in the planning stages of a project.

Clinical versus Statistical Significance A paper shows that there is a 0.8-degree difference in range of motion between two groups and that this difference is statistically significant. The question the reader might ask is whether this difference is of any practical significance and whether the reader’s methods are sensitive enough to detect such a small difference. One way that clinical significance is reported is by the use of confidence intervals. The clinician can more easily determine whether the detected difference is sufficient enough to warrant a change in practice. Part of the issue is the use of the word significant. To avoid confusion in medicine and research, the use of the

word significant should be reserved for statistical reporting and not be used as a synonym for important, critical, distinctive, or other similar terms.

EPIDEMIOLOGY TERMS Epidemiology is a special class of statistics that deals with the incidence, distribution, and control of disease in a population. The basic concepts are, whether one is aware of it or not, a common feature of dealing with colleagues and patients. Many basic epidemiologic statistics are based on the simple 2 × 2 table (Table APP-16) that asks, in this case, if a diagnostic procedure actually does find the existence of a clinical condition.

Sensitivity versus Specificity These are generally used to describe the utility of a test. Sensitivity is the fraction of true positives detected by a test (in Table APP-15, that would be a/a + c). Specificity is the fraction of true negatives detected by a test (as d/b + d). Assume that a test has a sensitivity of about 70% and a specificity of 95%. This means that about 70% of people with the condition will have a positive test (the remaining 30% will have a false-negative test) and that 95% of people without the condition will have a negative test (the remaining 5% will have a false-positive test).

Relative Risk and Odds Ratio Relative risk (routinely abbreviated as RR) estimates the degree of an association between exposure to a condition and the likelihood of developing the condition in the exposed group relative to those not exposed.32 In general, the relative risk is computed for prospective, randomized clinical trials, or cohort studies. Change the rows in Table APP-16 to Exposure to some risk and the columns to Outcome. RR is calculated as [a/(a + b)]/[c/(c + d)]. Consider prior ankle injury as a risk for subsequent injury. A sample of athletes has their records reviewed for injury history and then followed over the course of some predefined duration. Exposure would be prior ankle sprain (yes/no), and the outcome would be a new sprain (yes/no). At the end of the study, if the calculated relative risk were 4.5, this would mean an athlete with a prior ankle sprain is 4.5 times as likely to suffer another sprain than the player with no prior ankle sprain. A relative risk of 1.0 means the rates of disease in the exposed and unexposed groups are similar. An RR of more than 1.0 is a positive association (a prior ankle sprain is associated with a subsequent ankle sprain), and an RR of less than 1.0 is a negative association (greater flexibility is associated with fewer strain injuries). The odds ratio is a special case of relative risk and is usually used in retrospective or case-control studies.31    Table APP-16  The Basic 2×2 Table Disease present? Yes Test positive? Yes Test positive? No

a c

Disease present? No b d

Appendix 2219

Prevalence versus Incidence Prevalence is the proportion of individuals in a population with the particular condition at a specific point in time (as of today, how many residents of a nursing home have osteoporosis?). Incidence is the number of new cases that develop in a population during a defined time interval (how many new cases of osteoporosis are diagnosed in this nursing home over a 3-year period?).31

R eferences Please see www.expertconsult.com

I n d e x

Note: Page numbers followed by f refer to figures; page numbers followed by t refer to tables; page numbers followed by b refer to boxes.

A A band, 4f, 5, 6f, 13f, 208f A priori power analysis, 111 ABCDE mnemonic, 517, 519–520, 521f Abdomen on-field injury to, 526–527 pain in, 1451, 1452f preparticipation examination of, 512 Abdominal bracing, in core training, 280, 280f Abductor digiti quinti, nerve to, 2044, 2044f, 2045f entrapment of, 2044, 2044f, 2059–2060, 2060f release of, 2052 Abductor pollicis longus tendon, injury to, 30 Ablation procedures, in sudden death ­prevention, 170 Abrasion, 201–202 Abrasion procedure, arthroscopic, 52–53 Abscess, 387–388, 387f methicillin-resistant Staphylococcus aureus in, 395–397, 396b, 396f subperiosteal, 587, 589f Accessory lateral collateral ligament, 1301–1302, 1303f Accessory navicular, 1963, 1963f, 2162–2165, 2163f–2164f excision of, 2165, 2165f injury to, 2022–2023, 2024f Accuracy, 100, 108, 109f Acebutolol, 160t Acetabular labrum anatomy of, 1452–1453, 1453f tears of, 1470–1471, 1470f in children, 469, 470f magnetic resonance imaging in, 581–582 Acetabulum abnormal femoral head contact with, 1471–1472, 1471f–1472f anatomy of, 1452–1453, 1453f, 1500 chondral injury of, 1472 fracture of, 1464 Acetaminophen, in osteoarthritis, 1791 Acetylcholine, 353 Achilles tendon, 1997–2011 anatomy of, 1997, 2031 augmented end-to-end repair of, 2006t, 2007–2008, 2008f end-to-end repair of, 2006, 2006t, 2010–2011, 2010f–2011f examination of, 629–630 excision of, 2000 flexor hallucis longus reconstruction of, 2000, 2039f–2040f, 2041 imaging of, 630 inflammation of, 30 injury to, 1997–2002 anatomic considerations in, 1997 classification of, 29 evaluation of, 1998 flexibility and, 2182–2183 magnetic resonance imaging in, 563, 565f

Achilles tendon (Continued) nonoperative treatment of, 1998–1999 operative treatment of, 1999–2002 overuse, 29–30, 628–630, 1975, 1997, 2000–2001 playing surface and, 2198 rehabilitation after, 2001 magnetic resonance imaging of, 563, 565f, 2034, 2034f overuse injury of, 29–30, 628–630, 1975, 1997, 2000–2001 percutaneous repair of, 2008–2009 peritendinitis of, 30, 563, 1975, 1997, 2000, 2001 reconstruction of, 2000, 2039f–2040f, 2041 in retrocalcaneal bursitis, 2030, 2034, 2034f, 2036–2037 rupture of, 2002–2011 evaluation of, 2002–2004, 2003f external fixation in, 2009 imaging in, 2003–2004, 2003f needle test in, 2003 neglected, 2009 nonoperative treatment of, 2004–2005, 2004t operative treatment of, 2004–2011, 2004t, 2005f, 2006t, 2007f–2008f, 2010f–2011f percutaneous treatment of, 2008–2009 Thompson’s test in, 2003 stretching exercise for, 1999 three-tissue bundle suture repair of, 2006–2007, 2006t, 2007f ACL. See Anterior cruciate ligament (ACL) Acne, 205, 415 Acne mechanica, 202 Acromioclavicular joint anatomy of, 769, 769f, 826–828, 826f–827f, 831f arthrosis of, 829–830 biomechanics of, 826–828 dislocation of. See Acromioclavicular joint injury innervation of, 831f intra-articular fracture of, 854–855 kinematics of, 775–776, 776f, 828, 828f osteoarthritis of, 956, 959f osteophyte of, 956, 959f palpation of, 996, 997f pediatric anatomy of, 783, 783f biomechanics of, 786–787, 787f radiography of, 835–836 anteroposterior view for, 835–836, 949 normal, 835–836 posteroanterior, 949 stress view in, 835, 837f Stryker notch view in, 835, 838f, 949 Zanca view for, 835, 836f, 854 resection of, SLAP lesion and, 1024 Acromioclavicular joint injury, 826–856 anatomy in, 826–828, 826f, 827f classification of, 828, 829f, 830b, 2215, 2216t

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Acromioclavicular joint injury (Continued) coracoid fracture with, 880, 882, 882f, 883f, 884 evaluation of, 828–836 cross-arm adduction test in, 830, 832f imaging in, 835–836, 836f–������� 838f O’Brien’s test in, 830, 832f Paxinos’ test in, 830 radiography in, 831b, 833f–834f, 835–836, 836f–838f glenoid neck fracture with, 863, 865b, 865f mechanism of, 828, 830f pain pattern in, 828–829, 831f pediatric and adolescent, 855–856, 855f physical examination in, 828–834, 831b, 831f, 832f radiography in, 831b, 833f–834f, 835–836, 836f–838f, 949 return to play for, 854, 854t treatment of, 836–845 acromioclavicular ligament repair in, 840, 841f, 846–851, 847f–853f arthroscopic, 844 biceps transfer in, 840–841, 841f in children, 856 complications of, 851–853 coracoclavicular ligament reconstruction in, 842, 843f, 846–851, 847f–853f coracoclavicular ligament repair in, 841, 841f coracoclavicular ligament transfer in, 841–842 distal clavicle resection in, 842–844, 843f, 843t dynamic muscle transfer in, 840–841, 841f modified Weaver-Dunn procedure with augmentation in, 842 nonoperative, 838–840, 839f–840f, 844–845 operative, 838t, 840–846, 845f–846f osteolysis after, 853 pain after, 853 pin-related complications of, 851–852 return to play after, 854, 854t type I, 829f, 830b, 831, 833t type II, 829f, 830b, 831, 833f, 833t type III, 829f, 830b, 831–832, 833t, 834f type IV, 829f, 830b, 832, 833t, 834f type V, 829f, 830b, 832, 833t, 835f, 847f type VI, 829f, 830b, 833, 833t Acromioclavicular ligaments, 775, 776f, 826–827, 826f, 827f injury to. See Acromioclavicular joint injury repair of, 840, 841f, 846–851, 847f–853f Acromion accessory ossification center of, 781, 781f, 859, 860f, 955–956, 958f unfused, 781, 781f anatomy of, 859, 990, 990f, 992f, 1034t down-sloping of, 955, 957f enthesophyte of, 955, 958f fracture of, 866–867, 867b classification of, 867b



ii

Index

Acromion (Continued) magnetic resonance imaging of, 866f radiography of, 861, 861f, 866f return to play after, 873 stress, 867 treatment of, 866–867, 867f, 873f magnetic resonance imaging of, 955–956, 956f, 957f, 959b ossification centers of, 859, 860f type I, 955, 956f, 959b, 2211, 2211f type II, 955, 956f, 959b, 2211, 2211f type III, 955, 956f, 959b, 2211, 2211f Acromioplasty, 1008–1009 Actigraph, 451 Actin, 4, 5t Acute anterior cervical spinal cord injury, 674, 678 Acute lymphocytic leukemia, foot in, 1974 Acute mountain sickness, 504 Acyclovir, in herpes gladiatorum, 197 Adapalene, in acne, 205 Addison’s disease, 77t Adductor brevis, 1454t, 1455f, 1485 strain of, 1460–1461, 1490–1493 in adolescents, 1493 complications of, 1493 evaluation of, 1490–1491, 1490b, 1491f prevention of, 1492b return to play after, 1493, 1493b treatment of, 1491–1492, 1491b–1492b, 1492t Adductor canal syndrome, 1496b, 1497–1499 return to play after, 1498, 1498b treatment of, 1497–1498, 1498b, 1498f Adductor longus, 1454t, 1455f, 1485 strain of, 1460–1461, 1490–1493 in adolescents, 1493 complications of, 1493 evaluation of, 1490–1491, 1490b, 1491f prevention of, 1492b return to play after, 1493, 1493b treatment of, 1491–1492, 1491b–1492b, 1492t Adductor magnus, 1454t, 1455f, 1485 strain of, 1460–1461, 1490–1493 in adolescents, 1493 complications of, 1493 evaluation of, 1490–1491, 1490b, 1491f prevention of, 1492b return to play after, 1493, 1493b treatment of, 1491–1492, 1491b–����������������������� 1492b, 1492t Adenosine triphosphate, in energy metabolism, 210, 210f Adhesive capsulitis, 1094–1103 athroscopic release in, 1096–1102 anterior capsule in, 1098, 1099f–1100f bursectomy in, 1100, 1102f complications of, 1103, 1103b contraindications to, 1096–1097, 1097b indications for, 1096, 1097b inferior capsule in, 1099–1100, 1101f joint entry for, 1097–1098, 1097f–1098f physical therapy after, 1102, 1103b repeat, 1103 results of, 1103 rotator interval in, 1098, 1098f subacromial space in, 1100, 1102f closed manipulation in, 1095–1096 evaluation of, 1095, 1095b in female athlete, 489–490, 490t imaging in, 967, 967t, 968f, 1095 methylprednisolone in, 1096, 1103b nonoperative treatment of, 1095, 1095b open release in, 1096 pathoanatomy of, 1094 patient history in, 1095, 1095b

Adhesive capsulitis (Continued) physical examination in, 1095 treatment of, 1095–1096, 1095b. See also ­Adhesive capsulitis, arthroscopic release in Adolescents. See Children/adolescents Adrenal gland, exercise effects on, 217–218, 217t α-Adrenergic receptor agonists in complex regional pain syndrome, 365 in hypertension, 160t, 161 β-Adrenergic receptor agonists, in exercise-­induced bronchospasm, 181 α-Adrenergic receptor antagonists in complex regional pain syndrome, 363t in hypertension, 160t, 161 β-Adrenergic receptor antagonists in complex regional pain syndrome, 363t, 365 in hypertension, 159, 160t in sudden death prevention, 170 Adrenocorticotropic hormone, exercise effects on, 217, 217t Adson’s test in thoracic outlet syndrome, 1131, 1132f in vascular injury, 1138, 1138f Advanced dynamic training, in shoulder ­rehabilitation, 244–246, 245f–248f AED. See Automatic external defibrillator Aerobic exercise, 400 caffeine effects on, 421 Aeromonas hydrophila infection, 397, 398t Age/aging ACL injury and, 1651–1652 allograft properties and, 38 articular cartilage changes with, 40 bone changes with, 72, 78 immune function and, 148 osteochondral repair and, 52 tendon changes with, 25, 30 thermal balance and, 499 Aggrecan, 44 throwing-related accumulation in, 1215 Aggression, anabolic-androgenic steroids and, 417 Airway assessment in cervical spine injury, 667, 668f emergency, 519, 520t, 520f–521f Akin procedure, 2070, 2072b, 2072f–2073f, 2076, 2078f Albright’s disease, 73t Albuterol, in exercise-induced bronchospasm, 182 Alcohol matrixectomy, 2099–2100 Alcohol use/abuse, 401, 424–425 in hypertension, 159t Aldosterone, exercise effects on, 217, 217t Alendronate, in complex regional pain ­syndrome, 363t, 365 Alertness drug enhancement of, 451–452, 452t impairment of, 453 Alertness zones, 454, 455f performance and, 454–456, 455f Alkaline phosphatase, 74t–76t, 77t serum, in elbow heterotopic ossification, 1292 Allen’s test, 1320–1321, 1321f, 1357, 1359 Allergy, anaphylactic response in, 530 Allodynia, 356t in complex regional pain syndrome, 354, 356 Allograft(s), 137–145. See also Graft(s) donor screening for, 138–139 historical perspective on, 137–138 infection risk with, 141–142 ligamentous, 38, 142–143

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Allograft(s) (Continued) meniscal, 143–144, 1620–1622, 1621f osteochondral, 144–145 glenohumeral joint, 1111, 1111f, 1115, 1115f knee, 1775, 1777, 1777t, 1781–1783, 1782f, 1784t talus, 2146–2148, 2150–2152, 2150t, 2151f–2152f, 2153 procurement of, 138–139 sterilization of, 139 storage of, 140–141 ALPSA (anterior labroligamentous periosteal sleeve avulsion) lesion, 974, 974f Altitude. See High altitude AMBRI (atraumatic, multidirectional, bilateral, rehabilitation, inferior capsular shift) syndrome, 616 Amenorrhea, 478, 478f American Association of Tissue Banks, 137–139 American Heart Association, preparticipation examination recommendations of, 508, 509, 512t Amino acids, supplemental, 405, 409 Amitriptyline, in complex regional pain ­syndrome, 362, 363t Amlodipine, 160t Anabolic-androgenic steroids acne and, 415 adverse effects of, 415–417 athletic performance and, 414–415 body image disorders with, 417 cardiac effects of, 416 in children/adolescents, 467–468 complications of, 413–414 doses of, 415 efficacy of, 414 estrogen-related side effects of, 415–416 female use of, 412 hepatic effects of, 416–417 historical perspective on, 411–414 legal regulation of, 413 mechanism of action of, 414–415 musculoskeletal effects of, 415–416 opioid abuse and, 417 psychiatric effects of, 417 testicular effects of, 416 testing for, 413 Anaerobic exercise, 399–400 Analytical research, 107–108 Anaphylaxis, exercise-induced, 205, 530 Ancylostoma braziliense infection, 201 Androgen(s), 414. See also Anabolic-androgenic steroids adverse effects of, 415–417 athletic performance and, 414–415 mechanism of action of, 414 Androstenedione, 418 Aneurysm popliteal artery, 1838, 1844 throwing-related, 1226 Aneurysmal bone cyst, 983–984 Angiography in axillary artery injury, 1139, 1139f in hypothenar hammer syndrome, 1357, 1357f in knee dislocation, 1751, 1753f in popliteal artery entrapment, 1840, 1842, 1843f, 1844f in thoracic outlet syndrome, 1133 Angiotensin-converting enzyme inhibitors, in hypertension, 159, 160t Angiotensin receptor blockers, in hypertension, 159, 160t Anisocoria, preparticipation examination for, 509

Index Ankle. See also Foot (feet); Subtalar joint; Talus anatomy of, 338, 1865, 1866f arthroscopy of, 127, 127f in impingement, 2159–2160, 2159t, 2160f indications for, 127 normal anatomy on, 127, 127f portals for, 127, 127f positioning for, 127 biomechanics of, 1865–1867, 1865f–1867f, 2178–2179, 2178f, 2178t–2179t bone bruise of, 1929, 1931 brace for, 340, 1921–1922, 1921t, 1922f dislocation of. See also Ankle sprain with fracture, 1945, 1946f without fracture, 1945–1946, 1946f return to play after, 1946 foot linkage of, 1870–1872, 1872t fracture of dislocation with, 1945, 1946f pediatric, 597, 598f, 1964–1969 evaluation of, 1964 outcome of, 1966–1969, 1967t, 1968f, 1969f Salter-Harris type I, 1965, 1965f Salter-Harris type II, 1965, 1965f, 1968f Salter-Harris type III, 1965, 1965f, 1966f Salter-Harris type IV, 1965f, 1966, 1967f, 1969f stress, 2017–2018, 2017f��������������������� –�������������������� 2020f. See also Foot (feet), stress fracture of inferior extensor retinaculum of, 338 injury/instability of, 338–341, 1870–1873. See also Ankle sprain bracing in, 340 diagnosis of, 339 incidence of, 2174–2175, 2175t intra-articular pathology and, 339 mechanism of, 338–339 medial, 339 operative treatment of, 340–341 playing surface and, 2192–2204. See also Playing surface prevention of, 340 shoewear and, 2183–2192. See also Shoes, injury and taping in, 340 treatment of, 339–341 ligaments of, 338, 1912–1914, 1913f, 1914t, 1915f, 1935, 1935f, 1938–1940, 1939f. See also specific ligaments biomechanics of, 1914–1915, 1915t, 1935 injury to. See Ankle sprain magnetic resonance imaging of, 571, 573, 574f magnetic resonance imaging arthrography of, 537 motion of, 1912, 2178–2179, 2178f, 2178t, 2179t muscle function at, 1867, 1869, 1870f osteochondral lesions of, 2142–2153, 2142t. See also Osteochondrosis (osteochondroses), talar rehabilitation of. See Ankle rehabilitation stress fracture of, 2017–2018, 2017f–������������������ 2020f. See also Foot (feet), stress fracture of taping of, 340, 1934 tendons of. See Achilles tendon; Flexor ­hallucis longus tendon; Peroneus brevis tendon; Peroneus longus tendon; Tibial tendon Ankle brace, 340, 1921–1922, 1921t, 1922f Ankle-brachial index in knee dislocation, 1751 in popliteal artery entrapment, 1839

Ankle impingement, 2156–2161 anterior, 2157, 2157f anterolateral, 1931, 1931f–1932f, 2157, 2158f meniscoid lesion and, 1931, 1932f anteromedial, 2157–2158 evaluation of, 2157–2158, 2157f, 2158f, 2159t posterior, 2160, 2161t posteromedial, 2160 return to play after, 2161 treatment of, 2158–2160, 2159t, 2160f Ankle rehabilitation after bifurcate sprain, 1955 after dislocation, 1946 after lateral sprain, 1928–1929, 1928f–1929f, 1930f, 1931 after subtalar dislocation, 1953 after subtalar sprain, 1951 after syndesmosis sprain, 1944–1945 therapeutic exercise for, 272–276 eccentric training in, 275–276, 276f evertor muscle training in, 275 gastrosoleus training in, 273, 275f neuromuscular control training in, 273, 275, 275f neuromuscular training in, 273, 274f proprioceptive training in, 273, 275, 275f, 298 single-leg training in, 273, 275f Ankle sprain, 338–341, 1912–1974 in children/adolescents, 1963, 1964f grading of, 2216, 2217t high (syndesmosis), 1938–1945 anatomy of, 1938–1940, 1939f arthrography in, 1942 chronic, 1944 computed tomography in, 1942 direct eversion maneuver in, 1940, 1941f evaluation of, 1940, 1940f, 1941f external rotation stress test in, 1940, 1941f grade I, 1940, 1942, 1945 grade II, 1940, 1942, 1945 grade III, 1940, 1942–1944, 1943f, 1945 history in, 1940 latent, 1940 magnetic resonance imaging in, 1942, 1943f physical examination in, 1940, 1940f, 1941f radiography in, 1940–1942 radionuclide imaging in, 1942 rehabilitation for, 1944–1945 return to play after, 1945 squeeze test in, 1940, 1940f stress radiography in, 1942 treatment of, 1942–1944, 1943f lateral, 1912–1935 anatomy of, 1914–1915 anterior drawer stress radiography in, 1919, 1919f anterior drawer test in, 1916–1917, 1917f anterior lateral ankle impingement and, 1931, 1931f, 1932f arthrography in, 1919–1920 bifurcate ligament sprain and, 1918, 1933 biomechanics of, 1914–1915, 1915t bone bruise and, 1929, 1931 Broström procedure in, 1926–1928, 1927f Chrisman-Snook procedure in, 1925–1926, 1925f chronic (recurrent), 1924–1929, 1934 rehabilitation for, 1924 surgical treatment of, 1924–1928, 1925f, 1926f, 1927f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

iii

Ankle sprain (Continued) classification of, 1916, 1916t cold therapy in, 1920 distribution of, 1915, 1915t evaluation of, 1916–1918, 1916t Evans procedure in, 1924 fifth metatarsal base fracture and, 1933–1934 grade I, 1916, 1916t, 1920, 1934 grade II, 1916, 1916t, 1920, 1934 grade III, 1916, 1916t, 1920, 1934 history of, 1916 magnetic resonance imaging in, 1920, 1921f mechanism of, 1913, 1914f nerve palsy and, 1933 neural features of, 1915 osteochondral lesion and, 1929 peroneal tendon instability and, 1931–1933, 1932f physical examination in, 1916–1918, 1917f prevention of, 1934–1935 radiography in, 1918–1919, 1918f–1919f rehabilitation for, 1928–1929, 1928f, 1929f, 1930f, 1931f return to play after, 1934–1935 risk factors for, 1912 stress radiography in, 1918–1919, 1918f stress testing in, 1916 subtalar coalition and, 1933 talar tilt test in, 1917–1918, 1917f taping in, 1934 tibiofibular synostosis and, 1933, 1934f treatment of, 1920–1925, 1934 brace in, 1920, 1921–1922, 1922f cast immobilization in, 1920–1922, 1922t early mobilization in, 1922–1923 surgical, 1923–1924, 1923f untreated, 1915 Watson-Jones procedure in, 1924–1925, 1925f magnetic resonance imaging of, 571, 573, 574f, 1920, 1921f, 1937, 1942, 1943f medial, 339, 1935–1938 anatomy of, 1935, 1936f arthrography in, 1937 evaluation of, 1935–1936 grade I, 1937 grade II, 1937–1938 grade III, 1937–1938 history in, 1936 magnetic resonance imaging in, 1937 mechanism of, 1935 physical examination in, 1936 radiography in, 1936–1937, 1936f–1937f rehabilitation after, 1938 return to play after, 1938 treatment of, 1937–1938 vs. neoplasm, 2161–2162, 2161f prevention of, 338–341 talar osteochondral lesions and, 2144 vs. tarsal coalition, 2161, 2161f Ankle syndesmosis sprain. See Ankle sprain, high (syndesmosis) Anterior circumflex humeral artery, 1071–1072, 1072f Anterior cruciate ligament (ACL), 1644–1683. See also Anterior cruciate ligament (ACL) injury; Anterior cruciate ligament (ACL) reconstruction; Knee rehabilitation anatomy of, 1644–1646, 1645f–1648f, 1748 attachments of, 1644, 1646, 1646f biomechanics of, 1646–1647 during active range of motion, 1586, 1586f anteromedial bundle in, 1584, 1585–1586

iv

Index

Anterior cruciate ligament (ACL) (Continued) in arthroscopic meniscectomy, 1585–1586 autograft-related, 1584 during flexor-extensor exercises, 1586–1587 force measurements in, 1581–1582, 1585 gender and, 1584 with internal rotation, 1583, 1583f during isometric quadriceps contraction, 1586 during kinetic change exercises, 1586 during leg extension exercises, 1587 load-elongation curve in, 1583–1584, 1583f meniscectomy and, 1585–1586, 1590–1591 observational studies of, 1581 posterolateral bundle in, 1584 stabilizing function in, 1584–1588 strain measurements in, 1582, 1586–1588, 1586f, 1587t in weight-bearing flexion, 1585 blood supply to, 1645 collagen of, 1644–1645 fetal, 1645–1646, 1646f innervation of, 1645 insertion of, 1582 ligament unloading exercises for, 222–224, 223t loading of, 93f, 94–95, 1583–1584, 1583f exercise-related, 222t rehabilitative exercise effects on, 221–222, 222t squat exercise effects on, 223, 223t, 224, 224f, 263 stabilizing function of, 1584–1588 stair climbing effects on, 223t, 224 stationary cycling effects on, 223, 223t Anterior cruciate ligament (ACL) injury, 1644–1683 anterior drawer test in, 1650 biologic response to, 1647–1648 classification of, 1648, 2216 clinical presentation of, 1648–1649 cytokines in, 1648 epidemiology of, 1648 female athlete, 483–485 etiology of, 333–334 incidence of, 333 ligament reconstruction for, 485, 486t–487t mechanism of, 485, 488t prevention of, 333–334, 485, 487b risk factors for, 484–485, 484b functional testing in, 1649, 1652, 1674 history in, 1648–1649 KT-100 arthrometry in, 1650, 1652 Lachman test in, 1585–1586, 1649–1650, 2213 magnetic resonance imaging in, 553, 557f, 569, 571f, 572f, 1650–1651, 1653–1654, 1654t malalignment in, 1649 mechanism of, 1648–1649 medial collateral ligament injury with, 1632–1633, 1633f, 1634–1635, 1634f, 1635f, 1636, 1653, 1654 meniscal injury with, 1653–1654, 1653t epidemiology of, 1601 rehabilitation for, 1672–1673 treatment of, 1605–1606, 1654, 1654t meniscectomy and, 1590–1591 natural history of, 1654–1655 palpation examination in, 1649 partial, 1652–1653, 1652t pediatric, 1679–1683 iliotibial band tenodesis in, 1680

Anterior cruciate ligament (ACL) injury (Continued) incidence of, 1676–1677 nonoperative treatment of, 1679–1680 operative treatment of, 1680–1682, 1682f physical examination in, 1677 tibial eminence fracture with, 1677–1679, 1678f, 1679f physical examination in, 1649–1650 pivot shift test in, 1650, 2213–2214, 2214t posterolateral corner injury with, 1732, 1743 proprioceptive function after, 294. See also Knee rehabilitation, proprioceptive exercises in radiography in, 1650 swelling in, 1649 treatment of, 1655. See also Anterior cruciate ligament (ACL) reconstruction age and, 1651–1652 gender and, 1651 nonoperative, 1655 varus angulation with, 1801–1803, 1802t, 1803f clinical presentation of, 1803–1804 evaluation of, 1803–1814, 1805f, 1806t gait analysis in, 1808–1809, 1809f, 1810f, 1811f imaging and calculations in, 1809–1814, 1811f, 1812f, 1813f, 1814f, 1814t, 1815t physical examination in, 1804–1808, 1805f, 1806t, 1807f, 1808f treatment of, 1814–1835. See also High tibial osteotomy Anterior cruciate ligament (ACL) ­reconstruction, 1655–1676 arthroscopic examination for, 1662–1663, 1663f biology of, 1648 in children, 1681–1683, 1682f graft for, 38, 1656–1661, 1656t allograft, 142–143, 1656–1657, 1656t autograft, 37–38, 1584, 1656–1657, 1656t biology of, 1648 biomechanical properties of, 1584, 1656–1657 bone–anterior cruciate ligament–bone allograft in, 38 bone–patellar tendon–bone autograft, 37–38, 1584, 1656–1657, 1656t, 1659, 1659t, 1661, 1662f disease transmission with, 1657 donor site complications with, 1657 fixation of, 1658–1660, 1659t, 1660t, 1666–1668, 1667f, 1668f healing of, 1657 infection of, 391–393, 392t, 393t outcomes of, 1661–1662, 1662t placement of, 1666–1668, 1667f, 1668f preparation of, 1664, 1664f removal of, 393, 393b selection of, 1656, 1656t synthetic, 1655 tension of, 1657–1658 xenograft in, 38 insertion site marking for, 1663–1664, 1663f, 1664f in knee dislocation, 1758, 1759f ligament augmentation device for, 1655 medial collateral ligament injury repair with, 1632–1633, 1633f, 1634–1635, 1634f, 1635f, 1636 meniscal repair with, 1605–1606 outcomes of, 1661–1662, 1662f positioning for, 1662, 1663f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Anterior cruciate ligament (ACL) ­reconstruction (Continued) posterolateral corner injury and, 1743 rehabilitation after, 1668–1676, 1669t. See also Knee rehabilitation biofeedback in, 1671–1672 bracing in, 1672 cryotherapy in, 1670 electrical stimulation in, 230–233, 231f, 232f, 233f, 234f, 1671 functional testing in, 1674 kinetic chain exercise in, 1668–1670, 1669f muscle training in, 1671, 1674 proprioceptive training in, 1672, 1674 protocol for, 1672, 1673b range of motion in, 1670–1671 return-to-play plyometric training in, 300–321 contraindications to, 303 criteria for, 301–303, 302f glossary for, 324–330 stage I (dynamic stabilization and strengthening), 303–307, 304f, 305f, 306f, 322 stage II (functional strength), 307–312, 308f, 309f, 310f, 311f, 312f, 322–323 stage III (power development), 312–316, 313f, 314f, 316f, 323 stage IV (sport performance), 316–320, 317f, 318f, 319f, 320f, 324 RICE in, 1670 strain measurements for, 1586–1588, 1586f, 1587t unloading exercises in, 222–224, 223t weight-bearing in, 1671 return to play after, 301–303, 302f, 320–321, 1674, 1675t giving away and, 301 IKDC subjective rating and, 301 screw for, 1659 strain gauge analysis after, 1586f, 1587–1588 tibial osteotomy with, 1792, 1814–1816, 1815t, 1833, 1835. See also High tibial osteotomy results of, 1817, 1825–1830, 1828f, 1829f timing of, 1656 tunnel placement for, 1664–1666, 1665f, 1666f, 1667f Anterior drawer stress radiography, in lateral ankle sprain, 1919, 1919f Anterior drawer test in anterior cruciate ligament injury, 1650 in lateral ankle sprain, 1916–1917, 1917f in medial ankle sprain, 1936 in medial collateral ligament injury, 1629, 1630t Anterior interosseous nerve syndrome, 1317, 1317f Anteromedial tibial tuberosity transfer, in patellar dislocation, 1567 Antibiotics, 386–387 in erythrasma, 194t, 196 in folliculitis, 194, 194t, 195 in impetigo, 194, 194t in Lyme disease, 155 in septic olecranon bursitis, 1213 Anticoagulation physiologic, 372, 373f prophylactic, 378–384, 379t–381t in venous thromboembolism, 384–385, 384t Anticonvulsants, 188–189 cognitive effects of, 190, 191t in complex regional pain syndrome, 363t after head injury, 661

Index Antidepressants, in complex regional pain syndrome, 362, 363t Antidiuretic hormone, exercise effects on, 217, 217t Antihypertensive drugs, 158–161, 160t Antithrombin III, 372, 373f deficiency of, 372, 374t Anulus fibrosus, 717. See also Intervertebral disk biomechanics of, 718 tear in, 740–741 Aorta coarctation of, 158 rupture of, 167 Aortic insufficiency, 158 Apley’s test, 1601–1602 Apnea definition of, 448 sleep, 448–449 Apophysitis calcaneal, 1973–1974, 1974f, 2053–2054, 2143f, 2162, 2162f fifth metatarsal, 2167–2169, 2168f–2169f iliac, 1475 tibial tubercle, 599, 1526f, 1527–1529 Apprehension test in glenohumeral joint instability, 914, 915f, 939, 939f, 941 in patellar dislocation, 1538, 1539f, 1558 in rotator cuff disorders, 996, 999f in SLAP lesion, 1025 Arcuate ligament, 1585, 1588–1589, 1723 Arginine, 408, 421 Arginine α-ketoglutarate, 421 Arm. See also Elbow; Glenohumeral joint; Wrist anatomy of. See also specific structures muscular, 1157–1158, 1158f, 1160f–1161f neurovascular, 1157, 1159–1161, 1160f–1161f osseous. See Humerus; Radius; Ulna Arm curls, 88f, 213, 213f, 252, 252f Arrhythmias, inhalant-related, 430 Arrhythmogenic right ventricular ­cardiomyopathy, 166, 167f Arteriography. See Angiography Artery, 371f. See also specific arteries Arthritis degenerative. See Osteoarthritis patellofemoral, in female athlete, 489 pediatric inflammatory, 602–603, 602f septic, 602 rheumatoid. See Rheumatoid arthritis Arthrodesis first metatarsophalangeal joint, 2075 lunotriquetral, 1331, 1331f in sternoclavicular joint dislocation ­treatment, 810 triple, in posterior tibial tendon injury, 1980 Arthrofibrosis after high tibial osteotomy, 1832 knee rehabilitation and, 225 Arthrography, 535–537. See also Computed tomography (CT) arthrography; Magnetic resonance arthrography (MRA) ankle, 537 in high (syndesmosis) sprain, 1942 in lateral sprain, 1919–1920 in medial sprain, 1937 in subtalar sprain, 1950 hip, 536–537 knee, 536, 537f shoulder, 535–536, 536f, 949–950, 949f, 1000 wrist, 536

Arthropathy capsulorrhaphy, 1104b, 1105–1106, 1105t, 1114 dislocation, 1104–1105, 1104b Arthroplasty elbow for arthrosis, 1278 in distal humeral fracture nonunion, 1258 heterotopic ossification after, 1290, 1291f hip, 1499–1512 anatomy for, 1500–1502 approaches to, 1505 bearings for, 1505–1506 biomechanical aspects of, 1501–1502 high-offset femoral component in, 1502 minimally invasive, 1505 return to play after, 1507–1508, 1508t stability with, 1502 technique of, 1508–1512, 1509f, 1510f, 1511f, 1512f infection with, 394–395, 394f, 394t interphalangeal, 2121–2122, 2123b, 2123f–2124f knee, 1787–1801 athletic activity after, 1787–1789, 1789b clinical evaluation for, 1789–1790 imaging for, 1789–1790, 1790f mobile-bearing prostheses in, 1797–1798 vs. nonoperative treatment, 1790–1792 patient history and, 1789 physical examination in, 1789 quadriceps rupture with, 1521 total, 1794–1797, 1795t, 1796f, 1797f bearing surfaces in, 1800 gender-specific, 1799 high-flexion, 1798–1799, 1798f, 1799f indications for, 1794 minimally invasive, 1799–1800, 1800f prior proximal tibial osteotomy and, 1792 results of, 1795, 1795t unicompartmental, 1792–1794, 1793f metatarsal head, 2124, 2130f shoulder in fracture, 1046, 1048f in instability arthropathy, 1114 in osteoarthritis, 1110, 1113–1114, 1113f, 1114f, 1115–1118, 1116f, 1116t, 1117f, 1118f in rheumatoid arthritis, 1114 simulation of, 1151, 1153, 1153f–1155f Arthroscopy, 121–131 in acromioclavicular joint reconstruction, 844 ankle, 127, 127f in impingement, 2159–2160, 2159t, 2160f indications for, 127 normal anatomy on, 127, 127f portals for, 127, 127f positioning for, 127 in biceps tendon evaluation, 1009–1010 complications of, 122–124 deep vein thrombosis with, 123–124, 124b, 383–384, 383t elbow, 129–130, 129f in capitellar osteochondritis dissecans, 1243–1244, 1243f, 1245f in heterotopic ossification, 1296–1297 indications for, 129 in lateral epicondylitis, 1203–1204 normal anatomy on, 130, 130f in olecranon bursitis, 1212 portals for, 129–130, 129f positioning for, 129 in recurrent instability, 1307–1310, 1309f in valgus extension overload, 1224–1225

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.



Arthroscopy (Continued) equipment for, 121–122, 122f, 123f complications with, 123 sterilization of, 121 hip, 125–126, 126f, 1473–1474, 1473t, 1474f in acetabular labral tears, 1470, 1470f complications of, 1474 in femoroacetabular impingement, 1472 indications for, 125–126, 1473, 1473t in ligamentum teres rupture, 1473, 1473f in loose body removal, 1472, 1472f normal anatomy on, 126, 126f portals for, 126, 126f positioning for, 126 technique of, 1473–1474, 1474f image interpretation in, 122, 122f intra-articular injury with, 122–123 irrigation during, 122, 123f knee, 124–125, 124f, 125f in abrasion procedure, 53 in anterior cruciate ligament ­reconstruction, 1662–1668, 1663f, 1664f, 1665f, 1666f, 1667f. See also Anterior cruciate ligament (ACL) reconstruction in arthritis, 1792 deep vein thrombosis with, 123–124, 124b indications for, 124 in medial collateral ligament injury, 1632–1633, 1633f, 1634–1635, 1634f, 1635f in meniscal injury, 1602–1603, 1606, 1606f, 1618 normal anatomy on, 125, 125f in patellar dislocation, 1545, 1570–1572 portals for, 124–125, 125f positioning for, 124, 124f in posterior cruciate ligament and ­posterolateral corner injury, 1706–1710, 1707f in posterior cruciate ligament ­reconstruction, 1705, 1706f, 1707f in posterolateral corner injury, 1706–1710, 1707f, 1730–1731, 1732f in varus malalignment, 1808, 1808f pediatric in elbow injury, 468 in hip injury, 469, 470f in shoulder instability, 468, 469f in tibial spine fracture, 469, 472f in wrist injury, 468, 470f positioning for, 122 in scapulothoracic bursectomy, 890–891, 891f shoulder, 127–129, 128f, 1147–1151 in adhesive capsulitis, 1096–1102, 1097f–1102f, 1103b in anterior glenohumeral joint instability, 919–923, 921f–923f, 927–929 in anterior labral repair, 1151, 1152f in glenohumeral joint osteoarthritis, 1110–1111, 1115, 1115f, 1118, 1118t–1119t in glenoid detachment, 1220, 1220f indications for, 127 knot tying for, 1147, 1148f learning methods for, 1147, 1149f in multidirectional glenohumeral joint instability, 930–931, 930f normal anatomy on, 128–129, 129f portals for, 128, 128f positioning for, 127–128, 128f in posterior glenohumeral joint instability treatment, 924–926, 926f–927f in rotator cuff repair, 1009–1010, 1149, 1150f, 1151, 1218–1219, 1219f

vi

Index

Arthroscopy (Continued) in SLAP lesion, 1025, 1026–1031, 1027f–1031f suite for, 121 tourniquet effect during, 123 wrist, 130–131, 130f, 1430–1450 in children, 468, 470f in chondral lesions, 1447–1449, 1448f distal radioulnar joint portals for, 1434–1435 in dorsal ganglion cyst, 1444–1445, 1445f equipment for, 1430–1431, 1431f indications for, 130, 1430 in ligament instability, 1434, 1434t, 1442–1443, 1443f in ligament tears, 1442–1443, 1443f in loose bodies, 1447 normal anatomy on, 131, 131f in ostectomy, 1449–1450 portals for, 130–131, 130f, 1431–1435, 1431f 1-2, 1431, 1431f 3-4, 1431–1433, 1431f, 1432f 4-5, 1431f, 1433, 1433f 6-R, 1431f, 1433, 1433f 6-U, 1431f, 1433 distal radioulnar joint, 1434–1435 radial midcarpal, 1434, 1434f, 1434t scaphoid trapezium trapezoid, 1434, 1435f ulnar midcarpal, 1434 volar, 1434 positioning for, 130 in proximal row carpectomy, 1449–1450 radial midcarpal portal for, 1434, 1434f, 1434t in radial styloidectomy, 1449 scaphoid trapezium trapezoid portal for, 1434, 1435f in scapholunate ligament injury, 1325 in synovectomy, 1443–1444 in triangular fibrocartilage complex tears, 1435–1442. See also at Triangular fibrocartilage complex (TFCC) tears ulnar midcarpal portal for, 1434 in ulnar styloid impaction syndrome, 1445–1447, 1446f volar portal for, 1434 Arthrosis acromioclavicular joint, 829–830 elbow joint, 1278 glenohumeral joint, 929 Arthrotek tensioning boot, 1760, 1761f, 1762 Articular cartilage. See Cartilage, articular Artificial turf, 1210 Ascorbic acid (vitamin C) deficiency of, 72–73, 73t requirements for, 406b Askin’s tumor, 610 Aspiration (needle) of cyst, 585 in olecranon bursitis, 1248, 1248f Aspirin, in venous thromboembolism prevention, 378–384, 381t Asthma. See Bronchospasm Atenolol in complex regional pain syndrome, 363t in hypertension, 160t Athlete-ready position, 313f, 325 Athlete’s foot, 1964 Athletic pubalgia, 1463–1464 vs. adductor strain, 1490 Athletic trainer, 530 Atlanto-dens interval, 694, 695f in Down syndrome, 705, 707b

Atlanto-occipital fusion, 693 Atlanto-occipital instability, in children/ adolescents, 704–705, 705f Atlantoaxial complex, 677f Atlantoaxial instability, 514t, 677, 677f, 694, 695 in children/adolescents, 705, 707f Atlantoaxial subluxation, rotary, in children/ adolescents, 706–707, 708f Atlas (C1) fracture of, 677, 677f, 678f, 695–696, 705, 706f occipitalization of, 711 posterior arch absence in, 711, 711f ATPase, in muscle contraction, 8–9, 9f Atrophy, in complex regional pain syndrome, 358 Aura, with seizure, 186 Autograft(s). See also Graft(s) in anterior cruciate ligament reconstruction, 37–38, 1584, 1656–1657, 1656t, 1659, 1659t, 1661, 1662f articular cartilage, 53–54 bone, 81, 83t ligament, 36–38 fascia lata, 37 iliotibial band, 37 patellar tendon, 37 semitendinosus tendon, 37–38 osteochondral in knee, 1775, 1777, 1780–1781, 1781f in talus, 2146–2147, 2148, 2150–2152, 2150t, 2151f–2152f, 2153 periosteum, 53–54 Automatic external defibrillator, in sudden death, 171–172, 171t Autonomic nervous system, 353 in complex regional pain syndrome, 353–355, 354f, 356–357, 357f, 360–361 Avascular necrosis, 73, 77 femoral head, 1466–1467, 1467b, 1467t in children, 1476–1477 slipped capital femoral epiphysis and, 1476 histology of, 77 humeral head, 1049, 1105t, 1106–1107, 1107f arthroscopic treatment of, 1110–1111 magnetic resonance imaging of, 984, 985f physeal injury and, 1085 sesamoid, 2089 Avulsion fracture anterior process of calcaneus, 1954, 1954f, 2153–2156, 2154t, 2155t hip and pelvis, 553, 555, 1474–1475, 1475f, 1489 iliac spine, 553, 556f, 1475 imaging of, 553, 555, 556f, 557, 557f ischial, 1489, 1489f lesser humeral tuberosity, 1175–1176 lesser trochanter, 1475 rib, 895–896, 896f Axial load test in glenohumeral joint instability, 914–915 in rotator cuff disorders, 996, 999f Axillary artery anatomy of, 911–912, 1036f, 1071–1072, 1072f arteriography of, 1139, 1139f injury to, 1140–1141, 1141f anterior shoulder dislocation and, 1139–1140, 1140f, 1141f Axillary nerve anatomy of, 912, 1034, 1036b, 1036f, 1161f evaluation of, 1036–1037, 1038f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Axillary nerve (Continued) injury to in glenohumeral joint instability, 933, 934f paralabral cyst and, 980, 980t in proximal humeral fracture, 1049 quadrilateral space compression of, 1142 Axillary vein, effort thrombosis of, 1130, 1134, 1141–1142 Axis (C2), fracture of, 678, 694, 696, 707–708 Axis of rotation, 1190, 1191f Azoospermia, anabolic-androgenic steroids and, 416

B Back pain. See Low back pain Back squat, shoulder injury and, 249 Bagging, 430 Balance training, single-leg, in knee ­rehabilitation, 271 Ball exercises, in trunk stabilization, 342t, 345, 347–348 Band good morning exercise, 316f, 325 Band walking exercise, 258, 259f Bankart lesion, 616, 912, 933, 934f. See also Glenohumeral joint instability, anterior arthroscopic treatment of, 919–923, 920f, 921f–923f in children, 468, 470f, 940–941, 940f, 941f magnetic resonance arthrography of, 973–974, 973f, 974t magnetic resonance imaging of, 969, 969f, 973 radiography of, 968–969 reverse, 936, 942, 942f, 974t arthroscopic treatment of, 924–926, 926f–927f, 942, 942f in children, 942, 942f magnetic resonance arthrography of, 975, 975f variants of, 974, 974f, 974t Barefoot running, 1874–1875, 1874f, 2183–2184, 2183f Baseball. See also Overhead throwing circadian rhythms and, 457 pediatric. See also Little Leaguer’s elbow; Little Leaguer’s shoulder humeral fracture in, 1068, 1070f Baseball finger, 1388, 1420–1422, 1420f, 1421f Basketball, jet lag and, 458 Bathing, epilepsy and, 192 Battle’s sign, 525 Bayes’ network, 114 Bayonet sign, 1554 Bean bucket wrist and forearm training, 255, 255f Bear paw shoe, 1874, 1874f Behavior, anabolic-androgenic steroid effects on, 417 Behind-the-neck training, shoulder injury and, 248–249 Benazepril, 160t Bench press, shoulder injury and, 247–248, 248f Benign rolandic epilepsy, 187 Bennett’s fracture, 1402–1403, 1402f, 1403f, 1411, 1412 Bent leg side bridge exercise, in core training, 284, 284f Benzodiazepines, in complex regional pain syndrome, 363t, 364 Betazolol, 160t Bethesda Guidelines, in sudden death ­prevention, 169, 170t Bias, research, 100

Index Biceps brachii anatomy of, 858, 858f, 859f, 1157–1158, 1158f, 1160f pediatric, 785 long head of, tendon of. See Biceps tendon short head of, transfer of, 840–841, 841f Biceps curl, 88f, 213, 213f, 252, 252f Biceps femoris, 1485, 1485f, 1721–1722. See also Hamstring muscles Biceps load test, in SLAP lesion, 1025t Biceps tendon anatomy of, 771–772, 911, 989, 990, 991f, 992, 992f, 1018, 1018f arthroscopic evaluation of, 1009–1010 biomechanics of, 992, 992f, 1019–1021, 1019f, 1020b, 1021f detachment of. See SLAP (superior labrum, anterior to posterior) lesion distal rupture of, 1167–1170 classification of, 1168, 1168t evaluation of, 1168 imaging of, 1168, 1169f rehabilitation after, 1169–1170, 1170t treatment of, 1168–1170 fraying of, 1220 histology of, 1018, 1018f hourglass, 981, 982f magnetic resonance imaging of, 980–983, 981f, 982f pediatric, 790 popeye deformity of, 1165, 1166f rupture of, 1165–1167, 1166f, 1167t scapular attachment of, 858, 858f subluxation/dislocation of, 772, 981–983, 982t, 1009 evaluation of, 997 superior labral attachment of, 1018, 1018f tears of, 981, 992 clinical manifestations of, 996, 996f, 997, 1000f magnetic resonance imaging in, 566, 568f Biceps tenodesis, 1733–1734, 1736f Bicipital groove, 771–772, 1034t palpation of, 996, 997f Bifurcate ligament, 1913f, 1953, 1954f Bifurcate sprain, 1918, 1933, 1953–1955 calcaneal fracture and, 1954, 1954f evaluation of, 1954, 1954f rehabilitation for, 1955 return to play after, 1955 treatment of, 1955 Biglycan, 44 Binge drinking, 424–425 Biofeedback, 224–225, 233, 234f in anterior cruciate ligament rehabilitation, 1671–1672 in muscle atrophy, 224–225 with stretching exercises, 292f, 293 Biological rhythms. See Chronobiology Biomechanics, 85–96. See also at specific joints and structures angular kinematics in, 90, 91f degrees of freedom in, 89 dynamics in, 87, 89–93 force vectors in, 86, 90f free-body diagrams in, 87, 88f joint contact forces in, 87 joint motions in, 90, 93f kinematics in, 87, 90, 91f kinetics in, 90–91 linear kinematics in, 90, 91f loading-related, 93–95, 93f–95f mechanical properties in, 93–95, 93f–���������� 95f moment/torque vectors in, 86 Newton’s laws in, 86–87, 89t, 92–93

Biomechanics (Continued) relative motion in, 89–90 scalars of, 86 statics of, 86–87 structural properties in, 93, 94f units of measurement in, 85, 85t vectors of, 86 viscoelasticity in, 95–96, 95f Biomechanics of Distance Running (Cavanagh), 1896 Biopsy, CT-guided, 585, 586f Bioscaffolds, in tendon healing, 28 Bipartite patella, 1574 Bipartite scaphoid, 1366 Bipartite sesamoid, 2028, 2029f, 2088 Bird dog exercise, in core training, 282–283, 283f Bisoprolol, 160t Bisphosphonates in complex regional pain syndrome, 363t, 365 in tibial stress fracture, 1854 Bitter orange, 409 Black heel, 202, 202f Blackburne-Peel ratio, for patella height, 1523, 1523f, 1539, 1541f Bleeding. See Hematoma; Hemorrhage Bleeding disorder, 514t Blinding, statistical, 100, 104–105 Blisters, foot, 1964 shoe fit and, 2184–2185 Blood alcohol concentration of, 424, 449–450 venous stasis of, 371. See also Thrombosis Blood flow in complex regional pain syndrome, 361, 365 exercise effects on, 219–220 Blood pressure classification of, 156, 157t exercise effect on, 219–220, 220t measurement of, 157, 157b preparticipation examination for, 512, 513 Blount’s disease, 599 Blumensaat line, 1539, 1541f Body Blade training, in shoulder rehabilitation, 245, 246f Body clock. See Chronobiology Body image disorders, anabolic-androgenic steroids and, 417 Boils, 194–195, 194t Bone(s), 65–70. See also at specific bones age-related changes in, 72, 78 anabolic-androgenic steroid effects on, 416 biomechanics of, 77–78 blood supply of, 69, 69f, 70f calcification of, 67, 68f calcitonin effects on, 71, 71t calcium of, 70, 71t cancellous (spongy, trabecular), 65, 66f remodeling of, 68 cells of, 65–67, 67t cortical (compact), 65, 66f, 67t remodeling of, 68, 68f corticosteroid effects on, 72 cyst of radiography in, 552, 553f unicameral, 606, 606f fracture through, 552, 553f, 606, 606f, 1085, 1086f death of. See Avascular necrosis density of decrease in, 72–73 in female athlete, 478, 481 development of, 587–588, 588f ectopic. See Heterotopic ossification

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

vii

Bone(s) (Continued) endochondral, formation of, 78 estrogen effects on, 72 formation of, 78 distraction-induced, 81–82, 83f endochondral, 78 intramembranous, 78 growth factors of, 79–80 growth hormone effects on, 72 healthy, 406 infection of. See Osteomyelitis injury to. See Fracture(s) and at specific bones intramembranous, formation of, 78 matrix of, 67, 68f mechanical properties of, 77–78 mineral metabolism of, 67, 70–71, 71t disorders of, 72–73, 73t, 74t–75t, 76t–78t parathyroid hormone effects on, 71, 71t pathologic, 66f phosphate of, 70, 71t physis of, 78, 587–588 proteoglycans of, 67 remodeling of, 67–68, 68f, 79 thyroid hormone effects on, 72 tumor of, 74t, 77t, 547, 547f types of, 65, 66f, 67t vitamin D effects on, 71, 71t Bone bruise, lateral ankle sprain and, 1929, 1931 Bone cyst radiography in, 552, 553f unicameral, 606, 606f fracture through, 1085, 1086f Bone graft, 81, 83t allograft, 81, 83t, 137 autograft, 81, 83t cancellous, 81, 83t cortical, 81, 83t osteoarticular (osteochondral), 81 vascularized, 81 in Kienböck’s disease, 1377 in scaphoid nonunion, 1341–1345, 1342f–1344f Bone marrow, 69–70 Bone mineral density decrease in, 72–73 in female athlete, 478 Bone morphogenetic proteins in fracture healing, 79–80 in heterotopic ossification, 1289 Bone scan. See Radionuclide imaging Bone–anterior cruciate ligament–bone allograft, 38 Bone–patellar tendon–bone autograft, 1584, 1656–1657, 1656t, 1659, 1659t, 1661, 1662f Bony humeral avulsion of glenohumeral ligament (BHAGL) lesion, 555 Borrelia burgdorferi infection, 155–156 BOSU exercises, 305f, 308f, 315f, 325–327 Botox injection, in lateral epicondylitis, 1200 Bounding exercise, 315f, 327, 328 Boutonnière injury/deformity, 1388–1389, 1388f, 1389f pediatric, 1418, 1419f, 1422–1423, 1422f, 1424f pseudo, 1389–1390 treatment of, 1389, 1389f Bowstring sign, 722 Box drop-off exercises, 327 Box step-down exercise, 308f, 324 Boxer’s cast, 1413–1414, 1413f Boxer’s fracture, 1393–1394, 1393f, 1412–1413, 1413f Boxer’s knuckle, 1390

viii

Index

Brace/bracing in ankle instability, 340 in anterior cruciate ligament rehabilitation, 1672 in capitellar osteochondritis dissecans, 1242 in elbow dislocation, 1269 in knee osteoarthritis, 1790 in lateral ankle sprain, 1921–1922, 1921t, 1922f in lateral epicondylitis, 618–619, 618f, 1200–1201, 1201f in patellar dislocation, 1547, 1566 in posterior cruciate ligament rehabilitation, 1711 in tibial stress fracture, 1854 Brachial artery, 1159, 1160f, 1161f Brachial neuritis. See Parsonage-Turner ­syndrome Brachial plexus anatomy of, 671f, 1034, 1036f injury to, 670–673, 671f prevention of, 673, 673f, 674f Brachialis muscle, 1157–1158, 1158f Brain, 657 injury to. See Head injury Brainstem, 657 Breast cancer of exercise and, 481 radionuclide imaging in, 547, 548f injury to, 479 Breathing, assessment of, 519, 520t, 521f Bridging exercise in anterior cruciate ligament rehabilitation, 308f, 326 in core training, 281–284, 281f–����������������������� 282f, 285f in knee rehabilitation, 258, 258f, 308f, 326 in trunk stabilization, 342t, 344 unilateral, in core training, 281f Bright light therapy in jet lag, 458–459, 458f, 459t in seasonal affective disorder, 444–445 Brisement therapy, in retrocalcaneal bursitis, 2035 Broad jump, 314f, 324–325, 327 Bronchitis, 149–150 Bronchodilators, in exercise-induced ­bronchospasm, 182 Bronchospasm, exercise-induced, 180–185 clearance for participation and, 513 clinical manifestations of, 181, 181b complications of, 185 definition of, 180, 181t diagnosis of, 181–182, 183f differential diagnosis of, 181, 181b evaluation of, 182 on-field, 526 prevalence of, 180, 181t return to play criteria for, 184f, 185 risk for, 180–181 treatment of, 182–184, 183f nonpharmacologic, 183f, 184 pharmacologic, 182–183, 182t, 183f sideline, 184–185, 184b, 184f Broström procedure, 1926–1928, 1927f Brown-Séquard syndrome, 710 Bucket handle fracture, in child abuse, 595–596, 596f Buford complex, 575, 972, 972f, 1017–1018, 1017f Bulbocavernous reflex, 675 Bulls-eye lesion, in quadriceps strain, 1494, 1495f Bundle ridge, 1646, 1647f Bunion, 1962–1963

Bunionettes, 2132–2142 anatomy of, 2132, 2132b, 2133f in children, 1963 classification of, 2133, 2134b evaluation of, 2133–2134, 2134b imaging in, 2134, 2134b nonoperative treatment of, 2134, 2134b operative treatment of, 2134–2141 care after, 2139–2140, 2140f, 2141f diaphyseal metatarsal osteotomy in, 2135–2136, 2137, 2138b, 2139f, 2140f distal chevron osteotomy in, 1966f, 2135–2136, 2135b, 2137b distal metatarsal osteotomy in, 2135–2136, 2135b distal oblique osteotomy in, 2135–2136, 2135b, 2136b, 2136f, 2137f lateral condylectomy in, 2134–2135, 2134b, 2135f, 2140, 2140f return to play after, 2141 physical examination in, 2134, 2134b Burner syndrome, 670–673, 710–711 vs. acute herniated disk, 704, 711 cervical stenosis and, 671–672 in children/adolescents, 710–711 clearance for participation and, 513 prevention of, 673, 673f vs. spinal cord injury, 524 Burning hands syndrome, 710 Bursa (bursae) infraserratus, 886, 886f, 887b olecranon, 1246–1247 retrocalcaneal, 2031, 2031f, 2032f. See also Retrocalcaneal bursitis scapular, 886, 887b, 887f scapulotrapezial, 886, 887b, 887f subacromial, 990, 991f subscapularis, 771 supraserratus, 886f, 887b trapezoid, 886 Bursectomy olecranon, 1248 in rotator cuff tear repair, 1013 scapulothoracic, 890–891, 891f Bursitis iliopectineal, 1458 iliopsoas, 1457–1458, 1457f ischial, 1457 olecranon, 1209–1212, 1246–1249. See also Olecranon bursitis retrocalcaneal, 2030–2041. See also Retrocalcaneal bursitis scapulothoracic, 889–891, 891f trochanteric, 1455–1457, 1456f Bursography in iliopsoas bursitis, 1457, 1457f in snapping hip syndrome, 1458 Bursoscopy, scapulothoracic, 890–891, 891f Burst fracture atlas, 677, 677f–678f, 705, 706f C4, 681, 681f lumbar spine, 541f, 735, 735f Buschke’s disease, 2143f, 2166, 2166f γ-Butyrolactone, 409

C C protein, 4, 5t Caffeine, 401, 409, 421, 451–452, 453t Calcaneal angle, 2035, 2035f Calcaneal apophysitis (Sever’s disease), 1973–1974, 1974f, 2143f, 2162, 2162f treatment of, 2053–2054 Calcaneal nerve, medial, 2043, 2044f, 2045f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Calcaneocuboid joint biomechanics of, 1869, 1870f impairment of, 1872 Calcaneofibular ligament, 338, 1913–1914, 1913f, 1914f, 1915f biomechanics of, 1865–1866, 1866f injury to. See Ankle sprain, lateral; Subtalar sprain repair of, 1923–1924, 1923f Calcaneus anterior process of, avulsion fracture of, 1954, 1954f, 2153–2156, 2154t, 2155t fracture of, 2153–2156, 2154t, 2155t return to play after, 2156 posterior-superior pain in. See Retrocalcaneal bursitis stress fracture of, 556f, 646, 647f, 2016t, 2019–2020, 2022f Calcification. See also Heterotopic ossification; Myositis ossificans periarticular, 1289 Calcitonin in bone metabolism, 71, 71t in complex regional pain syndrome, 363t, 364–365 Calcium bone, 70, 71t deficiency of, cramps and, 12 in muscle contraction, 208, 209f requirement for, 70, 406, 406b in female athlete, 478, 478t serum, 74t–76t, 77t urinary, 74t–76t Calcium channel blockers in complex regional pain syndrome, 363t, 364 in hypertension, 160t, 161 Calcium hydroxyapatite, 67 Calf raise exercise, in ankle rehabilitation, 273, 275f Call to Action against Underage Drinking, 425 Callus bony, 79, 80f, 81t stratum corneum, 2108. See also Plantar keratoses shoe fit and, 2184–2185, 2184f Calories, requirements for, 402–403 Calorimetry, indirect, 212–213, 212f Campylobacter fetus infection, 397 Cancer, 514t bone, 74t, 77t, 547, 547f breast, 481, 547, 548f oral, smokeless tobacco and, 428 positron emission tomography in, 547 prostate, anabolic-androgenic steroid effects on, 416 Candesartan, 160t Candida infection, 397, 398t Cane, in knee osteoarthritis, 1790 Cannabis use/abuse, 425–426 Capitate anatomy of, 1319–1320, 1319f avascular necrosis of, 1375 fracture of adult, 1348–1349 pediatric, 1368 Capitellum fracture of, 1251, 1251f–1252f fixation of, 1256, 1257f ossification of, 1227–1228 osteochondritis dissecans of. See Osteochondritis dissecans, capitellum osteochondrosis of, 623, 625f, 1238 Capitolunate angle, 1322, 1322f Capsaicin, in complex regional pain syndrome, 364

Index Capsular ligament, 793–794, 793f, 794f Capsular shift procedure, in glenohumeral joint instability, 927, 945 Capsulorrhaphy arthropathy, 1104b, 1105–1106, 1105t, 1114 Captopril, 160t Carbamazepine in complex regional pain syndrome, 363t in epilepsy, 188, 191t Carbohydrate(s) after exercise, 404 before exercise, 404 during exercise, 404 metabolism of, 211 requirement for, 403–404, 403t Carbohydrate loading, 403–404 Carbonated beverage, 401 Cardiac arrest, 526. See also Sudden death Cardiac output, exercise effect on, 219, 219f, 220t Cardiomyopathy anabolic-androgenic steroids and, 416 dilated, 152 hypertrophic, 164–165, 164f, 170 right ventricular, arrhythmogenic, 166, 167f Cardiopulmonary resuscitation equipment for, 517, 518t guidelines for, 520, 520t, 521f Cardiovascular disease, preparticipation ­examination for, 509, 512, 512t, 514t Cardiovascular system anabolic-androgenic steroid effects on, 416 cocaine effects on, 429 exercise effects on, 218–220 hypothermia effects on, 501 inhalant effects on, 430 marijuana effects on, 426 nicotine effects on, 428 Carpal tunnel syndrome, 626, 1359–1361 clinical manifestations of, 1360 electrodiagnostic study in, 1361 physical examination in, 1360 return to play after, 1361 treatment of, 1360–1361, 1360f Carpometacarpal joint anatomy of, 1386–1387 dislocation of, 534f, 1386–1387, 1387f thumb, dislocation/subluxation of, 1398 Carpometacarpal ligaments, 1320 Carpus. See Wrist Carrying angle, 1190, 1192 Cartilage, articular, 40–55. See also specific joints age-related changes in, 40 biomechanics of, 48 calcified zone of, 41f, 42f, 46 cell-matrix interactions in, 45, 45f chondrocytes of, 41, 41f, 45, 45f, 47f collagen of, 42–43, 43f composition of, 40–42, 41f creep of, 48 deep zone of, 41f, 42f, 46 extracellular matrix of, 41–42, 42f, 45, 45f for graft, 55 regions of, 46–48, 47f fluid of, 41 glycoproteins of, 45 immobilization effects on, 228 injury to, 48–52, 49t. See also at specific joints arthroscopic abrasion for, 52–53 artificial matrix grafts for, 55 blunt trauma, 50–51 cartilage shaving for, 52 chondrocyte implantation for, 55 fracture, 49t, 51–52 grafts for, 53–55

Cartilage, articular (Continued) laceration, 50 matrix, 48, 49–50, 49t proteoglycan effects of, 44 subchondral bone abrasion for, 52–53 surgical treatment of, 52–53 tissue disruption, 49t, 50–51 interterritorial matrix of, 47f, 48 laceration of, 50 magnetic resonance imaging of, 576–582, 578f in children, 592–593 at hip, 581–582 at knee, 579–580, 579f–581f at shoulder, 580–581, 582f, 583f noncollagenous proteins of, 45 pediatric, 592–593 pericellular matrix of, 46, 47f proteoglycans of, 41, 42, 43f, 44–45 loss of, 49, 49t repair of, 48–52, 49t repetitive impact loads on, 50–51 shaving of, 52 stress relaxation of, 48 structure of, 41f–42f, 45–48 superficial zone of, 41f–42f, 46 territorial matrix of, 46, 47f transitional zone of, 41f, 42f, 46 viscoelastic properties of, 48 zones of, 41f, 42f, 46, 579–580 Cartilage graft(s), 53–55 allograft, 54 artificial matrix, 55 autograft, 53–54 perichondrial, 53–54 periosteal, 53–54 Cast/casting in Achilles tendon rupture, 2005 in calcaneal fracture, 2156 in lateral ankle sprain, 1920–1921, 1921t playing, 1362 in proximal humeral physeal fracture, 1076–1077 in talar fracture, 2156 Catecholamines, 353 in complex regional pain syndrome, 354, 354f exercise effects on, 217–218, 217t, 218f receptors for, 353 Caton-Deschamps ratio, 1539, 1541f Cauda equina, 717–718 Cauda equina syndrome, 743 Causalgia. See Complex regional pain syndrome Causation, 100–101 Cavus foot, 1963, 2183 in children/adolescents, 1963 shoe for, 1900 stress fracture and, 1850 Cefazolin, 386, 387t Cefuroxime, 387t Celecoxib, in tendon healing, 29 Cellulitis, 387–388, 387f, 547 Central nervous system, 353 cocaine effects on, 429 in force production, 7 inhalant effects on, 430 injury to. See Cervical spine injury; Head injury; Spinal cord injury Cephalosporins, 386, 387t Cerebellum, 657 Cerebral edema, high-altitude, 504–505 Cerebral palsy, 514t Cerebrospinal fluid (CSF) in head injury, 657 leak of, 525

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

ix

Cerebrum, 657 Cervical collar, in rotary atlantoaxial ­subluxation, 707 Cervical cord neurapraxia/quadriplegia, 681–686 developmental stenosis and, 693–694, 710 grade of, 685–686 os odontoideum and, 692, 692f pathophysiology of, 690–691, 691f prevention of, 686–690, 687f–690f recurrence of, 686, 686f stenosis and, 681–686, 683f–��������������������� 685f, 710 Cervical ligament, 1913f Cervical radiculopathy, vs. lateral epicondylitis, 618 Cervical spine. See also Cervical spine injury anterior subluxation of, C3-C4, 678–679, 679f canal:vertebral body ratio in, 694, 694f computed tomography of, 539–541, 540f congenital anomalies of, 711–712, 711f, 712b, 712f facet dislocation of C3-C4, 679, 679f, 707–709 C4-C7, 680, 707–709, 709f fracture of, 675–677. See also Cervical spine injury C1-C3, 677–678, 677f, 678f in children/adolescents, 704–708, 705, 706f C3-C4, 678–679 in children/adolescents, 708–709 C4-C7, 679–681, 680f–682f in children/adolescents, 708–709 return to play after, 696f, 697–698, 697f–���������� 698f compression, 680–681, 680f–���������� 682f in children/adolescents, 709, 709f instability in, 675–676, 676f management principles in, 675–677, 676f return to play after, 696f, 697–698, 697f, 698f surgical treatment of, 676–677 vertebral body, 680–681, 681f–682f, 697f, 698, 698f fusion of congenital, 693, 693f, 711–712, 712f surgical, 699, 699f, 700f intervertebral disk of herniation of, 674, 698–699, 699f in children/adolescents, 704 injury to, 674, 698–699, 699f rupture of, 674, 678 pediatric. See also Cervical spine injury, pediatric anatomy of, 702–703, 703f congenital anomalies of, 711–712, 712b, 712f pedicles of congenital absence of, 711 fracture of, 678, 707–708 scoliosis of, congenital, 711 sprain of, 673–674 stenosis of, 681–686. See also Cervical cord neurapraxia/quadriplegia developmental, 693–694, 695f, 710 measurement of, 683–686, 683f–685f, 694, 694f ratio method measurement of, 683–686, 683f, 684f, 685f in spear tackler’s spine, 694, 695f, 710 stingers and, 672 subluxation of atlantoaxial, 706–707, 708f C3-C4, 678–679, 679f without fracture, 674



Index

Cervical spine injury, 665–701 in ambulatory patient, 669–670 in diving, 690 emergency management of, 665–669, 691 airway in, 667, 668f facemask removal in, 667, 667f head tilt–jaw lift maneuver in, 667, 668f helmet removal in, 668–669, 670f imaging in, 669 immobilization in, 666–667, 666f jaw thrust maneuver in, 667, 668f logroll in, 666–667, 666f six-plus-person lift in, 667 spine board for, 666–667, 666f transport in, 669, 669f football-related, 665–669, 686–690, 687f, 688f, 689f, 690f airway in, 668f, 670f facemask removal in, 667, 667f helmet removal in, 668–669, 670f immobilization in, 666–667, 667f fracture, 675–677, 676f. See also Cervical spine, fracture of return to sports after, 696f, 697–698, 697f, 698f in hockey, 690 intervertebral disk, 674, 698–699, 699f ligamentous, 696–697 logroll for, 519, 519f, 666–667, 666f lower cervical spine fracture and dislocation, 679–681, 679f, 681f–683f, 696–697 midcervical spine fracture and dislocation, 678–679, 679f, 696–697 nerve root–brachial plexus, 670–673, 671f–674f on-field, 522–524, 524f assessment of, 523–524, 524f pathophysiology of, 690, 690f pediatric, 701–713 airway management in, 703 atlanto-occipital instability, 704–705, 705f atlantoaxial instability, 705, 707b, 707f emergency treatment of, 703 epidemiology of, 702 hangman’s fracture, 707–708 Jefferson fracture, 705, 706f odontoid fracture, 707 rotary atlantoaxial subluxation, 706–707, 708f soft-tissue, 703–704 sports-specific, 702 subaxial fracture, 708–709, 709f personnel training for, 665–666 planning for, 665 prevention of, 686–690, 687f–690f quadriplegia with. See Cervical cord ­neurapraxia/quadriplegia return to sports after, 691–701 atlanto-occipital fusion and, 693 developmental stenosis and, 693–694, 694f fracture and, 676, 676f, 696f, 697–698, 697f, 698f intervertebral disk injury and, 698–699, 699f Klippel-Feil deformity and, 693, 693f middle and lower spine trauma and, 696–697 odontoid anomalies and, 692, 692f spear tackler’s spine and, 694, 694f spinal fusion and, 699, 699f, 700f upper spine trauma and, 694–696, 695f, 696f spinal stenosis and, 681–686, 683f–686f, 710 spine board for, 665, 666–667, 666f

Cervical spine injury (Continued) sprain, 673–674 subluxation C3-C4, 678–679, 679f without fracture, 674 upper spine fracture and dislocation, 677–678, 677f–679f return to sports and, 694–696, 695f, 696f Cervical spine injury without radiography abnormality, 702, 703, 704 Cervical sprain, 673–674 Chest, flail, 526, 893. See also Rib(s), fracture of Chest radiography in myocarditis, 151 in pulmonary embolism, 376 Chevron procedure, in hallux valgus, 2070, 2073b, 2073f, 2074f, 2076 Chilblains, 203 Child abuse, 595–596, 596f humeral physeal fracture in, 1283 rib fracture in, 894 Children/adolescents, 463–474. See also Female athlete accessory navicular in, 1963, 1963f acromioclavicular injuries in, 855–856, 855f adductor injury in, 1493 anabolic-androgenic steroid use by, 467–468 ankle sprain in, 1963, 1964f anterior cruciate ligament injury in, 1679–1683, 1682f arthroscopy in, 468–473. See also Arthroscopy, pediatric avulsion fracture in at hip, 1474–1475, 1475f, 1489 ischial, 1489, 1489f bone development in, 587–588, 588f, 589f bunion in, 1962–1963 bunionettes in, 1963 calorie requirement for, 402–403 cavus foot in, 1963 cervical spinal cord injury in, 709–711 cervical spine injury in, 701–713. See also Cervical spine injury, pediatric chondroblastoma in, 604f, 605 complex regional pain syndrome in, 367, 369 computed tomography in, 590, 590f congenital cervical spine anomalies in, 711–712, 711f–712f, 712b dehydration in, 465–466 elbow of. See Elbow, pediatric endurance training in, 464 ergogenic drug use by, 412–413 Ewing’s sarcoma in, 607, 609f, 610 femoral shaft stress fracture in, 1481 fibrous dysplasia in, 606, 608f flexibility in, 464–465 fluid requirement for, 402 fracture in ankle, 597, 598f, 1964–1969, 1965f–1969f bucket handle, 596, 596f carpal, 1364–1371. See also at specific bones clavicular, 596, 596f distal femur, 595, 595f femoral shaft, 1481 fifth metatarsal, 1969–1970, 1971f glenoid, 872, 875 hip, 597, 598f, 1474–1475, 1475f, 1489 humeral. See Humeral fracture ­(distal), pediatric; Humeral fracture ­(proximal), pediatric imaging of, 593–596, 594f, 595f, 596f, 597f, 598f ischial, 1489, 1489f metatarsal, 1969–1971, 1970f–1971f nonaccidental, 595–596, 596f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Children/adolescents (Continued) patellar, 1576–1577, 1576f radius, 594f, 595, 597 rib, 894 scapular, 872, 875 tibial eminence, 1677–1679, 1678f, 1680f tibial spine, 469, 472f Tillaux, 597, 598f, 1965, 1966f, 1968–1969 toe, 1970, 1971f ulna, 594f, 595f, 597 gender differences in, 476–477, 477t. See also Female athlete giant cell tumor in, 605, 605f glenohumeral joint instability in, 932–946. See also Glenohumeral joint instability, pediatric hamstring injury in, 1489 heat-related injury in, 465–466 hip injury in, 469, 470f, 1474–1477, 1475f, 1476f, 1476t iliac apophysitis in, 1475 imaging in, 587–610. See also at specific ­imaging modalities juvenile idiopathic arthritis in, 602–603, 602f knee injury in, 469, 471f Legg-Calvé-Perthes disease in, 599, 600f, 1476–1477 magnetic resonance imaging in. See Magnetic resonance imaging (MRI), pediatric nonossifying fibroma in, 605–606, 605f nutrition in, 467 osteochondritis dissecans in. See ­Osteochondritis dissecans osteochondroma in, 606, 607f osteochondroses in, 599, 599f, 1972–1974, 1973f, 1974f osteoid osteoma in, 603, 603f, 605 osteomyelitis in, 599, 601, 601f–602f osteosarcoma in, 606–607, 608f Panner’s disease in, 623, 625f, 1238 performance-enhancing substance use by, 467–468 peroneal tendon subluxation in, 1963–1964 posterior cruciate ligament injury in, 1713–1718, 1714f–1715f, 1717f psychosocial development of, 466 pump bumps in, 1963 quadriceps contusion in, 1484 quadriceps strain in, 1497 radiography in. See Radiography, pediatric radionuclide imaging in, 591, 591f, 601f–602f, 1230 septic arthritis in, 602 shoulder injury in, 468, 468f shoulder instability in, 468, 469f slipped capital femoral epiphysis in, 597, 598f, 1475–1476, 1476f, 1476t sports participation by, 463 adult involvement in, 466–467 coaches in, 466–467 injury with, 463–464 psychosocial development and, 466 readiness for, 466 strength training in, 465 tarsal coalition in, 1960–1962, 1961f, 1962f thermoregulation in, 465–466 thoracolumbar spine injury in, 754–768. See also Thoracolumbar spine injury, pediatric ultrasonography in, 590, 590f, 592–593 unicameral bone cyst in, 606, 606f wrist injury in, 468, 470f, 1363–1377. See also Wrist injury, pediatric

Index Chitosan, 409 Chloroquine phosphate, in cramps, 12 Chlorothiazide, 160t Chlorthalidone, 160t Chondrocyte(s), 41, 41f matrix interaction with, 45, 45f–46f, 49 proteoglycan synthesis of, 45, 49 tissue-engineered, 1776 Chondrocyte implantation, 55, 1774–1775 matrix techniques with, 1775–1776 rehabilitation after, 1785t results of, 1777, 1777t technique of, 1778–1780, 1780f Chondroitin sulfate, 43f, 44, 409 in knee arthritis, 1774, 1791 Chondrolysis, glenohumeral, 984, 984f, 1105t, 1106 Chondromalacia knee, 579, 580f, 581f sesamoid, 2089 wrist, 1447–1449 Chondromatosis, synovial, 1473 Chondroplasty, abrasion, of wrist, 1448–1449, 1448f Chop-and-lift exercises, in core training, 287, 288f Chrisman-Snook procedure, 1925–1926, 1925f Chromium, requirements for, 406b Chronic exertional compartment syndrome, 14–15, 15f, 650–651, 1857–1863, 1857b anatomy of, 650, 650f, 1858, 1858b, 1858f complications of, 1863 evaluation of, 14, 15f, 1859–1860, 1859f, 1860t pain in, 1858, 1859 return to play after, 1863 treatment of, 14–15, 651, 1860–1863, 1861f–1862f, 1863f Chronic tendinitis syndrome, 29–30 Chronobiology, 441–445, 442f. See also Circadian rhythms; Sleep bright light treatment and, 444–445, 445f circadian rhythms in, 442f, 443–444, 443f, 444f consultation on, 450 seasonal rhythms in, 444 Cigarette smoking, 426–428, 428f Circadian rhythms, 443–444, 443f, 444f cortisol in, 457 definition of, 443, 443f, 444f in depression, 452 dim light melatonin onset in, 457 endogenous, 456–457, 456f hormonal markers of, 457 impairment in, 453 light effect on, 443, 444f markers of, 456–457 performance and, 442, 442f, 457 sleep-wake, 445, 446f, 447f suprachiasmatic nucleus and, 443–444 Circulation, emergency assessment of, 520, 521f Citric acid cycle, 211–212 Citrus aurantium, 409 Clam exercise, in knee rehabilitation, 258, 259f Classification, 2209–2217. See also at specific injuries of injury, 2214–2217, 2214t, 2215f, 2216t, 2217t of pathology-related anatomic changes, 2211–2212, 2211f, 2211t, 2212t of pathology-related physical findings, 2212–2214, 2213f, 2213t, 2214t Claudication neurogenic, 745, 745t vascular, 745, 745t in popliteal artery entrapment, 1837

Clavicle. See also Sternoclavicular joint distal fracture of. See Acromioclavicular joint injury osteolysis of, 854, 855f resection of, 842, 843f acromioclavicular ligament injury and, 844 coracoclavicular ligament reconstruction with, 842, 843f failure of, 844 instability after, 827 without ligament reconstruction, 842–844, 843t medial. See also Sternoclavicular joint dislocation of. See Sternoclavicular joint injury, severe sprain (dislocation) fracture of, 804f, 805f, 824 vs. sternoclavicular dislocation, 804, 804f osteotomy of, 810 physis of, 794, 796f, 811 injury to, 811, 824–825, 824f anterior, 811, 824–825 posterior, 811, 825 return to play after, 825 treatment of, 819 resection of, 810 motion of, 776, 828, 828f pediatric, 780–781, 781f, 783–784, 784f fracture of, 596, 596f Claw hand deformity, 1312, 1312f Claw toe deformity, 2118, 2118t, 2120b, 2129 evaluation of, 2119 operative treatment of, 2121–2125, 2122f Clay shoveler’s fracture, 709 CLEAR mnemonic, 757–758 Clearance for participation, 513–515. See also Preparticipation examination legal aspects of, 531 Cleats, injury and, 2184, 2190–2192, 2195f, 2196f Clindamycin, 387, 387t CLOCK gene, 442 Clomiphene citrate, anabolic-androgenic ­steroid interaction with, 415–416 Clonazepam, in complex regional pain ­syndrome, 363t Clonidine in complex regional pain syndrome, 363t, 364, 365 in hypertension, 160t Clothing, in cold environment, 499–500 Clotrimazole, in dermatophyte infection, 200t Coach, 530 Coagulation, 371–372, 373f Cocaine use/abuse, 428–429 marijuana use and, 426 Coccyx, 1451–1452 Codman’s point, palpation of, 996, 997f Cohort study, 101 Cold, common, 149 Cold-induced vasodilation, 500 Cold injury, 203, 498–502, 528–529. See also Hypothermia clothing and, 499–500 drug use and, 499 medical conditions and, 499 physiology of, 498–499 Cold intolerance, in complex regional pain syndrome, 356 Cold therapy. See Cryotherapy Collagen articular cartilage, 42–43, 43f bone, 67, 68f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xi

Collagen (Continued) breakdown of, myoglobin excretion and, 12 implant of, 39 ligament, 32–33, 34, 1644–1645 meniscal, 57, 57t, 58–59, 58f–59f, 1598–1599 physical properties of, 22–23 structure of, 22, 23f tendon, 22–23, 22f, 23f, 23t, 1515, 1517f age-related changes in, 25 corticosteroid effects on, 26–27 exercise-related changes in, 25, 26f immobilization-related changes in, 25–26, 26f NSAID effects on, 27 synthesis of, 27–28 types of, 22, 23t Collagen hydrolysate, 409 Collar cervical, in rotary atlantoaxial subluxation, 707 cowboy, 673, 673f, 674f Collateral ligaments fibular (lateral). See Fibular (lateral) collateral ligament (FCL) first metatarsophalangeal joint, 2064–2065, 2065f, 2082 lateral. See Fibular (lateral) collateral ­ligament (FCL) medial. See Medial collateral ligament (MCL) metacarpophalangeal joint, 1380 injury to, 1380, 1399–1401, 1400f, 1401f ulnar. See Ulnar collateral ligament (UCL) Common cold, 149 Common peroneal nerve, injury to, 1751–1752, 1752f Commotio cordis, 165, 165f, 526 Compartment(s) increased pressure in, 650. See also Compartment syndrome lower extremity, 650–651, 651f, 1858, 1858b, 1858f pressure measurement in, 14, 650–651, 1860, 1860t Compartment syndrome acute, 14, 527–528 diagnosis of, 14, 1859–1860, 1859f, 1860t exertional, 14–15, 15f, 650–651, 651f, 1857–1863. See also Chronic exertional compartment syndrome lower extremity, 650–651, 651f anatomy of, 1858, 1858b, 1858f classification of, 1858–1859 complications of, 1863 evaluation of, 1859–1860, 1859f, 1860t after knee dislocation treatment, 1765 thigh, 1483–1484 treatment of, 1860–1863, 1861f, 1862f, 1863f with on-field injury, 527–528 pathophysiology of, 14 Pedowitz criteria in, 1860, 1860t upper extremity, flexor-pronator ­hypertrophy and, 622 Complete Book of Athletic Footwear, The (Cheskin), 1875 Complex regional pain syndrome, 351–369, 353b clinical presentation of, 355–358, 356t definition of, 351, 353b diagnosis of, 352, 353b, 355–356, 358–361 bone scan in, 359 epidural blockade in, 359 hematologic tests in, 358 magnetic resonance imaging in, 359 paravertebral sympathetic ganglion blockade in, 359

xii

Index

Complex regional pain syndrome (Continued) phentolamine testing in, 359–360 radiography in, 358 regional intravenous sympathetic blockade in, 360 spinal blockade in, 359 sudomotor measurement in, 360–361 vasomotor measurement in, 360 motor abnormalities in, 357–358 pathophysiology of, 353–355, 354f patient education on, 369 pediatric, 367, 369 psychological issues in, 358 sensory disturbances in, 356, 356t sudomotor abnormalities in, 355 support groups for, 369 sympathetic dysfunction in, 356–357, 357f terminology for, 351–352, 352b, 352t, 353b treatment of, 361–367, 368, 368f α-adrenergic blockers in, 363t, 365 algorithm for, 368, 368f antidepressants in, 362, 363t benzodiazepines in, 363t, 364 bisphosphonates in, 363t, 364–365 calcitonin in, 363t, 364–365 calcium channel blockers in, 363t, 364 capsaicin in, 364 clonidine in, 365 corticosteroids in, 362 electroconvulsive therapy in, 365 membrane stabilizers in, 363t, 364 neuromodulation techniques in, 366–367 NSAIDs in, 363t, 364 opioids in, 363t, 364 patient education in, 369 pharmacologic, 362–365, 363t physical therapy in, 362 psychotherapy in, 365 surgery in, 367 surgical sympathectomy in, 366 sympatholysis in, 365 trophic changes in, 358 vasomotor abnormalities in, 355 Compression fracture cervical spine, 680–681, 680f–682f in children/adolescents, 709, 709f lumbar spine, 546, 547f thoracic spine, 546, 547f, 734–735, 734f, 755, 755f Compression-rotation test, in SLAP lesion, 1025t Computed arthrotomography, of glenohumeral joint, 950–951 Computed tomography (CT), 533, 539–543 in acromioclavicular joint injury, 834f in avulsion injury, 553 in biceps femoris strain, 16, 17f in bifurcate sprain, 1954 in cervical spine fracture, 698f in deep venous thrombosis, 376 in distal humeral fracture, 1250 in elbow heterotopic ossification, 1292f, 1298, 1298f in femoral neck stress fracture, 554f in femoral osteochondritis dissecans, 1773, 1773f in femoral stress fracture, 1479 in fracture, 552 of glenohumeral joint, 950, 950f in glenohumeral joint instability, 916 in glenoid neck fracture, 861, 862f, 864f in glenoid rim fracture, 868f, 871f helical, 539 in high (syndesmosis) ankle sprain, 1942 in hip arthroplasty, 542, 542f

Computed tomography (CT) (Continued) in hook of hamate fracture, 543, 544f in humeral fracture, 983 image quality in, 539 in infection, 543 of intervertebral disk, 727, 727f in intracranial hemorrhage, 660, 661f in Jefferson fracture, 705, 706f in Lisfranc sprain, 1957, 1957f in lumbar burst fracture, 541, 541f in lumbar isthmic spondylolisthesis, 748–749 of lumbar spine, 727 monitor for, 539, 540f in occult fracture, 552, 557 in odontoid fracture, 540–541, 540f in osteomyelitis, 543 in particle disease, 542, 542f in patellar osteochondritis dissecans, 1531 in patellofemoral disorders, 1564–1565, 1566t pediatric, 590, 590f in avulsion fracture, 599 bone on, 591–592 in fracture, 590f, 597, 598f in proximal humeral physeal fracture, 1074 radiation dose in, 588 of pelvis, 541–542 in popliteus tendon avulsion, 541–542, 542f in proximal humeral fracture, 1038–1039 in rotary atlantoaxial subluxation, 706–707, 708f in shoulder disorders, 543 of spine, 539–541, 540f, 541f in spondylolysis, 761, 762, 763f in sternoclavicular joint injury, 804–805, 805f in stress fracture, 552, 554f in subtalar sprain, 1950 in tarsal coalition, 1961, 1961f, 2161, 2161f in tarsal navicular stress fracture, 646, 646f of thoracic spine, 727 in tibial plateau fracture, 542, 543f in tibial stress fracture, 1853 in tracheal displacement, 821, 821f in trapezium fracture, 1348, 1349f windowing in, 539, 540f Computed tomography (CT) angiography in popliteal artery entrapment, 1839–1840, 1840f in pulmonary embolism, 378, 378f Computed tomography (CT) arthrography in glenohumeral joint osteoarthritis, 1108 of knee, 536, 537f Computed tomography (CT)–myelography in lumbar disk herniation, 743 in lumbar spine stenosis, 746 in thoracic disk herniation, 738 Concussion, 522, 658–662 classification of, 658, 659t, 2216–2217, 2217t clearance for participation and, 513 definition of, 658 evaluation of, 658b, 659–660 grade 1, 658, 659t grade 2, 658, 659t grade 3, 658–659, 659t grading of, 522, 523t, 658–659, 659t imaging in, 660 incidence of, 662 on-field evaluation of, 659–660 return to play after, 513, 522, 523t, 659, 659t seizure and, 660 signs and symptoms of, 658b treatment of, 660 Conditioning, team physician advice on, 516 Conduction, heat loss by, 493, 499

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Condylectomy in bunionettes, 2134–2135, 2134b, 2135f, 2140, 2140f in corns, 2116, 2117f, 2124 in intractable plantar keratoses, 2111, 2113f in mallet toe, 2122–2123, 2123f, 2124f Cone ambulation, in knee rehabilitation, 295, 296f Cone reaching, in knee rehabilitation, 296, 297f Confidence interval, 99 Confidentiality, athlete, 531 Confirmation, diagnosis, 110 Congenital anomalies cervical spine, 711–712, 711f, 712b, 712f lumbar spine, 746–747, 747f, 748f odontoid, 692, 692f renal, 712, 712f scaphoid, 1366 sternoclavicular joint, 799, 811 superior peroneal retinaculum, 1991 Conoid ligament, 827 Consciousness, alteration in. See Concussion; Head injury Construct validity, 100 Contact dermatitis, 202, 203f Content validity, 100 Continuous positive airway pressure, in sleep apnea, 448 Contracoup injury, 657, 657f Control group, 101 Contusion abdominal, 526 femoral, 1542, 1543f foot, 1964 groin, 1459–1460 iliac crest, 1459 muscle, 13. See also Quadriceps muscle, contusion of magnetic resonance imaging in, 558 myositis ossificans with, 13–14, 14f, 558 myocardial, 526 pulmonary, 526 quadriceps. See Quadriceps muscle, ­contusion of renal, 527 thoracic spine, 732–733 Convection, heat loss by, 499 Coracoacromial arch, anatomy of, 990, 990f–991f Coracoacromial ligament, 828 anatomy of, 773–774, 990, 990f magnetic resonance imaging of, 956, 959f resection of, 1009 Coracobrachialis muscle anatomy of, 1158, 1158f scapular attachment of, 858, 858f transfer of, 840–841, 841f Coracoclavicular ligament, 827, 827f injury to. See Acromioclavicular joint injury reconstruction of, 842, 843f, 846–851 approach to, 846, 847f–848f clavicle bone tunnel preparation for, 849–850, 850f–851f closure for, 851 complications of, 851–853 coracoid bone tunnel preparation for, 848, 849f graft-clavicle fixation for, 850–851, 852f, 853f graft-coracoid fixation for, 849, 850f graft preparation for, 847–848, 849f management after, 851–853 repair of, 841, 841f in sternoclavicular joint injury treatment, 824 transfer of, 841–842

Index Coracohumeral distance, 956 Coracohumeral ligament. See also Rotator interval anatomy of, 773, 966 biomechanics of, 788 Coracoid anatomy of, 990, 990f fracture of, 876–885 acromioclavicular dislocation and, 880, 882, 882f, 883f, 884 classification of, 882 computed tomography in, 879, 880f electromyography in, 879 vs. epiphysis, 879, 881f exercises after, 884 glenohumeral dislocation and, 876, 877 mechanism of, 876–877, 877b neurologic injury with, 877, 878, 878f nontraumatic etiologies of, 877 patient history in, 877–878 pattern of, 877, 878f physical examination in, 878 radiography in, 861, 862f, 878–879, 878f, 879f, 880f, 881f stress, 877–878 Stryker notch view for, 861, 862f, 879, 880f, 881f suprascapular nerve involvement in, 880, 882f–883f treatment of, 880–882, 881f–883f, 884b ossification of, 859, 859f, 860f, 879, 881f pediatric, 782 Core training, 277–288, 277t, 289t abdominal bracing in, 280, 280f advanced functional training in, 288 AIR principles in, 279 bridging progression in, 281–282, 281f–282f curl-up progression in, 284–285, 285f gluteal training in, 285–286, 286f lateral flexion progression in, 283–284, 284f latissimus dorsi training in, 286 loading parameters in, 279–280 manual perturbation training in, 286, 287f program design for, 279 quadruped progression in, 282–283, 282f, 283f rhythmic stabilization training in, 286 rotation training in, 286–288, 288f–289f scapular training in, 286, 286f in thoracolumbar spine rehabilitation, 728–730, 730f Corn, 2116, 2117f, 2119 interdigital, 2116, 2118f, 2119, 2124 Coronary artery, anomalies of, 165, 165f, 166f Coronary heart disease, in diabetes mellitus, 175–176 Coronary ligament, 1722 repair of, 1733, 1734f Coronoid fracture of, 1263, 1265f, 1273 anteromedial, 1266, 1270–1271 basal, 1266 classification of, 1264, 1266 operative treatment of, 1266, 1267f–1268f, 1269–1270, 1270f, 1275–1276 tip, 1264, 1266, 1269–1270, 1270f stabilizing effect of, 1192, 1193f Correlation, 100–101 Corticosteroid(s) bone effects of, 72 in cervical fracture, 675, 691 in complex regional pain syndrome, 362 in epilepsy, 190 in exercise-induced bronchospasm, 182

Corticosteroid(s) (Continued) injection of. See Corticosteroid injection in low back pain, 731 pectoralis major rupture and, 902 in tendon healing, 29 Corticosteroid injection in Achilles tendon injury, 1999 in de Quervain’s tenosynovitis, 1355 epidural, 534, 535f, 584, 584f fluoroscopy-guided, 582–584, 584f in knee arthritis, 1774, 1791 in lateral epicondylitis, 618, 1200 in meniscal injury, 1605 in plantar fasciitis, 2048–2049 in retrocalcaneal bursitis, 2035, 2036 in rotator cuff disorders, 1007 tendon, 31 experimental studies of, 26–27 long-term effects of, 26–27 studies of, 26–27 Cortisol exercise effects on, 217, 217t levels of, 457 Corynebacterium minutissimum infection, 194t, 195–196, 196f Costoclavicular ligament, 793, 793f Costoclavicular maneuver, in vascular injury, 1138, 1138f Coumadin, prophylactic, in venous ­thromboembolism, 378–384, 381t Counterforce brace, in lateral epicondylitis, 618–619, 618f, 1200–1201, 1201f Coup injury, 657, 657f Cowboy collar, 673, 673f, 674f Coxa sultans, 1458–1459 Cramps, 11–12 creatine and, 418 heat, 495–496, 529 Crank test, in SLAP lesion, 1025t Creatine, 408, 418–419 nonresponse to, 418 Creep, 48, 59, 95–96, 95f Crepitus, scapulothoracic, 886–889, 887b, 887f Cricothyroidotomy, needle, 519, 520f Criterion-related validity, 100 Cross-arm adduction test, in acromioclavicular joint injury, 830, 832f Crossing sign, in femoral trochlear dysplasia, 1561, 1561f Cruciate ligaments. See Anterior cruciate ­ligament (ACL); Posterior cruciate ligament (PCL) Cruciate ridge, 1646, 1647f Crutches, removal of, in knee rehabilitation, 295 Cryopreservation, allograft, 140–141 Cryotherapy, 234–235, 235f in ACL rehabilitation, 1670 in heat illness, 529 in lateral ankle sprain, 1920 in medial ankle sprain, 1937 in retrocalcaneal bursitis, 2035 in thoracolumbar spine injury, 731 CT. See Computed tomography (CT) Cubital tunnel syndrome, 1311–1315 anatomy of, 1312, 1312f etiology of, 1311–1312, 1311f history in, 1312 physical examination in, 1312, 1312f–1313f treatment of, 1312–1315 anterior subcutaneous transposition in, 1313–1314, 1313f–1315f anterior submuscular transposition in, 1314 medial epicondylectomy in, 1314–1315

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xiii

Cubital tunnel syndrome (Continued) nonoperative, 1312–1313 operative, 1313–1315, 1313f in situ decompression in, 1313 Cuboid, stress fracture of, 646, 2023 Cullen’s sign, 526 Cumulative incidence, 103 Cuneiform osteochondrosis of, 2166, 2166f stress fracture of, 646, 2023 Curl-up exercise against abdominal brace, 285, 285f beginner’s, 285, 285f in core training, 284–285, 285f Cyclist’s palsy, 1361 Cyst(s) aspiration of, 585 bone aneurysmal, 983–984 radiography in, 552, 553f unicameral, 606, 606f fracture through, 552, 553f, 606, 606f, 1085, 1086f ganglion, suprascapular nerve compression with, 617, 1121, 1121f, 1123 humeral, 983–984 image-guided aspiration of, 585 meniscal, 1613 paralabral, 580–581, 582f, 964–965, 965f, 980, 980t popliteal (Baker’s), 537–538, 538f rotator cuff, 964–965, 964f synovial, suprascapular nerve compression with, 617 ultrasonography of, 537–538, 538f Cytokines in complex regional pain syndrome, 355 in infection, 147

D Dacron prosthesis, in ligament injury ­treatment, 36 De Quervain’s tenosynovitis, 624–625, 1355, 1356f magnetic resonance imaging in, 569, 570f Dead bug exercise in core training, 285, 285f in trunk stabilization, 342t, 343 Death, sudden. See Sudden death Débridement, arthroscopic in glenohumeral joint, 1110, 1110f, 1115, 1115f in knee, 1774, 1792 in SLAP lesion, 1027, 1028 Decentralization supersensitivity, in complex regional pain syndrome, 355 Decision analysis, 99, 112 Decorin, 44 Deep peroneal nerve, entrapment of, 2061–2062, 2061f Deep venous thrombosis, 370–385 age-related factors in, 373, 375f arthroscopy and, 123–124, 124b, 383–384, 383t evaluation of, 374, 376–378, 376t, 377f high tibial osteotomy and, 1833 treatment of, 378–385, 384t prophylactic, 378–384, 379t, 381t, 382f Defibrillator, implantable, in sudden death prevention, 169–170 Degrees of freedom, 89 Dehydration, 401 in children/adolescents, 465–466 exercise-related, 494–495 hyponatremic, 495

xiv

Index

Dehydroepiandrosterone (DHEA), 414, 417–418 Delayed-onset muscle soreness, 12–13, 13f, 215–216 Delayed union, scaphoid fracture, 1366, 1367f Deltoid ligament, 338, 1913f, 1935, 1935f injury to, 1935–1938, 1936f, 1937f magnetic resonance imaging of, 573 ossicles within, 1938 repair of, 1938 Deltoid muscle anatomy of, 771, 1034, 1035f, 1035t, 1036b, 1063, 1063f pediatric, 784, 785f biomechanics of, 1063 denervation edema of, 965, 965f, 980 humeral attachment of, 1071 neurologic evaluation of, 1036–1037, 1038f rupture of, 1063–1064 clinical evaluation of, 1064 physical examination in, 1064 treatment of, 1064 scapular attachment of, 859f, 886f Dementia pugilistica, 663b Denervation supersensitivity, in complex regional pain syndrome, 355 Dependency, of variables, 100–101 Depression anabolic-androgenic steroids and, 417 bright light therapy in, 444–445, 445f sleep disorders in, 452 Dermatan sulfate, 44 Dermatitis, contact, 202, 203f Dermatologic disorders. See at Skin Dermatophyte infection, 198–200, 198f, 199f, 200f Desipramine, in complex regional pain ­syndrome, 362, 363t DEXA (dual-energy x-ray absorptiometry), 72 Dexamethasone, iontophoresis for, 234 DHEA (dehydroepiandrosterone), 417–418 Diabetes mellitus, 172–179, 514t coronary heart disease in, 175–176 evaluation of, 175–176, 176b, 179 exercise in, 172–175, 173f–175f glucose regulation in, 174, 174f hyperglycemia in, 175, 175f, 177–178, 177b hypoglycemia in, 174–175, 174b, 177, 177b acute management of, 178–179, 179b insulin in, 176–178, 176t, 177b, 177t, 178f peripheral neuropathy in, 176 pre-participation guidelines in, 179 treatment of, 176–179, 176t, 177b, 177t, 178b, 178f Dial test in posterior cruciate ligament injury, 1691 in posterolateral corner injury, 1727, 1728f Diaphysis, 587–588 Diarrhea, 152, 514t Diet. See also Nutrition fad, 406 in female athlete, 477–479, 478f, 478t, 479b at high altitude, 505 in hypertension, 158, 159t for pediatric athlete, 467 for weight gain, 406, 408b for weight loss, 406, 407b Dietary supplement(s), 406–409. See also specific supplements definition of, 417 ergogenic, 417–422 Dietary Supplement and Health Education Act (1994), 407, 468 Dieter’s Tea, 409 Dihydrotestosterone, 414

Diltiazem in complex regional pain syndrome, 363t in hypertension, 160t Diphosphonates, in heterotopic ossification, 1294 Direct eversion maneuver, in high ­(syndesmosis) ankle sprain, 1940, 1941f Disability, emergency assessment of, 520, 521f Diskectomy in congenital lumbar stenosis, 748f in lumbar disk herniation, 744, 744f, 751f Diskography percutaneous, 584–585 thoracolumbar, 728 Dislocation. See at specific joints Distraction osteogenesis, 81–82, 83f Diuretics, in hypertension, 158–159, 160t Diving, cervical spine injury in, 690 Doping, 413 Dorsal intercalary segmental instability (DISI), 1322, 1323f Double-bundle posterior cruciate ligament reconstruction, 1705–1706, 1705f–1706f in combined injury, 1706–1710, 1708f, 1709f, 1710f evaluation of, 1696, 1699–1700, 1699t Double-leg jumping exercise, in knee ­rehabilitation, 296, 298f Double PCL sign, in meniscal injury, 1602, 1603f Down syndrome activity restrictions in, 705, 707b atlanto-occipital instability in, 705, 705f atlantoaxial instability in, 705, 707f Doxazosin, 160t Doxepin, in complex regional pain syndrome, 363t Drawer test anterior in anterior cruciate ligament injury, 1650 in lateral ankle sprain, 1916–1917, 1917f in medial ankle sprain, 1936 in medial collateral ligament injury, 1629, 1630t anterior-posterior, in glenohumeral joint instability, 943–944, 944f posterior, in posterior cruciate ligament injury, 1690, 1690f posterolateral, in posterolateral corner injury, 1727, 1729f Drive-through sign, 1731, 1732f Drop finger (mallet finger), 1388, 1420–1422, 1420f, 1421f Drowning, epilepsy and, 192 Drug(s) alertness-enhancing, 451–452, 453t emergency assessment of, 520, 521f ergogenic, 410–423, 2173–2174. See also Anabolic-androgenic steroids and specific drugs historical perspective on, 410–411 iontophoresis for, 233–234, 235f recreational, 424–431.See also specific drugs tendon effects of, 1516 thermal balance and, 499 Dual-energy x-ray absorptiometry (DEXA), 72 Dumbbell floor press, in shoulder ­rehabilitation, 247, 248f Dumbbell fly exercise, shoulder injury and, 249 Duncan loop, 135, 135f Durkan’s compression test, 1360 Dusting, 430 Dynamic shift test, in posterolateral corner injury, 1728 Dysesthesia, definition of, 356t

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Dyskinesia scapular, 1007 scapulothoracic, 891–892

E Ear, on-field injury to, 525 East German athletes, drug use by, 412 Eating disorders, 478, 478f, 514t Eccentric training, in ankle rehabilitation, 275–276, 276f Ecchymosis, in pectoralis major rupture, 902, 903f ECG. See Electrocardiography (ECG) Echocardiography screening, 168 in sudden death prevention, 168 Eczema, 204, 204f Edema cerebral, high-altitude, 504–505 in complex regional pain syndrome, 355, 356 heat, 530 postexercise, 216 pulmonary, high-altitude, 504 Effort thrombosis axillary vein, 1141–1142 on-field, 527 subclavian vein, 1129, 1131f, 1134 Elastic deformation energy, 94 Elastin ligament, 34 meniscal, 58 tendon, 23 Elbow. See also Olecranon anatomy of, 1301–1303, 1302f–1303f on arthroscopy, 130, 130f normal variants in, 1229 arthroplasty of for arthrosis, 1278 in distal humeral fracture nonunion, 1258 heterotopic ossification after, 1290, 1291f arthroscopy of. See Arthroscopy, elbow arthrosis of, 1278 axis of rotation of, 1190, 1191f biomechanics of, 1189–1197 axis of rotation in, 1190, 1191f carrying angle in, 1190, 1191f, 1192 forces in, 1194–1195, 1196f, 1230 humerus in, 1189, 1189f normal, 1190 radial head in, 1190, 1190f ulna in, 1189–1190, 1189f carrying angle of, 1190, 1192 continuous passive motion of, after fracture treatment, 1277 effusion of, 535f in children, 596–597, 597f entrapment neuropathy at. See Cubital tunnel syndrome; Pronator syndrome; Radial tunnel syndrome forces across, 1194–1195, 1196f, 1230 fracture-dislocation of, 1262–1271, 1304. See also Elbow dislocation evaluation of, 1263–1267, 1264f, 1265f, 1266f–1267f mechanisms of, 1262 operative treatment of, 1263, 1267–1271 in anteromedial coronoid facet fracture, 1270–1271 complications of, 1271 in coronoid tip fracture, 1269–1270, 1270f in radial head fracture, 1269 stability testing in, 1268–1269, 1269f

Index Elbow (Continued) patterns of, 1263 terrible triad, 1259, 1263, 1264f, 1266, 1269 fracture of. See Capitellum, fracture of; Humeral fracture (distal); Olecranon, fracture of; Radial head, fracture of; Radial neck, fracture of golfer’s, 620–621, 1205–1206, 1205f, 1206f heterotopic ossification of. See Heterotopic ossification, elbow instability of postoperative, 1277–1278 recurrent, 1307–1310, 1307f–1310f stages of, 1263 trauma-related, 1262–1271. See also Elbow, fracture-dislocation of; Elbow­dislocation ligaments of. See also Radial collateral ligament; Ulnar collateral ligament anatomy of, 1301–1302, 1303f magnetic resonance imaging of, 576, 577f, 578f stabilizing effect of, 1193, 1195f, 1231, 1231f Little Leaguer’s. See Little Leaguer’s elbow miner’s. See Olecranon bursitis motion of, 1190–1192, 1191f extension, 1190, 1191f, 1196f flexion, 1190, 1191f, 1196f functional, 1192 normal, 1190 pronation, 1190 supination, 1190 nursemaid’s, 1301, 1302f, 1303–1305, 1305f ossification of, 1227–1229, 1228f, 1229f, 1279, 1280f osteochondritis dissecans of. See Osteochondritis dissecans, capitellum overuse injury of, 617–624, 618f lateral, 617–619, 618f medial, 619–624, 620f, 624f pediatric dislocation of. See Elbow dislocation, pediatric fracture of, 1279–1288. See also Humeral fracture (distal); Olecranon, fracture of; Radial head, fracture of; Radial neck, fracture of gymnastics-related injury to, 1236 injury to, 468, 469f, 470f, 596–599, 597f, 598f, 623–624, 624f, 1236–1240 gymnastics-related, 1236 lateral, 1238–1239 medial, 1236–1238, 1236f, 1238f posterior, 1239–1240, 1239f tennis-related, 1236 throwing-related, 1221–1225, 1223f, 1225f. See also Little Leaguer’s elbow; Overhead throwing injury, pediatric loose body in, 470f ossification of, 1227–1229, 1228f, 1229f, 1279, 1280f radiography of, 1279–1288 tennis-related injury to, 1236 throwing-related injury to, 1221–1226, 1223f, 1225f. See also Little Leaguer’s elbow; Overhead throwing injury, pediatric rehabilitation of. See Elbow rehabilitation stability of, 1192–1197, 1230–1231 coronoid in, 1192, 1193f intraoperative testing of, 1268–1269, 1269f lateral collateral ligament complex in, 1193, 1195f, 1231 medial collateral ligament in, 1193, 1194f, 1230–1231, 1231f

Elbow (Continued) muscles in, 1194, 1196f olecranon in, 1192, 1192f radial head in, 1192–1193, 1193f stiffness of, differential diagnosis of, 1293 student’s. See Olecranon bursitis tennis. See Epicondylitis, lateral (tennis elbow) terrible triad of, 1259, 1263, 1264f, 1266, 1269 Elbow dislocation, 1262–1271, 1300–1310 anatomy of, 1301–1303, 1302f, 1303f capsuloligamentous injury in, 1263 classification of, 1303–1304, 1303f closed reduction for, 1262, 1304–1305, 1305f, 1306 complications of, 1307–1310, 1307f–1310f coronoid fracture with, 1263, 1265f evaluation of, 1304 external fixation in, 1306, 1306f fracture with. See Elbow, fracture-dislocation of incidence of, 1300 instability after, 1307–1310 lateral collateral ligament reconstruction in, 1308, 1310, 1310f physical examination in, 1307–1308, 1307f, 1308f treatment of, 1308–1310, 1309f mechanisms of, 1262, 1300–1301, 1301f, 1302f nerve injury with, 1263, 1304 nonoperative treatment of, 1262, 1304–1305, 1305f, 1306 olecranon fracture with, 1263–1264, 1266f, 1267f–1268f, 1273 open reduction for, 1305 operative treatment of, 1305–1306, 1306f patterns of, 1263, 1264f–1267f pediatric, 1288, 1300–1310, 1302f classification of, 1303–1304 closed reduction for, 1304–1305, 1305f complications of, 1307–1310, 1307f evaluation of, 1304 incidence of, 1300 mechanism of, 1300–1301, 1301f–1302f post-reduction care in, 1306–1307 surgical treatment of, 1305–1306 treatment of, 1304–1306, 1305f post-reduction care in, 1306–1307 posterior, 1263, 1264f, 1269 rehabilitation after, 1306–1307 return to play after, 1310 stages of, 1263 treatment of, 1304–1306 vascular injury with, 1263, 1304 Elbow rehabilitation, therapeutic exercise for, 250–255 extensor training in, 253–254, 253f flexor training in, 251–253, 252f–253f forearm muscle training in, 254–255, 254f, 255f rhythmic stabilization training in, 255, 255f total arm strengthening in, 251 Elderly people. See Age/aging Electrical stimulation, 229–233 in anterior cruciate ligament rehabilitation, 230–233, 231f–234f, 1671 in complex regional pain syndrome, 366–367 in fracture healing, 80 for functional restoration, 230–233, 231f–234f in pain modulation, 229–230, 230f in swelling, 230, 231f Electrocardiography (ECG) in long Q-T syndrome, 166–167, 167f in myocarditis, 151 in pulmonary embolism, 376

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xv

Electrocardiography (ECG) (Continued) screening, 168, 169t in sudden death prevention, 167–169, 169t Electroconvulsive therapy, in complex regional pain syndrome, 365 Electromyography (EMG) of biceps tendon, 1019, 1019f, 1024 in brachial plexus injury, 673 in carpal tunnel syndrome, 1361 in coracoid fracture, 879 in entrapment neuropathy, 1311 in knee dislocation, 1752 in lateral epicondylitis, 1199 in Parsonage-Turner syndrome, 1144 in suprascapular nerve injury, 1122, 1123–1124 in tarsal tunnel syndrome, 2058 in ulnar neuropathy, 623 in valgus instability, 620 Elmslie-Trillant procedure, 1567 Elson’s test, 1389 Embolism, pulmonary. See Pulmonary ­embolism Embryo, endochondral bone formation in, 78 Emergency. See On-field emergency EMG. See Electromyography (EMG) Emissary vein, 69, 70f Emphysema, subcutaneous, sternoclavicular joint injury and, 821–822, 823f Empty-can exercise, in shoulder rehabilitation, 243–244, 243f, 244f Enalapril, 160t Endomysium, 3, 207, 208f Endorphins antiepileptic effects of, 187 exercise effects on, 217, 217t Endothelium, disruption of, 370, 371f Endurance training in ACL rehabilitation, 1671, 1674 in children/adolescents, 464 in female athlete, 479–480 muscle effects of, 10 Energy exercise-related requirements for, 498 muscle metabolism of, 210–213, 210f sources of, 498 Energy loss/return playing surface and, 2198–2199, 2199f shoe and, 1905–1906 Enthesophyte, of acromion, 955, 958f Enzyme-linked immunosorbent assay, in deep venous thrombosis, 376 Eosinophilic granuloma, 2161, 2161f Ephapse, in complex regional pain syndrome, 353–355, 354f Ephedra, 409 Epicondylitis lateral (tennis elbow), 30, 31, 566, 576, 612, 617–619, 618f, 1197–1205 chair test in, 1199 in children/adolescents, 1236 coffee cup test in, 1199 counterforce brace in, 618, 618f, 1200–1201, 1201f demographics of, 1197 differential diagnosis of, 1199 electromyography in, 1199 evaluation of, 1198–1199 focal hyaline degeneration in, 1198, 1198f hand exercises in, 1200 history in, 1198–1199 magnetic resonance imaging in, 566, 568f, 1199 nonoperative treatment of, 1199–1201, 1200f–1201f

xvi

Index

Epicondylitis (Continued) operative treatment of, 619, 1201–1205 arthroscopic release in, 1203–1204 extra-articular lateral epicondylar release in, 1204–1205, 1204f open release and resection in, 1201–1203, 1202f percutaneous release in, 1203 pathology of, 1197–1198, 1198f physical examination in, 1199 racquet selection and, 1200, 1201f radial nerve compression in, 1199 radiography in, 1199 treatment of, 1199–1205 medial (golfer’s elbow), 566, 620–621, 1205–1206 evaluation of, 1205, 1205f nonoperative treatment of, 1205 operative treatment of, 1205–1206, 1206f Epidemiological terminology, 2218–2219, 2218t Epidural blockade, in complex regional pain syndrome, 359, 366 Epidural corticosteroid injection, 534, 535f Epidural hematoma, 660, 661f Epilepsy, 185–192, 522–523 accidental death and, 187, 192 bathing and, 192 evaluation of, 190–192 exercise effects on, 187–188 historical perspective on, 185–186 immunotherapy in, 190 seizure frequency in, 187–188 surgery in, 190 swimming and, 192 terminology for, 186–187, 186b treatment of, 188–192, 191t vagal nerve stimulator in, 189–190 Epimysium, 3, 207, 208f Epinephrine, 353 exercise effects on, 217–218, 217t, 218f Epiphysis, 587–588, 588f of distal femur, 1638 of medial clavicle, 794, 796f, 811 of proximal humerus, 1069–1070, 1070f of proximal tibia, 1638 Epistaxis, on-field, 525 Epitenon, 20, 21f Eprosartan, in hypertension, 160t Epstein-Barr virus infection, 150–151 Epworth Sleepiness Scale, 450, 450f Ergogenic drugs, 410–423, 2173–2174. See also Anabolic-androgenic steroids and specific drugs historical perspective on, 410–411 ERMI Extensionator, 292, 292f ERMI Flexionator, 225, 225f, 292, 292f Error, statistical, 100, 111 Erysipelothrix rhusiopathiae infection, 397, 398t Erythrasma, 194t, 195–196, 196f Erythropoietin high-altitude effects on, 420, 503 recombinant, 420–421 Essex-Lopresti lesion, 1259, 1260 Estivation, 444 Estrogen, 414 bone effects of, 72 exercise effects on, 217t, 218 Estrogen replacement therapy, in female athlete, 478 Ethics, for team physician, 530–531 Ethylene oxide, in allograft sterilization, 139 Eucapnic voluntary hyperventilation challenge, 182 Evans procedure, 1924, 1925f

Evaporation, heat loss by, 493, 499 Excessive daytime sleepiness, 448–449 Exercise. See Exercise physiology; Rehabilitation; Therapeutic exercise(s) Exercise physiology, 207–220 cardiorespiratory response in, 218–220, 219f, 220t hormonal adaptation in, 217–218, 217t, 218f maximum oxygen uptake in, 218, 219f muscle, 10–11, 208–213, 209f, 210f, 212f. See also Muscle(s) neuromuscular adaptation in, 214–216, 216t respiratory response in, 220 training response in, 213–214, 214t, 215f Exertional compartment syndrome. See Chronic exertional compartment syndrome Exertional myositis, 496 Exertional rhabdomyolysis, 496–497 Exposure. See also Hypothermia emergency assessment of, 520, 521f Extensor carpi radialis brevis tendon, tear of, 1198. See also Epicondylitis, lateral Extensor carpi ulnaris tendinopathy, 1351, 1354–1355 classification of, 1354 clinical manifestations of, 1354 physical examination in, 1354 radiography in, 1354 return to play after, 1354 treatment of, 1354–1355, 1355f Extensor carpi ulnaris tendon inflammation of, 625 subluxation of, 625 Extensor hallucis longus tendon, 2065, 2065f Extensor pollicis longus tendon, injury to, 1401–1402 Extensor training, in elbow rehabilitation, 253–254, 253f External fixation in Achilles tendon rupture, 2009 in knee dislocation, 1754, 1754f External oblique muscle, strain of, 1461 External rotation exercise, in shoulder ­rehabilitation, 242–243, 242f, 243f External rotation recurvatum test, in ­posterolateral corner injury, 1726, 1726f External rotation stress test, in high ­(syndesmosis) ankle sprain, 1940, 1941f Extracorporeal shock-wave therapy in Achilles tendon injury, 1999 in lateral epicondylitis, 1200 in plantar fasciitis, 2049–2050, 2052 in retrocalcaneal bursitis, 2035 Eye(s) loss of, 514t on-field examination of, 525 on-field injury to, 525 raccoon, 525

F Fabellofibular ligament, 1722, 1731f FABERE test, in degenerative hip disease, 1503 Face validity, 100 Facet joint dislocation of C3-C4, 679, 679f, 708–709 C4-C7, 680, 708–709 in low back pain, 741 Factor V Leiden, 372, 374t Factor VIII, increase in, 372, 374t False-negative rate, 108, 109f False-positive rate, 108, 109f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Fas-T-Fix device, in meniscal repair, 1612–1613, 1612f Fascia lata, in sternoclavicular joint dislocation treatment, 810 Fascicle, 3, 4f Fasciectomy, 651 Fasciitis, plantar. See Plantar fasciitis Fasciotomy, 651 in chronic exertional compartment ­syndrome, 1860–1863, 1861f–1863f in plantar fasciitis, 2050, 2051–2052, 2053 in thigh compartment syndrome, 1483, 1484 Fat body, in female athlete, 476 dietary, metabolism of, 211 Fatigue muscle, 214, 214t nutritional deficits and, 401 seizures with, 187 Fatigue fracture. See Stress fracture Feet. See Foot (feet) Felodipine, 160t Female athlete, 475–491 anatomic parameters in, 476–477, 477t anterior cruciate ligament injury in, 483–485, 484b, 486t–487t, 487b, 488t, 1651 prevention of, 333–334, 485, 487b bone density in, 481 breast cancer in, 481 calcium intake by, 478, 478t conditioning for, 479–480, 480b forefoot problems in, 490–491, 491b frozen shoulder in, 489–490, 490t heart rate in, 476–477 iron deficiency in, 478–479, 479b knee replacement in, 1799 menopause in, 480–481, 481b nutrition in, 477–479, 478f, 478t, 479b patellofemoral arthritis in, 489 patellofemoral dislocation in, 487, 489, 489b patellofemoral joint injury in, 485, 487, 489 patellofemoral pain in, 487 physical examination of, 479 physiologic parameters in, 476–477, 477t pregnancy in, 481–483, 482b shoes for, 490–491, 491b, 1886–1887 shoulder instability in, 489 sport participation by, 475–476, 475f, 475t, 476f stress fracture in, 483, 483b, 483f, 632–633, 1478, 1849–1850, 1851–1852, 1856 Female athlete triad, 478, 478f, 633 Femoral artery, 1453, 1455f, 1514 occlusion of, 1497–1499 Femoral canal, 1463 Femoral condyles, osteochondritis dissecans of. See Osteochondritis dissecans, femoral condyle Femoral epicondylar axis, 1581 Femoral head abnormal acetabular contact with, 1471– 1472, 1471f, 1472f anatomy of, 1452, 1500 avascular necrosis of, 1466–1467, 1467b, 1467t in children, 1476–1477 slipped capital femoral epiphysis and, 1476 blood supply to, 1452, 1501 chondral injury of, 1472 fracture of, 1464 resurfacing procedures for, 1506 Femoral neck fracture of, 1464 stress fracture of, 638–640, 640f, 1465–1466, 1465f

Index Femoral neck (Continued) classification of, 639, 641f complications of, 639 compression-side, 639, 641f, 643f displaced, 639, 642f magnetic resonance imaging in, 554f tension-side, 639, 641f, 642f Femoral nerve, 1453, 1455f entrapment of, 1469 Femoral nerve stretch test in lumbar disk herniation, 743 in thoracolumbar spine injury, 722, 723f Femoral nerve tension sign, in thoracolumbar spine injury, 722 Femoral shaft, stress fracture of, 641, 1477–1481 anatomy of, 1477–1478 in children, 1481 classification of, 1478, 1478t clinical presentation of, 1478 grade of, 1478, 1478t imaging of, 1479, 1479f, 1480f physical examination in, 1478–1479, 1479b return to play after, 1481, 1481t shoewear and, 1898–1899 treatment of, 1479–1481, 1480f Femoral triangle, 1453 Femoral vein, 1453, 1455f Femoral version, 1500 Femoroacetabular impingement, 1471–1472, 1471f, 1472f, 1503 Femorotibial joint, biomechanics of, 1514–1515 Femur. See also at Femoral contusion of, 1542, 1543f distal. See also Knee anatomy of, 1549–1550, 1550f epiphyseal fracture of, 1641–1642, 1641b, 1641f, 1642f, 1643–1644, 1643f, 1644b epiphyseal ossification of, 587, 588f growth plate of, 1638 stress fracture of, 555f proximal. See also Femoral head; Femoral neck; Hip anatomy of, 1452, 1549, 1549f anteversion of, 1549, 1549f, 1565 torsion of, 1549, 1549f, 1565 version of, 1549, 1549f Femur–anterior cruciate ligament–tibia complex, tensile loading of, 93, 94f Fertility, anabolic-androgenic steroid effects on, 416 Fetal warfarin syndrome, 381, 382f Fever, 514t Fibrin clot augmentation, in meniscal injury, 1608 Fibroblasts ligament, 33–34, 35 tendon, 20, 22, 24, 27 Fibrochondrocytes, of meniscus, 1598–1599 Fibroma, nonossifying, in children, 605–606, 605f Fibronectin ligament, 34 meniscal, 58 Fibula, stress fracture of, 644, 644f, 1849, 2016t, 2018, 2019f Fibular (lateral) collateral ligament (FCL). See also Posterolateral corner anatomy of, 1685, 1719–1720, 1719f, 1720f, 1722f, 1731f, 1748 biomechanics of, 1724 evaluation of, 1692 reconstruction of, 1736–1737, 1737f high tibial osteotomy and, 1815t, 1833, 1834f stabilizing function of, 1588–1589

Field-exercise challenge test, in bronchospasm, 182 Finger(s). See also Thumb boutonnière deformity of, 1388–1389, 1388f, 1389f flexor pulley system injury of, 1392 foreign body of, 538, 539f fracture of, 1393–1398 metacarpal, 1393–1394, 1393f, 1394f, 1395f, 1411–1414, 1411f, 1412f, 1413f phalangeal, 1338, 1394–1398, 1395f, 1396f, 1397f, 1406–1411, 1406f–1410f Jersey, 1390–1392, 1391f pediatric, 1424–1428, 1424f–1427f ligamentous injury of, 1379–1387. See also Carpometacarpal joint; Interphalangeal joint; Metacarpophalangeal joint mallet, 1388, 1420–1422, 1420f, 1421f tendon injury to, 1387–1392, 1388f, 1389f, 1391f trigger, 626 Fingernail injury, 1428, 1428f, 1429f–1430f Fingertip injury, 1428–1430, 1428f, 1429f– 1430f Finkelstein’s test, 1355, 1356f Fisher exact test, 114 Fist test, in femoral stress fracture, 1479 Flail chest, 526, 893. See also Rib(s), fracture of Flake sign, in triceps tendon rupture, 1170– 1171, 1171f Flatfoot acquired, 631, 1981 medial tibial syndrome and, 15 orthotic devices for, 2044, 2046f Flexibility in children/adolescents, 464–465 excessive. See Hypermobility in foot and ankle injury, 2176–2183, 2176b, 2182f of shoes, 1909, 1910f, 2188–2189, 2189f in turf toe, 2182 Flexor carpi radialis tendinitis, 625–626, 1356–1357 Flexor carpi ulnaris tendinitis, 625, 1356–1357 Flexor digitorum longus tendon transfer, 1980 Flexor digitorum profundus tendon avulsion of (Jersey finger), 1390–1391, 1391f neglected, 1391–1392 pediatric, 1424–1428, 1424f–1427f disruption of, 1392 Flexor hallucis brevis tendon, 2065, 2066f Flexor hallucis longus tendon in Achilles tendon reconstruction, 2039f–2040f, 2041 anatomy of, 1984 injury to, 30, 1983–1987 in ballet dancer, 1984, 1985–1987 differential diagnosis of, 1984–1985 evaluation of, 1984–1985 magnetic resonance imaging in, 563, 564f nerve injury with, 1984, 1985 treatment of, 1985–1987, 1986f repair of, 1985–1987 Flexor-pronator injury, 622 Flexor pulley system, disruption of, 1392 Flexor training, in elbow rehabilitation, 251–253, 252f, 253f Floating ribs, 895–896, 896f Floating shoulder, 863, 865f Fluconazole, in dermatophyte infection, 199–200, 200t Fluid(s) for adolescent, 402 articular cartilage, 41

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xvii

Fluid(s) (Continued) for children, 402 cold/cool, 401 in cramps, 12 after exercise, 402 before exercise, 401 during exercise, 401 glycerol in, 402 guidelines for, 401, 402t in heat illness prevention, 494 at high altitude, 505 ligament, 34 loss of, 401 for pediatric athlete, 402, 465–466 requirements for, 401–402, 402t types of, 401 Fluoroquinolones, tendinopathy with, 30 Fluoroscopy, 534, 535f computed tomography, 585, 586f in corticosteroid injection, 582–584, 584f in methacrylate cement injection, 585, 585f Folate, requirements for, 406b Folliculitis, 194–195, 194t, 195f, 196f Fondaparinux, in venous thromboembolism prevention, 378–384, 381t Food(s) carbohydrate in, 403, 403t, 404 fat in, 405, 405t glycemic index of, 403, 403t protein in, 405, 405t thermic effects of, 498 Foot (feet). See also Ankle; Hallux; Metatarsal(s); Toe(s) ankle linkage of, 1870–1872, 1872t cavus, 1850, 1900, 1963, 2183 deformity of, patellar stability and, 1550 entrapment neuropathy of, 2057–2063 deep peroneal nerve, 2061–2062, 2061f nerve to abductor digiti quinti, 2059–2060, 2060f posterior tibial nerve, 2057–2059, 2057f, 2058b, 2059f superficial peroneal nerve, 2062–2063, 2062f sural nerve, 2060–2061, 2061f fasciitis of. See Plantar fasciitis in female athlete, 490–491, 491b frostbite of, 203, 203f hyperpronation of, 2183 injury to, 2171–2205. See also specific injuries and conditions environment and, 2173 extrinsic factors in, 2172, 2172t, 2173f flexibility and, 2176–2183, 2176b, 2182f. See also Hypermobility incidence of, 2174–2175, 2175t intrinsic factors in, 2171–2172 overuse, 628–631 performance-enhancing drugs and, 2173–2174 personality and, 2172 playing surface and, 2192–2197. See also Playing surface risk factors in, 2171–2174, 2172t, 2173f shoewear and, 2183–2192. See also Shoes, injury and training techniques and, 2172–2173, 2173f keratosis of. See Plantar keratoses ligament injury in, 1947–1960. See also ­Bifurcate sprain; Lisfranc sprain; Subtalar sprain; Turf toe osteochondral lesions of, 2142–2171, 2142t, 2143f, 2143t calcaneal, 1973–1974, 1974f, 2054, 2143f, 2162, 2162f

xviii

Index

Foot (feet) (Continued) cuneiform, 2143f, 2166, 2166f fifth metatarsal base, 2143f, 2167–2169, 2168f, 2169f metatarsal head, 599, 1973, 1973f, 2143f, 2166–2167, 2167f, 2168t navicular, 599, 1972–1973, 1973f, 2143f, 2162, 2163f talar, 2143–2153, 2143f. See also Osteochondrosis (osteochondroses), talar overuse injury to, 628–631 pitted keratolysis of, 196, 196f plantar fasciitis of. See Plantar fasciitis plantar keratosis of. See Plantar keratoses retrocalcaneal bursitis of. See Retrocalcaneal bursitis stress fracture of, 646–650, 648f, 2012–2017, 2012t. See also specific fractures age and, 2014 anatomic factors in, 2012–2014 ankle, 2017–2018, 2017f, 2018f, 2019f, 2020f biomechanics of, 2013 diagnosis of, 2014–2015 gender and, 2014 hindfoot and midfoot, 2018–2024, 2021f, 2022f, 2023f history in, 2014 imaging in, 2014–2015, 2015f, 2016f metatarsal, 2024–2030, 2025f, 2026f, 2027f, 2028f physical examination in, 2014 previous surgery and, 2014 risk factors for, 2013–2014 treatment principles in, 2015–2017, 2016t in systemic illness, 1974 tendon injury of. See Achilles tendon; Flexor hallucis longus tendon; Peroneus brevis tendon; Peroneus longus tendon; Tibial tendon Football cervical spine injury in, 665–669, 686–690, 687f, 688f, 689f, 690f airway in, 667, 668f, 670f facemask removal in, 667, 667f helmet removal in, 668–669, 670f immobilization in, 666–667, 667f transport in, 669, 669f jet lag and, 458 pediatric, humeral fracture in, 1068–1069 posterior shoulder dislocation in, 937, 937f spear tackler’s spine and, 694, 695f, 710 sternomanubrial dislocation in, 897, 898f Force at joint, 87 patellofemoral. See Patellofemoral joint reaction force shear, 94f, 95 vector, 86, 90f Forearm training, in elbow rehabilitation, 254–255, 254f, 255f Foreign body, ultrasonography of, 538, 539f Formoterol, in exercise-induced bronchospasm, 182 Fosinopril, 160t Fovea capitis, 1452 Fracture(s). See also Stress fracture acromion, 866–867, 866f, 867b, 867f, 873 ankle, 1964–1969, 1965f, 1966f, 1967f, 1968f, 1969f dislocation with, 1945, 1946f atlas (C1), 677, 677f, 678f, 695–696, 705, 706f avulsion anterior process of calcaneus, 1954, 1954f, 2153–2156, 2154t, 2155t

Fracture(s) (Continued) hip and pelvis, 553, 555, 1474–1475, 1475f, 1489 iliac spine, 553, 556f, 1475 imaging of, 553, 555, 556f, 557, 557f ischial, 1489, 1489f lesser humeral tuberosity, 1175–1176 lesser trochanter, 1475 rib, 895–896, 896f axis (C2), 678, 695–696, 707–709 Bennett’s, 1402–1403, 1402f, 1403f, 1411, 1412 boxer’s, 1393–1394, 1393f, 1412–1413, 1413f bucket handle, in child abuse, 595–596, 596f calcaneal, 2153–2156, 2154t, 2155t capitate, 1348–1349, 1368 cervical spine C1-C3, 677–678, 677f, 678f, 695–696 in children/adolescents, 704–708, 705, 706f C3-C4, 678–679, 696–697 in children/adolescents, 708–709 C4-C7, 679–681, 680f, 681f, 682f, 696f, 697–698, 697f, 698f in children/adolescents, 708–709 compression, 680–681, 680f, 681f, 682f, 683f in children/adolescents, 709, 709f instability in, 675–676, 676f management principles in, 675–677, 676f surgical treatment of, 676–677 vertebral body, 698, 698f in child abuse, 595–596, 596f clay shoveler’s, 709 deep venous thrombosis after, 384 epiphyseal distal femur, 65f, 1641–1642, 1641b, 1641f, 1642f, 1643–1644, 1644b proximal tibia, 1642–1644, 1643f, 1643t, 1644b fatigue. See Stress fracture and at specific bones glenoid neck, 861f, 862–863, 862f, 864f, 865b, 865f hamate, 1340–1347, 1346f, 1347t hangman’s, 678, 707–708 healing of, 78–79, 79t, 80f, 81t blood supply and, 69 callus formation in, 79, 80f, 81t electricity effect on, 80 growth factors in, 79–80 problems of, 80–81, 82f, 83f ultrasound effect on, 80 humeral. See Humeral fracture Jefferson, 677, 677f, 678f, 705, 706f Jones’, 1969–1970, 1971f lumbar spine, 733–736, 733f, 735f lunate, 1350, 1353f, 1370, 1370f march, 1849. See also Tibia, stress fracture of metacarpal, 1393–1394, 1393f, 1394f, 1395f nasal, 525 nonunion of, 80–81, 82f, 83f occult, imaging in, 546, 552, 557 odontoid, 677–678, 678f in children/adolescents, 707 olecranon, 1271–1276. See also Olecranon, fracture of on-field, 525, 527 osteochondral. See Osteochondral fracture osteoporotic imaging of, 546 radionuclide imaging in, 547f treatment of, 585, 585f pars interarticularis, 726, 726f, 755, 756, 756b patellar, 1572–1577, 1574f, 1576f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Fracture(s) (Continued) phalangeal, 1394–1398, 1395f, 1396f, 1397f pisiform, 1350, 1352f, 1370–1371 rib, 525, 893–896, 893t, 895f, 895t, 1187 Rolando, 1403 sacral, 736, 736f scaphoid, 1335–1340, 1335f, 1336f, 1336t, 1338f, 1341f–1344f, 1364–1368, 1365f, 1366f, 1367f Segond’s, 1650 sternum, 896–900, 897f, 898f, 899f stress. See Stress fracture and at specific bones talar, 2153–2156, 2154t, 2155f, 2155t, 2156f thoracic spine, 733–736, 733f, 734f in child/adolescent, 755, 755f trapezium, 1345–1348, 1349f, 1370 triplane, 1966, 1967f, 1968–1969 triquetrum, 1349–1350, 1351f, 1368–1369, 1369f Free-body diagrams, 87, 88f Free fatty acids, caffeine effects on, 421 Free-running rhythm, 443, 444f Freeze drying, of allograft, 140–141 Freiberg’s infraction, 599, 1973, 1973f, 2143f, 2166–2167, 2167f, 2168t treatment of, 2166, 2168f Friction playing surface, 2198, 2204, 2204f shoe-related, 1906 Friction-induced injury cutaneous, 202, 202f, 203f iliotibial band, 560, 562f, 627–628, 629f, 630f Frostbite, 203, 203f, 528 Frostnip, 528 Frozen shoulder. See Adhesive capsulitis Fulcrum test, in femoral stress fracture, 1479 Full-can exercise, in shoulder rehabilitation, 244, 244f Fungal infection, 198–200, 198f, 199f Furunculosis, 194–195, 194t

G Gabapentin in complex regional pain syndrome, 363t, 364 in epilepsy, 189, 191t Gadolinium, 550 Gage’s sign, 1476 Gait, 2012–2013 ankle joint in, 1866–1867, 1867f, 1870–1872, 1872t metatarsal break in, 1870, 1871f orthosis effects on, 1902–1904, 1903f plantar aponeurosis in, 1870, 1871f in posterior cruciate ligament injury, 1692 in recurrent patellar dislocation, 1554–1555 subtalar joint motion in, 1870–1872, 1872t in varus malalignment, 1808–1809, 1809f, 1810f, 1811f windlass mechanism in, 1870, 1871f Gait training, in knee rehabilitation, 295, 296f Game of Shadows (Fainaru-Wada and Williams), 410 Gamekeeper’s thumb, 1399–1401, 1400f, 1415 Gamma irradiation, in allograft sterilization, 139 Ganglion cyst suprascapular nerve compression with, 617, 1121, 1121f, 1123 wrist, 1444–1445, 1445f Ganglionectomy, 1445 Gap junctions, 24 Gap test, in varus malalignment, 1808, 1808f Gardner-Wells tongs, in cervical spine fracture, 675

Index Gastrocnemius muscle antagonist action of, 1586 release of, in popliteal artery entrapment, 1844–1845, 1845f, 1846b, 1846t stretching exercise for, 291, 291f Gastrointestinal system cocaine effects on, 429 preparticipation examination of, 512 Gastrosoleus muscle, strengthening exercises for, 273, 275f Gemellus muscle, 1454t, 1455f Gender. See also Female athlete anterior cruciate ligament biomechanics and, 1584 anterior cruciate ligament injury treatment and, 1651 foot stress fracture and, 2014 knee arthroplasty and, 1799 pediatric/adolescent differences in, 476–477, 477t popliteal artery entrapment and, 1837 proximal humeral fracture and, 1067 strength and, 476 wrist injury and, 1363 Gene therapy, 422–423 in knee cartilage lesions, 1776 Genetic testing, in hypertrophic cardiomyopathy, 165 Genicular artery, 1645 Genie stretch, 290f, 291 Genitourinary system on-field injury to, 526–527 preparticipation examination of, 512 Giant cell tumor, 605, 605f Giardiasis, 152 Gilmore’s groin, 1463–1464 GIRD (glenohumeral internal rotation deficit) disorder, 979–980, 980f Girls. See Female athlete GLAD (glenolabral articular disruption) lesion, 976, 978f Glading, 430 Glasgow Coma Scale, 662t Glenohumeral internal rotation deficit (GIRD) disorder, 979–980, 980f Glenohumeral joint. See also Shoulder anatomy of, 769, 769f, 770–775, 932 bony, 771–772, 771f, 772f capsular, 772–773, 772f, 910 labral, 774 ligamentous, 773–774, 909, 910f muscular, 770–771, 770f. See also Infra­ spinatus; Subscapularis; Supraspinatus; Teres minor pediatric, 782–783, 783f vascular, 911–912 arthritis of. See Glenohumeral joint ­osteoarthritis arthrography of, 969 arthrosis of, 929 biomechanics of, 777–778, 787–788, 787f, 788f, 932–933, 990–994. See also Overhead throwing capsule of, 772–773, 772f, 782–783, 783f, 910 chondrolysis of, 984, 984f, 1105t, 1106 computed arthrotomography of, 950–951 computed tomography of, 950, 951f, 969 conventional arthrography of, 949–950, 949f degenerative disease of. See Glenohumeral joint osteoarthritis dislocation of. See Glenohumeral joint instability infection of, 984 instability of. See Glenohumeral joint ­instability

Glenohumeral joint (Continued) intracapsular pressure of, 774 kinematics of, 777–778, 2212–2213, 2213f, 2213t ligaments of. See Glenohumeral ligament(s) magnetic resonance arthrography of, 953–954 magnetic resonance imaging of, 953, 953t, 960t, 969, 969f, 970f osteoarthritis of. See Glenohumeral joint osteoarthritis pediatric anatomy of, 779–780, 779f, 782–783, 783f biomechanics of, 787–788, 787f, 788f dynamic stability of, 789–790, 790f static stability of, 788–789, 789f radiography of, 915–916, 947–949, 947b, 968–969 anteroposterior view for, 947, 948f, 1037, 1039f–1040f axillary lateral view for, 947–948, 948f, 1038, 1042f Grasbey view for, 947, 948f scapular Y view for, 948–949, 948f, 1037, 1041f Stryker notch view for, 916, 948f, 949 Velpeau axillary view for, 1038, 1042f West Point view for, 916 rheumatoid arthritis of, 1105t, 1106, 1107f arthroplasty in, 1114 arthroscopic treatment of, 1110–1111 rotations of, 769, 770f, 778 stability of, 912, 932–933 active, 774–775, 789–790, 789f passive, 771–774, 771f, 788–789, 789f translations of, 769, 777–778, 2212–2213, 2213f, 2213t ultrasonography of, 951–953, 952f, 953t Glenohumeral joint instability, 616, 769, 909–931 anterior, 914t. See also Glenohumeral joint instability, pediatric, anterior arthroscopic treatment of, 919–923, 920f, 921f–923f, 927–929 Bankart lesion with, 912, 919f–923f capsular laxity with, 912 clinical presentation of, 913–916, 913b complications of, 929–930 magnetic resonance arthrography in, 973–974, 973f, 974f, 974t nonoperative treatment of, 916f, 917–919, 917f, 918f in older patient, 974–975 open stabilization for, 923–924, 928–929 operative treatment of, 917f, 919–924, 920f–923f, 927–929 physical examination in, 914–915, 915f thermal capsulorrhaphy for, 929–930 vascular injury with, 1139–1140, 1140f anterior apprehension test in, 914, 915f, 939, 939f arthropathy with, 1104–1105, 1104b, 1114 arthroscopic anterior stabilization in, 919–923, 920f–923f arthroscopic posterior stabilization in, 924–926, 926f–927f classification of, 913, 919b, 967–968 clinical presentation of, 913–916, 913b, 914t computed tomography in, 543, 916 in female athlete, 489 Jobe’s relocation test in, 914, 915f load and shift test in, 914–915 magnetic resonance arthrography in, 973–975, 973f, 974f, 974t, 975f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xix

Glenohumeral joint instability (Continued) magnetic resonance imaging in, 916, 969, 969f, 970f multidirectional, 616, 913 arthroscopic treatment of, 930–931, 930f magnetic resonance arthrography in, 973 pain pattern in, 913–914 pathoanatomy of, 912–913 pediatric, 468, 469f, 932–946 anterior, 933–936 axillary nerve in, 933, 934f clinical presentation of, 933, 934f imaging in, 934, 935f nonoperative treatment of, 935, 935f, 936f operative treatment of, 935–936 physical examination in, 933, 934f recurrent, 914, 914t, 938–941, 939f, 940f, 941f reduction for, 935, 935f, 936f return to play after, 936 atraumatic, 942–946, 943f, 944f nonoperative treatment of, 945, 945f operative treatment of, 945 return to play after, 946 classification of, 933, 933b incidence of, 933 posterior, 936–938, 937f, 938f recurrent, 941–942, 942f return to play after, 938 physical examination in, 914–915, 915f posterior, 913 arthroscopic treatment of, 924–927, 925f, 926f–927f magnetic resonance arthrography in, 975, 975f open capsular shift procedure for, 927 operative treatment of, 924–927, 926f–927f pediatric, 936–938, 937f, 938f, 941–942, 942f radiography in, 915–916, 968–969, 969b. See also Glenohumeral joint, radiography of recurrent anterior, 914, 914t, 938–941, 939f, 940f, 941f posterior, 941–942, 942f sulcus sign in, 914, 915f treatment of, 916–929, 916f, 917f algorithms for, 916f, 917f arthroscopic anterior stabilization in, 919–923, 920f–923f arthroscopic posterior stabilization in, 924–927, 925f–927f complications of, 929–930 historical perspective on, 912 immobilization in, 917–919 in multidirectional instability, 930–931, 930f nonoperative, 917–919, 917f, 918f operative, 919–927, 920f–923f, 925f–927f reduction in, 916–917, 917f return to play after, 929 unidirectional, 616 vascular injury with, 1139–1140, 1140f, 1141f venous thrombosis with, 1140, 1141f Glenohumeral joint osteoarthritis, 929, 1104–1119 chondrolysis-related, 1104b, 1105t, 1106 classification of, 1104, 1104b, 1108, 1109f computed tomography arthrography in, 1108 evaluation of, 1107–1108 glenoid version in, 1108, 1109f instability-related, 1104b, 1105–1106, 1105t, 1106f, 1114

xx

Index

Glenohumeral joint osteoarthritis (Continued) physical examination in, 1108 post-traumatic, 1104–1105, 1104b, 1105t primary, 1104, 1105f, 1105t radiography of, 1108, 1108f secondary, 1104–1108, 1104b treatment of, 1109–1116 algorithm for, 1115f–1116f arthroplasty in, 1113–1114, 1113f, 1114f, 1115, 1116f rehabilitation after, 1116–1118, 1116t, 1117f, 1118f arthroscopic, 1110–1111, 1115, 1115f complications of, 1118, 1118t, 1119t rehabilitation after, 1116, 1116t arthroscopic débridement in, 1110, 1110f, 1115 biologic resurfacing in, 1112–1113, 1112f glenoidplasty in, 1110 humeral head resurfacing in, 1111–1112, 1112f joint-resurfacing, 1111–1112, 1111f, 1112f joint-sparing, 1110–1111, 1110f nonoperative, 1109–1110 osteochondral allograft in, 1111, 1111f, 1115, 1115f return to play after, 1118, 1118t Glenohumeral ligament(s), 772–773, 772f humeral avulsion of. See also Glenohumeral joint instability, anterior bony, 555 magnetic resonance arthrography in, 978–979, 979f magnetic resonance imaging in, 575, 575f inferior anatomy of, 772f, 773, 774, 789, 789f, 910, 910f, 911f, 911t function of, 911f, 2194 functional loosening of, 1020, 1020f magnetic resonance arthrography of, 575, 575f, 971, 971f magnetic resonance arthrography of, 573, 575, 575f, 971–972, 971f, 972f, 973f middle anatomy of, 772f, 773, 789, 789f, 910, 910f, 911f, 911t, 1017–1018, 1017f cord-like, 1017–1018, 1017f, 1021, 1023f function of, 910, 911f magnetic resonance arthrography of, 971, 972f superior anatomy of, 772f, 773, 788–789, 910, 910f, 911f, 911t, 966, 990, 991f function of, 910, 911f magnetic resonance arthrography of, 971–972, 973f Glenoid. See also Glenoid labrum; Glenoid neck anatomy of, 772, 772f biologic resurfacing of, 1112–1113, 1112f epiphyseal line of, 860f fracture of. Glenoid neck, fracture of in children, 872, 875 classification of, 857, 858f comminuted, 872, 872b computed tomography in, 861, 862f vs. epiphyseal line, 860f intra-articular, 867–872, 868f, 869f, 870f, 871f, 872b radiography of, 861, 861f return to play after, 873, 875 type I (rim), 867–868, 868f, 869f, 870f, 872, 872b, 873–875 type II-V (fossa), 871f, 872, 872b, 873, 874f

Glenoid (Continued) ossification of, 859 posterosuperior impingement of. See ­Shoulder impingement, internal version of, in glenohumeral joint ­osteoarthritis, 1108, 1109f Glenoid labrum anatomy of, 774, 909–910, 969–971, 970f, 1016–1019, 1017f normal variations in, 970f, 972, 972f anterior avulsion of. See Bankart lesion anterosuperior, 1017–1018, 1017f biceps tendon insertion on, 1018, 1018f biomechanics of, 1016, 1019–1021 bumper effect of, 1016 cartilage undermining of, 970f, 972 cyst of, 580–581, 582f, 964–965, 965f, 980, 980t histology of, 1017, 1017f, 1018, 1018f inferior, 1017, 1017f magnetic resonance arthrography of, 969–971, 970f sublabral hole and, 1017, 1017f superior, 1016–1017. See also SLAP (superior labrum, anterior to posterior) lesion throwing-related injury to, 1219–1221, 1220f vascularity of, 1018–1019 Glenoid neck, fracture of, 862–863, 862f, 864f, 865b, 865f computed tomography in, 861, 862f, 864f return to play after, 873 treatment of, 873, 874f Glenoidplasty, in osteoarthritis, 1110 Glenolabral articular disruption (GLAD) ­lesion, 976, 978f Glenopolar angle, 1713 Glucagon exercise effects on, 217t, 218 parenteral, 179 Glucosamine, 409 in knee arthritis, 1774, 1791 Glucose for hypoglycemia, 178–179, 179b metabolism of, during exercise, 173–175, 174f, 174t γ-Glutamyltransferase, in anabolic-androgenic steroid user, 416 Gluteal artery, 1454, 1501 Gluteal muscle raise exercise, in knee ­rehabilitation, 267, 268f Gluteal nerve, 1454, 1501 Gluteus maximus, 1454t, 1455f neuromuscular activation exercises for, 285–286 Gluteus medius, 1454t, 1455f neuromuscular activation exercises for, 257–258, 258f, 259f, 285–286 Gluteus medius tendon, 567–568 Gluteus minimus, 1454t, 1455f Gluteus minimus tendon, 567–568 Glycemic index, 403, 403t Glycerol for hyperhydration, 402 ingestion of, 402 Glycogen, 211 depletion of, 403 muscle, 400–401 depletion of, 403 repletion of, 403, 403t, 404 Glycolysis, 210–211, 210f aerobic, 211 anaerobic, 211 Glycosaminoglycans immobilization effects on, 228 tendon, 23–24

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Godfrey’s test, in posterior cruciate ligament injury, 1690–1691, 1691f Golfer’s elbow, 620–621, 1205–1206, 1205f treatment of, 1205–1206, 1205f, 1206f Gracilis, 1454t, 1455f, 1485 strain of, 1460–1461 Graft(s). See also Allograft(s); Autograft(s) in ACL reconstruction. See Anterior cruciate ligament (ACL) reconstruction, graft for articular cartilage, 53–55. See also ­Osteochondral allograft/autograft allograft, 54 autograft, 53–54 preservation of, 54 bone, 81, 83t allograft, 81, 83t, 137 autograft, 81, 83t cancellous, 81, 83t cortical, 81, 83t osteoarticular (osteochondral), 81 vascularized, 81 in Kienböck’s disease, 1377 in scaphoid nonunion, 1341–1345, 1342f–1344f ligament, 35–39 allograft, 38, 142–143 autograft, 36–38 Dacron augmentation of, 36 xenograft, 38 meniscal, 63–64 allograft, 64, 143–144 synthetic, 64 nail bed, 1429, 1430f osteochondral allograft, 141, 144–145 in capitellar osteochondritis dissecans, 1244, 1245f tendon, 35–39 in ulnar collateral ligament reconstruction, 1401, 1401f Great toe. See Hallux Great vessels, injury-related compression of, 821, 822f Grey Turner’s sign, 526 Griseofulvin, in dermatophyte infection, 198–199, 200t Groin, contusion of, 1459–1460 Ground reaction force, 86 Growth factors in bone, 67 in fracture healing, 79–80 in knee cartilage lesions, 1776 in muscle injury treatment, 16 in tendon injury treatment, 31 Growth hormone, 419–420 bone effects of, 72 exercise effects on, 217, 217t, 421 Growth plate. See Epiphysis; Physis Guanfacine, in hypertension, 160t Gymnasts elbow disorders in, 1236, 1246 ulnocarpal impingement in, 1376 wrist disorders in, 1375–1376, 1376f Gynecomastia, anabolic-androgenic steroids and, 415

H H band, 6f H zone, 4f, 5 HAGL (humeral avulsion of glenohumeral ligament) lesion, 575, 575f, 978–979, 979f bony, 555

Index Haglund’s deformity, 2030, 2031f. See also Retrocalcaneal bursitis magnetic resonance imaging of, 2034, 2034f treatment of, 2036, 2038, 2038f Hair follicles, bacterial infection of, 194–195, 194t Hallux metatarsophalangeal joint injury of. See Turf toe sesamoids of. See Sesamoid(s) subungual exostosis of, 2105–2107, 2105t, 2106b, 2106f, 2107f Syme amputation of, 2099, 2104f taping of, 2090, 2093f Hallux interphalangeal angle, 2066, 2066t Hallux rigidus, 2182, 2182f Hallux valgus, 2064–2081 anatomy of, 2064–2068, 2064f, 2065f, 2066f, 2066t, 2067f biomechanics of, 2067–2068, 2067f, 2068f classification of, 2068, 2068b distal metatarsal articular angle in, 2066, 2066t, 2072 evaluation of, 2069 hallux interphalangeal angle in, 2066, 2066t hallux valgus angle in, 2066, 2066f, 2066t history in, 2069, 2069b joint congruency in, 2066, 2067f mild, 2069b, 2071f moderate, 2069b, 2071f nonoperative treatment of, 2069–2070 1-2 intermetatarsal angle in, 2066, 2066f, 2066t operative treatment of, 2070–2081, 2070b, 2071f, 2072b, 2079b Akin procedure in, 2070, 2072b, 2072f, 2073f, 2076, 2078f arthrodesis in, 2075 care after, 2080–2081 Chevron procedure in, 2070, 2073b, 2073f, 2074f, 2076 combined multiple first ray osteotomies in, 2071–2072, 2078f, 2079, 2080f complications of, 2081, 2081b distal soft tissue realignment in, 2070, 2074b, 2075f, 2076, 2078, 2079f in high-performance athlete, 2081 Keller procedure in, 2072, 2075 proximal first metatarsal osteotomy in, 2070–2071, 2076b, 2077f, 2078–2079 return to play after, 2081, 2081b salvage procedures in, 2072, 2075–2076 physical examination in, 2069, 2069b radiography in, 2069, 2069b risk factors of, 2064, 2064f sagittal sulcus in, 2068, 2068f sesamoid dysfunction and, 2090 severe, 2069b, 2071f Hallux valgus angle, 2066, 2066f, 2066t Hallux varus after fibular sesamoidectomy, 2093 mild, 2079 moderate, 2079 postoperative, 2078, 2079f sesamoid dysfunction and, 2090 severe, 2079 Halstead’s maneuver, in thoracic outlet syndrome, 1131, 1132f Hamate, anatomy of, 1319–1320, 1319f Hamate fracture adult, 1340, 1345 classification of, 1340 clinical manifestations of, 1340, 1345 computed tomography in, 543, 544f physical examination in, 1345, 1346f

Hamate fracture (Continued) radiography in, 1345, 1346f return to play after, 1348 treatment of, 1345, 1347t pediatric, 1369–1370 Hammer toe deformity, 2117, 2118t evaluation of, 2119–2120, 2119f, 2120b, 2121f nonoperative treatment of, 2121, 2121f operative treatment of, 2121–2125, 2122f, 2123f, 2124b Hamstring curl, 266, 317f, 325, 329–330, 337 Hamstring muscles anatomy of, 1485, 1485f strain of, 1461–1462, 1486–1489 in children/adolescents, 1489 classification of, 1485, 1485t clinical presentation of, 1486 complications of, 1489 imaging of, 1486f, 1487, 1487f magnetic resonance imaging in, 558, 559f mechanisms of, 335, 335f nonoperative treatment of, 1487 operative treatment of, 1488–1489 physical examination in, 1486–1487, 1486b, 1486f prevention of, 336–337, 337f, 337t, 1488–1489 previous, 336 rehabilitation protocol for, 1488, 1488t return to play after, 1489, 1489b risk factors for, 335–336, 336b treatment of, 1487–1489, 1487b stretching exercises for, 289–290, 290f, 292f, 293 Hamstring raise exercise, in knee rehabilitation, 267, 268f Hand injury. See also Finger(s); Thumb; Wrist adult, 1379–1403 biomechanics of, 1379 epidemiology of, 1379 fracture, 1393–1398 metacarpal, 1393–1394, 1393f, 1394f, 1395f phalangeal, 1394–1398, 1395f, 1396f, 1397f ligamentous, 1379–1387. See also ­Carpometacarpal joint; Interphalangeal joint; Metacarpophalangeal joint management of, 1379 tendon, 1387–1392, 1388f, 1389f, 1391f pediatric, 1404–1430 anatomy of, 1404, 1404f evaluation of, 1404–1405, 1405f fracture, 1405–1414 metacarpal, 1411–1414, 1411f, 1412f, 1413f phalangeal, 1405f, 1406–1411, 1406f–1411f ligamentous, 1414–1420. See also Carpometacarpal joint; Interphalangeal joint; Metacarpophalangeal joint radiography in, 1405 tendon, 1420–1428, 1420f–1422f, 1424f–1427f Hangman’s fracture, 678, 707–708 Hawkins’ sign, 997, 1000f Head injury, 657–663. See also Cervical spine injury concussive, 658–662, 658b, 659t. See also Concussion contracoup, 657, 657f coup, 657, 657f evaluation of, 522, 662t

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxi

Head injury (Continued) Glasgow Coma Scale in, 662t hematoma with, 660–661, 661f hemorrhage in, 660–661, 661f incidence of, 662 on-field, 522–524 prevention of, 662 return to play after, 522, 523t, 659t, 662, 663b risk for, 662, 662b seizure and, 661 treatment of, 660, 662, 662t, 663b Head tilt–jaw lift maneuver, in cervical spine injury, 667, 668f Heart anabolic-androgenic steroid effects on, 416 cocaine effects on, 429 exercise effects on, 218–220 hypothermia effects on, 501 inhalant effects on, 430 nicotine effects on, 428 Heart murmurs, preparticipation examination for, 512 Heart rate exercise effect on, 219, 219f, 220t in female athlete, 476–477 maximal, 219 Heat conductive transfer of, 493, 499 dissipation of, 493, 499 evaporative transfer of, 493, 499 physiologic adaptations to, 493–494 production of, 498–499 radiative transfer of, 493, 499 Heat cramps, 495–496 Heat edema, 530 Heat exhaustion, 494–495, 529 Heat illness/injury, 493–497, 514t, 528, 528f, 529–530 in children/adolescents, 465–466 risk factors for, 494 Heat stroke, 496, 529 Heat syncope, 494, 529–530 Heat therapy, in thoracolumbar spine injury, 731 Heavy-load eccentric training, in Achilles tendon injury, 1999 Heel black, 202, 202f pain in, 2030–2056. See also Plantar fasciitis; Retrocalcaneal bursitis in children, 2037 classification of, 2053 in entrapment neuropathy, 2059–2060 HLA-B27 in, 2047 referred, 2047 in sarcoidosis, 2051 in seronegative spondyloarthritis, 2033–2034, 2047 spur of, 2042–2043, 2042f. See also Plantar fasciitis Heel cups in plantar fasciitis, 2053f in retrocalcaneal bursitis, 2035 Heel lift, in Achilles tendon injury, 1998 Heel pad in retrocalcaneal bursitis, 2037, 2038f for shock absorption, 2052 Heel touch exercise, 308f, 324 Height, in preparticipation examination, 509 Helmet in cervical spine injury, 687, 687f, 689, 689f removal of, in cervical spine injury, 668–669, 670f

xxii

Index

Hematoma epidural, 523, 660, 661f heterotopic ossification treatment and, 1297–1298 intracerebral, 660–661, 661f in osteochondral fracture, 51 rectus sheath, 526 septal, 525 subdural, 660, 661f subungual, 2098 Hemiarthroplasty, shoulder, 1046, 1048f Hemodialysis, cramps during, 12 Hemoglobin A1c, 176–177, 176t Hemorrhage. See also Hematoma intracranial, 659, 660–661, 661f muscle, 17–18, 18f pelvic fracture and, 541 Hemothorax, on-field, 526 Heparin prophylactic, in venous thromboembolism, 378–384, 381t, 382f in venous thromboembolism, 384–385, 384t Hepatitis A virus infection, 154 Hepatitis B virus infection, 154–155 Hepatitis C virus infection, 155 Hepatitis D virus infection, 155 Hepatitis E virus infection, 154–155 Hepatocellular carcinoma, anabolic-androgenic steroids and, 416–417 Hepatomegaly, clearance for participation and, 513 Hernia, 1462–1464 clearance for participation and, 513 femoral, 1463 inguinal, 1462–1463 sports, 1463–1464 Herpes gladiatorum, 197, 197f Herpes labialis, 197, 197f Herpes simplex virus infection, 197, 197f Heterophil antibody test, 150 Heterotopic ossification, 82 elbow, 1289–1300 alkaline phosphatase levels in, 1292 anatomy of, 1290, 1293 anterior, 1290 anterolateral, 1296 anteromedial resection in, 1296–1297 chemoprophylaxis in, 1294, 1297 classification of, 1293–1294 clinical presentation of, 1290 differential diagnosis of, 1293 distal humeral fracture and, 1256 etiology of, 1289 functional classification of, 1293–1294 genetic factors in, 1290 history in, 1290 intra-articular, 1296 magnetic resonance imaging in, 1293 neurologic injury and, 1291 nonoperative treatment of, 1295 operative treatment of, 1295–1297, 1295f, 1298f–1299f complications of, 1297–1298 management after, 1297 pathophysiology of, 1289–1290 physical examination in, 1291–1292 posterolateral, 1290, 1296 posteromedial, 1296 prevention of, 1294–1295, 1297 radiation therapy in, 1294–1295, 1297 radiography in, 1292–1293 radioulnar, 1297 risk factors for, 1290–1291, 1290f–1292f trauma and, 1277, 1290, 1290f–1292f ultrasonography in, 1293

Heterotopic ossification (Continued) hip, 82, 1460 interosseous talocalcaneal ligament, 1933 muscle. See Myositis ossificans after proximal humeral fracture, 1050 proximal radioulnar joint, 1297 radiation in, 82, 1277, 1294–1295, 1297 High altitude, 502–503 adverse effects of, 504–505 cerebral edema at, 504–505 competition at, 503–504 definition of, 502 diet and, 505 fluids and, 505 physical characteristics of, 502 physiologic effects of, 503 pulmonary edema at, 504 simulation of, 504 training at, 220, 504, 505 erythropoietin and, 420 High tibial osteotomy, 1804–1835 arterial injury with, 1832 arthrofibrosis after, 1832 closing wedge, 1821–1824, 1824f–1826f complications of, 1830–1833 outcomes of, 1816–1818, 1817t, 1825–1829, 1828f, 1829f complications of, 1830–1835 contraindications to, 1816 correction wedge determination in, 1810–1811, 1812f, 1813f deep venous thrombosis after, 1833 delayed union after, 1831–1832 gap angle in, 1812, 1813f iliac crest harvest site pain after, 1833 indications for, 1814–1816, 1815t, 1833, 1835 nonunion after, 1831–1832 opening wedge, 1818–1821, 1819f–1823f complications of, 1830–1833 open wedge angle measurement in, 1812–1814, 1813f, 1814f, 1814t, 1815t outcomes of, 1816–1818, 1817t, 1829–1830, 1830f patella infera with, 1832–1833 patellar height measurement in, 1811 peroneal nerve injury with, 1832 rehabilitation after, 1824–1825, 1827t results of, 1816–1818, 1817t return to play after, 1824–1825 teeter effect after, 1830–1831 tibial plateau fracture after, 1832 timing of, 1814–1815, 1815t varus deformity recurrence after, 1831 weight-bearing line ratio in, 1810, 1811f, 1812f High-voltage galvanic stimulation in inflammation, 230, 231f in lateral epicondylitis, 618 Hill-Sachs lesion, 912–913, 920f magnetic resonance arthrography of, 973, 973f radiography of, 968–969 Hindfoot, stress fracture of, 2018–2024, 2021f, 2022f, 2024f Hip. See also Acetabular labrum; Acetabulum; Femur anatomy of, 1452–1453, 1453f arthroplasty of, 1504–1512 anatomy for, 1500–1502 approaches to, 1505 bearings for, 1505–1506 biomechanical aspects of, 1501–1502

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Hip (Continued) computed tomography after, 542, 542f high-offset femoral component in, 1502 minimally invasive, 1505 return to play after, 1507–1508, 1507t–1508t stability with, 1502 technique of, 1508–1512, 1509f–1512f arthroscopy of, 1473–1474, 1473t, 1474f. See also Arthroscopy, hip in acetabular labral tears, 1470f indications for, 1473t in ligamentum teres rupture, 1473f in loose body removal, 1472f avulsion fracture of, in children, 1474–1475, 1475f, 1489 blood supply to, 1501 chondral injury of, 1472, 1473t computed tomography–guided biopsy of, 585, 586f degenerative disease of, 1467, 1467b classification of, 1502, 1502b clinical presentation of, 1502–1503 etiology of, 1500 imaging of, 1503 nonoperative treatment of, 1503–1504, 1504b, 1504f operative treatment of. See Hip, ­arthroplasty of pain in, 1503 physical examination in, 1503 primary, 1500 range of motion testing in, 1503 secondary, 1500 dislocation of, 1464 femoroacetabular impingement at, 1471–1472, 1471f–1472f forces on, 1453 fracture of, 1464–1466, 1465f avulsion, 1474–1475, 1475f in children/adolescents, 597, 598f, 1474–1475, 1475f, 1489 computed tomography of, 542 stress, 1464–1466, 1465f injury to, 1455–1474, 1456b. See also specific injuries bone, 1464–1467, 1465f in children/adolescents, 469, 470f, 1474–1477, 1475f–1476f, 1476t intra-articular, 1469–1474 nerve, 1467–1469 soft tissue, 1455–1464, 1456b innervation of, 1453–1455, 1455f ligaments of, 1452–1453, 1453f loose bodies of, 1472, 1472f magnetic resonance imaging arthrography of, 536–537 muscles about, 1453, 1454t, 1455f, 1500–1501 heterotopic ossification of, 1460 strains of, 1460–1462, 1460t, 1485–1497. See also at specific muscles neonatal, 590, 590f nerve entrapment at, 1467–1469 pain in, 1452f range of motion of, 1452t resurfacing procedures for, 1506–1507 snapping, 1458–1459 stress fracture of, 1464–1466, 1465f synovial disease of, 1473 vasculature of, 1453–1455 Hip abduction exercise, in knee rehabilitation, 258, 258f Hip extension exercise, in knee rehabilitation, 258, 258f, 267

Index Hip hyperextension exercise, in knee ­rehabilitation, 267 Hip lift, single-leg, in core training, 281 Hip pointer, 1459 Hippocampus, marijuana effects on, 426 HIV infection. See Human immunodeficiency virus (HIV) infection Hockey, cervical spine injury in, 690 Homans’ sign, 374 Hook of hamate fracture. See Hamate fracture Hop test, in femoral stress fracture, 1479 Hopping, single-leg, in knee rehabilitation, 296, 298f, 315f, 329 Horn blower’s sign, in glenohumeral joint osteoarthritis, 1108 Horseback riding, proximal humeral fracture in, 1067–1068, 1068f, 1069 Hot-tub folliculitis, 194–195, 196f Hug exercise, dynamic, 241, 241f Human immunodeficiency virus (HIV) ­infection, 153–154, 514t allograft transmission of, 141 exercise and, 154 prevention of, 153–154 testing for, 154 Human papillomavirus (HPA) infection, 197–198 Humeral avulsion of glenohumeral ligament (HAGL) lesion, 575, 575f, 978–979, 979f bony, 555 Humeral circumflex artery, 1033–1034, 1034f, 1036f Humeral condyles lateral, fracture of, 1280t, 1284–1285, 1284f medial, fracture of, 1280t, 1286 ossification of, 1228 Humeral epicondyles fracture of, in children, 1238 lateral, fracture of, 1256 pediatric, 1284–1285 medial avulsion fracture of, 1183–1186, 1183b, 1183f, 1184f, 1185f fracture of, 1285–1286, 1286f ossification of, 1228 Humeral fracture (distal), 1250–1258 evaluation of, 1250–1251, 1250f–1252f nonunion of, 1278 operative treatment of, 1251–1256 complications of, 1256, 1258 countersunk threaded screw in, 1255 exposure for, 1251–1253, 1252f extensile lateral exposure in, 1253, 1253f fixation techniques in, 1253–1256, 1253f–1255f, 1257f heterotopic ossification with, 1256 home run screw in, 1254–1255, 1255f implant-related complications of, 1258 locking screw plating in, 1254–1255 metaphyseal comminution and, 1255 nonunion after, 1256, 1258 olecranon osteotomy in, 1252–1253, 1252f orthogonal plating in, 1253–1254, 1253f outcomes of, 1256 parallel plating in, 1254, 1254f, 1255–1256 rehabilitation after, 1277 stiffness after, 1256 triple plating in, 1254, 1254f ulnar nerve injury with, 1256 pediatric, 1280t, 1281–1283 cosmetic deformity after, 1282–1283 lateral condyle, 1280t, 1284–1285, 1284f lateral epicondyle, 1284–1285 medial condyle, 1280t, 1286

Humeral fracture (distal) (Continued) medial epicondyle, 1285–1286, 1286f supracondylar, 596–597, 597f, 1280t, 1281–1283, 1282f T-condylar, 1283 transphyseal, 1283 vascular injury with, 1282 Humeral fracture (proximal), 1033–1056 anatomy of, 1033–1034, 1033f, 1034t classification of, 1035–1036, 1037f clinical evaluation of, 1036–1038 four-part, 1037f, 1048f, 1053–1054 greater tuberosity, 1037f, 1050, 1051f–1052f head-splitting, 1037f, 1053–1056 incidence of, 1035 lesser tuberosity, 1037f, 1050, 1053f, 1175–1176 pediatric, 1090, 1090b mechanism of, 1036 neurovascular examination in, 1036–1037 nonoperative treatment of, 1040 operative treatment of, 1040–1049, 1043b avascular necrosis after, 1049 complications of, 1049–1050 in four-part fracture, 1048f, 1053–1054 humeral head replacement for, 1046, 1048f intramedullary nailing for, 1043, 1045–1046, 1047f in lesser tuberosity fracture, 1050, 1053f locking plates for, 1040–1041, 1043b, 1044f loss of motion after, 1050 malunion after, 1049–1050 myositis ossificans after, 1050 neurovascular injury after, 1049 nonunion after, 1049 open reduction and internal fixation for, 1046–1047 percutaneous reduction and pinning for, 1041–1042, 1043b, 1045f plate fixation for, 1042–1043 rehabilitation after, 1056–1057, 1057f–1058f retrograde Ender nails for, 1043, 1046f return to play after, 1059 suture tension band fixation for, 1043, 1043b in three-part fracture, 1052–1053, 1054f, 1055f–1056f in two-part greater tuberosity fracture, 1050, 1051f–1052f in two-part surgical neck fracture, 1050–1052, 1054f wire fixation for, 1043, 1043b pediatric, 594, 594f, 1066–1093 age and, 1068 baseball and, 1068, 1070f epidemiology of, 1067–1069 epiphyseal, 1172–1175. See also Little Leaguer’s shoulder football and, 1068–1069 gender and, 1067 in high-performance athlete, 1066, 1081, 1081f–1082f horseback riding and, 1067–1068, 1068f, 1069 incidence of, 1067–1069, 1072 lesser tubercle, 1090, 1090b macrotrauma and, 1068, 1069f metaphyseal, 1085–1089, 1089t age and, 1086 anatomic characteristics of, 1085–1086, 1085f, 1086f through bone cyst, 1085, 1086f classification of, 1087

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxiii

Humeral fracture (proximal) (Continued) displaced, 1087, 1088f extra-articular, 1086 greenstick-type, 1085, 1085f, 1087 incidence of, 1086–1087 intramedullary pins for, 1088, 1088f– 1089f muscle forces in, 1085–1086, 1086f nondisplaced, 1087 olecranon traction for, 1087–1088 open reduction for, 1088 percutaneous stabilization for, 1088, 1088f, 1089f radiography in, 1087 remodeling in, 1085, 1087, 1087f, 1088f, 1088 signs and symptoms of, 1087 treatment of, 1087–1089, 1087f–1089f in non–high-performance athlete, 1067 in nonorganized sports, 1067–1068 in organized sports, 1067 patterns of, 1072, 1072b physeal, 779, 780f, 1072–1085, 1089t anatomic characteristics of, 1073 avascular necrosis after, 1085 callus formation in, 1073, 1078f cast for, 1076–1077 classification of, 1074 closed treatment of, 1076–1077, 1077f, 1080, 1081f–1082f complications of, 1084–1085 computed tomography in, 1074 dislocation and, 1085 displacement with, 1073, 1073f, 1074, 1075b, 1075f–1076f fragment displacement with, 1073, 1073f growth arrest after, 1084 in high-performance athlete, 1081, 1081f–1082f incidence of, 1073 intracapsular, 1073 intramedullary nail fixation for, 1078, 1080f ipsilateral injury with, 1074 malunion of, 1082f, 1085 mechanism of, 1074 nerve injury with, 1074 nonskeletal complications of, 1084 open reduction for, 1078–1079, 1081f, 1084 operative treatment of, 1077–1084, 1079f–1082f, 1083f percutaneous fixation for, 1077–1078, 1079f, 1080f periosteum in, 1072 postfracture care in, 1084 radiography in, 1074–1075, 1076f reduction for, 1076–1077, 1078–1079, 1078f, 1081f, 1083–1084, 1083f remodeling after, 1076, 1078f return to play after, 1084 Salter-Harris I, 1073, 1085 Salter-Harris II, 1073, 1075f, 1085 Salter-Harris III, 1073, 1085 Salter-Harris IV, 1073, 1085 signs and symptoms of, 1073–1074 stress, 1070 structural aspects of, 1072–1073, 1073f swelling with, 1073 traction reduction for, 1077 treatment of, 1075–1084, 1083f management after, 1084–1085 nonoperative, 1076–1077, 1077f, 1080, 1081f–1082f operative, 1077–1084, 1079f–1083f

xxiv

Index

Humeral fracture (proximal) (Continued) sequelae of, 1067 sport-specific, 1068–1069, 1069f–1070f stress, 1090–1093. See also Little Leaguer’s shoulder causative factors in, 1091, 1092f imaging of, 1092, 1093f signs and symptoms of, 1091–1092, 1093b treatment of, 1092 treatment of, 1066–1067 physical examination in, 1036–1037, 1038b, 1038f radiography in, 1037–1038, 1038b, 1043f–�������������������� 1046f, 1048f anteroposterior view for, 1037, 1039f–1040f axillary view for, 1038, 1042f lateral view for, 1037–1038, 1041f Velpeau axillary view for, 1038, 1042f return to play after, 1059 soft tissue injury with, 1035 surgical neck, 1037f, 1040, 1044f–1045f, 1050–1052, 1054f three-part, 1037f, 1043, 1046f, 1047f, 1052–1053, 1054f, 1055f–1056f treatment of, 1038t, 1039–1056, 1043b. See also Humeral fracture (proximal), ­operative treatment of two-part, 1037f, 1040, 1044f–1045f, 1050–1052, 1051f–1052f, 1054f Humeral fracture (shaft), 1176–1183 anatomy of, 1176–1177 biomechanics of, 1176–1177, 1177f classification of, 1177, 1177f complications of, 1182 evaluation of, 1177–1178 imaging of, 1178, 1179f, 1180f, 1180t, 1181f return to play after, 1182 in snowboarders, 1183 stress, 634–635, 1176–1177, 1178–1179, 1180t, 1182 throwing-related, 1226 treatment of, 1162b, 1178–1182, 1179f–1181f, 1182b anterolateral approach for, 1159 medial approach for, 1161 posterior approach for, 1159, 1161 triceps-splitting approach for, 1161 Humeral fracture-dislocation, proximal, 1037f Humeral head, 1034t angulation of, 771–772, 771f articular surface of, 771–772, 771f–772f avascular necrosis of, 1105t, 1106–1107, 1107f arthroscopic treatment of, 1110–1111 magnetic resonance imaging of, 984, 985f physeal injury and, 1085 post-traumatic, 1049 blood supply to, 1033–1034 magnetic resonance imaging of, 983 osteochondral allograft for, 1111, 1111f pediatric, 782 replacement of, 1046, 1048f resurfacing of, 1111–1112, 1112f Humeral tuberosity greater, 1033, 1033f, 1034t lesser, 1033, 1033f, 1034t avulsion fracture of, 1090, 1175–1176 fracture of, 1037f, 1050, 1053f Humeroscapular articulation, 990 Humerus anatomy of, 1157–1158, 1157b, 1158f, 1160f, 1161f developmental, 780, 782, 782f aneurysmal cyst of, 983–984

Humerus (Continued) distal articular surface of, 1189 fracture of. See Humeral fracture (distal) ossification of, 1227–1228, 1228f–1229f supracondylar process of, 1229 fracture of, 596–597, 597f, 1186, 1186f, 1280t, 1281–1283, 1282f proximal anatomy of, 1033, 1033f, 1034t biomechanics of, 1034–1035 blood supply to, 1033–1034, 1034b, 1034f, 1071–1072, 1072f capsular attachments of, 1071, 1071f epiphysis of, 1069–1070, 1070f–1071f injury to, 1173, 1173b. See also Little Leaguer’s shoulder fracture of. See Humeral fracture (proximal) metaphysis of, 971f, 1070–1071, 1071f metastases of, 983, 983f muscle attachments of, 1070–1071, 1071f muscular anatomy of, 1034, 1035f, 1035t nerve supply of, 1034, 1036f pediatric, 782, 782f physis of, 779, 779f, 1070, 1070f closure of, 782, 782f, 1070 shaft of, fracture of. See Humeral fracture (shaft) Humphrey, ligament of, 1597, 1685, 1686f Hyaluronic acid, 43f, 44 in knee osteoarthritis, 1791–1792 Hydralazine, in hypertension, 160t Hydration, 401–402 with creatine use, 418 glycerol in, 402 guidelines for, 401–402, 402t optimization of, 402 Hydrochlorothiazide, 160t Hydrotherapy, after shoulder arthroplasty, 1116t, 1117–1118 β-Hydroxy-β-methylbutyrate, 422 γ-Hydroxybutyrate, 409, 422 Hydroxycitrate, 409 Hyperabduction maneuver, in vascular injury, 1138–1139, 1139f Hyperalgesia, 356t in complex regional pain syndrome, 354, 356 Hypercalcemia, 73t Hypercoagulability, 371–372 primary, 372–374, 374f Hyperesthesia, 356t in complex regional pain syndrome, 356 Hyperglycemia, 175, 175f, 177b, 178 Hyperhomocysteinemia, 374t Hyperhydration, 402 Hypermobility, 2176–2178 criteria for, 2176, 2177f historical perspective on, 2177–2178 vs. instability, 2176–2187 lower extremity injury and, 2182–2183 Hyperparathyroidism, 73t, 74t, 77t Hyperpathia, 356t in complex regional pain syndrome, 356 Hyperplasia, muscle, 214, 214t Hypertension, 156–161 classification of, 156, 157t 158, 157t–158t diagnosis of, 157–��������������������������� evaluation of, 158, 161 monitoring of, 161 secondary, 158, 158t treatment of, 158–161, 159t–160t α-adrenergic receptor agonists in, 161 α-adrenergic receptor antagonists in, 160t, 161

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Hypertension (Continued) angiotensin-converting enzyme inhibitors in, 159, 160t angiotensin receptor blockers in, 159, 160t β-adrenergic receptor antagonists in, 159, 160t calcium channel blockers in, 160t, 161 combination, 161 diuretics in, 159, 160t white coat, 157 Hyperthermia, 528, 528f, 529–530 Hyperthyroidism, 74t, 77t Hypertrophy, muscle, 214, 214t Hyperventilation, epileptiform discharges with, 187 Hypoalgesia, in complex regional pain ­syndrome, 356 Hypocalcemia, 73t Hypoesthesia, in complex regional pain ­syndrome, 356 Hypoglycemia acute management of, 178–179, 179b postexercise, 174–175, 174b, 177–178, 177b Hyponatremia, 402, 495 in heat illness, 529 Hypoparathyroidism, 73t–74t Hypophosphatasia, 73t, 76t–77t Hypopnea, definition of, 448 Hypotension, sternoclavicular joint injury and, 822, 823f Hypothenar hammer syndrome, 1357–1359 clinical manifestations of, 1357 physical examination in, 1357 radiography in, 1357, 1357f return to play after, 1359 treatment of, 1358–1359, 1358f Hypothermia, 500–502, 528–529 cardiovascular effects of, 501 mild, 500–501 moderate, 500–501 neurologic effects of, 500–501 prevention of, 501–502 renal effects of, 501 severe, 500–501 treatment of, 501 Hypothesis, 99, 110–112 alternate, 110 null, 110 Hypoxia high-altitude, 502–503 seizures with, 187

I I band, 4f, 5, 6f, 208f Ice test, in complex regional pain syndrome, 356 Ice therapy. See Cryotherapy Iliac apophysitis, 1475 Iliac crest, 1452 contusions of, 1459 graft harvest–related pain at, 1833 Iliac spine, avulsion fracture of, 556f, 1475 Iliofemoral ligament, 1452, 1453f Ilioinguinal nerve, entrapment of, 1469 Iliopatellar band, 1551, 1552f, 1719 Iliopectineal bursitis, 1458 Iliopsoas, 1454t, 1455f strain of, 1461 Iliopsoas bursitis, 1457–1458, 1457f Iliotibial band anatomy of, 1719 for ligament autograft, 37 stretch for, 291–292, 291f tightness of, 1558

Index Iliotibial band friction syndrome, 627–628, 629f magnetic resonance imaging in, 560, 562f Noble compression test in, 628, 628f Ober’s test in, 628, 629f treatment of, 628, 630f Iliotibial band tenodesis, in pediatric ACL injury, 1680 Ilium, 1452 Imaging. See specific imaging modalities Immobilization, 77t cartilage effects of, 228 in cervical spine injury, 666–667, 666f glycosaminoglycans effects of, 228 tendon effects of, 25–26, 26f, 28 in thoracolumbar spine injury, 720 Immune system cartilage allograft and, 54 exercise effect on, 147–148, 148f, 148t Immunoglobulin A, secretory, exercise effects on, 147, 148t Immunoglobulin G, in Epstein-Barr virus infection, 150 Immunotherapy, in epilepsy, 190 Impetigo, 193–194, 193f–194f, 194t Impingement. See Ankle impingement; Femoroacetabular impingement; Shoulder impingement Impingement test, 998 Implant chondrocyte, 55, 1774–1780, 1777t, 1780f, 1785t collagen, 39 infection with, 394–395, 394f, 394t Implantable defibrillator, in sudden death prevention, 169–170 Incidence, 102, 102t, 103, 2219 Indapamide, in hypertension, 160t Indirect calorimetry, 212–213, 212f Indomethacin in heterotopic ossification, 1294 in tendon healing, 29 Infection, 147–156. See also specific infections allograft transmission of, 141–142, 1657 anterior cruciate ligament graft, 391–393, 393b, 393t, 1657 blood-borne, 153–156 bone. See Osteomyelitis Borrelia burgdorferi, 155–156 cardiac, 151–152 cutaneous, 193–200 bacterial, 193–196, 193f–194f, 194t clearance for participation and, 513 fungal, 198–200, 198f–200f, 200t return-to-play guidelines for, 195t viral, 196–198, 197f–198f epidemiology of, 149 Epstein-Barr virus, 150–151 gastrointestinal, 152 hardware-related, 394–395, 394f, 394t hepatitis, 154–155 human immunodeficiency virus, 153–154 intestinal, 152 after knee dislocation treatment, 1765 pericardial, 152 respiratory, 149–150 return to play after, 150 risk for J-curve theory of, 148, 148f reduction of, 148 shoulder, 389–391, 390f, 391t, 392f Staphylococcus aureus, 395–397, 396b, 396f methicillin-resistant, 193, 194–195, 194t, 195f, 395–397, 396b, 396f prevention of, 396–397

Infection (Continued) superficial, 387–388, 387f, 389f toenail. See Ingrown toenail treatment of, 386–387, 387t unusual, 397–398, 398t urinary tract, 152–153 Infectious mononucleosis, 149, 151 Inference, 99, 112 Inferior lateral genicular artery, 1723 Inferior peroneal retinaculum, 1987 Infertility, anabolic-androgenic steroids and, 416 Inflammation in complex regional pain syndrome, 354–355 cutaneous, 204–205, 204f–205f in delayed-onset muscle soreness, 216 electrical current therapy in, 230, 231f in ligament healing, 34–35 in tendon healing, 27–28 Infrapatellar tendon injury, 560, 561f Infraspinatus anatomy of, 770–771, 770f, 911, 989, 989f, 1035t pediatric, 785, 785f atrophy of, 770 fatty infiltration of, 989, 989f function of, 991 humeral attachment of, 1070–1071, 1071f scapular attachment of, 859f, 886f strengthening exercises for, 242–243, 242f–243f Ingrown toenail, 2096–2104 anatomy of, 2096–2097, 2096b, 2097f classification of, 2097, 2097b evaluation of, 2097–2098, 2098b history in, 2097 imaging in, 2098 nonoperative treatment of, 2098, 2098f–2099f operative treatment of, 2098–2105, 2098b alcohol matrixectomy in, 2099–2100 care after, 2101–2102 complete nail plate avulsion in, 2099, 2101f complete onychectomy in, 2099, 2103f complications of, 2100–2101 partial nail plate avulsion in, 2099, 2100f partial onychectomy in, 2099, 2102f phenol matrixectomy in, 2099–2100, 2105f plastic nail wall reduction in, 2099, 2101f return to play after, 2102–2103, 2105b Syme amputation in, 2099, 2104f physical examination in, 2098, 2098b Inhalant use/abuse, 429–430 Injury, 431–441. See also specific structures and injuries cognitive response to, 436–437 coping skills in, 435–436 depression and, 453 disruptive impact of, 435 emotional nourishment in, 435 in older athlete, 433 prevalence of, 432–433 psychological response to, 431–432, 433–436 three-foci contextual considerations in, 434–435 treatment team in, 436 Innominate bone, 1451–1452 Insall-Salvati ratio, 1522, 1523f, 1539, 1541f in patellar rupture, 1522–1523, 1523f Insomnia, 447–448, 448b Insulin, 176–178, 176t, 177b, 177t, 178f excessive, exercise and, 174–175, 174b exercise effects on, 217t, 218 insufficient, exercise and, 175, 175f physiology of, 173–175, 174f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxv

Insulin-like growth factor–1, 20, 419–420 Insulin pump, 178, 178b, 178f Intercalary segmental instability dorsal (DISI), 1322, 1323f volar (VISI), 1322, 1324f Interclavicular ligament, 793, 793f International Knee Documentation ­Committee, subjective rating form of, 301 Interosseous talocalcaneal ligament, 1913f heterotopic ossification of, 1933 reconstruction of, 1951, 1951f Interphalangeal joint distal, dislocation of, 1384, 1386, 1418–1420, 1419f proximal anatomy of, 1381–1382, 1382f dislocation of, 1381–1384 in children, 1417–1418, 1417f–1418f classification of, 1382, 1382f clinical presentation of, 1382 complex dorsal, 1383, 1383f dorsal, 1382–1383, 1417–1418, 1417f, 1418f fracture with, 1383, 1383f–1385f palmar, 1384 treatment of, 1382–1384 type I, 1382 type II, 1383, 1383f type III, 1383, 1383f–1385f volar, 1384, 1386f, 1417–1418, 1417f fracture-dislocation of, 1383, 1383f–1385f fracture-subluxation of, 1383–1384, 1385f sprain of, pediatric, 1414 subluxation of fracture with, 1383–1384, 1385f palmar, 1384 volar, 1384 Interquartile range, 112 Intersection syndrome, 624–625, 1356 Intertubercular groove, pediatric, 783 Interventional procedures, imaging for, 582–586 at peripheral joints, 582–584, 584f at spine, 584–585, 584f–585f Intervertebral disk biomechanics of, 718–719 cervical spine herniation of, 674, 698–699, 699f in children/adolescents, 704 injury to, 674, 698–699, 699f rupture of, 674, 678 lumbar, 717, 718–719. See also Lumbar spine, degenerative disk disease of herniation of, 741–744, 744f, 751f, 752 in children/adolescents, 764–766, 765f, 767–768, 767b thoracic anatomy of, 717, 717f biomechanics of, 718–719 herniation of, 738, 739f, 755–756 Intra-articular disk ligament, 792–793, 792f–793f Intracerebral hematoma, 660–661, 661f Intraclass correlation coefficient, 100, 118 Intracranial hemorrhage, 659, 660–661, 661f Intramedullary nailing in proximal humeral fracture, 1043, 1045–1046, 1047f in proximal metaphyseal humeral fracture, 1088, 1088f–1089f in proximal physeal humeral fracture, 1078, 1080f in tibial stress fracture, 1854–1855 Intubation, endotracheal, 519 Iontophoresis, 233–234, 235f in plantar fasciitis, 2049

xxvi

Index

Irbesartan, in hypertension, 160t Iron deficiency of, in female athlete, 478–479, 479b requirements for, 406b Irrigation, during arthroscopy, 122, 123f Ischial bursitis, 1457 Ischial tuberosity, avulsion fracture of, 1475 Ischiofemoral ligament, 1452, 1453f Ischium, avulsion fracture of, 1489, 1489f Iselin’s disease, 2143f, 2167–2169, 2168f–2169f Isotretinoin, in acne, 205 Isradipine, in hypertension, 160t Itraconazole, in dermatophyte infection, 200t

J J sign, in patella dislocation, 1556 Jaw thrust maneuver, in cervical spine injury, 667, 668f Jefferson fracture, 677, 677f–678f, 705, 706f Jerk test, in glenohumeral joint instability, 941 Jersey finger, 1390–1392 classification of, 1390 neglected, 1391–1392 pediatric, 1424–1428, 1424f–1427f physical examination in, 1390 treatment of, 1391, 1391f Jet lag, 457–460, 458f adjustment avoidance and, 460 bright light exposure and, 458, 458f melatonin and, 459–460 performance and, 460 room light and, 458–459, 459t–460t Jobe’s relocation test in glenohumeral joint instability, 914, 915f, 939, 939f in glenohumeral joint osteoarthritis, 1108 in rotator cuff disorders, 996, 999f Jogger’s foot, 2053 Joint(s). See also specific joints motion at, 90, 93f, 2178–2182, 2178f preparticipation examination of, 512–513 Joint capsule, glenohumeral, 772–773, 772f Joint reaction force, 86, 89f. See also ­Patellofemoral joint reaction force Jones’ fracture, 1969–1970, 1971f, 2025–2027, 2026f, 2027f Jones (tendon-sling) procedure, 1994–1995, 1994f Juggling, time of day and, 457 Jugular vein, 794, 797f Juices, 401 Jumper’s knee, 30, 31, 626–627, 626f evaluation of, 1518–1519 imaging in, 560, 561f, 1519, 1519f–1520f pathophysiology of, 1515–1518, 1517f–1518t stages of, 1519 treatment of, 1519–1521, 1520f Jumping exercise, double-leg, in knee ­rehabilitation, 296, 298f Juvenile idiopathic arthritis, 602–603, 602f

K Karolinska Sleepiness Scale, 450 Kehr’s sign, 526 Keller procedure, in hallux valgus, 2072, 2075 Kelly’s bone block procedure, 1993, 1993f Keratan sulfate, 43f, 44 Keratolysis, pitted, 196, 196f Ketamine, in complex regional pain syndrome, 364

Ketoconazole, in dermatophyte infection, 200t Kicking, dynamic analysis of, 90, 91f Kidner’s procedure, 2164, 2165, 2165f Kidneys absence of, 514t congenital anomalies of, 712, 712f hypothermia effects on, 501 on-field injury to, 527 Kienböck’s disease adult, 1350, 1353f pediatric, 1376–1377, 1377b, 1378f Kinematics, 90 acromioclavicular joint, 775–776, 776f, 786–787 angular, 90, 91f glenohumeral joint, 777–778, 787–788, 787f–788f linear, 90, 91f scapulothoracic joint, 776–777, 776f, 787–788, 788f sternoclavicular joint, 775, 775f, 786 Kinetic chain, 221–222 Kinetic chain exercise after ACL reconstruction, 232, 232f, 1668–1670, 1669f, 1669t upper extremity, 232, 232f Kinetic muscle testing, in thoracolumbar spine, 728 Kinetics, 90–91 Kleinman’s shear test, 1330–1331 Klippel-Feil syndrome, 693, 693f, 711–712, 712f Knee. See also Patella anatomy of, 1548–1553, 1579 bony, 1549–1550, 1549f–1550f, 1747–1748 dynamic restraints in, 1548, 1548f lateral, 1718–1723, 1719f–1722f ligamentous, 1551–1553, 1551f–1553t, 1748 medial, 1624–1627, 1624f–1626f, 1638, 1638b posterolateral, 1718–1723, 1722f neurovascular, 1748–1749 posteromedial corner, 1625, 1627f soft tissue, 1550–1553, 1551f static restraints in, 1548, 1548f arthroplasty for. See Knee arthroplasty arthroscopy of. See Arthroscopy, knee biomechanics of, 1579–1596, 1747 bumper model of, 1580 compound hinge model of, 1580, 1581f experimental studies of, 1581 femoral epicondylar axis in, 1581 four-bar cruciate linkage model of, 1580, 1580f ligament, 1581–1589, 1583f, 1586f, 1587t. See also at specific ligaments mathematical model of, 1580 meniscal, 1589–1591. See also at Meniscus (menisci) models of, 1580–1581, 1580f–1581f patellofemoral, 1591–1596, 1592f–1595f rotation, 1579–1580, 1579f shear forces during, 222 three-dimensional tibiofemoral joint model of, 1580–1581 translation, 1579–1580, 1579f cartilage lesions of, 1771–1786 classification of, 1772, 1772t, 2212, 2212t clinical presentation of, 1773 débridement for, 1774 etiology of, 1771–1772 history in, 1773 imaging in, 1773, 1773f–1774f marrow stimulation techniques for, 1774 nonoperative treatment in, 1773–1774

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Knee (Continued) operative treatment in, 1774–1786 age and, 1786 algorithm for, 1778f autologous chondrocyte ­implantation for, 1774–1775, 1777, 1777t, 1778–1780, 1780f, 1785t complications of, 1783 contraindication to, 1786 gene therapy for, 1776 growth factors for, 1776 matrix-associated techniques for, 1775–1776 microfracture for, 1776, 1777–1778, 1779f, 1784t osteochondral allograft transplantation for, 1775, 1777, 1777t, 1781–1783, 1782f, 1784t osteochondral autograft transplantation for, 1775, 1777, 1780–1781, 1781f, 1784t outcomes of, 1783 rehabilitation after, 1783, 1784t–1785t return to play after, 1786 stem cells for, 1776 synthetic plugs for, 1776 tissue-engineered cartilage for, 1776 physical examination in, 1773 chondral lesions of. See Knee, cartilage ­lesions of chondromalacia of, 579, 580f–581f computed tomography arthrography of, 536, 537f cysts of, 537–538, 538f deep complex of. See Arcuate ligament; ­Popliteus tendon dislocation of. See Knee dislocation effusion of, in patellar dislocation, 1556 Fairbank’s changes of, 2211–2212, 2211t jumper’s. See Patellar tendinosis ligaments of. See Anterior cruciate ligament (ACL); Fibular (lateral) collateral ligament (FCL); Medial collateral ligament (MCL); Popliteofibular ligament; Posterior cruciate ligament (PCL) magnetic resonance arthrography of, 536 menisci of. See Meniscus (menisci) motions of, 1579–1580, 1579f. See also Knee, biomechanics of multiligament injury to. See Knee dislocation osteochondritis dissecans of. See Osteochondritis dissecans, patellar overuse injury of, 626–628, 626f, 629f pain in, after tibial stress fracture treatment, 1855 physeal fracture of, 595, 595f, 1640–1644, 1640b distal femur, 1641–1642, 1642b, 1642f, 1643f proximal tibia, 1642–1644, 1643f, 1643t popliteal (Baker’s) cyst of, 537–538, 538f posterolateral corner of. See Posterolateral corner proprioceptive function in, 294. See also Knee rehabilitation, proprioceptive exercises in prosthetic. See Knee arthroplasty radiography of, 2211–2212, 2211t rehabilitation of. See Knee rehabilitation replacement of. See Arthroplasty, knee rotation of, 1579–1580, 1579f stretching exercises for, 292–293, 292f translation of, 1579–1580, 1579f ultrasonography of, 537–538, 538f varus-aligned, 1801–1803, 1802t, 1803f. See also Varus malalignment

Index Knee arthroplasty, 1787–1801 athletic activity after, 1787–1789, 1789b clinical evaluation for, 1789–1790 imaging for, 1789–1790, 1790f mobile-bearing prostheses in, 1797–1798 vs. nonoperative treatment, 1790–1792 patient history and, 1789 physical examination in, 1789 quadriceps rupture with, 1521 total, 1794–1797, 1795t–1797f bearing surfaces in, 1800 gender-specific, 1799 high-flexion, 1798–1799, 1798f–1799f indications for, 1794 minimally invasive, 1799–1800, 1800f prior proximal tibial osteotomy and, 1792 results of, 1795, 1795t unicompartmental, 1792–1794, 1793f Knee dislocation, 1747–1764 bony injury with, 1752–1753, 1752f classification of, 1749, 1749t clinical presentation of, 1749 imaging of, 1753, 1753f inspection in, 1750, 1750f Lachman test in, 1750, 1750f nerve injury with, 1751–1752, 1752f neurovascular examination in, 1750 nonoperative treatment of, 1753–1754, 1754f operative treatment of, 1754–1765, 1762b ACL reconstruction in, 1758, 1759f, 1760b Arthrotek tensioning boot in, 1760, 1761f, 1762 without Arthrotek tensioning boot, 1761–1762 compartment syndrome after, 1765 complications of, 1764–1765 fixation technique in, 1760, 1761b graft selection for, 1755, 1757 infection after, 1765 instability after, 1765 nerve injury with, 1764–1765 vs. nonoperative, 1754–1755 platelet-rich fibrin matrix clot in, 1763, 1763f position for, 1757, 1757f posterior cruciate ligament reconstruction in, 1755–1756, 1758, 1759f, 1762–1763, 1763f posterolateral reconstruction in, 1758–1759, 1760f preparation for, 1757, 1758f rehabilitation after, 1756, 1763–1764, 1764b results of, 1761–1762 return to play after, 1765 stiffness after, 1765 timing of, 1755, 1756–1757, 1757f vascular injury with, 1764–1765 physical examination in, 1749–1750, 1750f vascular injury with, 1750, 1751, 1753, 1753f, 1764–1765 Knee rehabilitation, 221–222, 222t adhesions and, 225 arthrofibrosis and, 225 articular cartilage protection in, 225–228, 226f–227f biofeedback in, 224–225, 233, 233f ERMI Flexionator in, 225, 225f gait pattern in, 225 kinetic chain in, 221–222, 222t muscle inhibition and, 224–225 neuromuscular stimulation in, 231, 231f, 232–233, 232f–234f patellar mechanics in, 226–228, 226f–227f patellofemoral joint reaction forces and, 226, 226f, 1595–1596, 1595f proprioceptive exercises in, 294–296

Knee rehabilitation (Continued) cone ambulation, 295, 296f cone reaching, 296, 297f double-leg jumping, 296, 298f lunging, 296, 297f plyoball toss, 294, 294f side-to-side weight shifting, 294–295, 295f single-leg hopping, 296, 298f sports cord lunges, 296, 297f return-to-play phase of, 300–321 criteria for, 301–303, 302f stage I, 303–307, 304f–306f stage II, 307–310, 308f–312f stage III, 312–316, 313f–314f, 316f stage IV, 316–320, 317f–320f return-to-play plyometric training in, 300–321 contraindications to, 303 criteria for, 301–303, 302f glossary for, 324–330 stage I (dynamic stabilization and strengthening), 303–307, 304f–306f, 322 stage II (functional strength), 307–312, 308f–312f, 322–323 stage III (power development), 312–316, 313f–314f, 316f, 323 stage IV (sport performance), 316–320, 317f–320f, 324 therapeutic exercise for, 255–272 ACL loading with, 221–222, 222t ACL strain measurements in, 1586–1588, 1586f, 1587t acute phase of, 255 advanced phase of, 255 band walking, 258, 259f gluteal muscle raise in, 267, 268f gluteal musculature, 257–258, 258f–259f hamstring curls in, 266 hamstring raise in, 267, 268f hip abduction, 258, 258f hip extension in, 258, 258f, 267 hyperextension in, 267 kinetic chain, 1668–1670, 1669f, 1669t leg extension, 227–228, 227f leg press, 227, 227f, 259–260, 260f ligament unloading, 222–224, 223t, 224f lunges in, 264–266, 265f, 266f manual perturbation training in, 270–271, 271f multiplanar squats in, 270, 270f neuromuscular activation, 256–258 neuromuscular control training in, 268–270, 269t, 270f PCL loading with, 221–222, 222t program design for, 255–256 proprioceptive training in, 268–270, 269t, 270f quadriceps, 256–257, 257f quadriceps dominant squatting, 258–263, 260f–263f return-to-activity phase of, 256 Romanian deadlift in, 267–268, 269f single-leg balance training in, 271 single-leg presses in, 263, 264f single-leg Romanian deadlift in, 268, 269f single-leg strength training in, 263–266, 264f–265f single-leg wall squats in, 264, 265f slide board leg curls in, 266–267, 267f sport cord activities in, 271, 272f, 296, 297f squat and reach in, 270, 271f squats in, 258–259, 260–263, 261f–263f stability ball bridges in, 266, 267f stability ball curls in, 266–267, 267f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxvii

Knee rehabilitation (Continued) step-ups in, 264, 264f straight leg, 267–268 subacute phase of, 255 Total Gym in, 259–260, 260f unstable surface training in, 271, 272f wall squats in, 260, 260f, 264, 265f transcutaneous electrical nerve stimulation in, 229–230, 230f, 233, 233f Knot/knot tying, 135–136, 135f Duncan loop, 135, 135f Nicky’s, 135, 135f Roeder’s, 135, 135f for shoulder arthroscopy, 1147, 1148f for SLAP lesion repair, 1028, 1030–1031, 1030f, 1031f KOH testing, 198, 200f Köhler’s disease, 599, 1972–1973, 1973f, 2143f, 2162, 2163f Krebs’ cycle, 211–212 KT-100 arthrometry in anterior cruciate ligament injury, 1650, 1652 in posterior cruciate ligament injury, 1692–1693 Kyphoplasty, lumbar, 585, 585f Kyphosis, 736–737 classification of, 737 Scheuermann’s, 737 thoracic, 718

L Labetalol, in hypertension, 160t Laceration cartilage, 50 muscle, 11, 11f skin, 201–202 Lachman test in anterior cruciate ligament injury, 1585–1586, 1649–1650, 2213 in knee dislocation, 1750, 1750f in meniscal injury, 1602 Lactate, high-altitude effects on, 503 Lactate threshold, 211, 218 Lactic acid, creatine buffering of, 418 Lamotrigine, in epilepsy, 189, 191t Landau-Kleffner syndrome, 190 Larva migrans, 201, 202f Larynx, on-field injury to, 525 Lasègue’s sign in lumbar disk herniation, 743 in thoracolumbar spine injury, 722 Laser therapy contraindications to, 236 in rehabilitation, 235–236, 236f Lateral collateral ligament. See Fibular (lateral) collateral ligament (FCL) Lateral femoral cutaneous nerve, 1454, 1455f entrapment of, 1469 Lateral flexion progression exercise, in core training, 283–284, 284f Lateral gastrocnemius tendon, 1719f, 1723 Lateral genicular artery, 1723 Lateral patellofemoral ligament, 1552 Lateral pull test, in patellar dislocation, 1557, 1557f Latissimus dorsi muscle, 771 actions of, 907 anatomy of, 907, 1034, 1035t, 1065 pediatric, 785 contusion of, 1065 functional training of, in core training, 286 press-up exercise for, 241, 242f

xxviii

Index

Latissimus dorsi muscle (Continued) rupture of, 907–908, 907b–908b, 1065 scapular attachment of, 859f, 886f strain of, 1065 strengthening exercise for, 1003, 1006f Latissimus dorsi tendon injury, 1065 Learning naps and, 449, 449f sleep deprivation and, 449–450 Left ventricular hypertrophy, athletic ­participation guidelines for, 161 Leg extension exercise, patellar effects of, 227–228, 227f Leg-length inequality after epiphyseal fracture, 1643–1644 stress fracture and, 632 Leg press exercise in knee rehabilitation, 259–260, 260f patellar effects of, 227, 227f single-leg, in knee rehabilitation, 263, 264f Legal issues, 531 Legg-Calvé-Perthes disease, 599, 600f, 1476–1477 Lennox-Gastaut syndrome, 189 Leptospirosis, 152 Lesser trochanter, avulsion fracture of, 1475 Leukemia, magnetic resonance imaging in, 610 Leukocytes, radiolabeled, in osteomyelitis, 547 Leukoplakia, smokeless tobacco and, 428 Leukotriene modifiers, in exercise-induced bronchospasm, 182 Levator scapulae pediatric, 786, 786f scapular attachment of, 859f, 886f Levetiracetam, in epilepsy, 191t Liability, 531 Lice, 200–201, 201f Lidocaine, in complex regional pain syndrome, 364 Lidocaine injection test, in thoracic outlet syndrome, 1133 Lift-off test, in subscapularis evaluation, 1065 Ligament(s), 32–39. See also specific ligaments cells of, 33–34 collagen of, 32–33, 34 elastin of, 34 fibroblasts of, 35 grading of, 2212, 2212t, 2214, 2214t healing of, 34–39 allografts in, 38, 142–143 autografts in, 36–38 collagen implants in, 39 Dacron augmentation in, 36 grafts in, 35–39 inflammation in, 34–35 remodeling in, 35 tissue engineering in, 39 variables in, 35 xenografts in, 38 insertions of, 33 magnetic resonance imaging of, 569, 571–576 at ankle, 571, 573, 573f at elbow, 576, 577f at knee, 569, 571, 571f–573f at shoulder, 573, 575, 575f at wrist, 575, 575f matrix of, 34 noncollagenous proteins of, 34 proteoglycans of, 34 remodeling of, 35 structure of, 32 types of, 32 water of, 34 Ligament of Testut, 1320, 1432, 1432f Ligament unloading exercise, 222–224, 223t

Ligamentum teres, 1452, 1453 rupture of, 1472–1473, 1473f Light-touch testing, in thoracolumbar spine injury, 722, 722f Linear regression, 113, 113f Link protein, meniscal, 58 Linoleic acid, conjugated, 409 Lisfranc joint, 1956, 1956f. See also Lisfranc sprain computed tomography of, 542–543 fracture-dislocation of, 1955, 2180, 2181f Lisfranc sprain, 1955–1960 anatomy of, 1956, 1956f computed tomography in, 1957, 1957f evaluation of, 1956, 1957f history in, 1956 magnetic resonance imaging in, 1957 physical examination in, 1956, 1957f radiography in, 1956–1957, 1957f rehabilitation after, 1958–1960 return to play after, 1960 treatment of, 1957–1958, 1959f Lisinopril, 160t Little Leaguer’s elbow, 468, 469f, 597, 1234–1235, 1238–1239 clinical manifestations of, 1236, 1236f complications of, 1186 diagnosis of, 1183, 1183b, 1184f–1185f, 1235 history in, 1234–1235 pain in, 1235 past medical history in, 1235 prevention of, 623–624 treatment of, 623, 1183–1185, 1183b, 1184f, 1185f, 1237 Little Leaguer’s shoulder, 468, 468f, 634–635, 635f, 1090–1093, 1172–1175 anatomy of, 1172–1173 classification of, 1173, 1173t complications of, 1175 evaluation of, 1091–1092, 1093b, 1093f, 1173, 1173b, 1174f prevention of, 1175 return to play in, 1175 treatment of, 1092, 1173–1175 Liver anabolic-androgenic steroid effects on, 416–417 on-field injury to, 527 Liver function test, in infectious ­mononucleosis, 150 Load and shift test in glenohumeral joint instability, 914–915 in rotator cuff disorders, 996, 999f Load-elongation curve, 93, 94f, 95–96, 95f Log-linear analysis, 114 Logistic regression, 113–114, 114f Logroll, 519, 519f in cervical spine injury, 666–667, 666f Long arc quad exercise, in knee rehabilitation, 257 Long QT syndrome, 166–167 β-adrenergic blockers in, 170 Long thoracic nerve anatomy of, 1125, 1125f injury to, 1130–1131, 1130f�������������� –������������� 1131f Loose bodies of elbow, 470f, 621–622 of hip, 1472, 1472f of wrist, 1447 Lordosis, 718 Losartan, 160t Low back pain. See also Lumbar spine, ­degenerative disk disease of. ­Thoracolumbar spine injury in athlete, 741 in child/adolescent, 756–757, 757f, 765, 766

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Low back pain (Continued) cold therapy in, 731 diagnostic injection in, 731–732, 731t heat therapy in, 731 lumbar spine stabilization in, 728–730, 730f muscle spasm and, 733 muscle strain and, 733 pharmacologic treatment of, 731 trigger point injection in, 732 tumor and, 757, 758f Low-molecular-weight heparin in venous thromboembolism, 384–385, 384t in venous thromboembolism prevention, 378–384, 381t, 382f Lumbar kyphoplasty, 585, 585f Lumbar spine. See also Thoracolumbar spine anatomy of, 717, 724–726, 725f, 754–755 computed tomography of, 541, 541f, 727 degenerative disk disease of, 740–744, 740f annular tear in, 740–741 athletic factors in, 741 in children/adolescents, 764–766, 765f, 767–768, 767b conservative treatment of, 743 evaluation of, 740, 740f, 743 facet joint in, 741 herniation in, 741–744, 744f, 751f, 752 pathophysiology of, 740, 740f return to play and, 744, 744f surgery for, 741, 742f, 744 diskography of, 728 fracture of, 733–736, 733f, 735f burst, 541f, 735, 735f compression, 546, 547f transverse process, 734 injury to. See Thoracolumbar spine injury ligaments of, 717 lordosis of, 718 magnetic resonance imaging of, 727, 727f muscles of, 717 radiography of, 724–726, 725f radionuclide imaging of, 726–727, 726f scoliosis of, 737 spondylolisthesis of degenerative, 747–748, 748t, 752 isthmic, 748–750, 748t, 749f–750f, 752 spondylolysis of, 725, 726f sprain of, 733 stenosis of, 744–748, 744b, 745t, 746f central, 744, 745 congenital, 746–747, 747f–748f conservative treatment of, 746 imaging in, 746 lateral, 744 physical examination in, 746 surgery for, 746, 747f strain of, 733 stress fracture of, 635–636, 636f–637f transverse process fracture of, 734 zygapophyseal joints of, 718, 719f Lumbar spine stabilization, 728–730, 730f Lumbosacral plexus, 1453 Lunate anatomy of, 1319–1320, 1319f dislocation of adult, 1322, 1323f pediatric, 1371 fracture of adult, 1350, 1353f pediatric, 1370, 1370f Lunges, 264–266, 265f, 296, 297f lateral, 266, 266f, 328 reverse, 265 static, 265, 265f walking, 266, 330

Index Lungs exercise effect on, 220 preparticipation examination of, 512, 515t Lunotriquetral arthrodesis, 1331, 1331f Lunotriquetral ligament, 1320 injury to, 1330–1331, 1331f Lyme disease, 155–156 Lymphatics, lower extremity, 1453 Lymphoma, 77t spinal, 757, 758f

M M line, 4f, 5, 6f Ma huang, 409 Macrophages in delayed-onset muscle soreness, 216 exercise effects on, 147, 148t Magnesium deficiency of, 12 requirements for, 406b Magnetic resonance angiography in knee dislocation, 1753 in popliteal artery entrapment, 1840, 1842f in vascular shoulder injury, 1139 Magnetic resonance arthrography (MRA), 550–551, 551f in age-related rotator cuff lesions, 974–975, 975f in ALPSA lesion, 974, 974f of ankle, 537, 551 in Bankart lesion, 973–974, 973f, 974t in biceps tendinopathy, 981, 981f of Buford complex, 972, 972f of elbow, 551 in GLAD lesion, 976, 978f of glenohumeral joint, 949 in glenohumeral joint instability, 969–975, 970f–974f, 974t, 975f of glenohumeral ligaments, 573, 575, 575f of glenoid labrum, 969–971, 970f in glenoid labrum cyst, 580–581, 582f of hip, 536–537, 551 of inferior glenohumeral ligament, 971, 971f of knee, 536, 551 of middle glenohumeral ligament, 971, 972f in multidirectional glenohumeral instability, 973 in post-traumatic anterior glenohumeral instability, 973–974, 973f–974f, 974t in posterosuperior glenoid impingement, 978–979, 979f in rotator cuff tear, 565–566, 567f of shoulder, 535, 536, 536f, 551 in SLAP lesion, 976, 977f of sublabral foramen, 972, 972f of sublabral recess, 970f, 972 of superior glenohumeral ligament, 971–972, 973f of wrist, 536 in wrist injury, 1321 Magnetic resonance imaging (MRI), 533, 547–552 in acetabular labrum tear, 581–582, 1470 of Achilles tendon, 2034, 2034f in Achilles tendon injury, 563, 565f, 630, 2003 in acromioclavicular osteoarthritis, 956, 959f of acromion, 955–956, 956f–957f in adductor strain, 1490–1491, 1491f in adhesive capsulitis, 967, 967t, 968f in ankle dislocation, 1945 in ankle impingement, 2157, 2157f in ankle sprain, 571, 573, 574f, 1920, 1921f, 1937, 1942, 1943f

Magnetic resonance imaging (MRI) (Continued) in anterior cruciate ligament injury, 553, 557f, 569, 571f, 572f, 1650–1651, 1653–1654, 1654t in anterior tibial tendon tear, 560, 562f in avulsion injury, 553, 557f of biceps tendon, 980–983, 981f, 982f in biceps tendon rupture, 1165 in biceps tendon tear, 566, 568f in bifurcate sprain, 1954 in calcaneal stress fracture, 556f, 646, 647f in calcific supraspinatus tendinopathy, 566, 567f in calcific tendinitis, 963, 963f in capitellar osteochondritis dissecans, 1241–1242, 1242f in cartilage injury, 576–582, 578f at hip, 581–582 at knee, 579–580, 579f–581f at shoulder, 580–581, 582f–583f in cervical intervertebral disk injury, 698–699, 699f in chondral defect, 1773, 1773f in chondromalacia, 579, 580f–581f in compartment syndrome, 651, 1859, 1859f in complex regional pain syndrome, 359 contrast agents for, 550–551, 551f of coracoacromial ligament, 956, 959f of coracohumeral distance, 956 in de Quervain’s tenosynovitis, 569, 570f in deltoid denervation atrophy, 965, 965f in denervation edema, 965, 965f, 980 in distal biceps tendon rupture, 1168, 1169f in distal biceps tendon tear, 566, 570f in distal femoral stress fracture, 555f of elbow, 1311 in elbow heterotopic ossification, 1293 fat suppression for, 550 in fatty rotator cuff infiltration, 989, 989f in femoral contusion, 1542, 1543f in femoral neck stress fracture, 554f, 639, 640f, 642f, 1465, 1465f in femoral shaft stress fracture, 1478, 1478t, 1479 in femoroacetabular impingement, 1471, 1472f in flexor hallucis longus tendon rupture, 563, 564f in frozen shoulder, 967, 967t, 968f of glenohumeral joint, 953, 953t, 954, 960t in glenohumeral joint instability, 916, 920f, 925f, 937, 938f in glenohumeral joint osteomyelitis, 984 in glenoid rim fracture, 870f gradient echo, 549 in hamate fracture, 1345, 1346f in hamstring rupture, 557, 558, 559f in hamstring strain, 1486f, 1487, 1487f in high (syndesmosis) ankle sprain, 1942, 1943f in humeral head avascular necrosis, 984, 985f in humeral metastases, 983, 983f in humeral shaft fracture, 1178, 1180f, 1182 in iliotibial band friction syndrome, 560, 562f, 628 image archiving for, 552 image quality in, 551–552 in intra-articular cartilage fragment in ­shoulder, 581, 583f inversion recovery, 549 in Kienböck’s disease, 1353f in knee arthritis, 1790 in knee dislocation, 1753 in lateral ankle sprain, 1920, 1921f in lateral epicondylitis, 566, 568f, 1199

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxix

Magnetic resonance imaging (MRI) (Continued) in lateral meniscal tear, 579, 579f of lateral meniscus, 1598f in Legg-Calvé-Perthes disease, 599, 600f in ligament injury, 569, 571–576 at ankle, 571, 573, 574f at elbow, 576, 577f at knee, 569, 571, 571f–573f at shoulder, 573, 575, 575f at wrist, 575, 576f in Lisfranc sprain, 1957 in lumbar disk herniation, 743 in lumbar isthmic spondylolisthesis, 749, 749f in lumbar kyphoplasty, 585, 585f of lumbar spine, 727, 727f in lumbar spine stenosis, 746 in malignant fibrous histiocytoma, 557, 560f in medial ankle sprain, 1937 in medial collateral ligament injury, 569, 571, 573f, 1629–1631, 1631f in medial malleolar stress fracture, 644–645, 645f in medial meniscal tear, 579, 580f in medial patellofemoral ligament disruption, 1542, 1542t in meniscal deficiency, 1619, 1619f in meniscal injury, 1601, 1602, 1603f in muscle contusion, 558 in muscle denervation, 558 in muscle injury, 557–558, 559f–560f in muscle strain, 16, 557, 559f in occult fracture, 552, 554f, 557 in olecranon bursitis, 1247 of os acromiale, 955–956, 958f in Osgood-Schlatter disease, 1527 in osteochondral fracture, 578–579, 578f in overhead throwing injury, 1217 in Panner’s disease, 625f in paralabral cyst, 964–965, 965f in Parsonage-Turner syndrome, 1145 in partial-thickness rotator cuff tear, 960–961, 961f in PASTA lesion, 961, 962f in patellar dislocation, 1541, 1542, 1542t, 1565–1566, 1565t–1566t in patellar osteochondritis dissecans, 1531 in patellar tendinopathy, 560, 561f of patellar tendon, 1519, 1519f, 1520f in patellofemoral tracking, 1565 in pectoralis major rupture, 902, 903f, 1060–1061, 1061f, 1163, 1164f pediatric, 590, 593f in avulsion fracture, 599 of bone, 591–592, 593f of cartilage, 593 in chondroblastoma, 604f, 605 in elbow injury, 469f, 1230 in Ewing’s sarcoma, 607, 609f, 610 in fracture, 593–596, 594f, 595f in giant cell tumor, 605, 605f in humeral fracture, 594, 594f in infection, 599, 601–603, 601f, 602f in leukemia, 610 in nonossifying fibroma, 605–606 in osteochondritis dissecans, 471f in osteochondroma, 606, 607f in osteochondroses, 599, 601f in osteoid osteoma, 603, 603f, 605 in osteomyelitis, 601, 601f in osteosarcoma, 606–607, 608f of physis, 592, 593f in posterior cruciate ligament injury, 1716 in slipped capital femoral epiphysis, 597, 598f

xxx

Index

Magnetic resonance imaging (MRI) (Continued) of soft tissues, 593 in thoracolumbar spine evaluation, 761–762 in peroneal brevis tendon tear, 560–561, 563f in peroneal tendinitis, 1989 physics of, 548–549 in pisiform fracture, 1352f in popliteal artery entrapment, 1840, 1841f–1842f in posterior cruciate ligament injury, 569, 572f, 1693 in posterior glenohumeral joint instability, 937, 938f in posterior tibial tendinitis, 1981 in posterior tibial tendon tear, 562, 563f in posterolateral corner injury, 1730, 1730f–1731f proton density images with, 550 pulse sequences for, 549–550 in quadriceps strain, 1494–1495, 1494f–1495f in quadriceps tendinopathy, 560 in quadriceps tendon rupture, 1523, 1524f in radial collateral ligament tear, 576, 577f in radial head fracture, 535f in recurrent anterior glenohumeral joint instability, 940, 940f in rhabdomyolysis, 558 of rotator cuff, 564–566, 566f–568f, 953, 953t, 958–965, 960t in rotator cuff cyst, 964–965, 964f in rotator cuff disorders, 1001, 1001f of rotator interval, 966, 966t, 967f in scaphoid fracture, 552, 554f, 1336, 1336f, 1365 in scaphoid nonunion, 1339–1341, 1341f in scapholunate ligament tear, 575, 576f in SLAP lesion, 581, 583f, 1025, 1026f in soft tissue tumor, 557–558, 560f spatial resolution in, 551–552 spin echo, 549 in spinal lymphoma, 758f in sternoclavicular joint injury, 804 STIR images with, 550 in stress fracture, 552–553, 554f–556f, 633, 634t in subcalcaneal heel pain, 2047 in subperiosteal abscess, 589f in subscapularis tendon tear, 963–964, 964f in subtalar sprain, 1950 in suprascapular nerve injury, 1121–1122, 1121f in supraspinatus fatty atrophy, 963, 963f, 980 in supraspinatus tendinopathy, 960, 960f T1-weighted, 549 T2-weighted, 549–550 in talar osteochondral lesions, 2144–2145, 2145t, 2147f in tendon injury, 558, 560–569 at ankle, 560–563, 562f, 563f, 564f, 565f at elbow, 566–567, 569f, 570f at knee, 560, 561f, 562f at shoulder, 564–566, 566f–568f at wrist, 568–569, 570f in thoracic disk herniation, 738 in thoracic outlet syndrome, 1133 in tibial stress fracture, 1852–1853, 1853f, 2015, 2016f in tibial tendon tear, 562, 563f in Tillaux fracture, 597, 598f in triangular fibrocartilage complex tear, 1436 in trochanteric bursitis, 1456, 1456f in turf toe, 2084, 2084f in ulnar collateral ligament injury, 576, 577f in valgus instability, 620 in wrist injury, 1321

Magnetic resonance neurography, in ­Parsonage-Turner syndrome, 1145 Magnetic resonance venography, in deep venous thrombosis, 376 Malingerer, 358 Malleolus, medial fracture of, 1969, 1969f stress fracture of, 644–645, 645f, 2016t, 2018, 2020f Mallet finger, 1388, 1420–1422, 1420f operative treatment of, 1420–1422, 1421f pediatric, 1407, 1407f–1408f Mallet thumb, 1401–1402 Mallet toe, 2118–2119, 2118t, 2119f, 2120b, 2128–2129 operative treatment of, 2121–2125, 2123f, 2124f Malocclusion, on-field, 525 Malunion proximal humeral fracture, 1049–1050 proximal physeal humeral fracture, 1082f, 1085 scaphoid fracture, 1368 Manual perturbation training in core training, 286, 287f in knee rehabilitation, 270–271, 271f Manual resistance training, in shoulder ­rehabilitation, 246, 248f Marathon running immune system effects of, 148 physician staffing of, 171 March fracture, 647 Marfan syndrome, 167 Marijuana use/abuse, 425–426 Marrow stimulation techniques. See also ­Microfracture in knee cartilage lesions, 1774 Mast cell stabilizers, in exercise-induced ­bronchospasm, 184 Matrixectomy, in ingrown toenail, 2099, 2101, 2102, 2105f McBride procedure, 2070, 2074b, 2075f, 2076, 2078, 2079f McMurray’s test, in meniscal injury, 1601 Mean, statistical, 112–113, 113f Medial collateral ligament (MCL) allograft with, in ligament injury, 38 anatomy of, 1624–1627, 1624f–1627f, 1748 biomechanics of, 1588–1589, 1625–1626 stabilizing function of, 1588–1589, 1625–1626 Medial collateral ligament (MCL) injury anterior cruciate ligament injury with, 1632–1633, 1633f, 1634–1635, 1634f, 1635f, 1636, 1653, 1654 anterior drawer test in, 1629, 1630t classification of, 1625, 1629t, 2216 complications of, 1637 evaluation of, 1627–1631 grade of, 1629, 1629t, 2214 history in, 1627–1628 magnetic resonance imaging in, 569, 571, 573f, 1629–1631, 1631f nonoperative treatment of, 1631–1632, 1636, 1637 palpation in, 1629 pediatric, 1638–1640 evaluation of, 1638–1639, 1638b grades of, 1639 imaging in, 1640 physical examination in, 1639, 1640b treatment of, 1640, 1640b physical examination in, 1628–1629, 1629t radiography in, 1629, 1630f rehabilitation in, 1636–1637, 1636t

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Medial collateral ligament (MCL) injury (Continued) swelling in, 1629 treatment of, 1631–1635, 1633f in grade III injury, 1634–1635, 1634f–1635f nonoperative, 1631–1632, 1636–1637 rehabilitation after, 1636–1637 return to play after, 1637, 1637b valgus stress testing in, 1629, 1630t Medial patellofemoral ligament, 1551, 1551f, 1552 anatomy of, 1534–1536, 1535f, 1551, 1551f in patellar dislocation, 1535–1536, 1536f, 1541–1542, 1542t reconstruction of, 1569–1570, 1571t repair of, 1546 Medial patellomeniscal ligament, 1551, 1551f, 1552 Medial patellotibial ligament, 1551, 1551f, 1552 Medial tibial stress syndrome, 15, 545, 545f, 652, 652t, 1857. See also Chronic exertional compartment syndrome Median, statistical, 112 Median nerve anatomy of, 1159 injury to, 1359–1361 clinical manifestations of, 1360 at elbow, 1317–1318 electrodiagnostic study in, 1361 physical examination in, 1360 in supracondylar fracture, 1282 treatment of, 1360–1361, 1360f Medicolegal issues, 531 Melatonin dim light onset of, 457 in jet lag, 459–460 phase response curve to, 458–459, 459f sleep and, 457 Membrane stabilizers, in complex regional pain syndrome, 363t, 364 Memory declarative, 449 procedural, 449 sleep-related consolidation of, 449–450, 449f Meniscal arrow device, 1611–1612, 1611f Meniscal tear, 56–57, 62–64, 1596–1623 ACL injury and, 1653–1654, 1653t epidemiology of, 1601 rehabilitation for, 1672–1673 treatment of, 1605–1606, 1654, 1654t acute, 1603 Apley’s test in, 1601 arthroscopic examination in, 1602–1603 asymptomatic, 1601 bucket handle, 579, 580f, 1603–1604, 1603f classification of, 1603–1604, 1604f clinical presentation of, 1601 complex, 1604 degenerative, 1603, 1604f double PCL sign in, 1602 epidemiology of, 1600–1601 fascicular, 1613 grade I, 1602 grade II, 1602 grade III, 1602, 1603f historical perspective on, 1596–1597 horizontal, 1604, 1604f Lachman maneuver in, 1602 magnetic resonance imaging in, 579, 579f, 580f, 1602, 1603f McMurray’s test in, 1601 nonoperative treatment of, 1604–1605 oblique, 1604, 1604f

Index Meniscal tear (Continued) operative treatment of, 1605–1615 ACL injury and, 1605–1606, 1654, 1654t all-inside repair in, 1610–1613, 1611f–1612f, 1617 arthroscopic evaluation for, 1606, 1606f biologic enhancement techniques in, 1608 complications of, 1618–1619. See also ­Meniscus (menisci), deficiency of cyst and, 1613 discoid meniscus and, 1613–1615, 1614f–1615f equipment for, 1606 Fas-T-Fix device in, 1612–1613, 1612f, 1617 fascicular tears and, 1613 fibrin clot augmentation in, 1608 indications for, 1605 inside-out repair in, 1608–1610, 1609f, 1610f outcomes of, 1615–1617 outside-in repair in, 1610, 1611f rehabilitation after, 1618 repair in, 1607–1613 resection in, 1606–1607, 1607f. See also Meniscectomy return to play after, 1618 root avulsion and, 1613 trephination in, 1608 palpation examination in, 1601 patient history in, 1601 physical examination in, 1601–1602 pivot-shift test in, 1602 radial (transverse), 1604, 1604f radiography in, 1602 red-red, 1604 red-white, 1604 squat test in, 1602 terminology for, 1603–1604 vascular, 62–63 vertical/longitudinal, 1603–1604, 1604f white-white, 1604 zone classification of, 1604, 1604f Meniscectomy, 1606–1607, 1607f arthroscopic, 1606–1607, 1607f ACL biomechanics in, 1585–1586 in discoid meniscus, 1615 historical perspective on, 1589 instability after, 1590 load transmission and, 1589 meniscal deficiency after, 1619–1623 allograft transplantation for graft procurement for, 1620–1621 indications for, 1619–1620 outcomes of, 1622 technique of, 1621–1622, 1621f evaluation of, 1619, 1619f osteoarthritis and, 1772 outcomes of, 1615–1617 partial, 1606–1607, 1607f ACL biomechanics in, 1585–1586 biomechanics of, 1589 outcomes of, 1615–1617 Meniscofemoral ligaments, 1597–1598, 1685, 1686f accessory, 1597–1598, 1598f Meniscus (menisci), 56–65. See also Meniscal tear allograft, 1620–1622, 1621f anatomy of, 58–59, 58f–59f, 1597–1599 attachments of, 1597, 1598f tears of, 1613 biomechanics of, 59–61, 60f, 61f–62f, 1589–1591 load transmission and, 1589, 1599–1600

Meniscus (menisci) (Continued) stabilizing function and, 1590–1591, 1600, 1600f blood supply of, 61–62, 1599, 1599f cells of, 57 collagen of, 57, 57t, 58–59, 58f–59f composition of, 57–58, 57t compression of, 60–61, 62f creep of, 59 cysts of, 1613 deficiency of, 1619–1623 allograft transplantation for contraindications to, 1619–1620 graft procurement for, 1620–1621 indications for, 1619–1620 outcomes of, 1622 technique of, 1621–1622, 1621f evaluation of, 1619, 1619f discoid, 1613–1615 classification of, 1613–1614, 1614f examination of, 1614–1615, 1615f resection of, 1615 saucerization of, 1614–1615, 1615f Wrisberg ligament variant in, 1597–1598, 1614 elastin of, 58 embryology of, 1597 extracellular matrix of, 57–58, 57t fibrochondrocytes of, 1598–1599 functions of, 56, 1599–1600 healing of, 62–63 hoop stress in, 60 lateral anatomy of, 1597, 1598f attachments of, 1597, 1598f stabilizing function of, 1591–1592 load transmission through, 1589, 1599–1600 medial anatomy of, 1597, 1598f attachments of, 1597 stabilizing function of, 1590–1591 microstructure of, 1598–1599, 1599f microvasculature of, 1599, 1599f nerve supply of, 62, 1599 noncollagenous proteins of, 58 proteoglycans of, 58, 1599 radial tie fibers of, 59, 59f regeneration of, 63 repair of, 62–64, 1607–1613 all-inside, 1610–1613, 1611f, 1613f in avascular regions, 63–64 Fas-T-Fix device in, 1612–1613, 1612f fibrin clot augmentation with, 1608 grafts in, 63–64, 143–144 inside-out, 1608–1610, 1609f–1610f meniscal arrow device in, 1611–1612, 1611f outcomes of, 1616–1617 outside-in, 1610, 1611f trephination with, 1608 in vascular regions, 62–63 resection of. See Meniscectomy shear properties of, 61, 62f stabilizing function of, 1590–1591, 1600, 1600f stiffness of, 60, 61f, 1600 stress relaxation of, 59 structure of, 58–59, 58f–59f ultrastructure of, 61, 62f viscoelastic properties of, 59–60, 60f, 1600 water of, 57, 57t Menopause breast cancer and, 481 exercise and, 480–481, 481b Menstrual cycle, 445 ACL injury and, 1651

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxxi

Meralgia paresthetica, 1469 Metabolic rate, basal, 498 Metacarpal fracture adult, 1393–1394, 1393f–1395f oblique, 1394, 1395f transverse, 1394, 1395f pediatric, 1411–1414, 1411f–1413f Metacarpophalangeal joint anatomy of, 1380 dislocation of, 1379–1381 in children, 1415–1416, 1416f dorsal, 1380–1381, 1380f, 1381f volar, 1381 ligaments of, 1380 sagittal band rupture over, 1390 sprain of, in children, 1414–1415 thumb, dislocation of, 1398–1399 Metal dermatitis, 202, 203f Metastases, 74t, 77t humeral, 983, 983f radionuclide imaging in, 547, 547f Metatarsal(s) fifth base fracture of, 1933–1934, 2024–2025, 2026f diaphyseal fracture of, 1969–1970, 1971f, 2025–2027, 2026f–2027f stress fracture of, 648, 649f traction apophysitis of, 2167–2169, 2168f–2169f first combined multiple osteotomies in, 2071–2072, 2078f, 2079, 2080f plantar flexion of, 2108, 2109f fracture of, in children, 1969–1971, 1970f–1971f motion of, 2180, 2180f osteotomy of, 2111, 2111b, 2112f–2114f plantar flexion of, 2108, 2109f stress fracture of, 646–648, 648f, 2024–2030 base, 2027, 2028f basilar, 2027 flexibility and, 2182 malunion of, 2024 pediatric, 1970–1971 treatment of, 2016t, 2024–2030 Metatarsal break, 1870, 1871f Metatarsal head, osteochondrosis of, 599, 1973, 1973f, 2166–2167, 2167f, 2168t treatment of, 2166, 2168f Metatarsal osteotomy bunionette treatment with, 2135–2136, 2135b diaphyseal, 2135–2136, 2137, 2138b, 2139f, 2140f distal chevron, 2135–2136, 2135b, 2137b, 2138f distal oblique, 2135–2136, 2135b–2136b, 2136f–2137f hallux valgus treatment with, 2070–2071, 2076b, 2077f, 2078–2079 intractable plantar keratosis treatment with, 2111–2115, 2111b, 2112f–2114f Metatarsal pad, 2110, 2111f, 2121 Metatarsocuneiform joint, first, 2066, 2067f Metatarsophalangeal joint(s) first, 2082f. See also at Hallux anatomy of, 2064–2066, 2065f–2066f, 2081–2082, 2082b, 2180, 2181f arthrodesis of, 2075 biomechanics of, 2082, 2082f congruency of, 2066, 2067f dislocation of, 2084–2085, 2086f injury to. See Turf toe motion of, 2178–2179, 2179t

xxxii

Index

Metatarsophalangeal joint(s) (Continued) sesamoids of. See Sesamoid(s) subluxation of. See Hallux valgus impairment of, 1872 lesser. See also Toe(s), lesser anatomy of, 2115–2116, 2116f arthroplasty of, 2124, 2130f biomechanics of, 2115–2116, 2116f, 2180–2181, 2182f flexibility/inflexibility of, 2182 instability of, 2119, 2120b, 2120f medial deviation of, 2116, 2117f, 2119 treatment of, 2123–2124, 2128f plantar drawer test of, 2119, 2120f radiography of, 2121 release of, 2123–2124, 2125b, 2126f subluxation/dislocation of, 2116, 2119, 2120f treatment of, 2123–2124, 2125b, 2126f–2130f Methacrylate cement, fluoroscopy-guided injection of, 585, 585f Methandrostenolone, 411 Methicillin-resistant Staphylococcus aureus ­infection, 194–195, 195f, 395–396, 396b Methyldopa, 160t Methylprednisolone, in adhesive capsulitis, 1096, 1103b Metolazone, 160t Metoprolol, 160t Mexiletine, in complex regional pain syndrome, 364 MIBI (methoxyisobutyl isonitrile) perfusion scan, in compartment syndrome, 651 Microfracture, in knee cartilage lesions, 1776, 1777–1778, 1779f, 1784t Micronutrients, requirements for, 405–406, 406b Mile-high effect, 503–504 Milk-alkali syndrome, 77t Milligram test, in thoracolumbar spine injury, 722, 723f Miner’s elbow. See Olecranon bursitis Minoxidil, in hypertension, 160t Moccasin, 1873–1874 Mode, 112 Model building, 113–114 Moexipril, 160t Molluscum contagiosum, 197, 198f Monitored Rehab Systems Cable Column, 233, 233f, 296, 299f Monitored Rehab Systems Functional Squat, 224, 224f, 295–296, 297f Mono-Spot test, 150 Mononucleosis, 149, 151 Monster walk exercise, in knee rehabilitation, 258, 259f Monteggia fracture, 1259, 1260, 1260f, 1273, 1287 Mood anabolic-androgenic steroid effects on, 417 disorders of, sleep disorders and, 452–453 Morning alertness zone, 454, 455f Morningness-Eveningness Scale questionnaire, 456, 456f Morton’s neuroma, 2184, 2184f Moses’ sign, 374 Motion, joint, 90, 93f, 2178–2182, 2178f. See also Range of motion and at specific joints Motor system in complex regional pain syndrome, 357–358 in thoracolumbar spine injury, 722, 722t Motor unit, 209, 214, 216f adaptability of, 9 contractile properties of, 8–9, 9f

Motor unit (Continued) fast-twitch, 8 recruitment of, 8, 9f Hanneman principle of, 209 slow-twitch, 8 Mountain sickness, 504 Mouth leukoplakia of, smokeless tobacco and, 428 on-field injury to, 525 MRI. See Magnetic resonance imaging (MRI) Multiple baseline research design, 105, 105f Multiple endocrine neoplasia, 77t Multiple myeloma, 77t Multiple Sleep Latency Test, 450–451, 451f Muscle(s), 3–20, 207–213. See also specific muscles adaptability of, 9, 10–11, 19 aerobic capacity of, 212–213, 212f ATP-phosphocreatine system of, 210, 210f atrophy of, rehabilitation for, 224–225 carbohydrate metabolism of, 210f, 211 compensatory hypertrophy of, 10 contraction of, 5, 7–10, 8f–9f, 208–210, 209f concentric, 213, 213f eccentric, 213, 213f electrical current reeducation of, 230–233, 231f–233f force generation in, 7–8, 8f–9f isometric, 213 motor unit in, 8–9, 9f sarcoplasmic reticulum in, 5, 7f sliding filament model of, 5 contusion of, 13–14, 14f. See also Contusion cramps in, 11–12 creatine effects on, 418 cross-reinnervation of, 9–10 cross-sectional area of, 3, 5f delayed-onset soreness of, 12–13, 13f, 215–216 endurance training response of, 215, 216t energy metabolism of, 210–213, 210f exercise effects on, 10–11, 213–214, 214t, 215f fat metabolism of, 210f, 211 fatigue of, 19, 214 fibers of, 3, 4f, 5f, 207, 207t. See also Muscle fiber(s) function of, 3 glycolytic system of, 210–211, 210f growth hormone effects on, 419 hemorrhage within, 17–18, 18f hyperplasia of, 10, 214 hypertrophy of, 10, 214 injury to, 11–19. See also specific injuries and disorders response to, 11 intracompartmental pressure of increase in. See Chronic exertional ­compartment syndrome; ­Compartment syndrome measurement of, 14, 650–651, 1860, 1860t laceration of, 11, 11f magnetic resonance imaging of, 16, 557–558, 559f, 560f myofibrillar proteins of, 3–5, 5t, 6f, 11 one-joint, 3 oxidative system of, 210f, 211 physiology of, 7–11, 8f–9f protein metabolism of, 210f, 211–212 reflex effects in, 19 resistance training response of, 214–215, 216t sliding filament model of, 5 strain injury of, 15–16. See also at specific muscles active stretch and, 17, 17f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Muscle(s) (Continued) classification of, 1485, 1485t, 2214–2215, 2214t clinical studies of, 15–16, 17f computed tomography of, 16, 17f hemorrhage with, 17–18, 18f laboratory studies of, 16–18, 17f–18f magnetic resonance imaging of, 16, 558, 559f mechanism of, 15 vs. muscle soreness, 15 nondisruptive, 17–18, 18f passive stretch and, 17, 17f prevention of, 16, 19 structural changes with, 15–16, 17f treatment of, 16 stress relaxation of, 18, 19f structure of, 3–7, 4f–7f, 5t, 7t, 207–208, 208f tension-length curve for, 7–8, 8f tetanus of, 7–8, 8f training response of, 213–214, 214t, 215f tumors of, 558, 560f twitch of, 7–8, 8f two-joint, 3 ultrastructure of, 4f, 5, 6f in delayed-onset soreness, 12, 13f viscoelastic properties of, 18, 19f warm-up effects on, 19 Muscle fiber(s), 3, 4f, 5f, 207, 207t in athlete, 7, 10 exercise effects on, 10–11, 209–210 fast-twitch to slow-twitch transition in, 10–11 nuclei of, 5 type I, 5, 6–7, 7t, 207, 207t type II, 5, 6–7, 7t type IIA, 6, 7t, 207, 207t type IIB, 6, 7t, 207, 207t Musculocutaneous nerve, 1034, 1159 Musculotendinous unit, 3, 4f, 20. See also Muscle(s); Tendon(s) Mycobacterium infection, 397, 398, 398t Mycobacterium kansasii infection, 398, 398t Mycobacterium ulcerans infection, 397 Myocardial infarction, anabolic-androgenic steroids and, 416 Myocarditis, 151–152 sudden death and, 165–166, 166f Myocardium anabolic-androgenic steroid effects on, 416 contusion of, 526 inhalant effects on, 430 Myofibrillar proteins, 3–5, 5t Myosin, 3–4, 5t, 6f, 208, 209f ATPase of, 8–9, 9f Myositis, exertional, 496 Myositis ossificans, 13–14, 14f, 82, 1460 adductor, 1493 after proximal humeral fracture, 1050 quadriceps, 1483, 1484b, 1484f, 1495–1496 Myostatin, 422

N Nadolol, 160t Nail plate injury, 1428, 1428f, 1429f–1430f Naps motor skill learning and, 449, 449f power, 453, 454b Natural killer cells, exercise effects on, 147, 148t Naturalistic research, 106–107 Navicular accessory, 1963, 1963f, 1978, 2162–2165, 2163f–2164f

Index Navicular (Continued) excision of, 2165, 2165f injury to, 2022–2023, 2024f stress fracture of, 2016t, 2020–2022, 2023f Near-infrared spectroscopy, in compartment syndrome, 651 Neck on-field injury to, 522–524. See also Cervical spine injury tension of, head injury and, 658 Neck check, 150 Necrosis aseptic. See Avascular necrosis cutaneous, after patellar fracture, 1576 Needle cricothyroidotomy, 519, 520f Needle test, in Achilles tendon rupture, 2003 Neer’s sign, in rotator cuff disorders, 997, 1000f Nerve(s). See also specific nerves afferent, 353 efferent, 353 injury classification for, 2215, 2215f on-field injury to, 527 transection of in complex regional pain syndrome, 354, 354f Nerve conduction study in carpal tunnel syndrome, 1361 in entrapment neuropathy, 1311 in suprascapular nerve injury, 1122–1123 in ulnar neuropathy, 623 Nerve root block, fluoroscopy in, 534, 535f Nerve root injury, cervical, 670–673, 671f Neuralgic amyotrophy. See Parsonage-Turner syndrome Neuritis, brachial. See Parsonage-Turner syndrome Neuroblastoma, 610 Neurologic examination in elbow heterotopic ossification, 1291–1292 preparticipation, 513, 514t–515t in thoracolumbar spine injury, 721–723, 722f, 722t, 723f, 759, 759b in ulnar neuropathy, 622–623 Neurologic injury. See also at specific nerves in lumbar disk herniation, 743 in pediatric thoracolumbar spine injury, 758 Neuroma, Morton’s, 2184, 2184f Neuromodulation therapy, in complex regional pain syndrome, 366–367 Neuromuscular activation exercises in ankle rehabilitation, 273, 274f in core training, 285–286 Neuromuscular control exercises in ankle rehabilitation, 273, 275, 275f–276f in female ACL injury prevention, 334 in knee rehabilitation, 268–271, 269t, 271f Neuromuscular stimulation. See Electrical stimulation Neuropathy. See at specific nerves Newton’s laws first, 89t, 92 second, 89t, 92–93 third, 86–87, 89f, 89t Niacin, requirements for, 406b Nicardipine in complex regional pain syndrome, 363t in hypertension, 160t Nickel dermatitis, 202 Nicky’s knot, 135, 135f Nicotine, 427–428 Nifedipine, in complex regional pain syndrome, 363t, 364 Nisoldipine, 160t Nitric oxide, ergogenic effect of, 421 Noble compression test, 628, 628f

Nocardia infection, 398t Nonossifying fibroma, in children, 605–606, 605f Nonsteroidal anti-inflammatory drugs in complex regional pain syndrome, 363t, 364 in glenohumeral joint osteoarthritis, 1109 in heterotopic ossification, 1294, 1297 in hip degenerative disease, 1504 in knee arthritis, 1774 in knee osteoarthritis, 1791 in lateral epicondylitis, 1200 in olecranon bursitis, 1248–1249 in plantar fasciitis, 2048 in retrocalcaneal bursitis, 2035 in shoulder pain, 1006–1007, 1011 in SLAP lesion, 1026 in tendon disorders, 27, 29, 31 in tibial stress fracture, 1854 Nonunion distal humeral fracture, 1256, 1258, 1278 olecranon fracture, 1276, 1278 proximal humeral fracture, 1049 proximal ulnar fracture, 1276, 1278 radial head fracture, 1278 scaphoid, 1339–1340, 1367–1368 Nordic hamstring lower exercise, 337, 337f, 337t Norepinephrine, 353 exercise effects on, 217–218, 217t, 218f Nose, on-field injury to, 525 Nosebleed, on-field, 525 NS-398, 16 Nuclear medicine. See Radionuclide imaging Nucleus pulposus, 717. See also Intervertebral disk biomechanics of, 719 Nursemaid’s elbow, 1301, 1302f, 1303–1304, 1305, 1305f Nutrient artery, 69, 69f–70f Nutrition, 399–410 bone health and, 406 calorie requirements in, 402–403 carbohydrate requirements in, 403–404, 403t in children/adolescents, 467 fat requirements in, 405, 405t in female athlete, 477–479, 478f, 478t, 479b goals of, 399 hydration in, 401–402, 402t micronutrient requirements in, 405–406, 406b protein requirements in, 404–405, 405t screening form for, 399, 400f sodium requirements in, 402 supplement use and, 406–409, 409b weight management and, 406, 406b, 407b, 408b

O Ober’s test, 628, 629f, 1458, 1558 Obesity, 515t arthritis and, 1773–1774, 1790 O’Brien’s test in acromioclavicular joint injury, 830, 832f in rotator cuff disorders, 997 in SLAP lesion, 915, 1025, 1025t, 1216, 1216f Observational study, 101–103, 101f, 102t–103t Obturator externus, strain of, 1460–1461 Obturator internus, 1454t, 1455f Obturator nerve, entrapment of, 1468 Odds ratio, 2218

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxxiii

Odontoid congenital anomalies of, 692, 692f, 711 fracture of, 677–678, 678f in children/adolescents, 707 computed tomography of, 540–541, 540f Olecranon, 1189–1190, 1189f fracture of, 1271–1276 classification of, 1272–1273, 1272f dislocation with, 1263–1264, 1266f, 1267f–1268f, 1273 evaluation of, 1271–1273, 1272f nonunion of, 1276, 1278 operative treatment of, 1271, 1273–1276 arthrosis after, 1278 complications of, 1276, 1277–1278 heterotopic ossification after, 1277 incision in, 1273 infection after, 1278 instability after, 1277–1278 nonunion after, 1278 rehabilitation after, 1276–1277 tension band wiring in, 1273–1275, 1274f–1275f ulnar nerve injury after, 1277 wound problems after, 1278 pediatric, 1280t, 1286–1287 stress, 1225, 1225f ossification of, 1228 osteophytes of, 1239, 1239f, 1240 resection of, 1192, 1192f Olecranon bursitis, 1209–1212, 1246–1249 acute, 1210–1211 anatomy of, 1209–1210, 1210f, 1246–1247 arthroscopic treatment of, 1212 chronic, 1211–1212, 1211f classification of, 1210–1211 clinical evaluation of, 1247, 1247f crystal-induced, 1247 operative treatment of, 1211–1212, 1248, 1249 radiography of, 1247, 1247f septic, 1212–1213, 1247, 1248–1249 traumatic, 1247 treatment of, 1248–1249, 1248f Olecranon osteotomy, 1252–1253, 1252f Oligospermia, anabolic-androgenic steroids and, 416 Olmesartan, 160t Olympic athletes, drug use by, 410–411, 412–413 Omohyoid, scapular attachment of, 858, 858f–859f, 886f On-field emergency, 516–530 abdominal, 526–527 clavicular, 526 cold-related, 528–529 ear, 525 environmental, 528–530, 528f equipment fort, 517, 518t–519t extremity, 527–528 genitourinary, 526–527 head and neck, 522–524, 523t, 524f heat-related, 528, 528f, 529–530 laryngeal, 525 logroll for, 519, 519f medical bag for, 516–517, 518t–519t musculoskeletal, 527–528 nasal, 525 ocular, 525 oral, 525 pelvic, 526–527 preparation for, 517 primary surgery in, 517, 519–520, 520f, 520t, 521f rib, 525, 526

xxxiv

Index

On-field emergency (Continued) secondary surgery in, 520, 521f thoracic, 525–526 Onychectomy complete, 2099, 2101, 2102, 2103f partial, 2099, 2102f Opioids abuse of, anabolic-androgenic steroid use and, 417 in complex regional pain syndrome, 363t, 364 Oral cancer, smokeless tobacco and, 428 Organ loss, clearance for participation and, 513 Organ transplantation, 515t Orthotic devices. See also Shoes, orthotic devices for for flatfoot, 2044, 2046f injury and, 2189–2190 for intractable plantar keratoses, 2110, 2111f for plantar fasciitis, 2049 for retrocalcaneal bursitis, 2035 for tarsal tunnel syndrome, 2058 Os acromiale, 859, 860f magnetic resonance imaging of, 955–956, 958f unfused, 781, 781f Os calcis, 2031, 2032f fracture of, 2051 ostectomy of, 2000 superior tuberosity of enlargement of, 2031, 2031f. See also ­Retro­calcaneal bursitis excision of, 2038, 2038f Os odontoideum, 692, 692f, 711 Os peroneum fracture of, 1988 stress fracture of, 2024 Os subfibulare, 1918 Os tibial externum, 1978 Os trigonum, displacement of, 2019, 2021f Os vesalianum, 2168, 2169f Osgood-Schlatter disease, 1526f, 1527–1529 diagnosis of, 1528 etiology of, 1528 imaging in, 599 natural history of, 1528 treatment of, 1529 OssaTron, for plantar fasciitis, 2049–2050 Ossification acromion, 859, 860f appositional, 78 capitellum, 1227–1228 clavicular, 780 coracoid, 859, 859f–860f, 879, 881f distal femur, 587, 588f distal humerus, 1227–1228, 1228f–1229f elbow, 1227–1229, 1228f–1229f, 1279, 1280f glenoid fossa, 859 heterotopic. See Heterotopic ossification humeral condyle, 1228 olecranon, 1228 proximal humerus, 779, 782, 782f proximal radius, 1228 scapular, 781, 781f Osteitis pubis, 1466, 1466f vs. adductor strain, 1490, 1491f Osteoarthritis acromioclavicular joint, 956, 959f glenohumeral joint, 929, 1104–1119. See also Glenohumeral joint osteoarthritis hip, 1467, 1467b classification of, 1502, 1502b clinical presentation of, 1502–1503 etiology of, 1500 imaging of, 1503

Osteoarthritis (Continued) nonoperative treatment of, 1503–1504, 1504b, 1504f operative treatment of. See Arthroplasty, hip pain in, 1503 physical examination in, 1503 primary, 1500 range of motion testing in, 1503 secondary, 1500 knee. See also Knee, cartilage lesions of arthroplasty for. See Arthroplasty, knee arthroscopic treatment of, 1792 bracing in, 1790 evaluation of, 1789–1790 exercises in, 1790 imaging in, 1789–1790, 1790f lifestyle modification in, 1790 medical management of, 1791–1792 meniscectomy and, 1772 osteotomy in, 1792 physical examination in, 1789 posterolateral corner injury and, 1725, 1743–1744 proximal tibial osteotomy in, 1792 support devices in, 1790 sesamoid, 2089 sternoclavicular joint, 811–812, 820 Osteoblasts, 65–66, 78 Osteocalcin serum, 67 urinary, 67 Osteochondral allograft/autograft glenohumeral joint, 1111, 1111f, 1115, 1115f knee, 1775, 1777, 1777t, 1780–1783, 1781f, 1782f, 1784t talus, 2146–2147, 2148, 2150–2152, 2150t, 2151f–2152f, 2153 Osteochondral autologous transfer system, in talar osteochondral lesions, 2146–2147, 2148, 2150–2152, 2151f–2152f Osteochondral fracture, 49t, 51–52 magnetic resonance imaging in, 578–579, 578f Osteochondritis, sesamoid, 2089, 2090, 2092f, 2169–2170, 2170f Osteochondritis dissecans capitellum, 623, 624f, 1238–1239, 1241–1246 arthroscopy in, 1243–1244, 1243f, 1245f classification of, 1241 evaluation of, 1241–1242 grade of, 1241 history in, 1241 imaging in, 1241–1242, 1242f–1243f nonoperative treatment of, 1242 operative treatment of, 1244–1245, 1244f osteochondral grafting in, 1244, 1245f physical examination in, 1241 return to play after, 1244, 1246 treatment of, 1242–1244, 1242f, 1245f femoral condyle, 1766–1771 arthroscopy in, 469, 471f, 1767 classification of, 1766–1767, 1767f, 1767t epidemiology of, 1767–1768, 1768f etiology of, 1768 loose bodies in, 1770 MRI classification of, 1767, 1767t natural history of, 1768–1769 nonoperative treatment of, 1770 scintigraphic classification of, 1767, 1767t treatment of, 1769–1770, 1769f, 1773f zonal classification of, 1766, 1767f femoral trochlea, 1768, 1768f patellar, 1530–1533 diagnosis of, 1531, 1768, 1768f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Osteochondritis dissecans (Continued) etiology of, 1530–1531 natural history of, 1531 vs. osteochondrosis, 1526–1527, 1527f treatment of, 471f, 1531–1533 radiography in, 599, 599f talar, in children/adolescents, 1971–1972 Osteochondroma scapular, 887, 888f subungual, 2105, 2105t Osteochondrosis (osteochondroses), 77, 599, 599f, 1972–1974, 1973f, 1974f, 2153 calcaneal, 1973–1974, 1974f, 2143f, 2162, 2162f capitellar, 623, 625f, 1238 classification of, 2142–2143, 2142t–2143t cuneiform, 2143f, 2166, 2166f distal tibia, 2153, 2154f fifth metatarsal base, 2143f, 2167–2169, 2168f–2169f metatarsal head, 599, 1973, 1973f, 2166–2167, 2167f, 2168t treatment of, 2166, 2168f patellar inferior pole, 599, 1526f, 1529–1530 superior pole, 1530 talar, 2143–2153 anatomy of, 2143–2144 classification of, 2144, 2145t, 2146f evaluation of, 2144, 2146b in high-level athlete, 2153 history in, 2144 imaging in, 2144–2145, 2147f lateral sprain and, 1929 mechanisms of, 2143–2144 nonoperative treatment of, 2145–2146, 2148 operative treatment of, 2146–2147 care after, 2152 complications of, 2152, 2153f drilling in, 2146, 2148, 2149, 2149f–2150f, 2150t internal fixation in, 2146 lavage and débridement in, 2146, 2148 microfracture in, 2146, 2148 osteochondral autograft or allograft in, 2146–2147, 2148, 2150–2152, 2150t, 2151f–2152f, 2153 return to play after, 2152 physical examination in, 2144 tarsal navicular, 599, 1972–1973, 1973f, 2143f, 2162, 2163f tibial tubercle, 599, 1526f, 1527–1529 Osteoclasts, 66–67 Osteocytes, 66, 66f Osteogenesis, distraction, 81–82, 83f Osteogenesis imperfecta, 73 Osteoid osteoma, in children, 603, 603f, 605 Osteolysis, of distal clavicle, 854, 855f Osteoma, osteoid, in children, 603, 603f, 605 Osteomalacia, 72, 73t, 79b Osteomyelitis, 82–84 acute, 82 chronic, 83–84, 84f computed tomography in, 543 of foot, 1974 glenohumeral, 984 hematogenous, acute, 82 multifocal, chronic, 84 positron emission tomography in, 547 radiolabeled leukocyte imaging in, 547 radionuclide imaging in, 547 sclerosing, chronic, 84 spinal, 757, 760f subacute, 84

Index Osteonecrosis. See Avascular necrosis Osteopenia, 73t in female athlete, 478 Osteopetrosis, 73t Osteophyte(s) acromioclavicular joint, 956, 959f elbow, 621–622 olecranon, 52, 1239, 1239f proximal ulna, 1223–1224, 1223f Osteoporosis, 72, 73t in female athlete, 478, 481 periarticular (Sudeck’s atrophy), 358 stress fracture and, 632 Osteotomy medial clavicle, 810 metatarsal bunionette treatment with, 2135–2136, 2135b diaphyseal, 2135–2137, 2138b, 2139f–2140f distal chevron, 2135–2136, 2135b, 2137b, 2138f distal oblique, 2135–2136, 2135b, 2136b, 2136f–2137f hallux valgus treatment with, 2070–2071, 2076b, 2077f, 2078–2079 intractable plantar keratosis treatment with, 2111–2115, 2111b, 2112f–2114f olecranon, 1252–1253, 1252f tibial, 1804–1835. See also High tibial ­osteotomy in knee arthritis, 1792 in posterior cruciate ligament and ­posterolateral corner injury, 1744–1747, 1746f trochlear, in dysplasia, 1568–1569 Ovary, absence of, 515t Overhead throwing phases of, 993, 993f, 1091, 1092f, 1214–1215, 1215f in children/adolescents, 623, 624f, 790–791, 790f, 1231–1232, 1232f scapulothoracic motion in, 891 secondary adaptation to, 1215 Overhead throwing injury bursal, 890, 890f elbow, 1221–1226, 1223f, 1225f. See also Overhead throwing injury, pediatric evaluation of, 1215–1217, 1216f–1217f glenoid labrum, 1219–1221, 1220f humeral, 1226 inflammatory, 1218 internal impingement, 986, 987f, 992–993, 1001–1002, 1002f, 1016, 1217, 1218 evaluation of, 1216, 1216f magnetic resonance imaging in, 978–979, 979f nonoperative treatment of, 1010 magnetic resonance imaging in, 1217 neurovascular, 1226 O’Brien active compression test in, 1216, 1216f olecranon fracture, 1225, 1225f pediatric, 1227–1240. See also Little Leaguer’s elbow biomechanics in, 1231–1232, 1232f–1233f epicondylar avulsion fracture and, 1233, 1233f glenohumeral internal rotation deficit and, 1233–1234 in gymnasts, 1236 lateral articular compression and, 1233, 1233f lateral extension overload and, 1234, 1234f medial epicondylar, 1238

Overhead throwing injury (Continued) patterns of, 1232–1234, 1233f–1234f physical examination in, 1229 posterior articular damage and, 1234, 1234f posterior elbow, 1239–1240, 1239f radiography in, 1229–1230, 1230f in tennis players, 1236 ulnar collateral ligament, 1237, 1238f physical examination in, 1216–1217, 1216f, 1217f radiography in, 1217 rib fracture, 895–896, 896f rotator cuff, 1217–1219. See also Rotator cuff disorders; Rotator cuff tear(s) suprascapular nerve, 1221 triceps tendon, 1221, 1226 ulnar collateral ligament, 1221–1224, 1223f ulnar nerve, 1226 valgus extension overload, 621–622, 1224–1225 valgus stability testing in, 1217, 1217f Overuse injury, 29–30, 611–652. See also Chronic exertional compartment ­syndrome; Stress fracture and specific tendons classification of, 29–30 elbow lateral, 617–619, 618f medial, 619–624, 620f, 624f pediatric, 623–624, 623f–625f extrinsic factors in, 611 foot and ankle, 628–631. See also Achilles tendon iliac, 1475 intrinsic factors in, 611 knee, 626–628, 626f, 629f, 630f. See also ­Patellar tendinopathy; Patellar tendinosis low back, in adolescent, 766 radionuclide imaging in, 545–546, 545f–547f shoulder, 614–617, 615b. See also Rotator cuff disorders wrist, 624–626 Oxcarbazepine, in epilepsy, 189, 191t Oxidative phosphorylation, 211 Oxygen, uptake of, 212–213, 212f maximum, 218, 219f in children, 464 in female athlete, 476–477 Oxygen therapy, in high-altitude illness, 504

P P value, 110 Paget-Schroetter syndrome, 1129–1130, 1131f, 1134 Paget’s disease, 73t Pain. See also Complex regional pain syndrome back. See Low back pain in compartment syndromes, 14 definition of, 356t in delayed-onset muscle soreness, 12–13 in exertional compartment syndrome, 650 heel. See Heel, pain in knee, after tibial stress fracture treatment, 1855 leg, 1857–1863, 1857b in lumbar disk herniation, 743 sympathetically independent, 351, 360 sympathetically maintained, 351–352, 352f. See also Complex regional pain syndrome diagnosis of, 359–360 terminology for, 338t

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxxv

Pain (Continued) in thoracolumbar spine evaluation, 719–720 transcutaneous electrical nerve stimulation for, 229–230, 230f Pain dysfunction syndrome, 351 Pain provocative test, in SLAP lesion, 1025t Pancreas exercise effects on, 217t, 218 on-field injury to, 527 Panner’s disease, 623, 625f, 1238 Paralabral cyst, 964–965, 965f, 980, 980t Parasitic infestations, 200–201, 201f–202f Paratenon, 20, 21f Achilles, 563 Parathyroid hormone, 71, 71t, 77t Pars interarticularis acute fracture of, 755–756, 756b fatigue fracture of (spondylolysis), 635–636, 756, 756b, 762–763 computed tomography in, 546, 546f, 762, 763f radiography in, 635, 636f, 760, 760f radionuclide imaging in, 726, 726f, 761f return to play and, 767, 767b single-photon emission computed tomography in, 546, 546f, 635, 636f, 761, 761f Parsonage-Turner syndrome, 616, 1143–1146 anatomy of, 1143 differential diagnosis of, 1144 electromyography in, 1144 evaluation of, 1144–1145 genetics of, 1143, 1145, 1146 imaging in, 1145 long thoracic nerve in, 1124–1125, 1125f outcomes of, 1145 physical examination in, 1144 return to play and, 1145–1146 treatment of, 1145 Particle disease, 542, 542f PASTA (partial articular-sided supraspinatus tendon avulsion) lesion, 961, 962f Pasteurella multocida infection, 398t Patella, 1513–1576 anatomy of, 1513–1514, 1514f, 1548–1553 bony, 1549–1550, 1549f–1550f, 1573 ligamentous, 1534–1536, 1535f–1548f soft tissue, 1513–1514, 1514f, 1548f, 1550–1553, 1551f–1552f, 1553t vascular, 1573, 1573f ballotable, 1556 biomechanics of, 1514–1515, 1515f contact areas in, 1591–1593, 1592f exercise-related, 226, 226f force transmission in, 1593–1596, 1593f, 1594f–1595f bipartite, 1574 chondral injury of, 1538 contact areas of, 1514, 1515f, 1550, 1550f, 1591–1593, 1592f dislocation of. See Patellar dislocation exercise-related mechanics of, 226, 226f facets of, 1550 force transmission at, 226, 226f, 1593–1596, 1593f–1594f fracture of, 1572–1577 in children, 597, 1576–1577, 1576f classification of, 1573 complications of, 1576 evaluation of, 1573–1574, 1574f lag screw fixation of, 1575 multifragment, 1575 nonoperative treatment of, 1574, 1575f operative treatment of, 1574–1575 radiography in, 1574, 1574f

xxxvi

Index

Patella (Continued) return to play after, 1576 stress, 641–643 gliding movement of, 225, 225f height of Blackburne-Peel ratio for, 1523, 1523f, 1539, 1541f, 1559f, 1560, 1560t Blumensaat line for, 1539, 1541f, 1559, 1559f Caton-Deschamps ratio for, 1539, 1541f, 1559f, 1560, 1560t after high tibial osteotomy, 1832–1833 Insall-Salvati ratio for, 1522, 1523f, 1539, 1541f, 1559f, 1560, 1560t Labelle-Laurin measurement for, 1559–1560, 1559f, 1560t measurement of, 1522–1523, 1523f, 1539, 1541f non–weight-bearing exercise effects on, 226, 227–228, 226f, 227f osteochondral injury of, 1538, 1540f osteochondritis dissecans of, 1530–1533 diagnosis of, 1531, 1768, 1768f etiology of, 1530–1531 natural history of, 1531 vs. osteochondrosis, 1526–1527, 1527f treatment of, 471f, 1531–1533 osteochondrosis of, 1526–1533 inferior pole (Sinding-Larsen-Johansson disease), 599, 1526f, 1529–1530 vs. osteochondritis dissecans, 1526–1527, 1527f proximal pole, 1530 superior pole, 1530 palpation of, 1556–1557, 1556f radiography of, 1558–1564 anteroposterior view for, 1559 axial view for, 1561–1564, 1561t, 1562f, 1563f lateral view for, 1559–1561, 1559f–1560f, 1560t, 1561f Laurin’s view for, 1562, 1563f Merchant’s view for, 1562, 1562f stress, 1564 range of motion of, 1557 Sinding-Larsen-Johansson disease of, 599, 1526f, 1529–1530 stress fracture of, 641–643 subluxation of. See Patellar dislocation sulcus angle of, 1540 tracking of, 1556 abnormal, 1595–1596, 1595f weight-bearing exercise effects on, 226–227, 226f, 227f Wiberg classification of, 1513, 1514f, 1550 Patella alta, 1555, 1555f, 1560 in Osgood-Schlatter disease, 1528 Patella baja, 1555, 1576 Patella infera, 225 high tibial osteotomy and, 1832–1833 Patellar angle, 1528 Patellar dislocation, 1534–1547 apprehension test in, 1538, 1539f arthroscopy in, 1545 chondral injury with, 1538, 1540t clinical presentation of, 1536–1537, 1537f direct mechanism of, 1536–1537, 1537f effusion with, 1556 epidemiology of, 1534, 1534t in female athlete, 480, 487, 489, 489b hemarthrosis in, 1538 iatrogenic, 1545, 1567 imaging in, 1538–1542, 1541f, 1542t, 1543f incidence of, 1534 indirect mechanism of, 1536–1537, 1537f

Patellar dislocation (Continued) joint aspiration in, 1538 lipohemarthrosis in, 1538 mechanisms of, 1536–1537, 1537f, 1552–1553, 1552f medial patellofemoral ligament in, 1535–1536, 1536f, 1541–1542, 1542t, 1546 nonoperative treatment of, 1542–1544, 1544t operative treatment of, 1544–1546, 1544t lateral release in, 1545 medial patellofemoral ligament repair in, 1546 medial retinacular repair in, 1545–1546 rehabilitation after, 1546–1547 osteochondral injury with, 1538, 1540f, 1540t palpation in, 1538 physical examination in, 1538, 1539f recurrent, 1548–1572 apprehension test in, 1558 bayonet sign in, 1554 clinical presentation of, 1553, 1553t computed tomography in, 1564–1565, 1566t foot examination in, 1554 gait examination in, 1554–1555 J sign in, 1556 lateral pull test in, 1557, 1557f lower extremity examination in, 1558 magnetic resonance imaging in, 1565–1566, 1565t–1566t miserable malalignment in, 1554 nonoperative treatment of, 1566 operative treatment of, 1566–1572 arthroscopic, 1570–1572 distal realignment procedures in, 1566–1567 lateral retinacular release or lengthening in, 1567–1568 medial patellofemoral ligament ­reconstruction in, 1569–1570, 1571t proximal realignment procedures in, 1568 trochlear osteotomy in, 1568–1569 trochleoplasty in, 1568–1569 patella-trochlea compression test in, 1558 patellar glide in, 1558, 1558f patellar tilt in, 1557–1558, 1557f physical examination in, 1553–1558, 1554f–1558f quadriceps angle in, 1553–1554, 1554f radiography in, 1558–1564, 1563f, 1566t anteroposterior, 1559 axial, 1561–1564, 1561t, 1562f, 1564t lateral, 1559–1561, 1559f, 1560f, 1560t, 1561f stress, 1564 sitting examination in, 1555–1556, 1555f–1556f standing examination in, 1554–1555 supine examination in, 1556–1558, 1557f, 1558f tubercle sulcus angle in, 1555, 1556f risk factors for, 1534–1536, 1534b, 1534t Patellar glide, 1558, 1558f Patellar tendinopathy, 560, 561f, 626–627, 1516–1517 metabolic disease and, 1516, 1517b pathophysiology of, 1515–1518, 1517b, 1518t Patellar tendinosis, 30, 31, 626–627, 626f, 1516–1518, 1518t evaluation of, 1518–1519

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Patellar tendinosis (Continued) imaging in, 560, 561f, 1519, 1519f–1520f pathophysiology of, 1515–1518, 1517f–1518t stages of, 1519 treatment of, 1519–1521, 1520f Patellar tendon anatomy of, 1514, 1514f autograft with, in ligament injury treatment, 37 bioscaffold effect on, 28 collagen of, 1515, 1516f–1517f drug effects on, 1516 excision of, 1520, 1520f histology of, 1515, 1516f inflammation/degeneration of. See also Patellar tendinopathy; Patellar tendinosis evaluation of, 1518–1519, 1519f imaging in, 1519, 1519f–1520f pathophysiology of, 1515–1518, 1517f–1518f treatment of, 1519–1521, 1520f magnetic resonance imaging of, 1519, 1519f–1520f mechanobiology of, 1516 metabolic disease effects on, 1516, 1517b rabbit, 28 rupture of, 1521–1522, 1522t complications of, 1525 evaluation of, 1522–1523, 1523f–1524f histology of, 1516 treatment of, 1523–1524 Patellar tilt, 1539, 1555, 1557–1558, 1557f Patellofemoral joint. See also Knee; Patella arthritis of, in female athlete, 489 biomechanics of, 1514–1515, 1515f, 1591–1596 contact area in, 1591–1593, 1592f force transmission in, 1593–1596, 1593f–1595f contact regions of, 1591–1593, 1592f instability of. See Patellar dislocation pain in in female athlete, 480, 487 patellofemoral joint reaction force and, 1595, 1595f Patellofemoral joint reaction force, 226, 226f, 1591–1593, 1592f, 1593f with descending stairs, 227 in rehabilitation programs, 1595–1596, 1595f with squatting, 227, 1592–1593, 1592f with walking, 227 Patellofemoral ligaments, 1534–1536, 1535f, 1551–1552, 1551f. See also Lateral patellofemoral ligament; Medial patellofemoral ligament Pathologic fracture, of humeral metaphysis, 1085, 1086f Paxinos’ test, in acromioclavicular joint injury, 830 Pearson’s product-moment correlation, 100, 119 Pectineus, 1485 strain of, 1460–1461 Pectoral nerve, 901 Pectoralis major, 900–907 actions of, 901 anatomy of, 900–901, 901f, 1034, 1035f, 1059–1060, 1059f, 1162 pediatric, 785, 785f biomechanics of, 1162 humeral attachment of, 1071, 1071f innervation of, 901 press-up exercise for, 241, 242f rupture of, 900–907, 1059–1063, 1162–1165 classification of, 901–902, 902t, 1060, 1162, 1162t

Index Pectoralis major (Continued) clinical evaluation of, 1060–1061, 1162–1163 differential diagnosis of, 902 in elderly patients, 1164 evaluation of, 902, 902b, 903f, 1162–1163 magnetic resonance imaging in, 902, 903f, 1060–1061, 1061f, 1163, 1164f mechanism of, 1060 nonoperative treatment of, 903–904, 904t, 1062–1163 operative treatment of, 904–906, 904f, 904t, 906f, 1061–1063, 1062f–1064f, 1163–1165 physical examination in, 1060, 1061f, 1163 radiography in, 902, 1060 rehabilitation after, 1064, 1163–1165, 1164t retraction with, 905 return to play after, 907, 1064 shoulder dislocation and, 902 steroid use and, 902, 1164–1165 suture anchor repair of, 904, 905f, 906 trough repair of, 904, 904f, 906 Pectoralis minor, scapular attachment of, 858, 858f, 886f Pedicles cervical spine, congenital absence of, 711 lumbar spine, in congenital stenosis, 747 Pediculosis capitis, 200–201, 201f Peliosis hepatis, anabolic-androgenic steroids and, 416 Pellegrini-Stieda lesion, 1629, 1630f Pelvic rami, stress fracture of, 1465 Pelvis anatomy of, 1451–1452 avulsion injury of, 541–542, 542f, 553, 555 in children, 599, 1474–1475, 1475f computed tomography of, 541–542, 542f fracture of, 1464 hemorrhage with, 541 functions of, 1452b on-field injury to, 526–527 radiography of, 541 stress fracture of, 638, 1464–1466 Penbutolol, 160t Peptic ulcer disease, 77t Per gene, 442 Performance-enhancing substance use, in children/adolescents, 467–468 Pericarditis, 152 Perichondrium, autograft with, in articular cartilage injury, 53–54 Perilunate dislocation, 1332–1335, 1332f classification of, 1332 clinical manifestations of, 1332 fracture with, 1332, 1333f pediatric, 1371, 1372f, 1373f physical examination in, 1332 radiography in, 1332 return to play after, 1335 treatment of, 1332–1335, 1333f–1334f Perimysium, 3, 207, 208f Perindopril, 160t Periosteum, 69 autograft with, in articular cartilage injury, 53–54 Peripheral neuropathy. See at specific nerves Peripheral vascular resistance, exercise effect on, 219–220, 220t Peritendinitis, 30 Achilles, 30, 563, 1997, 2000–2001 Peritendon, 20, 21f Permethrin in pediculosis capitis, 200 in scabies, 201

Peroneal groove deepening of, 1932–1933, 1993–1994, 1994f tendon subluxation within, 1995–1996 Peroneal nerve common, injury to, 1751–1752, 1752f deep, entrapment of, 2061–2062, 2061f superficial, 15f, 2062, 2063f entrapment of, 2062–2063, 2062f injury to, 1933 tibial osteotomy–related injury to, 1832 Peroneal retinaculum inferior, 1987 superior, 338, 1931, 1987 congenital absence of, 1991 disruption of, 1932 imbrication of, 1932–1933 reconstruction of, 1995 Peroneal tunnel compression test, 1989 Peroneus brevis tendon anatomy of, 1987–1988 injury to, 1987–1988 magnetic resonance imaging of, 560–561, 563f instability of, 1931–1933, 1932f, 1963–1964 subluxation of, 1990–1996 anatomy of, 1990–1991 biomechanics of, 1991 evaluation of, 1991–1992, 1992f nonoperative treatment of, 1992–1993 operative treatment of, 1993–1996, 1993f–1994f treatment of, 1992–1996 tendinitis of, 1988–1990 Peroneus brevis tenodesis, in lateral ankle sprain, 1924–1926, 1925f Peroneus longus, exercise for, 275, 275f–276f Peroneus longus tendon anatomy of, 1987–1988 injury to, 1987–1988 instability of, 1931–1933, 1932f, 1963–1964 magnetic resonance imaging of, 560–561 subluxation of, 1990–1996 anatomy of, 1990–1991 biomechanics of, 1991 evaluation of, 1991–1992, 1992f nonoperative treatment of, 1992–1993 operative treatment of, 1993–1996, 1993f, 1994f treatment of, 1992–1996 tendinitis of, 1988–1990 Perthes’ lesion, 974, 974t Phalangeal fracture adult, 1394–1398, 1395f–1397f anatomy of, 1395 comminuted, 1394, 1395f metaphyseal, 1396, 1396f–1397f oblique, 1396, 1396f transverse metaphyseal, 1396, 1396f– 1397f pediatric, 1406–1411 distal, 1405f, 1406, 1406f–1407f mallet, 1407, 1407f–1408f middle, 1408–1410, 1408f–1410f proximal, 1411 Phalen’s test, 1312, 1360 Pharyngitis, 149 Phenobarbital, in epilepsy, 188, 191t Phenol matrixectomy, 2099–2102, 2105f Phenoxybenzamine, in complex regional pain syndrome, 363t, 365 Phentolamine testing, in complex regional pain syndrome, 359–360 Phenytoin, in epilepsy, 188, 191t Philosophical research, 108

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxxvii

Phosphate bone, 70, 71t serum, 74t–76t Phosphocreatine, in energy metabolism, 210, 210f Phosphorus blood, 77t requirements for, 406b Physical examination. See Preparticipation examination Physical therapy, in complex regional pain syndrome, 362 Physis, 78, 587–588 fracture of, 595, 595f. See also at specific bones imaging of, 591–592, 592f–593f proximal humerus, 779, 779f Pigmented villonodular synovitis, 1473, 2161, 2161f Pindolol, 160t Piriformis, 1454t, 1455f Piriformis syndrome, 1468 Pisiform anatomy of, 1319–1320, 1319f fracture of adult, 1350, 1352f pediatric, 1370–1371 Pitch angle, calcaneal, 2035, 2035f Pitching, 620. See also Overhead throwing phases of, 790–791, 790f, 1091, 1092f in children, 623, 624f, 1232, 1232f–1233f Pitted keratolysis, 196, 196f Pittsburgh sleep quality index, 450 Pituitary gland, exercise effects on, 217, 217t Pivot shift test in anterior cruciate ligament injury, 1650, 2213–2214, 2214t in meniscal injury, 1602 reverse in posterior cruciate ligament injury, 1692, 1692f in posterolateral corner injury, 1727–1728, 1729f in varus malalignment, 1807–1808 Pivot test, posteromedial, 1691 Plantar aponeurosis, 1870, 1871f, 2043–2044, 2043f Plantar digital nerves, 2088–2089, 2089f Plantar drawer test, 2119, 2120f Plantar fasciitis, 631, 2042–2056 anatomic considerations in, 2043–2044, 2043f–2045f biomechanics of, 2044 calcaneal taping in, 2049 corticosteroid injection in, 2048–2049 counterstrain treatment in, 2049 evaluation of, 2044–2046 exostosectomy in, 2051 fasciotomy in, 2051–2052, 2054 heel spur and, 2042–2043, 2042f history in, 2044 iontophoresis in, 2049 magnetic resonance imaging in, 2047 night splints in, 2049 nonoperative treatment of, 2048–2050, 2052–2053, 2053f, 2054–2055 NSAIDs in, 2048 operative treatment of, 2050–2056, 2050b, 2051f os calcis drilling in, 2051 OssaTron in, 2049–2050 physical examination in, 2044–2046 plantar fascia release in, 2051, 2052, 2055–2056, 2055f radiography in, 2046–2047, 2047f radionuclide imaging in, 2046–2047, 2047f

xxxviii

Index

Plantar fasciitis (Continued) return to play after, 2056 shock-wave therapy in, 2049–2050, 2052 Plantar keratoses, 2089, 2089f, 2107–2115, 2107b anatomy of, 2108–2109, 2109b, 2109f–2110f classification of, 2109, 2110b evaluation of, 2109, 2110b history in, 2109 imaging in, 2110 nonoperative treatment of, 2110–2111, 2111f operative treatment of, 2110–2115 care after, 2115 metatarsal osteotomy in, 2111–2115, 2111b, 2112f–2114f plantar condylectomy in, 2111, 2113f return to play after, 2115, 2115b orthotic devices in, 2110, 2111f physical examination in, 2109 vs. wart, 2108f, 2109 Plantar nerves, 2044, 2044f, 2046f Plasmin, 372 Platelet(s) in coagulation, 370–371, 371f–372f in osteochondral fracture, 51 Platelet-derived growth factor, in fracture healing, 80 Platelet-rich fibrin matrix clot, in knee ­dislocation treatment, 1763, 1763f Playing cast, 1362 Playing surface, 2192–2197 for baseball, 2196–2197 compliant, 2199 frictional properties of, 2198, 2204, 2204f hardness of, 2197–2198, 2197f–2198f historical perspective on, 2193, 2196 for indoor sports, 2196 injury and, 2192–2204 clinical studies of, 2203–2204, 2203f–2204f, 2204t energy loss and, 2198–2199, 2199f experimental studies of, 2199–2203, 2200t, 2201f, 2202t, 2203f friction and, 2198, 2204, 2204f hardness in, 2197–2198, 2197f–2198f incidence in, 2196–2197 for playgrounds, 2197 for tennis, 2193–2196, 2194t–2195t, 2203, 2203f, 2204t for track, 2196, 2204, 2204f Plyoball exercises in core training, 287–288, 289f in knee rehabilitation, 294, 294f in shoulder rehabilitation, 296, 298f Pneumonia, 150 Pneumothorax, on-field, 525–526 Point estimate, 99 Polycystic ovarian syndrome, valproate and, 189 Polysomnography, 450 Polythiazide, 160t Popeye deformity, 1165, 1166f Popliteal artery, 1686, 1687f, 1748, 1838f aneurysm of, 1838, 1844 embryology of, 1836 entrapment of, 1836–1847 anatomy of, 1836, 1838f angiography in, 1840, 1842, 1843f–1844f ankle-brachial index in, 1839 biomechanics of, 1836 classification of, 1836–1837, 1837t clinical presentation of, 1837, 1838b computed tomographic angiography in, 1839–1840, 1840f gender and, 1837

Popliteal artery (Continued) historical perspective on, 1836 imaging in, 1839–1842, 1839f–1844f incidence of, 1836 magnetic resonance angiography in, 1840, 1842f magnetic resonance imaging in, 1840, 1841f–1842f patient history in, 1837 physical examination in, 1837–1839, 1838b treatment of, 1842, 1844–1847, 1845f, 1846b, 1846t complications of, 1846–1847 graft thrombosis after, 1847 medial approach to, 1844–1845, 1846t posterior approach to, 1844, 1845f, 1846t return to play after, 1847, 1847b ultrasonography in, 1839, 1839f traumatic injury to, 1750, 1751, 1753, 1753f Popliteal (Baker’s) cyst, 537–538, 538f Popliteal plexus, 1686 Popliteofibular ligament, 1685, 1687, 1719f, 1720–1721, 1721f, 1722f. See also ­Posterolateral corner Popliteus muscle-tendon complex, 1687–1688, 1720–1721, 1721f, 1731f Popliteus tendon anatomy of, 1719f, 1720–1721, 1721f, 1731f avulsion of, 541–542, 542f recess procedure for, 1732–1733, 1733f reconstruction of, 1736, 1736f stabilizing function of, 1585, 1588–1589 Population, statistical, 99 Positron emission tomography, 547 Post hoc power analysis, 111 Postconcussion syndrome, 659 Posterior circumflex humeral artery, 1071f, 1072 Posterior cruciate ligament (PCL). See also ­Posterior cruciate ligament (PCL) injury anatomy of, 1684–1688, 1684f–1686f, 1748 in children, 1714, 1714f anterolateral bundle of, 1684, 1685f–1686f, 1687, 1687t avulsion of, 1695, 1702 magnetic resonance imaging in, 569, 572f pediatric, 1716–1717, 1717f biomechanics of, 1687–1688, 1687t force measurements in, 1581–1582, 1585 observational studies of, 1581 stabilizing function in, 1584–1588 strain measurements in, 1582 in weight-bearing flexion, 1585 blood supply to, 1686, 1687f innervation of, 1686 ligament unloading exercises for, 222–224, 223t loading of, rehabilitative exercise effects on, 221–222, 222t posteromedial bundle of, 1684, 1685f–1686f, 1687, 1687t squat exercise effects on, 223, 223t, 224, 224f, 262 stabilizing function of, 1584–1588 stair climbing effects on, 223t, 224 stationary cycling effects on, 223, 223t Posterior cruciate ligament (PCL) injury, 1683–1712 acute vs. chronic, 1689 avulsion, 1695, 1702 magnetic resonance imaging in, 569, 572f pediatric, 1716–1717, 1717f chronic, 1689 treatment of, 1702–1703, 1702f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Posterior cruciate ligament (PCL) injury (Continued) classification of, 1689, 2216 clinical presentation of, 1688–1689, 1688f–1689f collateral ligamentous examination in, 1692 combined, 1689, 1700 treatment of, 1703–1704, 1706–1710, 1708f–1710f dial test in, 1691 gait evaluation in, 1692 grading of, 1689 incidence of, 1688 isolated natural history of, 1693, 1695, 1695f nonoperative treatment of, 1693–1695, 1694t, 1695f, 1701–1702, 1701f operative treatment of, 1701–1703, 1701f, 1705–1706, 1705f–1707f review of, 1695–1700, 1697t–1699t kinematic changes with, 1687–1688, 1687t limb alignment evaluation in, 1692 magnetic resonance imaging in, 1693 mechanisms of, 1688–1689, 1688f–1689f pediatric, 1713–1718, 1714f avulsion, 1716–1717, 1717f classification of, 1715 evaluation of, 1715–1716 magnetic resonance imaging in, 1716 mechanism of, 1714, 1715f multiligament, 1717 natural history of, 1716 radiography in, 1716 treatment of, 1716–1717, 1717f physical examination in, 1690f–1691f, 1691–1692 posterior drawer test in, 1690, 1690f, 2214 posterior sag test in, 1690, 1691f posterolateral corner injury and, 1732–1744, 1745f double-bundle reconstruction in, 1706–1710, 1708f–1710f proximal tibial osteotomy in, 1744–1747, 1746f posteromedial pivot test in, 1691 quadriceps active test in, 1691, 1691f radiography in, 1692–1693 radionuclide imaging in, 1693 reverse pivot shift test in, 1691f, 1692 thumb sign test in, 2214 treatment of, 1693–1710 nonoperative, 1693–1695, 1694t, 1695f, 1700, 1701f operative, 1695–1710, 1701f, 1702f. See also Posterior cruciate ligament (PCL) reconstruction complications of, 1712 outcomes of, 1711–1712 rehabilitation after, 1711 return to play after, 1712 valgus stress test in, 1692 varus stress test in, 1692 Posterior cruciate ligament (PCL) ­reconstruction in combined injury, 1700, 1706–1710, 1708f–1709f, 1710f complications of, 1712 double-bundle, 1705–1706, 1705f–1706f in combined injury, 1706–1710, 1708f–1710f evaluation of, 1696, 1699–1700, 1699t evaluation of, 1695–1700, 1697t–1698t fixation techniques in, 1700 in knee dislocation, 1755–1756, 1758, 1759f, 1762–1763, 1763f

Index Posterior cruciate ligament (PCL) ­reconstruction (Continued) outcomes of, 1711–1712 posterolateral corner injury and, 1706–1710, 1708f–1710f, 1743–1744 rehabilitation after, 222–224, 223t, 1711. See also Knee rehabilitation return to play after, 1712 single-bundle, 1688, 1705–1706, 1705f, 1707f evaluation of, 1696, 1699–1700, 1699t Posterior drawer test, in posterior cruciate ­ligament injury, 1690, 1690f, 2214 Posterior humeral circumflex artery, ­quadrilateral space compression of, 1142 Posterior interosseous nerve injury, in radial head fracture, 1261–1262 Posterior oblique ligament (POL), 1625, 1627f Posterior sag test, in posterior cruciate ligament injury, 1690, 1691f Posterior tibial tendon. See Tibial tendon, posterior Posterolateral corner (PLC). See also ­Posterolateral corner (PLC) injury anatomy of, 1685, 1718–1723, 1721f–1722f, 1748 biomechanics of, 1687–1688, 1687t, 1723–1725, 1723f dial test of, 1691 stabilizing function of, 1724–1725 Posterolateral corner (PLC) injury, 1718–1747 arthroscopy in, 1730–1731, 1732f biceps femoris in, 1722 in children, 1744 classification of, 1725, 1725t, 2216 clinical presentation of, 1725, 1725t, 1726b dial test in, 1727, 1728f dynamic shift test in, 1728 external rotation recurvatum test in, 1726, 1726f history in, 1725, 1726f incidence of, 1743 magnetic resonance imaging in, 1730, 1730f, 1731f mechanisms of, 1725, 1726f nonoperative treatment of, 1731–1732, 1744 operative treatment of, 1732–1741 biceps tenodesis in, 1733–1734, 1735f complications of, 1742, 1742f cruciate ligament reconstruction with, 1732, 1743 fibular collateral ligament reconstruction in, 1736–1737, 1737f intrasubstance repair in, 1733, 1734f PLC reconstruction in, 1737–1741, 1737t, 1738f–1740f popliteus tendon recession procedure in, 1732–1733, 1733f popliteus tendon reconstruction in, 1736, 1736f rehabilitation after, 1741–1742, 1741f, 1742t return to play after, 1742, 1742b two-tailed reconstruction in, 1734, 1736, 1736f varus correction and, 1744–1747, 1745f–1746f osteoarthritis and, 1725, 1743–1744 physical examination in, 1725–1728, 1726b, 1726f–1729f posterior cruciate ligament injury and, 1689, 1700, 1703–1704, 1744, 1745f double-bundle reconstruction in, 1706–1710, 1708f–1710f proximal tibial osteotomy in, 1744–1747, 1746f

Posterolateral corner (PLC) injury (Continued) posterior tibial translation in, 1728 posterolateral drawer test in, 1727, 1729f radiography in, 1728, 1730f reverse pivot shift test in, 1727–1728, 1729f treatment of, 1731–1741 varus stress test in, 1727, 1727f, 1727t Posterolateral corner (PLC) reconstruction, 1737–1741, 1741b biceps tenodesis for, 1733–1734, 1735f complications of, 1742, 1742f exposure for, 1737–1739, 1738f fixation technique for, 1739–1740, 1739f–1740f graft preparation for, 1739, 1739f in knee dislocation, 1758–1759, 1760f preparation for, 1737, 1737t rehabilitation after, 1741f, 1742t return to play after, 1742, 1742b two-tailed technique of, 1734, 1736, 1736f varus laxity after, 1742, 1742f Posterolateral drawer test, in posterolateral ­corner injury, 1727, 1729f Posteromedial corner, 1625, 1627f Posteromedial pivot test, 1691 Posterosuperior glenoid impingement. See Shoulder impingement, internal Postphlebitic syndrome, 376 Postural exercises, in trunk stabilization, 348–349 Potassium, caffeine effects on, 421 Power, statistical, 2218 Power analysis, 111–112, 116–117 Power nap, 453, 454b Prazosin in complex regional pain syndrome, 363t, 365 in hypertension, 160t Pre-experimental research, 103–104 Precision, measurement, 100 Predictive value, test, 108, 109f, 110t Pregabalin, in complex regional pain syndrome, 363t, 364 Pregnancy, exercise and, 481–483, 482b Preparticipation examination, 508–515, 512b abdominal examination in, 512 athlete history in, 509 cardiovascular examination in, 512 clearance for participation and, 513–515 frequency of, 508 genitourinary examination in, 512 logistics of, 508–509, 509t musculoskeletal examination in, 512–513 neurologic examination in, 513 objectives of, 508, 508t physical examination in, 509–513, 510f–511f, 514t–515t pulmonary examination in, 512 skin examination in, 513 station-based setup for, 508–509, 509t timing of, 508 visual examination in, 509 written health history in, 508–509 Press-up exercise, in shoulder rehabilitation, 241, 242f Pressure, compartment, 650–651, 1860, 1860t. See also Compartment syndrome Prevalence, 102, 102t, 103, 108, 109f, 110t, 2219 Priapism, androstenedione and, 418 Primary surgery, in on-field emergency, 517, 519–520, 520f, 520t, 521f Primidone, in epilepsy, 188 Profunda femoris artery, 1501 Progesterone, exercise effects on, 217t, 218

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xxxix

Prolactin, exercise effects on, 217, 217t Pronator syndrome, 1317–1318 Pronator teres, hypertrophy of, 622 Prone exercises, in trunk stabilization, 342t, 345 Prone horizontal abduction exercise, in ­shoulder rehabilitation, 242–243, 243f Propionibacterium acnes infection, 415, 457 Propranolol in complex regional pain syndrome, 363t in hypertension, 160t Proprioceptive exercises ankle rehabilitation, 273, 275, 298, 340 in female athlete, 479–480 knee rehabilitation, 294–296, 1672, 1674 in ACL rehabilitation, 1674 cone ambulation, 295, 296f cone reaching, 296, 297f double-leg jumping, 296, 298f lunging, 296, 297f plyoball toss, 294, 294f side-to-side weight shifting, 294–295, 295f single-leg hopping, 296, 298f sports cord lunge, 296, 297f shoulder rehabilitation, 296, 298 MR Systems Cable Column, 298, 299f plyoball perturbations, 296, 298f plyoball Rebounder, 298, 299f wall bounce, 298, 298f Prostate cancer, anabolic-androgenic steroid effects on, 416 Protective equipment, in sudden death ­prevention, 170–171 Protein, dietary metabolism of, 211–212 for pediatric athlete, 467 powder, 409 requirements for, 404–405, 405t Protein C, 372, 373f deficiency of, 372, 374t Protein S, 372, 373f deficiency of, 372, 374t Proteoglycans articular cartilage, 41, 42, 43f, 44–45 loss of, 49, 49t bone, 67 ligament, 34 meniscal, 58 structure of, 43f tendon, 23–24 Prothrombin G20210A, 372, 374t Proximal first metatarsal osteotomy, in ­hallux valgus, 2070–2071, 2076b, 2077f, 2078–2079 Proximal row carpectomy, 1449–1450 Pseudoboutonnière deformity, 1389–1390 Pseudohypoparathyroidism, 74t, 77t Pseudoparalysis, in complex regional pain syndrome, 357–358 Psoriasis, 204–205, 204f Psychological factors/disorders in complex regional pain syndrome, 358 in pain dysfunction syndrome, 351, 352t team physician treatment of, 516 Psychological response, rehabilitation-related, 433–436 cognitive component of, 434–437 coping component of, 435–436 disruption and, 435 emotional nourishment and, 435 past coping and, 435–436 perceptual component of, 435 physiological component of, 434 Psychologist. See Sport psychologist Psychomotor vigilance task, 451, 452f

xl

Index

Psychotherapy, in complex regional pain ­syndrome, 365 Puberty. See also Children/adolescents flexibility at, 465 psychosocial aspects of, 466 Pubic rami, stress fracture of, 638, 638f Pubic symphysis, inflammation of, 1466, 1466f Pubis, anatomy of, 1452 Pubofemoral ligament, 1452, 1453f Pudendal artery, 1455f Pudendal nerve, 1454–1455 entrapment of, 1469 Pull-down exercise, in shoulder rehabilitation, 245, 247f Pulmonary contusion, on-field, 526 Pulmonary edema, high-altitude, 504 Pulmonary embolism, 370–385 age-related factors in, 375 evaluation of, 374, 376–378, 376t, 378f, 379f treatment of, 384–385, 384t Pump bump, 1963, 1998 Push-up exercise, in shoulder rehabilitation, 241, 241f, 245, 247f Pushing and pulling exercises, in core training, 287, 288f Pyelography, intravenous, in cervical spine anomalies, 712, 712f

Q Quadratus femoris, 1454t, 1455f Quadriceps active test, in posterior cruciate ligament injury, 1691, 1691f Quadriceps (Q) angle, 1553–1554, 1554f, 1595 Quadriceps muscle anatomy of, 1485, 1485f, 1550–1551, 1551f contraction of, for patellar evaluation, 1564 contusion of, 13, 1459, 1481–1484, 1482b in adolescents, 1484 biomechanics of, 1481 classification of, 1481–1482, 1481t clinical presentation of, 1482 complications of, 1483–1484, 1484b, 1484f imaging of, 1482 myositis ossificans with, 1483, 1484b, 1484f physical examination in, 1482 return to play after, 1484, 1484b treatment of, 1482–1484, 1482b, 1482f isometric strengthening of, in ACL-injured knee, 1586 myositis ossificans of, 1483, 1484b, 1484f, 1495–1496 neuromuscular activation exercise for, in knee rehabilitation, 256–257, 257f strain of, 1462, 1493–1497 in adolescents, 1497 clinical presentation of, 1493–1494, 1494f complications of, 1495–1496 grade of, 1494 imaging of, 1494–1495, 1494f–1495f physical examination of, 1494, 1494b, 1494f rehabilitation protocol for, 1496, 1496t return to play after, 1496, 1497b treatment of, 1495–1496, 1495b, 1496t strengthening of in ACL rehabilitation, 1671 in arthritis, 1774 Quadriceps tendon anatomy of, 1514 magnetic resonance imaging of, 560 repair of, 1524–1525, 1524f–1525f rupture of, 1521–1522, 1522t

Quadriceps tendon (Continued) complications of, 1525 evaluation of, 1522–1523, 1522f treatment of, 1523–1524, 1524f–1525f Quadrilateral space syndrome, 1142 Quadriplegia. See Cervical cord neurapraxia/ quadriplegia Quadruped progression exercise, 282–283 alternating arm and leg, 282–283, 283f drawing-in maneuver in, 282 hip extension, 282, 283f in trunk stabilization, 342t, 346 upper extremity lift, 282, 282f Qualification to play, 507. See also ­Preparticipation examination Qualitative research, 106–107 Quantitative sudomotor axon reflex test, 360–361 Quercetin, 409 Quinapril, 160t Quinine sulfate, in cramps, 12

R Raccoon eyes, 525 Radial artery, 1160f Radial collateral ligament anatomy of, 1301–1302, 1303f stabilizing effect of, 1193, 1195f, 1231, 1231f tear of, 576, 577f Radial head articular surface of, 1190, 1190f fracture of, 1258–1262 capitellar fracture with, 1259 classification of, 1260 coronoid process fracture with, 1259 evaluation of, 1258–1260, 1260f fragments in, 1260, 1260f magnetic resonance imaging in, 535f medial collateral ligament rupture with, 1259 nonunion of, 1278 operative treatment of, 1260–1262, 1275–1276 complications of, 1262 exposure in, 1260–1261, 1261f failure of, 1258, 1259f–1260f fixation in, 1261 prosthetic replacement in, 1261–1262 pediatric, 1286–1287 posterior dislocation and, 1259, 1261, 1263, 1264f, 1269 resection in, 1260, 1262 prosthetic, 1258, 1260f, 1261–1262 stabilizing effect of, 1192–1193, 1193f Radial neck, fracture of, 1280t, 1287–1288 Radial nerve anatomy of, 1159, 1161, 1161f injury to, 1315–1316. See also Radial tunnel syndrome in humeral shaft fracture, 1182 in lateral epicondylitis, 1199 vs. lateral epicondylitis, 618 in pediatric supracondylar fracture, 1282 Radial osteotomy, in Kienböck’s disease, 1377 Radial shortening, in Kienböck’s disease, 1377 Radial styloidectomy, 1449 Radial tunnel syndrome, 1315–1316 anatomy of, 1315–1316 etiology of, 1315 history in, 1316 nonoperative treatment of, 1316 operative treatment of, 1316 physical examination in, 1316

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Radiation, heat loss by, 493, 499 Radiation exposure, in children, 588 Radiation therapy, in heterotopic ossification, 1277, 1294–1295, 1297 Radiocarpal ligaments, 1320 Radiofrequency ablation, in Wolff-ParkinsonWhite syndrome, 170 Radiofrequency sympathectomy, in complex regional pain syndrome, 366 Radiography, 533–534 in accessory navicular, 1963, 1963f in acetabular labral tears, 1470 in Achilles tendon injury, 1998, 2003, 2003f in acromioclavicular joint injury, 831b, 833f, 834f, 835–836, 836f–838f in acromion fracture, 861, 861f, 866, 866f–867f in ankle dislocation, 1945, 1946f in anterior cruciate ligament injury, 1650 in anterior glenohumeral joint instability, 934, 935f in anterior tibial stress fracture, 643, 644f in atlanto-occipital instability, 705, 705f in atlantoaxial instability, 705, 707f of atlas-dens interval, 694, 695f in atraumatic glenohumeral joint instability, 945 in avulsion fracture, 553, 556f in Bennett’s fracture, 1402, 1402f in bifurcate sprain, 1954, 1954f in bone cyst, 552, 553f in Buschke’s disease, 2166, 2166f in C1 posterior arch absence, 711, 711f in C3-C4 facet dislocation, 679, 679f in C3-C4 subluxation, 678–679, 679f in C5 fracture, 697, 697f in calcaneal apophysitis, 2162, 2162f in calcaneal stress fracture, 646, 647f in capitate fracture, 1348–1349 in capitellar osteochondritis dissecans, 623, 624f, 1241, 1242f–1245f in capsulorrhaphy arthropathy, 1106f in carpometacarpal joint dislocation, 533, 534f, 1387, 1387f in cervical fusion, 699, 699f, 700f in cervical spine compression fracture, 680–681, 680f, 681f–683f, 709, 709f in chondral defect, 1773 in complex regional pain syndrome, 358 in congenital lumbar stenosis, 747, 747f in coracoid fracture, 861, 862f, 878–879, 878f–879f, 880 in distal clavicular osteolysis, 855f in distal femoral stress fracture, 555f in distal fibular stress fracture, 644, 644f in distal humeral fracture, 1250 in distal radioulnar joint injury, 1374 in elbow heterotopic ossification, 1292–1293, 1292f, 1298f in elbow joint effusion, 534, 535f in extensor carpi ulnaris tendinopathy, 1354 in femoral neck stress fracture, 638, 640f–642f, 1465 in femoral shaft stress fracture, 1478t, 1479, 1479f–1480f in femoroacetabular impingement, 1471 in fifth metatarsal stress fracture, 648, 649f in fracture, 552–553 in Freiberg’s infraction, 1973, 1973f, 2166, 2167f in ganglion cyst, 1444 of glenohumeral joint, 947–949, 947b, 948f in glenohumeral joint instability, 915–916 in glenohumeral joint osteoarthritis, 1105f, 1108, 1108f

Index Radiography (Continued) in glenohumeral rheumatoid arthritis, 1106, 1107f in glenoid fracture, 861, 861f in glenoid rim fracture, 868f–870f in gymnast wrist, 1375 in hallux rigidus, 2182, 2182f in hallux valgus, 2069, 2069b, 2073f, 2075f in hamate fracture, 1346f in high (syndesmosis) ankle sprain, 1940–1942 in hip degenerative disease, 1503 in humeral cyst–related fracture, 552, 553f in humeral epicondyle fracture, 1286f in humeral head avascular necrosis, 1107, 1107f in humeral shaft fracture, 1178, 1179f–1181f in iliac spine avulsion fracture, 556f in ingrown toenail, 2098 in Iselin’s disease, 2167, 2168f–2169f in Jefferson fracture, 705, 706f in Jersey finger, 1425, 1425f in Kienböck’s disease, 1377, 1377b, 1378f in Klippel-Feil syndrome, 693, 693f, 711–712, 712f in knee arthritis, 1789–1790, 1790f in knee dislocation, 1753, 1752f–1753f in Köhler’s disease, 1973, 1973f, 2162, 2163f in lateral ankle sprain, 1918–1919, 1918f–1919f in lateral epicondylitis, 1199 in Lisfranc sprain, 1956–1957, 1957f, 2181f in lumbar degenerative spondylolisthesis, 748 in lumbar isthmic spondylolisthesis, 748, 749f of lumbar spine, 724–726, 725f in lumbar spine burst fracture, 735, 735f in lumbar spine stenosis, 746f in lunate fracture, 1353f in lunotriquetral ligament injury, 1331 in mallet finger, 1420f in medial ankle sprain, 1936, 1936f–1937f in medial collateral ligament injury, 1629, 1630f in medial malleolar stress fracture, 644–645, 645f in meniscal deficiency, 1619 in meniscal injury, 1602 in metacarpophalangeal joint dislocation, 1380, 1380f–1381f in metatarsal stress fracture, 648, 648f in myositis ossificans, 13, 14f in olecranon bursitis, 1210f, 1247, 1247f in olecranon fracture, 1271–1272 in olecranon stress fracture, 1225, 1225f in os acromiale, 859, 860f in os odontoideum, 692, 692f in Osgood-Schlatter disease, 1528 in osteitis pubis, 1466, 1490, 1491f in overhead throwing injury, 1217 in Panner’s disease, 623, 625f in pars interarticularis defects, 760, 760f in Parsonage-Turner syndrome, 1145 in patellar dislocation, 1538–1540, 1541f, 1558–1564, 1559f–1563f, 1560t–1561t, 1564t–1566t in patellar fracture, 1574, 1574f in patellar tendon rupture, 1523, 1523f in pectoralis major rupture, 1060 pediatric, 590 in ankle fracture, 1964 in avulsion fracture, 599 bone on, 591–592, 592f of cervical spine, 702, 703f of clavicle, 781f

Radiography (Continued) in clavicular fracture, 596, 596f in distal tibial epiphyseal fracture, 1641, 1642f, 1643f in elbow injury, 1229–1230, 1230f, 1279–1280 in Ewing’s sarcoma, 607, 609f, 610 in fibrous dysplasia, 606, 607f, 608f in fracture, 593–599, 594f–598f in hemangioma, 606 of hip, 587, 588f in inflammatory arthritis, 602–603, 602f in juvenile idiopathic arthritis, 602–603, 602f in Little Leaguer’s elbow, 1183, 1184f–1185f in Little Leaguer’s shoulder, 1173, 1174f in nonossifying fibroma, 605–606, 605f in osteochondroses, 599, 600f in osteomyelitis, 601, 602f in osteosarcoma, 606–607, 608f of physis, 591–592, 592f in posterior cruciate ligament injury, 1716 in proximal humeral physeal fracture, 1074–1075, 1075f–1076f in proximal humeral physeal stress ­fracture, 1092, 1093f of proximal humerus, 782, 782f in proximal tibial epiphyseal fracture, 1642, 1643f radiation dose in, 588 in slipped capital femoral epiphysis, 597, 598f, 1476, 1476f in supracondylar fracture, 596–597, 597f in thoracolumbar spine injury, 759–761, 760f in unfused os acromiale, 781, 781f in unicameral bone cyst, 606, 606f in Pellegrini-Stieda lesion, 1629, 1630f in perilunate dislocation, 1332 in peroneal tendinitis, 1989 in peroneal tendon subluxation, 1992 physics of, 533–534 in pisiform fracture, 1350, 1352f in plantar fasciitis, 2046–2047, 2047f in plantar keratoses, 2110, 2110f in posterior cruciate ligament injury, 1692–1693 in posterior glenohumeral joint instability, 937, 941–942 in posterior sternoclavicular joint dislocation, 799f in posterior tibial tendinitis, 1981 in posterior tibial tendon injury, 1979 in posterolateral corner injury, 1728, 1730f in proximal humeral fracture, 1037–1038, 1038b, 1039f–1042f in proximal metaphyseal humeral fracture, 1087 in proximal ulna fracture, 1271–1272 in quadriceps myositis ossificans, 1483, 1484f in quadriceps tendon rupture, 1523, 1523f in radial head fracture, 1259–1260, 1259f in radial osteomyelitis, 83f in recurrent patellar dislocation, 1558–1564, 1559f–1563f, 1560t–1561t, 1564t–1566t in recurrent posterior glenohumeral ­instability, 941–942 in retrocalcaneal bursitis, 2032–2034, 2032f, 2033f in rheumatoid arthritis of wrist, 1444 in rib fracture, 637, 637f, 894, 1187 in rotary atlantoaxial subluxation, 706–707, 708f in rotator cuff disorders, 615, 954–955, 955f, 998, 1000

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xli

Radiography (Continued) in scaphoid fracture, 554f, 1333f, 1335f–1336f, 1336, 1365 in scapholunate ligament injury, 1325 in scapular fracture, 861, 861f–862f in scapular osteochondroma, 887, 888f in sesamoid dysfunction, 2090, 2091f–2092f in sesamoid osteochondrosis, 2169, 2170f in sesamoid stress fracture, 649, 649f in Sever’s disease, 1973, 1974f in shoulder impingement syndrome, 954–955, 955f in Sinding-Larsen-Johansson disease, 1529 in snapping hip syndrome, 1458 of soft tissues, 534, 535f in spear tackler’s spine, 694, 695f in spinal deformity, 737 in spinal lymphoma, 758f in spondylolysis, 635, 636f, 725, 726f, 762 in sternoclavicular joint injury, 798f, 802–805, 803f–804f in sternomanubrial dislocation, 898–899, 899f in stress fracture, 552–553, 633, 634t in subtalar dislocation, 1952–1953 in subtalar sprain, 1949–1950, 1949f in subungual exostosis, 2106, 2106f in supracondylar process fracture, 1186, 1186f in suprascapular nerve injury, 1121, 1122f in talar osteochondral lesions, 2144–2145, 2145t, 2147f in tarsal coalition, 1961, 1961f–1962f in tarsal navicular stress fracture, 646, 646f in thoracic compression fracture, 755f in thoracic outlet syndrome, 1133, 1133f of thoracic spine, 724, 724f in tibial fracture nonunion, 82f–83f in tibial stress fracture, 1852, 1852f, 2014, 2015f–2016f in trapezium fracture, 1347–1348, 1349f in triangular fibrocartilage complex tear, 1436 in triceps tendinitis, 1207 in triceps tendon rupture, 1170–1171, 1171f in triquetrum fracture, 1351f in turf toe, 2083–2084, 2084f, 2086f in ulnar collateral ligament injury, 1400, 1400f in ulnar neuropathy, 623 in ulnar styloid impaction syndrome, 1446, 1446f in ulnocarpal impaction syndrome, 1440 in valgus instability, 620 in varus malalignment, 1809–1814, 1812f, 1813f in vertebral body fracture, 698f in wrist injury, 1321, 1322, 1322f–1323f in wrist loose bodies, 1447 Radiolabeled leukocyte imaging, in ­osteomyelitis, 547 Radiolunate ligament, 1320 Radionuclide imaging, 543–547 in bone tumors, 547, 548f in cellulitis, 547 in complex regional pain syndrome, 359 in femoral neck stress fracture, 639, 641f, 642f in femoral shaft stress fracture, 1479, 1480f gamma camera for, 544 in high (syndesmosis) ankle sprain, 1942 in infection, 547 of lumbar spine, 726–727, 726f in metastatic disease, 547, 547f normal, 545f, 591f

xlii

Index

Radionuclide imaging (Continued) in occult fracture, 546, 552, 557 in osteitis pubis, 1466, 1466f in osteomyelitis, 547, 601, 602f in osteoporotic fracture, 546, 547f in pars interarticularis stress fracture, 726, 726f pediatric, 591, 591f in elbow injury, 1230 in osteomyelitis, 601, 602f in plantar fasciitis, 2046–2047, 2047f in posterior cruciate ligament injury, 1693 in pubic rami stress fracture, 638, 638f in quadriceps myositis ossificans, 1483 radiopharmaceuticals for, 544 in sacral stress fracture, 736, 736f in shin splints, 545, 545f in stress fracture, 552, 633, 634t in tarsal navicular stress fracture, 646, 646f in thoracic compression fracture, 734, 734f of thoracic spine, 726–727, 726f three-phase protocol for, 544 in tibial stress fracture, 545–546, 545f, 1852, 1853f, 2014–2015, 2015f in trauma, 545–546, 545f–547f in turf toe, 2084, 2084f whole-body, 544, 545f, 591, 591f in wrist injury, 1321 Radioscaphocapitate ligament, 1320, 1432, 1432f Radioscapholunate ligament, 1320, 1432, 1432f Radioulnar joint distal, injury to, in children, 1374–1375, 1374b proximal, heterotopic ossification at, 1297 Radius distal arthritis of, after scaphoid fracture, 1339–1340 physeal injury of, in gymnast, 1375–1376 vascular supply of, 1339 fracture of, in children, 594f, 595, 597 osteomyelitis of, 83f proximal. See also Radial head ossification of, 1228 Ramipril, 160t Randomized controlled trials, 101 Range of motion of acromioclavicular joint, 828, 828f in anterior cruciate ligament rehabilitation, 1670–1671 of hip, 1452t, 1503 of patella, 1557 after proximal humeral fracture, 1050, 1056 in rotator cuff disorders, 996, 998f of sternoclavicular joint, 794, 796f in thoracolumbar spine, 721, 759 in wrist injury, 1320 Rapid strep test, 150 Rasmussen’s syndrome, 190 Rate, 102, 102t false-negative/false-positive, 108, 109f Ratio method, in cervical spine stenosis ­evaluation, 683–686, 683f–685f Rebounder, in shoulder rehabilitation, 296, 299f Recreational drugs, 424–431. See also specific drugs Rectus femoris, 1454t, 1455f, 1481, 1550–1551. See also Quadriceps muscle Rectus sheath hematoma, 526 Red blood cells erythropoietin effects on, 420–421 high-altitude effects on, 503 Reduction-association of scapholunate (RASL) procedure, 1325, 1326f, 1327

Referred pain, to heel, 2047 Reflex(es) bulbocavernous, 675 in thoracolumbar spine injury, 722–723 Reflex sympathetic dystrophy. See Complex regional pain syndrome Regional intravenous sympathetic blockade, in complex regional pain syndrome, 360 Rehabilitation. See also at specific joints and disorders articular cartilage protection in, 225–228, 226f biofeedback in, 224–225, 233, 234f cryotherapy in, 234–235, 235f electrical currents in, 229–233 for functional restoration, 230–233, 231f–234f for pain modulation, 229–230, 230f for swelling, 230, 231f exercise prescription in, 238–293. See also Therapeutic exercise(s) iontophoresis in, 233–234, 235f in joint stiffness, 225, 225f laser in, 235–236, 236f in muscle atrophy, 224–225 psychological response in, 433–436 cognitive component of, 434–437 coping component of, 435–436 perceptual component of, 435 physiological component of, 434 sport psychologist in, 437–440 ultrasound in, 236–237 Rehabilitation cycle, 728, 728f Relative incidence rate, 102, 102t Relative motion, 89–90 Relative rate, 102, 102t Relative risk, 2218 Reliability measurement, 101f statistical, 100, 2217–2218 Relocation test in glenohumeral joint instability, 914, 915f, 939, 939f in glenohumeral joint osteoarthritis, 1108 in rotator cuff disorders, 996, 999f Renal osteodystrophy, 73t, 75t, 77t Repetitive trauma. See Overuse injury Research, 97–99, 98f. See also Statistics analytical, 107–108 clinical, 99–100 controls in, 101–102, 104–105 experimental, 103–106, 105f, 106f–107f hypothesis testing in, 110–112 philosophical, 108 pre-experimental, 103–104 qualitative, 106–107 question development for, 97 rate measures in, 102, 102t risk measurement in, 103 study design for, 99, 101–110, 101f–109f analytical, 107–108 blinding in, 104–105 control group in, 101–102, 104 experimental, 101f, 103–106, 105f matching in, 101 observational, 101–103, 101f, 103t, 106–107 power analysis in, 111 pre-experimental, 103–104 qualitative, 106–107 single-subject, 105–106, 105f–107f subjects for, 97 team for, 97 variables in, 97, 99

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Reserpine in complex regional pain syndrome, 365 in hypertension, 160t Respiratory distress, 184–185, 184b. See also Bronchospasm Respiratory system cocaine effects on, 429 exercise effect on, 220 infection of, 149–150 marijuana effects on, 426 Resuscitation, in sudden death, 171–172, 171t Retinaculum patellar lateral, 1548, 1548f, 1551–1552 release of, 1545, 1567–1568 medial, 1548, 1548f, 1551–1552, 1551f release of, 1568 repair of, 1545–1546 peroneal inferior, 1987 superior, 1931, 1987 congenital absence of, 1991 disruption of, 1932 imbrication of, 1932–1933 reconstruction of, 1995 Retinoids, in acne, 205 Retrocalcaneal bursa, 2031, 2031f–2032f aspiration of, 2033 palpation of, 2032 Retrocalcaneal bursitis, 2030–2042 Achilles tendinosis with, 2030–2031, 2034, 2034f, 2036 anatomy of, 2031, 2032f biomechanics of, 2031 brisement therapy in, 2035 calcaneal exostosectomy in, 2037 endoscopic débridement in, 2037 evaluation of, 2032, 2032f, 2035f, 2036 Haglund’s deformity and, 2030, 2031f, 2038, 2038f nonoperative treatment of, 2035, 2037–2038, 2038f operative treatment of, 2035–2041, 2037f, 2038f–2040f orthotic devices in, 2035 os calcis in, 2031, 2031f–2032f parallel pitch line measurement in, 2036 radiography in, 2032–2034, 2033f–2034f return to play after, 2041 shock-wave therapy in, 2035 vs. soleus muscle anomaly, 2036 treatment of, 2035–2041 Retroclavicular Spurling’s test, 1132–1133, 1133f Reversal research design, 105, 105f Reverse pivot shift test in posterior cruciate ligament injury, 1692, 1692f in posterolateral corner injury, 1727–1728, 1729f in varus malalignment, 1807–1808 Reverse straight leg raising test, in ­thoracolumbar spine injury, 722 Rewarming therapy, 501, 528, 529 Rhabdomyolysis exertional, 496–497 magnetic resonance imaging in, 558 Rheumatoid arthritis glenohumeral joint, 1105t, 1106, 1107f arthroplasty in, 1114 arthroscopic treatment of, 1110–1111 wrist proximal row carpectomy in, 1449–1450 radial styloidectomy in, 1449 synovectomy in, 1443–1444

Index Rhomboid fossa, 793, 793f Rhomboid ligament, 793, 793f Rhomboids, anatomy of, 786, 786f, 859f, 886f Rhythmic stabilization training in core training, 286 in elbow rehabilitation, 255, 255f Rib(s) first, stress fracture of, 636–637, 637f floating, avulsion fracture of, 895–896, 896f fracture of, 893–896, 893t, 1187 avulsion, 895–896, 896f in children, 894 on-field, 525 return to play after, 894–895, 895f, 895t, 1187 stress, 895–896, 895t, 896f traumatic, 893–895, 895t middle, stress fracture of, 637 Riboflavin, requirements for, 406b Rickets, 72–73, 73t, 75t–77t, 79b Ring test, for CSF, 525 Ringworm, 198, 199f Risk, 102–103, 102t Roeder’s knot, 135, 135f Rolando fracture, 1403 Rolling side bridge exercise, in core training, 284, 284f Romanian deadlift exercise, 267–268, 269f single-leg, 268, 269f Roos’ test, in thoracic outlet syndrome, 1132, 1133f Rotary atlantoaxial subluxation, in children/ adolescents, 706–707, 708f Rotation training, in core training, 286–288, 288f–289f Rotator cuff, 986–1015. See also Infraspinatus; Subscapularis; Supraspinatus; Teres minor age-related changes in, 974–975, 975f anatomy of, 989–990, 989f–991f pediatric, 783, 784–785, 785f, 789–790, 789f arthrography of, 958 biomechanics of, 990–994, 992f–993f cyst of, 964–965, 965f, 980, 980t denervation of, 965, 965f, 975 disorders of. See Rotator cuff disorders; ­Rotator cuff tear(s) fatty atrophy of, 963, 963f, 965, 980 intramuscular cyst of, 964–965, 964f magnetic resonance imaging of, 564–566, 953, 953t, 958–965, 960f–964f, 960t pseudorupture of, 860 radiography of, 955f, 957 in shoulder rehabilitation, 241–243, 242f, 243f, 250t strengthening exercises for, 242–244, 242f, 243f, 244f, 1003–1006, 1004f–1007f, 1007 stretching exercises for, 1003, 1003f, 1007 ultrasonography of, 951–953, 952f, 953t vascular anatomy of, 989–990 Rotator cuff disorders. See also Rotator cuff tear(s) biomechanics of, 990–994, 992f denervation, 964–965, 965f, 975 disability with, 995, 1002 etiology of, 1001–1002 evaluation of, 994–1001 apprehension test in, 996, 999f arthrography in, 949–950, 949f, 958, 1000 arthroscopic, 1009–1010 chief complaint in, 994–995 Hawkins’ sign in, 997, 1000f impingement test in, 998 inspection in, 995, 996f

Rotator cuff disorders (Continued) load and shift test in, 996, 999f magnetic resonance imaging in, 958–965, 960f–965f, 1001, 1001f Neer’s sign in, 997, 1000f palpation in, 996, 997f patient history in, 995 physical examination in, 995–998, 996f–1000f radiography in, 955f, 957–958, 998, 1000 range of motion in, 996, 998f relocation test in, 996, 999f Speed’s test in, 997, 1000f ultrasonography in, 1000–1001 Yergason’s test in, 997, 1000f historical perspective on, 986, 986f nonoperative treatment of, 1006–1010, 1011 activity modification in, 1006 corticosteroid injection in, 1007 pharmacological, 1006–1007 strengthening exercises in, 242–244, 242f–244f, 1003–1006, 1004f–1006f, 1007, 1008f stretching exercises in, 1003, 1003f, 1007 ultrasound in, 1007 operative treatment of, 1008–1031. See also at Rotator cuff tear(s) arthroscopic, 1009–1010, 1012–1014, 1013f–1015f open, 1008–1009, 1014–1015 overuse-related, 612, 614–615, 615b, 1217–1219 peritendinous, 30 prevention of, 1003–1006, 1003f, 1004f–1006f severity of, 1002–1003 SLAP lesion and, 1024. See also SLAP (­superior labrum, anterior to posterior) lesion Rotator cuff tear(s). See also Rotator cuff ­disorders aging and, 988 arthrography in, 949–950, 949f, 958 arthroscopic repair of, 1012–1014 arm positioning for, 1012 bursectomy with, 1013 care after, 1015 double-row technique for, 988, 988f, 1013, 1014f–1015f evaluation for, 1013 portals for, 1012–1013, 1013f subacromial decompression with, 1013, 1014f epidemiology of, 986–988 evaluation of. See Rotator cuff disorders, evaluation of fatty infiltration with, 770, 989, 989f full-thickness, 994 arthroscopic repair of, 1012–1014, 1013f–1015f magnetic resonance imaging in, 960–963, 960t, 961f–962f open repair of, 1014–1015 overhead throwing–related, 1218 postoperative care in, 1015 glenohumeral joint force in, 775 graft repair of, 989 historical perspective on, 986, 986f magnetic resonance arthrography in, 565–566, 567f natural history of, 988 in older patient, 974–975, 975f, 988 overhead throwing–related, 1218–1219 partial-thickness, 994 magnetic resonance imaging in, 960–961, 960t, 961f–962f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xliii

Rotator cuff tear(s) (Continued) overhead throwing–related, 1218–1219 treatment of, 1012 radiography of, 954–955, 955f, 957 recurrence of, 565–566, 567f, 988 treatment of, 615, 987–989, 988f, 1010–1015 nonoperative, 991, 1010 operative, 1011–1015, 1013f–1015f tendon transfer/graft in, 989 Rotator interval anatomy of, 773–774, 966, 966t, 990, 991f function of, 966, 966t magnetic resonance imaging of, 966, 966t, 967f tear of, 966, 967f Roux-Elmslie-Trillat procedure, 1595–1596 Rowing exercises on cable column, 240, 240f in core training, 286, 286f, 287, 288f with elastic resistance, 240, 240f prone, 240, 240f shoulder injury and, 249 in shoulder rehabilitation, 245, 247f in trapezius training, 240, 240f Running shoe. See Shoes Running Shoe Book, The (Cavanagh), 1875 Russian hamstring curl, 317f, 325

S Sacroiliac ligaments, sprain of, 1462 Sacrum anatomy of, 1451–1452 stress fracture of, 638, 736, 736f, 1465 Sagebrush bark sandal, 1873, 1874f Sagittal band bridge, 1390 Sagittal band rupture, 1390 SAID (specific adaptations to imposed ­demands) principle, 213–214 Sail sign, 706, 708f Salty sweater, 402 Sample/sampling, 99, 2218 consecutive, 99 convenience, 99 error and, 111 judgmental, 99 nonprobability, 99 nonrandom, 99 random, 99 Saphenous nerve, entrapment of, 1497–1499, 1497b, 1498f Saphenous vein graft, in popliteal artery ­entrapment, 1844–1845, 1845f, 1846b, 1846t, 1847 Sarcoidosis, 77t heel pain in, 2051 Sarcoma, radiation-induced, 1295 Sarcomere, 5, 207–208, 208f–209f Sarcoplasmic reticulum, 5, 7f Sartorius, 1454t, 1455f Satellite cell, 5 Scabies, 201, 201f Scalars, 86 Scalp in head injury, 658 pediculosis capitis of, 200–201, 201f Scaphoid. See also Scaphoid fracture anatomy of, 1319–1320, 1319f bipartite, 1366 humpback deformity of, 1340–1341 Scaphoid fracture adult, 1335–1340 classification of, 1335, 1335f, 1336t clinical manifestations of, 1335–1336 displaced, 1335, 1337, 1339

xliv

Index

Scaphoid fracture (Continued) magnetic resonance imaging in, 552, 554f nondisplaced, 1335, 1336–1337, 1338f nonunion of, 1339–1345, 1341f–1344f perilunate dislocation with, 1332, 1333f physical examination in, 1336 radiography in, 1336, 1336f return to play after, 1337, 1339 treatment of, 1336–1339 pediatric, 1364–1368 delayed union of, 1366, 1367f epidemiology of, 1364 evaluation of, 1364–1365 malunion of, 1368 nonunion of, 1367–1368 radial physeal injury with, 1365 stress-type, 1365–1366, 1366f treatment of, 1366 types of, 1365, 1365f Scaphoid graft, in scaphoid nonunion, 1341–1345, 1342f–1344f Scapholunate advanced collapse (SLAC) wrist, 1327–1330 stage I, 1328, 1328t stage II, 1328, 1328t, 1329f stage III, 1328–1329, 1328t, 1329f–1330f Scapholunate angle, 1322, 1322f Scapholunate dissociation, 1371 Scapholunate interosseous ligament, 1320 Scapholunate ligament, 1432–1433, 1432f injury to (Mayfield I), 1324–1330 acute, 1324, 1326 chronic, 1324, 1327 classification of, 1324 clinical manifestations of, 1324 magnetic resonance imaging in, 1325 pediatric, 1371–1374, 1374t physical examination in, 1324–1325, 1325f radiography in, 1323f, 1325 return to play after, 1326–1327 salvage procedures in, 1327–1330, 1328t, 1329f–1330f subacute, 1324 treatment of, 1325–1327, 1326f–1327f magnetic resonance imaging of, 575, 576f Scapula anatomy of, 858–859, 858f–859f pediatric, 781–782, 781f bursae about, 886, 887b, 887f dyskinesia of, 1007 fracture of. See Scapular fracture muscle attachments of, 858, 858f–859f, 885, 885b, 886f neck of, 782 neurovascular anatomy at, 858–859 osteochondroma of, 887, 888f in shoulder rehabilitation, 240 spine of, 782 strengthening exercises for, 1007, 1008f superomedial angle of, resection of, 888–889, 889f winging of, 616, 1124–1125, 1125f crepitus with, 888 differential diagnosis of, 1125 Scapular exercises protraction, 241, 241f, 250, 251t retraction, 245–246, 247f, 250, 251t strengthening, 1007, 1008f Scapular fracture. See also Acromion, fracture of; Coracoid, fracture of; Glenoid, fracture of of body, 863, 865–866 in children, 872, 875 classification of, 857, 858f evaluation of, 860–861 patient history in, 861

Scapular fracture (Continued) physical examination in, 860–861 radiography in, 861, 861f–862f incidence of, 857 malunion of, 865 mechanism of, 857, 865–866 nonunion of, 865 suprascapular nerve injury with, 1121 Scapular lag, 994, 1001 Scapular retraction test, 892 Scapulothoracic joint, 769, 885–893 anatomy of, 885–886, 885b, 886f pediatric, 784 biomechanics of, 776–777, 776f, 885–886 in children, 787–788, 788f bursitis of, 889–891, 891f crepitus of, 886–889, 887b, 888f–889f dissociation of, 1139 dyskinesia of, 891–892 endoscopy for, 890–891, 891f kinematics of, 776–777, 776f in shoulder impingement, 777 in shoulder rehabilitation, 239–240 Scapulothoracic rhythm, 776 Scheuermann’s disease, 737, 765, 766f Sciatic nerve, 1454, 1501, 1748–1749 entrapment of, 1468 vs. ischial bursitis, 1457 Scintigraphy. See Radionuclide imaging Scoliosis, 736–737 adult, 737 classification of, 737 degenerative, 737 idiopathic, 479, 757, 757f Scottie dogs, 725, 726f, 760, 760f Screening, 109–110 cardiovascular, 167–169, 168t Scrotum, on-field injury to, 527 Scrumpox, 197 Scurvy, 72–73, 73t Seasonal affective disorder, 444 bright light treatment of, 444–445, 445f Seasonal rhythms, 444 Second impact syndrome, 659 Secondary surgery, in on-field emergency, 520, 521f Segond’s fracture, 1650, 1722 Seizures. See also Epilepsy exercise effects on, 187–188 head injury and, 660–661 management of, 190 on-field, 522–523 terminology for, 186–187, 186b Semimembranosus, 1485, 1485f, 1625, 1625f–1627f. See also Hamstring muscles biomechanics of, 1626, 1628f Semitendinosus, 1485, 1485f. See also ­Hamstring muscles Semitendinosus tendon autograft, in ligament injury treatment, 37–38 Sensitivity, test, 108, 109f, 110, 110t, 2218 Sensory system in complex regional pain syndrome, 356 in thoracolumbar spine injury, 722, 722f Septal ablation, in hypertrophic ­cardiomyopathy, 170 Seronegative spondyloarthritis, heel pain in, 2033–2034, 2047 Serratia marcescens infection, 398, 398t Serratus anterior palsy of, 1130–1131, 1131f pediatric, 786 scapular attachment of, 858, 858f, 886f therapeutic exercise for, 240–241, 241f–242f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Sesamoid(s) anatomy of, 2087–2088, 2088f, 2169, 2170f arthritis of, 2089 bipartite, 2028, 2029f, 2088 dysfunction of, 2087–2096 classification of, 2088–2089 clinical presentation of, 2089 history in, 2089 imaging in, 2090, 2091f–2092f nonoperative treatment of, 2090, 2092b, 2093f operative treatment of, 2090–2095, 2092b, 2093b, 2094b, 2094f–2095f care after, 2096 complications of, 2096 return to play after, 2096, 2096b physical examination in, 2090, 2090b excision of, 2090–2095, 2093b–2094b, 2093f–2094f fracture of, 2088 nonunion of, 2093 function of, 2087f, 2088 keratosis beneath, 2114 nerve compression at, 2088–2089, 2089f osteochondritis of, 2089–2090, 2092f, 2169–2170, 2170f pain in, 2088–2089, 2088b shaving of, 2092–2095, 2094b, 2095f, 2096 stress fracture of, 648–650, 649f, 2016t, 2027–2028, 2030f vascular supply of, 2087–2088, 2088b Sesamoid ligaments, 2064–2065, 2065f, 2087–2088 Sesamoidectomy, 2090–2095, 2093b, 2093f, 2094b, 2094f care after, 2096 complications of, 2093–2094, 2096 Sesamoiditis, 2089 Sever’s disease, 1973–1974, 1974f, 2053–2054, 2143f, 2162, 2162f Shear force, 94f, 95 Shin splints, 15, 652, 652t, 1857. See also Chronic exertional compartment ­syndrome radionuclide imaging of, 545, 545f Shock-wave therapy in Achilles tendon injury, 1999 in lateral epicondylitis, 1200 in plantar fasciitis, 2049–2050, 2052 in retrocalcaneal bursitis, 2035 Shoes, 1873–1911 Achilles tendon protector of, 1889 advertising and, 1873, 1873f, 1875 air encapsulation in, 1892 alignment features of, 1900–1905, 1901f biomechanical aspects of, 1896–1906 alignment and control in, 1900–1905, 1901f, 1903f, 1904t–1905t energy return in, 1905–1906 friction and torque in, 1906 shock absorption and, 1897–1899, 1898t, 1899f, 2185–2187, 2185t, 2186f for cavus foot, 1900 cleats on, 2184, 2191–2192, 2195f–2196f collar of, 1889 components of, 1877–1878, 1878f, 1886–1887, 1887f, 1888f bottom, 1889–1890 glossary for, 1879–1886 upper, 1887–1889, 1888f–1889f copolymers for, 1895–1896, 1896t cork for, 1894 energy return from, 1905–1906 ethylene vinyl acetate for, 1892 eyelet stay of, 1888

Index Shoes (Continued) eyelets of, 1888, 1889f for female athlete, 490–491, 491b, 1886–1887 fit of, 1906–1911 flexibility test in, 1909–1910, 1910f guidelines for, 1910, 1910t individual variables in, 1908, 1908f, 1909 injury and, 2184–2185, 2184f kick test in, 1909 length in, 1909, 1909f manufacturing process and, 1908 measurements for, 1908, 1908f pinch test in, 1909, 1909f shape in, 1910 sizing for, 1906–1907, 1906t, 1907f flared heel for, 1902 flexibility of, 2188–2189, 2189f forefoot stabilizers of, 1889 gel encapsulation in, 1892 heel counter of, 1889, 1892 historical perspective on, 1873–1876, 1873f–1877f hyperpronation and, 2183 injury and, 2183–2192 cleats and, 2184, 2191–2192, 2195f–2196f control/support and, 2187–2189, 2187f–2189f cushioning and, 2185–2187, 2185t, 2186f historical perspective on, 2183–2184 incidence of, 2184 orthotics and, 2189–2190 outsole design and, 2190–2192, 2190t, 2191f, 2195f playing surface in, 2199–2203, 2200t, 2201f, 2202t, 2203f rear foot stability and, 2187–2189, 2187f, 2188f shoe fit and, 2184–2185, 2184f inlay of, 1893–1896, 1895t–1896t insert of, 1893–1896, 1895t–1896t insole board of, 1889, 1890t insole of, 1889 materials for, 1893–1896, 1895t–1896t shock absorption and, 1897–1899, 1898t, 1899f, 2185–2187, 2185t, 2186f last for, 1886–1887, 1887f–1888f leather for, 1892–1893, 1892f, 1894 lining of, 1889 materials for, 1890–1896 heel counter, 1892 inlay and insert, 1893–1896, 1895t–1896t insole, 1893–1896, 1895t– 1896t shock absorption of, 1897–1899, 1898t medial heel wedge for, 1872 midsole of, 1889–1890, 1890t midsole width for, 1902, 1903f neoprene for, 1894, 1895t nylon mesh for, 1893, 1893f nylon-weave uppers for, 1892–1893, 1893f orthotic devices for, 1872, 1893–1894, 1894f biomechanical aspects of, 1896–1906 injury and, 2189–2190 for pediatric athlete, 1974 rearfoot stability and, 1902–1904, 1903f shock absorption and, 1897–1899 outsole of, 1890 injury and, 2190–2192, 2190t, 2191f, 2195f pads for, 1893 for pediatric athlete, 1974 plantar pressure distribution of, 1899, 1899f plastics for, 1891–1892, 1891t, 1893, 1894–1895, 1895t–1896t playing surface interaction with. See Playing surface

Shoes (Continued) polyethylenes for, 1894–1895, 1895t polymers for, 1890, 1891t polyurethanes for, 1891–1892, 1895, 1895t polyvinyl chloride for, 1895 price of, 1911 pronation with, 1900–1905, 1901f quarter of, 1888 rearfoot control with, 1889, 1900–1905, 1901f injury and, 2187–2189, 2187f–2189f rubber for, 1891–1892, 1894 running injury and, 1904–1905, 1904t–1905t shock absorption of, 1897–1899, 1898t, 1899f, 2185–2187, 2185t, 2186f stress fracture and, 633, 1850 styrene-butadiene rubber for, 1894 toe box of, 1887 toe cap of, 1887 tongue of, 1889 torque testing of, 2200–2203, 2201f torsional flexibility of, 1902 turf toe and, 2082, 2082f, 2085 vamp of, 1888, 1888f viscoelastic materials for, 1895, 1895t Short arc quad exercise, in knee rehabilitation, 257 Short radiolunate ligament, 1320, 1433 Short stature anabolic-androgenic steroids and, 416 growth hormone in, 419 Shoulder. See also Glenohumeral joint; Rotator cuff arthroplasty of in fracture, 1046, 1048f in instability arthropathy, 1114 in osteoarthritis, 1113–1118, 1113f–1114f, 1116f–1118f, 1116t in rheumatoid arthritis, 1114 simulation of, 1151, 1153, 1153f–1155f arthroscopy of. See Arthroscopy, shoulder definitions for, 769–770 dislocation of. See Glenohumeral joint instability frozen. See Adhesive capsulitis hypermobility of, 616 infection of, 389–391, 390f, 391t, 392f instability of. See Glenohumeral joint ­instability intra-articular cartilage fragment in, 581, 583f laxity of, 769 ligaments of. See Glenohumeral ligament(s) Little Leaguer’s. See Little Leaguer’s ­shoulder magnetic resonance arthrography of, 535, 536, 536f, 565–566, 567f osteoarthritis of. See Glenohumeral joint osteoarthritis overuse injury to, 614–617, 615b pediatric, 779–786, 779f anatomy of, 780–786, 781f–785f biomechanics of, 786–790, 787f–789f development of, 780 kinesiology of, 790–791, 790f rehabilitation of. See Shoulder rehabilitation stiff, 1094, 1094b. See also Adhesive capsulitis stretching exercises for, 290f, 291, 1003, 1003f, 1007, 1011 subluxation of, 769 vascular anatomy of, 1137–1138, 1137f vascular injury of, 1137–1142 anatomy of, 1137–1138, 1137f axillary artery, 1140–1141 axillary vein, 1141–1142

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xlv

Shoulder (Continued) clinical presentation of, 1138 dislocation-related, 1139–1140, 1140f, 1141f imaging of, 1139 physical examination in, 1138–1139, 1138f, 1139f scapulothoracic dissociation and, 1139 sternoclavicular dislocation and, 1140 subclavian artery, 1140–1141 trauma-related, 1139–1140, 1139f Shoulder impingement, 986–1015. See also Rotator cuff disorders; Rotator cuff tear(s); SLAP (superior labrum, anterior to ­posterior) lesion diagnosis of, 614–615, 954–955, 955f internal, 992–993, 1001–1002, 1002f, 1016, 1217, 1218 arthroscopic appearance of, 986, 987f evaluation of, 1216, 1216f glenohumeral internal rotation deficit disorder and, 979–980, 980f magnetic resonance imaging in, 978–979, 979f nonoperative treatment of, 1010 magnetic resonance imaging in, 955–956, 956f–959f, 959b osteophyte formation in, 956, 959f pathophysiology of, 954, 1016, 1016f radiography in, 954–955, 955f scapulothoracic joint in, 777 secondary, 1001, 1010, 1016 scapular lag and, 994, 1001 stages of, 614, 954, 1001 Shoulder rehabilitation, 231–232, 232f biofeedback in, 224–225, 233, 233f neuromuscular stimulation in, 231–232, 232f proprioceptive exercises for, 296, 298 MR Systems Cable Column, 298, 299f plyoball perturbations, 296, 298f plyoball Rebounder, 298, 299f wall bounce, 298, 298f therapeutic exercise for, 239–250 acute phase of, 249 after adhesive capsulitis release, 1102, 1103b advanced dynamic training in, 244–246, 245f–248f advanced phase of, 249 anterior capsule stresses in, 249 after arthroplasty, 1117–1118, 1117f–1118f behind-the-neck training in, 248–249 bench press in, 247–248, 248f gym training in, 247–249, 248f infraspinatus, 242–243, 242f–243f kinetic chain, 232, 232f latissimus dorsi, 241, 242f manual resistance training in, 246, 248f overhead press in, 248 pectoralis major, 241, 242f program design for, 249–250 after proximal humeral fracture, 1056–1057, 1057f–1058f return-to-activity phase of, 249–250, 250t–251t rotator cuff, 241–242 scapulothoracic joint, 239–240 serratus anterior, 240–241, 241f–242f after SLAP lesion repair, 1031, 1031f–1032f, 1031t subacute phase of, 249 subscapularis, 244 supraspinatus, 243–244, 243f–244f teres minor, 242–243, 242f–243f trapezius, 240, 240f–241f

xlvi

Index

Shoulder rehabilitation (Continued) upright rows in, 249 transcutaneous electrical nerve stimulation in, 229–230, 230f SI System, 85, 85t Sickle cell disease, 515t Sickle trait, exertional rhabdomyolysis and, 497 Side bridge exercise, in core training, 284, 284f Side-lying abduction exercise, in shoulder rehabilitation, 244, 244f Side-to-side weight shifting, in knee ­rehabilitation, 294–295, 295f Significance, statistical, 2218 Sinding-Larsen-Johansson disease, 599, 1526f, 1529–1530 Single heel rise test, 1979, 1979f Single-leg hopping exercise, in knee ­rehabilitation, 296, 298f, 315f, 329 Single-leg strength training in ankle rehabilitation, 273, 275f in core training, 286 in knee rehabilitation, 263–266, 264f–265f Single-leg toe raise test, in posterior tibial tendinitis, 1981 Single-photon emission computed tomography (SPECT), 544 in isthmic spondylolisthesis, 749, 749f in spondylolysis, 546, 546f, 635, 636f, 761, 761f in stress fracture, 633 Single-subject research design, 105–106, 105f–107f Sinus tarsi syndrome, 1951–1952, 1952t Sinusitis, 149 Sit-ups, partial, in trunk stabilization, 342t, 343 Skier’s thumb, 1399–1401, 1400f Skin abrasion of, 201–202 acne of, 205, 415 anatomy of, 193 chilblains of, 203 contact dermatitis of, 202, 203f eczema of, 204, 204f environmentally induced injury to, 202–203 erythrasma of, 194t, 195–196, 196f folliculitis of, 194–195, 194t, 195f–196f friction-induced injury to, 202, 202f–203f frostbite of, 203, 203f furunculosis of, 194–195, 194t herpes simplex virus infection of, 197, 197f human papillomavirus infection of, 197–198 impetigo of, 193–194, 193f–194f, 194t infection of, 193–200 bacterial, 193–196, 193f–194f, 194t fungal, 198–200, 198f–200f, 200t preparticipation examination of, 513, 515t return-to-play guidelines for, 195t viral, 196–198, 197f–198f inflammatory disorders of, 204–205, 204f–205f injury to, 201–204, 202f–203f laceration of, 201–202 larva migrans of, 201, 202f lesions of, 193–206 methicillin-resistant Staphylococcus aureus infection of, 395–397, 396b, 396f molluscum contagiosum of, 197, 198f necrosis of, after patellar fracture, 1576 parasitic infestations of, 200–201, 201f–202f pediculosis capitis of, 200–201, 201f pitted keratolysis of, 196, 196f preparticipation examination of, 513 psoriasis of, 204–205, 204f scabies of, 201, 201f sunburn of, 204

Skin (Continued) tinea infection of, 198–200, 198f–200f, 200t urticaria of, 205, 205f warfarin-related necrosis of, 381–382, 382f warts of, 197–198 SLAC (scapholunate advanced collapse) wrist, 1327–1330 stage I, 1328, 1328t stage II, 1328, 1328t, 1330f stage III, 1328–1329, 1328t, 1329f–1330f SLAP (superior labrum, anterior to posterior) lesion, 616, 1016, 1016f acromioclavicular joint resection failure and, 1024–1025 arthroscopic repair of, 1027–1031 complications of, 1032 débridement for, 1027–1028 knot-tying for, 1028, 1030–1031, 1030f–1031f outcomes of, 1031–1032 portals for, 1029–1030, 1029f postoperative prescription for, 1031, 1031t, 1031f–1032f return to play after, 1032 superior glenoid preparation for, 1027, 1027f suture anchors for, 1027–1028, 1030, 1030f, 1032 Bankart lesion in, 1021, 1023f biomechanics of, 1019–1021, 1019f–1021f classification of, 1021, 1022f, 2211, 2211f continuous detachment in, 1021, 1023f cord-like middle glenohumeral ligament with, 1021, 1023f evaluation of, 1021–1025, 1025t arthroscopic, 1025, 1027 flap-type, 1021, 1023f magnetic resonance arthrography in, 976, 977f, 978b magnetic resonance imaging in, 581, 583f, 1025, 1026f mechanisms of, 1020, 1020f–1021f, 1022–1024, 1024t, 1028 nonoperative treatment of, 1026 O’Brien active compression test in, 915, 1216, 1216f operative treatment of, 1026–1031, 1027f, 1029f–1030f evidence for, 1028–1029, 1029f physical examination in, 1024–1025, 1025t reciprocal cable model of, 1020, 1020f rotator cuff disorders and, 1024. See also Rotator cuff disorders torsional peel-back mechanism of, 1020, 1021f, 1024 type I, 976, 977f, 1021, 1022f type II, 976, 977f, 1021, 1022f, 1024f, 1026f, 1029f type III, 976, 977f, 1021, 1022f type IV, 976, 977f, 1021, 1022f, 1026f type V, 976, 1021 type VI, 976 type VII, 976 Sleep, 445–450, 446f–447f circadian process of, 445, 446f–447f in depression, 452 disorders of, 447–449 circadian impairment and, 453 homeostatic impairment and, 453 mood disorders and, 452–453 gates to, 453–454, 455f increased appetite for, 453 melatonin secretion and, 457 memory consolidation during, 449–450, 449f morningness-eveningness scale and, 456–457, 456f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Sleep (Continued) naps for, 453, 454b nocturnal window in, 454 nonpharmacologic promotion of, 447–448, 448b, 454b NREM, 446, 447f objective measurement of, 450–451, 451f–452f performance and, 454 postprandial, 453–454 refractory periods for, 454, 455f REM, 446, 447f stages of, 446–447, 447f subjective measurement of, 450, 450f Sleep apnea, 448–449 Sleep deprivation, 449–450 Sleep gates, 453–454, 455f Sleep inertia process, 445–446, 447f Sleep-wake cycle, 445, 446f Sleeper stretch, 290f, 291, 1003, 1003f Slide board leg curls, in knee rehabilitation, 266–267, 267f Slipped capital femoral epiphysis, 597, 598f, 1475–1476, 1476f, 1476t Small intestinal submucosa, in tendon healing, 28 Small intestine, on-field injury to, 527 Smokeless tobacco, 427–428 Snapping hip syndrome, 1458–1459 Snapping triceps tendon, 1226 Sneaker, 1876 Snowboarders, fracture in, 1183, 2155 Soccer, time of day and, 457 Sodium deficiency of, cramps and, 12 in hypertension, 159t loss of, 402 requirements for, 402 Soft tissue infection, 387–388, 387f, 388b, 389f, 389t classification of, 388 Soleus muscle anomaly of, vs. retrocalcaneal bursitis, 2036 stretching exercise for, 71, 71f Solvent abuse, 429–430 Soviet athletes, drug use by, 411–412 Spalding Company shoe, 1876, 1876f Spasm bronchial. See Bronchospasm muscle, delayed-onset soreness and, 12 Spear tackler’s spine, 694, 695f, 710 Spearman’s rank correlation, 113, 113f Specificity, test, 108, 109f, 110, 110t, 2218 Speed’s test in rotator cuff disorders, 997, 1000f in SLAP lesion, 1024–1025 Spencer shoe, 1875, 1875f Sperm, anabolic-androgenic steroid effects on, 416 Spinal accessory nerve anatomy of, 1125–1126 injury to, 1125–1126, 1126f Spinal bifida occulta, 693 Spinal blockade, in complex regional pain syndrome, 359, 366 Spinal canal:vertebral body ratio, in cervical spine, 694, 694f Spinal cord vs. cauda equina, 717–718 cervical decompression of, 675 injury to. See Spinal cord injury, cervical thoracic, 717–718 Spinal cord injury, cervical, 681–686 acute anterior, 674, 678 vs. burners/stingers, 524 in children/adolescents, 709–711

Index Spinal cord injury, cervical�� (Continued) grade of, 685–686 pathophysiology of, 690–691 prevention of, 686–690, 687f–690f recurrence of, 686, 686f spinal stenosis and, 683f–685f, 693–694, 710 squid axon injury model of, 690, 691f Spinal cord stimulation, in complex regional pain syndrome, 366–367 Spinal fusion in cervical spine injury, 699, 699f–700f in lumbar isthmic spondylolisthesis, 750, 750f, 764, 764f Spine. See also Cervical spine injury; ­Thoracolumbar spine injury computed tomography of, 539–541, 540f–541f core training for, 277–288, 277t, 289t abdominal bracing in, 280, 280f advanced functional training in, 288 AIR principles in, 279 bridging progression in, 281–282, 281f–282f curl-up progression in, 284–285, 285f gluteal training in, 285–286, 286f lateral flexion progression in, 283–284, 284f latissimus dorsi training in, 286 loading parameters in, 279–280 manual perturbation training in, 286, 287f program design for, 279 quadruped progression in, 282–283, 282f, 283f rhythmic stabilization training in, 286 rotation training in, 286–288, 288f–289f scapular training in, 286, 286f stress fracture of, 635–636, 636f–637f trunk stabilization program for, 341–349, 342t aerobic exercise in, 342t, 349 ball exercises in, 342t, 345 bridging exercise in, 342t, 344 dead bug exercise in, 342t, 343 partial sit-ups in, 342t, 343 postural exercises in, 348–349 prone exercises in, 342t, 345 quadruped exercises in, 342t, 346 stabilization exercises in, 343–348 wall slide exercises in, 342t, 347 water running in, 349 weight training in, 349 Spine board in cervical spine injury, 665, 666–667, 666f in thoracolumbar spine injury, 720, 734 Spinoglenoid ligament, 1120, 1120f Spinoglenoid notch, suprascapular nerve ­compression at, 1121, 1121f, 1123 Spirometry, in exercise-induced bronchospasm, 182 Spleen on-field injury to, 526–527 rupture of, infectious mononucleosis and, 151 size of, 151 Splenomegaly, clearance for participation and, 513, 515t Splint/splinting in Achilles tendon rupture, 2005 in boutonnière deformity, 1389, 1389f in elbow heterotopic ossification, 1295 in Jersey finger, 1426, 1427f, 1428 in mallet finger, 1388 in plantar fasciitis, 2049 in retrocalcaneal bursitis, 2035 in sagittal band rupture, 1390 in wrist disorders, 1361–1362, 1443

Splinter, ultrasonography of, 538, 539f Spondyloarthritis, seronegative, heel pain in, 2033–2034, 2047 Spondylolisthesis, 635, 637f cervical, traumatic, 678, 707–708 degenerative, 747–748, 748t, 752 isthmic, 748–750, 748t, 749f, 752 in children, 756, 756f, 763–764, 764f return to play after, 750 pediatric, 756, 756f, 762–764, 767, 767b surgery in, 763–764, 764f Spondylolysis, 635–636, 756, 756b, 762–763 computed tomography in, 546, 546f, 762, 763f radiography in, 635, 636f, 760, 760f radionuclide imaging in, 726, 726f return to play and, 767, 767b single-photon emission computed ­tomography in, 546, 546f, 635, 636f, 761, 761f Sport cord exercise, in knee rehabilitation, 271, 272f, 296, 297f Sport psychologist as clinician, 437–438 as consultant, 438–439 definition of, 439–440 as educator, 438 as facilitator, 438 Sport psychology, 437–440 Sports drinks, 401 Sports shoes. See Shoes Sprain ankle. See Ankle sprain bifurcate, 1953–1955, 1954f cervical, 673–674 Lisfranc, 1955–1960, 1956f–1957f, 1959f lumbar, 733 sternoclavicular joint. See Sternoclavicular joint injury subtalar, 1947–1952, 1947f–1949f, 1951f, 1952t thoracic, 732–733 wrist. See Scapholunate ligament, injury to Spring ligament, 1913f Spurling’s test in cervical spine injury, 672, 672f in thoracic outlet syndrome, 1132–1133, 1133f Squat and reach exercise, in knee rehabilitation, 270, 271f Squat exercise back, 262, 263f in core training, 286, 286f cruciate ligament effects with, 223, 223t, 224, 224f, 262–263 front, 262, 262f goblet, 261, 261f hex bar, 261–262, 262f in knee rehabilitation, 260–261, 261f multiplanar, 270, 270f patellar effects with, 226–227 patellofemoral joint stresses with, 262–263 quadriceps dominant, 258–263, 259f–260f, 261f–263f squat and reach, 270, 271f standing, 260–263, 261f–263f sumo, 309f, 329 wall, 260, 260f, 330 single-leg, 264, 265f Squat test, in meniscal injury, 1602 Squatting, patellofemoral joint reaction force with, 227, 1592–1593, 1592f Squeakers, 625 Squeeze test, in high (syndesmosis) ankle sprain, 1940, 1940f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xlvii

Squid axon injury, 690, 691f Stability ball bridges, in knee rehabilitation, 266, 267f Stability ball leg curl, in knee rehabilitation, 266–267, 267f Stability ball training, in shoulder rehabilitation, 245, 246f Stabilization exercises. See Trunk stabilization program Standard deviation, 112 Standing side bridge exercise, in core training, 284, 284f Stanford Sleepiness Scale, 450 Staphylococcus aureus infection, 395–397, 396b, 396f methicillin-resistant, 193, 194–195, 194t, 195f community-acquired, 194–195, 195f prevention of, 395–396, 396b Statistics, 97–119. See also Research accuracy in, 100 bias in, 100 causation in, 100–101 comparison of means, 112–113, 113f correlation in, 100–101 decision analysis in, 99 descriptive, 112–114, 113f error in, 100, 111, 118–119 hypothesis testing in, 99 inference in, 99 linear regression, 113, 113f logistic regression, 113–114, 114f P-value in, 110 population in, 99 power analysis in, 111–112, 116–117 precision in, 100 reliability in, 100 sample in, 99 sensitivity in, 110 specificity in, 110 study design in, 99, 101–110, 101f, 111 table analysis in, 114 terminology of, 2217–2218 two-by-two table analysis in, 108–109, 109f, 114 validity in, 100 variables in, 97, 99, 100–101, 112–114 Status epilepticus, 523 Stem cells, in knee cartilage lesions, 1776 Step-up exercises in ankle rehabilitation, 273, 275f in knee rehabilitation, 264, 264f Sterilization, allograft, 139 Sternoclavicular joint, 791–825. See also ­Sternoclavicular joint injury anatomy of, 769, 769f ligamentous, 792–794, 792f–795f pediatric, 783–784, 784f surgical, 792–794, 792f–795f vascular, 794, 797f arthritis of, 811–812, 820 computed tomography of, 804–805, 805f congenital disorders of, 799, 811 dislocation of. See Sternoclavicular joint injury, severe sprain (dislocation) hyperostosis of, 812 iatrogenic instability of, 824 injury to. See Sternoclavicular joint injury kinematics of, 775, 775f magnetic resonance imaging of, 804 pediatric anatomy of, 783–784, 784f biomechanics of, 786, 787f radiography of, 802–805, 803f–804f anteroposterior view for, 802

xlviii

Index

Sternoclavicular joint (Continued) Heinig view for, 803, 803f Hobbs view for, 803, 803f serendipity view for, 803–804, 804f range of motion of, 794, 796f spontaneous subluxation/dislocation of, 799, 801f, 811, 820 tomography of, 804, 805f Sternoclavicular joint injury, 792–825 atraumatic, 799, 801f, 811–812, 820–821 classification of, 798–799, 799f–800f anatomy-based, 798, 799f–800f cause-based, 798–799, 801f clavicular epiphysis in, 794, 796f complications of, 821–824, 821f–823f computed tomography in, 804–805, 805f historical perspective on, 791–792 hypotension with, 822, 823f iatrogenic instability after, 824 incidence of, 799, 801–802 K-wire migration with, 823 magnetic resonance imaging in, 804 mechanism of, 794–797, 798f direct force, 795 indirect force, 795–796, 798f mediastinal vessel compression with, 821, 822f mild sprain, 798, 802, 805, 812 moderate sprain (subluxation), 798, 802, 805–806, 806f, 812 physeal, 811 medial, 824–825, 824f anterior displacement, 811, 824–825 posterior displacement, 811, 825 treatment of, 819 return to sport after, 825 pin-related complications with, 823–824 radiography in, 802–805, 803f–804f anteroposterior view for, 802 Heinig view for, 803, 803f Hobbs view for, 803, 803f serendipity view for, 803–804, 804f range of motion in, 794, 796f severe sprain (dislocation), 806–810, 813–818 acute, 798 anterior, 798, 798f–799f, 802, 815–816 closed reduction for, 806, 812 incidence of, 799, 801 incidence of, 799, 801–802 mechanism of, 795–796, 798f nonoperative treatment of, 806–809, 812–814, 813f on-field, 526 operative treatment of, 809–810, 810f, 814–815, 814f–818f posterior, 798, 799f, 807–810, 816 abduction traction reduction for, 808, 808f adduction traction reduction for, 808, 808f closed reduction for, 798, 800f, 807–809, 808f, 812–813, 813f complications of, 821–823, 821f, 822f, 823f computed tomography of, 807, 807f evaluation of, 807 incidence of, 799, 801 operative complications of, 823–824 operative treatment of, 809–810, 810f, 814–816, 814f–818f postreduction care of, 809 signs and symptoms of, 802 ultrasonography of, 807f recurrent, 799, 809, 815 signs and symptoms of, 802

Sternoclavicular joint injury (Continued) unreduced, 799, 809, 814–816 vascular injury with, 1140 signs and symptoms of, 802 subcutaneous emphysema with, 821–822, 823f surgical anatomy in, 792–794, 792f–795f, 797f tomography of, 804, 804f tracheal displacement with, 821, 821f Sternomanubrial dislocation, 897, 898f–899f reduction of, 899, 899f Sternum anatomy of, 896–897, 897f fracture of, 896–900, 898f–899f mechanisms of, 897, 897f–898f return to play after, 899 treatment of, 899 Steroids. See Anabolic-androgenic steroids; Corticosteroid(s) Stiffness biomechanical, 93 elbow, 1293 joint, 225, 225f, 2176 meniscal, 60, 61f, 1600 muscle, 2177–2178 Stimson’s maneuver, 935, 936f Stimulants, 451–452, 452t Stinchfield test, in hip degenerative disease, 1503 Stingers clearance for participation and, 513 vs. spinal cord injury, 524 Straight leg raise, tibia translation with, 222 Straight leg raise exercise, in knee ­rehabilitation, 256–257, 257f, 267–268, 269f Straight leg raising test in hip degenerative disease, 1503 in lumbar disk herniation, 743 in thoracolumbar spine injury, 722, 723f Strain lumbar, 733 mechanical, 94 muscle. See Muscle(s), strain injury of and at specific muscles thoracic, 732–733 Strengthening exercise/training, 10. See also Core training; Therapeutic exercise(s) in ankle instability, 340 in children/adolescents, 465 in complex regional pain syndrome, 362 in female athlete, 479–480, 481 gastrosoleus, 273, 275f in hamstring strain prevention, 336–337, 337f in iliotibial band friction band syndrome, 628, 630f after proximal humeral fracture, 1057, 1058f quadriceps, 256–257, 257f, 258–263, 260f–263f rotator cuff, 242–244, 242f–244f, 1003–1007, 1004f–1007f scapula, 1007, 1008f serratus anterior, 240–241, 241f–242f shoulder, 246, 248f, 1003, 1004f–1006f, 1007, 1011 single-leg, 263–266, 264f–265f, 273, 275f, 286 total arm, 251–255, 251t, 252f–254f Streptococcus sanguis infection, 398, 398t Stress, seizures with, 187 Stress fracture, 631–650, 1851b acromial, 867 calcaneal, 556f, 646, 647f cuboid, 646

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Stress fracture (Continued) cuneiform, 646 diagnosis of, 633 differential diagnosis of, 633 in endurance athlete, 632 in female, 483, 483b, 483f, 632–633, 1849–1850, 1851–1852, 1856, 2014 femoral neck, 554f, 638–640, 640f–644f femoral shaft, 641, 1477–1481. See also Femoral shaft, stress fracture of fibular, 644, 644f, 1849 fifth metatarsal, 648, 649f first proximal phalanx, 2028 foot and ankle, 2012–2030. See also Foot (feet), stress fracture of and at specific fractures great toe, 649–650, 649f humeral, 634–635, 635f, 1176–1177, 1178, 1179, 1180t, 1182 physeal, 1090–1093, 1093b, 1093f. See also Little Leaguer’s shoulder imaging of, 552–553, 554f–555f, 633, 634t incidence of, 632–633 leg-length inequality and, 632 lower extremity, 638–650, 1849–1856. See also Femoral shaft, stress fracture of; Tibia, stress fracture of magnetic resonance imaging in, 552–553, 554f–555f malleolar, 644–645, 645f, 2016t, 2018, 2020f metatarsal, 646–648, 648f–649f foot type and, 1904 metatarsal length and, 2182 shoewear and, 1898–1899 pars interarticularis (spondylolysis), 635–636, 756, 761, 761f, 762–763, 763f patellar, 641–643 pathogenesis of, 632–633 pelvic, 638 prevention of, 406, 634 proximal humeral physis, 1090–1093, 1093b, 1093f pubic rami, 638, 638f radionuclide imaging in, 545–546, 545f–547f, 552 recurrence of, 1856 rib, 636–637, 637f, 895–896, 895t, 896f sacral, 638 scaphoid, 1365–1366, 1366f sesamoid, 649–650, 649f shoe selection and, 633 spinal, 635–636, 636f–637f systemic factors in, 632 talar, 645 tarsal navicular, 645–646, 646f tibial, 643–644, 644f, 652, 652t, 1849–1856. See also Tibia, stress fracture of treatment of, 634 Stress-relaxation, 18, 19f, 95–96, 95f articular cartilage, 48 meniscal, 59 Stress-strain curve, 94, 94f Stretching animal model of, 19 in cramps, 12 plantar fascia, 2049 Stretching exercises, 288–293 biofeedback with, 292f, 293 gastrocnemius, 291, 291f hamstring, 289–290, 290f, 292f, 293 iliotibial band, 291–292, 291f knee, 292–293, 292f after proximal humeral fracture, 1056, 1057f shoulder, 290f, 291, 1003, 1003f, 1007, 1011 soleus, 71, 71f

Index Stroke volume, exercise effect on, 219, 219f, 220t Strychnine, 410 Stryker Intra-Compartmental Pressure ­Monitor, 1860 Student’s elbow. See Olecranon bursitis Study design, 101–110, 101f analytical, 107–108 blinding in, 104–105 control group in, 101–102, 104 experimental, 101f, 103–106, 105f matching in, 101 observational, 101–103, 101f, 102t–103t, 106–107 power analysis in, 111 pre-experimental, 103–104 qualitative, 106–107 single-subject, 105–106, 105f–107f Subacromial bursa, 990, 991f Subarachnoid hemorrhage, 661, 661f Subcalcaneal pain syndrome, 2042–2043. See also Plantar fasciitis Subchondral bone, 41f abrasion of, 52–63 Subclavian artery, injury to, 1140–1141 Subclavian vein, effort thrombosis of (­Paget-Schroetter syndrome), 1129–1130, 1131f, 1134 Subclavius tendon, in sternoclavicular joint dislocation treatment, 810 Subdural hematoma, 660, 661f Sublabral foramen, 972, 972f Sublabral recess, 970f, 972 Subscapularis anatomy of, 770, 770f, 771, 910, 989, 989f, 1035f, 1035t, 1064–1065 fatty infiltration of, 989, 989f function of, 991 humeral attachment of, 1070–1071, 1071f pediatric, 785, 785f rupture of, 1064–1065 clinical evaluation of, 1065 treatment of, 1065 scapular attachment of, 858, 858f, 886f strengthening exercise for, 244 Subscapularis bursa, 771 Subscapularis tendon, 771. See also Rotator cuff; Rotator cuff disorders entrapment of, 957 tear of, 566, 963–964, 964f. See also Rotator cuff tear(s) Subtalar joint anatomy of, 1947–1948, 1947f–1948t biomechanics of, 1867–1869, 1868f–1870f, 1947–1948, 1948f dislocation of, 1952–1953, 1953f ligaments of, 1947–1948, 1947f, 1948t injury to. See Subtalar sprain motion of, 2178–2179, 2179f, 2179t muscle function at, 1868–1869, 1870f Subtalar sprain, 1947–1952 anatomy of, 1947–1948, 1947f, 1948t arthrography in, 1950 chronic, 1950–1951 computed tomography in, 1950 evaluation of, 1948–1949, 1949f grade I, 1948, 1950 grade II, 1948, 1950 grade III, 1948, 1950 history in, 1949 magnetic resonance imaging in, 1950 physical examination in, 1949, 1949f radiography in, 1949–1950, 1949f rehabilitation after, 1951 return to play after, 1952

Subtalar sprain (Continued) sinus tarsi syndrome and, 1951–1952, 1952t tarsal coalition and, 1933 treatment of, 1950–1951, 1951f Subtalar varus tilt test, 1949–1950 Subungual exostosis, 2105–2107 classification of, 2105, 2105t evaluation of, 2105–2106, 2106b, 2106f imaging in, 2106, 2106f nonoperative treatment of, 2106 operative treatment of, 2106–2107, 2107f care after, 2107 return to play after, 2107 Subungual hematoma, 2098 Subungual osteochondroma, 2105, 2105t Sudden death, 162–172 arrhythmogenic right ventricular ­cardiomyopathy and, 166, 167f causes of, 162–167, 163f, 163t commotio cordis and, 165, 165f coronary artery anomalies and, 165, 165f–166f definition of, 163 hypertrophic cardiomyopathy and, 164–165, 164f long QT syndrome and, 166–167, 167f myocarditis and, 165–166, 166f prevention of, 167–171 ablation procedures in, 170 β-adrenergic blockers in, 170 Bethesda Guidelines in, 169, 170t echocardiography in, 168 electrocardiography in, 168–169, 169t equipment in, 170–171 implantable defibrillators in, 169–170 preparticipation evaluation in, 167–168 screening for, 167–169, 168t, 509, 512, 512t resuscitation for, 171–172, 171t Sudeck’s atrophy, in complex regional pain syndrome, 358 Sudomotor function, in complex regional pain syndrome, 355 Sulcus angle in femoral trochlea dysplasia, 1563–1564, 1564t patellar, 1540 Sulcus sign, in glenohumeral joint instability, 914, 915f, 943, 943f Sunburn, 204 Sunscreen, 204 Superficial femoral artery, occlusion of, 1497–1499 Superficial peroneal nerve, 15f, 2062, 2063f entrapment of, 2062–2063, 2062f injury to, 1933 Superior peroneal retinaculum, 1931, 1987 congenital absence of, 1991 disruption of, 1932 imbrication of, 1932–1933 reconstruction of, 1995 Superior shoulder suspensory complex, 857, 863, 864f Supplement. See Dietary supplement(s) Suprachiasmatic nucleus, 443–444, 444f Supracondylar fracture, 596–597, 597f, 1186, 1186f, 1280t, 1281–1283, 1282f Suprascapular artery, 858 Suprascapular nerve anatomy of, 858–859, 1120–1121, 1120f, 1122f in coracoid fracture, 880, 882f–883f Suprascapular nerve injury, 616–617, 1120–1124 electromyography in, 1122, 1124 evaluation of, 1121–1123, 1122f ganglion cyst and, 1121, 1121f, 1123 magnetic resonance imaging in, 1121–1122 muscle atrophy with, 1121, 1122f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

xlix

Suprascapular nerve injury (Continued) nerve conduction study in, 1122–1123 paralabral cyst and, 980, 980t radiography in, 1121, 1122f scapular fracture and, 1121 throwing-related, 1221 treatment of, 1123–1124 ultrasonography in, 1122 Suprascapular notch, suprascapular nerve ­compression at, 1120, 1120f, 1123 Supraspinatus anatomy of, 770, 770f, 910–911, 989, 989f, 1035t atrophy of, 770 fatty atrophy of, 963, 963f, 980 fatty infiltration of, 989, 989f humeral attachment of, 1070–1071, 1071f pediatric, 785, 785f scapular attachment of, 859f, 886f strengthening exercise for, 243–244, 243f–244f, 1003, 1006f Supraspinatus tendon, 770. See also Rotator cuff; Rotator cuff disorders calcification of, 566, 568f, 963, 963f tear of. See Rotator cuff tear(s) tendinopathy of, 30, 566, 568f, 960, 960f Sural nerve, 2060–2061, 2061f injury to, 2060–2061 Suture(s), 132–136 abrasion of, 133–134 biologic characteristics of, 132–133 deformation of, 134 demands on, 132 Duncan loop with, 135, 135f handling properties of, 136 knotting properties of, 135–136, 135f mechanical characteristics of, 133–134 Nicky’s knot with, 135, 135f Roeder’s knot with, 135, 135f selection of, 134, 136 strength of, 133–134 tensile load on, 134 Sweat, salt in, 402 Sweat glands, in children, 465 Sweating in complex regional pain syndrome, 355, 360–361 evaluation of, 360–361 Swimmer’s ear, vs. thoracic outlet syndrome, 1130 Swimming core body temperature in, 455f epilepsy and, 192 time of day and, 455f, 457 Syme amputation, of great toe, 2099, 2104f Sympathectomy, in complex regional pain syndrome, 366 Sympathetic ganglion blockade, in complex regional pain syndrome, 359, 365 Sympatholysis, in complex regional pain ­syndrome, 365 Synapse, in complex regional pain syndrome, 353–355, 354f Syncope, heat, 494, 529–530 Synephrine, 409 Synovectomy, of wrist, 1443–1444 Synovial cyst, glenohumeral, 617 Synovial disease, of hip, 1473

T t-test, 112–113, 113f, 116–117 Table lateral crunch exercise, 316f, 330 Taking-off-shoes test, in hamstring strain, 1486f, 1487



Index

Talalgia, 2034 Talar tilt, on stress radiography, 1918–1919, 1918f Talar tilt test in lateral ankle sprain, 1917–1918, 1917f in medial ankle sprain, 1936 in subtalar sprain, 1949 Talcum powder sclerosant, in olecranon ­bursitis, 1248 Talocalcaneal coalition, 1872, 1962 Talocalcaneal ligament, 1913f Talocrural joint. See Ankle Talofibular ligament anterior, 338, 1913–1914, 1913f–1915f biomechanics of, 1865–1866, 1866f injury to, 1866. See also Ankle sprain, lateral magnetic resonance imaging of, 573, 574f repair of, 1923–1924, 1923f, 1926–1928, 1926f, 1927f posterior, 338, 1913–1914, 1913f, 1914f injury to. See also Ankle sprain, lateral Talon noir, 202, 202f Talonavicular joint biomechanics of, 1869, 1870f impairment of, 1872 Talus eosinophilic granuloma of, 2161, 2161f fracture of, 2153–2156, 2154t–2155t, 2155f–2156f osteochondroses of. See Osteochondrosis (osteochondroses), talar pigmented villonodular synovitis of, 2161, 2161f stress fracture of, 645, 2016t, 2018–2019, 2021f Taping in ankle instability, 340, 1934 in lesser toe deformity, 2121, 2121f for plantar fasciitis, 2049–2050 for play, 1362 in sesamoid dysfunction, 2090, 2093f Tarsal coalition, 1933, 1960–1962, 1961f, 1962f, 2161, 2161f Tarsal joint, transverse biomechanics of, 1869, 1870f, 2179 impairment of, 1872 medial swivel syndrome of, 2179 Tarsal navicular osteochondrosis of, 599, 1972–1973, 1973f, 2162, 2163f stress fracture of, 645–646, 2016t, 2020–2022, 2023f Tarsal tunnel, 2057, 2057f release of, 2058–2059, 2059f Tarsal tunnel syndrome, 631–632, 2047, 2057–2059 anterior, 2061–2062, 2061f clinical features of, 2057, 2057f differential diagnosis of, 2058, 2058b electrodiagnostic studies in, 2058 etiology of, 2058 physical examination in, 2058 treatment of, 2058–2059, 2059f Tartan syndrome, 2198 Tazarotene, in acne, 205 Team physician, 507 emergency function of. See On-field ­emergency ethical responsibilities of, 530–531 examining function of. See Preparticipation examination institutional relationships of, 530 legal responsibilities of, 531 medicolegal responsibilities of, 531 minor care provision by, 531

Team physician (Continued) out-of-state practice by, 531 rewards of, 532 roles of, 507 supervisory function of, 516 support relationships of, 530 Teardrop fracture, 680–681, 681f Telmisartan, 160t Temperature, body clothing effects on, 499–500 disease effects on, 499 disorders of. See Cold injury; Heat illness/­ injury; Hypothermia drug effects on, 499 Ten (10) test, in thoracic outlet syndrome, 1131 Tendinitis, 29–30, 1975, 1516 Achilles, 1997–2002. See also Achilles tendon, injury to flexor carpi radialis, 1356–1357 flexor carpi ulnaris, 1356–1357 flexor hallucis longus, 1983–1984, 1986, 1986f peroneal, 1988–1990 posterior tibial tendon, 1981–1983, 1982f triceps, 1207–1209 Tendinopathy, 29, 611–614, 612t, 613f. See also Overuse injury and at specific tendon disorders calcific, of supraspinatus, 566, 567f distal biceps, 566, 570f fluoroquinolone-related, 30 insertional, 30 magnetic resonance imaging in, 558 patellar, 560, 561f, 626–627, 626f, 1515–1518, 1517b, 1518t posterior tibial tendon, 563, 563f quadriceps, 560 rotator cuff, 565, 566, 566f, 567f, 770. See also Rotator cuff disorders; ­Rotator cuff tear(s) triceps tendon, 567 ultrasonography of, 538 Tendinosis, 611–612, 612t, 613f, 1975 Achilles, 30, 1975, 2030–2031, 2034, 2034f, 2036 patellar. See Patellar tendinosis wrist extensor, 617–619, 618f Tendodermodesis, in mallet finger, 1421–1422, 1421f Tendon(s), 20–31. See also specific tendons age-related changes in, 25, 30 biochemistry of, 22–24, 22f, 23f, 23t biomechanics of, 24–25, 25f, 26f blood supply of, 21 bone insertions of, 20–21 cells of, 20, 22, 24 interaction among, 24 collagen of, 20, 22–23, 22f, 23f, 23t adaptability of, 25–27 age-related changes in, 25 corticosteroid effects on, 26–27 exercise-related changes in, 25, 26f immobilization-related changes in, 25–26, 26f NSAID effects on, 27 synthesis of, 27–28 corticosteroid injection effects on, 26–27, 29, 31 degeneration of. See Tendinopathy and at specific tendon disorders elastin of, 23 exercise-related changes in, 25, 26f fibroblasts of, 20, 22, 24, 27 glycosaminoglycans of, 23–24 ground substance of, 23–24

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Tendon(s) (Continued) healing of, 27–30 active mobilization in, 28 biomechanics of, 28 corticosteroid effects on, 29 inflammatory response in, 27–28 mechanical stress in, 28–29 nonsteroidal anti-inflammatory drug ­effects on, 29 passive mobilization in, 29 primary, 27–28 immobilization-related changes in, 25–26, 26f inflammation of. See Tenosynovitis injury to. See also at specific tendons diagnosis of, 30 exercise for, 31 healing after. See Tendon(s), healing of mechanisms of, 27 overuse, 29–30. See also Overuse injury trauma-induced, 30 treatment of, 30–31 innervation of, 22 load-deformation curve for, 25, 26f load-elongation curve for, 24–25, 25f magnetic resonance imaging of, 558, 560–569 at ankle, 560–563, 562f–566f at elbow, 566–567, 569f, 570f at hip, 567–568 at knee, 560, 561f, 562f at shoulder, 564–566, 566f, 567f, 568f at wrist, 568–569, 570f mechanical properties of, 24–25, 25f, 26f nonsteroidal anti-inflammatory drug effects on, 27, 29 proteoglycans of, 23–24 strength of, after injury, 28 stress relaxation of, 18, 19f stress-strain curve for, 25, 26f structure of, 20–22, 21f, 1515, 1517f syncytium of, 20 ultrasonography of, 538, 538f viscoelastic properties of, 24–25, 25f Tendon surface cells, 20, 24 Tendoscopy in flexor hallucis longus tendinitis, 1984 in peroneal tendinitis, 1990 in posterior tibial tendinitis, 1982–1983 Tennis elbow. See Epicondylitis, lateral Tennis racquet, 1199, 1200f Tennis shoulder, 1130 Tenosynovitis, 1975 de Quervain’s, 569, 570f, 624–625, 1355, 1356f magnetic resonance imaging in, 558 posterior tibial tendon, 538, 538f Tenosynovium, 20 Tenovagina, 20 Tensile loading testing, 93, 93f Tension pneumothorax, 525 Tensor fascia lata, 1454t, 1455f Terazosin in complex regional pain syndrome, 363t, 365 in hypertension, 160t Terbinafine, in dermatophyte infection, 200t Teres major, 771, 1034, 1035t scapular attachment of, 859f, 886f Teres major tendon, 771 Teres minor anatomy of, 770–771, 770f, 989, 989f, 1035t pediatric, 785, 785f function of, 991

Index Teres minor (Continued) humeral attachment of, 1070–1071, 1071f scapular attachment of, 859f, 886f strengthening exercises for, 242–243, 242f, 243f Teres minor tendon, 770. See also Rotator cuff tears of, 566 Terminology, 2207–2209, 2208t, 2209t epidemiological, 2218–2219, 2218t statistical, 2217–2218 Testis (testes) anabolic-androgenic steroid effects on, 416 disorders of, 515t on-field injury to, 527 Testosterone in endurance athlete, 633. See also Anabolic-androgenic steroids exercise effects on, 217t, 218 historical perspective on, 411–414 Tetanus, muscle, 7–8, 8f Thenar hammer syndrome, 1359 Therapeutic exercise(s), 238–293 in acromioclavicular joint injury, 838–840, 839f, 840f AIR acronym for, 238 ankle, 272–276 eccentric training in, 275–276, 276f gastrosoleus training in, 273, 275f neuromuscular control training in, 273, 275, 275f neuromuscular training in, 273, 274f proprioceptive training in, 273, 275, 275f single-leg training in, 273, 275f application concepts for, 238–239, 238t, 239t in atraumatic glenohumeral joint instability, 945, 945f after coracoid fracture, 884 core training in, 277–288, 277t, 289t abdominal bracing in, 280, 280f advanced functional training in, 288 AIR principles in, 279 bridging progression in, 281–282, 281f, 282f curl-up progression in, 284–285, 285f gluteal training in, 285–286, 286f lateral flexion progression in, 283–284, 284f latissimus dorsi training in, 286 loading parameters in, 279–280 manual perturbation training in, 286, 287f program design for, 279 quadruped progression in, 282–283, 282f, 283f rhythmic stabilization training in, 286 rotation training in, 286–288, 288f–289f scapular training in, 286, 286f elbow, 250–255 extensor training in, 253–254, 253f flexor training in, 251–253, 252f, 253f forearm muscle training in, 254–255, 254f, 255f rhythmic stabilization training in, 255, 255f total arm strengthening in, 251 glossary of, 324–330 knee, 255–272 ACL loading with, 221–222, 222t ACL strain measurements in, 1586–1588, 1586f, 1587t acute phase of, 255 advanced phase of, 255 gluteal muscle raise in, 267, 268f gluteal musculature, 257–258, 258f, 259f hamstring curls in, 266 hamstring raise in, 267, 268f

Therapeutic exercise(s) (Continued) hip extension in, 258, 258f, 267 hyperextension in, 267 leg press in, 259–260, 260f lunges in, 264–266, 265f, 266f manual perturbation training in, 270–271, 271f multiplanar squats in, 270, 270f neuromuscular activation, 256–258 neuromuscular control training in, 268–270, 269t, 270f PCL loading with, 221–222, 222t program design for, 256 proprioceptive training in, 268–270, 269t, 270f quadriceps, 256–257, 257f quadriceps dominant squatting, 258–263, 260f, 261f–263f return-to-activity phase of, 256 Romanian deadlift in, 267–268, 269f single-leg balance training in, 271 single-leg presses in, 263, 264f single-leg Romanian deadlift in, 268, 269f single-leg strength training in, 263–266, 264f, 265f single-leg wall squats in, 264, 265f slide board leg curls in, 266–267, 267f sport cord activities in, 271, 272f, 296, 297f squat and reach in, 270, 271f squats in, 260–263, 261f–263f stability ball bridges in, 266, 267f stability ball curls in, 266–267, 267f step-ups in, 264, 264f straight leg, 267–268 subacute phase of, 255 Total Gym in, 259–260, 260f unstable surface training in, 271, 272f wall squats in, 260, 260f loading parameters in, 277, 277t after proximal humerus fracture, 1057f–1058f shoulder, 239–250 acute phase of, 249 after adhesive capsulitis release, 1102, 1103b advanced dynamic training in, 244–246, 245f–248f advanced phase of, 249 anterior capsule stresses in, 249 after arthroplasty, 1117–1118, 1117f, 1118f behind-the-neck training in, 248–249 bench press in, 247–248, 248f gym training in, 247–249, 248f infraspinatus, 242–243, 242f–243f kinetic chain, 232, 232f latissimus dorsi, 241, 242f overhead press in, 248 pectoralis major, 241, 242f program design for, 249–250 after proximal humeral fracture, 1056–1066, 1057f, 1058f return-to-activity phase of, 249–250, 250t, 251t rotator cuff, 241–242 scapulothoracic joint, 239–240 serratus anterior, 240–241, 241f–242f after SLAP lesion repair, 1031, 1031t, 1031f–1032f subacute phase of, 249 subscapularis, 244 supraspinatus, 243–244, 243f–244f teres minor, 242–243, 242f–243f trapezius, 240, 240f, 241f upright rows in, 249 stretching, 288–293, 290f–292f warm-up for, 276

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

li

Thermogenesis, thermoregulatory, 498–499 Thermography, in complex regional pain syndrome, 360 Thermoregulation, 498–499 in children/adolescents, 465–466 Thiamin, requirements for, 406b Thigh pad with ring, 1482, 1482f Thomas test, in hip degenerative disease, 1503 Thompson-Terwilliger procedure, 2099, 2104f Thompson’s test, in Achilles tendon rupture, 2003 Thoracic outlet syndrome, 1127–1136, 1128b, 1142 Adson’s test in, 1131, 1132f anatomy of, 1127–1129, 1128f–1130f arteriography in, 1133 differential diagnosis of, 1128b fibrous muscular bands in, 1128, 1129f–1130f Halstead’s maneuver in, 1131, 1132f historical perspective on, 1127 lidocaine injection test in, 1133 magnetic resonance imaging in, 1133 neurologic symptoms in, 1129 nonoperative treatment of, 1134–1135, 1135f operative treatment of, 1134, 1135–1136, 1135f pain in, 1129 physical examination in, 1131 radiography in, 1133, 1133f retroclavicular Spurling’s test in, 1132–1133, 1133f return to play after, 1134 Roos’ test in, 1132, 1133f symptoms of, 1129–1131, 1130f–1131f ten (10) test in, 1131 Wright’s hyperabduction test in, 1132, 1132f Thoracic spine. See also Thoracolumbar spine injury anatomy of, 715–717, 715f–716f, 724, 724f, 754–755 blood supply of, 716, 716f computed tomography of, 541, 727 contusion of, 732–733 diskography of, 728 fracture of, 733–736, 733f, 734f in children/adolescents, 755, 755f compression, 546, 547f, 734–735, 734f, 755, 755f injury to. See Thoracolumbar spine injury innervation of, 716, 716f intervertebral disk of, 717, 717f herniation of, 738, 739f radiography of, 728 kyphosis of, 718, 736–737 ligaments of, 715 magnetic resonance imaging of, 727 muscles of, 715–716, 715f radiography of, 724, 724f, 728 radionuclide imaging of, 726, 726f sprain of, 732–733 stenosis of, 738–739, 739f strain of, 732–733 venous drainage of, 716–717 zygapophyseal joints of, 718, 719f Thoracolumbar spine. See also Thoracolumbar spine injury anatomy of, 715–717, 715f–717f biomechanics of, 718–719, 719f vs. cauda equina, 717–718 extension of, 718 flexion of, 718 lateral flexion of, 718 rotation of, 718 zygapophyseal joints of, 718, 719f

lii

Index

Thoracolumbar spine injury, 714–768 cold therapy for, 731 diagnostic blocks in, 731–732, 731t evaluation of, 719–728 diagnostic blocks in, 731–732, 731t diagnostic testing in, 723–728, 724f–727f imaging in, 723–728, 724f–727f inspection in, 721 neurologic examination in, 721–723, 722t, 722f–723f pain in, 719–720 palpation in, 721 patient history in, 719–720 physical examination in, 720–723, 722t, 722f–723f range of motion in, 721 straight leg raising test in, 722, 723f fracture, 733–736, 733f–����������� 735f heat therapy for, 731 immobilization for, 720 lumbar spine stabilization for, 728–730, 730f medications for, 731 on-field, 524, 720 pediatric, 754–768 anatomy in, 754–755 classification of, 755–757, 755f, 755t, 756b, 756f–758f evaluation of, 757–762 Adam’s forward bend test in, 759 imaging in, 759–762, 760f, 761f neurologic examination in, 759 palpation in, 759 patient history in, 757–759, 759b physical examination in, 759, 759b range of motion in, 759 overuse syndrome and, 766 return to play after, 767–768, 767b treatment of, 762–766, 763b evidence for, 766–767 nonoperative, 762–763, 764–765, 765f operative, 763–764, 764f, 765–766 rehabilitation cycle in, 728, 728f return to play after, 750–752, 751f, 752t spine board for, 720, 734 sprain, 732–733 strain, 732–733 trigger point injections for, 732 Thorax, on-field injury to, 525–526 Thrombocytopenia, heparin-induced, 381 Thrombophilia, 372–374, 374f, 374t Thrombosis axillary artery, 1140–1141, 1141f effort axillary vein, 1141–1142 on-field, 527 subclavian vein, 1129, 1131f, 1134 throwing-related, 1226 venous. See also Deep venous thrombosis formation of, 370–374 endothelial damage in, 370–371, 371f, 372f hypercoagulability in, 371–372, 374f, 375f venous stasis in, 371 shoulder dislocation and, 1140, 1141f Throwing. See Overhead throwing Thumb, 1398–1403. See also Finger(s) Bennett’s fracture of, 1402–1403, 1402f–1403f, 1411–1412 carpometacarpal joint of dislocation of, 1398 subluxation of, 1398 dislocation of, 1398–1399 extensor tendon injury of, 1401–1402 fracture of, 1402–1403, 1402f pediatric, 1411, 1411f

Thumb (Continued) gamekeeper’s (skier’s), 1399–1401, 1400f, 1415 interphalangeal joint of, dislocation of, 1384, 1386 ligamentous injury of, 1398–1401, 1400f, 1401f, 1415 mallet, 1401–1402 metacarpophalangeal joint of collateral ligament injury at, 1399–1401, 1400f, 1401f, 1415 dislocation of, 1398–1399 radial collateral ligament injury at, 1401 ulnar collateral ligament injury at, 1399–1401, 1400f–1401f, 1415 Rolando fracture of, 1403 ulnar collateral ligament injury of, 1399–1401, 1400f pediatric, 1415 treatment of, 1400–1401, 1401f volar dislocation of, 1399 Thumb sign test, in posterior cruciate ligament injury, 2214 Thumb spica splint, in scaphoid fracture, 1337, 1338f Thyroid hormone, bone effects of, 72 Tiagabine, in epilepsy, 189, 191t Tibia anatomy of, patellar stability and, 1550 epiphyseal fracture of, 1642–1644, 1643f, 1643t, 1644b external rotation of, in posterior cruciate ligament injury, 1691 fracture of, in knee dislocation, 1752–1753, 1752f growth plate of, 1638 loading of, 1849 osteochondral lesions of, 2153, 2154f stress fracture of, 1849–1866 anatomic factors in, 1850 anterior, 643–644, 644f anterior knee pain with, 1855 biomechanics of, 1849 bone scan of, 1852, 1853f classification of, 1851, 1851t complications of, 1855–1856 computed tomography of, 1853 distal, 2017–2018, 2017f–2018f in children/adolescents, 1970, 1972f magnetic resonance imaging in, 2015, 2016f radiography in, 2014, 2015f–2016f radionuclide imaging in, 2014–2015, 2015f treatment of, 2016t dreaded black line in, 1852, 1852f evaluation of, 1851–1853, 1852f in female, 1849–1850, 1851–1852, 1856 magnetic resonance imaging of, 1852–1853, 1853f medial, 643, 652, 652t in military recruit, 1849, 1850, 1856 nonoperative treatment of, 1854 operative treatment of, 1854–1856 pain in, 1851 pathogenesis of, 1850–1851 physical examination in, 1852 radiography in, 1852, 1852f radionuclide imaging in, 545–546, 545f recurrence of, 1856 return to play after, 1856 risk factors in, 1849–1850 shoewear and, 1850, 1898–1899 sport-specific factors in, 1850 surface-related factors in, 1850

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Tibia (Continued) training errors and, 1851 treatment of, 1854–1856 in untrained athlete, 1856 stress syndrome of (shin splints), 15, 652, 652t, 1857. See also Chronic exertional compartment syndrome radionuclide imaging of, 545, 545f translation of with non–weight-bearing knee extension, 222 in posterolateral corner injury, 1728 with straight leg raise, 222 Tibial artery injury, high tibial osteotomy and, 1832 Tibial eminence, fracture of, 1677–1679, 1678f, 1679f treatment of, 1677–1679 Tibial intercondylar artery, 1645 Tibial nerve entrapment of, 631–632, 2044, 2045f, 2047, 2057–2059. See also Tarsal tunnel syndrome injury to, 1752 posterior, 2043–2044 decompression of, 2058–2059, 2059f entrapment of, 2044, 2045f Tibial osteotomy high. See High tibial osteotomy in knee arthritis, 1792 in posterior cruciate ligament and posterolateral corner injury, 1744–1747, 1746f Tibial plateau anatomy of, 1748 fracture of computed tomography of, 542, 543f high tibial osteotomy and, 1832 in knee dislocation, 1752–1753 Tibial spine fracture, in children/adolescents, 469, 472f Tibial tendon anterior anatomy of, 1976 blood supply of, 1976 injury to, 1975–1977 evaluation of, 1976–1977 magnetic resonance imaging in, 560, 562f treatment of, 1977 longitudinal split tears in, 1977 posterior anatomy of, 1978–1979 blood supply of, 1978–1979 dysfunction of, 631–632 end-to-end repair of, 1980 injury to, 30, 1977–1981 end-to-end repair in, 1980 evaluation of, 1979, 1979f flexor digitorum longus transfer in, 1980 magnetic resonance imaging in, 562, 563f single heel rise test in, 1979, 1979f too-many-toes sign in, 1979, 1979f treatment of, 1980–1981 triple arthrodesis in, 1980 tendinitis of, 1981–1983, 1982f tenosynovitis of, 538, 538f Tibial tubercle osteochondrosis of (Osgood-Schlatter ­disease), 599, 1526f, 1527–1529 position of, 1555, 1555f, 1556f stress apophysitis of, 599 Tibiocalcaneal ligament, 1913f Tibiofemoral rotation test, in varus ­malalignment, 1806–1807, 1807f

Index Tibiofibular ligament anterior-inferior, 1913f, 1931, 1931f magnetic resonance imaging of, 573, 574f Tibiofibular syndesmosis, 1938–1940, 1939f injury to. See Ankle sprain, high (syndesmosis) Tibionavicular ligament, 1913f Tibiotalar ligament, 1913f Tillaux fracture, 597, 598f, 1965, 1966f, 1968–1969 Timolol, in complex regional pain syndrome, 363t Tinea capitis, 198, 198f, 199–200, 200t Tinea corporis, 198, 199, 199f, 200t Tinea cruris, 198, 199, 199f, 200t Tinea gladiatorum, 198 Tinea infection, 198–200, 198f–200f, 200t Tinea pedis, 198, 199f, 200t Tinel’s sign in cubital tunnel syndrome, 1312 in sesamoid dysfunction, 2090 Tissue Banking Project Team, 138 Tissue engineering, in ligament injury ­treatment, 39 Titin, 4–5 Tobacco use/abuse, 426–428, 428f Toe(s) fracture of pediatric, 1970, 1971f stress, 2028 great. See Hallux deviation of. See Hallux valgus; Hallux varus metatarsophalangeal joint injury to. See Turf toe sesamoids of. See Sesamoid(s) lesser, 2115–2132 anatomy of, 2115–2116 bunionettes of, 2132–2142 anatomy of, 2132, 2132b, 2133f classification of, 2133, 2134b evaluation of, 2133–2134, 2134b imaging in, 2134, 2134b nonoperative treatment of, 2134, 2134b operative treatment of, 2134–2141 care after, 2139–2141, 2140f, 2141f complications of, 2140–2141 diaphyseal metatarsal osteotomy in, 2135–2137, 2138b, 2139f, 2140f distal chevron osteotomy in, 1966f, 2135–2136, 2135b, 2137b distal metatarsal osteotomy in, 2135–2136, 2135b distal oblique osteotomy in, 2135– 2136, 2135b, 2136b, 2136f, 2137f lateral condylectomy in, 2134–2135, 2134b, 2135f, 2140, 2140f return to play after, 2141 physical examination in, 2134, 2134b corns of, 2116, 2117f treatment of, 2117f, 2124, 2129 deformity of, 2115–2132 capital oblique osteotomy for, 2114f, 2131, 2132f classification of, 2117–2119 claw, 2118, 2118t, 2119, 2120b, 2121–2125, 2122f, 2129 evaluation of, 2119–2121 flexor tendon transfer for, 2123, 2125f hammer, 2117, 2118t, 2119f, 2120b, 2121–2125, 2122f, 2123f, 2124b imaging in, 2121 mallet, 2118–2119, 2118t, 2119f, 2120b, 2121–2125, 2123f, 2124f, 2128–2129 metatarsal head arthroplasty for, 2124, 2130f

Toe(s) (Continued) metatarsal osteotomy for, 2124 nonoperative treatment of, 2121, 2121f, 2122f operative treatment of, 2121–2125, 2122f, 2123b, 2123f–2125f management after, 2130–2132, 2131b partial proximal phalangectomy for, 2124, 2129f physical examination in, 2120–2121, 2120b, 2120f, 2121f recurrence of, 2131 return to play after, 2132, 2132f interphalangeal joint motion in, 2181, 2182f medial deviation of, 2116, 2117f, 2119 treatment of, 2123–2124, 2128f metatarsophalangeal joint subluxation/dislocation of, 2116, 2119, 2120f, 2126f treatment of, 2123–2124, 2125b, 2126f–2130f molding of, 2131 taping of, 2121, 2122f turf. See Turf toe Toe cap, 2121, 2121f Toenail, 2096, 2096b, 2097f alcohol matrixectomy for, 2099–2100 complete avulsion procedures for, 2099, 2101f ingrown. See Ingrown toenail partial nail plate avulsion of, 2099, 2100f phenol matrixectomy for, 2099–2100, 2101, 2102, 2105f plastic nail wall reduction for, 2099, 2101f Thompson-Terwilliger procedure for, 2099–2100, 2104f Winograd procedure for, 2099, 2102f Zadik procedure for, 2099, 2103f Tomography, 534. See also Computed ­tomography (CT) in medial clavicular fracture, 804, 804f in sternoclavicular joint injury, 804 of thoracolumbar spine, 728 Too-many-toes sign, 1979, 1979f Tooth (teeth), on-field injury to, 525 Topiramate, in epilepsy, 189, 191t Total Gym, in knee rehabilitation, 259–260, 260f Trachea displacement of, sternoclavicular joint injury and, 821, 821f on-field injury to, 525 Traction in cervical spine fracture, 675 in proximal humeral physeal fracture, 1077 in rotary atlantoaxial subluxation, 707 Traction spurs, of elbow, 621–622 Training. See also Therapeutic exercise(s) altitude, physiological effects of, 220 endurance, physiological effects of, 215, 216t immune system effects of, 148, 148f resistance, physiological effects of, 214–215, 216t skeletal muscle response to, 213–214, 214t, 215f Tramadol, in complex regional pain syndrome, 364 Trandolapril, 160t Transcutaneous electrical nerve stimulation, in pain management, 229–230, 230f Transforming growth factor-β, in fracture ­healing, 79–80 Transforming growth factor-β1, in muscle injury, 16 Transverse process, lumbar, fracture of, 734

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

liii

Transverse scapular ligament, 1120, 1120f Trapezium anatomy of, 1319–1320, 1319f fracture of, 1345–1348, 1349f pediatric, 1370 Trapezius lower, therapeutic exercise for, 240, 240f–241f middle, therapeutic exercise for, 240, 240f pediatric, 785–786, 786f scapular attachment of, 859f, 886f strengthening exercise for, 240, 240f–241f, 1003, 1005f Trapezoid, anatomy of, 1319–1320, 1319f Trapezoid ligament, 827 Traveling, 531 jet lag with, 457–460, 458f adjustment avoidance and, 460 bright light exposure and, 458, 459f melatonin and, 459–460 performance and, 460 room light and, 458–459, 459t, 460t Trazodone, in complex regional pain ­syndrome, 363t Trendelenburg gait, 1555 Trephination, in meniscal injury, 1608 Tretinoin, in acne, 205 Triangular fibrocartilage complex (TFCC), 1433, 1433f, 1435, 1435f Triangular fibrocartilage complex (TFCC) tears, 575, 1435–1442 classification of, 1435–1436, 1436f history in, 1436 pediatric, 1374–1375, 1374b physical examination in, 1436 radiography in, 1436 return to play after, 1437, 1438, 1439, 1441 treatment of, 1436–1442 type 1A, 1435–1437, 1436f, 1437f type1B, 1435–1438, 1436f–1438f type1C, 1436, 1436f, 1438–1439, 1439f type 1D, 1436, 1436f, 1439 type 2, 1440–1441, 1441f Triceps dip exercise, shoulder injury and, 249 Triceps muscle anatomy of, 1158, 1160f, 1161f, 1170 rupture of, 1297 scapular attachment of, 858, 858f–859f, 886f Triceps tendinitis, 1207 diagnosis of, 1207 treatment of, 1207–1209 Triceps tendon magnetic resonance imaging of, 567 rupture of, 1170–1172, 1207–1209 complications of, 1172 diagnosis of, 1170–1171, 1171f, 1207 mechanism of, 1207 rehabilitation after, 1172, 1172t, 1208 treatment of, 1171–1172, 1171f, 1207–1209, 1208f, 1209f snapping, 1226 throwing-related injury to, 1221, 1226 Trigger, in pain dysfunction syndrome, 351 Trigger finger, 626 Trigger point injection, in low back pain, 732 Trimethoprim-sulfamethoxazole, in urinary tract infection, 153 Triplane fracture, 1966, 1967f, 1968–1969 Triquetrum anatomy of, 1319–1320, 1319f fracture of adult, 1349–1350, 1351f pediatric, 1368–1369, 1369f Trochanteric bursitis, 1455–1457, 1456f

liv

Index

Trochlea, femoral anatomy of, 1549–1550, 1550f, 1560, 1560f depth of, on radiograph, 1539, 1561, 1561f dysplasia of, 1550, 1560–1561, 1561f classification of, 1561, 1561f sulcus angle in, 1563–1564, 1564t treatment of, 1568–1569 Trochlear osteotomy, in trochlear dysplasia, 1568–1569 Trochleoplasty, in trochlear dysplasia, 1568–1569 Tropomyosin, 4, 5t Troponin, 4, 5, 5t Trunk stabilization program, 341–349, 342t aerobic exercise in, 342t, 349 ball exercises in, 342t, 345 bridging exercise in, 342t, 344 dead bug exercise in, 342t, 343 partial sit-ups in, 342t, 343 postural exercises in, 348–349 prone exercises in, 342t, 345 quadruped exercises in, 342t, 346 stabilization exercises in, 343–348 wall slide exercises in, 342t, 347 water running in, 349 weight training in, 349 TT-TG measurement, in patellofemoral disorders, 1565 Tubercle sulcus angle, 1555, 1556f Tuck jumps, 312, 312f, 330 Tumor bone, 74t, 77t, 547, 547f soft tissue, 558, 560f spinal, back pain and, 757, 758f Turf. See Playing surface Turf toe, 2081–2087 anatomy of, 2081–2082, 2082b, 2082f classification of, 2082, 2083b, 2083t complications of, 2087 evaluation of, 2083 flexibility/inflexibility and, 2182 footwear and, 2087 grade of, 2082, 2083b, 2083t, 2084 history in, 2083 imaging in, 2083–2084, 2084f, 2086f nonoperative treatment of, 2084–2085, 2085f, 2086b operative treatment of, 2085, 2086b physical examination in, 2083, 2083b playing surface and, 2198 return to play after, 2087, 2087b shoe design and, 2082, 2082f, 2087, 2188–2189 treatment of, 2083t Twitch, 7–8, 8f Two-by-two table analysis, 108–109, 109f

U Ulna fracture of, in children, 594f, 597 proximal articular surface of, 1189–1190, 1189f fracture of, 1271–1276 evaluation of, 1271–1273 nonunion of, 1278 operative treatment of, 1273–1276 arthrosis after, 1278 complications of, 1276, 1277–1278 infection after, 1278 instability after, 1277–1278 nonunion after, 1278 plate and screw fixation in, 1275 rehabilitation after, 1276–1277

Ulna (Continued) ulnar nerve injury after, 1277 wound problems after, 1278 Monteggia fracture of, 1259, 1260, 1260f, 1273, 1287 osteophyte of, 1223–1224, 1223f Ulnar artery, 1160f Ulnar collateral ligament (UCL) anatomy of, 1301, 1302f reconstruction of, 1222–1224, 1223f, 1237, 1238f, 1308, 1310, 1310f stabilizing effect of, 1193, 1194f, 1230–1231, 1231f Ulnar collateral ligament (UCL) injury, 619–620. See also Thumb, ulnar collateral ligament injury of magnetic resonance imaging in, 576, 577f pediatric, 1237, 1238f vs. tendinitis, 1221 throwing-related, 1221–1224, 1237 pediatric, 1237, 1238f Ulnar nerve anatomy of, 1159 injury to, 622–623, 1311–1315, 1361. See also Cubital tunnel syndrome in coronoid fracture fixation, 1271 in distal humeral fracture repair, 1256, 1277 after elbow trauma, 1277 in pediatric supracondylar fracture, 1282 throwing-related, 1226 Ulnar shortening, in gymnast, 1376, 1376f Ulnar styloid, fracture of, 468, 470f Ulnar styloid impaction syndrome, 1445–1447, 1446f return to play after, 1447 Ulnar styloid process index, 1446, 1446f Ulnar variance, in gymnast, 1376, 1376f Ulnocarpal impaction syndrome, 1440–1441 Ulnocarpal impingement, in gymnast, 1376 Ulnohumeral instability, after proximal ulnar fracture, 1276 Ulnolunate ligament, 1320 Ulnotriquetral ligament, 1320 Ultrasonography, 537–538 in Achilles tendon injury, 630, 1998, 2003–2004 in Baker’s cyst, 537, 538f in cyst aspiration, 585 in deep venous thrombosis, 376, 377f, 383 in elbow heterotopic ossification, 1293 in foreign body, 538, 539f of glenohumeral joint, 951–953, 952f in hamstring strain, 1487 in knee dislocation, 1751 in lateral epicondylitis, 1199 of neonatal hip, 590, 590f in Osgood-Schlatter disease, 1528 in patellar dislocation, 1541 of patellar tendon, 1519, 1519f pediatric, 590, 590f cartilage on, 592–593 soft tissues on, 593 in peroneal tendinitis, 1989 in popliteal artery entrapment, 1839, 1839f in quadriceps strain, 1494 of rotator cuff, 951–953, 952f, 953t, 1000–1001 in suprascapular nerve injury, 1122 in tibial tenosynovitis, 538, 538f transducer for, 537 in vascular injury, 1139 Ultrasound therapy in Achilles tendon injury, 1998 in fracture healing, 80 in lateral epicondylitis, 618

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

Ultrasound therapy (Continued) nonthermal, 237 in rehabilitation, 236–237 in rotator cuff disorders, 1007 thermal effects of, 237 in tibial stress fracture, 1854 Underage drinking, 424–425 United States Food and Drug Administration, 137, 138 Universal precautions, 153 Unstable surface training, in knee ­rehabilitation, 271, 272f Urethra, on-field injury to, 527 Urinary tract infection, 152–153 Urticaria, 205, 205f

V Vaccination, Lyme disease, 155–156 Vagal nerve simulator, in epilepsy, 189–190 Valacyclovir, in herpes gladiatorum, 197 Valgus extension overload, 621–622, 1224–1225. See also Overhead throwing injury Valgus instability, 619–620, 620f Validity, 100, 101f, 2217–2218 Valleix phenomenon, 2057 Valproate, in epilepsy, 188–189, 191t Valproic acid, in complex regional pain ­syndrome, 363t Valsartan, in hypertension, 160t Vancomycin, 387, 387t Variable, statistical, 2218 Variance, 112 analysis of, 112–113, 113f, 118 Varus malalignment, 1801–1803, 1802t, 1803f clinical presentation of, 1803–1804 correction wedge determination in, 1810–1811, 1812f, 1813f evaluation of, 1803–1814, 1805f, 1806t gait analysis in, 1808–1809, 1809f–1810f, 1811f gap angle in, 1812, 1813f imaging and calculations in, 1809–1814, 1811f–1814f, 1814t–���������� 1815t open wedge angle measurement in, 1812–1814, 1813f–1814f, 1814t–1815t patellar height measurement in, 1811 physical examination in, 1804–1808, 1805f, 1806t, 1807f, 1808f treatment of, 1814–1835. See also High tibial osteotomy weight-bearing line ratio in, 1810, 1811f, 1812f Varus recurvatum test, in varus malalignment, 1807–1808 Varus stress test, in posterolateral corner injury, 1727, 1727f, 1727t Vascular injury in distal humeral fracture, 1282 in elbow dislocation, 1263, 1304 in knee dislocation, 1750, 1751, 1753, 1753f, 1764–1765 on-field, 527 shoulder, 1137–1142, 1138f, 1139f, 1140f in sternoclavicular joint sprain, 1140 ultrasonography in, 1139 Vasoconstriction, in complex regional pain syndrome, 355, 360 Vasodilation cold-induced, 500 in complex regional pain syndrome, 355, 360 Vastus medialis obliquus neuromuscular activation exercise for, in knee rehabilitation, 256–257, 257f in recurrent patellar dislocation, 1556

Index Vastus muscles, 1485, 1485f, 1550–1551, 1551f. See also Quadriceps muscle Vectors, 86 force, 86, 90f moment/torque, 86 Vegetarianism, iron deficiency and, 479 Vein. See also specific veins anatomy of, 370, 371f circulatory stasis in, 371. See also Thrombosis Venography, in deep venous thrombosis, 376, 377f Venous thromboembolism. See Deep ­venous thrombosis; Pulmonary embolism; Thrombosis Ventilation, high-altitude effects on, 503 Ventilation-perfusion scan, in pulmonary embolism, 378, 379f Ventricular fibrillation, 163, 164f. See also ­Sudden death Verapamil in complex regional pain syndrome, 363t in hypertension, 160t Verruca plantaris, 2107b, 2108f Vertebral injury. See Cervical spine injury; Thoracolumbar spine injury Vertebroplasty, 585 Viral infection, cutaneous, 196–198, 197f, 198f Virchow’s triad, 370–374, 371f, 375f Viscoelasticity, 95–96, 95f articular cartilage, 48 meniscus, 59–60, 60f skeletal muscle, 18, 19f tendon, 24–25, 25f Visual acuity on-field examination of, 525 preparticipation examination for, 509 Vitamin A, requirements for, 406b Vitamin B6, requirements for, 406b Vitamin B12, requirements for, 406b Vitamin C deficiency of, 72–73, 73t requirements for, 406b Vitamin D in bone metabolism, 71, 71t deficiency of, 72–73, 73t, 75t, 77t, 79b excess of, 74t, 77t requirements for, 406b Vitamin E, requirements for, 406b Vitamin K, requirements for, 406b Vitamin K antagonist prophylactic, in venous thromboembolism, 378–384, 381t in venous thromboembolism, 384–385, 384t Volar intercalary segmental instability (VISI), 1322, 1324f Volleyball players, nerve entrapment in, 616–617

W Wake maintenance zone, 450, 454–456, 455f Walking. See also Gait ankle joint motion in, 1866, 1867f vs. running, 1865, 1865f, 1866, 1866f subtalar joint motion in, 1868, 1869f transverse tarsal joint motion in, 1869, 1870f Wall slide exercises, in trunk stabilization, 342t, 347 Warfarin prophylactic, in venous thromboembolism, 378–384, 381t, 382f in venous thromboembolism, 384–385, 384t Warm-up exercise in exercise-induced bronchospasm, 184

Warm-up exercise (Continued) in injury prevention, 19 for therapeutic exercises, 276 Warming therapy, in frostbite, 203 Warts, 197–198, 2107b, 2108f Water. See also Fluid(s) cartilage, 41 intake of, 401 with creatine use, 418 requirements of, 401–402, 402t meniscal, 57, 57t Water intoxication, 529 Water running exercise, in trunk stabilization, 349 Watson-Jones procedure, 1924–1925, 1925f Watson’s maneuver, in scapholunate ligament injury, 1324–1325, 1325f Weakness in complex regional pain syndrome, 357 in rotator cuff disorders, 996 Weight gain of, 406, 408b loss of, 406, 407b, 409 in hypertension, 159t making, 406, 406b in preparticipation examination, 509 Weight-bearing, in ACL rehabilitation, 1671 Weight-bearing exercise in female athlete, 481 patellar effects of, 226–227, 226f, 227f Weight training. See also Strengthening ­exercise/training in trunk stabilization, 349 Wet bulb globe temperature, 528 White blood cells, exercise effects on, 147, 148t White coat hypertension, 157 Windlass mechanism, 2043 in gait, 1870, 1871f Winograd procedure, 2099, 2102f Wolff-Parkinson-White syndrome, 170 Women. See Female athlete World Anti-Doping Agency, 413 Wright’s hyperabduction test, in thoracic outlet syndrome, 1132, 1132f Wrisberg, ligament of, 1597–1598, 1598f, 1614, 1685, 1686f Wrist. See also Wrist injury Allen’s test of, 1320–1321, 1321f anatomy of, 1319–1320, 1319f, 1320f arthroscopy of. See Arthroscopy, wrist chondral lesions of, 1447–1449, 1448f chondromalacia of, 1447–1449 computed tomography of, 543, 544f development of, 1364, 1364t gymnast, 1366f, 1375–1376, 1376f Kienböck’s disease of, 1376–1377, 1377b, 1378f ligaments of, 1319–1320 injury to, 1321–1335. See also Wrist injury, ligament and at specific ligaments magnetic resonance imaging of, 575, 576f loose bodies of, 1447 magnetic resonance imaging arthrography of, 536 overuse injury of, 624–626 playing cast for, 1362 rheumatoid arthritis of proximal row carpectomy in, 1449–1450 radial styloidectomy in, 1449 synovectomy in, 1443–1444 splints for, 1361–1362 taping of, 1362 tendons of injury to, 1351, 1354–1357, 1355f, 1356f

Volume 1: pages 1 to 1156; Volume 2: pages 1157 to 2220.

lv

Wrist (Continued) magnetic resonance imaging of, 568–569, 570f vascular anatomy of, 1320, 1320f Wrist extensor tendinosis, 617–619, 618f Wrist injury adult, 1319–1362 fracture, 1335–1351. See also specific carpal bones ligament, 1321–1335. See also at specific ligaments capitolunate angle in, 1322, 1322f clinical manifestations of, 1322, 1324 Mayfield classification of, 1321–1322, 1323f radiography of, 1322, 1322f–1323f, 1324f scapholunate angle in, 1322, 1322f magnetic resonance imaging of, 1321 neural, 1359–1361 palpation in, 1320 physical examination of, 1320–1321, 1321f playing casts for, 1362 radiography of, 1321 range of motion in, 1320 rehabilitation for, 1362 splints for, 1361–1362 taping for, 1362 tendon, 1351, 1354–1357, 1355f–1356f vascular, 1357–1359, 1357f–1358f pediatric, 468, 470f, 1363–1377 epidemiology of, 1363 fracture, 1364–1371. See also specific carpal bones gender and, 1363 ligamentous, 1371–1377, 1372f, 1373f, 1376f overuse and, 1363–1364 physeal, 1364, 1365, 1375–1376 risk factors for, 1363–1364 sport type and, 1363 types of, 1364 Wrist roller training, in elbow rehabilitation, 254, 254f Wry neck, 706–707, 707f

X Xenograft, 38 Xeroradiography, in retrocalcaneal bursitis, 2033

Y Yergason’s test in rotator cuff disorders, 997, 1000f in SLAP lesion, 1024 Yohimbe, 409 Youth Risk Behavior Survey, 463, 466, 467 YoYo flywheel ergometer, in hamstring strain prevention, 337

Z Z band, 4f, 5, 6f, 13f Zadik procedure, 2099, 2101, 2102, 2103f Zeitgeber, 444 Ziegler, John, 411 Zinc, requirements for, 406b Zonisamide, in epilepsy, 191t Zygapophyseal joints, 718, 719f